In The Workshop of History - Furet, François, 1927-1997 - 1984 - Chicago - University of Chicago Press - 9780226273365 - Anna's Archive

Download as pdf or txt
Download as pdf or txt
You are on page 1of 286

net{as

àNi a st

=
frere
fat

reer f
ue there sees
re HU
#4
fete
Digitized by the Internet Archive
in 2023 with funding from
No Sponsor

https://archive.org/details/inworkshopofhist000Ofure
*

‘ - ae

| oo ri É
. :* | +a 7

i
7 bd
ol |
. ?
| 4
La
-
A . Tr.

ee

,
Le *
Li * L

,
. * L
FRANCOIS
FURET
Translated by Jonathan Mandelbaum

The University of Chicago Press - Chicago and London


FRANCOIS FURET is president of the Ecole des Hautes Etudes en Sciences Sociales, Paris. His
published books include Livre et société dans la France du XVIIIe siècle (1965) and Penser la
Révolution frangaise (1978).

This book was first published as L’ Atelier de l’histoire, © 1982, Flammarion, Paris.

The University of Chicago Press, Chicago 60637


The University of Chicago Press, Ltd., London
© 1984 by The University of Chicago
All rights reserved. Published 1984
Printed in the United States of America
93 92 91 90 89 88 87 86 85 84 54321

Library of Congress Cataloging in Publication Data


Furet, François, 1927—
In the workshop of history.
Translation of: L’atelier de l’histoire.
Includes bibliographical references and index.
1. Historiography—France—History. 2. Historiography
—History. I. Title.
D13.5.F8F8713 1984 907'.2 84-2638
ISBN 0-226-27336-9
CONTENTS ES

Introduction

Part One: History Today

French Intellectuals: From Marxism to Structuralism


Quantitative History
From Narrative History to Problem-oriented History
Lh
©
D
= History and Ethnology

Part Two: History in Seventeenth- and Eighteenth-Century


European Culture

5 The Birth of History in France "#4


6 Book Licensing and Book Production in the Kingdom of 99
France in the Eighteenth Century
7 Two Historical Legitimations of Eighteenth-Century 125
French Society: Mably and Boulainvilliers
8 Civilization and Barbarism in Gibbon’s History 140

Part Three: America and the Idea of Democracy

9 From Savage Man to Historical Man: The American 153


Experience in Eighteenth-Century French Culture
10 The Conceptual System of Democracy in America 167
11 Today’s America: A Metamorphosis of the Idea of 197
Equality
vi Contents

Part Four: Aspects of Modern Jewish History

To the Reader 209


12 Jews and French Democracy: Some Recent Books 211
13 Gershom Scholem and Jewish History 225
14 Israel and the French Left: The Misunderstanding 230
15 Israel, Zionism, and the Diaspora 234

Notes 243

Index 253
INTRODUCTION”

French historians of my generation have led a happy existence. They


have easily found positions; they have had time to read and write; their
work has been well received both in France and abroad; and they consti-
tute a relatively homogeneous group, in which relations are more often
friendly than not. Above all, they are the heirs to a historiographical
tradition that is itself a success story. Renewed between the wars by the
Annales under the aegis of Marc Bloch and Lucien Febvre, the old
discipline arrived at virtual unanimity during the 1950s and 1960s among
historians of my age. This consensus expressed itself in a collective
agreement to extend the traditional boundaries of history, as “historical”?
projects were launched in areas previously staked out by other disci-
plines, such as the social sciences. As the historians’ guild has little taste
either for epistemology or for the history of its own history, it has usually
satisfied itself with this conquering spirit of unanimity, which does con-
tinue to stimulate very active, and sometimes brilliant, research. But such
unanimity also encompasses, by definition, an almost infinite variety of
scientific practices: the sociological unanimity of the discipline masks its
epistemological fragmentation. It is this contrast that I have sought to
investigate in depth in the course of my professional experience and that
informs this book.
Compared to the social sciences, history can claim greater seniority
and legitimacy, and its recent renewal has not invalidated its credentials
in these respects. History remains a discipline inseparable from the
nation, essential to the meaning of nationhood. Hence, history still has its
rules, usages, university chairs, and learned societies, all of which provide

Pages 1-19 of this Introduction first appeared as “En marge des Annales: histoire et sciences
sociales,” Le Débat, no. 17 (December 1981), 112-26. A slightly different form of the
English version was published in the Journal of Modern History.
2 Introduction

it with a common language and a professional consensus. Its prestige has


not been impaired by the fact that, in the course of the fifty years that
have elapsed since the first issue of the Annales, it has gradually ceased to
see itself as the interpreter of the national phenomenon. Having
espoused “modern” concerns, history now explores the secrets of
societies rather than those of nations; it has become that much harder to
teach, but also more rewarding to construct and write. Moreover, history
has still kept an eye on the nation, even if it reconstructs the nation’s past
from new standpoints, using society to shed light on nationhood. In
France, history has never been so solidly entrenched institutionally and
never enjoyed so great an audience among the educated public, whatever
the temporary problems raised in secondary education by the increasing
indeterminacy of the field.
There is, moreover, a certain logic to the fact that the only major
research and teaching institution founded in France since the war in the
field of social and human sciences owes its existence precisely to the
Annales historians: the Sixth Section of the Ecole Pratique des Hautes
Etudes, founded by Lucien Febvre in 1948, which in 1975 became the
Ecole des Hautes Etudes en Sciences Sociales. The transformation of a
journal into an institution expresses and codifies history’s sponsorship of
the social sciences—one of the functions that history has taken on by way
of rewarding itself for its renewal. This new institution has certainly
played an important part in perpetuating history’s guiding role.
But the academic crystallization of what was more than a journal—and
less than a doctrine—has above all accredited the mistaken notion that
the Annales historians as a group share a common and unified concept of
the discipline, a concept in conflict with the traditional approach. Institu-
tions have their own logic; there is nothing like their rivalries to confer an
imaginary degree of intellectual coherence to political, collective, profes-
sional, or personal differences. In this respect, the petty institutional war
and the great symbolic clash that from the very outset have characterized
the relations between the former Sixth Section of the Ecole Pratique des
Hautes Etudes and the former Facultés des Lettres in the universities—
chiefly the Sorbonne—have concealed a twofold intellectual phe-
nomenon, one that would have helped to moderate the conflict had it not
become more institutional than intellectual. First, during the 1950s and
1960s, the topics and research methods advocated by the founders of the
Annales gradually spread well beyond their initial home. They became
the common property of historians, regardless of their institutional alle-
giances. Second, historians of the Ecole stricto sensu were working in
directions that were too diverse for those historians to be easily assem-
bled under a single intellectual banner.
By their diversity, they were at least being faithful to the undogmatic
approach of the Annales since its inception. They were taking the utmost
advantage of the journal’s specific feature, which was a simple and
powerful idea: history was to be freed to wander in every field. Ulti-
Introduction 3

mately, that is the intellectual heritage that historians such as Alain


Besançon, Emmanuel Le Roy Ladurie, Jacques Ozouf, Denis Richet
(just to mention those who were closest to me in my generation), and
myself received from Lucien Febvre and Mare Bloch via Fernand
Braudel. It is a crucial heritage, since it gave us, from both the intellectual
and the professional points of view, exceptional opportunities and free-
dom. But this, simply, was our only legacy. We settled into a historiogra-
phy that was already dominant when we were taking the agrégation
examination and whose very success, embodied by the existence of the
Sixth Section, gave it an even greater receptive capacity. Thus, the
Annales fulfilled its role more than ever, since that role is synonymous
with hospitality and openness. From our point of view, the Annales
offered an almost boundless range of topics and methods—a heaven-sent
oasis on the path away from Stalino-Marxist historicism, whose power to
mystify we had only recently come to recognize.
Indeed, when I try to define, twenty-five years later, what we still have
in common, apart from reminiscences and feelings, I wonder whether our
prolonged adolescence in the ranks of the Communist party did not play a
greater role than our career as historians in the Ecole des Hautes Etudes.
From the former, we acquired not only a treasure trove of rather embar-
rassing memories but a set of intellectual and political references that
mark and date our common inoculation. In the course of the latter, we
received the blessing of a providentially welcome and open institution,
but we did not reach a new historiographical consensus. When abroad, I
am often asked what constitutes the Annales School. It is a difficult
question for me, because it evokes the somewhat ritual aspect involved in
simply managing cultural capital; I am tempted to answer that the ques-
tion is practically meaningless. Lucien Febvre and Marc Bloch would
have provided—and did provide—brilliant answers, as Fernand Braudel
did later. But the point is, they did reply. All the battles won against
professional narrow-mindedness and self-satisfaction—battles that even-
tually led to the founding of an institution—have gradually eliminated
their own raison d’étre. The Annales School is at the apogee of its
reputation in France and abroad; for lack of specific adversaries, it meets
with nothing but widespread praise. It is lauded for being both conceptual
and narrative, interested in individuals and searching for laws, open to
Marxism and well disposed toward Marxism’s opponents, and so on.
There is some truth to these statements, since they can all be supported
by good examples; but by now they describe only a hegemony of in-
fluence and reputation, not a school of thought nor even, any longer, a
collective spirit.
There was, in fact, no school of thought to begin with. It would be vain
to search for the traces of a doctrine or a favorite mode of explanation in
the prewar Annales. History, a polymorphous discipline whose object
cannot be defined, let alone circumscribed, remains faithful to its fun-
damental indeterminacy down to the very intellectual revolution that is
4 Introduction

supposed to have transformed its methods and subject matter. Nor is this
revolution itself easy to define, and it emerges more sharply by contrast to
its adversaries’ profile than by virtue of its founders’ concepts. That is
perhaps one of the reasons for the frequency of polemical arguments in
the prewar and postwar Annales, especially when Lucien Febvre was
involved. The martial vocabulary, the call to battle in defense of worthy
causes, the sense of adventure intellectually heightened by the breathless
prose—all these are telling signs of the extent to which the opening of the
discipline to new objects was a reaction against traditional historiogra-
phy. Indeed, now that these new objects have become the common
property of our profession, it is hard for us to imagine how eccentric they
once appeared. What, then, is the Annales School if the territory it
claimed for history has become a public domain?
It seems to me that the simplest way of describing the ambition and
development of the school is to consider jointly two objectives—two
ideas: first, that history should add to its subjects and methods by borrow-
ing from neighboring disciplines and even by the temporary abolition of
divisions between disciplines; and, second, that it should nevertheless
remain an all-embracing and ecumenical discipline, bringing together the
conditions required for the fullest understanding of social phenomena.
Neither of these two ideas is simple or clear; nor is their compatibility
self-evident, unless one regards it as a wish, a program, or an ultimate
ambition. Instead of paving the way for its intellectual unity, history’s
borrowings from the social sciences only add to the diversity of the
discipline by expanding its field of inquiry almost to infinity. But, while
these borrowings provide history with greater methodological eclecti-
cism, they do not necessarily guarantee additional knowledge—unless, of
course, history would precisely exploit this greater conceptual indeter-
minacy in order to assert its superiority over the social sciences. By
borrowing right and left, history does not tie itself to any one of its
creditors; on the contrary, it claims to bestow its own preeminent dignity
on all of them. Thus, in its comprehensive embrace, history absorbs the
social sciences instead of simply representing their temporal dimension.
In order to examine the situation of history today, a half century after
the founding of the Annales, one can therefore begin by examining its
relation to the social sciences. This is probably the least unreliable
instrument for a methodical assessment.

The social sciences aim, like Marxism, to explain social facts. But they
do not constitute a unified body of doctrines, nor even a corpus of partial
interpretations that could be unanimously accepted. The social sciences
delimit specialties, or objects of study, or approaches to a problem,
without ever securing the consensus of specialists. Consequently, they
have renounced the ambition of providing comprehensive explanations
and scientific forecasts of contemporary phenomena. Indeed, the social
sciences define their positivity by their distrust of the philosophy of
Introduction 5

history; to this extent, they cannot escape—any more than any other
intellectual examination of our world—the fact of being rooted in their
time. They express this rootedness by recognizing their limitations and no
longer by voicing vast ambitions.
However, the social sciences have not easily accepted this renuncia-
tion of their capacity, or of their claim, to provide a total explanation. All
they need is to adopt the singular form—social science—in order to
regain the prestige of the scientistic prophecy: from Auguste Comte to
Durkheim, the sociologist propounded and embodied the notion that not
only is it possible to study man in society, but the analysis of his social
determinants can alone provide the key to the celebrated “essence” of
man. More recently, the same hopes have been pinned on structural
anthropology, in circumstances that one of the essays in this book
attempts to elucidate: the search for constants within the diversity of
social manifestations in space (rather than time) can or must lead to a
unitary theory of man.
In its renewed form, history remains actively engaged in that rivalry
among the social sciences in which each aims, by claiming a higher
objectivity and an all-embracing knowledge, to subsume the others. Its
claims are no longer “historicist.” It no longer regards the time scale as
the yardstick for stages in human progress, stages punctuated by the
creation of nation-states and the spread of ‘‘civilization,”’ that is, of the
European model. On the contrary, history today is no longer endowed
with the primal, implicit significance attributed to time; it has abandoned
the linear vision that made it a royal discipline whose task was to weigh
the merits of the various ‘‘periods” of the past. At the same time, history
has turned to a broader range of topics. It confers its dignity less selec-
tively: everything has become historical, from an apparently ‘“‘nonhistor-
ical” gesture to the bill of fare in an inn to the division of a plot of land.
Thus, history has constantly encountered the social sciences. Indeed, it is
hard to say whether it is the discovery of new research areas that has led
history to an apprenticeship in more specialized disciplines or whether,
on the contrary, it has borrowed from social sciences such as economics
or demography certain objects of research that have already been defined
outside history. The two phenomena have been concurrent and have
influenced each other.
This process, which corresponds very roughly to the establishment and
growth of what is known as the Annales School of history, has been
accompanied by a claim on the part of history to exercise hegemony over
the social sciences, as if history alone, armed with the partial forms of
knowledge developed by neighboring disciplines, were destined to pre-
sent a unified vision of man under the name of “‘total history.” Such a
claim is, in a sense, a way of making up for the fragmentation and
specialization of the objects of historical analysis. Even as it ceases to act
as an organized representation of time, providing mankind with the
vision of its progress, history still maintains the notion of its superiority
6 Introduction

on the grounds that it alone can study certain objects exhaustively, from
every angle—whereas in actual fact, like other disciplines, it deals with
limited objects. Thus, in the midst of a crisis in historicist history, the
historian has fallen back on a less charismatic notion of his learning
without suffering any epistemological loss.
Yet the idea of total history is elusive. Naturally, it no longer denotes
the nineteenth-century view of history as a privileged discipline, designed
to encompass all the significant manifestations of man in society and to
interpret the development of these manifestations as so many necessary
stages in a history of mankind. Total history merely expresses the ambi-
tion of providing a fuller perspective, a more exhaustive description, a
more comprehensive explanation of a given object or problem than can
be provided by the social sciences whose conceptual and methodological
innovations it has borrowed. But if the secret of the flow of time is no
longer vested by definition in history, how can history still be regarded as
a superior discipline? Is this privilege justified by the temporal dimension
that history is thought to bring to the study of social phenomena? All the
social sciences—no less than history—deal of necessity with problems
whose past constitutes the only available laboratory in which they can be
studied. Admittedly, history’s central purpose is to analyze the develop-
ment of any number of those problems, sometimes over very long
periods; but this temporal dimension, however far back it may reach,
does not in itself justify a peculiar privilege of “totality.” It enables
historians to describe and analyze the series of changes and transforma-
tions of the phenomenon in question and thus to introduce, with the aid
of time, an implicitly comparative perspective that is essential, since it
enables one to enrich the concept or concepts used in the analysis—but
no more: temporality cannot lead us to what does not exist, that is, to a
concept capable of explaining every aspect of the object under study.
Social science tout court remains the goal of all the social sciences, but
it cannot constitute the “‘true”’ essence of any of them. If history, after
sociology and along with anthropology, has staked a claim to such intel-
lectual authority, that claim must not be taken for granted; such pre-
sumption stems both from the pathological tendency to generalize that
contaminates every important study of society undertaken in France and
from the rivalry among university disciplines for supreme legitimacy.
Having been enticed out of its position as the soothsayer of national
destiny and prophet of human progress, history has studied the social
sciences only the better to appropriate for its own advantage the ambition
of Durkheimian sociology.
But the important feature of history’s recent development is not this
illusory ambition; it is, on the contrary, what that ambition masks and
perhaps serves to make up for: the fact that the historian, like the
demographer or the anthropologist, constructs the object of his research.
Having abandoned the naive notion that facts speak for themselves—a
Introduction of

naive notion because it assumes without ever saying so that time has a
predetermined meaning—the historian has by the same token abandoned
the superstition of chronological divisions and periodization. If phe-
nomena of every sort are no longer primarily interpreted in a temporal
framework, one must first define objects of research and then determine,
by trial and error, the relevant time span for describing and analyzing
them. Hence, specialization in a given period, for so long a basic feature
of the profession, has become a lesser imperative compared to familiarity
with a particular issue or problem. The historian of what used to “‘consti-
tute” the reign of Louis XIV is less a historian of the seventeenth century
than, let us say, a specialist on absolutism or on the court phenomenon.
In other words, history is no longer wholly contained in a genealogical
construct in which earlier events explain later ones and must conse-
quently be reconstituted with painstaking attention down to their re-
motest origins. Indeed, history used to be a form of learning obsessed
with origins, seeking through erudition to assign an ever-earlier date to
the great secret of beginnings. Moreover, the entire process that it
ennobled was, by definition, a genealogy of the nation or of mankind: in
its bibliographical classifications, seventeenth- and eighteenth-century
European culture distinguished between the inventory of time—the ob-
ject of history—reserved for European peoples and states, and the inven-
tory of space, the realm of travelers, an immense repository of immobile
societies. In the non-European portion of the inhabited globe, these
societies also testified, in a certain sense, to the origin of man, since the
savage was the ancestor of civilized man. Thus, in the order of knowl-
edge, space was subordinated to time, and travel to history. This was a
way of depicting human progress. Somewhat later, when ethnology and
anthropology emerged as disciplines, they dealt only with societies whose
culture was unwritten, leaving outside their purview, for instance, the
great oriental civilizations. Here again, the division was based on the
same principle: China and India share with European nations the priv-
ilege of historicity; their time, too, has meaning.
If all societies today have regained a history, it is not only in the sense
that universal history, through the world market, colonization, and de-
colonization, has integrated them into its vast currents; it is also because
history, as a form of knowledge rather than an actual process, is no longer
the preserve of a small number of societies or nations but a discipline that
can apply to any one of them. While totally distinct, these two phe-
nomena are not in fact unconnected, for the spread of nationalism to the
entire planet has led to a resurgence of the genealogical history of
nations, on the old European model. But, in order to assess the present
situation of history, it is better not to focus on this contamination of the
present by the past but to concentrate on what is truly new: the fact that
everything has been historicized.
Such an assertion seems obvious, but in fact it is not. For example,
8 Introduction

immobility was for centuries regarded as nonhistorical. Only change was


historical; history was considered to be the movement that carried human
communities toward progress through the development of nation-states.
And “ancient” history, too, which dealt with remoter periods, was meant
to study and bring to life the civilizations that had been the ancestors and
guides of Europe’s nation-states. Even as the guiding thread of history
snapped, history at last opened itself up to all the languages that Euro-
pean societies have been constructing for several centuries in order to
interpret, understand, or govern themselves; by the same token, history
opened itself up to a wide variety of phenomena and a truly infinite
number of events. The size of an age cohort, the rules of marriage, the
structure of a rural estate—these have become historical facts. What has
remained motionless for centuries deserves examination no less than
what has not lasted.
This seems to me to be the fundamental change that has taken place in
historiography over the last few decades. The change, linked, as always,
to phenomena outside the realm of scholarship, cannot be understood
without reference to the decline of Europe in the twentieth century and
the crisis of its dominant nations, the traditional agents of the progress of
“civilization.” But the historiographical change has drawn on crisis and
decline: a thorough reorganization of history’s inner realm has been
carried out, both by an extension of the range of its objects of study and
by an at least partial transformation of its language.
The two shifts were inseparable from the outset. They both reflect the
fact that history, taking advantage of the immense flexibility of its bound-
aries, has fallen back on, and reconquered, the social sciences. While
everything has become “historical,” the need has arisen to define, within
the historical field, a certain number of objects—or, less accurately,
research themes—whose historical development had not already given
them a predetermined meaning. As history’s inventory was being re-
newed, the greatest number of suggestions came from the specialized
social sciences, such as economics, demography, and ethnology. But,
because these sciences deal with their objects not only in terms of specific
concepts but, more broadly, in a language alien to the historian’s familiar
mode of explanation, it is the intellectual landscape of historiography that
has gradually been transformed.
Traditional historical explanation obeys the logic of narrative. What
comes first explains what follows. Since history thus defined has a mean-
ing that predates, so to speak, the set of phenomena that it encompasses,
historical facts need only be arranged on the time scale to become
meaningful within a process known in advance. The very selection of
those facts is predicated upon the same, implicit logic: the period takes
precedence over the object being analyzed; events are chosen according
to their place in a narrative defined by a beginning and an end. Little does
it matter, therefore, if the events are unique, not comparable, or even
Introduction 9

lacking in homogeneity with one another, since their significance derives


from an external source. But, as that significance tends to turn events into
markers of change, signaling the dips and breaks along the imaginary
route of time, they are chosen for their capacity to embody change and
the phases of change. The model of this type of history is political history,
both because nation-states are held to be the major agents of change and
because politics, in a wide sense, constitutes the prime repertoire of
change: it is the main mode in which modern societies experience and
interpret their transformations.
Thus, too, political history is primarily a narrative of human freedom
as seen through change and progress. Although political history does
describe the framework, that is, the constraints within which men act, its
major function is to describe the thoughts, choices, and actions of men—
primarily great men. Politics is the quintessential realm of chance, and so
of freedom. It gives history the structure of a novel, except that its plot
must be composed of authentic facts verified according to the rules; and
this history is indeed the true novel of nations.
In contrast, the language of the social sciences is founded on the search
for the determinants and limits of action. Since it borrows from natural
science the aim of isolating constants, if not laws, this language seeks by
definition to describe and study objective behavior, regardless of the
deliberate intentions of the actors. That is why it prefers to select, among
the traces of human activity, sets of internally homogeneous data that are
repetitive and comparable, allowing it to measure the value or simply the
relevance of a question. The social sciences cannot attain the purity of
hypothetical-deductive reasoning, but they strive to do so. By building
their hypotheses and data explicitly and moving constantly from one to
the other in order to experiment and correct, the social sciences do not
assign to the materials on which they work a prior significance deter-
mined by time. They remain inside an intellectual circuit defined by this
experimentation through reconstruction, without attributing to it, a
priori, a third dimension containing its secret.
This temporal dimension, far from dominating research, is subservient
to it, for the period chosen for the study of a question depends on the
nature of that question; in other words, time is no longer homogeneous
and no longer has an all-embracing significance. Birthrates do not change
at the same pace as systems of international relations, and a “great”?
period of history, defined for example by the temporary hegemony of a
given nation, can be devoid of significance in the chronology of an
economic cycle. The historian’s celebrated ‘‘dates”’ depend on the prob-
lem that interests him. And the conditions of his analysis are all the more
productive when the object studied ‘‘moves” less: in order to identify
with the greatest possible accuracy the variables that explain it or are
linked to it, the object must be considered over a long and stable time
span before the reasons for its transformations can be understood. This is
10 Introduction

why the language of the social sciences is partly in contradiction with that
of ‘“‘historicist” history. The social sciences focus primarily on unchanging
phenomena so as to simplify—that is, to make possible—the mode of
reasoning that underlies their language. And it is no accident if histo-
rians, thanks to the social sciences, have rediscovered the long term—the
longue durée—and even “‘immobile”’ history. For historians, this is a way
of saying in their own language that there exists a category of historical
objects on which time has only very gradual effects. Concentrating on
these categories of objects, historians harmonize their practice with the
concepts they borrow from the social sciences.
In so doing, historians have been led to give up not only the major form
of their discipline—narrative—but also its favorite subject matter—poli-
tics. Just as narrative was the logical way of presenting an interpretation
based on temporal significance, so the analysis of a problem in the light of
data developed for that particular purpose obeys the logic of demonstra-
tion. The only common element in the two genres is the apparatus of
scholarly notes that establishes or discusses the nature of the data or their
veracity. But the internal structure of the two discourses is radically
different, even when they happen to mingle and when the old one lives on
by force of habit after the appearance of the new. Finally, their respective
contents have also, by force of circumstance, been dissociated. As we
have seen, the historical narrative genre is chiefly concerned with the
realm of political facts, the uppermost ridge of the past, the memory of
human action, the patrimony of nations, but is also a laboratory for
change and progress. It is in the study of politics, of human choice, that
history best expresses its capacity to create and to constitute the arena for
the study of man. However, the price paid by history for remodeling itself
on the pattern of the social sciences is that it focuses primarily on what
underlies those choices, on what determines them and makes them
inevitable despite the appearance of freedom. It prefers to analyze
deeper trends rather than superficial changes, to study collective be-
havior rather than individual choices, to examine economic and social
determinants rather than institutions or government decisions. Thus,
demography, economics, and sociology have taken over a field in-
creasingly deserted by its traditional inhabitants—kings, notables, na-
tions, and the theater of power around which they never ceased to
gravitate.

If one wants to understand the fundamental difference between his-


tory in its traditional form and history reshaped on the pattern of the
social sciences—a distinction that must not be confused with what the two
types of history have to say about themselves—the best approach is to
take “pure” examples borrowed from traditional history and use them in
order to analyze its mechanisms.
My first example is taken from one of the most classic areas of histori-
Introduction 11

ography, the French Revolution. This is a blessed event for historicist


history, almost custom-made, punctuated by unavoidable dates, and
admirably suited to narrative, for it combines change and progress like
inseparable twins, establishing politics as the fountainhead and instru-
ment of freedom. The Revolution is not so much a topic in modern
history as one of its chief manifestations, in that it embodies a mode of
change and human action that is fundamental to its significance. While it
is true that, from its very outbreak, the Revolution has occupied a central
position in the historical imagination, all the histories of the Revolution
deal with a topic far vaster than the history of the Revolution: they are
actually constructing a meaning for time itself. That is probably why
history-as-a-social-science has dealt so cautiously with this historical ob-
ject. It does not know how to tackle such a short period, how to grapple
with a succession of events so difficult to simplify, and how to avoid a
practically spontaneous chronological—that is, narrative—approach.
One need only consult a set of the prewar and postwar Annales: the
French Revolution is virtually absent, as if this Jocus classicus of national
history were the special preserve of the “‘other’’ history. Even the partial
transfer of the theme to social history, carried out in Georges Lefebvre’s
early work, does not invalidate this apparent rule.
The social dimension of the event has been absorbed, whittled down,
smothered by its political dimension. The social history of the Revolution
does not consist in describing or analyzing the changes that took place in
late eighteenth-century French society in order to measure the relation
between the political and institutional upheaval and the actual fabric of
society. That relation is taken for granted as being wholly deducible from
political history, evidenced by the fact that the social history of the
Revolution deals either with the analysis of contradictions and conflicts
within prerevolutionary society or with the study of revolutionary person-
nel in the broad sense—crowds, militants, and leaders. In the first case,
the aim is to assemble or expand the collection of “causes” within a
chronological framework whose turning point is 1789; in the second case,
the aim is instead to examine, after 1789 and within brief time spans, who
made the French Revolution. But the common element in these two
approaches is that both implicitly subscribe to a vision of the Revolution
as a political event that cuts French national history in two. If the second
approach extols more often than not the newfound sense of initiative of
the popular classes, the first has conferred on the people’s freedom the
additional blessing of historical necessity. In the historian’s trade, the
French Revolution is the professional specialty in which the period
chosen strictly determines the content of possible ‘‘subjects”’ by virtue of
an implicit notion of the event that filled that period. However varied,
these subjects are designed to describe or explain the revolutionary
break, since that is the prior meaning ascribed to the period.
Hence, the outstanding feature of the type of history that I call tradi-
12 Introduction

tional—to distinguish it from the so-called Annales history—is not that it


is a ‘history of events” or even a strictly political history; for every study
of the past reconstructs events, and there is no logical criterion enabling
certain categories of facts rather than others to be enshrined as “events.”
The most general characteristic of this mode of historical writing is the
notion that the period is more important than the problem studied and
that, since the study of a “period” establishes such a priority, the task of
historians is to add new facts or to correct the interpretation of historical
change within the period. The historian becomes a genre painter: the
frame (that is, the period) and the subject (that is, what he will choose
within the period) are supplied to him more by chance than by intellectual
effort. But the frame and subject tell us nothing about his art, which is the
real secret of his craft and does not consist in imagining connections
between various sets of facts hitherto thought to be independent or in
asking new questions about known facts; what the historian must do
above all is to bring back to life, by the magic of his narrative, all the lives,
thoughts, and passions lost in old manuscripts that no one, until he did so,
had ever read again. |
Perhaps I shall be accused of exaggerating. Here, then, is a second
example, taken not from a historiographical field but from the work of a
contemporary historian, Richard Cobb. Admittedly, he is British, but the
field of historical studies common to British and French historians is
sufficiently homogeneous for them to be included in the same analysis;
moreover, Richard Cobb has devoted his life to French history; and,
finally, his loathing for the Annales School makes him all the more
interesting to me. Straddling both countries, he remains just enough of an
Englishman to hate what he might well have been had he been born a
Frenchman. Indeed, every time I read him I am reminded of the chapter
in the second volume of Democracy in America in which Tocqueville
explains why the English hate abstraction: as products of an aristocratic
civilization, they are interested only in the particular, in the concrete and
unique individual, whereas democratic equality leads men to generalize
and to formulate abstract judgments or laws that apply to classes of
individuals, if not to all of mankind. Cobb would have been a perfect
Englishman for Tocqueville, for he is interested only in individuals and
abhors general ideas. And since, for lack of a theory—an absence based
on principle—he displays a jealous passion in making an example of the
type of history that he produces and loves, he provides me with the best
example of what I am trying to define. '
At the core of this concept of the historian’s craft and discipline lies the
archive, preferably a local one, a holy of holies that the historian must
gradually occupy if his activity is to have any meaning. The archive is an
unpredictable depository, its only reason for existence being the fact that
it is there; the patient examination of its contents gently guides the
researcher toward his subject. The archive cannot provide answers to a
Introduction 13

question or questions raised externally, since its contents are the fruit of
chance and, by definition, cannot be compared to those of a neighboring
depository despite the uniform system of classification. And the histo-
rian’s activity par excellence is the researcher’s colloquy with his archives,
provided he accepts beforehand that they are to have the “‘last word”’ (p.
15). This does not simply mean that, where he hesitates about a fact or an
interpretation, the document itself will decide; it means, more pro-
foundly, that the archives mined by the researcher are ultimately re-
sponsible for determining the content and horizon of his research. Hence
the sort of professional priority assigned not to the object of research but
to the geographical boundaries within which it is to take place and to the
site, that is, the depository, that is to be its center stage.
This concept of history and of the historian conceals a number of
elements that are worth bringing to light. First, at the most invisible level,
there is the existential choice of a way of life and a culture. The manu-
script document and the archive are the images and instruments of
cultural differentiation; they allow the historian to embark on the same
imaginary voyage that physical travel offers to the anthropologist. The
traveler is taken to an apparently exotic field, which he can master by
learning the ways of otherness. Indeed, in Richard Cobb’s mind, the two
varieties of exoticism—temporal and spatial—are combined, since the
British historian must become familiar with the “‘savage”’ territory called
France, and the sign of his success is his being told by the local archivist
that he has become an ‘excellent joueur de boules” (p. 24). France is a
museum, both in its past and in its present, and one can “learn” it by
studying not only its archives but also its present customs, provided they
have not changed too much. The passion for history is rooted here in a
choice at once aesthetic and moral: Oxonians do not like all the fuss being
made over change and modernity.
At the same time, the subjects and even the spirit of this history are
determined and bounded by that choice: it is a history that aims at
reconstituting the past, not at interpreting it. For that matter, the term
reconstituting itself contains an ambiguity: if it means that the historian’s
work must bring the past “back to life,” it is stating an ambition that is
either absurd or solely dependent on the techniques of the art—in which
case, the existence or nonexistence of documents and the historian’s
faithfulness to them are not crucial. Reconstitution makes sense only if it
means that history reconstructs not only what actually happened but the
way in which what happened was perceived by contemporaries. The first
part of the program is intellectually the simpler; the second, having an
ill-defined object, contains an irreducible element of uncertainty. But it is
true that the modes of reaction to (and, thus, of perception of) an event
are essential historical phenomena; it remains to be determined where
and how to find them.
This type of history-as-reconstitution, which I am distinguishing here
14 Introduction

from history-as-interpretation, involves no special methodological inno-


cence. The dialogue between the historian and his archive, which enables
“the stranger in a strange land” to become “a friend in a familiar one”
(p. 25), can never in itself provide him with a subject. If the aim of the
dialogue is not the publication of manuscript material (a goal and an
activity indispensable to history but insufficient to constitute history), this
dialogue or, rather, this apprenticeship must necessarily be determined
by earlier intellectual choices, whether explicit or not; otherwise, it can
lead only to a collection of unpublished anecdotes that struck our Martian
as memorable. That is not what happens, however. The Martian is not a
faux naif, and he has come to the archives equipped with questions and
his chosen period. Granted, the questions may be vague and the period ill
defined, but the tyranny of vagueness and imprecision is worse than that
of clarity, for it conceals its own existence from itself. It never ceases to
control the apparently unprejudiced dialogue between the historian and
his documents and to impose its constraints on him. This type of archival
research seems to dispense with the need for genuine conceptual plan-
ning. It provides mental indolence not only with the alibi of period flavor
but with the semblance of intellectual work. It is the paradise of
documentary platitude. That such research, as in the case of Richard
Cobb, can be rescued from its demons by literary talent in no way alters
the nature of the intellectual activity it implies. Literary talent gives life to
what is a rewriting of archives; it does not add a single idea.
What rewriting does, however, is to reveal the traces, not of hypoth-
eses or concepts, but of what must necessarily substitute for them in order
to permit a minimum of selection in the ocean of documents. That
substitute consists primarily of a period, as well as a particular form of
curiosity, a set of questions, or all these ingredients combined more or
less explicitly. What is interesting about Richard Cobb and his books is
the sort of rigor with which the British historian constructs and applies the
doctrine of his preferences. Carrying his hatred of ideas and of intellec-
tualism to the extreme, he turns history into a laboratory for a purely
existential preference, combing the archives for material of literary value
and transforming the quest for knowledge into a passion for novelistic
narrative. Cobb is a historian of society for whom only individuals exist—
pure empirical entities independent of all intellectual constructs.
Consequently, there is only one way of organizing these empirical
entities into historical material. The controlling assumption becomes the
randomness of human existence, the flow of time, or the life story. The
interesting feature is that our historian makes narrative a necessity for his
brand of history, for he clearly sees the link between the fact—conceived
as a “pure” fact, uncontaminated by any idea (whether prior or subse-
quent to it) and speaking, as it were, about and for itself—and history as
narrative. The two professions of faith (to deal only with the facts and to
tell a story) are presented as distinct yet inseparable. They are the sole
Introduction 15

constituent elements of the profession and of the secret of the profession,


handed down as such and thus learned or repeated more than thought
about. In truth, the specific feature of these tenets is that they are both
mistaken if stated separately. Each holds true only in the presence of the
other. As historians are periodically told—without their really heeding
the warning—there is no such thing as a ‘“‘pure”’ fact. A historical fact is an
intellectual choice—and this choice is what distinguishes history from the
wholesale publication of manuscript material. Moreover, narrative is not
a sufficiently distinctive feature of the historian’s art, since the novel, too,
adopts the narrative form in most cases. But here comes the miracle. The
most general and most superficial definition of the historical fact—that “it
happened,” that it actually took place—provided it is combined with
narrative, constitutes history. Under the guise of a mere form, narrative
is in reality assigned the task of making a fact understandable. For want of
its being integrated into an analytical framework of questions (since it
derives its falsely objective character—supposedly imparted to it by
reality itself—from the absence of such integration), the criteria for
selecting and explaining the fact are derived from chronology alone,
which is the fabric of narrative.
In this game, the historian wins on every count. He dresses up the
professional respect for facts and dates in a form that resorts only to the
“logic” of before and after in order to explain them. Narrative endows
scholarship and archival research with the charm and even the pleasure of
a novel. Because it is built on a succession of concrete and unique facts, it
calls on the historian’s evocative powers more than on his specifically
intellectual ability, on his art more than on his mind, on his sensitivity
more than on his intelligence. The ease with which historical material can
fit into the narrative mold explains why a learned discipline can also be a
popular genre—and why it is sometimes difficult to distinguish between
the various levels of historical output.
Indeed, acquaintance with historical writings reveals that nothing
blends so well as erudition and an affective commitment. The narrative of
the past, if well produced—that is, not only “‘true”’ (as regards the facts
recounted), but put together with even a modicum of depth—is insepar-
able from the historian’s sympathy with the “life” of the period whose
events he is describing, that is, with the way in which contemporaries
perceived and experienced the events that constitute the substance of the
historian’s narrative. But this sympathy, which makes possible, if not the
reconstruction, at least a reconstruction of what has disappeared, belongs
to the realm of affection, of ideology, or of the two combined. It replaces
the explicitly formulated question as a link between past and present; it
creates the vacuum that the historian’s craft, in this case, is supposed to
fill.
Much of what has been somewhat loosely called histoire des mentalités
is informed by this dialectic between the experience of strangeness and
16 Introduction

the act of familiarization. It is no accident if this type of history has


enjoyed its greatest popularity in the past decade or two, in a French
society violently torn from its past by economic growth and feeding in
compensation on a world of nostalgia. My generation has seen a transfer
of the values of rootedness, tradition, and the soil away from a residual
right-wing mythology to a vast left-wing consensus. The histoire des
mentalités thrives on this feeling as the history of nations does on patrio-
tism. It is linked to the past less by a series of specific questions than by a
passionate desire to convey to us, as if they could come alive again, the
emotions, beliefs, and mental universe of our ancestors. The secret of this
history is, above all, a secret of sensitivity. It views the past as a mirror
image of our present, but this mirror image, too, is our present.
At the same time, this history, owing to the vagueness of the word that
gives it a label if not a content, presents an almost infinite range of
methodological possibilities. The study of mentalités in a society or group
can, for example, be based on the contrast between conscious and uncon-
scious manifestations, on the distribution of psychic and intellectual
activity according to cultural levels, on the Freudian notion of repression,
or on a number of other similar investigative tools. In another intellectual
context, and because it embraces both the history of objective behavior
and the perceptions of such behavior, the histoire des mentalités can
create the illusion that it enables us to grasp a sort of comprehensive
social entity—a fusion of infrastructure and superstructure. Thus, in the
random pattern of its various applications, the histoire des mentalités
blurs the classic distinctions observed in the study of individuals and
societies all the more effectively as it gives the impression of transcending
those distinctions. All too often it is merely a Gallic substitute for Marx-
ism and psychoanalysis. This semantic prestidigitation—a good illustra-
tion of the casual manner with which historians have traditionally treated
their conceptual tools—adds no real explanatory power. However, it
presents French historiography with the danger of self-satisfaction in a
vacuum, since the word that it holds up like an emblem—mentalités—has
no equivalent in other languages.
Finally, the uncontrollable expansion of what is regarded as the “‘his-
torical” field threatens to increase the quotient of insignificance. The lack
of definition of this vast field—an area that is supposed to signal the
emergence of a “new” history—leads to the unending pursuit of new
research topics, turned up by the accidents of life and having no other
basis than a passing intuition or an ephemeral fashion. In reality, such
topics are found, not built; they pave the way for a proliferation of
histories that study the psychological realities of the past on the basis of
arbitrary curiosity, without a prior theory, using ambiguous sources and
yielding endlessly debatable results. As one of the essays in this book
shows, the use of past demographic behavior to reconstruct underlying
attitudes is fraught with uncertainty.
That is why there is no longer an automatic link between a history of
Introduction 17

the ‘‘new”’ variety, however misused the adjective, and the broadening of
the historiographic field. That this broadening is related to curiosity or to
feelings rooted in the present does not suffice to make it a novelty; the
study of the past has always been inseparable from the historian’s relation
to the present. And the example of history a la Cobb shows that, to the
extent that this relation remains purely emotional and devoid of ques-
tions and concepts, the wholesale conferral of historical dignity on the
lower classes and marginal individuals in no way alters the nature of this
variety of history, which still remains a narrative of bygone times.
For the most part, the so-called histoire des mentalités, even when it
styles itself ‘‘ethnological,”’ still belongs to such a breed of history. The
reference to ethnology, when it simply indicates that the historian is
dealing with subjects that used to be considered outside his province and
were invented by that neighboring discipline (marriage, kinship, the
constraints of sociability, for example), is only a token homage to the
interdisciplinary approach. Far from establishing or even maintaining the
cultural distance between the observer and the observed—a distance that
lies at the very heart of ethnological analysis—it reflects precisely the
opposite: a return to Peter Laslett’s “‘world we have lost,” the search for
an anchorage in the deepest stratum of the familiar, below the sediments
modernity has deposited in us. The reference to ethnology extends the
scope of genealogical research, adapting it to our feelings toward the
present, but it does not alter the nature of that research; history has
moved from the genealogy of nations to the genealogy of societies. To be
convinced of this, one need only consider the fact that ethnohistory as
practiced in France does not represent a comparative form of knowledge
but remains confined to French rural society. Furthermore, as it seeks
less to explain the unfamiliar than to discover the familiar behind the
illusion of the unfamiliar, it is subjected to the temptation of the pictur-
esque, which constitutes its link with the general public.
Thus, for history to proclaim its indebtedness to another discipline is
not enough to turn it into a form of knowledge different from what it was
fifty or a hundred years ago. This claim or assertion can mean only that
history has broadened its scope, not that it has transformed its intellectual
approach to the past. In this respect, one can understand the irritation of
an aggressively traditionalist historian such as Richard Cobb when con-
fronted with the “‘interdisciplinary”’ inflation that characterizes the pres-
ent-day vocabulary of historical studies in France. This almost compul-
sory terminology can sometimes apply, in fact, to a type of research or
exposition entirely similar to those practiced and cherished by the British
historian: old events, found in unpublished documents, used in a narra-
tive to reconstruct the particular atmosphere of a period through the life
of one or more individuals. Like all forms of extremism, Cobb’s reaction-
ary radicalism—in addition to its literary charm—has the advantage of
exposing the duplicity of words.
Actually, the distinction I am trying to draw between two types of
18 Introduction

history—the sort of history recommended by Cobb at one extreme, the


type represented, for example, by historical demography at the other,—
is an imaginary boundary, in that it is constantly being crossed by mixed
genres. But it is always useful, for clarity of argument, to visualize
extremes: on the one hand, periodized history, chronological narrative,
reconstruction of human experience, the empiricism of “facts” as
opposed to preconceived ideas; on the other, problem-oriented history,
the analytical examination of a single question over reputedly heter-
ogeneous periods, the interpretation of human experience with the aid of
a theory or idea. This theoretical opposition defines the space within
which our discipline floats today, in the context of a general transforma-
tion represented by the broadening of its curiosity and fields of research.
It is fairly easy to reconstruct the phases of this transformation. At
times the expansion of the field was accompanied by a marriage between
history and another previously constituted discipline—such as demogra-
phy—already occupying a given portion of the wider field. History
brought to demography a precious store of additional data even as it
borrowed demographic methods and concepts. That is how historical
demography—or demographic history—was born. Because of its simplic-
ity, it offers the purest example of the renewal of the study of the past with
the help of a particular social science. But, in the course of its expansion,
history at times also overlapped with disciplines that were far less clearly
defined, such as sociology, or even certain research topics—such as
mentalités—that had yet been honored by any social science in particular.
In these instances, the historian had to tinker, borrowing a concept here,
a method there, a technique elsewhere. There is no “‘sociological history”?
or “ethnological history” in the same sense as there is a ‘‘demographic
history.” In the latter specialty, history cast itself entirely into the mold of
another discipline, without modifying its objects, concepts, or research
methods. The renewal of demography stimulated by history stems from
the deepening of the chronological range and from the possibilities for
comparison, and thus for new hypotheses, offered by such an expansion.
The study of shifts in demographic equilibriums in particular has been
enriched by crucial insights. In the case of sociological or ethnological
history, the historian who practices it generally borrows not so much a
rigid set of research topics and concepts as a sort of general guideline for
his curiosity. In other words, he will stress, for example, problems of
social stratification or attitudes toward death in a specific society at a
given period. But this type of curiosity may imply no other commitment
than a particularly strong belief in the determination of the individual by
society or a particularly keen desire to discover the elementary and the
identical underneath the diversity of social phenomena, even under the
mask of modernity.
Thus, between good old narrative history, which reconstructs true
facts according to the chronological framework of a novel, and the history
Introduction 19

that calls itself new because it borrows part of its equipment from nearby
disciplines, the opposition is neither as sharp nor as genuine as the second
type of history would have us believe. The confusion stems from the fact
that the renewal and extension of history’s subject matter—phenomena
on which all historians now agree—do not necessarily entail a change in
the way of treating that subject matter. Richard Cobb, too, is a ‘“‘new”’
historian, in his fashion—even though he loathes the Annales—since he
devotes himself with such ardent solicitude to beggars and vagrants, the
forgotten individuals of traditional historiography. He has swapped
dukes for tramps, respectable folk for the destitute, great men for small
fry, deeds for daily life. Although staged on another scene, the show he
produces nonetheless observes conventions that Georges Lenôtre would
not repudiate. To open the noble realm of history to other heroes and
other actions—a gesture that all historians applaud—does not necessarily
imply that one looks at them differently or that one bestows on them an
original intellectual treatment.

As for the history I like, now defined not negatively but positively, the
studies that follow aim precisely at delineating its contours. I am deliber-
ately using this vague expression, for these texts do not pretend to be
normative. History, even scholarly history, is not and never will be an
exact discipline, as one speaks of the exact sciences. There will never be a
consensus among those who write history as to the criteria that distin-
guish scientific from nonscientific history. The only rules of the trade are
the procedural rules developed by European scholars between the six-
teenth and nineteenth centuries. These rules remain essential, indeed
indispensable, but all they define is a profession not a science. To estab-
lish a ‘“‘true”’ fact in its complexity is one thing; to find the law that governs
its emergence or existence is another. And if historians attach an almost
fetishistic importance to mastering professional “skills,” which concern
only the documentary aspect of their work, it is because they have no
other criteria for defining their guild.
Scholarship must serve as the basis for all imaginable types of history,
from the most narrative to the most conceptual. In other words, scholar-
ship makes it possible to distinguish between good and bad history but
not to establish a hierarchy of historical genres. It constitutes a minimum
requirement, beyond which the range of intellectual preferences remains
open almost to infinity. I believe that one must remain firmly attached
both to that obligation and to that freedom—to that obligation because
the damage caused by amateurism and shoddiness has perhaps never
been so visible as today; to that freedom because it is periodically being
called into question by conflicting but equally superficial dogmatic
approaches, one of which clings to the intransigent cult of narrative, the
other to the pretense of a “‘scientific’’ form of knowledge.
In the relatively vast area within which the discipline is fluctuating
20 Introduction

today, my preferences, as the essays that follow indicate, go toward an


intellectualist history that builds its data explicitly on the basis of concep-
tually developed questions. History, it seems to me, can advance only by
paying far closer attention to the formulation and reformulation of prob-
lems and by making a clearer distinction in historical writing between
documentary evidence and interpretation. To that extent, it is indeed
true that narrative is to my mind a somewhat lazy mode of writing history,
since it avoids questions and mingles facts and ideas in the magic flow of
time. But there are two illusions of “‘truth” in the historical field. First,
there is the illusion stemming from almost spontaneous adherence to the
temporal logic of a series of events that appears as an objective datum,
existing since all eternity and simply revealed by the historian. Second,
there is the illusion that consists in the rational consent given to the ex
post facto reconstruction of a necessity: nothing was possible except what
took place. As for problem-oriented history, it unmasks both scholarly
rationalization and the pitfalls of spontaneous understanding. By con-
structing its own hypotheses and data, it endows them with an auton-
omous existence that restores freedom to the actors it studies and open-
endedness to historical situations; thus, conceptual history protects its
analytical power against the scientistic illusion.
History represents a particularly valuable form of knowledge in that
the social sciences are just emerging from a period in which anthropology
and sociology were dominated by the notion that society speaks and acts
autonomously. According to this view, social agents simply enact their
society’s rules of operation and reproduction without knowing that they
do so, and enjoy no other freedom than the possibility of entertaining the
illusion of freedom. Whether Marxist or Freudian, the notion that the
intentions and actions of historical agents are intelligible only in relation
to an external source of meaning has played an indispensable role in a
French intellectual milieu long hostile to, and protected against, the
theoretical spirit. But history, because it deals with human action at the
level closest to the freedom of invention, is the best antidote against the
misleading simplifications and illusory rigor inherent in the notion of a
science of society. It is all the more effective an antidote for having
dropped its traditional qualms about hypotheses and ideas. Indeed, as
history has borrowed hypotheses and ideas from the social sciences, it can
test their explanatory power.
At the same time, history has never lost sight of the fact that part of its
curiosity is rooted in the present. Contrary to what the positivists be-
lieved, history’s relation to the present is one of the ingredients of its
relation to truth—an ingredient of greater or less importance, depending
on the problems and periods examined, but an ever-present one. A
searching formulation of a question allows the historian to avoid being
trapped inside a period and enables him to use the past as a repository of
experiments that are in some ways comparable, even if not concomitant.
Introduction 21

In the same way, the explicit discussion of his relation to the present is an
intellectual exercise that enables him to understand his own ‘‘objectiv-
ity.” The celebrated point “from which” he speaks is constantly being
explored in the shuttling comparison between past and present. How-
ever, the present must not be reconstructed as the only possible outcome
of that past but analyzed as what turned out to be the most probable
outcome, independent of the intentions or predictions of historical
agents. Thus, the historian can rediscover the gap between intention and
real process, a gap that characterizes his type of inquiry since it defines
the two poles of historical investigation.
There are no concepts for explaining the past that do not bear the
stamp of the present and do not therefore date the historian. Conversely,
however, without an analysis of the present there can be no concepts at
all. The first proposition is not enough to define a good conceptual
history, for it does not guarantee a choice of sound analytical tools; but it
does lay down a minimum requirement. It accepts the limits of historical
objectivity the better to abandon the fallacy of “bringing the past back to
life” or the temptation to tell a mere story.
From this point of view, the distinction of sorts sometimes drawn
between contemporary history and plain “history,” reputedly more reli-
able, seems to me to have no other justification than the vantage point
involved: the closer one gets to the present, the less “‘distance”’ one has.
For example, the consequences of the Reformation for the history of
mankind are easier to understand than those of the Bolshevik revolution,
since there are four and a half centuries of temporal distance in which to
study them. But this perceptual advantage, which, moreover, varies
according to the subject treated, does not in itself confer intellectual
merit: depending on the historian or the work, one can encounter super-
ficial histories of Protestantism and insightful histories of Bolshevism.
And the contemporary world can offer the attentive observer a host of
considerations that will help him understand the past, if only the historian
displays an interest in the present.
Indeed, the contemporary world, in its natural state, provides him
with the central problem of historical analysis: the problem of change and
of interpreting change. As regards “what happens” every day—the
events rehashed by endless comment in the media—the historian’s famil-
iarity with the past enables him to come up with additional questions or to
establish new connections between events. The raw form of change that
constitutes what we call ‘‘news” must immediately be made as intelligible
as circumstances permit. But the historian is best equipped to restore
these developments to their unfamiliar dimension, precisely by focusing
on the features that make them comparable to past events. Thus, he can
isolate the repetitive element and the totally new aspects of current
events. Admittedly, as an actor in his own epoch the historian is not
immune to the temptation to adopt any one of the explanations of ‘what
29 Introduction

is happening” that belong to the conventional wisdom of his time. But,


although the privilege of immunity is not granted to the historian when he
takes up his trade, it becomes more readily accessible to him in that the
nature of his knowledge teaches him both the constraints of action and an
awareness of its ambiguities.
In addition to a series of essays on history yesterday and today, I have
included some pieces that do not concern the intellectual analysis of time
but deal with two communities, two histories that are constituent ele-
ments of my cultural environment—America and Judaism. My aim here
is to demonstrate that for me the work of learning about the past is
inseparable from the effort to understand the world in which we live. The
two enterprises are identical in nature; the first draws on the second for
some of its inspiration and even some of its techniques.
As we have seen, the central change in historical studies over the past
fifty years has come from the spread of this type of knowledge to the
whole of mankind and has thus involved the relativization of European
history, western European history in particular. The recourse to the
language of the social sciences—a process that has overlapped with the
quest for another universality—is easily explained by the crisis in the
evolutionary approach to time developed in the nineteenth century. As
one can see in the works of Claude Lévi-Strauss and Louis Dumont,
which are comparable from this angle, the relativization of European
history required comparisons with other civilizations and societies. But I
have never had much ethnographical curiosity. I know very little about
the great non-European civilizations, and I have worked only on what
was familiar to me. However, as I have at the same time measured the
value of “unfamiliarity” in the construction of the objects of our studies,
particularly of the topics that seem the most readily intelligible, it seems
to me that I have never ceased to be interested in the Jewish and the
American experiences as instruments for historical comparisons. Beyond
the accidents of life, it was the European features of these phenomena
that allowed me to find anchorage in them. Yet they also provided me
with reference points located outside my mental habits, since the core of
these experiences had been built outside of Europe and even in opposi-
tion to it.
European reflections on the United States have, over the past two
centuries, been inseparable from a comparative analysis, implicit or
explicit, whose second term of comparison is the history of Europe;
Tocqueville is the classic illustration. Having become the world’s most
powerful nation, America has lost none of its power to evoke a universal
future. Despite its brutal and almost involuntary appearance in global
power politics, it remains the society that comes closest to eighteenth-
century credos: it is the most pristine laboratory for our modern beliefs.
As for ‘‘the Jewish question,” as it used to be called, it is for all members
of my generation the very crux of the European tragedy of the twentieth
Introduction 23

century. Nazi anti-Semitism, the complicity it fostered or encountered,


and the massive extermination it performed give a pathological measure
of what our secular and democratic societies are capable of producing—
following the intolerance of the Catholic church and contrary to the
beliefs of the enlightened nineteenth century. As for the other aspect of
contemporary Jewish history—the events leading up to the state of
Israel—it offers exceptional material for historical analysis, not least
because so few of its developments conform to predictions. Jewish history
as a whole is endlessly surprising, in the tragic as in the happy sense. The
state of Israel provides an example of an unlikely nation of latecoming
immigrants, fleeing the European scourge—two or three centuries after
the English dissenters—in order to people a Promised Land blessed by
God. In this respect, Israel presents the French historian with a problem
similar to that posed by the United States: a model of nation building and
a type of society radically different from those with which he is familiar,
since they are founded on the recent choice of their makers and engen-
dered by a religion, by an ideology, or simply by misfortune—causes that,
moreover, are not incompatible. But the creation of Israel is even more
astonishing and thought-provoking than that of America, for it combines
an infinitely older belief with a far more recent historical scheme. The
founding of the Israeli nation is predicated on both the socialist notion of
a new society and an ancient belief in a “return”; in almost every Israeli
citizen there is a mixture of extreme modernity and extreme archaism,
and Israeli society illustrates the extraordinary diversity of the chemistry
of nationhood. Its existence confronts the observer with all the problems
of the world in which we live. These problems are present in Israel in a
condensed form but at a level deep enough for the historian to feel that he
is on familiar ground: the extraordinary survival of a religion and a people
scattered throughout the Christian and Arab worlds, the relations be-
tween democracy and pathological anti-Semitism in the twentieth cen-
tury, the historical and modern elements of a national consensus, the
clash between religion and secularization on the natural battleground of
the state, and, finally, the watertight confrontation between civilizations,
in which the Israelis, involuntary heirs to European colonialism, repre-
sent in the Arab mind the world that persecuted the Jews and from which
they fled.
By adding to some general considerations on history in French intel-
lectual life today a number of studies devoted to the American world and
the Jewish world, my aim has been not only to establish cultural bearings
in relation to problems that interest me; I have also wanted to express my
belief that history is inseparable from an understanding of the contem-
porary world—a world that supplies history with its questions and its
raison d’étre.
.
Ai a aA

foA al Var silanes:


: eee “2
2.
hal ins
es tito | -~
pe.
=
7
i ee:
idUE tsrl Pa
ee
dint
|
a 2m re
| CPE

Wi Cape Ara
! a Ge re see
Er” 2 ne Je 7
_
| Eu 14 Ate van
= ETS = v
e > ur
ell UT AI Ter CASE
POTN 2 LN stl > 5
vArs
i | LEE Cure
:
Ds ed
Val
,<8
x :
— 7 RER
>
= AR |
sont |
‘hr
of 5 a 4
; ea
TA
me |
Alaa
Stu ap TO
j .
N Sh
Do ad

SH
à CE" LES
de 6
= <i
« ai
L2
a? LE
ad :

a]
PF
s =

Le vine

é f°
an
5 oh

cy :

P4
1

&

PU Aged
Le

«ge nl
z 7 Leon
t
: "
~~
-
PART Sens

LR
French Intellectuals:
From Marxism to Structuralism

The ‘‘end of ideologies,”’ as diagnosed by Raymond Aron, for instance,


in one of his books,' refers generally to the developed societies of Amer-
ica and western Europe. It connects prosperity, economic growth, and
social integration on the one hand with the progressive decline of political
extremisms on the other; the car, the refrigerator, and television are said
to have killed revolution. This type of analysis has already inspired
endless comment on neocapitalism, the Gaullist regime, and the kind of
political torpor that has characterized France since the end of the Alge-
rian war, as if one of the functions of Gaullist nationalism were now to
counterbalance an objective process of French ‘‘Americanization.”
But does the end of ideologies mean the end of ideologues? If it is true
that present-day France tends in its social depths to be dozing off into a
society of abundance and social integration, is the diagnosis also true of
the groups and the individuals whose profession is thinking and writing?
It may be objected that this question implies a certain modification of our
initial proposition, for the relations of intellectuals to ideologies differ in
nature from those of the general public; at any rate, they are more
complicated (even if they are apparently simple or deliberately sim-
plified). But French intellectuals, because of the authority they have
enjoyed since the Enlightenment, have often been indicative of the
problems and choices of society as a whole. There can be no greater
historical oversimplification than deifying the intelligentsia’s purely pro-
testing role; claiming a long line of great but damned ancestors is a cheap
way of entering into an exceptional heritage. But Voltaire was the most

First published in Preuves, no. 92 (February 1967). An English translation entitled “The
French Left: From Marxism to Structuralism” appeared in Survey, no. 62 (January 1967),
72-83. The present text is a slightly revised and expanded reprint of that translation.

27
28 History Today

acclaimed hero of the eighteenth century, the Revolution put Rousseau


in the Pantheon, and Victor Hugo’s coffin was followed by an enormous
crowd. The intellectual Left has rarely governed the France of its time,
but it has supplied its universal values. Neither the Dreyfus affair nor the
Popular Front nor the postwar spread of communism is intelligible if the
part played by the intellectuals and their ideological elaboration is left out
of account.
That is why ideology is far from consisting exclusively of a theory of
history, though in the France of yesterday Marxism-Leninism was the
most widespread and extreme form it took. Ideology is born of the feeling
that a great historical problem can and must be solved by individual
commitment. Hence the passion inseparable from it, the proselytism, the
condemnation of the enemy and even of the indifferent, the identification
of personal morality with historical necessity. The class struggle has
different aims according to Guizot or Marx. The former hails the triumph
of the bourgeoisie, and the second the advent of the proletariat; but both,
even if this is denied, as it is by Marx, imply a moral view of politics,
distinction between good and evil, and commitment to the side of the
good.
Thus, the French Left has no monopoly of ideology; on the contrary,
for nearly two hundred years the conflict between Right and Left has
been the warp on which ideologies have been woven. The Frenchman of
the Right, armed with Barrés versus Zola, Maurras versus Romain
Rolland, and Drieu La Rochelle versus Aragon, also has his cultural
pedigree, whose assumptions and political hypocrisy the Left has fre-
quently denounced. The national masochism of the Pétainist bourgeoisie
or, fifteen years later, the nationalist enthusiasm for /’Algérie française
refers back to the same system of intellectual and moral arguments, only
the terms of which vary with the event. Why, then, the privileged position
enjoyed today by the ideology of the Left and by the left-wing intellec-
tual? The answer is that the last great battle of right-wing ideology was
fought by fascism—and lost. Since the end of the war, ideological elabo-
ration has consequently been a quasi monopoly of the Left.
How the celebrated Hegelian tribunal, which had become the ‘‘opium
of the intellectuals,’ was wildly misused by the victorious Left after it had
been “proved right” by history was described by Raymond Aron even
before the politico-moral tribunal of Stalinism broke down with the death
of Stalin. Up to that moment, historical certainty and moral judgment
had combined and reinforced each other; now both factors in the apogee
of the ideological age were simultaneously punctured by developments in
the contemporary world.
Destalinization raised the issue of justice and truth in the Communist
world; there turned out to be other enemies besides the bourgeoisie, and
it came to be seen that the Soviet Union was not always, necessarily, and
inherently in the advance guard of human history. This led to a new
diaspora of Communist and progressive intellectuals; a whole world
French Intellectuals 29

collapsed, as I well remember, because I was a part of it. Warsaw,


Budapest, the Chinese schism have only accentuated the process, provid-
ing the last rites for a Marxism-Leninism that had been both incarnate
and universal. But during those very years a new universalist mirage, an
ersatz messianism, presented itself to the revolutionary intellectuals,
namely, the struggle for independence of the Third World or, in the
French context, support for the FLN in the Algerian war. This additional
experiment in ideological extremism was the more characteristic of left-
wing intellectual circles in that it was experienced in social isolation,
disclaimed by the Communist party, and not understood by the working
class. The Western intellectual looked overseas for his (mythical) tie with
the oppressed and the agents of world revolution, in the ranks of an
enemy who was assumed by definition to be socialist and internationalist.
The Muslim fellah celebrated by Fanon had become the latest ally of
revolutionary defeatism of the Leninist type. The outcome is familiar.
The left-wing intellectual had invested his liquid revolutionary capital in
religious nationalism; in seeking the Bolshevik party of 1917, it found
Islam; and, instead of Lenin, it found Boumediene.
The disconcerting outcome in Algeria and the very victories of the
colonial peoples have slowed down left-wing intellectual investment in
the Third World, for the difficulties in the way of economic “takeoff” are
often too technical in nature to nourish passion, and the love affair
between the FLN’s French wartime friends and the Algeria of Colonel
Boumediene has ended in disappointment. Nevertheless, the economic
setbacks that have taken place in most of the recently decolonized coun-
tries have in a way restored to honor the idea of a Leninist—or Maoist—
dictatorship. For lack of suitable soil for experimentation of this type at
home, some Western intellectuals proclaim its necessity in the underde-
veloped countries as the only way of demolishing the many obstacles in
the path of mobilization of labor and national saving for capital forma-
tion. A few years ago it was chiefly the romantic and antibureaucratic
character of the Cuban revolution (which contrasted so sharply with
socialism of the Soviet type) that roused enthusiasm, but the Chinese
model now seems to be restoring priority to economic rationality, and
this classic slant may again enable it to be used in defense of totalitarian-
ism. But it is too obviously inapplicable to European conditions, its
culture being too “‘exotic’’ and its antecedents too well known, and so it is
unlikely to rally a great deal of support; and, finally, the hostility of the
French Communist party deprives it of a great deal of influence.
The irruption into history of the nations of the Third World, instead of
promoting a ‘“‘Chinese” version of Marxism, has in fact contributed to
hastening the end of the ideologies in contemporary French culture.
Superficially and for a brief moment, it seemed to revive great universal
visions of social transformation, but in reality it served to discredit
lastingly and in depth the philosophies of history of the nineteenth
century. This phenomenon could be crudely summed up by saying that,
30 History Today

in the context of French intellectual life—so sensitive to fashion and so


prompt to generalize—the prestige of structural ethnology has been
partly due to the fact that it offered an ‘‘antihistory.”
A disappointed intellectual Left, demoralized by history, turned to
primitive man, not to decipher mankind’s infancy—which would have led
back to history—but to find man’s “‘true”’ situation. The approval of the
“savage” by a society that considers itself to be saturated with wealth and
“civilization,” to use the term already current in Rousseau’s time, is not a
new phenomenon. But the interesting feature of the present vogue is that
the savage has become, for a time, the model for the human sciences. It is
probably not by chance, nor is it the mere result of an incidentally only
too obvious cultural chauvinism, that structuralism, which has dominated
European linguistic research since the first postwar period, was popular-
ized in France not by linguistics but by ethnology. Nor is its success solely
attributable to the work of Lévi-Strauss, whose Structures élémentaires de
la parenté dates from 1948—nearly ten years before he achieved notori-
ety. It took the dislocation of Marxist dogmatism between 1955 and 1960
for ethnology to fulfill a social expectation and meet a historical situation.
Decolonization revealed to all the secrets of the ethnologists, those
pioneers of anticolonialism: namely, that cultures are multiple, are
equally worthy of respect, and manifest themselves in terms of perma-
nence rather than of change. French colonization (which, it must not be
forgotten, was often Left in origin) claimed to place them in Western
“time,” putting them through European stages of progress whether they
liked it or not. Now, however, there is perhaps a trace of expiatory
masochism in the revaluation of these extra-European worlds.
Furthermore, these exotic and impoverished worlds have the virtue of
focusing all the distaste for, and rejection of, the ‘‘affluent society”’: even
if they are no longer centers of revolution and are gradually sinking below
the survival line, they are at least pure and innocent in the eyes of a Left
that is at heart moralistic and more Christian than it thinks. They remain
a geographical refuge for the frustration with the historical immobility of
a nonrevolutionary West. Provided the flame of revolution keeps
smoldering in the Third World (as in South America) or flares up there
(as in China), the students of prosperous Europe will someday be able to
transmute their despair into hope. Above all, a profound change has
taken place in the French intellectual’s idea of the world and the part
played in it by his country. The transference of his revolutionary hopes to
the Soviet Union and then to the Third World was in itself an implicit
confession of impotence concerning the possibilities in his own country.
But it also betrayed the survival of the Jacobin tradition, nostalgia for the
France of 1793, a patriotism temporarily frustrated but nevertheless
optimistic: France would one day again pick up the torch of revolutionary
history. Now even this dream is fading. Not only has the Soviet flame
gone out and the Third World disconcerted or disappointed its friends of
French Intellectuals 31

the heroic age, but France itself is no longer France. The French intellec-
tual, heir to a prestige less fragile than power and unconsciously used to a
universal radiation of his culture, is not yet reduced to the sad state of the
Belgian, but he is increasingly aware of being a citizen of a country that,
in spite of Gaullist rhetoric, senses it is no longer making human history.
Having been expelled from history, the France in which he lives consents
the more willingly to expel history. It can look at the world with eyes no
longer veiled by its own example and civilizing obsession, with eyes now
sceptical about the “‘lessons”’ and “‘meaning” of history. Valéry admira-
bly foresaw this phenomenon after the First World War in his Regards sur
le monde actuel.
Thus, the recent disappointments of the French intellectuals and the
general political situation combine to cast doubt on history—for so long a
tyrannical mistress before she became an unfaithful one.

It is without doubt this situation that explains the great impact of the
thinking of which Lévi-Strauss is generally regarded as both a model and
a representative. It is, however, uncertain—even unlikely—that Lévi-
Strauss, who has a passion for accuracy, would accept this vague and
general attribution of paternity affixed to him under the label of “‘struc-
turalism.” But, from the point of view of the social character of his
audience, that matters little. On the other hand, it is significant that a
work as specific and as technical as his should have had so widespread an
impact on men of letters, art critics, and philosophers. To enumerate its
great themes and aspirations is not to evade the significance of the
intellectual movement that reflects the scope of its influence, since that is
precisely what has to be understood.
In the first place, however abstract and high-powered it may be, the
effect of his work is to revive something like the great Rousseauist
paradox that split the eighteenth century. One feels it to be suffused by
love of nature, the countryside and flowers, a fondness for “‘primitive”’
man that inevitably stimulates nostalgia for a happiness lost by industrial
societies. Lévi-Strauss writes of a world in which the eagle, the bear, and
meadow sage still exist, but he believes no more than Rousseau, who is
one of his favorite authors, in a return to a primitive happiness that has
disappeared forever. The heir to Boas and Mauss knows that there is no
such thing as a state of nature but, rather, myriad communities and
cultures, each representing a different form of man’s confrontation with
nature; and he also realizes that there is no objective way of establishing
any hierarchy among them. Thus, the industrial society of Europe or
America loses the privileged position to which it believes itself entitled in
relation to tribes buried in the forests of the Amazon. “A great deal of
egocentricity and naiveté is required if one is to believe that the whole of
man has taken refuge in a single one of his historical or geographical
modes of being, while the truth about man lies in the system of their
32 History Today

differences and common properties” (La Pensée sauvage, p. 329). Thus,


the savage does not provide us with an image of man’s childhood, as was
believed in the eighteenth and nineteenth centuries in accordance with a
naively European model of human history; he does not even provide us
with one adequate image among others. As soon as it is granted that he
has conceived quite varied societies and uses a logic as profound—or
simple—as that of modern science, he no longer exists as a ‘“‘savage” or as
a “primitive.” He merely offers us, as the so-called developed societies
do, a multiplicity of cultural solutions of the eternal conflict between man
and nature.
The role of the ethnologist is thus to classify cultural systems, to
undertake for each of them an objective analysis of their systems of
symbols and their articulation. The objectives of the psychoanalyst are
implicitly transferred from the individual to the collective, from clinical
analysis to the ‘“decoding”’ of a social language such as myth. In his last
years, Freud was greatly attracted by the interpretation of ethnological
data (which he had not completely mastered). Lévi-Strauss rejects the
methodological confusion implied in this illicit ethnology-psychoanalysis
relationship; he never extrapolates into his own field the procedures of
psychoanalysis. But his ideas about myths extend to the collective sphere
the objectives of individual psychoanalytic treatment; they are intended
to bring to the surface the subconscious structure of social beliefs, the
code that underlies man’s thinking and ultimately determines what he
does think of himself and of others. On this, Lévi-Strauss agrees with
Marx and Freud. Like that of myth, the realm of ideologies is the realm of
false consciousness of reality, which must be explained at another level.
But, unlike Marx, he sees no historical remedy for this false conscious-
ness, no reconciliation of man and his true history.
Moreover, structural ethnology at long last promises an advance of the
human sciences to a methodology as rigorous as that of the exact sciences,
for in one sense the work of ethnological analysis resembles a laboratory
experiment. In both cases the object under observation is treated as a
natural object, is laid out in space, is amenable to any amount of experi-
mentation, and there is a great awareness of the distance between the
observer and the observed. In another sense, however, the methodologi-
cal quarrels of the ethnologists threaten the assimilation of ethnology to
the exact sciences. Lévi-Strauss is constantly aware both of the kinship
and of the gap still to be filled, hence the originality of his work and no
doubt also its wide range, which contrasts with the deliberately restricted
nature of ethnological description. Working with a few societies into
which history has not introduced its chaos and in which reduction to a few
variables is relatively easy, he has been able to satisfy his concern for
scientific rigor and his obsession with the linguistic ‘‘model” in the best
possible conditions.
It is true, as he often repeats, that he has never ventured beyond,
French Intellectuals 33

though, as always happens whenever a book becomes fashionable, im-


prudent or overzealous disciples tend to extend his method toward build-
ing up a general theory of societies. Frivolous or perhaps illusory though
this anticipation may be, it underlines the methodological influence of his
work. Is not this entomologist of human behavior, capable of the feat of
emerging from his own cultural world, the reverse of Sartrian man, on
whom engulfment in history and the emergence of a revolutionary prac-
tice impose his celebrated commitment? The fact that he belongs to the
same generation as Sartre and has lived through practically the same
history and is also considered to belong to the Left is thus immaterial; to
him these things belong to the domain of opinion and not to that of
science. And in what way does he belong to the Left? The term hardly has
meaning to a man who believes that, in their state of babbling infancy, the
human sciences to which he has devoted his life and outside which he
wishes to say nothing have for the time being nothing serious, still less
useful, to offer to the polis and its struggles—except a relapse into
ideology. In fact, their chance of one day deserving their name specifi-
cally depends on their keeping silent. “One may legitimately ask the
exact and natural sciences what they are, but the social and human
sciences are not yet in a state to render accounts. If these are required of
them, or if on political grounds it is considered clever to pretend to do so,
it will not be surprising if all one gets is phony balance sheets” (Revue
internationale des sciences sociales, 1964).
Is this scruple to be regarded as a real break with history? It would be
highly significant for our argument if it were so. But Lévi-Strauss con-
tinually denies it and pays explicit tribute to history, allowing it even to
“reserve its rights” (Leçon inaugurale at the Collège de France, 1960).
But until when? And what is the history to which he refers? Is it the
meaning of history created by Sartre’s dialectical reason? Evidently not,
for, except at the price of unscrupulous falsification, looking back cannot
demonstrate in the present the necessary outcome of the past. To Sartre a
man can always draw logical conclusions; to Lévi-Strauss he can only
think that he does so. A stubborn illusion grants the present a rather
foolish privilege without seeing that the present itself introduces into
successive events a connecting link drawn from its own depths. Thus, the
assurance of a retrogressive movement of truth vanishes under the eth-
nologist’s eye, and the philosophy of history becomes a myth, the neces-
sity of which only emphasizes its lack of substance. Lévi-Strauss, criticiz-
ing the fabrication of historical meaning by Sartre’s dialectical reason,
wrote this significant passage on the history of the French Revolution:
“Those described as men of the Left still cling to a period of contempor-
ary history that granted them the privilege of consistency between prac-
tical imperatives and patterns of interpretation. Perhaps that golden age
of the historical conscience has already passed; and the fact that one can
at least imagine the possibility shows that this was merely a fortuitous
34 History Today

situation, resembling the chance ‘focusing’ on a heavenly body of an


optical instrument with which it happened to be in a state of relative
motion. We are still ‘focused’ on the French Revolution; but, if we had
lived earlier, we might have been focused on the Fronde”’ (La Pensée
sauvage).
Once this pseudoscience has been exposed, however, the fact remains
that societies change and there are histories—if not a single history—that
are entitled to their observers. History, like Le Verrier’s planet, thus
becomes a perpetual ‘disturbing factor” introducing structural disequi-
libriums. Now, it is impossible to grasp everything at once; synchronism
and diachronism cannot be taken in at the same glance. Increasing one’s
knowledge in one field proportionally diminishes one’s chances of acquir-
ing it in another. Thus, structural ethnologists are needed to study order,
and historians to study disorder. This division of labor is equal only in
appearance, however. The study of structures has a double advantage,
chronological and logical: chronological because the description of struc-
tures provides it with a starting point, and this gives complete autonomy
to the work of the structuralists, while the opposite is not true (the work
of the historian is dependent, ornamental, or, at all events, relegated to a
distant future); and logical because, contrary to structural analysis, his-
torical analysis pulverizes norms when it deals with events. Moreover,
history is very difficult—perhaps impossible—to rationalize.
From this point of view, which is perhaps the most significant for our
argument, Marx and Sartre are on the same side of the barricade, that of
history, while Lévi-Strauss is on the other, that of structure. Like Hegel
and Marx, Sartre still describes an advent, a history of human fulfillment,
while Lévi-Strauss reduces multiple man to common mechanisms, dis-
solves him in a universal determinism, in the last resort displays him like a
natural object. His books, written with a rather precious rigor, are a
pitiless commentary on man’s nothingness, marking a probably fun-
damental epistemological breach with the “ideological era.’’ This is sys-
tematized by Foucault in Les Mots et les choses.
I do not wish to lapse into overly facile identifications, unduly confus-
ing books and authors, but from the point of view of this analysis it is
permissible to compare the work of Barthes and Foucault with that of
Lévi-Strauss, particularly considering the connection that intellectual
opinion has spontaneously established among the three. Their fields of
work are very different, but the methodological inspiration is the same: it
is to try to obtain an ethnological view of contemporary societies and
cultures. Foucault, imitating the use of the ethnologist’s cultural tele-
scope and reversing it, tries to gain more light from it that way. Lévi-
Strauss mingles the Jivaro world with his European outlook, while
Foucault sets out to consider European culture from a Jivaro angle in
order to conjure away its presence at last and turn it into a scientific
French Intellectuals 35

object. He tries to describe not individual patterns, which pertain to the


study of opinions, but the conceptual structures that in each period make
those opinions possible; the present intellectual revolution consists in his
view in the breach with historicism and the end of humanist anthropo-
centrism. This makes Sartre the last ‘‘nineteenth-century philosopher’—
which cannot be pleasing to him. Foucault’s methodological aggressive-
ness, probably one of the clues to the success of his book, is of interest in
that it tries to systematize the general significance of structuralism in
present-day European culture; the analysts of man’s ‘‘dissolution’’ have
succeeded the prophets of his advent.
If what Foucault says is indeed true and if structuralism confines Marx
to a nineteenth century intellectually dominated by history, it is very
curious and sociologically extremely interesting that structuralism should
have developed in France so systematically and so late and in the same
left-wing intellectual circles that (in the broad sense of the term) had been
Marxist since the liberation. This leaves us with the task of trying to
describe and understand this paradoxical phenomenon as well as the
curious and, I think, specifically French mutual contamination that has
taken place between Marxism and structuralism.
At the first level of analysis, it is evident that if Marxism continues to
lie at the heart of the French intellectual debate, it is less as a theory than
as an ideal, less as an intellectual tool than as a political heritage. Twenty
years have now passed since Sartre tried to reconcile the existentialist self
with Marxist determinism, that is, his theory of personal liberty with his
progressive opinions. It is this that forced him in his last philosophical
book to substitute his dialectics of individual liberty for Marxist dialecti-
cal materialism while quoting as ‘‘self-evident” and approving without
the slightest critical analysis the sum total of the propositions put forward
in Das Kapital, that is to say, the essentials of the Marxist philosophy of
history.’ By this subterfuge, he reconciles his philosophic conscience with
his political progressiveness but at the same time illustrates the profound
duality of his work and the uneven quality of his intellectual rigor. What
interests him is the establishment of new existential foundations for
human history and the revision of what he considers the mark of “‘scien-
tism”” in Marx. The respectful and distant doffing of his hat to Das Kapital
does not derive from the same level of analysis; it signifies merely his
allegiance to the left-wing intelligentsia, the resistance of yesterday, and
the present-day struggles against imperialism. It is the historical symbol
of an age that is the ideological age. Sartre speaks and will always speak as
an elder brother to all those who lived profoundly through the times of
fascism and communism. The genius of this professional philosopher
resides, paradoxically perhaps, in the secret of his sensibility and art
rather than in the clarity of his thought. In vain he refuses to be what he
has become, for he is caught up in the pitiless history of literature. He has
36 History Today

turned into a patriarchal figure, a revered elder aging in glory, a Nobel


Prize winner in spite of himself, but a Nobel Prize winner all the same.
That is the last trick played on him by “words.”
It is true that Lévi-Strauss has also emphasized his debt to Marxism,
but his debt is of a different kind. Having deliberately abstained from
postwar political struggles, shut in his academic ivory tower, he has felt
no need to state his position in relation to communism or anticom-
munism; the few interviews that—obviously without pleasure—he has
given the press betray more than prudence: a professional desire to keep
his distance from his own cultural world and the chaotic history that
introduces its disorders into it. Nevertheless, he can perhaps be consid-
ered more faithful than Sartre to Marx’s philosophic premises and mater-
ialist determinism. He has inherited from Marx the scientific ambition of
interpreting in intelligible terms the ideas that men form about the
natural and social worlds by another system underlying them; he accepts
the idea of a homogeneous society in which the determining factors will in
the last resort be the relations between man and nature. But he trans-
forms it profoundly by drawing up a veritable theory of superstructures.
One of the plainest symptoms of the end of the ideological age among
French intellectuals is, incidentally, this passion for the study of super-
structures, as if they desire to track down, unveil, and understand the
intellectual products of men and groups from their most hidden motiva-
tions. This is also the weakest point in Marx’s analysis. To Lévi-Strauss,
who made himself plain on the subject in La Pensée sauvage, the primacy
of infrastructures is like a hand dealt at cards; what societies do with the
hand imposed upon them depends on man’s cultural inventiveness. But
this inventiveness does not suggest an unlimited number of variables. On
the contrary, it is governed by structures and logical systems whose
appearance and mutations, far from being necessary and inevitable as so
many stages in an identical human evolution, depend on a calculation of
probabilities; this accounts for the concomitant multiplicity of societies
and cultures.
Admittedly, from other passages in his books, less materialist and less
Marxist interpretations can be drawn, for one can never be sure whether
the-logical structure brought to light by the analysis is the same nature as
the material produced by it or whether it informs reality. In fact, the
problem of whether it is materialist or Kantian (a Kantianism without
transcendental subject, Paul Ricoeur has said} is of little interest to
Lévi-Strauss, who seems to accept both hypotheses; that is to say, he
takes little interest in his philosophical relations with Marxism.
Moreover, he has always refused all extrapolation of his analytic
procedures to mythologies, beliefs, or “historical” societies; he did so
explicitly in reply to Edmund Leach, the British anthropologist, who
suggested a Lévi-Straussian ‘‘decoding” of the Book of Genesis—and the
contrast between the theoretical ambition implicit in his ideas and the
French Intellectuals 37

narrowness of their field of application rouses the mistrust of many of his


colleagues in the English-speaking world, who remain attached to the
empirical accumulation of facts.’ In France, however, it is this ambition,
or rather its special contribution to South American ethnology, that has
roused the enthusiasm of intellectuals and led rapidly to discussions at the
most general level—existentialism, Marxism, and structuralism. The left-
wing periodicals bear witness to this; the ethnologist has been conse-
crated a philosopher rather in spite of himself. Perhaps this should be
regarded, as do certain English-speaking anthropologists, as a special
feature of the French national tradition. But in this instance the fascina-
tion exercised by Lévi-Strauss over many Marxist or formerly Marxist
intellectuals seems to me to have a more specific explanation. It is born
neither of fraternity of political opinion (for Lévi-Strauss is the opposite
of a “‘committed” man) nor of philosophical kinship (which is highly
problematical, if it exists at all, and is in any case a matter of indifference
to Lévi-Strauss) but from an inverse relationship in which nostalgia for
Marx has been able to insert itself. Stated simply, the structural descrip-
tion of man as object has, in every respect, taken the place of the
historical advent of the man-god.
There is yet another and still more surprising aspect of the relationship
between the Marxists and the structuralists. It is significant that a whole
trend of Communist thought implicitly relies on structuralism, not to
break with Marxism but to renovate it. That is the meaning of the work of
Althusser and his friends, who are trying to restore the theoretical value
of the work of Marx and Lenin by rigorous analysis of their operative
concepts in order to free it of the banally humanist ideology with which
Garaudy diluted it. Instead of being concerned, like Sartre, with reconcil-
ing an epistemology with progressive ideas, they aim to marry structural-
ist method and Marxist theory. From this there emerges a Marxism
purged of its Hegelian paternity and of all contamination by bourgeois
humanism and differing from Marx’s ideas about his own doctrine, since
the latter is redefined by the bringing to light of its fundamental concep-
tual structures. It is on this condition alone, in Althusser’s view, that
Marxism can again become what it really is (which was masked by the
huge amount of social and historical sedimentation that has taken place):
that is to say, the theory, the science par excellence, as opposed to the
ideologies. This explains the curious path taken by Althusser. Trying as
he is to ‘“‘deideologize”’ Marxism, he nevertheless uses it both as an object
of study and as his sole point of scientific reference. Structural analysis,
an attempt to extend the methods of natural science to the “human
sciences,” is thus subtly diverted in the direction of Marxist dogmatism,
which is assumed as a self-evident premise—for the Marxist model is
assimilated from the outset to the scientific model. Hence the epistemo-
logical contradiction and political ambiguity of the works of Althusser
and his friends, which are devoted to restoring life to Marx by rigorous
38 History Today

analysis but are blocked by bigoted adherence to a conceptual apparatus


that, brilliant though it may have been, dates from another age and
another world.
Perhaps on another plane this ambiguity is significant of the develop-
ment of some Communist intellectuals who have remained in the party
since the destalinization crisis and of younger ones who have since joined
it in a climate that has become much more tolerant and much more
critical than that of the fifties. Althusser simultaneously offers structural-
ism and Marxism, critical study and doctrinaire intransigence, loyalty to
the French Communist party and reservations about it. There are many
reasons to suppose that a sociological study of his present audience would
disclose these factors, which are so characteristic of the present and
recent past of Communist intellectuals. No matter that they are contra-
dictory; it is precisely because they are contradictory that a structuralist
interpretation of Marx is able to offer them a temporary home. Intellec-
tual and sociological contradictions mutually explain and reinforce each
other; the structuralist ‘“deideologizing” of Marxism undoubtedly offers
a way of living through the end ofthe ideologies inside the Communist
world.
If this analysis is as a whole correct, if there is indeed a link between a
general phenomenon like the end of ideologies and the attractiveness of
structuralism in the special environment of the French intellectuals, it will
strike one as surprising that the disintegration of ideological certainties
and of the “meaning” of history has not led to a return to favor of
empirical research and the gathering of factual information in the manner
of the English-speaking world. Not that such research and the accumula-
tion of factual information is not growing in France; on the contrary,
there as elsewhere sociological surveys, public-opinion polls, and large-
scale investigations of archives are multiplying. But everything is more
than ever subordinated, even more than it was yesterday, to the search
for a general theory. Everything is happening as if the crisis of Marxism
had cleared the way for a methodology of another kind but at the same
level, heir to the same ambition for total, systematic understanding.
What for lack of a better term is called the Parisian ‘‘passion”’ for
structuralism—that is, its social success and its timing—would thus be
explained by its deep relationship, both contradictory and homogeneous,
to Marxism. The natural-sciences model has displaced the history model,
man as object has taken the place of man as subject, structure has taken
the place of determinist ambition and derives from the same desire to
decipher what underlies the apparent or merely conscious meaning of
human behavior.
It is probable that the end of the ideological age among the French
intellectuals embraces two distinct phenomena differing in nature. Desta-
linization, the Sino-Soviet schism, the crisis of the Third World—and
French and European prosperity—have punctured the progressivism of
French Intellectuals 39

the fifties that was so characteristic of the ideological age. Hence, there
was a receptiveness of intellectual opinion, a kind of expectation—rather
as a century ago the inglorious failure of the romantics of 1848 preceded
and facilitated the formation of the realist and positivist generation. But
it helped this generation to emerge; it did not create it: this intellectual
change does not itself call for sociological explanation. At the present
time, the political disappointments suffered by progressivism have pro-
foundly weakened the influence of Marxism among the left-wing intellec-
tuals, but it is Lévi-Strauss and not Raymond Aron who reigns in the void
thus created. What prevails is not a liberal and empirical criticism of
Marxism but a hyperintellectual and systematic way of thinking that aims
at a general theory of man. Not only have Marxists or former Marxists
been able to invest their past in this without repudiating it; they have also
rediscovered in it their ambition for a comprehensive science of man and
their old all-embracing dream, “‘deideologized” and freed from the
naivetés of commitment and of the meaning of history. To that extent
perhaps the case of the French intellectuals deserves to become a classic;
the end of the ideological age has found its doctrinaires.
Quantitative History

Quantitative history is fashionable just now both in Europe and in the


United States. Since the 1930s, historical research has been making
rapidly increasing use of quantitative sources and of calculation and
quantification procedures. But, like all fashionable phrases, quantitative
history has come to be used so sweepingly that it covers almost every-
thing, from critical use of the simple enumerations of seventeenth-
century political arithmeticians to systematic application of mathematical
models in the reconstruction of the past. Sometimes quantitative history
refers to a type of source, sometimes to a type of procedure; always, in
some way or other, explicit or not, to a type of conceptualization of the
past. It seems to me that if one goes from the general to the particular and
tries to pinpoint the specific nature of historical knowledge in relation to
the other social sciences, one can distinguish three groups of problems
relating to quantitative history.
1. The first group concerns the methods of treating the data: problems
having to do with the formation of different families of data, the geo-
graphical unity of each family, and its internal subdivisions; with correla-
tions between different series; with the values, in relation to the data, of
different models of statistical analysis; with the interpretation of statisti-
cal relationships; and so on.

First published in an English translation by Barbara Bray as “Quantitative History,”


Daedalus, Winter 1971, 151-67. Reproduced by permission of Daedalus, Journal of the
American Academy of Arts and Sciences, Boston, Mass. A revised version in French was
published as “L’Histoire quantitative et la construction du fait historique,” Annales:
économies, sociétés, civilisations, January-February 1971, 63-75, reprinted as “Le Quanti-
tatif en histoire” in Jacques Le Goff and Pierre Nora, eds., Faire de l’histoire (Paris:
Gallimard, 1974), vol. 1, Nouveaux problèmes, 42-61. The present text is a slightly revised
and expanded reprint of the earlier English translation.

40
Quantitative History 41

These problems belong to the technology of research in the social


sciences. It is true, they may also include questions of methodology, not
only because no technique is “‘neutral,’’ but because, more specifically,
all statistical procedures are bound to raise the problem of whether, and
to what extent, historical or sociological knowledge is compatible with, or
can be dealt with exhaustively by, mathematical conceptualization of a
probabilistic kind. But neither the technical nor the theoretical debate is
specific to history: both arise in connection with all the social sciences. In
this respect, quantitative history is no different from, for example, what is
now called ‘“‘empirical sociology,” which in this context is simply contem-
porary quantitative history.
2. Quantitative history also refers, at least in France, to the aims and
researches of certain economic historians,’ who attempt to turn history
into a kind of retrospective econometrics,’ or, in other words, on the basis
of modern national accounting, to fill in all the columns of an imaginary
input-output table for past centuries. The champions of this econometric
history advocate total and systematic quantification, in their view indis-
pensable both for the elimination of arbitrariness in selecting data and for
the use of mathematical models in their processing. This processing is
based on the concept of general equilibrium as imported into economic
history from political economy.
According to this argument, genuine quantitative history would be the
result of a twofold reduction of history: first, at least provisionally, the
reduction of its field to economics; and, second, the reduction of its
descriptive and interpretative system to the one worked out by the most
rigorously constituted of the social sciences today, political economy.
The same analysis could be applied to demography and demographic
history: here again, a conceptually constituted science indicates the data
and supplies the methods for a particular historical discipline, the latter
thus becoming a sort of by-product of the other discipline, whose ques-
tions and concepts it merely transposes into the past.
Of course, there have to be data for the past just as for the present; or
at least it has to be possible to work them out with a sufficient degree of
accuracy or to reconstruct or extrapolate them. This necessity sets the
first limit to the complete quantification of historical data. Complete
quantification, even if possible at all before the nineteenth century, could
not go back beyond the introduction of the statistical or protostatistical
recording of data, which coincides with the centralization of the great
European monarchies. But history did not begin with William Petty or
Sébastien Vauban.
Moreover, there is no reason why the historian should agree, even
provisionally, to have his field of research reduced to economics or
demography. There are two alternatives. Either history is only the study
of a previously determined, limited sector of the past, into which mathe-
matical models established by certain social sciences are imported in
42 History Today

order to be tested. In which case, we come back again to contemporary


political economy, which seems to me the only one of the social sciences
with such models at its disposal. History then becomes nothing more than
an additional field of data. Or—the second alternative—one takes history
in the widest sense, that is, as a discipline not strictly reducible to a set of
concepts and with countless different levels of analysis, and then ad-
dresses oneself to describing these levels and establishing simple statisti-
cal connections between them on the basis of hypotheses that, whether
original or borrowed, depend on the intuition of the researcher.
3. This is why, even if one qualifies history as ‘‘quantitative,” one
cannot escape what is the specific object of historical research: the study
of time, of the diachronic dimension of phenomena. But, looked at from
this point of view, quantitative history’s most general and elementary
object is to form historical fact into temporal series of homogeneous and
comparable units, so that their evolution can be measured in terms of
fixed intervals, usually years. This fundamental and logical operation
constitutes what Pierre Chaunu has called ‘serial history,” a necessary
though not sufficient condition of strictly quantitative history as defined
above. For serial history offers the conclusive advantage, from the scien-
tific point of view, of substituting for the elusive “event” of positivist
history the regular repetition of data selected or constructed by reason of
their comparability. It does not, however, claim to give an exhaustive
account of the whole body of evidence nor to be a comprehensive system
of interpretation nor to be a mathematical formulation. On the contrary,
the division of historical reality into series leaves the historian confronted
with his material broken down into different levels and subsystems,
among which he is at liberty to suggest internal relationships if he
chooses.
Defined in this way, quantitative and serial history emerge as at once
connected with and distinct from each other. But they share an
elementary basis in that both substitute the series for the event, both
make a construction from historical data in terms of probabilistic analy-
sis. To the classic question What is a historical fact? they both give a new
answer that transforms the historian’s raw material—time. It is about this
internal transformation that I should like to put forward a few ideas.

To avoid any misunderstanding, let me say at once that this essay does
not set out to prescribe quantitative history as the only kind permissible.
During the last ten or twenty years, serial history has turned out to be one
of the most fertile approaches in the advancement of historical knowl-
edge. It has also the immense advantage of introducing into the ancient
discipline of history a rigor and efficiency superior to those of qualitative
methodology. But it is nevertheless true that there are important sectors
of historical reality that it is by nature unable to treat or even to approach,
either for circumstantial reasons such as irremediable lack of data or for
fundamental reasons such as the irreducibly qualitative nature of the
Quantitative History 43

phenomenon concerned. This explains why, for example, historians of


antiquity, who work with data very discontinuous in time, or specialists in
intellectual biography, concentrating particularly on what is unique and
incomparable in creativeness, are usually less attracted by serial history
than, say, historians of the agrarian structures of modern Europe.
From this point of view, another, and perhaps more basic, problem
should be raised. Serial history undoubtedly provides accurate methods
for measuring change, but to what extent does it enable the historian to
interpret change? By definition, data series are composed of identically
constructed units, in order to make comparisons possible. The long-term
variation of these units in time, when it takes the form of cycles, brings us
back to what one could call change within stability and thus to an analysis
in terms of equilibria. But sometimes the temporal variation of one or
more series takes the shape of an open-ended growth trend, that is, a
cumulative process. If one breaks down this trend into relatively small
units (annual or ten-year periods, for example), it becomes harder to
define the threshold beyond which the time structure and the rates of
change are transformed. This raises tremendous problems of dating and
periodization. Moreover, the decisive historical change cannot be in-
serted into any series endogenous to a given system; instead, it must be
the product either of an innovation unrecorded in any previous quan-
tification or of an exogenous factor that upsets the centuries-old equilib-
rium of the system. These methodological problems lie at the heart of the
present debate over the question of industrial takeoff.’ In other words, if
it is true that no methodology is innocent, serial history, by concentrating
on the long term and on the equilibrium of systems, seems to me to put a
premium on conservation. It is a good antidote to the identification of
history with change—a nineteenth-century inheritance—and thus a cru-
cial phase in the development of history as a discipline. Nevertheless, the
premises and limits of serial history must also be realized.
But the fact that serial history has limits—a subject that might be
discussed on another occasion—does not excuse intellectual indolence or
uncritical observance of tradition. If a number of historians today are
breaking away from narrative in order to concentrate on problems, this is
largely due to changes in the pieces of the puzzle used by historians to
reconstruct the picture of the past. Because of serial history, they are now
confronted with a new panorama of data and a new awareness of the
premises of their profession. It is probable that we have not yet exhausted
the possibilities offered by quantitative methods.

The Historian and His Sources

Quantitative history presupposes the existence and elaboration of long


series of homogeneous and comparable data, and the first problem that
presents itself in new terms is that of sources. In general, European
44 History Today

archives were formed and classified in the nineteenth century, in accord-


ance with procedures and criteria reflecting the ideological and methodo-
logical preoccupations of the period. This meant that national values
predominated and that priority was given to politico-administrative
sources. It also meant that documents were preserved and classified in
accordance with the special and limited purpose of a particular inquiry:
archives were built up to witness to events rather than to time. They were
constituted and criticized in themselves and not as factors in a series; the
point of reference was external. What was in question was the historical
“fact” of the positivists, the naive mind’s illusory sheet anchor in what is
supposed to be real, as distinct from mere testimony—a particular,
discontinuous, elusive sequence within either an indefinite flux or a
chronology preestablished in terms of centuries, reigns, and ministers. In
short, archives are the memoirs of nations, just as the letters a person
keeps show what an individual has chosen to remember.
But the data of quantitative history refer not to some external, vaguely
outlined “‘fact’’ but to internal criteria of consistency. A fact is no longer
an event selected because it marks’a high spot in a history whose meaning
has been predetermined, but a phenomenon chosen and sometimes
constructed by reason of the recurrence that makes it comparable with
others in terms of some unit of time. The whole conception of history
based on archives is radically transformed at the very time when its
technical possibilities are multiplied by the electronic processing of in-
formation. This simultaneous and interconnected revolution in method-
ology and technique enables us to think in terms of a new kind of archive
preserved on perforated tapes. Such archives would not only be built up
according to a deliberately planned system; their criteria would also be
quite different from those of the nineteenth century. Documents and data
exist no longer for themselves but in relation to the series that in each case
precedes or follows: it is their comparative value that becomes objective,
rather than their relation to some elusive ‘‘real’’ substance. Thus, in-
cidentally, the old problem of the critique of historical documents moves
on to different ground. “External” criticism is no longer based on credi-
bility as derived from contemporaneous texts of another kind, but on
consistency with a text of the same kind occurring elsewhere in the
temporal series. “Internal” criticism is simplified inasmuch as many of
the necessary cleaning-up operations can be entrusted to the memory of
the computer.
Consistency is introduced at the outset, when the data are first sorted
out, by a minimal formalization of each document that makes it possible
to retrieve, over a long period and for each unit of time, the same data in
the same logical order. From this point of view, the historian’s use of
computers is not only an enormous practical advance in the time it saves
(especially when the sorting is done verbally by tape-recording, as in the
Couturier method);° it is also a very useful theoretical discipline, in that
Quantitative History 45

the formalization of a documentary series that is to be programmed


forces the historian from the very beginning to abandon epistemological
naiveté, to construct the actual object of his research, to scrutinize his
hypotheses, and to make the transition from implicit to explicit. The
second critical process, this time an internal one, consists in testing the
consistency of the data themselves in relation to those that come before
and after—in other words, in eliminating errors. It thus emerges as a sort
of consequence of the first process and can in fact be done largely by
automation through programmed methods of verification.
Naturally enough, serial history in its manual form began by using
those historical series that were easiest to handle, that is, economic,
fiscal, and demographic documentation. The revolution introduced by
the computer into the collection and processing of data has steadily
multiplied the extent to which such numerical series can be explored. The
technique can now be applied to any kind of historical data reducible to a
language that can be programmed—not only tax rolls and market price
lists, but also series of relatively homogeneous literary collections such as
medieval chartularies or the cahiers of the states general of monarchical
France.
Thus emerges the first task of serial history, the imperative of its
development: the constitution of its subject matter. Classical historiogra-
phy was constructed from archives worked on and processed according to
the critical rules bequeathed to us by the Bendictines of the age of
Enlightenment and the German historians of the nineteenth century. The
serial history of today has to reconstruct its archives in terms of the dual
methodological and technical revolution that has transformed the rules
and procedures of history.
This being so, the question arises of the problematical nature of
history’s subject matter, the hazards of its survival, its partial destruction
and sometimes total disappearance. I am not sure this question distin-
guishes history as much as is sometimes alleged from the other human
sciences whose objects are more specifically defined. The characteristic
feature of history is the extraordinary and almost unlimited elasticity of
its sources. As the researcher’s curiosity roves further and further, huge
dormant areas of documentation are revealed. What nineteenth-century
historian bothered with the parish registers that have now become,
especially in England and France, one of the surest sources of our
knowledge of preindustrial society?
Moreover, if the researcher invests them with a new significance,
sources already exploited once can be used again for other purposes.
Descriptions of price movements can lead to sociological or political
analyses; Georges d’Avenel is followed by Ernest Labrousse. Demo-
graphic series studied from the point of view of, for example, the use of
contraception by married couples can also throw light on problems of
mental attitude or religious practice.° Signatures to legal documents can
46 History Today

give statistics about the spread of literacy. Biographies systematically


grouped in terms of common criteria, on the basis of a given working
hypothesis, can build up documentary series imparting an entirely new
life to one of the oldest kinds of historical narrative.
Hitherto, history has been almost exclusively based on the written
traces of men’s existence. No doubt live interrogation, which provides
empirical sociology with so much of its data, will always be beyond the
reach of the historian except for the period in which he lives. But how
much unwritten evidence there is still to be cataloged and systematically
described. The physical conditions of rural life, the divisions of the land,
iconography sacred and profane, the layout of early towns, what the
houses were like inside—one could go on forever listing the elements of
civilization that, once cataloged and classified in detail, would make it
possible to establish new chronological series and put at the historian’s
disposal the new subject matter that the conceptual enlargement of
history demands. For it is not the sources that determine the approach,
but the approach that determines the sources.
Of course, this type of argument must not be pressed too far. To the
documentary demands of certain contemporary social sciences, history
can answer only with irreparable gaps. It is difficult to see what substitu-
tions or extrapolations could ever fill in the columns of an input-output
table of the French economy in the time of Henry IV, not to mention
periods even more distant. But all this means, really, is that, concep-
tually, history is not reducible to political economy. The problem of
sources, for the historian, lies not so much in absolute lacunae as in series
that are incomplete, and this not only because of the difficulties of inter-
and extrapolation, but because of the chronological illusions they may
lead to.
Take the classic example of popular revolts in France at the beginning
of the seventeenth century. Because of the great abundance of adminis-
trative documents relating to the subject at that time, this period has
become the best-known chronological sector in the history of peasant
risings between the end of the Middle Ages and the French Revolution.
The hazards of survival have even seen to it that a large part of these
archives, the Fonds Séguier, ended up in Leningrad, enabling Soviet
historians to advance a Marxist interpretation of France’s ancien régime.
The subsequent controversy has enhanced the interest of the documents
still further. But another problem arises before that of interpretation, and
this concerns the presupposition common to both interpretations here:
that is, that there really was, during the period when the absolutist state
was coming into being and there was probably a rapid increase in taxa-
tion, a special chronological concentration of that classic phenomenon in
French history, the jacquerie. The existence of such a concentration could
be definitely established only by the study of a long homogeneous series
and comparison between this section of it and those before and after. But
Quantitative History 47

for several reasons such a series cannot be constructed. In the first place,
there is no unique and homogeneous source for such revolts over a long
period. Moreover, there is every reason to believe that the survival of
such a collection as the Fonds Séguier in Leningrad, a collection espe-
cially rich in this respect but limited to the papers of one family and thus
subject to the hazards and possible distortions of individuals’ careers,
falsifies our chronological perception of the subject. In any case, a
Jacquerie is a story without direct sources, a rising of illiterates. We can
glimpse it today through the medium of administrative or legal archives,
but by this very fact, as Charles Tilly has remarked, every revolt that
escapes repression escapes history. The relative richness of our sources
during a given period may be a sign of changes that are institutional
(reinforcement of the apparatus of repression) or purely individual (spe-
cial vigilance on the part of a particular official), rather than of any
unusual frequency in the phenomenon itself. The difference between the
number of peasant risings under Henry II and under Louis XIII may
reflect first and foremost the progress of monarchical centralization.
Therefore, in handling serial sources the historian is forced to think
carefully about the influence that the way they were constituted may have
on their quantitative application. I think we may distinguish between
such sources as follows, in order of increasing complexity in their conver-
sion into series.
1. Structurally numerical sources, grouped together as such and used
by the historian to answer questions directly connected with their original
field of investigation; for example, French parish registers for the demo-
graphic historian, prefectoral inquiries into industrial or agricultural
statistics in the nineteenth century for the economic historian, the data on
American presidential elections for the specialist in sociopolitical history.
These sources sometimes need standardizing (as when there is a variation
in local units or a modification of the classifying criterion); also, when
there are gaps in the documentary sequence, one may have to extrapolate
certain elements. But in such cases both operations are carried out with
the minimum of uncertainty.
2. Sources that are structurally numerical but used by the historian
substitutively, to find the answers to questions completely outside their
original field of investigation; for example, the analysis of sexual behavior
on the basis of parish registers, the study of economic growth through
price series, the socioprofessional evolution of a population through a
series relating to taxes. Here the historian encounters a double difficulty.
He has to define his questions all the more meticulously because the
documentary material was not assembled with them in mind, and the
question is constantly before him of the relevance of such material to such
questions. He usually has to reorganize the material completely in order
to make it usable and in so doing makes it more arbitrary and so more
open to objection.
48 History Today

3. Sources that are not structually numerical but that the historian
wants to use quantitatively, by a process involving two substitutions. In
such cases he has to find in his sources a univocal significance in relation to
the question he is asking. He also has to be able to reorganize them in
series, that is, in comparable chronological units, and this demands an
even more complex process of standardization than in the preceding
paragraph. Data of this third type, which become more and more fre-
quent the further one goes back into the past, can be subdivided into two
classes: first, nonnumerical sources that are nevertheless serial and thus
easily quantifiable, such as modern European marriage contracts drawn
up by notaries, which, according to the historian’s choice, can give
evidence about endogamy, social mobility, income, literacy, and so on;
second, the sources that are strictly qualitative, and therefore not serial,
or at least particularly difficult to standardize and arrange in series, such
as the administrative and legal series referred to above or iconographical
survivals of forgotten faiths.
Whichever kind of source he is dealing with, the historian of today has
to rid himself of any methodological naiveté and devote a good deal of
thought to the way in which his knowledge is to be established. The
computer gives him the leisure to do so by freeing him from what used to
take up most of his time—the recording and card-indexing of data—but
at the same time it demands from him rigorous preliminary work on the
organization of series and their meaning in relation to the inquiry. Like
all the social sciences, but perhaps with a slight time lag, history is passing
from the implicit to the explicit. The encoding of data presupposes their
definition; their definition implies a certain number of choices and
hypotheses, made all the more consciously because they have to conform
to the logic of a program. And so the mask finally falls away of that
historical objectivity that was supposed to lie concealed in the facts and to
reveal itself at the same time as them. Henceforward the historian is
bound to be aware that he has constructed his own “facts” and that the
objectivity of his research resides in the use of correct methods for
elaborating and processing them and in their relevance to his hypotheses.
So serial history is not only, or even primarily, a transformation of the
raw material of history; it is a revolution in the historiographical con-
sciousness.

The Historian and His ‘‘Facts’’

The historian, working systematically on chronological series of


homogeneous data, is really transforming the specific object of his knowl-
edge—time, or rather his conception and representation of it.
1. The so-called histoire événementielle is not to be defined by the
preponderance it gives to political facts. Nor is it made up of a mere
Quantitative History 49

narrative of certain selected “‘events’’ along the time axis. First and
foremost, it is based on the idea that these events are unique and cannot
be set out statistically and that the unique is the material par excellence of
history. That is why this kind of history paradoxically deals at one and the
same time in the short term and in a finalistic ideology. Since the event, a
sudden irruption of the unique and the new into the concatenation of
time, cannot be compared to any antecedent, the only way of integrating
it into history is to give it a teleological meaning. And as history, espe-
cially since the nineteenth century, has developed primarily as a mode of
interiorizing and conceptualizing the sense of progress, the ‘‘event”
usually marks some stage in the advent of a political or philosophical
ideal—republic, liberty, democracy, reason, and so forth. The historian’s
ideological consciousness can assume very subtle forms. It may group
knowledge relating to a certain period around unifying schemas not
directly linked to political options or values; for example, the spirit of an
age, its Weltanschauung. But basically the same compensating mecha-
nism is at work: in order to be intelligible, the event needs a general
history apart from itself and independently determined. Hence the classic
conception of historical time as a series of discontinuities described in the
mode of continuity—that is, as narrative.
Serial history, on the other hand, describes continuities in the mode of
discontinuity: it is a problem-oriented history instead of a narrative one.
Because it has to distinguish between the levels of historical reality, it
breaks down all previous conceptions of general history, calling in ques-
tion the old postulate that all the elements of a society follow a
homogeneous and identical evolution. The analysis of series has meaning
only if it is done on a long-term basis, so as to show short or periodic
variations within trends. The series reveals a time that is no longer the
mysterious occasional spurt of the event but an evolutionary rhythm that
is measurable, comparable, and doubly differential in that it can be
examined within one series or as between two or more.
A wedge has thus been driven into the old carefully enclosed empire of
classical historiography, and this by means of two distinct but connected
operations. First, by the analytical breakdown of reality into different
levels of description, serial history has opened history in general to
concepts and methods imported from the more specifically constituted
social sciences, such as political economy. This opening has probably
been the operative factor in the recent historical revival. Second, by
quantitatively analyzing the different evolutionary rhythms of the differ-
ent levels of reality, it has at last turned into a scientifically measurable
object the dimension of human activity that is history’s raison d’être—
time.
2. Now that the historian’s hypothesis has shifted from the level of the
philosophy of history to that of a series of data both particular and
homogeneous, it usually reaps the advantage of becoming explicit and
50 History Today

formulable; but, at the same time, historical reality is broken down into
fragments so distinct that history’s classic claim to give a universal view of
things is endangered. Must the claim be abandoned?
I would say it may probably be kept as a goal on the horizon but that, if
history wants to go forward, it should abandon this ambition as a point of
departure. Otherwise, it might fall once more into the teleological illu-
sion described above. Present-day historiography can progress only in-
sofar as it delimits its object, defines its hypotheses, and constitutes and
describes its sources as carefully. as possible. This does not mean it has to
restrict itself to microscopic analysis of one chronological series. It can
group several series together and put forward an interpretation of a
system or subsystem. But today a comprehensive analysis of the “system
of systems” is probably beyond its power.
We may take as examples demographic and economic history, the
most advanced sectors now in France and probably elsewhere. It so
happens that for the past twenty years or so the period that in France is
called ‘‘modern’’—that is, the period between the end of the Middle Ages
and 1789—has been the subject of the largest number of studies in serial
history, both demographic and economic. So it is the one we are least
ignorant about from this point of view. French historiography,’ starting
out from reconstructed commodity price series, has compared these with
the evolution in the number of people as shown in demographic series.
Thus, there has gradually been built up the concept of an “economic
ancien régime,” based on the preponderance of a cereal production
exposed to the vagaries of meteorology and on the periodic purging of the
system by recurrent crises. These crises are indicated by sudden steep
rises in price curves and the collapse of those indicating size of popula-
tion.
But the price series, the meanings of which can be quite varied and
ambiguous, have been supplemented by more specific indications con-
cerning volume of production and by the use of series relating to the
evolution of supply and demand, itself a factor in the evolution of prices.
On the subject of production, though the tithe records concern the same
percentage of the harvest every year and so tell us nothing absolute, they
are valuable because of their relative comparability. For production we
also have the protostatistical sources brought together by the administra-
tion of the ancien régime and possibly reorganized on a national scale. On
the subject of demand, in addition to general demographic records, we
can also turn to the reconstruction of the great masses of liquid money:
the treasuries of communes and seigneuries, tithes, rents, profits, wages.
The combination of many demographic and economic series has re-
cently enabled Emmanuel Le Roy Ladurie to make a wider analysis of
the old agrarian economy.‘ His book gives a sampling of data covering the
whole of the Languedoc, a long-term chronology (from the fifteenth to
the eighteenth century), and a rich and varied quantitative documenta-
Quantitative History 51

tion. Thanks to the cadastral surveys, the latter makes possible a study of
rural property owning. The fifteenth to the eighteenth centuries tell the
story of a very long agrarian cycle characterized at the same time by a
general equilibrium and by a series of states of disequilibrium. The
general equilibrium roughly corresponds to the Malthusian model—the
model that Malthus discovered and made immortal at the very moment it
ceased to be true, at the time of England’s takeoff. The economy of early
rural Languedoc was dominated in the long term by the relation between
agricultural production and the number of men. Society’s inability to
raise agrarian productivity and the absence of an unlimited reserve of
cultivable land, together with the famous “monetary famine” beloved of
the price historians, presented structural obstacles to decisive growth.
Though the monetary explanation loses its central role, it is integrated
into an interpretation both more complex and more unified.
The structure of the old economy acted in the long term as an internal
regulator. Nevertheless, within the system, the different variables—
number of men, evolution of property, distribution of income from rents,
fluctuations in productivity and prices, and so on—make it possible to
distinguish separate periods in accordance with the position each variable
occupies in relation to the whole, in terms of the annual rhythms and
cycles of each particular curve. The complete structure thus chronologi-
cally comprises several types of combinations of series, that is, several
different situations. And, in fact, it is through the careful examination of
these successive situations, and the features they have in common or in
which they differ, that the structure itself emerges. This, incidentally,
may shed some light on the dispute over synchronic and diachronic,
which often divides anthropologists and historians and which at present is
at the heart of the evolution of the social sciences. The short- or medium-
term periodic movement that constitutes an “event” on the economic
plane does not necessarily clash with the theory of general equilibrium.
On the contrary, an empirical description of such movements may make
it possible to define the theoretical conditions of the equilibrium, whose
elasticity indicates the limits within which it operates.
3. The Languedoc example quoted above is a special case in that the
correlation there between the different demographic and economic series
is made within a comparatively homogeneous region and a limited sphere
of human activity (agrarian economy). The sectional application of serial
history to different areas usually leads to the analysis of regional or
national disequilibriums. And general, or would-be general, serial his-
tory, even when restricted to a limited geographical area, tends to lead to
the analysis of temporal disequilibriums between the different evolution-
ary rhythms of each level of human activity.
The first point is now well known, thanks to the increasing number of
studies on regional economic history. The specialist is used to the idea of
there being measurable differences between different countries and be-
52 History Today

tween areas influenced in different degrees by the same situation or


reacting in different ways to similar situations occurring at different
times. There are countless examples, some of which raise problems that
have become classic in European history. Such, for instance, are the
recently revived question of the comparative growth of France and
England in the eighteenth century;’ the antithesis between the rise of
agriculture in Catalonia in the eighteenth century and its decline in
Castile, which has been shown by Pierre Vilar," and the contrast in
seventeenth-century France between the Beauvaisis revealed by Pierre
Goubert," poor and seriously stricken by the middle of the century by
economic and demographic recession, and the Provence described by
René Baehrel,” comparatively fortunate, or at least not affected by the
downturn of expansion until much later. More generally, the date of this
reversal, this plunge into the “‘tragedy”’ of the seventeenth century, varies
considerably according to region and to the nature of the local economy.
It becomes increasingly unlikely that there was only one economic con-
joncture for both urban and rural economies.”
Thus, serial history opens out at once into the analysis of situations
either differential or simply separated from one another in time (in other
words, into what might be called the geography of serial history’s chro-
nology) and into the study of structural differences that may be indicated
by chronological discrepancies. Cycles occurring at different times in the
same or different regions, but fundamentally comparable internally,
exhibit only geographical variations of the same theme. Contradictory
developments, on the other hand, whether within the same geographical
area (for example, between town and country) or between two different
areas, may present the historian with differences in economic structure.
But history cannot be reduced just to the description and interpreta-
tion of economic activity. If it has a specific character distinguishing it
from the other social sciences, this consists precisely in having no specific
character and in claiming the right to explore time in all its dimensions. It
is easy to see why economics has been the primary sphere of quantitative
history —because of the necessarily measurable character of its indica-
tors; by the preciseness of the concepts it makes available; and by its
theoretical approach in terms of growth, the favorite image for historical
change in Western thought today. But man is not merely an economic
agent. The contemporary world offers too many examples of cultural
resistance to the general adoption of growth on the Western model for
the historian not to mistrust the Manchester-school approach to progress
(or its Marxist inversion). He is bound to want to analyze the societies of
the past in terms of politics and ideology as well as economics.
Even so, he does not and cannot revert to the old teleological history of
progress, which extrapolates into cultural life the rhythms of economic
development, whether this is supposed to occur by a kind of peaceful,
natural adaptation or through the necessary medium of revolution. These
Quantitative History 53

ideological postulates of another age are useless now; it is not by clinging


to them that the historian can preserve the universality of history. The
only way to do so is by setting out to list and describe by the methods of
serial history the other levels of human activity besides the objective
processes of economics; by starting from the hypothesis that chronologi-
cal rhythms, and the researcher’s attitude to time, may vary according to
the different levels of reality or to the particular part of the system being
analyzed.
On the practical plane, almost everything still remains to be done. The
historian must look for the possible indicators, quantifiable or not, of
what may be called a “‘politico-ideological’’ society; he must establish its
documentation and what constitutes representativity and comparability
from one period to another. For all this, there are sources as abundant
and series as homogeneous as in the field of economics or demography:
they exist for popular literacy," the sociology of education and religious
sentiment,’ the absorption of ideas by elites, the manifest or latent
contents of political ideologies, and so on. On the theoretical plane, the
main task, of course, is to build up gradually the components of a
comprehensive history but first and above all to analyze the different
rhythms of development at various levels of a historical complex. This is
the only way to achieve two of the priorities of historiography today.
These are:
1. To revise the traditional general periodizations, which are mainly
an ideological inheritance from the nineteenth century and which presup-
pose precisely what is still to be demonstrated, that is, the roughly
concomitant development of the most diverse components of a historical
complex within a given period. Instead of beginning from a set of
periodizations, it is probably more useful to start by examining the
components concerned. It is probable, for example, that while the con-
cept of the Renaissance is relevant to many of the indicators of cultural
history, it is devoid of meaning in relation to the data for agricultural
productivity.
2. Then to define, within a complex of data of different kinds, which
levels are developing rapidly or changing decisely and which in the
medium or long term are in a state of inertia. It is not clear, for example,
that the dynamism of French history from, say, the great expansion of the
eleventh and twelfth centuries is economic in character; it may be that
education, culture (in the broad sense), and the state (the latter through
the various public offices) played a more fundamental role here than the
increase in the national product. Perhaps I shall be allowed to conclude
on this bold hypothesis if I add that it will remain unverifiable until
general history has sat at the feet of serial history.
From Narrative History to
Problem-oriented History

History is the child of narrative. It is not defined by an object of study but


by a type of discourse. To say that history studies time is just another way
of saying that it arranges all of its objects of study in a temporal
framework: to produce history is to tell a story.
A narrative, then, is an account of ‘“‘what happened”: to someone or
something, to an individual, to a country, to an institution, to the people
who lived before the moment of the narrative, and to the products of their
activity. Narrative history brings to life the tangle of events that make up
the fabric of an existence, the thread of a lifetime. Its model is naturally
biography, because the latter describes something that man can view as
the quintessential image of time: the clean-cut duration of a lifetime from
birth to death and the identifiable dates of the major events that took
place between its beginning and end. The choice of a chronological
segment is inseparable here from the empirical nature of the “‘subject” of
the story.
A history of France or of any other country basically follows the same
logic: it can begin only with the country’s origins, followed by an account
of the stages of its growth and the adventure of the nation, illustrated by
chronological divisions. The only difference is that such a history leaves
the future open; however, the narration of the past—the description of
the nation’s patrimony—is also intended to give some indication of this
future and thus to freeze time.

First published as “De l’histoire-récit à l’histoire-problème,” Diogéne, no. 89 (January-


March 1975), 113-31. The present text is a substantially revised version of the translation by
Susanna Contini, “From Narrative History to History as a Problem,” Diogenes (English-
language edition of the same journal), no. 89, 106-23. Reproduced by permission of the
International Council for Philosophy and Humanistic Studies.

54
From Narrative History to Problem-oriented History o>

Historical narrative, therefore, must follow a division of time inherent


in the raw datum of experience. Essentially, it records the recollections of
individuals and communities. It keeps alive what they have chosen of
their past, or of the past in general, without taking apart or reconstructing
the objects within that past. In other words, it deals with moments, not
objects. Even when it discusses or tries to discuss, “‘civilizations,”’ this
kind of history cannot avoid the rule. When Voltaire compares the age of
Pericles or of Augustus to that of Louis XIV, the concrete incarnations of
these successive periods of greatness are proof enough that he is compar-
ing periods, not concepts.
No doubt that is one of the reasons why narrative history has been
primarily, although not exclusively, biographical or political. Within the
collective experience of humanity, what witnesses have found most fasci-
nating and what lends itself best to narrative is the saga of great men and
of states. Little wonder, then, that history, first in Greek and Roman
antiquity, then in modern Europe, developed into chronicles of power
and war. The divisions of the narrative tended to underline the misfor-
tunes and victories of mankind—the great moments of history.
The events in such a history consist precisely of moments. Their
ephemeral nature is what characterizes them above all else. Events are
the unique points in time in which something happens that cannot be
assimilated to what has come before it or what will come after it. That
“‘something’’—the historical fact promoted to the rank of event—can
never be compared, strictly speaking, to a preceding or subsequent fact,
since it is its empirically unique nature that determines its importance.
The battle of Waterloo and Stalin’s death occurred only once; they
cannot be likened to any other battle or any other death, and they have
transformed world history.
And yet an event, if considered in isolation, is unintelligible. It is like a
pebble picked up on a beach—meaningless. For it to acquire significance,
it must be integrated into a pattern of other events, in relation to which it
will become meaningful. That is the function of narrative. Waterloo can
acquire significance in the context of a history of Napoleon’s life, the First
Empire, or nineteenth-century Franco-British rivalry. Stalin’s death be-
comes important in the context of the history of twentieth-century Rus-
sia, international communism, or any other imaginable chronological
constellation of facts. Thus, in narrative history, an event, even though it
is by definition unique and not comparable, derives its significance from
its position on the axis of the narrative, that is, on the axis of time.
Since an event is not an object intellectually created to be studied, it
cannot acquire significance by means of an analysis of its relationship to
other comparable or identical objects within a system. As it belongs to
the realm of experience, of what has happened, it cannot be organized or
even simply named except in relation to the external and general signifi-
cance of the historical period of which it is one of the features. All
56 History Today

narrative history is a succession of origin events, or, if one prefers, a


history of events. And all history of events is teleological history: only the
“ending” of the history makes it possible to choose and understand the
events that compose it.
That ending can differ considerably from one historian to another and
according to their chosen topics. For a long time endings were enveloped
in religious apologetics or moral edification, which is no longer fashion-
able. The same cannot be said for the glorification of national power or
national consciousness, which is still one of narrative history’s most
important functions, after having been, no doubt, its initial mainspring.
All peoples need an account of their origins and a memorial to their times
of greatness that can serve at the same time as guarantees of their future.
Just as the ability to write brings power, so our archives are the memories
or symbols of power. Not even transnational history, generally referred
to as history of civilizations, can escape the inevitable obligation to assign
a prior meaning to time. In our secular world, narrative history, apart
from emphasizing national consciousness, more often than not embodies
the other great collective experience of mankind since the eighteenth
century: the feeling of progress. Progress assumes different names and
aspects; it sometimes refers to the development of material goods, but
more often to the problematic advent of reason, democracy, freedom, or
equality. Confronted with the uncertainties that such a list brings to
mind, we must recognize at the same time the full ambiguity of the deeds
and values that characterize the contemporary world and the impossibil-
ity of not summoning them up as implicit foundations of a particular
history. The narrator must, after all, place his own world at the end of the
period he is describing.
In short, narrative history reconstructs an experience along a temporal
axis. This reconstruction requires some conceptualization, but the latter
is never made explicit. It is concealed within the temporal finality that
structures and gives meaning to all narrative.
Yet the recent evolution of historiography seems to me to be charac-
terized by the possibly definitive decline of narrative history. While it still
flourishes in productions destined for mass consumption, it is being
increasingly abandoned by professionals in the field. In my view, there
has been a sometimes unconscious shift from narrative history to prob-
lem-oriented history, at the cost of the following changes:
1. The historian has surrendered before the immense indeterminacy
of the object of his knowledge: time. He no longer claims to describe past
events, not even important events, whether in the history of mankind or
in that of a part of mankind. He is aware that he is choosing what to
examine of the past and that in the process he is raising certain problems
relative to a particular period. In other words, he constructs his own
object of study by defining not only the period—the complex of events—
but also the problems that are raised by that period and by those events
From Narrative History to Problem-oriented History 57

and need to be solved. He therefore cannot avoid a minimal amount of


explicit conceptual elaboration: a good question or a well-formulated
problem is becoming more important—and is still less common!—than
the skill or patience needed to bring to light an unknown but marginal
fact.
2. As he breaks away from narrative, the historian also breaks away
from his traditional source material: the unique event. If, instead of
describing a unique, fleeting, incomparable experience, he seeks to ex-
plain a problem, he needs historical facts less vague than those to be
found in human memory. He must conceptualize the objects of his
inquiry, integrating them into a network of meanings and thus making
them nearly identical, or at least comparable within a given period of
time. Quantitative history provides the easiest—though not the only—
means for this kind of intellectual task.
3. In defining his object of study, the historian must also “‘invent’’ his
sources, for, in their original form, historical sources are usually unsuited
to his inquiry. Naturally, he may come across a set of records that not only
will be usable in themselves but will lead him to new or more valuable
ideas and theories. History does provide such blessings, but the opposite
is more often the case. Yet the historian who is trying to formulate and
solve a problem must find pertinent sources and organize them into
comparable and interchangeable units in order to be able to describe and
interpret the phenomenon he is studying on the basis of a certain number
of conceptual hypotheses.
4. The fourth change in the historian’s profession derives from the
above. The conclusions of a study are becoming ever more closely bound
up with the verification procedures upon which they are based and with
the intellectual constraints imposed by those procedures. Narration’s
particular kind of logic—post hoc, ergo propter hoc—is no better suited to
the new type of history than the equally traditional method of generaliz-
ing from the singular. Here the phantom of mathematics takes form.
Quantitative analysis and statistical procedures, provided they are suited
to the problem and sensibly applied, are among the most rigorous
methods for “‘testing”’ data.
Before proceeding further, one ought to look at the possible reasons
for these changes in historiography. They are probably related to factors
external to the discipline itself, such as the general crisis affecting the idea
of progress—a crisis that is challenging not only the concept of an evolu-
tion dominated by the nineteenth- and twentieth-century European
model but the very notion of an all-embracing and linear history. How-
ever, the changes in historiography are also related to internal, intellec-
tual factors such as the widespread influence of Marxist theory on the
social sciences; the brilliant development of the social sciences dealing
with limited and defined objects (economics, demography, anthropol-
ogy); or the impact of computer technology, which makes it possible to
58 History Today

carry out hitherto unimaginable calculations, provided the problems to


be solved and the hypotheses to be tested have been rigorously formu-
lated beforehand. Instead of discussing this vast problem at length, I
should like to confine myself here to examining a few of the consequences
of these changes on our profession and our historical knowledge.
The archives that serve as the basis for historiography are no longer
collections of documents but data constructed in series. If historians are
to work on conceptually clear objects of inquiry while remaining faithful
to the specific character of their discipline—the study of the evolution of
phenomena over time—they will need pertinent data (seldom available
ready-made) that can be compared over a relatively long period. Histor-
ical facts no longer consist of the explosion of important events that
shatter the silence of time but of chosen and constructed phenomena
whose regularity makes them easier to identify and examine by means of
a chronological sequence of identical data comparable within given time
intervals. Such data no longer exist independently but as parts of a system
that also includes earlier and later data. An examination of their internal
consistency (by establishing their comparability within the system to
which they belong) is a better test of their validity than an external
assessment of their probability (by comparing them to other sources for
the same period).
The intellectual process for defining the data is thus twofold. First, one
must determine their significance in order to apply them correctly. For
example, the major pre-nineteenth-century sources available to the
historian interested in literacy are signature counts. But what significance
does the ability to sign one’s name have in relation to the usual criteria for
measuring literacy: the ability to read and write? To take another exam-
ple, the historian who studies crises—particularly the different kinds of
economic crises of modern times—makes considerable use of price
series. However, he must first answer the question, What does a price
mean? For what movements or levels of economic life does it serve as an
indicator? Once the significance of the data has been established, the
historian must arrange them in serial form, make them comparable to
one another, decide what time unit they concern, what statistical methods
are appropriate to use, and so on. All these procedures are more than
mere techniques; they require methodological choices at each step of the
process.
One can raise a preliminary objection to this view of historical re-
search, namely, that the historian’s sources often contain gaps, are frag-
mentary, or simply do not exist, depending on the hazards of survival. In
any case, the difference between history and other social sciences is not
one of principle but of situation. There are undoubtedly problems,
particularly relating to the more remote periods of the past, for which
source material has disappeared. However, it must be stressed that such
material was not developed once and for all in the public archives in the
From Narrative History to Problem-oriented History 59

nineteenth century. On the contrary, the range of potential sources is


almost infinite, and quite often their existence is revealed by the nature of
the historian’s curiosity or by the problem he sets himself. The classic
example in this field is parish registers, which lay dormant for centuries in
French towns until the advent of historical demography in the 1950s led to
the discovery of their immense value. Moreover, a historian who is
unable to find immediately pertinent data to answer the question he has
set himself can, in most cases, get around the obstacle by treating what-
ever data are available in such a way as to be able to use them indirectly.
From this point of view, there is always the possibility of a ‘‘substitu-
tional” use of historical data. In a recent article,' I distinguished three
kinds of serial data. The first kind is the simplest and the easiest to
manipulate. It consists of the available quantitative data organized in a
way that provides a direct answer to the question at hand. Births, mar-
riages, and deaths listed in parish registers can be used in this way by the
demographic historian. The classic demographic rates can be calculated
from such data by means of minimal and standardized treatment (the
technique used to reconstruct families). The historian of political atti-
tudes can use the same technique with election results. The second kind
of source also includes quantitative data. These, however, are used in a
substitutional way to answer questions quite different from those for
which the data had been originally assembled. An example is the histo-
rian’s use of the calculation of time intervals between births to study the
spread of contraception and the patterns of sexual behavior in the past.
The specialist of economic growth also uses this method when examining
price series. In these cases the main problem involved in handling the
data is their relevance and the possibility of reorganizing them in relation
to the problem examined. Finally, there is a third type of source, which
requires even more careful handling: nonnumerical data that the histo-
rian nevertheless wants to treat quantitatively. In order to do so, he must
not only establish the relevance and value of the data, as in the preceding
case, but also rearrange them systematically into conceptually and chro-
nologically comparable units. Two examples are the use of notarized
marriage contracts to study endogamy, social mobility, fortune, or liter-
acy; and the use of wills to analyze attitudes toward death.
Thus, if one wanted to classify the most recent advances in contempo-
rary historiography according to their mathematical rigor, one would
have to take into account both the type of conceptualization applied to
the problems studied and the quality of the sources used to study those
problems. For example, it is clear that historical demography and eco-
nomic history are, on those two counts and at least for the so-called
modern period, the best-equipped fields. First, they can draw on concepts
developed in specific disciplines, such as demography and political eco-
nomics—concepts that can be readily imported into history at the cost of
only minor adaptations; second, the objects of those studies are easier to
60 History Today

abstract, define, and measure than most products of human activity.


Furthermore, most European states have been creating and preserving
data in these fields for many centuries.
Nevertheless, even within these ‘‘advanced”’ sectors of history, the
situation is not as simple as one might deduce from the ranking above,
which is based on the academic classification of our disciplines. History,
by virtue of its open-endedness, always tends to overflow the boundaries
of the sectional advances in these specialized fields. The question that
arises is whether, and to what extent, by borrowing some of those
advances and integrating them into its own practices, history has estab-
lished a knowledge of the past that could qualify as scientific.

The best way to tackle this very old problem is to study some examples, in
increasing order of complexity and uncertainty. I shall borrow them from
the field of historical demography, one of the most studied areas of
French historiography in the past twenty years. It is also one of the fields
that provide the greatest opportunities for formulating problems mathe-
matically. This special position is-due to the particular nature of the
discipline and to the sacrifices it has made for the sake of defining its
object clearly. Demography is entirely predicated on the principle of
abstract equality, according to which Napoleon’s birth has exactly the
same importance as that of any one of his future soldiers. Having adopted
a hypothesis that sacrifices all the particular aspects in the life of indi-
viduals—in other words, the essence of their history—it transforms his-
torical individuals into interchangeable and measurable units, by means
of unvarying and comparable events: birth, marriage, and death.
Stripped of the layers of meaning that each civilization has in its own way
given them, these events are reduced to their most fundamental charac-
teristic: the stark fact that they took place.
I am deliberately describing them all as “events,” since I do not see, a
priori, what distinguishes one particular historical fact from another—for
example, a birth, however anonymous, from a battle, however famous.
In this respect, the current distinction between structural and narrative
history is irrelevant to historical data themselves: there is no such thing as
a difference between facts that are events and facts that are not events.
History is a permanent event. However, some classes of events can be
more easily conceptualized than others, that is, integrated into an intel-
ligible system—as in the case of demographic events.
The raw and particularly simple data on births, marriages, and deaths
have become the object of a specific discipline: demography. They can
thus form the basis for a certain number of calculations and analyses,
which themselves are prefabricated objects of historical research. In
other words, they are objects or concepts elaborated by a discipline other
than history—in this case, demography, for which history, however, also
supplied primary material in the form of birth, marriage, and death
From Narrative History to Problem-oriented History 61

records. To the limited degree that it works with reliable or verified


data—though that “limited” degree is actually a considerable one, since
the verification of numerical sources is no easy task—historical demogra-
phy produces results comparable to those of demography itself: the set of
relationships that allow one to measure the elements of a given popula-
tion and the way in which they are evolving.
These elements, measured year by year, provide results that are
unambiguous and certain—unlike their interpretation. Let us examine a
century-long decline in the general mortality rate, for example in eigh-
teenth-century France. In order to determine when the definitive drop in
the mortality rate occurred, the rate must be calculated by age cohort,
particularly to obtain the infant or juvenile mortality rate. Let us assume
that there was a spectacular increase in the survival of newborns (from
birth to one year). Such a phenomenon could be explained by any
number of hypotheses, from an increase in the number of midwives
throughout the countryside to a transformation of the nursing system, not
to mention a sudden victory of medicine over a children’s disease. How
can one choose without testing each of these theories as well as several
others?
Admittedly, one can proceed otherwise, starting not from a single
variable but from the set of variables of a demographic system. Such an
approach is more properly demographic than historical. It uses or con-
structs a reproduction model for a population that is considered stable,
with the time factor provisionally set aside. Let us suppose that all the
“blanks” in the model have been filled; one is left with the question that
the historian must examine: How did the system evolve? Admittedly, by
studying what happened or even simulating what might have happened if
a given variable of the system had been missing or had been quite
different, one can diagnose at what point the system changed—for exam-
ple, how it expanded or regressed. However, the analysis of these
strategic variables refers back, as in the previous case, to elements that
are exogenous to the system and at the same time influence it: in other
words, to hypotheses that lie outside the demographic field. These
hypotheses involve not only concepts that have not yet been organized
into a scientific discipline, but also indicators most of which remain to be
invented.
Let us examine the problem of age at marriage, the main variable of
demographic control in preindustrial Europe between the twelfth and
nineteenth centuries. Although I cannot go into the question in detail
here, the postponement of marriages seems to have been the basic
endogenous factor that contributed to stabilizing the size of populations.
External agents (such as famine, epidemics, and wars) also took their
toll. However, their impact diminished during the period. The regulatory
mechanism worked in two ways. In the long term, the gradual rise in the
age at marriage, up to its classic ‘‘plateau’’ of twenty-five or twenty-six
62 History Today

(for women), eliminated ten years of potential fertility and, indepen-


dently of any recourse to contraception, reduced the number of children
per “complete” family. In the short term, the considerable variation in
mortality rates conditioned by particular historical situations was made
up for by variations in the age at marriage. When a population experi-
enced a demographic crisis (whatever the cause), marriages were put off
and the age at marriage increased; but, once the crisis was over, mar-
riages in younger age cohorts were added to delayed marriages. Thus, the
temporary lowering of the age at marriage brought the population back
to its precrisis size. It is therefore easy to devise and apply a demographic
model enabling one, on the basis of variations in the age at marriage (all
other factors being equal), to study population changes: in what condi-
tions a population grows, and in what conditions it diminishes.
With this type of simulation one can isolate the role of a variable within
a system, and even in the evolution of that system, but one cannot
identify the causes at work. In other words, simulation allows one to
describe, not to interpret, and still less to explain. As soon as one goes
beyond it to ask which factors are capable of influencing a cultural
behavior pattern such as age at marriage, one is confronted with any
number of possible interpretations. In the long run, the rise in the age at
marriage in seventeenth- and eighteenth-century Europe, up to twenty-
five or twenty-six years, can be interpreted as an optimal adjustment of
population density to available resources: witness Chaunu and Le Roy
Ladurie rediscovering Malthus! In the rich, ‘““developed”’ Europe of the
period—a belt of high agricultural productivity extending from the Lon-
don basin to northern Italy, passing through the Low Countries, open-
field France, and the Rhine valley—the population density apparently
stabilized at about forty inhabitants per square kilometer. But this state-
ment, even if approximately true—which is not certain, since data on
productivity and agrarian output for that period are difficult to handle—
tells us nothing about the modes in which the adjustment of the age at
marriage was experienced. To the extent that it was not accompanied by
an increase in births of illegitimate children, did it mean a greater accept-
ance, during a longer adolescence, of the rules of sexual austerity? Or
should one speak primarily of a socioeconomic adaptation—children
waiting to get married and settle down until the preceding generation had
turned over the family landholding to them?
It might be argued that one should begin by the easiest phenomena to
interpret and that short-term variations in the age at marriage involve
fewer uncertainties. Why, in a period of crisis, does a population post-
pone marriages? The answer is relatively clear: because of doubts about
the future, which stem from the sight of the present. Historical awareness
is determined by events in the short term; optimism or pessimism about
the future is conditioned by the immediate situation. When a historian
encounters reactions of this kind, which are conscious responses to
From Narrative History to Problem-oriented History 63

specific events, it is fairly easy for him to reconstruct their progression by


means of the traces they left behind; he is, after all, merely exposing the
motives expressed by the historical agents themselves. Unfortunately,
such redundancy does not lead very far. A crisis will delay marriages,
prosperity will increase marriages, and the next crisis will cause a new
drop—that much is clear. But one is left with the basic problem of
understanding how, over a period of successive upward or downward
adjustments, the age at marriage increased to such an extent that it
slowed “‘natural’’ demographic growth in preindustrial Europe.
At this point a descriptive discovery such as the one above necessarily
leads the historian to venture explanatory hypotheses that are fragile in
two ways: first, because they were by nature out of reach of the people
whose behavior he is studying, so that there are no directly usable written
traces; second, because he will have to abandon a purely demographic
analysis, and with it the factual and conceptual precision it requires. He
will have to understand the mechanism by which the probability of
collective behavior indicated in the analysis of data concerning age at
marriage is embodied in the multiplicity of individual behavior.
Let us return to the two hypotheses mentioned earlier. As they are of a
different nature, they are not incompatible. Their common feature is that
the behavior patterns they describe would have made it easier for the
people who lived in that period to harmonize their expectations with their
actual opportunities. This process is one of the conditions of social life; it
is the somewhat melancholy mechanism by which men predict and con-
struct the most probable future for themselves. But the first hypothesis is
of a psychological nature, the second of an economic nature. The first
indicates a morality, the second a strategy. The first cannot be measured;
the second can be, for the historian can establish a relationship between
the demand represented by younger generations and the supply of farm
estates or job vacancies generated by deaths in the ranks of older genera-
tions. If he does not have enough data at hand to work on a macroeco-
nomic scale, he can at least tackle the problem through a series of
monographs on family estates, which will enable him to see how genera-
tions succeeded one another on a single estate. This is an objective
process that can lead, at least theoretically, to a clear conclusion. In
contrast, the notion of the spread of a puritanical superego (on a sexual
level) throughout seventeenth- and eighteenth-century Europe is a
hypothesis that can lead only to ambiguous answers. It is easy enough to
identify the factors that make it plausible: the Protestant ethic, the
Counter-Reformation, or Norbert Elias’s “‘civilization.”? However, it
cannot actually be verified or invalidated.
Why is that so? To begin with, the superego is an indemonstrable
psychological concept. It is used to interpret behavior that could be
interpreted in any number of other ways. For example, Weber’s notion of
individual self-control can be replaced by that of a reinforcement of
64 History Today

external constraints imposed, in this case, by church and clergy. But there
are no appropriate data—nor will there ever be—that can provide
answers to hypotheses concerning the psychology of historical agents.
The agents are dead, and very few of them, even among the fraction who
wrote about themselves, bothered with the part of their being that,
before Freud, they had neither the means nor even the curiosity to
explore. Thus the historian of what is now referred to by the very vague
term of mentalités must either base his investigation on scattered or
ambiguous texts or find an indicator not of his subjects’ psychology but of
their behavior in order to deduce the psychological roots of this behavior.
In the first approach, the historian encounters the difficulty of assess-
ing the significance of a both subjective and exceptional piece of evi-
dence. Admittedly, all historical data (except the vestiges of men’s mate-
rial existence) are to a certain extent subjective. Even the registration of a
birth or the accounts of an estate were, at a certain moment in time, put
down on paper by an individual. But the constraints that govern the
recording of an event differ considerably according to the phenomenon
observed, the nature of the observation, and that of the observer; accord-
ing to whether the event is normal and repetitive—that is, comparable to
an earlier one—or extraordinary and therefore recorded precisely be-
cause it lies outside the norm of habit; according to whether one is dealing
with a systematic observation governed by certain rules or with a chance
testimony, a census, or an impression; finally, according to whether the
relationship linking the observer to the object observed is or is not in the
nature of knowledge.
With regard to my example, the historical evidence that can tell us
about the psychological roots of behavior dating back several centuries is
naturally of a literary nature. I use literary in the broad sense to include
certain texts that posterity has not elevated to that rank: the handful of
unpublished private diaries and of old manuscripts that can cast some
light on the subject. But such evidence is scarce, impossible to use in
systematic temporal series, and limited to a very narrow social environ-
ment. In order to bypass its random character, one will have to consult a
different kind of documentation, this time of a normative type: manuals
of good manners or specialized treatises on religious morality, such as
penitentials. However, texts of this kind present the same ambiguity as
government legislation. They prescribe an optimal conduct, but one can
never determine to what extent it was accepted, obeyed, or internalized.
Is the reiteration of rules of conduct over a long period a sign of their
having penetrated into society or, rather, of the resistance they have
provoked? The second hypothesis is as probable as, if not more probable
than, the first. In this case, the normative text is more interesting for its
“preamble” and whatever observations of actual behavior it may contain
than for what it forbids or orders. But it serves primarily as evidence
From Narrative History to Problem-oriented History 65

about the institutional environment in which it was produced, the state or


the church.
That is why the historian of mentalités, who is trying to investigate the
more common forms of behavior, cannot be content with the traditional
literature of historical testimony, which is inevitably subjective, untypi-
cal, and ambiguous. He must examine behavior itself, that is, the objec-
tive indications of behavior. The hypothesis discussed above concerning a
““Weberian”’ superego extending its control over the souls of seventeenth-
and eighteenth-century Europe can be tested for many of the signs of a
given behavior; for instance, the number of illegitimate births, the num-
ber of premarital conceptions, or the use of contraception. A decrease in,
or a low incidence of, illegitimate births or premarital conceptions in a
society characterized by a delayed age at marriage is indeed a sign of a
long period of accepted chastity. However, for these indications to be
meaningful, one would have to prove that the use of contraceptives in
Europe during that period was not widespread. How can that be proved?
Not by means of literary evidence—which is very rare in this realm par
excellence of the unexpressed—but by measuring the spacing of chil-
dren’s births during the married life of couples. The statistical technique
for measuring the behavior of this variable is well known. For a popula-
tion of married women old enough to give birth, a ratio is established
between the number of births and mothers’ ages. If the fertility of couples
decreases very rapidly after the first children and in relation to the
mothers’ ages, then the use of contraception can be deduced. Otherwise,
one finds a succession of births, slowed down only by the nursing period
for the latest-born child and the biological decrease in fertility as the
potential mother grows older.
The conditions for this kind of experiment seem clear and simple. For
example, graphs unambiguously prove that eighteenth-century Cana-
dians were ignorant of contraception, while French dukes and peers of
the same period practiced it. However, between these two extremes, the
conclusions remain ambiguous. Precisely because the spacing of births
during the life of a couple is also influenced by factors other than con-
traception, it is impossible to assess the specific role of contraception. An
increased interval between births, if it is not abrupt, could be due to a
change in nursing methods and a later weaning of the latest-born child.
That is why categorical conclusions are difficult to reach, as witnessed by
the discussions concerning this problem that have been going on since the
mid-1960s.
To summarize the methodological aspects of the discussion, it seems to
me that we are faced with insuperable difficulties on three levels. First,
the concept of the superego as a sort of austere collective moral con-
science governing individual behavior cannot be actually proven.
Second, the subjective historical data and firsthand accounts are scarce,
unrepresentative, and ambiguous. Third, the objective indicators are
66 History Today

equally ambiguous. The hypothesis put forward is more in the realm of


plausibility than in the realm of truth.
It would therefore be incorrect to believe that the passage from narra-
tive history to problem-oriented history (or, if one prefers, to conceptual-
ized history) suffices to enter ipso facto into the scientific domain of the
demonstrable. From the epistemological point of view, conceptualized
history is probably superior to narrative history because it replaces an
understanding of the past based on the future by explicitly formulated
explanatory factors. It also unearths and constructs historical facts in-
tended to support the proposed explanation; thus, while carving out
specific entities within the realm of history, conceptual historiography
expands it considerably. Perhaps Max Weber chose the wrong path with
his Protestant Ethic, but what a landmark it was! A conceptual discovery
can be judged by the areas of research it opens up and by the traces it
leaves behind.
Nevertheless, we still have not arrived at a scientific history. First,
there are some questions and concepts that do not lead to clear, unambig-
uous answers. Second, there are some questions that in principle lead to
clear-cut answers yet cannot be solved, either because of a lack of data or
because of the nature of the data: the indicators may be ambiguous or
impossible to subject to rigorous analytical techniques.
As we have seen—and one could give many more examples—these
techniques are suitable to handling data that are clear (or have been made
clear), available in chronological series, and capable of answering un-
ambiguous questions generally formulated by the most advanced con-
temporary social sciences, such as demography and economics. To that
extent, even history can lead to definite conclusions. For example, one
can calculate the major variables in demographic behavior in western
Europe since the seventeenth century. One can measure the increase in
prices in eighteenth-century France or the takeoff of agrarian productiv-
ity in the nineteenth century. In other words, this kind of history, with its
potential for extrapolating into the past a number of very specific ques-
tions usually formulated in other disciplines, is both highly profitable and
very limited. It enables one to arrive at clear-cut conclusions and to
obtain a good description ofthe particular phenomenon chosen for study.
However, the interpretation of these findings does not offer the same
degree of certainty as the findings themselves. In problem-oriented his-
tory, interpretation is basically the analysis of the objective and subjec-
tive mechanisms by which a probable pattern of collective behavior—the
very one revealed by data analysis—is embodied in individual behavior in
a given period; interpretation also studies the transformation of these
mechanisms. Thus, it goes beyond the level of described data in order to
relate it to other levels of historical reality. It generally requires addi-
tional data that belong to another field and are neither necessarily avail-
From Narrative History to Problem-oriented History 67

able nor necessarily clear. Interpretation more often than not involves
hypotheses that are not proven or not provable.
The problem posed by recent historiographical developments, partic-
ularly by the use of strict demonstration procedures, is not to determine
whether history itself can become a science. Considering the indeter-
minacy of history’s object of study, the answer to that question is un-
doubtedly negative. The problem is to determine the limits within which
those procedures can be of use in a basically unscientific discipline.
Although these limits are obvious, one should not deduce that history
should revert to its former function as fortune-teller. Instead, the un-
reasonable ambitions of “total history” should be lowered and, in our
exploration of the past, we should make the utmost use of the sectorial
discoveries and methods of certain disciplines and of the conceptual
hypotheses emerging from the great contemporary potpourri called the
human sciences. The cost of this change is the breakup of history into
many histories and the renunciation by historians of their role as social
authorities. However, the epistemological gains may be worth such sac-
rifices. History will probably always oscillate between the art of narrative,
conceptual understanding, and the rigor of proofs; but, if its proofs are
more solid and its concepts more explicit, knowledge will stand to gain
and narrative will have nothing to lose.
History and Ethnology

History and ethnology developed during the eighteenth century as both


related and contradictory disciplines, whose relationship to each other
was determined by two primal categories: time and space. Both disci-
plines were the instruments for describing the human universe; but
history drew up an inventory of time, and ethnology an inventory of
space. Thus, in the old bibliographical classifications of Baconian
Europe, accounts of travels to faraway places come under the heading of
“history” books; they represent a subcategory devoted to the description
of foreign, and especially of exotic, countries. By providing the reader
with an account of the mores of remote peoples, travelers were not only
trying to popularize the picturesque aspect of otherness; they were also
bringing back an image of the past situated in contemporary space. The
savage state represented the infancy of civilized man. Thus, two readings ~
of the same vision of man were brought together.
No doubt brought together is somewhat of an exaggeration, for the
advent of the nation-state already stood between ethnology and history
and played a decisive role in the emergence of history as a distinct form of
knowledge. This process began in the seventeenth century with the
secularization of time. The old apocalyptic chronology of the four monar-
chies was gradually abandoned. According to Daniel’s prophecy, the four
monarchies were supposed to succeed one another in an order of increas-
ing degeneracy: the Roman Empire constituted an interminable ending

Originally delivered as a paper at an international seminar, “The Historian: Between


Ethnology and Futurology,” sponsored by the International Association for Cultural
Freedom, the Giovanni Agnelli Foundation, and the Giorgio Cini Foundation (Venice, 2-8
April 1971). First published, in a revised form, as “L'Histoire et ‘l’homme sauvage,’ ” in
Méthodologie de l'Histoire et des sciences humaines; Mélanges en l'honneur de Fernand
Braudel (Toulouse: Privat, 1973), 2:227-33.

68
History and Ethnology 69

to human history—a final phase that would perpetuate itself through the
Holy Empire of medieval Christendom. The Old Testament legend suc-
cumbed first to humanist pressure, then to Protestant pressure, and
finally to the rise of the nation-state. Humanist philosophers were too
fascinated by antiquity to develop a historicist vision of the past. But
humanism provided at least the intellectual groundwork for such a vision
be inventing scholarship and distinguishing between ‘‘fables”’ and “‘true”’
history. In any event, the religious schism clouded the meaning of Scrip-
ture; it shifted the locus of the narrative of origins at the same time as it
transformed eschatology. But the Protestant rejection of the Apocalypse
and the church was a sign that the activity of deciphering the past could
now focus on other objects—on the state, the notion of irreversibility,
and the obsession with origins. History became secularized even as it
turned to the study of nations. This phenomenon is clear in the French
case, with Jean Bodin, La Popeliniére, and the Chancelier Pasquier.
Gallican history emerged as a reaction by French jurists against “‘Italian”’
humanism, which they suspected of embodying a double form of Roman
imperialism: the imperialism of antiquity and that of the papacy. The
French jurists were concerned with justifying royal power, particularly in
the struggle against the extremism of the Ligues.'
Thus, between the sixteenth century and the Enlightenment, history—
at least profane history, which was carefully distinguished from sacred
history—was above all the history of nations, that is, of the states and
peoples of Europe. Even Voltaire, who tried to adopt a broader perspec-
tive, used Louis XIV’s state—the apogee of civilization—as the implicit
point of reference for his universal history. The history of human prog-
ress is signposted by the nation-state. Fledgling anthropology was re-
served for undeveloped peoples.
Accordingly, by the eighteenth century, when the description of
“primitive” peoples was becoming a fairly systematic discipline, a single
hierarchy of values ranked the study of time above travel literature. The
national societies explored by the historian became the ideal model for
the future of the groups described by the traveler.
The nineteenth century probably deepened the rift between the two
disciplines by abolishing their complementary character. The imaginary
temporal and spatial link between two universal phases—infant man and
adult man, primitive society and civilized society—was destroyed by the
notion that these were elusive and distinctive phenomena. The nation-
state ceased to embody the collective progress of mankind and became
the locus par excellence of antagonism and differentiation. Moreover,
history was no longer dependent on a frame of reference that charted the
stages of human development. It was now torn by an endlessly shifting
balance of power and by conflicting justifications. The ideology of pro-
gress, which somewhat hastily extrapolated the extraordinary rhythms of
economic change into the whole scope of human activity, transformed
70 History Today

time into a sort of unending current, urged on by nationalistic rivalries.


Progress was both an instrument and a stake of power politics, a breeding
ground for confrontation between national histories. History dealt
henceforth only with a handful of nations: those that produced, those that
changed—in a word, those that mattered. The remainder of the inhabited
world was thus abandoned as a historical nonentity, and travel lost its
bibliographical and scientific status. Ethnology became a form of residual
knowledge, defined negatively in relation to the history of western
Europe and North America. Not only the attitudes of the great liberal
economists and historians but also the extraordinary Europocentrism of
Marx point to the fact that ethnology had turned into a subordinate and
secondary discipline, an ill-defined by-product of European expansion, a
combination of blindness and guilty conscience. Since the historian had
turned an exception into a model, ethnology was left to reign over a
mirror image of history, over a marginal yet vast realm: the henceforth
distinct realm of the the unwritten as against the written, of immobility as
against change, of the primitive as against progress.
Although this dichotomy is still-with us, it is becoming increasingly
difficult to live with, because it has been severly undermined by two series
of contemporary events. The first series is external to the two disciplines,
while the second relates to changes in the social sciences.
1. Many major events in contemporary history have called in question
the idea of progress. Neither the Hitlerian apocalypse nor the trans-
formation of the Russian revolution into a bureaucratic and ideological
terror is easy to fit into a march of mankind toward the advent of
rationality or freedom. The modern world, while increasing at an unprec-
edented rate man’s mastery over nature, presents an ever-greater num-
ber of insoluble problems, historical impasses, and manifestations of
social violence. Technical and economic progress is making its way amid
political irrationality and planetary disorder, laying down a new chal-
lenge to the notion of total history, a notion according to which all levels
of history evolve at the same rhythm along a single time scale. In this
respect, two phenomena have probably played a fundamental role in
breaking down the traditional barrier between ethnology and history:
human space has become homogeneous even as time has ceased to be so.
The inhabited world has not only been explored, inventoried, and
bounded; during the last few decades, particularly in the course of
decolonization, it has been the locus of the irresistible spread of the
European political model, that is, of the appropriation of nationalism.
Thus, the entire inhabited world has been raised, like Europe, to the rank
of national history. The countries belonging to what, on the basis of
economic criteria, we call the Third World are all feverishly engaged in
glorifying their origins, in which they are seeking a definition of their very
identity, for these countries have ceased to conceive their differences in
terms of space and have taken to enhancing them in terms of history.
History and Ethnology 71

Even the West, when it tries to counteract the intellectual spread of its
most radical ideas, has no clear-cut frontiers left other than those pro-
vided by economic history, such as Rostow’s model of takeoff. Perhaps
this is an underhand manner of continuing to put oneself forward as an
example by constructing a linear schema in which “‘postindustrial’’ so-
ciety constitutes a horizon for mankind. At least it is a way of bringing
mankind back into the realm of history.
Space has thus become historicized, but at the cost of a fragmentation
of time. As history gradually absorbs the whole of mankind and becomes
less Europocentric, it faces the ethnological challenge represented by the
multiplicity of societies and cultures. This multiplicity undermines the
notion of homogeneous time. Not only do societies evolve at different
rhythms, but even within a given society the different levels of reality that
compose it are not governed by an all-embracing and homogeneous
temporality. “Change” has become a concept measurable in economic
terms, in its various aspects; but it also reveals resistance to change.
“Takeoff,” “modernization,” and the universalization of material prog-
ress and economic growth are thought of as the keys to the significance of
contemporary history; but they clash with heritage, tradition, and the
entire range of sociocultural patterns of inertia. Thus, by extending its
field to the human world, history has discovered that it is also nonhistory:
change reveals immobility. The great nineteenth-century historical tradi-
tion, inspired by both Manchester and Marx and predicated upon univer-
sal progress through economic development, is consequently being
undermined by the crises of the contemporary world and by the applica-
tion of its hypotheses to non-European areas. It is therefore no surprise
that, while desperately trying to preserve its hegemony as a standard-
bearer of “modernization,” history is returning to ethnology as if it were
aware of its own failures.
2. A second and probably less visible series of changes in the relations
between history and ethnology is inherent in history itself and in its
development as a discipline over the last thirty or forty years. I am
referring to what is called in France, using a vague and moreover negative
term, the replacement of a “history of events” (histoire événementielle)
by a “‘non-event-oriented history” (histoire non événementielle).
Let us try to define these terms by abstracting them from the context of
a pointless dispute. The history of “events” seems to me to be both a way
of describing the past and a way of selecting events. Such history, which is
informed by the desire to reconstruct “what happened” and to recount it
in the narrative mode, chooses its material in accordance with these aims;
in other words, the material—the celebrated events—is selected and
arranged along the time axis in such a way as to give substance to the
narrative as it unfolds. Change thereby becomes inherent in events, and
the chain of events is supposed to give a meaning to the succession of
changes. That is why history of this kind is marked, in contradictory
72 History Today

fashion, both by the short term and by teleology. Since an event, a sudden
irruption of the unique and the new into the concatenation of time,
cannot be compared with any antecedent, the only way to integrate it into
history is to endow it with a teleological significance. And as history has
developed as a mode of internalizing and conceptualizing the feeling of
progress, “events” in most cases constitute stages in an advent, for
instance, of the Republic, of freedom, of democracy, or of reason.
However patiently one tries, in accordance with the strictest rules of
scholarship, to reconstruct an “‘event’”—the quintessential form of histor-
ical ‘“‘fact”—that event derives its significance only from a total history
external to the event and independently defined. The temporal structure
of this type of history consists of a series of discontinuities described in the
mode of continuity: such is the classic substance of narrative.
History of the non événementielle kind rejects narrative—at least this
literary type of narrative—as its primary concern is to define problems.
Borrowing from the contemporary social sciences such as demography,
geography, and sociology, it has renewed and redefined the scope of
historical inquiry. First, it isolates the various levels of historical reality in
order to focus on a small number of them, or ona single one, and describe
them (or it) as systematically as possible, that is, separately. That is why it
constructs historical ‘‘facts” that differ on two counts from those of
“event-oriented”’ history: in most cases, they lie outside the classic field
of major political changes, and they are defined no longer by their
uniqueness but by the possibility of their being compared with earlier and
later phenomena. A “fact” is no longer an event chosen because it
coincides with a high point in a history whose ‘meaning’ has been
determined beforehand, but a phenomenon chosen and constructed on
the basis of its recurrent character, which makes it comparable to other
phenomena of the same kind over a given period. Documents and “‘facts”’
no longer exist independently but in relation to the series that precedes
and follows them; it is their relative value that becomes objective and not
their relation to an elusive “true” substance. By the same token, new
topics and methods have been developed. Historical corpora are by
definition so diverse that the historian can assemble them according to his
preferences and skills and deal with them from an economic, demo-
graphic, sociological, ethnological, or linguistic point of view.
But there is no such thing as an innocent methodology. By his new
approach, the historian has changed the nature of the problems tradi-
tionally dealt with by his discipline. The unique and noncomparable
phenomenon cannot be handled by such a methodology, and the special-
ist of intellectual biography will not find it satisfactory, either. Nor will
the historian of antiquity, who lacks and probably always will lack the
serial data indispensable for a systematic quantitative treatment.
Moreover, history that relies on serial data concentrates on a particular
type of source, of problem, and above all of temporality. The sources
History and Ethnology 73

must be, if not numerical, at least reducible to homogeneous and compa-


rable units, a requirement that both restricts and broadens their scope,
for while the ‘thapax”’ is no longer usable, an entire portion of the vast
“reserve” of unwritten material that the historian has until now used so
sparingly can be arranged in series; when handled in this way, even
iconographic data such as photographic land surveys can represent more
valuable historical material than the sempiternal literature of sempiternal
witnesses. It is also natural that this type of serial source should serve as
the basis for investigations or hypotheses that are more economic or
ethnological in character than strictly political; such sources imply the
reduction of individuals to equal units as economic or sociocultural
agents. Incidentally, this explains how, by virtue of a misunderstanding,
an encounter took place between the most “‘scientific-minded”’ type of
history—the one with the fewest value judgments—and the most ‘‘demo-
cratic-minded” type of history, which aimed at rehabilitating the anony-
mous “‘little men,” dwarfed by the great heroes of politics and yet
indispensable to their existence.
Finally, the serial history described here is a history of long-term
phenomena. Fernand Braudel, in his books and in an article that is now a
classic, has demonstrated this better than anyone. Because it essentially
describes repetitive and regular phenomena, serial history is of value only
if it brings to light examples of historical evolution over a sufficiently long
period for these repetitive and regular phenomena to change and fluctu-
ate. Did the growth of the rural economy in the first two-thirds of the
sixteenth century—a growth that can be charted by means of various
indicators—betoken a decisive takeoff or merely a recovery from the
terrible crisis of the period from 1350 to 1450? Only an analysis in the
longer term, encompassing earlier and later periods, can provide an
answer. Thus, by choosing identical indicators over a long period of time,
historians can isolate phenomena according to their duration—short-
term crises, longer recessions, cycles, and trends—and integrate them
into a general interpretation. But this mode of selection puts greater
emphasis on the factors that help to preserve a given system than on those
responsible for qualitative changes. By adopting this approach, historians
have rediscovered the long periods of economic stability and the social
and cultural inertia that for a long time characterized societies studied by
ethnologists. Historians, too, as Lévi-Strauss would say, have their ‘“‘cold
societies.”
One could thus imagine a double-column inventory of what history
owes to ethnology. The first column would list the changes in historical
methodology, such as the systematic recourse to unwritten sources, the
increasing use—as far as written sources are concerned—of statistical or
prestatistical documents, or textual analysis of the “structural” kind
(whether the model is provided, for example, by linguistics, psychoanaly-
sis, or content analysis). The second column would list the new objects of
74 History Today

history, developed in the wake of the momentous shift in historiographi-


cal curiosity described above: the substitution of anonymous men for
great men, of nondevelopment for a model based on change alone, of the
rudimentary forms of cultural life for the testimony provided by “‘great”’
literature, and so on. This list would, in fact, cover a considerable portion
of contemporary historiography in France and elsewhere.
Actually, the dividing line between history and ethnology was never
determined by epistemological criteria. It was based far more on the
external conditions that shaped the development of these two fields and,
later, on the conventional academic separation established between
them. Today this boundary is disappearing like so many others and giving
way to a configuration no easier to define than the disciplines of
yesterday.
The Birth of History in France

A Dual Tradition

The absence of history from the educational system, and thus its non-
existence as a school discipline, in early modern Europe is due to the fact
that it simply did not exist as a discipline at all. It was torn between two
intellectual activities that ignored or even scorned each other: erudition
and philosophy. The first was in the hands of the antiquarii, also known in
old French texts as antiquaires, that is, specialists of things ancient and,
naturally, of antiquity—men who confined themselves to a narrow,
esoteric, and recondite area of knowledge and who were proficient in
dead languages. This tradition gave birth, not to history in the
nineteenth-century sense of the word, but to the notion of historical facts
as the building blocks of history.
This old tradition, whose roots go back to the Renaissance, was not
originally a critical tradition.' It did not deal with Scripture but with
Greco-Roman antiquity, in which sixteenth-century Europe passionately
sought a new identity. However, the antiquarians’ interest in antiquity
did not stem from a desire to rewrite the history of that period. The
ancients had already written it, and who could improve on Thucydides,
Livy, or Tacitus? The task of the “moderns” was simply to comment on
ancient historians, to work in their margins. That was the meaning of
belles-lettres. Alternatively, when they wanted to escape this mirror game
they would write Roman (or Greek) ‘‘antiquities,” not “histories.”’ But
these “‘antiquities’”’ were themselves doubly marginal with respect to the
royal way of history: they described nonliterary sources, and they ex-
humed coins, stones, inscriptions, and portions of monuments—the ran-

First published as ‘“‘La Naissance de l’histoire,” H-Histoire, no. 1 (March 1979), 11-41.

dig
78 Seventeenth- and Eighteenth-century Culture

dom remains of a fatal shipwreck. These vestiges provided subject matter


for commentaries and studies that did not really qualify as history, since
they dealt with customs, institutions, and art, whereas history consisted
of the chronological analysis of political systems and governments.
Thus, the antiquarian was not a historian. But the second half of the
seventeenth century witnessed the downfall of the notion of a universal
history within which every particular history had been written for all
eternity. At the same time, the antiquarian turned into a historical critic.
The area covered by his “‘art”’ (the ars antiquaria) spread beyond classical
antiquity to take in, for example, biblical antiquity. Above all, the
antiquarian freed himself from the tutelage imposed on him by the
historiography of antiquity, that is, by the model of the ancients. The
antiquarian had not yet become a historian. But since the past had not
been fixed for all eternity by Livy and Plutarch, he was now free to write
about history.
In other words, the materials he unearthed and classified were no
longer marginal. No less than literary sources, they now constituted the
ingredients of history. And literary sources themselves were now sub-
jected to scholarly criticism. The purpose of philology was no longer only
to reconstruct them but to discuss their value. Coins, inscriptions, and
fragments of arches and columns now served to corroborate the informa-
tion contained in literary sources. Internal and external documentary
criticism was born when the various types of sources were all integrated
into the search for truth.
Thus, the second half of the seventeenth century did not mark the
invention of history but rather a reworking of historical materials that led
to a shift in the apparently immutable boundaries of the field. Bossuet
was still able to write a Universal History, but he had difficulty making
room in sacred chronology for the profane history of ancient peoples, a
history whose limits had been pushed back by the discoveries of ‘‘anti-
quarians.”’ Even biblical history itself, an intangible block standing
motionless in the unending flow of time, was reexamined by modern
chronologists. The Oratorian Richard Simon published in 1678 an His-
toire critique du Vieux Testament that caused him to be expelled from the
order.
Yet it was the church itself that set an example for scholarly research,
even if it was not always able to control its development. Not only was the
church caught up in the spirit of the age, but, in its struggle against
Protestantism, it found itself obliged to undertake a systematic descrip-
tion of and to glorify the entire Christian tradition, particularly the first
six centuries that constitute Christian antiquity and provide the basic
interpretation of Scripture. Ecclesiastical scholarship focused primarily
on the church fathers. It flourished at Port-Royal and above all in the
work of Tillemont. Its central impulse came from the Benedictines of
The Birth of History in France 79

Saint-Maur, who, more than a century before the advent of German


historiography, laid down the canons of historical criticism.
The antiquarian’s art thus culminated, at the end of the century, in the
methodical undertaking of the monks of Saint-Germain-des-Prés, who
set out to distinguish between truth, likelihood, and falsehood. In con-
formity with Mabillon’s motto, “True piety loves nothing but what is
grounded in truth,” modern historical research was born of the applica-
tion of the procedures of critical reasoning to the exploration of Christian
antiquity. Consequently, pagan antiquity, which from this point of view
was inseparable from Christian antiquity—because both were part of the
same chronology—could henceforth be dealt with in the same manner.
But if the canonical division between sacred and profane history was
thus becoming increasingly blurred, history itself remained distinct from
historical research. It constituted a literary genre, one of whose rules was
precisely to exclude all reference to a critical apparatus and to “‘proofs.”
Antiquarians published chronologies, ‘‘annals,” ‘collections,’ and
“memoirs.” In contrast, history was a continuous narrative unencum-
bered by original documents. It was characterized not only by a moral
lesson but by a regular and ornate form. While losing its rigidity of
content, history preserved all its aesthetic and moral rules. It remained a
writer’s pursuit.
When Tillemont set out to publish what were to be his Mémoires pour
servir à l’histoire ecclésiastique, he hesitated over the title to give to his
work. In settling for Mémoires, he was guided by the fact that his
expository method was that of the antiquarians: “It seems to be the most
reliable and the most solid. It is like producing evidence in court: the
reader need only judge for himself. But this method leads to lengthy
exposition and frequent repetition. . . . Itis rather the material of history
than history itself.” The same author, however, accepted the word his-
tory for the profane part of his work when, three years earlier, in 1640, he
had published his Histoire des empereurs, under a title that is worth
quoting in full: Histoire des empereurs et des autres princes qui ont régné
durant les six premiers siècles de l’Eglise, des persécutions qu'ils ont faites
aux chrétiens, de leurs guerres contre les juifs, des écrivains profanes et des
personnes les plus illustres de leur temps, justifiée par les citations des
auteurs originaux, avec des notes pour éclaircir les principales difficultés de
l’histoire (“History of the emperors and other princes who reigned during
the first six centuries of the Church, of the persecutions they conducted
against the Christians, of their wars against the Jews, of the most note-
worthy profane writers and individuals of their time, illustrated by quota-
tions from original authors, with notes to explain the main difficulties in
this history.”’) Thus, Tillemont was one of the first to combine history and
erudition. But how apologetic he was in his foreword! Let us quote him
once more in order to appreciate the tyranny of ‘‘genres” in the age of
80 Seventeenth- and Eighteenth-century Culture

classicism: “For a long time the author wondered whether this work
should not be entitled Mémoires, which is certainly the title that would
suit it best, both because of the manner of its composition and because of
the purpose for which it was compiled. The title of Annales has also been
envisaged, because in fact the work follows as much as possible the
sequence of periods and is even divided almost always according to year;
moreover, a plain and unadorned style, such as will be found here, seems
more suited to annals than to history. Nevertheless, the title of History
was finally adopted as it is the one requiring the least justification,
because it is the simplest, and because all narrative is in a sense a History.
But the readers are asked to consider the title only in that sense of the
word, and not to expect to find an ordered history here. The author never
contemplated writing a history of that sort, and he would like to make it
clear that he has always regarded such an undertaking as exceedingly
difficult in itself and far beyond the scope of whatever talent and knowl-
edge he may possess.’”

The Impossibility of Teaching History in the Eighteenth Century

One might, however, infer from this excessive display of modesty that the
gap between historians and antiquarians was narrowing. On the contrary,
eighteenth-century French culture widened the gap by creating a vogue
for ‘‘philosophical history,” which drew away from antiquarian research,
an activity it disdained.
Actually, it was the very advances in this research that backfired on it
in the late seventeenth and early eighteenth centuries. Antiquarian re-
search had criticized some traditional historical beliefs, for instance by
debunking miracles, reducing the number of Christian martyrs, and
revamping biblical chronology. As a consequence, rationalist individual-
ism systematically challenged the very notion of historical fact. Bayle
devoted an entire dictionary, from A to Z, to the destruction of the
historical foundations of religious beliefs, but all he left to the rational
individual was uncertainties. Fontenelle bluntly observed that it was
impossible to write truthful history: ‘We have been so accustomed during
our childhood to Greek fables that when we have become capable of
reasoning, it does not occur to us to find them as astonishing as they really
are. But if we shake off our habitual ways, we cannot help being fright-
ened at the realization that the entire ancient history of a people is merely
a heap of fantasies, dreams, and absurdities. Is it possible that all this
could have been passed off as true? Why would it have been presented to
us as false? How could men have been fond of blatant and ridiculous
falsehoods, and why should they have stopped being fond of them?”
But, above all, this historical defeatism was rooted in an obsession
with modernity, that is, with the present. European elites had lived since
The Birth of History in France 81

the Renaissance with a cultural identity borrowed from antiquity, a


period whose artists and authors represented unsurpassable models and
whose literary genres constituted the authoritative canons of beauty and
truth. Now Europe was raising the question of its cultural autonomy: the
academic quarrel between ‘‘ancients” and “moderns” in France at the
end of Louis XIV’s reign ultimately centered on the notion that classical
culture was not a past but a present and that history was not a new
beginning but a progression. Consequently, history, too, began to struc-
ture itself around the perception of the present, thus making antiquarian
curiosity obsolete.
Moreover, the philosophe historians incorporated nonliterary sources
and “‘evidence”’ into their new history. Having shaken off the tyranny of
political history as bequeathed by antiquity, and having abandoned the
sequence of emperors, they turned to art, religion, and institutions. They
were writing the history of “‘civilization,”’ but in order to understand their
own time. Montesquieu sought in Roman history the secrets of political
stability or decadence. Voltaire compared the age of Pericles to that of
Louis XIV. The eighteenth century turned to the history of the world’s
peoples not only to witness the spectacle of religions and customs in all
their diversity but to find the key to a temporal flow emancipated from
Holy Writ and now indefinitely open to progress.
In addition to the progress of civilization, philosophical history had
another conceptual pole: the origins of the nation. Eighteenth-century
Frenchmen searched their national history for the source of their ‘‘con-
tract” with the king and of the nobility’s legitimacy. The Germanic
invasions were thought to have introduced an elective monarchy and an
aristocracy of warriors into Roman Gaul. The dispute about Clovis is thus
an expression, in its way, of French society’s dramatic search for a
self-representation. But Boulainvilliers’s history—like Voltaire’s—no
longer had anything to do with ‘antiquarian’ history. Eighteenth-
century France did not have a Gibbon. An insurmountable barrier still
separated French philosophes and érudits: the former were glorified, and
the latter relegated to the ghetto of the Académie des Inscriptions. The
tradition of critical research and the tradition of the great philosophical
and literary narrative were to be reconciled only by the historians of the
Restoration.
One need only glance at the bibliographies of the period to realize to
what extent history represented a heterogeneous and fast-changing
genre. Library classifications, for example, grouped under the heading of
“history” an extremely vast area of knowledge. History encompassed
everything that had to do with knowledge concerning human societies.
As an epistemological heading, history expanded, thanks to all the schol-
arly or even simply descriptive contributions of European culture since
the Renaissance. History presided over the whole, with its canonical
division between sacred and profane, the cultural preponderance of
82 Seventeenth- and Eighteenth-century Culture

antiquity, and the model of a moral narrative in the manner of Livy. But
history subsumed not only the techniques and achievements of the ars
antiquaria—chronology, paleography, archaeology, and so on—but also
the inventory of space, that is the activity that preceded geography and
was known as “travel.” For the non-European societies, scattered across
the surface of the earth and gradually described by travelers, also tes-
tified, in their own way, to history: the “savage” was man in his infancy.
Space and time thus provided complementary perspectives for the de-
veloping study of evolution. And it is on the basis of this epistemological
connivance that, in the later reforms of the French educational system,
geography and history moved forward in stride and as if intertwined.
But in the eighteenth century, the ill-defined state of history as a field is
sufficient proof of the extent to which the study of the past was still far
from being a school discipline: if history was not taught, it is because it
was not yet a teachable subject.
The two types of intellectual activity that composed history were too
foreign to each other to constitute a homogenous discipline. Erudition
was both too uncertain and too scholarly an art to be taught in the
classroom. It was indulged in by gentlemen and by a narrow circle of
specialists who discussed their discoveries far from even a cultivated
public. How could one teach numismatics in primary or secondary
school? As for philosophical history, it attracted many readers but repre-
sented too modern a genre, in every sense of the word modern, for it not
to be, educationally speaking, a dangerous product. In the eighteenth
century, it was still too recent a creation for its legitimacy to be recog-
nized. Thus, it did not command the respect normally associated with
subjects taught at school. Above all, philosophical history ran counter to
the classical tradition, in which history was no more than an annex to
belles-lettres, in other words, a fine story modeled on Livy or Tacitus.
The Jesuit secondary schools (colléges) remained faithful to their late
sixteenth-century charter, in which the model of antiquity constituted the
cultural identity of Europe. Students learned history—apart from sacred
history—only in the margins of Cicero.
There had been and still were exceptions to this rule. The elementary
schools of Port-Royal turned history into a core discipline to which part of
the daily schedule had to be devoted. But these schools were ephem-
eral—they were closed when Louis XIV persecuted the “Messieurs” of
Port-Royal—and strictly elitist, since they took in only the children of the
Jansenist upper bourgeoisie. The Port-Royal example, therefore, illus-
trates the exceptional rather than the regular presence of history in school
curricula. The Oratorians, too, later regretted not having made enough
room for history in their schools. And the military academies founded by
the monarchy in the third quarter of the century, with a view to training
professional soldiers, tried to include history in their curriculum. But,
until the expulsion of the Jesuits from the kingdom, in 1762, it was their
The Birth of History in France 83

colléges that set the pace for secondary education. Although they re-
mained conservative in their curricula—introducing Cartesianism, for
example, only in the eighteenth century—it would be wrong to think that
they were particularly ‘“reactionary.” The universities of the same
period—especially the University of Paris—were still far more insensitive
to the shifting boundaries of intellectual fields. At the end of the eigh-
teenth century, the professors of rhetoric of the faculty of liberal arts in
Paris found nothing to change in their traditional blend: a smattering of
ancient history tacked onto sacrosanct Latin discourse.
The fact remains that the expulsion of the Jesuits signaled the begin-
ning of a vast intellectual debate over the nation’s educational system.
The celebrated Jesuit schools, bereft of their teachers, were placed under
the control of the Parlement of Paris, whose task was now to fill them with
new professors and new ideas. Hence a flowering of educational schemes,
the most famous being the one put forward by La Chalotais. In a report of
1768, the president, Rolland d’Erceville, tried to make a synthesis of
these schemes. This was, in a sense, a Jansenist revenge, for the eigh-
teenth-century parlementaires had never really accepted the condemna-
tion of Jansenism by Rome, and they relished the political aspects of what
had been a rare example of resistance to Louis XIV’s absolute authority.
Thus, it was also a revenge for history, which had been held in such high
esteem at Port-Royal. But, for the parlementaires, the most detestable
feature of the Society of Jesus was the fact that it was a foreign order,
alien to the kingdom and totally in the hands of the pope. The par-
lementaires wanted a “‘national’” educational system controlled by the
state. This upper nobility of the robe, a passionately Gallican group,
expressed in its own manner and in its own cultural terms the strong
upsurge of a feeling of national identity rooted in a very distant past.
These men wanted school curricula to include not history in general so
much as the history of France, for it was this history that guaranteed the
original contract between the nation and its king and was the depository
of an imprescriptible tradition.
Actually, if one looks at the French secondary schools revamped at the
end of the ancien régime, one finds that history had made occasional
inroads into the curricula. At the celebrated Collège Louis-le-Grand on
the Montagne Sainte-Geneviéve in Paris, a compulsory half hour of
history on “days off, Sundays, and holidays” was introduced in 1769. A
number of historical topics were set as subject matter for “exercises,” the
public contests held on days off with the aim of testing the pupils’
expository and reasoning skills. In 1772, an exercise at the collége of
Arras was conducted in order to demonstrate that ‘‘the study of the
history of France in particular can alone impress in the lawyer’s mind the
true principles of our government.” In Lille, French history was included
in the curriculum itself, beginning with the class of troisiéme, during
which Gaul, the Germanic invasions, and the first two dynasties (Mero-
84 Seventeenth- and Eighteenth-century Culture

vingians and Capetians) were studied, and continuing in seconde up to


the sixteenth century.”
The best example, because it is perhaps the earliest, is that of the
famous Collége de Juilly, the showcase of Oratorian education, attended
by children of the kingdom’s elite. As old rivals of the Jesuits, the
Oratorians prided themselves on offering their ‘‘clientele” a more “‘mod-
ern” curriculum. By the end of the seventeenth century, an Oratorian
was recommending that “‘heraldry, geography, some chronology, and
history be taught with particular care, following the custom of this
academy.” These disciplines thus already enjoyed, in his view, an inde-
pendent educational status. They were, moreover, emancipated from the
exclusive relationship to antiquity. At Juilly, both the history of France
and the geography of America were taught. The history syllabus was
unusual for its time in that it was arranged in a chronological sequence
culminating, so to speak, in the history of France. The syllabus pro-
gressed from sacred history to the history of France via Greco-Roman
antiquity, from the junior to the senior grades. But it remained somewhat
marginal in that it did not form part of the formal curricula of the grades
themselves. History was taught in the chambres—that is, in the rooms
where the various groups of students lived and studied (Juilly was a
boarding school)—and outside official class hours. It also provided mate-
rial for the public exercises, which, although held on rest days (Thursdays
and Saturdays), were compulsory. However, they seem to have been very
popular with the students during the eighteenth century.*
These examples—to which one could add a few more—show that, by
the end of the ancien régime, the teaching of history had made some
progress in the secondary-school system and was gradually ridding itself
of the dual tyranny of sacred history and classical antiquity. But the
change was slow and advances were modest. As a teaching subject,
history remained in most cases a stowaway aboard official curricula. It
was a source for dissertation topics; it was not a self-sufficient discipline.
Moreover, history was absent from elementary schools, even from the
most advanced of the period, the establishments run by the Fréres des
Ecoles Chrétiennes. At the other end of the educational system, even the
highest teaching institution in Paris, the Collége de France (which practi-
cally specialized in innovation, since it was created in the sixteenth
century to counterbalance the Sorbonne’s inertia), had no chair in spe-
cialized history in the eighteenth century. The only innovation was a
course entitled ‘History and Morals,” which survived into the nineteenth
century under the sixth subject-category of the program, ‘Moral and
Political Sciences,” alongside “Natural and Human Law,” ‘History of
Comparative Legislation,” and “Political Economy.” History at the Col-
lege was thus freed from the yoke of ancient languages (which formed the
second category of professorial chairs), but it remained philosophical
history, divorced from erudition. The indirect and, so to speak, negative
The Birth of History in France 85

contribution of the Collége de France to the emergence of history as a


distinct discipline resides instead in the definition of specific cultural
areas. These were gradually detached from the common trunk of history
because of their marginal position with respect to the European experi-
ence. An example of this is sinology.
The turmoil generated by discussions about educational reform and
modernization contrasts with the slowness of pedagogic development.
This shows that it would be dangerous to confuse the history of ideas on
education with the history of education itself. The two histories do not
advance at the same rhythm, nor do they follow the same chronology. It is
true that they are not confronted with the same forms of inertia.

The Revolution: Break and Continuity

The preceding observation is particularly applicable to the period of the


French Revolution. The Revolution enacted laws concerning national
education far more than it lastingly transformed teaching institutions.
That is easily explained both by the brevity of the revolutionary period
and by the illusion—which was precisely typical of the period—that
people and things were being completely transformed. In actual fact,
secondary schools emerged nearly unchanged from the apparently uni-
versal collapse of the old institutions, and nothing resembles a collége of
the ancien régime more than a lycée under the Empire. In particular,
history remained a simple appendage to classical studies and to the
teaching of Latin.
Yet it is worth casting an eye over revolutionary legislation in order to
measure the development of attitudes and the aspirations of the new
political elites. The Constituent Assembly waited until the last days of its
existence (September 1791) to hear Talleyrand’s huge report on educa-
tion. The bishop worked into his report the ideas of the parlementaires of
the end of the ancien régime. While preserving the structure of the
classical secondary-school curriculum (grammar, humanities, rhetoric,
logic), Talleyrand introduced history and geography. Condorcet, who
took over from Talleyrand under the Legislative Assembly, was a direct
heir to the men of the Enlightenment; he had a wide-ranging and pene-
trating mind; a mathematician and philosopher, he was obsessed by the
division of knowledge into compartmentalized disciplines and the need to
recognize the unity of human learning. The scheme he devised was
guided by a totally different ambition, which was to reorganize the entire
education system along the lines of a “philosophical” classification of
knowledge, so as to place it at the forefront of intellectual innovation.
From the very first year of secondary school, pupils were to study ‘the
elements of all human knowledge,” divided into four categories:
(1) mathematical and physical sciences, (2) moral and political sciences,
86 Seventeenth- and Eighteenth-century Culture

(3) applied sciences (for example, comparative anatomy, midwifery, the


art of warfare, and the principles of arts and crafts), and (4) literature
and the fine arts (a category in which one finds—reduced to a very modest
share—the humanities as taught in the old colléges). Thus, history was
included under “‘moral and political sciences,” which also comprised the
analysis of sensations and ideas, morals, natural law, social science,
political economy, public law, and legislation. History in this context was
precisely what the eighteenth century had christened ‘“‘philosophical his-
tory’: a meditation on the evolution of peoples and civilizations, a study
of the past that was indispensable for analyzing humanity’s progress along
the path of reason. It was accompanied by chronology and geography,
two complementary tools for deciphering time and space.
From Condorcet, one can turn to Lakanal,° for the debates devoted to
education during the Montagnard period were characterized by an obses-
sion with the purely political aspect of the issue; moreover, they contrib-
ute nothing to our discussion. It was after 9 Thermidor that revolutionary
political society reclaimed its rights, temporarily relinquished in favor of
the Committee of Public Safety. The education act of Frimaire, year III
(December 1794), presented by Lakanal, established a two-tier educa-
tional system, with primary schools for all pupils and écoles centrales to
train the nation’s elite. The écoles centrales, combining secondary and
higher education, did away with the track system formerly used from the
sixiéme to the classe de philosophie. The new schools introduced a set of
parallel courses aimed at covering the entire span of human knowledge;
pupils were free to move from one course to another. Each school
comprised fourteen teaching chairs, representing the encyclopedia of
knowledge imagined by Condorcet and revised by the Idéologues. One of
these chairs was devoted to the “‘philosophical history of peoples.” In the
following years, the Thermidoreans, anxious to reestablish a higher-
education system in the form of special schools each concentrating on a
particular discipline, planned to set up a number of such schools devoted
to history. Legislation, political economy, philosophy, criticism, and
antiquities would also be taught there.
Thus, the Revolution, in the phases that preceded and followed the
Robespierrist dictatorship, engineered the triumph of a concept of his-
tory developed by the Encyclopedists and systematized by Condillac and
Condorcet. The ultimate aim was to turn history into a special testing
ground for demonstrating the meaning of social life. Philosophical history
was a secularized “discourse on universal history.’’ The question arises as
to why the revolutionary bourgeoisie, which committed itself with such
energy and patriotic fervor to waging war on Europe, did not display
greater fondness for the version of national history that had been de-
veloped by parlementaires and jurists and was very much alive at the end
of the ancien régime. In my view, there are several explanations for this
indifference. Some are of an epistemological nature: for Condorcet and
for the Idéologues alike, history belonged to the realm of scientific
The Birth of History in France 87

reasoning, and the glorification of national identity did not square with a
scientistic vision of the universe—a vision within which that identity
constituted a sort of irreducible residue. The other explanation involves
political ideology: the French revolutionaries did not regard their action
as taking place within a narrow, national framework. In its struggle for
liberty and equality, Jacobin and Thermidorean France constituted the
vanguard of mankind itself. Even when its armies ransomed conquered
countries, revolutionary France never abandoned the concept of univer-
sal democracy. Last, and perhaps not least, what was revolutionary
France to do with the nation’s interminable past, which belonged to the
monarchy and to feudalism? The Revolution was associated only very
briefly with the notion of a restoration of a golden age in the relationship
between the monarchy and the nation. All the ideas about a popular
contract, original rights, and a primitive constitution vanished as soon as
the Revolution displayed its true colors. It was the Revolution itself that
marked the starting point, established the original contract and constitu-
tion, and founded a new national history by wresting the French from
their past. Since the Revolution cut French history in half, why did it have
to recount that damnable part that belonged to its enemies? But the other
part was still too short to constitute a past; it was only the celebration of
origins.
Thus, for the French revolutionaries, history was not a genealogy, as it
was to be for nationalist ideologies of the nineteenth century. Instead, it
provided a universal reference chart against which the excellence and
supreme rationality of the French experience could be measured. The
French Revolution saw history as a laboratory for social science and as
the agent responsible for organizing the materials for that science—not as
a form of knowledge in its own right, built around the chronological study
of the nation’s annals. Consequently, the French Revolution bequeathed
no enduring doctrine of history to posterity. The notion of “social sci-
ence” lived on in the work of Saint-Simon and Auguste Comte but as a
marginal, illegitimate, and suspect current in our culture. The encyclo-
pedist concept was already obsolescent under the Directory, and it never
passed into the educational system. As for national history—the prime
area for the emergence of history as a discipline and for its legitimation as
a school subject—the French Revolution turned it into a battlefield for an
intellectual civil war. The French during the nineteenth century were a
people who could cherish only one-half of their history: who could not
love the Revolution without despising the ancien régime, nor love the
ancien régime without despising the Revolution.

History Becomes a Discipline

Thus, history became an all the more burning educational problem and
issue as its development—both epistemological and institutional—led it
88 Seventeenth- and Eighteenth-century Culture

to focus on the genealogy of the nation. History therefore became less


easy to teach to the French people as a whole. The First Empire was able
to ignore the question and to reinstate the study of history as a mere
annex to the study of Latin. This bracketing amply demonstrated the
imperial regime’s capacity and determination to forget, but it was not a
durable solution. Even as the Restoration, personified by Louis XVIII
and the émigrés, was bringing back to power the tangible forces of the
past, history was consecrated as a full-fledged discipline both in France’s
school system and in its intellectual life. Guizot, nearly half a century
after Gibbon (whom he translated into French), blended the antiquarian
and the historical traditions. At the same time, he reconciled national
history with the history of civilization. “‘Philosophy”’ had been the tribu-
nal of the eighteenth century. History became the magisterium of the
nineteenth century.
But precisely what history was it? The Restoration, the first regime to
introduce a systematically chronological history syllabus, sought to give
pride of place to its own genealogy, that is, to the monarchical tradition.
A decree of 1814 drafted by Royer-Collard, and thus informed by consti-
tutionalist and moderate attitudes, divided the history syllabus into chro-
nological sections for each grade in the lycées and collèges: sacred history
in sixiéme, Egypt and Greece in cinquiéme, Rome (up to the Empire) in
quatrième, Augustus to Charlemagne in troisiéme, the Middle Ages in
seconde, and the modern period and French history in première. It was an
attempt to combine the old curriculum based on sacred history and
antiquity with the requirements of a laicized chronology, more modern
and more “national” in spirit. The new syllabus sought to emphasize
France’s dual tradition—Catholic and dynastic—and to train young
minds in the ways of the monarchy as embodied in the Charte. History is
never innocent; it was less so than ever in nineteenth-century French
culture. But it is significant that the constitutional monarchy should have
regarded history as an indispensable medium for diffusing its own values.
Indeed, history had difficulty surviving the hardening of the regime in
1820 and the fall of the Constitutionnels. Even if the cutoff in the syllabus
was set in 1789, an explanation still had to be divised for the terminus ad
quem represented by the French Revolution, which dominated the entire
sweep of history that preceded it. But among the concepts that could have
been used for such an explanation, the ultraroyalist Right was incapable
of making intellectual room for progress, democracy, or the nation. Allit
could suggest was divine right and Providence—a return to Bossuet.
Thus, when the ultraroyalists came to power, history became a suspect
discipline, to be strictly watched over by the authorities, not only in the
secondary-school system but also in the faculties of liberal arts, where
lectures became political and social events. While history vegetated in
secondary schools, Guizot restored the Sorbonne’s prestige by attacking
Villèle’s government in the name of the third estate, the old monarchy,
The Birth of History in France 89

and the progress of civilization. Guizot’s dismissal in 1822 represented a


new blow struck against the established upper bourgeoisie of the third
estate, against the Protestant tradition, against freedom, and against the
values of 1789. The fall of Villéle in 1827 was thus the crowning point of
history’s revenge. Shortly after, history was completely emancipated
from the tutelage of the humanities by the appointment of specialized
teachers (soon to be recruited by a newly created agrégation in history) in
the secondary schools.
But it was in 1830, with the July Monarchy, that the teaching of history
entered a decisive period. No doubt one of the reasons was that two of the
greatest French historians of the nineteenth century, Guizot and
Michelet, were at the peak of their careers: Guizot was in the government
(a fact that was to impair—unjustifiably—his reputation as historian);
Michelet was a member of the intellectual and republican opposition at
the Collége de France. But the main reason was that the Orleanist
regime, a product of the Parisian insurrection, had no other legitimacy
than the one it could derive both from the ancien régime and from the
French Revolution. Unlike Bonapartism, Orleanism could not conceal
the flimsiness of its legal foundations behind a legend or behind the
notion of a prior consent given to despotism. It therefore had to take up a
position at the exact point where the two liberal traditions of French
national history—that of the nobility and that of the bourgeoisie—still
stood side by side and strengthened each other. In other words, Orlean-
ism had to set up 1789 as a new starting point, but this time as a link
between past and future, not as a dividing line or a stake in a civil war.
Louis Philippe turned the Chateau de Versailles into a museum of na-
tional glories and brought the emperor’s coffin back to the Invalides. The
history of France thus became the regime’s highest legitimating author-
ity. The July Monarchy lavished its attention upon history as if it were a
beloved child, as evidenced by the extraordinary effort made during this
period to preserve the nation’s archival heritage.
This political determination also expressed itself in education policy.
In 1838, Salvandy, the education minister, revamped the history syllabus
by a “downstream” chronological shift: sacred history, Asia, and Greece
in sixiéme; Magna Graecia, Macedonia, and the Jews in cinquiéme;
Rome in quatriéme; the Middle Ages in troisiéme; the modern period
(1453-1789) in seconde; and the history of France from 406 to 1789 in
première. One can already discern the outline of the academic chronology
to which we are still subjected, since in the French historiographical
consciousness the ‘“‘modern’”’ period ends in 1789, as if that were a
universally valid terminus. Moreover, premiére was devoted entirely to
the history of France-—a demonstration that history had come to occupy a
central pedagogic role, in contrast to the humanistic tradition. When
Louis Philippe, in 1838, invited pupils from two Parisian colléges to the
Château de Versailles, granting them the honor of visiting their past in his
90 Seventeenth- and Eighteenth-century Culture

company, he made no mystery of his aims: “I wanted you to profit from


all these fine examples of our history, all these glorious relics of the old
French monarchy. The monarchy was certainly the equal of the Athenian
and Roman republics, which take up perhaps too much of your time.”
In this process, history as a whole, and not just the history of France,
became one of the key issues in the French political and intellectual
debate. Guizot’s ecumenical and middle-of-the-road history encountered
opposition on the Right from reactionary traditionalists haunted by the
prospect of a politicized youth and on the Left from the two great
democratic interpreters of national and European history, Michelet and
Edgar Quinet. It would be outside the scope of this essay to recount the
two professors’ celebrated struggle against the Jesuits and against clerical
control of the university; but, for the purpose of my argument, it is
important to realize to what extent this battle destabilized a history that
the July Monarchy would have liked to center on 1789 and its “‘remake,”
1830. Guizot had seen the July Revolution as a new anchorage point for
the accomplishments of 1789, as a sort of French 1688 that would open up
for France an era of prosperity comparable to that ushered in by the
Glorious Revolution in England. Michelet and Quinet countered with
other references drawn from history: the dynamics of the Reformation,
the unfinished character of the Revolution, and the boundless expecta-
tions aroused by democracy. The consensus of the French on their history
does not seem to have run any deeper than their agreement about the July
Monarchy. Both that consensus and that agreement vanished in 1848.
Yet the process begun under Louis Philippe was irreversible: however
much history and the history syllabus continued to provoke strident
political conflicts, they remained central to every change in education
policy. Finally, the Second Republic (despite the grumblings of conserva-
tives) and the Second Empire (despite its distrust of critical thinking)
continued along the lines of Salvandy’s reform. In 1848, Carnot added
the 1784-1814 period to the syllabus of seconde and of the classe de
rhétorique: the Revolution and the Empire had found their way into
secondary education. In 1852, the decree establishing the ‘‘fork”’ between
literary studies and scientific studies (another aspect of educational mod-
ernization) also reorganized the history syllabus. Biblical history was
henceforth taught to the very young, in huitième and septième. Sixième,
cinquième, and quatrième were devoted entirely to French history up to
1815. Finally, in troisième, seconde, and première, the syllabus consisted
of the by-now-classic triad: antiquity, the Middle Ages, and modern
times. But this reform, authored by Fortoul, was transformed in turn by
the syllabus changes carried out by Victor Duruy in 1865. Biblical history
was now taught in huitième, French history in septième. The introduction
of French history into elementary grades was the sign of a growing
consensus on the pedagogic need for this subject and on its content. In the
higher grades, Duruy introduced a very modern division: antiquity from
The Birth of History in France 91

sixième to quatrième, the Middle Ages in troisième, the early modern


period in seconde, the 1661-1815 period in première, and the nineteenth
century up to the Second Empire in the classe de philosophie. Thus,
Duruy won the battle for contemporary history, which he brought into
secondary education. Not only did he raise the nineteenth century to the
noble rank of history, but, in so doing, he broadened the history syllabus
to include economic and social issues. For history was no longer simply
the genealogy of the nation; it was now also the study of mankind’s
scientific and material progress. Thus, the groundwork was being laid, in
these new circumstances, for the reconciliation of the nationalist and the
encyclopedist philosophies.
For two sets of reasons, history appeared as the ideal medium for such
a reconciliation. The first set was scientific in nature: reasons concerning
the status of history as a type of knowledge and as a discipline. The
mid-nineteenth century (or the last third of the century) was a decisive
period for the development of historical studies in France—the most
important, perhaps, since the Benedictines of Saint-Maur. Taine, Renan,
Fustel de Coulanges, and Gabriel Monod put history once again on a
scientific footing. In 1866, Victor Duruy founded the Ecole Pratique des
Hautes Etudes with the aim of introducing German research practices
into France, that is, of substituting the transmission from master to
student of a rigorous method and of critical learning for the “social
occasion”’ lecture favored by the universities.
But, if history appeared to be draped in the intellectual prestige of
science, it essentially continued to represent—as far as social expecta-
tions were concerned—not what society knew about itself but what the
nation knew about its past. This was another aspect of history’s towering
prestige. But after the few years of the ordre moral, a period that revived
reactionary fears about the inherent dangers of history, the victorious
Republic gave the French not only a regime on which they could reach a
durable consensus, but a cumulative interpretation of their conflicting
traditions. Unlike the July Monarchy, which sought above all to find a
geometrical point common to the country’s ruling classes, the Third
Republic integrated Michelet’s vision into Guizot’s and provided the
entire nation with a democratic history of itself. In this history, not all the
kings of France were depicted as models of virtue or conscientiousness;
but, on average, they were shown to have built France and to have
contributed to its progress and prestige. The Revolution itself was guilty
of excesses, but the Declaration of the Rights of Man and the armies of
the year II made French history a sort of universal model. Thus, the two
halves of French history were less conflicting than complementary—not,
as Tocqueville had written, because they had in common a centralized
administrative state, but because they shared the cult of the nation-state
as instrument of progress. The fledgling Third Republic adopted all of the
national heritage in the name of the people, because the Republic itself,
92 Seventeenth- and Eighteenth-century Culture

at last, after nearly a century, was the French Revolution in power. That
Republic, a provisional entity that nevertheless proved relatively long-
lived, represented a conservative authority governing in the name of
revolutionary values.

History’s Magisterium in the Nineteenth Century

Henceforth, history was no longer just a subject in the secondary- or


higher-education syllabi. It was also indispensable for very young chil-
dren, who had to be trained early in patriotism and critical judgment.
Once the general pattern of history had been made familiar—a pattern
based on a history of France fitted into a universal history marked by the
material and moral progress of mankind—the schoolmaster could forgo
abstract ideas in the elementary grades. The general “philosophy” of
evolution could be conveyed just as well by an anecdote, a detail, or, to
use the very apt phrase, “by a story.”’ Here, for example, is Lavisse, in his
famous “Instructions” of 1890, recommending the teaching of ancient
history (one can measure the road traveled since the days of the Jesuit
collèges): ‘The history of Greece and Rome is already our history, since
it already contains the origins of modern intelligence and of modern
politics. These origins must be shown and explained to pupils, but practi-
cally without their being aware of it. They must not be treated to philo-
sophical considerations or encumbered with details about institutions.”
Antiquity was no longer a model but an introduction to the history of
Europe and France. It no longer gave the modern world its meaning but
derived its own meaning from the modern world.
Lavisse’s ability to write history at all levels—not in the same manner
but with the same assurance (and, incidentally, with a great felicity of
expression )—testifies to the extent to which history had become a form of
knowledge and a discipline bordering on a sort of pedagogic classicism.
Not that Lavisse was ever superficial—he was extremely well read—but
he always knew where he was headed. A great admirer of the Ency-
clopédie and the eighteenth century, Lavisse wrote his own variety of
“philosophical” history, dominated by an enlightened and learned
bourgeoisie that was gradually emancipated from the church and the
monarchy and soon brought to the entire world the conquests of science
and progress. But there were two fundamental differences between this
“philosophical history” and its predecessor. The new history had in-
corporated the ars antiquaria in the consummate form of positivism, and
it had made the nation-state the central protagonist of evolution. In
short, it had a method and an object. It had become what one calls a
discipline.
I shall spare the reader a commentary of the famous “petit Lavisse,”’
the textbook in which several generations of French people have acquired
The Birth of History in France 93

a lifelong knowledge of the basics of their history. It is certainly a crucial


piece of evidence as regards the pedagogic and social usefulness of history
in late-nineteenth century republican France, for it is at this level that the
effects of historical writing must be particularly visible in order to have
the maximum impact on young minds. But in his celebrated text of 1885
on history instruction in primary schools, Lavisse himself wrote a most
explicit metahistory of his history of France for children. In this text he
provided the best possible explanation of what he was doing. The
fledgling Republic did not have a guilty conscience. Never was the
general vision of the nation’s history more limpid. First came the slow
birth of France, made possible by the kings’ struggle against feudal chaos.
The decisive turning point was the Hundred Years’ War.

Once the English had been driven out, our France emerged.
But, in this France, the central figure was the man in whom
Joan of Arc put her hopes: the king. Because of the very fact
that he had united his kingdom and reconquered it from the
enemy, he, so to speak, concentrated all of France in his
person. And here is what pupils must be clearly told: by the
fifteenth century, when there were no more powerful vassals,
when Louis XI had annexed the last great independent prov-
inces, and when the communes had been disorganized by the
king’s agents and ruined by war, the king was no longer a
suzerain and protector; he was a master.

The next phase was the development of absolute power—an ambig-


uous history, since it led France to European hegemony but also under-
mined the nation and oppressed the French people. The Revolution
extended the “‘good side”’ of the monarchy while eliminating the bad side.

It is an indisputable truth that the French Revolution made a


heroic effort to replace the old monarchy by the reign of
justice and reason. It is an indisputable truth that it ushered in
a new era in the world and that nearly all of Europe was, as it
were, reshaped by the Revolution. Thus, the schoolmaster
can wound no conscience when he expounds the principles of
this Revolution and shows how, by the force of our weapons
and ideas, absolute governments were transformed every-
where and how new peoples, in the course of our contempo-
rary history, acquired the right to exist.

But beware! The following warning was crucial for future citizens:

It is an indisputable truth that the ideal regime dreamed of by


the French Revolution is the most difficult of all to put into
practice. The Revolution itself and the succession of coups
d’état show this quite clearly. It is an indisputable truth [the
94 Seventeenth- and Eighteenth-century Culture

repetition of this introductory clause is in itself a telling indica-


tion of the fact that all these truths are precisely contested and
should not be] that these ‘revolutions and coups d’état have
weakened France, and that further occurrences of them
would be fatal to her. Thus, the schoolmaster cannot mislead
any conscience when he teaches that any act of violence
against the law is an attack on the country, and that the
condition for France’s survival is political stability.

Lavisse concludes with a thought for Alsace-Lorraine: “The school-


master who has recounted to his pupils the destiny of France, of all of
France—old and new—will be well aware of what one should think of,
and say about, the mutilation suffered by France fifteen years ago.” The
aim of teaching history was as clear as the fact that school had become
secular, compulsory, and free. It was to train ‘‘citizens imbued with a
sense of duty, soldiers who love their rifles.”
In secondary education, syllabi were broader and guidelines more
subtle. Above all, directives were designed to take into account the
pedagogic changes that were due not to republican ideology but to the
development of the discipline itself. In this field, the Third Republic
began by consolidating the work of Victor Duruy, which had been
imperiled during the ordre moral. Most notably, with the reform of 1902,
which reorganized the whole of the secondary-school system, the Repub-
lic once again modified the history syllabus. History was divided into two
cycles at the secondary level—a division that has survived to this day:
antiquity, the Middle Ages, the modern age (up to 1789), and the
contemporary period from sixième to troisième; and, once again, modern
history in seconde and première (but up to 1815: the no-man’s-land
between 1789 and 1815 was still hard to label) and contemporary history
in philosophie. An interesting commentary on the background to this
reform was written by the man who played a crucial role in its prepara-
tion, Charles Seignobos. In fact, it is a general introduction to his
“course,” which provided the substance for textbooks for all grades from
sixiéme to philosophie.
Seignobos does not distinguish between what he calls the “revolution”
in the approach to history teaching—a change dating from Lavisse’s
celebrated “‘Instructions’’—and the development of the discipline itself.
He does not clearly separate the two sets of phenomena—a separation
that strikes us today as obvious and necessary. For Seignobos, the peda-
gogic autonomy of history overlaps with its emancipation as a form of
knowledge. In other words, the triumphant entry of history into the
school syllabus in this period represents a consecration for a fully consti-
tuted discipline with its specific method, object, and social utility—the
three aspects being indissolubly linked.
History’s “‘subject matter’ was no longer confined to commentary on
great Greco-Roman literature, as in the Jesuit colléges, or to the analysis
The Birth of History in France 95

of treaties and wars, as in the tradition of the Ecole Militaire. It no longer


trained for a particular career. Its purpose was to make every Frenchman
a citizen. ‘“The study of science teaches us about the material world. The
study of letters opens up the world of forms and ideas. History introduces
the pupil to the social and political world. The humanities of old were
oblivious to the world—and the world kept them at arms’s length. But a
Frenchman, who is destined to live in a democracy, needs to understand
that democracy.” The purpose of teaching history was, therefore, to
build a general social science that would teach schoolchildren both the
diversity of past societies and the general direction of their evolution. But
the past remained “‘genealogical,” chosen in terms of what it was sup-
posed to portend or prepare: classical antiquity, the Christian Middle
Ages, modern and contemporary Europe. The other societies, scattered
in space, were abandoned to other disciplines. History honored by its
interest only those societies that participated in ‘‘evolution’’-—another
name for progress.
Hence the emphasis on the contemporary period, to the detriment of
antiquity and the Middle Ages. Not only was there a desire to underscore
history’s hard-won independence from the humanities, but the contem-
porary period was felt to make the past meaningful and thus to justify its
study. ‘Modern times from the sixteenth century onward now form the
substance of the history syllabus. Most of the facts that need to be known
in order to understand the present state of the world date from this
period.” But even within “modern” history the traditional proportions
were reversed: the seventeenth century, “‘in which no deep change took
place outside of the revolutions in England,” was whittled down—to the
advantage of the eighteenth century, “‘which witnessed the formation of
the great contemporary states, the Russian Empire, Prussia, the United
States, parliamentary Britain, and revolutionary France,” and the
nineteenth century, “during which material and intellectual life was
revolutionized by the definitive establishment of science, and political life
was transformed by representative government and democratic equal-
ity.” History was not only a genealogy but also the study of change, of
“upheaval” and “transformation,” an area singled out for greater atten-
tion than the stable features of the past. Moreover, genealogy and change
were twin images: the search for the origins of contemporary civilization
became meaningful only through the study of the successive stages of its
development.
This enclosure of history’s field involved a change in the nature of the
facts to be studied and taught. It became imperative to abandon intermi-
nable chronological lists, in particular the enumerations of kings, minis-
ters, generals, battles, and treaties that needlessly cluttered pupils’
memories. The main task was to emphasize two categories of facts: those
pertaining to material civilization—because it formed the basis for civi-
lization itself—and those that made it possible to understand the specific
96 Seventeenth- and Eighteenth-century Culture

character of a period with respect to another, that is, to understand


change. And these facts were, of course, to be located, dated, and
described according to the method introduced by positivism—a method
aimed at awakening pupils’ critical faculties instead of appealing to their
memory alone. Finally, the new way of teaching history, which was no
longer to be a moral lesson or the occasion for literary commonplaces,
had to give up rhetorical or philosophical style: ‘Now that history has
begun to establish itself as a science, the time has come to break with the
Roman and academic rhetorical tradition and to adopt the language of
natural science.”’

A School for Citizenship

History’s emergence as a full-fledged teaching subject in the late


nineteenth century was due to an inseparable combination of a scientific
method, a concept of evolution, and the chronological and spatial focus
on Europe as a field of study. The basic rules of the ars antiquaria,
codified by the positivists, made their way into secondary education via a
provisional consensus on the meaning of history. To obtain this consen-
sus, Lavisse and Seignobos revived the two themes of philosophical
history since the eighteenth century: history is the nation; history is
civilization. But they united the two themes in a far more organic manner
than Guizot and the men of the July Monarchy had been able to do fifty
years earlier. The Orleanists had remained trapped within a strictly
bourgeois concept of evolution and “civilization”; by cutting off history
in 1830—that is, in 1789—or even at the establishment of habeas corpus
in England, they were offering too restricted a base for the historical field
opened up by the French Revolution. The French were never sufficiently
enthusiastic about representative government to make it the culmination
of universal history.
In contrast, “the evolution of mankind,” a la Seignobos or à la
Lavisse, offered the French a series of historical configurations on which
they could more easily reach a consensus. The internal arrangement of
these configurations involved three successive levels. First came ‘‘civiliza-
tion,” another name for the scientistic prophecy that prevailed in the late
nineteenth century. Through intellectual advances, civilization led man-
kind to mastery over nature. The chief historical agent of this intellectual
and material progress was the nation, or, more precisely, the nation-state
invented by modern Europe. And the history of France, through the
medium of absolute monarchy and the French Revolution, offered the
prime example of the nation-state as vehicle of progress. For it is inaccu-
rate to say that the republican historiography of the period was narrowly
chauvinistic. With all its nationalism, it constantly made reference to the
universal democratic ideal, following the Jacobin example. What made
The Birth of History in France 97

the history of France a choice topic was not only its particular value but its
universal value. It took a hundred years for Mably and Condorcet to be
united through the agency of the republican school system.
The other facet of this analysis would consist in looking at why and how
this consensus has since broken down, especially after the Second World
War, under both internal and external pressure—resulting from the
evolution of historiography and of the social sciences in general and from
the end of European ascendancy in the world. Although, or because,
school syllabi always live on long after the circumstances that explain
their birth have disappeared, there is a general awareness today that the
teaching of history in France once again needs to be reexamined. And
perhaps the first step, before putting forward any proposals, would be to
understand what has come undone in the past hundred years. But one
must understand beforehand the various elements of the total picture.
One must first journey back through time. The following are what seem
to me to be the main stages of the journey.
In order to become a school discipline, history has had to undergo
several changes, so as to constitute an intellectually autonomous, socially
necessary, and technically teachable field of knowledge. For history, by
nature, lacks a particular object (since everything is ‘‘historical’’), an
autonomous language (since history is narrative), and assignable limits
(since it exists everywhere and nowhere). It is, therefore, particularly
hard to conceive history as a discipline, and all the more so as a school
discipline. Either it is not teachable or else it is taught—as was the case for
many centuries—as an appendage to classical letters. Even when it
became a school “subject,” its boundaries were laid down with painstak-
ing care, out of fear that the pupil would be lost in the ocean of “‘historical
facts” and fail to learn the language or method of history.
Since the seventeenth century, history’s progress toward autonomy
has followed two parallel and consequently independent tracks. Philo-
sophical history won the battle of the ‘““modern”’ against the “‘ancient”’
and eventually developed, thanks to Condorcet and the Idéologues, a
doctrine of progress. For its part, the ars antiquaria—from Port-Royal to
the Benedictines of Saint-Maur via the Académie des Inscriptions—
constructed a method for locating and evaluating historical facts. But in
the absence of a French Gibbon, the Enlightenment never united the two
learned traditions. It bequeathed to the Revolution and to the nineteenth
century a secularized discourse on universal history, along with a set of
distinct techniques and descriptive forms of knowledge—chronology,
paleography, travel, and so on.
It was these dissociated traditions that the nineteenth century thor-
oughly reshaped and redefined in order to produce, at the beginning of
the Third Republic, the history taught to French youth. Reshaping
meant, first of all, excluding. The age of classicism had shown the way by
beginning to place outside history certain sectors of the vast spectacle
98 Seventeenth- and Eighteenth-century Culture

offered by human societies. “Travel” began as an inventory of space,


before developing into geography and anthropology. The study of past
and present non-European civilizations, which required particular lin-
guistic skills, generally evolved into distinct specialties. This trend con-
tinued under the Restoration and the July Monarchy, as evidenced by the
highest teaching institution, the Collége de France. At the same time, the
history of religion was also breaking away from the main trunk of history
to become a special field of scholarship. Conversely, in the wake of the
decline of Latin in the school curriculum, learned history gradually
appropriated Greco-Roman antiquity as a subject to be taught as some-
thing other than a literary model. What had been a cultural identity for
the Europe of letters gradually became its genealogy.
Indeed, the great transformation of history in the nineteenth century,
and particularly in the 1820s and 1830s, lies in this shift. History was now
the genealogical tree of European nations and of the civilization they
bore. Guizot still regarded both France and England as models; for
Michelet, France alone was the model. Once the encyclopedic discourse
of the eighteenth century had been vested with that particular signifi-
cance, national history was delivered from the curse of “feudalism”
inflicted on it by the Revolution. The history of France represented both
a choice image (but not the only one) of mankind’s progress and a
“subject” for study, a heritage of texts, sources, and monuments that
made it possible to reconstruct the past accurately. It was at the con-
fluence of these two notions that the positivist ‘revolution’ took hold,
giving to both of them the blessing of science. Henceforth, history had a
field and a method of its own. On these two counts, it had become the
main school for citizenship.
Book Licensing and Book Production
in the Kingdom of France in the
Eighteenth Century

For the historian, the book is always a perplexing object. Wrapped in its
title as if in a timeless definition, the book is forever enclosed, and yet it
never ceases to take on successive meanings. Although it is the product
par excellence of individual effort, it presupposes a community of lan-
guage and a complex system of social connections. It is doubly myste-
rious—as an invention and as a familiar object. Indeed, the study of the
book crystallizes all the difficulties of the historian’s craft: the passage
from the individual to the collective, the relation between the intellectual
and the social, time’s judgments on time, the measure of innovation and
inertia. Even though it has accumulated so many layers of criticism,
human writing is far from having been deciphered in terms of history.
For the past century and a half, traditional literary scholarship has
been tracking down the secrets of the book on two levels: internally,
through the study of the texts themselves, and externally, through the
biographical approach. Thus, traditional literary scholarship has been led
to make assumptions about social and collective phenomena on the basis
of evidence about individuals. It is precisely this approach that the
historian would like to reverse. Not that he is obliged to settle beforehand
the old debates about collective versus individual behavior or infrastruc-
ture versus superstructure. But the historian’s raison d’étre is to integrate
accidental phenomena into a quantitative and intelligible framework. He
must therefore isolate—in the extraordinarily bewildering mass of

First published as “La ‘Librairie’ du royaume de France au XVIIIe siècle,” Comité


International des Sciences Historiques, XIIe Congres International des Sciences Histo-
riques (Vienna, 1965), in Rapports, ed. Hanns Leo Mikoletzky (Horn-Wein: Verlag F.
Berger, n.d.), vol. 1, Grands thèmes, 423-S0. Reprinted in G. Bolléme, J. Ehrard, F. Furet,
D. Roche, J. Roger, and A. Dupront, Livre et société dans la France du XVIIIe siècle, vol. 1
(Paris and The Hague: Mouton, 1965).

99
100 Seventeenth- and Eighteenth-century Culture

creativity represented by books—a certain number of trends and con-


stants. He must begin with those which, because they are external and
measurable, are easiest to identify.
One can then venture a comparison with recent advances in economic
history. Following in its footsteps, historians should be able to combine
macroscopic and microscopic analysis, drawing up a general census of
literary production in a society,' while conducting an ever-greater num-
ber of specific investigations into sociocultural circles and groups. It is
only if the two types of research can be combined that one can attain a
degree of historical certainty about major trends and their mechanisms.
But, in both cases, the analytical method involved is deliberately
external to the unique melody of each book. These methods may there-
fore be suspected of leading to oversimplification. To avoid a pointless
debate, suffice it to reply that if nothing can replace the study of the texts
themselves, this type of research can bring to light with greater force—
and with fewer theoretical presuppositions—the major points of conver-
gence of a society and its written output. These investigations must be
understood simply as a preface to, and a framework for, literary analysis
proper—as nothing more, but nothing less.

As luck would have it, the historian of the printed book in the eigh-
teenth century has a series of highly valuable quantitative sources at his
disposal: the records of the royal book-control bureau, the Administra-
tion de la Librairie, recently unearthed by Robert Estivals.* By the
Renaissance, the French monarchy had secured control of the kingdom’s
literary output. It kept a careful record of printed matter at the Chancel-
lerie, which delivered printing licenses, and at the Dépôt Légal, which
registered published books. This extremely thorough and complex
bureaucratic accounting of book production has the posthumous advan-
tage of making it possible to conduct surveys on a very large scale and
thus to improve on later, incomplete bibliographies. It also offers the
historian a more sophisticated perspective: the relation between a society
and its own literary works. In this respect, the eighteenth century is an
ideal period for study.
In actual fact, the French monarchy under Louis XV and Louis XVI
was not the crude force of repression and censorship described a pos-
teriori by the revolutionary liberation movement. Not that it is inaccu-
rate, in a certain sense, to regard the Age of Enlightenment as a long
struggle between the forces of intellectual innovation and those of con-
servative resistance. But the role of inquisitor devolved primarily, not on
the central government, but on the Parlements, which happened at the
same time to be the staunch adversaries of the royal bureaucracy. The
king of France allowed more books to be published than the Sorbonne or
the great magistrates would bear, as is shown by the crisis over the
Encyclopédie. The truth is that the monarchic state, which put
Book Licensing and Book Production 101

Malesherbes in charge of the book trade for twelve crucial years as the
first directeur général de la librairie, followed prevailing trends more than
it directed them. The monarchy was not only very sensitive to pressures
from civil society but receptive to contemporary ideas and to the concept
of a more rational government. In short, it had become both weaker and
more modern. This change puts into clearer perspective the official
sources one can use to study the control of ideas. The technical and social
value of these sources makes them a rather exceptional document pro-
duced by a society on the subject of its own writings.
Most of this society’s literary output can be inventoried, since each
work had to be granted a printing license by the chancelier. But even
books that were refused a license were recorded on the register of
requests, and these victims of censorship do not elude our general count.
Traditionally, there were two types of licenses: priviléges and permissions
du Sceau. The first, which was costlier,’ also gave the applicant a monop-
oly on the work for a specified period. The second was not exclusive, but
it saved the expense of a privilége. Both licenses were public and were
explicitly mentioned in the books concerned; they were tantamount to
government decrees. The book-trade code (Code de la Librairie) of 1723
restated this procedure, which was more than a century old. Both types of
licenses were revocable, temporary, and renewable. When his privilége
expired, the printer who wanted to keep his monopoly or reissue a work
had to ask for a renewal or an extension of his privilége. The new license
was recorded in the same register and is, thus, a likely indication of the
work’s success.
The tendency for printing licenses to be renewed was a boon for
Parisian libraires (publisher-booksellers), who were in closer contact
than their provincial colleagues with the authorities and with authors. It
also fueled a long controversy in the eighteenth century. Provincial
libraires were opposed not to licenses but to their renewal. In a famous
text, Diderot came out in favor of the Parisian libraires in the name of
property rights.‘ This debate, from which the modern notion of copyright
gradually emerged, was resolved on 30 August 1777 by Louis XVI, who
issued a series of important decrees. Henceforth, an author who secured
a license in his own name and sold his work himself could hand down to
his heirs a perpetual copyright for his work. But the transfer of the
manuscript to a third party reduced the copyright to a life annuity, for the
license granted to the libraire was valid only during the author’s lifetime.
(However, a ten-year minimum was stipulated if the author died within
ten years of the issue date.) This marked the end of the renewal of
privilèges, and these renewals ceased to appear in our registers.
In addition, the 1777 decrees seem to have led to a new type of permit,
which did not bear the stamp of the Grand Sceau: the permission simple.
The Manuel de l’auteur et du libraire of 1777 does not mention it, while
the Almanach de la librairie of 1778 gives the following definition: ‘The
102 Seventeenth- and Eighteenth-century Culture

permission simple gives no other right than that of publishing a work for
which no license exists or for which the license has expired, under the
provisions of the Council’s decree of 30 August 1777 on licenses. Permis-
sions simples are issued with the simple signature of M. le Directeur
Général [de la Librairie].
“Any libraire or printer in any town can obtain a permit of this kind,
but he must have it recorded within two months on the registers of the
Chambre Syndicale [de la Librairie—the libraires’ guild] in the arron-
dissement corresponding to his place of residence, otherwise his permit
will be null and void.”
Thus, the decrees of 1777, by placing more books out of copyright,
made it easier to reprint them, especially for provincial libraires. How-
ever, these books slip through our survey, since the requests for permis-
sions simples were handled not by the Chancellerie but by the directeur
général alone.
Another type of printing license also appeared in the eighteenth
century: the permission tacite (‘‘tacit permit”). The best comment on it
was made by Malesherbes, in his fifth Mémoire sur les problèmes de la
librairie, written in 1759. First he discusses its origins: ““As the tendency
to publish on all sorts of subjects has become more widespread, and as
private individuals, particularly the powerful, have also become more
sensitive to being mentioned in print, circumstances have arisen in which
the authorities have not dared to grant public permission to publish a
book, and in which, nevertheless, they have felt that it would be impossi-
ble to forbid it. This is how the first tacit permits originated. . . .”
Malesherbes added that they had been on the increase “for the past thirty
years,” but in 1788 he wrote that he did not know when they had begun.°
His predecessor as supervisor of the book trade, the comte d’Argenson,
who had been chief of police during the Regency and had been, since his
very birth, familiar ‘‘with all the secrets of government,” had always
resorted to tacit permits. “I think,” Malesherbes added, ‘‘that they began
at about the time of Louis XIV’s death.” But the fact remained that they
were illegal—under the laws then in force—because they were not public:
“The only difference between these illegal permits and the others is that
they are not handled by the Sceau and that the public does not see the
censor’s name. This practice was probably introduced to give libraires
and authors some token authorization while protecting censors from
complaints by hostile [i.e. unsuccessful] applicants. But a record of these
permits is kept and the censor at fault is not immune from the rigors of
official punishment.”
These records are indeed preserved in the manuscript division of the
Bibliothèque Nationale under the title ‘Registre des livres d’impression
étrangère présentés à Monseigneur le Chancelier pour la permission de
débiter.” The first volume begins in 1718, but the series kept its fictitious
title until 1772 in order to conceal the illegal character of the new practice
Book Licensing and Book Production 103

by presenting these works as imports. The tacit royal license was confined
to the distribution of the book in question; officials refused to accept any
responsibility with regard to the act of printing. For the same reasons, a
great many eighteenth-century works printed in Paris by tacit permission
listed Amsterdam, London, Geneva, or even Peking as place of publica-
tion. It is only in 1772 that the registers of these licenses were labeled as
what they really were: an interesting sign that the royal bureaucracy was
no longer afraid of its own past and its own laws.
But, by the early eighteenth century, at the cost of a falsehood collec-
tively agreed upon, the bureaucracy authorized and kept count of a
literature that it itself designated as bearing the stamp of presumed
nonconformism. The state accepted this literature as inevitable, that is, as
the expression of a social and intellectual current that could no longer be
repressed but only channeled. When in 1758 the weight of scandal led the
chancelier to revoke the license for the Encyclopédie by government
decree, Malesherbes got around the difficulty by issuing a tacit permit.
But, more generally, authors or libraires who were seeking to avoid
difficulties and reduce costs or who were speculating on the diminishing
resistance of the authorities bypassed the administrative channel of the
Sceau Public de la Chancellerie; they contented themselves with asking
for a gesture of tolerance whose only guarantee lay in a consensus
between public opinion and government. Among these somewhat sus-
pect new titles, whose character was not concealed from the censor,
manuscripts formed the largest but—as Diderot noted—not the only
group: ““One must distinguish two sorts of titles appearing by tacit per-
mission alone: the first are the work of foreign authors and have already
been published outside the kingdom; the second are the work of authors
living in France and are in manuscript or have been published under
spurious titles.’
Thus, there is a qualitative difference between the two major series of
records relating to applications for printing licenses. In the eighteenth
century itself, an external dividing line was drawn between two types of
literature—a partition that is valuable for the historian, since it indicates
the origin of the defenses put up by a society against its own culture. This
aspect alone would be a sufficient justification for maintaining that divid-
ing line and thus for studying priviléges and tacit permits separately. But
the quantitative analysis of these sources confirms the need for this
methodological approach.’
Between 1723—the opening date for the registers, following a gap
between 1716 and 1723—and 1789, one finds 31,716 works submitted for
a privilège or permission du Sceau. This figure is higher than the one given
by R. Estivals, because he counts only published books, not licenses. The
fact is that libraires or printers would sometimes include several books in
a request for a single privilège or permission du Sceau.* Actually, the same
is true for tacit permits, but there are far fewer works covered by this
104 Seventeenth- and Eighteenth-century Culture

series for the whole century: only 12,610, or just over a third of the
number of books in the first series.
This numerical gap is essentially due to the chronological difference
between the two sources. Even though they dated from the Regency,
tacit permits did not flourish in earnest before the 1750s. The practice did
not become widespread until Malesherbes was appointed to the Direc-
tion Générale de la Librairie in 1751. The first register of tacit permits
(1718-47) lists a total of 713 books over thirty years, that is, a very low
average of 24 a year. The second register (December 1750-March
1760)—which, no less than the first one, is impossible to analyze on a
yearly basis—includes 714 titles for a period of nine years and four
months, that is, 126 a year. This almost decennial average is in itself only
the starting point of a rising trend, since the next two registers reveal a
very rapid increase in the annual average: 156 books a year between
March 1760 and October 1763; 396 between October 1763 and November
1766. From 1767 on, the sources make it possible to calculate annual
figures showing that there were nearly as many requests for tacit permits
as for standard licenses (see chart).
While the two trends converged from the late 1760s on, only the
second trend presents features enabling internal comparisons to be made
from the beginning of the century onward, because of the relative stabil-
ity of the annual figures. The yearly average for the period from 1723 to
1789 was 463 books; the figure works out to 456 before 1750 and 469 from
1750 to 1789. There is no trace here of the discontinuity that characterizes
the registers of tacit permits. Thus, one cannot combine the evidence for
two administrative practices that are not comparable, for one was an
institutionalized tradition, and the other, which was for long time un-
avowed and unavowable, did not flourish before the 1760s.
Moreover, the problem remains of determining if and how, in earlier
decades, the hundreds of titles a year belonging to a category officially
recognized only later were actually printed. In his fourth Mémoire sur les
problèmes la librairie, which can be dated to early 1759, Malesherbes
noted that ‘over the past thirty years the use of tacit permits has become
almost as widespread as that of public licenses.” “Over the past thirty
years”? Since the 1730s? The statement is surprising if one compares it
with the huge disparity in the annual figures revealed by the two series of
licensing records spanning the first half of the century. Yet it is difficult to
reject this statement out of hand, coming as it does from a man who was
ideally placed to know. One can assume both that the statement was
exaggerated’ and that nevertheless, during this period, a considerable
number of books were granted permits so “‘tacit’’ as to leave no written
trace. Indeed, Malesherbes himself implied as much in his fifth Mémoire
sur les problèmes de la librairie, quoted earlier, when, having distin-
guished between tacit permits and simple acts of tolerance of which no
trace remained, he added, “The first tacit permits to be granted were no
Book Licensing and Book Production 105

doubt of this sort; similar permits are sometimes still given, owing to the
lack of well-established principles enabling the censor to regard himself
as immune from reproach. But true tacit permissions differ considerably
from these acts of tolerance or perhaps of connivance.”
Thus, it is probable that until the 1750s a wide variety of illegal
literature—distinct, however, from clandestine literature proper, hunted
down by the royal police—was simply tolerated by the authorities, with-
out our being able to find a written trace of it in the licensing registers.
An example dating from the midcentury shows the extent of juris-
prudential uncertainty surrounding the matter. In 1748, Montesquieu
had L’Esprit des lois printed in Geneva without an author’s name. Its
immediate impact in France indicates how widely it was distributed. Both
the Jesuits, in the Mémoires de Trévoux, and the Jansenists, in the
Nouvelles ecclésiastiques, devoted two articles to it the following year.
The Jansenist journal, which fulminated against the book, concluded its
second article, dated 16 October 1749, by openly accusing the authorities:
“Orders will be given for the executioner to burn the Nouvelles ecclésias-
tiques, whose sole and perpetual aim is to confirm men in the possession
of truths that make both a true Christian and a faithful subject of the king;
and permission will be granted for the sale of a disgraceful work that
teaches men to regard virtue as a useless principle for a monarchy, and all
religions, even the true religion, as political issues, as being solely the
product of climate, etc. May we be allowed to say so: is not one act a
punishment for the other?” In August 1750 the Sorbonne intervened in
its turn to make up post eventum for the official clearance that had not
been requested for the work. The Sorbonne suggested a certain number
of cuts, which Montesquieu refused by appealing to public opinion: “‘All
of Europe has read my book, and everyone agrees that it was impossible
to tell whether I preferred republican government or monarchical gov-
ernment... .’’ Actually, if Malesherbes is to be believed,” Montesquieu’s
book, in its original form, obtained a tacit permit that immediately paved
the way for numerous reprints, none of which bear any traces of the
Sorbonne’s demands.

The very complexity of this example shows that it is hard to assess the
precise bibliographical value of the two series of licensing records de-
scribed above. As comparisons cannot be made with the two most readily
available sources—the Dépôt Légal, which is less exhaustive, or the
alphabetical author catalog of the Bibliothéque Nationale, which does
not list anonymous titles—the simplest procedure is to summarize the
contents of the two registers in order to point out what they do not
contain.
1. The registers of privilèges and permissions du Sceau contain three
sorts of applications. The largest category concerns new manuscripts. But
before the decrees of 1777, which put an end to the need to apply for
106 Seventeenth- and Eighteenth-century Culture

BELLES-LETTRES
1793 books 2728 books 2285 books Re books 821 books 724 books
12 unidentified 72 unidentified 49 unidentified ~ - Miscellanies
100 % ye Almanacs
> Journals
< Humor
Orators

Poetry

Sciences and Arts (Novels)

Grammar
Philology
Belles-lettres
ar Dictionaries
0 nai 1723-27 1750-54 1784-88
Various
rio LESS 00S: = = Specialized arts
e Liberal arts

PUBLIC LICENSES = Agriculture

=
aed -
Politics

Sciences

20

Philosophy

330 books 686 books 742 books


SCIENCES AND ARTS

1082 books 1863 books 2165 books BELLES-LETTRES


55 unidentified 71 unidentified 94 unidentified Miscellanies
' ET Theology i Almanacs
xy Law Eu
Journals
4 History ~ Humor
Orators

\ Sciences and Arts Poetry

(Novels)
Belles-lettres
Grammar
Dictionaries
0 1750-59 1770-74 1780-84
1750-59 1770-74 1780-84
100 %; Specialized arts
Various + =
Mechanical arts
Liberal arts
TACIT PERMITS
PEN
Politics

Sciences
\

Philosophy

SCIENCES AND ARTS


Book Licensing and Book Production 107

100

80

Tacit
permits

Sciences
71] and arts

20
Politics

1770071 7200073 1164 "85" 66" G7 REG

Books published with a privilege


" BOOKS published by tacit permission
DISSEES DE Books published by tacit permission
books “entered via the Chambre
Syndicale de la Librairie”

1720 30 40 50 60 70 80 88
108 Seventeenth- and Eighteenth-century Culture

renewal of priviléges, one also finds applications for such renewals by


authors and printers wishing to prolong their monopoly. These requests
indicate the likelihood that the book was popular and that the libraire
intended to reprint it in the near future. Finally, new editions of previ-
ously published books—as distinct from manuscripts—also had to be
vetted by the royal censors: these instances would involve a libraire
basing his claim on a license that had already been issued for an earlier
edition and had expired or a competitor invoking the original character of
the new edition in order to seek permission to publish an author who was
not yet on his list. The latter sort of applicant could save money by
requesting a permission du Sceau. The registers for the first half of the
century make fairly frequent mention of these reprints (maked by an R),
which were considerably greater in number than applications for renewal
of priviléges. But the records were kept too carelessly—the R markings
actually disappeared in the 1760s—for an accurate count to be made.
Finally, from 1778 onward, the introduction of permissions simples led to
less frequent listings for reprints.
New editions almost always concerned the same categories of books:
Latin classics, seventeenth-century masterpieces (especially plays), the
great legal handbooks, devotional and liturgical works, and what can be
called, for lack of a better term, ‘“‘popular literature’—almanacs and
short novels issued by the Bibliothèque Bleue de Troyes and other
specialized libraires. While the frequency of reprints can lead to double
counts in the general census, they are valuable indicators of book con-
sumption. In this respect, the registers of priviléges not only make it
possible to measure the volume of traditionalist literary output; they also
reveal the areas of consensus in public taste and the sense of a shared
culture, that is, of a common past.
But these records also teach us that even the present could become
“classical.” Time and success could temper the subversive character of a
book first published abroad. Dangerous ideas could become common-
place. In some cases the book in question was granted a license later on
and so found its way onto the very official licensing records. Such was the
case with Voltaire’s Henriade and Siécle de Louis XIV, and the same
happened to the Lettres de Madame de Maintenon, a sort of secret history
of Louis XIV’s reign, which, according to Malesherbes, was first pub-
lished abroad and was allowed to circulate in France by the chief of
police: “For if the authorities had not wanted the book to appear, the
persons most connected with the court would not have provided the
publisher with information. . . . One had to be in the know in order to
make sure that the king would not object to the publication of the book,
in which Louis XIV’s secret marriage, whose existence had until then
been in doubt, was fully described.” In short, already under Fleury’s
premiership, the court was secretly taking its petty revenge for its great
humiliation. But from the 1750s on the book was quite openly mentioned
Book Licensing and Book Production 109

in many guises in applications for licenses—which, moreover, it readily


obtained.
2. Nevertheless, tacit permits were the refuge par excellence of new
titles, since that was their very function. Some tacit permits were also
issued for reprints—generally described in such a way as to give the
impression that the books in question had been totally rewritten—but
instances of this are far less frequent than in the registers for priviléges.
For the most part, tacit permits concerned manuscripts and new books
for which anonymous approval would compromise neither the censor nor
the authorities. That is why these permits sometimes concerned books
that had been removed from the public circuit of priviléges. While the
registers tell us next to nothing about the first half of the century, they are
increasingly informative from Malesherbes onward. Helvetius, Condil-
lac, Mably, Condorcet, and Beaumarchais, for example, all get a men-
tion. These sources are all the more valuable for listing rejected manu-
scripts.
Finally, this paralegal administrative bibliography also gives a chrono-
logical listing of another category of books—those that came from abroad
already printed, did not succeed in avoiding controls, and reached the
Chambre Syndicale de la Librairie in sealed packets. These works are
mentioned in applications for tacit permits between 1767 and 1778;" as
with “domestic”? manuscripts, permission was being sought for their
distribution in France. A separate count was made for these volumes, in
most cases at the end of each register; these figures can therefore be
added to the total of authorized publications.

One can also reverse the question: What sort of publication bypassed
applications for printing permits, whether public or tacit? Basically, three
categories of books.
No doubt the least important one includes a share of provincial pub-
lications. In the eighteenth century, most new books published in the
provinces with official permission seem to be works of local interest—
items commissioned by local government bodies, bishoprics, courts of
justice, academies, or universities. That, at least, is the conclusion drawn
from Madeleine Ventre’s study of the Languedoc,” and her data seem to
correspond to what we know about the rapid decline in provincial pub-
lishing as early as the seventeenth century.
Actually, in the eighteenth century, a considerable number of provin-
cial libraires and printers had taken to dealing directly with the central
government, especially for matters that did not concern official orders
from local authorities. The prevailing trend was toward centralization,
and the appropriate habits were spreading. But the fact that our registers
contain few applications for licenses from the provinces substantiates and
justifies in retrospect the endless complaints by eighteenth-century pro-
vincial libraires against their Parisian colleagues, who exercised a dual
110 Seventeenth- and Eighteenth-century Culture

hegemony over the book market owing to the renewal of their licenses
(which prolonged their monopoly of previously published books) and
owing to the proximity of most authors (which gave them a monopoly of
new titles). The example of the Languedoc shows that the decrees of 1777
gave a small stimulus to provincial publishing without seriously threaten-
ing Parisian ascendancy.
In practice, a Parisian libraire would be deprived of a manuscript only
if the author, fearing even the censorship of tacit permits, preferred the
circuitous route of having his book printed in the provinces or abroad and
deposited at the Chambre Syndicale. But even in this case the work
would be mentioned in the registers of tacit permits under the heading of
“books introduced through the Chambre.” Thus, the Parisian and cen-
tralized character of the sources concerning official control of the book
trade is probably not a major drawback, at least for the eighteenth
century.
The two other categories of missing books pose a far greater problem
for an exhaustive census of eighteenth-century publications. First, some
books were simply tolerated by the police without ever being mentioned
in official documents. ‘I do not know,’’ wrote Malesherbes, who dis-
cussed these publications at length, “what to call these sorts of licenses,
which have become common practice. Strictly speaking, they are only
assurances of impunity.” It was, for example, an authorization of this
kind that Montesquieu’s Paris publisher claims to have obtained—prior
to any tacit permit—from the comte d’Argenson for L’Esprit des lois as
early as the beginning of 1749.”
We have seen that the very graph of the number of requests for tacit
permits is a sufficient indication of the extent of this practice before
Malesherbes took office. But, conversely, the increase in tacit permits
from 1751 onward strongly suggests that the first directeur général suc-
ceeded in establishing as a standard administrative practice the doctrine
he described to the dauphin in 1759, when he asserted that he knew ‘‘only
one way of enforcing prohibitions, namely, to enact as few of them as
possible.” One can also observe that his successors remained faithful to
his doctrine.
However, it is unlikely that simple tolerance disappeared altogether.
Too many anecdotes testify to the contrary, and in 1761 Malesherbes
himself found no other solution than to allow Rousseau’s Emile to
circulate at its own risk, without being able to grant it a tacit permit. But
when one considers the spread of tacit permits from the 1760s onward and
when one observes that Restif de la Bretonne and Mirabeau requested
and obtained them, one can assume that this practice very largely re-
placed the secret acts of tolerance of the first half of the century.
Nevertheless, the Parlements and the clergy continued to threaten
even authorized books with scandal and legal prosecution. They intimi-
dated not only authors but also censors and royal officials. It is their
Book Licensing and Book Production 111

pressure that explains to a great extent the existence of the last category
of books to elude our grasp—those that enjoyed no permission or toler-
ance whatsoever and, whether printed in France or (as was more often
the case) abroad, were quite simply clandestine. Police records and legal
sources for the history of the book trade under the ancien régime" are so
scattered that it is virtually impossible to make a count of such books.
Thus, by definition, they fall outside the scope of the present study.”
All the same, by using the applications for priviléges and tacit permits,
we have been able to count up slightly more than 44,000 book titles. This
is a considerable figure if one compares it, for example, with the 25,000
titles published in Paris during the sixteenth century (according to Lucien
Febvre and Henri-Jean Martin), even if it is true that we are dealing with
requests for publication and not with the published books themselves.
But we have seen how the statistical value of this count is unevenly
distributed: nearly 13,000 licensed books up to 1750, and only a few
hundred tacit permits. Thus, it is above all on the strength of the 30,000
works dating from the second half of the eighteenth century that the
sources studied here accurately reflect their titles and provide a realistic
census of the publications of their period.

Let us imagine for a moment that all these books could be placed in a
library arranged in a temporal sequence—a doubly fictive library, in that
successive reprints would stand alongside never-published manuscripts
and in that no cultured individual in the eighteenth century would have
deprived himself of the major works published abroad and banned in
France. But this imaginary library remains to a great extent, as we have
seen, that of an entire society. It is richer and more representative than
the partial bibliographies compiled late in the century. We have tried to
reconstruct it according to subject by means of chronological samplings.
Works have been classified according to eighteenth-century criteria.
The Bibliothéque Nationale contains a wealth of inventories of eigh-
teenth-century private libraries, in which books are arranged according
to the five major categories in use at the time: (1) theology and religion,
(2) law and jurisprudence, (3) history, (4) sciences and arts, and
(5) belles-lettres. Moreover, bibliographical manuals such as those of
Durcy de Noinville or Cels-Martin provide detailed information on the
guidelines for such inventories, thus making our own classification easier.
The following is the scheme adopted here:"

I. Theology and religion”


A. Scripture, Bible, interpreters of the Bible”
B. Church fathers, conciliar literature”
C. Theology and apologetics
1. Catholic”
2. Non-Catholic
112 Seventeenth- and Eighteenth-century Culture

D. Liturgy and devotion”


II. Law and jurisprudence”
A. Canon law and ecclesiastical law”
B. Civil law”
1. Ancient law
2. Natural and public law
C. Jurisprudence and practice
III. History”
A. Ecclesiastical history”
B. Profane history
1. Ancient
2. Modern (by state)
3. Ancillary sciences (genealogy, numismatics, paleography, etc.)
C. Geography, travel, and cartography
IV. Sciences and arts”
A. Philosophy
1. Ancient
2.LOgic
3. Ethics
4. Metaphysics*
B. Science
1. Physics
2. Mathematics
a) Astronomy
b) Mechanics
c) Algebra, arithmetic, geometry
d) Applied sciences and mathematics
3. Natural sciences
a) Botany
b) Mineralogy
c) Zoology
d) Chemistry
4. Medicine, surgery, pharmacology”
. Political economy”
. Agriculture and agronomy”!
. Liberal arts”
Mechanical arts”
. Specialized arts“
LATE. Various
V. Belles-lettres*
A. Dictionaries”
B. Grammar and philology
Cre Poetry
1. Poetry
2. Drama
Book Licensing and Book Production 113

3. Novels
4. Correspondence”
D. Orators
E. Humor
F. Journals and periodicals
G. Almanacs*
H. Miscellanies

As with all classifications, the rigidity of the scheme above involves


certain difficulties. The first problem is the relation of the title to the work
itself. The tradition of long and detailed titles often yields sufficiently
accurate information about a book’s contents, but occasionally, and in
particular in the first half of the century, a neglectful scribe would omit
part of the title, replacing it by a mere etc. In such cases, we are generally
left without the explanatory subtitle—the semantic equivalent suggested
by the author himself—which was so common in the eighteenth century
and is so valuable for the historian. For example, in 1750 an application
for a privilége was submitted for a work entitled L’Art de vérifier les dates,
etc., which was actually published in Paris that year by the libraire
Desprez. The abbreviation of the title conceals the fact that the work was
a polemical theological treatise, written from a Jansenist and Gallican
point of view and based on a chronology of the errors of various popes,
such as Honorius and Liber. Patouillet’s dictionary classifies it among the
more pernicious works.
Generally speaking, these incomplete titles have been easier to iden-
tify—using the catalogs of the Bibliothéque Nationale—than simply
ambiguous but complete titles such as Promenades d’un solitaire or
Lettres de Monsieur X à Monsieur Y. In all these cases, several hundred
checks have been made, enabling us to establish the virtual absence of
any misleading title aimed at concealing the true contents of a book.
Besides, the fact that the censor was obliged to read the work eliminated
any possibility that such a subterfuge might be effective. However, the
practice does seem to have been commonly applied to prohibited books
circulating clandestinely. But our listings still contain a few ambiguous
titles that we have been unable to identify at the Bibliothéque
Nationale—whether in the author catalog, the catalog of anonymous
titles, or the eighteenth- and early-nineteenth-century bibliographies.”
Consequently, these books, which were either lost or never published,
are impossible to classify. They are slightly more common in the records
of tacit permits than in the registers of public licenses, but never exceed 5
percent.
After these investigations, a final difficulty remains, owing to the
imprecision of bibliographical criteria for a certain number of works. The
title of the book itself can cause hesitation: is Rousseau’s Discours sur les
sciences et les arts (1750) “philosophy” or “‘politics”? At least it clearly
114 Seventeenth- and Eighteenth-century Culture

belongs to category IV. A very small number of titles raise the problem of
whether form or content should be the criterion. For example, should Le
Bonheur, a poem by Helvetius, come under the heading of “‘poetry”’ or
“ethics”? In all these cases we have been guided by eighteenth-century
criteria: the theme for Rousseau’s Discours was suggested by a resident
member of the Classe de Morale of the Académie de Dijon, and most
contemporary libraries classified the work under “ethics.” In the catalog
of Malesherbes’s library, Le Bonheur is listed under ‘ethics and morals,”
an extremely rare exception to the tyranny of nomenclature—a product
of classical aesthetics—which quite often manifested itself by the juxta-
position of a genre label to the title.
The slowness of identification procedures for these books naturally
made it impossible to classify all the 45,000 works mentioned. In order to
eliminate annual fluctuations and attempt to measure the general trend,
we have made three five-year samplings (1723-27, 1750-54, and 1784-
88), each involving some 2,000 works, in the registers of privilèges. The
registers of tacit permits, which can be used from 1750 on, have also been
divided into three periods: the first isa ten-year period (1750-59), so as to
obtain a reasonable number of books; the other two are five-year periods
(1770-74 and 1784-88). All the figures have been converted into percent-
ages to allow for comparisons. In absolute terms these figures represent
highly diverse units—since nothing is more different from one book than
another book—but in relative terms they are quite comparable, if indeed
reckoning in large numbers eliminates similar disparities within each time
period.

The bibliography of the privilége registers of the 1720s reveals the


fundamental importance of the supernatural underpinnings of the social
world—in consonance with the hierarchy of the time. More than a third
of the works were concerned with religion“ in the broad sense, However,
under this general heading, they were very unevenly distributed among
the traditional categories. One finds few commentaries on Scripture, and
even fewer on the church fathers, in an age dominated by the Augustinian
obsession. But it would be wrong to conclude that the Jansenist current
had been eliminated by official fiat. A society cannot enforce an adminis-
trative ban on its own religious sensibility. When examined individually,
the theological and devotional works of this period show the extraordi-
nary penetration of Jansenist ideas, which one must be careful to distin-
guish—as the royal censors did—from Jansenist politics. The Unigenitus
Bull was a law of the realm; all comment on it, even favorable, was
forbidden. Similarly, censors would invariably refuse all titles containing
such words, expressions, and names as missal for laymen, grace, predes-
tination, Port-Royal, knowledge, council, exposition, or even Embrun,
Tencin, Senez, Soanen, Auxerre, Montpellier, and so on. However, one
can observe the authorized development of an abundant popular reli-
Book Licensing and Book Production 115

gious literature of a Jansenist strain, representing more than half the


religious works in our sample.
This enduring tyranny of the sacred was the source for the legal and
jurisprudential regulation of the two realms of human activity: first, the
church, as a temporal organization; second, the civil and political world.
Hence the considerable importance of canon law, ecclesiastical law, and
jurisprudence. The first two categories constituted a model and, as it
were, a guarantee for the rest. Jurisprudence was a matter of publishing,
updating, and adapting legal procedure; it attests to the justifications
gradually elaborated by learned royal jurists with a view to unifying
customs and customaries, defining status and rank, and building the
foundations of political society. These books were by definition weightier
than small devotional works. They were also more esoteric and, thus, far
less numerous, but they continued to embody in exemplary fashion the
monarchic absolutist civilization.
The following categories involve an ultimately less fundamental area
of knowledge, since it was in a certain sense of secondary importance.
This area was the ornament and almost the pleasure of life; it did not
cover the rules of life. But it already represented a majority of published
or reprinted books. Most of the history books in our sample deal with
profane history; only a quarter of them are devoted to ecclesiastical
history. The accent is on the modern period, but the outlook is both quite
international and almost exclusively European (France, England, Spain,
Italy, Russia, Poland, Sweden, and so forth). The extra-European world
was the province of travel accounts, which transformed the view of the
present not through time but through space. But geographical curiosity
was also linked to history by a kinship more secret than the fraternity of
faraway time and faraway space: travel unveiled the past contained in the
present by revealing the existence of infant mankind in other parts of the
human world. As Servan explained in 1781: “It has always seemed to me
that the discovery of America has contributed not insignificantly to the
progress of ethics. . . . For previously we knew nothing about the infancy
of our species.” Thus did history and geography begin their arduous
cohabitation. The interest they aroused goes back a long way. By the
early eighteenth century, these two fields accounted for a considerable
number of works.
But profane knowledge was essentially constituted by “sciences and
arts,” a classic label designating every intellectual activity connected with
knowledge and beauty. Even before Boileau, the parallelism between
arts and sciences—one of the fundamental theses of French classicism—
had been declared to be the logical consequence of their common origin:
reason. ‘‘The common feature of arts and sciences,’’ wrote Le Bossu at
the beginning of his treatise on epic poetry (1675), “is that they are
founded on reason, and that one must be led to them by the wisdom that
nature has given to us.” Thus, belles-lettres themselves were only an
116 Seventeenth- and Eighteenth-century Culture

especially important area of sciences and arts: it was their volume and
above all their degree of ‘“‘nobility,” more than their special character,
that justified their being classified separately.
The privilége registers of 1723-27 point to the great social importance
of sciences, arts, and belles-lettres, all of which together represent nearly
half the total. Within the “sciences and arts” category, ethics and
metaphysics predominate in philosophy, and medicine dominates the
sciences. The “‘liberal arts,’’ consisting mainly of music, have a crushing
numerical superiority over agriculture and mechanical arts, which retain
the stigma of manual activities. Apart from the abrégés de paix perpetu-
elle, almost all “political” books consisted of manuals of commercial
practice.
In belles-lettres, classical aestheticism still prevailed: many grammar
and philology books perpetuated and taught the science and rules of
noble language. “Poetry,” in the wide sense of the word, was by far the
largest category, bearing witness to the persistence of the main classical
genres: verse, drama, and correspondence. There were also many books
in Greek and even more in Latin. But one can already note the relative
importance of novels, which were more common than verse compositions
and tragedies.
Such, then, were the broad bibliographical outlines of a great tradi-
tional classical culture in the early eighteenth century and the balance
between its various elements. One can proceed to a comparison with the
two samplings for the 1750s and the late eighteenth century based on the
privilége registers alone.

The later samples reveal continuities as striking as the signs of change.


One finds a comparable mass of works in law, history, and belles-lettres,
indicating a persistence throughout the century of a classic style of writing
and of a social demand for such writing. The “law” category expanded to
take in the developments of jurisprudence, attributable both to the
increase in individual litigations and procedural ‘‘briefs”’ and to the great
national effort to rationalize juridical practice. In the course of the
century, history lost part of its substance. Books on ecclesiastical history
fell from 25 percent in 1724-28 to 15 percent in the 1750s and 11 percent in
the 1780s. But history remained faithful to the general principles visible
at the very outset: a common past made Europe an intelligible and
worthy object of historical study, while the rest of the world was in most
cases dealt with by ahistorical travel accounts. As the century wore on,
the increasing number of books devoted to early France, particularly to
the Carolingian period, pointed to the rise of an antiabsolutistic national
consciousness, which originated as a nobiliary nostalgia for Frankish
assemblies and turned into a lesson in constitutional government.
‘“Belles-lettres” remained an extraordinarily stable category, not only
with respect to the others, but also in terms of the distribution of works
Book Licensing and Book Production 117

according to classical literary genres: orators, poetry, theater, novels,


grammar, and so forth. The aesthetic formalization of classicism survived
the century unscathed, Diderot and Rousseau notwithstanding; it had
become too rooted in public taste for it not to enjoy a long lease of life.
Classical aestheticism revealed the persistence of the style noble in which
revolutionary eloquence was to flourish before dying out.
Yet there were two new elements in late-eighteenth-century belles-
lettres. One was the proliferation of dictionaries, a well-known aspect of
the encyclopedist fanaticism of the Enlightenment and of the zealous
efforts to classify and fence off the realms of knowledge. The other
element is more surprising and can be discerned only by reading the titles
themselves: it was the virtual disappearance during the 1780s of Latin
classics, which were still quite common in midcentury.“ It is hard to find a
cultural reason for this change in an age still dominated by neoclassical
aestheticism. More probably we are dealing with an administrative side
effect of the law of 1777 concerning the renewal of priviléges and with the
effects of the spread of the new permissions simples.
But although, through its books, the whole of eighteenth-century
French culture gives us the impression of continuity and of well-
established social habits, two bibliographical categories exchanged their
respective dimensions between 1724 and 1789: “‘theology” and “‘sciences
and arts.” This dual trend seems to have been very gradual and fairly
steady. It does not bear out the hypothesis of Daniel Mornet, for whom
the great battle against religion was confined to the first half century.”
The trend we have observed was already established by the middle of the
century and, if anything, it accelerated until the end of the ancien ré-
gime—an interesting sign of the rhythm of desacralization in the eigh-
teenth-century world.
In actual fact, the main types of religious books to disappear were
liturgical and devotional works. Catholic theology and apologetics were
characterized until the end of the century either by a Jansenist approach
or by a traditionalism that in the 1780s seems to have been infected with
“philosophy.”
‘“Philosophically demonstrated” Christian truths were adapted to pre-
vailing fashions. Moreover, such works were by then hardly ever written
in Latin. But the relative scarcity of liturgical books and devotional tracts
commissioned by dioceses is perhaps an indication of a slackening of
public demand. Does this not confirm the existence of an urban anticler-
icalism that was mentioned by so many eighteenth-century authors since
the great crisis over the vingtième tax and, thus, seems to have developed
well before the Civil Constitution of the clergy voted in 1790? The
Jansenist, or even Richerist, roots of this movement no doubt explain a
contrario the persistence of a sizable theological culture.
But, by the end of the century, ‘‘sciences and arts” had become the
largest category of books produced. The two terms were all the more
118 Seventeenth- and Eighteenth-century Culture

significant for having taken on new meanings: they no longer designated,


as in the seventeenth century, a harmony between social activities and
divine order—a harmony variously expressed by the notions of truth and
beauty or nature and reason. In 1750, in the very middle of the century,
Rousseau brutally divorced society from nature. Sciences and arts lost
their innocence; henceforth they were damned or blessed. They had
come to resemble the neoclassical façades built at the end of the century
and lost in nature à l’anglaise—an expression used to characterize na-
ture’s alien and secret essence. Sciences and arts had become the prime
agents of history and mankind, or, as Condorcet was to say, “of the
progress of the human mind.” The controversy over sciences and arts, as
well as the naturally connected debate about luxury, was thus at the very
heart of Enlightenment society, for which these debates symbolized the
split in its own self-awareness. Hence the extraordinary impact of Rous-
seau’s first discourse, which, from the literary point of view, was more
commonplace than the intuition that informed it.
The privilége registers bear witness to the increasing number of books
in the “‘sciences and arts’ category—the messengers par excellence of a
more civilized, richer, and more human world. The percentage of these
books doubled during the course of the century. These figures represent,
as it were, a vast collective compensation for the disappearance of de-
votional works. However, the distribution within the ‘‘sciences and arts”
category remained fairly stable. One finds the same percentages for
“philosophy,” although “‘ethics” increased with respect to ‘“‘metaphys-
ics.”” The percentages for scientific books remained comparable: the
major heading was “medicine,” dominated by the obsession with
venereal disease—an obsession whose psychological hold was fully re-
vealed here. Books about agriculture and especially about politics were
increasingly numerous. The nature of political works also changed, shift-
ing from economics to politics proper. One can see by their titles the
effect of the crisis of the last years of the ancien régime.
Among the “‘liberal arts,” there was a proportionate increase until the
1750s in the number of books on painting, architecture, and the art of
warfare. At the end of the century, the decrease in the total percentage
for this category was due mainly to the decline in music books, which
were very common between 1750 and 1754 and very scarce between 1780
and 1784. It may also be that, after the decrees of 1777, the regulations
concerning public licenses became laxer for works as innocuous as music
books.
The most surprising feature in this classification of arts and sciences is
the consistently negligible share of books about the ‘‘mechanical arts’”’-—
arts that the eighteenth century often prided itself on having restored to
their proper rank. But our bibliographical statistics point perhaps less to a
falsehood than to a confusion. If Enlightenment society did indeed
attempt to revive the mechanical arts, it was not so much through a study
Book Licensing and Book Production 119

of the manual tradition of a preindustrial world that did not have much to
teach it. Instead, the Enlightenment sought to raise these arts to the level
of the style noble and to include them—in the name of universality—in a
hedonistic utopia. The cities of happiness depicted by Ledoux are
perhaps more significant than the famous plates of the Encyclopédie.

But, in contrast to this vast traditional output, what do the tacit


permits tell us?
Comparisons can be made only for the second half of the century, but
the three soundings carried out in the tacit-permit records are propor-
tionately larger than the samplings of public licenses for the same period.
The tacit-permit samples indicate that by the 1750s the share of religious
and legal works dropped to about 2 to 3 percent of the total—a figure that
remained unchanged until 1788. This confirms both the traditionalist
character of books in these categories and the particular purpose of tacit
permits. One finds an artificially larger share of intellectual innovation
and of what we would today call “fashion,” as opposed to intellectual and
social habits.*
Why, then, does one find religious works at all, even in very small
numbers? Liturgical literature disappeared, and devotional works be-
came very scarce. ‘““Theology”’ was the dominant category and was gener-
ally contending with the “prevailing errors” of the age and with the
philosophes. The numerous samplings carried out among these books—
when they exist in the Bibliothèque Nationale—generally reveal once
again the influence and vocabulary of Jansenism. For example, a book
entitled Pensées morales adaptées aux figures de l’Ancien Testament qui
représentent Jésus-Christ, which was granted a tacit permit in 1788, turns
out to have been written by an anonymous author in a rigorist style
evocative of the great period of the solitaires. The violence of his anti-
philosophical polemic is also reminiscent of the Nouvelles ecclésiastiques.
But apart from these religious and legal works, too scarce to warrant
anything but individual comment, tacit permits almost exclusively con-
cerned history, sciences and arts, and belles-lettres. The boundaries of
these fields are more elastic than in the privilége registers, not only
because there are greater variations in the percentages obtained for each
major category, but especially because the internal structure of each field
seems to have varied considerably from one sampling to another and to
have been more dependent on time—that is, on fashions. Within an
enduring pattern of great classical aesthetics, one can read the exagger-
ated signs of short-term variations and of longer-term changes.
The main feature of these trends is already familiar: the rapid increase
in the number of books about sciences and arts, up from 25.6 percent in
the 1750s to more than 40 percent in the 1780s. The prerevolutionary
character of this trend is evidenced by the specific increase in the number
of political works, which by the end of the century represented more than
120 Seventeenth- and Eighteenth-century Culture

half the titles in this category. Naturally, we are dealing with the crisis of
the ancien régime—that is, the short-term crisis, for if one looks at the
1784-88 period year by year, one finds a fairly constant proportion of
books in the sciences and arts, while the percentage of these books that
were devoted to politics rose sharply from 1787 onward, specifically after
the Assembly of Notables in February, which triggered the spate of
revolutionary literature. A similar but opposite phenomenon no doubt
explains the relatively shrinking volume of political books between 1770
and 1774 in comparison with the 1750s: the early 1770s were the period of
the Triumvirate, the last great attempt at neoabsolutist government by a
king of France. At any rate, the percentage for “politics” rose suddenly in
1774 with the death of Louis XV, Turgot’s liberalization, and the return
of the Parlements. Thus, the registers of tacit permits not only confirm
the exuberance of science and art so deeply characteristic of the Enlight-
enment; they also record the shorter-term variations of this phe-
nomenon, which reflect the turning points charted by traditional history.
Hence the greater instability and the more marked character of per-
centages within the category. The most revealing perspective is obtained
by adding philosophical and political works together; their proportionate
volumes vary, but they have many internal connections in that the philo-
sophical category is dominated by the moral works of contemporary
philosophers. Indeed, it is not the twentieth century but precisely the
eighteenth century—with Abbé Baudeau—that invented the expression
“moral and political sciences.”’ One can thus measure their numerical
superiority over conventional sciences, a superiority that does not appear
in the public priviléges. The tacit permits forcefully demonstrate the
upsurge of interest in a wide range of social issues. However, certain
features are common to both sets of sources: the predominance of medi-
cine in the scientific category,“ the relative importance of agronomy, and
the infinitesimal percentage of works dealing with mechanical arts.
Toward the middle and end of the century, history seems to have
represented a percentage of tacit permits very close to its share of
privilèges. It is hard to explain the intervening drop during the 1770-74
period. However, an examination of the titles clearly shows that the
historical works submitted for tacit approval were of a slightly different
character from those listed in the privilége records: alongside genuine
historical narratives, one can observe an increase in historical
documentary literature, especially in more or less authentic memoirs of
more or less famous figures of the past. The term memoirs is in itself a
sufficient indication of the ambiguity and novelistic overtones of this
genre. During the eighteenth century, which novelized history and his-
toricized the novel, heated discussions took place about the similarity
between the two genres.
Moreover, the novel, a perennial suspect for classical aesthetics, was
taking shelter behind history. For this was one of the means it had
Book Licensing and Book Production 121

found—along with the artificial exoticism of Persian or Turkish travel and


the device of ‘‘letters’’—for freeing itself from the burden of poetry and
for setting out in its quiet way to describe the real world. Titles like
Mémoires de Monsieur X or Histoire de Mademoiselle Y, which were so
common, often betray a move away from pastora! toward realism and
from collective idealization to particular truth. Such was the wish ex-
pressed by Diderot in his éloge of Richardson: “This author does not
make blood drip from the walls; he does not expose you to the danger of
being devoured by savages; he does not shut himself up in clandestine
dens of iniquity; he never gets lost in fairyland. His setting is the world in
which we live. His drama is rooted in truth; his characters are as real as
can be.” There is no better way of saying that, in the eighteenth century, a
certain type of English-style novel destroyed the aesthetic formalization
of classicism.
Thus, it is not surprising that tacit permits were a shelter par excellence
for the novel. The statistics point unequivocally to the invasion of the
“poetry” section by the novel. Novels represent between 25 and 50
percent of belles-lettres titles as against 15, 13, and again 15 percent in the
three soundings in the privilège registers.* Admittedly, the share of
belles-lettres declined among tacit permits, owing to the sharp drop in
ancient and modern classics and, above all, to the expansion of sciences
and arts. But the internal proportion of novels remained very high.
These figures prompted us to undertake a comparison for the periods
1740-45 and 1750-55 of the novels recorded in the licensing registers and
those listed by Daniel Mornet in his classic edition of La Nouvelle
Héloise.* We have had to rearrange in chronological order the works
classified by Mornet according to genre. But the comparison remains
shaky, since Mornet’s bibliography lists novels that were actually pub-
lished or reissued (and of which copies are still extant), whether printed
in France or abroad. In contrast, applications for printing licenses con-
cerned only the novels for which permission was sought in Paris from the
chancelier. Some of these titles were reissues, but they also included, in
the mass of manuscripts, a certain number of novels that were simply
planned or that may possibly have been lost. Certain of these books can
be identified only tentatively or with difficulty.” Moreover, there is an
irreducible time lag between the two sources being compared, because of
the variable delay between the request for permission to print and actual
publication. Nevertheless, the figures seem interesting enough to warrant
inclusion here:

Manuscripts
Priviléges Tacit Permits Mornet
1741-45 74 18 205
1751-55 123 195 199
122 Seventeenth- and Eighteenth-century Culture

Thus, the novels listed in the manuscript records of the royal


bureaucracy were less numerous than those in Mornet’s bibliography in
the 1740s and considerably more numerous during the 1750s. This con-
trast is primarily due to the rise in tacit permits, whose representative
character is therefore demonstrated for the post-1751 period. Many—but
naturally not all—of the novels listed by Mornet as having been published
abroad are to be found in the requests for tacit permits. Conversely,
Mornet’s list omits many titles of novels mentioned in the official records.
But quite apart from this comparison—which will have to be examined
in a less general perspective than that of the present study“—the com-
parison between the two series of licensing records underscores the
considerable proportion of novels among books submitted for tacit
approval. This relative importance is all the more striking as, in midcen-
tury, the number of books per year submitted for tacit permits was three
times lower than the figure for privilèges. Thus, in the 1750s, France does
not seem to have been exclusively obsessed, as Voltaire claimed in a
well-known passage, with the debate about grain. One can only conclude
that the novel must have been a much-demanded commodity, since it
seems to have monopolized the “‘belles-lettres” category. The origins of
this phenomenon are more remote, and must also be sought in the long
aesthetic and literary quarrel chronicled by Georges May. But the tacit
permits of the second half of the century bear witness to what was now the
dominant role of the novel in literary innovation.

The soundings we have carrried out confirm the basic features of the
two sources—features emphasized in all contemporary accounts: without
tacit permits, the book-licensing records of the ancien régime would give
no indication of the fascination exercised by the novel over French
society at large. But, without the public licenses, these records would
obliterate all trace of the Latin culture and traditional aesthetics on which
every eighteenth-century generation—even that of the Revolution—was
brought up.
These two radically divergent series are valuable for the historian. Not
only do they help to mark the boundary between innovation and tradi-
tion, but, by assembling at the outset the conditions for their contradic-
tory development, they appear as complementary at the end of the
process. The proportions differ, but the trends are identical.
The persistence of legal works indicates the vitality of a political
society engendered by the great royal jurists. This continuity points both
to the proliferation of private interests and to the yearning for the
abstract and univocal arbitration of legal codes. The persistence of books
on history is as ambiguous to interpret as history itself, which in the
eighteenth century was, all at the same time, a conventional form of
recitation, an undiscriminating form of curiosity about time and space,
and an awakening of national consciousness. The Age of Enlightenment
Book Licensing and Book Production 123

preserved the traditional balance between history as rhetoric, as knowl-


edge, and as justification.
The bookshelves of our imaginary libraries also reveal quite clearly the
continuing importance of belles-lettres and the survival of the major
genres. The eighteenth-century world was still that of the grammarians
and normative critics of the French classicist period. The culture of
antiquity, oratory, poetry, and theater retained its privileged status and
its audience. It is through the novel that Beauty sometimes took on a new
form, freed from the constraints of le grand style. But novels often
preserved the appearances of that style by disguising themselves under
the traditional headings of “‘pastoral,”’ ‘“‘morals,”’ “travel,” or “history.”
Such devices were no doubt less a precaution than an homage of sorts to
prevailing public taste. This general pattern is not without significance, in
that the quest for what the Revolution owed to Enlightenment philoso-
phy still goes on. Perhaps it would also be interesting to examine what the
Revolution owed to classical rhetoric: Robespierre and Saint-Just wrote
verse before their speechmaking days.
Finally, throughout the century, there is the large-scale decline of
religious books coupled with the rise of ‘“‘sciences and arts”; the varia-
tions in this trend are visible indications of shifts in collective interests and
in receptiveness to new ideas. At the most general and unsophisticated
level of analysis, this trend expresses the well-known phenomenon ex-
perienced by contemporaries as the attempt by the philosophes to elimi-
nate the supernatural from the human world. It is no coincidence if so
many eighteenth-century texts speak of “‘sciences and arts” and “‘mores
and religion” as opposing categories. The desacralization of French
society and culture was expressed through the old unifying concept of
classical nomenclature.
But sciences and arts were not only the instruments of secularization.
Well before the spectacular successes of industrial efficiency, they
already appeared to be vested with the attributes later assigned to them
by optimistic liberalism and its Marxist inversion: sciences and arts were
no longer ornaments or even mere forms of knowledge, but mankind’s
specific implements, the equipment that would make its adventure a
success. The great idea that presided over the rationality of the modern
world thus originated in classical culture from the very accumulation of
reflections on sciences and arts.
The fact that this idea emerged well before the process of industrial
transformation no doubt explains the specific features that it displayed in
French history: the weakness of technological research, the relatively
slow development of science itself, the persistence of classical modes of
thought, the predominance of social welfare as an intellectual preoccupa-
tion. Sciences and arts were no longer a joint study of Truth and Beauty
as concepts spontaneously attributed to the eternal order of the world.
Sciences and arts already expressed the awareness of a dichotomy and of
124 Seventeenth- and Eighteenth-century Culture

a history—not so much the history of man’s relationship to nature as the


history of human knowledge, the progress of which was both recognized
as a fact and transfigured as a value.
In this advancement of learning, ethics and politics carved out the
lion’s share. The two fields were not only concerned with technical
observation and with correcting ‘‘abuses”’; they also aimed at reconstruct-
ing the entire polity. A vast social upsurge expressed itself through the
twofold language of experience and vision. The most innovative currents
of Enlightenment thought espoused both an ambition and a utopia: the
ambition of power and the utopia of happiness. The Enlightenment was a
conquest, an enlargement of human knowledge, but at the same time a
clandestine effort to forestall the dangers of a now open-ended history. It
bequeathed to modern France a pair of alternatives that for the past two
centuries have never ceased to infuse passion into her cultural and
political history.
Two Historical Legitimations of
Eighteenth-Century French Society:
Mably and Boulainvilliers

The eighteenth century asked two questions of history: What is a civiliza-


tion? What is a nation?
The two questions were independent of one another and never over-
lapped. The first was connected with the strong feeling of progress that
characterized the century and with the development of a linear scheme of
human history. ‘‘Civilization” was seen as the peak of moral, literary, and
artistic perfection, its model provided by Greek and Latin antiquity. The
Europe of the Enlightenment was a new illustration of the benefits of
such perfection. Between Voltaire and Condorcet, civilization even
ceased to be thought of as a cyclical embodiment of the model of antiquity
and came to be regarded simply as the more advanced phase of the
“progress of the human mind.” All societies had to go through civiliza-
tion at some point. Civilization was an ideal for societies that had not yet
attained it and a blessing for those that were flourishing in it. In any case,
history—now that its direction was clear—insured that all societies, soon-
er or later, would experience civilization.
The second question had nothing to do with this prophetic rational-
ism—at least not until the French Revolution. On the contrary, it origin-
ated in the uneasy relationship of French society with absolutism. It is
therefore understandable that the first signs of preoccupation with the
question can be discerned during the wars of religion and the Fronde, that
is, during the two earlier crises of absolutism. A civil society that was
asking itself “What is the nation?” was expressing a sense of autonomy
with respect to government. This question also involved the notion of
rights: if a kingdom was a gathering of subjects, a nation was a community

First published as “Deux légitimations historiques de la société française au XVIIIe siècle:


Mably et Boulainvilliers,” Annales: économies, sociétés, civilisations, May-June 1979,
438-50. I thank Mona Ozouf, coauthor of this essay, for allowing me to reprint it here.

125
126 Seventeenth- and Eighteenth-century Culture

of citizens. A kingdom was a property; a nation was a contract. The


function of history was to rediscover the clauses of that contract, to assert
its imprescriptibility, to insure that it was being carried out, or to de-
nounce its abandonment. Acting for, and on behalf of, defaulting Provi-
dence, history was entrusted with the broad task of verifying titles: those
of the king and those of the nation. History was either the repository of
the original contract or it was merely a chronology of despotism—another
way of saying that the nation embodied freedom.
But just what was that nation? That was the core of the problem. For
although French society stridently displayed very early in the century—
by 1715—its determination to transform its relationship with royal au-
thority, it was unable to think of itself as a political community.
French society had been not just tamed but truly shattered by Louis
XIV’s tyranny, whose memory it never exorcised. For intellectual, poli-
tical, and fiscal reasons, and by a variety of means—which were based on
different principles but converged in their results—the Grand Roi mixed
all ranks of society. Through wholesale servitude, the sale or renegotia-
tion of titles and privileges, and arbitrary bureaucratic promotions, the
nobility was robbed less of its role than of its very essence. But, while the
absolute monarchy destroyed the principles of the society of orders, it
preserved and even ‘‘castified” the appearances of this society. The
inflexible court protocol actually dissimulated a state and a society that
had ceased to share a common legitimacy.
Moreover, the intellectual conditions for developing a new legitimacy
had changed. By the early eighteenth century, politics and society had
been included in the sphere of critical thought. The providentialist and
organicist justification of social hierarchy was no longer sufficient to
rebuild a consensus. This hierarchy had ceased to be taken for granted.
Rather, it had to be made compatible with natural equality among men.
It had to reveal and express the social contract, and no longer a divinely
ordained plan. Lastly, it had at all costs to lay the basis for participation in
government, that is, for a rudimentary form of citizenship.
These complex imperatives, at once social, psychological, and cultu-
ral, could be justified by arguments drawn from the boundless resources
of French history. The history of France had replaced God as the guar-
dian ofthe original contract, of the rights of the French, and ofthe secrets
of the social compact. It was now the supreme legitimating authority,
whose task was to answer the question, What is a nation? The reference
to history served as the basis for political theory, for claims to citizenship,
and for a discourse on inequality. It is striking to see history being
invoked in the same terms by men as different as Boulainvilliers and
Mably, so often contrasted as the exponent of nobiliary ideology versus
the exponent of democratic ideology. We should like to show in this essay
that their constructs are contradictory but comparable, having been
obtained by the handling of identical historical data and identical concep-
tual building blocks. Through Boulainvilliers and Mably, one discovers
Mably and Boulainvilliers 127

that in the eighteenth century the same history could be written from
contradictory points of view—all it took was the demands of immediate
circumstance for that history to become overladen with unexpected signi-
ficance.
As observers of their own times, Boulainvilliers and Mably were
struck by the debasement of political freedom. What they regarded as the
most decisive development of their age was, however, stubbornly over-
looked by men around them whose task was to understand the present.
The intendants, for example, were supposed to describe conditions in
France, but their reports, which were filled with references to royal
rights, were silent about the rights of the French. “They do not conceive,
or refuse to conceive, of any principle of government other than the
despotism of a prince and his ministers,! wrote Boulainvilliers indig-
nantly, having resolved there and then to condemn the authorities. As for
Mably, he was distressed to read a history of France by Pére Daniel:
“From Clovis to our times,’’ Père Daniel led his reader on “‘as if he were
dealing with the very same monarchy.’” Only those who took the present
for granted could display such peace of mind; but a present experienced
in such a manner was, of necessity, unintelligible. Only dissatisfaction
with the present could give meaning to the past.
For both Boulainvilliers and Mably, the present in all its features
bespoke the abuses of despotic kingship, while contemporary observers
did not even seem aware of the fact. This twofold diagnosis lay at the
heart of the two authors’ pursuit of history, which they regarded as a
necessity. So similar was their need for history that Mably had to resist
those of his friends who wanted him to title his observations Histoire de
notre gouvernement. Histoire de l’Ancien Gouvernement de la France is
what Boulainvilliers had called his great work a half century earlier—a
telling sign that the two men shared an identical belief in history as a
meditation on the origins of power and as a remedy for fascination with
the present.
To confess to such beliefs was obviously to break with the providential-
ist view of history. Boulainvilliers did so more forcefully than Mably,
because for him the model of Bossuet, however repulsive (“one of the
most shameful demonstrations of the indignity of our time’’),*’ was also
closer and more constraining. Boulainvilliers still felt it necessary, before
undertaking a historical project, to state that God may have given man
natural wisdom without controlling the use man would make of it. “He
has abandoned the world to our contest”: this strong statement, which
emancipated history from metaphysics, opens the A brégé d’ Histoire uni-
verselle, written around 1700 for his children. Fifty years later, Mably did
not regard this laissez-passer—this laissez-penser—as indispensable,
even as a rhetorical precaution.
There was still another kind of history from which the very logic of
their undertaking diverted them: the history of princes written for
princes. Nevertheless, in keeping with tradition, Boulainvilliers wrote
128 Seventeenth- and Eighteenth-century Culture

history primers for the duke of Burgundy and Mably for the infante of
Parma. But neither author remained faithful to the content or tone of this
type of history, designed for a traditional purpose. Erudite minutiae
concerning ‘“‘idle and lazy kings who contributed nothing to human
happiness’’* seemed superfluous to both. There should be no trace of
sycophancy in a work aimed at being thoroughly critical. As a revenge on
despotism, history could bring monarchs before its tribunal. Mézeray and
Pére Daniel, for their part, had been gripped by panic when popes and
kings appeared on the stage. But their pusillanimity was precisely a
telling sign of what their work was worth.
Neither Mably nor Boulainvilliers wanted to break with erudite his-
tory—another variety of history for which tradition provided a model.
Following the teachings of Bayle, they still regarded history as a con-
tinuing effort to correct errors. But this demanding form of history,
requiring vast quantities of archives and preoccupied with checking facts,
was wholly dependent on the present. It was the present that guided the
historian in the necessary selection of events from the abundant stores of
the past. It was in the name of the present that the two authors disdained
the “‘succession of feats of arms and wars,” “quirks,” and “‘prejudice,” in
order to focus on the crucial issues: the source of authority, the extent and
limits of power, the rules of government in France, and, lastly, the
“principles” that determined the entire sequence of events leading up to
the detestable present.
For both Boulainvilliers and Mably recounted a process of deteriora-
tion. Theirs was a history of how ‘“‘self-evident principles” were worn
down, adulterated, and transformed beyond recognition. The history of
France was a story of decay. Boulainvilliers made the point in his empiri-
cal and realistic way by excusing himself—when, for example, he had to
discuss “‘the most unpleasant reign in the history of France,’ that of
Charles VI—for having to dwell so often on the disgrace of the state.
Mably described the same situation with the more tragic accents of a man
who viewed all of history as a decline. “History is an almost unbroken
succession of miseries, disasters, and calamities.’’’ What was there to be
gained from such a black tale? Mably and Boulainvilliers shared a some-
what intermittent belief in the pedagogic value of misfortune.
Yet the true value of history did not lie in these negative lessons, but in
giving prominence to a continuous transformation, in which the past was
constantly present in the present. Mably and Boulainvilliers—unlike
Voltaire—firmly believed in history as a process of perpetual change.
Neither thought it possible to divide the flow of time into periods that
would constitute strongly individualized and relatively autonomous
wholes. Thus, their history, which linked interminable transformations to
one another and displayed the ubiquitous role of differences, was both a
war machine against centralizing government and a remedy for the
doctrines of centralization and fixity. Its peculiar virtue was that it en-
abled the historian, thanks to the well-established connections between
Mably and Boulainvilliers 129

successive transformations, to work his way backward to a point of


origin.
Hence, for Mably and Boulainvilliers alike, the very essence of the
historian’s activity was to provide the French with a description of “their
old government.” They were counting on the powerful emotional signifi-
cance of this return to origins to counteract the superficial notion—so
widespread among the French—according to which the present was
mistakenly credited with the solidity of what had always existed. They
wanted to cure the French of the frivolity that prevented them from
concentrating on “ideas about government" and to combat their com-
patriots’ prodigious ability to forget. Indeed, the French were guilty only
of having forgotten. The severity displayed by Mably and Boulainvilliers
toward a nation inconsiderate enough to have forgotten that “it once had
a Charlemagne’’—their positive and shared hero—was thus tempered by
hope, for forgetfulness was not quite the same thing as ignorance. Even if
stifled, the “old notions of society and order”’ did not have to be instilled
so much as revived. History was a reminiscence. Thus, it was hardly
necessary to offer the French a new interpretation of their past. It was
enough to show them their past for their memory to be reawakened.
After so many “memoirs for a perfect history,” the task was to write
histories for a perfect memory.
It is easier to understand now the role played within their interpreta-
tion of history by the flattering description of primitive barbarity, a very
strange attitude on the part of men so obviously attached to the contem-
porary values of the Enlightenment. How could the unpolished customs
of a tribe of fully armed roughnecks roaming the Germanic forests
constitute the breeding ground for liberty and equality? Why did Mably
and Boulainvilliers, like their ‘“Germanist’’ contemporaries,’ so readily
equate barbarism with energy and regard savagery as a portent of inde-
pendence? What was their purpose in describing cruelty as the very mark
of a ‘‘proud soul”? One can understand such sympathy only in terms of
their need to evaluate contemporary abuses. Without going back to the
primitive savagery of the fledgling nation, no one could judge the present
for what it was or be fully aware of the “gaps” between French society
and its historical traditions. The unsophisticated comradeship of the
Franks and their jealous attachment to their independence helped to
highlight the illegitimate growth of despotism and its retinue of servility.
The concept of origins, which led to a lesson in comparative history, was
thus a governing concept.
But it did not imply any sanctification of origins. Neither Mably nor
Boulainvilliers considered a nation’s first laws to be its fundamental laws.
No one, said Mably, is so ignorant as to confuse the former with the
latter. And he demystified the Golden Age. Mably’s “early Greeks” did
not live bedecked by myrtle wreaths in pastoral peace. They, too, like the
Franks, advanced in arms, with determination. The tale of origins was
thus curiously tempered by realism. This approach is more explicit in the
130 Seventeenth- and Eighteenth-century Culture

work of Boulainvilliers, who, feeling a greater need than Mably did to


legitimize the society of orders, was obliged to admit that violence was
indeed the root cause of social distinctions. Without a doubt, society
originated in viciousness. The Angouléme manuscript notes that the
nobility ‘is hateful, so to speak, because of the fearsome violence that
desolated the earth.”’ But this violence in no way justified going beyond
its assertion to seek for an innocent and more primitive starting point.
The feeling that retrogressive historical analysis had to stop somewhere
certainly determined Boulainvilliers to draw the line where he did. But he
was also guided by a sense of realism free from any illusion: “The
antiquity of origins thus gave way, with reason, to the superior force of
conquest. . . .”"' That is why the strenuous efforts of certain historians to
camouflage the Capetian usurpation seem to Boulainvilliers to be totally
absurd: was a 700-year-old tenure such a mediocre claim to legitimacy?”
To temper the surprise one feels at seeing a man so representative of
the ‘early Enlightenment” blithely equate law and fact, one ought to
remember that he was an assiduous reader of Spinoza. Familiarity with a
system in which God’s right is equated with his power and in which every
individual enjoys rights exactly proportionate to his might undoubtedly
taught Boulainvilliers not to put right on a par with ideal ends. Like the
Spinozist polity, Boulainvilliers’ society was the outcome of an interplay
of forces. Historical description was for him only the expression of an
empirical situation, and it excluded axiology altogether. Boulainvilliers’s
realism expressed itself most fully in the Angoulême manuscript: ‘There
is in fact no more authentic nobility than that acquired by right of
conquest, just as there is no greater distinction among men than that
between victor and vanquished... .”
Mably, on the other hand, was not prepared to grant any right to the
victor. While he agreed about the state of the French before the con-
quest, he refused the crude statement that seemed to him to summarize
the history of origins as writen by Boulainvilliers: ‘Every Frenchman was
a nobleman, and every Gaul was a commoner.” But Mably was no more
convinced than Boulainvilliers that one could proceed backward in his-
tory indefinitely. One could never move from history back to nature. The
“most ancient monuments of history” represented already decayed
societies. Nevertheless, they had to be dwelt on.
But the purpose of such attention was to improve one’s understanding
of history, not to relive it. Tradition was a stimulus for intellectual and
moral energy, but not an invitation to reverie. There was no nostalgic
note in the description of origins, for the description of each period
balanced its advantages and drawbacks, so that even the fallen nobility
depicted by Boulainvilliers could not simply complain ‘of having lost the
superior and incommunicable rank that it enjoyed for so long during the
centuries of ignorance and coarseness.” Similarly, Boulainvilliers
thought it would be indecent to criticize the third estate and the magis-
Mably and Boulainvilliers 131

tracy indefinitely for what they initially were; it would be insane to


believe that one could downgrade the representatives of the third estate
“to the rank of serfs’; it would be absurd to attempt to restore the
turbulent equality practiced by the pillaging horde." Mably was some-
what less sure: with a little luck, and by acting in time, one could have
succeeded in going back to origins. Sometimes Mably indulged in a vision
of such a return, playing the role of the legislator intent upon reforming a
society that had not yet reached the point at which the accumulation of
vice and prejudice becomes irreversible. His description then takes on
the ethereal quality of utopia: “I have visions of citizens divided into
different classes. . . . everywhere I can see public stores containing the
riches of the Republic . . . And the magistrates, true fathers of the
homeland. . . .” But the opportunity for returning to the primitive state
had unquestionably vanished. “Wisdom comes too late, when mores
have been corrupted.”’"
Prediction, however, was not impossible. The unfolding of a tight-knit
temporal sequence in effect committed the historian to ‘““horoscopes”—
the term is Mably’s. The fact that each period was so closely linked to the
past and enmeshed in a network of effects and causes that left so little
room for chance was no doubt an encouragement to make forecasts. But,
by the same token, this vision quashed the hope of a leap forward as
effectively as it eliminated the prospect of a return to the past. Social
change could never take place except gradually and unobtrusively.
Boulainvilliers contented himself with expressing a desire to see the states
general reinstated and endowed in particular with financial powers.
Mably wanted to revive the assemblies of the Champs de Mars and the
Champs de Mai under the name of the Estates General. But the recourse
to “this forgotten practice” would create only an illusion of change if the
degenerate nation was unfit to ‘“‘take advantage of the event.” The
prophetic overtones ascribed to Mably’s work owe a great deal to its
proximity to the Revolution. His Droits et devoirs du citoyen, written in
midcentury, was published in 1789. But in reality there was no messianic
spirit in Mably any more than in Boulainvilliers. Neither author regarded
the present as the locus of an imminent restoration—and even less of a
revolution.

Thus, Mably and Boulainvilliers shared a common notion of history. But


they also used the same material and developed identical themes. It was
not to retrace the progress of civilization that Mably and Boulainvilliers
examined the passing of time, but to discover the lost origins of the nation
and the reasons and the stages of this eclipse. Only the Germanist thesis
with its emphasis on Frankish conquest gave them the means to conduct
this investigation, for it alone enabled them to wrest the history of the
nation from the yoke of the Roman imperium. It alone could conjure up
from the forests of Germany the warriors who brought the original
132 Seventeenth- and Eighteenth-century Culture

contract to France. It alone turned the French people into a free nation.
Everything had already been said in the late seventeenth century with the
words free nation.
History was thus the repository of the greatness of origins and the
secret of usurpations. It constituted a purely immanent reference system
that substituted for the unfolding of a providential plan. But Boulainvil-
liers used history to justify a ‘‘discourse on inequality” that was exactly
the opposite of Mably’s.
Yet for the comte de Boulainvilliers—and this once again proves how
much of an eighteenth-century man he was—men were originally free
and equal in rights. The problem for this intellectual so obsessed with his
noble status was, therefore, how to interpret and justify the traditional
social hierarchy according to the values of a cultural world already
dominated by the Enlightenment—that is, a world cut off from any
transcendental reference point and suffused with the concept of natural
equality among individuals. The task of history, in this cultural setting
marked by the notion of natural equality, was to explain the emergence of
inequality, the product of a conquest that turned the victors into nobles
and the vanquished into non-nobles. But Boulainvilliers did not stop
there, for history could undo what it had wrought. Race was the element
that would preserve history’s achievement. The original distinction be-
tween statuses, which was by and large accidental, became a necessity
when perpetuated through the blood. Thus, far from being archaic, the
concept of “‘race,”’ as Louis Dumont has suggested, seems to have been
linked to the birth of egalitarian ideology. Indeed, this concept made it
possible to replace the old organicist justification of inequality with a
historico-biological theory that was not only compatible with “natural”
equality but allowed the notion of equality to be integrated into an
inegalitarian construct. Nobles were different from (that is, superior to)
commoners, but all nobles were equal among themselves. Equality was a
basic element in Boulainvilliers’s vision of the nobility, and it was what
made him so different from Saint-Simon, obsessed with the preeminence
of dukes and peers.
For Boulainvilliers, the military caste created in the wake of the
conquest was a society of equals. He never failed to stress that among the
free conquerors who had come from Germany, ‘‘the now-prevalent dis-
tinctions based on titles were totally unknown.” But this egalitarian
society was also a closed society. While conquest conferred a rank then
perpetuated through the blood, nothing else could confer it in its place—
not even, and least of all, royal favor. Boulainvilliers was therefore intent
on emphasizing the closed character of the nobility (the idea of estab-
lishing once and for all a general inventory of noble families of the realm
was another aspect of that intention)" and the strict equality prevailing
within that exclusive group.
Mably and Boulainvilliers 133

This model, which reinstated equality in an inegalitarian framework,


did involve a major difficulty, for the argument based on heredity seemed
to Boulainvilliers as unconvincing when applied to the king (all the
misfortunes that befell the state were due to the hereditary monarchy) as
when applied to the people (while virtue could indeed “‘appear” among
commoners, it remained strictly personal and could never be transmitted
to their offspring). Yet heredity guaranteed “incommunicable nobil-
ity’ —a privilege that Boulainvilliers justified by some shaky statistical
arguments: virtue was more widespread in good races than in others;
furthermore, noble birth was the most common way of exploiting and
honoring virtue.
In addition, two contradictory dangers threatened this curious
arrangement. First, the inequality of status might be challenged from the
outside: it was therefore necessary to resist the rise of the third estate and
of the magistracy. Second, equality might be compromised within the
privileged group itself: it was vital to prevent the proliferation of new
forms of inequality within the system of second-degree equality estab-
lished amid inequality. Thus, the nobility had as much to fear from
expansion (that is, from the sovereign’s power to ennoble, which intro-
duced spurious equals into the group of equals) as from separation (that
is, from the prospect of new and more marked distinctions within the
group of equals). That is why Philip the Fair was the great villain of
French history, according to Boulainvilliers. Philip the Fair was the first
to “claim the power to ennoble the blood of commoners and, by a
somewhat similar abuse—although different in kind—he was the first to
create new peerages.”’" It is significant that the two actions should be
described as identical and as being the product of the muddle-headedness
inseparable from despotism. Philip the Fair was guilty on two counts: for
having introduced, first, a measure of equality into inequality (since
“newly ennobled commoners came to believe that they had attained
genuine equality”) and, second, a measure of inequality into equality, by
establishing a peerage that had no historical justification (since ‘‘peers,
princes, and hereditary nobles alike all sat with the nobility in the assem-
blies of state”). The only argument capable of tempering Boulainvilliers’s
opposition to Saint-Simon in the dispute over dukes and peers was that,
since the possibility of purchasing office had opened the doors of the
magistracy to ‘‘even the most modest bourgeois,” there was no reason for
“royal favor not to be displayed toward some nobles as well. . . The fact
remained that the two practices were, as far as principles were concerned,
equally heinous.
What was to be done, then? Boulainvilliers ultimately aimed at coun-
teracting the apologetics of absolutism by building another model of
national history, one founded on “‘feudal”’ society. But it must be noted
that Boulainvilliers was not a “‘feudalist” in the sense in which the term
134 Seventeenth- and Eighteenth-century Culture

was used during the Enlightenment and the Revolution. He criticized the
church and scholasticism, as well as the crass manners and the ignorance
of his time. He hailed the sixteenth-century Renaissance of arts and
letters’—an uncommon periodization in histories of France of this epoch
and in eighteenth-century culture in general.
In truth, what Boulainvilliers sought to give back to a nobility that had
so completely forgotten its origins as to be unable to define itself was the
awareness of its own existence. The fact was that the nobility would
nearly have lost that awareness were it not for the greed of commoners.
The “inordinate passion” of the latter for ‘“‘entering the nobility by royal
patent” still kept alive within the nobility the feeling of corporate iden-
tity, with its dual aspect of exclusion and equality. Boulainvilliers
dreamed of a less external basis. But the feudal regime that provided him
with the model was simply an ideal type of the relationship between civil
society and the king of France. This ideal type presented two features: a
society founded on inequality of status and on equality within each status;
the independence of nobles from the king, conceived as primus inter
pares.
Mably’s problem was quite different. Like Boulainvilliers, he postu-
lated a primitive state of equality, rooted in “a nature that had no
enclosed fields.” But he was not out to legitimize at any price a metaphys-
ically unacceptable social hierarchy, for he regarded it, in any case, as
having no genuine historical foundation. On the contrary, Mably de-
veloped a radical critique of heredity. He was challenging the concept not
as applied to kings alone, but as applied to all of mankind. As long as
vassals were granted personal distinctions, all was well in Mably’s history:
nobility was not transmitted through the blood, the children of the
nobility remained in the common class of citizens, and there was a single
order with two classes. History went wrong when these two classes (which
were constantly disintegrating and being rebuilt according to individual
talent and virtue) were turned into fixed orders. Mably’s villain was not
Philip the Fair, but Clothar II, who, by establishing hereditary benefices,
introduced a privileged order into the nation.
This order at once became the mortal enemy of equality—if only by
the fascination it exerted on the third estate, which it prevented from
having a sense of its own corporate identity (one can observe here the
complete reversal of Boulainvilliers’s thesis). Mably dated its birth to a
still remote period. But his basic aim was nevertheless to demonstrate
that the nobility had not always existed, thereby depriving the privileged
order of the prestige of an original state. This point was so crucial for
Mably that, in order to prove it, he did not hesitate to challenge the
authority of Montesquieu and to call to the rescue—only once, it is
true—the arguments of the opposite camp, that of Abbé Dubos: “I do
not imagine that Monsieur le Président de Montesquieu believes that
nations had noblemen from the start.” Thus, Mably found no trace of
Mably and Boulainvilliers 135

political feudalism in the origins of the nation. In his borrowings from the
germanistes, what counted for Mably was the structure of the ‘‘old gov-
ernment.’’ A king who was more of a captain than a monarch and who
commanded soldiers far more than he ruled subjects; a collegial executive
power exercised by the king and the notables in council; a legislative
power in the hands of a general assembly, the Champ de Mars—these
were the basic features of Mably’s system.
One must add to it a vision of the conquest borrowed from Montes-
quieu, halfway between the absurd idyll of the relations between Gauls
and Franks imagined by Dubos and the watertight separation imposed by
Boulainvilliers. For Mably, as for Montesquieu, the vanquished, having
been freed from the Roman imperium, rallied to their conqueror. Mably
saw this rallying less as an effect of habit than as the result of a voluntary,
conscious, and public choice—a solemn declaration, an oath. Official
renunciation of Roman law was the price that the Gauls had to pay in
order to “‘begin to enjoy the same prerogatives,” to obtain the same legal
settlements as the victorious Franks, and finally—the obvious triumph of
assimilation!—to be admitted to the Champs de Mars.
Thus, Mably rejected inequality on two counts: it was neither the
product of race (since, unlike Montesquieu, he did not believe in the
existence of an original Frankish nobility) nor the fruit of conquest (since,
unlike Boulainvilliers, Mably postulated the intermingling of the victors
and the vanquished). Inequality was a perversion, the first link in a long
chain of vices. It stimulated the growth of private wealth, the loss of
attachment to freedom, and—worst of all!—irregular attendance at the
Champs de Mars. In Mably’s view, therefore, the only possible founda-
tion for society was equality.
But one must be clear about the exact nature of equality as Mably saw
it. Mably did not believe, any more than Boulainvilliers did, in an
undiscriminating form of equality, whose model was once again provided
by despotism: “The closer one gets to despotism, the more social ranks
become indistinguishable.”’'* Mably—who also invoked the authority of
Montesquieu—inherited from Boulainvilliers the notion of a farsighted
despotism that saw things from too far and too high to perceive any
differences. But Mably did not advocate the type of equality that would
be fatal to dependency, for “‘only in Turkey or some other despotic state
can one believe that the subordination necessary in society is incompati-
ble with equality.”
This time the aim was therefore to reestablish inequality within an
egalitarian system, and Mably did his best to devise such a scheme.
Actually, a few adjustments were enough to make inequality tolerable.
Could one preserve inequality of wealth? The Swiss example showed that
wise sumptuary laws succeeded in preventing the anarchic proliferation
of wealth. Could one admit the inequality of power? Yes, if it were based
on competence, that is, if one claimed power less as a sovereign than as a
136 Seventeenth- and Eighteenth-century Culture

magistrate. Therefore, since it was impossible to return to primitive


equality, one could preserve all the features of unequal status, provided
one took certain precautions, some of which were reminiscent of the
“hidden hand” in Rousseauism: subjects would obey “without sorrow
and without humiliation sovereigns who, by copying their simple, unos-
tentatious, and thrifty bourgeois habits, would conceal the fact that they
constituted a privileged order.’’” As for the nobility, in order for it to
enjoy a fully justified status, all it had to do was to bear in mind that “the
spirit of servitude is the most demeaning behavior for a nobleman.”
One category was excluded from the benefits of this system: the
“populace, which lacked all credit, esteem, and fortune and was incap-
able of acting on its own.” Admittedly, some of Mably’s writings imply
that every man should belong to an order and enjoy genuine rights within
it: every man should be declared free. But it did not necessarily follow
that every man should participate in government. While the artisans in
the Entretiens de Phocion were regarded as free men in the eyes of the
law, they had no political existence and did not take part in assemblies.
“It is important for the very masses themselves, who are demeaned by
their work and occupations, not to take over the government.’ This is
not a farcry from the terms used by Boulainvilliers to describe the subject
nation, similarly excluded from political assemblies “‘because its role was
to work, to cultivate the land, and not to partake of the honors of
government.” Boulainvilliers excluded a larger category than did Mably,
but both used the same criterion for exclusion.
In other words, Mably’s version of history is essentially a justification
of the society of orders. Like Boulainvilliers, he was obsessed by the
feeling that societies were perpetually deteriorating, and he sought to halt
the interminable drift of history. He therefore wanted to preserve social
orders but also to prevent the encroachments, the challenges, and even
the transfer of individuals from one order to another that heralded new
upheavals. This aim could be achieved by assigning a fixed amount of
wealth to each order:” individual wealth could increase or decrease, but
the total wealth of each order had to remain unchanged; thus, none of
them would “fall prey to contempt.”’ The effort to provide social orders
with an insurpassable ‘constant condition” was to be accompanied by the
education of the community. Mably was as severe as Boulainvilliers
toward those who tried to infringe upon the rights of the neighboring
order. For example, Mably criticized the parlementaires, who stubbornly
advocated the notion that “the clergy and nobility were admitted only as
a favor to the assemblies of notables, primarily composed of magis-
trates.” He was not optimistic about the future of a kingdom “‘in which no
one wants to remain in his place and in which everyone’s ambition is to
join an order that refuses to let him in,”’ for vanity would then become the
major concern of every citizen.
Mably and Boulainvilliers 137

Thus, Mably’s great innovation was not the reversal of the concept of
nobiliary liberalism—what one might call his democratic version of Ger-
manism—but its extension to include the third estate alongside the clergy
and the nobles in the nation’s assemblies. Mably also linked the ideal of
equality to the preservation of ‘“ranks.” This linkage—which, inciden-
tally, shows how risky it would be to oppose the liberal current and the
democratic current in the eighteenth century—could be achieved,
according to Mably, thanks to a legislation devised as a harmonious
system.
Indeed, the idea that is new in Mably, and that one would look for in
vain in Boulainvilliers, is that of harmony. The organicist metaphor is
obviously useless in describing a society split between victors all enjoying
equality through privilege and vanquished all equal in servitude. How-
ever, the metaphor is indispensable to a pluralist notion of the body
social. It guarantees the dignity of each section of society (‘‘each class of
citizens understands that it can be happy without oppressing the others’’)
and insures that they will take part in the effort to reach a common
goal—the “public good’—which can be attained providing there is a
general acceptance of a hierarchy of social functions. That is why, for
Mably, the most admirable sovereign in the history of France was Charle-
magne, the orchestrator of harmony, the mediator who brought social
orders closer together and made them forget their old quarrels. The work
of the ideal legislator is to hold the balance between the various orders—a
balance that “‘is today the only imaginable equality among men.” Conse-
quently, the legislator’s task is easier when there are more orders: they
balance each other out and, in addition, it is a finer achievement to make
them all contribute to common happiness. Four orders are better than
three, and five better than four. Such multiplicity would therefore,
according to Mably’s wish, mean the triumph of harmonious plurality,
which he regarded as the ultimate constituent factor of the nation.
Thus, the discussion on the history of France was one of the liveliest
features of eighteenth-century French culture, because it was rooted in
civil society’s drive toward power and in the search for a legitimacy
intended not as a substitution—since absolutism was a usurpation—but
as a restoration. Boulainvilliers and Mably represent two decisive mo-
ments in the debate, for they were writing respectively the nobiliary
version and the roturier version of the same history, that of the nation.
What separates them is important as far as their opinions are concerned:
Boulainvilliers was an exclusionist aristocrat, while Mably brought the
third estate back into the national fold. But in other respects their
differences were virtually negligible. Both authors joined in the social
demand for a liberalization of the regime—a demand for which the
nobility, at the end of Louis XIV’s reign, acted as a natural spokesman.
Above all, both authors were writing what was intellectually the same
138 Seventeenth- and Eighteenth-century Culture

history, built on the same institutions, and enacted by the same cast.
Their concept of “nation” enabled them to deal simultaneously with the
social and the national dimensions, unlike their nineteenth-century suc-
cessors, who restricted their vision to the national dimension alone. By
combining the social sphere and the national sphere, Boulainvilliers and
Mably wedded society to its own myth.
In the years immediately preceding the French Revolution, the histor-
ical theme of nationhood acquired a particularly momentous political
significance. It occurs very often in the prerevolutionary pamphlets pub-
lished between the Assembly of Notables and the opening of the Estates
General (late 1787—April 1789). These pamphlets, most of which were
anonymous, testify to the spread of the Boulainvilliers-Mably version of
French history among the educated classes of the eighteenth century.
They contain many references to the origins and history of the nation.
In these improvised writings, so closely connected to contemporary
events, such references present the advantage of making the political
demands quite explicit; they also show the extent of the triumph of
Mably’s ‘‘democratic”” thesis over-Boulainvilliers’s aristocratic thesis in
public opinion. When one of these pamphlets defends the orders of
society as one of the building blocks of the monarchy, it is in the name of
tradition or of the Constitution, never in the name of race. Far from being
a possible foundation for a new legitimacy that would have benefited the
nobility, the notion of race was used as an argument against the nobles in
Sieyés’s Qu'est-ce que le Tiers Etat? ,in which he advocated sending back
“into the forests of Franconia all the families that cling to the absurd
pretention of being issued from the race of conquerors and of having
inherited their rights.”
In actual fact, “Germanic liberties” had become the mythical heritage
of the entire nation, now waging the decisive battle for the restoration of
those liberties. The centerpiece of this construct was the king-and-nation
dyad—two powers that were not defined by a conflict but were the two
indispensable elements of legitimate public authority. Their theoretical
definition was not contradictory but complementary. The king, enjoying
a historical right attested by filiation, embodied the state. The nation was
both a historical and a mythical human community—a depository of the
social contract, a general will whose origins were shrouded in the mists of
time, a promise of fidelity to those origins. Between king and nation there
existed an apparent relationship of subordination, which was actually a
binding obligation to cooperate. The king was the head of the nation, but
his rule was legitimate only if he respected the rights and laws of the
nation.
When he did not respect them, it was because nefarious elements
thwarted that cooperation. The culprits were either ministerial despo-
tism, as had been the case since Richelieu, or usurpation by the nobility,
as during the feudal period. Thus, the battle waged during the years from
Mably and Boulainvilliers 139

1787 to 1789 was thought of as a campaign for the nation’s reconquest of


its own rights and for the reestablishment of communication between
king and nation, to be symbolized by the rule of law. An age-old legalistic
current triumphed in this new secular religion, intended as the foundation
of a revamped monarchy.
Thus, the French Revolution, before it became the zero point of
French history, the myth of origins of our contemporary society, was the
culmination of our remotest past, the restoration of the national com-
munity to the fullness of its rights. Ultimately, the two visions—the
Revolution as a culmination or as a point of origin—are informed by an
identical desire to abolish time (an aim inseparable from the revolution-
ary ideology) and by the same emotive power that created modern
nations.
Civilization and Barbarism in
Gibbon’s History

In the Paris of the Enlightenment,-from which Gibbon drew much of his


intellectual inspiration, the century had begun with the so-called quarrel
between the ‘‘ancients” and the “‘moderns,”’ a literary controversy that
seems to have run its course by Gibbon’s time but that nevertheless went
to the heart of his intellectual life. At issue in the quarrel as it began
toward the end of Louis XIV’s reign was the nature of the cultural
identity between ancient and modern thought that had constituted the
common heritage of the European intelligentsia since the Renaissance.
The moderns did not deny their ties with antiquity and its heritage, but
they rejected the notion that their own contribution should be limited to
the rediscovery of a Greco-Roman model. They declared the modern to
be superior to the ancient, especially in regard to the progress of knowl-
edge, the rigor of reasoning, and the quest for truth; in doing so, they
broke with the classical conception of history as cyclical and replaced it
with a belief in the creative value of time, which would progressively
separate truth from error.
This detachment of the concept of “modern” from its ties with antiq-
uity gradually allowed the development in the course of the eighteenth
century of an evolutionary history and a theory of progress. After the
providential histories of churchmen, after the cyclical histories of the
humanists and the Reformation, now history could be opened up infi-
nitely into the future. The fears that were latent in the notion of an infinite
and unknown future were conjured away by the spectacle of a continuing
advance in the arts and sciences, which seemed to promise a more general
progress.
First published in English, under the present title, in Daedalus, Summer 1976, 209-16.
Reprinted here with corrections, by permission of Daedalus, Journal of the American
Academy of Arts and Sciences, Boston, Mass.

140
Gibbon’s History 141

The ideas of the moderns were given their definitive form for the
eighteenth century in d’Alembert’s Discours préliminaire and Condor-
cet’s Esquisse d'un tableau historique des progrès de l’esprit humain. In
them we find that ‘‘civilization” was a process before it was a state; its
purpose was to make “civil,” to “‘police” the uncivilized. The word
civilization itself, which dates from this period in both French and En-
glish, was invented to express this drive of enlightened society toward
what ought to be, the conviction of being on the right road, the certitude
that the future was in fact infinite and that history had a purpose.
But, in many respects, Gibbon wrote a history very different from
others of his century; his history was at once more “‘ancient”’ and more
“modern”: more ancient because Gibbon was a professional antiquarian
and familiar with the learned and the scholarly societies of Europe in the
seventeenth century, into whose proceedings his mastery of French per-
mitted him easy access; more modern because he integrated this erudi-
tion with the art of retelling the past and, in so doing, invented the
historical panorama that was to become so important in the romantic era.
Gibbon was somewhere between Tillemont and Renan, but this ‘“‘some-
where” constituted more than just the philosophical consensus of the
Enlightenment; it was also an original vision of the past and present in
Europe.
His own biography already reveals some of this vision in the studious,
retiring years in Lausanne undertaken by an English gentleman who had
tasted the charms of the salons of Paris and London. For, on the banks of
Lake Geneva, in those Franco-German confines that were so crucial for
the transformations of European culture, Gibbon saw only a retreat
where he could continue his work. A place of exile in his youth, Switzer-
land became a shelter in his maturity. By choosing Switzerland, he
avoided belonging to any of the nations of Europe and thus signified that
his only loyalties lay with his two universal homelands: historical erudi-
tion—that most international of cultures—and Rome, the mother of
Europe. The revelation of 15 October 1764 had given a meaning to his life
amounting almost to a religious conversion. Freud entered Rome only
after he had already become Freud and was ready to confront the classical
studies of his youth, which constituted at the same time the world of the
gentile. But Gibbon did not become Gibbon until that day in October
when he trod “with a proud foot upon the ruins of the Forum.” He
received from Rome more than his idea of civilization; he received from it
his cultural identity.
Between Rome and Gibbon lay the same link that connected the
European intellectuals of the Renaissance to antiquity. It was a romance,
complete with the same delight in discovery and the same freshness and
joy in the recapture of lost secrets. Yet, since the Renaissance, education
had put Rome into everyone’s store of knowledge: the Latin and history
that one learned from Cicero and Tacitus continued to be the basis of the
142 Seventeenth- and Eighteenth-century Culture

scholarly apprenticeship expected of future gentlemen. Thus, when Gib-


bon discovered Rome, he was familiar with it already. It was the most
classical topos of European culture, the basis for its theater, its art, its
moral philosophy, its historical reflection. What is strange is that his
meeting with a place already so heavily charged with significance should
still contain any surprise at all, much less one amounting almost to an
existential conversion. In contrast, when Montesquieu had visited Rome
in 1729, though he could hardly have been accused of indifference to
Roman history—he had written his Essai (1716) before his famous
Considérations (1734)*—precisely because ancient Rome was for him a
legitimate subject of study, he retained in his reactions to pontifical Rome
something of the attitudes of an intelligent tourist—an observer of cus-
toms, politics, and the arts.‘ Thirty-five years later, Gibbon saw only the
urbs in Rome: “Each memorable spot where Romulus stood, or Tully
spoke, or Caesar fell, was at once present to my eye; and several days of
intoxication were lost or enjoyed before I could descend to a cool and
minute investigation.’
This traveler’s sentiment, which anticipates the emotional investment
the romantic writers were to bring to history, also reflects Gibbon’s
devotion to European humanism and to the tradition of the sixteenth and
seventeenth centuries. According to this tradition, Rome was a model of
civilization that had never been surpassed. For Gibbon, however, this
was not a philosophical proposition. He did not theorize about man in
society, about natural law, about the social contract; he had no interest in
that essential question of his time, What is a “‘savage,’’ and where does
one place him in the history of humanity? He was instinctively and totally
the historian; that is, he was an empiricist, a narrator, and completely and
unquestioningly Europocentric in the midst of a world where only Greco-
Roman antiquity, the Judaeo-Christian tradition, and their offspring—
Europe—counted.
The superiority of Rome was, therefore, not something that had to be
demonstrated. It was in the order of evidence, a fact. It was a unique
experience that had only to be described and had nothing to do with
historical laws. Gibbon had read and admired Montesquieu, but in the
end the two works had little in common. For Montesquieu, Roman
history was merely a ‘case study” for a general typology of political
regimes. The reasons for Rome’s greatness were at the same time the
reasons for its fall: the expansion of the Empire necessitated a monarchi-
cal government incompatible with the laws that had nonetheless made it
necessary. Gibbon, though he intermittently echoed his predecessor’s
theory, was not given to rigorous conceptualizations. He was eclectic,
and he multiplied not only the possible explanations for Rome’s greatness
and fall, but even the kinds of explanations. For him, Roman history was
not just another collection of human experiences; it was quite simply, in
the second century after Christ, the highest point in human history, “the
Gibbon’s History 143

period in the history of the world, during which the condition of the
human race was most happy and prosperous.’” It was a unique moment,
which did not fit into any general concept of historical change.
This view of second-century Rome as representing a privileged mo-
ment in human history was impossible to integrate into a linear view of
humanity’s progress, such as the one Condorcet wrote later on (1793).
But such a conception was typical of cyclical histories, according to which
civilization had no purpose toward which the cumulative progress of
mankind was heading but rather consisted of a series of intermittent
“happy and prosperous” periods. In this conception, second-century
Rome represented the crowning of history’s most splendid cycle—and to
such a degree that the hope Gibbon assigned to eighteenth-century
Europe was not to surpass this model and the values it bore but to
constitute a less fragile vessel for them.
It is difficult to determine very precisely or securely the reasons behind
Gibbon’s worshipful admiration for the Empire. Nowhere does he pre-
sent a systematic description of Roman civilization. He is so permeated
with the idea of its superiority that he feels no need to discuss the reasons
for it nor to take stock of its elements. The famous judgment from
chapter 3 of the Decline and Fall quoted above, regarding the excep-
tionally happy state of humanity during the Antonine period, is sup-
ported only by some rather brief justifications relating to the virtue and
wisdom of the emperors who came after the civil unrest in the first
century. The loss of freedoms—those famous Roman freedoms that
nourished so many of the century’s books—was more than compensated
by the exercise of an equally enlightened despotism.
Gibbon was, in fact, indifferent to the political philosophy that so
excited many of his contemporaries. He wrote, as did his masters in
antiquity—Thucydides, Cicero, and Tacitus—moral history. For a period
to be great, it was necessary, but also sufficient, for it to have produced a
certain kind of man. The emperors of the second century were his
exempla; they were at once a culture and a moral.
Nevertheless—and not surprisingly—Gibbon’s history of Rome is also
a history of Gibbon. The cult of the Antonines reflected the views the
historian held toward his own present. In this respect, Gibbon was an
entirely original writer. He did not, as did the men of the Renaissance,
have to rediscover Rome beneath the medieval sediment, for that work
had been done for him. But neither did he believe, as did the artists and
men of letters of the seventeenth century, that Rome could truly be
imitated. He was already too much of a historian, in the nineteenth-
century sense of the word, not to conceive of history as flux, never
representing the same situations—or the same successes—twice. The
cycles of civilization that history presented were not comparable, and
they were transient. The Decline and Fall of the Roman Empire is
dominated throughout by the romantic notion of the uniqueness and
144 Seventeenth- and Eighteenth-century Culture

transience of the great periods of history. It is an extraordinary example


of that moment when the old cyclical conception of history hung in
balance with nineteenth-century historicism.
Gibbon, we must remember, did not write a history of the Roman
Empire; he chose to write a history only of its decline and fall. In the very
years when impeccably neoclassical ruins contributed their note of con-
trolled sadness to the parks of aristocratic castles, Gibbon joined his
powerful voice to the melancholy chorus of European scholars: far from
being history’s promise, civilization was the historian’s nostalgia. The
clearest expression of this comes at the end of chapter 38, in the famous
“General Observations on the Fall of the Roman Empire in the West,”
which forms the only genuinely analytic commentary on the problem that
gave the Decline and Fall its title. As we know, Gibbon, after having
summarized the causes of the fall of Rome, proposed to draw from them
lessons for his own time. Does this mean that he found a similarity
between eighteenth-century Europe and the Roman Empire in its years
of fading splendor? Yes and no. Yes because he points out, at the risk of
appearing unpatriotic, that he considers Europe a great republic, united
precisely by its participation in the same civilization, for all its inhabitants
have attained ‘‘almost the same level of politeness and cultivation.” By
this he means that the European elite of his time, regardless of the
vicissitudes in power relations among nations, displays a “‘state of happi-
ness” and a “system of arts, and laws, and manners” common to all
nations, and these are what constitute a civilization. There is nothing in
this attitude that cannot be found in Enlightenment thought of the most
classic kind. At the same time, however, the comparison with the civiliza-
tion of the great Roman period was made only with references to external
dangers: Roman history no longer constituted the basis of Europe’s
cultural identity, but a lesson for Europe’s defense.
In short, what fascinates Gibbon about Roman civilization is not so
much that in it lay the foundations of Europe but that it was so fragile, as
fragile perhaps as Europe’s civilization now was. As he looks upon the
ruins of Rome, the humanist gentleman asks the future about the chances
of survival for the things he loves. Gibbon’s feeling of belonging to a
special but threatened moment in history does not come from an analysis
of the inner contradictions of European civilization. On the contrary, he
says that the fact that the majority of nations carry on this civilization and
emulate it is, in his view, a source of strength, not of weakness. He does
not even suspect that, within this community of European culture,
nationalism would become an element of disintegration. He reasons only
in terms of a possible barbarian (or, more precisely, ‘‘savage’’) menace,
as if the invasions had been the sole cause of the dismemberment of the
Western Empire (though several pages earlier he expressly blamed its
excessive size and the disintegrating influence of Christianity) and as if
the multinational Europe of the eighteenth century could be compared to
Gibbon’s History 145

the extended Roman frontiers (though a few pages later he points out
that northern Europe, once a stronghold of barbarians, had little by little
been civilized).
The comparison, in fact, comes out in favor of eighteenth-century
Europe. Gibbon lists the elements militating in favor of a relative stability
for the civilization it has attained—its geographic extent, its national
diversity, its mastery of the art of war, the undeniable distinction of its
inventions and basic technology—and in the end he even advances the
idea of a continuous progress of humanity “‘since the first discovery of the
arts.” But it was still the case that Europe’s civilization, even if it was
probably indestructible, remained subject to the challenge of the “‘savage
nations of the globe.” It could be seriously threatened, forced to retreat,
to become expatriated ‘“‘in the American world.” In a century that had
witnessed a rapid acceleration in the Europeanization of the globe, a
phenomenon that had characterized history everywhere since the Ren-
aissance, Gibbon discussed the reverse hypothesis drawn from the Ro-
man example. It was not that he thought it likely, but that his concept of
civilization already included a threatening reversal, a permanent exterior
menace—the savage world.
Savage or barbarian? Generally speaking, the Enlightenment distin-
guished between these two terms by defining them as two different steps
in the evolution toward civilization. The Encyclopédie (1751) still con-
fused them, however, for in it ‘“‘savages” were described as ‘“‘barbaric
people who live without laws, without government, without religion, and
who have no fixed habitation.”’ But it then adds, ‘There is this difference
between savages and barbarians, namely, that the first form scattered
little nations that have no desire to unite, whereas the barbarians often
unite, and this happens when a chief submits to one of the others. Natural
liberty is the sole concern of government among savages; along with this
liberty, nature and climate are almost the sole forces that prevail among
them. Occupied by hunting or pastoral life, they do not burden them-
selves with religious observances and do not make their religion a basis
for organizing their lives.”
Some time later, the Dictionnaire of Trévoux (1771), although it
followed the Encyclopédie very closely, took pains not to use the term
barbarian as a synonym for savage. The latter term “‘is also used for those
people who wander in the forests, without fixed habitation, without laws,
without government, and almost without religion.” Thus, the two dic-
tionaries agree with Montesquieu in distinguishing between savages and
barbarians by stating that the first “live scattered about, retreating into
the forests and the mountains, without uniting, while the second often
unite and sometimes live under a chief to whom they have submitted.”
Ferguson (1767) writes that the savage, in America, has neither property
nor government nor judges, while the barbarian, in Europe, had property
and obeyed a chief.’ De Pauw (1768) clearly distinguished different stages
146 Seventeenth- and Eighteenth-century Culture

in history when, for example, he differentiated between the treatment of


prisoners by “the most savage,” “ordinary savages,” ‘“‘semibarbarous
peoples,” and ‘‘the least barbarous”’ nations.’
Thus, the Enlightenment, in France and in England, constructed a
three-stage progressive scheme of history: ‘‘savage-barbarian-civilized.”’
Just after Démeunier, Robertson, in 1778, defined the same three stages
of evolution but this time according to more materialistic criteria: the
savages had neither writing nor metals nor domesticated animals (Amer-
ica); the barbarians had metals and domesticated animals (Europe, Mex-
ico, and Peru); the civilized nations had industry and the arts." To this
materialistic classification, destined for a great future in the following
century, the eighteenth-century thinkers usually added philosophico-
political criteria: the savage belonged to the natural order. He was
without fixed habitat, without religion, without laws, without customs,
the embodiment of human origins. Yet he was capable of acceding to
history and to a policed society so long as his natural character was not
corrupted by contact with Europeans. The barbarian, however, already
belonged to history: he formed nations and he established states; but he
did not enjoy the protection of regular laws, and his knowledge and
customs remained, or reverted to being, crude. For barbarism was what
preceded, threatened, or came after civilization. “‘Nations have all oscil-
lated,” wrote Diderot, ‘“‘from barbarism to the policed state, from the
policed state to barbarism, until unforeseen causes have brought them to
an equilibrium that they never perfectly maintain.”’”
Gibbon, however, did not distinguish between the savage and the
barbarian. In this respect he was behind the times, still fixed on the
historical dichotomy of the beginning of the century. For he was not
really interested in man in nature, the concept of the savage that so
engaged the philosophers of the Enlightenment. His perception was
more historical than philosophical; it was perhaps entirely historical. The
fall of the Roman Empire is played out between a civilized society, on the
one hand, and those outside this society, on the other. Ultimately this
former reader of Thucydides and Tacitus adopts the classical distinction
between Greek and barbarian or Roman and barbarian: the barbarian is
the one who is on the other side of the frontier. But Gibbon also calls him
“savage,” in part at least because the word was so commonly used in the
eighteenth century, but also to emphasize the gap that separated him
from civilized man.
One has only to read chapter 9 of Gibbon’s history, which is devoted to
the Germanic tribes, to become convinced of this. Here Gibbon closely
follows Tacitus, his master and model. He wants to understand what has
made those “‘wild barbarians of Germany’? Rome’s most formidable
enemy. No cities, no letters, no arts, no monetary system—such are the
negative traits that define what he calls a “‘savage state.” When he comes
to the famous theme of the Germanic freedoms, so important in the
historiography of the period, especially in Montesquieu and Mably—
Gibbon’s History 147

whom Gibbon had read attentively—the description of the system of


assemblies and the independence of the soldiers do not arouse any
“democratic”? sympathy in him. To his way of thinking, the backward
state of German customs, letters, and arts carries its own condemnation;
on several occasions he refers to the Germanic tribes as “‘savages,”’ until
he arrives at the following perception, which more precisely defines his
thought: ‘Modern nations are fixed and permanent societies, connected
among themselves by laws and government, bound to their native soil by
arts and agriculture. The German tribes were voluntary and fluctuating
associations of soldiers, almost of savages.”
An extraordinary judgment for a writer so passionate about the histo-
riography of his time, a witness of the central controversy of French
historiography over the origins of the nation: were they Roman or
Frankish? At no time did Gibbon seem interested in what had constituted
in Europe since the sixteenth century one of the raisons d étre of history
and the fundamental impetus for it: the quest for origins, the original
contract from which a nation arose. Of the two questions that the eigh-
teenth century asked of history—What is a nation? What is civiliza-
tion?—Gibbon was interested only in the second. He had read Boulain-
villiers, Montesquieu, Dubos, and Mably, not as their heir but purely as
an erudite, as an ethnologist of the Franks. As a result, he deprived the
Germanic people of their basic dignity; they existed only as “‘near-
savages.”
The same sort of judgment can easily be found when Gibbon deals
with other peoples whose movements threatened Rome. In chapter 26,
for example, Gibbon describes the nomadic tribes of the Far East, whose
growth would ultimately affect the Empire by driving the Goths to the
West, along with “‘so many [other] hostile tribes more savage then
themselves.”’ In discussing those populations of nomadic shepherds,
Gibbon says that what makes the study of them so simple is their proxim-
ity to animality:

. . it is much easier to ascertain the appetites of a quadruped


than the speculations of a philosopher; and the savage tribes
of mankind, as they approach nearer to the conditions of
animals, preserve a stronger resemblance to themselves and
to each other. The uniform stability of their manners is the
natural consequence of the imperfection of their faculties.
Reduced to a similar situation, their wants, their desires, their
enjoyments, still continue the same; and the influence of food
or climate, which in a more improved state of society, is
suspended or subdued by so many moral causes, most power-
fully contributes to form and to maintain the national charac-
ter of Barbarians.

Consequently, barbarians are savages. Gibbon recognizes neither


natural man nor the ‘noble savage.” There is only historical man, and
148 Seventeenth- and Eighteenth-century Culture

certain of his manifestations, unchecked by reason, remain bogged down


in a dependency predicated upon natural conditions, stagnation, and the
absence of a policed society. At the other end is civilized man, who is not
necessarily the conqueror: the fall of the Empire proves that point. The
paradox is that Gibbon was so interested in the victory of barbarism over
civilization.
This paradox is resolved when one realizes that Gibbon wrote a second
history alongside the first, but distinct nonetheless; this second history is
that of Christianity. Gibbon was the first historian to treat the history of
Rome and the history of religion together. This innovation lies behind the
chronological distortion to which he subjected Roman history: he was
trying to comprehend not just the secrets of the greatness of Rome, not
just the collapse of that greatness, but beyond that the passage from
imperial Rome to papal Rome. If the fall of the Empire encompassed all
the Middle Ages, it was because the historian sought to describe more
than the fall of a civilization invaded by barbarians. Ultimately he says so,
with a disarming directness: “I have described the triumph of barbarism
and religion” (chapter 71).
“Barbarism and religion’’—the phrase clearly indicates that if, in the
dramatic history of Rome’s fall, the two phenomena contributed to the
same result, they nevertheless remained distinct. Religion helped barbar-
ism to win, but it was not itself barbaric. For the religious phenomenon
was multiform because it was rooted in that fear and ignorance which
were inseparable from human society. Gibbon was an enthusiastic disci-
ple of Bayle, who perfectly reflected Enlightenment thinking on the
subject. But he goes further than Bayle. He was the first historian
systematically to place religion in a relative position in human events,
which hardly means that he reduced its importance—on the contrary, he
paid particular attention to it—but that he integrated it in all its many
forms into the societies and empires whose history he outlined. There is,
thus, in Gibbon a historian of paganism, a historian of the cults of the
ancient Germans, a historian of Islam, and a historian of Christianity.
Religion is a cultural phenomenon that he examines with great care, even
to details of its refinements, as the chapters devoted to the theological
controversies of the first centuries of Christianity demonstrate.
Religion thus becomes a part of the great social and historical drama of
the fall of the Roman Empire. But when Gibbon writes that he has
described “the triumph of barbarism and religion,” he means only one
religion: Christianity. The analysis of the religious phenomenon, of the
generic, becomes specific. In the same way that, in the eyes of this
gentleman-scholar, there have been several societies in history that have
attained the status of, and embodied, civilization, though none so per-
fectly as the Roman Empire, so, too, inversely, have there been many
religions in human history, though probably none of them quite so
noxious as Christianity. Here we reach the second major theme in the
Gibbon’s History 149

Decline and Fall. The first revolved around the external confrontation
between Rome and the barbarians, the second around the internal disin-
tegration of the Empire at the hands of the Christians. It is a Rome
weakened by “the spirit of Christianity,” we might say, borrowing from
Montesquieu, that is finally conquered by the barbarian invasions. And
the real victor, as the history of both the Eastern and Western Middle
Ages shows, was Christianity.
This is perhaps the reason—or the existential impulse—that led Gib-
bon to write, not a panorama of Roman civilization, but an account only
of its fall. Of course, like his contemporaries, Gibbon was sensitive to the
transience of history’s great successes, but this feeling hardly justified his
having extended his account as far as the fifteenth century! If he wished to
encompass the entire Middle Ages into the fall of Rome and under what
was, in that context, a very strange title, it was because Christianity’s
history fascinated him as much as, perhaps even more than, that of
Rome. He invested his account with his hatred, not of the Christian faith,
but of the church, the priests, and monks; he committed to it his struggle
against intolerance and fanaticism; he deployed for the purpose the
whole anticlerical tradition of the French Enlightenment.
He was too good a historian, however, to say—and he never did
say—that the Christianity of the established churches for which he pro-
fessed no love was a form of ‘“‘barbarism.”’ On the contrary, he seized
every opportunity to show how Christianity was in fact the important
reconstructive principle of his historical world—the Europe built upon
the debris of the Roman Empire. On the one hand, he wanted to express,
through the three-dimensional historical space in which civilization, bar-
barism, and Christianity evolved, his preference for a pre-Christian civi-
lization rather than a Christianized barbarism. But, on the other hand, he
constantly—and more radically—showed that what motivated his
worship of Rome, its values, its “‘spirit,’’ and its moral figures was the
existence of a civilization in its chemically pure state—and that meant
without the Church.
ALU OR
ae we 4.6 ee
oR a
=
SAUTER
Le
ars
ios tr fee La
*
D in : cues

the 2 TES a

|4
DEMOCRACY
| |
F4
Fe
From Savage Man to Historical Man:
The American Experience in
Eighteenth-Century French Culture

In order to interpret America, along with the rest of the non-European


world, the Europeans of the seventeenth and eighteenth centuries re-
sorted to the conceptual opposition between the civilized world and the
savage world. This device enabled the Europeans to translate into episte-
mological terms the new regions discovered since the Renaissance, the
new-found peoples of the globe—in short, a world that offered the three
unusual features of being neither European nor Christian nor national.
The “‘savage”’ world was described and inventoried by travelers, not by
historians. It was a world without laws, without arts, without govern-
ments—in a word, without history. For an implicit scale of values still
piaced time, the dynamic creator of laws and nations, above space, the
passive distributor of immobile societies.

The Savage Man

The misfortune of the savage was sometimes equated simply with his
backwardness. Such was the view expressed by travelers who, in provid-
ing the reader with accounts of the mores of distant peoples, were seeking

Originally delivered as a lecture in October 1976 at the William L. Clements Library, Ann
Arbor, Michigan. First published in French in La Révolution américaine et l'Europe,
proceedings of the Colloques Internationaux du Centre National de la Recherche Scienti-
fique, no. 577, Paris-Toulouse, 21-25 February 1978 (Paris: Editions du CNRS, 1979),
93-105; reprinted in Annales: économies, sociétés, civilisations, July-August 1978, 729-39.
This study, as will be seen, is greatly indebted to two classic works: Bernard Fay, L'Esprit
révolutionnaire en France et aux Etats-Unis (Paris, 1924); and D. Echeverria, Mirage in the
West: A History of the French Image of American Society to 1815 (Princeton: Princeton
University Press, 1957). My analysis is an attempt to modify the conclusions reached by
these two authors, but it is based very largely on the material they have provided.

153
154 America and the Idea of Democracy

to popularize the picturesque aspects of otherness but were also bringing


back an image of the past from an area of the contemporary world: the
savage was man in his infancy. But the curse that afflicted the savage
could also be the incurable consequence of permanent factors such as
climate, natural environment, or race. This famous and influential thesis
was advanced in the late eighteenth century by Cornelius de Pauw, who
invoked Buffon’s authority when presenting his Recherches philosophi-
ques sur les Américains! This extravagant compilation of geographical
and biological data was aimed at combating two ideas: that the European
colonization of America was a good thing and that the inhabitants of
America were happy and thriving in the savage state. De Pauw, on the
basis of a fundamentally pessimistic analysis of American savagery,
called instead for a systematic European withdrawal.
For de Pauw, that savagery, an object of study for the “history of
natural man,” was not a social state; it was a curse of nature—and more
precisely of climate—that had led to the appearance of biological features
of inferiority with respect to European man. De Pauw regarded America
as a recently emerged continent, colder and moister than the Old World,
covered with forests and swamps, and filled with degenerate, and thus all
the more dangerous, animals. Its human inhabitants suffered from the
same handicaps stemming from the same causes. They were lazy, amoral,
sexually and intellectually idle, hairless, and devoid of intelligence. They
represented not a sort of happy childhood of historical man but, on the
contrary, a degenerate state in comparison with European man:

Is it not surprising that the only men found on one half of the
globe should be beardless, devoid of intelligence, contami-
nated by venereal disease (for syphilis, which Europe calls the
French disease, has become the American disease), and so
fallen from the human state as to be incapable of being disci-
plined—which is the complement of stupidity? The fondness
that Americans have always displayed, and still display, for
savage life proves that they hate the laws of society and the
constraints of education, which, by taming the most intemper-
ate passions, are alone capable of raising man above the
animals. . . .”

De Pauw’s theory of the nefarious influence of the American nation on


man also applied to the colonizers—first to the Spaniards since the
sixteenth century, then to the emigrants who peopled the English col-
onies. The Dutch theoretician cited the case of the Creoles, but his
demonstration applied also to any American of European origin—
whether immigrant, son, grandson, or great-grandson of Europeans. The
‘“debased genius of the Americans” (title of section 1, part 5) had a rapid
effect and caused immigrants to degenerate, although somewhat more
slowly than animals taken from the Old World to the New, for men could
From Savage Man to Historical Man 155

“protect themselves against the immediate impact of climate”; all the


same, the end result was inevitable.
Thus, the immigrant enjoyed no long-lasting privilege. All Americans
were or had become savages. The European who left Europe was aban-
doning civilization for savagery. The most astonishing feature of this
bleak conceptualization of the American enigma is its blindness in face of
the facts, its ignorance of the most recent historical developments. De
Pauw was writing at about the time that the decisive political crisis
erupted between the British colonies in America and their motherland, a
crisis that brought typically European values into play and turned the
principles of the European Enlightenment against Europe itself. But de
Pauw’s blindness fed primarily on the tremendous cultural and affective
commitment of European elites to their civilization, which they quite
simply regarded as civilization tout court. In the European consciousness,
this civilization presupposed the existence of an external or inverted
world always regarded as somewhat threatening: savagery. De Pauw
carried the commitment to civilization to its most extreme—and also its
most archaic—form, for he argued that civilization was underwritten for
all time by nature, that it was a product of natural history and not of
human history. If climate and geography were indeed the determining
factors of civilization, Europe was, for all eternity, on the right side of the
fence.
During the same period, however, two factors of a very different
nature modified the outline of this traditional image, which comforted the
good conscience of Christian or seventeenth- and eighteenth-century
Europe. The first factor was political: the American Revolution. The
second was intellectual: the mirage of “nature” in European culture.
On the American Revolution and its impact on opinion in France (and
elsewhere), I shall not dwell at length. This is the best-known aspect of
the history of ideas about America, and it has also become a topos for a
certain type of Franco-American rhetoric. The political and military
events that marked American independence crystallized images of
America that emerged on different levels. For enlightened opinion, these
events embodied both the political ideas of the philosophes (conveyed in
particular by the Quakers) and the agrarian state envisaged by the
physiocrats—a fusion of which Benjamin Franklin was the living symbol.
But, on a broader level, no doubt—and especially in France, in the
climate of anglophobia—the accession of the thirteen colonies to nation-
hood destroyed forever the image of a savage continent and integrated
America into the history of Europe, that is, into history tout court.
At the same time, the concept of ‘‘savagery,”’ with reference to the
new nation, gave way to the concept of nature. Actually, since the
Renaissance, the concept of “‘savage’’ in European culture had never
ceased to be ambiguous. It referred either to man bereft of the benefits of
civilization or to man in the state of nature that preceded the social
156 America and the Idea of Democracy

contract. In this second usage, the savage embodied primitive natural


law, whose principles had been corrupted by social life and had to be
revived and respected by the social contract. The savage was a matrix for
true civilization. For Rousseau, this civilization had degenerated by
virtue of its very development (arts and sciences, luxury, and pleasure);
for most other authors, civilized man remained the symbol of perfection,
but the barbarian replaced the savage as his antonym. Thus, in the
tripartite scheme being developed in the eighteenth century—the savage,
the barbarian, and civilized man—the savage represented the origin of
mankind (in the temporal sense and in the principial sense), civilized man
the end of his history, and the barbarian the cyclical corruption of this
history. The savage, if he emerged from the natural order without having
been corrupted by the Europeans, combined the benefits of nature with
those of history and served as an example to mankind.

The French View of America

It is precisely this fundamental notion of a natural order turning into a


historical order—that is, of a history for once congruent with nature—
that the young American nation came to embody in the eyes of enlight-
ened eighteenth-century opinion and more particularly, owing to the
Franco-American alliance, in French opinion.
First, America had to be removed from the category of “travel”
accounts—from the inventory of space—and accede to the higher, tem-
poral order of history. This shift was brought about by the War of
Independence and the break with England. This war in itself already had
a familiar ring for Europeans. But, because it was waged against a
European power (England) with the help of another European power
(France), it was a doubly effective agent for integrating America into the
exclusive circle of “historical” nations. Finally, the war was the high point
in the birth of a nation—and the nation was the supreme form of civilized
societies. On all these counts, the war constituted a historical entity, that
is, the description of a memorable event and the testing ground for a
primordial experience.
Accounts of the American war proliferated in France, especially as the
participation of the French and the victorious outcome of the conflict
provided national pride with revenge on England and on the 1763 treaty.
On this elementary level, the history of France provisionally identified
itself with that of the thirteen colonies. The high points in the progress
toward American independence—the struggle with Parliament over
taxation, the Boston Tea Party, the Declaration of Independence, the
major military battles—found their way very early into French histori-
ography: a common fund of reminiscences was being constituted at the
From Savage Man to Historical Man 157

same time as a corpus of full-fledged historical material. The earlier


period of Franco-Indian collusion against American settlers was totally
forgotten.
This common fund relied—even more than on military victory or
anglophobic glee—on philosophical ideas. For educated opinion in
France, independent America embodied an alternative society and an
alternative state, a world whose presence was so keenly felt only because
America existed by default, as an imaginary recourse against what was
soon to be called the ancien régime and as a compensation for the
impossibility of destroying or transforming that old order. America did
not make a sudden and glorious appearance on the stage of history
merely in the role of a new nation; its chief role was that of a model
country, which, because of its youth, its expanse, and the circumstances
surrounding its birth, had succeeded in escaping from the curses of the
Old World. America was both a Europe and an anti-Europe: a Europe
because it was founded on the ideas of the Enlightenment—a language
common to both sides of the Atlantic—and an anti-Europe because its
inhabitants had precisely fled from religious and political intolerance in
order to invent a new historical memory, free from persecution and
injustice. The sentimental attachment of the French to the fledgling
United States was thus a disguised way of expressing concern with domes-
tic politics, one of the manifestations of the vast drive of civil society
toward political power that characterized the last years of the ancien
régime. The ‘American discourse”’ that flourished in those years was in
fact a discourse on French society and on the French Constitution.
As this is one of the best-known aspects of this history, I shall not dwell
on it here. From La Fayette to Chastellux, from Mirabeau to Brissot,
from Mably to Condorcet, from subtle argument to polemic, the com-
mentary on the young United States became a new discourse on human
equality and on the right way to govern societies. America, which had
embodied the infancy of mankind, had not yet become mankind’s future
but was already an example to it. America existed primarily as a semi-
imaginary, semi-real phenomenon discussed by the French in a prefigura-
tion of the debates on their revolution. It is especially striking to see how
the mechanism of rhetorical escalation in the name of democracy—a
process that was to become so important during the struggles for power
between 1789 and 1794—was already at work, though as yet without a
political dimension, in discussions about America. Brissot’s critique of
the book by the marquis de Chastellux is a case in point: it reads like a
girondin attack on the feuillants!’ The unfortunate marquis, despite his
having fought for America, was soundly rebuked for having contrasted
the American people, who enjoyed a freedom equal to that of ancient
republics, with the people of Europe, more inclined toward inequality
and servitude:
158 America and the Idea of Democracy

What is clearest in my mind is that your argument is far more


favorable to the cause of mankind’s true oppressors; for by
telling them that the people today are neither worthy nor
capable of freedom, that men and nations are degenerating at
an increasing rate, and that this degeneration would not cease
even if the circumstances that provoked it were to disappear—
by imparting to them these alleged truths, you will encourage
them to keep their slaves in irons.
But nothing is more contrary to the truth. The American
Revolution proves it. Those who made the Revolution de-
scend either from Englishmen tyrannized by the Stuarts, or
from enslaved Germans.
You think you are doing the Americans a great honor by
likening them to the Romans and Greeks; actually, I think
they are far superior to these ancient peoples. But this is not
the place to prove it. I shall limit myself here to advancing this
proposition, whose demonstration will be found elsewhere, in
order to show you that men today, far from being degenerate,
can, in the right circumstances, surpass their ancestors.*
But independent America was also, and at the same time, a laboratory
for constitutional experiment. The discussion about principles and the
heated arguments concerning the best possible society were to lead to
considerations on the mechanism of power: the history of America was
an object lesson. The debates that later took place in the Constituent
Assembly originated in the debates on the government of the United
States. On the subject of voting rights and eligibility, for example, the
French debates of 1789-91 contributed nothing beyond what was already
contained in a book by Mazzei, a friend and correspondent of Jefferson.
Mazzei’s work, published in Paris in 1788, was probably the best compen-
dium of historical and political information on the United States to
appear in its time. Between the American described by Cornelius de
Pauw and the one described by Mazzei, a nation and a people had entered
history.
But the redemptive character of this history stemmed from the fact
that it was inseparable from nature. Until its appearance, the culture of
the Enlightenment had oscillated between two incompatible beliefs: the
superiority of history over nature and the corruption of nature by history.
The young United States came as the unexpected reconciliation between
human industry and fidelity to nature; America was the first to give an
impulse, in this laicized form, to a feeling that was destined to a great
future: revolutionary expectation.
No book demonstrates this better than that of Crévecceur, a Norman
nobleman for many years an expatriate.° Letters of anAmerican Farmer,
written in English and translated into the French style sensible then
fashionable, gave a Marie Antoinette-like description of the “frontier”
From Savage Man to Historical Man 159

epic. But the agrarian idyll of the American smallholder was considerably
more than an additional variant of the old theme of natural happiness.
Far from being a sort of exotic reference lost in space—that is, harking
back to a very ancient past—it inaugurated a new historical epoch and
constituted a promise. The union of the thirteen colonies represented “‘an
unforgettable moment in the annals of the universe. This moment can be
regarded as a rebirth of nature, as a new gift it has bestowed on the Old
World, as a second creation; for everything we see today bears the stamp
of youth and is only just unfolding.’”’
Thus, America was a new origin of humanity, but a thoroughly histor-
ical origin, since it was contemporary. This logical paradox was possible
because the American homo novus brought about a radical change in
European man—a rebirth at once geographical and cultural. It was
geographical because it was initially a consequence of the availability of
space, in contrast to the ‘‘overcrowding” of European societies, but it was
especially a consequence of man’s return to the earth, the mother of
moral and civic virtues. Crévecceur’s American foreshadows Robespierre
and Saint-Just’s “patriot,” who also owned a parcel of his homeland’s
soil. Landownership and agriculture enabled man to be at one with
nature and with history. Indeed, the man of the “frontier,” living on the
outer limit of cultivated land, was for Crévecceur a sort of transitional,
ambiguous figure, caught between the virtues of new man and the two
sources of corruption threatening those virtues: the European heritage of
violence, conveyed by every newcomer, and contamination by the savage
state, which wrested the settler from agriculture and drew him toward
hunting.
For only educated and enlightened men could fully enjoy the blessings
of space and land. The United States was doubly fortunate in having been
peopled by European dissenters, fleeing from intolerance, prejudice, and
persecution, and in having been born precisely in a later age, enlightened
by science and philosophy. Far from being, like the Europeans, burdened
with and corrupted by the weight of the past, Americans were predis-
posed by that very past to rid themselves of its inertia, since they were the
first to have fought and escaped its violence. Their relationship to history
was in the nature of a rebirth, in which their late appearance on the
American scene made them benefit fully from the progress of knowledge
and reason. It is this shortcut that brought them back to the state of
nature, which, instead of acting as an abstract reference, thus became a
historical period, the rebirth of society on a rational basis. Provided it
remained faithful to this exceptional origin, America was destined to be
the central laboratory for human history. It was thus to America, and not
to Greece and Rome, that the “‘learned men” of Europe were invited to
turn in order to acquire new knowledge—that is, to the “primeval germ
... of an enlightened and new people,” since it is “more enjoyable to find
160 America and the Idea of Democracy

oneself at the origin of things than at a sad review of the fragments of the
past.”
In this happy history, what was the role of the two figures excluded
from the new society, the Negro and the Indian? The Negro was either
happy under the patriarchal authority of a good master or on his way to
emancipation. He had no real existence in a world built on equality;
indeed, there was almost no conceptual room for him. The Indian played
an ambiguous but crucial role. On the one hand, as a primitive, he
bestowed nature’s blessing on the settlers. While de Pauw had regarded
the settler as being trapped in the brutish state of savages, for Crèvecœur,
on the contrary, the settler imitated their simple and natural virtues. On
the other hand, although the Indian was untainted by European corrup-
tion, he embodied a society inferior to that of the settlers: Indian society
was fragile and unenlightened; it was based on hunting and had no
agriculture, arts, or sciences. Europe represented history without nature.
The Indian embodied nature without history. Only the American recon-
ciled the two principles.
Therein lies the secret, in my view, of the assimilation of the ‘ Amer-
ican dream” into French culture. Before the outbreak of the French
Revolution, which acted as the focal point for the belief in a new begin-
ning—a regeneration, as the French were to put it—of mankind, it was
American independence that crystallized the idea of a history-as-origin,
by means of which society would once again conform to nature and
reason. That is the deep connection between the two revolutions. The
idea of revolution in its 1789 sense originated for the French in the birth of
the United States.

The End of the ‘‘American Dream”?

But this link also explains precisely why the French Revolution put an
end to the “American dream” in French opinion. Not that America
ceased to exist as a political and philosophical point of reference. The
American example was still cited in revolutionary speeches and newspa-
pers, particularly in connection with constitutional problems. A system-
atic study’ would probably show that references to America were specifi-
cally made by the girondins, and in particular by Brissot and Condorcet,
who remainded faithful to their youthful passion. But that the ‘ American
discourse” should have been a feature of the most cosmopolitan intellec-
tual and political family of the Revolution—the only family, in fact, with
any “international” character at all—already indicates that this discourse
had become quite alien, if not suspect, to a great many actors in the
French drama, from the members of the Constituent Assembly to the
montagnards. |
The reason is that from 1789 on, the French had their own dream,
From Savage Man to Historical Man 161

which made a substitute unnecessary. From the very onset of the Revolu-
tion, they were obsessed with their own example, narcissistically fasci-
nated by their own history. If one reads the countless pamphlets that
flooded the old kingdom between 1787 and 1789 or if one examines the
interminable and monotonous evidence of the cahiers de doléances, one
will find few references indeed to foreign examples. Tocqueville, who was
well versed in this literature, had understood the phenomenon:

Among all the proposed schemes that had just blossomed at


the time the government seemed to put the Constitution up
for contest, one finds hardly a single one that deigns to imitate
what was happening abroad. The major concern was not to
take lessons, but to give them. . . . Thus every single French-
man was convinced that the Revolution would not only
change the French system but also bring to the world new
principles of government that could be applied to all peoples
and would reshape the entire face of human affairs; and every
single Frenchman was also convinced that he held the key not
only to his country’s fate, but also to the very future of his
species."

Accordingly, the American experience lost the central historical posi-


tion it had occupied since independence. American became once again
peripheral. It continued to represent the opportunity afforded by a tabula
rasa and by an absolute beginning, but for that very reason it became
relatively alien to the French revolutionaries, whose primary concern was
to articulate the new nation’s connections with its own past. And the two
modes of dealing with this past—compromise and refusal—removed
America from the French Revolution.
The moderate revolutionaries understood this at once, and Mounier,
on 9 July 1789, provided the best statement of their position in the course
of the first debates on the Constitution:

We shall never relinquish our rights, but we shall not use them
to excess. We shall not forget that the French are not a new
people just emerged from the depth of the forest to form an
association, but a great society of twenty-five million men—a
society intent on strengthening the ties that exist between all
its parts, and on regenerating the kingdom, a kingdom for
which the principles of genuine monarchy will always be
sacred."

Mounier, like Mirabeau somewhat later, and like so many members of


the Constituent Assembly, was hoping for a compromise with the king,
that is, with the country’s past. On this point, the United States of
America could offer Mounier no guidance. The Revolution was a face-to-
face encounter between the French and their own history.
162 America and the Idea of Democracy

But as the Revolution unfolded, entering into conflict with the Catho-
lic church, waging war on the kings of Europe, and sending Louis XVI to
his death, it became an expression of the French people’s rejection of
their own history. There are two reasons why the Jacobin break with—or
eradication of—the nation’s past was incompatible with an admiration of
the American example. First, Jacobinism was predicated on a total
psychological and ideological commitment to the French Revolution,
now the sole embodiment of liberty and equality, of the great beginning
of human emancipation. There was no room in Jacobin thinking for two
pilot nations. Second, the secret of social happiness rediscovered by
France was no longer thought of as a gift of nature but as the stake in a
domestic and international conflict. In America, equality and indepen-
dence resulted from consensus. In revolutionary France, they were seen
as the outcome of a long and bloody conflict. This, it seems to me, is
where the two myths of origins parted ways and made the political and
cultural life of the two nations so alien to each other after several years of
considerably fraternal relations. Furthermore, from 1793 on, it is what I
would call the “conflictual messianism”’ of French Jacobinism that, in its
turn, had an impact on American politics via the Republicans. However,
Jacobinism failed to constitute an effective ideological model for Amer-
ica, for the two experiences had become as remote from each other as the
physical distance that lay between them.

America, the Observation Ground

Even as America ceased to be a myth for the French Jacobins, the


shattering of the revolutionary political illusion after 9 Thermidor led the
survivors of the terrorist adventure to a critical reexamination of that
illusion. The Thermidoreans, as exemplary heirs to the Enlightenment,
put the burden of blame not only on Robespierre’s Supreme Being but
also on the American dream, which had fed their hopes during the
Constituent Assembly. Back in 1791 and 1792, liberals such as La Roche-
foucauld, Démeunier, and Talleyrand, disappointed by the turn of events
in France, had forsaken the homeland of freedom for a rough and difficult
America, where they led an unhappy and, incidentally, not always blame-
less existence. After Thermidor they returned fairly promptly to France,
where they became additional witnesses to the end of the revolutionary
dream on both sides of the Atlantic. For the moderates, too, America
was no longer a dream; it was a friendly but foreign country. And for the
intellectuals among them, such as the Idéologues, America became an
object for dispassionate study.
A representative of this attitude is Volney, former member of the
Constituent Assembly, who sailed for the United States in the year III
(summer 1795). Volney was a seasoned traveler and scientifically minded
From Savage Man to Historical Man 163

observer, a pioneer in the use of questionnaires and social statistics, and a


member of the Institut—of the academy of Idéologues who became the
spiritual mentors of the Thermidorean republic. Less than ten years had
elapsed since Brissot’s visit, yet how the tone had changed! Volney was
the first Frenchman to visit the United States not in search of nature or
freedom, or the two together, but in order to study one of the manifesta-
tions of historical man. In passing, Volney counseled diplomatic caution
to his Parisian friends, at a time when the Directory wanted to concede
Louisiana to Spain. But his journey was above all that of a postrevolu-
tionary intellectual: America had entered the social science of the Idéo-
logues. Consequently, it was no longer a model, but a varied ground for
observation:

Roughly speaking [wrote Volney to a friend on 14 January


1797], my itinerary has included Virginia, Kentucky, Wabash,
Fort Detroit, Lake Erie and some of what is called here the
North-East: Genesee country, Mohawk, and the Hudson; in
particular, I have been able to sample the four most diverse
areas of this great continent. In the South, I have seen the
slave system and its moral effects; in the West, the condition
of a nascent society; in Canada, Frenchmen of the age of
Louis XIV who have become half-Indian; in the East, En-
glishmen of the last century—already an old nation, the only
one here to have a developed character. That, my friend, is
the knowledge I have acquired for you and for our Institut
National.”

Thus, Volney’s America had ceased to be a unique and comprehensive


image of the human condition. It was fragmented into distinct objects of
knowledge, according to the character, history, and situation of the
various societies that it accommodated. America was no longer a focus of
ideological commitment but a mosaic of historical experience.
The only book that Volney actually wrote as a result of his American
travels was Tableau du climat et du sol des Etats-Unis (1803), a study
confined to physical geography. Of the projected second part of the
book—which was never written and was to have been devoted to the
population, mores, and history of the United States—Volney left us a
general outline, the preface, and some detailed analyses in the appen-
dixes.
Volney’s general argument was that the young United States was not
free from the drag of European history, in which it had originated. Hence
a frontal attack against the founding myth of origins, constructed and
kept alive by certain writers of the 1780s. Volney had specifically planned
to study—after the climate and soil—‘‘the size of the population; its
territorial distribution; its composition according to work and occupa-
tion; the habits, that is, the mores, resulting from these occupations; the
164 America and the Idea of Democracy

combination of these habits with the ideas and prejudices dating from the
earliest origins.’ There follows this crucial passage:

By retracing these origins through history, language, laws,


and customs, I exposed the fanciful mistake of writers who
described as a new and virgin people a gathering of inhabitants
of old Europe—Germans, Dutch, and above all Britons from
the three kingdoms. The organization of these traditional and
diverse elements into political bodies led me to outline the
history of the formation of each colony; to show that the
character of their founders contained the spiritual ferment
that acted as the driving force for almost the entire pattern of
behavior of their successors, in keeping with the insufficiently
recognized moral truth that, in corporate bodies as in indi-
viduals, the earliest habits exert a predominant influence on
the remainder of their existence.

In other words, the British American remained a Briton, the German


American a German, the French Canadian a Frenchman, and so on. The
origin of the United States was no longer the blessing of liberty bestowed
on a new people but the weight of European history, that is, the various
“spirits’’ (in the sense in which Montesquieu spoke of the ‘spirit of laws’’)
of nations and peoples as constituted by their respective traditions.
Thus, for Volney, America had lost its political superiority and its
exemplary value. Instead, America had a history in which, as in the
European history it reproduced, founding principles were corrupted by a
combination of money, ambition, and passion. The American Revolu-
tion, exactly like the French Revolution between the Constituent Assem-
bly and the Convention, had regressed from its initial promise to a
situation of strife and civil war.

I should then have examined [Volney went on, describing the


section he never wrote] from a moral point of view the con-
duct of this people and of its government, from 1783 to 1798;
and I should have proved by incontrovertible facts that, in
proportion to the population, the volume of business, and the
multiplicity of combinations, there existed in the United
States neither a greater restraint in financial matters, nor a
greater degree of good faith in transactions, nor a greater
decency in public morality, nor a greater moderation in the
partisan spirit, nor a greater care displayed in upbringing and
education, than in most of the states of old Europe; that
whatever good and useful achievements the United States has
to its credit, whatever has been accomplished in the realm of
civil liberty, and in protecting persons and property, has been
due more to popular and individual habits, to the need to
work, and to the high cost of all forms of labor, than to any
From Savage Man to Historical Man 165

ingenious measure or policy devised by the government; that


on nearly all these counts, the nation has regressed with
respect to its founding principles; that in 1798 only the ab-
sence of a particular set of circumstances prevented a certain
party from usurping power and resorting to violence of a
totally counterrevolutionary character—in a word, that the
United States has owed its public prosperity, the affluence of
its state and its inhabitants, far more to its insular position, to
its distance from powerful neighbors and from all theaters of
war, and, finally, to the general ease of its situation, than to
the fundamental goodness of its laws or to the wisdom of its
administration.

Thus, the United States came once again under the common-law
jurisdiction of history. The American Revolution was subject to the same
erosion of principle by passion as the French Revolution had been, and it
became the object of the same disenchantment. Whatever advantage still
accrued to America was due solely to geography and chance.
In order to complete his demonstration, Volney also had to demolish
the last American myth: that of a harmonious and happy savage state,
enshrined in the natural order before the institution of property. In short,
Volney had to exorcise Rousseau. He did so in one of the very few
fragments of this part of his book that he actually wrote. These fragments
were published as appendixes to the Tableau under the title Observations
générales sur les Indiens ou Sauvages de l’Amérique du Nord. Volney’s
savage no longer played the leading role on the philosophical stage where
the natural order and civilization confronted each other. Although, as in
Rousseau, the savage predated the creation of property, he ceased to be
an abstract entity a la Rousseau and became the object of empirical
observation. Volney’s savage embodied not a principle but a historical
society, whose weakness lay precisely in its having predated property and
agriculture; consequently, it was vulnerable to the slightest contact with
“civilization.”’ The savage was neither a utopian creature, as in Rous-
seau, nor a foil, as in de Pauw. Savagery was a primitive society, a
still-rudimentary history, destined—condemned, even—to become civi-
lized through the introduction of property. Thus, it was one of the several
elements in American history, but a secondary element, set for assimila-
tion or destruction, and subject to the merciless law of development.
Savagery had been removed from the poetic world of origins and re-
turned to the prosaic world of social life. By losing its ‘savages,’ America
was all the more thoroughly reabsorbed into the ordinary flow of history.
By achronological coincidence, at almost the same time, or just a short
while earlier, the greatest of all French writers went—or pretended to
have gone—on the same journey as Volney. Actually, it is of little
importance that the itinerary was for the most part invented, for
Chateaubriand introduced America, not into French philosophy or poli-
166 America and the Idea of Democracy

tics, but into French literature—and with what a stir! Atala and René—
together with Les Natchez, published later but cast in the same mold—
revolutionized the novel and aesthetic sensibility. Admittedly,
Chateaubriand incorporated traditional descriptive elements such as the
good savage, the antithesis between savagery and civilization, and the
luxuriance of nature; but he added the feeling that Europe would never
again move out of its confines, that it was a prisoner of its past, and that
the American dream of happiness lay beyond its grasp. Of all the imagi-
nary travelers in search of the new man, René was the first to experience
the burden of time and the impossibility of casting off the old man. France
under the Consulate shared in this feeling. All it discovered in the
American forest was the “‘genius of Christianity’ —its own past.
Thus, between 1780 and 1800—twenty years during which French
history moved at such a rapid pace—the American dream shifted from
revolutionary messianism to social science and met its death in literature.
The myth of a reborn humanity at last faithful to its origins—a myth so
enduring in American culture—did not survive in France after the Jaco-
bin experience. The French replaced it by the myth of an egalitarian
nation, composed of both peasants and soldiers, a nation engaged in an
endless battle to deliver mankind from an invincible past. I do not feel
that, since this period, the two nations have ever again experienced the
ephemeral sense of cultural community that had brought them so close
together for ten or fifteen years. That is perhaps why these ten or fifteen
years are being celebrated with such loquaciousness and such solemnity.
Deep ties are generally silent.
The Conceptual System of
“Democracy in America”

The genesis of Tocqueville’s visit to American is shrouded in mystery.


When did the idea first occur to him? When did the project materialize?
And why America?
Neither the known facts nor the available documents provide a satis-
factory answer to these basic questions. The facts are clear, but they
illustrate only the secondary aspect of the American voyage: the mission
to study the American penal system. When Tocqueville and his friend
Beaumont sailed from Le Havre in April 1831, the two young magistrates
had been entrusted with the mission of examining American penal
institutions.’ This mission, requested by the two magistrates themselves,
was unpaid but official; a formal report was duly presented to the author-
ities and later published.’ But this investigation, whatever appeal it may
have had for Tocqueville (who displayed an abiding interest in French
prison reform), was obviously, on the intellectual level, only a side aspect
of his momentous visit.
The available sources do not provide irrefutable evidence regarding
his deeper motives. For example, the correspondence between Toc-
queville and Gustave de Beaumont contains only one allusion to the
matter in a letter of Tocqueville’s dated 14 March 1831, that is, shortly

First published as a preface to Alexis de Tocqueville, De la démocratie en Amérique (Paris:


Flammarion ‘“‘G.F.” paperback series, 1981), 1:7-46. References to the original of De la
démocratie are to this edition. References to Tocqueville’s other writings are to the Œuvres
complètes, in progress (Paris: Gallimard, 1951-). [ The passages from De la démocratie
quoted here are taken from Alexis de Tocqueville, Democracy in America, ed. J. P. Mayer,
trans. George Lawrence (Garden City, N. Y.: Doubleday, 1969). References to this edition
are given in brackets and preceded by the letters GL.—TRans.] I should like to thank my
friend André Jardin, the foremost contemporary specialist on Tocqueville, for having read
this essay in draft and having made many valuable suggestions.

167
168 America and the Idea of Democracy

before his departure.’ Moreover, this letter mentions only the incidental
reasons connected with the 1830 revolution—an event that placed the two
applicants for the mission, as scions of legitimist families, in a “difficult
position.” Besides, even if one were to accept this type of “diplomatic”
explanation, why America? Many other countries could satisfy the
curiosity of two friends and provide a legitimate excuse for their absence.
At the time, the young American republic represented the ideal for an
intellectual orientation quite alien to the tradition in which the two young
aristocrats had been brought up: the orientation of the liberal opposition
under the Restoration—liberals of every hue, freemasons, and republi-
cans, all assembled in a pro-American camp under the symbolic aegis of
La Fayette.’ But it is true that Beaumont was distantly related to La
Fayette and that in family circles the young Alexis had met former
“Americans” such as Chateaubriand, Hyde de Neuville—the agent of
princes during the Revolution, former ambassador to Washington, and
close friend of the comte Hervé—and Monseigneur de Cheverus,
archbishop of Bordeaux and former bishop of Boston.
If Tocqueville and Beaumont were searching for the secret of free
institutions, they could have gone to study them in countries less radically
alien to their tradition and more spontaneously dear to their hearts—
Switzerland and England in particular. But Switzerland, according to
classic political theory, was considered a republic only in consequence of
its smallness; England, then regarded as being on the verge of collapse,
was not, in any case, a democracy. However, Tocqueville did visit En-
gland in later years, and his correspondence from the pre-1830 period,
even as he was attending with passionate interest Guizot’s course at the
Sorbonne, shows his concern with the comparative history of France and
England.® Why, then, America?
To this question, at least, there is an answer provided by Tocqueville
himself, well after his return, in a letter of January 1835 to his friend
Kergorlay, written just after the publication of the first volume of Democ-
racy in America.’ Tocqueville begins by stating that, since the march
toward equality is inevitable, the central problem of the age is to deter-
mine whether equality is compatible with freedom; he adds:

Thus it is not without having given the matter much thought


that I resolved to write the book that I am now publishing. I
am not at all blind to the awkwardness of my position, which
will not elicit anyone’s heartfelt sympathy. Some will find that
basically I have no love for democracy and that I criticize it
severely, others will think that I am recklessly encouraging its
development. The best thing for me would be if no one read
my book; and perhaps I shall be so lucky. I know all this, but
here is my reply: it is now nearly ten years that I have been
reflecting on some of the matters I shall tell you about later on.
Democracy in America 169

My only reason for going to America was to clarify this point.


The prison system was a pretext: I used it as a passport to gain
admittance everywhere in the United States. In this country,
where I came across countless unexpected objects, I gleaned
several insights into the questions I had so often asked myself.

“Tt is now nearly ten years. . . .” Tocqueville wrote these words in


1835, and he was born in 1805: he was therefore about twenty years old
when he conceived the question that was to lead him to America and,
more generally, to inform his entire intellectual and political life. This is,
it seems to me, a very rare instance in intellectual history—a system that
crystallized so early, in the mind of a young man brought up, moreover,
in a confined environment and hardly trained in any subject other than
law. One is naturally reminded of Sainte-Beuve’s remark: “‘He began to
think before he had learned anything.’’* Repeating the idea in a different
form, one could say that Tocqueville is an extreme example of an intellec-
tual who never “learned” anything outside of a conceptual framework
that he had developed beforehand. Apart from the time and energy
saved, this gave him exceptional narrowness and exceptional depth:
Tocqueville recorded nothing haphazardly, for the simple pleasure of
knowledge. The visit to America, like the history of France or England,
was an element in the systematic experimentation carried out by this
deductive mind. It remains to be explained why Tocqueville proceeded in
this manner, and one must therefore retrace the antecedents of his
“thinking.”
I would argue that if his “system” was constituted so early, it was
because even in its explicit aspect it rested on a base that was not intellec-
tual, but purely existential. Tocqueville belonged to the world defeated
by the French Revolution—an event from which he deduced, as all his
generation did, the feeling of the irreversible march of history. But, as he
was given to abstraction, he gave the well-known romantic notion of
“fate” the form of a concept, drawn directly from the experience of his
milieu. This concept was the victory of the democratic principle over the
aristocratic principle. Tocqueville’s entire work can be regarded as an
endless reflection on the nobility.
This reflection was his point of departure, both existentially and intel-
lectually—a meditation begun in adolescence and focused on himself, his
family, his life, the historical significance of what his parents had lived
through and what he himself was living, or reliving, through the failure of
the Restoration and the events of 1830. For his father, Hervé de Tocque-
ville, saved from the guillotine by 9 Thermidor, the same questions had
been an ever-present preoccupation. In 1847, he even published a histor-
ical study of the causes of the Revolution, under an antiquated title that
testified the generation to which he belonged.’ A disciple of Montes-
quieu, the old count centered his analysis on the worsening relations
170 America and the Idea of Democracy

between absolute monarchy and nobility and on Louis XV’s incapacity to


adapt the regime to the aristocracy’s demands for liberalization:

Richelieu and Louis XIV imposed absolute authority to the


detriment of public liberties because the nation was weary of
the strife that had bloodied earlier reigns. Louis XV mis-
judged the spirit of his age. Words of freedom were repeated
everywhere: they echoed under the vaulted ceilings of the
courts of justice and were uttered even by courtiers. The weak
hand of a discredited monarch could hardly sustain the edifice
built by Louis the Great. The Revolution had already made its
appearance in the upper class; it was gradually to descend to
the lower class."

Thus, Hervé de Tocqueville, in order to understand what he had lived


through, invoked the “spirit of the age,’’ embodied by the nobility and
misjudged by the monarchy. From his cradle, Alexis encountered, as an
integral feature of his milieu, this questioning about the great historical
drama experienced and conceived as inevitable and yet as being linked to
two specific agents: the nobility and the king of France.
This questioning had lost neither vigor nor relevance. Since 1815, the
restored monarchy had actively fostered it by pressing the fight against
the “‘spirit of the age” in the name of the restored alliance between the
king and his nobility. This increasingly aristocratic monarchy led to the
Journées of July 1830: the French Revolution continued. And it is no
coincidence that this date crystallized Alexis de Tocqueville’s deep
choices by revealing the manner in which he invested his inheritance in
new issues and new reflections. The affair of the oath to the new king—an
act made compulsory by the law of 31 August 1830—turned Tocqueville
into something of an outcast from his milieu. His closest friend, Louis de
Kergorlay, left the army and was later involved in the adventure of the
duchesse de Berry. Tocqueville, instead, took the oath, without joy (‘‘it is
an unpleasant moment,” he observed in a letter), but also without any
real pang of conscience. He merely regretted that his gesture might be
interpreted as having been dictated by self-interest, whereas in fact it was
a token of resignation. And the decision to visit America (even if this long
absence may have been calculated to efface the memory of the somewhat
awkward position in which the affair of the oath had placed Tocqueville
with respect to his milieu) reveals the same indifference, but on a theoret-
ical level. For Tocqueville’s decision was connected with an already
constructed intellectual system, which gave a powerful new impetus to his
father’s query, stripping it of the conventions of tradition and nobiliary
rhetoric.
In this system—about whose development we know next to nothing,
Democracy in America 171

since Tocqueville was already Tocqueville at the age of twenty—it mat-


tered little whether the reigning dynasty was legitimate; indeed, it mat-
tered little whether there was a dynasty at all. The central question did
not concern the relations between nobility and monarchy; it concerned
the compatibility between nobility and democracy. Out of the three
diverse elements that his milieu used to explain historical misfortune—
monarchy, nobility, spirit of the age—Tocqueville built a very simple,
two-dimensional system. He kept nobility as one pole: it was the neces-
sary starting point, the primordial social experience, the taproot of his
theory. As a type of government, of society, or of culture, the “‘aristoc-
racy’ was a model for the nobility. At the other pole, the heir to a
defeated principle placed the victorious principle of democracy, at once a
government by the people, an egalitarian society, and, to borrow his
father’s vocabulary, the “spirit of the age.”
This intellectual process did involve a certain degree of fatalism, an
acquiescence in the inevitable that corresponded to the historical experi-
ence of Tocqueville’s milieu: it was the march toward an ever more
complete democracy that defined the direction of the trend revealed by
the French Revolution. But Tocqueville was not searching for the causes
of this trend, at least not at the time. Unlike Marx, for example, who
thought that one could demonstrate the march of history and that the end
of capitalism could be deduced from the economic laws that governed it,
Tocqueville posited as an axiom, or a self-evident truth, that mankind
was moving in great strides toward the age of democracy. This conclusion
was not the outcome of a systematic inquiry; it was merely the abstract
expression—in keeping with the nature of Tocqueville’s genius—of the
actual experience of Tocqueville and his milieu. Nor was it a new idea
(although Tocqueville played a considerable part in spreading it); one
finds it in many authors of the period, and even in his own milieu (witness
his relative Chateaubriand). But Tocqueville, having established it as a
starting point, was the only one who sought to develop the idea and
explore all of its aspects. For him, the idea was to be understood at
various levels—cultural social, and political—but only the cultural and
social levels defined that which was inevitable. While the societies of his
day seemed to him to be drawn by a sort of fate toward an increasingly
widespread belief in equality, and toward a greater equality of condi-
tions, the political forms that could accompany this evolution continued
to depend on human choice. Thus, the problem that came to dominate his
entire intellectual life was not so much that of the causes of equality as
that of its consequences on political civilization. Once again, this
approach and this formulation of the problem placed him at the opposite
pole from Marx. Marx was interested in the laws of economic structure
and the relationship between economics and society, from which he
tended to ‘“‘deduce” political phenomena. Tocqueville explored the con-
172 America and the Idea of Democracy

nections between the principle that governed societies and the type of
political system that could issue from it, but he never regarded those
connections as inevitable.
Thus, Tocqueville constantly combined two types of analysis and two
types of beliefs. In his reasoning, he juxtaposed a logic of typology—
based on the opposition between aristocracy and democracy—and a logic
of evolution, based on the inevitable triumph of democracy. As for his
general world view, he compensated for his rational acceptance of
democracy by fighting for the values that were inseparable from the
aristocratic world, freedom foremost. His entire life, from an early age,
revolved around this problem—a blend of theory and experience—or
around this theory of family experience, which constantly combined facts
with values, conceptual richness and exhaustive documentation with
political convictions.
If the visit to America was already implicit in this exploration, it was
because America provided the young, systematically minded aristocrat
with an existential and conceptual laboratory. A homeland founded on
the negation of the nobility (a country, therefore, where Tocqueville
himself could not have existed), America offered the example of a
chemically pure experiment in democracy. Tocqueville used this discov-
ery—truly a brilliant one in its simplicity and boldness—to verify and
elaborate on an idea. He must have had a premonition, when he set sail,
of the great secret that he later shared with Comte Molé on his return
from his second visit to England in 1835, in a letter that belatedly answers
the question of 1831, Why America?

One would have to display considerable philosophical pre-


sumption to imagine that one could judge England in six
months. One year has always seemed to me too short a time in
which to appreciate the United States properly, and it is
infinitely easier to gain clear ideas and accurate notions about
the American Union that about Great Britain. In America, all
laws issue so to speak from a single idea. All of society, as it
were, is founded on a single fact; everything flows from one
principle. One could compare America to a great forest witha
myriad of straight roads built through it, all of them leading to
the same point. One need only find the crossroads and every-
thing will become visible in a single glance.”

Thus, America was the ideal site for analyzing in vivo the democratic
principle at work, that is, the risks it involved and the advantages it
offered to freedom. As for the European nations, they stood midway
between aristocracy and democracy; they were torn by the conflict be-
tween two principles and two worlds and often prey to the extreme form
of democracy represented by revolution. The American example did not
constitute Europe’s future—which was necessarily different—but it gave
Democracy in America 173

Europe some basis for conceiving its future so as to ensure the most
favorable conditions and the least number of obstacles to freedom.
Tocqueville was prepared to resign himself to the end of the nobility,
provided the aristocratic legacy of freedom were capable of surviving the
democratic age.
Actually, in order to understand the close connection that exists
between Tocqueville’s visit to America and his analysis of France, one
can also look directly at Democracy in America, which is entirely aimed at
a comparison between America and Europe. The most explicit text on
this point is the end of chapter 9 of part 2 of volume 1. Tocqueville begins
by examining the importance of laws and mores in preserving American
democracy, as opposed to what he calls the ‘material causes,” that is, the
specific features of the New World and the advantages it offers to man in
his relation to geographical space. Tocqueville is confronted with a classic
problem—perhaps the central problem—of social sciences: how to iso-
late the role and influence of a variable or of a limited set of variables on a
complex process. That he was well aware of the difficulty is proved by his
attempt to find a point of comparison in a country situated outside the
American continent, and therefore deprived of the geographical advan-
tages inseparable from America, yet possessing comparable laws and
mores. However, he cannot find any such country. He draws the conclu-
sion that, in the absence of objects of comparison, “we . . . can only
hazard opinions.’’”
This is a typical passage in that it illustrates Tocqueville’s fundamental
mode of reasoning and demonstration: the comparative method. Having
found one or more ideas as explanatory hypotheses for a phenomenon he
was attempting to elucidate, Tocqueville tested them on various
‘“grounds.” When he could not’find relevant grounds for the problem at
hand, he ‘“‘hazarded opinions’’—that is, to his mind, indemonstrable and,
at best, plausible propositions. And America represents one of the poles
of the intellectual two-way journey that forms the fabric of his entire
book.
But, as always in the human sciences, there were no terms capable of
being compared rigorously. Not only was Europe deprived of the
“physical advantages” enjoyed by the Anglo-Americans but its historical
traits made it radically different from the New World. Tocqueville cites
Europe’s large population, its big cities, its armies, and its ‘““complicated
politics.” These legacies would be enough to prevent a transfer of the
laws of American democracy to Europe, since these laws would clash
with other mores, other ideas, and other religious beliefs. For Tocque-
ville, who never clearly distinguished between facts and values, the
United States did not constitute for Europe either a conceptually compa-
rable experience or a model to be followed. For “‘one can imagine a
democratic nation organized in a different way from the American
people.”
174 America and the Idea of Democracy

Yet what confers a universal value to his analysis of American democ-


racy is the existence of a problem common to the American people and to
European peoples; this common problem stems from the fact that men
are not different on the two continents and that they experience in
America and Europe the same characteristic passions of the democratic
social state: impatience with their condition, anxiety over promotion,
yearning for superior status. From this state of mind, Americans have
derived the very nature of their society—its driving force—but they have
channeled it through law, religion, institutions, and mores. The Euro-
pean peoples, gripped as they are by identical social passions, are faced
with the same institutional problem, in the broadest sense: how can these
passions be organized into laws and mores?
This problem is all the more difficult to resolve as the European
peoples are less ‘democratic’ than ‘revolutionary.’ Tocqueville, re-
fining his analysis, makes this point in the second volume of his book (part
3, chapter 21); he thereby introduces an essential distinction that runs
through the entire second volume and serves to explain American con-
sensus and political stability. The democratic social state is shown to be
unpropitious to revolution, owing to the tight-knit fabric of conservative
microinterests that it is constantly creating and maintaining. On the
contrary, it is inequality that leads to revolution; it was to destroy the
aristocratic social state and the ideology of inequality that the French
carried out their revolution. But the passions and state of mind that have
characterized the French since the Revolution have not been conducive
to institutional stability. ‘‘In America, there exist democratic ideas and
passions; in Europe, we still have revolutionary ones.’’*
But even if the European problem is harder to solve than the Amer-
ican one, the fact remains that only a comparison can enable one to define
the elements of the European problem: “The great problem of our time is
the organization and establishment of democracy in Christian lands. The
American have certainly not solved this problem, but they have furnished
useful lessons to those who wish to solve it.”"* The pages that follow this
statement provide the key to Tocqueville’s intentions. His discussion of
the situation in Europe anticipates some of the analyses presented in
L’Ancien Régime et la Révolution twenty-five years later. For Tocque-
ville, what was taking place or had taken place in the Old World was the
creation of historical conditions exceptionally favorable to the establish-
ment of a truly absolute central authority: therein lay the peril that had to
be seen in order to be exorcised. For although the old monarchies were
regarded as absolute, they were not so in practice. Political institutions
(especially corporate bodies and local communities), intellectual and
moral traditions (especially the “family” tie between the king and his
subjects and the aristocratic values of independence and honor), and
religion had prevented these monarchies from becoming absolute.
Democracy in America 175

But these institutions, these traditions, and even religion itself had
disappeared or were withering away, to be replaced by a society in which
classes were less and less differentiated, individuals increasingly alike and
isolated from one another, and public opinion ever more malleable and
indistinct.

[N]owadays, when monarchic honor [writes Tocqueville, in a


tone reminiscent of Montesquieu] has almost lost its sway
without being replaced by virtue, and there is nothing left
which raises a man above himself, who can say where the
exigencies of authority and the yielding of weakness will
stop?”

In other words, the nations of Europe had acquired a democratic social


state and a democratic state of mind without possessing the correspond-
ing institutions or even, in their absence, political and religious traditions
capable of counteracting democracy. This explained why their history
was characterized by the silence of passive and demoralized peoples in
the face of strong and organized governments—a prelude to a situation
comparable to the end of the Roman Republic.

For my part, when I consider the state already reached by


several European nations and that toward which all are tend-
ing, I am led to believe that there will soon be no room except
for either democratic freedom or the tyranny of the Caesars.
Is this not worth thinking about? If men must, in fact, come
to choose between all being free or all slaves, all having equal
rights or all being deprived of them, if the rulers of societies
are reduced to this alternative, either gradually to raise the
crowd up to their own level or to let all citizens fall below the
level of humanity, would not that be enough to overcome
many doubts, to reassure many consciences, and to prepare
each man readily to make great sacrifices?
Should we not, then, consider the gradual development of
democratic institutions and mores not as the best but as the
only means remaining to us in order to remain free? And,
without loving democratic government, would one not, then,
be disposed to adopt it as the readiest and most honorable
remedy against the present ills of society?”

This is to my mind a crucial passage, because it links the visit to


America not only to Tocqueville’s fundamental intention, to the goal of
his life, but also to the internal structure of his thought. Tocqueville—in
the facet of his analysis that he does not make explicit—is a fatalist. He
believes in the inevitable, namely, in the march of societies toward
“‘democracy.’’ This process is common to the Old and the New World,
176 America and the Idea of Democracy

but it appears in its pristine purity only in the American experience.


However, the American people have adopted mores and laws suited to
this social and cultural state, while the European peoples have inherited
centralized states that are in contradiction with the development of
democratic political institutions or democratic national mores. In the first
case, history has subordinated the state to society. In the second, it has
handed society over to the state.
But the second process is not inevitable. This is the second facet of
Tocqueville’s thought, the facet that gives his books an almost militant
character. His aim is to encourage the evolution of the laws and mores of
the old European nations—and above all of France—in harmony with the
material and spiritual inroads of democracy. Such an evolution is indis-
pensable in order to avoid the dictatorship of a single individual in control
of the state. For an aristocrat like Tocqueville, there was a price to pay,
feelings and interests had to be sacrificed, but he gave his prior consent to
this in view of the stakes involved:

The will of a democracy is changeable, its agents rough, its


laws imperfect. I grant that. But ifit is true that there will soon
be nothing intermediate between the sway of democracy and
the yoke of a single man, should we not rather steer toward
the former than voluntarily submit to the latter? And if we
must finally reach a state of complete equality, is it not better
to let ourselves be leveled down by freedom rather than by a
despot?”

Thus, Tocqueville went to the United States in search, not of a model,


but of a principle to study and a question to illustrate and resolve: In what
conditions does democracy, if it is a state of society, become what it must
also be if it is not to lead to dictatorship—a state of government?
Tocqueville’s system, which he developed very early, is essentially
built around a few simple oppositions of which he makes sophisticated
use through a constant dialectic between the cultural, the social, and the
political spheres. On the social and cultural level, there are two histori-
cally conceivable states: the aristocratic state and the democratic one.
The first is inseparable, on the political level, from local government,
while the second leads to centralized government. But here a second,
purely political choice presents itself, for not all centralized governments
are necessarily oppressive. They can either be tyrannical or respect their
citizens’ freedom. On a first level, Tocqueville’s reasoning is constructed
on the opposition between aristocracy and democracy. On a second level,
it revolves around the alternative between democratic Caesarism and
democratic freedom, that is, around the analysis of the conditions for
compatibility betweeen democracy and freedom. This explains why—as
has often been said”—Tocqueville is constantly shifting from the social to
Democracy in America 177

the political meaning of the term democracy, and vice versa, according to
which one of these two conceptual levels he is exploring.
Socially and culturally, American provides him with an example of
pure democracy and of a government derived from this pure democracy.
In both respects, America is an anti-Europe, with no aristocratic inheri-
tance, no legacy of absolutism, and no revolutionary passions. Instead, it
has a tradition of local community freedom. Because of all these features,
mutatis mutandis, America is a crucial topic for Europeans to reflect
upon.

Democracy in America was published in two stages. The first volume,


mainly devoted to an analytical description of American institutions,
appeared in 1835. The second, which studies in more abstract fashion the
influence of democracy on national mores and habits, using America as
an example, was published in 1840. The commentary on the most intelli-
gent journey of the nineteenth century demanded of Tocqueville nearly
ten years of additional study and of intense intellectual work. The major
seminal ideas, particularly those of the first volume (the more specifically
“American” of the two), were already contained in his travel notes—a
proof that the traveler arrived with his “‘system”’ in mind. But, if he took
his time to write, it was not only because of a fondness for stylistic
elegance; it was to “lay out” his object of study—a preparation that
involved reading widely, in particular in the constitutional, political, and
legal fields.” Furthermore, and above all, Tocqueville wanted to “think
through” what he had “learned,” to refine his conceptual scheme with
the help of American material, and to work out the “lessons” it contained
for European peoples. Tocqueville was a thinker who worked over the
same ideas indefinitely and always uncovered new facets of them. The
second volume of Democracy in America is the best example of this type
of intellectual patience.
In addition to its simplifying virtues, America gave him, as a bonus,
even though he was hardly interested in the question, the secret of its
origins. In the New World, democracy was not shrouded in the mists of
time or in the designs of Providence; it was brought by immigrants—New
England Puritans or Pennsylvania Quakers—as a founding religious prin-
ciple for the new homeland. Thus, American democracy had a cultural
matrix, a logical evolution embedded in the history of its origins—a
history that for once was clear and known. Better still, America offered
on its very territory the antithesis of this history: the South, inhabited by a
subaristocracy of planters who lacked genuine power over the free
population, because they ruled slaves. On the one hand, one found the
spirit of religion and the spirit of freedom indissolubly linked; on the
other, a civilization founded on slavery, a principle destructive of the
social state.
178 America and the Idea of Democracy

But Tocqueville did not dwell on the genealogy of American society;


he put far less emphasis on this aspect of reality than did, for example, his
contemporary Michel Chevalier in his Lettres sur ’Amérique du Nord.”
Here again, Tocqueville was less interested, at the time, in the historical
question than in understanding the present and making a prognosis about
the future. The center of his analysis is the “‘social state” of the Anglo-
Americans, not their history. The social state “‘is commonly the result of
circumstances, sometimes of laws, but most often of a combination of the
two. But once it has come into being, it may itself be considered as the
prime cause of most of the laws, customs, and ideas which control the
nation’s behavior; it modifies even those things which it does not cause.”
Tocqueville thus avoids a scholastic debate about prime causes. He
methodically holds fast to his analytical system. He does not move from
the celebrated crossroads in the forest, from which he is supposed to see
all the paths.
This aspect is discussed in the chapter on the “social state of the
Anglo-Americans” (volume 1, part 1, chapter 3), a state characterized by
an extreme form of democracy, a tradition of origins, the uprising for
independence, and a series of laws, particularly the law providing for
equal sharing of inheritance. Equality, the dominant feature of democ-
racy, does not mean that fortunes are equal or that Americans wish that
they were; on the contrary, it simply means that fortunes are not rooted in
a system of family inheritance and that money circulates very rapidly.
One could even say that “democracy” is legitimized by innate talent
alone, with no consideration for preexisting hierarchy. That is why
democracy has attained a sort of absolute level in an area where society
does not yet exist but where the social passions of the inhabitants are
nevertheless intensified by familiarity, or impatience, with an earlier
social tie: this area is the frontier of European colonization. But the reign
of equality does not manifest itself only through the mobility of fortune
and the distribution of wealth. It also creates uniformity in educational
levels and even in mental endowments. For equality provides a basic
instruction to all without creating a privileged class characterized by the
leisure or taste for devoting itself exclusively to intellectual pursuits.
This social situation—absolutely unique in history—can have one of
two incompatible political consequences: freedom or servitude, the
sovereignty of the people or that of a master. We thus reach the second
level of Tocqueville’s conceptualization, a level clearly deduced from the
first (since he is dealing explicitly with the “political consequences of the
social state”) but not determined by it, since this time a choice is offered:
the Anglo-Americans have been able to construct political democracy on
the basis of social democracy.
By understanding how and why they were able to do so, one can
examine the series of connections that bridge the gap between the two
levels of analysis. Tocqueville does not—at least immediately—spell out
Democracy in America 179

these connections, since he goes from the chapter on the “‘social state” to
the famous and very detailed description of the American political sys-
tem. But the connections can be found in chapter 9 of part 2 of volume 1
and also in Tocqueville’s travel notes, which testify to the fact that the
elements of his explanation were already in place by 1831. For Tocque-
ville, what has preserved American political democracy, after historical
circumstances gave it birth, is not only a fidelity to origins but something
resembling a state of mind, so widespread and so deeply rooted that one
can call it “national mores.” It is this factor that is responsible, day after
day, for maintaining society’s independence and its preeminence over
politics.
First, religion plays an admirable regulatory role in two ways: by what
it prescribes and what it forbids. While Catholicism, according to
Tocqueville, encourages equality and obedience (except when Catholi-
cism is separated from the state, in which case it acquires new characteris-
tics), Protestantism, especially in its sectarian and pluralist form, encour-
ages equality and independence. American religion is made of a set of
republican varieties of Christianity. But religion also sets limits on what
can be known to—and transformed by—man; hence, American boldness
is tempered by a sort of collective moderation. By preventing citizens
from “imagining certain things,” religion acts as an obstacle, in other
words (in words that Tocqueville does not use), to the revolutionary
spirit, which is the negation of democracy in the name of democracy.
Tocqueville wondered about this paradox all his life without ever fully
mastering its breathtaking implications; but, in America, religious con-
sensus had destroyed its attractiveness.
Another key factor in the independence of American society was its
high level of culture. Here Tocqueville parts company with Montesquieu:
the mainspring of republics is not virtue but instruction, by which
Tocqueville means the democratization of knowledge, particularly in
political matters. Speculating on the “thousand reasons that make repub-
lican freedom tolerable in the United States,” he wrote in his travel
notebooks:

There is one major reason that towers over all the others and,
after one has enumerated all of them, carries the day. The
American people taken as a whole are not only the most
enlightened in the world but—and I consider this a far more
important advantage—the people with the most advanced
education in practical politics. It is this truth, in which I
strongly believe, that arouses in me the only hope I have for
the future happiness of Europe.”

Despite their lack of refinement, the vulgarity of their manners, and their
obsession with money, the American people were, for the French aristo-
180 America and the Idea of Democracy

crat, the most civilized on earth. Tocqueville thus measured the worth of
the human investment that America represented—with eighteen centu-
ries of European history behind it. America was a nation of town dwellers
out to conquer nature, bypassing the unending curse of the peasant
condition in Europe. The pioneer was the unlikely breed between the
height of civilization and the height of savagery—the quintessence of
history and the quintessence of nature. Tocqueville’s travel notebooks
contain some unforgettably beautiful passages on this theme,” but he
used only a small portion of them in his book.
It is this high degree of civilization that supplied the Union—setting
aside the South, undermined from within by slavery—with one of the
basic ingredients in democratic mores: uniformity. Tocqueville, in his
notes, admits to being surprised by this feature of American life. Accus-
tomed as he was to observing differences of ‘“‘several centuries” between
the provinces—or parts of provinces—of European nations, he expected
to find an all the greater disparity within the New World as it was still
being colonized and ought to have presented, according to the locality,
‘‘an image of society at every age . .. from the wealthy urban patrician to
the desert savage.’ But Tocqueville observed the opposite. As the
Americans who were settling on the ‘frontier’ came not directly from
Europe but from the oldest settled territories, they had already been
exposed to the spirit of equality and subjected to the uniformization of
tastes and mores. “You will find the same man you left behind in the
streets of New York in the almost inaccessible wilderness: same clothing,
same attitude, same language, same habits, same pleasures.’’* Tocque-
ville did not believe, as Michel Chevalier did,” that the frontier consti-
tuted a third America, after that of Puritan businessmen and that of
slaveholding planters. The Saint-Simonian’s criteria were economic,
while Tocqueville’s were moral and cultural.
Actually, in analyzing the West and the American spirit in general (for
the West merely reveals its quintessence), he borrows—albeit with some
modifications—a concept dating from the previous century: ‘“‘civiliza-
tion.” Like the eighteenth-century philosophers, Tocqueville uses it to
refer to a set of cultural features that make societies both more regulated
and more active; but, unlike Voltaire, he does not regard it as the upper
ridge of a succession of cycles or, like Condorcet, as the terminus of a
linear progress. He wrests the concept from history—an undefinable
vector, another name for Providence—and incorporates it into his own
intellectual system. For Tocqueville, “civilization” is the particular form
of social activity that democracy, in its free state, makes available to all
citizens. By the same token, this definition allows him to relativize the
field of his analysis and to confine the benefits of civilization to the
European stock of mankind. Hence the well-known passage on the
genocide of the Indians, a passage that I cannot resist quoting at length,
not because its tone corresponds to present-day attitudes, but because it
Democracy in America 181

gives a good illustration of the exceptional virtues of the interpretative


system analyzed here:

The Indian races melt away in the presence of European


civilization like snows in sunshine. Their efforts to resist their
fate only accelerate their destruction at the hand of time.
Every ten years or so, the Indian tribes that have been pushed
back into the western deserts realize that they have gained
nothing by retreating and that the white race is advancing
even faster than they are withdrawing. Irritated by the very
feeling of their helplessness, or incensed by some new affront,
they assemble and swoop down on the areas they used to
inhabit, which are now dotted with European settlements,
rustic pioneer cabins here, the first villages there. They scout
the country, burn down settlements, kill herds, and take a few
scalps. Civilization retreats, but it retreats like the flow of the
rising sea. The United States sides with the last of its settlers
and declares war on these wretched tribes. A regular army is
sent out to fight them; not only is American territory recon-
quered, but the whites, driving the Indians in front of them,
destroy their villages, seize their herds, and establish the
outermost limit of their possessions one hundred leagues
beyond the previous boundary. Deprived of this newly
adopted homeland by what educated and enlightened Europe
has felt like calling the right of war, the Indians resume their
westward march, until they stop in a new wilderness, where
the sound of the white man’s hatchet will soon be heard again.
In the area they have just plundered, now safe from invasion,
new villages rise and will soon turn into teeming cities (at least
so their inhabitants believe). Marching ahead of the vast
European family, of which he is, as it were, the vanguard, the
pioneer takes over in turn the forests recently inhabited by the
savages. There he builds his rustic cabin, and waits for the
next war to open the way toward new deserts.”

Thus, ‘“‘democracy” is not the end of history or one of its universal


phases, or even less the reconciliation of humanity with itself. It is a
concept that enables Tocqueville to imagine a state of society and mores
specific to Europe, and more particularly specific to the extension of
Britain in the New World represented by the American Republic. In-
deed, in America the historical experience of democracy has been so
radical in the social and cultural spheres that the entire realm of politics
has been suffused and taken over by democracy to the point of being
completely subordinated to it. For one of Tocqueville’s strongest im-
pressions during his visit was that of the virtual absence of “politics” from
American society. He speaks in his notes of the ‘absence of
government,* an asset that man can find ‘‘only at the two extremes of
civilization,” either in the savage state, when he is alone in the face of his
182 America and the Idea of Democracy

needs, or, after the formation of society, when the individuals that
compose it are sufficiently enlightened and sufficiently independent of
their passions (or law-abiding, which amounts to the same) to do without
government. In a later passage,” he distinguishes two clearly separate
“social states”: “In the first, the people are sufficiently enlightened and
they find themselves in sufficiently favorable circumstances for them to be
able to govern themselves. Here society acts upon itself. In the second, a
power external to society acts upon it, forcing it to march along a given
path.” America, of course, fits the first description—that of a self-
administered society. It offers the example not only of what could be
called ‘‘pure democracy,” but also, up to a point, of ‘pure society,” with
politics excluded.
Naturally, my analysis goes too far, since a great portion of the first
volume of Democracy in America is devoted to a description of the
American political system. But it does serve to emphasize Tocqueville’s
intellectual approval of political democracy, despite what he regards as its
‘‘excesses.”” Political democracy, which provides for the freedom and
responsibility of citizens through administrative decentralization (which
Tocqueville is careful to distinguish, as we know, from governmental
decentralization), basically embodies the advantages of an aristocratic
political system while increasing the number of its beneficiaries. Conse-
quently, the problem, on both the administrative and the governmental
levels, is to compare the advantages and drawbacks of the two systems,
one in the hands of society as a whole, the other controlled by a heredi-
tary elite. This fascinating and justly famous part of the book is of little
import here, in that it is informed throughout, explicitly and implicitly, by
the conceptual opposition described above. What makes these chapters
interesting is the exceptional thoroughness with which Tocqueville ex-
plores this contrast from every angle, using American institutions as the
basic example; but he does not modify the central structure of the
dichotomy.
However, there is one chapter in the first volume of Democracy in
America—the last chapter—in which he is compelled to abandon his
conceptual “‘core.” In the extraordinary chapter 10 of part 2, he specifi-
cally leaves American ‘‘democracy” aside in order to examine the future
of what he calls the “three races” that inhabit the United States, namely,
the Anglo-Americans, the Indians, and the Negroes. Hence, Tocqueville
feels obliged, for once, to discard his analytical system, which has no
relevance to the two marginalized “‘races’—Indians and Negroes—the
two peoples who by definition are not ‘“‘democratic’’; and, as their exis-
tence is not without consequence on the future of the Union, this future
itself is not entirely contained in the prognosis concerning democracy.
Curiously enough, and almost regretfully, Tocqueville adds to the list of
subjects left aside by his type of analysis ‘the commercial activity prevail-
ing in the Union” and its importance for the future, as if his task were to
Democracy in America 183

enumerate, even in an intellectually chaotic fashion, the questions he has


not treated and to say why: “These topics are like tangents to my subject,
being American, not democratic, and my main business has been to
describe democracy. So at first I had to leave them on one side, but now at
the end I must return to them.’”®
With regard to the Indians, we have already seen from Tocqueville’s
travel notebooks that, in analyzing their society, he resorts to a modified
version of the concept of “civilization” inherited from the eighteenth
century. He uses it to describe not so much the historical ‘‘backwardness”’
of this savage world as its insulation from the civilized world, namely,
from American democracy. Once again, he proceeds from the “‘social
state’’ of the Indians—a world of nomads, tribes, warriors, and hunters—
to their mores and beliefs, which remind him, mutatis mutandis, of those
of the ancient Germans according to Tacitus. Thus, his vision of history,
far from being restricted to a belief in mankind’s inevitable evolution,
links up with his typological analysis. Tocqueville does not believe that
Indian societies, by coming into contact—even conflictually—with
Anglo-American “‘civilization,”’ will gradually be raised to its level. On
the contrary, he thinks that their nomadic social state, consolidated by
their system or mores and beliefs, isolates them thoroughly and per-
manently. Either the Indians react by waging war, in which case they are
defeated and driven westward, or they consent to become “‘civilized,”’ in
which case they enter an alien world where they are irremediably in-
ferior, exploited, and doomed. Both recourse to war and submission to
laws will seal their fate. Tocqueville’s sociological genius guards him
against the humanist illusion concerning the benefits brought by Euro-
pean civilization to other continents.
As for the Negroes, their fate cannot be sealed as dramatically, for it is
tied to that of the whites. But it poses a twofold problem for the Union,
and here, too, Tocqueville’s outlook is not optimistic. The conjunction of
an institution as anachronistic as slavery in a century of equality with its
confinement to a particular race among the population—the Negroes—
seems to Tocqueville to entail catastrophic consequences for the activity
of white society in areas where slavery exists, as well as an inevitable
struggle for its abolition, particularly in places where slavery does not
exist. But this struggle, which is rooted in the self-interest of the whites,
not of the Negroes, does not imply the disappearance of race prejudice;
on the contrary, the emancipation of the Negroes would not be followed
by interbreeding—vehemently refused by the Anglo-American settlers—
or by equality, impossible to achieve between two peoples alien to each
other. It is more than likely that the abolition of slavery would be offset
by the intensification of color prejudice, of what today we would call
racism. Thus, even if slavery were abolished by the masters, and not
through Negro violence, the Union’s cohesion would be seriously
threatened.
184 America and the Idea of Democracy

In these brilliant pages on Indians and Negroes, Tocqueville sets aside


his concept of ‘‘democracy,” since it would be incapable of explaining the
two societies he is discussing. Nevertheless, his analysis here is linked to
his general intellectual system by the priority he assigns to the notion of
“social state” and to its defining principle. For Tocqueville, Indians and
Negroes are not so much races distinct from the Anglo-Americans as
social groups organized according to principles incompatible with the
democracy prevailing in the Union. But they are incompatible in differ-
ent ways. Indians form a distinctive and introverted society whose “‘sav-
age” rules, beliefs, and mores are less an expression of humanity in its
primal state—as it was thought in the eighteenth century—than features
of a nonagricultural and nonsedentary mode of social organization. As
slaves, Negroes form a nonsociety, since servitude is by definition a
relationship based exclusively on power and not a “‘social”’ tie. But the
principle of slavery compromises the existence of the free society that has
established it. Having introduced and maintained slavery, this society is,
therefore, undermined from within by slavery itself. The Indians can and
will be destroyed by the law, as they are a society situated outside it. The
Negroes, on the other hand, owe their presence to an institution of
American democracy that is contradictory with that democracy but a
deliberate product of it. They are at the same time indispensable and
impossible to assimilate, necessary and destructive of the basic social
compact. Egalitarian America has incorporated an unavowable and nox-
ious principle into its life; and if this paradox puts its very existence at
stake, it is because it is even more destructive of white democratic society
than of the black slave population.
This pessimistic diagnosis of the black problem is accompanied, as we
know, by some doubts about the survival of the American federal system.
But Tocqueville’s pessimism in no way diminishes his admiration for the
“spirit” in which the British settled in America, the social state to which it
led, and the usages and ideas engendered by this social state. It has often
been said that Tocqueville the aristocrat had “resigned himself” to
democracy. Democracy in America does not seem to me to justify such an
expression, no doubt more appropriate to describe his judgment on the
European situation. In actual fact, Tocqueville is a critical admirer of
American democracy, and one could quote many passages from his book
to show that on the whole, and despite its shortcomings, he prefers it to
the aristocratic system, both as a social state and as a form of government.
On these two counts, America showed him the power a society could
exercise over itself. It is precisely this lesson—this example—that he had
gone to America to seek.

Five years later, in 1840, Tocqueville published the second volume of


Democracy in America. He could assume that his readers were familiar
with the American political system, since it was the main topic of the 1835
Democracy in America 185

volume. Tocqueville’s problem now was to explore in greater depth the


question that he had already touched on and outlined in the first volume
without discussing it systematically: the influence of the democratic social
state on the American spirit and, more generally, on the spirit of the
peoples among whom it prevails. Tocqueville does not want to treat this
“democratic social state’—that is, equality—as the sole cause of the
characteristics of American society or “of everything that is happening
now.’ He is well aware that the opposite is true and that one must take
into account a host of circumstances independent of equality. However, it
is this trait, common to the Oid and the New World, that enables him to
compare America and Europe. Moreover, this trait is his subject, the
starting point for his examination of society and history, because he
regards it as the most likely key to an understanding of the present state of
the world. The most surprising aspect of Tocqueville’s intellectual ob-
session with the concept of equality is the very obviousness of this
concept’s capacity to explain the present state of societies and their future
as well. As for its capacity to explain the present, the least one can say is
that the societies of Tocqueville’s time—and particularly his own society,
that of France under the July Monarchy—did not exhibit equality. Fur-
thermore, his own existence, in its outer aspects, was entirely governed
by the spirit of conformity to the values of his family environment and by
the strictest observance of the principles of social hierarchy:* one need
only glance at his correspondence to see this. Even after the July Revolu-
tion, France under Louis Philippe was a country in which men like
Tocqueville, the descendants of the old aristocracy, continued to exer-
cise, without having to seek it, an almost natural ideological and tangible
authority. What, then, is the source of Tocqueville’s intellectual convic-
tion, which his very existence contradicted daily?
The origins of his belief lie less in the present state of society than in its
history—that is, in his own past. It is probably difficult to imagine today
what the thunderburst of the French Revolution represented for a family
such as his—the mass of recollections and dramatic stories that filled his
childhood, the horrified fascination that he must have felt at a very early
age. By the end of his adolescence, he had succeeded in transforming this
family experience into an intellectual problem. His genius lies in the
precocious appropriation of his inheritance on another level and in
another manner. It is from a tale of misfortune, handed down by family
tradition, that he constructed both the idea of equality and the idea of
inevitability.
These two ideas are inseparable precisely because they are rooted in
the same existential soil: in the feeling that the French Revolution is part
of the course of history—in other words, that it is not over—and that,
having manifested itself by such a radical hostility toward the nobility and
even toward all notion of social superiority, it can be defined only by the
passion for equality. Little does it matter that this equality was not
186 America and the Idea of Democracy

“realized” in postrevolutionary societies. If it is true that Tocqueville


conceptualized only his own experience (and this is probably what dis-
tinguishes him from most great philosophical thinkers, trained primarily
in the abstract study of doctrines and ideas; it is also what explains his
obstinate investigation of a single idea, which one can properly speak of,
as one says of a woman, as the idea of his life), this experience told him, in
effect, that the French Revolution continued—through the Empire, the
Restoration, and 1830—and that equality remained at the heart of politi-
cal debates and events. Actually, this conviction was Tocqueville’s way of
appropriating the celebrated melancholy of the romantic generation, of
which he was also a son. But at the same time this feeling is a concept.
Hence its constant ambiguity, situated on two distinct levels. The use
of the word democracy as an approximate substitute for equality, but in
an even larger sense, leads Tocqueville to use the word in different
semantic ways—ways that, in particular, do not distinguish between the
social sphere and the political sphere. But the ambiguity also exists—and
perhaps at a deeper, because more hidden, level—in the specific use of
democracy to designate an egalitarian social state (in fact, this is the most
frequent use). What does our author mean by a society or a social state
characterized by equality? It is not easy to answer this question.
The simplest answer is a commonsensical one: a society in which
equality reigns is a society in which class barriers between individuals
have disappeared. Sometimes Tocqueville does give such a definition, for
example when he writes, ““When a people has lived for centuries under a
system of castes and classes, it can only reach a democratic state of society
through a long series of more or less painful transformations.’’* A little
later, in a note to chapter 26 of part 3 of volume 2, there is an even more
explicit statement, since the words that is introduce a definition: ‘When a
people has a democratic state of society, that is to say, when there are no
longer any castes or classes in its community . . .’’” These quotations are
wholly consistent, as they equate social classes and castes, at least with
respect to their character, which is contradictory to democracy and makes
the classless society the condition for equality. This type of definition,
both maximalist (equality as a real social state) and naive (equality
corresponding to the way in which social actors conceive it), occurs
elsewhere: for example, in the comparison Tocqueville makes between
aristocratic marriage and democratic marriage.* The first is a socially
programmed union, offering, in any event, little freedom of choice; it is a
union of properties rather than of people. The second, in contrast, results
from a free choice; it is based on the spouses’ feelings and the similarity of
their tastes (also, in this respect, it makes greater demands on marital
fidelity). One can see from this example how, in order to comply with a
mode of reasoning that proceeds by opposition, Tocqueville can be led to
confuse what he calls the democratic social state with this social state’s
self-image and the image of itself that it wants to project. In an age when
Democracy in America 187

bourgeois marriage as a union of wealth was one of the favorite themes of


novelistic literature, Tocqueville, because of the requirements of his
conceptual system, was oddly and temporarily blind to the obvious reality
of class marriage, a practice masked by the apparent freedom of indi-
vidual sentiment.
His blindness was both odd and temporary, for he was obviously well
aware—as he wrote elsewhere—that “there has not yet been a society in
which conditions were so equal that there was neither rich nor poor, and
consequently neither masters nor servants.”” How does the democratic
social state change this situation? “Democracy in no way prevents the
existence of these two classes, but it changes their attitudes and modifies
their relations.” This passage is crucial for understanding what he most
often meant by democracy—not a real social state, but the egalitarian
perception of social relationships, which are usually hierarchical (at least
judging by human history), by the actors in these relationships. This
egalitarian perception in turn modifies the nature of the relationship,
even when it has remained totally grounded in inequality. An example of
this is the master-servant relationship: it exists in the democratic society
of the United States but does not constitute, as in aristocratic societies, a
principle of the social order. Hence, in America, this relationship does
not create a separate, subject people whose mores and mentality are
handed down from father to son. Instead, it is the outcome of a freely
consented contract, by means of which the interested party negotiates its
provisional subservience and the limits of this subservience. “Equality [of
conditions], ”” says Tocqueville, ‘makes new men of servant and of master
and establishes new connections between them.”” Thus, “equality of
conditions,” one of his favorite expressions to characterize democracy,
does not mean that master and servant are truly equal but that they can
be; in other words, the relationship of temporary subordination does not
constitute a ‘‘state”’ that provides a complete definition of both parties,
since the relationship can, for example be reversed as a consequence of
their individual careers. As the servant can become a master, and wants
to become one, he is no different from the master. Furthermore, outside
the sphere of the revocable contract that ties him to his master, the
servant is a citizen enjoying the same rights as his master. “‘Equality of
conditions” must not, therefore, be understood in the material sense of
the expression but as a constituent principle of the democratic social
order, in contrast to the aristocratic world—as a norm and not an empiri-
cal observation.
Tocqueville is aware that this norm is a sort of unattainable and
endlessly elusive objective, never a reality: hence the perpetual instabil-
ity of democratic individuals and societies. As it is impossible for a people
to achieve perfect equality and as there will always be dominant positions
coveted by ambitious individuals, the presence of an egalitarian rule in
social life, by opening all careers in principle to all men, sharpens their
188 America and the Idea of Democracy

desires and struggles. This expiains why Tocqueville shifts so often from
equality as a dominant social state (that is, as a norm) to equality as a
passion (that is, as both a feeling and an ideology). One could almost say
that the democratic social state is characterized more by the passions it
engenders than by their fulfillment, since actual equality of conditions is
never achieved but always yearned for. “When inequality is the general
rule in society, the greatest inequalities attract no attention. When every-
thing is more or less level, the slightest variation is noticed. Hence the
more equal men are, the more insatiable will be their longing for
equality.’”*!
In other words, this genuine equality—an avowed but elusive aim, a
reference to a norm but also an object of desire—is subdivided into a host
of partial attempts to establish social differences. Tocqueville understood
that the belief in equality as a value does not eliminate what he calls
“private individual pride,” that is, the passion for distinguishing oneself
from others. On the contrary, the belief in equality intensifies this pas-
sion, while modifying and increasing its scope. The tendency to “‘distin-
guish”’ oneself is all the more legitimate as classes are less defined by
custom or law, as is the case in aristocracies. In addition, democracy
increases the number of signs of differentiation because, by making
conditions equal, and even more by making citizens uniform, it gives
added value to the slightest advantage; and, by increasing social mobility,
it replaces traditional hierarchies with recent and temporary privileges,
conspicuously displayed by their beneficiaries while they enjoy them.
There is in Democracy in America (particularly in chapters 14 and 16 of
part 3 of volume 2) an outline of a theory of how equality breeds symbolic
inequality and how this process gives rise to the restlessness and envy that
are the sentiments characteristic of democracies.
Nevertheless, Tocqueville believes that this inherent instability is com-
pensated by certain factors enabling the system to function in harmony.
First, there is religion, which he regards as a key element of the social
consensus and as a protection for ‘‘democratic” citizens against the
inordinate pretense to know everything and thus to change everything.
But there is also a built-in corrective, hidden within the very desire for
equality and the passion for social promotion: in a democratic state,
where careers are relatively slow and where individuals have become
used to making a daily effort to climb up the social ladder step by step,
there is a tendency for desires to become commensurate with means, and
ambitions with chances. Tocqueville was the first to discover this basic
law of modern societies, according to which men internalize by their
desire only a probable future, in the statistical sense of the term. They
look forward only to what can happen to them, thus avoiding both
exaggerated ambitions and insurmountable disappointments.
Moreover, it is the absence of these correctives for equality that
characterizes an intermediate social state between aristocracy and
Democracy in America 189

democracy: the state of revolution. It is an intermediate state in the


chronological sense, since revolution marks the transition from aristoc-
racy to democracy, but also in the sociological sense, since revolution
suddenly creates the ingredients of democracy without, however, en-
abling it to function: it introduces the notion of a break while unwittingly
embodying a tradition. ‘“Revolutionary” individuals inherit the immod-
eration of aristocratic ambitions; their feeling of creating a totally new
world is part of this immoderation, and the emergence of new men and
new laws prevents desires from becoming adjusted to chances. Thus,
revolution is characterized more by the explosion of egalitarianism than
by that of equality; it thrives on a vast discrepancy between what indi-
viduals want and what society can offer them. It creates in turn a tradition
that outlives the exceptional years and explains postrevolutionary insta-
bility: ‘The passions roused by revolution by no means vanish at its close.
A sense of instability is perpetuated amid order. The hope of easy success
lives on after the strange turns of fortune which gave it birth.’’”
Thus, for Tocqueville, ““democracy”’ in France is not the “normal”
form of this social state: it is its revolutionary form. This fundamental
distinction runs through the entire second volume of Democracy in
America and constitutes the leading strand of the comparison whose
elements the traveler had come to seek. In the United States, he found a
pure democracy, that is, one that resembled his ideal type—a democracy
since its inception, which had been generated from a religious matrix that
extolled its values and which, therefore, had never had to struggle against
an earlier aristocratic state. The concept of revolution, which Tocqueville
constructs entirely on the basis of the French experience, thus strikes him
as alien to American history (except as a potential civil war between the
southern aristocracy and Yankee democracy). More important, he re-
gards it as incompatible with the functioning of democracy, in that
equality constitutes a far more homogeneous and resistant social fabric
than aristocratic society. Tocqueville devotes a chapter“ to a discussion of
why the democratic social state, because of the social and cultural uni-
formity it induces and because of the mesh of conservative microinterests
in which it involves citizens, is a poor breeding ground for revolution; on
the contrary, it is the destruction of aristocratic inequality that provides a
pretext and an objective for revolution, as the European example demon-
strates.
Thus, the idea of democracy, in all the aspects that Tocqueville never
ceases to examine, does indeed merge, in most cases, with the idea of
equality; but it also takes on the ambiguities and various connotations of
the latter. There is just one aspect of equality that does not interest
Tocqueville: that of its objective reality. He is satisfied with the belief that
real social conditions have already begun to level off and are continuing
to do so gradually. But he never tries to be more specific about this
assumption, much less to measure it: it is an existentially, not a statisti-
190 America and the Idea of Democracy

cally, obvious fact. When he discusses this aspect of the question, he


speaks as a descendant of a great family of the ancien régime: he has in
mind the leveling of the material conditions of the nobility and the middle
class (exactly like Guizot, who agrees with his analysis but takes the
middle class as a starting point).
But Tocqueville’s greatest and most abiding interest is both the role of
equality as a norm of social life and the mental mechanism of which the
emergence of this norm is both the consequence and the cause: the
passion for equality. With regard to the first aspect, democracy never
ceases to carry within itself, as a consubstantial and inseparable element,
a threat to its own future, a prospect for the sake of which democracy
must allow itself to be continually challenged. Equality is a value that
cannot be realized by any social state (just as democracy, in the com-
monest sense of the term—government of the people by itself—is an ideal
type of power whose conditions cannot be met by any modern commu-
nity); and it is in the inevitable gap between values and facts, between a
society and its norm, that equality as a social passion takes root. This is
one of the features of Tocquevillian democracy. Aristocratic societies do
not suffer from such a discrepancy, since they are not in conflict with their
own values of subordination and hierarchy. If these societies happen to
be overturned, it is in the name of an inverse legitimacy, that of equality,
which appropriates for its own purposes a set of passions that are similar
in nature and directed toward the same goal. But an egalitarian society,
once established, cannot abide by its own values without being in con-
stant contradiction with its real state; and the passions that it has inten-
sified by its victory or by its mere existence pose a challenge to it in its
everyday functioning.” That is why democracy, while it has the advantage
over Other states of society of mobilizing its citizens through their egali-
tarian passions, also poses a problem unknown to aristocracies: the
problem of its daily viability.
In any event, this is a difficult problem to solve. It becomes dramatic
when the passion for equality takes precedence over all others, and
particularly over men’s fondness for freedom. It is this hierarchy of
preference that constitutes the chief danger for democracies. If the two
passions were of equal strength and were equally widespread, they would
combine their effects, and each citizen would actually enjoy an equal
right to participate in government. However, experience suggests that
equality and the passion for equality can exist in civil society and yet be
absent from political society; such is the case in a system based on
property franchise. One can also find equality and the passion for equal-
ity in a political society without freedom; such is the case with despotism.
The relation between the passion for equality and the other passions in
democratic life thus emerges as one of the essential elements of demo-
cratic society. Basically, Tocqueville believes that the passion for equality
is the overriding and distinctive passion among democratic peoples and
Democracy in America 191

that the fundamental problem it raises is precisely how it can be kept


within limits compatible with freedom. Why is the passion for equality
stronger than all other political feelings? The first chapter of the second
part of volume 2 offers a series of reasons: prevailing conformity, rooted-
ness in the deep-seated habits of the social state, and, above all, the fact
that the passion for equality is consonant with the logic of democracy,
since it can be shared by all, whereas the benefits of freedom can be
perceived by only a very few. Conversely, the excesses of freedom
(anarchy, for example) are obvious to all, whereas those of equality are
imperceptible and can be discerned by only a few. Finally, Tocqueville
does not forget the French example, since it is his lifelong concern. In
France the passion for equality is all the stronger for its being consider-
ably older than freedom and for having been fostered by the leveling
policies of absolute monarchs. The tradition of freedom is fragile, inter-
mittent, and limited; the tradition of equality is a constitutive element of
the nation.
This is not the case in America, which Tocqueville sees as a democracy
where the passion for equality is kept in check both by religious consen-
sus, which places man’s ultimate destiny in the hands of the divinity, and
by political institutions, which give precedence, and even power, to
society over the state: the celebrated chapter on associations® shows that
the latter play in democratic society a role comparable to that of the
aristocracy in aristocratic society, in that they each constitute a corporate
body expressing a social initiative independent of the state. Conse-
quently, Tocqueville’s analysis does not simply consist in studying the
passion for equality—even if this is a crucial phenomenon—but also aims
at understanding how, in the case of America, democracy has spun a web
of feelings, ideas, and mores that have given society its distinctive fea-
tures and its particular way of life. Thus, the intellectual architecture of
the 1840 volume becomes clearer. Tocqueville is not out to reconstruct
the history of American democracy, its origins, or its causes; instead, he
regards democracy as the central element in American history, as the key
to its interpretation, both because of its normative function in social life
and because of its existence as a complex of passions and individual
desires. In the second volume, Tocqueville is interested not in causes but
in consequences—in how democracy tends to produce what we could call
a “public spirit” sui generis, that is, a set of ideas and mores that, in turn,
help to strengthen the system.
To be frank, the second volume is so brilliant and dense that it is quite
impossible to look at its analyses in detail here. The reader must be left
with the enjoyable yet arduous task of discovering the depth and com-
plexity that lie underneath the apparent clarity of style. Nowhere else
does Tocqueville’s conceptual genius stand out more vividly than in this
text, which deals with the central concern of his life at its highest level of
abstraction and intellectual ambition.“ By studying in succession the
192 America and the Idea of Democracy

“intellectual movement,” the ‘‘sentiments,’’ and the ‘“‘mores”’ of the


American people, Tocqueville grapples with the most important question
in the social sciences, the question we have been asking ourselves ever
since: What is the connection between the production of ideas—of men-
tal images—and the other levels of social life? This is the question that the
young Marx was also trying to solve, at almost the same time, by positing
a connection between ideas and the social state in general. But Marx
characterizes this social state only by objective and, so to speak, material
elements—the productive forces and the production relationships that
they engender. Tocqueville, instead, focuses directly on social phe-
nomena without prior examination of economic factors, which are absent
from his type of analysis; and the social phenomena he examines are in
reality cultural.
When Tocqueville analyzes equality in its objective sense, he is actu-
ally discussing only the leveling of conditions, a process whose terminus
cannot be predicted, since its goal is elusive. Equality is not a state but a
history that gives a meaning to individual behavior and to the notions that
govern this behavior. Equality exists-more by virtue of the significance it
confers on social relations than by virtue of the changes it has brought
about in these relations. As a source of legitimacy, it imparts to all of
society—including people’s perceptions of society—the movement of
conflictual autonomy that characterizes American democracy. One does
not find in Tocqueville’s work the gap—which Marx was unable to
fill—between the production of material life and the production of ideas.
As Tocqueville’s chief analytical tool is not so much equality as notions
about equality, both as social norms and as the focus of individual
passions, he has no trouble descending from that level to the production
of ideas and of moral and intellectual traditions.
This “‘descent”’ does not necessarily imply a single or constant central
link between social states and ideas. Indeed, this link is even less exclu-
sive in more “‘intellectual” fields. On the subject of literature, for exam-
ple, Tocqueville warns the reader against a determinism based on a social
state—democratic or aristocratic:

I should say more than I mean if I asserted that a nation’s


literature is always subordinated to its social state and political
constitution. I know that, apart from these, there are other
causes that give literature certain characteristics, but those do
seem the most important to me.
There are always numerous connections between the social
and political condition of a people and the inspiration of its
writers. He who knows the one is never completely ignorant
of the other."

Thus, whatever precautions Tocqueville takes in order to temper its


systemic character, the nature of his thought is deductive: from ‘“‘democ-
Democracy in America 193

racy” stem the intellectual traits, the mental habits, and the mores of the
Americans. This proposition is partly tautological, since the definition of
democracy includes equality as a norm and as a passion. But Tocqueville
manages to compose variations of almost infinite subtlety and richness on
this proposition, in that his aim is to rework his central definition un-
ceasingly by examining its correlates.
Thus, it is not of the greatest importance, in my view, that the plan of
the admirable second volume has only the appearances of rigor; that the
SHC
distinction between ‘“‘ideas,’”’ “sentiments,” and ‘‘mores”’ is often de-
batable; that even within each of the parts, especially the third, the
arrangement of topics is not very logical. What does matter is the extraor-
dinary congruence between America as an object of study and the defini-
tion of democracy that Tocqueville uses in order to explore it. By virtue
of their origins, at once recent and homogeneous, their social state, and
their political institutions, the Americans fulfill the requirements for a
laboratory experiment in democracy. The assertion that every aspect of
their life is a consequence of democracy is a convenient expository device
for saying that every feature of America can be understood by reference
to a social consensus on equality. The association of new men for the
purpose of exploiting a virgin territory—an association with no other
historical foundation than this common belief—provided Tocqueville
with a field of study providentially suited to his conceptual genius, which
was both profound and simple.
Profound and simple: a comparison with Marx“ will perhaps enable us
to add a final word of explanation concerning these predicates.
Marx’s concepts, in every period of his intellectual development, are
never simple. They are reworkings of concepts inherited from German
philosophy or borrowed from English political economy; they owe noth-
ing to Marx’s subjective experience and almost everything to his reflec-
tion on the reflections of others. In contrast, Tocqueville’s system is
founded not on an intellectual construct but on a self-evident empirical
notion transposed to the abstract level, where it is defined as the irreversi-
ble progress of equality. This notion becomes the centerpiece of Tocque-
ville’s analysis, which aims at deducing its consequences.
Marx delves behind the equality proclaimed by the French Revolu-
tion, a value that has become the source of social legitimacy in the West.
He unmasks and denounces actual inequality, the contradiction between
facts and values. He writes a genealogy of the cause of this contradiction,
a cause that he locates on the economic level and in the social relations
arising from the production of wealth. Equality becomes no more than an
ideological lie, or a higher goal that cannot be achieved without the
destruction of capitalism. This dichotomy destroys all the features of
Tocquevillian equality, that is, equality as a social norm and as a passion.
Marx is interested in the workings of economic life and in the connections
between economics and society. Tocqueville explores the relation be-
194 America and the Idea of Democracy

tween the social principle and human behavior: by bringing his analysis to
bear on the motives of the individual and collective action of his contem-
poraries, Tocqueville has no need to reduce the political sphere to
another order of reality that is supposed to underlie and determine it. He
thus breaks with the obsessive search for the foundations of society so
characteristic of the eighteenth century and of Marx, who in this respect is
the heir to that century. Tocqueville deliberately focuses on what lies
“‘downstream’’: he regards the founding principle as a sort of historical
datum that is both evident and impossible to reduce to a causal demon-
stration; what counts is to deduce its consequences on social life. For
Marx, freedom lies in the abolition of surplus value; for Tocqueville, in
the intelligent handling of the belief in equality.
In this sense, Marx’s thought is infinitely more complex; it approaches
politics only through a series of mediations (in which, as it turns out,
Marx gets lost). But, in another sense, Tocqueville’s ‘“‘simplicity” goes
deeper. Despite appearances, Marx’s entire system is directed toward the
realization of the promises of equality. In denouncing inequality through
the analysis of capitalist social relations, he rediscovers equality in the
guise of the historical necessity of socialism but never ceases to regard it
as the basic value of the social compact. He criticizes the capitalist version
of equality only the better to demonstrate that it must act once more as a
founding principle for history, through the new departure represented by
revolution. Tocqueville, on the contrary, does not internalize equality as
a value: he is an aristocrat. But he does observe the fact that equality has
become the legitimating principle of modern societies, and he tries to
measure this phenomenon. By using a method that is both spatially and
temporally comparative, he confers a relative value to equality as a
legitimating principle and a new belief that must be compared to older
principles and beliefs in order for its immense impact to be appreciated.
Tocqueville is thus the first to view modernity in an anthropological
perspective.
Paradoxically, the ‘“‘simpler’ his thought and the less bookish his
theory, the more the two—thought and theory—are directly fed by a
contemporaneous psychological experience and the more they enable
experience to be dissociated from its conceptualization. Tocqueville went
from the aristocratic world to the democratic world, and it is this very
passage that constitutes the thread—and the anguish—of his life. With a
foot in each world, he regards as a self-evident truth that equality is only
one of the modes of social life. It is out of his archaic existential position
that he constructs his modern conceptual inquiry.
Marx’s thought, on the contrary, aims at a totally scientific investiga-
tion into the most hidden sources of inequality, which it unmasks under
the misleading disguise of a free contract between equal individuals. But
the more it tries to be scientific, independent of contemporary experi-
ence, and intent on discovering reality under the mask of ideology, the
Democracy in America 195

more it “adheres” unknowingly to equality as a choice, forbidding the


slightest deviation and without conveying the slightest doubt as to the
moral necessity of equality as a social principle. Marx’s thought dresses
up the quintessentially modern value of equality in the language of
science. No doubt it is this juxtaposition of prophecy and analysis that
explains the great posthumous destiny of a weighty treatise on political
economy.
Marx remains inside the system of belief in equality; Tocqueville
subjects it to a comparative analysis. This surely provides the fullest and
most satisfactory explanation of why their works had such a different
impact or, rather, why one enjoyed universal glory and the other a
somewhat muted success. But this contrast may also explain Tocque-
ville’s superiority over Marx in the realm of predictive accuracy: it would
not be difficult to show that the French aristocrat’s forecast of the con-
temporary world (I am referring to what we call today the ‘‘developed’’
European world, since the two authors did not study any other) is
infinitely closer to our present experience than the forecast of the Ger-
man socialist. The predictive value of Tocqueville’s work remains ex-
traordinary, even in its detail, and the famous pages (not very original for
their time) on the imperial future of America and Russia are in this
respect less extraordinary than a host of observations on the ideology that
still informs our existence. Moreover, the disparity in the predictive
accuracy of Tocqueville and Marx is all the more striking as it seems to be
inversely proportional to the veracity of their empirical points of depar-
ture. The social equality with which Tocqueville seems obsessed does not
appear to the historian to be a characteristic feature of European
societies of the first half of the nineteenth century. In contrast, the misery
of the working classes, which forms the backdrop to the theory of capital-
ist exploitation, is indisputable.
But, in constructing the doctrine derived from this situation, Marx
focuses on an economic mechanism that explains practically nothing
about the great events of the nineteenth and twentieth centuries. Not
only does it ultimately reduce the political sphere to the economic sphere,
but it ‘‘freezes”’ the evolutionary process by equating it with the develop-
ment of an ‘“‘objective”’ contradiction in production relations. This scien-
tistic view of the future as the extinction of surplus value through the
proletarian revolution turned out to be less prophetic than the study of
the connection between equality and the administrative state. The reason
for this is not only that a subjective feeling of evolution—provided it is
developed in an abstract mode—can turn out to be more accurate than
the pretense of a scientific knowledge of history. It is, above all, that by
analyzing equality not as a situation, but as a principle, a complex of
passions, and an open-ended political dynamic, Tocqueville has a double
advantage over Marx. First, his analysis is wholly compatible with an age
of expectation ushered in by the French Revolution, an expectation
196 America and the Idea of Democracy

whose outlines he can discern in the United States; second, he tries to


understand, not the causes, but the consequences of this phenomenon. In
so doing, he made a prediction that proved to be correct, namely, that the
world of equality and the forms of behavior it engendered were durable
and irreversible phenomena that would decisively affect the future. To
that extent, Tocqueville was already analyzing the world in which we are
still living.
Today’s America:
A Metamorphosis of the
Idea of Equality

For any traveler in modern America, the most glaring phenomenon is the
urban crisis. It has hit hardest in the oldest industrialized areas of the
United States—the Northeast and the Great Lakes—the areas that are
the most traditional symbols of America’s success. These chosen lands of
democracy, affluence, and individual promotion have become the sites of
collective revolt and hopeless poverty. Yesterday, they embodied the
nation’s promise; today, the frailty of basic values and principles: equal
opportunity, individual success, free enterprise, and liberty.
It all began when the blacks, driven from the South by farm mecha-
nization, migrated to the North, where they were attracted by industrial
wages. This movement dates from just after the First World War, but it
assumed massive proportions only during and after the Second World
War. The great American cities had grown in the nineteenth century,
thanks to European immigrants, who contributed to their expansion,
their construction, and their rising prosperity. By the mid-twentieth
century, these cities were in the hands of the only peasants in American
history: the blacks who had lived in the old plantation economy.
The blacks have indeed settled in the inner cities. For a European
accustomed to the gloomy housing complexes of working-class suburbs
and the concentration of wealth in the inner cities, it is extraordinary to
observe the reverse phenomenon in the United States: the poor, that is,
the blacks, in the heart of the city, surrounded by a vast middle-class area
within a radius of twenty, thirty, or fifty kilometers. What is extraordi-
nary is not that blacks and whites cannot live together. After all, the
urban history of Europe also involves the separation of rich neighbor-
hoods and poor neighborhoods and even, more recently, the ethnic

First published in Le Nouvel Observateur, 24 January 1977.

197
198 America and the Idea of Democracy

segregation of North Africans, Pakistanis, or Turks. It is the specific


features of segregation in America that are intriguing: its inverse geo-
graphical distribution relative to the European pattern and its extraor-
dinarily rapid rate of increase.
The first of these features is often attributed to economic reasons, such
as the availability of low-interest building loans for new ‘“‘suburbanites”
anxious to become homeowners or the flight of businesses from midtown
districts in order to escape excessively high municipal taxes. But these
factors accompany and accelerate this phenomenon, rather than inducing
it. I suspect that, behind the white exodus and the abandonment of the
inner city to the poor, there lies a cultural choice consonant with the
historical origins of the nation: the detached wood-and-brick house ver-
sus the skyscraper; a small parcel of nature versus the pollution and
corruption of the big city; the homogeneous community, the “nice neigh-
borhood,” clustered around its churches and supermarket, versus the
faceless, atomized individuals of the metropolis.
The heart of America has always been made of small and medium-
sized towns—barely distinguishable from what we Europeans call “‘the
countryside’’—with their wooden houses so widely scattered that they
have no real geographical or architectural center; yet these towns are the
home of living communities that act as guardians of a domesticated
pioneer tradition in which lawns have replaced forests and the Sunday-
morning lawn mower the first settlers’ axes. For the past twenty years, the
American middle class of the big cities has been rediscovering and
rebuilding its traditional habitat. In a country so close, in many respects,
to eighteenth-century philosophy, ecology is more an ideology than a
fashion, and one would not have to delve very far to find a lingering
notion that rural life predisposes to virtue and that the metropolis is the
mother of all vices.
The case of the blacks provides a counterexample of this. For, if the
white exodus toward the wealthy green suburbs has been so massive, it is
also because the increasingly rapid urbanization of blacks has upset all
the traditional equilibriums of the environment, starting with the most
fundamental of all: the family. The mechanisms and causes of this process
are still not fully understood, but its effects are clear: a pauperization of
vast numbers of blacks in the inner cities, the precariousness of family
units, a standard of living too low to provide durable funding for ade-
quate municipal services, the rapid growth of crime (especially juvenile
delinquency). The combined effect of these mutually aggravating phe-
nomena has been the creation of sinister ghettos situated in the heart of
the big cities and more or less ‘‘self-managed” by gangs whose chief
victims are the blacks themselves.
Even during the period of great prosperity and full employment of the
1960s, many black mothers with dependent children managed to survive
only with the help of a weekly municipal welfare check. Today, with
Today’s America 199

unemployment particularly high among young blacks, the welfare system


has become both more necessary and more costly. As a poor palliative for
an unsuccessful urbanization and an economy in a state of semicrisis, it
helps to perpetuate an evil it was supposed to cure. It makes taxpayers
unhappy while demoralizing its beneficiaries. It ruins municipal finances
without preventing urban blight.
New York’s bankruptcy has become the best-known example, but
New York is not alone. Gary, Newark, Cleveland, and Detroit, for
instance, also face the same deadlines for the same reasons. The example
of New York will mark a milestone, for better or for worse.
The problem, in economic terms, is simple. In a system of decentral-
ized municipal management, as in the United States, the accounts of a
city like New York must obey the same rules as the accounts of any
company: receipts must balance expenditures. But, on the receipts side,
the production of wealth, which provides the basis of local tax assess-
ment, has steadily declined. Only twenty years ago, New York was still a
great industrial metropolis, with innumerable small and medium-sized
firms producing consumer goods (beer and clothing, for example). The
big city created jobs and managed fairly successfully to absorb an im-
migration that brought its population up to eight million inhabitants.
Today, industry is leaving New York: none of the advantages the city
used to offer still exist. The transportation system is deplorable, the
railroads are in a state of ruin, and trucks are bogged down in traffic jams;
today, relatively cheap labor can be found everywhere; production costs
in the inner cities are constantly rising owing to traffic problems, the
increasing cost of energy, and industrial disputes.
When a municipality tries to make up for these drawbacks by offering
tax breaks to new industries, it comes into conflict with the old industries,
who see no justification for this unequal treatment, and with ecological
activists, who accuse city hall of favoring business at the expense of the
people’s interests. The deterioration of urban living conditions even
threatens to reduce the volume of tertiary activities in the near future.
Nor can it encourage the development of tourism, already hit by the
world economic depression. Hence a structural employment crisis
(worsened by the cyclical recession since 1975), a shrinking of the tax
base, and an impoverishment of New York.
Simultaneously, the city has to increase its spending considerably, not
only to meet the rising cost of municipal services, but to finance welfare
for the poor. Municipal services are far more wide-ranging than in a
centralized country such as France; in addition to public order, sanita-
tion, and firefighting, they include the school system, from grade school
to university, and the hospital system. Thus, they entail considerable
expenditure on wages, including pensions—an all the more formidable
obligation as the latest mayors of New York, who were unable to grant
the wage increases regularly demanded by municipal employees, have
200 America and the Idea of Democracy

guaranteed in exchange an increase in pensions, thus handing over the


problem to their successors.
But to this “‘normal’’ expenditure were added the rising costs of
welfare during the 1960s and especially under the Lindsay administra-
tion, owing to the influx of an uprooted and impoverished black pop-
ulation. These costs involved municipal support of families without
resources (particularly the countless families where the father had dis-
appeared without a trace) and a minimum of reimbursement for medical
expenses. This huge program spawned a new municipai bureaucracy,
poorly suited to the complex tasks of registration and selection; it also
gave rise to widespread fraud among eligible (and less than eligible)
claimants.
In short, everything conspires to plunge the New York administration
into a deepening chasm of debt: decreasing tax revenues, enormous
operating expenses, and an urban welfare state that feeds, or helps to
feed, several million people. As things now stand, all the elements of the
problem are tending to make it worse. Economic difficulties are increas-
ing the population of welfare beneficiaries while reducing tax revenues
even further. The municipal authorities are cutting expenditures, laying
off policemen, and consequently reducing police duties. In certain neigh-
borhoods, they are even closing down schools and laying off sanitation
teams. But the deterioration of all these city services, which further
weakens the mechanisms for maintaining social order, leads to even
greater delinquency, which in turn stimulates the middle-class exodus
and thus the impoverishment of the city.
There are some areas of New York, in the Bronx, for example, that
now resemble the blocks of a bombed city. The smashed windows and
broken-down doors testify to the desolation of what was once a living
neighborhood. Even in Manhattan, when one walks south from 38th or
36th Street to the Village, the gradual “‘desertification” of the city is
patent. There are several hundred thousand abandoned dwellings in New
York, while in the suburbs there has been a speculative real-estate boom
for the past ten or fifteen years. New York’s financial bankruptcy can be
plainly read in the urban landscape as one of the great social and political
crises that America has experienced.
One cannot accuse the United States of not being honest about its
crises. This reformist country is a violent country. What it is displaying so
forcefully, what it is paying the price for so visibly, is still the black
problem. The old curse of the American dream, which civil-rights mili-
tants thought they had exorcised in the 1960s, has now returned in
American history, in a new, unexpected, and diabolically reversed form.
It is now a problem for the North, not an element of the civilization of the
South; a threat to the very fabric of Yankee America, not the expression
of the immemorial ties of hatred and connivance that had produced the
old plantation world.
Today’s America 201

Southerners have lived, as it were, on familiar terms with their crime


against equality; they are aware of its extent and historical roots; they
know the value of time and that it will take long to make amends. In
contrast, the American civil-rights generation—the young students who
went down south in the sixties to spread the good word—believed in
regeneration through ideas. This generation overturned, reinstated, and
revitalized the lessons of the great founding principles; it at last extended
to the blacks, who had been excluded from the original promise, the
blessings of Enlightenment philosophy. Thus, it created both a move-
ment toward progress and an illusion. The movement toward progress
was represented by the Johnson legislation, the legal ban on racial
discrimination, the massive registration of black voters, the forced pro-
motion of an entire people in American public opinion. The illusion was
that this moral pedagogy would be enough to change the fate of the
blacks—even as they were being uprooted on a vast scale and their living
conditions were radically modified. The power of ideas not only came up
against the inertia of history; it also encountered market forces.
At this point, some misunderstandings must be avoided. There is no
lack of “radical” intellectuals on the American far Left to say that the
moral puritanism of the young, the antiracist campaign, and the Johnson
laws in no way changed the real situation of blacks, that all this is a new
ruse of the whites’ good conscience. I disagree. Political maximalism
always leads to the belief that if not everything is transformed, then
nothing is. That is the mirror image of the idealist illusion. In this
particular case, the maximalist position fails to take into account the
tremendous change wrought in American public opinion, an undertaking
unparalleled in the contemporary world: the blacks have been integrated
into the nation’s egalitarian ideology. This is not to say that racism as
ideology or that every form of racial discrimination (in hiring, for exam-
ple) has disappeared; but there is a huge difference between a society in
which these phenomena are tolerated or even encouraged by a public
consensus and a society in which they are condemned and campaigned
against at the explicit insistence of the authorities, the media, and the
law.
This difference is all the greater because American society is more
flexible, more receptive to the lessons of the media, and also more used to
espousing and extolling the democratic consensus. Nowhere else in the
world can one find a more integrative social machinery. It owes this
capacity to its adaptability, its malleability, its traditions, and its founding
creed—to its qualities and defects, to its combination of conformism and
innovation. And, in the 1960s, Americans decided to integrate the black
minority and to pay the ideological, political, and moral price at once.
This is hardly an insignificant change.
The fact remains that at the same moment, and for reasons of another
nature, the black population changed civilizations and posed novel prob-
202 America and the Idea of Democracy

lems. The two phenomena were not unconnected: it was by becoming a


problem for the North that the black minority escaped from the old
southern segregationism and succeeded in forcing the mental barrier that
excluded it from the Yankee egalitarian creed. This vast migration, this
massive urbanization, many of whose features belong to the sphere of
social pathology, has also provided an opportunity for collective eman-
cipation. Since the late nineteenth century, Yankee America has been
the refuge of poor immigrants in search of a happier identity. Southern
blacks—along with Puerto Ricans and Jamaicans in New York and
Chicanos in the West—represent the massive final wave of these poor
immigrants. This time they have come from within the country, but from
its defeated and accursed part; they thus bring with them a special
challenge, which reaches down to the roots of the nation. Will the
America that was the image of paradise and the land of real opportunity
for so many poor Europeans be—can it be?—a belated providence for its
blacks?
This problem is infinitely more difficult than that of giving a new future
to the Greeks, the Jews, the Irish, or the Sicilians. For the blacks never
chose America: it was forced on them. Even when they “went up” in
droves to the North and its big cities, their expectations played a far lesser
role than the crude mechanism of market laws: farm mechanization in the
South versus demand for labor in the North. Blacks are both considerably
more alien to America than the Greeks or the Irish, and much closer to it.
They are totally foreign to the white Anglo-Saxon Protestant cultural
model of meritocracy based on work, and yet they are totally involved in
its history, of which they represent the tragic side. They are just as
impossible to assimilate as to forget.
The sort of obstinacy with which every branch of officialdom in the
United States, echoed by the press and television, preaches antiracism
and equality is a sign of this psychological obsession, as well as of the
traditional faith in the reassertion of great integrationist ideals. But the
ghettoization of the blacks, delinquency, urban decay, and the dramatic
conflicts over school integration indicate that, with respect to the black
question, other forces are at work than those that enabled a pluralist
nation, from its very inception, to find a consensus. This crisis also
indicates that other solutions are required and that the assertion of
equality is only a preliminary step. In other words, it will not be enough to
resort to tradition.
One of these solutions, dating back to the 1930s, has been precisely the
recourse to social legislation, culminating in today’s welfare system. This
solution involves the transfer of revenue from the federal level to states
and municipalities, with the aim of guaranteeing a minimum standard of
living to the poorest strata of the population. Initially, at the time of the
New Deal, this system was designed to protect the poor against the
Today’s America 203

economic crisis. Both the Kennedy legislation and Johnson’s ambitious


programs—through their impact on housing, employment, education,
and hospitals—were basically aimed at hastening the urban integration of
the black minority. However, they did not release local, that is, munici-
pal, authorities from the obligation of paying most of the bill.
We have seen that this system is expensive and has missed its objec-
tives. At the federal or state level, it is limited by American taxpayers’
hostility to big spending, that is, to the increase in public expenditure and
thus in taxes. At the municipal level, the system is supported by its
beneficiaries, mainly blacks, and by the municipal workers for whom it
provides jobs. These groups represent the bulk of the Democratic clien-
tele in the big cities. But, as it has not benefited from massive federal
aid—an unlikely prospect—the system is worn out, debt-ridden, and
bankrupt. It was just efficient enough to accelerate a vast black immigra-
tion but not efficient enough to integrate it.
The price paid by American society to try to solve “its” black question
is therefore twofold: on the one hand, the existence of a very large
“ghettoized”’ and subsidized urban population; on the other, a phe-
nomenal “‘social-democratization” of big-city government. The second of
these new features was originally designed to obviate the first; but now
they are inseparable and complementary, as if the benefit of a Swedish-
style welfare system were granted—on a paradoxical yet permanent
basis, in the name of equality, and within a society founded on principles
contradictory to that policy—to an ethnic minority.
This development is worth pondering over, for it probably marks an
important date in the history of the United States. First, the state—or at
least a certain number of political institutions—has taken charge of
society. Although Washington still confines its intervention to short-term
social programs, the fact remains that the survival of the various munici-
pal welfare systems will depend, and already depends, on federal help.
However, the intervention of political forces in the social sphere, the
systematic redistribution of revenues by democratically elected institu-
tions, and the ensuing inflation of bureaucracy are all features of Amer-
ican society that run counter to its tradition, according to which free
social initiative offers the best solutions for poverty and inequality.
The conviction that each individual could find in the new nation the
success and wealth corresponding to this merit has been consubstantial
with the United States. It made it possible to conceive of poverty as a
purely individual and temporary condition, to be alleviated by the charity
of the rich and by the philanthropy of private foundations, religious
associations, or local communities. The state had no jurisdictional pow-
ers over society: the hatred of “bureaucracy” is still one of the basic traits
of American political consciousness, as both Carter’s and Reagan’s suc-
cessful campaigns have plainly shown. A testimony to another age, this
204 America and the Idea of Democracy

attitude has survived the extraordinary growth of federal government


over the past thirty years and also what Alain Touraine calls the “social-
democratization” of the big cities.
But, by making its way in such an adverse human environment and by
running counter to such lively historical traditions, the urban welfare
state of Yankee America has transformed the old doctrine of equal
opportunity and of the connection between success and merit. If the
United States has belatedly implemented—along with a legislative ban on
racial discrimination—a massive welfare program for the very poor, that
is, for the population of non-European origin, it is because the introduc-
tion of the blacks into the so-called Yankee melting pot has revealed the
falsehood, or rather the limits, of the equal-opportunity principle. If this
principle does not in fact lead to a statistical distribution of success and
employment among blacks comparable to that of the rest of the popula-
tion, it is because a certain number of elements, at the level of the ethnic
group, are working against equality.
What is important is not that this sociological fact has been estab-
lished; it was established long ago. What matters is that this fact has been
integrated into the American egalitarian ideology and that it has led so
quickly to a new consensus aimed at guaranteeing equal opportunity, not
for individuals, but for groups. This is an ambitious project that, in many
respects, goes against the ideas on which America was founded: antidis-
crimination laws and their jurisprudence, for example, challenge the
freedom of employers and landlords. In a deeper sense, these measures
lead to the notion of preferential treatment of individuals belonging to an
oppressed minority—that is, to the notion of reverse discrimination,
which is expected to have a corrective effect.
In this respect, American society is still what it was for Tocqueville a
century and a half ago: a laboratory for the idea of equality. The secret of
this experimental longevity is that, even after two hundred years, Amer-
ica has not given to injustice the traditional consent based on the argu-
ment that “there is nothing new under the sun.” For America has
steadfastly continued to regard itself as the embodiment of a set of values
and as a complex of statistically measurable performances—beliefs one
could describe as a religion and a sociology of equality. Consequently, it
has now come to grips with the truly awesome idea of real equality.
This idea was imposed by the blacks as well as by the young students of
the 1960s, at a time when equality began to be conceived at the collective
level as the establishment of equal employment opportunities for the
various ethnic groups. This change introduced a tremendous dynamic in
such a diverse society and a new stake in social conflicts over housing,
employment, school integration and so on. I am not suggesting that
Americans woke up one fine morning as antiracists determined, as one
says, to ‘make amends.” Many of them, as in any society, did not like, or
simply did not go along with, the new desegregationist policy. This group
Today’s America 205

included even democrats, even members of traditionally liberal minor-


ities: one of the factors—but not the only one—in the latent hostility
between blacks and Jews in the United States is the attachment of Jews to
the meritocratic, individualistic model and their distrust of the notion that
advantages should be guaranteed or offered on the basis of ethnic ties. I
am not suggesting, therefore, that everything has been solved but, quite
to the contrary, that everything has been ‘‘destabilized”’ by the definition
of new rules for the game of equality. One can qualify these rules as
asymptotic, utopian, unattained, unattainable, or whatever other adjec-
tive one cares to use; nevertheless, in a certain sense, they have changed
everything.
This is the price that has been paid not for integrating blacks but for
admitting the possibility of their integration, which remains American
society's major problem. On the one hand, there is the urban welfare
state; on the other, an extension of egalitarian legitimacy to groups: in
short, the end of liberal individualism. None of the mechanisms of the
ideal America outlined by these choices is functioning properly. Welfare
costs too much for an enfeebled economy, and it perpetuates urban decay
as well as, no doubt, demoralization among blacks. The policy of non-
discrimination favors the promotion of a black bourgeoisie more than it
changes the lot of the poor. Finally, the general orientation of these
measures has aroused hostile reactions in Reaganite heartland America.
Yet, even though reality still lags far behind ideas and even though
reforms, as always, are tremendously costly in financial and psychological
terms in relation to their actual impact, it seems to me that the dynamic of
the new American consensus is irreversible. It has been fueled less by
government (since it has survived the eight years of domestic immobility
under Nixon and Ford) than by society itself and by the numerous local
associations and institutions that society continues to control and orga-
nize. However much American society may seem, on the surface, to have
returned to the conservative consensus of the 1950s, after the turmoil of
the 1960s, I believe that the opposite is true—that the demands for
greater democracy voiced in the 1960s have been assimilated into the
body social and that they serve as a model for the new reformist con-
sensus.
Let us take, for example, the women’s movement, which is infinitely
more powerful than in Europe. It follows the black model so closely that
American women are demanding, among other things, equal access to all
jobs as members of a minority group. This is an absurd demand, if one
examines it literally; it acquires force and significance only in the context
of the new American democratic consensus, arising in response to black
nationalism. Feminism has been able to rush into the breach opened by
ethnic demands because public opinion is less sensitive to the absurdity of
this analogy than it is familiarized with the notion that a certain number of
collective inequalities have to be set right. By shifting from the free
206 America and the Idea of Democracy

competition of individual merit to a determination to introduce social


equality at the group level, the dominant ideology, so crucial to the
survival of this segmented society, provides the famous melting pot with
an almost infinite range of objectives and standards—as infinite as there
are imaginable social groups and as long as inequality will persist among
them.
Thus, America continues to live in conformity with its original mes-
sage, while adapting it to the circumstances of the second half of the
twentieth century, that is, to the uprooting and reintegration of the
blacks.
MODERN
JEWISH
HISTORY
To the Reader

I have assembled in the last part of this book four studies on modern
Jewish history, three of which have been published in a different form by
Le Nouvel Observateur (the fourth, “Israel, Zionism, and the Diaspora,”
is here published for the first time).
The first of these studies is devoted to the Jewish problem in modern
French history; the second is a presentation for the French public of the
work of the Israeli historian Gershom Scholem; the third deals with the
relations between the French Left and Israel. The third essay is the only
one to have appeared more or less in its present form in Le Nouvel
Observateur (15 May 1978). The first two have been somewhat revised
and rewritten.
References to the original articles are given in the notes. I should like
to take the opportunity of thanking my friend Jean Daniel for having
published them in the first place and for allowing me to use them in this
book.

209
Ltée fe eliascd fs ei ir
couch
sae FAN 9.1 sop
; eG, Ÿ :

Teeny de Cyr last oti Ow treks tan bath


us We iidacp | ag: À eodylag tae ce: an sciohd fe
at he CS th ft sin ive tT aie
tomate ct) nid age ion) bee ee sroelsa om
Veen, Gin! il Iyer
eel a) eat an sent tiing’ 2
Lov -rTdg monty nom EM (98 tl of) ICE 4 LU Le
QUE a

Lin tarts: covet diam


esti get
-

rane
KA eave nr27 Samet von anitaee
H LE _

D dt nt Cad si io@) Saber el ré Les Cas wit «dt . Sug |


Jews and French Democracy:
Some Recent Books

The history of French Jews has been affected by the repercussions of its
characteristic feature: the willingness to submit to Jacobin assimilation.
The reason why this history has been so unobtrusive is that it was also
almost pointless. Since the French Revolution, the Jews were a part of
French history. Even if the Right denied them this title to joint ownership
of the French patrimony, the Jews had chosen to receive at least precar-
ious tenure rights to it from the liberals. What was the use, then, of
dwelling on the factors that made them a distinctive category of citizens?
In this respect, the increase over the past few years in the number of
works on the history of the French Jewish community from 1789 to today
is the sign of a major change, many of the reasons for which are obvious.
First, the impact of the war, which has taken several forms: the Vichy
persecutions, the existence of an officially anti-Semitic French govern-
ment, and the massacre of Jewish refugees from Germany and eastern
Europe have reactivated the residue of bad conscience inseparable from
any successful assimilation. And then there was Israel, the unlikely and
tragic victory of the destitute Jews scorned by the French consistories, the
survivors par excellence of the Holocaust, the outcasts of Europe pro-
moted to the central role in Jewish history, the proletarians of the
community who had become its national flag.
I am writing as if this were an exclusively Jewish history. Of course, it is
also our history, that of the gentile French, and it is not a particularly
glorious one. But it derives its true weight and universal value from the
way the Jews experienced it, suffered from it, acted on it, and interpreted
it. The French Left has been all too prone to deal only with the outer

First published in Le Nouvel Observateur, 5 April 1976, 17 September 1979, and 28 April
1981.

211
212 Aspects of Modern Jewish History

aspect of the Jewish question, even with the best possible intentions
(which was not always the case), even when trying to reduce it to anti-
Semitism, and especially when treating it as a capitalist diversion. Now
that the Jews have gained a national history, that of the French Jews—
which seemed frozen, withdrawn, and dependent—is stirring again. For a
fairly long time, the history of French Jews has fluctuated according to the
variations of our “‘thermometer’’: is it not time we looked at theirs?
First, I should like to take this opportunity to pay tribute to a series of
books that are an honor to postwar French publishing. The “Diaspora”
series, edited by Roger Errera, is precisely an expression of the renewed
interest in Jewish history in France. It has brought to the attention of the
French public a first-rate collection of authors and books, such as Ger-
shom Scholem’s admirable Messianisme juif (The Messianic Idea in Juda-
ism). On a less fundamental but no less valuable level, this series also
aims at providing a survey of modern French Judaism.
Patrick Girard’s book concerns the oldest period, 1789 to 1860.' While
it does not contain any new information on the history of French Jews
between these two dates, it gives an excellent summary of their progress
from revolutionary emancipation to gradual integration into French soci-
ety. The tremendous impulse of liberal universalism marked the con-
tribution of the Revolution to the Jewish question. By granting full
citizenship first to the Sephardic Jews of Bordeaux and of the Comtat
Venaissin (1790), then to the Jews of Alsace (1791), the Revolution
brought these old ghetto communities back into the mainstream of his-
tory. The Revolution offered them the historical model of French “‘re-
generation”; and the consequent substitution of a politically oriented
form of messianism for traditional values became a key factor in the
fragmentation of the old Jewish communities. No Jew had an ancien
régime to mourn. All Jews responded to both the most modern idea and
the oldest idea introduced by the new regime: the rights of the citizen and
the right to participate in a just undertaking.
Yet this ‘‘assimilation” was to take half a century, owing to the
persistence of traditional French attitudes, first in the form of popular
anti-Semitism during the Terror, then in the policy of the Empire. Napo-
leon reversed the Revolution’s perspective on the Jews. Whereas revolu-
tionary emancipation had consisted in replacing their religious status by a
political status, Napoleon, in convening the Great Sanhedrin—that is,
the notables of the Jewish community—in 1807, made their French
citizenship conditional on a religious pact with the Christians.
What the revolutionaries had regarded as the cause of the moral decay
of the Jews became for the emperor the guarantee of their good conduct.
This renewed focus on religious identity explains the confiscation of some
of the benefits acquired by the Jews through the Revolution. In particu-
lar, it explains the “infamous decree”’ of 1808, which reestablished the
juridical inferiority of Jews before the courts, with the exception of the
Jews and French Democracy 213

Jews of Paris and Bordeaux (but, at the time, the Alsatian Jews, against
whom the decree was directed, were by far the largest community).
The decree did not survive the fall of the Empire for very long. The
July Monarchy definitively granted the Jews equal status in law and
recognition of their denominational rights, by creating a budget for the
Jewish religious establishment. French Jews, now legally assimilated,
entered a relatively happy phase of their history as a dynamic community
of bourgeois, of shopkeepers and artisans dominated by eminently suc-
cessful bankers and professionals. This community was increasingly
urban, Parisian, and—in short—bourgeois in character, cemented by a
secularized complex of traditional religious beliefs, by a strong minority
feeling, and by a deep attachment to France’s liberal image.
Was it an ‘assimilated’? community? Patrick Girard’s whole book is
centered on this basic question. In his view, assimilation is a dynamic
process “in which individuals belonging to a community with a distinctive
way of life attempted to formulate a doctrine combining a very old
religious heritage with the values of the surrounding society, values
intended to justify the preservation of an identity reduced to its denomi-
national aspect.” This somewhat weighty vocabulary, nonetheless useful
for its relative precision, expresses a positive judgment on the cultural
crossbreeding arising from the encounter between the Jewish tradition
and Western liberalism.
The author feels no idealized regret for the old, closed community. He
does not project on these mid-nineteenth-century French Jews the catas-
trophes of the twentieth century. Surprisingly, for such a young man (he
was born in 1950), he does not detest the ‘“‘middle-of-the-road” Judaism
of moderate and prudent notables, who preserved traditional practices in
the privacy of family life while adapting themselves to the modern world
in their professional and public life. He admires them for having handled
so skillfully the intercommunity tensions between the orthodox, who
rallied round L’Univers israélite, and the liberals, who published the
Archives israélites.
In both camps, French Jews had become israélites. This was their way
of saying that they were no longer Jews like those of the eastern Euro-
pean ghettos but that they nevertheless remained Jews in their attach-
ment to a past and a religion. These ‘‘assimilated” communities, products
of the encounter between Judaism and the French Revolution, were not
national communities, but neither did they mark the end of Judaism.
They represent a phase of its history that was “just as Jewish and collec-
tive as other [phases].”
This analysis echoes—without ever saying so explicitly—a fun-
damental debate in Jewish historiography and particularly a book in the
same series, published in 1973; the book of a Canadian professor,
Michael Marrus, devoted to French Jews during the Dreyfus affair.* The
two authors, while following each other chronologically, do not have the
214 Aspects of Modern Jewish History

same opinion on the nineteenth-century “‘assimilation” of French Jews;


nor, it is true, do they use the same analytical tools.
For Michael Marrus, assimilation is “the mechanism by which indi-
viduals of Jewish origin would assume a basically French identity.”’ The
period studied is obviously an ideal terrain for a definition of this type. As
the period was characterized by an extraordinary flare-up of nationalism,
it enables one to measure the extraordinary acceptance of this national-
ism by French Jews, even during the Dreyfus affair, in the name of their
patriotism. By a series of fascinating portraits and texts, the book shows
that, among the factors at work in this internalization of the opponent’s
values, there was more than the pressure of the surrounding society,
more than the desire not to add fuel to the flames. One could almost say
that there was anti-Semitism, particularly toward immigrant Jews, sym-
bols of a repressed Judaism. In fin-de-siécle France as described by
Michael Marrus, assimilation was the guiding thread of a Jewish anti-
Semitism. One could also find some grounds for this accusation in the
book by David H. Weinberg,’ which shows to what extent, in a later
period, between the wars, the French Jews of the Sixteenth Arrondisse-
ment and the Polish Jews of the Marais constituted in every sense two
watertight, if not antagonistic, societies.
Is one to conclude that the successful assimilation of the nineteenth
century led, thirty or forty years later, to an impasse? Of course there is
some truth to this statement, as evidenced, for example, by the explosion
in the decade from 1880 to 1890 of a massive and organized anti-Semitism
unknown during the July Monarchy or the Second Empire. But the gap
between the analyses of Girard and Marrus is not only due to the periods
they examine. I am inclined to believe that, on the contrary, these periods
have been chosen because of the two authors’ conceptual and political
divergences regarding assimilation. Patrick Girard has written a liberal
history of French Jews; Michael Marrus is searching that history for the
hidden sources of a national history. The first admires French Jews for
their extraordinary adaptation to modern civilization; the second cele-
brates only one fraternal voice in the Dreyfus affair, that of Bernard
Lazare, the first Jewish nationalist born in the French community.
Thus, Jewish history is drawn toward the dilemmas of the present.‘ If
the case of French Jewry takes on in this respect an exemplary value, it is
because it marks the invention of the Diaspora Jew—no longer the exiled
Jew, cut off from his land, yet still hoping to return to it at the end of time,
but the Jew of the Diaspora, comfortably settled in a foreign land, a new
metamorphosis of collective Jewish destiny. Do these “privileged Jews,’”®
to borrow the title of an article by Hannah Arendt—who inspired
Michael Marrus—represent the end of Judaism or just one of its
varieties?
Perhaps it is more useful to examine the question by starting from the
past rather than from the future assigned to that past as a hypothetical
Jews and French Democracy 215

outcome. Annie Kriegel, in an article in L’Arche,° reminds us that it was


the French Revolution that invented the two responses to the Jewish
question: the individual solution and the national solution. Naturally, it
preferred the first, because revolutionary thinking could not imagine a
better path to the “‘regeneration” of the Jews than individual success
within the model nation, namely, France. But, at the same time, the
Revolution spawned the notion that the emancipation of the Jews—like
that of any oppressed people—could assume a collective or national
form. The right of peoples to self-determination flows from the Declara-
tion of the Rights of Man. The conquest of legal equality and individual
liberty by the French Jews, by exposing them to the values of the sur-
rounding society, certainly secularized them but also contributed to their
fondness for democratic, national, and socialist ideas. These ideas also
played a part in the Jewish national awakening at the end of the
nineteenth century.
The ambiguity of Jewish “‘assimilation” stems perhaps from the ambi-
guity of the idea of Zionism itself. Unlike nineteenth-century “‘national”
movements, Zionism stands at the crossroads of a secular notion (na-
tional emancipation, the nation-state) and a religious notion (the return
to Jerusalem). ‘Assimilation’? weakens the second while promoting the
first. For that matter, one would have to examine closely what “‘assimila-
tion” abandoned or preserved of the spiritual tradition by “‘privatizing”’
and secularizing it. This analysis is somewhat lacking in the books I have
discussed. In the unstable mixture of community and individualism, of
national and religious ingredients, out of which the destiny of the Jewish
people was built in the nineteenth and twentieth centuries, nothing is as
simple as the peculiar hatred that accompanied the Jews through time.

The Vichy regime is the most tragic illustration of that hatred in


modern French history. Although it is true that it was established only in
the wake of the German victory of 1940, it represents a very distinctively
French type of regime, in many respects independent from its Nazi
protectors. This is the paradox once again explained by the American
historian Robert Paxton, in a new work devoted this time not to the
general history of Vichy but to its policy toward the Jews.’ No doubt this
explains why he enlisted the help of Michael Marrus, an expert, as we
have just seen, on “modern” French anti-Semitism.
France’s defeat brought to power a set of groups and individuals who
were not directly appointed by the Germans (even though, without them,
they would have stood no chance of coming to power) but owed their
success to their determination to revenge themselves on the Republic.
Their action was informed, not by foreign ideas, but by a national
tradition. Hitler’s victory provided them, not with the substance of their
policy, but with an opportunity for implementing it. The Jewish question
provides a virtually perfect example of this nationalist exploitation of
216 Aspects of Modern Jewish History

France’s defeat. The proof of this is that no clear connection exists—


either before or during the war, in Vichy or in Paris—between ideological
sympathy with Nazi Germany and the intensity of anti-Semitism: Doriot
was a fascist and not very prone to Jew hating; Xavier Vallat, an anti-
German of long standing, was an anti-Semitic minister in the Vichy
government; Céline, a Nazilike anti-Semite, was hostile to Vichy. The
politics and ideology of the first two years of Vichy were, in fact, charac-
terized less by racism, in the biological and Nazi sense, than by considera-
tions or pretexts related to the cultural cohesion of the national com-
munity.
One of the reasons for this is that, during the 1920s and 1930s, France
took in many foreigners, more than any other country in the world, the
United States included. The first wave of immigrants came in search of
work; they were followed by a stream of political refugees. France
displayed first a self-interested hospitality, then an increasing reluctance,
owing to the economic depression and the political situation. The “‘for-
eigner’’ became the scapegoat for the international complications that an
appeasement-minded France saw as portents of war. And the Jew be-
came the prime embodiment of this “foreigner,” since he represented
one of the most prominent issues of the coming war. After France lost the
war, he thus remained one of the most obvious ‘‘culprits.” Not only a part
of the Left but also a part of French Jewry were prisoners of this insane
belief. One need only read what a man as intelligent and refined as
Emmanuel Berl wrote in 1938 and 1939 to appreciate to what extent the
tragedy of the period was unfolding in a veil of deep mystery.
Furthermore, Paxton and Marrus cite the two statutory orders of the
spring and autumn of 1938, which reinforced control measures against
aliens—in particular by extending the obligation to maintain an assigned
residence—and prescribed the establishment of “‘special centers’’ for
those aliens (Spanish Republicans in particular) who were deemed to be a
threat to public order. Thus, Vichy was to inherit concentration camps
dating from the Daladier period; it was to use them systematically against
foreign Jews even before Germany adopted the procedure. One of the
most dishonorable episodes of this story was the internment of several
thousand German Jews who were fleeing the occupied zone in the sum-
mer of 1940 and were made so desperate by their internment that some
turned to the Nazis for protection against France! The fact is that, until
late 1941, Hitler’s policy aimed less at exterminating the Jews than at
chasing them out of Germany. In putting them—women and children
included—behind barbed wire at Gurs and elsewhere, in frighteningly
overcrowded and unsanitary conditions, Vichy was somewhat ahead of
the German timetable. Unknowingly, but without, however, being able
to rule out the hypothesis, the Vichy regime carefully laid the ground-
work for the tragic outcome.
Jews and French Democracy 217

In actual fact, the French anti-Semites did not advocate forced emigra-
tion. Very few Jews managed to leave the occupied zone between 1940
and 1942. Apart from internment, a fate reserved for foreigners, the aim
was to set up a special status, an idea borrowed from the monarchic
tradition and from the Action Frangaise. A second-class citizenship
within the bounds of the law was to be defined for French Jews. This was
the purpose of the decree of 18 October 1940, drafted by the minister of
justice of the first Laval government, Raphaél Alibert, former member of
the Conseil d’Etat and a faithful disciple of Maurras. But this decree,
which barred Jews from the civil service, was reinforced on 14 June 1941
by a new statute, extending the ban to the professions and to business;
this was the work of the commissaire général aux questions juives under
the Darlan government, Xavier Vallat. And what a highly typical figure
Vallat was: a right-thinking member of parliament under the Third
Republic, a former teacher in the Catholic school system who had be-
come a professional war veteran after the First World War, a Catholic
monarchist who had made the rounds of all the right-wing parties; not a
fascist, even less a germanophile, but simply a symbol of a milieu and a
culture that had finally taken its revenge on democracy. Professing an
anti-Semitism “‘in the national interest” for which he was actually at-
tacked by Parisian “collaborators,” Xavier Vallat was the perfect
embodiment of something worse than Vichy’s submissiveness: its duplic-
ity. On every level, he pushed anti-Semitic policy further than the Ger-
mans were asking, but in the name of his convictions against theirs. He
cloaked the unequal bargaining with the victor, which was both Vichy’s
principle and its excuse, with the domestic flag of French nationalism.
Thus, he was able to stake out an original position for himself in the
process whereby Vichy outrivaled the Nazis in their own policies.
But the most surprising feature in Paxton and Marrus’s merciless
description is not the existence of anti-Semitic ministers or doctrines at
Vichy, nor the relative passivity of public opinion, dazed by a defeat as
sweeping as that of June 1940. After all, this type of policy, this brand of
minister, and this spineless acquiescence of public opinion can ultimately
be explained by the Nazi victory and the state of French politics. What is
more surprising, and frightening, is the absence of any reaction on the
part of the civil servants who enacted all these measures, prosecutions,
and confiscations. The Conseil d’Etat examined the statutes concerning
Jews as if they were ordinary laws. Law professors demonstrated their
validity and novelty. The prefects enforced them. The police carried out
orders. Never, nowhere, did an official resign or even protest! The entire
French administrative machinery obeyed the new ministers as if they
derived their authority from the people in the normal way.
All these officials had served republican France. Their behavior can-
not easily be accounted for either by invoking their prewar attitudes (as in
218 Aspects of Modern Jewish History

the case of Alibert or Vallat) or by the all too easy recourse to the oldest
exorcistic myth of Catholic France (as in the case of the average French-
man). This sinister history nevertheless seems to demonstrate two things:
first, that what one can call the French elites of the period were much
more imbued with deep and lively anti-Semitic feelings than one would
be led to believe by the number of far-right deputies in the 1928, 1932,
and 1936 parliaments; second, that the incantatory recourse to the Jewish
plot or poison was perhaps more vital to bourgeois France than to
grass-roots France, in that the defeat provided the first with a twofold
revenge for the Dreyfus affair and for the Léon Blum government, while
the second was not asking for that much. Consequently, this recourse was
more a technique for conquering power and public opinion than an active
ideological and emotional act of communion. Of the two classical poles of
anti-Semitism, Vichy embodied manipulation far more than preaching.
This feature, it seems to me, gives it a distinctive—and, moreover,
particularly vile—character in the annals of this inexhaustible passion.
The next period is better known and tragically simple. In late 1941 and
early 1942, when Hitler decided to exterminate the Jews and inaugurated
the policy of deportation to the East, a policy extended to France in June,
Vichy’s “independence”’ in anti-Semitic matters came to an end. Dar-
quier de Pellepoix, the racketeering hoodlum, replaced Xavier Vallat,
the Catholic member of parliament. A card-index census of the Jewish
population was patiently put together by French civil servants and police
between 1940 and 1942. It became the matrix for the deportation of Jews,
which began with the massive roundup of the Velodrome d’Hiver (indoor
sports stadium) on 16 and 17 July 1942, carried out by 9,000 French
policemen. In its camps, Vichy held in reserve thousands of other foreign
Jews, whom Laval gradually handed over, on request. By the end of
1944, slightly more than 75,000 Jews (out of 300,000) had been deported
from France to the concentration camps located on former Polish terri-
tory. Some 2,500 returned.
Thus, from the autumn of 1942 on, the mechanism of wholesale
deportation continued on its lethal course, with the help of the censuses,
identity checks, and internments carried out by the French authorities. In
February 1943, in the southern zone, a series of arrests of foreign Jews
second in magnitude only to the July and August 1942 roundups was
carried out exclusively by the French police. Yet, as the liberation
approached, this task was assumed by the German police, with the help,
it is true, of the Milice. This is probably one of the reasons why the
deportation figures for 1943 and 1944 are lower than those for 1942, when
all the resources of the French police were at the Germans’ disposal.
Some 33,500 Jews were deported in 1943 and 1944 as against 42,500 in
1942 alone.
Unfortunately the relative decline in these sinister figures does not
mean that the Vichy government had finally taken a stand against Ger-
Jews and French Democracy 219

man policy. Paxton and Marrus cite only two examples of genuine
opposition. The first dates from the spring of 1942, when Vichy refused to
extend the obligation to wear the yellow star to the Jews of the still-
unoccupied zone; the second dates from the summer of 1943, when
Pétain resisted pressures to ‘‘denaturalize”’ recently naturalized Jews (by
virtue of a law of 1937) and to add to the categories of ‘“‘deportable”’ Jews.
But neither of these refusals led Vichy to protest against the deportation
of Jews or to withdraw its consent and support for the measures that
brought it into effect. The deportation of foreign Jews, or of French Jews
arrested for having violated the legal restrictions imposed on them,
ceased only with the liberation of the country.
Finally, what can one say about the balance sheet? A classic argument
in favor of the thesis that Vichy was a “‘shield” is that three-quarters of the
Jewish community living in France in 1939 survived; moreover, it is
known that the overwhelming majority of the 75,000 Jews deported from
France to the camps where they were exterminated was composed of
foreign Jews. And this dismal distinction, borrowed from Vichy, still
serves as its posthumous justification. In reality, the argument would
have a demonstrative value—within its own anti-Semitic logic—only if it
could be accompanied by proofs showing that the Vichy government
actively protected French Jews from deportation after having persecuted
them by its laws. But these proofs do not exist, and no one has ever been
able to produce any. Furthermore, Paxton and Marrus show that there is
no link between the number of surviving Jews and the various policies of
occupied or satellite countries. The tragically essential factor in the
extermination of the Jews, and in its magnitude in each European state,
was the means implemented by the Nazis for that purpose.
Thus, one cannot give Vichy policy the credit for what was actually due
to circumstance. If three-quarters of the French Jewish community suc-
ceeded in escaping deportation and genocide, it was owing to a shortage
of manpower in the German police operating in France and to the
possibilities for shelter both in the Italian occupation zone and in rural
France. Public opinion in these areas became increasingly sensitive to the
Jewish plight as the prospect of Nazi defeat became clearer and the
Service du Travail Obligatoire simultaneously deported greater numbers
of workers to Germany. But none of this is attributable to a determina-
tion or to a policy—even at a clandestine level—of the Laval government.
As far as Vichy’s action proper is concerned, the historian can only
document its original and autonomous contribution to Nazi anti-
Semitism, particularly between 1940 and 1942, and its objective contribu-
tion to extermination, through its own concentration camps, its police,
and the assistance it provided the Germans in their roundups, particularly
in connection with the foreign Jews who took refuge in France between
the wars.
The evidence marshaled by Paxton and Marrus thus leads the reader
220 Aspects of Modern Jewish History

less to a condemnation of France in general—which would not make


much sense—than to a harsh verdict on the section of its governing class
that found in France’s defeat an opportunity for taking a long-awaited
revenge on the Republic. That people do not behave heroically in misfor-
tune is not a new phenomenon, nor one peculiar to France in 1940. But
that a certain part of the Right should have availed itself of the national
debacle to take its revenge for 1789 and the Dreyfus affair is instead
characteristic of our history. Not that the process was determined solely
by indigenous factors, since, on the contrary, it took the Nazi victory to
make such revenge possible. But this episode does belong to our history
in that its protagonists represented the repressed forces, the camp of
those who had been defeated a century and a half earlier, or at least since
the birth of the Third Republic. Vichy France revealed the obverse side
of democracy inseparable from our history, and inseparable, too,
perhaps from democracy itself. In this respect, the work of Paxton and
Marrus does more than explain the facts and establish responsibilities; it
also teaches a moral lesson.

Since the war and the inglorious disappearance of the Vichy regime,
my generation especially remembers having lived through a period
haunted by the return from the camps—a return that also meant the
discovery of the camps. But the taboo that consequently weighed on
anti-Semitism also led to a somewhat furtive burial of the Jewish ques-
tion. A spontaneous conspiracy of silence, as it were, was hatched to
exorcise this specter. Gentiles tried not to think of it, and Jews carried a
secret too heavy and too peculiar to talk about.
But this uneasy equilibrium between a guilty conscience and the
willingness to efface was upset ten or fifteen years ago. The generation
that reached twenty at the end of the 1960s—Jews and Gentiles alike—
rediscovered the Jewish problem. To understand how and why this
occurred, one can start with the family photograph taken in 1979 by two
journalists, Harris and Sédouy.* From this collection of interviews con-
ducted by them in all the milieus of French Jewry, only a single common
feature emerges: anxiety. No doubt this is an age-old trait, but, in the
midst of the apparently reassuring uniformity of our prosperous society,
itis more alive, more entrenched, and more varied than ever. And what if
France were to become once again a country where Jews could not live?
This question is always latent in the answers given in the interviews; it is
even implicitly voiced by French Jews who have “always” been assimi-
lated.
This patent erosion of the democratic certainties of assimilation can be
clearly explained by certain structural reasons. French Jewry (600,000 to
700,000 French Jews) is a population replenished by a recent influx of
North African immigrants, whose arrival has modified the balances
within the community. The assimilated bourgeoisie is less predominant
Jews and French Democracy 221

among the North African Jews, and the assimilationist creed is not
necessarily accompanied by the abandonment of religious practices and
traditional ways of life. Consequently, the face of French Jewry has
changed and become more visibly Jewish. This greater capacity to assume
the particular Jewish heritage (even in its secularized form, that is, as a set
of habits more social than cultural in nature) poses novel problems for all
French intellectual traditions, both left-wing and right-wing. To under-
stand this, one can listen to Rabbi Guedj expressing regret—in the
unlikely setting of Sarcelles—for the secular environment of Constantine
and answering the question “‘Is it possible to live in a country without
espousing its dominant philosophy and its way of life?” in the following
words: “Tell me, please, what is the dominant philosophy in France
today? I cannot go along with those who think like that. That’s how
Nazism started... .”
The thoroughgoing renewal of French Jewry was marked by two major
and equally destabilizing events (if one takes assimilation as a reference
point): the Holocaust and the birth of the state of Israel. The first
exterminated nearly all the eastern European Jews who had recently
immigrated into France. The role of the Vichy authorities also struck a
deep blow at the belief of assimilated Jews in the safety of their own
destiny. It showed them that even in France, a hundred and fifty years
after the Revolution, their future was not settled forever.
On top of that, after the war, came the birth of Israel and, twenty years
later, the Six Day War, which seems to have played a crucial role in
crystallizing the attitude of French Jews toward the new state. French
Jewry, no longer so assured of being able to stake everything on demo-
cratic assimilation, could at least refer to a new image and at most count
on the help of a new homeland. Through innumerable channels, Zionism
transformed the Jewish consciousness in the Diaspora by confronting it,
on the strength of Zionism’s own historical success, with a series of
ultimatum-questions that, in accordance with the nature of Zionism, led
from the religious sphere to the national sphere. At the same time, the
international dimension of Israel’s existence led the Diaspora communi-
ties to establish forms of solidarity that are probably in turn transforming
anti-Semitism into a worldwide phenomenon, in an unprecedented form
and on an unprecedented scale.
Hence the new tremors in Franco-Jewish identity that one can discern
in the interviews assembled by Harris and Sédouy. Thirty-five years after
the war, one can measure the impact of a twofold evolution: on the
gentile side, the ban on anti-Semitism imposed when death-camp survi-
vors returned is gradually weakening, not because it has disappeared
from living memory, but because the young generations do not have that
memory; and, on the Jewish side, the community is much more willing
today than before or just after the war to assert and accept its distinctive-
ness, as regards its special ties with Israel, its traditional way of life, its
222 Aspects of Modern Jewish History

culture, or its religion. Even the intellectual climate in which we have


been living for the past several years has contributed to this evolution by
extolling ethnic minorities, criticizing the Jacobin state’s passion for
leveling, and somewhat naively rediscovering the religious experience
behind the illusions of Western nationalism. Thus, fashion confers its
transient appeal on the age-old problem of Jewish identity. The Jewish
people, miraculously adapted to ‘“‘modernity,” have never ceased,
however, to be simultaneously. rooted in archaism.
Thus, French Jews have rediscovered that their prospects are always
uncertain: on the religious level, on the national level, and on the level of
individual destinies. They have consequently reinvented one of the clas-
sic accusations leveled at them by anti-Semitism: that of being a mobile
and universal people, present everywhere and nowhere. But it is to their
credit that they accept this role and do not shun it. For they know that the
history of the last half century has made them, as it has made European
Judaism, the prime symbols of modern tragedies. The Holocaust has
tragically deprived them of their happy belief in assimilation, and the
calamities of communism have ruled out a recourse to revolutionary
messianism. Apart from seeking refuge in archaism and traditional soli-
darity, the only choices left in ““modernity” are difficult, uncertain, or
disheartening.
One should read, for instance, in Harris and Sédouy, the reasons given
by Maurice Kriegel for his departure to Israel. A former French leftist of
1968, a descendant of an Alsatian Jewish family deeply assimilated ‘‘on
the Left” through Jacobin patriotism and socialism, this young history
professor at the University of Haifa sternly criticizes his adopted home-
land. For example, he detests its all too often anti-intellectual and paro-
chial reactions; but his choice was dictated by exactly the same reasons
that led his ancestors, in the nineteenth and twentieth centuries, to
choose the France of the Revolution: so that his children’s happiness (or,
for that matter, their misfortunes) should no longer be ‘“‘specifically
Jewish” but determined by other factors—by so-called normal factors.
What led his ancestors to Paris led him to Haifa.
By the hazards of publishing, the appearance of this self-portrait of
French Jews coincided with that of a short book by Alfred Fabre-Luce
titled Pour en finir avec l’antisémitisme.’ Let us be thankful for this
coincidence, which enables us to compare Jewish anxieties about the way
of the world and the voice of a French intellectual whose ambition is to
soothe them. In this particular case, the parallel offers a nearly perfect
example of lines that never meet. To soothe, one would have to lend an
ear. But what does Fabre-Luce hear? Next to nothing.
And yet he has an intuition of the basic phenomenon—of the fact that
the assimilationist creed is in a state of crisis and that the two phases of
this crisis were the experience of the Second World War and Zionism.
Jews and French Democracy 223

But this notion, which should mark the starting point of any discussion of
the topic, is where Fabre-Luce leaves off. Taking up the idea of assimila-
tion precisely where the Jewish community has left it, he turns it into a
sort of no-man’s-land protected from Christian anti-Semitism and Jewish
proselytism. Assimilation becomes an abstract, unreal zone where pas-
sions are extinct and anxiety has vanished because only the author’s
limpid ideas prevail. In the typically French realm of blind intelligence,
Alfred Fabre-Luce finds a refuge from the historical density of a problem
that he finds more shocking than intriguing.
His idea is to propose to French Jews a sort of pact that would define a
minimal consensus, which both parties would endeavor to maintain: the
gentile French would act to prevent the rebirth of anti-Semitism as a way
of drawing attention away from national difficulties, and French Jews
would abide by a sort of code of good conduct, characterized by the
abandonment of cultural proselytism and Zionist militancy. In this cu-
rious deal, two concrete clauses could serve as guarantees: the search for
a common position of the French on the Middle East and the establish-
ment of a “‘good”’ version of the history of Vichy.
The mind boggles at these chapter headings. The first belongs to the
realm of wishful thinking—an idea that is unworkable in practice and,
moreover, not very interesting in the abstract. But the second is more
revealing of Fabre-Luce’s polemical obstinacy: he wants to rewrite the
history of Vichy.
Indeed, the bulk of his short book is devoted to an analysis of Vichy, a
crucial period in the relationship of Jews with France, especially if one
focuses exclusively on assimilation. For Fabre-Luce, there is a Jewish
version of the history of Vichy, characterized by an emphasis on the
purely French responsibility for the persecutions. Curiously, the pro-
tagonist in this enterprise is the American professor Robert Paxton (a
Gentile, and rather of the southern aristocratic sort). I shall not dwell on
this curious assertion, since Fabre-Luce knows that, naturally, one finds
Gentiles in every Jewish lobby. But what I find bizarre, and even some-
what disturbing, in this type of historiographical analysis, is the suspicion
that there exists a Jewish determination to produce and circulate a history
of Vichy that is contrary to the truth and serves the interests of an
avenging minority.
I have no objection—far from it—to discussing the editing of Le
Chagrin et la Pitié or certain excessively systematic allegations made by
Paxton. But what seems less admissible to me is to pin the same yellow
star on Ophiils, Paxton, Stanley Hoffmann, and Léon Poliakov, whom
Fabre-Luce regards as a lobby of historians and film makers entrusted
with the task of ‘‘presenting to the French the history of their war years
through a Jewish prism.” Indiscriminate amalgamations seldom proceed
from a form of thinking that is not somewhat crude. In this instance, the
224 Aspects of Modern Jewish History

notion of a Jewish plot is being reintroduced into the interpretation of our


modern history, turning the victim into a culprit. Instead of “putting an
end to anti-Semitism,” this argument feeds and even expresses it.
Consequently, such an approach fails to deal with the problem it
raises. Let us for a moment give Fabre-Luce the benefit of the doubt and
accept his hypothesis of “Pétain the shield.” Even if that were true, one
aspect of the history of French-Jewish relations would still require ex-
planation: why a defeated people resorted to a familiar demon to exorcise
its defeat, under the authority of a national glory. The enactment of the
statute of 1940 concerning the Jews was an act of French politics; the
statute was drafted and implemented in the name of a national tradition
by men who were not forced to do so by any foreign pressures. And if the
generation of assimilated Jews, even of those who supported Pétain for a
few months between June and the autumn, saw his gesture as a break by
the French with the tradition of 1789, it was because Vichy proclaimed it
to be just that. Politics in a democratic age is a simple art, in which
ambiguity is never long-lived. In this case, the ambiguity ceased within
three months. |
Alfred Fabre-Luce does not feel this wound, which is still unhealed
and cannot be healed by any comparative quantification of the history of
Jewish suffering in Europe. More generally, his treatment of the Jewish
problem reveals a deliberate insensitivity to the somewhat overwhelming
nature of the question. This is both strange and typical. It is strange on the
part of a knowledgeable and discerning mind, but typical of the way the
much-vaunted French intelligence grapples with a question whose histor-
ical and psychological implications are almost infinite. Finally, one must
take into account prevailing currents of thought: at present, in late 1979,
Paris is under the sway of an arrogant variety of “liberal” thought that
seems to me contradictory to the spirit of liberalism. The impassioned
attempt to “normalize” the Jewish problem and the collective existence
of French Jews (or Jews in France) is a good example of this arrogance—
shallow thoughts on a vast question. The fact is that the Jewish problem
cannot be “‘normalized.’’ One must live with it, such as it is, in the present
day. Everyone’s freedom depends on this acceptance, because it is one of
the criteria of democracy.
Gershom Scholem and
Jewish History

Gershom Scholem, born in Berlin in 1897 and settled in Palestine since


1923, belongs neither to the Jewry of eastern Europe that supplied troops
to Zionism after the First World War nor to the community of German
Jews who fled their country after Hitler came to power. He has had no
direct experience of the Polish or Belorussian ghettos, nor of their
traditionalist rabbis, who unwittingly fostered the growth of Zionism as
an angry modernist movement. On the contrary, it is from within this
modernity—that is, the assimilation into German culture—that he be-
came aware at a very early stage, even before the First World War, of the
irreducible specificity of Judaism. The Zionism he espoused as a young
man was thus—as was Theodore Herzl’s Zionism, for that matter—a
resurgence, a rediscovery of sources. It was not a revolution, even less a
revolt, but a history.
In this respect, the long interview in one of his books,’ in which the old
historian talks about himself, is immensely fascinating. Scholem was born
into a small dynasty of Berlin printers that went back four generations.
For the Scholems, the Friday-evening family dinner represented the last,
infinitesimal vestige of old religious observance. The father, and his
milieu, believed in assimilation, in the German identity, in German
patriotism. This petty Jewish bourgeoisie, whose outlook and social life
remained strictly Jewish, lived its cultural life on a borrowed tradition.
Scholem perceived this rift by his adolescence, at the age of fourteen or
fifteen; his reaction was more one of unease or revolt than of analysis. He
wanted to be a Jew, not a German. “So you want to return to the ghetto?”
asked his father. “But it is you who are living in a ghetto, even though you
think you are not,” answered the son.

First published in Le Nouvel Observateur, 19 February 1979.

225
226 Aspects of Modern Jewish History

As if to round out the story and give the example of family the
pedagogic simplicity of an ancient tragedy, Gershom Scholem had three
brothers. Two of them sided with his father, and one of these two was
even more German—in fact, an outspoken German nationalist. The
fourth brother, in contrast, soon joined the Social-Democratic party,
became a Communist member of parliament after the war, was expelled
from the party as a Trotskyite in 1927, and died at Buchenwald at the
hands of the Nazis precisely for having been a Communist member of
parliament. Thus, the Scholem children entered the twentieth century
each under his own flag: assimilation, revolution, and Zionism. Of the
three options for solving the famous “‘Jewish question” that accompanied
their childhood, the third seemed at the time the most outlandish and the
most shaky; it alone was to withstand the century.
Moreover, Gershom Scholem’s books continually explore this variety
of Zionism, for it represents not just a youthful awakening but the central
question of a lifetime. It was less a political commitment than a moral
choice. The young Scholem was motivated not by the desire to build a
Jewish state but by the refusal of the duplicity inherent in the daily
existence of the Jewish bourgeoisie in Germany. Basically, it is the very
same feeling that led one of his brothers—as if by chance, the only one
with whom he had close ties—toward revolution and socialism. Admit-
tedly, it is hard to explain what drove one brother to Marxism and
Leninism, the other to study with a professor of Hebrew and to take
Talmud courses: such are the mysteries of individual existence. But if one
accepts the idea that their contradictory choices had a common origin,
one cannot help observing that the choice of the future historian was
better suited to the problem at hand than that of the future Communist
deputy of the Reichstag. The first wanted to identify the elements and
sources of the problem, while the second sought to eliminate its very
terms and thus began by negating them.
In this respect, the Zionism of the young Gershom Scholem, because it
was a postassimilation Zionism, a return and not just a breach or a vision
of the future, was inseparable from the central paradox of the entire
movement—from the chemical process that produced national ideas
from religious ideas, modernity from archaism. In the Zionist movement
to which its youth rallied at the turn of the century despite their rabbis’
hostility, the ghetto Jewry of eastern Europe saw only its new, secular,
modern, socialist, and nationalist aspects—Zionism as a break with the
past. In contrast, because he was the rebel child of a historical Jewish
experience rooted in the cult of modernity and in the repression of
tradition, Scholem knew in advance that the Jewish national movement
conveyed, took hold of, and revitalized an age-old cultural heritage
through the agency of modern political categories. What the Jews have in
common—and what provides the least unsatisfactory explanation for
their extraordinary survival and their national ambition to return to
Gershom Scholem and Jewish History 227

Zion—is a religious anthropology. Gershom Scholem’s entire work is


contained within the historical exploration of this proposition.
Naturally, this meant that the historian had to define his relationship
with his native cultural soil, Germany. Scholem discusses this subject at
length in several of the studies collected in Fidélité et Utopie: he sees no
cross-fertilization between Jewish culture and German culture. The
assimilated Jewish bourgeoisie gradually gave up its own tradition, but
the Germans never recognized the Jewish tradition. Modern Jewish
historiography, founded by Leopold Zunz in the nineteenth century,
never obtained a university chair, and the discipline was never granted
academic or social legitimacy. Nor can one find in nineteenth-century
German literature the equivalent of the well-known passages of Notre
Jeunesse in which Charles Péguy hails Bernard Lazare as a descendant of
the prophets of Israel—even as French Jews, out of prudence or igno-
rance, remained silent about one of their most daring sons. Replying in
1962 to a young German writer’s question on the Jewish-German dia-
logue, “‘whose core is indestructible,’ Scholem denied that such a dia-
logue ever existed: “The unbounded exhilaration of Jewish enthusiasm
never aroused a corresponding note that had the slightest connection with
a creative response to the Jews, that is, a response that would have been
made to them on the basis of what they could contribute as Jews, and not
on what they had to abandon as Jews.”’ For Scholem, the Nazi genocide
did not, as it were, accidentally interrupt an exchange or even the
beginnings of an exchange. Rather, it was a tragic sign of its absence. It
crowned the violence wrought on Jewish minds by annihilating their
bodies.
Yet it so happens that there was one exception to Gershom Scholem’s
militant demonstration: himself. The great German historiographical
tradition of the nineteenth century, when grafted onto the study of the
Jewish tradition, produced at the very end, just before the catastrophe, a
practitioner of a history of Judaism emancipated from its overbearing
tutors. In reaction against most of his predecessors and their tendency to
“flatten,’’ to secularize, the Jewish tradition in order to make it accept-
able to the assimilated bourgeoisie of the nineteenth century, Scholem
has devoted most of his work and his life to studying the aspect of this
tradition that may seem the most alien to modera rationalism: the Judeo-
Spanish Cabala mystique between the thirteenth and sixteenth centuries,
and especially the extraordinary offspring of the cabalistic teachings of
Isaac Luria, the wise man of Safed in Galilee—the Sabbatian movement
of the mid-seventeenth century. The entire Jewish world, from Yemen to
Morocco, from Poland to Holland, went wild over the pseudo-Messiah
Sabbatai Zevi, who ended his life with an inglorious conversion to Islam.
In his scholarly way, Scholem dissects this episode to show the infinite
plasticity of the religious phenomenon at work, from its adjustment to the
expectations contained in the messianic scenario down to its later
228 Aspects of Modern Jewish History

ramifications, leading to the philosophy of the Enlightenment. This ex-


traordinarily learned thousand-page book displays the intellectual eccen-
tricity characteristic of Scholem’s talent. The young man who scandalized
his milieu of good German bourgeois by learning Hebrew has become the
historian unearthing before the Israeli rabbinical establishment the awe-
some treasures of Judaism.
Indeed, if, by devoting his life and intellectual powers to the endless
reconstruction of the expressions of the Law and the promises it contains,
Scholem sought to épater le bourgeois—the twentieth-century bourgeois,
in whatever guise—then, he has succeeded only too well! His historical
work does not fit easily into any of the intellectual traditions prevailing in
Europe. Moreover, what are known as “Jewish studies” are concentrated
today in the United States and Israel alone. But, while my lack of
competence in the field prevents me from giving a specialist’s introduc-
tion to his work (on this point, I can only recommend the admirable
Jewish Messianism, published in French a few years ago), one can
probably examine in some depth here what makes his work so closely
connected to the world in which we live.
Unlike his immediate predecessors, Buber or Rosenzweig (whom he
constantly uses as reference points in defining his own position), Scholem
does not study Jewish mysticism from the inside, as a closed and timeless
universe. He is neither a theologian nor even, strictly speaking, a be-
liever, but purely and totally a historian. He manifests toward Judaism
not only a filial affection but an admiration such as romantic historiogra-
phy displayed toward the inexhaustible wealth of cultural or national
phenomena. The underlying feature of his entire work is basically the
simple and baffling question that we all ask ourselves as contemporaries
of the genocide and of the birth of the state of Israel: After centuries of
misfortune, culminating in the absolute tragedy, how is it that Judaism
still exists? Scholem retraces, strand by strand, the thread of this survival,
through its cultural—that is, religious—forms, and the endless variants of
the notions of God, the Law, and the Messiah devised by the Jewish
people in its exile throughout Europe.
An old people, endowed with an age-old culture, ceaselessly rework-
ing its heritage in order to construct a distinctive identity in every cir-
cumstance—such are the Jews, according to Scholem. Theirs is the
history of a faith more than of a developing society. But if one treats
Judaism as a religious culture, while excluding from its definition both
rabbinical orthodoxy and modern criteria of national identity, where
does one put Zionism and the state of Israel? Scholem relegates them to
the profane, that is, the political level. A man who left Berlin for Jeru-
salem in 1923 cannot be suspected of underestimating their importance.
Zionism, in his view, represents the advent of the Jewish people in
political history, the assumption by a Jewish state of its historical duties.
But this order of realities does not encompass the spiritual and religious
Gershom Scholem and Jewish History 229

sphere. Zionism fulfills no promises; it realizes no messianic expecta-


tions. It marks a renewal in Jewish history, but precisely at the historical
level, not at the eschatological level.
If Scholem watches so zealously over the dividing line between the
religious and the profane, between the spiritual and what he calls “‘the
historical,” it is primarily because he is a religious man, with a religious
view of the continuity of Judaism. For this old opponent of the rabbis, the
human adventure has no meaning other than that conferred on it from the
outside by revelation. But if he puts such emphasis on the hidden mean-
ing of Israel—a meaning that remains to be discovered—it is because he
knows the dangers that threaten a ‘‘historicized’’ Jewish people: the
dangers of being a nation like the others, of contributing acts of injustice
to history, of rebuilding the Bronx in Palestine, of turning the descend-
ants of the Hasidim and the Cabalists into “‘modern”’ petty bourgeois. At
the close of his life, the old historian recognizes in the Jewish state itself
the specter of the demon he fought in his father’s house in Berlin at
fifteen: Jews in danger of being dispossessed of their past.
In the words of R. P. Blackmur, quoted by Leon Wieseltier in an
article devoted precisely to Gershom Scholem: “‘Beholden to his fathers,
the Jew is still in search of a son.’”
Israel and the French Left:
The Misunderstanding

As long as Israel was only the reflection of Jewish destiny, that is, of
Jewish suffering, it was not hard for the Left to love Israel. Indeed, the
Left saw the new state as the embodiment of its traditional conception of
the Jewish problem as the sign of capitalist injustice, of the Jews as the
exemplary victims of a system, and of Jewish distinctness as the obverse
of persecution. The Second World War and the Holocaust brought this
feeling to a peak—a feeling theorized by Sartre: what “‘coagulated” the
Jewish problem was the anti-Semitic hatred displayed by Christian
Europe; the men and women who in those years were settling in their new
homeland could not be colonizers, since they were wanderers, survivors
of the tragedy in the camps. Even this dignity was not enough to preserve
them from suffering, since they were being rejected from shore to shore
by their new persecutors: the British. In fact, the French Left had not yet
learned to hate colonizers, whom it was often fond of regarding as the
overseas exporters of its own values. It felt far more guilty about the
plight of the Jews than about Arab or Vietnamese suffering.
This democratic universalism, so popular and yet so sectarian in the
immediate aftermath of the war, was dealt a heavy blow by the decol-
onization wars. Once again, the force and legitimacy of nationalism
became manifest, in their simplest and yet most conceptually elusive
aspects. A new Left began to define itself with reference to this struggle,
which was helping to put an end to Jacobin Europocentrism. The new
Left gradually exerted its pressure on the two great traditional parties by
waging intellectual guerrilla warfare on the outskirts of the Communist
party and by provoking a definitive split within the Section Francaise de

First published in Le Nouvel Observateur, 15 May 1978.

230
Israel and the French Left 231

Internationale Ouvrière (SFIO) through the creation of the Parti


Socialiste Unifié in 1959.
Israel was not only left out of this anticolonialist movement; it was also
a logical scapegoat for it, as evidenced by the Suez crisis, when Ben
Gurion’s troops came to the aid of the paratroopers sent by Guy Mollet
and Eden against Nasser’s Egypt. The short war of 1956—which, with
hindsight, still seems to me the major error in Israeli foreign policy—
nevertheless established, with a sort of miraculous malevolence, the new
image of the Jewish people for the outside world. It was a political image,
in which the Jew became the ally, the accomplice, of the old colonial
powers; but also a moral image, which destroyed the stereotype of
humiliation and created the victorious Jew.
This intellectual and emotional turning point was not always easily
accepted by the French Left, which, as we have seen, needed the pres-
ence of calamity in order to conceive the Jewish identity—or, if not
calamity, at least the threat of calamity. This is the secret of the explosion
of sympathy, and even affection, for Israel in French public opinion on
the eve of the Six Day War: the Jewish state was threatened with annihila-
tion, and the words of Shukairy conjured up the ghost of Hitler. But the
fact remains that from Suez onward, Israel’s image was also that of a
soldiering people, strong and victorious. But, just as in the nineteenth
century the French Left had had the greatest difficulty in evoking the rich
or simply “successful” Jew without sinking into anti-Semitism or an
outright denial of the problem, so it had no intellectual tools for conceiv-
ing Israel without evoking the tragedies of European history and the
French Left’s own share of guilt in them.
On the contrary, the intellectual tools developed by the French Left in
the course of its struggles may well be an obstacle to its understanding, no
longer the Jewish destiny, but the Israeli destiny. Once oblivious to
colonial realities, it has now made anticolonialism a dominant element of
its ideology. Once obsessed with Jacobin narcissism, it has made anti-
imperialism the key interpretative concept in a now-global history.
Through this prism, Zionism appeared to be moving against the historical
current: it was an “objectively” colonial movement at a time of world-
wide decolonization, an “‘objectively” imperialist force at a time when
the struggle against imperialism was a watchword the world over. The
passengers of the Exodus had once again become colonizers, admittedly
unwilling ones, but colonizers all the same. And the suburban Chicago
Jew who paid an annual contribution to his second homeland unknow-
ingly revived one of the oldest images of anti-Semitism, that of all-
powerful international financiers fighting against the freedom of peoples.
Naturally, in order to discourage this type of analysis, Iam more or less
caricaturizing it. But my reason for doing so is that I do not think it
derives its strength from any conceptual rigor. Rather, it seems to me to
be a rearrangement of old elements drawn from French political cul-
232 Aspects of Modern Jewish History

ture—most of which, incidentally, are common to both Right and Left.


First, there is a latent anti-Semitism, which is never as widespread as
when it is directed against the ‘“‘successful” Jew: in the new configuration,
the victorious Israeli takes the place of the Jewish plutocrat in the
nineteenth-century right-wing or left-wing imagination. An even stron-
ger ingredient is the hostility toward the United States, one of the
greatest passions of the French Left since the end of the war: Israel
became a substitute object for this hostility, enabling France to reap an
easy benefit as interpreter of the Third World. Thus, the Jewish state
took on the function traditionally assigned to the Jews in our political life,
by serving after a fashion as a witness to French national greatness. The
best expression of this amalgam of ideas and feelings can be found, in my
view, in what has been called “‘left-wing Gaullism,” located precisely at
the confluence of Right and Left; but it would not be hard to find other
traces of it in the Communist party, the PSU, and on the left of the
Socialist party.
For the Left still bases its vision and understanding of Israel on the
impulse that led it unanimously, thirty years ago, to hail the birth of the
new state as an act of reparation. It never tried to develop a doctrine
corresponding to that feeling, as if the latter were simply a powerful sense
of guilt, irreducible to rationality. In truth, the Left has never had very
much to say about the two elements that would need to be analyzed and
confronted in order to attempt to interpret the history of Israel: the
national phenomenon and the religious phenomenon. In all its versions,
Marxism ignores these phenomena, or, rather, it conjures them away
instead of explaining them. The nearly irresistible reaction of the French
Left to national phenomena is to give them a universal dimension, by
means of which they can be inscribed in the great book of world history
(as phenomena pointing in the right ‘‘direction’’). Thus, the slightest
tribal conflict is dignified with the status of democratic universalism, and
every Chadian maquis becomes the successor to the Spanish Republi-
cans. But Israel is precisely excluded from this universalist vision in that it
is the product of a latter-day national European movement, impossible to
fit into the dialectical process of colonial liberation, and even (if one
wants to cling at all cost to this frame of reference, which is more than
sufficient to bias the analysis) contradictory to that process.
As for the religious aspect of Israel as a historical phenomenon, it quite
obviously contributes to the Left’s incomprehension, since, in the Euro-
pean tradition, national emancipation and secularization of the state have
always gone together. But there is more. When the debate over Israel has
been freed of the unsavory variety of metaphysics that sometimes encum-
bers it, the fact remains that the (re)creation of the Jewish state in the
mid-twentieth century, in biblical Palestine, is not an easy event to
understand if one clings to a materialist interpretation of history. The
circumstantial factors—above all the Holocaust—are easy enough to
Israel and the French Left 233

perceive, but they do not provide an exhaustive explanation of the


phenomenon, which is inseparable from Jewish messianism. In this field,
Gershom Scholem is a better guide than Marx, but the Left is in no way
prepared by its culture for a reading of that great work! When the Left is
secular, it is Marxist. When it is not secular, it is Catholic. In both cases,
its approach to Jewish history is conditioned by intellectual traditions that
negate that history.
That is why the Left, ever since Israel donned the features of a strong
and victorious country, has never stopped oscillating between its emo-
tionally charged memories and the notion that Israel is “a country like
any other.” When the Jewish state is threatened, the specter of final
extermination looms up again. After an Israeli victory, café strategists
take out their maps and start talking about “rightful boundaries.” Nei-
ther of these attitudes contributes one iota to a better understanding of
the problem: neither that of identification with guilt nor that of “rational”’
diplomatic arbitration. The first attitude will eventually fade away, ex-
cept from Israel’s memory; moreover, it is incomprehensible to the Arab
world. The second presupposes that the Israeli-Arab conflict is in a sense
“normal,” whereas it lies precisely outside the norms of international
politics, since its roots are exceptionally deep and complex.
In truth, there is no conceptual or emotional shortcut to a full under-
standing of a country that embodies mankind’s oldest written memory as
well as the miracle and tragedies of modern Europe—a country that is the
offspring of religion, nationalism, and socialism. Thirty years after the
birth of the state of Israel, nothing is yet clear in its history except to the
simpleminded. There is still only one way of approaching Jewish history
without avoiding its singular destiny: it is to confront its specific features,
which alone can shed light on its universal import.
Ae
Israel, Zionism,
and the Diaspora

The fascinating feature of the history of Zionism is that it reveals better


than any other episode the radically unpredictable character of twentieth-
century European history.' Let us imagine for one moment that an
opinion poll could have been carried out at the turn of the century—when
the notion of Zionism was just emerging—on the following question:
During the next fifty years, which of these two events seems to you most
likely to occur: the advent of socialism in Germany (or France, or
England) or that of a Jewish state in Palestine? Public opinion, and
particularly among educated people, would have overwhelmingly chosen
the first hypothesis. By indulging in this exercise in imagination, public
opinion would have based its erroneous prediction on an optimistic vision
of the relationship among democracy, socialism, and nation in the
Europe of the future. Zionism, instead, invented another future by
placing itself on the fringe of this triangular system, or, if one prefers, on
its obverse side: it took as its starting point the situation of the Jews in
Europe, but, in order to understand and ‘‘resolve” this situation, it
adopted the three key elements of European political civilization. The
Zionist movement advocated a break in the name of an identity, a
separation from the very environment that had shaped it. And it is
through this contradiction that Zionism provides the observer not only
with the secrets of its success but with the best vantage point for analyzing
its current problems.
Zionism was born at the confluence of two currents, of two visions of
European Jewry, whose combined effect was to anathematize the way of
Published in Le Nouvel Observateur, 27 February 1982, to coincide with the appearance of
L'Atelier de l'histoire. The documentation for this essay has been drawn primarily from
Joseph Heller, The Zionist Idea (New York: Schocken Books, 1949), and Arthur Hertz-
berg, The Zionist Idea: A Historical Analysis and Reader (New York: Doubleday, 1959).

234
Israel, Zionism, and the Diaspora 235

life to be abandoned. The first vision was that of Herzl, for Zionism was
born in western Europe, specifically in Austria, where the Jewish com-
munity had made brilliant progress in the nineteenth century but where
anti-Semitism, too, had flared up violently. Lueger, the leader of the
Christian Social party—the man whose anti-Semitic diatribes fed the
young Hitler—was elected mayor of Vienna in 1897. The fact is that
Jewish assimilation had apparently provoked a rejective reaction in the
body social. Through Theodore Herzl, the Zionist dream emerged in this
society at the crossroads of democratic progress and resistance to that
progress. Like many memorable ideas or great events, it was predicated
on a revolution made of expectations, a revolution thwarted by the
counterrevolutionary force of reality.
Thus, the starting point of Zionism was both the success and the failure
of assimilation. The success, since Herzl belonged to a completely secu-
larized milieu that was cut off from the Jewish tradition. He was the
perfect representative of the intelligentsia integrated into German cul-
ture since the Enlightenment—a milieu whose fundamental belief was
that democracy was the necessary and sufficient condition for emancipa-
tion. But, at the same time, Herzl witnessed the failure of this central
creed, since the development of a new anti-Semitism within democratic
culture, in Austria, Germany, and France, proved to him that his youth-
ful convictions were mistaken. Thus, “‘assimilated’’ Jews were the first to
understand that democracy did not solve the Jewish problem, as all of
enlightened Europe had believed since the French Revolution. And it
was a “modern” anti-Semitism, born of democracy, that triggered Zion-
ism. By founding a national state and a homogeneous Jewish society,
Zionism aimed at enabling modern Judaism to lead a collective existence
modeled on European nations and yet in opposition to them, or at least to
the ideology of individual Jewish assimilation that they professed.
It is probable that Austro-German Jewry was the only branch of
European Jewry whose intellectual sensibility at the time made it a
suitable breeding ground for the Zionist idea: it was sufficiently inte-
grated into modern culture (unlike the traditionalist eastern European
ghettos) to adopt the idea of nationality and yet close enough to its origins
(unlike the French-style embourgeoisé varieties of Jewry) to preserve a
connection, if not with a religious belief, at least with the idea of a
separate culture. But, as evidenced by the episode of Herzl’s temporary
endorsement of the West African compromise as a substitute for
Jerusalem,’ that connection is easier to define in negative terms, as a
consequence of anti-Semitic rejection, than in positive terms, as a taking
into account of the cultural heritage.
Unlike that of Herzl, the second vision of the European Jew that
contributed a contrario to the birth of Zionism was not informed by the
need to put an end to anti-Semitism or by the impossibility of assimila-
tion. It was centered not on the external aspect of the Jewish problem,
236 Aspects of Modern Jewish History

but on its internal one—on the cultural, and consequently national,


identity of the Jews. In this respect, it was more radically opposed to the
Diaspora than the first vision, since it did not reject the exile, the galuth,
on empirical grounds relating to anti-Semitism and the impossibility of
assimilation: it rejected it for reasons having to do with the very essence
of Judaism. This vision did not originate in a factual observation of the
failure of assimilation. On the contrary, it regarded assimilation as a
natural and inescapable consequence of the existence of Jews as indi-
viduals in the democratic states of Europe; hence, assimilation would
eventually put an end to anti-Semitism by depriving it of sustenance. The
only way to avoid the democratic elimination of the Jewish problem (that
is, the elimination of the Jews themselves), and thus the only way to keep
Judaism alive, was to create the conditions in which Jews could exist as a
nation.
This argument, the most systematic expression of which can be found
in the articles published by Jacob Klatzkin on the eve of the First World
War,’ treated the modern anti-Semitism of European states as in some
sense a rational phenomenon. Such anti-Semitism was the obverse side of
nation building, the will to reject whatever was alien to national culture,
and consequently the decisive proof that Judaism continued to exist as a
potentially national tradition, since it was perceived as such by the
European nations. The enemy, for Klatzkin, was not the anti-Semite; it
was the assimilating liberal. The Jewish national identity was to be
salvaged in the name of the same values that led the anti-Semite to an
exclusive worship of his own nation—that is, because the Jews formed a
distinctive people with an identity of its own.
Thus, of the two visions of the Diaspora Jew the negation of which led
to Zionism, one originated in the anti-Semitic rejection that followed
assimilation—a rejection regarded as insurmountable; the other, on the
contrary, stemmed from the propositions that a Jewish nationhood ex-
isted in a positive sense—on a par with European nationalities that
emerged in the modern age, particularly in the nineteenth century—and
that this nationhood had to be saved from the perils of assimilation.
These analyses of the Jewish problem at the turn of the twentieth century
were, therefore, contradictory. But both of them, in their opposite ways,
sought to resolve the central paradox of Zionism: how to abandon
Europe and its culture while at the same time holding on to them. Herzl’s
brand of Zionism regarded contemporary anti-Semitism as a pathological
aspect of modernity, a nationalist deviation of the national idea, an
unexpected but indestructible product of democratic ideology. On the
contrary, Klatzkin’s brand of Zionism regarded it as the natural offshoot
of European nationhood—as consecutive to, and not incompatible with,
the existence of modern democratic states. The two approaches were
probably rooted in different perceptions of the Jewish experience at the
time: the first was totally emancipated from tradition and patterned
Israel, Zionism, and the Diaspora 237

negatively on the modern nationhood of European states; the second was


closer to tradition and defined in positive terms by a cultural specificity
that had to be saved because it was threatened. But these two experi-
ences, these two analyses, ultimately led to the same conclusion: the need
for a Jewish state. For they shared at least the same opinion about the
“abnormality” of Jewish life in the Diaspora and the need to put an end to
this predicament.
Naturally, the picture is not as simple as this point-by-point opposition
between the two types of Zionist ideology would suggest. For example,
one could discern in the “‘spiritual’’ Zionism of Ahad Haam the desire to
find a definition based neither on external factors—on the existence of
anti-Semitism—nor on a purely national view, but rather on the Jewish
cultural heritage, in the widest sense of the word. To this extent, although
he shared with Klatzkin the idea that the situation of the Diaspora was no
longer conducive to the advancement of Jewish culture, Ahad Haam had
an infinitely more positive opinion of Diaspora Judaism in general. He
felt it essential for the Jewish renaissance in Palestine to maintain a
constant dialogue with what would consequently have become a
peripheral form of Judaism but would have remained an indispensable
element of Jewish national history.
Yet one of the factors that gradually brought cohesion to the Jewish
population in Palestine, even more forcefully than the condemnation or
analysis of modern anti-Semitism, was precisely the radical rejection of
the situation of the Diaspora Jew. Anti-Semitism was in everyone’s
memories, but the sort of intellectual quasi acceptance of it in advance
eliminated all justification for the survival of the Diaspora from the
moment Judaism existed in Palestine. One need only consider the ideol-
ogy and practice of the Zionist movement in the early twentieth century,
at the time when the outlines of Israel were emerging, to understand the
extent to which the Zionist ““Western” saga, in a blend of Russian
populism and Marxist social democracy, manufactured images of the
Jewish future that were tantamount to final condemnations of the Jewish
past. The return to Zion represents a new version of Judaism, based on
European ideas and on a rejection of the ghettos and the Diaspora.
From the outset, the success of the Zionist project was due to the
extraordinary activity and determination of a few individuals: Weizmann
was a new Cavour, and Jabotinsky a reincarnation of both Mazzini and
Garibaldi. The first, a visionary but adroit diplomat, abided by the official
plan for a national home and kept a watchful eye on the international
scene; the second, a flamboyant, romantic man in a hurry, sought a
historical shortcut to the state of Israel. Between the two there existed the
same secret connection as between the Haganah, the Jewish settlers’
self-defense army, and the Irgun, which preached and practiced terrorist
tactics—a violent conflict but a fundamental agreement on what all
parties regarded as the Jewish revolution.
238 Aspects of Modern Jewish History

The same situation prevailed somewhat earlier in the “‘working-class”


Zionism that, in combination with the Jewish tradition, constituted the
most important—and, incidentally, the most touching—cultural ingre-
dient of the future state. The great pre-1914 aliyah (several tens of
thousands of Lithuanians, Poles, Ukrainians, and Russians) was an im-
migration of qualified labor. It brought together—as the American
adventure in the seventeenth century had done—a body of dynamic and
faithful individuals. These latter-day pioneers were militants who had
often had to overcome the hostility of their rabbis—staunch believers in
the age-old curse—and of their comrades, locked into the internationalist
dogma of the labor movement. They brought to Palestine the peasant
myth of Russian culture and the interminable discussions of the Euro-
pean working-class Left—Tolstoy and Marx united in the fraternal
atmosphere of absolute new beginnings. On this new “‘frontier’”’ of
Europe, they planted the flag of a new agrarian and military Rome. On
this infertile soil, these impoverished settlers invented an unpolished and
egalitarian civilization. They had left the urban ghetto; they founded the
rural antighetto. The Jew had been landless; now he was to be a farmer.
He had been a usurer; now he would scorn money. He had been dis-
armed; now he would be a soldier. He had been humiliated; now he
would be a victor. One could go on indefinitely contrasting these two
images point by point. By defining itself as both a radical break with the
past and a return to origins, Zionism unites the two cardinal but divergent
notions contained in the modern idea of revolution in seventeenth-
century England and late eighteenth-century America and France. In so
doing, it categorically condemns—in a manner not unsimilar to that of
the anti-Semites—the existence of Jews in exile, cut off both from their
origins and from their future.
The Nazi extermination engraved these features of the Zionist revolu-
tion in tragedy. It gave them the force of an earthquake. The reasons why
the extermination has constituted, since the inception of the state, one of
the foundations of Israel’s national consciousness go far beyond the
material fact that it drove toward what was not yet the Jewish state the
survivors of the genocide. In the Jewish historical consciousness, but also
in gentile public opinion and in the political world at state level, the vision
of Jewish destiny offered by Zionism—for so long a marginal vision—has
assumed since Hitler the status of a self-evident truth. Inseparable from
its three preconditions—the German will to annihilate, the consent or
passivity of gentile Europe, and the weakness of Jewish resistance—the
genocide has become the Diaspora’s truth, while the state of Israel and
the law of return have become the only guarantees against its reoccur-
rence.
Thus, Hitler’s genocide lies at the origin of Israeli society for factual
reasons, since so many Israelis are its survivors, but even more so because
of the basic truth that the state of Israel was born of the conjunction
Israel, Zionism, and the Diaspora 239

between the irrepressible violence of anti-Semitism and Zionism’s pre-


ventive response to it. The massive aliyah of European Jews beginning in
the 1920s and 1930s only served to drive home the original message of the
founders about the impossibility of being European and Jewish and the
need to leave Europe in search of a Jewish nation in order to live as
Europeans. In the light of the genocide, acceptance of the Diaspora
became tantamount to a warrant for crime; condemnation of the Di-
aspora, on the contrary, turned out to be justified by the stark tragedy of
history.
Everything, therefore, has conspired to turn the Israelis into a people
set against its past sufferings. The contemporary experience has cut
Jewish history into two parts, the interminable exile representing its
negative half, the Zionist revolution representing its renaissance. But the
words exile and renaissance imply a political world wholly informed by
the search for a “‘past”’ that is supposed to give them a meaning. We are
confronted here with the exceptional character—by the standards of
European history—of the constitution of the Israeli nation, despite the
fact that it was built with two classic elements of the European repertoire:
the idea of revolution and the idea of nationhood. The reason for this
peculiarity is that the revolution of the Jewish people, carried out in the
name of a secular ideal, with the aim—similar to that of the French
Revolution—of rebuilding society on the basis of principles originating in
society itself, was at the same time a means of reuniting the Jews with an
archaic history and a religious promise. Jewish nationhood, reinvented
not on the basis of a collective presence but on that of a scattered people,
found its symbolic cohesion and the shared treasure-trove of its memories
only in the oldest of old books, the Hebrew Bible. Thus, the two great
mooring points for the pedagogy of nationhood and for the national
consensus are the two extremes of Jewish history, the most ancient and
the most recent: the Bible and Zionism. What lies in between—that is,
nearly all of the Jewish past—is the object of a collective semicensorship,
inherent in the very nature of Israeli national feeling. To put it differ-
ently, the central figure of Israel is Ben Gurion, not Ahad Haam.
However, it seems to me that these characteristic traits of Israeli
national consciousness, which can be explained both by the history of
Zionism and by that of twentieth-century Europe, have contributed to
the present predicament of Israeli society—at least to the share of its
predicament for which it is alone responsible. The European obsession in
reverse that constitutes the basis of the Zionist movement has long made
it blind to the Arab problem. But, at the other end of the Jewish
revolution, the biblical point of reference has not made the problem
easier to solve, ever since the “problem” of the 1920s took on the
inevitable dimensions of a question of life or death. The idea of election
and of a promised land, clashing with that of Palestine as a holy land of
Islam, has again turned this patch of barren land, whose inhabitants once
240 Aspects of Modern Jewish History

eked out an existence in the shadow of the Turkish pashas, into the stake
of a battle between two messianic movements. More generally, as far as
Israel is concerned, the emphasis on biblical origins turned religion into a
cofounder of the state, inseparable from public authority. Religion has
made Israeli society a hostage to a conservative rabbinate, despite the
deliberately modernist aims of that society’s founders.
But it is probably the relationship between Israel and the Diaspora
that provides the best insight into the Israelis’ curious relation to the
Jewish past. It is true that the converse analysis is no less interesting and
that the attitude of Diaspora Jews toward Israel is composed of varying
degrees of love and hatred, from which a guilty conscience is never
absent. The very existence of Israel has transformed the general condi-
tion of Diaspora Jews owing to the way they are perceived by Gentiles;
yet these Diaspora Jews have remained in the Diaspora. At the other
end, Israel’s vision of the Diaspora is most often centered either on the
early-twentieth-century eastern European ghettos or on the Nazi
tragedy—that is, on the negation of the new Jewish farmer and soldier.
Occasionally, one can even discern a certain anti-intellectualism in this
spontaneous reconstruction of the past as a foil. Furthermore, as in the
early days of Zionism, the unfortunate and endangered Diaspora is only
an instrument or a source of recruitment: its fate is to become assimilated
or to enlist. While awaiting the inevitable outcome, Israel contributes to
the process by propagating a vision of the Diaspora’s natural submission
to Israel as its potential state. Having normalized the Jewish destiny
within its borders, Israel expects the same to happen to Judaism as a
whole. Because it has altered the image of Judaism the world over, Israel
assumes that it has unified Judaism from every point of view and thus no
longer feels the need to be aware of its diversity. Hence, geographical
space confronts Israel with the same enigmas and the same impasses as
historical time, and for the same reasons: the Jewish nation, whatever the
intentions or beliefs of Zionism, is not ‘‘like the others.”
Indeed, a modification of this vision could well be the first step toward
the general reappraisal of Zionism that everyone in Israel regards as a
necessity. To begin with, a very simple observation: aside from its ex-
traordinary historical achievement, Zionism has not realized its chief
objective, which was to bring all of the Jewish people together in Pales-
tine. Immigration has virtually ceased, and Israel’s population growth is
now solely dependent on its domestic birthrate. The annual migratory
balance is actually negative: this is a sign of the relatively high emigration
rate, but the phenomenon is hard to evaluate since the Israelis who leave
their country present their departure as a temporary absence. Emigration
apparently involves both the poorest social strata and the professions
(doctors, lawyers, and engineers): in both cases, the American dream
remains the natural pole of attraction. This phenomenon, which is quite
“normal”—and, moreover, easy to understand in a country where wages
Israel, Zionism, and the Diaspora 241

are relatively low, taxes very high, and military service interminable—has
become, no less understandably, the object of anxious concern in a
country founded on the law of return. Apart from its statistical impact,
emigration is a vivid sign of the psychological limits of the Zionist idea.
In fact, these limits have never ceased to exert a constraining effect,
since, from the very outset, at the turn of the century, emigration to
America was infinitely greater than the aliyah to Jerusalem; but these
constraints had been masked both by the spectacular triumph of the
creation of the state of Israel and by the massive immigration tragically
engendered by Hitler’s genocide. Today, more than thirty years after the
birth of Israel, no Zionist leader can seriously base a policy on the
assumption that Israel will absorb the entire Diaspora. The Diaspora
Jews, if they were supposed to have come, would already be there.
Russian Jews will perhaps continue to emigrate to Israel, since the Israeli
visa is the only one that allows them to leave the Soviet Union. However,
these immigrants will be few and far between, and there is no guarantee,
as experience has shown, that they will settle easily or definitively in
Israel. Barring an unpredictable catastrophe in the United States or
western Europe, Zionism does not solve the problem of the Diaspora. It
is condemned to coexist with the Diaspora.
The simplest way of rethinking Zionism, on the basis of what it has
accomplished and what it has become, would probably be to take into
account the abandonment of a universal solution to the Jewish problem—
an abandonment that is evident in the facts. Like all great historical
movements, Zionism has not fulfilled all the predictions of its founders or
achieved all the desired results. However, if one compares the degree to
which Zionism has carried out its initial project with the degree to which
other great ideas that have mobilized mankind in the twentieth century
have carried out theirs, Zionism does not come out too badly: all the
more reason to limit its scope and objectives according to the historical
tasks imposed on it by circumstances, while preserving its fidelity to the
spirit of its founders. Paradoxically, this redefinition implies both the
secularization of the state and the restitution to the Jewish people of its
entire past. It is by emancipating the national feeling from the historical
shortcut between the Bible and the founding of Israel that one will spare
it from the temptation of a debased form of messianism. In this respect,
Israel’s relation with the Diaspora constitutes a test.
AA hotel
Veo marca Ge Cee é
ar aR wi ALT Le
Mr malt, Préfet aS
"PEN \ar PRESS L
| Dre DU 2e
Li Le ALES PR
é… MATE Ayo des
ve CRD rien *des
23: | bret eae
mia te et ar ù
IV | oi
arr oes pee F4
. a ine
NOTES:

Introduction

1. To illustrate my argument, I have relied particularly on Richard Cobb’s “Modern


French History in Britain,” the Raleigh Lecture on History for 1974 (London: Proceedings
of the British Academy, 60 [published as a separate pamphlet], 1974).

Chapter 1

1. Raymond Aron, Trois essais sur l’âge industriel (Paris: Plon, 1966).
2. See Raymond Aron’s article on Sartre’s Critique de la raison dialectique in Le Figaro
littéraire (19 October 1964).
3. In Esprit (November 1963).
4. See Edmund Leach, “‘Genesis as Myth,” Discovery (May 1962); idem, ‘The Legit-
imacy of Solomon,” Archives européennes de sociologie (1966); idem, ‘Claude Lévi-
Strauss, Anthropologist and Philosopher,” New Left Review (1966).

Chapter 2

1. J. Marczewski, general ed., Histoire quantitative de l'économie française, 10 vols. to


date (Paris: ISEA, 1961-68), esp. vol. 1, Histoire quantitative: buts et méthodes, by J.
Marczewski.
2. The expression is Pierre Vilar’s.
3. Pierre Chaunu has defended and used the term in a number of studies. See esp.
Histoire quantitative ou histoire sérielle, Cahiers Vilfredo Pareto (Geneva, 1968), and
“THistoire sérielle: bilans et perspectives,” an article published simultaneously in Revue
historique (April-June 1970) and Revue roumaine @histoire, no. 3 (1970).
4. See esp. P. Deane and W. A. Cole, British Economic Growth, 1688-1959: Trends
and Structure (Cambridge: Cambridge University Press, 1962); David Landes, The Un-
bound Prometheus (Cambridge: Cambridge University Press, 1969); François Crouzet,
“Angleterre et France au XVIIIe siècle: essai d’analyse comparée de deux croissances
économiques,” Annales: économies, sociétés, civilisations (March-April 1966).

243
244 Notes to pages 44-84

5. Marcel Couturier, “Vers une nouvelle méthodologie mécanographique: la prépara-


tion des données,”’ Annales:économies, sociétés, civilisations (July-August 1966).
6. E. Le Roy Ladurie, “Révolution française et contraception: dossiers languedo-
ciens,” Annales de démographie historique (1966), and “Révolution française et funestes
secrets,” Annales historiques de la Révolution française (October-December 1965). See
also A. Chamoux and C. Dauphin, ‘‘La Contraception avant la Révolution frangaise:
l'exemple de Châtillon-sur-Saône,” Annales: économies, sociétés, civilisations (May-June
1969).
7. The bibliography is too vast even to be summarized.
8. E. Le Roy Ladurie, Les Paysans de Languedoc (Paris: SEVPEN, 1966; abridged ed.,
Flammarion, 1969). The account I give here is a shorter version of my article “Sur quelques
problèmes posés par le développment de l’histoire quantitative,” Information sur les
sciences sociales 7 (1968): 71-82.
9. Crouzet, ‘Angleterre et France.”
10. Pierre Vilar, La Catalogne dans I’Espagne moderne (Paris: SEVPEN, 1962), esp.

11. P. Goubert, Beauvais et le Beauvaisis de 1600 a 1730 (Paris: SEVPEN, 1960).


12. R. Baehrel, Une croissance: la Basse-Provence rurale, fin du XVIe seicle-1789
(Paris: SEVPEN, 1961).
13. Denis Richet, “Croissance et blocages en France du XVe au XVIIIe siècle,”
Annales: économies, sociétés, civilisations (July-August 1968).
14. Michel Fleury, ‘“‘Les Progrès de l’instruction élémentaire de Louis XIV à Napoléon
III,”’ Population (1957). See also Lawrence Stone, “‘Literacy and Education in England,
1640-1900,” Past and Present (February 1968): 69-139; Carlo Cipolla, Literacy and De-
velopment in the West (Harmondsworth and Baltimore: Penguin Books, 1969).
15. Gaby and Michel Vovelle have given a brilliant demonstration of how iconographi-
cal series can be used to study religious sentiment. See their Vision de la mort et de l'au-delà
en Provence, Cahiers des Annales (Paris, 1970).

Chapter 3

1. Le., the essay “Quantitative History,” reprinted above as chapter 2.


2. Norbert Elias, Uber den Prozess der Zivilisation, vol. 1 (Basle, 1939).

Chapter 4

1. See esp. Donald R. Kelley, Foundations of Modern Historical Scholarship (New


York: Columbia University Press, 1970); George Huppert, The Idea of Perfect History
(Urbana: University of Illinois Press, 1970).

Chapter 5

1. The best discussion of this question is to be found in A. Momigliano, ‘Ancient


History and the Antiquarians,” Journal of the Warburg and Courtauld Institutes 13 (1950):
285-315.
2. B. Neveu, Un historian à l’école de Port-Royal: Sébastien Le Nain de Tillemont,
1637-1698 (The Hague, 1966): 182-85.
3. These examples are taken from an unfortunately unpublished essay by Louis Tré-
nard, ““L’Enseignement de l’histoire en France de 1770 à 1885,” delivered as a lecture in
June 1968 under the aegis of the Fédération Belge des Professeurs d’Histoire. [French
classes are numbered backward with respect to the American and British systems: onziéme
Notes to pages 84-103 245

is the first grade of elementary school, sixiéme the first grade of secondary school, premiére
(formerly classe de rhétorique) and classe de philosophie (or terminale) the last two years of
secondary school. Before the Revolution, collèges was the general term for secondary
schools. After the Revolution, the secondary schools were divided into /ycées and collèges,
the former ranking somewhat above the latter.—TRans. ]
4. This information on the Collége de Juilly is taken from a recent and as yet unpub-
lished dissertation by Etienne Broglin, “De l’Académie royale à l’institution: le Collège de
Juilly, 1745-1828” (University of Paris, 1978).
5. The major speeches on education policy by the members of the revolutionary
assemblies can be found in C. Hippeau, L’Instruction publique en France pendant la
Révolution (1881). For more details, see J. Guillaume, ed., Procés-verbaux du Comité
d'Instruction publique de la Convention nationale, 6 vols. (Paris: Imprimerie Nationale,
1890-1907).

Chapter 6

1. The example has been given in this field by the work of L. Febvre and H.-J. Martin,
L’Apparition du livre, L’Evolution de l’humanité, 49 (Paris: Albin Michel, 1958).
2. R. Estivals, “La Statistique bibliographique de la France sous la monarchie au
XVIIIe siècle” (unpublished dissertation, University of Paris, 1962); idem, Le Dépôt légal
sous l’Ancien Régime de 1537 à 1791 (Paris: M. Rivière, 1961).
3. The verso of the last leaf of Bibliothèque Nationale Ms. Fr. 22 001 lists the following
fees: privilège général, 101 livres and 2 sous; permission for six years, 61 livres and 18 sous;
permission for three years, 30 livres.
4. Diderot, Lettre sur le commerce de la librairie, reprinted as Lettre sur la liberté de la
presse, ed. J. Proust (Paris: Editions Sociales, 1963).
5. Malesherbes, Mémoire sur la liberté de la presse (1788).
6. Diderot, Lettre.
7. The registers for printing license requests used as the basic source for this study are
the following:
Priviléges et permissions du Sceau
Ms. Fr. 21 995 1723-28
Ms. Fr. 21 996 1728-38
Ms. Fr. 21 997 1738-50
Ms. Fr. 21 998 1750-60
Ms. Fr. 21 999 1760-63
Ms. Fr. 22 000 1763-68
Ms. Fr. 22 001 1768-74
Ms. Fr. 22 002 1774-84
Ms. Fr. 21 978 1784-89
Permissions tacites
Ms. Fr. 21 990 1718-46
Ms. Fr. 21 994 1750-60
Ms. Fr. 21 992 1760-63
Ms. Fr. 21 991 1763-66
Ms: Er. 21,993 1766-72
Ms. Fr. 21 983 1772-82 (first occurrence of the
title permissions tacites)
Ms. Fr. 21 986 1782-88
Ms. Fr. 22 003 1788-89 (This last register of requests for tacit permits has been
mistakenly listed in the Bibliothèque Nationale manu-
script catalog among the privilège registers.)
246 Notes to pages 103-112

8. The itemized count of several books included in the same request for a privilége
generally presents no problems. Prints and engravings, often submitted in very great
numbers in a single request, have not been included in our census. However, bishoprics
often applied for a single privilège for a whole series of liturgical or devotional handbooks,
of which a summary list would be provided: breviaries, diurnals, missals, antiphonaries,
directories, and so on. In this case alone, it has been impossible to distinguish between
licenses and books. We have therefore adopted the method of counting arbitrarily as one
book the set of diocesan handbooks (usages), which artificially reduces their number.
9. Malesherbes did write in his fifth Mémoire sur les problèmes de la librairie that tacit
permits ‘‘have multiplied to the point of being today as common as public licenses.”
10. Malesherbes makes this statement in his Mémoire sur la liberté de la presse of 1788.
Moreover, a letter from the Parisian libraires (publisher-booksellers) Huart and Moreau to
Montesquieu, dated 8 January 1749, cites the rumor that L'Esprit des lois had been granted
a tacit permit. This fact cannot be verified on our registers, which are missing from late 1746
to late 1750.
11. After 1778, the “books entered via the Chambre Syndicale” no longer appear on the
registers of requests, but only in the listings of tacit permits.
12. Madeleine Ventre, L’Imprimerie et la librairie en Languedoc au dernier siècle de
l’Ancien Régime, 1700-1789 (Paris and The Hague: Mouton, 1958).
13. Letter from Huart to Montesquieu, March 1749.
14. Most of which are to be found in the manuscript division of the Bibliothèque
Nationale and in the Bibliothèque de l’Arsenal.
15. Forbidden books escape our census, but the book-licensing records invariably
preserve the traces of the most significant among them: Emile and Le Contrat social, for
example, spawned a long series of commentaries, which testify to their social impact.
16. This classification has been prepared with the help of Daniel Roche.
17. The division into five major categories is attested by the practices followed in library
catalogs. See for ex. Louis Desgraves, ed., Catalogue de la bibliothèque de Montesquieu
(Geneva: Droz, 1954); “Catalogue des bibliothèques des parlementaires parisiens,” in
François Bluche, Les Magistrats du Parlement de Paris au XVIIIe siècle, 1715-1771 (Paris:
Les Belles Lettres, 1960), 291; Durcy de Noinville, Dissertation sur les bibliothèques (Paris,
1758), Bibl. Nat. call no. Q. 3507; J.-M. Cels-Martin, Coup d'œil éclairé d’une bibliothèque
(Paris, 1773), Bibl. Nat. call no. Q. 5346.
18. See library catalogs such as Catalogue des livres de feu Monsieur le Maréchal de
Lautrec (Paris, 1762), Bibl. Nat. call no. Q. 8138; see also references in note 17 and esp.
Cels-Martin, Coup d'œil. We have classified under this heading the Histories of the Old and
New Testaments, the Lives of Jesus Christ, and similar works, for they all deal with biblical
or sacred history as distinct from ecclesiastical history.
19. See references in note 18.
20. The division of theological works into Catholic and non-Catholic (orthodox and
heterodox) was practiced in public libraries. See the “Catalogue général de la Bibliothèque
du Roi” (1739-42), reprinted in /ntroduction au catalogue général auteurs of the Biblio-
thèque Nationale (1897). This section naturally includes sermonizing, polemical, catechis-
tic, ascetic, and mystical theologians. See Cels-Martin, Coup d'œil.
21. The “liturgy” category appears in the bibliographical guides mentioned above,
where it includes devotional and catechistic literature.
22 See noted,
23. See Cels-Martin, Coup d'œil; Catalogue Montesquieu; Catalogue de la bibliothèque
Malesherbes (1797), Bibl. Nat. call no. 8° INV. Q. 9128; Catalogue Lautrec.
24. See M. Camus, Lettre sur la profession d'avocat, avec un catalogue raisonné des
livres de droit qu’il est le plus utile d’acquérir et de connaitre; Catalogue Malesherbes, vol. 2;
Cels-Martin, Coup d'œil.
25. Biographies and memoirs of famous men, literary history, history of theater,
heraldry, archaeology, and so forth, have been classified under “ancillary sciences [of
history]’’ (III.B.3). See Cels-Martin, Coup d'œil; Catalogue Malesherbes.
Notes to pages 112-121 247

26. See Catalogue de la Bibliothèque du Roi (1688) s.v. histoire ecclésiastique; Cels-
Martin, Coup d'œil, s.v. histoire sacrée; Catalogue Lautrec. In all of these works and
catalogs, geographical studies are grouped together with historical works.
27. See note 17: certain authors equate this category with philosophy, by virtue of the
tradition that philosophia comprehendit artes et scientias. For ex., Formey, Conseils pour
former une bibliothèque peu nombreuse mais choisie. However, in the catalog of books
belonging to M. Augry, a lawyer in Vendôme (Bibl. Nat. call no. Q. 3026), the entire
category is included under the heading “mathematics.” Pedagogical works are listed here
under “‘ethics.’” Theodicy, ontology, anthropology, magic, and cabalistic literature have
been classified under “‘metaphysics.”’ The science of navigation is classified here, in accord-
ance with Cels-Martin, under astronomy; alchemy under “‘chemistry.”’
28. Cels-Martin, Coup d'œil; Catalogue Lautrec; Catalogue Malesherbes.
29. The distinction between physical, mathematical, and natural sciences can be found
in Cels-Martin, Coup d'œil; Catalogue Lautrec; Catalogue Malesherbes. See also the Encyc-
lopédie, vol. 1, “Système figuré des connaissances humaines.”
30. Includes works on government, politics, commerce, and finance. See Catalogue
Malesherbes; Encyclopédie méthodique (1784), s.v. art; on political economics, see Cels-
Martin, Coup d'œil.
31. Catalogue de la Bibliothèque du Roi (1688) Catalogue Malesherbes; Cels-Martin,
Coup d'œil, Formey, Conseils; Maïeul de Chaudon, Bibliothèque d’un homme de goût
(Paris, 1772-77), Bibl. Nat. call no. 5340-41 and Q. 5542-45; Almanach de la librairie
(1781).
32. Music, painting, sculpture, drawing, etching, dance, architecture, art of warfare,
epistolary art, decoration; see Cels-Martin, Coup d'œil.
33. Cels-Martin, Coup d'œil; Encyclopédie, vol. 1, “Système figuré”; Catalogue
Malesherbes. Wood, silk, gems, iron, copper, clocks, wool, pyrotechnics, fire, and so on.
34. See note 17. Sports and pastimes.
35. Belles-lettres: see note 17. Essays, criticism, rhetoric, and so on, have been clas-
sified under “grammar and philology” (V.B), in accordance with Cels-Martin.
36. See note 17.
37. Formey, Conseils; Maieul de Chaudon, Bibliothéque.
38. Formey, Conseils; Maïeul de Chaudon, Bibliothèque; Cels-Martin, Coup d’eil;
Catalogue de la Bibliothèque du Roi (1688); Catalogue Malesherbes; Catalogue Lautrec.
39. The assistance of my friends and colleagues Jean-Louis Flandrin and Daniel Roche
has been invaluable in helping me to identify a great number of these books.
40. As explained in note 8, collections of diocesan handbooks (usages) have been
counted in this category as single units—an arbitrary but unavoidable choice.
41. This phenomenon is the chief explanation for the relative decrease in the number of
“poetry” books between 1784 and 1788.
42. Daniel Mornet, Les Origines intellectuelles de la Révolution française (Paris, 1932).
43. This share is all the larger as the three samplings, for the sake of consistency and
comparability, have not taken into account the “books entered via the Chambre Syndi-
cale”’—a category mentioned only between 1767 and 1778. These already printed books,
sent in from the provinces and abroad, cover a wider range of genres than the manuscripts
deliberately submitted for tacit permits.
44. In accordance with the criteria of the time, we have listed under ‘‘medicine,”’ in the
tacit-permits category, the abundant polemical literature of the 1780s concerning Mesmer’s
theory of animal magnetism. More generally, the bibliographical grid scatters among
different headings (metaphysics, chemistry, medicine) a basically homogeneous occultist
philosophy whose importance in the late-eighteenth century is well known.
45. We have excluded from these percentages the short popular novels published by the
Bibliothèque Bleue and other specialized libraires.
46. Jean-Jacques Rousseau, La Nouvelle Héloise, ed. Daniel Mornet (Paris, 1923),
vol. 1.
47. The Bibliothéque universelle de romans is a good instrument for identification.
248 Notes to pages 122-130

48. The manuscript registers of book licenses make it possible to reexamine certain
problems in literary history, such as the one raised by Georges May—in a remarkable work
(Le dilemme du roman au XVIIIe siècle: étude sur les rapports du roman et de la critique,
1715-1761 [Paris, 1963])—concerning the ban on novels imposed by Chancelier d’Agues-
seau in 1738. Citing the bibliography compiled by S. Paul Jones for the first half of the
eighteenth century (A List of French Prose Fiction, New York, 1939), May rests his
argument on the paucity of new novels published in Paris in 1738. According to May, only
six novels appeared, of which one, in fact (Essai sur la nécessité et les moyens de plaire, by
Paradis de Moncrif), is essentially a moral treatise followed, it is true, by a tale. But an
examination of the registers of printing-license requests shows that between 20 February
1737—the date given by May as that of the enactment of the ban—and the end of 1738, the
number of new, authorized novels was considerably greater. Admittedly, during the same
period, the censors turned down some twenty novels, including both new titles and reprints.

Chapter 7

1. Comte Henri de Boulainvilliers, Histoire de l'Ancien Gouvernement de la France


(The Hague and Amsterdam: Aux dépenses de la compagnie [des libraires], 1727).
2. Gabriel Bonnot de Mably, De la manière d'écrire l'histoire (Paris: A. Jombert Jeune,
1783).
3. Boulainvilliers, Histoire.
4. Mably, De l’étude de l'histoire, à Monseigneur le Prince de Parme (Maéstricht:
Cavelier, 1778). Exhorting his pupil to choose a model, Mably adds, “I warn you lest it
should be a prince. The lives of the greatest kings are always tainted by some sort of
counterfeit and overweening ambition. Choose as your model a simple citizen of Greece or
Rome.”
5. Ibid.:.“‘Observe attentively how the same laws, the same mores, the same virtues,
and the same vices have always produced the same effects. Thus the fate of states is
governed by fixed, immutable, and manifest principles. Discover these principles, Monsei-
gneur, and politics will hold no more secrets for you. . . .”
6. Boulainvilliers, Histoire.
7. Mably, De la Législation ou Principes des Lois, in Oeuvres complètes (London,
1789). For Boulainvilliers, history provided more examples to avoid and abhor than models
to imitate: see his “Lettre à Mademoiselle Cousinot sur l'Histoire et sa méthode,”
Angouléme library, Ms. 23.
8. Onthis theme, Mably often seems to copy Boulainvilliers: ‘This can convince us with
increasing force of the indisputable fact that, of all the nations of the world, ours is
distinguished by its thoughtlessness and inattentiveness; as a consequence, from one
century to another, the French have always forgotten what their forefathers did.” And
Mably: “France is the most heedless nation, and the easiest to mislead, because it is the least
attentive to an examination of the past.”
9. The best account of these theories is provided by Elie Carcassonne, Montesquieu et le
problème de la Constitution française au XVIIIe siècle (Paris: Presses Universitaires de
France, 1927).
10. Mably, Observations sur les Grecs (Geneva: Par la Compagnie des Libraires, 1749):
“Today we know what to make of this greensward, these wreaths of flowers, these concerts,
this sweet leisure . . . etc.”
11. Boulainvilliers, “Mémoire sur la noblesse de France,” Angoulême library, Ms. 23.
Boulainvilliers has a word of commiseration for the free men whom the conquest suddenly
turned into slaves. But he immediately adds that, ‘as they defended themselves poorly, they
were rightly subjected to the law of the victor.”
12. Boulainvilliers, Histoire: “As if an undisputed possession of 700 years’ standing
could be thought of as such a mediocre claim! This weakness is all the more dangerous for
Notes to pages 130-146 249

having bred the false and ridiculous argument of those who say that Hugh Capet handed
over his land, fiefs, and real estate to his new subjects to reward them for having made him
king. From this argument a most abominable conclusion has been drawn, namely, that all
property belongs to the king, and that he is entitled to leave to his subjects whatever part of
it he pleases.”’
13. In the “Dissertation abrégée sur les premiers Français et leur origine,” which
concludes the Essais sur la noblesse de France (Amsterdam, 1732), Boulainvilliers goes so
far as to reassure those whom despotism has ennobled: ‘“‘As for the new nobles and the
anoblis, they have nothing to fear from our undertaking. Our views are summary, simple,
and innocent. They never descend explicitly to an examination of particular cases. Such
persons can therefore peacefully enjoy their metamorphosis.”
14. Mably, De la Législation.
15. Boulainvilliers, “Mémoire pour la construction d’un nobiliaire général,” Ecole
Supérieure de la Guerre (Paris), Ms. 25-26: “The old nobility has no means for guaran-
teeing its status; some of these nobles have lost their patents and charters in the course of
foreign and civil wars, and, with the exception of those families whose names remain in
history, they are left with only partial evidence, which is bound to perish through an infinity
of accidents.” Moreover, Boulainvilliers makes his purpose quite clear: ‘“‘the old nobility
[must be able to] distinguish itself from the new nobility, the nobility granted by individual
patent or consecutive to the ownership of certain offices. . . .”
16. Boulainvilliers, Histoire.
17. Boulainvilliers, Abrégé de l'Histoire de France (The Hague: Gesse et Néaulme,
1733): ‘‘Belles-lettres, after languishing for a long time, buried under the bad taste of earlier
centuries, reemerged in this century in all their brilliance... .”
18. Mably, Observations sur l'Histoire de France (Geneva: Par la Compagnie des
Libraires, 1765).
19. Mably, De la Législation.
20. Mably, De l’étude de |’Histoire: Mably cites the case of certain Swiss cantons as an
example of this ingenious arrangement.
21. Mably, Entretiens de Phocion (Amsterdam, 1763).
22. This is the solution Mably advocates in De la Législation: ‘The nobility must have a
patrimony that it cannot increase; on no account must the nobility be allowed to possess land
or inheritances intended for another order of citizens. . . .”

Chapter 8

1. Emile Benveniste, “Civilisation, contribution à l’histoire du mot,” in Hommage à


Lucien Febvre (Paris, 1954).
2. Politique des Romains dans la religion (1716).
3. Considérations sur les causes de la grandeur des Romains et de leur décadence (1734).
4. Voyage de Graz a La Haye, ed. de La Pléiade 1: 663,669.
5. Memoirs of My Life, ed. Georges A. Bonnard (London, 1966), 134.
6. The History of the Decline and Fall of the Roman Empire, chap. 3.
7. Esprit des lois, 18: 2.
8. History of Civil Society (1767), 124,185.
9. Recherches philosophiques sur les Américains (1768), 1: 218.
10. L’Esprit des usages et des coutumes. This discussion owes a great deal to the
attention that Edna Lemay has devoted to Démeunier’s book in her study “Naissance de
V’anthropologie sociale en France au XVIII‘ siècle: Jean Nicolas Démeunier et l'Esprit des
usages et des coutumes” (unpublished dissertation, Paris, 1972).
11. Histoire de l'Amérique, passim.
12. Histoire des Indes, cited in Michéle Duchet, Anthropologie et histoire au siécle des
Lumières (Paris, 1971).
250 Notes to pages 154-169

Chapter 9

1. 2 vols. (1768-69).
2. Ibid., 2:176.
3. J.-P. Brissot, Examen critique des voyages dans l'Amérique septentrionale de M. le
marquis de Chastellux (Paris, 1786).
4. Ibid., 105-6.
5. Mazzei, Recherches historiques et politiques sur les Etats-Unis de l'Amérique septen-
trionale ... , 4 vols. (Paris, 1788).
6. Crévecceur, Lettres d’un cultivateur américain (Paris, 1784). [Retranslated here from
the French edition: the passages quoted do not correspond precisely to the original English
text.—TRANS. ]
7. Ibid., ‘‘deuxième lettre.”
8. Ibid.
9. This study is in progress, thanks to the able survey of parliamentary speeches of the
French revolutionary period being conducted by Mme Edna Lemay, maitre-assistant at the
Ecole des Hautes Etudes en Sciences Sociales.
10. Tocqueville, L’Ancien Régime et la Révolution (Paris: Gallimard, 1953), vol. 2,
chap. 7: 131-32.
11. Reprint of Le Moniteur, 1:142.
12. Letter quoted in Gilbert Chinard, Volney et l'Amérique d’après des documents
inédits (Baltimore and Paris, 1923), 63-66.
13. Volney, Tableau du climat et du sol des Etats-Unis (Paris, 1883), preface. The
quotations that follow are also taken from the preface.

Chapter 10

1. The fundamental work on Tocqueville’s journey to the United States is still that of G.
W. Pierson, Tocqueville and Beaumont in America (New York: Oxford University Press,
1938).
2. Tocqueville and Beaumont’s plan for a fact-finding visit was set out in Note sur le
système pénitentiaire et sur la mission confiée par Monsieur le Ministre de l'Intérieur à MM.
Gustave de Beaumont et Alexis de Tocqueville (1831). Their findings were published in 1833.
A second, two-volume edition appeared in 1836, with a long introduction: Du système
pénitentiaire aux Etats-Unis et de son application en France, suivi d'un appendice sur les
colonies pénales et de notes statistiques. A third edition, expanded to include Tocqueville’s
legislative work on the penitentiary problem, was published in 1845.
3. Correspondance d’Alexis de Tocqueville et de Gustave de Beaumont (Œuvres com-
plètes, 8), 1:105-6.
4. René Rémond, Les Etats-Unis devant l'opinion française, 2 vols. (Paris: Armand
Colin, 1962), esp. vol. 2, chap. 7.
5. André Jardin, “L’ Amérique et les Américains vus par Tocqueville,’ H-Histoire, no.
4 (March 1980): 227—-40.
6. See esp. Tocqueville’s letter to Gustave de Beaumont of 5 October 1828 (Corres-
pondance Tocqueville-Beaumont, 1: 47-71).
7. Correspondance d’Alexis de Tocqueville et de Louis de Kergorlay (Œuvres complètes,
13), 1:373-75.
8. Actually, Saint-Beuve attributes this statement to “a very judicious and very respect-
able person,” adding a commentary of his own: ‘consequently, [Tocqueville’s] thoughts
were sometimes shallow.” See Sainte-Beuve, Causeries du lundi, 3rd ed., vol. 15.
9. H. de Tocqueville, Histoire philosophique du régne de Louis XV, 2 vols. (Paris,
1847).
Notes to pages 170-193 251

10. Ibid., 2:405.


11. Correspondance Tocqueville-Kergorlay, 1:214.
12. Tocqueville, Voyages en Sicile et aux Etats-Unis (Œuvres complètes, 5, vol. 1), letter
quoted p. 26.
13. Démocratie, 1:418 [GL 309].
14. Ibid. [GL 309].
15. Ibid., 2:316 [GL 639].
16. Ibid., 1:420 [GL 311].
17. Ibid., 1:423 [GL 313].
18. Ibid., 1:424 [GL 314-15].
19 GS 15]:
20. See esp. Seymour Drescher, Dilemmas of Democracy: Tocqueville and Moderniza-
tion (Pittsburgh: University of Pittsburgh Press, 1968).
21. Tocqueville, Voyages.
22. See G. W. Pierson, Tocqueville and Beaumont.
23. Michel Chevalier, Lettres sur l'Amérique du Nord, 2 vols. (1836).
24. Démocratie, 1:107 [GL 50].
25. Tocqueville, Voyages, 205-6.
26. Ibid., 342-87, esp. the account of ‘‘fifteen days in the wilderness,” Tocqueville and
Beaumont’s expedition to the outermost edge of the “frontier” near Lake Michigan.
27. Ibid., 346.
28. Ibid., 347.
29. Michel Chevalier, Lettres, esp. vol. 1, chap. 10, p. 149.
30. Tocqueville, Voyages, 155-56.
31. Ibid., 89.
32. Ibid., 258-59.
33. Démocratie, 1:426 [GL 316].
34. Ibid., 2:5 [GL 417].
35. Except, it is true, for his marriage, which his family regarded as a misalliance.
36. Démocratie, 2:311 [GL 634].
37. Ibid., 2:346 [GL 660].
38. Ibid., 2:256-61 [GL 595-600].
39Mibid-2221[GES78)
40. Ibid., 2:225 [GL 575].
41. Ibid., 2:174 [GL 538].
42. Ibid., 2:300 [GL 628].
43. Ibid., vol. 2, pt. 3, chap. 21.
44. By discussing the question of equality chiefly in terms of the passions conveyed by
egalitarian ideology and in terms of the relative frustrations engendered by equality,
Tocqueville once again reveals his astonishing modernity. An increasing number of contem-
porary sociological studies have been devoted to this theme, focusing in particular on the
concept of ‘‘reference group.” For a good summary of these studies, see Philippe Bénéton,
“Les Frustrations de l’égalité: contribution aux recherches sur la relativité des aspirations et
la perception des inégalités,” Archives européennes de sociologie 19 (1978).
45. Démocratie, vol. 2, pt. 2, chap. 5 [GL 513-17].
46. This aspect of Tocqueville’s genius was repugnant to Sainte-Beuve, who, in the
article in Causeries du lundi quoted above, criticized the abstract and systemic character of
the second volume of Démocratie. Indeed, generally speaking, Sainte-Beuve fails to com-
prehend Tocqueville’s thought, which he criticizes for its very nature.
47. Démocratie, 2:74-75 [GL 474].
48. This comparison is the object of an infinitely more systematic analysis in an article by
Raymond Aron, “La Définition libérale de la liberté: Alexis de Tocqueville et Karl Marx,”
Archives européennes de sociologie 5 (1964).
252 Notes to pages 212-228

Chapter 12

1. Patrick Girard, Les Juifs de France de 1789 à 1860: de l'émancipation 4 l'égalité,


Diaspora series (Paris: Calmann-Lévy, 1976).
2. Michael R. Marrus, Les Juifs de France à l’époque de l'affaire Dreyfus, Diaspora
series (Paris: Calmanr:-Lévy, 1972).
3. David H. Weinberg, Les Juifs de Paris de 1933 à 1939, Diaspora series (Paris:
Calmann-Lévy, 1974).
4. On this point, see Pierre Vidal-Naquet’s preface to Marrus, Les Juifs de France.
5. Hannah Arendt, ‘Privileged Jews,”’ Jewish Social Studies (January 1946).
6. Annie Kriegel, ‘““Révolution française et judaïsme,” L’Arche (March 1975).
7. Michael R. Marrus and Robert O. Paxton, Vichy et les Juifs, Diaspora series (Paris:
Calmann-Lévy, 1981).
8. André Harris and Alain de Sédouy, Juifs et Français (Paris: Grasset, 1979).
9. Alfred Fabre-Luce, Pour en finir avec l'antisémitisme (Paris: Juillard, 1979).

Chapter 13

1. Gershom Scholem, Fidélité et Utopie: essais sur le judaïsme contemporain, Diaspora


series (Paris: Calmann-Lévy, 1978). |
2. Gershom Scholem, Le Messianisme juif: essais sur la spiritualité du judaïsme, Di-
aspora series (Paris: Calmann-Lévy, 1974), trans. of The Messianic Idea in Judaism and
other essays on Jewish spirituality (New York: Schocken Books, 1971).
3. New York Review of Books 24, no. 6 (April 1977).
Abrégé d Histoire universelle (Boulain- Avenel, Georges d,’ 45
villiers), 127
Académie des Inscriptions, 81, 97 Baehrel, René, 52
Action Frangaise, 217 Barbarian, versus savage, 144-49, 156
Age at marriage, 61-65 Barrés, Maurice, 28
Agrarian economy, 50-51, 66, 73 Barthes, Roland, 34
Alembert, Jean Le Rond d,’ 141 Baudeau, Abbé, 120
Algerian war, 27, 29 Bayle, Pierre, 80, 148
Alibert, Raphaél, 217-18 Beaumarchais, Pierre-Augustin
Almanach de la librairie, 101-2 Caron de, 109
Althusser, Louis, 37-38 Beaumont, Gustave de, 167-68
America, 22, 23, 151-206; and French Belles-lettres, 77, 82, 111-23
Left, 232; and Israel, 240-41; savage Benedictines of Saint-Maur, 45, 78-79,
in, 145-46, 153-56, 181 91, 97
American Indian, 160, 180-84 Ben Gurion, David, 231, 239
American Revolution, 155-58, 164-65 Berl, Emmanuel, 216
Ancien Régime et la Révolution, L’ Besancon, Alain, 3
(Tocqueville), 174 Bible: commentaries on, 111, 114; and
Ancients and moderns, quarrel be- history, 78; and Zionism, 239-41
tween, 81, 97, 140-41 Bibliothèque Bleue de Troyes, 108
Angouléme manuscript, 130 Bibliothéque Nationale, 102-3, 105,
Annales, 1-5, 11-12, 19 GL, Wass sky
Anthropology, 5-7 Biography, and history, 43, 54-55
Antiquarians, 77-81 Blackmur, R. P., 229
Antonines, 143 Blacks, in America, 160; Tocqueville
Arabs, 230, 233, 239-40 on, 177, 182-84; and urban crisis,
Aragon, Louis, 28 197-206
Arche 160 OAS) Bloch, Marc, 1, 3
Arendt, Hannah, 214 Blum, Léon, 218
Argenson, comte d,’ 102, 110 Boas, Franz, 31
Aron, Raymond, 27, 28, 39 Bodin, Jean, 69
Art de vérifier les dates, etc., L,’ 113 Boileau, Nicolas, 115
Atala (Chateaubriand), 165-66 Bonheur, Le (Helvetius), 114

253
254 Index

Bossuet, Jacques-Bénigne, 78, 88, 127 Considérations (Montesquieu), 142


Boston Tea Party, 156 Contraception, 61-62, 65
Boulainvilliers, Comte Henri de, 81, Creoles, 154
125-39, 147 Crèvecoeur, Michel-Guillaume-Jean de,
Boumediene, Houari, 29 158-60
Braudel, Fernand, 3, 73 Criticism, 44
Bretonne, Restif de la, 110 Cuban revolution, 29
Brissot, Jacques-Pierre, 157, 160
Buber, Martin, 228 Daniel, Pére, 127, 128
Declaration of Independence, 156
Cabalists, 227, 229 Declaration of the Rights of Man, 91,
Capitalism, 171, 194-95, 230 215
Carnot, 90 Decline and Fall of the Roman Empire,
Carter, Jimmy, 203 The (Gibbon), 140-49
Cartesianism, 83 Démeunier, 162
Catholicism, 179, 233 Democracy: and anti-Semitism, 235;
Cavour, Count, 237 Tocqueville on, 167-96
Céline, Louis-Ferdinand, 216 Democracy in America (Tocqueville),
Cels-Martin, 111 12, 167-96
Charlemagne, 129, 137 Demography, 5-10, 17-18, 41-53,
Charles VI of France, 128 59-66, 72
Chastellux, marquis de, 157 De Pauw, Cornelius, 145-46; Re-
Chateaubriand, Vicomte de, 165-66, cherches philosophiques sur les Améri-
168, 171 cains, 154-55, 158, 160, 165
Château de Versailles, 89-90 Dépôt Légal, 100, 105
Chaunu, Pierre, 42, 62 Dictionnaire (Trévoux), 145
Chevalier, Michel, 178, 180 Diderot, Denis, 101, 103, 117, 121; En-
Cheverus, Jean-Louis Lefebvre de, 168 cyclopédie, 92, 100, 103, 119, 145-46
China, 7, 29 Discours préliminaire (d’ Alembert), 141
Christianity: and fall of Rome, 148-49; Discours sur les sciences et les arts
and history in France, 78-79; (Rousseau), 113-14
Tocqueville on, 179 Doriot, Jacques, 216
Cicero, 82, 141-42, 143 Dreyfus affair, 28, 213-14, 218, 220
Civilization: in America, Tocqueville Droits et devoirs du citoyen (Mably),
on, 179-80; as eighteenth-century 131
problem, 125-26, 141, 147; and sav- Dubos, Abbé, 134, 147
age state, 144-49, 153-56, 165; as Dumont, Louis, 22, 132
subject of history, 5, 8, 55 Durkheim, Emile, 5, 6
Clothar II, 134 Duruy, Victor, 90-91, 94
Clovis, 81
Cobb, Richard, 12-19 Ecole des Hautes Etudes en Sciences
Code de la Librairie, 101 Sociales, 2-3
Collège de France, 84-85, 89, 98 Ecole Pratique des Hautes Etudes, 2, 91
Collége de Juilly, 84 Economics, 5, 10, 41-53, 59-66
Collége Louis-le-Grand, 83 Eden, Anthony, 231
Collége of Arras, 83 Elias, Norbert, 63
Collegès, 82-85 Emile (Rousseau), 110
Communist Party, 3, 28-29, 38, 230-32 Encyclopédie, 92, 100, 103, 119, 145-46
Computers, 44-45, 48, 57-58 England, and Tocqueville, 168
Comte, Auguste, 5, 87 Entretiens de Phocion (Mably), 136
Condillac, Étienne Bonot de, 86, 109 Equality: in eighteenth-century writings,
Condorcet, marquis de, 97, 109, 118, 126, 132-39; in present-day America,
160; and ‘“‘civilization,” 125, 141, 143, 204-6; Tocqueville on, 168, 178,
180; curriculum of, 85-87 186-96
Index 255

Erceville, Rolland d,’ 83 Gibbon, Edward, 81, 88, 97; The De-
Errera, Roger, 212 cline and Fall of the Roman Empire,
Erudition, 77-82, 128 140-49
Esprit des lois, L’ (Montesquieu), 105, Girard, Patrick, 212-14
110, 164 Girondins, 157, 160
Esquisse d'un tableau historique des pro- Glorious Revolution, 90
grès de l'esprit humain (Condorcet), Goths, 147
141 Goubert, Pierre, 52
Essai (Montesquieu), 142 Great Sanhedrin, 212
Estivals, Robert, 100, 103 Greece, ancient: America versus, 158,
Ethnology, 7-8, 17-18, 29-39, 68-74 159; and history, 55, 77, 92; Mably
Event, historical, 11-12, 49, 51, 55-57, on, 129
71-72 Guedj, Rabbi, 221
Guizot, François, 28, 168, 190; and
history in France, 88-91, 96, 98
Fabre-Luce, Alfred, 222-24
Family, and urban crisis, 198
Fanon, Frantz, 29 Haam, Ahad, 237, 239
Fascism, 28, 35 Haganah, 237
Febvre, Lucien, 1-4, 111 Harris, André, 220-22
Ferguson, Adam, 145 Hegel, Georg Wilhelm Friedrich, 34, 37
Feudalism, 133-35 Helvetius, Claude-Adrien, 109, 114
Fidélité et Utopie (Scholem), 227 Henriade (Voltaire), 108
Fleury, André-Hercule de, 108 Henry II of France, 47
Fonds Séguier, 46-47 Henry IV of France, 46
Fontenelle, Bernard Le Bovier, Sieur Herzl, Theodore, 225, 235-37
de, 80 Histoire critique du Vieux Testament
Ford, Gerald, 205 (Simon), 78
Fortoul, 90 Histoire de l'Ancien Gouvernement de la
Foucault, Michel, 34-35 France (Boulainvilliers), 127
Franklin, Benjamin, 155 Histoire des empereurs (Tillemont),
Franks, 129, 131, 135, 147 79-80
French Revolution, 27-28, 30, 46, 131; History, definitions of: histoire des men-
and American Revolution, 157-60, talités, 15-18, 64-65; histoire
164-65; and Enlightenment, 123; and événementielle, 48-49, 71-72; narra-
Jews, 211, 215, 235, 239; Lavisse on, tive, 8-9, 54-56; political, 9; serial,
93-94; Lévi-Strauss on, 33-34; Marx 42, 58, 73; ‘‘total,”” 5-6, 67, 70
on, 193; and nationhood, 125, Hitler, Adolf, 70, 215-18, 231, 235-41
138-39; and teaching of history, 85- passim
98; and Tocqueville, 169-71, 185-86, Hoffman, Stanley, 223
195-96; and two types of history, Holocaust, 211, 221, 230, 232-33,
10-11 238-41
Fréres des Ecoles Chrétiennes, 84 Hundred Years’ War, 93
Freud, Sigmund, 20, 32, 64, 141 Hugo, Victor, 27-28
Fronde, 34, 125 Humanism, 69, 142
Fustel de Coulanges, Numa-Denis, 91 Hyde de Neuville, Jean-Guillaume,
Baron, 168

Garaudy, 37
Garibaldi, Giuseppe, 237 Ideologies, end of, 27-31
Gaullism, 27, 31, 232 Idéologues, 86-87, 97, 162-63
Gauls, 130, 135 India, 7
Genesis, Book of, 36 “Instructions” (Lavisse), 92, 94
Geography, 72, 82, 85, 115 Irgun, 237
Germanic tribes, 129-38, 146-47 Islam, 29, 148, 227, 239-40
256 Index

Israel, 23, 211; and French Jews, Lévi-Strauss, Claude, 22, 30-39, 73; La
221-22; and French Left, 230-33; and Pensée sauvage, 31-36
Scholem, 228-29; and Zionism, Libraires, 101-22
234-41 Lindsay, John, 200
Livy, 77, 78, 82
Jabotinsky, Vladimir, 237 Louis Philippe, 89-90, 185
Jacobins, 30, 96, 166, 230, 231; and Louis XI of France, 93
American Revolution, 162; tradition Louis XIII of France, 47
of, and French Jews, 211, 221 Louis XIV of France, 7, 81, 102, 140,
Jacquerie, 46-47 163, 170; and nobility, 126, 137;
Jansenists, 82-83, 105, 114, 117, 119 and Port-Royal, 82, 83; and Voltaire,
Jefferson, Thomas, 158 55, 69
Jews, Judaism, 22-23, 207-41; and Louis XV of France, 100, 120, 169-70
blacks in America, 205 Louis XVI of France, 100, 101
Jivaro, 34-35 Louis XVIII of France, 88
Johnson, Lyndon, 201, 203 Lueger, Karl, 235
July Monarchy: development of history Luria, Isaac, 227
during, 89-91, 96, 98; and equality,
185; Jews under, 213, 214 Mabillon, Jean, 79
Mably, Gabriel Bonnet de, 97, 109,
125-39, 146-47
Kant, Immanuel, 36
Malesherbes, Chrétien-Guillaume de
Kapital, Das (Marx), 35
Lamoignon de, 100-110, 114
Kennedy, John F., 203
Malthus, Thomas, 51, 62
Kergorlay, Louis de, 168, 170
Manchester, 71
Klatzkin, Jacob, 236-37
Manuel de l'auteur et du libraire, 101
Kriegel, Annie, 215
Marriage: age at, 61-65; Tocqueville
Kriegel, Maurice, 222
on, 186-87
Marrus, Michael, 213-20
Labrousse, Ernest, 45 Martin, Henri-Jean, 111
La Chalotais, Louis René de Caradeuc Marx, Karl, 28, 70, 71, 233, 238;
de, 83 Das Kapital, 35; Lévi-Strauss com-
La Fayette, marquis de, 157, 168 pared with, 32, 36; Sartre compared
Lakanal, Joseph, 86 with, 34; Tocqueville compared with,
Languedoc, 50-51, 109-10 171-72, 192-96
La Popeliniére, 69 Marxism, 3, 4, 16, 20, 46, 57, 123; of
La Rochefoucauld, 162 French intellectuals, 27-39; and
La Rochelle, Drieu, 28 Israel, 230-33
Laslett, Peter, 17 Maurras, Charles, 28, 217
Laval, Pierre, 218-19 Mauss, Marcel, 31
Lavisse, Ernest, 92-94, 96 May, Georges, 122
Lazare, Bernard, 214, 227 Mazzei, 158
Leach, Edmund, 36 Mazzini, Giuseppe, 237
Le Bossu, René, 115 Mémoires de Trévoux, 105
Legon inaugurale (Lévi-Strauss), 33 Mémoires pour servir à l'histoire ecclé-
Ledoux, Claude-Nicolas, 119 siastique (Tillemont), 79
Lefebvre, Georges, 11 Mémoire sur les problèmes de la librairie
Lenin, 28, 29, 37 (Malesherbes), 102, 104
Lenôtre, Georges, 19 Mentalités, 15-18, 64-65
Le Roy Ladurie, Emmanuel, 3, Messianisme juif (Scholem), 212
50-51, 62 Mézeray, François Eudes de, 128
Lettres de Madame de Maintenon, 108-9 Michelet, Jules, 89-91, 98
Lettres sur l'Amérique du Nord (Cheva- Mirabeau, 110, 157, 161
lier), 178 Molé, Louis-Mathieu, Comte, 172
Index 257

Mollet, Guy, 231 Pensée sauvage, La (Lévi-Strauss),


Monod, Gabriel, 91 31-36
Montesquieu, 134-35, 146-47, 149; Pensées morales adaptées aux figures de
L’Esprit des lois, 105, 110, 164; and l’Ancien Testament qui représentent
Rome, 81, 142; on savage and barbar- Jestis-Christ, 119
ian, 145; and Tocqueville, 169, 175, Pericles, 55, 81
179 Periodization of history, 43, 53
Mornet, Daniel, 117, 121-22 Permissions, 101-22
Mortality rate, 61 Pétain, Philippe, 28, 219, 224
Mots et les choses, Le (Foucault), 34 Petty, William, 41
Mounier, Jean-Joseph, 161 Philip the Fair, 133, 134
Myth, 32 Philology, 78
Philosophes, 80-81, 119, 123, 155
“Philosophical history,” 80-84, 86, 92
Napoleon, 55, 60, 212
Philosophy: eighteenth-century pub-
Nasser, Gamal Abdel, 231
lication of, 112, 113-14, 117, 118;
Natches, Les (Chateaubriand), 165-66
erudition versus, 77; of history, 4-5,
Nationalism: and civilization, 144;
33, 49
and French Jews, 214-16; and history,
Plutarch, 78
7, 69-71; and new Left, 230-31
Poliakov, Léon, 223
Nation-state: as eighteenth-century
Political economy, 41—42, 46, 49
problem, 125-26, 138, 147; and
Popular Front, 28
French Revolution, 92, 96; as subject
Port-Royal, 78, 82-83, 97
of history, 44, 68-69
Positivism, 20, 42, 44
Nazis, 23, 215-24, 238
Pour en finir avec l'antisémitisme
Negro. See Blacks
(Fabre-Luce), 222-24
New Deal, 202-3
Privilèges, 101-22
New York City, urban crisis in, 199-200
Progress, idea of, 49, 52, 56-57, 69-71,
Nixon, Richard, 205
140-41
Nobility: eighteenth century, 126, 130,
Protestant Ethic, The (Weber), 66
132-39; and Tocqueville, 170-71
Protestantism, 69, 78, 179
Noinville, Durcy de, 111
Psychoanalysis, 16, 32
Notre jeunesse (Péguy), 227
Puritans, 177, 180
Nouvelle Héloise, La, 121
Nouvelles ecclésiastiques, 105, 119
Novel, 120-23, 166 Quakers, 155, 177
Qu'est-ce que le Tiers Etat? (Sieyès),
138
Observations générales sur les Indiens ou
Quinet, Edgar, 90
Sauvages de l'Amérique du Nord
(Volney), 165
Old Testament, 68-69 Reagan, Ronald, 203, 205
Ophüls, Max, 223 Recherches philosophiques sur les
Oratorians, 78, 82, 84 Américains (de Pauw), 154-55,
Ordre moral, 91, 94 158, 160
Orleanism, 89, 96 Reformation, 21, 90, 140
Regards sur le monde actuel
Parlements, 83, 85, 100, 110-11, 120, (Valéry), 31
136 Religion: in eighteenth-century pub-
Parti Socialiste Unifié, 230-31 lishing, 111-17, 123; and fall of
Pasquier, Étienne Denis, Duc, 69 Rome, 148-49; Tocqueville on, 179,
Patouillet, 113 188, 191
Paxton, Robert, 215-20, 223 Renaissance, 53, 77, 80-81, 100, 134
Péguy, Charles, 227 Renan, Ernest, 91, 141
Pellepoix, Darquier de, 218 Restoration, 88, 98, 168, 186
258 Index

Revolution: and democracy, 174-75, ““Social-democratization,” 203—4


188-89; and Zionism, 238-39. Social sciences, 173; and history, rela-
See also American Revolution; tionship between, 1-74
French Revolution Society of Jesus, 82-84, 90, 105
Revue internationale des sciences Sociology, 5-10, 41, 72
sociales, 33 Sorbonne, 2, 84, 88-89, 100, 105
Richardson, Samuel, 121 Soviet Union, 28, 30, 46
Richelieu, Cardinal de, 138, 170 Spinoza, Benedict de, 130
Richet, Jacques Ozouf Denis, 3 Stalin, 28, 55
Ricoeur, Paul, 36 State of nature. See Savage state
Robertson, William, 146 Structures élémentaires de la parenté
Robespierre, 86, 123, 159, 162 (Lévi-Strauss), 30
Rolland, Romain, 28 Style noble, 117, 119
Rome: America versus, 158, 159; and Suez crisis, 231
Gibbon, 140-49; and history, 55, Superego, of Weber, 63-66
68-69, 77, 92 Switzerland, 135, 141, 168
Rosenzweig, Franz, 228
Rostow, Walt, 71 Tableau du climat et du sol des Etats-
Rousseau, Jean-Jacques, 27-28, 117, Unis (Volney), 163-65
118, 136; Discours sur les sciences et Tacitus, 77, 82, 141-43, 146, 183
les arts, 113-14; Emile, 110; ‘‘savage” Taine, Hippolyte, 91
Of, 30) 81156 065 Talleyrand, 85, 162
Royer, Collard, Pierre-Paul, 88 Thermidoreans, 86-87, 162-63
Russian Revolution, 21, 70 Third Republic, 91-92, 94, 97, 220
Third World, 29-31, 38, 70-71, 232
Sainte-Beuve, Charles-Augustin, 169 Thucydides, 77, 143, 146
Saint-Germain-des-Prés, 79 Tillemont, Sébastien Le Nain de,
Saint-Just, Louis, 123, 159 78-80, 141
Saint-Simon, Comte de, 87, 133, 180 Tilly, Charles, 47
Salvandy, 89, 90 Tocqueville, Alexis de, 12, 22, 91, 204;
Sartre, Jean-Paul, 33-36, 230 Democracy in America, 167-96; on
Savage state: barbarism versus, French Revolution, 161
144-19, 156; and civilization, 144-49, Tocqueville, Hervé de, 169-70
153-56, 165-66; as infancy of man, 7, Tolstoy, Leo, 238
68, 82; Mably and Boulainvilliers on, Touraine, Alain, 203-4
129-30; and structuralists, 30, 31-32; Travel: accounts of, 115; versus history,
Volney on, 165 7, 68-70, 82, 156
Sceau Public de Chancellerie, 100-103 Trévoux, Dictionnaire of, 145
Scholem, Gershom, 212, 225-29, 233 Turgot, Anne-Robert-Jacques, 120
‘Sciences and arts,’ 111-20, 123-24
Scripture. See Bible Unigenitus Bull, 114
Second Empire, 90-91, 214 United States, 232. See also America
Second Republic, 90 Universal History (Bossuet), 78
Section Française de l’Internationale Universities, 83
Ouvière, 230-31 University of Paris, 83
Sédouy, Alain de, 220-22
Seignobos, Charles, 94-96 Valéry, Paul, 31
Servan, 115 Vallat, Xavier, 216-18
Shukairy, 231 Vauban, Sébastien, 41
Siècle de Louis XIV (Voltaire), 108 Ventre, Madeleine, 109
Sièyes, Emmanuel-Joseph, 138 Vichy regime, 211, 215-24
Simon, Richard, 78 Vietnamese, 230
Six Day War, 221, 231 Vilar, Pierre, 52
Slavery, black, 177, 180, 183-84 Villèle, Joseph, comte de, 88-89
Index 259

Volney, Constantin-François de Chase- Weiseltier, Leon, 229


boeuf, comte de, 162-65 Weizmann, Chaim, 237
Voltaire, 27-28, 108, 122; on civiliza- Welfare system, in America, 199-205
tion, 55, 69, 81, 125, 180; on history, Women’s movement, 205-6
128
Zevi, Sabbatai, 227
Wars of religion, 125 Zionism, 215, 221-23, 225-29
Waterloo, battle of, 55 Zola, Emile, 28
Weber, Max, 63-66 Zunz, Leopold, 227
Weinberg, David H., 214
Vo RÉ este) re
Len (bé loue At
LEO -

D * En .

€ 7

reas QE 1e
'
To (arr Re «

PCIe & <a,


in
ire 4 COR
iad, |

L TT CE

perty vs

==
zs

D. LA d te

» en

| A

=. .

: Coque.
mA

Tr

ve

e D.
SRE

ÈS
à

RSR RE Pistes trs)


MANU
=
at
sare
Rss

sepsnree
sectgeeecti=

cé > Te +
frà yee $2 SFIS
<: i
Sterne SS: RS, ase Der Pise
=e
PARIS 3 NS:
+
Foie orme
ES: nier es AE
SE DR easasrso te LE
HE
HAE FEDEIRE HSE
=

SRE
BUS
HSE
Eat

Peat ettees ee
ae

HR

He
RITES

re
ERNST RER
LT:
FE

a
de
ost Gesen ee

FE
RES

You might also like