ALGERIAN, FRENCH, REFUGEES, REPATRIATES, IMMIGRANTS

The Pennsylvania State University
The Graduate School
College of the Liberal Arts
ALGERIAN, FRENCH, REFUGEES, REPATRIATES, IMMIGRANTS?
HARKI CITIZENS IN POST-IMPERIAL FRANCE (1962-2005)
A Dissertation in
French
by
Jeannette E. Miller
© 2012 Jeannette E. Miller
Submitted in Partial Fulfillment
of the Requirements
for the Degree of
Doctor of Philosophy
May 2012
The dissertation of Jeannette E. Miller was reviewed and approved* by the following:
Willa Z. Silverman
Professor of French and Jewish Studies
Dissertation Adviser
Chair of Committee
Jennifer A. Boittin
Associate Professor of French, Francophone Studies, and History
Tobias A. Brinkmann
Malvin and Lea Bank Associate Professor of Jewish Studies and History
Thomas A. Hale
Edwin Erle Sparks Professor of African, French, and Comparative Literature
Sophie de Schaepdrijver
Associate Professor of History
Patrick Weil
Directeur de recherche, CNRS/Université de Paris-I
Special Member
Jean-Claude Vuillemin
Professor of French
Head of the Department of French and Francophone Studies
*Signatures are on file in the Graduate School.
ii
ABSTRACT
Through an analysis of the national creation and the local implementation of French government
policies toward the harki population, “Algerian, French, Refugees, Repatriates, Immigrants? Harki
Citizens in Post-Imperial France (1962-2005)” examines this group’s post-imperial citizenship. To
escape the violence in Algeria, 20,000 of these native Algerians who fought for France during the
1954-1962 Algerian War migrated to France with their family members (totaling 100,000 people)
during the 1960s. The government placed half of the “repatriated” population in camps, choosing to
hide from public view these reminders of the end of France’s colonial dominance. Using two local
case studies—the Rivesaltes camp near Perpignan and a rural housing development in the Provencal
forest—this dissertation focuses on harki citizens’ nationality, exile, integration, protests, and
memorialization. The post-imperial French state, indelibly marked by colonial methods of
governance and the loss of French Algeria, marginalized harki citizens from both French society and
access to some rights accorded other citizens. Government officials enacted policies that prolonged
the colonial Algerian practice of differentiating populations based on ethnicity and attributed to the
harkis an unclear status, which questioned their belonging on French soil. Grounded in documents
from twenty different national and local public archives and private archival collections, my study is
balanced with interviews of harki population members and government officials, memoirs,
periodical articles, and judicial texts.
iii
TABLE OF CONTENTS
LIST OF ABBREVIATIONS
vi
LIST OF FIGURES
viii
LIST OF TABLES
ix
ACKNOWLEDGEMENTS
x
Introduction Harki Citizens: Between Algeria and France, Between the Colonial and Postcolonial
1
Harki Citizens: A Contradiction
Ethnic Soldiering and the Harkis
Algeria, France, and Harki Citizens
The Colonial, the Postcolonial, and Harki Citizens
Writing about the Harki Population: From Actors to Observers
A Note on Sources
Overview
5
10
15
21
26
32
36
Section 1 Introduction: The Harkis’ Nationality Amidst a Franco-French War
38
Chapter 1 Algerians? The Harkis’ Deeply “Rooted” Colonial Status
44
The “Nationalité Dénaturée” of Algerian Muslims
The “Nationalité Dénaturée” of the Harkis
The Harkis’ Future in an “Algerian Algeria”
The Harkis’ Future in Post-Imperial France
Conclusion: Algerians in a Franco-French War
Chapter 2 French? Movements of Post-Imperial Citizens
Distancing Harki Soldiers from France
Violence and “Repatriation”
“Harkis” as “Rapatriés”?
The OAS, the Harkis, and “Repatriating” the Franco-French War
“Security” over Movements of (Post) Empire
Violence and Nationality
Becoming “French” Citizens
Conclusion: Harki Citizens
48
52
56
61
68
70
77
81
94
104
111
118
131
141
Section 2 Introduction: Harki Citizens’ Exile in France
144
Chapter 3 Refugees? The Harkis’ “Univers Concentrationnaire” at the Rivesaltes Camp
161
A Camp for “Foreigners”
Tents and Barracks
The Power of Encadrement
Community of Exile: An Isolated Population
Social Advancement: Toward the Reclassement of the Harkis
Conclusion: Leaving the Rivesaltes “Univers Concentrationnaire”
iv
166
173
194
207
218
225
Chapter 4 Repatriates? The Post-Imperial Government’s Forest Hamlet “Integration” Policy
The First Phase: Developing the Government’s Contradictory Forest Hamlet Policy
The Second Phase: Evaluating and “Extend[ing] the Forest Hamlet Experiment”
Envisioning Permanent Forest Hamlets & Ending the Ministry of Repatriates’ Mission
Shifting Responsibilities and a Permanent Forest Hamlet Policy
Conclusion: Failed Integration Policy for the Harki “Repatriates”
Chapter 5 Immigrants? Isolation in Rural Southern France
Fuveau’s “Foreigners”: From Miners to Harkis
The First Decade in Fuveau: Isolation and Encadrement
The First Decade in Fuveau: Isolation and Integration
Chapter 6 Refugees, Repatriates, and Immigrants. The “Harki Problem”
Reevaluating Government Policies
Protests and Revising Government Policies
The Failure of Revised Policies in Fuveau
Conclusion: Refugees, Repatriates, Immigrants?
Conclusion Harki Citizens. Remembering a Forgotten Population
From 1975 to 1991: Protests, Status Quo, and More Protests
Harki Citizens and Commemoration
From Algeria to France, From the Colonial to Postcolonial
228
236
248
267
273
288
293
299
309
319
337
340
350
361
369
373
377
385
391
Appendix A Interview with Monsieur le Chancelier de l’Institut Pierrre Messmer
400
Appendix B “Annexes relatant les principales exactions connues”
407
Appendix C “Audition du nommé AA”
410
Appendix D List of Forest Hamlets in France
413
Appendix E Population of Forest Hamlets
416
Appendix F “Règlement intérieur concernant l’hébergement dans les hameaux forestiers”
417
BIBLIOGRAPHY
418
v
LIST OF ABBREVIATIONS
ACF: Archives Communales de Fuveau
ACR: Archives Communales de Rivesaltes
ACRA: Archives Communales de la Roque d’Antheron
ACNMF: Archives du Comité national pour les Musulmans Français
ADAF: Amicale des Algériens en France
ADBR: Archives Départementales des Bouches-du-Rhône (Marseille)
ADPO: Archives Départementales des Pyrénées-Orientales (Perpignan)
ADV: Archives Départementales du Var (Draguignan)
AFN: Afrique du nord
ALN: Armée de Libération Nationale
AN: Centre Historique des Archives Nationales (Paris)
ANP: Armée nationale populaire (d’Algérie)
AONACBR: Archives du Service Départemental de l’Office national des Anciens Combattants et Victimes
de Guerre des Bouches-du-Rhône (Marseille)
AONFBR: Archives de l’Office National des Forêts, Département des Bouches-du-Rhône (Aix-enProvence)
ASRBR: Archives du Service des Rapatriés des Bouches-du-Rhône (Marseille)
ASANHLMM: Archives de la Société Anonyme Nouvelle d’HLM de Marseille
ASSRA: Assistante sanitaire et sociale rurale auxiliaire
ATOM: Aide aux travailleurs d’Outremer
BIAC: Bureaux d’information, d’aide administrative et de conseils pour les Français musulmans
CAA: Corps d’Armée d’Alger
CAC: Centre des Archives Contemporaines (Fontainebleau)
CAC: Corps d’Armée de Constantine
CAO: Corps d’Armée d’Oran
CARA: Cité d’Accueil des Rapatriés Algériens (Bias)
CEMJ: Centre d’Entraînement des Monitrices de la Jeunesse Algérienne (Nantes)
CIC: Certificat d’Initiation de Construction
CIMADE: Service œcuménique d’entraide, formerly Comité Inter-Mouvements auprès des Evacués
CFMRAA: Confédération des Français musulmans rapatriés d’Algérie et leurs amis
CFPA: Centre de Formation Professionnelle Accélérée (Rivesaltes)
CMFP: Centre Militaire de Formation Professionnelle (Rivesaltes, Fontenay-le-Comte)
CNMF: Comité nationale pour les Musulmans Français
CRA: Centre de Rétention Administrative
CSFA: Commandant Supérieur des Forces en Algérie
EMAT: Etat-major de l’Armée de Terre
EMI: Etat-major Interarmées
FAFA: Forces Armées Françaises en Algérie
FFFLN: Fédération française du front de libération nationale
FLN: Front de libération nationale
FMA: Français musulman d’Algérie
FMR: Français musulman rapatrié
FNRFCI: Front National des Rapatriés Français de Confession Islamique
FPA: Force de police auxiliaire
FSNA: Français de souche nord-africaine
FSE: Français de souche européen
GAD: Groupe d’autodéfense
GMPR: Groupe mobile de protection rurale
GMS: Groupe mobile de sécurité
GPRA: Gouvernement Provisoire de la République Algérienne
vi
HLM: Habitation à loyer modéré
IGAME: Inspecteur Général de l’Administration en Mission extraordinaire
INA: Institut national de l’audiovisuel
INSEE: Institut national de la statistique et des études économiques
JORF: Journal Officiel de la République Française
MADRMAN: Le Mouvement d’assistance et de défense des rapatriés musulmans d’Afrique du Nord
MAE: Archives du Ministère des Affaires Etrangères (La Corneuve, formerly in Paris)
MRP: Mouvement Républicain Populaire
OAS: Organisation de l’Armée Secrète
ONAC: Office national des Anciens Combattants et Victimes de Guerre
ONF: Office national des forêts
PCF: Parti Communiste Français
PSU: Parti Socialiste Unifié
RONA: Rapatriés d’origine nord-africaine
RPF: Rassemblement pour la France
SANHLMM: Société Anonyme Nouvelle d’HLM de Marseille
SAS: Sections administratives sociales
SCINA: Service de coordination et d’information nord-africaine
SEAA: Secrétariat d’Etat chargé des Affaires algériennes
SFIM: Service d’accueil et de reclassement des Français d’Indochine et des Français musulmans
SHAT: Service Historique de l’Armée de Terre (Vincennes)
SONACOTRA: Société nationale de construction pour les travailleurs
SPDP: Sciences Po Dossiers de Presse (Paris)
UN: United Nations
vii
LIST OF FIGURES
Figure 1: “L’Algérie de demain” pamphlet
66
Figure 2: Map of northern Algeria
71
Figure 3: Refugee camps on the northern coast of Algeria
89
Figure 4: The Bourg-Lastic and Larzac transit camps located in rural southern France
108
Figure 5: Harkis arriving in Marseille on June 12, 1962
109
Figure 6: Harki families waiting in the Marseille harbor to be transported to the Bourg-Lastic
camp June 23, 1962
109
Figure 7: Harki families residing in canvas military tents at the Bourg-Lastic camp in
summer 1962
112
Figure 8: Makeshift courtroom set up for nationality ceremonies at the Rivesaltes camp
133
Figure 9: A harki in front of the magistrate during a nationality ceremony in summer 1963
133
Figure 10: Copy of Mohamed M.’s nationality declaration
135
Figure 11: Present-day map of Fuveau
150
Figure 12: Ferhat K. preparing kesra
152
Figure 13: Geographical location of Rivesaltes
167
Figure 14: Image of the CMFP Nº1 from a pamphlet issued by the Ministry of Labor
171
Figure 15: Harkis arriving in Rivesaltes from Bourg-Lastic on September 16, 1962
175
Figure 16: Monthly population statistics of the harkis residing at the Rivesaltes camp
178
Figure 17: Housing of harkis at the Rivesaltes camp in tents versus buildings
180
Figure 18: Sketch of the Rivesaltes camp in the fall of 1962
182
Figure 19: Hollowed out barrack at the Rivesaltes camp
187
Figure 20: The dimensions and layout of the barracks, as captured in July 1963
188
Figure 21: Outhouses with Turkish toilets on the Rivesaltes camp grounds
189
Figure 22: Carnet familial de rapatrié of M’hamed H.
205
Figure 23: Number of weekly arrivals and departures of harkis to and from the Rivesaltes camp 209
Figure 24: Percentage of reclassement by number of jobs
viii
248
Figure 25: Percentage of reclassement by number of people
248
Figure 26: Number of forest hamlets per department
260
Figure 27: Number of forest hamlet inhabitants from December 15, 1962 to January 1, 1965
272
Figure 28: The town of Fuveau
295
Figure 29: Satellite image of Fuveau in 2011
301
Figure 30: Map of Fuveau in 1974
303
Figure 31: Standardized architectural drawing of the “type A” forest hamlet buildings
306
Figure 32: Officials welcoming harkis and their families to the sylvan surroundings in Fuveau
310
Figure 33: A harki and his family walking to their new home in the Fuveau forest hamlet
321
Figure 34: Children of harkis blocking the entrance to the Saint-Maurice-l’Ardoise camp in 1975 355
Figure 35: Masked harki sons bearing rifles and dynamite on the balcony of the Saint-Laurentdes-Arbres town hall
356
Figure 36: Riot police outside of the Bias camp on August 18, 1975
358
Figure 37: Blueprint of the Fuveau forest hamlet
362
Figure 38: Ali H. pictured in front of one of the original forest hamlet buildings in 1996 with a
former GMS soldier who resides in Fuveau
364
Figure 39: Row of four reconstructed residences at the Fuveau forest hamlet
366
Figure 40: Protest at the Fuveau forest hamlet, July 1, 1991
382
Figure 41: Peaceful march led by harkis and their families on July 13, 1991
385
Figure 42: Aissa B. and Abdelkader B. in front of le monument aux morts in Fuveau on September 398
25, 2009
ix
LIST OF TABLES
Table 1: Number of auxiliary soldiers serving in the French Army (January 1957-March 1962)
14
Table 2: Population of European descent in Algeria from 1833 to 1926
95
Table 3: Harkis age eighteen and older requesting, granted, and refused French nationality from 130
1962 to 1970
Table 4: List of forest hamlets in France
252
Table 5: Forest hamlets and forestry worksites open from December 1962 to December 1975
267
Table 6: Population of the Fuveau forest hamlet from 1964 to 1976
335
Table 7: Size and number of the thirty-one new houses in the Fuveau forest hamlet
364
x
ACKNOWLEDGMENTS
While studying for my comprehensive exams I came across the first time that I heard the word
“harkis.” I wrote in my October 10, 2002 notes from Emmanuelle Saada’s “Colonization,
Immigration, and National Identity” class:
les harkis (indigènes who sided w/ Fr Army) – who were massacred, or if they moved to Fr
they were treated horribly – refugee camps  in last couple of yrs they have been recognized
The same day Emmanuelle instructed us to begin thinking about a topic for our final papers. My
first final paper in graduate school, “The Recognition of the Harkis,” set me off on a quest to
understand the contradiction exposed in the two lines of my class notes. Over nine years later,
having conducted research in twenty different national, departmental, local, and personal archives as
well as having and interviewed dozens of harkis, civil servants, local and national politicians who
oversaw harki citizens during the last 50 years, I submit this dissertation.
To begin my journey, faculty, courses, colleagues, and speakers at New York University’s Institute of
French Studies instilled in me an intellectual curiosity about France’s largely occulted immigration
history and the workings of the French Fifth Republic. The encouragement that I continue to
receive from Ed Berenson, Herrick Chapman, and Françoise Gramet is precious to me. During my
PhD training at Penn State University, I was grateful to the Department of French and Francophone
Studies for offering me many occasions to present my work. Each time, I received thoughtprovoking feedback from fellow graduate students and professors.
My dissertation was enriched by the good fortune to work with two fabulous directors. Willa
Silverman’s perceptive reading of many drafts of my work helped me to become a more critical
thinker and a better writer. Your support and compassionate belief in me are greatly cherished.
Patrick Weil encouraged me to pursue my final paper for his Fifth Republic France class as my
dissertation topic under his co-direction. I thank you for pushing me to leave no archival or
interview stone unturned and for our many probing conversations over the years. My committee
members—Jennifer Boittin, Tobias Brinkmann, Thomas Hale, and Sophie de Schaepdrijver—made
valuable contributions to my development as a scholar and were devoted advocates in my academic
pursuits. Tobias, I particularly appreciate your astute reading of chapters in draft form and our
invaluable conversations about migration.
The intellectual support I received extended outside of my departments and universities. From my
first semester of graduate school through the end of the dissertation project, I benefited from many
stimulating discussions on both sides of the Atlantic. I particularly wish to acknowledge Françoise
de Barros, Vincent Crapanzano, Greg Eghigian, Claire Eldridge, François-Xavier Hautreux, JeanJacques Jordi, Abderahmen Moumen, Ed Naylor, Denis Peschanski, Stéphanie Ponsavady, Todd
Shepard, Alexis Spire, Benjamin Stora, and Sylvie Thénault. I extend heartfelt appreciation to my
dear friends and colleagues Melissa K Byrnes and Yann Scioldo-Zürcher for providing me with
excellent comments on my work and seeing me through the ups and downs of research, writing, and
life as a graduate student.
I am extremely grateful to the following organizations and institutions for the financial support that
made my schooling and research possible. The Penn State University College of Liberal Arts and the
xi
New York University Graduate School of Arts and Sciences funded my graduate training and travel
to conferences. A pre-dissertation grant from the Florence J. Gould Foundation allowed me to
explore the French archives for the first time. The generous support of the William J. Fulbright
Foundation, la Société des professeurs français et francophones d’Amérique, the Society for French Historical
Studies, the Western Society for French History, and l’Institut français de Washington made possible my
eighteen months of research in France, without which this dissertation could not have been written.
The Africana Research Center at Penn State University provided me with a teaching release to begin
writing. The Camargo Foundation in Cassis offered me ideal and picturesque conditions for my
work. And the CNRS/NYU Transitions Research Center welcomed me as a fellow to participate in
its seminars, which strengthened my foundation in memory studies.
In France, I wish to thank the many archivists, librarians, and individuals who facilitated my access
to the thousands of documents that I consulted. In particular, Claudie Le Cossec at the Archives
Nationales persistently hunted down officials at numerous ministries to help me acquire derogations
and François Gasnault and Géraldine Gall at the Archives départementales des Bouches-du-Rhône
were particularly helpful as I searched for documents that had not been deposited into the archives.
A special thank you to the following people who allowed me to consult their personal archives and
interview them: Slimane Djera, Marie-Joseph Haupte, Saïd Merabti, Prime Minister Pierre Messmer,
Senator Roger Romani, and André Wormser. My thanks as well as to the following associations and
public offices for granting me access to their papers: the Office National des Forêts departmental
office in Aix-en-Provence, the Direction de la Cohésion Sociale et Emploi (successor to the Service
des Rapatriés) in the Marseille Prefecture, the Mission Interministérielle aux Rapatriés in Paris, the
Société Anonyme Nouvelle d’HLM de Marseille, the Office National des Anciens Combattants et
Victimes de Guerre departmental office in Marseille, and the ADOMA (successor to the
SONACOTRA) in Marseille. Municipal officials and civil servants in Fuveau, la Roque d’Anthéron,
and Rivesaltes shared with me their experiences interacting with the harki population living in their
towns. I particularly wish to recognize la Roque d’Anthéron mayor, Dr. Jean-Louis Turcan; former
social counselor in la Roque d’Anthéron forest hamlet, Raymonde Guillet; Fuveau town councilman
and former GMS soldier, Maklouf Habbaz (who also served as a translator); and especially former
Fuveau town councilman, Michel Tholomier, whose dedication to the harki population at the Cité
Brogilum was touching. Full library privileges at Sciences Po in Paris greatly facilitated my access to
its rich press dossiers, its collection of books and periodicals, and French dissertations from across
the country as well as provided me with a tranquil library workspace. The Génériques Association,
with the permission of André Wormser’s brother, Marcel Wormser, generously allowed me to
consult the archives of the Comité National pour les Musulmans Français just after its archivist,
Quentin Dupuis, finished cataloguing them. I am grateful for the subsidized housing provided by
the Fondation Paul-Albert Février in Aix-en-Provence. My thanks to Amy Tondu at the Fulbright
Commission in Paris, the équipe at NYU in Paris, and the staff and directors at the Camargo
Foundation for their enthusiastic support of my work and administrative help.
Sharing the most delicate, difficult parts of one’s past is challenging. This project would have been
much less meaningful, and much less interesting, without the opportunity to meet members of the
harki population in Fuveau, Jouques, Marseille, Paris, Perpignan, Rivesaltes, and la Roque
d’Anthéron. I profusely thank all those—harkis, harki wives, harki children, and the harkette I met in
Marseille on a sunny June afternoon—who shared their past and present with me and entrusted me
with their histories. I am particularly grateful to Aïcha, the first member of the harki population
whom I met in November 2006, her husband André, her brother Bouziane, her mother Ferhat
xii
whose melodious voice and humor enlightened our interviews, and her father Mohammed who
served ten years in the French Army during the wars of decolonization.
Some magnificent friends made France feel like home. David Bonnaffé and Yann Scioldo-Zürcher
gave me a home base in Paris and welcomed me into their family to experience my first Noël français.
Patrick, Pascale, and “le petit” Rémi Bosch housed me during my research in Perpignan and spoiled
me each time I returned to visit them. Françoise, Elliott, and Gaspard Murphy provided me with a
home on the rue Beauregard since my first year of graduate school. Dinners chez Scott Gunther and
Enrique Avila were a lovely way to spend many Sunday evenings in Paris during my first year of
research. Yamina Boudellal was helpful each time I returned to Aix-en-Provence since first studying
abroad there in 1995. The story of her father, a former FLN leader in France who, unbeknownst to
his family, made it a mission at the end of his life to visit hospitalized harkis with few visitors
poignantly contradicted the myth that all Algerians believe that the harkis were “traitors.” Paul and
the late Lucy Schwartz generously lent me their apartment in Aix multiple times during the last
phase of my dissertation research. The camaraderie of the “glamoisie”—Chris Dietz, Eddie Kolla,
and Libby Murphy—at Camargo (et après) was a beautiful gift of laughter, friendship, and hours in
the library capped off with Mediterranean dips.
Graduate school is an odyssey. During this time some wonderful people laughed with me in the silly
moments, walked with me during the difficult times, and always believed in me. Alan Baehr, Amy
Blanch, Keren Baltzer, Connie Blackburn, Amanda Dalola, Wendy Moran, Leslie Ray, Erin
Sorenson, Russell Spinney, Barbara Thompson, Aurélie Van de Wiele, and Amy Wendholt: your
love and support played an invaluable role in helping me to accomplish things great and small. For
this and your many other gifts, I thank you de tout cœur.
xiii
In appreciation of the love and support of my parents,
Doug and Joyce Miller, who made it with me to the end of this journey.
Dedicated to the memory of my “big sister” Jill E. Hungerford,
whose words and example continue to inspire me each day.
xiv
The historian’s task is not to disrupt for the sake of it,
but it is to tell what is almost always an uncomfortable story
and explain why the discomfort is part of the truth
we need to live well and live properly.1
~Tony Judt
Donald A. Yerxa, “Postwar: An Interview with Tony Judt,” Historically Speaking VII no. 3 (Jan.-Feb. 2006): 20.
President Bouteflika’s comment came in response to President Chirac’s question about granting visas to visit Algeria to
former harkis who were French citizens. Algerian officials have routinely denied anyone who (or is suspected to have
served) as a harki during the Algerian War for Independence entry into his former homeland. As will be discussed below,
during the Algerian War for Independence the term “harki” referred to a specific type of native Algerian auxiliary soldier
in the French Army. At the end of the war the Algerian and French governments applied the label to any native Algerian
who had fought for France or was perceived to have been “pro-French.” Throughout this dissertation, I use “Algerian
War for Independence,” a term that situates the war in a transnational space between France and Algeria. Historian
Sylvie Thénault argues that “Algerian War for Independence” is a marriage of the French appellation “Algerian War”
(emphasizing the French government’s fight to keep its sovereignty over Algeria) and two Algerian designations, “War
1
1
xv
INTRODUCTION
Harki Citizens: Between Algeria and France,
Between the Colonial and Postcolonial
Je crois que les conditions [ne] sont pas encore
venues pour des visites de harkis [en Algérie]… C’est
exactement comme si on demandait à un Français de
la Résistance de toucher la main à un collabo…
- President Abdelaziz Bouteflika, Paris,
June 16, 2000
In a live television interview with French President Jacques Chirac during his first state visit
to France, Algerian President Abdelaziz Bouteflika pronounced these words about harkis—native
Algerians who had fought for France during the 1954-1962 Algerian War for Independence,
approximately 100,000 of whom subsequently migrated to France.1 Bouteflika’s choice of the label
“collabos”—in other words, traitors complicit with a murderous regime—sparked an unprecedented
questioning by the harki population about its past and present relationships to France. After all,
many of these Algerian natives had supported the French cause (responsible for the death of four
hundred thousand Algerians) not because of their allegiance to France, but rather owing to economic
necessity or (sometimes forced) recruitment by the French Army.2 Could the harkis’ service to
President Bouteflika’s comment came in response to President Chirac’s question about granting visas to visit Algeria to
former harkis who were French citizens. Algerian officials have routinely denied anyone who (or is suspected to have
served) as a harki during the Algerian War for Independence entry into his former homeland. As will be discussed below,
during the Algerian War for Independence the term “harki” referred to a specific type of native Algerian auxiliary soldier
in the French Army. At the end of the war the Algerian and French governments applied the label to any native Algerian
who had fought for France or was perceived to have been “pro-French.” Throughout this dissertation, I use “Algerian
War for Independence,” a term that situates the war in a transnational space between France and Algeria. Historian
Sylvie Thénault argues that “Algerian War for Independence” is a marriage of the French appellation “Algerian War”
(emphasizing the French government’s fight to keep its sovereignty over Algeria) and two Algerian designations, “War
of Liberation” (the resurrection of an Algerian nation stifled by the French colonizer) and “Algerian Revolution” (a
radical transformation of the country and its society). Sylvie Thénault, Histoire de la guerre d’indépendance algérienne (Paris:
Flammarion, 2005), 14.
2 The figure of four hundred thousand Algerian deaths is cited in: Kamel Kateb, Européens, “indigènes” et juifs en Algérie
(1830-1962). Représentations et réalités des populations (Paris: Éditions de l’Institut national d’études démographiques, 2001),
359.
1
1
France during the Algerian War for Independence be reduced to “collaborationism”? Considering
this service, when Algeria became independent why did harkis lose their French nationality and need
to formally request French nationality? Why had the state not yet publicly recognized the harkis’
contributions to France during the war and their subsequent personal sacrifices, including fleeing
their homeland fearing for their lives and being placed by the government in camps exiled from the
general population in France? Why had Chirac and other French presidents not defended the
community against claims of treason from Algerians on both sides of the Mediterranean? Most
importantly, why have successive post-imperial French governments, reminiscent at times of those
in colonial Algeria, persistently treated the harki population as though they are not full French
citizens?
Collabos was powerful epithet in a French nation that was confronting what historian Richard
Golsan has termed “Vichy’s afterlife” and had only just begun to publicly contend with the plural
memories of the Algerian War for Independence.3 Five years earlier, on July 16, 1995, Chirac had
symbolically chosen to dedicate his first public allocution after his election as Fifth Republic
president to acknowledging the role that the French State had played in the persecution of Jews and
victims of the Third Reich’s persecution of minorities and genocidal policies. On the fifty-third
anniversary of the Vél d’Hiv roundup, during which over 13,000 Jews—including 3,500 children—
In 1998, following a court case that had received much national and international media attention, former Vichy official
Maurice Papon was convicted for crimes against humanity for his role in deporting 1,500 Jews from Bordeaux to
Auschwitz during the Second World War. During the proceedings, prosecutors presented evidence concerning Papon’s
role as prefect of the Paris police during the October 17, 1961 peaceful anti-war demonstrations by thirty thousand
Algerians, during which as many as two hundred people were killed and over ten thousand arrested. Richard Golsan
argues that incorporating this information about the Algerian War for Independence into a trial about the Second World
War distorted and conflated the history of two distinct moments. Golsan, Vichy’s Afterlife: History and Counterhistory in
Postwar France (Lincoln: University of Nebraska Press, 2000), 161. On the effects of plural memories of Vichy France on
French society, see: Julie Fette, “The Apology Moment: Vichy Memories in 1990s France,” in Taking Wrongs Seriously:
Apologies and Reconciliation, ed. Elazar Barkan and Alexander Karn (Stanford, CA: Stanford University Press, 2006), 265310; Golsan, Vichy’s Afterlife; Tony Judt, “Epilogue. From the House of the Dead: An Essay on Modern European
Memory,” in Postwar: A History of Europe Since 1945 (New York: Penguin Books, 2005), 813-31; Tony Judt, “À la
recherche du temps perdu: France and Its Pasts,” in Reappraisals: Reflections on the Forgotten Twentieth Century (New York:
Penguin Press, 2008), 196-218; Henry Rousso, La hantise du passé: entretien avec Philippe Petit (Paris: Textuel, 1998); and
Henry Rousso, Le syndrome de Vichy: De 1944 à nos jours (Paris: Seuil, 1987). The events of October 17, 1961 will be
treated on p. 117.
3
2
had been arrested and held in an indoor cycling arena before being deported to extermination camps,
Chirac admitted the French state’s complicity: “[L]a folie criminelle de l’occupant a été secondée par
des Français, par l’État français… Pour toutes ces personnes arrêtées, commence alors le long et
douloureux voyage vers l’enfer. Combien d’entre elles ne reverront jamais leur foyer? Et combien, à
cet instant, se sont senties trahies?”4
Immediately following Chirac’s speech, ten children of former harki soldiers from the
Marseille region traveled to Paris to meet with the new president. They asked him to recognize the
role of the French government in the “deportation” of harki soldiers at the hands of members of the
National Liberation Front (FLN) independence movement. In a press release, the harki children
referenced the unique colonial relationship between France and Algeria, whereby the latter was
incorporated into the former’s system of departments. They contended: “[M]aintenant que le
président de la République a rendu responsable l’Etat français dans la déportation des juifs, nous
espérons qu’il définira la même responsabilité de la France dans la déportation des Français
musulmans de leur département d’Algérie.”5 The harki children’s comparison of Jewish French
citizens and Muslim Algerian French citizens was not entirely accurate since the latter group never
achieved full French citizenship, with rights equal to those of the settler population in Algeria. Nor
was their claim of the “deportation” of harkis and their family members correct. It was instead a
forced migration of individuals who feared for their lives owing to violent acts perpetrated against
them because they had fought for France. Nevertheless, the harki children’s efforts to link two
events did touch on a sensitive topic at a time when plural memories of the so-called guerre sans nom
had begun to enter public consciousness and Vichy France was deeply entrenched in public debates.
On July 16, 1995 during a ceremony in Paris commemorating the Vél d’Hiv roundup President Chirac delivered this
speech, which is reprinted in: Jacques Chirac, Mon combat pour la France: textes et interventions, 1995-2007 (Paris: Odile Jacob,
2007), 31.
5 “Des enfants de harkis demandent à Chirac de reconnaître leurs droits,” Libération, July 20, 1995. Right-tilting Le Figaro
and centrist Le Monde did not cover this story, which shone an arguably negative light on the new center-right president
for failing to address the responsibility of the 1962 French government in the deaths of harkis by other Algerians.
4
3
The ensuing uproar in the harki community was not entirely due to these scathing remarks
by the Algerian president. Despite the equation of French citizens who had been harkis with those
who had collaborated with the Nazi regime during France’s “Black Years,” Chirac took four weeks
to publicly denounce Bouteflika’s remarks. As French citizens since they had arrived in France
during the 1960s, numerous harki population members claimed that the silence of their own
president was even more injurious.6
Chirac’s delay during summer 2000 in defending the harki community—French citizens—
after Bouteflika’s words reawakened its sense of abandonment and marginalization from French
society as a result of both imperial and post-imperial government policies. The harki population
believed its abandonment began in winter 1962 when the French Army disarmed harki soldiers,
leaving them to fend for themselves against FLN insurgents who believed that the “Algerian
Revolution” (the FLN appellation for the war) did not end with the March 18, 1962 Evian Accords.
Despite consistent threats of retaliation from FLN members who believed the harkis were traitors to
their native Algeria, this peace treaty negotiated by representatives of the French government
contained no specific mention of the harkis. French government treatment of harkis as second-class
citizens was not limited to the imperial era, during which this practice aligned with their French
citizenship status, inferior to that of Algerian residents of European and Jewish origins. The harkis
who migrated to post-imperial France felt forsaken because almost all harki repatriates—like other
native Algerians—needed to undergo a thorny administrative process established by President
Charles de Gaulle to regain their French nationality, thereby questioning their legal relationship with
In his recent ethnographic study, Vincent Crapanzano writes about the general response to this incident of the harkis,
their wives, and their descendants whom he interviewed: “The Harkis were furious at the Algerian president and, if one
can compare levels of fury, even more so at the French president for not having immediately responded to Bouteflika’s
insult.” Vincent Crapanzano, The Harkis: The Wound that Never Heals (Chicago, IL: The University of Chicago Press,
2011), 172. Several of the seventeen members of the harki community whom I interviewed brought up this event. In
each case the interviewee spoke of the pain and anger he or she felt owing to President Chirac’s failure to defend the
harkis.
6
4
the country for which they had risked their lives.7 Approximately half of the harkis who arrived in
France found housing on their own, through the aid of French soldiers who had fought in Algeria or
through familial or social networks. However, the other half, who relied on the French government
for assistance, were placed in camps—many for months, some for over a decade—which isolated
them from other French citizens. Finally, many in the harki population felt symbolically abandoned
by each successive president because none had officially recognized, on the one hand, the French
government’s failure during the chaos that ensued after the ceasefire to protect and rescue the tens
of thousands harkis massacred by FLN members and, on the other, the harkis’ contributions to
French war efforts.8
Harki Citizens: A Contradiction
The paradoxical relationship between the harki population and post-imperial French
governments has foremost evoked questions about its treatment as French citizens. Successive
government policies and discourses concerning the harki population—bearing resemblance to those
relative to native populations in colonial Algeria who possessed an inferior citizenship status and
foreign immigrants in metropolitan France—have repeatedly marginalized its members from French
society. Nearly all harkis and their family members who migrated to France obtained French
nationality, after which they legally possessed equal rights to other French citizens. Yet in practice
the government did not always treat the harki population as full French citizens. They were—and
are—“harki citizens.” My dissertation seeks to unravel this contradiction. Why did the government
Journal Officiel de la République Française (hereafter JORF), “Ordonnance n˚ 62-825 du 21 juillet 1962 relative à certaines
dispositions concernant la nationalité française, prises en application de la loi n˚ 62-421 du 13 avril 1962,” July 22, 1962,
7230.
8 Precise figures of the number of harkis tortured and massacred remain the object of what historian Charles-Robert
Ageron calls a bataille de chiffres. This debate can never be resolved since neither the French nor the Algerian governments
kept such records. Estimates of the number of deaths range from tens of thousands by Ageron to 150,000 by harki
association leaders, such as Abd-El-Aziz Méliani. This topic will be addressed later in this chapter and in the
introduction to chapter 2.
7
5
not consider harki citizens like its other French citizens? How did the French government
marginalize harki citizens from French society? What are the long-term consequences for the harki
population and contemporary French society of this treatment—which began as soon as they
arrived in 1962 on French soil and, more significantly, continued for decades?
Decisions that the imperial French government made in the waning months of French
Algeria about whether and how to include the harkis and their family members into French society
differentiated them from settlers of European and Jewish descent living in Algeria, commonly
known as the pieds-noirs, who retained their French nationality.9 The choices to not include the
harkis into the repatriation process established for the pieds-noirs and not bestow upon them
automatic French nationality set a precedent which post-imperial governments would follow for
harki citizens. As Todd Shepard has contended, at this time “race and ethnicity appeared as
meaningful markers” to determine who could be considered French (the one million settlers and a
handful of Algerian Muslims who had renounced Sharia law) and who would not be (most of the
nine million other Algerian Muslims who held an inferior citizenship status).10 Almost all harkis fell
into this latter colonial category of native Algerians. Ethnicity—and more precisely, colonial
citizenship status as French with local civil status (Français de statut civil de droit local) predicated on
their Arab and Berber roots—indeed determined whether the harkis were granted automatic French
nationality when they arrived in France. However, ethnicity alone cannot account for the postimperial French government continuing to treat the harkis as inferior citizens for decades.
I argue that a complex interplay of three factors explains the French government’s
marginalization of its harki citizens from society. First, government officials held onto a vision of the
Throughout this dissertation to describe inhabitants of European and Jewish descent living in Algeria I use the term
“pied-noir.” This generic descriptor is a cultural term that references their past or present residence in Algeria. It does
not designate their legal relationship with the French state, which before Algerian independence was “Français d’Algérie”
or “Français de statut civil de droit commun” and after arriving in France was “rapatriés d’Algérie” or “rapatriés français.” When
referring to their legal status, I use one of the expressions listed in the previous sentence.
10 Todd Shepard, The Invention of Decolonization: The Algerian War and the Remaking of France (Ithaca, NY: Cornell University
Press, 2005), 2-3.
9
6
harkis as inferior citizens and even foreigners (a legal status signifying the absence of French
nationality) owing to their former colonial status as French Muslims from Algeria (Français musulmans
d’Algérie or FMA). Consequently, state agents created and implemented policies that prolonged
practices used in colonial Algeria of differentiating populations based on ethnicity, despite the end of
the empire, and those used in metropolitan France of differentiating citizens from foreigners, despite
the harkis’ French nationality. Second, government officials challenged harki citizens’ belonging in
post-imperial French society by not ascribing to them a clear de facto status that matched their de
jure status as French citizens and repatriates. Instead, officials’ interactions with members of the
harki population sometimes implied that they were Algerians, refugees, or immigrants. Since the
government did not consistently treat them as French citizens and repatriates, it placed the
population in a power dynamic whereby harki citizens had to fight for the rights associated with
these categorizations. 11 This played out notably through the responsibility for managing issues
concerning harki citizens continually being shifted between various ministries, most of which usually
handled immigrant affairs. Finally, harki citizens were inextricably linked with the Algerian War for
Independence in a nation whose government sought to immediately forget—and continued to
suppress the memory of—the war. Even the recent fiftieth anniversary of the ceasefire, in March
2012, was not officially celebrated in France. Harki citizens on French soil were visible reminders of
the end of France’s nineteenth and twentieth-century colonial empire, whose memorialization the
French government is still grappling with.
Once harkis migrated to French soil and even after they became French citizens in a country
did not legally have different classes of French citizenship (as existed in French Algeria), French
government policies continued to differentiate them from the pieds-noirs and other French citizens.
As Gérard Noiriel maintains about the politics of naming, “les enjeux de nomination sont des enjeux de pouvoir.”
Gérard Noiriel, “Introduction: De l’histoire sociale du politique à la socio-histoire des relations de pouvoir,” in Etat,
nation et immigration. Vers une histoire du pouvoir (Paris: Belin, 2001), 10.
11
7
First, whereas the government housed French repatriates requesting its assistance upon their
migration to France in emergency shelters and requisitioned hotels, it hid harki repatriates from view
in former refugee and prisoner camps located in sparsely populated corners of France.12 In total,
nearly half of the 100,000 harkis and their family members repatriated to France resided, at least
briefly, in two transit camps (camps de transit) open between June and October 1962 and two housing
camps (camps d’hébergement) open from September 1962 to December 1964 under military
surveillance.13 Indeed, the government treated them like refugees and not the repatriates that they
were. Through a series of circulars concerning more permanent lodging, the government mandated
that French repatriates had priority access over harki citizens to integrated public housing buildings
(HLM).14 To rehouse the harkis after the camps closed, the state provided two arrival centers (cités
d’accueil) for those it deemed “unhouseables,” two thousand residences in seventy-two forest hamlets
(hameaux de forestage) in rural France, and two thousand HLM apartments in cités urbaines.15 All of
these options—originally intended to be transitional spaces—contained exclusively harki families
except for the social counselors and camp directors, most of whom had supervised FMAs in the
colony and metropole, charged with their oversight. This impeded harki citizens’ social interactions
with other French citizens. The government subsequently confirmed its exclusion of harki citizens
who remained under state supervision from French society by enacting policies that made these
For an analysis of the French repatriates’ emergency housing situation, see: Yann Scioldo-Zürcher, Devenir métropolitain:
Politique d’intégration et parcours de rapatriés d’Algérie en métropole (1954-2005) (Paris: Éditions EHESS, 2010), 197-215.
13 The transit camps comprised Larzac (Aveyron) and Bourg-Lastic (Puy-de-Dôme) and the housing camps included
Rivesaltes (Pyrénées-Orientales) and Saint-Maurice-l’Ardoise (Gard).
14 See, for example, Archives du Comité National pour les Musulmans Français (hereafter ACNMF), 15/13, Circular
from the Ministry of Repatriates to Departmental Prefects and Regional Delegates of the Ministry of Repatriates,
“Objet: Attribution de logements H.L.M. aux anciens harkis,” 1, Nº 64/19, Jan. 31, 1964.
15 The two arrival centers were Bias (Lot-et-Garonne) and Saint-Maurice-l’Ardoise (Gard). By July 1965, the state had
built 2,043 residences in forest hamlets. Archives Nationales (hereafter AN), 5 AG 1/22 (Archives of President Charles
de Gaulle), Minister of the Interior, “Rapport sur l’application de la loi du 26 décembre 1961 transmis au Secrétaire
général de la Présidence de la République,” 119, July 26, 1965. After this time, it constructed an additional five additional
forest hamlets containing approximately two hundred residences. At the end of 1966 the government housed 2,089
families in cités urbaines (also called ensembles immobiliers) built under a special project, “Le Programme ‘Harki.’” Archives
Départementales des Bouches-du-Rhône (hereafter ADBR), 1451 W 115, ONASEC, “Français Musulmans Rapatriés
Information. 1962-1981: La Politique d’Accueil,” 3, Apr. 18, 1985.
12
8
temporary housing solutions into permanent ones. The last so-called arrival center remained open
until 2000, twenty forest hamlets still existed in 1980, and several cités urbaines continue to house
harki families today, though now mixed with foreign migrants. Meanwhile, the last boarding house
for the pieds-noirs, one located in Paris and lodging approximately twenty individuals, had
disappeared in 1979. Finally, the laws the French government passed before and after Algerian
independence for all repatriates to provide social welfare assistance, furnish indemnities for lost
property and goods in Algeria, and commemorate their sacrifices marginalized harki citizens in one
way or another. In some cases, they encountered difficulties accessing the rights outlined by the
legislation.16 In the recent February 23, 2005 law, under political pressure from the forceful pied-noir
lobby, commemoration of this repatriate community overshadowed that of harki repatriates, much
like the pieds-noirs’ interests obscured those of the harkis in 1962.17
To dissect the complicated relationship between the post-imperial French state and its harki
citizens this dissertation analyzes the creation and implementation of the French government’s
nationality, repatriation, housing, integration, and commemoration policies concerning the harki
population. My examination begins with exchanges during summer 1961 between harki soldiers and
Army officials concerning the former’s postwar relationship to France in terms of nationality and
residence. It concludes with the passage of the February 23, 2005 legislation that lawmakers initially
conceived to recognize and compensate former harki soldiers who became French citizens for their
contributions to France during the war and the failures of successive French governments to foster
harki citizens’ economic and social integration. To investigate the implementation of national
policies at the local level, this dissertation incorporates two case studies that focus on a subsection of
JORF, “Loi n°61-1439 du 26 décembre 1961 relative à l’accueil et à la réinstallation des Français d’outre-mer,” Dec. 28,
1961, 11996-7 and JORF, “Loi nº 70-632 du 15 juillet 1970 relative à une contribution nationale à l’indemnisation des
Français dépossédés de biens situés dans un territoire placé antérieurement sous la souveraineté, le protectorat ou la
tutelle de l’Etat,” July 17, 1970, 6651.
17 JORF, “Loi n° 2005-158 du 23 février 2005 portant reconnaissance de la Nation et contribution nationale en faveur
des Français rapatriés,” Feb. 24, 2005, 3128.
16
9
the harki population exiled in camps under state supervision. The Rivesaltes camp outside
Perpignan—previously used for refugees, soldiers, and prisoners—housed approximately one-fifth
of the harki population repatriated to France and is the site where approximately 7,500 harkis and
their wives became French citizens.18 The Fuveau forest hamlet—a microcosm of the Rivesaltes
camp, which was constructed for fifty-four harki families—was a transitional solution that
government officials maintained would help these newly-minted citizens integrate into French
society. This forest hamlet, located thirty-five kilometers northeast of Marseille, remains open today,
though is no longer run by the state. While there is no homogeneous “harki experience,” the camp
situations are consequential for each harki citizen in France. Even though these spaces, labeled
“camps” and “ghettos” in numerous press articles, housed less than 10 percent of the harki
population in 1975 they became the symbol of all harki citizens’ marginalization by the French
government. 19 Moreover, the post-imperial government’s most significant policies affecting the
entire population (enacted in 1975, 1991, and 2005) resulted from protests that took place in the
camps and/or representations of these spaces used in public demonstrations to sway popular
opinion.
My study differs from the majority of works on colonial and postcolonial immigration in two
significant ways. First, recent scholarship investigating colonial-like practices in metropolitan France
during and after the colonial period focuses on immigrant populations chiefly, if not entirely,
composed foreigners. 20 In contrast, this dissertation analyzes marginalizing treatment for a
Centre des archives contemporaines (hereafter CAC), 19920149/5/12, Chef du 13º Bureau bis de la Sous-Direction
des Naturalisations (Ministry of Public Health and Population), “Etat numérique des déclarations enregistrées pendant la
période du 1er juillet au 31 juillet 1964,” Aug. 10, 1964.
19 In a report submitted on Jan. 1, 1972, sociologist Jean Servier estimated that the total harki population comprised
162,000 individuals, 16,000 of which were “regroupés” in the arrival centers, forest hamlets, and cités urbaines. ACNMF, 6,
Servier, “Enquête sur les Français Musulmans,” Part 1, 1 and Part 2, 1.
20 For books published on this topic, see: Mary Dewhurst Lewis, The Boundaries of the Republic: Migrant Rights and the Limits
of Universalism in France, 1918-1940 (Stanford, CA: Stanford University Press, 2007) and Clifford Rosenberg, Policing Paris:
The Origins of Modern Immigration Control Between the Wars (Ithaca, NY: Cornell University Press, 2006). For dissertations,
see: Françoise de Barros, “L’État au prisme des municipalités. Une comparaison historique des catégorisations des
18
10
population of French citizens. Second, the bulk of studies on French immigration policy focus on
either the national level (primarily those published during the 1990s) or the local one (those
published since the beginning of the twentieth-first century).21 Notable exceptions include Clifford
Rosenberg’s monograph about immigration control in the interwar period, that by Mary Dewhurst
Lewis concerning the rights of migrant workers during the same timeframe, and Yann ScioldoZürcher’s work on government policies toward French repatriates from Algeria since the war.
Similarly, my dissertation investigates within the same analytical field both the national-level creation
of state policies toward the harkis and their local implementation. Examining the interrelation
between the local and national allows me to determine their reciprocal influence and examine the
points of tension and consensus between national and local officials.
Ethnic Soldiering and the Harkis
Throughout France’s imperial history in North America and the Caribbean, encompassing
Napoleonic expansion in Europe, and in Africa and Asia its army incorporated native soldiers.
During the nineteenth and twentieth-century empire, the history of colonial natives fighting with
French forces began during France’s conquest of Algeria with the 1830 formation of the Zouaves
military corps, which owes its name to the Kabyle tribe the Zouaouas.22 In 1842 the government
created an additional corps of soldiers, known as tirailleurs algériens, to recruit natives to aid in quelling
étrangers en France (1919-1984),” (PhD diss., Université de Paris 1, 2004); Melissa Byrnes, “French Like Us? Municipal
Policies and North African Migrants in the Parisian Banlieues, 1945-1975” (PhD diss., Georgetown University, 2008);
Choukri Hmed, “Loger les étrangers ‘isolés’ en France. Socio-histoire d’une institution d’État: la Sonacotra (1956-2006)” (PhD diss.,
Université de Paris 1, 2006); and Amelia Lyons, “Invisible Immigrants: Algerian Families and the French Welfare State in
the Era of Decolonization (1947-1974)” (PhD diss., University of California, Irvine, 2004).
21 For national-level studies, see: Ralph Schor, Histoire de l’immigration en France de la fin du XIXe siècle à nos jours (Paris:
Armand Colin, 1996); Vincent Viet, La France immigrée. Construction d’une politique (Paris: Fayard, 1998); and Patrick Weil,
La France et ses étrangers: L’aventure d’une politique de l’immigration de 1938 à nos jours (Paris: Calmann-Lévy, 2005). For locallevel ones, see: Marie-Claude Blanc-Chaléard, Les Italiens dans l’Est parisien (années 1880-1960) (Rome: Ecole française de
Rome, 2000); Emmanuel Blanchard, La police parisienne et les Algériens (1944-1962) (Paris: Nouveau Monde éditions, 2011);
Byrnes, “French Like Us?;” and Lyons, “Invisible Immigrants.”
22 For information about the Zouaves, see: Henri duc d’Orléans Aumale, Les zouaves et les chasseurs à pied: esquisses historiques
(Paris: M. Lévy frères, 1855) and a special issue of the journal Carnet de la Sabretache, “Les Zouaves” no. 80 (1985).
11
resistance of tribal leaders toward the increasing French dominance in Algeria given the small
population of European settlers at this time (37,374 in 1841).23 As France extended its dominance
southward in Africa, in 1857, under the auspices of Emperor Napoleon III, Governor Louis
Faidherbe of Senegal created a corps of tirailleurs sénégalais to help facilitate the French Empire’s
conquest of Africa. These Sub-Saharan conscripts—who hailed from all regions of French West
Africa (AOF) and French Equatorial Africa (AEF)—were soon expanded to include soldiers from
all over greater France. More generally known as troupes indigènes, colonial subjects subsequently
became a foundation of the French Army during non-colonial conquests, including the 1870-71
Franco-Prussian War and both World Wars, and the 1946-54 French-Indochinese War.24 After these
conflicts, the former soldiers (other than Indochinese natives who fought in their homeland, many
of whom fled south of the seventeenth parallel) returned from foreign theaters to their countries and
villages, which were still under French colonial rule and protection.25 Yet, the “return home” of
native Algerian soldiers—notably the “harkis”—after the end of the Algerian War for Independence
was different.
Who were the harkis? Today “harkis” is synonymous with traitors among Algerians and
connotes all native Algerian repatriates to the French, which are distortions of the term’s original
John Ruedy, Modern Algeria: The Origins and Development of a Nation (Bloomington: Indiana University Press, 1992), 69.
On the progressive colonization of Algeria, see pages 51-79. Following the establishment of the tirailleurs algériens, the
Zouaves subsequently developed into regiments pulled from Algeria’s European and Jewish populations.
24 On colonial soldiers fighting in the French Army from the Franco-Prussian War through the wars of decolonization,
see: Myron Echenberg, Colonial Conscripts: The Tirailleurs Sénégalais in French West Africa, 1857-1960 (Portsmouth, NH:
Heinemann, 1991); Richard S. Fogarty, Race and War in France: Colonial Subjects in the French Army, 1914-1918 (Baltimore,
MD: Johns Hopkins University Press, 2008); Nancy Ellen Lawler, Soldiers of Misfortune: Ivoirien Tirailleurs of World War II
(Athens: Ohio University Press, 1992); Joe Lunn, Memories of a Maelstrom: A Senegalese Oral History of the First World War
(Portsmouth, NH: Heinemann, 1999); and Gregory Mann, Native Sons: West African Veterans and France in the Twentieth
Century (Durham, NC: Duke University Press, 2006).
25 In 1953, 108,000 Indochinese, approximately half of whom were auxiliary soldiers, served in the French Army. The
number of Indochinese soldiers massacred following the war is unknown. Pierre Messmer, Les Blancs s’en vont: Récits de
décolonisation (Paris: Albin Michel, 1998), 49. Much scholarly work remains to be accomplished on Indochinese auxiliary
soldiers during and after the French-Indochinese War. The few existing studies include: Gérard Brett, Les supplétifs
d’Indochine, 1951-1953 (Paris: L’Harmattan, 1996); Gérard Brett, La tragédie des supplétifs. La fin des combats 1953-1954 (Paris:
L’Harmattan, 1998); and Trinh Van Thao, “Le retour des rapatriés d’Indochine. L’expérience des Centres d’accueil
(1954-1960),” in Marseille et le choc des décolonisations, ed. Jean-Jacques Jordi and Emile Temime (Aix-en-Provence: Edisud,
1996), 29-38.
23
12
meaning. During the war, harkis were native Algerian soldiers in harkas, a derivative of the Arabic
word for movement, haraka, which aptly represented their function of patrolling the nearby
countryside for fellaghas or supporters of the FLN. The harkis composed the greatest number of the
five types of Muslim civilian and military auxiliary forces, which also included moghaznis, groupes
mobiles de sécurité, groupes d’auto défense, and aassès. Each formation performed tasks to aid French
soldiers or French civilian administrators in their war efforts and protect rural populations from
attacks by National Liberation Army (ALN) troops.26 The French Army recruited auxiliary forces to
provide a source of manpower to reinforce French troops, receive aid from individuals with
knowledge of the countryside and native languages, and demonstrate that native Algerians supported
its efforts to maintain sovereignty over Algeria. 27 As General Maurice Challe, who became
commander of French forces in Algeria in December 1958 stated, “Nous ne pacifierons pas l’Algérie
sans les Algériens.” Accordingly, as demonstrated in table 1, the number of auxiliary forces increased
dramatically between January 1959 and January 1960.
Harkis
Groupes d’autodéfense
Moghaznis
Groupes mobiles de sécurité
Aassès
Auxiliary Forces (Total)
1/1957
2,186
10,000
15,000
4,800
-31,986
1/1958 1/1959
16,902
28,021
12,000
30,000
16,000
17,191
6,100
6,100
--51,002
84,312
1/1960 11/1960
57,800
61,400
50,000
62,000
19,100
19,100
7,500
7,500
7,600
2,300
142,000
152,300
7/1961 3/1962
59,000
42,100
62,000
17,000
19,100
18,300
8,000
8,500
3,060
1,100
151,160 87,000
Table 1: Number of FMA auxiliary soldiers serving in the French Army (January 1957-March 1962). The aassès only
existed between 1960 and early 1962. Source: Maurice Faivre, Les combattants musulmans de la guerre d’Algérie: Des soldats
sacrifiés (Paris: L’Harmattan, 1995), 252.
The first harkas were created in January 1955 in the Aurès Mountains. Their status was made official in April 1956 by
Governor General Robert Lacoste, who defined their mission as “formations temporaires dont la mission est de
participer aux opérations de maintien de l’ordre.” Groups of twenty to twenty-five moghaznis protected the sections
administratives sociales (SAS), a corps of French doctors, nurses, educators, and social workers charged with “pacifying”
native populations and promoting French Algeria. The groupes mobiles de sécurité were rural police forces and groupes d’auto
défense were tasked with protecting rural villages. Finally the aassès were Muslim auxiliary policemen in unités territoriales,
which protected Algerian infrastructure and communication lines. For a more detailed description of each auxiliary
force’s role see, Charles-Robert Ageron, “Les supplétifs algériens dans l’armée française pendant la guerre d’Algérie,”
Vingtième siècle. Revue d’histoire, no. 48 (December 1995): 6-8 and François-Xavier Hautreux, “L’engagement des harkis
(1954-1962). Essai de périodisation,” Vingtième siècle. Revue d’histoire 90 (June 2006): 33-37.
27 Hautreux, “L’engagement des harkis,” 33-34.
26
13
Within these ranks the French Army considered—and treated—the harkis as second-class
auxiliary troops, with auxiliary troops already being inferior to FMA soldiers enlisted in the French
Army. Unlike other formations, the harkis’ contracts had to be renewed each day. The harkis had
fewer opportunities for military promotions, received less technologically advanced weapons, and
had access to fewer social benefits.28 This inferior relationship with other auxiliary forces (known as
soldats irréguliers) and FMAs in the French Army (called soldats réguliers) would become significant in
the postcolonial context. In part owing to the harkis’ daily contracts, the aggregate number of FMAs
who served as auxiliary troops during the war is difficult to determine. The only definite statistics
concern the number of individuals who served at precise moments. Table 1 shows these figures for
FMA auxiliary soldiers in the French Army from January 1957 to the March 1962 ceasefire.
Despite claims from pro-French Algeria former Army officers who asserted, primarily based
on their personal experiences, that Muslims joined the French cause owing to their allegiance to
France, scholarship by Tom Charbit demonstrates that only a small minority of Algerian natives
joined the auxiliary forces for this reason.29 The need for jobs in the face of the war-ravaged agrarian
Algerian economy, forced recruitment by French soldiers (enrôlement) with orders from the
government to promote the appearance that Algerians were opposed to independence, and the quest
for weapons to protect their families against FLN violence proved more salient factors in the
decision of Algerian natives to fight for France.30 In fact, some auxiliary soldiers and policemen
fought alternately for France and the FLN over the course of the war. Therefore, the beginning of
Ageron, “Les supplétifs algériens,” 6-8.
Tom Charbit summarizes this point: “Si l’engagement dans le camp français peut s’apparenter à un choix de nature
politique ou idéologique, celui-ci ne concerne donc qu’une minorité de ‘musulmans’…” Charbit, Les harkis (Paris: La
Découverte, 2006), 30-31. Charbit’s positions contrast with arguments from former Army generals who, based on their
personal experiences, maintain that Algerian natives joined auxiliary forces primarily owing to their allegiance to France.
See, for example, Faivre, Les Combattants musulmans, and François Meyer, Pour l’honneur… avec les harkis: De 1958 à nos jours
(Paris: Éditions CLD, 2005).
30 Mohand Hamoumou discusses these reasons for the harkis’ enlistment in Et ils sont devenus harkis (Paris: Fayard, 1993),
129-211.
28
29
14
many of their relationships with imperial France was fraught with contradictions. So, too, was the
end.
Algeria, France, and Harki Citizens
The difficult relationship between the post-imperial French government and its harki citizens
in France was generated by not only the Algerian War for Independence but also the nature of
France’s colonial relationship with Algeria. Therefore, examining the situation of harki citizens in
post-imperial France first requires looking at elements of the history of French Algeria. Since the
1997 publication of Frederick Cooper and Ann Laura Stoler’s edited volume, Tensions of Empire:
Colonial Cultures in a Bourgeois World, colonial scholarship has focused on placing within the same
historical and analytical field the colony and the metropole. In their introduction, the editors
advocate an approach that scrutinizes the dialectical relationship between European nation-states
and their colonies, arguing: “Europe was made by its imperial projects as much as colonial
encounters were shaped by conflicts within Europe itself.”31 Much like the history of such colonial
situations, the harki population’s history took place in a transnational space between (former) colony
and (former) metropole and indelibly marked populations, cultures, and institutions on both sides of
the Mediterranean.32 To place my study into the context of the reciprocal influence between Algeria
and France, I examine the specificity of colonial Algeria, which was simultaneously an exceptional
“colony” in relation to France’s other possessions and exemplary of methods of governance across
France’s empire. In particular, I focus on the categorization of Algerian inhabitants, the shifting legal
relationships between these categories and the French government, and the practices of
Frederick Cooper and Ann Laura Stoler, “Between Metropole and Colony: Rethinking a Research Agenda,” in Tensions
of Empire: Colonial Culture in a Bourgeois World, ed. Cooper and Stoler (Berkeley: University of California Press, 1997), 1.
32 Studies examining specific colonial situations and demonstrating that imperial governance affected France in addition
to the colonies include: Laurent Dubois, A Colony of Citizens: Revolution and Slave Emancipation in the French Caribbean, 17871804 (Chapel Hill: University of North Carolina Press, 2004); Gregory Mann, Native Sons: West African Veterans and France
in the Twentieth Century (Durham, NC: Duke University Press, 2006); and Emmanuelle Saada, Les enfants de la colonie: Les
métis de l’Empire français entre sujétion et citoyenneté (Paris: La Découverte, 2007).
31
15
administering colonial populations. These demonstrate a divide between legal situations and
practices.
Algeria was foremost an exception in comparison with France’s other possessions during
what scholars have widely-termed the “second empire” (1830 to 1962), coinciding with the
territory’s colonization and decolonization.33 As France’s premier settler colony, Algeria’s sizable
population of settlers influenced strategies and decisions of how the French government, and by
extension colonial administrators, ruled the territory and its peoples. As a result, Algeria’s legal
relationship with France different from that with all of France’s other possessions during the second
empire. In 1848 following the collapse of the July Monarchy, the Second Republic Constitution
declared Algeria to be an integral part of France. A decree later that year incorporated the regions
around Algiers, Constantine, and Oran into the French system of departments. Therefore, unlike
France’s settlements such as Senegal and Mali that were colonies (under its political control, but not
ruled by metropolitan laws) and those such as Morocco and Tunisia that were protectorates (with
which France maintained a relationship of partial control and protection), Algeria was to be
administered by French laws in “metropolitan fashion.”34 But in practice this would not happen:
Algeria was a de facto colony.
Efforts by French regimes to integrate Algeria into France foremost engendered unique—
and shifting—legal relationships between residents of Algeria and the French government, distinct
from those in other colonial holdings. Following an 1865 decree (sénatus-consulte), all residents of
Algeria held French nationality, which colonial subjects of France’s other possessions only attained
A recent collection of essays edited by Kate Marsh and Nicola Frith seeks to challenge the extant division in French
historiography of France’s imperial project into two discrete periods (seventeenth-eighteenth centuries and nineteenthtwentieth centuries). The goal of the project is to highlight how the perceptions of territorial loss sustained continuities
between the first and second empires by informing subsequent manifestations of French imperialism. Marsh, “Territorial
Loss and the Construction of French Colonial Identities, 1763-1962,” in France’s lost empires: fragmentation, nostalgia, and la
fracture colonial, ed. Frith and Marsh (Lanham, MD: Lexington Books, 2011), 2.
34 Charles-Robert Ageron, Modern Algeria: A History from 1830 to the Present, ed. Michael Brett, trans. Michael Brett
(Trenton, NJ: African World Press, 1991), 28-29.
33
16
in 1946 through the Fourth Republic Constitution. Moreover, besides a limited number of
inhabitants of four Senegalese communes, Algeria was the only colony where natives had the
possibility for full French citizenship, theoretically equal to that of metropolitan residents.35
Although all Algerians were French citizens, the definition of (male) citizen in Algeria broke
with that in metropolitan French definition since, in colonial Algeria, French nationality did not
always ensure French citizenship.36 Citizenship in France, in its most restricted definition, denotes
political participation or voting rights, though the term can also be interpreted to include economic
and social rights such as access to social welfare and civil service employment.37 While Algeria was
purportedly an extension of metropolitan France, its native inhabitants—referred to as natives, or
indigènes, in legal documents through the end of the Second World War—were not full French
citizens unless they underwent a procedure to renounce Muslim Sharia law. Algerian natives’
classification as French with local civil status (Français de statut civil de droit local) under an 1865 decree
set them apart from the population of colonists from France and other European countries, the
French with common civil status (Français de statut civil de droit commun), who had access to the same
rights as metropolitan citizens. (The 1870 Crémieux decree subsequently included Jews residing in
Algeria into the category of French with common civil status without the requirement of renouncing
their faith.) Even though the 1865 legislation bestowed French nationality upon Algerian natives,
unless they abandoned their Koranic civil status they held an inferior form of nationality depriving
them of all citizenship rights until a 1919 law allowed certain natives limited voting rights and a
The French government placed substantial restrictions on access to full French citizenship in Senegal. The French
government only bestowed these rights on évolués—educated, acculturated inhabitants of the Four Communes (Dakar,
Gorée, Rufisque, and Saint-Louis) who renounced their personal status as Muslims. For a discussion of this unique
citizenship status, see: Catherine Coquery-Vidrovitch, “Nationalité et citoyenneté en Afrique occidentale française:
originaires et citoyens dans le Sénégal colonial,” The Journal of African History, 42 (2001): 285-305.
36 The government did not grant suffrage rights to women in France until 1944.
37 The definition of citizenship in France is often unclear because the words “citoyenneté” and “citoyen” have not
appeared in any French legal text since the 1789 Declaration of the Rights of Man and Citizen. For a discussion about
defining citizenship, see: Fred Constant, La citoyenneté (Paris: Montchrestien, 2000); Danièle Lochak, “La citoyenneté: un
concept juridique flou,” in Citoyenneté et nationalité: perspectives en France et au Québec, ed. Dominique Colas et al. (Paris:
Presses Universitaires Françaises, 1991): 179-207; and Pierre Rosanvallon, Le sacre du citoyen: Histoire du suffrage universel en
France (Paris: Gallimard, 1992).
35
17
series of legislation from 1947 onward progressively opened the door to increased rights for all
Algerian natives.38 However, these expanded rights did not amount to a fully equal citizenship
because Algeria’s legislature comprised a two-college system with unequal representation, which was
later abolished in 1958 with the objective of integrating Muslim populations.39 Nevertheless, this
legal inequality and treatment of “native” populations during the colonial era in Algeria influenced
government and popular perceptions of Algerians in France during and after the colonial era,
including the harkis.40
At the same time Algeria was an exemplary colonial situation. There was not one French
colonial situation, but rather many colonial situations with similar characteristics that varied across
time and space. Though Algerian natives technically were citizens of France (they had French
nationality), their inferior French citizenship resembled the limited rights of imperial citizens in
France’s colonies and protectorates. Further, many of the colonial methods of governance that
originated in Algeria—such as the Native Code (Code de l’indigénat)—were transmitted, in modified
forms, to other imperial holdings. The framework for administering Algerian natives, whose
implementation varied within different regions, vacillated between inclusion and exclusion. This was
the case for populations in other colonies, whether ruled by France or other imperial powers.
Frederick Cooper argues that one of the central facets of imperial governance of nineteenth and
On the distinction between nationality and citizenship in colonial Algeria, the evolution of Algerian natives’ rights in
Algeria, and citizenship rights of FMAs in metropolitan France, see: Emmanuel Blanchard, La police parisienne, 19-49;
Laure Blévis, “La citoyenneté française au miroir de la colonisation: Étude des demandes de naturalisation des ‘sujets
français’ en Algérie coloniale,” Genèses, no. 53 (Dec. 2003); Shepard, The Invention of Decolonization, 19-54; Alexis Spire,
“Semblables et pourtant différents. La citoyenneté paradoxale des ‘Français Musulmans d’Algérie’ en métropole,” Genèses,
no. 53 (Dec. 2003): 48-68; and Patrick Weil, Qu’est-ce qu’un Français? Histoire de la Nationalité française depuis la Révolution
(Paris: Grasset, 2002), 225-44.
39 In 1944, the first college comprised non-Muslims and approximately sixty thousand naturalized Muslims (those who
had undergone the process outlined in the 1865 decree to become French with common civil status and évolués)
representing one million Algerian inhabitants. The second college comprised 1.2 million voters—Muslim men aged at
least twenty-one years old with local civil status—and represented the balance of the population. Weil, Qu’est-ce qu’un
Français?, 243.
40 Works that examine the governance of Algerian natives residing in France during and after the colonial era include:
Blanchard, La police parisienne; Tom Charbit and Françoise de Barros (ed.), “La colonie rapatriée,” special issue of Politix
19 no. 76 (2006); and Lyons, “Invisible Immigrants.”
38
18
twentieth century was negotiating “a balance between poles of incorporation (the empire’s claim that
its subjects belonged within the empire) and differentiation (the empire’s claim that different
subjects should be governed differently).”41 The laws and administration of Algerians reflected
France’s continual desire to differentiate European and Jewish versus Arab and Berber populations.
This treatment, along with that in other colonies, has led scholars to underscore that republican
France’s claims of universalism were indeed bounded by inherent unequal treatment of its citizens
and nationals in the colony and the metropole.42
Despite certain similarities between Algeria and France’s other nineteenth and twentieth
century colonial possessions, this close relationship made the “decolonization” struggle and
postcolonial relationship unique. The difference between Algeria and France’s other colonial
possessions resided not only in the fact that the former claimed its independence following a war.43
France’s war strategies (including the use of torture), the presence of between 300,000 and 450,000
so-called FMAs in metropolitan France during the war, and the violence that occurred in the months
following the March 1962 ceasefire, also caused repercussions for populations and institutions that
continue to reverberate on both sides of the Mediterranean today.44 By posing, in a 1997 article, the
provocative question of whether France and Algeria constitute one nation or two, Etienne Balibar
Frederick Cooper, Colonialism in Question. Theory, Knowledge, History (Berkeley: University of California Press, 2005), 154.
For such studies, see Jennifer Boittin, Colonial Metropolis: The Urban Grounds of Anti-Imperialism and Feminism in Interwar
Paris (Lincoln: Nebraska University Press, 2010); Alice Conklin, A Mission to Civilize: The Republican Idea of Empire in France
and West Africa, 1895-1930 (Stanford, CA: Stanford University Press, 1997); Lewis, The Boundaries of the Republic;
Rosenberg, Policing Paris; and Gary Wilder, The French Imperial Nation-State: Negritude and Colonial Humanism between the Two
World Wars (Chicago, IL: University of Chicago Press, 2005).
43 The only other war of decolonization took place in Indochina from 1946 to 1954, yet this war had less impact on
populations and institutions because Indochina—neither a settlers’ colony nor departments of France—was not
incorporated as intimately into France as Algeria. On the relationship between France and Indochina, see Kathryn
Robson and Jennifer Yee, eds., France and “Indochina”: Cultural Representations (Lanham, MD: Lexington Books, 2005).
44 The number of FMAs is taken from statistics kept by the North Africa division of the Etat-Major de l’Armée de Terre.
Service historique de l’Armée de Terre (hereafter SHAT), 10T 549/3a, EMAT, 2e bureau, Chart: “Évolution de la
population algérienne de 1945 à 1964.” For historical scholarship on these difficult legacies of the Algerian War for
Independence, see Raphaëlle Branche, La torture et l’armée pendant la guerre d’Algérie (Paris: Gallimard, 2001) and Sylvie
Thénault, Une drôle de justice: Les magistrats dans la guerre d’Algérie (Paris: La Découverte, 2001).
41
42
19
drew attention to the unique postcolonial rapport between France and Algeria owing to their
intertwined histories and populations.45
Recent historical studies have examined the immediate and long-term effects of the breakup
of this unique colonial relationship. Yann Scioldo-Zürcher recently published an analysis of French
government policies toward French repatriates from Algeria from 1954 to 2005. He demonstrates
that contrary to representations of repatriates from Algeria in national memory, which tend to single
their traumatic arrival in France to the point of obscuring subsequent state initiatives, the French
state indeed created (and most often followed through on) an integration policy for the population
upon their arrival in France. The French government’s treatment of the pieds-noirs contrasts with
that of the harkis. Scioldo-Zürcher analyzes both the creation of policies and their implementation at
the local level, similar to my study.46 While limiting his analysis to 1962, Todd Shepard focuses on
the impact of the end of French Algeria, which he refers to as a “revolutionary moment,” on Fifth
Republic government institutions, laws, and definitions of who was French. 47 He reveals how
French bureaucrats, politicians, and journalists at this time rewrote the history of French Algeria so
that decolonization was the predetermined endpoint, which allowed the post-1962 state to exclude
Algeria and Algerians from France and French history. My study seeks to contribute to this literature
by analyzing the actions of the post-1962 state toward harki citizens in France, thereby
demonstrating precisely how French Algeria exerted a long-term influence on citizenship in postimperial France.
Etienne Balibar, “Algérie, France: Une ou deux nations?,” Lignes 30 (February 1997): 7-21. Balibar asserts that the
France-Algeria relationship is unique, but not singular as the same question could be posed about England and the
United States as well as Germany and Austria (9).
46 Scioldo-Zürcher, Devenir métropolitain. Valérie Esclangon-Morin also published a work investigating state policies
toward repatriates of European descent after they arrived in France. However, her study examines repatriates from
across North Africa—Tunisia and Morocco in addition to Algeria. Esclangon-Morin, Les rapatriés d’Afrique du nord de
1956 à nos jours (Paris: L’Harmattan, 2007).
47 Shepard, The Invention of Decolonization, 12.
45
20
The Colonial, the Postcolonial, and Harki Citizens
The “uncomfortable story,” in the words of Tony Judt, of harki citizens’ paradoxical
marginalization from French society requires reexamining the “fracture coloniale” between the
colonial period in Algeria and the effects of its colonial situations on harki citizens in postcolonial
France.48 In their volume, La fracture coloniale: La société française au prisme de l’héritage colonial, editors
Pascal Blanchard, Nicolas Bancel, and Sandrine Lemaire encourage scholars to rethink the division
between the colonial and postcolonial periods. They explain “la fracture coloniale” as “un concept
[qui] voudrait à la fois signifier la tension et les effets de la postcolonialité: il recouvre des réalités
multiples et des situations hétérogènes, dans la mesure où ces réalités multiples et des situations
peuvent être éclairées, en partie, par des processus de longue durée, reliés à la situation coloniale.”49
Various scholars have critiqued this approach and I share some of their critiques. For
example, Emmanuelle Saada has argued that when applied to racism in contemporary society,
effacing the “colonial fracture” gives too much weight to colonial situations, to the point of negating
the longue durée of racist thought in France.50 By foremost insisting on continuities in structures, this
approach first overlooks the agency of individuals. Neither colonized populations (whether in the
colony or metropole) nor postcolonial migrants to France were powerless. Examples of
independence movements, violent and non-violent protests, as well as more subtle forms of
resistance that subverted domination proliferate in the colonial and postcolonial contexts.51 The
See the epigraph to this dissertation.
Pascal Blanchard, Nicolas Bancel, and Sandrine Lemaire, ed., La fracture coloniale: La société française au prisme de l’héritage
colonial (Paris: La Découverte, 2005), 13.
50 Emmanuelle Saada, “Un racisme de l’expansion. Les discriminations raciales au regard des situations coloniales,” in De
la question sociale à la question raciale? Représenter la société française, ed. Didier Fassin and Eric Fassin, 2nd ed. (Paris: La
Découverte, 2009), 63-79.
51 For a theoretical discussion of strategies of resistance, see: Ania Loomba, Colonialism/Postcolonialism (New York:
Routledge, 1998), 173–254. On independence movements, see: John Ruedy, “Chérif Benhabylès and Ferhat Abbas: Case
Studies in the Contradictions of the Mission civilisatrice,” Historical Reflections 28.2 (Summer 2002): 185-202 and Benjamin
Stora, Messali Hadj (Paris: Hachette, 2004). On protests in Algeria and France during the colonial period, see: Jim House
and Neil MacMaster, Paris 1961: Algerians, State Terror, and Memory (Oxford: Oxford University Press, 2006) and JeanPierre Peyroulou, Guelma, 1945: Une subversion française dans l’Algérie coloniale (Paris: La Découverte, 2009). On protests in
France undertaken by immigrants from its former colonies and their descendants, see: Saïd Bouamama, Dix ans de marche
48
49
21
focus on structures moreover fails to recognize the autonomy of administrators, which led to
discrepancies in policy implementation. When examining the relationship between the state and its
population, one must conceptualize the state as not only collective political actor, but also individual
representatives responsible for putting policies into practice. In the context of my study, by
classifying harkis as “French,” “Algerian,” “immigrant,” “refugees,” or “repatriates,” to define their
relationship with the state, this body—both the collectivity and its representatives who carry out
policies—dictates who has which rights and duties associated with French citizenship. Such
categorizations are particularly important in France given the centralized state, represented by a large
corps of civil servants, which uses its power to intervene in many aspects of its citizens’ and
nationals’ everyday life.
Despite such critiques, certain elements of Blanchard, Bancel, and Lemaire’s approach are
useful for my study. By reexamining the “fracture coloniale” for the postcolonial experiences of
harki citizens in France, three legacies from France’s 132-year rule in Algeria become particularly
salient for their history. First, I note the similarities between the de facto and de jure categorizations
of native populations in Algeria and those of the harki population after the March 1962 ceasefire.
During the colonial period, difficulties continually arose in defining the association between, on the
one hand, the French government and the European settlers and, on the other, native Algerians.
The attempts to legally differentiate these two categories of inhabitants resulted in new lexicons
seeking to clarify their relationships with the French government. In actuality, the appellations that
the government chose essentialized the heterogeneous population of Berbers and Arabs and
emphasized differences between the two sets of citizens in Algeria. For example, Français musulmans
d’Algérie (chosen in opposition to Français d’Algérie) highlighted the different religion—one often
viewed as inferior—of almost all Algerian natives. Similarly, in the post-imperial context, “harkis”
des Beurs: Chronique d’un mouvement avorté (Paris: Desclée de Brouwer, 1994) and Fadela Amara and Mohammed Abdi, La
racaille de la République (Paris: Seuil, 2006).
22
arose out of the state’s creation of a social category that expanded the prior definition of harkis as
one of the five types of auxiliary soldiers during the Algerian War for Independence.52 At the end of
the war, the French government mapped the term “harki” onto any native Algerian who had fought
for France or was perceived to have been “pro-French” (such as domestic servants for pied-noir
families) and whose life was consequently threatened by possible FLN retributions.53 The definition
of an already heterogeneous group (geographically, ethnically, and those who fought for France
because they wanted Algeria to remain French versus those who enlisted to save their families’ lives
from violence and starvation) expanded for those who crossed the Mediterranean. The government
lumped together all women and children who fled Algeria with their husbands and fathers under the
generic term “harki.” Whereas men had chosen to enlist in the French Army, for the host of reasons
listed above, this redefinition placed their wives, children, and descendants into a stigmatized
category associated with “traitors.”
This state categorization, moreover, presented a double bind for the harki population. It
simultaneously allowed them access to specific benefits from the state and placed them in an
unofficially inferior citizenship status. This inferiority in part derived from the post-1962
government’s de facto classification of harki citizens as refugees and immigrants, and not their de
jure status as repatriates. The term rapatriés in government discourse almost always referred
exclusively (and implicitly) to French repatriates. When government officials referred to harkis as
repatriates, they qualified this term with “French Muslim” to distinguish them from “French”
repatriates. Rapatriés Français musulmans was a reference to their former imperial status and not to
their different religion, which in the post-imperial context not only was perceived as inferior, but
unlike in Algeria (where Islam was the most practiced religion) highlighted their belonging to a
On the difference between social groups and social categories as well as the power dynamic that such classifications
create, see Gérard Noiriel, “Représentation nationale et catégories sociales. L’exemple des réfugiés politiques,” in Etat,
nation et immigration. Vers une histoire du pouvoir (Paris: Belin, 2001), 265-66.
53 The Algerian government equally used this term in a broader sense.
52
23
minority. Government classification as a “harki” meant placement into refugee-like conditions in
camps upon their arrival. Furthermore, the French government rehoused 25 percent of harki
citizens from the Rivesaltes and Saint-Maurice-l’Ardoise camps in forest hamlets, smaller scale
versions of these camps. 54 To address what officials termed “the harki problem”—the failed
integration of harki citizens—the government used many of the same social welfare offices and
agents as it did for immigrants, yet the latter were not citizens.
The second legacy of French Algeria relevant to my study of harki citizens in France is
exclusionary practices and discourses. While noting that some of these were not limited to colonial
situations, I analyze continuities between particular methods of governance used in French Algeria
and those for harki citizens in post-imperial France.55 The existence of these continuities in imperial
and post-imperial methods of governance, therefore, necessitates nuancing the idea of 1962 as an
abrupt rupture in the history of French government, as Frederick Cooper argues in a 2005 essay.
Cooper contends that France did not become a nation-state until 1962 because up to this moment it
was an “empire-state.” To demonstrate his point, Cooper offers the example of President de
Gaulle’s decision in 1946 to create a French Union, uniting France and all its colonies. Cooper
argues that because “[Algeria’s] population was not treated equally… France, in 1946, was not a
nation-state, but an empire-state.”56 An abrupt shift from an “empire-state” to a “nation-state” in
1962, as Cooper asserts, minimizes the effects of France’s (former) colonies on the state after this
time and does not account for the French government’s unequal treatment of harki citizens in postimperial France. My analysis of this treatment reveals that continuities in practices, discourses,
decision makers’ mentalities, and political actors themselves existed alongside the political ruptures
ascribed to the end of the imperial era. While the overseas empire is no longer geographically part of
AN, 5 AG 1/22, Minister of the Interior, “Rapport sur l’application…,” 112.
While some of these practices were present in France’s other colonial holdings, in this study, I primarily focus on
those in Algeria.
56 Cooper, Colonialism in Question, 153.
54
55
24
the French nation-state, remnants of colonial divisions continue to mark the post-1962 state’s
creation of policies for harki citizens and the policies’ implementation. Though the harki
community—originally one hundred thousand repatriates and estimated in a 2003 government
report at five-hundred thousand people—comprises a small percentage of French citizens, this
treatment is nonetheless significant owing to the importance of social groups at the margins of
society for shaping national identity.57
Finally, legacies of France’s defeat in the Algerian War for Independence influenced the
French government’s treatment of harki citizens. This war—which France won militarily, but lost
politically—signaled the end of its entire nineteenth and twentieth-century empire. The conquest,
moreover, came on the tail end of twenty-two years of humbling combat (notably the “strange
defeat” during the Second World War and the fall of Diên Biên Phu to native forces ending the
Indochinese War), making the loss of Algeria more glaring.58 Harki citizens were closely associated
with the Algerian War for Independence (as “les témoins gênants de la guerre d’Algérie,” according
to Benjamin Stora)59 in a nation whose government worked to immediately forget, and continued to
suppress, the event out of which this social category was created.60 Sociologist Ann Stoler argues
that the term “aphasia”—instead of the commonly-used “forgetting” or “amnesia”—better
characterizes the absence of colonial history as objects of scholarship until recently. She defines
aphasia in this context as “both loss of access and active dissociation.”61 This term is apropos for the
French government’s treatment of the war after Algerian independence—an active forgetting.
See Maurice Agulhon, “Le centre et la périphérie,” in Les lieux de mémoire, III. Les France, ed. Pierre Nora (Paris:
Gallimard, 1993), 2903 and Peter Sahlins, Boundaries: The Making of France and Spain in the Pyrenees (Berkeley: University of
California Press, 1989), 8-9.
58 See Marc Bloch, L’étrange défaite (Paris: Éditions Franc-Tireur, 1946).
59 Benjamin Stora, La gangrène et l’oubli: La mémoire de la guerre d’Algérie, 2nd ed. (Paris: La Découverte, 1998), 207.
60 On the suppression of the memory of the Algerian War for Independence, see Raphaëlle Branche, La Guerre d’Algérie:
une histoire apaisée? (Paris: Seuil, 2005); William Cohen, “The Algerian War, the French State and Official Memory,”
Historical Reflections 28, no. 2 (2002): 219-39; Jo McCormack, Collective Memory: France and the Algerian War (1954-1962)
(Lanham, MD: Lexington Books, 2007); and Stora, La gangrène et l’oubli.
61 Ann Laura Stoler, “Colonial Aphasia: Race and Disabled Histories in France,” Public Culture 23, no. 1 (2011): 125.
57
25
Nowhere was this aphasia more visible than in how French government officials characterized the
war for nearly half a century. Using “les opérations effectués en Afrique du Nord” and other
euphemisms in official documents and discourse, they belied the reality of the seven and a half year
bloody decolonization struggle as a “war,” until an October 1999 law replaced these vague terms
with “la guerre d’Algérie.”62
Writing about the Harki Population: From Actors to Observers
Writings about the harki population have been marked by three trends. First, the authors of
most existing books are not scholars. Indeed, the first books written by scholars did not appear until
the early 1990s. This turning point can be attributed to both the events leading up to the 1991
revolts by harki children, which awakened in many a self-identification as harki citizens and public
questions about their occluded history, and expiration of the thirty-year waiting period for military
archives in 1992, which gave scholars access to documents regarding the Algerian War for
Independence. Second, social scientists—and not historians—have published most of the scholarly
work. And finally, almost all scholarship is authored by individuals with links to the harki
community. Only recently have historians and outsiders to the population begun to produce work
on the harkis, which is where my dissertation is situated.
Former military commanders, soldiers, and politicians were the first to publish books about
the harki population. These texts appeared soon after the end of the Algerian War for Independence
and recounted the authors’ experiences interacting with harki soldiers during and after the war.63
While there has been a fairly steady stream of such publications over the last five decades, two
JORF, “Loi nº 99-882 du 18 octobre 1999 relative à la substitution, à l’expression ‘aux opérations effectués en Afrique
du Nord’, de l’expression ‘à la guerre d’Algérie ou aux combats en Tunisie ou au Maroc,” Oct. 20, 1999, 15647. Other
euphemisms present in government discourse and documents encompassed “les opérations du maintien de l’ordre” and
“les événements en Algérie.”
63 For example, Bachaga Boualam, Les harkis au service de la France (Paris: Éditions France-Empire, 1963); General
Georges Fleury, Harkis: les combattants du mauvais choix (Paris: Éditions Bellamy et Marlet, 1976); and Eric Taleb, La fin des
harkis (Paris: Éditions La Pensée universelle, 1972).
62
26
events served as catalysts for an upswing in volume. First, those linked to the harki population
sought to honor France’s forgotten harki citizens following the summer 1991 revolts by
predominantly male harki children, which highlighted their difficult history. 64 Second, President
Bouteflika’s equation of harkis with collabos in June 2000 and President Chirac’s subsequent delay in
defending France’s harki citizens sparked five members of the harki population to publish texts over
the next three years.65 In particular, within the span of six months in 2003, the first memoirs ever
written by harki daughters emerged into the public literary space. Four women whose families had
resided in refugees camps after the migration of their families to France and who had never met
before—Fatima Besnaci-Lancou, Hadjila Kemoum, Dalila Kerchouche, and Zahia Rahmani—broke
the silence surrounding their families and population’s taboo histories with firsthand testimonies.
This act was of particular significance since previously voices of harki children had taken the form of
violent protests and hunger strikes from their brothers, which served to point toward a homogenous
“harki experience.” By vividly and emotionally recounting their unique experiences, these authors
put forth completely new perspectives on the Algerian War for Independence and its ensuing
ramifications and forced the public to rethink “Who are the harkis?”
The first scholars to study the harki population were anthropologists, sociologists,
psychologists, and political scientists, some of who had attachments to the harki community. These
notably include a 1976 dissertation by sociologist Anne Heinis, who worked for the French Muslim
service, and compiled her research from ethnographic data commissioned by the French
See, for example, Maurice Faivre, Un village de harkis, des Babors au pays drouais (Paris: L’Harmattan, 1994) and Abd-elAziz Méliani, La France honteuse: le drame des harkis (Paris: Perrin, 1993). Other texts published in the mid-1990s were
authored by scholars and will be discussed below.
65 These texts comprise: Boussad Azni, Harkis: crime d’Etat. Généalogie d’un abandon (Paris: Ramsay, 2001); Fatima BesnaciLancou, Fille de harki: Le bouleversant témoignage d’une enfant de la guerre d’Algérie (Paris: Éditions de l’Atelier, 2003); Hadjila
Kemoum, Mohand le harki (Paris: Anne Carrière, 2003); Dalila Kerchouche, Mon père, ce Harki (Paris: Seuil, 2003); and
Zahia Rahmani, Moze (Paris: Sabine Wespieser, 2003). Since this time Besnaci-Lancou, Kerchouche, and Rahmani have
published additional books and Kerchouche has co-directed a made for television movie.
64
27
government and the Comité National pour les Musulmans Français (CNMF).66 Since this time, social
scientists continue to publish more books than historians. Their studies tend to evaluate identity
construction and the integration of harki citizens, and more specifically, harki children. In his 1999
Le Silence des harkis, sociologist Laurent Muller analyzed the harki population’s difficult integration
based on his experiences in the late 1980s working with harki children in Alsace during his military
service.67 Between 2000 and 2008 political scientist Stéphanie Abrial and sociologists Mohamed Kara
and Régis Pierret published monographs examining the difficulties encountered by second and third
generation harkis integrating into French society.68 Political scientist Tom Charbit offered a brief
overview of the harkis’ history from the war through 2005 in a volume of La Découverte’s “Repères”
series.69 The most recent text, published in 2011, is American anthropologist Vincent Crapazano’s
ethnographic study tracing the identity construction of the harkis and their children. Through an
analysis of the expressed words and emotions of his interviewees, Crapanzano argues that the role of
absent fathers, the history of the Algerian War for Independence, and the population’s treatment by
the French government have influenced harki children’s place in French society.70
To date, only three historians have published books focused on the harki population. Michel
Roux’s text, which summarized other secondary sources tracing the history of the harki population
since their arrival in France, was removed from the market for plagiarism shortly after its 1991
publication.71 In 1999 pied-noir historian Jean-Jacques Jordi co-authored with sociologist Mohand
Hamoumou an overview of the harkis’ history from their participation in the war to their
Anne Heinis, “L’Insertion des Français Musulmans: Étude faite sur les populations regroupées dans le Midi de la
France et les centres de ex-harkis” (Doctorat de Spécialité 3° cycle, Université Paul Valéry de Montpellier, 1977).
67 Laurent Muller, Le silence des harkis (Paris: L’Harmattan, 1999).
68 Stéphanie Abrial, Les enfants de Harkis: de la révolte à l’intégration (Paris: L’Harmattan, 2001); Mohamed Kara, Les
Tentations du repli communautaire: Le cas des Franco-Maghrébins et des enfants de Harkis en particulier (Paris: L’Harmattan, 2000);
and Régis Pierret, Les Filles et fils de harkis: Entre double rejet et triple appartenance (Paris: L’Harmattan, 2008).
69 Charbit, Les harkis.
70 Crapanzano, The Harkis. Another anthropologist’s study merits mentioning. Guilia Fabbiano’s dissertation analyzes the
complexities of harki identity through the writings of harki children. Giulia Fabbiano, “Des générations post algériennes.
Discours, Pratiques, recompositions identitaires” (PhD diss., Università degli studi di Siena/EHESS, 2006).
71 Michel Roux, Les harkis ou les oubliés de l’histoire (Paris: La Découverte, 1991).
66
28
repatriation to their living conditions in France. The authors strongly critiqued the French
government for having forgotten the harki population and speak of the various camps as the harkis’
“lieux de mémoire” in France. The text is largely based on oral interviews and press articles, though
also uses a small number of archival documents from the Bouches-du-Rhône departmental
archives.72 Historian Abderahmen Moumen’s published Masters thesis is the only study of the harkis
chiefly based on archival sources. Moumen focuses on the harki population residing in the Vaucluse
department and concludes that those who lived in forest hamlets or in neighborhoods with
concentrations of harki families were less likely to integrate into French society.73
Until the late 1990s most studies continued to be written by members of the population and
those associated with the population. These individuals, outside the French Academy, often sought
to positively portray their interest group (whether harkis or the French Army) and their conclusions
highlighted their personal involvement. Two examples are emblematic. In 1993 harki son Mohand
Hamoumou published a study, based on his dissertation, that defined who constitutes “French
Muslim Repatriates” (a name used by government officials for the harkis) and aimed to debunk the
myth that harkis had joined the French Army owing to their allegiance to France. His work primarily
drew from interviews with former Muslim auxiliary soldiers in the Clermont-Ferrand region. The
polemical concluding words of Hamoumou’s book represented a challenge to the French
government and not scientific conclusions based on his research: “Reste donc à la France, pour son
honneur, d’admettre officiellement sa responsabilité dans l’abandon – et donc le massacre – des
harkis et d’autres milliers de ‘musulmans français’ qui lui avaient fait confiance, afin que l’oubli ne les
assassine pas une seconde fois. Afin aussi que des vérités trop longtemps tues n’empoisonnent
Jean-Jacques Jordi and Mohand Hamoumou, Les harkis, une mémoire enfouie (Paris: Autrement, 1999).
Abderahmen Moumen, Les Français musulmans en Vaucluse 1962-1991. Installation et difficultés d’intégration d’une communauté
de rapatriés d’Algérie (Paris: L’Harmattan, 2003).
72
73
29
l’avenir.”74 The second example is a book about auxiliary soldiers during the war published in 1995
by Maurice Faivre, an Army general and former harka commander who personally repatriated a
village of harki soldiers and their families to the countryside near Dreux (Eure-et-Loir). Taking
advantage of the newly accessible Army Archives, his text provides useful statistics concerning the
participation of auxiliary forces in the war, but his conclusions seek to absolve the French Army of
claims that it abandoned the harkis in 1962 and reify the memory of the harki community. Faivre
writes: “Les responsables directs du massacre sont les chefs du FLN eux-mêmes…[Ils avaient] une
idéologie révolutionnaire fondée sur le terrorisme, opposée théoriquement au pluralisme…”75 He
goes on to laud the efforts of the French Army to help integrate the harkis in France into society:
“La réussite la plus notable est celle des garnisons françaises d’Allemagne.”76 These conclusions,
notably the latter, stray from the archival documents that his study is based upon.77
Since the early 1990s, due to some authors’ political and emotional investment with the harki
community and/or French Algeria, scholarship about the harkis has been entrenched heated debates
(sometimes including personal attacks) between authors of different political persuasions or points
of view. For example, after Hamoumou termed the post-March 1962 killings a “genocide” in his
monograph, historian Charles-Robert Ageron, who taught history in an Algerian high school when
the war broke out, responded in a 1994 article in Vingtième Siècle: Revue d’histoire. He argued: “Ces
chiffres [des morts] relèvent de la légende noire, au même titre que les accusations de l’auteur
stigmatisant ‘le déshonneur de l’armée abandonnant les siens’… M. Hamoumou est brouillé avec les
chiffres.”78 These debates continue today.
Hamoumou, Et ils sont devenus harkis, 322.
Faivre, Les combattants musulmans, 224-25.
76 Ibid., 235.
77 For a critique of Faivre’s earlier work about the repatriation to France of a village of harkis in Kabylia, see Branche, La
guerre d’Algérie, 59-62.
78 Charles-Robert Ageron, “Le drame des harkis en 1962,” Vingtième siècle. Revue d’histoire, no. 42 (June 1994): 5-6.
74
75
30
In 2005, following the passage of the law to honor the harkis, harki daughter Fatima
Besnaci-Lancou teamed with historian Gilles Manceron to create the Association Harkis et Droits de
l’Homme. The association’s website describes its mission this way: “mettre en œuvre tout travail
d’histoire et de mémoire, sous l’angle des droits de l’homme, pour faire connaître l’histoire des
harkis.”79 Toward this end, Besnaci-Lancou and Manceron have invited scholars to participate in
conferences and submit articles to their edited volumes. However, they advocate a specific version
of the harkis’ history, one indicts the French government for its responsibility in the killings and
torture after the ceasefire. Following the contribution of Algerian War for Independence historian
Sylvie Thénault to their 2008 volume, Besnaci-Lancou and Manceron critiqued Thénault’s article in
which she used a subtle linguistic difference in the title to argue that “des massacres de harkis” and
not “un massacre des harkis” took place across the territory in different regions with varied intensity
at different times.80 That is, there were multiple massacres of some harkis, and not one massacre of
the harki population. In their introduction, titled “En finir avec toutes les légendes,” the editors
write: “Nous remercions Sylvie Thénault d’avoir participé à cette réflexion, mais les directeurs de cet
ouvrage ne cachent pas leur désaccord avec le fait que son texte, à propos des pouvoirs publics
français en 1962, ne reprend pas à son compte les concepts d’abandon et de crime d’Etat, qui
résultent pourtant, à nos yeux, de la simple analyse de faits.”81 Such emotional debates rife with
polemical terms among scholars and members of the harki population over the precise number of
harkis killed, lexicon, and representations of the population have tended to overshadow harki
citizens’ history.
<http://www.harki.net/article.php?rubrique=5>.
My emphasis. Sylvie Thénault, “Massacre des harkis ou massacres de harkis?,” in Les harkis dans la décolonisation et après,
ed. Fatima Besnaci-Lancou and Gilles Manceron (Paris: Éditions de l’Atelier, 2008), 81-91.
81 Fatima Besnaci-Lancou and Gilles Manceron, “Introduction: En finir avec toutes les légendes,” in Les harkis dans la
décolonisation, 23.
79
80
31
Nevertheless, recently historians who are not linked to the harki population have undertaken
studies. Between 2007 and 2009, three doctoral students defended dissertations in France, the
United States, and the United Kingdom that focus entirely on the harkis or compare the harki and
pied-noir communities. François-Xavier Hautreux examined the French Army’s strategies of
recruiting and using auxiliary soldiers during the war.82 Sung Choi analyzed the influence of the
resettlement of pied-noir and harki repatriates on contemporary French national identity and politics
commemorating the colonial era.83 Claire Eldridge focused on similarities, differences, and overlaps
in memory construction in harki and pied-noir communities.84 These three recent dissertations in
history and American anthropologist Crapanzano’s book are important contributions in part owing
to the authors’ distance from their subjects.
A Note on Sources
This study draws on documents gathered from twenty different public and private archives.
The private archival collections include, notably, that of CNMF, whose sustained involvement in
aiding the harki community from 1962 to 2007 generated a wealth of documentation. These
meticulous records encompass reports with precise data about the harki population at the nationallevel and in the Bouches-du-Rhône department and correspondence among high-level government
officials. Catalogued in spring 2008 just before I finished my research in Paris, these documents
were a valuable resource for filling in gaps in materials absent from public archives.
François-Xavier Hautreux, “L’armée française et les supplétifs ‘Français musulmans’ pendant la guerre d’Algérie
(1954-1962): Expérience et enjeux” (PhD diss., Université de Paris X-Nanterre, 2009). See also, Hautreux,
“L’engagement des harkis.”
83 Sung Choi, “From Colonial Citizen to Postcolonial Repatriate: The Politics of National Belonging and the Integration
of the French from Algeria after Decolonization,” (PhD diss., UCLA, 2007). See also, Sung Choi, “The Muslim Veteran
in Postcolonial France: The Politics of the Integration of Harkis After 1962,” French Politics, Culture & Society 29, no. 1
(Spring 2011): 24-45.
84 Claire Eldridge, “The Mobilisation and Transmission of Memory within the Pied-Noir and Harki Communities, 19622007,” (PhD diss., University of St. Andrews, 2009). See also, Claire Eldridge, “‘We’ve never had a voice’: memory
construction and the children of the harkis (1962–1991),” French History 23, no. 1 (2009): 88-107.
82
32
I carried out my work in national government archives at the Centre historique des Archives
Nationales (AN), the Centre des Archives contemporaines (CAC), the Archives du Ministère des
Affaires Etrangères (MAE), and the Service historique de l’Armée de Terre (SHAT), which house
the Ministry of Defense’s archives. Due to the sensitive nature of the material and the recentness of
the history I was investigating, most of these archives required special permission (dérogations) to
access following the provisions of the January 3, 1979 law on archives.85 The challenge in this
process was not being accorded dérogations, as I received them for nearly every dossier. But rather,
the often vague descriptions in archival inventories rendered it difficult to determine which cartons
might contain documents relevant to my study, which I could not verify on my own without first
completing a dérogation request, and the length of time some ministries took to grant me access to
their dossiers (in one case, eighteen months).
The SHAT collections include documents about the harkis’ participation in the Algerian War
for Independence, correspondence among soldiers and commanders about the harkis’ situation in
Algeria, exchanges between officials in Algeria and those in France about repatriation and violence
toward the harkis, and information about the Rivesaltes camp, which was overseen by military
officials. Minister of Armies Pierre Messmer’s personal papers, to which he granted me special
access, include personal correspondence about the harkis’ repatriation and residence in camps in
France. Some of the documents he wrote in 1962 and 1963 or marginalia in documents he received
revealed his resistance toward repatriating harki soldiers given the overcrowding of these camps.
Much of my work in Paris was carried out at the CAC, which houses most ministries’ archives.
Cartons from the Ministries of Agriculture and Social Affairs contained particularly useful
documents concerning the governments’ policies for the forest hamlets. In addition to documents
On this law and the challenges faced by historians in accessing sensitive archival documents, particularly those
concerning Vichy France, see Sébastien Laurent, ed., Archives “secrètes”, secrets d’archives? Historiens et archivistes face aux
archives sensibles (Paris: CNRS Éditions, 2003).
85
33
about the personnel overseeing harki camps, the AN houses President Charles de Gaulle’s papers.
At the time of my research (fall 2006 to spring 2008), de Gaulle’s archives remained under a one
hundred year waiting period. The access that the curator of his archives granted me revealed only
official reports and no personal communication from the President. In fact, I only found two
documents signed by him. To fill in this gap, I relied on Alain Peyrefitte’s C’était de Gaulle, a two
thousand-word tome based on Peyrefitte’s extensive note-taking during Council of Ministers’
meetings and their private conversations. Therefore, the words I cite from de Gaulle come primarily
from this source.
Over the course of my research I interviewed three ministers: Minister for Veterans Affairs
Hamlaoui Mekachera, former Minister of Armies and Prime Minister Pierre Messmer, and former
Minister delegate of Repatriates Roger Romani. Each of these encounters allowed me to learn more
about the individuals responsible for legislation and policies. While I wrote to Minister Messmer for
an interview, Ministers Mekachera and Romani contacted me based on derogation requests I had
made for archives at the CAC. Moreover, while my American nationality and distance from the
harkis’ situation facilitated obtaining these interviews, the men were eager for me to share their
version of controversial events in which they had been involved. Appendix A contains a transcript
of my July 5, 2004 interview with Minister Messmer. I use sparingly these men’s words in sections of
my dissertation that describe events that took place decades ago. However, in my final chapter,
which focuses on the memory, I analyze how they remember the harkis’ history.
To investigate the implementation of government policies toward the harki population at the
Rivesaltes camp and the Fuveau forest hamlet, I began my research in the Pyrénées-Orientales
(ADPO) and the Bouches-du-Rhône departmental archives (ADBR). The ADPO houses very few
documents about the camp and the Rivesaltes municipal archives were equally disappointing.
However, given the wealth of documents at the SHAT, I was able to largely overcome this obstacle.
34
The ADBR contained more useful documents than the ADPO. I soon realized, though, that many
of the offices that interacted regularly with the population in the department, as well as
municipalities with sizeable harki populations, had not deposited their archives at the ADBR.
Therefore, I requested permission from each office to view these documents onsite. At the
departmental branch of the Office national des forêts in Aix-en-Provence, I found personnel files
for harkis who worked at the Fuveau forest hamlet and reports explaining the type of work they
carried out. Salary records allowed me to compile the precise number of harkis who resided in the
forest hamlet. The most glaring absence from the ADBR was that of any papers from the Service
des rapatriés; some of these documents had been destroyed and some were in an attic at the
Marseille Prefecture. Here, I cross-referenced the names of Fuveau harkis with the thousands of
social welfare files and was able to locate twenty-five of their files. The documents that these
dossiers contained gave texture to descriptions of individual harki families’ difficult financial
situation and illustrated their trajectories once they left Algeria. I also sought out personal dossiers at
the departmental office of the Office national des Anciens Combattants et Victimes de Guerre
(ONAC) in Marseille. In addition to information about individuals’ lives in France, these documents
included information about the men’s military service. While I did not use many of the documents
found in these files for the present study, they nonetheless provided a contrast to official
communications that spoke in generalities. For information about rebuilding of the Fuveau forest
hamlet in the mid-1970s, I located documents at the construction and management company office
in Marseille, the Société Anonyme Nouvelle d’HLM de Marseille (SANHLMM).
The municipal archives in Fuveau, where I had hoped to find reports from the forest
hamlet’s social counselor and documents giving a voice to the harki population, contained no
documents dating before the mid-1990s. While some of these other offices allowed me to fill in gaps,
I also interviewed members of the harki population and municipal officials. A town councilwoman
35
whose functions included relations with the camp residents provided me with personal documents
about her work with the population in the height of the early 1990s protests. I interviewed ten
members of Fuveau’s harki population and met several others. Although these interviews did not
figure prominently into my study, they were important to understanding harki citizens’ views, largely
absent from the archives. I was also able to recover the voices of some harki citizens from the
excellent collection of periodical articles in the Sciences Po dossiers de presse. Beginning in the mid1970s newspapers and magazines reproduced press releases and speeches from members of the
harki population who were active in protests. These articles also offer a glimpse into how harki
citizens’ were portrayed to the public.
Overview
In section one (“The Harkis’ Nationality Amidst a Franco-French War”), I argue that the
fractures created as a result of the Franco-French war in the early 1960s between the French
government and the Secret Army Organization (OAS), which fought violently to keep Algeria
French, impacted the harkis’ access to French nationality after Algerian independence. The former
colonial status of nearly all harkis meant that they were not automatically eligible for French
nationality, thereby restricting their access to the physical and metaphorical interior of France’s new
post-imperial geographic boundaries.
Section two (“Harki Citizens’ Exile in France”) relies on two case studies to demonstrate
that the government’s decision to relegate to camps those who succeeded in “returning” to France
set a precedent for long-persisting exclusionary policies. This treatment labeled the harkis not as
French citizens but as “harki citizens.” Already shunned from Algerian communities in France
because they had fought for France during the Algerian War for Independence, their different
ethnicity and the isolated living conditions of those in camps kept them exiled from their
36
surrounding French community. This section also demonstrates the significance of the local camp
situations for policy shifts affecting the entire harki population.
Finally, the conclusion turns to national legislative and memorial initiatives regarding harki
citizens. The commemorative acts to honor France’s harki citizens were influenced by both the
actions of male harki children who sometimes violently protested their plight and France’s evolving
relationship with the memory of its colonial past.
37
SECTION 1
The Harkis’ Nationality Amidst a Franco-French War
…L’intérêt de la France a cessé de se confondre
avec celui des pieds-noirs. Il ne faut pas laisser croire
que l’avenir de la France en Algérie c’est le leur…
Cette affaire d’Algérie aura démontré l’effroyable
infirmité de l’État, encouragée au fil du temps par
toutes sortes d’abandons et de faiblesses. Le tournant
était difficile, les habitudes étaient tellement
invétérées! Maintenant le tournant est pris. Cette
tournée, à tous égards, sera celle du grand tournant.1
-President Charles de Gaulle, Council of
Ministers meeting, May 4, 1962
“En l’an de grâce 1962, fleurit le renouveau de la France. Elle avait été menacée de guerre
civile. La faillite allait l’accabler. Le monde oubliait sa voix. La voici tirée d’affaire.”2 So opens the
second volume of President Charles de Gaulle’s memoirs. These sentiments about France’s internal
conflict during the Algerian War for Independence, written shortly before his death in November
1970 (eighteen months after he stepped down from the French presidency), resonate with the words
cited in the epigraph that he vociferated in a May 4, 1962 Council of Ministers meeting. De Gaulle
wanted to shed France’s former departments that, in his view, had weakened France for too long
and consign to the past the “affaire d’Algérie,” immediately. He wanted to witness the birth of a new
Algeria, one in which France would have only strategic and economic interests. He wanted to end
the violence of the OAS, formed in January 1961 by extremist elements of the pieds-noirs to fight to
keep Algeria French, violence that threatened security and stability in metropolitan France. De
Gaulle took this “grand tournant”—a new resolve to no longer abandon France’s interests by
surrendering to those of the pieds-noirs—to reassert France’s strength. This meant placing full
1
2
Quoted in Alain Peyrefitte, C’était de Gaulle (Paris: Gallimard, 2002), 140.
Charles de Gaulle, Mémoires d’espoir: L’Effort, 1962-…, tome II (Paris: Plon, 1971), 13.
38
attention on extricating the French government from the crippling “Franco-French war” with the
OAS that continued after the March 19, 1962 ceasefire had officially ended its military battle with
the FLN.
The first time that the expression “Franco-French war” appeared in academic discourse was
in American historian Stanley Hoffmann’s 1968 article “Collaborationism in France during World
War II.”3 Here, Hoffmann argues that historians must analyze the French state’s collaboration with
the German state and individuals’ collaborationism with the Nazis—topics hitherto absent from
scholarly treatment—from the viewpoint of Franco-French relations and not from that of FrancoGerman relations.4 He nuances the difference between the Vichy government’s collaboration with
Germany, necessitated by the June 1940 armistice, and its more extreme form, citizens’ willful
collaborationism with the Nazis.5 For him, the “Franco-French war” pitted French men and women
who “openly desired co-operation with and imitation of the Nazi regime,” many of who did not
“know much about what the Nazis were really like” and understood little about what fascism and
national socialism really were, against those opposed to collaboration and/or collaborationism with
the Nazi cause.6 This Franco-French war, which Hoffmann labeled a “civil war” once in his text,
took on an identity of its own, separate from the German Occupation.
Since Hoffmann’s identification of the internal conflict during the Vichy regime as a FrancoFrench war, the designation has been used as a descriptor and an analytical category for events such
as the French Revolution (notably the Terror), the Paris Commune, the Dreyfus affair, and the
Stanley Hoffmann, “Collaborationism in France during World War II,” The Journal of Modern History 40, no. 3 (Sept.
1968): 375-395. The term Franco-French war first appeared in print in 1950 with the publication of La guerre francofrançaise by Maréchal Pétain’s cabinet director, Louis-Dominique Girard. This book defended the Vichy regime’s
controversial policies against Jews. However, the author did not use this phrase to describe the Vichy era, but rather the
civil war in France taking place at that time caused by, on one hand, the repression of Pétain supporters and
collaborationists and, on the other, the Cold War. Girard, La guerre franco-française (Paris: Éditions André Bonne, 1950).
4 Hoffman, “Collaborationism in France,” 376.
5 Ibid.
6 Ibid. The actions by collaborationists—notably, their agency in round-ups and deportations—threatened the lives of
French civilians, particularly Jews and Resistance fighters.
3
39
Algerian War for Independence. In their own ways, all these clashes brought to the surface cleavages
between two camps that fought passionately—and often violently—for their own cause, which
necessarily meant fighting versus fellow French citizens and for or against the French government of
the time. Franco-French wars were offshoots of the primary conflicts taking place. For example, in
the case of the Second World War, France’s initial conflict was with Germany; the ensuing FrancoFrench war pitted collaborators against Resistance fighters. During the Dreyfus affair, the trial of
Jewish Army Captain Alfred Dreyfus for treason gave birth to a battle between Dreyfusards and
Anti-Dreyfusards. The former saw injustice in society if any individual’s human rights were
threatened, while the latter’s holistic conception of society (one unable to be reduced to its
individual members) led it to argue for protecting the French nation from the threats that the
Dreyfus case posed.7
Nearly two decade’s after Hoffmann’s article, French historians Jean-Pierre Azéma, JeanPierre Rioux, and Henry Rousso, in a special issue of Vingtième siècle. Revue d’histoire titled “Les guerres
Franco-Françaises,” underlined the importance of the notion of the French nation-state’s repeated
“fractures” in their analysis of Franco-French wars. In their introduction to the journal, the editors
write that Franco-French wars are ideological conflicts similar to geological faults. During periods
between two fractures, “sediments” (like peace treaties or amnesties) cover the faults, masking
previous internal conflicts. When faced with a new clash that falls along the same fault lines
(whether it be a war, international crisis, or internal difficulties), the newest layer of sediment cracks
and additional, similar fractures appear.8 The internal conflicts that have been present in France
since 1789 repeat themselves in different forms, yet the core issue in these Franco-French wars is
Michel Winock, “Les affaires Dreyfus,” Vingtième siècle. Revue d’histoire, no. 5 (Jan.-Mar. 1985): 23-24.
Jean-Pierre Azéma, Jean-Pierre Rioux, and Henry Rousso, “Les guerres Franco-Françaises,” Vingtième siècle. Revue
d’histoire, no. 5 (Jan.-Mar. 1985): 4.
7
8
40
whether to include within or exclude from the nation certain French citizens, often based on
conflicting interpretations of defining the nation.
By early 1962 a Franco-French war had emerged from the fracture between, one on hand,
French government officials who increasingly wanted to exclude Algeria and Algerians from France
and, on the other, the majority of pieds-noirs who vehemently believed that the Algerian territory
should remain French. During the last several years of the war, the French government’s attention
slowly shifted from defeating the FLN on the battlefield to quelling its burgeoning fight with
staunch supporters of French Algeria whose terrorist actions threatened the safety of metropolitan
and Algerian residents. The government’s policies throughout the spring and summer of 1962 when
it made crucial decisions about breaking from Algeria were inflected by these acts such as the
February 1962 bombing of Culture Minister Andre Malraux’s apartment building, attacks on
Algerian civilians around the March 1962 ceasefire, and an August 22, 1962 assassination attempt on
de Gaulle.
This Franco-French war had two consequences for the relationship between France and the
harkis, in the term’s broader definition of anyone who had fought for France or was perceived to be
“pro-French” and their family members. First, the question of inclusion versus exclusion that
predominates during and as a result of Franco-French wars yet again surfaced as the Algerian War
for Independence drew to a close. At issue for the harkis was whether they would retain their
French nationality after the war, particularly given their service to France and complicity with the
French cause. Ultimately, President de Gaulle issued legislation that would restrict access to French
nationality for nearly all native Algerians—including the harkis—owing to their classification as
French with local civil status (Français de statut civil de droit local).9 Meanwhile the French with common
civil status (Français de statut civil de droit commun)—pieds-noirs and the infinitesimal number of
9
JORF, “Ordonnance n˚ 62-825 du 21 juillet 1962…,” July 22, 1962, 7230.
41
Algerian Muslims who had renounced their local, or Koranic, civil status—retained their French
nationality. Determined by their Berber and/or Arab origins, the colonial classification of French
with local civil status would limit the harkis’ entry into France’s emerging post-imperial geographic
boundaries because, after Algerian independence, they were foreigners (a legal status in France). The
government’s decision not to automatically accord French nationality to these former Algerian
Muslims who had fought for France followed policies developed over nearly two centuries, most
visible at moments of Franco-French internal conflicts, to exclude certain populations from full
French citizenship owing to their class, gender, religion, or ethnicity.
Second, the government’s focus on ending its Franco-French war with the pieds-noirs
obscured the harkis’ interests, which foremost included protection by the French Army from violent
retributions by other Algerians and unimpeded access to France to escape these attacks. Since the
government regarded the harkis as Algerians after the ceasefire and subsequently legally codified that
they were not French citizens, it gave the Army orders not to rescue harkis whose lives were in
danger and limited repatriation to harkis judged to be “severely threatened.” Indeed, the harkis
found themselves on the outside looking in on the Franco-French war that consumed the
government in the winter and spring of 1962, because the government considered them to be
neither Franco (pied-noir) nor French (metropolitan). While the government was circumspect
regarding the pieds-noirs—its adversaries in the Franco-French war—it guaranteed their French
nationality after the ceasefire, and with it their freedom to cross the Mediterranean. The harkis, like
other Muslim Algerians, were not accorded this guarantee. Despite promises to the contrary as they
fought side by side with French military officers, they were not eligible to automatically become
French citizens after France relinquished its control over Algeria in July 1962. Once the military
fighting ended, the Franco-French war rendered it impossible for the harkis to escape their
imperial—and now post-imperial—categorization as “Algerians.”
42
This section comprises two chapters that analyze how the fractures created as a result of the
Franco-French war in the early 1960s—which defined who belonged within France’s reduced
geographical boundaries—impacted the harkis’ access to French nationality after Algerian
independence. Chapter 1 focuses on the obstacles to French nationality posed by the harkis’ colonial
status and chapter 2 examines those posed by “repatriation” to metropolitan soil and the
administrative procedure to become French citizens.
43
CHAPTER 1
Algerians? The Harkis’ Deeply “Rooted” Colonial Status
Le souci de la protection des musulmans fidèles à
la France conduit à des conclusions diamétralement
opposées: pour pouvoir admettre ces musulmans au
bénéfice des mesures de rapatriement, la France doit
souhaiter que tous les musulmans conservent de plein
droit la nationalité française, au moins pendant un
certain délai. Durant cette période, la liberté absolue
de quitter l’Algérie étant garantie, la France devrait
accueillir les musulmans sans aucune restriction.1
- Official communication from the Secretariat
of State for Algerian Affairs (SEAA),
December 11, 1961
On July 21, 1962, nearly three weeks after France relinquished control of Algeria
symbolizing the end of its 132-year colonial empire, President de Gaulle issued an ordinance
concerning French nationality for Algerian residents, which had serious consequences for the socalled “musulmans fidèles à la France.”2 This legislation differentiated individuals’ access to French
nationality based on their former colonial citizenship status: French with local civil status versus
French with common civil status. The French with common civil status—European settlers, Jews,
and an infinitesimal number of Algerian natives—kept their French nationality, no matter whether
they resided in France or Algeria after independence. However, the legislation required the French
with local civil status—an inferior citizenship category during the colonial era for native Algerians
that included almost all harkis—to request French nationality in front of a judge on metropolitan
French soil. The ordinance, therefore, induced three obstacles to French nationality for these harkis:
Archives du Ministère des Affaires Etrangères (hereafter MAE), SEAA, 97, “Note a/s Circulation des Musulmans,” 1,
Dec. 11, 1961.
2 JORF, “Ordonnance n˚ 62-825 du 21 juillet 1962…,” July 22, 1962, 7230.
1
44
their colonial status based on their non-white ethnicity, the need to be repatriated to France, and the
process of making a nationality declaration.3
The July 21 legislation signified that the harkis’ former colonial status proved more
important than their service to France in determining their nationality at the end of French Algeria.
The requirement put on the harkis of requesting French nationality demonstrated that the post-1962
government continued to view the Arab and Berber harkis as unequal to the other “pro-French”
population in Algeria, the pieds-noirs. 4 Just as the first law regarding nationality for Algerian
residents during the colonial period had attributed a superior French nationality status to the settler
population in 1865, the 1962 text accorded superior access to French nationality to this other group
of “pro-French” Algerian residents, who automatically became French citizens. Moreover, the
ordinance did not differentiate the harkis from other native Algerians who sought out French
nationality: the procedure to become French citizens after Algerian independence was identical for
harkis and for Algerians who had fought against France. De Gaulle’s ordinance reaffirmed the
March 18, 1962 Evian Accords between France and the FLN, as the peace treaty ending the
Algerian War for Independence made no special provisions for the harkis to become French citizens.
By placing the harkis within the category of “Algerians,” de Gaulle disregarded their actions to
protect France during the recently lost—and eager to be forgotten—Algerian War for Independence.
Instead, he privileged their colonial status as the primary factor determining whether they would
automatically be granted French nationality. In a discussion about the harkis’ access to metropolitan
soil during the July 25, 1962 Council of Ministers’ meeting, de Gaulle expressed a differentiation
between Muslim Algerian roots and French origins. He claimed that: “On ne peut pas accepter de
Historian Guy Pervillé provides an interesting discussion of the evolving meanings and uses of the myriad terms used
to describe natives, Jews, French, and other Europeans living in Algeria throughout the colonial era in “Comment
appeler les habitants de l’Algérie avant la définition légale d’une nationalité algérienne?” Cahiers de la Méditerranée 54 (June
1997): 55-60.
4 In contrast, the July 1960 law issued after Sub-Saharan African countries and Madagascar gained their independence
based the nationality of these countries’ indigenous residents on their place of birth and not on their personal status.
This difference, once again, highlights the uniqueness of Algeria.
3
45
replier tous les musulmans qui viendraient à déclarer qu’ils ne s’entendront pas avec leur
gouvernement! Le terme de rapatriés ne s’applique évidemment pas aux musulmans: ils ne retournent
pas dans la terre de leurs pères!”5 Even if de Gaulle’s statement was incorrect since the “French”
population in Algeria had migrated from different European countries, he was clear: the harkis’ lack
of ancestral heritage in France was a reason to limit their migration to France. In July 1962, after the
end of the empire, the harkis’ deeply “rooted” colonial status as “Muslims” continued to determine
on which side of the Mediterranean the French government believed that they belonged. The
rupture with exclusionary colonial methods of governance in Algeria did not end once France lost its
former “colony.” The harkis could not escape their imperial label as “French Muslims from Algeria”
or FMA.6
These measures at the height of political crisis 1962 evoked once again the question of
inclusion and exclusion based on difference (gender, religion, and ethnicity, for example), which had
been present during previous Franco-French conflicts. The meaning of the July 21 ordinance for the
harkis cannot be understood without taking into account the Franco-French war between the
French government and the OAS. De Gaulle’s decision to resort to an emergency legal text to
attribute French nationality to Algerian residents during decolonization is particularly consequential
because it reflects France’s crisis situation in the immediate aftermath of the Algerian War for
Independence. Ordinances, a special executive power granted by the 1958 Constitution rarely used
Quoted in Peyrefitte, C’était de Gaulle, 209. “Tous” and “rapatriés” are italicized in Peyrefitte’s text.
Since 1865 Algerian residents were divided into two categories based on their civil status. However, at the end of the
colonial era emerged additional taxonomy in government documents that focused entirely on individuals’ origins—or
ethnicity. The term Français musulman d’Algérie appeared in a March 7, 1944 ordinance for Algerians in metropolitan
France and subsequently was used in opposition to Français d’Algérie, which explicitly referred to Algerian residents of
European and Jewish origins. (SHAT, 21R 179/2, “Ordinance du 7 mars 1944 relative au statut des Français musulmans
d’Algérie,” Journal Officiel N˚ 24 du 18 mars 1944, 217.) According to Todd Shepard, this frequently employed term
lacked a clear definition about whether it included native Algerians who had renounced their Koranic status until a 1956
memo by the secretary of state for the Civil Service to his cabinet members. This document defined FMA as both
Algerians who conserved their local civil status and those who had renounced their Muslim religion to become full
French citizens. Shepard, The Invention of Decolonization, 49.
5
6
46
in the first decades of the Fifth Republic,7 authorize the president to expeditiously enact legislation
without securing Parliamentary approval in order for the law to take legal effect.8 Significantly, the
first time that de Gaulle used this power was in February 1960 after a week of insurrection and
barricade building by the pieds-noirs who revolted against his decision to dismiss Algiers Prefect and
Army General Jacques Massu, a staunch supporter of French Algeria. This ordinance, which
targeted the budding Franco-French conflict, authorized de Gaulle’s government to “prendre
certaines mesures relatives au maintien de l’ordre, à la sauvegarde de l’État, à la pacification et à
l’administration de l’Algérie.”9
The now raging Franco-French war was a direct factor in de Gaulle’s July 1962 decision to
enact the same type of emergency legislation to maintain order over which Algerian residents would
be French (the Franco pieds-noirs) and which would remain Algerian (all others). The FrancoFrench war—which obscured the interests of anyone who, like the harkis, was neither Franco (piednoir) nor French (metropolitan)—yet again played a determining role as de Gaulle tried to extricate
France from Algeria, or rather, tried to extricate Algerians from France. The need to officially
request French nationality had a great impact on the harkis in Algeria. As the December 1961
communication from the Secretariat of State for Algerian Affairs cited above predicted, the
restricted access of the “musulmans fidèles à la France” to French nationality limited the French
Army’s legal capability to protect them after Algerian independence because France would be
intervening in Algeria’s internal affairs.
According to a working paper by the French Senate, from 1960 to 1990 Presidents issued a total of 158 ordinances,
which is, on average, five times per year. “Les documents de travail du Sénat, Série Études Juridiques, Les Ordonnances,
Bilan au 31 décembre 2007.” <http://www.senat.fr/ej/ej_ordonnance/ej_ordonnance0.html.>
8 Article 38 of the Fifth Republic Constitution states: “Le Gouvernement peut, pour l’exécution de son programme,
demander au Parlement l’autorisation de prendre par ordonnances, pendant un délai limité, des mesures qui sont
normalement du domaine de la loi. Les ordonnances sont prises en Conseil des ministres après avis du Conseil d’État.
Elles entrent en vigueur dès leur publication mais deviennent caduques si le projet de loi de ratification n’est pas déposé
devant le Parlement avant la date fixée par la loi d’habilitation.” <http://www.assembleenationale.fr/connaissance/constitution.asp.>
9 “Les documents de travail du Sénat… Les Ordonnances.”
7
47
De Gaulle’s focus remained on French internal politics and French citizens. He issued the
July 21 ordinance as he looked to curtail an unanticipated massive influx from Algeria, which tested
France’s already saturated public housing system.10 In June 1962 alone, 354,914 pieds-noirs migrated
from Algeria to France and only 26,480 returned to Algeria.11 And in June and July 1962, 15,339
harkis arrived in Marseille with only a miniscule number returning to Algeria to bring their families
back to France.12 On the other hand, the reverse was true for Algerian labor migrants. According to
data kept by the Sûreté Nationale on the movements of Algerians (which did not include harkis), in
July 1962 only 7,510 people arrived in France and 23,478 migrated to Algeria and in August, 22,193
people arrived in France and 26,705 migrated to Algeria.13 The post-1962 government’s efforts to
slow down the exodus from Algeria by limiting the access of French with local civil status to French
nationality targeted, at least at the moment of independence, the harkis.
This chapter presents a brief history of French nationality law in colonial Algeria to
introduce an analysis of the language and significance of the 1962 text. Next, it examines the first of
three obstacles to becoming French citizens that the legislation created: the harkis’ colonial status.
Chapter Two will analyze the remaining two obstacles: the harkis’ limited access to the repatriation
process and the confusing logistical process the legislation required once they reached France.
The “Nationalité Dénaturée” of Algerian Muslims
The harkis’ nationality status created by the July 21, 1962 ordinance resembled the
exceptional nationality status of the inferior Algerian Muslims during the colonial period, which
The French government insisted on the principle of free circulation in the Evian Accords allowing the pieds-noirs
who had fled Algeria owing to violence at the end of the war to return to Algeria once peace was reestablished. However,
the population shift was not bi-directional. Instead, over one million former residents of Algeria migrated permanently
to France during the 1960s. These figures do not take into account their descendants born after they arrived on French
soil.
11 Jean-Jacques Jordi, De l’exode à l’exil: Rapatriés et Pieds-Noirs en France (Paris: L’Harmattan, 1993), 66.
12 CAC, 19910467/1, “Chiffre des harkis qui sont arrivés à Marseille en juin et juillet 1962.”
13 ADBR, 138 W 5, Service de coordination et d’information nord-africaine (SCINA), “Synthèse quotidienne de
renseignements, journées des 8, 9, 10 & 11 septembre 1962,” g, N˚ 1 750.
10
48
historian Patrick Weil labels a “nationalité dénaturée.” Playing on the idea of “naturalization”—a
“rebirth” according to the term’s Latin etymological root of “naturalis”—Weil uses this concept to
signify that Algerian Muslims’ nationality was “vidée de ses droits et de son sens.”14 Their rights were
inferior to those of French citizens and, more perplexingly, in most respects inferior to those of
foreigners living on French soil, which included Algeria. Muslim natives had a specific set of laws
about naturalization that applied explicitly to them, which highlighted their truly unique status.
Consequently, they were ineligible to become French citizens under the 1889 nationality law, which
put forth a republican version of jus soli—nationality based on one’s place of birth—as the dominant
criteria for determining the nationality of foreigners and children of foreigners residing on French
soil, including Algeria.15 The law greatly facilitated access to French nationality for the pieds-noirs
(many of whom had migrated to Algeria from countries such as Spain and Italy) since Algeria had
been incorporated as three departments of France in 1848. However, Algerian natives did not fall
into the category of “foreigners” residing on French soil, even though they were not considered
French citizens.
Instead, as mentioned in the introduction, Muslim natives’ access to French nationality was
determined by an 1865 sénatus-consulte. This judicial decree bestowed upon them the possibility to
Weil, Qu’est-ce qu’un Français?, 238. Weil and philosopher Pierre-André Taguieff maintain that naturalization cannot be
reduced to a juridical process of acquiring nationality because it must “être pensé comme une métamorphose, voire une
transsubstantiation, fondée sur une déculturation/acculturation (reculturation) des étrangers rendue possible par la
scolarisation, laquelle garantit l’homogénéité linguistique et l’inculcation de valeurs communes, ainsi la convergence des
mœurs, dont la mixité des mariages est, par exemple, témoin et facteur.” Taguieff and Weil, “‘Immigration’, fait national
et ‘citoyenneté’” Esprit (May 1990): 89. On naturalization also see: Abdelmalek Sayad, “La ‘naturalisation,’” La double
absence: Des illusions de l’émigré aux souffrances de l’immigré (Paris: Seuil, 1999), 319-71.
15 Jus soli, which established citizenship based on one’s place of birth, prevailed as the chief determinant of French
citizenship during the Old Regime and until the 1804 Civil Code issued by Napoleon. Therefore, the Frenchman was
defined by his link to the king and to the physical territory. The Civil Code broke with monarchy’s conception of
citizenship by instituting for the first time in Europe the idea of jus sanguinis or filiation. The June 26, 1889 law—known
as the first nationality law—reintroduced jus soli into nationality legislation. It also reaffirmed the principle of double jus soli
(a child born in France to a foreigner who was also born in France is automatically French) established in a February 7,
1851 law, but added the restriction that the child could not repudiate French citizenship. This republican version of jus
soli correlated nationality with socialization under the premise that those who were educated in French schools learned
how to become French citizens. The schooling provided at camps such as Rivesaltes for harki adults and children after
their repatriation follows this logic; see chapter 3 of this dissertation. On the 1889 law, see Rogers Brubaker, Citizenship
and Nationhood in France and Germany (Cambridge, MA: Harvard University Press, 1992), 85-113 and Weil, Qu’est-ce qu’un
Français?, 53-61.
14
49
obtain full French citizenship if they underwent what Weil denominates “une lourde procédure de
‘naturalisation’” culminating in the president of the Republic signing a decree of approval for each
candidate.16 The law’s first article pronounced: “[L’indigène musulman] peut, sur sa demande, être
admis à jouir des droits de citoyen français: dans ce cas, il est régi par les lois civiles et politiques de
la France.”17 In other words, in order to obtain French nationality and its accompanying civil rights,
duties, and social benefits, the Algerian native needed to renounce the five customs of Muslim
Sharia law that were incompatible with the Civil Code (such as polygamy and the male privilege in
matters of inheritance).18 The three million natives residing in Algeria in 1865 were technically
French, but, by virtue of being governed by Sharia law, they maintained an inferior status to the
Christian residents of Algeria and to Jews living in Algeria following their mass naturalization and
inclusion into the definition of French from Algeria with the 1870 Crémieux Decree.19 Most notably,
natives were subject to the Native Code. This doctrine, passed in 1881, created a different body of
laws for Algerian natives that severely restricted their freedoms by sanctioning them for offenses
such as speaking disrespectfully to or about French officials, traveling outside of their commune
without a permit, and refusing to perform obligatory forced labor tasks (la corvée) including fighting
forest fires and grasshoppers. The infinitesimal number of natives who became naturalized French
citizens (2,396 between 1865 and 1915)20 were able to escape what historian John Ruedy refers to as
Weil, Qu’est-ce qu’un Français?, 211.
Ministère de la Justice, “Le sénatus-consulte du 14 juillet 1865 sur l’état des personnes et la naturalisation en Algérie,”
in La nationalité française, Textes et documents (Paris: La Documentation Française, 1996), 228.
18 The other customs include: the right for a Muslim father to marry off his child until a certain age (djebr), the right of a
husband to break the conjugal bond at his own discretion, and the theory of the “sleeping child” (the possibility to
recognize the legitimate filiation of a child born more than ten months and as long as five years after the dissolution of a
marriage). Weil, Qu’est-ce qu’un Français?, 234.
19 Historian Daniel Rivet estimates the population of Algerian Jews at 50,000 individuals on the eve of colonization.
Daniel Rivet, Le Maghreb à l’épreuve de la colonisation (Paris: Hachette, 2002), 71.
20 Weil, Qu’est-ce qu’un Français?, 237.
16
17
50
“a humiliating regime of exception.”21 This tenet of the sénatus-consulte based on personal religious
status served as the first obstacle to the natives’ citizenship.
But, as Weil demonstrates, renunciation of their personal status as Muslims did not always
guarantee full French citizenship since its attribution hinged on a discretionary decision by French
magistrates, some of whom believed that “Muslim” was a permanent label that remained unchanged
even if an individual stopped practicing Islam.22 The authority over each individual dossier that the
government bestowed to civil servants underscores the subjectivity of this process and represents a
second obstacle toward gaining French citizenship. Each candidate needed to undergo an
administrative investigation, which scrutinized his morality, ancestors, and marital situation, thereby
giving colonial authorities significant power over the natives. According to sociologist Laure Blévis,
administrators in France and Algeria looked above all for proof of their possessing “worth” (valeur)
and “dignity” (dignité),23 that is, their exemplifying of “‘bonne vie et mœurs.’”24 One ideal means for
the natives to demonstrate loyalty and worth to the mother country was service in the military;
soldiers and former soldiers represented twenty-seven percent of the Muslim natives naturalized
between 1870 and 1919.25 During the nineteenth century when colonial policy focused on the theory
of assimilating the “natives” into Frenchmen, the total of rejected requests for naturalization was
quite small. However, starting around the turn of the century once the French government shifted
to a policy of association that focused less on the “francisation” of colonial populations, the figure
Ruedy, Modern Algeria, 89.
The court of appeals in Algiers handed down a 1903 ruling that “le terme musulman ‘n’a pas un sens purement
confessionnel, mais qu’il désigne au contraire l’ensemble des individus d’origine musulmane qui, n’ayant point été admis
au droit de cité, ont nécessairement conservé leur statut personnel musulman, sans qu’il y ait lieu de distinguer s’ils
appartiennent ou non au culte mahométan.’” Quoted in Weil, Qu’est-ce qu’un Français?, 235 and 334 n. 153.
23 Blévis, “La citoyenneté française,” 28. Sociologist Emmanuelle Saada calls attention to the ubiquity and frequency of
the terms “dignity” and “prestige” (as in “threats to [the] prestige” of the mère patrie) in legal documents produced by
colonial administrators, arguing that these codified legal terms were “privileged instruments of colonial domination.”
Emmanuelle Saada, “The Empire of Law: Dignity, Prestige, and Domination in the ‘Colonial Situation,’” French Politics,
Culture & Society 20.2 (2002): 101.
24 Blévis, “La citoyenneté française,” 39.
25 Ibid., 40.
21
22
51
grew markedly, reaching one third, one half, or even three fifths of the requests.26 This discretionary
dynamic between state representatives and the natives was indeed a facet of the colonial civilizing
mission, which sought to adjudicate who was “evolved” enough to be awarded full French
citizenship.
The last set of hurdles that Muslim natives needed to surmount to obtain French nationality
was logistical. Their dossier needed to include eight different documents, including the certificate of
“bonne vie et moeurs” from French local government officials. Each Algerian had to appear in
person before the mayor or administrative authority to declare that he had renounced his status as a
Muslim and would abide by French political and civil laws. Once accomplishing these tasks, the
candidate became the object of the administrative inquiry described above. Along with the opinions
of the prefect and the governor, administrators sent the dossier across the Mediterranean for
approval by the Ministry of Justice, the Council of State, and finally the president of the Republic
who signed a decree.27 This procedure would remain in place until the end of French Algeria:
between 1865 and 1962 approximately 7,000 Algerian natives would become French citizens.28
The “Nationalité Dénaturée” of the Harkis
Although the harkis did not have to withstand such strict scrutiny, the obstacles that they
encountered in acquiring French nationality as a result of the July 21, 1962 ordinance “relative à
certaines dispositions concernant la nationalité française” are reminiscent of the Algerian Muslims’
nationality procedure during the colonial period. The July legislation’s second article stipulated that
“les personnes de statut civil de droit local originaires d’Algérie ainsi que leurs enfants” who had
Moutet report, 1919, reprinted in Victor Piquet, Les réformes en Algérie et le statut des indigènes (Paris: Emile Larose, 1919),
59-64, quoted in Weil, Qu’est-ce qu’un Français?, 237.
27 Weil, Qu’est-ce qu’un Français?, 236.
28 Patrick Weil, “Histoire et mémoire des discriminations en matière de nationalité française,” Vingtième siècle. Revue
d’histoire 84 (Oct.-Dec. 2004): 8.
26
52
French nationality on July 3, 1962 (the date that France recognized Algerian independence)
“peuvent, en France, se faire reconnaître la nationalité française.”29 This category of Algerians—
which included the harkis—had been the inferior colonial status of almost all natives. According to
the text, they had to present themselves before January 1, 1963 in front of a metropolitan French
judge bearing the proper documentation and make a declaration affirming that they wished to
“remain”—or, more accurately “become”—French as outlined in the 1945 Nationality Code. If not,
they would lose the possibility of acquiring French nationality.
The obstacles that the ordinance created for the harkis to become French (their colonial
status, repatriation to France, and the administrative procedure) are significant in themselves.
However, the sometimes subtle wording of the text makes its impact even more significant. First,
the lexical choice of “personnes” to describe the Muslim natives in the ordinance indubitably
symbolized that these former French citizens were no longer considered French by the government.
Hitherto, the harkis and other Muslim natives in colonial Algeria had been legally categorized as
“French with local civil status.” This inferior class of French nationality was predicated on the
Algerian natives’ need to renounce their personal status as Muslims and the subsequent obstacleridden procedure they needed to endure before they could become naturalized French citizens. The
decision to employ “personnes” contrasts with Article One of the ordinance, which referred to the
pieds-noirs in Algeria as “French with common civil status,” a juridically superior class of French
citizens during the colonial era. The legislation held that all pieds-noirs were now entitled to
automatic French nationality without any legal proceedings, even if they still resided in Algeria,
which amounted to a double French-Algerian nationality.30 The inferior colonial status of harkis as
JORF, “Ordonnance n˚ 62-825 du 21 juillet 1962…,” July 22, 1962, 7230.
The first article of the July 21, 1962 ordinance decreed: “Les Français de statut civil de droit commun domiciliés en
Algérie à la date de l’annonce officielle des résultats du scrutin de l’autodétermination conservent la nationalité française
quelle que soit leur situation au regard de la nationalité algérienne.”
29
30
53
“Français/personnes de statut civil de droit local,” therefore, continued to legally differentiate them
from their non-Muslim counterparts after the end of the French empire.
Second, harkis who had been French with local civil status could only make a nationality
declaration if they resided “in France;” they could not ask for French nationality while they were still
living in Algeria. This stipulation did not exist for any of the pieds-noirs, who became French even if
they chose to reside in Algeria. Conversely, the harkis’ French nationality was contingent upon their
movement across the transforming borders of France, a migration that, as the next chapter
demonstrates, the French government tried to limit.
Lastly, the phrase “se faire reconnaître la nationalité française” and the process it entailed in
part induced ambiguities in the nationality status of harkis and other Muslim Algerians once they
arrived in metropolitan France. “Se faire reconnaître,” which is a less common phrase than “se faire
connaître” (to identify oneself or to make oneself known) and “se reconnaître” (to recognize oneself
or to consider oneself), does not have a straightforward meaning and, therefore, leaves many
questions about Muslim Algerians’ French nationality when they arrived in metropolitan France.
Were they French during the liminal moment between July 3, 1962 (the date they—retroactively—
lost French nationality) and the time that they made a nationality declaration? Were they acquiring or
reacquiring French nationality? After all, the question to which they needed to respond was
“Voulez-vous garder votre nationalité française?”
The process the harkis and other Algerian Muslims needed to undergo to acquire French
nationality after the end of the empire implicitly addresses these questions. The ordinance required
them to follow the procedure outlined by the 1945 Nationality Code applicable to foreigners: appear
before a judge with a lengthy list of appropriate documentation. This procedure in many ways
mimicked that implemented during the colonial period for Algerian subjects to become full French
54
citizens. The recourse to this policy implied that the harkis were foreigners like the Algerian natives
who had actively fought against France during the seven and a half year war.
The harkis’ nationality was indeed “denatured” because these “pro-French” Muslims could
only become French through an exceptional process, which discounted their military service to
France. The harkis’ defense of French Algeria—which put their lives in danger not only during the
war on the battlefield but also at that moment as members of the FLN and its military arm, the ALN,
tortured and massacred harki “traitors” by the thousands—did not earn them the right to French
nationality. Their military service arguably created a “blood debt” owed by France toward them, but
it did not repay the debt by welcoming the harkis into the French polity.31 The refusal to grant
automatic French nationality to the harkis mimicked the policies of the French government during
the imperial era, which had excluded colonial soldiers and veterans from full French nationality.
Even if this exclusion contradicted the republican principle of equality, such policies were also,
paradoxically, intrinsic features of republican France.32 Although citizenship was tightly linked to
defense of the nation since the formation of the National Guard on July 13, 1789 created a “citizenarmy” 33 and although the Third Republic embraced military service as one of the twin
In a recent study, Gregory Mann probes the notion of France owing a “blood debt” to French West African veterans.
He argues: “in twentieth-century West Africa, regional idioms and ideals of social exchange, mutual obligation, and
uneven reciprocity intersected with French ideas about the special relationship between a nation and its veterans… The
unfinished product of the intersection of these contentious French and West African ideals is a political language of
sacrifice and obligation that continues to inflect contemporary debates, notably about African immigration to France and
the idea of a ‘blood debt’ owed by France to its former colonies.” Gregory Mann, Native Sons: West African Veterans and
France in the Twentieth Century (Durham, NC: Duke University Press, 2006), 4.
32 On this argument, see Wilder, The French Imperial Nation-State. He contends on page 4 that an “antinomy between
universality and particularity… existed within both the metropolitan and colonial poles of the imperial nation-state and
… expressed itself in discourses as well as practices.”
33 On the National Guard “citizen-army” formed through individuals’ political and military mobilization during the
Revolution, see Pierre Birnbaum, “L’Étatisation de la nation: La levée en masse pendant la Révolution Française,” in La
Logigue de l’État (Paris: Fayard, 1982), 193-213. Julia Osman argues that while the National Guard has been widely
recognized as France’s first “citizen-army,” the notion of citizen soldiers was not an outcome of the French Revolution,
but rather the result of reforms that the French Army enacted after the loss of its North American empire during the
1756-63 Seven Years’ War. Julia Osman, “The Citizen Army of Old Regime France,” (PhD diss., University of North
Carolina at Chapel Hill, 2010), 12 and 64-65.
31
55
components—along with schooling—for building French citizens,34 republican France operated a
differentiation between les troupes indigènes and les troupes françaises in terms of their nationality status.
For example, despite a 1915 legislative proposal to reward colonial troops with French nationality,
the imperial nation-state never opted to bestow this “highest, noblest recognition” on its colonial
subjects who served as vital components of the French Army during France’s colonial conquests all
over the globe, and later in the World Wars, and even in the wars of decolonization in Asia and
Africa.35 After France’s second empire ended with loss of Algeria, the French government chose to
uphold imperial citizenship policies for its former soldiers. De Gaulle, eager to resign the “Affaire
d’Algérie” to the past, did not want to encourage the migration of Berber and Arab Algerians to
France. Despite continual promises that government officials made to the harkis in the war’s waning
months that they would be French citizens, once Algeria became independent the harkis’ former
colonial status trumped their having risked their lives for France.
The Harkis’ Future in an “Algerian Algeria”
The requirement for harkis to formally request French nationality simultaneously
contradicted public claims the government had made during the war that they would not encounter
obstacles to being French citizens and reinforced the policies of the imperial French government to
distance itself from them. As the war drew to a close, the French Army disarmed the harkis and
encouraged them to return to their villages, without its protection. Despite the discourse to mollify
the trepidation that many harkis expressed about potentially losing their French nationality, it is clear
For an analysis of the Third Republic’s defining citizenship based on the idea of socialization into the nation through
schooling and military service, see: Brubaker, Citizenship and Nationhood, 108-13; Gérard Noiriel, “Nations, nationalités,
nationalismes: Pour une socio-histoire comparée,” in État, nation et immigration: Vers une histoire du pouvoir (Paris: Belin,
1999), 128-37; and Eugen Weber, Peasants into Frenchmen: The Modernization of Rural France, 1870-1914 (Stanford, CA:
Stanford University Press, 1976), 292-302 and 330-38.
35 The legislative proposal presented before the Chamber of Deputies on April 1, 1915 stated: “It is an obligation for
France to seek to compensate the indigènes who fight for her, or who, simply but loyally, have fulfilled their military duty.
The highest, noblest recognition that France can perceive is to offer what she considers most precious, that is to say,
French nationality.” Quoted in Fogarty, Race and War, 230.
34
56
that the French government neither viewed the harkis as full citizens nor intended them to be
citizens after the war. During their service in the French Army, government documents referred to
the harkis as musulmans, Français de souche nord-africaine or its abbreviated form FSNA, and even
Musulmans fidèles à la France. Army officials most often used FSNA or harkis in their documents,
while Foreign Affairs’ officials employed musulmans, musulmans fidèles à la France, and musulmans attachés
à la France seemingly to include civilian pro-French Muslims. After the end of the war, and even
subsequent to the loss of colonial Algeria, government papers kept employing such colonial
classifications to define the harkis, thereby distinguishing them from other French citizens.
Once it became apparent that French Algeria would cease to exist, government officials
employed a discursive strategy of promising the harkis that they could be French citizens and that
France would not abandon them. The first negotiations to discuss the “new Algeria” and move
toward re-establishing peace in the war-ravaged country opened between an FLN delegation led by
Provisional Government of the Algerian Republic (GPRA) president Belkacem Krim and a French
government delegation headed by Minister of Algerian Affairs Louis Joxe in Evian, Switzerland on
May 20, 1961.36 On this date, France also declared a unilateral one-month truce in the fighting.
These talks and actions signaled to the Algerian natives fighting for the French cause that an FLN
victory was imminent. The auxiliary soldiers’ morale, which had been degrading since the previous
summer because of their fear of inevitable reprisals if the FLN was victorious, dipped even further
and they expressed their grave concerns to their superiors about their future in a new Algeria.37 One
harki soldier wrote in June to Colonel Guillard of the Akbou sector in Kabylia that the harkis’
“attitude… bonne volonté et loyauté” had considerably contributed to French military success in
Joxe opened the talks by exposing France’s double objective: “tenter de rétablir la paix après six années de guerre en
Algérie; étudier les principes selon lesquels peut se fonder une Algérie nouvelle.” Vers la paix en Algérie: Les négociations
d'Evian dans les archives diplomatiques françaises (15 janvier 1961 - 29 juin 1962) (Brussels: Bruyant, 2003), 40.
37 SHAT, 1H 1397/7, Le Chef d’ Escadrons de Bouillas, “Objet: État d’esprit des supplétifs – Problème de Solde,” 1, Nº
5100/EMI/2/EG, Sept. 20, 1960.
36
57
Algeria, but that “la France demeure encore sourde à notre égard et ignore totalement le sort de les
[sic] Harkas sans défense.”38 He also exposed growing discontent by harkis who worried about their
relationship with France. Colonel Guillard immediately met with the harki to reassure him that “en
aucun cas, la France n’abandonnerait tous ceux qui s’étaient engagés à ses côtés.” 39 Guillard’s
superior noted in a letter sharing this story with the Commander of the Western region of
Constantine that the Akbou harki’s concerns about a future relationship with France reflected the
concerns expressed for some time by practically all of the harkis.40
Guillard’s response aligns with instructions from Prime Minister Michel Debré that the
Army Chief of Staff in Paris relayed that same month to military commanders in Algeria with orders
to distribute to harki soldiers with “le plus grand discernement” given “l’extême sensibilité actuelle
du moral des supplétifs.” In other words, officials in Paris feared that harki soldiers might desert the
French Army, or worse, use their weapons to turn on French soldiers. Concerning their relationship
with France, this document stated, “Les supplétifs FSNA pourront en toutes circonstances se
réclamer de la citoyenneté Française et bénéficier de tous les droits correspondants. À cet égard le
Corps d’Armée a demandé aux autorités supérieures que soient indiqués les avantages concrets de
cette mesure et a prescrit en attendant, de préciser aux harkis qu’ils sont des citoyens et des soldats
français jouissant de tous les avantages de cette situation.”41 Based on these instructions relayed from
the Prime Minister, the French government was asking military officials to affirm to harki soldiers
that they were French citizens without question.
SHAT, 1H 1397/7, Harki serving in Harka 224 stationed in Tazmalt, handwritten letter to Colonel Guillad (Battalion
Chief of the II/2˚ RIMA) received on June 17, 1961, page 1. This text is attached to a letter from Colonel Charrier to the
General commanding Western region of Constantine, “Objet: Moral des Harkis,” N˚ 1655/SA/B 3/PH/SC, June 23,
1961. Colonel Charrier believed that the harki’s father, who was a local Muslim notable, might have written the letter.
39 Ibid.
40 Ibid.
41 SHAT, 1H 1260/1, Division General Perrotat, “Note de Service. Objet: Situation des Harkis et AASSES,” June 12,
1961.
38
58
Later that summer, on August 25, 1961, the Deputy Chief of Staff of the Army in Paris
issued a memorandum with the precise measures the government was taking to improve the harki
soldiers’ situations for the zone commanders in Algeria to communicate with all soldiers. The note
promised increased pecuniary benefits, more opportunities within each unit of harkis (harka) to earn
a higher military rank, and a forthcoming statute allowing for more harkis to enlist in the “regular”
army. Its final clause affirmed the harkis’ French nationality, but with a catch not present in the June
instructions: “Les supplétifs sont des personnes de nationalité française. Cette situation ne pourra
être modifiée que par l’effet d’une loi française.”42
The French government maintained a public discourse that it would not strip the harkis of
their French nationality and gave no indication that the state would issue such a law. For example, a
fall 1961 public announcement broadcast on Algeria’s Radio Arabe opened with the sentence “Les
Harkis vont toucher le prix de leur fidélité à la République Française.” Its first paragraph, seemingly
aimed at assuaging the harkis’ fears about being abandoned, elaborated on this point by promising
them that they would remain French citizens: “Les harkis qui servent sous les armes françaises
pourront se réclamer, en toutes circonstances, de la citoyenneté française et de la totalité des droits
qui s’y rattachent.”43 With these words, the government publicly equated the harkis’ service in the
French Army with the reward of French nationality, but enacted no measures to assure this status.
Officials in Algeria, well acquainted with the political climate, also supported guaranteeing
French nationality for the pro-French Muslims and opposed requiring them to ask for French
nationality. On October 4, 1961 and on November 2, 1961, the Secretariat of Algerian Affairs
circulated to the Prefects of each Algerian department questionnaires principally focused on the
SHAT, 1H 1397/1, General Lagarde (S/Chef d’État-Major de l’Armée) à MM. Les Généraux Commandant les
Régions Territoriales et Corps d’Armée d’Alger, d’Oran, de Constantine et M. le Général Commandant la Région
Militaire du Sahara, “Objet: Problème Harki. Note d’information générale,” 2, N˚ 2800/EMA/1.E., Aug. 25, 1961. The
Army sent 400 copies of the memo to hand out in all military zones.
43 SHAT, 1H 1397/1, Le Chef d’État-Major, “Note pour la Radio Arabe.”
42
59
nationality and future of the pieds-noirs in a new Algerian state, but each form also included one
question concerning the protection of Muslims. 44 Question nine of the October questionnaire
inquired: “Quelle est la meilleure manière d’assurer la protection des musulmans particulièrement
attachés à la France? Doit-on pour certains prévoir leur transfert en Métropole et pour le plus grand
nombre le maintien s’ils le veulent de la nationalité française pour qu’ils bénéficient en Algérie des
garanties accordées aux Algériens de souche européenne?” The summary of responses compiled by
Algerian Affairs revealed that the Prefects strongly advocated safeguarding “les Musulmans attachés
à la France” from future possible FLN retaliatory measures by offering them the same protections,
possibilities for double nationality, and right to migrate to France as the pieds-noirs would have.45 At
the same time, the Prefects recognized that assimilating the pro-French Muslims with the pieds-noirs
would provoke a de facto expatriation from their community of origin (as the Algiers prefect
contended, “On voit mal des Musulmans français dans une Algérie algérienne”) and, therefore, they
argued that French government must provide a place for these pro-French Muslims in the
repatriation process and guarantee their French nationality.46 Nevertheless, the Ministry of Algerian
Affairs opined in the summary’s conclusion that according the same dual nationality status to the
pro-French Muslims as to the pieds-noirs, which would allow both groups the same right to
repatriation, would be an illusory solution. This report concluded that most of these Muslims would
not opt for double nationality because it would place them in a minority position vis-à-vis other
Algerians.47 Opposed to these arguments and concerned for the harkis’ safety, local officials strongly
encouraged the government to allow the harkis to automatically keep their French nationality and, in
multiple notes to the Quai d’Orsay, warned of detrimental consequences if this were not the case.
The questionnaires, most of the responses, and a summary of responses can be found in MAE, SEAA, 97.
MAE, SEAA, 97 (dossier: Premier questionnaire – garanties), “Synthèse des réponses au premier questionnaire relatif
aux garanties,” Oct. 24, 1961.
46 Ibid.
47 Ibid.
44
45
60
One such document titled “Protection des Musulmans” contended: “Les Musulmans fidèles à la
France doivent pouvoir conserver à l’égard de la France leur nationalité française. Il serait injurieux
de leur imposer un acte positif pour que cette nationalité française leur soit reconnue.”48 These
words proved to be a harbinger of the future exile that the harkis would experience in France.
The Harkis’ Future in Post-Imperial France
As peace talks began, French civilian administrators and military personnel in Algeria warned
government officials in Paris that they must protect the pro-French Muslims from “certain danger”
by assuring them a future legal link with the French state. Nevertheless, the negotiations during the
fall of 1961 and winter of 1961-1962 between FLN representatives and French Foreign Affairs
Ministry officials completely neglected to reference civilian and military auxiliary forces in
discussions about according French nationality to Algerian populations. Moreover, the resulting
document outlining the future relationship between the two countries and their populations—the
March 18, 1962 Evian Accords—made this position official by not including any special provisions
for the harkis to retain their French nationality.
In fact, this treaty did not include one single line about the nationality status of the harkis (or
that of other native Algerians).49 Instead, France placed the onus of assuring the harkis’ safety onto
the FLN and the new Algerian state, which, after the Algerian population approved its creation, the
Evian Accords declared “exercera sa souveraineté pleine et entière à l’intérieur et à l’extérieur.”50 The
protection of the harkis rested on this state, and this sovereign state alone, to uphold the following
clauses from the Accords:
MAE, SEAA, 97 (dossier: Protection des musulmans), DL, “Note a/s Protection des Musulmans,” 1, Dec. 12, 1961.
In the entire text, the word “nationalité” was used twice and both times the term was used referred to the pieds-noirs’
nationality. The two occurrences are in: Déclaration des Garanties, Deuxième Partie, Chapitre 1.
50 Accords d’Evian, Déclaration Générale, Chapter 2.
48
49
61
Nul ne pourra faire l’objet de mesures de police ou de justice, de sanctions disciplinaires
ou d’une discrimination quelconque en raison: d’opinions émises à l’occasion des
événements survenus en Algérie avant le jour du scrutin d’autodétermination; d’actes
commis à l’occasion des mêmes événements avant le jour de la proclamation du cessez-le-feu.
Aucun Algérien ne pourra être contraint de quitter le territoire algérien ni empêché d’en
sortir.51
In a directive to the commanders of the Army, Navy, and Air Force outlining the probable terms of
the peace agreement ten days before the ceasefire Minister of Armies Pierre Messmer emphasized
this trust of the new Algerian state. He wrote: “L’Algérie, Indépendante et Souveraine, garantira à
tous ses habitants la sûreté de leur personne et de leurs biens, le respect des particularismes et des
droits publics et privés.”52 As will be shown in the next chapter, the French government officials’
faith in the new Algerian State was misguided.
By not insisting on French nationality for the harkis, voices from the Quai d’Orsay were
dissonant with those of Army and local government officials in Algeria. These latter groups
overwhelmingly advocated including the pro-French Muslims as legal members of the French postimperial state, citing real safety concerns if they returned to their status as “Algerians.” The available
transcripts and summaries from secret meetings on October 28-29 and November 8-10, 1961,53 on
December 9, 1961,54 on December 30, 1961,55 on January 28-29, 1962,56 on February 11-19, 1962,57
Ibid.
SHAT, 1K 744 (dossier: Directives), Minister of Armies Pierre Messmer to the Commanders of the Army, the Navy,
and the Air Force, “Annexe 1: Les termes probables de la négociation,” Nº 206/MA/CAB/INF/CA (DR), Mar. 8, 1962.
53 The October 28-29 and November 8-10 secret meeting was held in Bâle, Switzerland between Mohammed Ben Yahia
and Redha Malek representing the FLN and Bruno de Leusse and Claude Chayet of the French Ministry of Algerian
Affairs. A hunger strike by FLN leader Ahmed Bella caused the talks to come to an end before higher-level officials
joined in. Vers la paix en Algérie, xxxvii.
54 On December 9, de Gaulle told his Enarque confidant at the Quai d’Orsay, Alain Peyrefitte, that “France had to
unburden itself of Algeria.” Matthew Connelly, A Diplomatic Revolution: Algeria’s Fight for Independence and the Origin of the
Post-Cold War Era (New York: Oxford University Press, 2002), 257. Following this conversation, Minister of Algerian
Affairs Louis Joxe brought de Leusse with him to meet Saad Dahlab and Ben Yahia at the Yéti chalet in the French Jura
mountains. Vers la paix en Algérie, xxxviii.
55 The third secret meeting on December 30 between the same parties as the December 9 encounter took place in
Chalain, France. Vers la paix en Algérie, xxxix.
56 Joxe led the French delegation and Dahlab led that of the FLN at the January 28-29 meeting in Lons-le-Saunier. Vers
la paix en Algérie, xli.
57 The delegations for the final secret meeting on February 11-19, once again in Rousses, were headed by Joxe and by
GPRA leader Krim Belkacem. Vers la paix en Algérie, xlii-xliii.
51
52
62
and finally those of the public Evian discussions on March 7-18, 196258 reveal that the French
government’s primary, and only, concern about Algerian populations’ nationality was guaranteeing a
double French-Algerian nationality for the pieds-noirs. Each time that the question of nationality
surfaced, it was either in a debate over the pieds-noirs being given automatic Algerian nationality
(the French position) or over Algerians in metropolitan France not losing their French nationality
(the FLN position).
Minister Joxe and his team’s insistence on double nationality for the pieds-noirs clashed with
the FLN’s position and made their choice to neglect the French state’s relationship with the “proFrench Muslims” more glaring. FLN officials, led in the initial meetings by Saad Dahlab and in the
last clandestine talks by Krim Belkacem, fervently maintained that only Muslim Algerians should
automatically become—or rather, remain—Algerian citizens, believing that “le peuple algérien,
composé de musulmans, a toujours existé” and this people would be reborn when France released
its stranglehold on the colony.59 Therefore, not only did the FLN delegation question the place of
the pieds-noirs in the new Algeria, but it also disputed the pieds-noirs’ status since 1830. The
debates over who would be automatically Algerian in the new state reawakened a significant source
of contention between Muslim natives and the pieds-noirs dating back to the nineteenth century,
when the latter had appropriated the label “Algerians” for themselves.60 The FLN wanted to protect
post-imperial Algeria from a neo-colonial situation of according the pieds-noirs special privileges
that would replicate the state within a state that had been present on their territory throughout the
colonial era. They argued that granting the pieds-noirs automatic Algerian nationality would threaten
A transcript of the proceedings for the first nine sessions of meeting (March 7-11) is located in Vers la paix en Algérie,
316-377.
59 Note du Ministère d’État chargé des Affaires algériennes, “Projet de propositions en vue de la reprise des
négociations,” Nov. 21, 1961, in Vers la paix en Algérie, 291.
60 For a discussion of the evolution and definition of the term “Algériens,” see Guy Pervillé, “Comment appeler les
habitants.”
58
63
the stability in, and the definition of, the nascent independent Algeria.61 The FLN wanted people of
European origin to officially affirm their allegiance to the new state, which would ensure that these
individuals supported the new iteration of Algeria.
The French government would not cede to the FLN about double nationality for the piedsnoirs, which became, according to Todd Shepard, “one of the most charged issues that was
negotiated.”62 The progression of the peace talks seemingly contradicted the French government’s
opposition to the pieds-noirs’ vehement stance about keeping Algeria French because government
officials allowed this issue to dominate discussions, and ultimately to drag them out. In fact, Sylvie
Thénault contends that during the last two years of the war, the killings and coup attempts on
metropolitan French soil carried out by the partisans of French Algeria reoriented the French
government’s focus in the conflict. These individuals became France’s primary enemy and were
increasingly viewed as a greater threat than FLN operatives.63 Most feared of all was the OAS. The
OAS’s violent actions seriously threatened France’s stability not only in Algeria, but also in the
metropole. This militia regularly engaged in bombing campaigns of government buildings and public
venues in Algeria, reaching a fevered pitch as France and the FLN grew nearer to a peace agreement.
In February 1962 alone, OAS bombs or bullets killed 553 people.64 This same month, an OAS bomb
rocked the ground floor of Minister of Culture André Malraux’s Parisian suburb apartment building,
leaving his landlord’s four-year-old daughter Delphine with a bloody face that the next morning’s
newspapers spread across their front pages. In protest, the Communist Party (PCF), the far-left
Unified Socialist Party (PSU), and the centrist Republican Popular Movement (MRP), along with
several leftist labor unions, organized a march near the Charonne metro station in the twelfth
arrondissement on February 8. The Interior Ministry’s infamous Forces de l’Ordre (security forces)
“Compte rendu, Entretien L. Joxe – S. Dahlab du 30 décembre 1961,” in Vers la paix en Algérie, 311.
Shepard, The Invention of Decolonization, 156.
63 Thénault, Histoire de la guerre, 198.
64 Connelly, A Diplomatic Revolution, 260.
61
62
64
executed a violent repression over the thirty thousand protesters, which left nine of them dead.
Historians estimate that as many as one million people attended their funerals at Père Lachaise
cemetery on February 13.65 Meanwhile, the OAS also set its sights on toppling its archenemy,
President de Gaulle, who increasingly demonstrated that he wanted the war to end. In April 1961, its
members perpetrated a failed putsch (known as the Generals’ Uprising) to protest secret
deliberations between Prime Minister Debré and the FLN and, on September 8, 1961, it botched a
coup attempt on President de Gaulle in Pont-sur-Seine, France.66
The OAS acts of violence that transformed the Algerian War for Independence into a
Franco-French conflict, along with mounting international pressure on France to capitulate in the
wake of the anti-colonial Cold War context, beg the question of why French negotiators would fight
so earnestly for the pieds-noirs’ right to Algerian nationality. Over the course of the negotiations, the
FLN and the French delegation passed back and forth a total of nineteen proposals and
counterproposals about the pieds-noirs’ future in Algeria.67 The French government had a selfish
motivation for wanting the pieds-noirs to obtain Algerian nationality; namely, they would be more
likely to stay in Algeria—and not flee to France—if they were Algerian citizens. But, an internal note
emanating from Joxe himself reveals another important motivation. France’s calculated choice not
to yield over the question of the pieds-noirs’ Algerian nationality reflected French strategic interests
in the future Algerian state: “La coexistence des Français et des musulmans devenus Algériens au
sein de l’État permettra de maintenir la présence française en Algérie…”68 Understanding why the
French government favored an Algerian nationality for the pieds-noirs sheds light on why it did not
ensure automatic access to French nationality to the harkis.
For a detailed account of this episode and a description of the Franco-French war between the government and the
OAS, see: Rémi Kauffer, “OAS: la guerre franco-française d’Algérie,” in La Guerre d’Algérie: 1954-2004, la fin de l’amnésie,
ed. Mohammed Harbi and Benjamin Stora (Paris: Robert Laffont, 2004), 451-76.
66 The OAS violence continued even after Algerian independence. In August 1962, OAS members bombed de Gaulle’s
motorcade heading from Paris to his hometown of Colombey-les-Deux-Eglises, narrowly missing the president.
67 Connelly, A Diplomatic Revolution, 259.
68 “Projet de propositions,” Vers la paix en Algérie, 291.
65
65
Once the war was over and there were no more partisans of Algerian independence (fellaghas)
to “pacify,” the harkis could not offer any strategic advantage to France if they became French
citizens. France would incur the responsibility of protecting them, or repatriating them. At the same
time, the French government recognized that offering the harkis French nationality was the sole
assurance for them to escape the inevitable FLN retaliatory violence because the French state would
then have legal justification to protect its own citizens. If they remained Algerian, France would
need to violate international law sovereignty principles to protect them: “S’ils devenaient Algériens
en Algérie, la défense de leurs intérêts et la protection de leur sécurité constitueraient un problème
délicat. Confier cette tâche aux représentants diplomatiques et consulaires de la France, serait une
atteinte à la souveraineté intérieure algérienne, et en quelque sorte un retour aggravé au système des
‘capitulations’ pratiqué durant les siècles précédents.”69
Figure 1: “L’Algérie de demain” pamphlet, pages 10-11. Source: SHAT, 1H 2467/6.
MAE, SEAA, 97, “Formule possible pour la protection des musulmans attachés à la France et n’envisageant pas
l’émigration en métropole.”
69
66
The government, or rather President de Gaulle, distanced the harkis from France and sought
to instantly forget France’s seven and a half years of fighting to maintain the fiction that Algeria was
French and Algerians were French.70 This view also did not take into account the sentiments of the
GPRA leaders, including its president Benyoucef Ben Khedda who issued a public statement to the
“Peuple Algérien” on March 18 in which he stressed, “le cessez-le-feu n’est pas la paix.”71 The
Algerian Revolution was not over according to Algerian nationalists. Nevertheless, immediately after
the ceasefire, the French government published a pamphlet titled “L’Algérie de demain” to explain
the terms of the Evian Accords to the pieds-noirs.72 The centerfold, its sole image, displays a
swarthy Algerian girl and a blond-haired French boy looking affectionately at each other while
embracing and surrounded by the words “Pour nos enfants la paix en Algérie” (see figure 1). This
picture negates France and Algeria’s colonial relationship semiotically through its choice of children
who were born after the start of the war and linguistically through the prominence of the word
“peace.” Despite the appearance of harmony, the photograph seems to contain a deeper
inegalitarian message. By depicting the French boy as bigger than the Algerian girl, the image
suggests to the reader that France is still more powerful than Algeria in size and in gender. The
concluding words of this propagandistic tract perfectly capture France’s belief that it remained
dominant over Algeria:
Dans le monde d’aujourd’hui, apparaissent beaucoup de nations nouvelles. Les Algériens
de souche musulmane, qui sont la grande majorité, semblent vouloir constituer une nation.
La France qui a créé l’Algérie leur donne une chance.
La France est décidée à marquer son intérêt pour l’Algérie nouvelle et à continuer une
œuvre grâce à laquelle celle-ci existe aujourd’hui…73
Todd Shepard argues that “French responses to the Algerian Revolution gave birth to the certainty that
‘decolonization’ was a stage in the forward march of history, of the Hegelian ‘linear History with a capital H.’” Shepard,
The Invention of Decolonization, 2.
71 Quoted in Mohammed Harbi and Gilbert Meynier, Le FLN: Documents et Histoire, 1954-1962 (Paris: Fayard, 2004), 853.
72 SHAT, 1H 2467/6, “L’Algérie de demain,” Mar. 1962.
73 Ibid., 19.
70
67
This document, therefore, positioned France as a unifying force in Algeria that would “continue” to
provide paternalistic support to its “oldest daughter” and would serve as a peace monger that
altruistically “created” the opportunity for the Algerian masses to “form a nation.” The pamphlet
shows France’s efforts to actively forget the reality of the situation. It neglected to mention that
France had spent 132 years asserting its unwanted dominance over the native Algerian subjects and
that it had not just “created” the opportunity for an Algerian nation, but it also created enable
irreparable divisions within its Muslim population.
Conclusion: Algerians in a Franco-French War
With no provisions in the Evian Accords for granting French nationality to the harkis, they
would become “Algériens de souche musulmane” who were left to find their own place in the new
Algerian nation, including escaping the violent retributions of FLN members. De Gaulle wanted to
make the “affaire d’Algérie”—which had shown “l’effroyable infirmité de l’État” because France
continually ceded to the pieds-noirs’ interests—part of yesterday, and not of tomorrow.74 As Pierre
Messmer, Minister of Armies from 1960 to 1969, explained in a 2004 interview with me: “C’est que
de Gaulle veut sortir d’Algérie, veut sortir. Et aller chercher les Harkis là où ils se retrouvent… dans
les villages, c’est chaque fois une opération militaire et ce que de Gaulle pense c’est que ces
opérations militaires risquent de rallumer la guerre d’Algérie. Et ça, on veut pas [sic].”75
Reigniting the Algerian War would not only spark fighting with the FLN, but it also would
send a signal to the pieds-noirs that France was willing to continue to fight to keep Algeria French.
De Gaulle did not want to add any more fuel to the Franco-French war, which had become a larger
threat to him and his government than their battle with the FLN. After the March 1962 ceasefire,
the Franco-French war consumed the government’s attention and resources: transcripts of Council
74
75
Quoted in Peyrefitte, C’était de Gaulle, 140.
Pierre Messmer, Interview with the author, July 5, 2004, Paris.
68
of Ministers’ meetings demonstrate that discussions more often centered on the pieds-noirs and the
OAS, and not on the FLN. Because the harkis were outsiders to the Franco-French war—owing to
their deeply rooted colonial status as “Muslim Algerians”—France ignored them and their interests.
In the July 21, 1962 ordinance, de Gaulle advocated a position of ethnically differentiated
nationality. In response to a statement by Minister of Algerian Affairs Joxe that the harkis wanted to
leave Algeria “en masse” and reports about violence among the Muslim population, the President
maintained that, “La France ne doit plus avoir aucune responsabilité dans le maintien de l’ordre
après l’autodétermination…. Si les gens s’entre-massacrent, ce sera l’affaire des nouvelles
autorités.”76 Once independence was proclaimed, the massacres among Muslim Algerians, including
those carried out toward harkis because they had fought for France, would no longer be France’s
“business.” The harkis were not included in the new definition of French citizens, revisiting
fractures present during previous Franco-French wars and evoking new questions of inclusion
versus exclusion based on ethnicity.
The next chapter follows the harkis in their quest to escape the massacres and overcome the
last two obstacles for French nationality by focusing on their movement from the Algerian colonial
space to within the new geographical—and figurative—boundaries of France.
76
Quoted in Peyrefitte, C’était de Gaulle, 151.
69
CHAPTER 2
French? Movements of Post-Imperial Citizens
Primo: Epuration menée par les populations et
l’ALN envers ex-supplétifs, signalée par message
référence, s’est poursuivie avec une violence accrue
durant la semaine écoulée…
Secundo: Précarité petits regroupements locaux,
rendus obligatoires par saturation principaux camps
présente risque incidents graves avec l’ALN. De plus
situation pitoyable anciens compagnons d’armées
menacés dans leur vie par la population…
Tertio: Honneur vous demander instamment
autoriser embarquement vers métropole ex-supplétifs
menacés, tant que pouvoir central algérien se révélera
incapable de faire cesser violences a leur égard. En
tout état de cause, il y a urgence à transférer dès
maintenant les personnes regroupées…1
-General Michel de Brébisson (Senior
Commander of French Armed Forces in
Algeria) to Minister of Armies Pierre Messmer,
August 1, 1962
On July 1, 1962, Algerians went to the polls in droves to vote on the following referendum:
“Voulez-vous que l’Algérie devienne un État indépendant, coopérant avec la France, dans les
conditions définies par la déclaration du 19 mars 1962?” Three months after 90.7 percent of
metropolitan French citizens had approved a similar referendum, 99.72 percent of the Algerians
who voted replied with a resounding “oui.”2 Two days later during a brief ceremony near Algiers
SHAT, 1K 744 (dossier: Situation des harkis en Algérie), General de Brébisson to the Minister of Armies (Cabinet),
“Objet: Ex-supplétifs menacés,” N° 1820/CSFAFA/EMI/MOR, Aug. 1, 1962. General de Brébisson was named
Commandant Supérieur des Forces Armées Françaises en Algérie on July 24, 1962.
2 The referendum held in France on April 8, 1962 asked voters: “Approuvez-vous le projet de loi soumis au peuple
français par le président de la République et concernant les accords à établir et les mesures à prendre au sujet de l’Algérie
sur la base des déclarations gouvernementales du 19 mars 1962?” The bill referenced above, voted into law on April 13
(Loi n° 62-421), maintained: “Article premier: Le président de la République peut conclure tous accords à établir
conformément aux déclarations gouvernementales du 19 mars 1962, si les populations algériennes, consultées en vertu
de la loi du 11 janvier 1961, choisissent de constituer l’Algérie en un État indépendant coopérant avec la France. Article
2: Jusqu’à la mise en place de l’organisation politique nouvelle éventuellement issue de l’autodétermination des
populations algériennes, le président de la République peut arrêter, par voie d’ordonnances ou, selon le cas, de décrets
1
70
(the star on the map in figure 2), the French High Commissioner in Algeria, Christian Fouchet,
handed the President of the Provisional Executive Government, Abderrahmane Farès, a letter from
President Charles de Gaulle that recognized the independent Algerian state. The Algerian
government (GPRA) symbolically selected July 5—the same day that the dey of Algiers had signed
the act of capitulation that began, in 1830, the era of French colonial domination—to celebrate its
victory. Joyful Algerians flooded streets across the nascent nation-state draped in their new green,
white, and red flag holding signs with slogans such as “Sept ans, ça suffit!” However, some of the
festivities turned violent, particularly in the western city of Oran (see figure 2) where haphazard
gunfire by OAS members, ALN solders, Algerian civilians, and French soldiers left seventy-six
Algerians and twenty-five pieds-noirs dead.3 The Algerian War for Independence—during which,
according to historian Benjamin Stora, five hundred thousand French and Algerian civilians and
soldiers perished—had ended over three months earlier, but the killing in Algeria was not over.4
Figure 2: Map of
northern Algeria. The
country’s three principal
cities are Oran in the
west, Algiers (the
capital), and Constantine
in the east. Source:
http://www.lib.utexas.ed
u/maps/africa/algeria_p
ol01.jpg.
pris en Conseil des ministres, toutes mesures législatives ou réglementaires relatives à l’application des déclarations
gouvernementales du 19 mars 1962.” JORF, Apr. 14, 1962, 3843. The abstention rate for the April referendum in France
was 24.4 percent and that for the July referendum in Algeria was a paltry 8.77 percent. Benjamin Stora, Histoire de la guerre
d’Algérie (1954-1962), new ed. (Paris: La Découverte, 2001), 78 and 84.
3 Fouad Soufi, “Oran, 28 février 1962, 5 juillet 1962. Deux événements pour l’histoire, deux événements pour la
mémoire,” in La guerre d’Algérie au miroir des décolonisations françaises. En l’honneur de Charles-Robert Ageron: actes du colloque
international, Paris, Sorbonne (23, 24, 25 novembre 2000) (Paris: Société Française d’Histoire d’Outre-Mer, 2000), 669.
4 Stora, Histoire de la guerre, 91.
71
Most of the violence carried out by Algerians after independence targeted the harkis. A
French government report issued at the signature of the Evian Accords approximated that 2,500
harkis were killed and 3,900 more were injured during the war.5 Estimates from historians and
reports by government officials about the number of harkis killed after the ceasefire range from tens
of thousands to one-hundred thousand, which does not include those who survived acts of torture.6
Some harki association leaders advance the highly improbable figure of 150,000 harkis killed,
originating from a report from the prefect of the Akbou district near Sétif, where 750 harkis had
been slain between the ceasefire and November 1962. The prefect, Jean-Marie Robert, estimated
that other regions probably had more fatalities so he increased the figure to one or two thousand,
which he then multiplied by seventy-two, the number of districts in Algeria. By his calculation, at
minimum 72,000 harkis were killed, and at maximum nearly 150,000.7 Lacking concrete data, this
debate will never be resolved. However, based on documentary evidence (particularly that from the
Army Archives and Minister of Armies Pierre Messmer’s private archival collection), several facts
are certain, all of which the Senior Commander of French Armed Forces in Algeria, General Michel
de Brébisson, referenced in the urgent August 1, 1962 telegram to Messmer cited above.
First, retaliatory actions by Algerians who believed their brethren to be traitors led to
exponentially more harki deaths after the March 19, 1962 ceasefire than during the war. This
violence was in large part due to unstable government structures in Algeria and the belief by FLN
and ALN leaders that the “Algerian Revolution” did not end with the ceasefire.8 As Messmer wrote
Ibid., 89.
Charles-Robert Ageron, the first scholar to exploit documents about the harkis in the Army Archives after the thirty
year waiting period expired in 1993, summarizes the varied estimates of harkis massacred advanced by government
officials, journalists, and military officers in his article: “Le ‘drame des harkis,’” 9-11.
7 Ibid., 10-11. This report was reprinted in Anne Heinis, “L’Insertion des Français Musulmans,” 22-31. A copy of the
report is also located in: ACNMF, 4/3, “Rapport rédigé par M. Robert, ex sous-préfet d’Akbou, actuellement sous-préfet
à Sarlat,” undated.
8 For example, the ALN commander of the fifth military region in Algeria proclaimed in a directive on March 8 that “Le
cessez-le-feu ne peut être et ne représente qu’une sorte de trêve militaire seulement, étant donné que politiquement nous
ne cesserons en aucune manière notre lutte infinie qu’est la révolution par tous les moyens. Durant cette trêve,
5
6
72
in a January 23 letter to a National Assembly deputy: “Il n’est pas possible de préciser le nombre des
harkis assassinés. Avec le repliement de nos forces et dans un climat de désorganisation, nous avons
éprouvé les plus grandes difficultés à suivre ce qui se faisait dans le bled et même dans certaines
villes.”9 These killings started during spring 1962 with two particularly violent periods. The first
began once France transferred its power to the Algerian government after the July 1 selfdetermination vote and the French Army could no longer legally intervene in Algerian internal
affairs. As a result, ALN members and civilians carried out, in the words of General de Brébisson,
an “épuration”—the term used to describe the “purge” of Vichy collaborators after the 1944
Liberation of France. At this same time during July and August 1962, factions within the Algerian
independence movement struggled for control of the government for nearly two months. The
disorganization left a power vacuum in the Algerian countryside as anarchy surfaced with no stable
government in place. Then, after a relatively calm month and a half with a new unified Algerian
governing structure in place, French troops accelerated their withdraw from Algeria in mid-October.
The decreased French presence—even if soldiers had only served a defensive role since the July
independence—meant that control of the countryside increasingly fell to the newly-constituted
successor to the ALN, the Algerian National Popular Amry (Armée nationale populaire algérienne
or ANP), founded on August 30 by former ALN chief Houari Boumediène. Many ANP members
did not uphold the clauses of the Evian Accords to protect individuals from sanctions as a result of
opinions expressed or actions taken during the war. In fact, the ANP imprisoned a significant
l’endoctrinement de nos unités et leur instruction restent primordiales. Nous devons former de nos hommes des
militants au sens propre du mot, des éléments essentiels et fondamentaux de la construction d’une Algérie nouvelle,
prospère et développée.” SHAT, 1H 1786/1, Le commandement de la wilâya 5, Directive n° 403, Mar. 8, 1962. Quoted
in Harbi and Meynier, Le FLN: Documents et Histoire, 849.
9 SHAT, 1K 744 (dossier: Point de la situation des Harkis), Letter from Pierre Messmer to Pierre Bas (National
Assembly Deputy for Paris), Nº 1814 MA/CC, Jan. 22, 1963.
73
number of harkis in detention centers owing to their service to France, and violence toward the
harkis did not markedly decrease until early 1963.10
Second, in addition to general statements like the one in the above telegram, Army officers
and commanders in Algeria sent to government officials in France countless documents describing
precise acts of torture and slayings. The first accounts of violence and threats toward the harkis
trickled in from officers stationed in the countryside, who recognized that they could do little to
protect the former soldiers and their families. Beginning in April 1962, officers related to their
superiors cases of torture (sometimes carried out publicly in town squares), murders, disappearances,
and rapes of harkis’ wives. Some officers even wrote to government leaders such as Prime Minister
Michel Debré, who subsequently sent a handwritten note to Messmer on May 17 stating that he was
“un peu ému” by the number of letters he had received detailing acts of violence toward the harkis
and the “très profonde tristesse” of soldiers because they could not save their comrades-in-arms.11
As the violence escalated in summer 1962, Army officials sent directly to Messmer itemized lists of
acts of torture, arrest, and execution with the date, place, and sometimes the names of individuals
and the gruesome details. One such document from General de Brébisson—located in Messmer’s
personal archives—enumerated 50 separate instances of violence across Algeria between July 11 and
August 6, some of which were perpetrated against as many as 250 harkis at a time (see Appendix B).
De Brébisson and other Army officers in Algeria did not believe that the government reacted to
their reports swiftly enough with concrete actions, demonstrating a gap between officials on the two
sides of the Mediterranean.
Finally, the documentary evidence offered in this chapter demonstrates that President de
Gaulle’s government implemented a policy of restricting the harkis’ access to metropolitan French
Ageron, “Le ‘drame des harkis,’” 5.
SHAT, 1K 744 (dossier: Directives concernant les harkis), handwritten letter from Michel Debré to Pierre Messmer,
May 17, 1962.
10
11
74
soil. As the third paragraph of the epigraph indicates, each harki waiting in the relocation camps
(camps de regroupement) set up to protect those whose lives were in danger needed individualized
authorization from Minister of Armies Messmer or Minister of Algerian Affairs Joxe themselves to
cross the Mediterranean. Messmer even issued a warning in mid-May that any harkis repatriated to
metropolitan France without his authorization would be sent back to Algeria.12 The restriction of the
number of harkis allowed across French borders had been established immediately after the ceasefire
and was constantly repeated in official documents emanating from Paris.13 Many of these directives
also reminded military personnel in Algeria to “strictly limit” government repatriation to “severely
threatened” harkis.
With stringent regulatory controls in place for the official repatriation process, Army officers
and French residents of Algeria—some of who had links to the OAS—took it upon themselves to
repatriate harki families whose lives the FLN violence endangered. Government officials believed
that these actions potentially threatened the security of post-imperial French metropolitan territory,
a territory where Algerians—pieds-noirs or harkis—were not welcome. Officials feared that the
harkis provided a fertile recruitment ground for the OAS, whose terrorist actions protesting de
Gaulle’s decision to end French Algeria included bombing civilian targets on Algerian and French
soil and eradicating Algerian farmland and government buildings with the “scorched-earth policy.”
Therefore, the government was circumspect of the harkis who arrived on French soil through
individual initiatives. In a May 1962 telegram Minister of the Interior Roger Frey wrote to Messmer
that the possible link between the OAS and harkis provoked “certain danger.” 14 This security
concern—soon applied to other harkis who arrived in France—caused the government to impose
SHAT, 1H 3077/2, Lieutenant Colonel Pochat to COMSEC Autonome Oran, telegram n° 97/RT/CAO/MOR, May
14, 1962.
13 For example, SHAT 1K 744 (dossier: Directives concernant les harkis), Minister of Algerian Affairs Joxe to High
Commissioner of the Algerian Republic Christian Delaballe, “Objet: Situation des personnes engagées en Algérie aux
côtés de l’Administration ou de l’Armée,” Apr. 18, 1962.
14 SHAT, 1K 744 (dossier: Directives concernant les harkis), letter from Minister of the Interior Roger Frey to Minister
of Armies Pierre Messmer, 2, N° SN/CAB 2924, May 15, 1962.
12
75
stricter control over France’s post-imperial borders. After the March 1962 ceasefire ended the
fighting with the FLN, the government’s attention fully shifted to controlling the raging FrancoFrench war with OAS members. This myopic focus on its “internal” conflict, which persisted after
the Algerian War for Independence’s ceasefire, obscured the grave safety concerns facing the harkis.
As discussed in the previous chapter, almost all harkis fell outside of the Franco-French war
owing to their colonial status, predicated on their Berber and/or Arab ethnicity, which was the first
obstacle to their obtaining French nationality. As per the July 21, 1962 ordinance about nationality
for residents of Algeria—which codified the omission of the harkis from the Evian Accords’
passages about French nationality—the majority of harkis and all other French with local civil status
needed to migrate to France and undergo an administrative procedure to become French citizens.
Consequently, restricting the harkis’ access to metropolitan soil limited their access to French
nationality, thereby creating a double bind. Put differently, possessing the superior colonial
classification of French with common civil status like all pieds-noirs (which amounted to a double
French-Algerian nationality) facilitated state repatriation. The government only guaranteed
movements of post-empire from Algeria to France for French citizens. Nevertheless, the harkis
could not obtain full nationality status unless they migrated to France owing to their (former)
colonial status. Repatriation and the special administrative procedure, therefore, were the second and
third obstacles to French nationality for the harkis. Not being accorded automatic French nationality
had immediate consequences: it created a refugee situation for harkis in Algeria, it meant that the
French Army could not legally intervene in Algerian internal affairs to stop violence toward the
harkis, and it interfered with their repatriation. It also had long-term consequences: the need to ask
for French nationality created psychological scars for who lost their French nationality after having
fought for France and has led many in the harki population to blame the French government for
76
abandoning harki soldiers who were subsequently massacred, a responsibility that harki associations
continue implore the government to take today.
This chapter follows the harkis in their quest for repatriation to and nationality in an
unfamiliar, unwelcoming supposed safe haven in France in the aftermath of the Algerian War for
Independence and in the midst of the Franco-French war that had a stranglehold on the
government’s attention. At this crucial moment, French government officials were learning to
acclimate to France’s newly-reconfigured map with eraser marks where its “oldest daughter” had
once buttressed the other side of the Mediterranean. The determination of French nationality based
on their former colonial status as inferior French citizens questioned the harkis’ present and future
relationship to the state as “French” and as “citizens.”
Distancing Harki Soldiers from France
Beginning in the summer of 1961 the French government enacted a policy of reducing the
number of harkis serving in the Army, which distanced them from France not only in terms of their
connection with the nation but also, eventually, geographically. As it became increasingly evident
that Algeria would gain its independence from France, the government recognized that not having a
relationship with the harkis at the time of the imminent ceasefire would lessen its responsibility
toward repatriating them. Toward this end, in July 1961 the Army stopped recruiting new harkis and
outlined its plans for the gradual reductions of its auxiliary forces.15 After an August 23 meeting of
the Committee for Algerian Affairs, President de Gaulle released precise figures to decrease the
number of harkis from 65,000 to 45,000 men by the end of 1961: 3,000 harkis would shift to the
military police, 500 to 600 harkis would move to another type of auxiliary force, the groupes mobiles de
sécurité (GMS), 3,000 to 4,000 harkis would integrate into the Army, and each month an average of
SHAT, 1H 2467/3 bis, “Remarques concernant la fiche B.3 sur les supplétifs.” On July 10, 1961, the État-Major
Interarmées (3ème bureau) wrote the original “fiche,” which is stapled to this document.
15
77
850 harkis, a few of whom would receive from the Army training for civilian jobs, would be sent
back to their villages.16
As this information indicates, the government decided to proceed cautiously when cutting
ties with harki soldiers, preferring to give them new assignments and provide them with job training.
According to several internal memos, government officials’ motives were not benevolent. Instead,
they foremost feared that abruptly ending the relationship with the harkis could put their 100,000
arms into the hands of FLN members, who could then turn these weapons on “French” soldiers.17
A July 1961 memo from the État-Major interarmées stressed: “Il faut que chaque supplétif porteur
d’une arme de guerre soit sûr: d’être défendu par nous, lui et sa famille, d’être entièrement indemnisé
et recasé s’il nous suit, d’être récompensé si nous le démobilisons et s’il nous remet volontairement
son arme pour reprendre sa liberté.”18 The Army rationalized that paying each former harki ten
thousand old francs to trade in their weapons would be less costly because it would avoid “very
grave danger” if the FLN took possession of their arms: “C’est bien plus de 10 milliards que nous
perdrions en prestige international, en confiance nationale, en vies et en biens français.”19 The
government’s primary concern, therefore, was safeguarding its own image and its “French” soldiers’
lives.
Another Army document written five months later confirms the government’s skepticism of
the harkis’ allegiance to France by stating that its course of action to provide the harkis with
concrete benefits would prevent “une éventuelle ‘révolte des Cipayes.’”20 The government feared a
SHAT, 1H 1397/1, Charles de Gaulle, “Comité des Affaires Algériennes, Séance du Mercredi 23 Août 1961, Relevé
des décisions.”
17 SHAT, 1H 2467/3 bis, État-Major Interarmées, 3ème bureau, “Fiche,” July 10, 1961.
18 Ibid.
19 Ibid.
20 SHAT, 1H 1397/8, “Problèmes posés par les harkis au moment du ‘Cessez-le-feu,’” Dec. 14, 1961. Several newspaper
articles in the spring of 1962 referenced this revolt, including: “De l’entraide à l’utilisation des Harkis en métropole,” Le
Monde, May 23, 1962 and Jean Planchais, “D’anciens supplétifs et leurs familles devront être protégés lors des premiers
remous de l’indépendance,” Le Monde, June 21, 1962. Both articles can be found in the Sciences Po Dossiers de Presse
(hereafter SPDP), dossier: Harkis.
16
78
rebellion like the 1857-58 mutiny of the native Indians who fought alongside British troops for the
East India Company’s Army in its quest to conquer India—and against their fellow Indians. After
years of inequitable treatment from their British commanders, in May 1857 the native Indian Cipayes
(the Hindi word for “soldiers”) protested that placing the new bullets greased with cow and pig lard
into their mouths to remove the shell casing violated their Hindu and Muslim religions. The East
India Company Army and the Cipayes violently clashed for nine months until the British government
decided to dissolve the Company and take control of India. This comparison, which weighed on
some Army officials’ minds, demonstrates a fundamental mistrust of the harkis. Despite the
appearance of helping these soldiers, the government’s actions were primarily a means to continue
to watch over the harkis—and their weapons.
As the ceasefire neared the government remained reluctant to repatriate harkis and their
families to France, which would block them from French nationality. On March 15, 1962, the Army
headquarters in Algeria received a telegram from the National Defense Headquarters in Paris with
Prime Minister Michel Debré’s instructions explaining the crucial link between repatriation and
nationality for all French with local civil status, a category that included harkis. The telegram
responded to the following question that Army officials in Algeria had posed “Les français
musulmans de droit civil local perdront-ils automatiquement la nationalité française d’après résultat
autodétermination qui se prononcerait en faveur indépendance en Algérie?” by declaring “Oui, s’ils
demeurent en Algérie.” 21 These instructions clearly conveyed that without repatriation these
harkis—whose nationality status, according to the Evian Accords, was no different from that of
other Algerians—were ineligible to become French citizens.
While some government officials recognized that there needed to be a repatriation
procedure in place for the “most threatened” harkis, documents from February 1962 onward
SHAT, 1H 2467/6, DefNat Paris to Genesuper Reghaia, telegram Nº 231/DN/COD/A/TS, 1, received Mar. 15,
1962.
21
79
constantly stressed that this would be a last option. Minister of Interior Frey’s cabinet director
reported to his boss that Messmer had contended during a phone conversation in early February
that he did not believe it necessary to prepare a plan for the harkis’ repatriation.22 “Return to civilian
life,” the heading for a plethora of memos issued by the Army and other government officials in
Algeria and France, was clearly the government’s preference.23 On March 8, 1962, Messmer wrote an
informational memo giving the commanding officers of the Army, Navy, and Air Force instructions
about the course of action to take regarding their “subordonnées de tous grades et de toutes origines”
following the ceasefire.24 While he announced that the pieds-noirs and the “musulmans attachés à la
France” would have the opportunity to choose whether to settle in France or Algeria, his next
sentence contradicted this pledge: “…[I]l est hautement souhaitable que la grande majorité des
Algériens décident de continuer à vivre dans leur pays natal et d’assurer ainsi une réalité aux liens
d’association entre la France et l’Algérie.” 25 This internal government discourse demonstrated
apprehension over an influx of Algerians, whether pieds-noirs or harkis. Messmer maintained that a
mass departure of pieds-noirs would run counter to the principle of “cooperation”—a word that
appeared in the Evian Accords ten times—because such a migration would “vider l’Algérie des
cadres nécessaires à sa vie et à son développement.”26
On the contrary, the chief reason Messmer offered for why it was “highly desirable” for the
harkis not to migrate to France focused on their “adaptation brutale en France,”27 that is, the
presumed inability of these men—98 percent of whom were illiterate, according to an April 1962
AN, F1a 5140, R. Morice, “Note à l’attention de Monsieur le Ministre,” N˚ 166 AGA/SEC, Feb. 13, 1962.
For example, SHAT, 1H 2028/5, Chief of the General Delegation in Algeria, “Objet: Retour des supplétifs à la vie
civile,” Mar. 17, 1962.
24 SHAT, 1H 2467/6, Minister of Armies Pierre Messmer to the Commanding Officers of the Army, Navy, and Air
Force, 1, Nº 106/MA/CAB/INF/CS, Mar. 8, 1962.
25 Ibid., 2.
26 Ibid.
27 Ibid., Annexe 2, p. 3.
22
23
80
French government report 28 —to adjust to metropolitan life. In an August 1962 document
summarizing the French government policies toward the harkis since the ceasefire, the supreme
commander of French Forces in Algeria advanced a similar explanation for limiting their
repatriation: “Cette politique… est destinée à… empêcher l’arrivée en Métropole de personnes
inadaptables destinées à devenir des épaves.” 29 This disparaging assessment of the harkis as
“derelicts” clearly indicated that the government did not want the harkis to “return” to metropolitan
France because they could offer nothing to, and would be a drain on, French society. Or, as another
document from April 1962 put it, only those showing “worth” should be repatriated: “Autorité
militaire doivent [sic] être persuadées [sic] qu’avenir des personnes prises en charge doit être digne. Il
ne peut être question de déraciner des FSNA inadaptables appelés à devenir des épaves sur le sol
français.”30 The use of “digne” kept alive colonial discourse. “Dignité,” a codified legal term, had
been one requisite quality—and “an instrument of colonial domination,” according to Emmanuelle
Saada—for Algerian natives applying for full French nationality.31
Violence and “Repatriation”
Despite top-ranking government officials’ circumspection toward repatriating harkis and
their family members, on February 21, 1962 Messmer convoked an inter-ministerial commission
“concernant le rapatriement éventuel des personnels placés sous le contrôle des autorités militaires,”
which would report to Prime Minister Debré. Its members comprised Minister of Algerian Affairs
Joxe, Minister of Armies Messmer, Minister of Agriculture Edgard Pisani, Secretary of State for
Repatriates Robert Boulin, Minister of the Interior Frey, and Minister of Labor Paul Bacon. The
ACNMF, 5/7, “Procès-verbal de la réunion tenue le mardi 10 avril 1962 et concernant le rapatriement éventuel de
personnels musulmans placés sous le contrôle des autorités militaires,” 3.
29 SHAT, 1H 1793/1, General de Brébisson, “Objet: Situation des ex-harkis en Algérie depuis le cessez-le-feu,” Nº
5555/31/2d, Aug. 13, 1962.
30 SHAT, 1H 3077/2, Sergeant Vella, “Objet: Recasement en France supplétifs et civils FSNA engagés aux côtés F.
Armées,” Nº 1050/CSFA/EMI/MOR, Apr. 16, 1962.
31 Saada, “The Empire of Law,” 101.
28
81
Prime Minister’s selection for the president of the Commission fell to a pro-French Algeria Council
of State member who had impressed Gaullist figures, most notably Debré, by writing a pamphlet in
1957 defending the French government against charges that it condoned the French Army’s use of
torture.32 Michel Massenet had served since January 1959 in the interministerial post of “délégué à
l’action sociale pour les Français musulmans d’Algérie en métropole.” In this capacity, he was the
only high-level French government official representing the interests of Algerian Muslims. Massenet,
therefore, was familiar with the challenges facing Algerian natives as they tried to integrate into
French society. The Commission’s charter introduced its assignment to prepare “step by step” plans for rehousing the harkis in France.33 More specifically, Prime Minister Debré tasked the Commission
members with compiling information to respond to nine concrete questions about “the refugees,”
which can be divided into two broad categories.34 The first set of questions addressed the harkis’
present situation in Algeria: the number of harkis to be repatriated and how to protect them from
the likely violence perpetrated by FLN members. The second group focused on the harkis’ future in
France: their initial and permanent housing, what jobs would be appropriate given their lack of
professional skills, and how to facilitate their integration into French society. At the same time, the
charter made clear that the Massenet Commission’s authority was limited to “studying” potential
housing and jobs for the harkis. It stipulated that neither the Commission as a whole nor any of its
Sylvain Laurens, “La noblesse d’État à l’épreuve de ‘l’Algérie’ et de l’après 1962” Politix 76.4 (2006): 89. Massenet’s
pamphlet was titled Contrepoison, ou, la morale en Algérie (Paris: Grasset, 1957).
33 ACNMF, 5/7, “Procès-verbal de la réunion tenue le mardi 10 avril 1962…,” 1.
34 Ibid., 2. The specific questions comprised: how many harkis wanted to migrate to France; how to protect harkis from
FLN members who were threatening their lives; how to ensure the principle of free circulation for harkis to migrate to
France; how to transport them; where to house them when they first arrived in the metropole; how to facilitate social
networks to help them adapt to metropolitan life; what the modalities would be for providing job training for the men,
“adaptation” classes for the women, and schooling for the children; how to find them jobs appropriate for their
professional skills; and what type of permanent housing would be possible.
32
82
individual members would have the power to implement its own suggestions, which would later
serve as a point of contention between Massenet and Debré.35
The Commission met four times before, on April 10, Massenet issued an eleven-page singlespaced document summarizing the information gathered by each ministry to address the specific
questions in its domain. Massenet also presented his suggestions about how to remedy the “harki
problem.” The report’s level of detail demonstrated the extensive research that committee members
and Massenet himself had undertaken. According to surveys that the Army conducted, the harkis’
median age was twenty-nine years old and 52 percent of these men were heads of families with three
children on average.36 Massenet further relayed that the Minister of Agriculture estimated that 1,000
to 1,500 men could find permanent positions as farm hands; 1,000 to 1,500 men could be forestry
workers; and 1,000 men could become farmers.37 Massenet’s tone in the report reinforced the harkis’
exigent circumstances. In addition to employing terms such as “sévérité” and “situation dramatique”
several times in the document, derivatives of “grave” and “menacer” appeared four and eleven times
respectively, and the word “problème” was used eighteen times. His words were powerful indicators
of the fate that would befall not only the harkis in Algeria whose lives were threatened by FLN
members but also those who migrated to metropolitan France and were placed in camps by the
government.
In the report’s conclusion, Massenet’s words forcefully communicated three points. In the
first section, which he labeled “Gravité et urgence du problème à resoudre,” he argued that the
government was not taking seriously enough this time-sensitive “problème de vie et de mort.”38
Notably, Secretary of State for Repatriates Boulin, whose ministry was in charge of welcoming these
Muslim repatriates to France, declared at the end of the meeting that the “harki problem” did not
Ibid., 1-2.
Ibid., 3.
37 Ibid., 7.
38 Ibid., 9.
35
36
83
really exist because none would agree to come to France.39 This miscalculation by the official under
whose jurisdiction fell the harkis—as repatriates—had serious consequences for the government’s
organization of the their arrival in France. While his Ministry dedicated its resources to the
reclassement of the repatriates of European and Jewish origins by aiding them look for jobs and
housing, it did little to prepare for the harkis’ migration to France.
Massenet then exposed the “severity” of violence by FLN members toward the harkis and
instances of FLN operatives blackmailing harkis’ families, both of which the Army had failed to
protect them from. Without detailing specific incidents, Massenet had revealed earlier in the report
that “multiple reliable sources” had related the great extent of the threats to the harkis’ lives.40
Meanwhile, other accounts divulged concrete incidents of Algerian subprefectures routinely denying
harkis the necessary authorization they required to travel to France to escape this violence, a
violation the of Evian Accords’ free circulation clause. 41 According to Massenet, the French
government proved inefficacious in the face of this situation because it had neither committed any
money to fund, nor taken any steps to organize, repatriating harkis. He averred that “Il faudrait un
miracle pour que l’organisation et le financement du repli des réfugiés musulmans en France puisse
être pris en charge par le secrétariat d’État aux Rapatriés si nous ne disposons que d’un délai de 2
mois avant la date fatidique de l’autodétermination.”42 Put simply, the government was not following
through on its promise in the February 21 press release to provide a concrete plan to ensure, or
welcome, the harkis’ repatriation to France.
Massenet then presented the “Mesures immédiates à prendre pour redresser la situation”
since the government had failed to provide a concrete repatriation plan and allocate the funds to
CAC, 19910097/40 (dossier: Harkis: Correspondance 1962), “Compte rendu de séance sur les harkis, Commission
Interministérielle Massenet, 10 avril 1962,” 4.
40 ACNMF, 5/7, “Procès-verbal de la réunion tenue le mardi 10 avril 1962…,” 3.
41 Ibid., 4.
42 Ibid., 10.
39
84
officials in Algeria that would be necessary to remedy the harkis’ dire situation.43 For him, it was a
foregone conclusion that repatriation was the only option. However, the requisite means to achieve
this would challenge the harkis’ status as repatriates because these measures would define the harkis
as refugees. Massenet proposed that the Army open relocation camps in Algeria as way stations for
the harkis whose lives were in peril while they awaited safe passage to metropolitan France.
Significantly, these would be located on the same grounds—and be classified under the identical
name—as the relocation camps that had been used for FLN prisoners and Algerian refugees forced
from their homes during the war. Massenet acknowledged that these actions would provoke
“conditions certes regrettables sur le plan psychologique…” 44 During the war, native Algerian
auxiliary police forces in the French Army (groupes mobiles de protection rurale) had watched over FLN
prisoners in these very same camps.45 Now, French soldiers watched over the powerless former rural
“protection” police forces to safeguard them from FLN attacks.
The next step in developing a repatriation plan, according to Massenet, was to inventory
where to house the harkis and where to find them jobs in metropolitan France. With carefully
chosen words that in themselves related the urgency of the situation, he argued that the government
had not put enough time into finding satisfactory solutions to these serious problems. Massenet
wrote in the body of the report, “les moyens d’accueil disponibles doivent-ils être considérés comme
dérisoires si des mesures immédiates ne sont pas prises pour en créer de toutes pièces.”46 In the
weeks following this meeting Ministry of Agriculture officials inquired among prefects in southern
France about housing possibilities for harkis in their departments, to which Agriculture officials
received primarily negative responses. In the end, the national government failed to adequately plan
Ibid.
Ibid., 4.
45 Sylvie Thénault, “Personnel et internés dans les camps français de la guerre d’Algérie,” Politix 69 (2004): 71. Thénault’s
article provides an excellent analysis of internment camps in France and Algeria during the Algerian War for
Independence.
46 ACNMF, 5/7, “Procès-verbal de la réunion tenue le mardi 10 avril 1962…,” 6.
43
44
85
for the harkis’ migration to metropolitan France and local government officials were unable—or
refused—to offer land to house harkis. Consequently, the government opted to house them in
former FLN prisoner camps in metropolitan France, such as Larzac in the Aveyron department, and
former refugee and prisoner of war camps, such as Rivesaltes in the Pyrénées-Orientales. The
significance of relegating the harkis to such sites provoked a lasting exile owing to their isolation and
exclusion from the surrounding community, the subject of section two of this dissertation.
Massenet’s final, and most forceful, point bookended his two-page conclusion to the report,
which he believed would “clôt la première phase des travaux de la commission.”47 Secretary of State
for Repatriates Boulin, responsible for managing the harkis’ situation along with that of the piedsnoirs, could not adequately address the former problem because the harkis were “refugees” and not
“repatriates,” words that he underlined in the report’s second to last sentence: “…il faudrait placer
auprès du Haut-Commissaire un fonctionnaire chargé de traiter le problème des réfugiés en
attendant que ceux-ci reçoivent, peut-être seulement dans la suite et sur le territoire métropolitain le
statut de rapatriés.”48
Not only did the government reject Massenet’s twofold recommendation to organize a
repatriation plan to address the harkis’ unique situation and appoint him to ensure that there would
be someone representing the harkis’ specific interests but it also never convened the Commission
again. The first phase of its work was its last. One week after submitting his report to Prime Minister
Debré, Massenet attached a copy to a handwritten letter to the Director of the National Police
(Sûreté Nationale), Jacques Aubert, who had spent two years leading the police forces in Algeria
before returning to the metropole to serve as director of the Minister of Interior’s cabinet. Massenet
believed that Aubert, intimately acquainted with the brutality of FLN members in Algeria, was in a
better position than anyone else in the government to assess the “definite and real” threats that
47
48
Ibid., 1.
Ibid., 9.
86
faced the harkis and could convey these to government officials. 49 Massenet insisted that
humanitarianism was not the sole reason that the government needed to repatriate the harkis. Citing
the example of France having provided a safe haven to Hungarians when the Soviet Union invaded
their homeland in 1956, he argued, “…il s’agit avant tout d’accueillir sur notre sol de malheureuses
épaves, conformément à une constante tradition de notre pays.”50
Aubert’s response reveals important insights into the divergence between government and
Army officials about the risks to harkis’ safety from FLN members and the French government’s
motivations for not enacting a repatriation plan to welcome harkis to France. Massenet did not find
an ally in Aubert. The Director of the National Police minimized the FLN violence toward the
harkis, believed that only a few hundred harkis would need to be repatriated, and wanted any
operations to evacuate the harkis to be led “in secrecy.” 51 Aubert claimed that his previous
experiences in Algeria led him to believe that FLN members would not break the “détente entre les
familles politiques musulmanes,” that is, between them and former harkis. Reports that he recently
received from civilian administrators in Algeria attested “almost unanimously” to an easing of
hostilities.52 If Aubert was indeed accurately portraying the prefects and sub-prefects’ opinions, then
these reports reveal a discrepancy between Ministry of Interior employees and military men who
increasingly informed their superiors of FLN threats. For example, on April 26, one Gendarmerie
Corps commander wrote a memo to the generals in charge of the Gendarmerie in Algiers, Oran, and
Constantine, signaling that a questionnaire had been sent to each Muslim family, which needed to
answer questions about its pro-FLN or pro-French activities. According to this officer, “La
population entière vit un régime de terreur. Il y aurait de nombreuses exactions commises (roues de
coups, règlements de compte), mais les victimes n’osent pas se plaindre, tant est grande la peur du
CAC 19910467/2, handwritten letter from Michel Massenet to Jacques Aubert, Apr. 17, 1962.
Ibid.
51 CAC, 19910467/2, letter from Jacques Aubert to Michel Massenet, Apr. 25, 1962.
52 Ibid.
49
50
87
FLN.”53 This report, which reached the upper echelons of the French Army, contradicted not only
Aubert’s assertions about the détente, but also scores of firsthand accounts of violence toward the
harkis that officers related to their commanders, which will be discussed below.
The reasons that Aubert offered to Massenet for why the Army should call on authorities
from the newly-created Provisional Government of the Algerian Republic (GPRA) to curb the
violence instead of giving “the exodus signal” suggest that the government’s refusal to come to the
aid of harkis was a purposeful political maneuver.54 Aubert listed several motivations for not doing
so. First, he was concerned for the safety of the “troupes de choc” that would need to rescue the
harkis. Second, he underscored the importance of France demonstrating faith in the GPRA to
uphold the Evian Accords statutes that guaranteed the safety of Algerian inhabitants. Third, he
emphasized the Gaullist regime’s goal of building strong relations between France and the Muslim
world with the global post-imperial reconfiguration. And finally, he warned of the necessity for
“tomorrow’s Algeria” to become subdued.55 Aubert expounded on why a “spectacular manoeuver”
to bring the harkis to safety in the metropole would disrupt peace not only in Algeria but also, and
more importantly, in France:
Derrière nos collègues musulmans, une tendance raisonnable à ses chances de modeler le
visage de l’Algérie de demain. Ne commençons pas prématurément par annoncer notre
impuissance, décourager les adaptations qui ne demandent qu’à se faire et orienter
délibérément des ‘buveurs de soleil’ sur nos vertes mais humides campagnes.
Et puis, ne soyons pas dupe d’un parti du pire en France qui veut cela pour terminer sur
un constat de faillite le chapitre de l’histoire de notre République.56
If repatriated en masse to France, government officials would face the challenge of helping the
harkis adapt to a French society to which, many felt, they had little to contribute. Aubert exhibited a
contemptuous attitude for the harkis through his use of “buveurs de soleil,” a pejorative label given
SHAT, 1H 1793/1, Captain Marrel (Commandant la Compagnie de Gendarmerie) to the Generals commanding the
Gendarmerie in Algiers, Constantine, and Oran, Nº 232/4, Apr. 26, 1962.
54 CAC, 19910467/2, letter from Jacques Aubert to Michel Massenet, Apr. 25, 1962.
55 Ibid.
56 Ibid.
53
88
by the pieds-noirs to idle Algerian men who whiled away hours in the sun. This phrase oddly
contrasted with Aubert’s description of FLN members, whom he identified in the previous sentence
as “our colleagues” and in whom he was placing great faith to uphold the Evian Accords. Therefore,
he portrayed the harkis who fought for France in a more negative light than the fellaghas who fought
against France.
However, more significantly, Aubert’s words reinforced the notion that the French
government wanted to will a peaceful new Algeria—the “Algérie de demain” showcased in the
March 1962 pamphlet destined for the pieds-noirs—to quell the fallout from the two-front war, that
with the FLN and that with extremist elements of the pieds-noirs.57 And it was the Franco-French
conflict that Aubert was foremost concerned about. He argued that admitting the harkis’ lives were
in danger would present the appearance that France was weak and would give fodder to the piedsnoirs who vehemently opposed France’s decision to stop fighting for French Algeria. The violence
perpetrated by those that he designated as the “parti du pire” was rampant in Algeria, from the
“Rock and Roll” Operation in which OAS members set 120 bombs to explode over two hours on
March 6, to the slaying of six Ministry of Education inspectors on March 15, to the OAS’s arbitrary
decision to target pharmacy assistants for assassination on March 17.58 Any sign of the French
government’s failure or frailty would open the door to the “migration” of these widespread acts of
terrorism to metropolitan France.
Aubert’s prediction that the “détente entre les familles politiques musulmanes,” would
continue proved completely inaccurate. In fact, while chiefly targeting harkis, the FLN also
committed widespread terrorist acts across Algeria that affected all inhabitants, including Muslim
57
58
See pages 66 and 67 for an image and description of this pamphlet.
Thénault, Histoire de la guerre, 249-250.
89
civilians, pieds-noirs, and French Army soldiers.59 From the end of April onward, eyewitness reports
of the FLN’s repression of the harkis streamed in to superior Army officers in Algeria and
government officials across the Mediterranean. Initially, the FLN threatened these individuals and
tried to exclude them from the new Algerian society. For instance, one harki wife relayed to Army
officials in mid-May that her family and other harki families were unable to sustain themselves
financially because the FLN would not hire former harkis and forbade harki men to work for piedsnoirs “sous peine de sanctions graves.”60 Pieds-noirs who related that the Muslims refused to work
for them and would not provide an explanation corroborated these facts.61
Another harki who resided in Rouached, a town in the Aurès mountains fifty kilometers
southeast of Algiers, wrote a letter that reached Ministry of Interior officials in Paris detailing the
violence and threats in his region. In this May 4 letter, he repeatedly argued that it was impossible
for the harkis to remain in Algeria. He relayed that fellaghas requisitioned lands belonging to
members of the moghazni auxiliary forces and destroyed their houses, leading former auxiliary troop
members like himself to be certain that, without any defense, they would be killed. The harki told of
three harkis and two Muslim civilians from his village whom the FLN captured and tortured for four
days. These men were paraded naked through neighboring towns where, following the fellaghas’
orders, inhabitants beat and spat on them. He never saw four of these individuals again, although
one of the Muslim civilians was given his freedom after proving during a four-day interrogation that
he had never fought for the French Army. The letter further revealed that harkis’ wives were taken
For example, during the course of his rounds near the gendarmerie headquarters in Oran, one military policeman
found a threatening letter written on April 26. The document signed “FLNA” (the FLN of Algeria) and “ALNA” (the
ALN of Algeria) warned in error-ridden French: “FLN vous envoi un lettre, poure vou dire deux disparaetre ou si Non
Nou salon vou tue tous grand et petit. Set nautre preux miere avertise man separe l’ordre deux FLNA vive vie l’algerien
ondependant et vive FLN.” (“FLN vous envoie une lettre, pour vous dire de disparaître ou, sinon, nous allons vous tuer
tous, grand et petit. Cette nature première avertissement sépare l’ordre du FLN. Vive la vie algérienne indépendante et
vive le FLN.”) On the back, a drawing of a head with a blood flowing from its ear and a red cross on its forehead
promised future violence. SHAT, 1H 3077/2, Chef d’Escadron Coadic au Commandant du sous-secteur Ouest,
“Bulletin de Renseignements,” N˚ 958/B.2/SC, May 11, 1962.
60 SHAT, 1H 3077/2, Chef d’Escadron Coadic, “Objet: Agissement du FLN à l’égard des populations rurales,” N˚
997/B.2/SC, May 19, 1962.
61 Ibid.
59
90
each night into the countryside where fellaghas raped them and that FLN members demanded large
sums of money from former harkis.62
By means of such reports from both harkis and French soldiers throughout April and May,
French government officials were cognizant of the escalation of FLN violence. Yet the government
still did not start repatriating harkis until mid-June. As mentioned in the introduction to this chapter,
on May 17 Prime Minister Debré sent to Messmer a handwritten note, which signified a heightened,
personal level of attention about the harkis’ situation. He stated that he was moved by the number
of letters he had received describing violent acts committed against the harkis and relaying the
profound sadness of French military officers unable to defend the former FMA soldiers. 63
Messmer’s response two weeks later outlined the steps that the French Army was taking to address
the harkis’ precarious situation and once again emphasized that the Army was preparing to transfer
to metropolitan France only those who were judged to be “particulièrement menacés.” 64 This
phrasing that stressed repatriation as a last resort—for example, “dans les cas urgents,”65 “harkis et
civils FSNA gravement menacés,”66 “en nombre limité,”67 and “celles réellement menacées”68—was
continually repeated in memos emanating from Ministry of Foreign Affairs and Ministry of Armies
officials after the ceasefire.
The first step the Army took toward determining whether individuals’ lives were being
threatened was to conduct a census among as many harkis as possible to establish who wished to
CAC, 19910467/2 (dossier: Correspondance avec M. Massenet), Anonymous letter, Rouached, May 4, 1962.
SHAT, 1K 744 (dossier: Directives concernant les harkis), handwritten letter from Michel Debré to Pierre Messmer,
May 17, 1962.
64 SHAT, 1K 744 (dossier: Directives concernant les harkis), letter from Pierre Messmer to Michel Debré, June 2, 1962.
65 SHAT, 1K 744 (dossier: Directives concernant les harkis), letter from the Minister of State for Algerian Affairs to the
High Commissionner of the Algerian Republic, “Objet: Rapatriement éventuel des Français-Musulmans engagés aux
côtés des Forces de l’Ordre,” 2, N˚ 44 API/POL, Apr. 11, 1962.
66 SHAT, 3077/2, Sergent Vella, “Objet: Recasement en France supplétifs et civils FSNA engagés aux côtés Forces
Armées,” Message N˚ 1050/CSFA/EMI/MOR, Apr. 16, 1962.
67 SHAT, 1K 744 (dossier: Directives concernant les harkis), letter from the Minister of State for Algerian Affairs to the
Minister of Armies, “Objet: Situation des personnes engagées en Algérie aux côtés de l’Administration ou de l’Armée,”
N˚ 515 API/POL, Apr. 19, 1962.
68 SHAT, 1H 3077/2, Lt. Colonel Pochat to COMSEC autonome Oran, N˚ 97/RT/CAO/MOR, May 14, 1962.
62
63
91
migrate to France. This census was based both on the harki’s desire for repatriation and qualitative
assessments from military officers about his service. According to a memo from the Army chief of
staff, General Hublot, officers were to fill out a questionnaire for each male, including information
about his current employment and job skills, his marital status, the immediate family members to be
repatriated with him, his professional aptitude, and a succinct explanation of how his life was being
threatened.69 The government also solicited recommendations concerning the harki’s request for
repatriation both from officers who had worked directly with the former soldier and the officers’
superiors. The memo urged those deciding whether to endorse a candidate for repatriation to keep
in mind that “la réadaptation dans un pays, un climat, une ambiance sociale différente de ceux qu’ils
connaissent, constituera pour les FSNA une épreuve difficile et qu’elle ne doit être envisagée que si
le maintien en Algérie se révèle impossible.” 70 However, many of the recommendations for
repatriation from military officers of all grades did not address the harki’s potential for adaptation to
metropolitan France. Instead, in addition to describing the concrete threats to the harki’s life (such
as family members killed by FLN members), the officers emphasized the individual’s attachment to
France. For example,
Depuis juillet 1957 combat pour la France de toutes ses forces et de toute son âme.
Exemple parfait de Musulman Francophile.71
Avis très favorable à cette demande d’installation en France métropolitaine. Le loyalisme
et le dévouement de ce harki pour la France, se fasse de tous, les commentaires. Malgré les
injures et les menaces dont il est l’objet, il reste attaché à la France et à son drapeau. Cet
emblème pour lequel il a déjà tout donné.72
SHAT, 1H 4214/1, General Hublot (Chef d’État-Major) to the General commanding the “Région Territoriale” and
the Oran Army Corps, “Objet: Recasement en France de militaires supplétifs et civils FSNA,” 2, N˚
1013/CSFA/EMI/MOR, Apr. 11, 1962.
70 Ibid.
71 In accordance with confidentiality laws for classified archival documents, “CM” are the harki’s initials. SHAT, 1H
1260/3, Rapport du Lieutenant COMBESSI (16e RI de Marine, 1˚ Bataillon CCAS Commando V 221) sur les services
rendus par le Sergent Chef CM, N˚ 84/CDO, Apr. 9, 1962.
72 SHAT, 1H 1260/3, Général de Brigade MULTRIER Commandant la SEC et la 2˚ SIM à Monsieur le Général de
Division, Commandant la Région Territoriale et le CAC, May 15, 1962.
69
92
Fidèle serviteur de notre cause le harki MA fortement compromis risque d’être victime
de représailles.73
These appraisals underscoring the harkis’ devotion to France as a reason for repatriation—a crucial
step toward French nationality—echo the language in reports that civil administrators needed to
submit follow the 1865 sénatus-consulte to approve on a case-by-case basis, as explained in the
previous chapter, Algerian natives’ applications for French nationality. One such example from 1890
reads: “M. est un très bon officier indigène très digne, d’une conduite exemplaire, excellent serviteur
et animé de très bons sentiments français.”74 Civil administrators in colonial Algeria and Army
officers in 1962 alike argued that serving devotedly under the French flag should earn Algerian
natives access to French nationality.
After the three Army Corps (Algiers, Oran, and Constantine) and the Sahara region finished
the census in late April, a May 10 memo from Army headquarters in Algeria evaluated the maximum
number of “supplétifs et civils FSNA engagés aux côtés des Forces Armées” and their family
members to be repatriated at 7,006 (representing 1,334 heads of families).75 Approximately two
thousand individuals were moghaznis76 and civilians (such as domestics for pied-noir families), who
fell under the authority of the Ministry of Interior, and slightly under five thousand were harkis for
SHAT, 1H 1260/3, Le Lieutenant-Colonel Marois à Monsieur le Général Comandant la 14˚ DI – EM – B 3 PH, 1, N˚
572/SPH/3/PH, Apr. 30, 1962. “MA” are the harki’s initials.
74 Blévis, “La citoyenneté française,” 40.
75 SHAT, 1R 336/8 (dossier: Rapatriement des harkis), CSFA, EMI, Bureau du Moral, “Recensement des supplétifs et
civils FSNA engagés aux côtés des Forces Armées et dont le transfert en France est envisagé (à la date du 1er mai 1962),”
N˚ 1266/CSFA/EMI/MOR, May 10, 1962.
76 As mentioned in the introduction to this dissertation, the moghaznis were civilian forces charged during the war with
protecting and aiding the sections administratives sociales (SAS), local French civil administrators who served as mediators
between the Algerian population and state institutions through a dynamic of close police control or encadrement. The
functions of these Ministry of Interior employees comprised collecting taxes, monitoring elections, supplying free
medical care, providing schooling and job training to children, overseeing construction projects and improvements in
agricultural systems, running the relocation camps for displaced Algerians, and serving as intelligence agents seeking out
individuals suspected of anti-French actions. Owing to the moghaznis’ close daily contact with the Algerian population,
which made them easily identifiable, they were particularly at risk for retributions from Algerian insurgents due to their
complicity with the French cause. For more information about the role of the SAS and the moghaznis during the war, see:
Charbit, Les harkis, 17-22.
73
93
whom the Ministry of Armies was responsible.77 The memo further predicted that these numbers
would diminish because some of these men were likely to change their minds and decide to remain
in Algeria, which indeed was the case initially. Army officials completed a second census a mere 10
days later and determined that 2,298 less people, or one-third of the initial total, had reconsidered
their decision to leave Algeria as episodes of FLN violence were relatively isolated, at least then.78
“Harkis” As “Rapatriés”?
To facilitate and prepare the “repatriation” of harkis and their family members to France, the
Army followed Massenet’s suggestion to reopen the relocation—or refugee—camps. Over the
course of the Algerian War for Independence, these sites had housed three million displaced persons
(representing one-half of the rural population).79 Government discourse in spring 1962 consistently
referred to harkis as “refugees,” an incorrect classification that government officials continued to use
after harkis and their family members arrived in France. This descriptor had an impact on how state
agents and the general public viewed their French citizenship.
The term “refugees,” moreover, contradicted the legal truth that the harkis who migrated to
France were “repatriates.” The harkis qualified for the December 26, 1961 “Loi relative à l’accueil et
à la réinstallation des Français d’outre-mer,” named after Robert Boulin, who occupied the newlycreated cabinet post of Secretary of State for Repatriates since August 24, 1961.80 This legislation,
developed with French repatriates from Algeria in mind, allocated to French citizens returning from
Numerous documents, including Messmer’s June 2 letter to Debré, offered this approximate breakdown of “proFrench Muslims” to be repatriated. Another example includes AN, F1a 5140, Cabinet director for the Secretary of State
for Repatriates, “Note sur le rapatriement des Français-Musulmans,” 1, June 20, 1962.
78 SHAT, 1R 336/8 (dossier: Rapatriement des harkis), CSFA, EMI, Bureau du Moral, “Recasement en France des
supplétifs et civils FSNA. État d’avancement des opérations de regroupement initial au 10 mai 1962,” N˚ RR/YR-15.5,
May 16, 1962.
79 Kateb, Européens, “indigènes” et juifs, 315-16.
80 After a lengthy exchange, Boulin confirmed to Messmer in late March 1962 that the law legally applied to the harkis.
SHAT, 1R 367/7, Minister of Armies Messmer to the Senior French Commander in Algeria, Nº 10074/MA/CAB/DIR,
Mar. 29, 1962.
77
94
all of its former colonies and protectorates social and economic benefits to support their integration
into metropolitan society.81 However, the status of “repatriate” did not guarantee harkis—Algerian
Muslims—access to the repatriation process, a necessary step to becoming French citizens. Based on
colonial classifications, the French government physically separated the harki “repatriates” from the
pieds-noirs repatriates, both of who left Algeria in refugee-like conditions owing to FLN violence.
At the harki refugee camps, the government allowed only wives, dependent minor children,
and elderly parents with no means to provide for themselves to accompany the “pro-French
Muslims.” Moreover, it prohibited from entering heads of families who were planning to leave their
families when they migrated to France.82 The government planned to first group the refugee families
into camps and military barracks located in the local military sectors, where they would remain for a
minimum of one month. All the refugees underwent medical tests and received the vaccinations
required for residence in metropolitan France. The children attended school, the men attended
courses intended to help them adapt to their future French life, and the women took hygiene and
social welfare classes.83 Once government officials approved each individual harki family’s request to
migrate to France, its members were transported by military convoy to displaced persons camps in
towns near Algiers—such as Tefeschoun, Zéralda, and Reghaia (see figure 3)—from which they
JORF, “Loi n°61-1439 du 26 décembre 1961…,” Dec. 28, 1961. Article 1 of the law states:
Les Français, ayant dû ou estimé devoir quitter, par suite d’événements politiques, un territoire où ils
étaient établis et qui étaient antérieurement placé sous la souveraineté, le protectorat ou la tutelle de la France,
pourront bénéficier du concours de l’État, en vertu de la solidarité nationale affirmée de la Constitution de
1946 dans les conditions prévues par la présente loi.
Ce concours se manifeste par un ensemble de mesures de nature à intégrer les Français rapatriés dans les
structures économiques et sociales de la nation.
Ces mesures consisteront, en particulier, à accorder aux rapatriés des prestations de retour, des prestations
temporaires de subsistance, des prêts à taux réduit et des subventions d’installation et de reclassement, des
facilités d’accès à la profession et d’admission dans les établissements scolaires, des prestations sociales ainsi
que des secours exceptionnels.
Les programmes de construction de logements bénéficiant de l’aide de l’État seront complétés par
l’adjonction de contingents supplémentaires de logements pour les rapatriés…
82 SHAT, 1H 3077/2, Sergent Vella, “Objet: Recasement en France supplétifs et civils FSNA engagés aux côtés Forces
Armées,” Message N˚ 1050/CSFA/EMI/MOR, Apr. 16, 1962.
83 SHAT, 1H 4214/1, General Hublot (Chef d’État-Major) to the General commanding the Région Territoriale and the
Corps d’Armée d’Oran, “Objet: Recasement en France de militaires supplétifs et civils FSNA,” 1, N˚
1013/CSFA/EMI/MOR, Apr. 11, 1962.
81
95
would depart several days later for Marseille and Port-Vendres, located twenty-five kilometers south
of the Rivesaltes camp.
R
Z
T
Figure 3: Refugee camps on the northern coast of Algeria. From west to east on this one hundred kilometer stretch, “T”
marks the Tefeschoun refugee camp, “Z” marks the Zéralda refugee camp, the star marks Algiers, and “R” marks the
Reghaia refugee camp. Source: Google Maps.
In her memoir, harki daughter Fatima Besnaci-Lancou gives substance to harki families’
secluded life in the relocation camps through a narrative of her experiences during October and
November 1962 at the Béni-Messous military barracks on Algiers’ western outskirts. 84 In early
August her father fled his village because he feared for his life after two family members who had
served as harkis were savagely killed. His family faced uncertainty about whether he was alive. Two
months later, Besnaci, her mother, and her four sisters were able to join him in Béni-Messous, the
first in a succession of camps in Algeria and France where they would live hidden from the
surrounding community. Each family lived in pre-fabricated buildings comprised of four or five
one-room apartments, which contained bunk beds, a table, and two chairs. Since these dwellings had
no kitchens, the families received food rations from a canteen several times a day, just like the
Besnaci-Lancou, Fille de harki. Chapter 2 (“C’est ainsi que nous sommes devenus des harkis”) focuses on her family’s
passage through the relocation camps.
84
96
soldiers who lived on the grounds protecting them. In the mornings, a military officer led classes for
eight-year-old Besnaci and fifteen other children of all ages. Every afternoon at three o’clock, harki
men attended courses to help them transition to metropolitan life. Besnaci foremost emphasizes the
fear, disorientation, and uncertainty for their future that she and the other camp dwellers suffered:
“Nous étions coupés du monde extérieur. Nous n’étions nulle part. Les seules nouvelles de
l’extérieur nous venaient des nouveaux arrivants. Elles n’étaient pas joyeuses.”85 The isolation that
Besnaci expressed served as a harbinger for her family’s future life in camps in metropolitan France.
Isolation, though, was not the only problem induced by the camps.
On May 16 the Army headquarters in Algeria issued a report with information from all three
Army Corps and the Sahara revealing the camp’s principal difficulties. These complications did not
just reflect the immediate situation in the camps but, more importantly, they also signaled the two
principal factors that served to block the repatriation of harkis.
Army officials argued in the report that “un maintien prolongé dans les [camps de]
regroupements risque, en raison de la promiscuité et de l’inaction des personnes regroupées,
d’engendrer des incidents: affaires de mœurs, vols, manifestations collectives plus ou moins violentes
de mécontentement.”86 This view of the harki population as disposed toward violence, protests, and
laziness entered into the government’s new, broader definition of “harkis.” After the last units of
harkis (harkas) were dissolved in April 1962, the term transformed from its previous meaning of
(almost exclusively) male auxiliary soldiers to a word that encompassed entire families of Algerian
refugees forced to migrate from their homeland.87 As one Army report expressed, single women,
whether widows or unmarried adult daughters of former harki soldiers, should not be included into
Ibid., 53-54.
SHAT, 1R 336/8 (dossier: Rapatriement des harkis), CSFA, EMI, Bureau du Moral, “Recasement en France des
supplétifs et civils FSNA. État d’avancement des opérations de regroupement initial au 10 mai 1962,” N˚ RR/YR-15.5,
May 16, 1962.
87 Documents in the Army Archives indicate that there were very few female harkis also known as “harkettes.” I met one
of these women in June 2009 at the Cité les Tilleuls in Marseille.
85
86
97
the repatriation process for fear that they would resort to prostitution out of economic
desperation.88 Army officials who decried the “promiscuity” and “idleness” of the camp dwellers
imbued the emergent term “harki” (male and female) with the same stereotypes that government
officials had attributed throughout the imperial era to male colonial subjects, whom they held in
disdain for being lazy, promiscuous, and dangerous.89 To cite an example from the burgeoning
scholarship on government oversight of “natives” in the colony and the metropole during this time,
one 1917 government report described problems with Madagascan workers in France by averring:
“Conducted with firmness and goodwill at the same time, one can get good work from him…
Although intelligent, the Madagascan has little initiative, which explains the small number of
industries he has created… The taste for promiscuity which he often abuses, even from childhood,
is for him a cause of degeneracy and abbreviates his days.”90 Here we see linguistic continuities
between the essentializing assessments of Madagascan “colonial subjects” and those of the harkis.
Male and female harkis, adults and children alike, would encounter great difficulties to overcome
stereotypes developed from settlers and state agents’ contacts with “natives” during the colonial era,
even after they became French citizens in metropolitan France.
As the harkis sought approval to migrate to France from military and government officials,
extant colonial stigmas and classifications also challenged their inclusion into the definition of
repatriates—a necessary step toward French nationality—as much as the Army’s evaluations of the
situation in the relocation camps. The harkis could not escape the colonial specter of the label
“musulmans,” a statutory category of inferior citizens attributed to the native Algerians, to be
SHAT, 1H 1260/2, CSFA, EMI, Bureau du Moral, “Recasement des supplétifs et civils FSNA menacés,” 2, N° SP
87.000, May 26, 1962.
89 For a discussion of stereotypes of colonial subjects, see, for example: Fogarty, Race and War; Lewis, The Boundaries of the
Republic; Neil MacMaster, Colonial Migrants and Racism: Algerians in France, 1900-62 (New York: St. Martin’s Press, 1997);
Rosenberg, Policing Paris; Emmanuelle Saada, Les enfants de la colonie: les métis de l’Empire français entre sujétion et citoyenneté
(Paris: La Découverte, 2007); Sayad, La double absence; and Tyler Stovall, “Colour-blind France? Colonial Workers during
the First World War,” Race & Class 35.2 (1993): 35-55.
90 SHAT, 7N 997, “Rapport mensuel,” July-Aug. 1917. Quoted in Stovall, “Colour-blind France?,” 48 and 54 n. 37.
88
98
viewed by upper echelon government officials as “repatriates.” Indeed, when asked to give his
position about the place of the harkis in the repatriation process, President de Gaulle would not
include them into the category of “repatriates” because he continued to consider them “Muslims.”
He argued in an exchange with Minister of Armies Messmer at a July 25, 1962 Council of Ministers
that “repatriates” and “Muslims” were mutually exclusive categories:
Messmer: Des harkis et des fonctionnaires musulmans, les moghaznis, se disent menacés.
D’où des demandes qui viennent à la fois des autorités civiles et militaires. Il faut prendre
une position de principe.
De Gaulle: On ne peut pas accepter de replier tous les musulmans qui viendraient à
déclarer qu’ils ne s’entendront pas avec leur gouvernement! Le terme de rapatriés ne
s’applique évidemment pas aux musulmans: ils ne retournent pas dans la terre de leurs
pères!91
De Gaulle’s use of “musulmans” reflected an ethnic conception of “rapatriés.” According to him,
the harkis could never be “repatriates” because they were still, or always, Muslims, which necessarily
meant that they had no roots in the French “fatherland.”
However, the President’s reasoning was faulty. A significant portion of the pieds-noirs did
not fit into de Gaulle’s definition of repatriate, either. Scholars who study the pieds-noirs emphasize
that while almost all were French citizens on the eve of the Algerian War, France was not necessarily
“la terre de leurs pères.”92 Indeed, non-Muslim inhabitants of Algeria held diverse European origins
and some were Jews who had migrated from the Middle East.93 In addition to French settlers, there
were Maltese, Spaniards, Italians, Belgians, and even Alsatians who fled their homeland in 1871 to
Quoted in Peyrefitte, C’était de Gaulle, 209. “Tous” and “rapatriés” are italicized in Peyrefitte’s text.
For this reason, Ageron labels the two sections of a chapter about the economic and social development of Algeria
from 1930 to 1954 “The European Population” and “The Muslim Population,” and Shepard similarly titles his chapter
that contrasts France’s policy of “rejecting” Algerians Muslims from the repatriation process as “Repatriating the
Europeans.” Ageron, “Chapter 6: The Economic and Social Development of Algeria, 1930-1954,” Modern Algeria, 82-92,
and Shepard, “Chapter 8: Repatriating the Europeans,” Invention of Decolonization, 207-228.
93 The Vichy regime stripped Algerian Jews of their full French citizenship under an October 7, 1940 law. This
legislation repealed the 1870 Crémieux decree, which had included Jews into the category of French with common civil
status. As a result, the 110,000 Jews in Algeria returned to the pre-1870 status of “subjects” until an October 21, 1943
declaration issued by the Comité Français de Libération nationale gave them back full French nationality. Weil, “Histoire
et mémoire,” 9-10.
91
92
99
avoid becoming German citizens after the loss of Alsace-Lorraine. It is impossible to quantify the
repartition of national origins among the pied-noir population owing to a lack of statistical
information (the government’s ethnic distinctions for population groups in Algeria was limited to
“Europeans” and “Muslims”) and a high rate of intermarriages among settlers of different national
origins. Nevertheless, data from the Annuaire statistique de l’Algérie published in 1933 establish that
“foreigners” (i.e., non-French citizens) and naturalized French citizens had consistently comprised a
significant proportion of the European population since French colonization of Algeria in 1830.
Table 2, which reproduces these figures, demonstrates that, on average, forty-three percent of the
“European population” from 1833 to 1926 was comprised of foreigners, data that does not include
the proportion of French citizens who were naturalized.94 De Gaulle, therefore, was willing to
assimilate those of European origin whose ancestors had never lived in France into French
repatriates, but he refused to admit harkis into this category. He also included within this category of
“pieds-noirs” Jews, whom Todd Shepard describes as “French citizens whose supposed ‘group’
differences had obsessed and shaped French culturalist xenophobia” since the 1789 Revolution. The
opposition between Christians and Jews that had permeated recent Franco-French conflicts, namely
the Dreyfus Affair and Vichy France, were forgotten at the end of French Algeria. The French
government labeled them as “wholly French,” according to Shepard, now emphasizing the divide
between French and Muslims.95
Kateb, Européens, “indigènes” et juifs, 187.
Shepard, The Invention of Decolonization, 169-170. Chapter 6 of this book, “Repatriation Rather Than Aliyah: The Jews of
France and the End of French Algeria,” focuses on the repatriation of Jews from Algeria.
94
95
100
Year
1833
1836
1839
1841
1846
1847
1851
1856
1866
1872
1876
1886
1891
1896
1901
1906
1911
1921
1926
Total French
citizens (of
French origin and
naturalized)
3,478
5,485
11,000
15,497
46,339
42,274
66,050
92,738
122,119
164,175
189,677
261,666
315,131
366,900
421,389
514,065
542,871
602,609
657,641
Foreigners
Total
Europeans
Proportion
of Foreigners
4,334
9,076
14,000
20,230
49,780
67,126
65,233
66,544
94,781
115,516
155,072
203,154
215,793
211,580
212,461
166,198
209,172
188,761
175,718
7,812
14,561
25,000
35,537
96,119
109,400
131,283
159,282
216,990
279,691
344,749
464,280
530,924
578,480
633,850
680,263
752,043
791,370
833,359
55.5
62.3
56
56.9
51.8
61.4
49.7
41.8
43.7
41.3
45
43.7
40.6
36.6
33.5
24.4
27.8
23.9
21.1
Table 2: Population of European descent in Algeria from 1833 to 1926. The decreasing proportion of foreigners since
the end of the nineteenth century represents an increasingly aggressive policy to naturalize foreigners in Algeria as the
metropolitan French population stagnated. Source: Kateb, Européens, “indigènes” et juifs, 187.
The government did not organize a concrete repatriation plan for the harkis, the second
major problem reported by Army officials in the May 16 memo, both because officials did not
include the harkis into the definition of “French repatriates” and owing to the emergency situation
in Algeria. In a heated exchange during the May 4, 1962 Council of Ministers meeting President de
Gaulle vociferated his strong desire to celeritously sever all ties with Algeria, which he believed was
draining France’s strength:
L’État est profondément malade, et les événements n’ont fait qu’aggraver sa maladie,
dont ils sont la conséquence… Il faut annoncer la date de l’autodétermination. Que
personne ne doute que la France n’exercera plus aucune responsabilité, ni politique ni
maintien de l’ordre, au plus tard six mois après le cessez-le-feu! Que les musulmans
préparent le gouvernement de l’Algérie! Que les Européens se persuadent qu’il faut, ou bien
101
s’accommoder avec les musulmans sans que la France les protège, ou bien rentrer en
France!96
Now that Algeria was on its way to becoming independent, de Gaulle wanted to remove the vestiges
of France’s colonial empire that had caused it to be “profoundly sick” and return the governance of
Algeria to the “Muslims.”
Violence in Algeria forced a frenzied “exodus” of nearly one million pieds-noirs to France
during 1962 alone.97 The migration of pieds-noirs to France increased dramatically following the
April 30 arrest of ex-General Raoul Salan, the leader of the OAS. In its efforts to keep the piedsnoirs in their homeland to demonstrate the strength of its pro-French Algeria position, this
organization issued threats to those wanting to flee.98 After 173,816 departures during the first four
months of 1962, 629,980 pieds-noirs arrived in French ports and airports from May through July.
While 412,752 individuals did migrate from France back to Algeria over the course of the year, the
net gain in metropolitan France equaled 651,265 disoriented people who needed housing and jobs.99
The government neither expected nor wanted this influx. Secretary of State Boulin issued a press
release after the May 30, 1962 Council of Ministers, which succinctly exposed the government’s
position. It stated: “Nous disposons d’un filet protecteur, la loi de décembre 1961 sur les
rapatriements, qui nous permet de prendre en charge 70 000 rapatriés en 1962; mais on peut espérer
qu’on n’aura besoin d’y avoir recours et que la quasi-totalité des Européens qui reviennent
actuellement en métropole repartiront sans demander à bénéficier du statut de rapatriés.”100 The net
total of pieds-noirs who migrated to France in 1962 was more than nine times this ceiling. The
government was patently aware that along with the repatriates of European origins came security
Quoted in Peyrefitte, C’était de Gaulle, 139.
According to Todd Shepard, the term “exodus” was introduced in 1961 in reaction to concerns about Jewish
departures from Constantine. The Invention of Decolonization, 211.
98 For details about the OAS threats and propaganda, see Shepard, The Invention of Decolonization, 208-218.
99 Jordi, De l’exode à l’exil, 66. This figure represented two-thirds of the 984,000 Europeans who inhabited Algeria at the
outbreak of the war. Scioldo-Zürcher, Devenir métropolitain, 31.
100 Quoted in Peyrefitte, C’était de Gaulle, 151.
96
97
102
threats from recalcitrant OAS members who clung to their visions of a French Algeria. However,
despite these links there is no evidence that it refused to admit these repatriates into France.
Discussions about Algerian populations during the spring 1962 Council of Ministers
meetings focused overwhelmingly on OAS affiliates’ terrorist acts in Algeria and the serious security
risks that their migration to France could pose. After metropolitan French citizens overwhelmingly
ratified the April 8 referendum that paved the way toward Algerian self-determination OAS
extremists undertook the “scorched earth policy,” incinerating municipal buildings, ports, and
farmland in Algeria. Apprehension about the lingering Franco-French war consumed President de
Gaulle and his Ministers. Evidence from Council of Ministers meetings suggest that they feared that
the pied-noirs’ unfettered access to metropolitan France would allow dangerous OAS members to
migrate among them, two groups that were difficult to distinguish one from the other. As discussed
in chapter one, a few cases of OAS violence had indeed reached metropolitan France with the
bombing of Minister of Culture André Malraux’s apartment building and assassination attempts on
President de Gaulle (see pages 64-65). In this context, Prime Minister Georges Pompidou asserted
on April 25: “Le moment est proche où le seul recours de l’OAS sera le terrorisme désespéré; non
plus en Algérie, mais en métropole.”101 Interior Minister Roger Frey declared on May 4: “Nous
avons saisi de nombreux documents qui prouvent que l’OAS croit que, lorsque les rapatriés et les
militaires vont installer en métropole, elle pourra s’infiltrer partout et que la situation basculera.”102
And de Gaulle pronounced on May 9: “Il faut s’attendre à ce que le retour des Français d’Algérie
nous amène de grandes difficultés. Il y aura parmi eux beaucoup de tueurs, beaucoup de tueurs.”103
The government’s attention therefore focused squarely on quelling the Franco-French war.
Ibid., 136.
Ibid., 139.
103 Ibid., 142.
101
102
103
The OAS, the Harkis, and “Repatriating” the Franco-French War
Government officials believed that another, albeit less numerous, potential source of entry
for unwanted OAS “killers” existed: the harkis. Confidential documents written by French Ministers
and transcripts of Council of Ministers meetings in spring 1962 demonstrate that logistical reasons
(including a shortage of housing in France) and ethnicity alone cannot account for why the French
government did not want to repatriate the harkis. 104 Government officials feared these former
soldiers as potential means to “repatriate” OAS violence to France. For this reason, the government
sought to strictly control the harkis’ access to France, and even threatened to send them back to
Algeria if government orders were not followed. Amidst reports of soldiers with links to the OAS
taking the initiative to repatriate groups of harkis, on May 12, 1962 Pierre Messmer sent a secret
telegram warning the Commander of Military Forces in Algeria of dire consequences for harkis who
did not have his personal authorization to migrate to France:
Dès maintenant je vous prie:
Primo- d’effectuer sans délai enquête en vue déterminer conditions départ d’Algérie de ces
groupes incontrôlés et sanctionner officiers qui pourraient en être à l’origine.
Secundo- …d’informer vos subordonnées que, à compter du 20 mai, seront refoulés sur
Algérie tous anciens supplétifs qui arriveraient en métropole sans autorisation de ma part
accordée après consultation départements ministériels intéressés.105
Messmer asserted that the primary reason for these “mesures sévères” was the “[n]écessité absolue
éviter arrivée en métropole individus ou groupes d’individus incontrôlés qui pourraient constituer
masse manœuvre pour organisations subversives.”106 While Messmer maintained in a 2004 interview
that he never ordered any harkis be sent back to Algeria (and there is no documentary evidence of
In making this assertion, I question Todd Shepard’s claim that “Rather than security concerns, or any kind of explicit
embrace of the ideological terms of left-wing rejection of harkis as ‘collaborators’ and OAS ‘storm troopers,’ de Gaulle’s
government affirmed a racialized exclusion.” As I demonstrate below, some of the most damning contentions about the
OAS trying to use harkis toward political ends came from the right-tilting newspaper, Le Figaro. Shepard, Invention of
Decolonization, 239.
105 CAC, 19910467/2 (dossier: Harkis I), telegram from Pierre Messmer to Genesuper Algeria, N˚ 1334 MA/CAB/DIR,
May 12, 1962.
106 Ibid.
104
104
such cases), the extreme measures that he promised nonetheless signal that the harkis were feared as
potential operatives in the Franco-French war.107
Messmer was not alone in his concern about the OAS-harki connection. Significantly, one of
the sole instances that the harkis were mentioned during the postwar Council of Ministers meetings
was when Minister of Algerian Affairs Joxe posited a link between the OAS and the harkis. In the
context of FLN members increasingly threatening and carrying out violent retributions toward the
harkis he argued on May 24, “Enfin, les harkis veulent partir en masse; il faut évidemment combattre
une infiltration qui, sous prétexte de bienfaisance, aurait pour effet de nous faire accueillir des
éléments indésirables.” In response to Joxe’s statement and reports about strife among the Muslim
population, President de Gaulle declared to his cabinet members that that after Algeria became
independent, “[l]a France ne doit plus avoir aucune responsabilité dans le maintien de l’ordre…. Si
les gens s’entre-massacrent, ce sera l’affaire des nouvelles autorités.”108 Joxe’s statement perfectly
summarizes the situation: how to balance humanitarian concerns for the harkis’ safety in the wake of
myriad reports of killing, imprisonment, and torture by FLN members with political concerns over
the Franco-French war. And President de Gaulle’s response to place the harkis within the category
of “les gens [qui] s’entre-massacrent” evidences which way the scales tipped. His government’s
policy to limit the harkis’ repatriation demonstrated that political efficacy trumped humanitarian
considerations, neglecting the harkis’ welfare in the massacres.
Links between the harkis and the OAS did exist, but there is scant evidence that the harkis
actually participated in OAS activities. Nevertheless, government officials expressed concerns that
the harkis provided a fertile breeding ground for this seditious organization and if the OAS oversaw
the harkis’ migration to France, this would perpetuate the Franco-French war. It is unclear to what
extent a deliberate desire to believe in the threat of harkis joining the OAS (as opposed to a
107
108
Messmer, interview.
Quoted in Peyrefitte, C’était de Gaulle, 151.
105
miscomprehension of the few ties that actually existed between harkis and OAS members)
influenced the government’s evaluation that the harkis wanted to continue to fight for a French
Algeria. But it is apparent that this appraisal served as a reason to view repatriating harkis to France
as a potentially dangerous move and therefore to exercise tight controls over its boundaries to limit
the process. The perception of a dangerous harki-OAS link was primarily fueled by two factors. First,
throughout the war Army officers had relayed to their superiors that Algerian natives had willingly
joined the French cause because they supported a French Algeria. And second, after the ceasefire,
reports from military commanders supporting repatriation requests made by harkis whom they had
commanded, like those on pages 92 and 93 above, lauded these former soldiers devotion to the
French cause. In their quest to save these men’s lives against the impending FLN violence, soldiers
(a minority of whom was affiliated with the OAS) actually increased public and governmental
anxiety about the “repatriation” of the violent Franco-French war because officials chose to
interpret the harkis’ pro-French sentiments as extremist.
In spring 1962, Army reports fed government officials’ assumptions that some of the harkis
would cling to their supposed vision of a French Algeria and disturb the peace if repatriated to
metropolitan France. As explained in the introduction, the need for jobs in the face of the warravaged agrarian Algerian economy, forced recruitment by French soldiers (enrôlement) with orders
from the government to promote the appearance that Algerians were opposed to independence, and
the quest for weapons to protect themselves and their families against FLN violence proved more
important factors in the decision of Algerian natives to fight for the French cause.109 In fact, over the
course of the war an unquantifiable number of Algerian natives fought for France and later for the
FLN (or vice versa); some even served both sides concurrently.110 As mentioned in the previous
Charbit, Les harkis, 30-31 and Hamoumou, Et ils sont devenus harkis, 63.
For example, through her voyage to Algeria to gather information to write her memoir, harki daughter Dalila
Kerchouche learns that her father served simultaneously as a harki and as an FLN informant. Mon père, ce Harki.
109
110
106
chapter, from summer 1961 onward army officials expressed concerns over desertions, fearing in
particular that French weapons would be passed to ALN members who would then use them on
French soldiers. 111 The government distrusted the harkis’ loyalty to France and used this
apprehension as a reason for expeditiously disarming harkis who returned to their villages during the
war or after the ceasefire.
Nevertheless, when the government could no longer avoid taking a stance on repatriating
the harkis, it chose to privilege as exemplary the atypical case of staunchly pro-French BeniBoudouane tribal leader Bachaga Boualam over its concerns that the harkis would side with the
FLN.112 The majority of the men among the fifteen thousand tribe members living in the Bachaga’s
mountainous fief near Chlef (midway between Algiers and Oran) did engage as harkis.113 However,
the complex situation of customary obligations to follow their leader was a principal factor in these
men’s decisions. 114 The reason that the Beni-Boudouanes joined the harkis, like other Algerian
natives, could not be reduced to allegiance to French Algeria, in contrast to projections by French
Army and government officials in discussions about repatriating the harkis. Once the military
operations ended, these individuals, therefore, did not mourn the end of French Algeria in the same
way as the extremist pieds-noirs who prolonged the conflict with terrorist acts toward the French
government. Since few harkis held a strong conviction for the French cause during and after the war,
their conjectured participation in the OAS maquis was tenuous at best.
SHAT, 1H 2467/3 bis, État-Major Interarmées, 3ème Bureau, “Fiche,” July 10, 1961. Based on documents in the 1H
1598/1 carton, Charles-Robert Ageron contends that 10,893 Muslims deserted their positions as harkis or soldiers in the
French Army during the war, an undetermined number with their weapons. Ageron, “Les supplétifs algériens,” 17.
112 Bachaga is an Arabic word signifying high Arab dignitary. In late May, the military organized the evacuation of 140 of
the Bachaga Boualam’s family members and entourage to the Mas-Thibert farm outside of Arles (the first group of 67
immediate family members arrived on May 18). This was the sole case of a collective tribal repatriation by the
government. Guilia Fabbiano, “Les harkis du Bachaga Boualam. Des Beni-Boudouanes à Mas-Thibert,” in Les harkis
dans la décolonisation et après, ed. Fatima Besnaci-Lancou and Gilles Manceron (Paris: Éditions de l’Atelier, 2008), 117.
113 Bachaga Boualam, Mon pays… la France! (Paris: Éditions France-Empire, 1962), 79-80.
114 Fabbiano, “Les harkis du Bachaga Boualam,” 113-16.
111
107
Without a governmental repatriation plan in place, as FLN violence escalated in spring 1962,
some French soldiers and civilians in Algeria felt a moral obligation toward these former soldiers
because they had risked their lives for France. These Frenchmen used their own means to provide a
small number of harki families with safe passage to France.115 Among these French citizens were
troops in the sections administratives sociales (SAS) whom Algerian civilians employed by France (the
moghaznis) had protected during the war as they performed administrative, social, educational,
economic, and military duties in the Algerian countryside. Some of these SAS troops had OAS ties.
Press outlets across the political spectrum picked up on these isolated cases and published
articles in late May with titles such as “Comment l’OAS tente d’installer des harkis en métropole”
(Le Figaro, May 22) and “De l’entraide à l’utilisation des harkis en métropole” (Le Monde, May 23),
“Harkis transférés en France ‘ratonneurs’… expulsés: Des recrues pour l’O.A.S.” (Libération, May 24),
and “L’O.A.S. installe en France des ‘harkis’ pour préparer de nouveaux crimes” (Humanité, May 23).
These inflammatory pieces provided fuel for government officials’ anxieties over the potential for
the harkis to serve as a means to “repatriate” OAS violence to France, though none reported that
the repatriated harkis actually carried out subversive actions. Periodicals including France-Observateur
(the predecessor to Le Nouvel Observateur), Libération, and Le Monde honed in on two particular
instances. The leftist Libération reported in a May 26 article, “Des réserves pour l’O.A.S. dans la
Vienne?,” that ninety harkis had arrived on a property near Poitiers owned by a former secretary of
state under Vichy government leader Marshal Pétain whose son served as a colonel in the SAS and
A 1963 report by the Minister of Interior titled “Aspects du problème des harkis en France” presented detailed
biographical information for twenty-four individuals who had actively participated in these “individual initiatives.” Some
of these men led associations created to help the harkis to settle in France, such as the “Comité National de Solidarité
pour les Français Musulmans Rapatriés” founded on July 10, 1962, and presided over by Alexandre Parodi. The report
argued that such actions “complicated” the harkis’ adjustment to life in metropolitan France because: “Les intéressés
eux-mêmes [les harkis], peu au fait de la réglementation métropolitaine, ne connaissant pratiquement pas le territoire
français, enclins, comme la plupart de leurs coreligionnaires, à des pérégrinations fréquentes, ont échappé dans une
certaine mesure à un contrôle généralisé.” The harkis’ adjustment to life in metropolitan France, including living under
the strict supervision of camp officials, is the focus of Section Two of this dissertation. AN, F1a 5017, Direction des
Renseignements Généraux, Ministère de l’Intérieur, Report: “Aspects du problème des harkis en France,” 1, 1963.
115
108
reportedly disagreed with President de Gaulle’s policies on Algeria. The same piece related that 138
harkis were camped on the grounds of a chateau outside of Limoges whose proprietor, an excolonel in the SAS, had been imprisoned in April for publishing pro-OAS tracts on this site. In line
with Libération’s continual opposition to de Gaulle’s policies in Algeria (and his policies in general),
the article concluded by vilifying his government: “Et n’est-il pas scandaleux que les pouvoirs
publics continuent à fermer les yeux sur la constitution en France de groupes dont certains pourront
le cas échéant fournir des hommes ‘d’élite’ aux commandos O.A.S.”116 These words posited a strong
link between the OAS and the harkis, and suggested the former auxiliary soldiers’ potential for
violent actions.
These journalists were not alone in their concerns about the OAS trying to use harkis toward
political ends. High-level government officials demonstrate similar apprehensiveness in their
correspondence written at the same time. By examining side by side a pair of articles published on
May 22 and May 24 in Le Figaro and a confidential letter sent from Minister of Interior Frey to
Minister of Armies Messmer on May 15, this fact becomes clear. Based on interviews conducted
among harkis repatriated through unofficial channels, Le Figaro reporter Serge Bromberger, who had
recently returned from Algeria, pronounced that these ex-soldiers did not in fact feel threatened in
Algeria and only migrated because their superiors and pieds-noirs encouraged them.117 This claim
echoed concerns that Frey expressed in his letter just a week before: “Il me paraît même probable
que, dans certains cas, des harkis ont été incités à franchir la Méditerranée, sans qu’ils aient été
précisément menacés.”118 Bromberger’s May 22 article continued, “Dans un certain nombre de cas
précis, il est évident qu’il s’agit d’une tentation tendant à faire entrer en métropole des éléments
SPDP, dossier: Harkis, “Des réserves pour l’O.A.S. dans la Vienne?” Libération, May 26, 1962.
SPDP, dossier: Harkis, Serge Bromberger, “Comment l’OAS tente d’installer des harkis en métropole,” Le Figaro, May
22, 1962.
118 SHAT, 1K 744 (dossier: Directives concernant les harkis), letter from Minister of the Interior Roger Frey to Minister
of Armies Pierre Messmer, 2, N° SN CAB 2924, May 15, 1962.
116
117
109
indésirables ayant la pratique des armes, et destinés à servir de moyens… à l’O.A.S.”119 Similarly,
Frey’s letter continued, “Il y a là un danger certain, car il ne faut pas se cacher que la plupart des
personnes qui essaient, en ce domaine, de se substituer aux autorités, sont politiquement très
orientées et que le désordre qui préside, du fait de leur interférence, à la reconversion des harkis, ne
peut être que générateur de troubles.”120
Yet, while Bromberger and Frey expressed analogous concerns, their proposed solutions to
the problem differed and demonstrated the government’s apprehension over repatriating the harkis.
Whereas Frey simply asked Messmer to remind his officers that “individual initiatives” were
prohibited, in his May 24 article Bromberger argued that to thwart OAS efforts to turn the harkis
into “shock troops,” the government must step up its efforts to repatriate the harkis.121
The press’s assessment was correct. While these “individual initiatives,” as the government
called them, started in April, the first military vessel carrying harkis did not embark for France until
June 9.122 Instead of—and before—establishing and implementing a procedure for the harkis to
migrate to France, government officials focused on exerting control over how the harkis should not
be repatriated, that is, to ensure that the government’s OAS adversaries would not be involved in
transporting the harkis and providing them housing in France. In an April 18 letter, Minister Joxe
gave the High Commissioner in Algeria, Christian Fouchet, stern instructions to monitor the
repatriation process: “… il conviendra de veiller à ce qu’aucun retour ne soit effectué sans avoir mon
accord préalable afin que l’accueil en Métropole soit assuré dans des conditions satisfaisantes.”123
However, this letter and other documents failed to specify what these “satisfactory conditions”
Bromberger, “Comment l’OAS tente d’installer…”
SHAT, 1K 744, letter from Frey to Messmer, 2.
121 SPDP, dossier: Harkis, Serge Bromberger, “Aider au reclassement des harkis,” Le Figaro, May 24, 1962.
122 AN, F1a 5140, Cabinet Director of the Secretary of State for Repatriates, “Note sur le rapatriement des Français
Musulmans,” 1, June 20, 1962.
123 SHAT, 1K 744 (dossier: “Directives concernant les harkis”), letter from Minister for Algerian Affairs Louis Joxe to
the High Commissioner in Algeria, Christian Fouchet, Objet: “Situation des personnes engagées en Algérie aux côtés de
l’Administration ou de l’Armée,” Apr. 18, 1962.
119
120
110
would be, leaving the door open for “individual initiatives” by soldiers and civilians in Algeria who
believed that France had a moral responsibility not to “abandon” the harkis. As one Army official
from the Oran Army Corps wrote to the Zone Commanders in his region: “Généraux
Commandants Zones feront tout ce qu’il est possible de faire pour FSNA qui ont servi a nos côtés
afin qu’en aucun cas ils puissent avoir sentiment que nous les avons abandonnés – Agir avec
maximum sollicitude est un devoir pour nous et aussi dette reconnaissance dont il faut
s’acquitter…”124 Officials in France, fearful of “repatriating” the Franco-French war, had yet to form
a plan for the harkis, which led some soldiers to substitute themselves for the French authorities.
“Security” over Movements of (Post) Empire
In the face of this situation, the government sought to use what Michel Foucault terms a
“technology of security” over a sliver of its population whom it deemed a threat to metropolitan
safety. Foucault’s description of this technology inherent to modern governance illuminates why the
government did not prepare for and sought to limit the harkis’ repatriation. Foucault writes that the
technology of security “brings together the series of random events that can occur in a living mass, a
technology which tries to predict the probability of those events (by modifying it, if necessary), or at
least to compensate for their effects. This is a technology which aims to establish a sort of
homeostasis not by training individuals, but by achieving an overall equilibrium that protects the
security of the whole from internal dangers.”125
Interior Minister Roger Frey expressed his great concern to Messmer over reports of groups
of harkis arriving in Marseille and being directed to locales around southern France “dans des
SHAT, 1H 4194/1, “Notification n° 107/RT/CAO/MOR du Corps d’Armée d’Oran et R.T.” This notification is
written on the same document as a June 1, 1962 “message postalisé” from Army Headquarters in Reghaia to the Oran
Army Corps.
125 Michel Foucault, “Society Must Be Defended.” Lectures at the Collège de France, 1975-1976, ed. Mauro Bertani and
Alessandro Fontana, trans. David Macey (New York: Picador, 2003), 249.
124
111
conditions assez anarchiques” and over inquiries from military officers with possible OAS links to
prefects about housing groups of harkis and moghaznis in their department.126 To take control of the
situation, Frey and other cabinet members sought to exercise a “technology of security” (state
regulation of populations) because these officials believed that the harkis threatened metropolitan
safety.127 On May 12 Frey issued to prefects across France a telegram (following up on a telegram
written to these administrators several days prior), which gave detailed orders of how to police the
repatriation of groups of harkis by non-government channels:
Primo: Adresser à mon cabinet, lundi 14, liste nominative de tous éléments harkis arrivés
dans votre département en précisant exactement lieux et conditions leur hébergement et en
indiquant identité précise des personnes ayant pris initiative de ces installations –
Secundo: Adresser à mon cabinet tous éléments en possession vos services police et
administration sur projet telles installations –
Tertio: Ne pas manquer me donner toutes informations sur nouvelles initiatives de cet ordre
que vous pourriez déceler dans l’avenir –
Quarto: Vous renouvelle devez vous opposer à tous projets dans ce domaine en dehors
mesures adoptées par secrétariat aux rapatriés.128
Frey penned yet another telegram five days later with nearly identical language to reiterate his third
and fourth points, and added an additional preventative measure: “Vous demande d’informer de ces
directives toute personne ou groupe en cause.” 129 And a few weeks later, once the Army was
preparing to open a military camp to receive the harkis arriving through the government’s
repatriation procedure, he wrote an additional telegram to once again remind prefects to inform him
SHAT, 1K 744, letter from Frey to Messmer, 1. Examples of these letters include: SHAT, 1R 367/7, Sergeant André
Jekabsons to the prefect of the Alpes-Maritimes department (attached to the May 15 letter from Frey to Messmer) and
CAC, 19910467/2 (dossier: Harkis I), Lieutenant François Le Breton to the prefect of the Alpes-Maritimes department,
Apr. 28, 1962. Lieutenant Le Breton sent the same letter to the prefects of the Gers, Hérault, Pyrénées-Orientales,
Pyrénées-Basses, Tarn, and Garonne departments.
127 Foucault, “Society Must Be Defended,” 249.
128 Frey references a previous telegram (n° 263) in the present document. However, I have been unable to locate it in the
archives. The number of the telegram leads me to conjecture that Frey wrote it in late April or early May. ADBR, 137 W
460, Roger Frey to All Departmental Prefects, telegram n° 288, May 12, 1962.
129 ADBR, 137 W 460, Roger Frey to All Departmental Prefects, telegram n° 301, May 17, 1962.
126
112
and the Secretary of Repatriates of harkis arriving in their departments with the help of groups that
“vous paraîtront politiquement dangereux.”130
The succession of three telegrams over the space of approximately two weeks (and the
telegram in early June) as well as Frey’s words in themselves signal that the government sought to
use a “technology of security.” Part of controlling the harkis’ access to metropolitan soil was to hide
them from the general population during and after the repatriation process, thereby indicating that
harki repatriates were unwelcome in French society. When the government did finally start
repatriating the harkis, they were sent to the Larzac military camp located in the Midi-Pyrénées
region and the Bourg-Lastic military base in Auvergne, both located in sparsely populated corners of
France (see figure 4). On their way to these “transit camps,” the harkis imperceptibly passed through
Marseille after arriving from Algiers, Nemours, Bône, and Mers-el-Kébir on Navy vessels, a trip that
lasted upwards of eighteen hours. The first ship carrying 200 harkis and their family members
(totaling 651 people) arrived in Marseille on June 11, 1962, and the transfer was expected to
continue at a rhythm of approximately one boat per day until July 1, the eve of Algerian
independence.131 As depicted in figure 5, the harkis’ transport to Cap Janet, the section of the port
that is located the furthest from the city center and was used for container ships, most often
transpired at night—to hide them from public view. Indeed, the National Police reported that 960
harkis and their family members who had debarked in Marseille the evening of June 23 arrived at the
Clermont-Ferrand train station at two o’clock the following morning, where they were subsequently
transported by train and military trucks to the Bourg-Lastic camp.132 As the image in figure 6 shows,
after arriving at the port, the harkis waited in an adjacent hangar or on the dock itself for the trains
ADBR, 137 W 460, Roger Frey to All Departmental Prefects, telegram n° 347. This telegram is not dated, but based
on its reference to the Larzac camp being prepared by military authorities to receive the harkis arriving in France, I
estimate that Frey wrote the document in early June.
131 ADBR, 138 W 3, SCINA, “Synthèse… journée du 14 juin 1962,” N˚ 1 706.
132 ADBR, 138 W 3, SCINA, “Synthèse… journée du 27 juin 1962,” b, N˚ 1 715.
130
113
that would hurry them off to the camps. While the military held the sole responsibility for watching
over the harkis (in principle, the government sent three officers per one hundred persons),133 the
captain of the port, Albert Payan, deemed it necessary to put into place extra police officers to patrol
Cap Janet and the surrounding railroad tracks.134
Bourg-Lastic
Larzac
Figure 4: The Bourg-Lastic and Larzac transit camps located in rural southern France. Source: Google Maps.
SHAT, 1H 3077/2, Genesuper Reghaia to the Oran Army Corps, “Objet: Transfert en métropole supplétifs FSNA
menacés et leurs familles,” N° 1407/CSFA/EMI/MOR, June 1, 1962.
134 ADBR, 137 W 460, “Communication téléphonique de M. Payan,” June 9, 1962.
133
114
Figure 5: Harkis arriving
in Marseille on June 12,
1962. Source: Agence
France-Presse Archives.
Figure 6: Harki families
waiting in the Marseille
harbor to be transported
to the Bourg-Lastic camp
on June 23, 1962. Source:
Agence France-Presse
Archives.
115
The harkis’ repatriation contrasted with that of the pieds-noirs, also organized by the state.
Eighty percent of these French citizens were housed in the Phocaean city in HLM buildings, hotels
and school dormitories requisitioned by the government, emergency shelters run by the Prefecture,
and housing centers operated by charitable organizations such as the Catholic Relief (Secours
Catholique), and the Red Cross.135 In contrast, the government, heeding explicit concerns that the
prefect of the Bouches-du-Rhône department expressed on May 24 about the dangers posed if the
harkis settled in his capital, made no housing provisions for them in Marseille. As the Chief of
Muslim Affairs for the Interior Ministry wrote to the prefect on June 15: “Il est apparu qu’en raison
du grand nombre de rapatriés affluant à Marseille, le transit dans cette ville des refugiés musulmans
devrait être réduit à la plus courte durée possible et que ceux-ci seraient, en conséquence, acheminés
vers des centres de grande capacité situés dans d’autres départements.”136 Indeed, while national and
local government agencies and private organizations worked together to house the “repatriates” in
France’s third largest city, national and local government officials planned to hide the “Muslim
refugees” in the French countryside secured away from the rest of the French population.
The government initially planned to house the five thousand harkis whom it estimated
would briefly transit Marseille before July 1137 in tents and barracks on the grounds of the Larzac
military camp,138 which had originally been constructed to house a maximum of three thousand
Jordi, De l’exode à l’exil, 66-67. On the pieds-noirs’ arrival in Marseille, see pp. 68-97. Their living conditions were not
ideal, however, the government never housed them in tents on campgrounds. For an analysis of the pieds-noirs’
emergency housing situation, see Scioldo-Zürcher, Devenir métropolitain, 197-215.
136 ADBR, 137 W 460, G. Lamossoure (Chief of Service of Muslim Affairs) to the Prefect of the Bouches-du-Rhône,
“Objet: Transit à Marseille et dispersion en Métropole de réfugiés musulmans,” SAM N˚ 341200, June 15, 1962.
Lamossoure references a May 24 letter (N˚ 6901) that the prefect wrote to him.
137 SHAT, 9R 450/6, Corps du contrôle de l’Administration de l’Armée, “Rapport d’information sur l’hébergement des
Harkis au Camp du Larzac,” 2, N° 15 RC/62, Oct. 15, 1962. This report cites a May 29 document written by Army
officials in France (N° 9463 EMA/I.O./DR.SC.), which estimated that five thousand FSNA would arrive at the Larzac
camp by July 1. The first one thousand inhabitants were housed in thirty-two military barracks and all subsequent
arrivals resided in “villages” of tents (pp. 3-4 of the report).
138 Larzac had been the largest camp for FLN prisoners in France during the Algerian War for Independence, interning
approximately ten thousand of them from 1959 to 1962. Marc Bernardot, “Être interné au Larzac: La politique
d’assignation à résidence surveillée durant la guerre d’Algérie (1958-1962),” Politix 24, no. 69 (2004): 40. The analysis of
135
116
people.139 However, after just over one week, the camp, which had received the first harki families
on June 13, reached almost full capacity.140 To absorb what Messmer later described as an influx of
former auxiliary soldiers that reached much greater than expected proportions,141 on June 23 the
Army established a second “transit camp” on the Bourg-Lastic military base with space to house an
additional six thousand harkis in canvas tents (see figure 7).142 On June 25, with 8,500 harkis in the
camps or in transit to the camps,143 the Army predicted that by July 1 (the date that Algerians would
vote on the self-determination referendum), Larzac would house approximately six thousand people
and that Bourg-Lastic, opened just one week prior “un peu ‘en catastrophe,’” would reach nearly
five thousand people.144 These figures reflected more than double the initial estimate. And, there
were still harkis waiting in relocation camps to cross the Mediterranean as Algerian independence
loomed and violence toward the former French soldiers escalated.
the Rivesaltes camp in chapter three addresses continuities with the colonial era in the use of spaces to house harki
families and in methods of governance.
139 Charles-Robert Ageron, “Le ‘Drame des harkis,’” 4.
140 SHAT, 1K 744 (dossier: “Premières mesures d’accueil en France”), letter from Pierre Messmer to Louis Joxe, N°
19586 MA/CC, July 19, 1962.
141 SHAT, 1K 744 (dossier: “Premières mesures d’accueil en France”), letter from Pierre Messmer to Robert Boulin, N°
1710 MA/CC, June 21, 1962.
142 SHAT, 14T 92, Army Corps General Conze (Military Governor of Lyon), “Objet: Organisation de l’accueil des
supplétifs et de leurs familles au Camp de Bourg Lastic,” 1, N° 4996/EM.8/4.ETG, July 2, 1962.
143 SHAT, 1R 336/8, telegram from Genesuper Reghaia to the Minister of Armies (Paris), “Objet: Rapatriement harkis et
moghaznis,” N° 1536/CSFA/EMI/MOR, June 25, 1962.
144 SHAT, 1K 744 (dossier: Premières mesures d’accueil en France), “Note pour M. le Ministre,” June 28, 1962. The
phrase “un peu ‘en catastrophe’” appeared in SHAT, 14T 92, handwritten note by Ministry of Armies official, 1, July, 19,
1965.
117
Figure 7: Harki families residing in canvas military tents at the Bourg-Lastic camp in summer 1962.
Source: La Montagne regional newspaper.145
Violence and Nationality
In the midst of this violence, the new post-1962 governments on both sides of the
Mediterranean questioned the harkis’ French and Algerian nationalities. These uncertainties each
contributed, in their own ways, to harki deaths. The decisions that French government officials
made at this moment when the harkis’ lives were in mortal danger clearly indicated whom they
believed belonged—and did not belong—within the boundaries of the French post-imperial space.
The harkis, like all other Algerians with local civil status, were governed by the July 21, 1962
ordinance, which required them to make an official declaration in front of a metropolitan French
judge as the final step to becoming a French citizen. Consequently, the government did not give all
of the harkis who wanted to migrate to France the freedom to cross the Mediterranean, because they
145
The photograph appears in Leïla Sebbar, Mes Algéries en France (Paris: Bleu autour, 2004), 194.
118
were officially “Algerians” with no legal link to France after the self-determination vote. The fact
that the harkis were Algerian—and not French—citizens became particularly salient at this time. As
explained in chapter one, per international law, because the harkis did not have French nationality
the Army had no recourse to intervene in Algerian internal affairs to rescue the harkis from violence
perpetrated against them by their compatriots after independence.
Concurrently, FLN leaders questioned the harkis’ Algerian nationality and, in some cases,
stripped them of it owing to their role in torturing other Algerians during the war. This included the
actions of “harki police” in Paris (Force de police auxiliaire or FPA) in the October 17, 1961 “Paris
massacre” or “Battle of Paris,” which particularly marked Algerian consciousness on both sides of
the Mediterranean.146 On this fall evening, thirty thousand pro-FLN Algerians protested peacefully
against the new curfew imposed on North Africans in metropolitan France by marching from
shantytowns and lodging houses in the banlieues to the center of Paris. Under the orders of prefect
Maurice Papon, police killed as many as two hundred Algerians (the precise figure is elusive), some
of whom they pushed into the Seine, and rounded up approximately ten thousand others and
interned them in the Palais des Sports at the Porte de Versailles and the Stade Coubertin in the
sixteenth arrondissement.147 In the days that followed, FPA officers assisted French police as they
questioned and tortured the prisoners to extract information about FLN activities in metropolitan
France.
The place of the harkis in torturing their compatriots in France and Algeria, the extent of
which is not clear, caused a backlash from FLN members following the ceasefire, who questioned
whether the harkis merited Algerian nationality. One intelligence report in June 1962 disclosed
For historical accounts of the October 17, 1961 massacre, see: Joshua Cole, “Remembering the Battle of Paris: 17
October 1961 in French and Algerian Memory,” French Politics, Culture & Society 21.3 (2003): 21-50; Jean-Luc Einaudi, La
bataille de Paris: 17 octobre 1961 (Paris: Seuil, 1991); and House and MacMaster, Paris 1961. While Cole and Einaudi refer
to the October 17 event as a “battle,” House and MacMaster’s account labels it a “massacre” because “it was the
bloodiest act of state repression of street protest in Western Europe in modern history” (1).
147 On debates about the precise number of Algerians killed see: House and MacMaster, Paris 1961, 6-13 and Cole,
“Remembering the Battle of Paris,” 25-29.
146
119
information about a trial of a former harki soldier for “crimes contre le peuple algérien” presided
over by two FLN judges. Three former harkis who had recently joined the ALN—known as marsiens,
those who rallied to the FLN cause right before the March ceasefire—were tasked with searching
the countryside near Orléansville, one hundred miles west of Algiers, for their ex-comrades in arms.
After capturing this particular harki, the new ALN members showed him photographs of harki
cadavers with no eyes and ears. They then gathered the signatures of forty witnesses who attested
that the harki acted in opposition Algeria and they escorted him and three of the witnesses to testify
at the trial before FLN judges.148 Another report from the French National Police revealed the
existence after independence of an Algerian “purge commission” (commission d’épuration), which
revoked Algerian nationality from harkis convicted of minor offenses and forced them to leave the
territory.149 The Algerian government created this commission in November 1962 to adjudicate on
Algerians who exhibited “anti-national behavior” during the war.150 These actions were a direct
violation of the Evian Accords.
The salient question is not whether the new Algerian government violated the Evian
Accords by not stopping the violence carried out against the harkis and by forcing certain harkis to
leave Algeria while preventing others from emigrating. It undeniably breached in multiple ways
Chapter Two of the General Declaration, specifically the clauses seeking to safeguard “Muslims”
and “Europeans” who remained in Algeria after independence from imprisonment, prosecution, and
discrimination owing to their allegiances and actions during the war. But rather, what is pertinent for
this historical examination is whether President de Gaulle and his government knew about these
breaches of the Evian Accords, how they reacted, and what motivated their actions.
SHAT, 1H 1793/1, 10e Région militaire, CAA, ZOA, Secteur d’Orleansville, “Bulletin de renseignements:
Circonstances de l’enlèvement de [MB],” N° 479/SO/OPR/R.SC, June 21, 1962.
149 ADBR, 138 W 6, SCINA, “Synthèse… journée du 6 février 1963,” N˚ 1 840.
150 MAE, SEAA 152, Secretariat of State for Algerian Affairs, “Note sur les supplétifs de l’Armée française en Algérie,” 5,
Nov. 26, 1962.
148
120
In late March 1962 when the government mandated a policy of requiring approval from
Ministers Messmer or Joxe for each individual harki who requested repatriation to France, there
were minimal retributions against the harkis for their support of the French cause. These acts did
increase in the following months, according to reports from soldiers and harkis themselves, yet still
remained relatively isolated. In fact, this was a calculated strategy on the FLN’s part. An April 10
directive from the fifth military region in Algeria (wilâya 5, situated in the western third of Algeria)
cautioned ALN members against exacting revenge on the “traitors” too soon: “le cessez-le-feu
n’étant pas la paix, nous saurons user de tact et agir avec souplesse afin de les gagner provisoirement
pour ne pas leur donner l’occasion de faire encore le jeu de l’ennemi […]. Leur jugement final aura
lieu dans une Algérie libre et indépendante devant Dieu et le peuple qui sera alors responsable de
leur sort […].”151 FLN leaders wanted to lure the harkis into a false sense of security and have them
believe that their support of France during the war was forgiven so that they would not migrate to
France and could be brought to justice. However, once Algerian citizens voted for their state’s
independence on July 1, in direct violation of the Evian Accords, civilians and ALN members
escalated dramatically their assaults on the “enemy” harkis.
At this time, the prevailing struggles for control of the government among factions within
the Algerian independence movement became more pronounced. The GPRA (led by Benyoucef
Ben Khedda), which had represented the Algerian people during the peace negotiations as the
official FLN organ of government, was rivaled by the État-major général (EMG) headed by future
presidents Ahmed Ben Bella and Houari Boumediène. Inspired by Franz Fanon’s revolutionary
program, the EMG had a radical vision of the Algerian state. Its leaders desired an Algeria with a
strong Muslim Arab identity and whose independence would be accompanied by a popular social
SHAT, 1H 1786/1, Le commandement de la wilâya 5, Directive n° 442, Apr. 10, 1962. Quoted in Harbi and Meynier,
Le FLN: Documents et Histoire, 535.
151
121
and economic revolution to forge the new state.152 This conception was not adopted during the May
28 to June 7, 1962 meeting of the FLN’s governing assembly, the Conseil national de la Révolution
Algérienne (CNRA), which favored a program that omitted any reference to an Islamic republic. As
fighting among leaders of the EMG and GPRA intensified, Ben Khedda dissolved the EMG on
June 30. On July 22, this latter group’s leaders—which, in addition to Ben Bella and Boumediène,
included pre-war Algerian nationalist leader and ousted GPRA president, Ferhat Abbas—formed
the “Bureau politique,” also known as the Tlemcen group for the northwestern Algerian city where
the leaders declared the group’s charter. The Tlemcen group supplanted the GPRA in discussions
with the interim governing structure, the Exécutif provisoire, which on August 2 granted the
Tlemcen group the power to prepare lists for the upcoming elections. The GPRA floundered on the
sidelines, but its supporters across the country continued to combat the Tlemcen group’s dominance,
leading to bloody clashes, arrests, and pillaging. Order was finally restored after the September 20
elections for the Algerian Assembly. All of the victors were partisans of the revolutionary Tlemcen
group.153
While this factional strife played out in Algiers and in the countryside, the absence of a viable
governing structure allowed chaos to prevail among Algerian residents. In particular, some Muslim
Algerians fervently believed that the harkis did not merit a place in independent Algeria. Cleavages
existed both between Muslims and Europeans and within each of these groups that led to Algerian
residents killing other Algerian residents, violence reminiscent of that during previous internal
conflicts, which French Army officials readily cited. In the midst of this tumult in July and August,
civilians and ALN members carried out a “purge” (épuration) of the harkis, as General de Brébisson
worded the telegram to Messmer cited in the epigraph to this chapter. The harkis’ actions to support
Thénault, Histoire de la guerre, 260-61. Fanon outlined his program in Les Damnés de la Terre (Paris: Librairie François
Maspero, 1961).
153 Thénault describes in detail the complex factional strife for control of the Algerian government that began in June
1961 at the first Evian peace talks in Histoire de la guerre, 259-63.
152
122
France led certain Algerians to seek to purge the population of harki “collaborators” against Algeria,
much as occurred during the 1944-45 “épuration” of Vichy sympathizers in France. During this
bloody phase of the Franco-French Vichy conflict, between eight and nine thousand collabos were
extra judicially killed.154 The use of “épuration” was significant for the parallel top-echelon Army
officials drew between these two conflicts opposing populations of the same nationality (Resistance
fighters and collaborators, harkis and FLN members). These groups’ battles after the end of the
military conflict with Germany during the Second World War and that with France during the
Algerian War for Independence led to internal divisions that would long influence their respective
postwar societies. One officer reported on July 22 about the explosion of violence since July 2 by
using a reference to an earlier Franco-French conflict, the Saint Bartholomew massacre. In 1572
Catholics massacred thousands of French Protestants in Paris because they viewed the Protestant
religion as incompatible with being a French citizen. In retaliation, Protestants attacked their
Catholic aggressors after which the violence soon spread to the provinces where it claimed the lives
of thousands more. With this massacre in mind, the officer wrote: “…[C]’est une ‘Saint Barthelemy’
qui a une allure particulière en ce sens qu’elle peut s’étaler. Son accélération toutefois devient très
inquiétante. A ce train là, d’ici trois mois, tous les musulmans qui ont été compromis avec nous, et
qui n’ont pu rejoindre la Métropole, seront massacrés ou périront dans des camps.”155
The camps to which the officer referred were created by the ALN and its successor, the
ANP, which took over posts abandoned by the French Army and established detention camps.
Their members tortured, murdered, and forced ex-harkis to perform tasks such as loading and
unloading trucks and cleaning after what one report called rulings that were “a travesty of justice.”156
Henry Rousso, “L’Épuration en France: Une histoire inachevée,” Vingtième siècle. Revue d’histoire no. 33 (Jan.-Mar.
1992): 84.
155 SHAT, 1H 1793/1, Lieutenant Colonel Gallouet to the General Commanding the CAA and the General
Commanding the CAT, “Objet: Massacres des Musulmans Francophiles,” N° 831/2.S, July 22, 1962.
156 ADBR, 137 W 5, SCINA, “Synthèse… journée du 9 novembre 1962,” d, N˚ 1 778. Several documents in carton 1H
3155/5 of the Army Archives describe the harkis’ forced labor. See, for example, 24° Corps d’Armée, Groupement “G,”
154
123
As they carried out these punishments, some of the men bore shaved heads and wore vests with a
large “H” on the back, branding them as outcasts in the new Algerian state and thereby questioning
whether they belonged as “Algerians.”157
Indeed, alarming reports of imprisonment, torture, and killings by Algerians who sought to
“purge” their population of presumed traitors streamed into officers at the military barracks where
the harkis took refuge. The Army Archives include lists with the location of detention centers and
precise testimonials from harkis who were witnesses and victims of violence.158 These include an
account from a harki who escaped incarceration and torture and, on August 13, sought asylum at the
Lodi military base south of Algiers.159 He described a four and a half hour interrogation during
which two ALN leaders and eight or nine soldiers (djounouds) and policemen intermittently
questioned and tortured him and three other captured harkis. ALN members employed tactics such
as the supplice d’eau (forcing water down prisoners’ throats with a hose until they could not breathe).
He reported that the officers fired questions about how many civilians and ALN soldiers
(moudjahidine) the harkis had killed, how many women they had raped, and why they did not listen to
FLN calls to desert the French Army and turn over their weapons. The officers vowed that the
prisoners’ children would soon be orphans. In the middle of the questioning, one of the harkis was
led out of the room and was “presumably executed” after he admitted to killing six men and one
“Bulletin de renseignements,” Oct. 13, 1962. Another carton, 1H 2745/1, contains reports from doctors who examined
tortured harkis, including photographs of their wounds.
157 This situation is reminiscent of that of European Jews and other groups such as homosexuals branded as “enemies”
who were forced to wear a Star of David sewn to their upper left side of their shirts during the Holocaust. The French
Vichy regime took this symbolic action a step further for its Jewish population, even before the April 29, 1942 laws
requiring Jews in Occupied France to display the star. As mentioned in footnote 93 above, an October 7, 1940 law
stripped all 110,000 Jews residing in Algeria of their French nationality and a July 22, 1940 law resulted in 7,000 Jews in
metropolitan France becoming foreigners over the course of the war. Weil, “Histoire et mémoire,” 9-10. Another
resonance with the situation of Jews during the Holocaust was that guards shaved the heads of all prisoners upon their
arrival in the concentration and extermination camps in Central and Eastern Europe.
158 See, for example, SHAT 1H 3155/5, “Liste des Lieux de détention d’anciens supplétifs des F.A.F. par l’A.L.N,” Aug.
1962. This carton also contains a detailed map with the location of camps in the Oran region drawn by members of the
Oran Army Corps (CAO).
159 SHAT, 1H 1793/1, Colonel De Nadaillac (Commandant le 20º Division d’Infanterie Groupement E), “Audition du
nommé AA venu se réfugier le 13 août 1962 auprès des Forces Armées à Lodi,” Nº 1230/GE/2/S. The full text is
reproduced in Appendix C.
124
woman during a protest in Medea commemorating the November 1 outbreak of the Algerian War
for Independence (no year was given). After the official interrogation was completed and the ALN
leaders departed, the police once again beat the three remaining harkis and tied their hands and feet
with electric wire. However, when the guards fell asleep one of the harkis was able to escape and
immediately fled to the Lodi military base where officers recorded his story and sent it to their
superiors, who then transmitted the report to General de Brébisson.160
The document does not indicate whether the harki asked to be repatriated to France and it is
not clear whether he was aware of the procedure in place, that is, he would need to demonstrate that
his life was “severely threatened.” Convincing French authorities of, on the one hand, the
impending danger to himself and, on the other, the magnitude of the violence toward his fellow
harkis was a necessity for his own repatriation and a legitimization of rumors of violence. It is also
impossible to glean whether he told the story in French, Arabic, or Berber and, therefore, what was
lost—or added—in translation to French either by the military officers or the translator, who was
possibly an Algerian native working for France, a harki himself.
Nevertheless, this document and the myriad others like it demonstrate that the French
government received grisly reports of violence and needed, based on this information (in whatever
form it came), to adjudicate whether to allow individual harkis onto French soil.161 The Army
Archives contain summary reports, like that in Appendix B, which enumerated fifty separate
instances of violence with brief descriptions of the acts that occurred across Algeria between July 11
and August 6, 1962. A July 11 account from a small town near Algiers related, “Enlèvement d’un
Moghazni, son épouse et ses deux enfants. Relâché le 12, l’épouse a déclaré que son mari avait été
Ibid. Brigadier General Lève of the 23º Corps d’Armée transmitted this report to de Brébisson on Aug. 22, 1962
(document Nº 2246/23ºCA/B2).
161 French newspapers, tilting both right and left, occasionally printed articles about the harkis’ imprisonment and torture
by Algerians. See, for example, SPDP, dossier: Harkis, “Le cas des harkis,” Le Monde, Sept. 22, 1962; SHAT, 1H 1170,
“Un ancien harki fait le récit des tortures qu’il a endurées,” Le Figaro, Nov. 11, 1962; SPDP, dossier: Harkis, Pierre VidalNaquet, “La guerre révolutionnaire et la tragédie des harkis,” Le Monde, Nov. 11, 1962; SHAT, 1H 1170, “Livre blanc de
notre honte et de la passion des harkis,” La Nation Française, Nov. 14, 1962.
160
125
exécuté.” 162 An August 4 report from the eastern town of Lambèse revealed, “Environ 70
prisonniers ex-harkis, moghaznis et civils musulmans ayant coopéré avec la France sont actuellement
internés à la Centrale de Lambèse et gardés par des militaires de l’ALN. Ils sont maltraités et
astreints à des travaux de force. Leur état de santé paraît déficient et ils semblent sous-alimentés.
Certains portent des traces de sévices.”163 These reports were compiled into one document—located
in Messmer’s personal archives—addressed from General de Brébisson directly to Messmer himself.
Therefore, it is clear that top echelon army officials in Algeria kept their counterparts in Paris
apprised of these situations.
Despite such violations of the peace treaty, French soldiers and military police continued to
relay these instances of imprisonment, torture, and slayings and to try, often in vain, to negotiate
with local ALN leaders. One July 31 intelligence report encapsulates the Army’s lack of military
action to combat the violence toward the harkis. In the body of the document the author wrote,
sometimes offering precise details, that ALN soldiers were capturing harkis and putting them into
forced labor camps where some of them were tortured and executed. His concluding remarks,
“Mesures prises: Diffusion Autorités Gendarmerie,” suggest that the Army could now only assume a
passive role to hinder the brutality.164 In turn, according to the author of the July 22 report quoted
above, the FLN tried to exploit the Army’s lack of action to instill fear among the harkis. He
communicated to his superiors: “La terreur qui se lit sur le visage des musulmans, à leur arrivée dans
les postes militaires où ils se réfugient, est insigne. Ces musulmans sont persuadés que toute fuite
leur est rendue impossible. Le FLN fait en effet courir le bruit que, dans ce domaine, l’Armée
SHAT, 1K 744 (dossier: Situation des harkis en Algérie), General de Brébisson to Minister of Armies Messmer
(cabinet), “Objet: Situation des ex-harkis en Algérie depuis le cessez-le-feu,” Aug. 17, 1962.
163 Ibid.
164 SHAT, 1H 1793/1, Gendarmerie Captain Brachet, “Objet: Bulletin de renseignements sur exactions des éléments de
l’ALN contre les anciens harkis et moghaznis,” N° 381/4.R, July 31, 1962.
162
126
française n’a plus aucune possibilité d’action, et d’autre part, que les quais d’embarquement sont
tenus par l’ALN.”165
The Army’s inaction was based on orders from Messmer and ultimately de Gaulle, who did
not want to reignite the war, now that he had shifted his full attention away from the Algerian War
for Independence against the FLN and toward the Franco-French war with the OAS. He wanted to
forget immediately French Algeria, exclude Algerians from the new French nation-state, and focus
on building a strong post-imperial France devoid of vestiges of its colonial empire. The harkis
served as reminders of France’s fall from colonial dominance. They were symbols of both
collaboration—they continue to be associated with the word “traitor” and “collabos” on both sides of
the Mediterranean—and France’s loss of a Third World nation it claimed as a vital part of its own
territory for 132 years.
In a series of interviews conducted with journalist Philippe de Saint Robert between late
2001 and early 2003, Messmer revealed that he questioned de Gaulle about what action the Army
should take in the face of the “massacres de harkis”: “‘Est-ce que vous autorisez des opérations
coup de poing de l’armée française pour récupérer les harkis sur telle ou telle zone que nous avons
évacuée?’ Le général de Gaulle refuse! Il refuse en disant: ‘Vous allez rallumer la guerre d’Algérie’ –
ce qui est un risque qu’il refuse de prendre.”166 Messmer distanced himself from this position, and
from de Gaulle whom he claimed “n’a pas eu pour les harkis beaucoup de considération, ni de
commisération,” by contending, “A mon avis, on aurait pu faire ces opérations et courir ce risque. Il
y aurait sans doute eu quelques batailles rangées, mais je ne crois pas que la guerre d’Algérie aurait
recommencé, je ne le crois pas.”167 Messmer’s recounting of de Gaulle’s firm stance against French
military officers saving the harkis from danger resonated with words that the President had spoken
SHAT, 1H 1793/1, Lieutenant Colonel Gallouet, “Objet: Massacres des Musulmans Francophiles.”
Pierre Messmer, Ma part de France: Entretiens avec Philippe de Saint Robert (Paris: François-Xavier de Guibert, 2003), 119.
167 Ibid.
165
166
127
during the July 25, 1962 Council of Ministers meeting cited above: “On ne peut pas accepter de
replier tous les musulmans qui viendraient à déclarer qu’ils ne s’entendront pas avec leur
gouvernement!”168 In these conversations, it is clear that de Gaulle viewed the harkis—who, for him,
remained Algerian Muslims—as outsiders to France, much like the new Algerian government
viewed them as outsiders to Algeria. As a result of the president’s consistent position, the French
military was constrained in the help that it could offer the threatened harkis.
The French government’s policy of forbidding military officers from carrying out any rescue
operations that risked reigniting the war with the FLN aimed to minimize the number of harkis in
the relocation camps. For, once in these spaces, they only needed to overcome one last
administrative procedure. Upon the harkis’ arrival, Army information officers interviewed them and
submitted a “demande de rapatriement.” This form included the ages, birthplaces, birthdates, and
vocation of the harki and all of his family members to migrate with him, information about his
military service, the “motif détaillé des menaces,” his current place of residence, and the officer’s
opinion about whether the harki was “worthy” of being repatriated. In one case, the agent described
the situation of a harki in the Oran region: “AG semble sérieux dans ses déclarations… Il semble
que les menaces dont il est l’objet soient réelles. S’il est vrai, l’exemple de l’enlèvement des épouses
de ses amis [A] et [M] pourraient en confirmer l’importance. En conclusion, il semble que son cas
soit digne d’intérêt.”169 Other examples included mentions of the harki’s honorable service, his good
comportment, his ability to speak French, and his state of mind. These reports—with stamps such
as “très vraisemblable,” “recoupé,” “intéressant,” and “douteux”—were then passed onto the
officer’s superiors, and eventually worked their way across the Mediterranean for final judgment by
Ministers Messmer or Joxe. Therefore, similar to the process for Algerian Muslims to obtain French
Quoted in Peyrefitte, C’était de Gaulle, 209.
AG, A, and M are the initials of the harkis. SHAT, 1H 3229/6, Lieutenant Jimenez (20° Bataillon de Chasseurs
Portes), “Objet: Demande de Rapatriement, [AG],” 3, N° 737/O.R./20° BCP, May 7, 1963.
168
169
128
nationality during the colonial era described in chapter one, officers’ opinions about whether the
harki was sufficiently threatened played a role in whether he and his family could surmount the
second obstacle to French nationality.170
Despite reports like those cited above with references such as “Saint Barthelemy” and
“épuration” to relate the severity of the situation to de Gaulle’s cabinet members, as anarchy
increasingly reigned in the Algerian countryside, the government upheld its policy of strictly
monitoring the repatriation process. In summer 1962 high-level government officials continually
reminded officials in France and Algeria of the procedure that every harki required individual
authorization to migrate to France. For example, on July 23, 1962, Joxe wrote to the newlyappointed French Ambassador in Algeria, Jean-Marcel Jeanneney: “Je vous serais obligé de me tenir
étroitement informé, ainsi que le Secrétaire d’État aux Rapatriés, des demandes de migration de
musulmans qui vous seraient présentées et de ne mettre personne en route avant d’avoir les
directives du Ministre des Armées ou de moi-même.”171 And on August 11 Messmer wrote to the
Regional Army Commanders in France, “Aucun acheminement sur les camps d’ex-supplétifs ou de
familles venus d’Algérie hors-plan…”172 Put simply, the government did not welcome the harkis to
freely cross the Mediterranean to escape the violence.
The government’s response to the violence toward the harkis demonstrated that post-1962
France wanted to sever the connection between French and Algerian post-imperial spaces and
populations. The intensification of brutality in Algeria, which pushed the harkis who survived
imprisonment and torture to seek refuge at French military barracks, did not influence the French
government to open its borders to accept more harkis. Instead, the circumstances on Algerian soil,
Cartons 1H 3229/6 and 1H 1793/1 in the Army Archives (SHAT) contain many examples of the “demandes de
repatriement,” the majority of which were favorable to repatriation.
171 SHAT, 1R 336/8 (dossier: Rapatriement des harkis), Minister Joxe to Ambassador Jeanneney, “Objet: Envoi en
France des supplétifs musulmans,” N° 187/CAB, July 23, 1962.
172 SHAT, 1R 336/8 (dossier: Rapatriement des harkis), Minister Messmer to the Generals Commanding the First
through the Seventh Military Regions in France, “Objet: Ex-supplétifs FSNA et leurs familles repliées d’Algérie,” 1, N°
7736/EMAT/4-PO, Aug. 11, 1962.
170
129
of which government officials in France were fully apprised, engendered increased resistance to
allow the harkis safe passage to France. For example, in early July when the Bourg-Lastic and Larzac
camps reached beyond full capacity, exhausting the Ministry of Armies’ budget and resources,
Messmer—who received frequent updates about the perilous situation in Algeria from his
subordinates—relayed to his counterpart in the Ministry of Algerian Affairs that it was “out of the
question” for the Army to house any more harkis. But beyond the practical considerations of space
and human and physical materials, Messmer argued, in spite of the massacres, that “few individuals”
should be allowed access to metropolitan France, thereby ignoring the actual situation of violence on
Algerian soil and limiting the movements of these now Algerian citizens:
Il semble d’ailleurs que, les intéressés ayant eu le temps et la possibilité d’apprécier les
conditions de leur reconversion en Algérie, les options pour la Métropole qui pourraient
aujourd’hui se manifester, ne devraient être le fait que de quelques rares individus, pourvu
que soit vérifiée soigneusement la justification du départ sollicité et que soient déjouées
toutes manœuvres, inconscientes ou mal intentionnées, pour maintenir le courant des
départs.173
He then requested from Joxe a rapid intervention from Jeanneney to stop the flow of harkis to
France.174 Messmer’s words strongly discouraged the harkis’ migration to France and once again
demonstrated Messmer’s insistence on exerting control over the “courant des départs.”
A month later, after the many reports cited above about the surge of violence toward the
harkis, Messmer firmly held his position of limiting France’s responsibility toward the former
soldiers in response to an appeal from a top general stationed in Algeria. General Frat provided
details of ten separate episodes of torture and slayings, explained that Army officials tried
unsuccessfully to negotiate with local Algerian officials, and argued that the continuation of the
retributions would lead to the “extermination” of all the Muslims who had demonstrated sympathies
SHAT, 1K 744 (dossier: Premières mesures d’accueil en France), letter from Pierre Messmer to Louis Joxe, N°
019588 MA/CC, July 19, 1962.
174 Ibid.
173
130
for France.175 When he received the plea, Messmer handwrote at the top of the letter a note for one
of his cabinet members to relay to General Frat: “Nous atteindrons alors au chiffre de 20 000
supplétifs et parents transportés en France, ce qui est considérable à tous égards.”176 Messmer’s
words here, and those in the letter cited above, minimized France’s role in defending the harkis
against retaliatory actions perpetrated by Algerians, and distanced the harkis from French nationality.
Becoming “French” Citizens
Despite the government’s efforts to limit the harkis’ movement to within the new
geographical boundaries of France, an estimated 100,000 harki men, women, and children reached
metropolitan France during the 1960s, whether through government repatriation or “individual
initiatives.” After successfully arriving within the geographic boundaries France—thereby
overcoming the second obstacle to French nationality posed by the July 21, 1962 legislation—the
harkis needed to surmount one last hurdle to become French citizens, one that would allow them to
cross over into the figurative boundaries of the French polity. As per the ordinance, individuals who
had been French with local civil status needed to present themselves in front of a magistrate with a
series of documents and “se faire reconnaître la nationalité française.” While this administrative
procedure posed numerous logistical problems, it was arguably the least arduous of the three
obstacles to surmount. Of the nearly seventy thousand adult harkis who requested French
nationality between 1962 and 1970, the government accorded it to 86 percent of them (see table
3).177 However, the very existence of the process provoked its greatest defect, for it created a lasting
psychological distance between the harkis and other French citizens because they needed ask for
SHAT, 1R 336/8 (dossier: Rapatriement des harkis), letter from General Frat to Minister of Armies Messmer, 1, Aug.
25, 1962.
176 Ibid.
177 ACNMF, 6, Jean Servier, “Enquête sur les Musulmans Français,” Part 1, 13, 1972. The Ministry of Labor and the
CNMF commissioned sociologist Jean Servier to compile statistical data about the harki population’s size, geographical
distribution, living conditions, nationality, and social composition, information that he used to author a 114-page report.
175
131
French nationality in the first place. French with common civil status—the tiny number of Muslim
Algerians who had become full French citizens during the colonial era and pieds-noirs—were
automatically granted French nationality, no matter where they resided.
In contrast, the July 21 ordinance stipulated that Muslim Algerians living in Algeria lost
French nationality on July 2, 1962 (the day after the self-determination vote), and those living in the
metropole were French only until the arbitrary date of January 1, 1963. After this time, their
passports and identity cards expired and in order to become French citizens, they would need to
make a “déclaration recognitive de la nationalité française” in front of a magistrate (juge d’instruction)
in the metropolitan French department where they resided. The precise language in the July
ordinance left Algerian Muslims’ nationality status when they arrived in France up for interpretation:
were they becoming or re-becoming French citizens? The first paragraph of the article that
addressed Muslims with local civil status stipulated that they could “se faire reconnaître la nationalité
française” once in France and its second paragraph specified that they would not be able to “établir
leur nationalité française” unless they followed the conditions outlined in article 156 of the
Nationality Code.178 Both these phrases suggest that Algerian Muslims had lost French nationality.
When asked about the continuity in the French nationality status of those who made a declaration
after January 1, 1963, government officials offered varied responses. Most often they maintained
that these Algerian Muslims legally never had never been deprived of their French nationality.
However, the Minister of Repatriates wrote in a June 17, 1963 circular that Algerian Muslims who
had not yet undergone the process outlined in the July ordinance had “une présomption de
‘nationalité étrangère.’” 179 They held what amounted to a virtual nationality during this liminal
Article 156 is one of the six articles comprising Title VII of the French Nationality Code. This title, “Des effets sur la
nationalité française des transferts de souveraineté relatifs à certains territoires,” was added to the Code in 1960 to
address the nationality of French citizens born or residing in former colonial holdings.
179 CAC, 19770391/8, circular from the minister of Repatriates to departmental prefects and subprefects and regional
delegates of the ministry of Repatriates, “Objet: Conséquence de l’application de l’article 2 de l’ordonnance nº 62-825 du
178
132
moment until they appeared before the magistrate because they did not have the rights and benefits
bestowed on French citizens. Most notably, they were ineligible for diplomatic protections when
traveling outside of France (for example, if they returned to Algeria to get family members that they
had left behind) and social welfare laws including the 1961 Boulin Law that granted concrete
financial assistance to repatriates from former colonies and protectorates.180 Quite simply, as had
been the case during the colonial period, native Algerians had to wade through a legal quagmire in
order to acquire full French citizenship.
For over four months after de Gaulle issued the ordinance, no legal text followed to explain
the procedure. Finally, on November 27, 1962, Prime Minister Georges Pompidou issued a decree;
however, the text remained vague about which documents Algerian Muslims needed to procure.181 A
February 1963 circular from the Minister of Interior to the Departmental Prefects and Regional
Military Inspectors (IGAME, who were the interlocutors with the camp directors) worked toward
clarifying the procedure. It stipulated that the Algerian Muslims’ dossiers needed to include a birth
certificate to demonstrate that they were at least eighteen years old, an identity card to prove their
French nationality on the date of the self-determination vote, the birth or death certificate of at least
one parent, and proof of current residence in France. If applicable, harkis also had to provide a
marriage certificate and minor children’s birth certificates. Husbands and wives each had to furnish
their own documents and minor children were exempt from this process as their nationality was
determined by that of their father or their widowed mother. If the dossier was incomplete, the judge
gave the harki a certificate (attestation) valid for six months, which was not proof of nationality, and
21 juillet 1962, relatif à la reconnaissance de la nationalité française des personnes de statut civil de droit local originaires
d’Algérie pour l’attribution aux musulmans algériens la qualité de rapatrié,” 1, Nº 63-100 AL/Reg, June 17, 1963.
180 At the same time, some harkis delayed requesting French nationality until after their wives and children safely made it
onto French soil as they feared repercussions for these family members if Algerians from their villages learned that they
chose to be French citizens. ACNMF, 15/13, Colonel Schoen, Report: “Reconnaissance de la nationalité française par
les Musulmans d’origine algérienne,” 22, Jan. 21, 1964.
181 JORF, “Décret n° 62-1475 du 27 novembre 1962 relatif à la procédure de reconnaissance de la nationalité française
prévue à l’article 2 de l’ordonnance n° 62-825 du 21 juillet 1962,” Dec. 7, 1962, 12012.
133
instructed the harki to return with the appropriate documentation. If the dossier was complete, the
judge handed the harki a receipt (récépissé) and then he was required to send the dossier the same day
for approval to the Ministry of Public Health and Population, which had six months from this date
to make a decision.182
This ministry’s Naturalization Office had four options for the harkis’ applications. The first
was “refus en l’état.” Even though the local judge had accepted the harkis’ dossier as complete, the
Naturalization Office could determine that it was missing documentation. The harki was then asked
to submit the proper documents to the local court and the six-month waiting period would restart.
The second possibility was “un ajournement,” if officials disagreed about whether to accept the
request. In this situation, the Ministry could delay the decision until after July 22, 1965, during which
time no appeal was possible. Third, the state could oppose the request, “une opposition,” because
the harki was judged “indigne de la nationalité française,” for reasons such as financial debt or
inability to prove that he was a “repatriate” who first arrived in France after the end of the war. The
harki had six months to appeal this decision by furnishing evidence to the contrary. If the state
denied the appeal, the harki was indefinitely refused French nationality. For the first 40,000
applications (until April 30, 1965), the state handed down 1,800 refusals as is, 1,600 deferrals, and
800 oppositions.183 However, as was the case for the vast majority of the harkis, the state approved
their dossier and, often within six months, a nationality certificate was mailed to the local judge’s
office for them to pick up.184 As the chart in table 3 reveals, 59,864 of the 69,303 harkis who sought
French nationality between 1962 and 1970 became French citizens. The state rarely deferred
nationality requests and even more rarely deemed individuals unworthy of being French citizens.
AN, F1a 5125, Minister of Interior, Circular n° 104: “Notice relative à la Nationalité Française des personnes de
statut civil de droit local originaires d’Algérie,” Feb. 12, 1963.
183 CAC, 19920149/5/17, Colonel Schoen (CNMF), “Questions de nationalité (Renseignements donnés par M. Ahriche,
et commentaries,” 5-6, Nº 370/S, Apr. 30, 1965.
184 ACNMF, 15/13, Colonel Schoen, Report: “Reconnaissance de la nationalité française…,” 14-15.
182
134
Instead, the majority of the fourteen percent of refusals were linked to logistical impediments such
as dossiers not containing the required documents and confusion over the modalities of the
process.185
Year
Number of Harkis
Requesting French
Nationality
Number of Harkis
Granted French
Nationality
Number of Harkis
Refused French
Nationality
1962
1963
1964
1965
1966
1967
1968
1969
1970
{
Total from 1962 to 1965:
58,000
}
6,725
4,002
304
156
116
______
69,303
100
16,926
19,575
10,260
6,775
4,002
1,294
545
298
______
59,864
{
Total from 1962 to 1965:
8,099
}
50
0
990
298
182
______
9,619
PERCENTAGE
86%
14%
TOTAL
Table 3: Harkis age eighteen and older requesting, granted, and refused French nationality from 1962 to 1970 based on
data from the Ministry of Population’s Naturalization Office. A comparison of the first column (requests) with the
second and third columns (decisions) reveals that the delay was often greater than six months, particularly toward the
end of the decade. Source: ACNMF, 6, Servier, “Enquête sur les Musulmans Français,” 13.186
While the preceding paragraphs succinctly outline the process, in actuality it remained obtuse
to local government officials and the harkis, particularly those residing outside the camps. In some
cases the harkis did not realize that they did not have French nationality until they needed to provide
a nationality certificate for job applications, welfare requests, or other official procedures.187 In an
effort to aid local administrators understand the texts he believed to be “difficult to understand,” in
early 1964 the General Secretary of the CNMF issued a twenty-seven-page document.188 This text
also sought to address the principal problems that the harkis had hitherto encountered, the two most
Ibid., 26.
In an article written in 2000, Charles-Robert Ageron estimated that the number of harkis who requested French
nationality from 1962 to 1967 was 84,000, which is nearly 15,000 more than Servier. However, Ageron does not specify
a source for this figure, instead writing that 84,000 is a “generally accepted” estimate. Ageron, “Le ‘drame des harkis,’” 5.
187 Heinis, “L’Insertion des Français Musulmans,” 42.
188 ACNMF, 15/13, Colonel Schoen, Report: “Reconnaissance de la nationalité française…,” 1.
185
186
135
prevalent of which were trouble procuring official documents and not having a fixed residence. Both
of these complications required the harkis to interact more often with the state. These dealings were
particularly difficult owing to linguistic barriers, concerns over identifying themselves publicly as
harkis, and unfamiliarity with the French bureaucratic system.
The first problem arose because many harkis who fled in haste to military bases to escape
the violence did not bring their identity papers with them. Meanwhile, others who were imprisoned
in ALN or ANP camps were stripped of their papers. Recognizing that many individuals abruptly
leaving Algeria—whether pieds-noirs, harkis, or labor migrants—would encounter difficulties
requesting copies of official documents from the new Algerian government, President de Gaulle
issued an ordinance on July 16, 1962 to facilitate the procurement of civil records issued in Algeria.
This text allowed birth certificates and identity cards to be substituted by presenting their official
family record book (livret de famille) or their family civil status record (fiche familiale d’état civil) or
through an affidavit process whereby three people over the age of eighteen attested to the
individual’s identity in front of a magistrate (acte de notoriété).189 Many harkis needed to resort to this
latter option, which added an extra step to completing the nationality process.
Second, those harkis who did not have a fixed residence encountered several additional
difficulties in becoming French citizens. As noted in a 1963 report by the security branch of the
national police force (Renseignements Généraux), the harkis exhibited a “mobilité quasi continuelle,”
most often because they were in search of employment or felt threatened by other Algerians in the
community. The harkis did not always inform authorities when they relocated either because they
feared interacting with state agents or did not realize that they should keep officials apprised of their
movement.190 This mobility posed a problem if the Naturalization Office had any questions about
JORF, “Ordonnance n° 62-800 du 16 juillet 1962 facilitant la preuve des actes de l’état civil dressés en Algérie,” July
17, 1962, 7005.
190 AN, F1a 5017, Direction des Renseignements Généraux, Report: “Aspects du problème des harkis en France,” 13.
189
136
their dossier since officials contacted the harkis by mail at the address given on their application. If
the harkis did not respond within six months, then their request was refused. Moreover, when a
magistrate required harkis to procure additional documents for their initial request, if they had
moved, the harkis then needed to submit these papers to the magistrate in their former place of
residence. The same was true when picking up their nationality certificate as this document was
mailed to the magistrate in front of whom the harkis had made their request.191 Finally, frequent
movements posed a problem because to initiate the entire process the harkis had to furnish proof of
residence. Whereas a certificate from the camp director sufficed for the encamped harkis, the harkis
residing outside of the camps needed to ask their landlord for a letter or provide a utility bill in their
name. Furthermore, temporary housing such as emergency shelters and nursing homes—where
elderly and wounded harkis resided—were not considered permanent domiciles. These disabled
harkis, some of whom were among the most destitute and without possibility for employment owing
to their physical and/or psychological conditions, could not set in motion the nationality process,
and therefore were ineligible for the social welfare benefits they desperately needed.192
191
192
ACNMF, 15/13, Colonel Schoen, Report: “Reconnaissance de la nationalité française…,” 8 and 15.
Ibid., 10.
137
Figure 8: Makeshift courtroom
set up for nationality ceremonies
at the Rivesaltes camp. Source:
“C’étaient les harkis.”
Figure 9: A harki in front of
the magistrate during a
nationality ceremony in
summer 1963 at the
Rivesaltes camp. Source:
“C’étaient les harkis.”
Indeed, the harkis in the camps had an easier time logistically in becoming French citizens
than those outside of the camps. Camp officials kept them informed of the necessary steps they
138
would need to take; affidavits were easier to obtain since they lived among thousands of other harkis
who could act as witnesses; and the camp director furnished them with a certificate of residence and
submitted ahead of time the requisite dossier of documents. A report broadcast in July 1963 on
France’s public television station’s ninety-minute weekly political news journal, 5 colonnes à la une,
provides the only known film footage of the weekly nationality ceremonies.193 As pictured in figure 8,
the opening scene of the fifteen-minute piece titled “C’étaient les harkis” showed a makeshift
courtroom, unadorned except for a bust of Marianne symbolizing the Republic and a photograph of
President de Gaulle, set up in one of buildings at the Rivesaltes camp. A magistrate and a court
reporter traveled fifteen kilometers from Perpignan on Wednesday mornings to officiate the
procedure. In a process that often took less than a minute, the judge called each person and asked:
“Vous désirez garder la nationalité française?” Wives were summoned immediately following their
husbands and the men would sometimes serve as translators for those who could not understand
French. The documentary also showed that when all else failed, court officials and harkis interacted
with hand signals, such as motioning for a signature. After responding “Oui, Monsieur le juge,” as
did the harki in figure 9, the “new” citizen was asked to sign a declaration like that of Mohamed M.
pictured in figure 10.
“C’étaient les harkis,” 5 colonnes à la une, dir. Jean-Claude Bringuier, RTF (July 7, 1963). 5 colonnes à la une is a reference
to the traditional five columns appearing on the front page of newspapers. The program was journalistically innovative
as the chief reporter always appeared on site and his role was a combination of a director and a journalist. It ran from the
day after de Gaulle reascended to the presidency on January 9, 1959 to May 1968 when its directors stopped production
shortly before de Gaulle stepped down to protest the government’s actions during this period of social unrest. The series
often tackled delicate political issues, including the Algerian War for Independence and its aftermath, for which it was
censured several times. Francis James, “5 colonnes à la une ou le journalisme total,” Cinémaction no. 84 (1997). “C’étaient
les harkis” presented an overall positive vision of the harkis’ situation, which contrasts with that expressed in first-person
narratives by former camp residents. For example, the show included multiple interviews with harkis who avowed their
thankfulness to be living in France and none who said that they wished to return to Algeria.
193
139
Figure 10: Copy of Mohamed M.’s nationality declaration submitted on October 22, 1963, and approved on June 1, 1964.
Source: Archives du Service des Rapatriés des Bouches-du-Rhône (hereafter ASRBR), social welfare file of Mohamed M.,
Nº B338126.
This document, which would be required for future interactions with the state, included the
date, his name and birthplace, and the names and birthplaces of his parents. The judge’s signature
and his official stamp (surrounded by a black square on the image) attested to Mohamed M.’s
submission of an notarized deed in lieu of his birth certificate, proof of French nationality on July 3,
1962, and a certificate of residence from the Rivesaltes camp director. Since he did not know how to
write in French, Mohamed M. was not required to sign his name; instead, the declaration was
marked “ne sait pas signer” (“ne sait pas signer,” surrounded by a black oval on both panels of the
image). He then received a receipt for this document and the original was sent to the Naturalization
140
Office in Paris for approval. Although there is no indication of the reason for the delay, and the
judge approved that he had submitted the proper documentation, Mohamed M.’s request took eight
months instead of the six months prescribed by the legislation.
All harkis who took part in this ceremony faced a lasting psychological hurdle to being
included into the French community. The requirement to undergo the administrative procedure to
become French citizens and the question that they needed to answer—“Vous désirez garder la
nationalité française?”—confused the harkis’ past, present, and future relationship with the French
state. The harkis at the Rivesaltes camp and elsewhere were asked if they wanted to “keep” their
French nationality when, in fact, they had technically lost this designation on July 3, 1962 when
Algeria became independent. Therefore, when they arrived in front of the judge their legal
relationship with France was not clear. Moreover, the procedure to become French citizens again
was akin to the one in place for all foreigners and was the exact same for all former French with
local civil status, including those who had fought against France during the Algerian War for
Independence. The state did not view the harkis’ previous military service, long considered a crucial
component in building French citizens, as a reason to grant them automatic French nationality in
1962. After the end of the war when the French Army no longer needed their services on the
battlefield, in terms of their future legal relationship with the state, the harkis were foreigners and
“Algerians” like any others. They were neither “Franco” (the pieds-noirs who automatically became
French citizens) nor “French.” De Gaulle’s focus on rapidly ending the Franco-French war
obscured the harkis interests, and their inclusion within French society.
Conclusion: Harki Citizens
Government actors in the French government both before and after Algerian independence
distanced the harkis from French citizenship—legally and in practice—just as those in the imperial
141
French government had done with Algerian Muslims. The July 21, 1962 legislation required French
with local civil status, the colonial citizenship category to which almost all harkis belonged, to
undertake an obstacle-ridden process to become French citizens. High-level officials delayed
enacting a repatriation plan for harkis in spite of the extreme FLN violence. They enacted strict
control over which harkis could be repatriated. And government policy dictated placing harkis
asking for government assistance in refugee-like camps once they arrived in France.
In the aftermath of the loss of French Algeria, the last vestige of France’s colonial empire,
President de Gaulle and his government turned their attention inward to focus on “Franco-French”
problems (i.e., its lingering battle with the OAS and efforts to reassert France’s strength in the
world) and “Franco-French” citizens (i.e., French repatriates and metropolitan French).194 The harkis,
for de Gaulle and others in his cabinet, were “Algerians” and not “French.”
Muslim French citizens from Algeria, whether they fought for or against France during the
“affaire d’Algérie,” were legally indistinguishable after Algerian independence. For the French
government, they were “Algerians,” Berbers and Arabs with restricted access to the physical and
metaphorical interior of France’s new post-imperial geographic boundaries. Alluding to his
statements in May 1962 about “Muslims massacr[ing] each other,” de Gaulle one again reinforced
the divide between Algerians and “us” after the October 9, 1963 Council of Ministers meeting. This
time, in reference to the recent power struggle between Arabs and Kabyles, he declared to his
Minister of Information Alain Peyrefitte: “S’ils s’entre-tuent, ce n’est plus notre affaire. Nous en
sommes dé-bar-ras-sés, vous m’entendez? Les Arabes, les Kabyles, c’est une population
fondamentalement anarchique…”195 De Gaulle’s regarded all Algerians—Arabs or Kabyles, harkis or
FLN fellaghas—as “one population” (an uncontrollable one at that).
194
195
Robert Aldrich, Greater France: A History of French Overseas Expansion (London: MacMillan Press, 1996), 1.
Quoted in Peyrefitte, C’était de Gaulle, 1038. “Dé-bar-ras-sés” is italicized in the original text.
142
Those harkis who surmounted the three obstacles posed by the July 21 legislation—their
colonial status, repatriation, and the administrative procedure described above—and obtained
French nationality were often treated by government officials as second-class citizens. They were
“harki citizens” who would never be considered fully French owing to their ethnic differences and
association with Algeria. These “harki citizens” were marginalized in relation to the French
community because the government opted to tuck away those who needed housing assistance—
approximately half of the one hundred thousand harkis who migrated to France—in locales such as
those that are the subject of the next section: the Rivesaltes camp in French Catalonia and the
Fuveau forest hamlet near Aix-en-Provence. In these spaces, they lived only with other harkis. This
isolation impeded their inclusion into the surrounding community—that is, their social citizenship—
because it afforded them few opportunities to interact with other French citizens, except for the
camp directors who strictly oversaw the harkis’ everyday lives. The state agents employed governing
methods reminiscent of those in colonial Algeria, which treated “harki citizens” as inferior. The July
1, 1962 self-determination vote ended France’s governance of the Algerian territory and the
populations who remained there. De Gaulle’s France had “turned the corner” and ended French
Algeria, yet imperial legacies were immediately discernable in metropolitan France, perhaps nowhere
more visibly than the French government’s policies that marginalized its harki citizens. The next
section focuses on the daily lives of camp-dwelling harkis who remained under state supervision,
reminiscent of that during colonial Algeria.
143
SECTION 2
Harki Citizens’ Exile in France
La plupart de nos réfugiés musulmans [les harkis]
ont perdu tous leurs biens. Exilés de leur terre natale,
brinqueballés de droite et de gauche depuis plus de
dix-huit mois, souvent en deuil de certains de leurs
proches, traumatisés au physique comme au moral, ils
sont à la recherche d’une patrie, prêts à se donner à la
France.
Logement et travail, c’est ce qu’il leur faut avant
tout. Mais ils ont aussi besoin, pour retrouver le goût
de vivre et la force d’espérer, de se sentir pleinement
accueillis dans la cité française: sur le plan juridique,
c’est l’affaire des juges d’instance et du Ministère de la
Santé Publique et de la Population. Sur le plan moral,
l’administration n’y saurait suffire: il y faudra aussi
l’aide fraternelle et généreuse des Français de
souche…
Mais du point de vue sociologique, ils constituent
aussi pour nous, du fait de leur manque d’instruction
et de leur inadaptation aux conditions de la vie
moderne, une sorte de “minorité nationale”…1
- General Secretary Colonel Schoen, National
Committee for the Muslim French (CNMF),
January 21, 1964
As they sailed across the Mediterranean—which no longer served to geographically unify
“Greater France”—former harki soldier Mohamed K., his wife Ferhat, and their five children
arrived in an unwelcoming post-imperial France. Yet they could not return to Algeria without fear of
persecution. The family had been pushed from its mountainous region midway between Algiers and
Oran after witnessing FLN insurgents torture and execute other “pro-French” harkis, including two
of Mohamed’s brothers. Moreover, the war-ravaged economy had interfered with the patriarch’s
ability to provide basic necessities for his family, who often went hungry. They were pulled onto the
1
ACNMF, 15/13, Colonel Schoen, Report: “Reconnaissance de la nationalité française…,” 27.
144
unfamiliar shores of Marseille in July 1962, after a two-day boat ride from Algiers, buoyed by French
government officials’ promises of a safe haven from the increasing violence perpetrated by FLN
members who deemed Mohamed and his family traitors and “collaborators,” like the French citizens
who supported the Nazi enemy regime during the Second World War. With their Algerian
nationality in question, the K. family believed that by migrating to France they would obtain French
nationality and be treated like other French citizens. Nevertheless, as will be shown in this section,
their French citizenship would be unique.2
On the evening the K. family disembarked the cargo ship they had shared with other harkis
and livestock, Mohamed and Ferhat sought a place to live, a place to put down roots, and a
community of French citizens to join. Instead, in the words of the General Secretary of the CNMF
(a joint public-private rescue committee founded in 1962 to aid harkis in Algeria and in France),
harki families like this one were “brinqueballés de droite et de gauche” through a succession of
camps, where the government hid them from public view.3 The K. family moved from the BourgLastic transit camp in Auvergne, to the Rivesaltes camp in French Catalonia, to an abandoned village
at the foot of Mont-Lozère (Bagnols-les-Bains), before settling in December 1964 at the Fuveau
forest hamlet in Provence. Mohamed, Ferhat, and three of their nine children still live today in this
isolated development situated in a forest clearing.4
The K. family is but one of the nearly twenty thousand harki families who migrated to
France between 1962 and 1969. There is no universal “harki experience.” Nevertheless, each harki
who crossed the Mediterranean experienced an exile in France, an exile that was concurrently
Ferhat recounted her family’s story in an interview with the author on July 25, 2008 in Fuveau.
ACNMF, Colonel Schoen, Report: “Reconnaissance de la nationalité française…,” 27.
4 Mohamed’s personal file at National Office of Veterans and War Victims local branch in Marseille (hereafter
AONACBR) confirms the family’s places of residence, Mohamed’s service as a harki soldier from Oct. 1959 to July 1962,
and the date of their declaration to become French citizens. Payroll records at the National Forestry Office’s
departmental office in the Bouches-du-Rhône (hereafter AONFBR) certify that Mohamed began his employment as a
forestry worker in Fuveau on Dec. 10, 1964.
2
3
145
physical and cultural and that existed between them and France and between them and Algeria.5 I
emphasize exile from both nations because we must view the harki population within the geography
of the French colonial empire, even after it legally ceased to exist.
Harki citizens’ physical exile from Algeria arose out of their divide from the homeland they
were forced to escape and that from France resulted from their spatial isolation from the native
French population’s living spaces. Approximately 55,000 harkis and their family members resided, at
least briefly, in camps housing only other harki population members. The government offered three
practical reasons for their encampment. Foremost, the post-Second World War housing crisis,
compounded by the concomitant frenzied “exodus” of nearly one million pieds-noirs, meant that
there were few places to lodge the fleeing harkis.6 While the government did not want, or expect,
this massive migration of pieds-noirs either, it lodged these repatriates of European and Jewish
origins initially in requisitioned hotels, school dormitories, and emergency shelters and eventually in
newly-constructed HLM. Directives from local and national government officials consistently
affirmed a preference for using available housing for pieds-noirs over the harkis. For example, in
March 1963 the Prefect of the Pyrénées-Orientales department, home to the Rivesaltes camp, wrote:
“Parmi les difficultés qui freinent le recasement des harkis, la principale est, sans contestation
possible, celle du logement. L’arrivée dans le Département des Pyrénées-Orientales, de 6.200
familles rapatriées d’Algérie, a donné au problème de l’habitat une acuité toute particulière. Tout ce
qui pouvait constituer un abri, même précaire, a été aménagé et occupé et il reste peu de locaux
disponibles.”7 Government officials fervently maintained that the largely peasant harki population
By “physical exile” I mean the geographical or symbolic separation of a population and by “cultural exile” a separation
based on different customs, social behaviors, and physical characteristics of a population.
6 Moreover, during the previous decade the government resettled approximately 275,000 French repatriates from
Indochina, Egypt, Morocco, and Tunisia. Colette Dubois, “La nation et les Français d’outre-mer: Rapatriés ou sinistrés
de la décolonisation?” in L’Europe retrouvée. Les migrations de la décolonisation, eds. Jean-Louis Miège and Colette Dubois
(Paris: L’Harmattan, 1994): 91.
7 CAC, 19920149/4 (dossier: Travaux de la Commission départementale), Prefect of the Pyrénées-Orientales to the
Minister of Repatriates, “Objet: Reclassement des supplétifs musulmans réfugiés en France,” 2, Mar. 28, 1963.
5
146
had to be rehoused in rural settings, yet they found limited solutions. In a 2004 interview Pierre
Messmer, the Minister of Armies from 1960 to 1969 and Prime Minister from 1972 to 1974,
revealed to me his unwavering perception of the housing problem: “…on [ne] peut pas donner de
propriété rurale aux harkis parce qu’il y a des propriétaires. La France n’est pas un pays que l’on peut
coloniser.”8 At the same time, amidst the same housing crisis, the government found alternative
solutions for the pieds-noirs, thereby segregating the “French Muslims from Algeria” from the
“French from Algeria,” as it had during the colonial era.
Second, government officials viewed this “disoriented” population as unprepared to enter
directly into French society. (The French word dépaysé, with the root “country,” seems to more
accurately describe their situation.) In the same interview Messmer, whose Ministry was responsible
for transporting the harki population from Algeria and overseeing the camps, contended that the
harkis were “totalement incapables” of working the machines that the rural French men and women
as “très évolués,” in his opinion, used in farming techniques.9 The camps would therefore serve
transitional spaces to become “French citizens.” Here, instructors would teach harki population
members the French language, provide training in the necessary skills for France’s more
technologically advanced job sector, and assist them in grasping metropolitan “ways of life.” One
Ministry of Interior report explained the fundamental goal of the camps as allowing the harki
population to overcome its very superficial knowledge of “nos mœurs”—the same term used in
determining whether Algerian natives were worthy of French nationality during the imperial era (as
discussed in chapter one).10 Government discourse consistently was tainted with racial undertones.
Finally, officials believed that placing the harki population in concentrated spaces would be
the best means for military personnel, on the one hand, to protect it and, on the other, to control its
Messmer, interview.
Ibid.
10 AN, 5 AG 1/22, Minister of the Interior, “Rapport sur l’application de la loi du 26 décembre 1961…,” 104. On
French nationality requirements for Algerian natives, including exemplifying “bonne vie et mœurs,” see chap. 1.
8
9
147
potentially subversive actions. This military surveillance aimed to both curb reprisals by FLN
members in France against harkis because they had fought for France during the war and hinder the
infiltration of the camps by non-harki Algerians fleeing to the metropole (whom government
officials called “false harkis”). It also intended to inhibit the recruitment of harkis into pro-FLN
factions and limit their contacts with OAS members whom, as the previous chapter detailed, the
government and media outlets believed posed safety threats on metropolitan soil.
Beyond these practical reasons, two other unstated explanations emerge for physically exiling
harki families in camps. The first is the harkis’ inextricable link to the painful loss of the Algerian
War for Independence, which, as demonstrated in chapters one and two, the French government
sought to immediately forget. According to de Gaulle’s Minister of Information, Alain Peyrefitte, by
late August 1962 Algeria no longer was a weekly discussion topic at the Council of Ministers
meeting, though when the subject was evoked, “on avait toujours le cœur serré.”11 This ambivalent
relationship with a territory that France had incorporated into its system of departments for over
one hundred years led the government to push out of mind and sight reminders of the war. The
camps where the government chose to house harki families served this function.
The second explanation for putting harkis and their families in camps derives from their
ambiguous status when they arrived in France. Were they refugees or repatriates? Legally, the harkis
were “repatriates” and thus eligible for measures in the December 1961 Boulin law providing loans,
special access to HLM housing, and social assistance to all repatriates from former French colonies.
However, as Todd Shepard has demonstrated, the government rarely employed this term for the
harkis. Instead—and in contrast to the nomenclature government officials consistently used for the
pieds-noirs—the government almost exclusively used “refugees” to describe the harkis.12 To offer
Peyrefitte, C’était de Gaulle, 260.
For a detailed discussion of the harkis’ status after the Evian Accords, see Shepard, “Chapter 9: Rejecting the Muslims,”
in The Invention of Decolonization, 229-242.
11
12
148
just one example of this tendency, in 1963 the Conseil économique et social published Problèmes posés
par le rapatriement des réfugiés d’Algérie authored by professor of medicine Robert de Vernejoul. Despite
the word “refugiés” in its title, this term only appeared in the report for the pieds-noirs to describe
their arrival in the Bouches-du-Rhône department from Algeria13 as well as in the subject heading
“Le nombre de réfugiés,” a reference that included the harkis.14 For de Vernejoul, the pieds-noirs
were “rapatriés” and the harkis relegated to the status of “réfugiés.” Throughout the report he paid
careful attention to distinguish these two populations based on ethnic origin and status. For example,
in the section concerning the “number of refugees” he writes: “624.957 personnes de souche
européenne auraient quitté l’Algérie pour la France. Àce chiffre, il convient d’ajouter 27.000
musulmans réfugiés politiques.” 15 The government’s de facto categorization of the harkis as
“refugees”—even though they were never accorded legal refugee status—influenced what type of
housing the government provided for them both upon their arrival in France and, more importantly,
after their initial residence in the camps. An abundance of examples demonstrates that during the
twentieth century European host governments often housed refugees in camps. For instance, the
French government had encamped Armenian refugees who escaped the genocide during the early
1920s, Spanish refugees fleeing Francoist Spain in the late 1930s, Central Europeans fleeing the Nazi
threat in the early 1940s, and Indochinese refugees following the fall of Diên Biên Phu in 1954.16 By
classifying the harkis as refugees when they arrived in France, the government justified its choice to
house them in camps. At the same time, the harkis continued to be considered refugees beyond their
Conseil économique et social, Problèmes posés par le rapatriement des réfugiés d’Algérie, prepared by Robert de Vernejoul
(Paris: JORF, 1963), 176.
14 Ibid., 177.
15 Ibid.
16 On camps for Armenian refugees in France see Cyril Le Tallec, La communauté arménienne de France: 1920-1950 (Paris:
L’Harmattan, 2003). On camps for Spanish refugees in France see Michael R. Marrus, The Unwanted: European Refugees
from the First World War through the Cold War, 2nd ed. (Philadelphia, PA: Temple University Press, 2002), 189-93 and Schor,
Histoire de l’immigration en France, 142-50. For a detailed analysis of camps in France from 1938 to 1946, including those
for Spanish refugees and Jewish refugees from Central Europe, see Denis Peschanski, La France des camps. L’internement,
1938-1946 (Paris: Gallimard, 2002). On camps in France for Indochinese refugees see Van Thao, “Le retour des
rapatriés d’Indochine,” 29-38.
13
149
initial residence in France—and even after they became French citizens (12,597 acquired French
nationality while in the camps before August 1964).17 Living in these spaces linked to refugees
reinforced the government’s categorization of the harkis as such.
The harki population’s cultural exile ensued from its rejection by, and isolation from, the
Algerian community in France and from its categorization as “Arabs” by the French population.
This public opinion questioned whether the harkis were indeed French citizens. Aspects of the harki
population’s exile did allow individuals to retain—to varying degrees—their Algerian customs, which
mitigated their dépaysement in France. Nevertheless, the exile simultaneously excluded them from
their surrounding community (whether native French or Algerian), hindering their integration into
French society, and from Algeria, to which they could not safely return. This complex exile defines
the uniqueness of the harki population’s situation: no other population migrating to France from its
former colonies was placed into camps, unable to return home, and rejected by the French
population and their native community alike.18
The physical distance between the K. family, whose story is described above, and Algerian
soil constituted a painful exile that challenged its entire way of life. The couple recognized that
geographical separation from Algeria constituted the only way to protect their children’s and their
own lives. However, Mohamed had to leave behind his life as a farmer working the small parcel of
earth that he owned outside of Chlef, situated in the Ouarsenis Mountains between Algiers and
Oran.
CAC, 19920149/5/12, Chef du 13º Bureau bis de la Sous-Direction des Naturalisations (Ministère de la Santé
Publique et de la Population), “État numérique des déclarations enregistrées pendant la période du 1er juillet au 31 juillet
1964,” Aug. 10, 1964.
18 The only other colonial population that the French government placed into camps was the diverse group of refugees
from Indochina (Vietnamese, Franco-Vietnamese couples, and mixed race children, métis) who migrated to France
following the end of the Indochinese War in 1954. Upon their arrival, the government sent 12,000 of the 15,000
refugees in the first wave (35,000 total would arrive between 1954 and 1960) to centres de transit, including the same
Larzac camp where harki population members were sheltered in 1962. Van Thao, “Le retour des rapatriés d’Indochine,”
30-33. One of the camps, the Centre d’accueil des Français d’Indochine in Sainte-Livrade, still housed over half a
century later in 2010 one hundred Indochinese refugees—primarily elderly women, unemployed and handicapped
individuals. Dominique Rolland, “Indochine: les oubliés de Sainte-Livrade,” L’Histoire no. 356 (Sept. 2010): 77.
17
150
Mohamed and his family also needed to adapt to unfamiliar surroundings in France devoid
of family ties and the accustomed hierarchical structure of their society. In the Algerian bled, they had
lived on the Beni-Boudouane tribal territory. Colonial settlements had not reached this eighty by
fifty kilometer fief that the Bachaga Boualam administered.19 Their home was a cob hut (consisting
of a mixture of clay and straw) located among their fifteen thousand fellow tribesmen. In France
they first resided in canvas military tents, next in one of the four twenty square meter rooms in a
restored military barrack, then in an abandoned stone house. Finally, they settled permanently in a
37.5 square meter metal and wood prefabricated house grouped into clusters of four. Arabs, they
resided side-by-side with Kabyles, Chaouias, and other Arabs, who had been distanced from each
other by as much as one thousand kilometers in Algeria. Mohamed and his family were separated
from almost all of their one thousand fellow tribesmen who had also been repatriated. After their
passage through temporary camps, only two other Beni-Boudouane families would settle with them
at the Fuveau forest hamlet, thereby rendering impossible the recreation of their established social
life. The disruption of the hierarchical structure of the K. family’s social fabric was not unique,
however, as the government placed harki families into forest hamlets and public housing projects
constructed specifically for the population (known as cités urbaines) often with little regard for their
region of origin in Algeria. Arguably, this disruption could help foster harki families’ integration. But
the stated objective of the government’s policy for forest hamlets and cités urbaines was integrating
harki families into the French community, not into the harki community.
The K. family simultaneously sustained a physical exile from the neighboring French
community. Family members’ “déracinement” from Algeria was not accompanied by an
“enracinement” in France as during the first two and a half years in their adopted country they
resided in four camps located in sparsely-populated corners of southern France. This perpetual
19
Fabbiano, “Les harkis du Bachaga Boualam,” 114.
151
movement challenged their inclusion into the French community. In its efforts to immediately
consign the Algerian War for Independence to the past, the government, by pushing the harkis out
of public view in the camps, did not give them the opportunity to put down roots and familiarize
themselves with—or become a part of—their surroundings. Their provisional residences of tents
and military barracks at the Bourg-Lastic and Rivesaltes camps represented a rupture with the
surrounding architecture of rural farmhouses, which isolated them even further from the French
lifestyles. In addition to this literal, spatial distance, the harki population encountered a symbolic
separation from the French community when they arrived in France, which challenged whether it
belonged on French territory. The government impeded its members from being “pleinement
accueillis dans la cité française” because, as detailed in the previous section, it did not grant them
automatic French nationality. Mohamed and Ferhat, like approximately 7,500 other harki adults,
applied to become French citizens while living in the Rivesaltes camp.20
After spending eighteen months in an abandoned village with other harki families and a
camp director in the Lozère, the K. family settled on the outskirts of the 2,500-person village of
Fuveau.21 However, its situation did not improve as it continued to reside in a community entirely
comprised of harki families and to live under surveillance. Its frequent mobility reversed into
immobility and isolation at this microcosm of the Rivesaltes camp. The only native French people
ever to live side by side with the K. family were the military officers, social counselor, and camp
directors who watched over them.
The Fuveau forest hamlet—like the seventy-one other similar forest hamlets primarily
located in southern France—was imperceptible to town residents. This prefabricated development is
CAC, 19920149/5/12, Ministère de la Santé Publique et de la Population, Sous-direction des Naturalisations,
“Déclarations recognitives de la nationalité française souscrites au titre de l’article 2 de l’Ordonnance n° 62-825 du 21
juillet 1962 par d’ex-harkis hébergés dans des camps d’accueil,” Aug. 3, 1964. Mohamed and Ferhat made their
nationality declaration on Dec. 29, 1962; their request was not granted until May 1964. A copy of the certificate is
located in his personal file at the AONACBR.
21 K. family, interview.
20
152
located three kilometers south of the town center in a forest clearing off a two-lane highway, on land
previously used for coal mining. As the image in figure 11 shows, even to this day there are few
streets and houses near the harkis’ residences. While some forest hamlets were located less than one
kilometer from a town (such as Curcuron in the Vaucluse), others were situated as far away as
fourteen kilometers (Le Capelude in the Var).22 By isolating the harkis and obscuring them from
public view in camps, the government limited the possibility for any form of “aide fraternelle et
généreuse des Français de souche,” which Colonel Schoen argued was imperative to their integration
into French society. According to the social counselor in charge of the Roque d’Anthéron (Bouchesdu-Rhône) forest hamlet from 1969 until it closed in 1976 and the Fuveau town councilwoman who
served as a liaison to the harki population in the Cité Brogilum from 1989 to 1995, town residents
rarely if ever visited the forest hamlets.23 Despite the occasional “adaptation to metropolitan life
classes” from the on-site social counselors, the geography of the harki population’s camp living
arrangements did not facilitate its interactions with other French citizens to experience, or learn,
“French” ways of life.
ACNMF, 6, Servier, “Enquête sur les Musulmans Français,” Part 2, 38-39. In Les harkis, une mémoire enfouie, JeanJacques Jordi and Mohand Hamoumou write “aucun des soixante-quinze hameaux forestiers n’est implanté à proximité
du village métropolitain!” They proceed to cite Les Peyrouas as the forest hamlet closest to a town, at one and a half
kilometers from Saint-Maximin in the Var, and list the distance between the Fuveau forest hamlet and the town as six
kilometers. Jordi and Hamoumou, Les harkis, 97. These distances, ostensibly given to support their statement above, are
incorrect. My travels by car between the Fuveau town hall and the forest hamlet, on the same roads that existed in 1964,
showed that the distance is three kilometers. Moreover, Servier’s report, based on site visits, enumerates the distance
between seventeen of the thirty-six forest hamlets open in 1972 and their respective town centers. Nine of these are less
than 1.5 kilometers apart and he correctly lists the distance between Fuveau and the forest hamlet as three kilometers
(38-41).
23 Raymonde Guillet (monitrice sociale du hameau de la Baume), interview with the author, Dec. 18, 2007, la Roque
d’Anthéron and Marie-Josephe Haupt (adjointe spéciale aux harkis pour la commune de Fuveau), interview with the
author, Dec. 5, 2007, Fuveau.
22
153
Figure 11: Present-day
map of Fuveau. The forest
hamlet is represented by
the three rectangles
located near the bottom of
the photo. Source: Fuveau
village map, Boulevard
Loubet. Photograph by
the author in 2007.
This ignorance of “French customs,” which social counselors would teach them, was one of
the principal factors contributing to the K. family’s cultural exile from the French population. The
harki population encountered doubts about its “Frenchness” from its neighbors owing to what
Colonel Schoen described as “[son] inadaptation aux conditions de la vie moderne.” The
population’s cultural alterity and the fact that its members were, in the words of Colonel Schoen, a
“minorité nationale” justified secluding them to camps, which, in turn, challenged their belonging in
the French nation both upon their arrival, and even today.24 In Algeria, Mohamed and Ferhat’s life
rarely included electricity, cars, and other “modern” inventions. Their food primarily came from
their land, not grocery stores. When they arrived in France, their lack of knowledge about using
machines, speaking French, and customs for interacting with shopkeepers erected a barrier between
them and the rest of the French population that deemed their new neighbors’ ways of life inferior.
24
ACNMF, 15/13, Colonel Schoen, Report: “Reconnaissance de la nationalité française…,” 27.
154
Individual harki population members’ responses to their situation varied based on
numerous factors, including proximity to “French” communities, age, previous knowledge of
French language and French conventions, and personal nature. By and large, few adult harkis—
especially those in the camps—adopted metropolitan practices, and almost never inside their homes.
Women, in particular, were slow to adopt French language and customs, if they ever did. The former
social counselor who lived at the Roque d’Anthéron forest hamlet related during an interview with
the author that none of the approximately forty harki wives knew more than a few words of French,
nor had much interest in learning the language (her language classes were sparsely attended). She
conducted courses about metropolitan ways of cooking, cleaning, and childcare, yet during her
regular home visits she saw that few women used these practices.25 Indeed maintaining Algerian
customs served as a means to keep alive the link with—and memory of—their homeland and retain
their cultural identity. Further, it demonstrated that these ways of life remained deeply ingrained.
This preservation simultaneously represented a subtle form of revolt against the imposition of
French norms, which sometimes reached into the private sphere for harkis living under state control.
Some of the rules that the social counselors and camp directors enforced conflicted with Algerian
traditions. The photograph in figure 12 shows Ferhat K. dressed in classic Algerian garb baking the
traditional Algerian bread, kesra, over a fire in the wood shed behind her house at the Fuveau forest
hamlet in 2000 (almost a decade after the social counselor moved away from the development). This
ritual method of cooking over an open flame indoors was practiced in Algeria where her home was
not constructed with flammable materials.
Guillet, interview. Mrs. Guillet resided in Algeria from 1946 to 1962 and was a SAS during the Algerian War for
Independence.
25
155
Figure 12: Ferhat K. preparing
kesra. Photograph taken by and
used with the permission of her
daughter, Aïcha.
Harki population members faced an additional source of cultural exile from the French
community owing to their physical features since many of them—particularly those of Arab
descent—had a darker skin tone than that of their neighbors. While the majority of Algerians in
France distinguished between those who had served as harkis and those who had fought for the
FLN (two categories that were not mutually exclusive as pointed out in the previous chapter),
French natives most often saw all Algerians as “Arabs” (and “Berbers,” if aware that such a
distinction existed). Consequently, harkis such as Mohamed and Ferhat K., like other North African
immigrants on French soil, were often regarded with suspicion and became victims of racism owing
to their skin color, a “stigma” that was always perceptible. The “visibility” and perception of their
bodies’ difference shaped these non-white migrants’ social identity as the “Other” in their
156
communities.26 This posed a particular problem for the harki; whenever seen in public, he could not
lay claim to his “pro-French” link. He was viewed as “Algerian,” not French.
However, many Algerians believed Mohamed (like other “pro-French Muslims”) was a
traitor no longer worthy of Algerian nationality, even though he joined the French Army owing to
economic factors and tribal obligations, and not because he had an allegiance to France.27 Owing to
questions surrounding Mohamed’s October 1959 choice to engage as a harki soldier, the K. family
would also experience a lasting exile from other Algerians on both sides of the Mediterranean
because they were isolated from their native culture.28 Mohamed and Ferhat were indeed “en deuil”
for their family members—including an aunt, two uncles, and three siblings—whom FLN members
executed after the ceasefire. This history rendered difficult their assimilation into the same Algerian
community that believed they and their family members were traitors.
The killing of pro-independence Algerians and harkis at each other’s hands on Algerian soil
generated an enduring rift between the two groups, tensions that migrated with them across the
Mediterranean. In France, harkis like Mohamed still feared for their lives as members of the Amicale
des Algériens en France (ADAF), the metropolitan arm of the FLN, sought them out. The police
reported many incidents of ADAF members threatening—and even murdering—harkis, both inside
and outside of the camps, and Minister of Interior Frey even sent a circular in September 1962 to all
Department Prefects asking them to exercise a “surveillance vigilante” in areas where harkis settled,
The concepts of “stigma” and “visibility” are developed by Erving Goffman, Stigma: Notes on the Management of Spoiled
Identity, First Touchstone ed. (New York: Simon & Schuster, 1986 [1963]). For Goffman’s discussion of visibility see pp.
48-51. Sander Gilman also discusses the notion of “visibility” as it relates to differences, and perceptions of difference of
Jews’ bodies. See Sander Gilman, “Chapter 7: The Jewish Nose: Are Jews White? Or, the History of the Nose Job,” The
Jew’s Body (New York: Routledge, 1991), 169-93.
27 K. family, interview.
28 A Sept. 9, 1992 certificate of military service from the Bureau central d’archives administratives militaires in Pau
confirms Mohamed’s engagement as a harki soldier in the 2/10˚ RAMA harka from Oct. 1959 until July 1962. This
document is located in his personal dossier at the AONACBR.
26
157
given these reports of violence.29 The FLN in Algeria had ordered its metropolitan comrades to
make all Algerians in France, including the harkis, contribute money each month to rebuild their
homeland, which also served as a means to monitor their movements. To cite one example, on
February 6, 1963, several ADAF members visited the residence of nine harkis each of the first three
nights after they arrived from Rivesaltes to work in a factory in Reims. The ADAF members
questioned the harkis about their role in the French Army and why they had migrated to France, and
then issued the following threat: “Les dossiers des harkis sont à l’étude. Bientôt leurs photos nous
parviendront. Les jugements d’Algérie seront exécutés en France. Les traîtres seront égorgés.”
Therefore, the ADAF members (who at moments “[ont] feign[é] d’ignorer le passé des anciens
militaires”) tried to force the group of harkis to give thirty francs on the spot and pay subsequent
monthly dues of ten francs to join their organization. The harkis refused and noticed a few days later
that these men were closely watching over them; as a result, they decided to return to Rivesaltes to
escape the threat.30 Such events distanced the harkis from their native Algerian community.
Understanding the uniqueness of the harki population’s exile in France allows us to ascertain
the challenges it faced in its post-imperial French citizenship. The same factors that induced the
harki population’s exile in France simultaneously led to its fluctuating status, which contributed to its
exclusion from French society. That is, harki population members were an amalgam of repatriates,
refugees, and immigrants. Legally, they were repatriates like the pieds-noirs, even though this label
that colonial powers created during decolonization to identify this specific class of migrants was a
misnomer for all of the harkis and the vast majority of the pieds-noirs.31 They were rarely “returning”
to their home country. However, ethnic and cultural differences—their visible characteristics and
Archives Départementales du Bas-Rhin, 544 D 173, Minister of Interior to Department Prefects, Circulaire n˚ 545,
“Objet: Sécurité des anciens supplétifs musulmans,” SN/CAB N˚ 5444, Sept. 1, 1962.
30 ADBR, 138 W 6, SCINA, “Synthèse… journées des 16-18 février 1963,” 6-7, N˚ 1 847.
31 Andrea L. Smith, “Introduction,” in Europe’s Invisible Migrants: Consequences of the Colonists’ Return, ed. Smith (Amsterdam:
Amsterdam University Press, 2003), 19. According to Smith, it appears that several colonial powers began to use this
term independently of each other.
29
158
Algerian ways of life—confused their de jure status. The circumstances in which the harki
population left Algeria were in fact similar to those of the pieds-noirs (who experienced an
analogous physical exile from Algeria). Owing to the harki population’s alterity, however, the French
community most often viewed its members as Algerian immigrants. At the same time, government
officials and others, including Colonel Schoen, most often referred to the harkis as “réfugiés,” a
term inextricably linked to exile. This designation, however, is not correct for them because, among
other reasons, unlike refugees once the harki and their family members arrived in France they could
obtain French nationality through a pro forma procedure.32 Therefore, none of these three terms
alone—repatriate, immigrant, or refugee—can accurately describe the harki population’s status,
which is due in large part to its unique exilic situation. They were harki citizens.
This section analyzes government housing and integration policies for harki families and
their living conditions in two isolated locales supervised by French government representatives.
Exiled from Algerian, French, and pied-noir communities in France and geographically separated
from Algeria, they did not fully belong to French society. My inquiry goes beyond the local level to
investigate within the same analytical field both the national creation of state policies toward the
harkis and their local implementation. Chapter three focuses on the state’s use from 1962 to 1964 of
the Rivesaltes camp as a powerful physical and psychological tool to isolate, control, and re-educate
members of the harki population. Chapter four presents the rationale behind and development of
the government’s policy to rehouse harkis and their families in forest hamlets. Chapter five
investigates the daily lives of harki families living in the Fuveau forest hamlet, which opened in 1964
and remains open today. Chapter six places this case study within the national context by analyzing
To become French citizens, political refugees need to present a dossier to a special commission (Office français de
protection des réfugiés et apatrides), which judges whether they fall under the definition of refugee outlined in the 1951
Geneva Convention on the status of refugees. If so, they benefit from a special status that accords them all the same
rights as French citizens except for voting rights. This status is not the same as French nationality. Patrick Weil, La
France et ses étrangers: l’aventure d’une politique de l’immigration de 1938 à nos jours, new ed. (Paris: Seuil, 2004), 335-37.
32
159
new measures proposed by the government in the mid-1970s to address its admittedly failed
integration policies and the violent protests of harki children and supporters of the population. It
returns to the case study under examination in chapter five to evaluate the impact of these policy
changes on the Fuveau forest hamlet. These fours chapters explore harki citizens’ fluctuating
status—among refugees, repatriates, and immigrants—to demonstrate that despite the heterogeneity
of individual harki’s experiences in France, all endured a similarly unique exile. Government
practices and public perceptions challenged their legal status as French citizens.
160
CHAPTER 3
Refugees? The Harkis’ “Univers Concentrationnaire”
at the Rivesaltes Camp
Or les camps ne constituent pas une formule
d’avenir. Leur organisation fut-elle sans reproche, ils
restent marqués de la tare de tout “univers
concentrationnaires.” Non seulement le système
favorise le développement des épidémies…, mais il
contribue à la démoralisation des hommes [supplétifs],
qui y perdent leur raison d’être. En outre sur le plan
politique il offre une cible facile à l’opposition.1
- Minister of Armies Pierre Messmer,
March 2, 1963
Of the estimated 100,000 harkis who traversed the Mediterranean Sea, 20,000 experienced
internment in the Rivesaltes camp between September 1962 and December 1964. This space,
located on the outskirts of the Catalonian town of Rivesaltes, had previously housed refugees fleeing
over the Pyrenees from Francoist Spain and Jews prior to their deportation to extermination camps
during the Second World War. At its most densely populated, during the first week of December
1962, the Rivesaltes camp’s 8,885 harkis2 dwarfed the town’s 6,262 inhabitants3 from whom the
newly-arrived Algerians were separated by barbed wire. The French government’s encampment of
the harkis contributed to the complicated exile discussed in the introduction to this section: the
Rivesaltes camp physically and culturally separated them from their homeland and from their
adopted country, calling into question whether they belonged in or to either nation. Even after the
SHAT, 10T 549/3b, Minister of Armies Pierre Messmer to Minister of Rapatriates François Missoffe, “Objet:
Problème des ex-supplétifs,” 2, Mar. 2, 1963.
2 SHAT, 1R 337/1 (dossier: Camp de Rivesaltes et St. Maurice l’Ardoise). A graph in this carton shows the camp’s
weekly population statistics from Dec. 1, 1962 to Mar. 30, 1963.
3 Archives Communales de Rivesaltes (hereafter, ACR), “Evolution de la population depuis 1881.”
1
161
harkis surmounted the obstacles to becoming French citizens, the camp living space fostered an
exilic and isolated existence, which challenged their inclusion within French society.
In the epigraph to this chapter, Messmer—whose ministry oversaw the harkis’ daily life—
evoked a parallel between the harkis’ exile at the Rivesaltes camp and that of Nazi prisoners living in
concentration camps during the Second World War. For him, the Rivesaltes camp remained
“marqués de la tare de tout ‘univers concentrationnaires,’” a reference to the earliest firsthand
account written about Nazi camps, penned by French Resistant and Trotskyite David Rousset.4
Messmer’s comparison significantly positions the Rivesaltes camp into the Nazi concentration (and
extermination) camps’ “system” of excluding and isolating its inhabitants from society. 5 The
concentration camp universe, as defined by Rousset and to which Messmer alludes, was a system
linked to political and moral conceptions of racial superiority that imposed a forceful domination
over its internees. Rousset further contended that the camps were plagued by health epidemics,
unsanitary conditions, insufficient healthcare, promiscuity, and an elevated rate of mortality, which
demoralized their populations. Messmer’s choice of the polysemic term “tare”—which can be
defined as a grave defect, staining someone’s honor, diminishing someone’s or something’s value or
merit—perfectly captures this essence of concentration camps generally, and of the Rivesaltes camp.
The Rivesaltes camp did not include the starvation and the forced labor experienced by Nazi
prisoners and was not an extermination camp. However, the health epidemics, unsanitary conditions,
demoralization, and other characteristics listed by Rousset abounded in this space. The Rivesaltes
camp “concentrated” the harki population for several reasons. First, the French government viewed
them as unable to adapt to life in metropolitan France without first being given “une connaissance
David Rousset, L’univers concentrationnaire (Paris: Éditions du Pavois, 1946).
Rousset argues that “Entre ces camps de destruction et les camps ‘normaux’, il n’y a pas de différence de nature, mais
seulement de degré” (57).
4
5
162
plus approfondie de la langue française et de nos mœurs.”6 As the National Police claimed in an
internal report in October 1962, the re-education centers at the camp would “dégrossir les anciens
supplétifs,” in other words, to knock off the rough edges off these unrefined individuals and reeducate them.7 Second, the government sought to hide from sight any reminders of the painful loss
of French Algeria. Finally, the post-imperial government maintained these individuals’ inferior status
as “harkis”—an ethnic category that the colonial French government created for Muslim auxiliary
soldiers during the Algerian War for Independence—but it also added the element of social
exclusion as a key component in defining post-imperial harki citizens, which now included women
and children. The government treated them like refugees.
Yet the harkis were not legally refugees. Following mass migratory movements resulting
from World War Two, the 1951 United Nations Geneva Convention defined this term as a person
forced from his home country fearing persecution owing to race, nationality, membership in a social
group, or political beliefs.8 During the discussions leading up to the publication of this document, as
the European country with the greatest number of refugees in 1950, France actively advocated that
the denomination be restricted to forced migrants from Europe and those who were victims of
events taking place before January 1, 1951 (in opposition to Great Britain and the United States,
both of which pushed for a broader definition).9 While the French position prevailed on the latter
point, individual countries were permitted to choose whether to include only Europeans or those
AN, 5 AG 1/22, Minister of the Interior, “Rapport sur l’application…” 104.
ADBR, 138 W 5, SCINA, “Synthèse…, journées des 5, 6, 7 & 8 octobre 1962,” e-f, N˚ 1 765.
8 The Geneva Convention’s focus on refugees forced from their homeland for political and humanitarian reasons
represented a departure from the 1933 Bretton Woods system, concentrated on labor migration. Since 1951 the meaning
of refugee has continued to evolve to account for new types of forced migration, including those provoked by the end
of empires. While the meaning of refugee has changed, its legal definition has undergone only minor modifications.
Anthony M. Messina and Gallya Lahav, “Introduction: The Evolution of an International Refugee Regime,” in The
Migration Reader: Exploring Politics and Policies, ed. Messina and Lahav (Boulder, CO: Lynne Rienner Publishers, 2006), 199201.
9 Gérard Noiriel, Réfugiés et sans-papiers: La République face au droit d’asile (Paris: Hachette, 1998), 140-47.
6
7
163
from “Europe and elsewhere.”10 In 1961 when the High Commission for Refugees, the United
Nations body responsible for overseeing international refugee law, reevaluated its mission and
contemplated reorienting its focus toward Asian and African refugees, French officials restated their
opposition, fearing that this supranational commission would interfere with its colonial affairs.11 The
last major European power to decolonize, France received pressure from the Soviet Union and the
United States, which converged in their anti-colonial positions. French officials, however, wanted to
keep Algeria and Algerians an internal matter. Historian Matthew Connelly asserts that France’s
attempts to contain the Algerian War as an internal conflict actually caused it to have international
repercussions. He argues that the war “was distinctively diplomatic in nature, and… its most decisive
struggles occurred in the international arena.” 12 Attributing legal refugee status to the harkis (or the
pieds-noirs, for that matter) would have been a tacit admission that the Algerian War for
Independence was an international affair and allowed international organizations such as the High
Commission for Refugees to be involved in administering the harkis.
The government’s 1962 decision to relegate the fleeing harkis to camps isolated from the
general population was a powerful indicator of how government institutions and their
representatives wanted to interact with this population of Arabs and Berbers from that moment
forward. Faced with the crisis of the end of the Algerian War for Independence, the government
chose to view the harki population as a collective mass of foreigners and refugees whom it then
exiled and excluded from the national community, even after they became French citizens. While
maintaining the goal of rehousing the harkis as soon as possible, once it became apparent that this
task would not be easily accomplished government documents in fall 1962 and winter 1963
semantically shifted from “transit camp” to “housing camp” to describe the Rivesaltes camp (which
Ibid., 146.
Ibid., 150.
12 Connelley, A Diplomatic Revolution, 4-5.
10
11
164
did not close until December 31, 1964), thereby putting into question the temporary nature of their
living situation. The state often regarded the harkis with suspicion because of their Arabicity, that is,
their different customs and their perceived racial differences. The harkis could never escape
government and public categorization as “others” in their new homeland, which meant that they
could never fully belong within France. Their placement into the Rivesaltes camp symbolized that
they were outsiders, indeed de facto refugees, who required not only shelter, but also surveillance.
The encampment and the duration of the harkis’ stay were not the only problems posed by
the Rivesaltes camp. The harkis’ lived reality in this closed space, particularly in regard to the flow of
power between the population and the state, limited possibilities for the harkis to interact with other
metropolitan citizens and demonstrated continuities with colonial practices. Power in the Rivesaltes
camp flowed from the state representatives responsible for the harkis’ day-to-day encadrement, from
French government institutions, and from national-level politicians responsible for creating the
camp’s policies down to the harkis. Encadrement, the term the government employed to describe the
meticulous control over the harkis by state agents, left little space for the harkis to appropriate this
power and escape their exile.13 These officials oversaw all aspects of the harkis’ lives and taught
classes about job skills, French language and customs, and hygiene to prepare them for their
reclassement, a complex term signifying their integration into French society through access to housing
and employment.14
In multiple, and sometimes subtle, ways the Rivesaltes camp acted as a physical and
psychological tool for the state to control and educate its “undesirable” harki population. The
Rivesaltes camp’s architecture, insalubrious living conditions, isolation from other French citizens,
military oversight, and “reeducation” classes, beyond functioning as powerful symbols, reinforced—
Encadrement is most often translated into English as “surveillance,” “regimentation,” or “supervision.” These
translations do not fully capture the great extent to which army and civilian officials used military-style tactics to closely
supervise the harkis’ lives; therefore, encadrement and its verbal form encadrer are left in their original French.
14 Reclassement does not have a succinct English translation; for this reason, it is left in French throughout the text.
13
165
and contributed to—the government’s treatment of the harkis as unwanted residents of France.
Over the twenty-seven months that it functioned as a camp for harkis, it fostered an exilic existence
for the harkis and socially excluded them from French society. The organization and functioning of
the Rivesaltes camp set a precedent for exclusionary policies that would remain in place for harki
citizens in other locales long after it closed in December 1964.
A Camp for “Foreigners”
The Rivesaltes camp, also known as the Joffre camp, is located on a vast plot measuring four
by two kilometers straddling the Rivesaltes and Salses municipalities in rural Catalonia. The camp
takes its name from one of Rivesaltes’ most famous native sons, Maréchal Joseph Joffre, World War
One general and member of the Académie Française, born in a house that once stood on the land.
Its southernmost border sits five kilometers north of the Rivesaltes town hall, thirteen kilometers
north of downtown Perpignan, and thirty-five kilometers north of the Franco-Spanish border.
Fifteen kilometers eastward lies the Mediterranean Sea, the passageway to France’s former colony of
Algeria (see figure 13). Vineyards with Muscat grapes, the local specialty, skirt the camp’s periphery.
Today, all that remain of the camp are the crumbling cement and wood shells of eighty-square-meter
barracks, hollow outhouses with Turkish-style toilets, remnants of a flag pole where its internees
once stood at daily attention, occasional pieces of barbed wire intertwined with weeds, and a
dilapidated sign bearing “C mp J ffre de Rivesaltes.”
166
Figure 13: Geographical location of
Rivesaltes. Source:
http://www.cg66.fr/culture/memo
rial/renseignement.htm
The harkis were not the first, nor the last, population to inhabit the Joffre camp. However,
the government’s placement of harkis in the Rivesaltes camp signaled an important departure from
the legal categories of individuals the camp had previously sheltered. Three months after the Third
Republic government opened the first of the two hundred internment camps for “undesirable
foreigners,” refugees, and prisoners that would dot the French countryside during “the Black Years,”
it inaugurated Rivesaltes in May 1939.15 As a camp, by definition this space would separate and
distance its inhabitants from the rest of society. 16 Divided into military and civilian sections,
Rivesaltes was unique for the diversity of foreign populations that it would house. The camp’s initial
mission was to lodge colonial troops awaiting assignment during the war, yet its principal vocation
quickly shifted as the events of the war caused an international refugee crisis.17 General Franco’s
1936 coup d’état in neighboring Spain forced over half a million people to escape over the Pyrenees.
From 1940 to 1942 the military section of the Rivesaltes camp was primarily used as a training
The French government opened the first camp of this period, Rieucros (Lozère), in February 1939. Denis Peschanski,
La France des camps. L’internement, 1938-1946 (Paris: Gallimard, 2002), 15. On the French government’s policies and
rationale for opening camps at this time, see pp. 28-35.
16 For an analysis of the function, definition, and sociology of camps for foreigners, see Marc Bernadot, Camps d’étrangers
(Bellecombe-en-Bauges, France: Éditions du Croquant, 2008).
17 Peschanski, La France des camps, 111.
15
167
facility for groups of refugees (Compagnies de Travailleurs Etrangers) who would fill jobs vacated
by metropolitan workers enlisted in the French Army.18
The Rivesaltes camp’s vocation, however, very shortly shifted to housing prisoner and
refugee populations. One month after being broken into civil and military sections, on January 14,
1941, the “Centre d’Hébergement de Rivesaltes” run by the Ministries of Interior and Labor was
inaugurated on six square kilometers and would remain open until November 25, 1942. 19 The
change in the camp’s purpose followed the opening of ninety-three camps between spring 1939 and
May 1940 20 to absorb the massive influx of Spanish republican refugees—450,000 crossed the
Pyrenees in February 1939 alone21—and foreigners, notably Jews escaping the Nazi regime, fleeing
central Europe on the eve of the Second World War. On May 31, 1941 Spaniards accounted for
more than one half of the 6,475 internees; however, this group was rounded out by fifteen other
nationalities, the majority of whom were foreign Jews fleeing from Eastern Europe.22 These Jews,
whom the government classified as “refugees” and not as political prisoners, comprised
approximately one third of the camp’s population between May 1941 and June 1942.23 This changed
when a new portion of the camp, baptized “Le Centre de Rassemblement des Israélites de
Rivesaltes,” opened in August 1942 as a way station for Jews awaiting deportation. Historian Serge
Klarsfeld attributed to the camp the title “Drancy de la zone libre,” a weighty descriptor given that
A March 1939 letter from the Ministries of Interior and Labor stated the reason for creating the groups was to
“transformer cette masse inorganisée et passive que constituent ces réfugiés en éléments utiles à la collectivité nationale.”
This policy was officialized the next month by an April 12, 1939 decree, which required all foreign males between the
ages of twenty and forty-eight living on French soil whom the state deemed as refugees or stateless persons to serve on
work details equal in length to French citizens’ national service. Joël Mettay, L’Archipel du mépris: Histoire du camp de
Rivesaltes de 1939 à nos jours (Canet: Éditions Trabucaire, 2001), 19.
19 Archives départementales des Pyrénées-Orientales (hereafter ADPO), 1419 W 109, Service départemental des
Renseignements Généraux, Rapport n˚ 107 de la Sûreté Nationale, “Historique du camp de Rivesaltes,” Jan. 8, 1958.
20 For a map of “les camps de rassemblement et d’internement français pour étrangers” that opened during this period,
see Anne Grynberg, Les Camps de la honte. Les internés juifs des camps français, 1939-1944 (Paris: La Découverte, 1991), 8.
21 Nicolas Lebourg, “Histoire générale du Camp de Rivesaltes” (Journées du Patrimoine, Camp de Rivesaltes, Sept. 15,
2007), 1.
22 Ibid., 2.
23 Anne Boitel, Le camp de Rivesaltes, 1941-1942: Du centre d’hébergement au “Drancy de la zone libre” (Perpignan: Presses
Universitaires de Perpignan/Mare Nostrum, 2001), 219.
18
168
Drancy was the major transit point in France for the Auschwitz extermination camp. Of the 5,174
foreign Jews interned at the Rivesaltes camp between August and November 1942, 2,313 were sent
to the extermination camp Auschwitz via Drancy.24
During this time, Rivesaltes housed its foreign refugees and prisoners in 150 military-style
barracks, some without roofs, which divided the men from the women and children. Internees lived
in deplorable conditions including their only water supply being tainted by feces, no heat in the
barracks, and showers available only once every two weeks. 25 A web of Ministry of Interior
employees—including local civil officials, military officers, healthcare workers, police inspectors, and
schoolteachers—subjected prisoners to a strict discipline. These state representatives acted in
concert to oversee and control the day-to-day life of the prisoners.26
Following the 1942 German Occupation of the Free Zone, the Rivesaltes camp once again
changed vocations. At the same time, the camp continued its characteristic “Tower of Babel”
hodgepodge of nationalities and linguistic groups and its overlapping mini-camps within the camp,
which were structured by the sixteen housing blocks. Gypsy refugees fleeing from Eastern Europe,
German soldiers, and German and Italian prisoners of war comprised the majority of population in
the camp’s military portion. The German soldiers left on August 19, 1944, days after the landing of
Allied forces on the shores of Provence. With this evacuation, the military section of the camp
returned to its original vocation of a training facility, though few troops entered it over the next ten
years. As for the camp’s civil section, from September 1944 to December 1945 the Centre de séjour
surveillé de Rivesaltes detained a total of three thousand presumed Vichy collaborators as part of the
post-Liberation Purge.27 Still residing in this section of the camp at this time were German and
Ibid., 20. In existence from 1941 to 1944, the Drancy internment camp, located in the northeastern suburbs of Paris,
was France’s principal deportation camp.
25 Lebourg, “Histoire générale du Camp de Rivesaltes,” 2.
26 Peschanski, La France des camps, 19.
27 ADPO, 1419 W 109, “Historique du camp de Rivesaltes.”
24
169
Italian prisoners, who worked during the day outside of the camp in an effort to rebuild areas
destroyed by the war, and the Madagascan colonial soldiers who oversaw them.28
The civil section of the camp closed on May 1, 1948 and lay vacant until the Centre de
formation professionnelle accélérée pour les nords-africains opened on December 1, 1951 to
provide job training for North African colonial subjects living in the metropole. With racial tensions
flaring at the onset of the Algerian War for Independence, the center was reorganized in December
1954 to balance the North Africans with fifty percent “metropolitan” students.29 In January 1958, on
the initiative of the Minister of Algeria, the center further expanded its vocation to provide sixmonth job training courses for French Muslim military personnel returning to civilian life in France.
The Centre militaire de formation professionnelle (CMFP) N˚1 trained French Muslim and French
students in one of the following specialties: tiling, bricklaying, pouring concrete, plaster works,
rubble work, and surface coating.30 The Center’s motto, “Bâtis notre pays,” alluded to the vocation
of French Muslim migrant laborers during France’s Trente Glorieuses (a period of economic growth
and prosperity from 1944 to 1974). An image from a pamphlet issued by the Ministry of Labor
(figure 14) shows a uniformed soldier arriving at the enclosed center just north of the Pyrenees
Mountains prominently displaying a French flag and a graduate of the program in civilian clothes
with his diploma in hand ready to enter the metropolitan French job force.
Lebourg, “Histoire générale du Camp de Rivesaltes,” 5.
ADPO, 1419 W 109, “Historique du camp de Rivesaltes.”
30 For a pamphlet describing the center, see CAC, 19970391/3, “Centre militaire de formation professionnelle n˚ 1
Rivesaltes.”
28
29
170
Figure 14: Image of the CMFP
N˚ 1 from a pamphlet issued by
the Ministry of Labor.
CAC, 19970391/3.
During the Algerian War for Independence, the military portion of the camp served as a
stopping point for officers awaiting deployment to Algeria. In 1957, the state also envisaged opening
an internment camp here similar to those that had sprung up during the war in both the colony and
the metropole since the beginning of the war.31 However, the prefect of the Pyrénées-Orientales
fought against and succeeded in quashing this proposal. Talks of a new project to intern war
prisoners arose in late 1961 and thus was created a “centre pénitentiaire;” however, it would only
stay open for a little more than one month, from March 9, 1962 to April 18, 1962. Of the 527
incarcerated prisoners, 487 were “French Muslims,” 12 percent of whom were sentenced to death
for “anti-national activities” that threatened the security of the French state, that is, activities in favor
of Algerian independence.32 This prisoner population is particularly significant when contrasted to
the harkis who would arrive just five months later at the Rivesaltes camp owing to its relationship to
For a description of internment camps during the Algerian War for Independence, see Sylvie Thénault, “Personnel et
internés dans les camps français de la guerre d’Algérie. Entre stéréotypes coloniaux et combat pour l’indépendance,”
Politix 24, no. 69 (2004): 63-81.
32 Lebourg, “Histoire générale du Camp de Rivesaltes,” 6.
31
171
Algeria: “French Muslims” incarcerated for pro-Algerian independence activities instead of “proFrench Muslims” who aided French war efforts against an independent Algeria.
After the war ended, the Rivesaltes camp housed troops still in service who were awaiting
reassignment including the First Regiment of Tirailleurs Algériens who were repatriated from Algeria.
The 1,700 soldier regiment arrived at the end of May and in early June 1962; they were accompanied
by the families of 125 of these men.33 When the harkis arrived at the camp in September 1962, they
shared the grounds with a small number of these reserve troops and their families.
The Rivesaltes camp, therefore, came to be known for housing, on one hand, individuals
who were foreigners (étrangers)—a legal status defined as a person living in France who does not
have French nationality—and, on the other, French citizens whom the government removed from
society. The Joffre camp’s history reveals how France interacted with those it viewed as foreigners
during what historian Denis Peschanski labels “the century of camps.” 34 The first camps for
foreigners opened at the turn of the twentieth century and the peak of this phenomenon was during
the années noires of the Vichy regime when two hundred camps interned six-hundred thousand
people.35 According to sociologist Marc Bernadot, this new spatial relationship between France and
its foreigners had the principal objective of: “contenir ou… retarder l’entrée et la dispersion de
certains groupes d’individus dans les sociétés dites ouvertes.”36 With the exception of Jews who were
marked for deportation between August and November 1942, Bernadot’s statement well describes
the non-military populations in the Rivesaltes camp. As Bernadot reminds us, by creating this
physical barrier between from the general French population, the camp served to strip individuals of
their liberties, often in the name of preserving the pubic order. 37 The Rivesaltes camp was
ADPO, 1419 W 109, Bulletin des Renseignements Généraux du 22 mai 1962, “Implantation à Rivesaltes d’un
Régiment de Tirailleurs Algériens.”
34 Peschanski, La France des camps, 17.
35 Ibid., 15.
36 Bernadot, Camps d’étrangers, 13.
37 Ibid., 43.
33
172
exceptional in how long it remained open, spanning three Republics and the Vichy regime. It
continually housed transient populations who most often remained for less than one year and
sometimes for as little as a few weeks.
Examining the camp’s previous populations places the experiences of harkis at the Rivesaltes
camp within a larger context of marginalizing foreigners from French society, thereby providing a
clear indication of how France regarded the harkis—as foreigners and not as the citizens that they
were. In contrast to the prisoners and many of the refugees who were not free to leave the
Rivesaltes camp, the harkis were permitted to come and go while they resided there. But this
distinction between open and closed camps becomes less significant when considering the
continuities in the physical space, which maintained its function of separating of the camp’s
inhabitants from French society.38 As the rest of this chapter demonstrates, the Rivesaltes camp that
harkis knew had continuities with its previous iterations. First, the camp was transitory; the very use
of camp de transit and centre d’hébergement throughout its history underscores its ephemeral nature.
Second, it provided shelter in poorly-constructed buildings which dominated the populations they
housed and contained unsanitary living conditions. Third, the camp was almost exclusively used for
those the state deemed as “foreigners,” many of whom were labeled as “refugees.” Finally, all of the
camp’s populations were subject to strict military surveillance.
Tents and Barracks
On September 15, 1962, the Rivesaltes camp received eight hundred harkis arriving directly
from the Tefeschoun refugee camp located outside of Algiers.39 Over the next nine days, the entire
harki population from the Bourg-Lastic transit camp, which had numbered 5,083 people on August
On the similarities and differences between open camps and closed camps, see Caroline Intrand and Pierre-Arnaud
Perrouty, “La diversité des camps d’étrangers en Europe: présentation de la carte des camps de Migreurop,” Cultures &
Conflits no. 57 (2005): 73-78.
39 ADPO, 1419 W 109, “Arrivées d’Algérie au Camp de Rivesaltes.”
38
173
30, was sent by train to Rivesaltes and grouped into villages of tents on the camp’s vast plain (their
arrival at the Rivesaltes train station is depicted in figure 15). 40 Military officers responsible for the
encadrement of the harkis accompanied the disoriented population on their five-hundred kilometer
journey to Rivesaltes. Here, the soldiers continued to encadrer the same families they had supervised
at Bourg-Lastic. Other officers transferred to the Rivesaltes camp from Algeria—who were,
therefore, familiar with the “Muslim milieu”—met the arrivals at the train station and shuttled them
by military convoy the five kilometers from the center of town to their new home.41 Similar to
previous refugee and prisoner populations in this isolated corner of the commune, the harkis had
little or no contact with the town’s other residents. By October 3, with harkis from the Larzac transit
camp who had been transferred to Rivesaltes, the population swelled to 7,700 harkis and a small
number of tirailleurs algériens still stationed at the camp.42
SHAT, 1R 336/8 (dossier: Camps–Rivesaltes et St. Maurice l’Ardoise), Ministre des Armées Messmer (signed by
Général Lagarde) au Gouverneur Militaire de Lyon, au Général Commandant la 5ème Région Militaire (Toulouse) et au
Général Commandant la 9ème Région Militaire (Marseille), “Transfert des ex supplétifs du Camp de Bourg-Lastic au
Camp de Rivesaltes – Aménagement du Camp de l’Ardoise,” 2659/EMIT/1.O and 8621/EMAT/4.P.O, Sept. 11, 1962.
41 “Connaissances des ‘milieux musulmans’” is a phrase continually repeated in documents to specify the desired
credentials for agents to oversee the harkis. For example, a Dec. 19, 1962 letter from the Minister of Repatriates to all
seventy-five departmental prefects, the director of the Rivesaltes camp, and the Minister of Armies introduces the role of
the departmental inspectors of the chantiers de forestage housing the harkis with the following clause: “Ces
fonctionnaires choisis pour leur connaissance approfondie des milieux musulmans auront essentiellement pour tâche…”
CAC, 19920149/3 (dossier 1: Chantiers de forestage, octobre 62 – juin 65), 2, N˚ 2175 SFIM/MG.
42 ADBR, 138 W 5, SCINA, “Synthèse..., journées des 5, 6, 7 & 8 octobre 1962,” e, N˚ 1 765.
40
174
Figure 15: Harkis arriving in
Rivesaltes from Bourg-Lastic on
September 16, 1962. Source:
Agence France-Presse Archives.
On September 18, 1962 Minister of Armies Messmer issued a ten-page memo to the
Regional Military commander for southwestern France, which enumerated “general rules” for the
Rivesaltes camp’s structures, discipline, and cleanliness.43 Among the provisions, Messmer urged that
each tent house only one family, collective toilets and sinks be installed, lighting be provided inside
the tents, pipes be laid to vacate wastewater, and efforts be made to immediately reconstruct three
blocks of barracks, most recently used to house prisoners during the Second World War.44 Yet these
orders were not heeded.
SHAT, 7T 253/3, Minister of Armies to the General Commanding the 9th Military Region, “Objet: Camps
d’hébergement d’anciens supplétifs,” 7, Sept. 18, 1962.
44 Ibid., 3 and 5.
43
175
Below are accounts from a representative of the Ministry of Repatriates who visited the
camp on a rainy October 15, retired General Secretary of the Ministry of Armies General Jean Olié
who toured Rivesaltes one month later, and former Vice-President of the Algerian National
Assembly Bachaga Boualam who visited his comrades just before Christmas 1962. The Repatriates
official observed:
… Tous les harkis… sont sous des tentes, le camp est boueux, l’eau coule sous les tentes.
Une certaine amélioration de l’installation sanitaire a été effectuée mais les douches ne
marchent pas encore faute de personnels compétents pour les faire fonctionner…
Il ne semble pas que des ordres aient été donnés pour faire commencer les travaux de
restauration de baraques réservées aux harkis…
Il est impossible que cette situation soit maintenue plus longtemps…
Il importe que les travaux de restauration soient entrepris immédiatement: les questions
de devis ne pouvant huaminement [sic] retarder l’installation des familles dans des bâtiments
durs…45
General Olié reported at a November 19 meeting of the CNMF:
En résumé, Rivesaltes n’aurait jamais dû exister. Cette situation est indigne de la France;
pitoyable: des êtres sont malheureux de notre fait. Malgré les efforts déjà accomplis, ils
vivent dans des conditions matérielles navrantes, aggravées par l’oisiveté…
Il faut loger ces malheureux qui sont actuellement sous des tentes sans chauffage ni
électricité, dans des bâtiments en dur, éclairés et chauffés.
Hygiène: les installations sanitaires sont très insuffisantes. Il faut des W.C., des douches,
des salles de petits soins, un personnel médical plus nombreux (il y a une naissance par
jour)…
Les installations du camp s’améliorent… [mais] il reste néanmoins beaucoup à faire pour
résorber un lourd handicap logistique et pour transformer totalement la vie du camp.46
Finally, the Bachaga Boualam, a Muslim notable who fought for France during the Algerian War for
Independence, published the following account:
Mille tentes de campagne où se retrouvaient à chaque crépuscule, serrées les unes contre
les autres pour se préserver du froid…
ÀRivesaltes, 2 500 harkis, des maires, des notables vivaient et dormaient sous des tentes
déchiquetées par le vent avec ceux qu’ils ont pu sauver de leur famille, femmes, vieillards,
enfants. 12 000 pauvres bougres entassés par dix, par douze, sous une tente, sans feu, sans
CAC, 19920149/1 (dossier 6: Comptes rendus des tournées, décembre 1962-septembre 1965), Service d’accueil et de
reclassement des Français d’Indochine et des Français musulmans (SFIM), “Compte rendu de la tournée au camp de
Rivesaltes,” 1-2, Oct. 15, 1962.
46 André Wormser, Pour l’honneur des harkis, 1 an de combats, 45 années de lutte (Marseille: Éditionss Sillages, 2009), 12. The
words “indigne” and “pitoyable” were underlined in the original text, a ten-page summary of a Nov. 19, 1962 meeting of
the CNMF.
45
176
lumières. 5 000 gosses jouant avec des cailloux qui, le long des tentes, écorchaient leurs petits
pieds nus. Oui, ils étaient là grelottants, ne comprenant rien à ce qui leur arrivait, accrochés à
leur mère, en robe d’été…47
These firsthand testimonies differ in tone, which is necessarily linked to their authors. The first was
a representative of the government that was responsible for the placing the harkis into such abysmal
living conditions. The second was a member of an association created to aid the harkis and who had
served in Algeria alongside of these men. And finally, the emblematic “pro-French Muslim” was
trying to attract public attention to the plight of his brethren through the publication from 1962 to
1964 of three books, which decried the camps where the government housed harkis. Nevertheless,
all of the excerpts provide a snapshot of unhygienic, cold, dark, overcrowded, and indignant living
space, which they all agreed must be improved immediately.
The government was not oblivious to these defects. To address the harkis’ poor living
conditions, a group of local and national government and military officials who met on October 24,
1962 in Perpignan laid out a program for a joint effort to reconstruct by January 15, 1963 three
blocks of crumbling barracks comprising 1,400 housing units. 48 The officials at the meeting
underscored their desire to be able to use these reconstructed barracks for future populations which
might later inhabit the camp, thereby emphasizing what they hoped would be a transitory experience
by the camp’s harkis.49 However, in addition to a steady stream of new arrivals from Algeria, the
Service d’Accueil et de Reclassement des Français d’Indochine et des Français Musulmans (SFIM)—
the arm of the Ministry of Repatriates responsible for finding the harkis jobs and housing—
encountered greater than anticipated difficulties in the successful reclassement of the harkis, which is
detailed later in this chapter. As the population swelled from October to December 1962, which is
Boualam, Les harkis au service, 268.
SHAT, 1K 744 (dossier: Accueil à Saint-Maurice-l’Ardoise, Rivesaltes, et Bias), Bureau Spécialisé de la Défense
Nationale de la Préfecture des Pyrénées-Orientales, “Centre d’hébergement des supplétifs musulmans de Rivesaltes.
Séance plénière de travail tenue à la Préfecture des Pyrénées-Orientales le 24 octobre 1962 à 10 H,” 2.
49 Ibid., 4.
47
48
177
depicted in figure 16, these officials were forced to augment the initial number of barracks. In the
end, 1,554 twenty to twenty-five square meter housing units (cellules) were reconstructed between
November 1962 and February 1963.50
Rivesaltes Camp Population,
October 1962 - February 1964
10,000
9,000
8,000
Population
7,000
6,000
5,000
4,000
3,000
2,000
1,000
12
/2
10
10
/3
/6
2
9/
62
/8
12 /6
/3 2
1/
2/ 62
2/
6
3/ 3
2/
6
4/ 3
6/
5/ 63
12
/
6/ 63
8/
6
7/ 3
6/
8/ 63
3/
6
9/ 3
7/
10 63
/5
/
11 63
/2
11 /6
/3 3
0
12 /6
/2 3
8/
2/ 63
1/
64
0
Figure 16: Monthly population statistics of the harkis residing
51
at the Rivesaltes camp from October 3, 1962 to February 1, 1964.
However, torrential rain and floods in early November and initial difficulties finding masons
to construct the barracks pushed back the project’s start by two weeks. To compensate for the latter,
officials at the October 24 meeting decided to employ harkis in the renovation projects. This action
would also address concerns SFIM Director Yves Pérony had about “l’inactivité prolongée [qui]
SHAT, 13T 286, “Rapport du Colonel CASSAN Directeur des Travaux du Génie, relatif au Camp de Rivesaltes
Hébergement des ex-supplétifs. Bilan des dépenses. Demande de crédits supplémentaires,” 2, N˚ 4018/Te, May 13, 1963.
51 The statistics from Dec. 8, 1962 to Dec. 28, 1963 were gathered from the compte-rendus hebdomadaires in SHAT 1R,
336/8 and SHAT 1R 337/1. The Oct. 3, 1962 figure is from ADBR, 138 W 5, SCINA, “Synthèse..., journées des 5, 6, 7
& 8 octobre 1962,” e, N˚ 1 765. And the Feb. 1, 1964 figure is from ACNMF, 3/2, “Situation des Rapatriés Musulmans
au 1er février 1964.”
50
178
risque de dégrader le moral de ces personnes,”52 concerns that were later echoed many times by
Messmer when he referred to Rivesaltes as a “univers concentrationnaire.” The state decided,
however, that it would remunerate them at a rate inferior to that of local workers. Owing to their
lower daily pay, which two weeks later was set at six new francs (the equivalent of approximately
eight euros today when adjusted for inflation), 53 Pérony cautioned: “Il faut, cependant, faire
attention aux contacts avec les ouvriers qui travailleront sur les chantiers. Ceux-ci, auront un salaire
normal. La différence entre les deux systèmes pourrait être mal interprétée.”54 Separating the harkis
from metropolitan workers on the campgrounds further inhibited their contacts with and implied
their inferiority to other French citizens.
On November 21, 1962, Prime Minister Georges Pompidou penned a letter to Messmer
underscoring the persistence of problems. He asked about the specific measures that Messmer’s
ministry would pursue to ameliorate the harkis’ “conditions de vie… très défectueuses,” which were
aggravated by a particularly cold, rainy spell and a meager supply of warm clothing.55 At this time, all
of the harkis—unused to the harsher French climate—still lived in tents without running water and
electricity. Unsupervised harki men even resorted to scavenging wood from the adjacent vineyards,
which they used to build fires to keep their families warm.56 Furthermore, there was no means to
dispose of waste from the latrines, which led to problems with mosquitoes.57 This disorganization
SHAT, 1K 744, Bureau Spécialisé de la Défense Nationale de la Préfecture des Pyrénées-Orientales, “Centre
d’hébergement des supplétifs musulmans de Rivesaltes,” 3.
53 Letter N˚ 1738/SRIM/MG, Nov. 7, 1962 from the Minister of Repatriates, which provides this figure, is cited in:
SHAT, 14T 92, 4e Bureau de l’État-Major de l’Armée (Terre), “Fiche relative à la réunion au Cabinet du Premier Ministre
le 20/12/62 au sujet des Rapatriés (Camp de Rivesaltes),” 2, N˚ 309 EMAT/4/PFP, Dec. 21, 1962. Six new francs in
1962 was converted to eight euros in 2010 using a calculator on the website <http://inflation.free.fr/calcul.php>.
54 SHAT, 1K 744, Bureau Spécialisé de la Défense Nationale de la Préfecture des Pyrénées-Orientales, “Centre
d’hébergement des supplétifs musulmans de Rivesaltes,” 5.
55 SHAT, 1K 744 (dossier: Accueil à St. Maurice l’Ardoise, Rivesaltes et Bias), Prime Minister Pompidou to Minister of
Armies Messmer, Nov. 21, 1962.
56 Ibid.
57 SHAT, 1K 744, Bureau Spécialisé de la Défense Nationale de la Préfecture des Pyrénées-Orientales, “Centre
d’hébergement des supplétifs musulmans de Rivesaltes,” 6-7.
52
179
existed on multiple levels and did not end in January 1963 once all of the harkis were moved to
hastily renovated buildings two to three kilometers from the tents that once sheltered them.
Although the barracks that housed previous prisoner and refugee populations at Rivesaltes
were dilapidated and filth-ridden, historians who have written about these internees disclose that
they that lived in permanent structures (bâtiments en dur) during their time at the Rivesaltes camp.58 In
contrast, harkis resided in canvas military tents—which some sources indicated were holdovers from
the American military stationed in France during the Second World War—from September 15, 1962
until January 8, 1963 when the construction of permanent buildings was completed. As the chart in
figure 17 indicates, three months after the camp opened, 75 percent of the population was still
sheltered by tents, which offered little protection from December snowstorms, a rare occurrence for
this region.
Number of inhabitants
Harkis’ housing at the Rivesaltes camp
December 10, 1962 - January 8, 1963
8632
8062
7834
7032
5028
3588
Tents
Buildings
1763
837
823
12/10/62
12/15/62
12/22/62
12/31/62
0
1/8/63
Figure 17: Housing of harkis at the Rivesaltes camp in tents versus buildings.
59
Boitel, Camp de Rivesaltes; Emmanuel Filhol, La mémoire et l’oubli: L’internement des Tsiganes en France, 1940-45 (Paris:
L’Harmattan, 2004); and Peschanski, La France des camps.
59
Sources: SHAT, 1R 336/8 and SHAT 1R 337/1, Weekly bulletins for Dec. 12, 1962, Dec. 15, 1962, Dec. 22, 1962,
Dec. 31, 1962, and Jan. 13, 1963.
58
180
The tents were divided into two distinct areas in the military portion of the Joffre camp,
located in the western zone of the camp (see figure 18). Based on experiences at the Larzac and
Bourg-Lastic camps, Messmer issued a detailed memo to the Commander of the military region that
encompassed Rivesaltes enumerating the general rules for the organization and functioning of the
newly-opened camp. 60 He gave orders to break each area into villages, which were to house a
maximum of five hundred people. The villages were further divided into ilôts (housing blocks)
grouping two to three villages. The first set of villages (represented by the top two blue squares in
figure 18) grouped harki soldiers and their families who had been transferred from Bourg-Lastic
along with the new arrivals from Algeria who disembarked in Marseille or in nearby Port-Vendres. A
second group of villages contained those transferred in late September from the Larzac camp (the
blue quadrilateral located south of the Bourg-Lastic villages in figure 18). The placement of the
harkis into tents was significant because this physical separation reinforced the overriding
separateness and segregation of the harkis from their surrounding community.
SHAT, 7T 253/3, Minister of Armies to the General Commanding the 9th Military Region, (signed by General
Lagarde), “Objet: Camps d’hébergement d’anciens supplétifs,” 7, N˚ 8.864/EMAT/4-P, Sept. 18, 1962.
60
181
Figure 18: Sketch of the
Rivesaltes camp in the fall of
1962. Source: SHAT 1R
336/8. Photograph of the
document taken by the
author with special
permission.
Between the northern and southern groups of tents were an officer’s mess hall and two
villages made of permanent structures. Block M housed the 124th military company of metropolitan
conscripts who trained on the grounds. Block N sheltered active duty tirailleurs algériens and their
families who were awaiting transfer to the Franco-German border; they numbered 148 people on
November 10.61 Placing the harkis next to these soldiers proved perilous. A December 1-3, 1962
police report indicated that “un antagonisme certain” existed between the harkis and the Algerian
soldiers because some of the latter had allegiances to the FLN, whose members’ violence toward the
harkis was one of the principle factors that had pushed them from Algeria. The same report
additionally revealed that a group of tirailleurs raped an unspecified number of twelve to fifteen year
SHAT, 14T 92, General Le Puloch, Chief of Staff of the Army, to the Military Cabinet of the Minister of Armies,
“Objet: Camp de Rivesaltes,” 1, N˚ 10775 EMAT/CB, Nov. 10, 1962.
61
182
old harki daughters.62 The proximity to the tirailleurs algériens, who were housed in buildings and not
tents, also served a reminder of the harkis’ inferior status as auxiliary soldiers during the colonial era.
During the Algerian War for Independence, unlike the tirailleurs who had six-month contracts, the
April 1956 directive issued by the Resident Minister in Algeria Robert Lacoste officially creating
harkas cited these new troops as “formations militaries temporaires” under the Prime Minister who
were only eligible for renewable one-day contracts.63 Consequently, the harkis were not eligible for
valuable social advantages given to the tirailleurs, such as welfare benefits for their families (allocations
familiales) and health insurance (sécurité sociale).64
The camp’s civil section, under the authority of the Interior Ministry while being staffed by
military officers and civilian employees, was located across the road that bisected the camp, as
depicted in figure 18. The supplies corps (intendance), containing food, linens, and other supplies for
the harki families, was situated at the northernmost portion of the camp. Just to the south was the
camp direction (40e Compagnie de Camp). Located in block J, it contained amenities not present in
the buildings constructed for the harkis. The remnants of the director’s house and the barracks for
camp officers, which are still visible today, include tiled floors, building frames made of bricks,
painted yellow exterior walls, large windows, and sinks for running water. Deciduous trees protected
the officers’ houses from la tramontane—a violent, cold northwesterly wind gusting up to 120
kilometers per hour—and provided shade during sweltering summer days. Further, these barracks
were less densely populated than the tents or permanent structures the harkis inhabited, thereby
affording the residents more personal space.
ADBR, 138 W 5, SCINA, “Synthèse..., journées des 1, 2 et 3 décembre 1962,” d, N˚ 1 803
Ageron, “Les supplétifs algériens,” 5. The harkis were granted daily contracts for over five years until a decree issued
only four and a half months before the Mar. 1962 ceasefire increased the duration to one month. JORF, Prime Minister,
“Décret n˚ 61-1201 du 6 novembre 1961 portant réglementation applicable aux personnels des harkas en Algérie,” Nov.
7, 1961, 10164-65.
64 Ageron, “Les supplétifs algériens,” 6.
62
63
183
To block J’s south was the engineer corps (block P, Génie); it contained building supplies and
offices for those in charge of reconstructing the camp. Just to the west of block P was a general
infirmary and a maternity unit for women (Santé), which were constructed to reduce the number of
harkis who were sent to the Perpignan hospital.65 Nevertheless, according to weekly telegrams from
the camp director to Messmer, approximately twenty seriously ill were hospitalized per week in
Perpignan during the winter and the spring of 1963.66 As a result of the large number of patients
from the Rivesaltes camp, a military wing was created specifically for harkis at Perpignan’s Centre
Hospitalier Maréchal Joffre in October 1963. This special unit served to divide them from the rest
of the hospital’s population, just as the camp separated the harkis from their Rivesaltais neighbors.
The block labeled Mangin on figure 18 was destined for schooling harki children. It
contained twenty-six classrooms, an administrative building, living quarters for teachers, and four
groups of outhouses, all of which were converted in late 1962 and early 1963 from their previous
usage as barracks.67 Attending school buildings on the campgrounds geographically separated harki
children apart from other French children and served as a physical impediment to the harki
children’s inclusion into the French nation, for the school was recognized since the Third Republic
as a vital means of socialization allowing children to become French citizens.
Blocks O and Q housed the centres de formation professionnelle accélérée (CFPA),
mentioned earlier in the chapter, which provided job training for civilians and for military personnel
returning to civilian life. The harkis rarely attended these FPA centers that remained open until
January 1965. They had their own facility that opened in November in block K. Block K was a
centre de préformation, which separated the harkis deemed unfit, owing to their rudimentary
SHAT, 1K 744, Bureau Spécialisé de la Défense Nationale de la Préfecture des Pyrénées-Orientales, “Centre
d’hébergement des supplétifs musulmans de Rivesaltes,” 6.
66
This figure was calculated using weekly bulletins located in: SHAT, 1R 336/8 and SHAT, 1R 337/1.
67 SHAT, 14T 92 (dossier: Travaux), Minister of Armies to the General Commanding the 9th Military Region (signed by
General Pagès, Directeur Adjoint du Direction Centrale du Génie), 2, n˚ 5931 DCG/T2, Nov. 27, 1962.
65
184
educational background in Algeria and linguistic skills, to attend classes with other French citizens.
The block contained seventy barracks, six of which housed the center’s instructors.68 The préformation
classes are discussed later in this chapter.
Rounding out the civil camp were the three blocks in disrepair located on the camp’s
easternmost extremity (E, B, and F on figure 18) that were renovated to house harki families. These
had last been used during the Second World War for prisoners and refugees and in the year after
southern France’s liberation for suspected collaborators. The slow start to construction, which was
compounded by delays in available funds, however did not delay moving the harkis into these
permanent structures.69 According to Messmer, thanks to expedited efforts by local construction
workers in late December and early January, all harki families had migrated the one to three
kilometers from their tents into the renovated barracks by January 8, 1963, even though all buildings
were not completed until the end of February 1963. 70 Each distinct housing block measured
approximately one square kilometer with barracks laid out in rows with as many as thirteen buildings
next to each other, the same layout used during the camp’s previous iterations.71 The three blocks
were divided into villages housing around 1,000 people. Each village was further split into two or
three neighborhoods, each of which contained six to ten barracks.72 These structures, like the tents,
were both cause and effect of the power relationship between the harkis and the French government.
The cramped, unsanitary conditions in the barracks were simultaneously a means for camp officials
SHAT, 1K 744, Bureau Spécialisé de la Défense Nationale de la Préfecture des Pyrénées-Orientales, “Centre
d’hébergement des supplétifs musulmans de Rivesaltes,” 8.
69 SHAT, 14T 92 (dossier: Travaux), General Houssay, Commander of the 9th Military Region, “Compte rendu de
mission du Chef de Bataillon LE HENAFF de la Direction Centrale du Génie,” 7, N˚ 6239 DCG/T2, Dec. 17, 1962.
70 SHAT, 1K 744 (dossier: Point de la situation des Harkis), Minister of Armies Pierre Messmer to Deputy Pierre Bas
(Paris), N˚001814 MA-CC, Jan. 22, 1963.
71 AN, F1a 5140, Génie-Direction de Toulouse, “Petit atlas des bâtiments militaires, Camp Joffre à Rivesaltes (PyrénéesOrientales).”
72 ADPO, 1111 PER 73, L’Indépendant, “Nous sommes mieux que chez nous, disent les harkis (ET C’EST VRAI),” Mar.
22, 1963.
68
185
to exert dominance over the harkis and a consequence of the government’s decision to house them
in a camp.
The harkis’ new homes measured between twenty and twenty-five square meters in a
refurbished rectangular building, which consisted of three to five units, as pictured in figures 19 and
20. The walls dividing each unit, not shown in figure 19 (which was photographed in 2007), were
located where the diagonal wooden beams stretch from midway down the wall. Barrack walls were
made of painted cement blocks and the cathedral ceiling had a wood frame covered by asbestos
sheets, cement, and tiles.73 Each dwelling was accessible from the outside by a wooden door, which
had a small window above it; all but one per barrack had doors on the same long side of the building.
Opposite each of the doors on the long side were two windows measuring one-half meter by onehalf meter, which were four meters from the ground. The other unit, however, had its door at the
end of the barrack and one window on each of the walls on the long side of the building. The
windows’ placement limited the light that entered the homes and, according to one harki woman
named Zohra (speaking in a published 2004 interview), these small, high windows gave the barracks
a prison feel.74 Electricity supplemented the natural light in the units, while a small woodstove
located in a corner heated the harkis’ homes in the winter, often insufficiently. An October 15, 1963
letter to Messmer from a National Assembly deputy representing the Pyrénées-Orientales evidenced
this fact. André Tourné revealed that the harkis were cutting down adjacent almond trees in order to
keep themselves warm; for example, they destroyed 120 of the 300 trees on one local farmer’s
property, causing him to lose at least 150,000 francs.75 The woodstoves were also used to cook their
food rations since they no longer ate food that camp employees (including a small number of harkis)
or by harki women for their own families in the communal kitchens set up next to the tents
AN, F1a 5140, Yves Pérony (Directeur du SFIM), “Note pour M. le Directeur de l’Administration Générale et de
l’Accueil,” N˚ 1708/SFIM/MG, Nov. 2, 1962.
74 Fatima Besnaci-Lancou, Nos mères, paroles blessées. Une autre histoire de harkis (Lunay: Zellige, 2006), 34.
75 SHAT, 19T 257/2, Deputy André Tourné (Pyrénées-Orientales) to the Minister of Armies, Oct. 15, 1963.
73
186
prepared. According to a camp official quoted in a March 22, 1963 article in the local newspaper
L’Indépendant, which was based on the first visit ever made by the media to the camp, “Tous les jours,
chaque personne reçoit 150 grammes de viande, du pain à volonté, des légumes, des conserves, des
desserts. Nous distribuons donc l’équivalent de 3 NF par personne (tout étant obtenu au prix de
gros).”76 This official claimed that a total of one ton of meat, four tons of bread, and twenty-three
tons of wood were distributed each morning at the camp.77
Figure 19: Hollowed out
barrack. Photograph taken
by the author in January
2007.
ADPO, 1111 PER 73, “Nous sommes mieux que chez nous, disent les harkis (ET C’EST VRAI),” L’Indépendant, Mar.
22, 1963.
77 Ibid.
76
187
Figure 20: Still shot from the July 1963 television documentary, “C’étaient les harkis,” showing the dimensions and
layout of the barracks. Source: Cinq colonnes à la une, dir. Jean-Claude Bringuier, RTF (July 7, 1963).
The same article described the interior of the harkis’ homes and demonstrates their attempts
at appropriating these spaces. Rugs, pieces of embroidery, and the former soldiers’ military
uniform—and even pictures of camels, palm trees, and tents that they had drawn themselves—hid
the drab cement walls in some of the units.78 The harkis seemingly chose these items as a “tactic” (in
a Certelian sense), or a type of “mimicry” (à la Homi Bhabha), to turn the stark and monotonous
buildings, which exerted power over the harkis by their very architecture, into more familiar
surroundings.79 The men presumably hung their military uniforms—which they were not permitted
to wear on the camp grounds so that they would not be confused with enlisted French Muslim
soldiers—to recall their service to, and perhaps justify their belonging in, France.
The harkis’ homes had no internal plumbing. Instead, families carried water from spigots
located on the camp’s perimeter, which were as much as a kilometer away from the barracks.80
Turkish-style toilets in rows of ten constructed at the same time as the barracks were contained in
Ibid.
For a discussion of the notions of “tactics,” see Michel de Certeau, “Faire avec: Usages et tactiques,” L’Invention du
quotidien, 1. Arts de faire (Paris: Gallimard, 1990), 50-68. For a succinct explanation of mimicry, see Homi Bhabha, “Of
Mimicry and Man,” in Tensions of Empire, eds. Cooper and Stoler, 152-160.
80 AN, F1a 5140, Génie-Direction de Toulouse, “Petit atlas des bâtiments militaires…”
78
79
188
rudimentary outhouses. As the photograph in figure 21 demonstrates, the stalls afforded individuals
little privacy and had limited protection from the external elements. There was no sewage system for
the toilets. Alternatively, a Perpignan waste management company emptied the raw sewage every
two days.81 The hygiene concerns resulting from this insalubrious situation were compounded by the
outdated sewerage system in the adjacent military section of the camp. Effluents flowed untreated
through a pipe 250 millimeters in diameter and four kilometers in length into the nearby Agly River,
also provoking questions of potential sanitary problems for the surrounding community.82
Figure 21: Outhouses
with Turkish toilets
on the camp
grounds. Photograph
taken by the author
in January 2007.
Discussions among government officials about whether to build a water purification plant
and a sewerage system to combat these unsanitary conditions provide an example of how budgetary
constraints and disagreements between the Ministry of Armies and the Ministry of Repatriates
SHAT, 14T 92 (dossier: Travaux), General Houssay, Commander of the 9th Military Region, “Compte rendu de
mission du Chef de Bataillon LE HENAFF de la Direction Centrale du Génie,” 5, N˚ 6239 DCG/T2, Dec. 17, 1962.
82 SHAT, 13T 286, Brigadier General PLENIER (Directeur de la Section Technique des Bâtiments, Fortifications et
Travaux) to the Minister of Armies (Direction Centrale du Génie), “Objet: Camp de RIVESALTES (PyrénéesOrientales) Travaux d’hébergement des ex-supplétifs – Amélioration des installations sanitaires,” 2, July 16, 1963.
81
189
affected the camp’s physical structures and, consequently, the harkis’ living conditions. During the
aforementioned October 24, 1962 meeting at the Pyrénées-Orientales prefecture, the prefect was
charged with looking into constructing this system. National officials stressed that it would not only
serve the current population, but also future ones that might live on the camp grounds.83 However,
when the budget was submitted the next day, no previsions were made for a water purification plant
or a sewerage system.84 As the weather warmed in the spring and no improvements had been made,
Army officials were concerned that camp dwellers and locals alike would be susceptible to health
epidemics. In addition to problems provoked by the raw sewage that remained in the latrines, these
officials feared that the shallow Agly River would dry up in the summer, as it habitually did, thereby
fostering the spread of disease. In a May 10, 1963 letter, Messmer wrote to Pérony that it had
become urgently necessary to build a water purification plant and a sewerage system for the camp.85
With no money for these in the Army’s budget, Messmer requested 1.5 million francs from the
Ministry of Repatriates, underscoring that “[c]e besoin correspond uniquement à la situation actuelle
et n’est pas imposé par l’utilisation ultérieure du camp militaire qui ne saurait être ni prolongée ni
massive puisque la destination militaire retenue pour ce camp est celle d’un camp de transit.”86 This
request was complicated by the fact that it would take a year and a half to two years to build these
facilities. By this time, it was estimated that almost all, if not all, of the harkis would have evacuated
the Rivesaltes camp. Pérony replied a month later that the Ministry of Repatriates would not allocate
such funds from its budget, which led Messmer to ask the Director of the Army’s Health Services
whether the lack of a proper sewage system really posed a risk to camp dwellers and neighboring
Rivesaltes residents, whether the Army could resort to the massive use of disinfectants and, with a
SHAT, 1K 744, Bureau Spécialisé de la Défense Nationale de la Préfecture des Pyrénées-Orientales, “Centre
d’hébergement des supplétifs musulmans de Rivesaltes…,” 6.
84 SHAT, 13T 286, “Rapport du Colonel CASSAN Directeur des Travaux du Génie, relatif au Camp de Rivesaltes
Hébergement des ex-supplétifs. Bilan des dépenses. Demande de crédits supplémentaires,” 1, N˚ 4018/Te, May 13, 1963.
85 SHAT, 13T 286, the Minister of Armies to the Cabinet of the Minister of Repatriates, “Objet: Amélioration des
installations sanitaires du Camp de RIVESALTES,” 1-2, N˚ 012052 MA/CC, May 10, 1963.
86 Ibid., 1.
83
190
new wave of harkis expected to arrive, the maximum population the current facilities would allow.87
The Director responded after a June 18 visit to the camp with an Army epidemiologist that the
current system, even if it violated public hygiene laws, did not increase local residents’ risk of disease
and that recourse to massive quantities of disinfectants would be a viable solution.88
The possibility of building a sewerage system and water purification plant for the harkis’
section of the camp to conform to public hygiene standards and reduce the risk of disease was thus
swept away. Refusals by both the Ministry of Armies and Ministry of Repatriates to provide the
necessary funds to make what Messmer had referred to in his May 10, 1963 letter to Minister of
Repatriates Missoffe as “améliorations indispensables” ended the project.89 The uncertainty of the
length of the harkis’ stay at the Rivesaltes camp, much like their ambiguous status, led both
ministries to shift the responsibility for the camp’s upkeep onto each other. Ministry of Armies
officials believed that the improvements had to be made immediately purely to address the current
situation. On the contrary, Ministry of Repatriates officials claimed that they lacked the appropriate
funds for facilities that would only be realized after the harkis evacuated the camp and their ministry
no longer would be responsible for the camp’s maintenance. Once both ministries refused to
finance the improvements, reports from Army officials in summer 1963 turned toward minimizing
the health risks for the harkis and locals. For example, an army official reported after an early July
visit by technical officers, “Malgré ce système d’assainissement rudimentaire la situation sanitaire de
SHAT, 13T 286 and SHAT, 14T 92 (dossier: Travaux), Pierre Messmer, “Note pour le Médecin Général Directeur du
Service de Santé, Objet: Amélioration des installations sanitaires du Camp de RIVESALTES,” N˚ 015046/MA/CC,
June 12, 1963.
88 SHAT, 1R 337/1, Inspector general Dr. Debenedetti (Director of the Army’s Health Services), “Note pour le Cabinet
de M. le Ministre des Armées, Objet: Amélioration des installations sanitaires du Camp de Rivesaltes,” 2, N˚ 2244
2/T/DCSSA, June 29, 1963.
89 SHAT, 13T 286, the Minister of Armies to the Cabinet of the Minister of Repatriates, “Objet: Amélioration des
installations sanitaires du Camp de RIVESALTES,” 2, N˚ 012052 MA/CC, May 10, 1963.
87
191
ce camp est satisfaisante,” a statement in complete contradiction with the sanitary concerns
expressed since the harkis had arrived at the camp the previous September.90
The harkis’ cramped living situation in the tents and barracks led to health problems, which
the government ultimately chose to blame on the harkis’ inferior and foreign ways and not on the
buildings. In January 1963 Dr. Aujaleu, the Ministry of Health’s public health director, warned top
Repatriates official Yves Pérony of a tuberculosis epidemic sweeping through the camp. He
attributed the spread of the disease to “les conditions de promiscuité dans lesquelles vivent les
membres d’une même famille, et les familles de harkis entre elles, qu’elles habitent sous tente, ou
dans des bâtiments en dur,” as well as many harkis’ refusal to be hospitalized when infected.91 The
two-room barracks in which harki families lived well exceeded the legal population limit for public
housing facilities. According to article twenty of a June 30, 1961 decree, which enumerated
“population conditions” for residences constructed by the state, dwellings housing more than three
occupants must contain at least three rooms.92 This requirement was disregarded when housing
harkis, even after they were moved from the tents to the permanent structures. Between children
and the occasional extended family members, households exceeded the three-person limit almost
without exception.
While in January 1963 Aujaleu wrote that the camp’s physical structures provoked these
unsanitary conditions and were obstacles to his Ministry’s ability to “assurer une protection sanitaire
efficace,” a few months later with tuberculosis still propagating in the camp, he shifted the blame
SHAT, 13T 286, Le Général de Brigade PLENIER, Directeur de la Section Technique des Bâtiments, Fortifications et
Travaux à Monsieur le Ministre des Armées, Direction Centrale du Génie, “Objet: Camp de RIVESALTES (PyrénéesOrientales) Travaux d’hébergement des ex-supplétifs – Amélioration des installations sanitaires,” 2, July 16, 1963.
91 SHAT, 1R 336/6, Director of Public Health (Dr. Aujaleu) to the Minister of Repatriates (to the attention of M.
Pérony), “Objet: Lutte antituberculeuse dans les camps ou les chantiers forestiers qui groupent des harkis (dépistage,
hospitalisation, vaccination par le BCG),” DGSP/HS 6, Jan. 8, 1963.
92 This passage from the June 30, 1961 decree was attached to: CAC, 19920149/1/6, SFIM, “Compte rendu de la
tournée au camp de Rivesaltes,” Oct. 15, 1962.
90
192
from the physical space to the harkis themselves.93 The Director of the Army’s health services
claimed that the disease was spreading not because of a lack of medical treatment: “… [I]l est
nécessaire d’observer que les us et les coutumes de ces populations posaient des difficultés du même
ordre aux dispensaires qui œuvraient en Algérie, et que leur transplantation en Métropole n’a pas
modifié radicalement leurs traditions ancestrales.”94 Unlike the January letter that voiced concerns
about the difficulties that the physical dwellings engendered for the harkis’ health, this
correspondence contained no reference to their overcrowded living conditions in the barracks as a
reason for the spread of tuberculosis. Instead, health officials chose to turn the disease-prone
Muslims into scapegoats for the spread of tuberculosis. They would not recognize that such
outbreaks were endemic to the camp space where the government opted to house the harkis.
Epidemics had been consistently present throughout Rivesaltes’ history as a refugee and prisoner of
war camp.
The camp structures and the insalubrious living conditions they induced directly contributed
to the harkis’ physical exile from the surrounding French community and, in turn, stressed that the
harkis were refugees and foreigners on French soil. Portions of the Rivesaltes camp were encircled
by barbed wire, a material which functioned not only to prevent people inside from leaving, but also
to prevent those outside from being able to have contact with the harkis. 95 The physical and
psychological constraints created by the Rivesaltes camp echoed the division between Muslims and
French during the colonial era, divisions that were exemplified in North African cities by the
indigenous casbah or medina juxtaposed with the rebuilt modern European city96 and in metropolitan
SHAT, 1R 336/6, Dr. Aujaleu, “Objet: Lutte antituberculeuse dans les camps…”
SHAT, 1R 336/6, Central Administration of the Armies Health Service to the Minister of Armies’ Military Cabinet,
“Objet: Problème de la lutte antituberculeuse dans les camps,” Mar. 4, 1963.
95 On the political function and history of barbed wire, see Olivier Razac, Histoire politique du barbelé: la prairie, la tranchée, le
camp (Paris: Fabrique, 2000).
96 For a description of French architectural and urban policies in its colonies, see Gwendolyn Wright, “Tradition in the
Service of Modernity. Architecture and Urbanism in French Colonial Policy, 1900-1930,” in Tensions of Empire, eds.
Cooper and Stoler, 322-345.
93
94
193
France by the strictly monitored dormitories (SONACOTRAL foyers) for migrant workers.97 These
types of spaces often imposed a physical barrier—such as a wall or fence—as a marker
differentiating French and Muslim populations. The camp’s architecture continually reminded harkis
of their inferiority to other French citizens by relying on visible markers such as the barbed wire
fencing, tents, and barracks that disfigured the Catalan countryside and implicitly proclaimed their
power over the camp’s subjugated populations. However, how government officials used the camp’s
physical space was not the only way that they exerted control over the harki population residing in
Rivesaltes. An intricate network of military officers, policemen, and Ministry of Repatriates’
employees closely monitored their movements and daily life.
The Power of Encadrement
As mentioned in the introduction to this chapter, encadrement is the term the government
employed to describe state agents’ strict control over the harkis’ quotidian life. The French verb
encadrer, which means “to surround with a frame,” provides a fitting image of how the Rivesaltes
camp’s physical space, state agents’ authority over the harkis, and obligatory classes to prepare them
for metropolitan life created physical and psychological borders around the harkis. The military
structures, especially the unused watchtowers dotting the camp’s perimeter, suggest an “architecture
of domination” over the harki population who inhabited this space. In Surveiller et punir: Naissance de
la prison, Michel Foucault demonstrates that modern Western society shifted during the eighteenth
century from a “spectacle punitif” of torture and execution of the body to more disciplinary actions
whose focus was henceforth on the prisoner’s soul, to the very essence of his being.98 Employing the
example of the panopticon prison, Foucault contends that power is now more complex and
dispersed among actors representing the state who act as disciplinarians toward the populations they
97
98
For a discussion of the history of the SONACOTRA foyers, see: Hmed, “Loger les étrangers ‘isolés.’”
Michel Foucault, Surveiller et punir: Naissance de la prison (Paris: Gallimard, 1975), 14-15.
194
watch over: “… toute une armée de techniciens est venue prendre la relève du bourreau, anatomiste
immédiat de la souffrance: les surveillants, les médecins, les aumôniers, les psychiatres, les
psychologues, les éducateurs…”99 While the Rivesaltes camp was not a prison, the state nonetheless
employed “une armée de techniciens”—doctors, policemen, social workers, and teachers—to
exercise power over, and encadrer, the harkis.
The relations of power in the Rivesaltes camp echo Foucault’s notion of modern “biopower,”
that is, the simultaneous use of disciplinary actions over individuals exercised at the local level and
the state regulation of populations, which he calls a “technology of security.”100 The Rivesaltes harki
population fits into this Foucauldian paradigm since the state viewed it as an “internal danger” from
which the “security of the whole” needed to be protected.101 In a December 1962 letter, Minister of
Armies Messmer warned Prime Minister Pompidou that turning the camp into a more permanent
residence risked endangering “la sécurité des régions avoisinantes.” 102 Another way that the
biopower paradigm applies to the harki population is through the state’s housing policies that
differentiated repatriates based on their ethnicity. Foucault underscores that the basic mechanism of
biopower is racism, which he defines as: “The appearance within the biological continuum of the
human race of races, the distinction among races, the hierarchy of races, the fact that certain races
are described as good and that others, in contrast, are described as inferior.”103 Racism is a means for
the state to separate and hierarchize populations, which occurred when the French government
differentiated the returning pieds-noirs of European and Jewish descent from the Arab and Berber
Ibid., 18.
Foucault, “Society Must Be Defended,” 249-250.
101 Ibid., 249.
102 SHAT, 1R 336/8 (dossier: Camps–Rivesaltes et St. Maurice l’Ardoise), Minister of Armies Messmer to Prime
Minister Pompidou, N˚ 031038 MA/CC, Dec. 10, 1962.
103 Foucault, “Society Must be Defended,” 254-255.
99
100
195
harkis. It created different institutions to regulate their arrival and life in metropolitan France and
provided different spaces for them to live.104
The government advanced several reasons to justify the strict regime of encadrement at
Rivesaltes. Foremost, it believed that this type of surveillance would be a means to prevent harkis
from joining two potentially dangerous groups created during the Algerian War for Independence:
the FLN and the OAS. This fear—and treatment—predated the harkis’ arrival at the Rivesaltes
camp. Encadrement was repeatedly invoked as the dynamic structure of relations between harki
soldiers and their superiors in the French Army. François-Xavier Hautreux maintains that during the
Algerian War for Independence, “les directives de l’état-major insistent sur la nécessité de
l’encadrement et de la surveillance de ces nouvelles recrues [harkis], mais également de leurs familles,
par le biais desquelles le FLN est censé exercer sa propagande.”105 Concern about the OAS, which
was portrayed in various media outlets as actively recruiting the harkis who arrived in France (as
cited in chapter two), manifested itself when hiring camp staff. The government simultaneously
preferred to employ those who had lived in Algeria, whether European or Arab, because of their
familiarity with “le milieu nord-africain” yet acted circumspectly with these individuals because of
their possible link to the OAS. For example, an August 1962 note signed by a pied-noir Army
sergeant who was employed as a contract worker at the Bourg-Lastic camp (and subsequently
transferred to Rivesaltes) stated: “Je soussigné G. Angélo Agent Contractuel de la Mission des
Rapatriés de Bourg-Lastic (Puy-de-Dôme) déclare sur l’honneur ne jamais avoir été pris en compte
par un organisme quelconque à mon retour d’Algérie.”106 Fears about aftershocks of the recently lost
war in metropolitan France, therefore, provide the primary motivation for encadrement.
Frederick Cooper discusses the concept of differentiation in Colonialism in Question, 154. Additionally, Mahmood
Mamdani focuses his work, Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism, (Princeton, NJ:
Princeton University Press, 1996), on the regime of differentiation, which he defines as “institutional segregation.”
105 Hautreux, “Engagement des harkis,” 37-38.
106 CAC, 19970146/3 (dossier: personnel Angélo G.), handwritten note by Angélo G., Bourg-Lastic, Aug. 23, 1962.
104
196
The government further regarded this principle as a panacea for the grave problems
encountered during the first few months that Rivesaltes was open. In particular, episodes of violence
threatened the camp’s stability and endangered the security of neighboring regions, according to
Messmer.107 In addition to the group of tirailleurs that raped an unspecified number of twelve to
fifteen year old harki daughters mentioned above, police reports disclosed several troubling events.
In October one harki beat another to death108 and a grenade exploded in close proximity to a group
of harkis. 109 The next month the gendarmerie seized a 7.65 caliber pistol from a harki. 110
Furthermore, while the women spent their days tending to their children, the men did not have jobs
and, in the words of Pompidou, their idleness left them to “flân[er] aux environs” and to stray as far
as Perpignan.111 Pompidou directly attributed the “oisiveté” that provoked such dangerous incidents
and the ability for the harkis to the absence of a camp director, and a lack of “encadrement et …
contrôle de police appropriés,” that is, troops who had recently returned from Algeria.112 The tight
surveillance necessary for such control was complicated by a dearth of personnel. On October 30
there were only two social workers113 and four nurses working at the camp.114 By November 8, 1962,
the Army had assigned 69 military officers, some of who had not yet reported for duty, for the
camp’s 8,200 harkis.115 Moreover, from the time that the camp opened there was lack of military
police to guarantee the camp’s security, as the episodes of violence cited above demonstrate.
SHAT, 1R 336/8, Letter from Messmer to Pompidou, Dec. 10, 1962.
ADBR, 138 W 5, SCINA, “Synthèse…, journée du 17 octobre 1962,” 1, N˚ 1 772.
109 ADBR, 138 W 5, SCINA, “Synthèse..., journées des 27, 28 et 29 octobre 1962,” 1, N˚ 1 780.
110 ADBR, 138 W 5, SCINA, “Synthèse..., journée du 27 novembre 1962,” 1, N˚ 1 799.
111 SHAT, 1R 336/8 (dossier: Camps–Rivesaltes et St. Maurice l’Ardoise), Prime Minister Pompidou to the Cabinet of
the Minister of Armies, Nov. 21, 1962.
112 Ibid.
113 SHAT, 1R 336/8 (dossier: Camps–Rivesaltes et St. Maurice l’Ardoise), M. Moullet, “Objet: Le camp de Rivesaltes,”
N˚ 43219 MA/CM, Oct. 30, 1962.
114 SHAT, 1R 336/8 (dossier: Camps–Rivesaltes et St. Maurice l’Ardoise), M. Moullet, “Objet: Service de santé au camp
de Rivesaltes,” N˚ 43217 MA/CM, Oct. 30, 1962.
115 SHAT, 1R 336/8 (dossier: Camps–Rivesaltes et St. Maurice l’Ardoise), General Lagarde, “Note pour la Direction des
Personnels Militaires de l’Armée de Terre,” N˚ 3458 EMAT/1.E., Nov. 13, 1962.
107
108
197
However, the national Commander of the Gendarmerie was reticent to send more military
police to the Rivesaltes camp because he believed his men should be reserved for conflicts on
foreign soil: “les postes prévotales [sic] ne peuvent être installés en France, mais seulement dans des
pays occupés militairement.” 116 The Commander’s position reflects the camp’s fundamental
contradiction: were the space and the harkis foreign? Military police were deployed to watch over
the harkis at Rivesaltes, which implies that the camp was an “occupied foreign country” (following
the Commander’s reason), yet the space was located on metropolitan soil and it lodged French
citizens. This ambiguity created difficulties in how to run the camp.
Prime Minister Pompidou blamed these shortcomings squarely on the Ministry of Armies
because, according to a August 22, 1962 Interministerial Council meeting, this ministry was
responsible for providing the harkis with shelter, food, security, encadrement, leisure activities, and job
training.117 In a December 8 letter to Minister of Armies Messmer, Pompidou seemed less concerned
with the Ministry of Repatriate’s accountability for the camp’s failures, even though this ministry had
the responsibility of paying the camp personnel’s salaries and of placing the harkis in jobs around
France to ensure that the camp remained a transitory space.118 Pompidou argued that because “les
refugiés … sont d’ailleurs habitués à une organisation de type militaire comportant un encadrement
solide” they should be subject to stricter discipline.119 He also encouraged Messmer to implement at
CAC, 19920149/1 (dossier 6: Comptes rendus des tournées, décembre 1962-septembre 1965), SFIM, “Compte rendu
de la tournée au camp de Rivesaltes,” 2, Oct. 15, 1962.
117
SHAT, 1K 744 (dossier: Conseils interministériels), “Relevé de décisions du Conseil restreint du mercredi 22 août
1962, à 10h30, au sujet des mesures à prendre en ce qui concerne les rapatriés et les problèmes de l’ordre public,” 1. The
following ministers and secretaries of state were present at the meeting: Pompidou (Prime Minister), Joxe (Algerian
Affairs), Frey (Interior), Giscard d’Estaing (Finance and Economic Affairs), Sudreau (Education), Pisani (Agriculture),
Grandval (Labor), Maziol (Construction), Peyrefitte (Information), and Boulin (Repatriates). Oddly, Minister of Armies
Messmer was not present. Yet he was later critical of the decision to require his ministry to foster so much manpower
for the harkis’ sustenance and to split the costs with the Ministry of Repatriates for employing these personnel and for
constructing and restoring barracks.
118 For a breakdown of the division of finances between the two ministries, see SHAT, 1R 336/8, Ministry of Armies,
Financial Services Administration, “Note relative aux problèmes financiers posés par l’installation et le fonctionnement
des camps d’ex-supplétifs de Rivesaltes et de Saint-Maurice l’Ardoise,” 3-4, N˚ 2952 MA/DSF/1, Nov. 22, 1962.
119 SHAT, 1K 744 (dossier: Accueil à Saint-Maurice-l’Ardoise et à Rivesaltes), Prime Minister Pompidou to Minister of
Armies Messmer, Dec. 8, 1962.
116
198
the Rivesaltes camp the stricter disciplinary measures in place at the smaller Saint-Maurice-l’Ardoise
camp. 120 Messmer responded on December 10 by asserting that there were more problems at
Rivesaltes because it had double the population. Furthermore, he declared that unlike the SaintMaurice-l’Ardoise harkis, many of the Rivesaltes-dwelling harkis had arrived in France in early
summer 1962, before the massive FLN violence campaign began in Algeria. Therefore, he reasoned
that the atmosphere of terror that the more recently arriving migrants had experienced in Algeria
made them more amenable to military discipline. Messmer defended his ministry by revealing that he
had recently appointed a camp director who had served two tours of duty in Algeria and that he had
given orders for more gendarmes to serve on the camp grounds. Moreover, he evoked the Ministry
of Repatriates’ failure to find the harkis jobs as a major factor in the camp’s difficulties.121 Messmer
did agree, nonetheless, that the Rivesaltes harkis needed more encadrement. Although he maintained
that his ministry no longer considered the harkis as members of the French Army, he asserted in a
letter to the Chief of Staff of the Army that “il convient de leur appliquer une organisation et des
habitudes de vie militaire: fractionnement en groupes placés sous les ordres d’un Officier ou SousOfficier, entretien du camp et de ses abords, surveillance de la circulation, des entrées et des sorties,
les déplacements à l’extérieur devant être limités.”122 In this way, Messmer appeared to view the
Rivesaltes-dwelling harkis in the same optic as the Army had seen its colonial soldiers. A May 20,
1957 Army memorandum outlining the role of the harki units, or harkas, stressed: “En aucun cas les
harkas ne doivent être engagées isolément.”123 Like the May 1957 memorandum Messmer advocated,
in December 1962, that military authorities must closely watch over the movements of former harki
soldiers.
SHAT, 1K 744 (dossier: Accueil à Saint-Maurice-l’Ardoise et à Rivesaltes), Prime Minister Pompidou to Minister of
Armies Messmer, Dec. 8, 1962.
121 SHAT, 1R 336/8 (dossier: Camps–Rivesaltes et St. Maurice l’Ardoise), Minister of Armies Messmer to Prime
Minister Pompidou, N˚ 031038 MA/CC, Dec. 10, 1962.
122 SHAT, 1K 774 (dossier: Accueil à Saint-Maurice-l’Ardoise et à Rivesaltes), Pierre Messmer to the Chief of Staff of the
Army, “Objet: Camps d’ex-supplétifs Musulmans,” N˚ 48353 MA/CM, Dec. 10, 1962.
123 Hautreux, “Engagement des harkis,” 36.
120
199
The other measures taken to remedy the camp’s faults also hinged on the premises of
increasing the control over the harkis’ lives and of having “une conception moins ‘civile’” of the
harkis’ living situation, in essence one more in line with conceptions during the colonial era.124 The
most obvious way to “maintain order” was to augment the number of officers both inside and
around the camp.125 The Army reinforced this military hierarchy with a similar structure of harkis
who would thus participate in their own surveillance, a function that harkis had performed in the
harka units during the Algerian War for Independence and that kapos (Jewish prisoners) had
performed in concentration and extermination camps during the Holocaust.126 Men chosen for these
positions at Rivesaltes had to be able to converse in French with the military personnel in charge
and often acted as translators with those who only spoke Berber or Arabic dialects. French-speaking
harkis had fulfilled this same function during the war.
To compensate for his lack of on-site presence at the camp, on November 30, 1962
Messmer sent a memo to the camp director asking him to send to Paris each Monday morning a
telegram, which would serve as another tool of encadrement over the harkis. These comptes rendus
hebdomadaires (weekly memos) were to contain information about the number of men, women, and
children who arrived at and left the camp; their état d’esprit; their état sanitaire (including the number
of births, deaths, and hospitalizations); and an update on the progress of the construction of
barracks.127 The first weekly memo appeared on December 8, 1962 and by the end of month, they
included information about the number of children attending school, the number of men
participating in job training classes, and the number of women attending cooking, hygiene, and
sewing courses.
SHAT, 1R 336/8 (dossier: Camps–Rivesaltes et St. Maurice l’Ardoise), Ministry of Armies, Technical Adviser de
Christen, “Inspection à Rivesaltes et à St. Maurice l’Ardoise des 13, 14, et 15 décembre 1962.”
125 SHAT, 1K 744 (dossier: Accueil à Saint-Maurice-l’Ardoise, Rivesaltes, et Bias), Telegram from the Minister of Armies
to the Marseille Gendarmerie, “Maintien Ordre Camp Rivesaltes,” N˚ 19945/EMAT.3.EPO/DR, Dec. 29, 1962.
126 SHAT, 1R 336/8, De Christen, “Inspection à Rivesaltes….”
127 SHAT, 1R 336/8 (dossier: Camps–Rivesaltes et St. Maurice l’Ardoise), Minister of Armies Paris (signed by Moullet)
to the Commanders of the Rivesaltes and Saint-Maurice-l’Ardoise camps, Telegram N˚ 47233/MA/CM, Nov. 30, 1962.
124
200
However this means of surveillance proved faulty. Since the correspondence was in the form
of telegrams, it contained little description. Instead, information was elliptically conveyed with
phrases such as “effectifs personnes arrivées au camp 99.” Interestingly, every one of the fifty
telegrams found in the Army Archives—corresponding to fifty of the fifty-six weeks from
December 1-8, 1962 to December 21-28, 1963—was marked “État d’esprit Bon.”128 This one-word
qualification contradicts the plethora of government documents underscoring that there were grave
problems with the harkis’ morale while at Rivesaltes. Moreover, while the first telegram reported
that their health was “satisfaisant;” all forty-nine successive ones describe their health as “bon” or
“très bon.” These judgments demonstrate a dissociation between the Panglossian impression that
these telegrams gave to Parisian officials in the Ministry of Armies, including Messmer himself (who
marked many of the telegrams with “Vu, PM”), and other government documents that reported, for
example, the tuberculosis epidemic and unclean drinking water at the camp.
To provide Messmer with such information and to remedy what Pompidou had labeled as
“[des] conditions de vie très défectueuses,” the government primarily relied on a camp director.
Initially, the Ninth Military Region Commander, who commanded the Army stationed on the
territory stretching from Marseille to Perpignan, was charged with overseeing the camp.129 His ability
to provide the organization that the camp necessitated was complicated because he was based in
Marseille and was also in charge of the Saint-Maurice-l’Ardoise camp in the Gard Department.
Therefore, after close to three months without an on-site director, on December 7, the Army
The telegrams are divided between two cartons at the Service historique de l’Armée de Terre (SHAT): 1R 336/8
(dossier: Messages de Rivesaltes et Saint-Maurice-l’Ardoise) and 1R 337/1 (dossier: Camp de Rivesaltes et SMA).
129 SHAT, 7T 253/3, Minister of Armies to the General Commanding the 9th Military Region (signed by General
Lagarde), “Objet: Camps d’hébergement d’anciens supplétifs,” Annexe (“Note sur l’organisation et le fonctionnement
des camps d’ex. supplétifs réfugiés d’A.F.N.”), 1, N˚ 8.864/EMAT/4-P, Sept. 18, 1962.
128
201
designated a veteran field officer who had served two tours of duty in Algeria to manage the camp’s
daily operations.130
The camp director oversaw the two parallel structures that served to encadrer the harkis at the
Rivesaltes camp: a civil section comprised of medical personnel, social workers, teachers, and
administrative staff under the authority of the Ministry of Repatriates, and a military section
consisting of officers and policemen under the authority of the Ministry of Armies. On the military
end, the hierarchy of officers and policemen were responsible for the direct supervision and policing
of the harkis and for ensuring the camp’s day-to-day functioning.131 All of the officers hired in
September and October 1962 when the camp opened had served in Algeria, either in the military or
in the SAS.132 Some of the officers, such as Captain Raymond M., had been stationed at harki
relocation camps in Algeria133 while others, such as Pierre L., had accompanied the harkis from one
of their initial stopping points in France, the Larzac camp.134 Messmer defined these men’s principal
duty as “encadrement” of the “réfugiés,” and specified, “ils participent au maintien de la discipline et
ils ont à traiter des problèmes humains relatifs aux réfugiés, dont ils sont les intermédiaires obligés
vis-à-vis du Commandant du camp et du délégué du Ministère des Rapatriés.”135 A lieutenant aided
by several noncommissioned Algerian and French officers directed each five-hundred person village
of tents and a captain supervised two or three of these villages. The officers were charged with
130SHAT,
1R 336/8 (dossier: Camps–Rivesaltes et St. Maurice l’Ardoise), Minister of Armies Messmer to Prime Minister
Pompidou, N˚ 031038 MA/CC, Dec. 10, 1962.
131 SHAT, 1K 744 (dossier: Accueil à St. Maurice l’Ardoise, Rivesaltes et Bias), Prime Minister Pompidou to Minister of
Armies Messmer, Nov. 21, 1962.
132 ADBR, 138 W 5, SCINA, “Synthèse..., journées des 5, 6, 7 et 8 octobre 1962,” e, N˚ 1 765.
133 AN, F1a 5138, Personnel file of Capitan Raymond M.
134 AN, F1a 5137, Personnel file of Pierre L.
135 “Problèmes humains” is underlined in the original typewritten document. SHAT, 7T 253/3, Minister of Armies to
the General Commanding the 9th Military Region (signed by General Lagarde), “Objet: Camps d’hébergement d’anciens
supplétifs,” Annexe (“Note sur l’organisation et le fonctionnement des camps d’ex. supplétifs réfugiés d’A.F.N.”), 2, N˚
8.864/EMAT/4-P, Sept. 18, 1962.
202
providing for the harkis’ material needs, such as organizing the distribution of food and housing and
arranging medical care from the on-site personnel.136
The Army tasked the camp’s military police with maintaining “la salubrité physique et morale
du camp” and guaranteeing safety in the camp. Specifically, the officers confiscated weapons from
harkis who arrived directly from Algeria, collected the harkis’ valuables to store in a safe place,
surveyed residences and public spaces, disciplined those who disturbed the peace, reported the most
serious problems to the local police, sought out information about interactions between harkis and
the FLN or the OAS, and encouraged the men and children to participate in activities to keep them
occupied.137
The military structure worked in tandem with civilian employees of the Ministry of
Repatriates who staffed the headquarters and those who were course instructors. In December 1962
a director (Mr. Louis Couston), a deputy director (Mr. René Aucante), and a liaison with repatriate
offices in each department of France (Mr. Robert Bourgat) sat at the top of this hierarchy. All three
of these men had served in Algeria and bore the title “Administrateur des Services Civils d’Algérie”
while working at Rivesaltes.138 They oversaw the twenty-one workers employed in the headquarters’
six divisions: general services (management of the camp’s finances and personnel), pensions and
hospitalizations, social security, employment (matching harki men with potential employers), arrivals
(the creation of family dossiers to keep track of each harki who entered the camp), and departures
(the dissemination of the dossiers and train vouchers). 139 Some personnel members had been
transferred from the Larzac camp, including the employment division chief Jean Filhastre and
Ibid., 2.
Ibid., 3-4.
138 René Aucante and Robert Bourgat’s personnel files are respectively archived in CAC, 19970146/1 and CAC,
19970146/2.
139 For a breakdown of the different offices, see AN, F1a 5138, Bureau de gestion des camps d’ex-supplétifs,
“Organigramme des services extérieurs – Rivesaltes,” Dec. 1962. For a more detailed description of the responsibilities
of the Mission de Liaison, see: CAC, 19970146/2 (personnel file of Robert Bourgat), Robert Bourgat to the Cabinet of
the Pyrénées-Orientales Prefect, “Objet: Situation des ‘Harkis’ recasés dans les Pyrénées-Orientales,” 1-2, N˚
835/RB/RB/3, June 27, 1963.
136
137
203
reception office agent Angelo Garrouste. Other employees—such as the director of the General
Services division, André Perrault, and future director of the arrival center, Henri Tardy—were career
military men. Meanwhile, the head of the departures office, Jeanine Bouchet, had worked as a
municipal civil servant in Algeria.140
These individuals did not make daily contact with the harkis. Instead, state agents added to
the paper trail of documents initiated in Algeria to continue to categorize, and distinguish, the harkis.
Despite little and irregular in-person contact, the paper trail, which served to keep vigilant watch
over the harkis, thus was another form of encadrement. After arriving at the camp from the Rivesaltes
train station or from boats docking at Port-Vendres, each family passed through the “arrival center”
in block J. Here, repatriate service employees opened a dossier for each family to include
information such as the adults’ civil status, birthdates, birthplaces, and the males’ service in the
harkas or other auxiliary force. Often, the harkis were required to provide proof of their service, such
as a signed affidavit by a military official for former harkis or the SAS leader for former moghaznis,
a “carte de combattant,” or a “carnet individuel de harki.” Further, the harkis who had waited in
relocation camps in Algeria before setting sail for the metropole were expected to produce the
“carnet familial de rapatrié” that each family head had received. One such example—that of
M’hamed H. who passed through the Zeralda refugee camp outside of Algiers—is pictured in figure
22.
140
All of these individuals’ personnel files are located in CAC, 19970146/1-6.
204
Figure 22: Carnet familial de rapatrié of M’hamed H.
Source: ASRBR, social welfare file of M’hamed H., nº B374286.
As harkis remained in the camp, their files accumulated additional documents, which
thickened the paper trail and thus the state’s surveillance of them through its local-level agents who
implemented national policies. These papers included proof of military service by their commanding
officer, the nationality declarations that they signed during the ceremonies discussed in the previous
chapter, social security and family allowance cards issued by the state, and notes about job
preferences that they expressed during brief interviews with employment counselors. Finally, when
each harki family left the camp, state agents added to their dossiers a receipt for the train vouchers
that the central office issued to ensure they would be able to reach their destination and details about
their reclassement.141 Upon their departure, camp officials presented the harkis with a copy of these
In his discussion of Polish refugees during the July Monarchy, Gérard Noiriel compares the documents used to
monitor these refugees to the “feuilles de route” (travel warrants) that military personnel had to procure before going on
leave. The travel warrants outlined a detailed itinerary with specific destinations and allocated only small food allowances
to them at each stopover point. According to Noiriel, these “instruments essentiels de surveillance” aimed to prevent the
141
205
papers, which became surveillance instruments that the state used to closely monitor the harkis’
movements. Each time that they would subsequently apply for social welfare benefits and statesponsored financial assistance specifically for the harki population they would need to procure these
documents to the Departmental Repatriates Office or the Departmental Office Nationale des
Anciens Combattants et Victimes de Guerre (ONAC) in the locale where they currently resided.
From the time that the harkis presented themselves to military officials in Algeria in hopes of
state-sponsored repatriation to France to their arrival at the camp when they needed to produce
documentation to prove that they were indeed harkis, therefore, they were the subjects of what
historian Gérard Noiriel describes as a two-fold process of categorization. First, there was “une
opération taxonomique,” by which the state created, and refined, the post-imperial category “harki,”
which differed from the imperial definition of (almost exclusively) male auxiliary soldiers in harka
units. The government enlarged the category to include these men’s immediate family members,
other types of auxiliary forces, and even those who did not serve the French Army government, but
whose lives were threatened because their compatriots deemed them as pro-French. And secondly,
“une opération d’identification” placed certain individuals into this category and excluded others.142
This process asserted the state’s power over the harkis through its local agents’ power to apply the
categorization. For example, to receive a carnet familial de rapatrié like the one pictured above, those
wishing to migrate to France underwent an investigation by French soldiers in Algeria to determine
whether they were indeed former harkis or former members of other auxiliary units. Concerned
about non-harkis using French naval vessels to gain passage to France in the midst of the massive
influx of pieds-noirs and harkis and the turbulent political end of the war, Messmer gave Army
officials in Algeria the following orders: “Ce que l’on peut faire dans l’immédiat c’est apporter le
refugees from deviating from their prescribed route. The harkis’ train vouchers seem to follow the same logic of
ensuring that they reached their intended destination. Noiriel, Réfugiés et sans-papiers, 52-53.
142 Noiriel, “Représentation Nationale et Catégories Sociales,” 265.
206
maximum de soins aux enquêtes pour éviter qu’il ne se glisse dans les rangs des réfugiés des
indésirables ou des Algériens seulement en quête de travail.”143 Therefore, the investigations and
resulting documents represent the initial phase of a selection process to establish the metropolitan
category of “harki,” which operated through the principle of encadrement. Once the harkis arrived in
metropolitan France, these documents bearing signatures and stamps from military officials in
Algeria became a tool that the Rivesaltes employees used to separate out those who had served
under the French flag from “faux harkis,” a distinction that was complicated for those who fled
Algeria in haste and could not access their identity papers. Through this separation, local agents
implemented national-level directives, which continued to define, and refine, the social category
“harki” and how it fit into the broader category of repatriates.
As discussed earlier in this chapter, high-level government officials maintained that harki
repatriates, unlike their counterparts with European origins, necessitated the encadrement only
available at the camps. This view regarding the exigency for encadrement, therefore, served as a source
of division between the harkis and the pieds-noirs. The camps, however, identified harkis into a de
facto sub-class of repatriates, which closely resembled refugees. The pieds-noirs avoided relegation
to camps, which—by their very geography, sheer isolation from the surrounding community, and
encadrement—imposed exclusion into the definition of “harki.”
Community of Exile: An Isolated Population
In contrast to many of the Algerian labor migrants who flowed to and from France during
the interwar period and the Trente Glorieuses in a “va et vient” fashion (which sociologist
SHAT, 1K 744 (dossier: Directives concernant les harkis), “Accueil en France des anciens supplétifs et des Algériens
menacés,” Apr. 26, 1963.
143
207
Abdelmalek Sayad has termed “le premier âge de l’émigration”144), the harkis flowed into France but
rarely returned to Algeria.145 As discussed in chapter two, if the harkis returned to their homeland
they risked putting their lives into mortal danger because they were sought after by FLN members
who believed them to be traitors to their native Algeria, which all but erased this option. Harki
citizens’ isolation from the French nation, therefore, embodied a distinct, additional layer than that
experienced by other Algerians who migrated to France during the 1960s. Algerian labor migrants
were free to cross the Mediterranean to visit their families and return home without facing the death
and torture to which tens of thousands of harkis were subject in the aftermath of the Algerian War
for Independence.
While the 20,000 harkis who passed through the Rivesaltes camp came into contact with a
great number of harkis speaking multiple Arabic and Berber dialects, they had little association with
their neighbors. The camp had a revolving door of new arrivals and departures: on average, each
week between December 1, 1962 and December 28, 1963, 207 harkis arrived at the camp and 280
harkis departed from the camp (see figure 23).146 The transient population formed a community of
exile isolated from other French citizens. Even though individuals needed to overcome linguistic
barriers to communicate with the neighboring French-speaking community, in reality the state’s
policies inhibited the very possibility for them to interact with other French citizens, thereby
relegating language to an ancillary reason for their isolation. Harki citizens’ exile at the Rivesaltes
camp was primarily due to the heavy-handed state policies—limiting both their movements outside
Abdelmalek Sayad, “Les trois âges de l’émigration algérienne en France,” Actes de la recherche en sciences sociales 15 (Jun.
1977): 59-79.
145 Emmanuel Ma Mung, Kamel Dorai, Frantz Loyer, Marie-Antoinette Hily, “Bilan des travaux sur la circulation
migratoire,” Migrations Études 84 (Dec. 1998): 2. “Noria” is often used to describe this same phenomenon, though, as
sociologist Emmanuelle Saada cautions, this “image of the ‘noria’—of immigration as a circle or wheel—…persists as a
comfortable if erroneous representation of Algerian immigration long after its basic constituents have changed.”
Emmanuelle Saada, “Abdelmalek Sayad and the Double Absence: Toward a Total Sociology of Immigration,” French
Politics, Culture & Society 18, no. 1 (2000): 35.
146 These figures were calculated using weekly bulletins from Dec. 1-8, 1962 to Dec. 22-28, 1963 sent from camp
officials to Ministry of Armies officials in Paris. SHAT, 1R 336/8 and SHAT, 1R 337/1.
144
208
of the camp and outsiders’ entrance into the camps—and to the strict controls that the government
placed on divulging to the public information about their life at the camp. This situation provoked
myriad questions about their place in French society for harkis themselves, for the French public,
and for the French government.
Weekly Arrivals and Departures of Harkis
Rivesaltes Camp, December 1, 1962 - December 28, 1963
1200
Number of
Arrivals
Number of
Departures
1000
800
600
400
200
0
Figure 23: Number of weekly arrivals and departures of harkis to and from the Rivesaltes camp
147
from December 1, 1962 to December 28, 1963.
Ostensibly, putting the harkis in isolated camps with tight surveillance intended to protect
them from vengeful FLN members who deemed their compatriots who had joined the French cause,
whether in the colony or in the metropole, traitors.148 Nevertheless, in its efforts to restrict the
relationship between these two groups, the government in actuality placed the harkis into a category
This chart reflects each week from Dec. 1-8, 1962 to Dec. 22-28, 1963 except Apr. 28-May 5, 1963 and Dec. 8-14,
1963 since these telegrams could not be located in either carton at the SHAT. The spike in the number of departures
during the week of Sept. 1-7, 1963 is due to the 858 harkis who migrated to the camp’s newly-opened village civil. The
creation of the village civil is discussed later in this chapter. Source: SHAT 1R 336/8 and SHAT 1R 337/1, Comptes
rendus hebdomadaires (weekly memos), Dec. 1-8, 1962 to Dec. 22-28, 1963.
148 In his memoirs, former FLN leader Ali Haroun writes critically of the Force de police auxiliaire (FPA), Algerian
natives who served in a special Parisian police brigade with the goal of dismantling the FLN in France under the
authority of Prefect Maurice Papon. According to Haroun, these men, to whom he and others refer as “harkis,” “ont
commis des actes ignobles” during the war, including torturing FLN supporters and aiding the Paris police in the
October 17, 1961 battle of Paris. Ali Haroun, La 7e Willaya: La guerre du FLN en France, 1954-1962 (Paris: Seuil, 1986),
402.
147
209
over which the state exercised strict surveillance. The Service de coordination et d’information nordafricaine (SCINA), an arm of the Ministry of the Interior created at the height of the Algerian War
for Independence in 1958 to monitor “North African” immigrants in the metropole, issued daily
reports that described in eerily-close detail their movements and actions.149 Including the harkis in
these police bulletins to monitor their interactions with FLN members led to the arguably soughtafter objective of increasing police supervision of the harkis. The population was now lumped into
the broader—undesirable—category of North African immigrants,” which distanced them from the
category of “repatriates.”150
Five SCINA reports in November and December 1962 referenced interactions between the
FLN and the Rivesaltes-dwelling harkis; four of these attest to problems that arose when FLN
members permeated the camp’s boundaries. For instance, the December 12 bulletin includes a
report, in a rather aggravated tone, by the local military police about FLN members who arrived in
France “sous prétexte de chercher du travail, [et qui] poussent l’audace jusqu’à venir au camp
contacter leur famille.”151 On November 29 several Rivesaltes harkis identified as such two of these
FLN agents (whose full names and place of residence in Algeria were given in the report) and,
consequently, they were nearly lynched.152 The report also indicated that “les réfugiés de Rivesaltes”
had further considered creating clandestine vigilante organizations to protect themselves and their
“North African” was in reality a code word for “Algerians.” For a discussion of this conflation, see: Byrnes, “French
Like Us?,” 24-25.
150 The SCINA reports from June 1962 to Dec. 1963—which were daily initially but tapered off to weekly in late-1963—
were between four and eight pages in length. They contained precise details about Algerian terrorist activities, arrests,
and violent attacks; population movements (including figures of how many Algerians entered and left France by boat
and by airplane daily or weekly); the number Algerians requesting and granted French citizenship; and Algerian political
parties and associations deemed threats to French security. Information about the harkis was included under the rubrics
“divers,” “harkis,” or “réfugiés” (first used on Jan. 15, 1963), without any seemingly logical reason for the repartition
among different sections, other than it was difficult to classify the harkis. Though never used as a subject heading, the
harkis were even referred to as “rapatriés” on a few occasions (see, for example: SCINA n˚ 1 784, Nov. 3-5, 1962, 6). A
full of record these bulletins d’information from this period can be found in the Bouches-du-Rhône Departmental Archives
(138 W 3-8).
151 ADBR, 138 W 5, SCINA, “Synthèse..., journée du 12 décembre 1962,” d, N˚ 1 810.
152 Ibid.
149
210
families.153 Dangerous contacts with FLN members also took place when the harkis strayed outside
the camp. In the November 20 bulletin, the gendarmerie reported that local FLN operatives aimed
to approach all “North Africans” from the Rivesaltes camp spotted in Perpignan to convince them
to join the metropolitan arm of this organization, the ADAF. Once contact was made, “Ils sont
envoyés dans une gargotte [sic] ‘Au bon accueil’ avec un mot de passe qui varie suivant leur position:
- ‘Je veux un couscous’, pour ceux qui acceptent de se rallier à la cause. - ‘Je veux une eau gazeuse’,
pour les hésitants qu’il faut ‘travailler’ encore, - ‘Je veux une bière’, pour les récalcitrants qu’il faut
ficher et éviter, ou éventuellement éliminer.”154 Evidence collected by the local military police also
reveals that some harkis were involved in pro-FLN activities. The same November 20 report alleged
that one harki used a car left for him near the camp by an FLN member to communicate camp
activities each night around midnight to the leaders of the Perpignan FLN office who waited for him
at a local bar, l’Entr’acte.155
The government’s response to these threatening interactions between harkis and FLN
members was to increase efforts to seal off the camp. On December 8, 1962, Prime Minister
Pompidou wrote to Minister of Armies Messmer that in order to avoid such incidents in the future,
“il conviendra de soumettre les allées et venues à une certaine surveillance, les sorties du camp ne
devant être autorisées que pour des motifs sérieux.”156 By trying to close the camp to FLN members
(albeit with mixed success), the government simultaneously limited the harkis’ contact with another
group of “outsiders,” that is, their neighbors.
However, minimizing such contacts lies in stark contrast to local government policies toward
the pieds-noirs arriving from Algeria. For example, in May 1962 with a horde of repatriates
debarking in Port-Vendres just south of Perpignan, Pyrénées-Orientales Prefect Pierre Dubois
Ibid.
ADBR, 138 W 5, SCINA, “Synthèse..., journée du 20 novembre 1962,” b, N˚ 1 795.
155 Ibid.
156 SHAT, 1K 774, Letter from Pompidou to Messmer, Dec. 8, 1962.
153
154
211
addressed a letter to a host of union leaders and local politicians inviting them the first meeting of
the Comité départemental d’Accueil. The Prefect saw this organization as a means of both ensuring
the collaboration of public services and strengthening the link “entre nos populations et les
rapatriés.”157 In order to properly welcome the pieds-noirs to the department, Dubois believed that
local officials had “le devoir de les aider à retrouver la sécurité matérielle et la tranquillité morale
dans un climat d’humaine compréhension.”158
Conversely, the local government did provide such a welcome for the harkis. Its actions
challenged the harkis’ membership in the commune. For the entire twenty-seven months that the
Rivesaltes camp was home to the harkis, there is only one reference to the population in the
monthly Rivesaltes Town Council meetings. On June 21, 1963, the director of the Ministry of
Repatriates branch office at the camp, René Aucante, was named “l’adjoint spécial de la section
‘Agglomération du Camp de Rivesaltes.’” 159 Subsequent meetings contained neither discussions
about the camp nor reports from Aucante, which, along with his title, suggests that local officials did
not consider the camp a part of their municipality.
The decisions of the mayor, Doctor Émile Parès, reinforced that the harkis were outsiders.
He refused to allocate local funds to compensate housing costs for the instructors who worked at
the camp’s school, to which they were legally entitled, because he did not regard the school for harki
children as a municipal establishment. Therefore, he reasoned, it should not draw on the Rivesaltes
town budget.160 The mayor’s response to a problem of trash removal from Guinean families living
on the campgrounds in 1966—they had arrived as the last harkis were vacating the premises in late
1964—reinforced his position about the harkis. The mayor insisted in a letter to a public health
ADPO, 104 W 6, Letter from Prefect Pierre Dubois, May 29, 1962.
Ibid.
159 Commune de Rivesaltes, Délibérations du Conseil Municipal de Rivesaltes, June 21, 1963.
160 AN, F1a 5141, Jacques Toutain, “Objet: Matériel scolaire du Groupe scolaire du centre d’hébergement de harkis de
Rivesaltes (P.O.),” Feb. 11, 1963.
157
158
212
official: “Le Camp ne concerne, en effet, que géographiquement notre commune. Il est et reste un
Camp militaire. L’affaire des Guinéens, comme celle des Harkis dépasse donc le cadre communal.
C’est un problème national.”161 With such actions and words, the local government clearly signaled
that the harkis fell outside of the communal affairs and, therefore, not within the Rivesaltes
community.
Similarly, the few interactions between the Rivesaltais and the harkis were often marked with
strife, echoing the tone set by the government. For example, on October 28, 1962, approximately
one hundred harkis and their wives, twenty-five of whom were registered to vote, walked to the
Rivesaltes town hall to cast their ballot for the referendum on voting for the Fifth Republic
President by universal suffrage. According to a report from the National Police, the group
encountered four locals who “[ont] provoqué une certaine effervescence parmi le groupe” and tried
to prevent them from voting.162 These actions demonstrate suspicion in regard to the new residents,
who were, on the one hand, just another transient immigrant population succeeding those that had
inhabited the camp since 1939 and, on the other, French citizens who shared the same civic rights as
the Rivesaltais.
The difficult memories of the recent political break-up of Algeria and France, the territories
and the populations, undoubtedly was a trigger for the townspeople’s actions. However, this
tumultuous interaction can also be attributed to the local population’s ignorance of who their new
neighbors were. The harkis’ lives at the Joffre camp were enshrouded in mystery owing to the
government’s refusal to permit media and civilians onto the campgrounds until a March 18, 1963
visit by a team of reporters from the regional newspaper L’Indépendant. Between March 22 and 28,
the pro-Gaullist daily, undoubtedly selected for its political leanings, published a series of five articles
about the harkis, who had hitherto been hidden behind the barbed wire encircling the camp.
161
162
ADPO 1419 W 109, Letter from Mayor Emile Parès to Madame Witté (Public Health Doctor), Feb. 3, 1966.
ADBR, 138 W 5, SCINA, “Synthèse..., journée du 30 octobre 1962,” 1, N˚ 1 781.
213
Newspaper officials, whose curiosity was “aroused” by the secrecy enveloping the camp in what was
now the second largest city in the department, had requested permission numerous times over five
months before Minister of Armies Messmer acquiesced.163 The harkis’ first months in the camp—
when they lived in tents, had no working showers, and suffered through a particularly harsh fall and
winter—would, therefore, remain invisible to their fellow citizens. Nevertheless, the author of the
third article offered an apology for the government’s actions to shield the harkis from the public eye
by pronouncing: “Aux flottements d’une arrivée massive avec, à l’époque, un avenir pas tout à fait
défini, les autorités ont préféré nous montrer les choses bien en place et un ciel bleu par-dessus le
camp. Ce n’est peut-être pas très journalistique, mais c’est logique. Et quand on a vu le travail
énorme effectué en quelques mois, on ne peut honnêtement en vouloir à ceux qui nous ont fait
attendre.”164 Camp officials’ stranglehold on the harkis’ lives, coupled with the authors’ “not too
journalistic” ways and Gaullist bias, skewed the entire series of articles. This kept the reality of their
situation—and the harkis themselves—hidden from public view, which was the government’s goal.
When the readers of L’Indépendant were finally allowed to peer in on their new neighbors, the
articles painted a Panglossian picture of the well-oiled machine and the pleasant space that camp
officials wanted other French citizens to see. The third article claimed that the public would now
have a positive vision of the Rivesaltes camp and “ne considérer[a] pas plus cet ensemble comme
une ville morte, vestige d’un passé assez proche, mais comme le commencement d’un futur.”165 The
articles indeed projected a positive present and future for the harkis, and the last in the series even
likened the harkis’ memory of their time at Rivesaltes to the memories that a child has of his or her
formative school years.166 Some of the harkis’ homes were described as exquisitely decorated with
ADPO, 1111 PER 73, “Le camp des déracinés de Rivesaltes avec ses 10 000 harkis est devenu la 2e ville des Pyr.-Or.,”
L’Indépendant, Mar. 24, 1963.
164 Ibid.
165 Ibid.
166 ADPO, 1111 PER 73, “Prêt pour la vie civile, le harki s’en va,” L’Indépendant, Mar. 28, 1963.
163
214
luxuries such as Persian rugs and copperplates on the walls. The journalists witnessed women
preparing meals of steak and fresh green beans in a “first-class” cassoulet sauce. They quoted a camp
official as saying: “Est-ce qu’ils ont l’air malheureux? Croyez-vous qu’en Algérie ils avaient de la
viande à tous les repas?”167
While acknowledging that most of the residents had large families, the articles make no
reference to the cramped conditions the harkis endured in their twenty to twenty-five square meter
units. Instead, the series emphasized the vastness of the camp and its similar geography to Algeria,
thus giving the impression that these individuals had an abundance of space that felt like home. For
example, as the third article in the series related, “Une vaste plaine uniforme qui ressemble comme
une sœur à celles d’Algérie: une plaine qui vient mourir sur les contreforts des Corbières rappelant
les montagnes de l’Aurès…” 168 While the articles do refer multiple times to the harkis’
“déracinement,” they attribute a palliative function to the physical space and order that Rivesaltes
offered. For instance, the same article claims, “Ils sont des ‘déracinés’, que la terre de Rivesaltes,
cette terre rouge perçant à travers les cailloux, cultivée au maximum par l’armée, va aider à prendre
racine et à s’épanouir de nouveau au soleil.”169
A strict level of organization “que seule l’Armée pouvait mettre en place” was necessary to
run the camp.170 The first article explained in detail the organization of the barracks into “quartiers”
and “villages” and each successive article, whether referring to the infirmary or the children’s
daycare facilities, used phrases such as “admirablement bien organisées.” 171 In fact, the space’s
categorization as a “camp” was put into question and effaced by the second article, which preferred
to call it a city: “Ce n’est plus un camp. C’est vraisemblablement une ville comme aucune en France,
ADPO, 1111 PER 73, “Nous sommes mieux que chez nous, disent les harkis (ET C’EST VRAI),” L’Indépendant, Mar.
22, 1963.
168 ADPO, 1111 PER 73, “Le camp des déracinés de Rivesaltes…,” L’Indépendant.
169 Ibid.
170 ADPO, 1111 PER 73, “Prêt pour la vie civile…,” L’Indépendant.
171 ADPO, 1111 PER 73, “À Rivesaltes, l’école est un vrai plaisir pour les enfants de harkis tandis que leurs mères
découvrent les humbles joies de coudre et de tricoter,” L’Indépendant, Mar. 23, 1963.
167
215
et pour une même population aucune n’a une aussi belle organisation sociale.”172 The following
article in the series affirmed this classification by its very title, “Le camp des déracinés de Rivesaltes
avec ses 10 000 harkis est devenu la 2e ville des Pyr.-Or.”173
In addition to questioning what kind of space the camp constitutes, the articles attribute a
murky status to the harkis. The articles continually posit that the harkis feel a strong link to their
adopted country and that they unquestionably consider themselves French, as one article avers: “Et
vous demandez aux [harkis eux-mêmes] de quelle nationalité ils sont, ils se mettront presque au
garde-à-vous et dans leurs yeux vous livrez un reproche. Sèchement, avec fierté, ils répondront: ‘Je
suis Français’, comme une chose évidente.” 174 While peppering the articles with the harkis’
affirmations of being French, and even advancing the statistic that only forty of the ten thousand
harkis who appeared before the magistrate refused French nationality, the newspaper puts into
question their being French by continually emphasizing that the camp is a space where “les harkis
apprennent la vie à l’européenne,” to cite part of the fourth article’s title. According to this piece,
here they become “civilized” and learn to forego previous Muslim practices such as polygamy.175 In
addition to raising doubts about the harkis’ Frenchness, the articles question the harkis’ status as
repatriates. This word appears just once in the entire series, but then is immediately qualified with
“expatriés volontaires, faudrait-il dire.”176
Besides qualitative positive evaluations that concealed the disorientation that camp dwellers
have expressed in recently published memoirs and interviews, the articles contradicted the scores of
official documents written while the camp was open about hygiene problems.177 For example, the
Ibid.
ADPO, 1111 PER 73, “Le camp des déracinés de Rivesaltes…,” L’Indépendant.
174 Ibid.
175 ADPO, 1111 PER 73, “En un mois : 58 naissances, 3 décès. Bilan d’un camp où les harkis apprennent la vie à
l’européenne,” L’Indépendant, Mar. 25, 1963. The term “civilisé” appears in the Mar. 24, 1963 article.
176 ADPO, 1111 PER 73, “Le camp des déracinés de Rivesaltes…,” L’Indépendant.
177 Harki daughters Fatima Besnaci-Lancou and Dalila Kerchouche recount their and their family’s memories of the
Rivesaltes camp in Besnaci-Lancou, Fille de harki and Kerchouche, Mon père, ce Harki. Both women have also published
172
173
216
first article, entitled “Nous sommes mieux que chez nous, disent les harkis (ET C’EST VRAI),”
opened with an image contradicting the scores of letters written at the same time by government
officials about the camp’s hygiene problems: “‘Rien que pour ramasser les ordures, le matin, il faut
une véritable organisation, un bataillon de la salubrité’ nous a-t-on dit. Cette phrase est à l’échelle de
la ville-champignon que constitue le camp de Rivesaltes.”178 The fourth article similarly claimed that
sanitary conditions were excellent and that cases of illnesses indigenous to “pays chauds,” such as
trachoma (which is caused by the sexually-transmitted disease Chlamydia), were becoming less and
less common. 179 The article further emphasized that all harkis were x-rayed as a preventative
measure and that “[d]ès qu’un cas paraît suspect, les médecins préviennent le Dispensaire de
Perpignan qui prend le malade en charge.”180 As cited above, sealed government documents written
at precisely the same time demonstrate that the tuberculosis epidemic greatly concerned camp health
officials who encountered difficulties ensuring that the harkis sought out proper treatments.
The policies that national government officials created and Rivesaltes camp agents enacted
to strictly control the harkis’ daily life, therefore, masked their situation from public view and
inhibited the population from interacting with their Rivesaltais neighbors. In turn, local officials
declared the camp as outside the boundaries of their municipality, just as they and their predecessors
had done while the camp housed refugees and prisoners. By isolating the harkis at the Rivesaltes
camp from other French citizens, the post-imperial government unequivocally signaled to the harkis
themselves and to the French public that these Arabs and Berbers were neither repatriates nor
French citizens like the pieds-noirs.
interviews with former camp dwellers, which give texture to the population’s disorientation at Rivesaltes: BesnaciLancou, Nos mères, paroles blessés and Kerchouche, Destins de harkis (Paris: Autrement, 2003).
178 ADPO, 1111 PER 73, “Nous sommes mieux…,” L’Indépendant.
179 The reference to diseases indigenous to “pays chauds” recalls Montequieu’s theory of climates whereby he contends
that populations in “southern countries,” such as Algeria, are labeled as physically weaker than and intellectually inferior
to those from “northern countries,” such as France. Baron Charles de Montesquieu, De l’esprit des lois, 1748.
180 ADPO, 1111 PER 73, “En un mois…,” L’Indépendant.
217
Social Advancement: Toward the Reclassement of the Harkis
Despite the many limitations of the Rivesaltes camp—which, as indicated in the epigraph to
this chapter, Messmer summarized as “marqué de la tare de tout ‘univers concentrationnaire’”—the
government nevertheless upheld the view that it was a necessary intermediary step on the way
toward their reclassement.181 In fact, in the same letter, Messmer justified the camp’s existence solely
because of the opportunity it afforded the harkis to gain desperately lacking job skills: “Le séjour
dans les camps ne se justifie que dans la mesure où il permet de donner à ceux qui en sont capables
des rudiments de formation professionnelle.” 182 Transforming the harkis into citizens able to
contribute to metropolitan France’s economy recalls the training of the Compagnies de Travailleurs
Etrangers who were among the first groups to inhabit the camp in 1939.
To facilitate the harkis’ reclassement, military and civilian state representatives at the Rivesaltes
camp carried out a program of “Action de Promotion Sociale,” or social advancement. Messmer
claimed on September 24, 1962 that reclassement efforts were necessary because: “Les ex-supplétifs et
leurs familles proviennent du ‘bled’ algérien; leurs conditions traditionnelles de vie constituent pour
beaucoup un très grave obstacle à un reclassement aussi bien d’ailleurs dans les campagnes que dans
les villes françaises.”183 The experience gained from the first three months at the Bourg-Lastic and
Larzac camps had demonstrated that only a small number of harkis were “suffisamment évolué” to
immediately find work and could not even adapt to working in a rural setting. A slightly higher
number could be placed into jobs after a few weeks of job training classes, but the vast majority
needed to participate in a social advancement program for several months before they would be
SHAT, 10T 549/3b, Minister of Armies Messmer to Minister of Repatriates Missoffe, “Objet: Problème des exsupplétifs,” 2, Mar. 2, 1963.
182 Ibid.
183 SHAT, 23R 16/2, Minister of Armies to Minister of Repatriates, (signed by Bernard Tricot, Administrative General
Secretary), “Objet: Action de Promotion Sociale dans les deux nouveaux Camps d’ex-supplétifs musulmans en vue de
leur reclassement,” 1, N˚ 2.264 MA/DSF/1, Sept. 24, 1962.
181
218
qualified to seek jobs.184 Messmer underscored the urgency of this program: “si l’on veut éviter soit
de perpétuer en France l’existence de ‘camps palestiniens’, soit de voir dans les zônes rurales et
urbaines, où seront ultérieurement placés ces ex-supplétifs, se recréer des ‘bidonvilles’: ces ‘rapatriés’
y feraient de ‘parias’ mis à l’écart, avec leurs nombreux enfants, par tous les milieux.”185 Therefore,
the objective of the social advancement program at the Rivesaltes camp was to create well-adjusted
metropolitan citizens out of these less “evolved” harkis from the “‘bled’ algérien.” The government
wanted to avoid the fate of other immigrants in France who languished in communitarian groups in
shantytowns on the outskirts of France’s major cities.186
The underlying principal, indeed the motor, of social advancement was the deployment of
appropriate personnel to teach and encadrer the Rivesaltes-dwelling harkis. As Messmer claimed, “la
réussite de l’action de Promotion Sociale entreprise repose avant tout sur la valeur des Cadres qui en
seront chargés.”187 The first director of the social advancement program, General De Segonzac,
issued a directive in October 1962 outlining how the program would function. The body of this sixpage memo did not begin with a description of what activities and classes would be offered, but
rather with a description of the personnel who would encadrer the harkis. 188 De Segonzac
underscored in the memo’s conclusion that the program’s success hinged on the state
representatives keeping in constant close contact with the military camp commander and the head of
the camp’s Ministry of Repatriates’ office, that is, exercising a “technology of security” over the
harkis.189 The organization chart of the personnel employed for social advancement demonstrates an
Ibid., 1-2.
Ibid., 2.
186 For a discussion of immigrant bidonvilles in France, see Blanc-Chaléard, Les Italiens dans l’Est parisien; Byrnes, “French
Like Us?”; Natacha Lillo, La Petite Espagne de la Plaine-Saint-Denis, 1900-1980 (Paris: Autrement, 2004); Abdelmalek Sayad
and Eliane Dupuis, Un Nanterre algérien: terre de bidonvilles (Paris: Autrement, 1995); and Marie-Christine VolvovitchTavares, Portugais à Champigny: Le temps des baraques (Paris: Autrement, 1995).
187 SHAT, 23R 16/2, Minister of Armies, “Objet: Action de Promotion Sociale…,” 4.
188 SHAT, 14T 92, General de Segonzac, Directive N˚ 1, “Objet: Action de promotion sociale en faveur des exsupplétifs musulmans regroupés au Camp de RIVESALTES,” 1, N˚ 617 EMA/BIAE, Oct. 5, 1962.
189 Ibid., 6.
184
185
219
intricate web of inspectors, military officers, and secretaries in the central officers, and either nine or
ten on-site monitors for each one-hundred person foyer de jeunes, foyer jeunesses féminines, female adults’
section feminine de progrès and male adults’ centre de préformation.190
Social advancement was organized around the individual’s gender and age. First, school-aged
children under twelve years old attended schools set up on the camp grounds with other harki
children. In March 1963 there were forty-four classes staffed by fifty-two instructors, twenty-three
of whom were military men and twenty-nine of whom were civilians.191 The classes foremost had the
goal of teaching these young children how to speak, read, and write French, yet they also provided a
space for leisure and physical activities in the afternoons. Between the inception of the camp’s social
advancement programs in December 1962 and February 1964, the government furnished 2,300
athletic outfits, 2,465 pairs of athletic shoes, and 130 balls of all sorts (for example, soccer, handball,
and volleyball) for use by harki children.192 Weekly attendance tallies from February 1963 until
summer vacation began on June 30, 1963 hovered around 1,300 harki children. Nevertheless,
between one week and the next, attendance varied by as many as 172 students, reflecting the
transitory nature of the camp’s population.193 During summer 1963, the 1,200 school-aged children
continued leisure activities at an on-site summer camp where they played Monopoly, the traditional
SHAT, 1R 336/8 (dossier: Promotion Sociale, Reclassements), General de Segonzac (Délégué du Ministre des
Armées pour les Centres de Formation Professionnelle et les Centres de Jeunes à Monsieur le Ministre des Armées),
“Objet: Action de promotion sociale en vue du reclassement à l’intérieur des deux nouveaux camps de rassemblement de
supplétifs Musulmans,” Sous-dossier A, Pièce N˚ 3 and Pièce N˚ 5, N˚ 550 EMA.BIAE, Sept. 4, 1962.
191 SHAT, 1R 337/1, General Houssay to the Minister of Armies, E.M.A.T. – 1er Bureau, “Objet: Scolarisation des
enfants d’ex-supplétifs des Camps de RIVESALTES et de SAINT-MAURICE L’ARDOISE,” Annexe, N˚ 5221 EFF/1,
May 7, 1963.
192 AN, F1a 5140 (dossier: Gestion du camp de Rivesaltes, 1963-1966), 9th Military Region, Pyrénées-Orientales Military
Subdivision, 40˚ Compagnie de Camp, Center for Social Advancement, “Liste du matériel mis en place au centre de
promotion sociale, à sa création et existant à la date du 20 février 1964,” 4-5.
193 During the week of Mar. 17-23, 1963, 1,200 students attended school, whereas during the week of Mar. 24-30, 1963,
1,372 students were in attendance. The attendance figures can be found in the weekly bulletins cited above. SHAT, 1R
336/8 (dossier: Messages de Rivesaltes et Saint-Maurice-l’Ardoise) and SHAT, 1R 337/1 (dossier: Camp de Rivesaltes et
SMA).
190
220
local game of pétanque, volleyball, soccer, basketball, and dominoes.194 By remaining on the grounds
they continued to be divided from neighboring children.
Eight foyers de jeunes, also set up on the campgrounds, were created for children age twelve to
twenty. The boys’ foyers offered literacy classes, remedial school courses, “adaptation à la vie
métropolitaine” workshops, physical education, and leisure activities three-quarters of the day and
pre-professional classes for one-quarter of the day. While camp officials expected eight hundred
boys to participate in the activities offered by the eight foyers, only fifty-four students had shown up
by late January 1963.195 Young girls participated in similar foyers except the pre-professional classes at
their foyers jeunesses féminines were replaced by a “formation familiale et menagerie,” which they
attended with the harki women in the afternoons.
The classes at the eight centres féminins (also called sections féminines de progrès) focused on
“adaptation à la vie métropolitaine,” childcare, cooking, home economics, sewing, housekeeping,
and language skills. Daycare services were provided for infant and toddler children so that the
women could attend these afternoon courses, keeping with traditional gender roles. The camp
recruited instructors among metropolitan graduates of the Centre d’Entraînement des Monitrices de
la Jeunesse Algérienne de Nantes (CEMJ), former social monitors returning from Algeria, and
former Assistantes sanitaire et sociale rurale auxiliaire (ASSRA) who had worked as social assistants
in the Algerian countryside.196 Between February 16, 1963 and June 30, 1963 when three of the eight
foyers closed because some of the counselors migrated with the harki families to the newly-opened
forest hamlets, an average of 454 women and female adolescents attended classes at the centres
SHAT, 14T 92, Colonel GALEY, Commander of the Rivesaltes Camp and the 40ème Compagnie de Camp to the
Director of the 9th Military Region Supplies Corps, “Objet: Demande d’autoristation d’achat,” N˚ 1390/MAJ/MI, June
17, 1963.
195 CAC, 19920149/3 (dossier 6: Formation prof des harkis, avril-août 63), “Procès-verbal de la réunion tenue le 24
janvier 1963 dans le cabinet de M. Besson,” Jan. 25, 1963.
196 SHAT, 23R 16/2, Minister of Armies, “Objet: Action de Promotion Sociale…,” 3.
194
221
féminins.197 These venues had the dual goal of helping them to apply for admission to metropolitan
schools for nursing, daycare, and other domestic jobs, and to help them to adapt to metropolitan
ways of hygiene, child rearing, and housekeeping.198 The list of materials used at the centers includes
a wide assortment of cooking utensils, boilers for washing laundry, ironing boards and irons, sewing
machines, and plastic bathtubs.199
Meanwhile, adult men between twenty and thirty years old divided their time between
“Initiation à la vie métropolitaine” courses similar to the classes adolescent and female adult harkis
attended and remedial job-training classes at the centres de préformation professionnelle located in Block K
of the camp. A report issued by the Ministry of Interior explained that the former classes, like those
North African immigrants arriving in France during the colonial period participated in,200 aimed to
“donner une connaissance plus approfondie de la langue française et de nos mœurs” to this
population believed only to have “une connaissance très superficielle” of such qualities necessary for
their successful integration into French society.201 Instructors who had been trained at the Centre
d’Entraînement des Moniteurs de Jeunesse d’Issoire in the Puy-de-Dôme department, which had
been opened during the Algerian War for Independence to train French Muslim military personnel,
taught the harki men about French culture and history, how to adapt to “modern life” (including
using factory equipment) and how to read, write, and speak French.
At the centres de préformation professionnelle, harki men also attended remedial job-training classes
specializing in construction or in metal works, which had opened in early December 1962.
This average was calculated using the weekly bulletins located in SHAT, 1R 336/8 (dossier: Messages de Rivesaltes et
Saint-Maurice-l’Ardoise) and SHAT, 1R 337/1 (dossier: Camp de Rivesaltes et SMA).
198 SHAT, 1R 336/8 (dossier: Promotion Sociale, Reclassements), General de Segonzac (Délégué du Ministre des
Armées pour les Centres de Formation Professionnelle et les Centres de Jeunes à Monsieur le Ministre des Armées),
“Objet: Action de promotion sociale en vue du reclassement à l’intérieur des deux nouveaux camps de rassemblement de
supplétifs Musulmans,” Sous-dossier A, Pièce N˚ 2, N˚ 550 EMA.BIAE, Sept. 4, 1962.
199 AN, F1a 5140, 40˚ Compagnie de Camp, Centre de Promotion Sociale, “Liste du matériel…,” 5-6.
200 For a description of the classes offered to Algerian immigrants in the post-Second World War period, see Lyons,
“Invisible Immigrants,” 111-115.
201 AN, 5 AG 1/22, Minister of the Interior, “Rapport sur l’application de la loi du 26 décembre 1961 transmis au
Secrétaire général de la Présidence de la République,” 104, Jul. 26, 1965.
197
222
Instructors for this program hailed from the Centre Militaire de Formation Professionnelle (CMFP)
in Fontenay-le-Comte, which served to train French Muslim soldiers returning to civilian life during
the Algerian War for Independence. In January 1963, approximately four-hundred men had signed
up for the construction classes at the four centres de préformation bâtiment (though only three-hundred
regularly showed up) and fifty men attended the metal works courses at the three centres de
préformation métaux.202 Between February 2, 1963 and June 28, 1963, an average of 564 men per week
attended classes at both centers.203 However, government officials questioned their attendance at and
devotion to the classes, claiming that “leur zèle à s’instruire laisse souvent à désirer.”204
When the government put the social advancement program into place in the fall of 1962, it
anticipated that the remedial job classes and initiation to metropolitan life classes would adequately
prepare the harkis for entrance into full-time job training at the centres de formation professionnelle accélérée
located in blocks O and Q of the camp or for employment in the local Pyrénées-Orientales
construction sector of the economy. However, most harkis failed the psychotechnical entrance tests
evaluating their mental aptitude, their psychomotor skills, and their sensory functions that were
necessary for admission to the centres normaux de formation professionnelle accélérée run by the Ministry of
Labor. Of 393 harkis who took these tests in July 1963 upon their completion of the remedial jobtraining program, only 20 proved fit for the centres normaux de formation professionnelle accélérée. 205
Therefore, Minister of Repatriates Missoffe proposed opening centres spéciaux de formation professionnelle
accélérée also run by the Ministry of Labor to provide the harkis special training apart from the general
CAC, 19920149/3 (dossier 6: Formation professionnelle des harkis, avril-août 1963), Ministry of Repatriates, SFIM,
“Procès-verbal de la réunion tenue le 24 janvier 1963 dans le cabinet de M. Besson,” Jan. 25, 1963.
203 This average was calculated using figures in the weekly bulletins located in SHAT, 1R 336/8 (dossier: Messages de
Rivesaltes et Saint-Maurice-l’Ardoise) and SHAT, 1R 337/1 (dossier: Camp de Rivesaltes et SMA).
204 ACNMF, 15/13, Comité National pour les Musulmans Français, “Aide au logement des réfugiés musulmans,” 1, Oct.
10, 1963. This sentence is underlined in the original document.
205 CAC, 19920149/3 (dossier 6: Formation professionnelle des harkis, avril-août 1963), Ministry of Repatriates to the
Minister of Labor, “Objet: Formation professionnelle des harkis. Création de centres à RIVESALTES et LA RYE,” 4,
undated. The information contained in this letter leads the author to conclude that it was written between June 27 and
July 27, 1963.
202
223
population with whom they were not considered to be on par.206 Fifteen centers of fifteen to
seventeen students opened on September 1, 1963 in block K.207 After a six-month internship, the
graduates would be eligible to earn a Certificat d’Initiation de Construction, which would hopefully
aid their chances for employment. Ministry of Labor employees staffed the classes, though the
Ministry of Repatriates managed the facility.
The harkis’ participation in the classes offered at the centres spéciaux de formation professionnelle
accélérée was contradictorily considered a form of reclassement, even though they remained on the
camp grounds. In the midst of a general housing crisis, which was particularly severe in the
Pyrénées-Orientales and in neighboring departments, Missoffe opted to open a “civilian enclave” in
block F. He claimed that this solution would provide the benefits of not requiring the families to
move from the camp and of allowing their children to remain enrolled in the camp schools and
remedial job training classes, which he averred “fonctionnent dans d’excellentes conditions.”208 The
equation of reclassement with moving to the camp’s civilian enclave is reflected in the weekly memo
for September 1 to 7, 1963, which stated “Effectifs Personnes ayant quitté le camp suite
reclassement de chef de famille 1058 dont 866 ont été affectées village civil.”209 Even though the
government considered that these individuals were “reclassés,” it contradicted its own definition of
reclassement since the harkis continued to live at the Rivesaltes camp and they did not have jobs.
While the civilian enclave was indeed funded by and under the authority of the Ministry of
Repatriates, the military maintained the function of furnishing encadrement over the harkis. According
to SFIM director Yves Pérony, the military was tasked with “veiller au maintien de l’ordre, à la
CAC, 19920149/3 (dossier 6: Formation professionnelle des harkis, avril-août 1963), Ministry of Repatriates to the
Inspectors of the Chantiers de Forestage, “Objet: Formation Professionnelle des supplétifs musulmans réfugiés en
France,” 1, N˚ 1916/SFIM/MG, July 12, 1963.
207 ACNMF, 32/3, Pierre Durney, Civil Services Inspector, “Mission du 8 Août 1963 au Camp de LA RYE (Vienne) et
du 12 Août 1963 au Camp de RIVESALTES (P.O.).”
208 SHAT, 1R 337/1, Minister of Repatriates to the Cabinet of the Minister of Armies, “Objet: Hébergement provisoire
de 500 familles de harkis dans une partie du camp de RIVESALTES,” 2, N˚ 1078/SFIM/DIR, Apr. 23, 1963.
209 The spike in the number of departures is visible on the graph of the weekly statistics of harkis arriving at and
departing from the Rivesaltes camp after the reclassement of the head of the family depicted in figure 23 above.
206
224
salubrité et à la sécurité.”210 This camp within the camp comprised fifty barracks of seven families,
two barracks of shops with foodstuffs since these harkis were no longer eligible for provisions from
the Ministry of Repatriates, and one barrack with a clothing and supplies store. By October 15, 1963,
1,469 men, women, and children resided in this new section of the camp while waiting for jobs and
housing outside of the camp.211
Conclusion: Leaving the Rivesaltes “Univers Concentrationnaire”
Given the long-term repercussions of putting harkis in camps, the Rivesaltes camp must be
viewed through a more complex lens than just a physical space to park an exiled population. As
contended Minister of Armies Messmer—the government official with the ultimate responsibility
for the harkis’ living conditions at the camp—Rivesaltes demonstrated the same characteristics as
the Second World War concentration camp universe through its physical structures and its
psychological demoralization of its residents, who remained isolated from the local population. The
camp thus was a unique physical hurdle to their belonging to the French nation, a hurdle not faced
by other populations migrating from Algeria. The camp further served as a psychological obstacle to
the harkis’ membership in their local and national communities by treating them like refugees.
The decision to relegate the harkis to this space also opened the door to future placements in
other refugee camps. Leaving Rivesaltes did not ensure that the harkis would leave behind their
exilic refugee camp life. A small number would return to Rivesaltes, sometimes as many as four
times before definitively leaving the camp.212 Some proved unable or unwilling to perform their jobs,
SHAT, 14T 92, Minister of Repatriates (signed by Yves Pérony) to the Minister of Armies, “Objet: Centre de
Formation Professionnelle – Aménagements à réaliser au Camp de RIVESALTES,” 2, N˚ 2375 SFIM/MG, Aug. 13,
1963.
211 SHAT, 1R 337/1, “Situation des effectifs à la date du 15 octobre 1963.”
212 CAC, 19970146/2 (dossier: R. Bourgat), R. Bourgat to the Prefect of the Pyrénées-Orientales, “Objet: Situation des
‘Harki’ [sic] recasé dans les Pyrénées-Orientales,” 2, N 835/RB/Rb/3, June 27, 1963. The dossier “Retours au camp” in
AN, F1a 5142 contains a series of documents from Oct. 1963 to Feb. 1964 written monthly by René Aucante entitled
210
225
sometimes getting into disputes with other harkis or their supervisors at the forest hamlets. Others
encountered problems with their housing. And still others feared for their safety living near
Algerians in their new communities; one harki reported that an ADAF member warned him: “la
guerre était finie en Algérie, mais pas ici.”213 Starting in January 1963, those who were deemed as
incasables or irrecasables—most often widows, the ill, and elderly people—were transferred to the Bias
camp (also known as “La Cité d’Accueil des Rapatriés Algériens” or CARA) in the nearby Lot-etGaronne department, which had most recently sheltered refugees from French Indochina.214 The
day after the Rivesaltes camp was liquidated, on January 1, 1965, la Cité d’Accueil et d’Hébergement
opened on the site of the former Saint-Maurice-l’Ardoise camp for the 760 remaining “unhouseable”
harkis.215 According to a report submitted by the Minister of the Interior to the General Secretary of
the Parliament in July 1965, “ces milieux très fermés sur le monde extérieur” housed 1,800 harkis.216
Saint-Maurice-l’Ardoise remained open until December 31, 1976, and the site of the Bias camp
(officially closed in 1983) contained two complexes exclusively housing harkis today.217
The harkis who departed the Rivesaltes camp for destinations across metropolitan France
carried their nationality declaration, social security card, train voucher, certificates of completion of
job training programs, and a letter from René Aucante to the Prefect of the harki’s new home
department. This form letter’s first sentence underlined that because the harki opted for French
nationality, he must be considered a “citoyen français.” However, Aucante conceded that problems
could very well arise for the newly-minted citizen: “…En raison de son inadaptation à peu près
“État des ex-harkis de retour au camp.” During this period, an average of fourteen harki males, some of whom had a
wife and children, returned to the Rivesaltes camp each month.
213 ADBR, 138 W 7, SCINA, “Synthèse..., journée du 30 mai 1963,” 3, N˚ 1 908.
214 For a study of the Bias camp, see Aude Lanoizelez, “La Cité d’Accueil de Rapatriés Algériens (CARA) de Bias, du
camp au ghetto: socio-histoire d’un lieu d’hébergement de Harkis oubliés (1963-2000),” (M2 thesis, Université Paris I
Panthéon-Sorbonne, 2008).
215 CAC, 19980351/11, “Statistiques: Reclassement des rapatriés musulmans par les antennes civiles des camps d’accueil
à la date du 1/1/1965.”
216 AN, 5 AG 1/22, “Rapport sur l’application…,” 127.
217 Abderahmen Moumen gives this date of closure for Saint-Maurice-l’Ardoise in Les Français musulmans en Vaucluse, 194.
Lanoizlez discusses the two housing developments opened on the site of the Bias camp in “La Cité d’Accueil,” 99-105.
226
complète à la vie métropolitaine, de son manque de formation intellectuelle, de son ignorance des
lois et coutumes.” Therefore, it was the responsibility of the departmental social services office to
provide the harki with information about his rights and duties as a citizen and to intervene in any
conflicts that the harki encountered with his employer.218 Despite his emphasis that the harkis were
indeed French citizens in the letter’s opening sentence, the rest of Aucante’s words marked the harki
as an outsider to French society.
The most popular destinations for those with some job skills were factories, construction
sites in the Pyrénées-Orientales department, the SNCF worksites, and mines located in northeastern
France. The harki families were sometimes housed in cités urbaines under a special construction
project, “Le Programme ‘Harkis.’” Situated in urban settings, these public housing projects sheltered
exclusively harki families, whose heads of household worked in nearby factories, in the mining
industry, or for the SNCF. Social counselors responsible for the residents’ encadrement helped the
harkis adapt to metropolitan life and acted as their interlocutors with the state. Nevertheless, the
government’s preferred option for the harkis’ reclassement was the forest hamlets, whose policies are
examined in the next chapter.
ACNMF, 32/3, Form letter René Aucante to Departmental Prefects. “Citoyen français” and “En aucun cas” are
underlined in the original document.
218
227
CHAPTER 4
Repatriates? The Post-Imperial French Government’s
Forest Hamlet “Integration” Policy
Le reclassement des harkis dans le secteur
forestier entraîne depuis quelques semaines
l’ouverture d’un certain nombre de chantiers de
forestage et l’installation des hameaux d’habitation
destinés aux familles de ces harkis…
Nos services restent en liaison constante avec
vous [les préfets] pour essayer de résoudre les
problèmes posés par l’intégration de ces réfugiés dans
la communauté métropolitaine…
Il faut en effet éviter que ces groupes ethniques
ne se trouvent encore plus isolés en dépendant
uniquement de structure trop spécialisée.1
-Yves Pérony, Ministry of Repatriates,
December 19, 1962
Le problème des rapatriés d’origine algérienne
[sur les hameaux forestiers] sensibilise actuellement,
comme vous le savez, cette catégorie des Français,
dont la situation matérielle, morale et psychologique
laisse à désirer. Or, cet aspect humain semble avoir
été quelque peu perdu de vue.2
-Michel Poniatowski, Minister of Interior,
April 4, 1975
The Ministry of Repatriates’ reclassement service (SFIM) rehoused nearly 25 percent of the
40,582 harki population members from the Rivesaltes and Saint-Maurice-l’Ardoise camps, and 12
percent of the total harki population repatriated to France, in 72 forest hamlets concentrated
primarily in southern France.3 These spaces, hastily set up in emergency conditions, were located in
proximity to a village de rattachement, rural towns with populations numbering from several hundred to
Archives Départementales du Var (hereafter ADV), 746 W 62, Circular from Yves Pérony to Departmental Prefects, 1,
N° 2175 SFIM/MG, Dec. 19, 1962.
2 CAC, 19910097/43 (dossier: Harkis – Correspondance 1975), Minister of Interior Michel Poniatowski to Minister of
Agriculture Christian Bonnet, Apr. 4, 1975.
3 AN, 5 AG 1/22, Minister of the Interior, “Rapport sur l’application…,” 112 and 117.
1
228
several thousand. Men worked on forestry worksites (chantiers de forestage) in twenty-five to thirty
person teams, constituted exclusively of harkis and supervised by a Ministry of Agriculture official
and a military officer. Women spent their days in the forest hamlets tending to their household and
children. Most harki children attended nearby schools. Representatives of the state—a social
counselor and camp director—lived on-site with the harki families, acting as these French citizens
interlocutors with local government officials and ensuring that the forest hamlet inhabitants did not
disturb their neighbors. Despite setting up the forest hamlets as spaces that differentiated and
isolated the harki population, the Ministry of Repatriates maintained that the cornerstone of its
forest hamlet policy—as SFIM Director Yves Pérony refers to above—was integrating the “refugees”
into their surrounding community.
This housing situation was a form of what officials labeled collective reclassement (reclassement
collectif), intended to be a temporary step after the housing camps to prepare the harkis for insertion
into French society. 4 As explained in chapter three, reclassement is a complex term that the
government used to signify the harkis’ integration into French society through access to housing and
employment. While this word first appeared in the dictionary in the 1869 edition of Emile Littré’s
Dictionnaire de la langue française to describe the reclassification of British electoral districts following
the 1867 Reform Bill (which doubled the ranks of voters as a movement toward universal male
suffrage), its meaning soon shifted to describe a process by which people were transformed.5 The
Dictionnaire de l’Académie française initially included reclassement in its eighth edition (published from
1932 to 1935), defining the term as a new classification of people or things to “correct” a former
classification. This entry cited the example of the “reclassement des valeurs,” referring to changing a
As mentioned in chap. 3, the word reclassement in this context does not have a succinct translation in English. Therefore,
it is left in the original French throughout the text.
5 Littré’s Dictionnaire de la langue française defines reclassement as: “Classement nouveau, différent. Le reclassement des
districts électoraux exigé par la réforme en Angleterre.” Emile Littré, Dictionnaire de la langue française (Paris: 1863-73), s.v.
“reclassement,” <http://francois.gannaz.free.fr/Littre/xmlittre.php?rand=&requete=reclassement&submit=Rechercher
(accessed March 13, 2011).
4
229
person’s values or worth.6 At the beginning of the twentieth century, colonial administrators in
Indochina, Africa, and Madagascar had employed this term in this way, that is, to describe
circumstances of re-educating native populations. For example, officials in Indochina used “reclasser”
to characterize the process the government undertook with children who were born, most often
outside of marriage, to a French father and a mother of indigenous origin. According to one such
administrator, by taking these métis away from their mothers and placing them in “institutions
spécialisées,” they would be able to become “Français d’âme et de qualités.” 7 The process of
reclassement therefore required the métis children’s dependence on French officials and administrative
structures. This definition more closely reflects the term’s usage for the harkis in the 1960s and
underscores the importance of the physical spaces where the transformation took place.
In a similar fashion to the procedure of inculcating Indochinese métis children earlier in the
century, the forest hamlets for the harkis and their families were “specialized institutions” separating
them from other French citizens. The residents remained under the supervision of a social counselor
tasked with educating them about how to act as “French citizens” now that they legally had French
nationality. The government’s vision for the forest hamlets provoked two of their three principal
flaws: isolation and dependency. Indeed, these are precisely the faults that, in December 1962 after
the government had opened the first ten forest hamlets for the harkis, Pérony warned departmental
prefects could materialize, as cited in the epigraph to this chapter.
The third flaw endemic to these spaces was their longevity. This reflects discursive and
policy shifts from the government’s initial claims that the forest hamlets would merely be a
temporary step to help the harkis integrate into French society. As Minister of Interior Michel
Poniatowski wrote in the April 1975 letter to Minister of Agriculture Christian Bonnet cited in the
Dictionnaire de l’Académie française, 8th ed., s.v. “reclassement,”
http://atilf.atilf.fr/dendien/scripts/generic/cherche.exe?15;s=3292224930 (accessed March 13, 2011). The term’s verbal
form, “reclasser,” referred to the reclassification of archival documents.
7 Saada, Les enfants de la colonie, 79. On the definition of the term métis and its usages, see pp. 14-16.
6
230
epigraph to this chapter, by keeping open the isolated forest hamlets, whose living conditions had
severely degraded, government officials had lost sight of the harki population’s humanity in their
policy decisions.
Instead, the harkis—men who had risked their lives for France during the Algerian War for
Independence—were an unwelcome reminder to these officials of the end of France’s colonial
empire. Historian Benjamin Stora wrote in 1991 about the memory of this war: “Dans ce pays, où
les guerres de Vendée [pendant la Révolution française de 1789] sont encore des discordes
contemporaines, on cache ce passé tout récent. Tout un ensemble subtil de mensonges et de
refoulements organise la ‘mémoire algérienne’. Et cette dénégation continue à ronger comme un
cancer, comme une gangrène, les fondements mêmes de la société française.”8 By continuing to hide
the harkis in secluded camps, government officials seemed to try to physically push them out of
their sight, just as they tried to repress from their minds the loss of the Algerian War for
Independence.
The forest hamlets, moreover, challenged the harkis’ legal status as repatriates. This
classification entitled the harkis to special social and economic benefits; in practice, however, some
of these benefits were limited for them. In August 1961 when Prime Minister Michel Debré named
Robert Boulin the first Secretary of State for Repatriates, he tasked the new member of his cabinet
with designing an “arrival” and “settling in” policy (une politique d’accueil et de réinstallation) for the
increasing number of French repatriates from Algeria who arrived in France.9 The result was the
December 26, 1961 “Loi relative à l’accueil et à la réinstallation des Français d’outre-mer,” which in
theory applied to the harkis. A September 4, 1962 decree officially extended the provisions of the
Boulin law to “étrangers”—a legal denomination—who were forced to leave a French colony,
protectorate, or Algeria and fell under one of five categories. Since very few harkis were enlisted in
8
9
Stora, La gangrène et l’oubli, 8.
Scioldo-Zürcher, Devenir métropolitain, 107.
231
the French Army (variations of which were categories one through three), they and the other
demobilized auxiliary forces met the requirement for category five: “Avoir fait preuve de
dévouement à l’égard de la France, ou lui avoir rendu des services exceptionnels.”10 The Boulin law,
and the ensuing legislation with more precise instructions for its implementation, enacted measures
“de nature à intégrer les Français rapatriés dans les structures économiques et sociales de la nation.”
These included assistance finding jobs and housing, monthly allocations for a temporary period, an
allowance to settle in to a new residence (subventions d’installation), and priority access to housing
constructed by the state.11 However, the law was applied differently to the French repatriates and to
the harkis—mostly because of the government’s preference for collective reclassement to house the
harkis. Collective reclassement included camps such as Rivesaltes, the seventy-two forest hamlets, and
seventy-one cités urbaines built by the SONACOTRA under the special “Programme ‘Harkis.’”12
Collective reclassement challenged the harkis’ access to assistance intended for repatriates—
and, therefore, their status as repatriates—in three principal ways. First, the organization that the
Secretary of State for Repatriates put into place for doling out benefits to the French repatriates and
the harkis fostered two different categories of repatriates. SFIM officials at the Rivesaltes and SaintMaurice-l’Ardoise camps were responsible for giving the harkis in the camps any allocations for
which they were eligible and finding the harkis jobs and housing. Twenty percent of these positions
were located on isolated forestry worksites. Any harki who had ever lived in the camps would
remain under the jurisdiction of the camp SFIM offices for the purposes of receiving future social
JORF, “Décret nº 62-1049 du 4 septembre 1962 portant règlement d’administration publique pour l’application à
certains étrangers de la loi nº 61-1439 du 26 décembre 1961 relative à l’accueil et à la réinstallation des Français outremer,” Sept. 5, 1962. The fourth category was “Avoir perdu un descendant, un ascendant ou son conjoint mort pour la
France,” and therefore applied to widows and orphan children of French Muslim auxiliary forces.
11 JORF, “Loi nº 61-1439 du 26 décembre 1961.” The principal text with the modalities for implementing the Boulin
Law is: JORF, “Décret nº 62-261 du 10 mars 1962 relatif aux mesures prises pour l’accueil et le reclassement
professionnel et social des bénéficiaires de la loi nº 61-1439 du 26 décembre 1961,” 2521-33, Mar. 11, 1962. For a
description of this decree see Scioldo-Zürcher, Devenir métropolitain, 114-17. Government documents sometimes referred
to “subventions d’installation” as “indemnités de déménagement.”
12 CAC, 19920149/1/4, Ministry of Social Affairs, “Programme de Construction pour les ‘Harkis,’” undated. The author
intends to include a case study of the Cité les Tilleuls cité urbaine in Marseille in the manuscript version of this text.
10
232
welfare benefits and assistance finding housing and jobs should his initial reclassement fail.13 The harkis
who did not pass through the camps were instructed to address themselves to the French Muslim
division of the Repatriates Service in the department where they resided or the Centre National de
Répartition de l’Emploi et du Logement pour Rapatriés Musulmans in Jouques (Bouches-du-Rhône)
to receive employment aid.14 On the other hand, a repatriate service in each department handled the
French repatriates’ reclassement and distributed their allocations. Departmental officials gathered
information about French repatriates’ professions in Algeria, which, like with the harkis, often did
not correspond to the type of jobs that existed in metropolitan France.15 The state even created
quotas in some professions—such as pharmacists and taxi drivers—requiring that employers hire a
certain percentage of French repatriates. Departmental agents conducted surveys among local
businesses and municipal officials to try to find these migrants jobs and offered to employers
incentives to hire French repatriates.16 In the Bouches-du-Rhône department, for example, after
conducting one such survey among 8,700 potential employers in July 1962, the local branch of the
Association pour l’emploi dans l’industrie et le commerce (ASSEDIC) subsequently published a list
of repatriates in search of jobs every two weeks.17 This two-tiered system for the harkis and French
repatriates led a top official in the SFIM office at the Rivesaltes camps to label it “une sorte
‘d’apartheid’” owing to the government’s recourse to different offices for allocations and services
“[dûs] à une seule catégorie de citoyens.”18
The second way that collective reclassement challenged the harkis’ access to repatriate benefits
is that a portion of money due to each head of household as subventions d’installation was used to
CAC, 19970146/2 (dossier: Robert Bourgat), Robert Bourgat (Liaison Mission at the Rivesaltes Camp) to the Prefect
of the Pyrénées-Orientales, “Objet: Situation des ‘Harki’ recasés dans les Pyrénées-Orientales,” 1, nº 835/RB/RB/3,
June 27, 1963.
14 AN, 5 AG 1/22, Minister of the Interior, “Rapport sur l’application…,” 107, July 26, 1965.
15 Scioldo-Zürcher, Devenir métropolitain, 249.
16 For a detailed analysis of the government’s policy and practices for finding jobs for French repatriates, see ScioldoZürcher, Devenir métropolitain, 249-67.
17 See ADBR, 131 W 377 for various documents, studies, and pamphlets published by the ASSEDIC departmental office.
18 CAC, 19970146/2, Robert Bourgat, “Objet: Situation des ‘Harki’ recasés…,” 2.
13
233
finance the construction of forest hamlets and to furnish the cités urbaines.19 A July 1963 circular from
the Ministry of Repatriates specified that harki families who moved to the “temporary” forest
hamlets received no money to help them find a more permanent living situation, unless they lived
there (and the harki worked at the forestry worksite) for at least six months.20 As Pérony wrote to a
harki who had resided in a forest hamlet in the Basses-Alpes department and subsequently requested
his allowance to settle into a new residence outside of the forest hamlets, “J’ai l’honneur de vous
faire connaître que les rapatriés reclassés et logés sur les hameaux de forestage par les soins du
Ministère des Rapatriés ne pouvaient prétendre au bénéfice de cette subvention…”21 This exception
made it more difficult to for the harkis to leave the forest hamlets. Moreover, this information was
often difficult for the harkis to access since it was sent to departmental prefects and regional
delegates of the Ministry of Repatriates, who had the responsibility of passing it on to the social
counselors and camp directors to share with the harki population.
As a final example, while the government set aside places for the harkis in collective
reclassement housing situations—the forest hamlets and cités urbaines—they were not eligible for the
same public housing benefits offered to French repatriates. A January 31, 1964 circular from the
Ministry of Repatriates instructed local officials to give priority to French repatriates over harkis
when applying the August 18, 1962 ordinance reserving 30 percent of apartments in newlyconstructed HLM buildings for repatriates: “Vous ne devrez reloger les anciens harkis qu’après avoir
relogé tous les rapatriés demandeurs de logement et particulièrement mal logés.” The circular
CAC, 19920149/1/5, SFIM, “Note sur les harkis et les moghaznis,” 5, undated. While this copy of the memo is
undated, the author concludes that it was written in late September based on its reference to the completed transfer to
the Rivesaltes camp of the harki population from the Bourg-Lastic camp, which concluded on September 24 (ADPO,
1419 W 109), and to that from the Larzac camp, which the memo indicated would start on October 1.
20 CAC, 19770391/8, Circular from Minister of Repatriates to Departmental Prefects and Regional Delegates of the
Ministry of Repatriates, “Objet: Attribution des prestations de retour aux musulmans rapatriés,” 2, Nº 63-99, June 19,
1963.
21 CAC, 19920149/5/11, Yves Pérony to Mr. Teffahi A., Nº 4497 SFIM/MR, Dec. 3, 1964. This exception was outlined
in an Aug. 7, 1963 circular, a copy of which is located in: CAC, 19920149/1/5, Minister of Repatriates to Departmental
Prefects and Regional delegates of the Ministry of Repatriates, “Objet: Attribution de la subvention d’installation aux
musulmans reclassés dans un emploi salarié,” Nº 63.121/Cab., Aug. 7, 1963.
19
234
explained the rationale for this order, which impeded the harkis’ access to an integrated housing
situation, “…en raison des avantages particuliers prévus pour le logement des anciens harkis,
notamment du programme lancé par la SONACOTRA…”22 As mentioned above, the buildings
constructed under this collective reclassement program housed harki families exclusively.
This chapter traces the creation and evolution of the national government’s policies for the
forest hamlets. It examines how the shifts and stagnations in these policies affected both the
individuals residing in these spaces and the place of harki citizens in French society. The harkis and
their family members residing in the forest hamlets were legally French citizens. Nevertheless,
government officials treated them as an inferior class of repatriates—and citizens—to begin with by
devising different systems and legislation to implement the Boulin law, whose measures were created
to integrate all repatriates into the social and economic structures of the French nation. The
government’s recourse to forest hamlets as a central feature of its housing policy and the subsequent
modifications to this policy, which made these spaces permanent without improving their residents’
living conditions, solidified the harki population’s de facto status as refugees and immigrants. This
chapter ultimately demonstrates that the government’s choice to initiate and prolong a policy of
housing members of the harki population in isolated forest hamlets under the supervision of
encadrement agents resulted in the marginalization from French society of these reminders of the
Algerian War for Independence. As will be argued in this chapter and the next, the challenges
encountered in integrating harki citizens residing in the forest hamlets into their surrounding
communities did not arise from a divide between theory and practice. Instead, these challenges
resulted from the inconsistency, created at the outset, between Ministry of Repatriates officials’
discourse of integration and national government actors’ design and subsequent refinement of the
ACNMF, 15/13, Circular from the Ministry of Repatriates to Departmental Prefects and Regional Delegates of the
Ministry of Repatriates, “Objet: Attribution de logements H.L.M. aux anciens harkis,” 1, Nº 64/19, Jan. 31, 1964.
22
235
forest hamlet policy, which effectively impeded these harki citizens’ integration into the French
community.
The First Phase: Developing the Government’s Contradictory Forest Hamlet Policy
The policy that the French government officials established in 1962 for what they would
later term “the first phase of forest hamlets” was contradictory. On the one hand, Ministry of
Repatriates officials insisted that the goal of these temporary spaces was to integrate the harkis into
the French community. Yet, on the other, the instructions these administrators simultaneously gave
for the forest hamlets’ design effaced the possibility for the harki population’s integration. The
prescriptions necessitated that the forest hamlets be located in isolated areas kilometers away from
small towns in rural France. They included encadrement by soldiers and agents who lived in the forest
hamlet, which ensured that the harki population would be separated and differentiated from their
neighbors. Finally, these instructions depended on local officials, some of who did not want the
harkis in their communities, to facilitate their integration. In short, the government set up the forest
hamlet policy to fail in its purported goal of integrating the harki population.
The notion of encadrement, which was not specific to the harkis, is essential to understanding
the construction of the government’s forest hamlet policy. Since the late nineteenth century French
administrators had employed this practice, initially for single male French workers and more recently
for foreign—often colonial—populations.23 As detailed in last chapter’s analysis of the Rivesaltes
camp, encadrement for the harkis implied a two-pronged relationship of power with the state
resembling Michel Foucault’s notion of modern biopower. It was a method of social and political
control. On the one hand, a network of state agents with different functions (“une armée de
techniciens”) closely watched over individual camp inhabitants’ movements, disciplined them, and
On encadrement structures used for single workers, see Robert Castel, Les métamorphoses de la question sociale: Une chronique
du salariat (Paris: Fayard, 1995).
23
236
educated them about—and sometimes imposed on them—metropolitan ways of life.24 On the other,
the state enacted toward the harkis a regulatory technology, which Foucault called a “technology of
security.” This policy aimed to protect the security of “the whole” (the general French population)
from “internal dangers” (the harki population) by placing them in camps.25 Encadrement therefore
presents a fundamental contradiction with the Ministry of Repatriates’ goal of integrating the harkis
into their communities because it was an inherently exclusionary policy. Encadrement moreover
exacerbated the physical separation of the harkis from other French citizens already present given
the forest hamlets’ isolation.
Several sources indicate that the original idea to house the harkis in forest hamlets came
from SAS officers who had overseen Muslim auxiliary troops in Algeria, the same type of agents
who would subsequently be responsible for the harkis’ encadrement. Beginning in February 1962
associations of SAS officers developed a policy that advocated using the abandoned villages in
southern France resulting from the generalized rural exodus that began during the interwar period to
house groups of harkis. Conceived as a “solution à moyen terme/transitoire de trois années environ,”
harki men would work in nearby forestry or farming jobs and the houses would provide the families
with the possibility to, according to a report of the Comité national de solidarité pour les Français
Musulmans réfugiés, “vivre correctement.”26 In March a group of SAS members identified as a
prototypical abandoned village Ybours in the Basses-Alpes department. This town had no more
inhabitants, but the houses were connected to electricity and water and the harkis could cultivate
nearby fields.27 Officers, who knew firsthand the violence—and threat of violence—in Algeria (and
Foucault, Surveiller et punir, 18.
Foucault, “Society Must Be Defended,” 249. One can also argue another application for this paradigm: the harkis were
“the whole” and FLN insurgents were “internal dangers.”
26 ACNMF, 25/17, Comité national de solidarité pour les Français Musulmans réfugiés, “Conseil d’administration du 3
août 1962: Note sur les possibilités d’implantation de communautés de Français Musulmans dans les villages et dans les
chantiers de forestage,” 3 and 1. This document in the CNMF archives had the word “à moyen terme” scratched out in
pen and an unidentified person wrote “transitoire” in its place.
27 ACNMF, 25/17, Comité national de solidarité pour les Français Musulmans réfugiés, “Conseil d’administration,…” 2.
24
25
237
in France) by FLN members, believed this potential aggression outweighed the negative effect that
provisionally isolating the harkis would have on their integration into French society. For example, a
second lieutenant performing his military service as an SAS agent wrote in April to several prefects
in the Midi region, inquiring about abandoned villages in their departments where the “Muslim
soldiers” who had served under him could live while they worked the land. He warned that the
harkis “sont voués, eux et leur famille, à une sort atroce et certaine” if they remained in Algeria and
underscored this point: “L’isolement n’est pas un obstacle pour ces gens là.”28
The pressure that SAS associations and officers exerted on the government to consider
abandoned villages as a solution for housing repatriated harkis led Prime Minister Michel Debré to
put this issue on the agenda at a March 1962 Council of Ministers meeting.29 To sustain themselves,
the harkis would need to procure jobs in the forestry or agricultural sector. Consequently, Debré
asked Ministry of Agriculture officials to take the lead in exploring the possibility of temporarily
employing the harkis. This decision, which shifted the emphasis of their inquiries onto jobs over
housing, would have multiple future implications for the harki population, particularly for women,
who spent the most if not all their time in the forest hamlets.30 By mid-April a plan emerged to
constitute groups of twenty-five men to perform tasks such as reforestation and clearing paths, with
the initial goal of having two or three of these forestry worksites per department.31 In the end,
however, only twenty-six of France’s ninety departments would contain worksites. The men and
their families would live adjacent to the worksite in a forest hamlet, whether in abandoned or
SHAT, 1R 367/7 (dossier: Envoi en métropole des harkis et de leurs familles), S/Lieutenant François Le Breton, Apr.
20, 1962.
29 ACNMF, 25/17, Comité national de solidarité pour les Français Musulmans réfugiés, “Conseil d’administration…,” 1.
30 The particular effect that the forest hamlets’ isolation had on women will be addressed in next chapter’s case study.
31 CAC, 19910097/40 (dossier: Harkis – Correspondance 1962), Chief Forestry Expert Jolain to the Forestry
Commissioners in Gap and Digne, 1, Apr. 14, 1962. Some officials at the April 10 Massenet Commission meeting
evoked the possibility of using the harkis as a replacement for Italian lumberjacks (bûcherons). At this time, Italian
migrants were increasingly choosing to work in other European countries offering better pay and many were leaving the
profession. However, Ministry of Agriculture officials noted that it would be impossible for the harkis to gain the
necessary skills demanded by the profession and that limited housing possibilities existed. CAC, 19910097/40, “Compte
rendu de séance sur les harkis…,” 2.
28
238
prefabricated houses. The Ministry of Agriculture specifically instructed local officials to choose sites
“en dehors des agglomérations.”32
It is clear from the earliest discussions that encadrement would be a central feature in
implementing this plan. During what appears to be the first meeting devoted to employing the
harkis as forestry workers, attended by representatives of multiple ministries, on April 6 officials
underscored that the harkis would need to be “solidement encadrés sur le triple plan policier,
technique et social.”33 The recommendation that followed from the April 10 Massenet meeting was
to have SAS and other officers from Algeria familiar with their ways of life continue to oversee
them. 34 The organization of these work details echoed that of auxiliary force units during the
Algerian War for Independence. For example, harkas composed of one hundred harkis were
commanded by a French sergeant, “chef de harka,” and overseen by three to five officers, and
maghzens included twenty-five to thirty moghaznis overseen by one or two SAS officers. In a
November 1962 telegram SFIM Director Pérony would even refer to the group of harkis to be
transferred from the Rivesaltes camp to a forest hamlet as a “harka de 25 travailleurs.”35
The challenge to enacting the government’s plan resided in finding towns willing, on one
hand, to integrate the harkis and their families into their communities and, on the other, to provide
land or abandoned villages to house them. During late spring 1962, the Ministry of Agriculture took
an active role in inquiring among mayors and prefects in six southern French departments, yet found
few sites with work opportunities, adequate housing, and the approval of local officials. On May 28
Minister of Agriculture Edgar Pisani’s cabinet director sent a letter to a member of Secretary of State
CAC, 19910097/40, Chief Forestry Expert Jolain, 1, Apr. 14, 1962.
On April 6, 1962, Minister of Agriculture Edgar Pisani presided over a meeting with representatives from different
ministries (including Armies and Algerian Affairs), Secretary of State for Repatriates Boulin, and several of his
subordinates. The summary of the meeting can be found in: CAC, 19910097/40 (dossier: Harkis – Correspondance
1962), Paul Benda (Chief Forestry Expert), “Note pour Monsieur le Directeur Général: Compte-rendu de la réunion
consacrée aux rapatriements d’Afrique du Nord,” 1, EF/B Nº 106, Apr. 7, 1962.
34 CAC, 19910097/40, “Compte rendu de séance sur les harkis…,” 3.
35 CAC, 19920149/3/2, Telegram from Yves Pérony to M. Couston (Director of the Ministry of Repatriates office at the
Rivesaltes Camp), Nº 210/SFIM, Nov. 12, 1962.
32
33
239
for Repatriates Robert Boulin’s cabinet indicating that local agents in his Ministry’s Forestry
Division (Direction Générale des Eaux et Forêts) had studied the possibility for nineteen sites in the
Basses-Alpes, Hautes-Alpes, Gard, Hérault, Aveyron, and Tarn departments. Only four of these
would later open, one of which was in an abandoned village.36 According to the Director General of
the Forestry Division, one of the primary reasons that these sites never opened was because of
political objections by departmental prefects.37 A letter written two weeks later by the commissioner
of the Forestry Division in Valence to the director general confirms this assertion. He exposed the
litany of excuses he received from local officials in his region in response to previous offers to allot
land for forestry worksites, including one official’s concern about the decision to “implanter
définitivement ces cellules sociales sur le vieil organisme rural français.” The commissioner
concluded the letter by stating, “En résumé, le problème majeur, du moins au début, n’est pas celui
de l’emploi mais celui de l’Administration.”38
Meanwhile, to house harki families SAS officers continued to advocate the use of abandoned
villages, which would provide a more stable living environment. In July one group of officers
proposed three additional ones in the Aude department—Pujol de Bosc, Puylaurens, and La Camp.39
While the former was eventually converted to house harki families, it soon became apparent that the
best chance for finding the harkis homes lay in shifting the search to worksites with adjacent land
where lightweight prefabricated houses (préfabriqués légers) could be erected, as complications had
arisen with the potential abandoned villages. Namely, these houses had owners—some of whom had
CAC, 19910097/40 (dossier: Harkis – Correspondance 1962), Minister of Agriculture to Secretary of State for
Repatriates, “Objet: Emploi de Harkis à des travaux dans les forêts domaniales et dans les séries domaniales de
restauration de terrains en montagne,” EF/DR-1202, May 28, 1962.
37 CAC, 19910097/40 (dossier: Harkis – Correspondance 1962), Director General of the Forestry Division “Note pour
Monsieur le Ministre (Cabinet): Projets d’emploi de main-d’œuvre rapatriée, et notamment de Harkis, sur les chantiers de
l’Administration des Eaux et Forêts,” 4, EF/D-1625, June 19, 1962.
38 CAC, 19910097/40 (dossier: Harkis – Correspondance 1962), Forestry Commissioner in Valence to the Director
General of the Forestry Division (Paris), “Objet: Chantiers de harkis – Compte rendu général,” 1-2, Nº 1354, July 2,
1962.
39 ACNMF, 25/17, Comité national de solidarité pour les Français Musulmans réfugiés, “Conseil d’administration…,” 2.
36
240
moved, others of whom had passed away and left their property to family members.40 Moreover,
some houses needed many repairs, often more costly than constructing new prefabricated houses,
which the Secretary of State for Repatriates proved unwilling to fund.41
Throughout the summer of 1962 local and national officials in the Ministry of Agriculture
continued to canvas the French countryside, predominantly in southern France, for possible land to
construct forest hamlets. In July, Minister Pisani sent another round of letters to departmental
prefects asking whether their departments had a need for groups of twenty-five forestry workers to
employ for at least one year, thereby presenting the project as temporary. The worksite would need
to be close to an abandoned village whose houses the harkis could repair or empty land where the
harkis themselves could construct prefabricated homes for their families. The memo underscored
that potential houses must have access to water, be located in a mild enough climate for the harkis
to work almost year round, be situated close to schools for the harki children, and have accessible
roads to facilitate receiving supply shipments. What is striking about this brief memo is Pisani’s
choice to stress in three different places that soldiers provided by the Ministry of Armies would
closely watch over the harkis. Specifically, he detailed both the precise structure of the “personnel
d’encadrement” to oversee the workers on the forestry worksites and the type of agents and soldiers
who would direct the harkis if they repaired abandoned houses. The latter, he noted, would be the
“le même contrôle et encadrement” as the one that would be put in place if the harkis constructed
their own homes.42 Indeed, two letters written in June by high-ranking officials in the Ministry of
Agriculture to their superiors demonstrated their belief in the centrality of a well-defined plan of
encadrement to the forest hamlet’s success. The first, written on June 4 to the Director of the Forestry
Ibid., 2-3.
CAC, 19920149/3/1, “Réunion du lundi 7 janvier 1963,” 1.
42 CAC, 19910097/40 (dossier: Harkis – Correspondance 1962), “Ministre de l’Agriculture à Monsieur le Préfet de…,”
1-2, July 2, 1962. The copy of this form letter included in the archives is dated July 2. However, an unidentified person
wrote on it “25 juillet 1962 pour la,” leading the author to conclude that not all of the letters were sent on July 2.
40
41
241
Division, gave a three-page explanation of how many of the different types of agents (Ingénieurs,
Ingénieurs adjoints, Chefs de district, and Agents techniques) would be allotted per Forestry Commission,
which comprised one or two departments, and the agents’ specific roles.43 The second, addressed to
Minister Pisani and his cabinet on June 19, outlined the five principle issues that needed to be
resolved before opening the forestry worksites to ensure their success. Two of these five “impératifs”
related to encadrement: “UN ENCADREMENT SOCIAL EST ABSOLUMENT NECESSAIRE”
(underlined and placed in capital letters in the document) and “Un encadrement technique sévère
n’est pas moins utile.”44
Pisani’s plea received few favorable responses. Replies from prefects and forestry
commissioners who consulted with local officials demonstrated various reservations. The principle
concerns included a preference for using local laborers for forestry jobs; worries about their need to
finance costs associated with forest hamlets and worksites (including personnel to oversee the
harkis); and the absence of available housing. The prefect of the Hautes-Pyrénées, for example,
wrote that despite the rural exodus no villages were completely abandoned. Other reasons supplied
by local officials consisted of not enough forests in their department to provide work for the harkis,
a colder climate preventing the harkis from working several months out of the year, and concerns
about their adaptation to this climate.45 Although—and perhaps because—Pisani underscored how
the harkis would be closely supervised, only one objection relating to security concerns was raised in
these letters. The forestry commissioner overseeing the region around Bordeaux expressed
CAC, 19910097/40 (dossier: Harkis – Correspondance 1962), M. Doniol (Le Chef de Service des Domaines Soumis
au Régime Forestier), “Note pour Monsieur le Directeur Général des Eaux et Forêts: Organisation de l’emploi des harkis
– Evaluation des besoins en personnel forestier d’encadrement,” EF/D4-nº 1418, June 4, 1962.
44 CAC, 19910097/40 (dossier: Harkis – Correspondance 1962), Director General of the Forestry Division “Note pour
Monsieur le Ministre…,” 3. Two of the other three “impératifs” concerned funding and the last one focused on the
difficulty in finding sites.
45 The folder “Harkis – Correspondance 1962” in CAC, 19910097/40 contains response letters written in July and
August. Those cited above were written by prefects of the Corrèze, Haute-Pyrénées, and Savoie departments and
forestry commissioners in the Allier, Puy-de-Dôme (27th Commission), Gironde, Dordogne, and Lot-et-Garonne (32nd
Commission) departments.
43
242
uneasiness about housing the harkis near local seaside resorts: “…l’implantation des harkis sur le
littoral atlantique au voisinage des stations balnéaires et des nombreux camps d’estivants ne paraît
pas souhaitable.”46 In short, many local officials were not persuaded by the government’s vision to
billet the harkis on the fringes of rural towns in their regions, and particularly, the social and
economic costs that the forest hamlets could incur.
Agriculture Ministry officials chose to respond to some of the objections by addressing local
agents’ concerns and asking them to reconsider. For example, the official whom Pisani put in charge
of the mission to find potential forestry worksites urged the director of the Forestry’s Civil
Engineering Division in Toulon (Var) to meet with the director of the Forestry Commission in the
Var who had declined to accept groups of harkis owing to a lack of housing. This official provided
the Var director with nine talking points for the discussion, including ones concerning the amount
the government would provide the department to construct each house (4,000 new francs for
repairing an existing house and 8,000 new francs for erecting a new one); the possibility of sheltering
the harkis in tents temporarily while the houses were being constructed; and the potential to reuse
the houses once the harkis left. The memo again emphasized the temporary duration of the forest
hamlets—approximately two years.47
The Ministry of Agriculture nevertheless did receive some favorable responses from local
officials regarding constructing forest hamlets in their departments. On August 16, 1962 Pisani
wrote to Boulin with a list of seventeen forestry worksites likely to open soon. Ten of these would
later open.48 Boulin (who served as Secretary of State for Repatriates until September 11, 1962), his
CAC, 19910097/40 (dossier: Harkis – Correspondance 1962), Forestry Commissioner of the 32nd Region to the
Director General of the Forestry Division, “Objet: GIRONDE – Emploi de la main d’œuvre rapatriée et notamment
des harkis sur les chantiers d’amélioration du domaine forestier de l’État,” Nº 1258/7 b.C.7, Aug. 13, 1962.
47 While there is no trace of future correspondence in this archival carton to evaluate the effect of this letter, the Var
would contain more forest hamlets than any other department. ADV, 746 W 62, J. Montpied to the Director of the Civil
Engineering division in Toulon, “Objet: Recasement des harkis rapatriés d’Algérie,” 1-2, July 28, 1962.
48 CAC, 19910097/40 (dossier: Harkis – Correspondance 1962), Minister of Agriculture to the Secretary of State for
Repatriates, “Objet: Recasement des harkis,” Aug. 16, 1962.
46
243
successor Alain Peyrefitte (who briefly served as Minister of State for Repatriates until Prime
Minister Pompidou reorganized his government on November 28, 1962), and Peyrefitte’s successor
François Missoffe (the last Minister of State for Repatriates, serving until the Ministry was eliminated
on July 23, 1964) addressed a series of memos to prefects in the departments that had agreed to
create forest hamlets. These three documents—dated August 27, October 29, and December 19—
outlined practical information concerning how the forest hamlets would be financed, the structure
of encadrement, and details about housing.49 The October memo reaffirmed the temporary nature of
the forest hamlets by assuring local officials that any land they provided would only be needed for a
“temps très limité.”50
Foremost, these documents consistently underscored the government’s vision for the
functioning of the forest hamlets, a vision that centered on integrating the harkis into the French
community. As the August memo succinctly explained, “Il est indispensable que ces groupes de
harkis soient intégrés dans la communauté française…,”51 while the December one offered the
government’s aid toward this end: “Nos services restent en liaison constante avec vous pour essayer
de résoudre les problèmes posés par l’intégration de ces réfugiés dans la communauté
métropolitaine.”52 It must be underscored, however, that the isolated locations that government
officials chose for the forest hamlets would call into question the very possibility of successfully
integrating the harkis into French society before they even opened. Moreover, the forest hamlets’
distance from their villages de rattachement, rural towns with small populations, ranged from one
kilometer to as many as fourteen kilometers, as was the case with the Capelude forest hamlet paired
with the town of Collobrières in the Var department.
CAC, 19920149/3/1, Secretary of State for Repatriates to the Prefect of…, “Objet: Chantiers de forestage,” Aug. 27,
1962; CAC, 19920149/3/1, Minister of State for Repatriates to the Prefect of…, “Objet: Reclassement des harkis –
chantiers de forestage,” Nº 1681/SFIM/DIR, Oct. 29, 1962; and ADV, 746 W 62, Minister of State for Repatriates to
the Prefect of the Var, Nº 2175 SFIM/MG, Dec. 19, 1962.
50 CAC, 19920149/3/1, Minister of State for Repatriates, 1, Oct. 29, 1962.
51 CAC, 19920149/3/1, Secretary of State for Repatriates, 1, Aug. 27, 1962.
52 ADV, 746 W 62, Minister of State for Repatriates, 1, Dec. 19, 1962.
49
244
Integration nevertheless permeated each aspect of the Ministry of Repatriates’ proposed
organization for the forest hamlets. Indeed, Ministry of Repatriates officials viewed the forest
hamlets in themselves as a means to integrate the harkis into the French community. The October
memo urged the prefects to open the forest hamlets as soon as possible, “[e]n raison de l’importance
que le Gouvernement attache à l’intégration des harkis dans la communauté française…”53 The
August memo already had exposed the twofold goal of the forest hamlets as furnishing the male
harkis with work and integrating them “little by little into the French community.”54 Boulin argued
that a crucial element of this policy was that the harki children attend schools with neighboring
children and proposed that the regional education authority (Inspecteur d’Académie) make a special
plan to address the needs of the many who did not speak French.55
All of the memos underscored that the collaboration of a network of local officials was
essential to the success of the harkis’ integration. The December memo, in particular, focused on
this theme and put into place a unique administrative structure to implement this policy. Ministry of
Repatriates officials emphasized to the prefects the importance of giving the harkis the same access
to social services as any other departmental residents. Referring to this formula as the best method
to assure the integration of the “refugees,” this memo warned, as cited in the epigraph to this
chapter, “Il faut en effet éviter que ces groupes ethniques ne se trouvent encore plus isolés en
dépendant uniquement de structure trop spécialisée…”56 It continued by asking the prefects to give
the directors of the various social services (such as health care and employment) in their department
orders to be available to the “refugees” for whatever problems or concerns might arise.57 The
Ministry of Repatriates moreover advocated a close relationship between local officials and the
CAC, 19920149/3/1, Minister of State for Repatriates, 2, Oct. 29, 1962.
CAC, 19920149/3/1, Secretary of State for Repatriates, 3, Aug. 27, 1962.
55 CAC, 19920149/3/1, Secretary of State for Repatriates, 1, Aug. 27, 1962.
56 ADV, 746 W 62, Minister of State for Repatriates, 1, Dec. 19, 1962.
57 Ibid., 1-2.
53
54
245
individuals responsible for the harkis’ encadrement. The memo strongly encouraged departmental
officials to visit the forest hamlets often to “prendre connaissance des difficultés d’ordre moral,
matériel qui peuvent exister.”58 To facilitate these interactions, the Ministry of Repatriates created a
network of forest hamlet inspectors, who oversaw several forest hamlets in the same region. The
inspectors were instructed to contact department officials on a weekly basis to inform them how the
forest hamlets were functioning.59 The memo presented the role of the inspector, who was selected
for the position owing to his “connaissance approfondie des milieux musulmans.” He was to
monitor the setup of forest hamlets in coordination with local officials; relay instructions about how
to run the forest hamlets to the military officers who served as camp directors; provide guidance to
the on-site social counselor; and “préconiser toutes les mesures susceptibles d’améliorer la vie du
hameau et de faciliter l’intégration des familles musulmanes au sein de la communauté française…”60
According to a future inspector for social policy (a position that was created in 1965 to oversee the
social counselors), this three-tiered structure used for managing the forest hamlets—composed of
the SFIM director, an intermediary level of inspectors, and the on-site camp directors and social
counselors—was unlike the organization of any other administrative unit responsible for social
welfare. Given this structure, officials responsible for the forest hamlets, therefore, did not have a
blueprint for how to resolve problems that arose and consequently dealt with them on an ad hoc
basis. This inspector explained, “Partagés entre la joie de croire s’accroître l’importance de leur
service et l’effroi causé par les problèmes inhabituels qui se posaient à eux, ceux-ci n’optèrent jamais
pour une politique définie et se contentèrent d’atermoyer indéfiniment chaque fois qu’il aurait fallu
prendre une décision.”61
Ibid., 2-3.
Ibid., 3.
60 Ibid., 2.
61 Heinis, “L’insertion des Français-Musulmans,” 89-90.
58
59
246
From the three documents written in 1962, the government’s vision for the forest hamlets
concretized into a set of five guidelines to be implemented by local agents. This vision was the
government’s integration policy for the harki population. The forest hamlets would be isolated so as
to protect the harkis from FLN insurgents, yet close enough to the small rural towns to allow the
harkis to interact with their neighbors. The harki children would attend local schools, which would
facilitate not only their integration but also that of their parents through their interactions with
“French” people. With the active investment of a network of departmental and local officials, the
harkis would not need to rely exclusively on forest hamlet agents and therefore would become
integrated into their communities. Agents familiar with “milieux musulmans,” that is individuals who
had lived in Algeria, would encadrer the harki population. This encadrement would help them learn
about French ways of life, protect them from FLN members, and control and discipline the harkis
who threatened the stability of the forest hamlet or the surrounding area. Finally, the integration
policy’s success hinged on the forest hamlets being a temporary solution while government officials
sought more permanent jobs for the harkis.
From the outset, this policy contained several inherent flaws, which would later be
compounded as the forest hamlets continued to remain open. The instructions that Ministry of
Repatriates’ officials gave for the forest hamlets’ set up structured them as mini versions of the
housing camps. Isolating the harkis from their surrounding community—both requiring the forest
hamlets to be linked to sparsely-populated towns and then choosing sites on the outskirts of these
small villages—would necessarily inhibit their interactions with other French citizens. The encadrement
structure, which sought to control their movements and differentiated them from other French
citizens not subject to such surveillance, served as an additional impediment to their interactions
with townspeople. Therefore, if the harkis were given little opportunity to interact with their
neighbors, integration could not take place. Moreover, the national government’s design of the
247
forest hamlet policy required “un maximum de concours” from local and departmental officials,
which concentrated the responsibility for its successful implementation into the hands of a few
individuals. This design assured variable results once the national government devolved its
responsibility to local officials and the inspectors.
The Second Phase: Evaluating and “Extend[ing] the Forest Hamlet Experiment”62
On October 7, 1963, sixteen months after the government repatriated the first harkis to
metropolitan soil, the SFIM released a five-page report summarizing the principal problems it faced.
(The government had created this office in July 1962 to run the harki camps and those housing
Indochinese refugees (Noyant-d’Allier and Sainte-Livrade) as well as to find jobs and housing for the
harkis.63) At this time, four housing camps (Bias, Rivesaltes, la Rye, and Saint-Maurice-l’Ardoise)
sheltered 9,376 harki population members and the government had found housing and jobs for
approximately 25,188 harkis, 5,082 of whom it placed in forest hamlets.64 On the first page of the
report, SFIM Director Yves Pérony posed the question, “Pourquoi tous les harkis ne sont-ils pas
encore reclassés?” and proceeded to furnish three reasons, which outlined how his office would
proceed henceforth with the harkis’ reclassement. His first answer justified the delay by advancing
what had become the SFIM’s preferred method for the harkis’ reclassement: “[D]’abord parce que le
Ministère des Rapatriés a tenu à développer au maximum l’expérience des chantiers forestiers…”
The second reason Pérony offered was that reclassement in the industrial sector was proceeding slowly
because the Ministry of Repatriates’ policy was to find harkis jobs and housing on a case-by-case
basis. Finally, Pérony conceded that the Ministry of Repatriates was waiting for the SONACOTRA
This quotation comes from the subject line of the following letter: CAC, 19920149/3/1, Minister of State for
Repatriates to the Minister of Finance and Economic Affairs, “Objet: Reclassement des anciens supplétifs musulmans.
Extension de l’expérience des chantiers forestiers,” May 14, 1963.
63 AN, 5 AG 1/22, “Note concernant le service des Français Musulmans,” May 6, 1966.
64 CAC, 19920149/1/13, J. Besson (Ministry of Repatriates), “Note pour Monsieur Pérony,” 3, Oct. 9, 1963.
62
248
to complete its construction of 2,500 HLM apartments, primarily in urban areas, for the population
under the “Programme ‘Harkis.’”65
As the charts in figures 24 and 25 illustrate, the government increasingly relied on the forest
hamlets to rehouse the harkis from the camps, preferring collective reclassement on these isolated sites
to that in more populous areas and individual reclassement. Only 13 percent of the jobs that the SFIM
found for the harkis until November 15, 1963 (representing the rehousing of 15 percent of the harki
population) were on forestry worksites. The other 87 percent of jobs (representing 85 percent of the
harki population members the government rehoused) were in other sectors—including factories,
mines, agriculture, the SNCF, public works, and domestic work. After this time, reclassement on the
forestry worksites would increase dramatically, even though 1,734 apartments had opened under the
“Programme ‘Harkis’” by the end of 1964, thereby providing the potential for a greater reliance in
other job sectors.66 Between November 15, 1963 and January 1, 1965, 36 percent of the jobs the
government found for the harkis were on forestry worksites (representing 45 percent of the harki
population the government rehoused between these dates) and 64 percent of the jobs (representing
55 percent of the population) were in all other professions combined. The discrepancy between the
number of people and number of jobs demonstrates that the SFIM tended not to place single harkis
(either unmarried or whose family was still in Algeria) in the forest hamlets.
ACNMF, 15/13, Yves Pérony (SFIM), “Problèmes posés,” 1, Oct. 7, 1963.
CAC, 19920149/1/4, “Programme de Construction pour les ‘Harkis.’ I. – SONACOTRA et Conventions – Situation
au 1er Janvier 1965.” These apartments were constructed by the SONACOTRA and its subsidiaries, the SNCF, local
authorities, and associations such as “Comité Accueil aux harkis.”
65
66
249
Percentage of reclassement by number of jobs
100%
80%
60%
Forestry
worksites
40%
All other
professions
20%
0%
7/1/62-11/15/63
11/16/63-1/1/65
Figure 24: The reclassement of the harkis on the forestry worksites accounted for 1,392 jobs between September 7, 1962
(the date the first forest hamlet opened in St-Etienne-des-Orgues in the Basses-Alpes department) and November 15,
1963. An additional 797 jobs were created by January 1, 1965, for a total of 2,189 jobs on the forestry worksites. The
harkis’ reclassement in all other professions accounted for 9,397 jobs between July 1, 1962 and November 15, 1963 and an
67
additional 1,415 jobs by January 1, 1965, for a total of 10,812 jobs.
Percentage of reclassement by number of people
80%
70%
60%
Forestry
worksites
50%
40%
All other
professions
30%
20%
10%
0%
7/1/62-11/15/63
11/16/63-1/1/65
Figure 25: Between September 7, 1962 and November 15, 1963, the government placed 6,081 harkis and their family
members in the forest hamlets. By January 1, 1965, this number had increased by 3,734 to a total of 9,815 people. The
harkis’ reclassement in all other professions affected 23,772 people between July 1, 1962 and November 15, 1963. An
additional 4,531 harkis and their family members were rehoused outside of the forest hamlets by January 1, 1965, for a
68
total of 28,303 people.
The data for this chart is pulled from: AN, F1a 5142, “Ex-supplétifs musulmans passés par les camps et reclassés par
secteurs d’activité à la date du 15 novembre 1963” and CAC, 19980331/11, “Reclassement des rapatriés musulmans par
les antennes civiles des camps d’accueil à la date du 1/1/1965.” A chart in the SFIM archives cites July 1, 1962 as the
date this office began its reclassement of the harkis. CAC, 19920149/2/12, “État des effectifs des ex-supplétifs musulmans
reclassés du 1er juillet 1962 au 1er juin 1963.”
68 Ibid.
67
250
This section examines the government’s contradictory integration policy for the forest
hamlets and reveals why these spaces became the national government’s preferred method of
reclassement for the harkis. By choosing isolated locations, using a regime of encadrement separating and
differentiating the harkis’ from their community, and needing to rely on substantial help from local
officials—some of who were disinterested in the project—to assure the policy’s success, the
government’s integration policy was inherently flawed. The modest potential for this policy to lead
to integration would, nevertheless, not prevent national government officials from praising it, while
simultaneously maintaining a discourse of the importance of integrating the harkis into French
society. Minister of Repatriates Missoffe claimed in May 1963 that his ministry’s integration policy
for the harkis residing in the forest hamlets, created only several months before, was proving to be a
“véritable réussite.”69 This assessment revealed at times a divide between national and local officials
about the success of the forest hamlets. The conflicting evaluations between national and local
officials underscore the problem with relying heavily on local officials, some of who vested little
interest in the harkis’ integration. Meanwhile, the government’s policy of selecting agents to run the
forest hamlets among those who had lived and served in Algeria impeded the harki population’s
integration because these agents treated them as colonial subjects, and not citizens or repatriates as
these agents themselves were classified.
Though inclement weather and construction difficulties delayed the inauguration of thirtyone of the forty-two forestry worksites (representing thirty-seven forest hamlets) the government
intended to open in 1962,70 only one of the forestry worksites in the first phase would never open.71
Indeed, it appeared that the government preferred to open the forest hamlets in poor living
CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…,” 1.
CAC, 19920149/3/2, Ministry of Repatriates (SFIM), “Reclassement des harkis dans le secteur agricole,” 1, undated.
While this document is undated, the author estimates that it was written in February 1963 owing to the verb tenses used
when discussing the number of forest hamlets opened and those to be opened in spring 1963.
71 To determine the number of forestry worksites and forest hamlets in the first phase that opened, the author compared
table 4 with the following document: CAC, 19920149/3/2, “État des monitrices de promotion sociale—Liste des 42
chantiers (1ère tranche),” undated. The one forest hamlet that never opened was Lunas in the Hérault department.
69
70
251
conditions rather than abandon a site, given the difficulty of finding municipal officials willing to
allow the harkis to be domiciled in their towns. For example, the government did not wait until the
houses were completed before opening the Mirande forest hamlet in the Gers department; on
December 4, almost three months after the harkis arrived, they still resided in tents. 72 This
determination demonstrated the increasing importance that the Ministry of Repatriates accorded to
this form of reclassement. By May 15, 1963, 24 first phase forest hamlets sheltered 3,345 members of
the harki population,73 with plans to “extend the forest hamlet experiment,” in Missoffe’s words.74
National government officials—led by Missoffe—predominantly deemed the first phase of
forest hamlets reclassement a success. Missoffe based his assessment primarily on the potential for, and
early signs of, the harkis’ integration into their communities. He argued in a May 14, 1963 letter
asking Minister of Finances and Economic Affairs Valéry Giscard d’Estaing for funds to open
additional forest hamlets: “Toutes les conditions y sont en effet réunies pour permettre une
intégration durable et réelle des anciens harkis dans la communauté nationale” and followed this
statement with a list of conditions.75 The local officials responsible for implementing the forest
hamlet policy did not always share Missoffe’s opinion that the policy would lead to the harkis’
integration. In fact, they sometimes painted a very different picture, with precise examples, about the
forest hamlet’s lack of success in specific domains.
Missoffe’s sometimes inaccurate portrayal of the forest hamlets’ potential for integrating the
harkis into their surrounding communities, the primary goal his Ministry had outlined the previous
fall, can be in part attributed to his objective of convincing Giscard d’Estaing to allocate funds to
construct and hire encadrement agents for additional forest hamlets. Indeed, following Ministries of
CAC, 19920149/3/1, Ministry of Repatriates, “Compte-rendu de la réunion du 4 Décembre 1962 à 10 heures du
matin au Ministère des Armées,” 3, Dec. 4, 1962.
73 CAC, 19910097 /40 (dossier: Harkis – Correspondance 1963), Ministry of Agriculture, “Situation des chantiers
ouverts au 15 mai 1963.”
74 CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…”
75 Ibid., 1.
72
252
Agriculture and Repatriates officials’ approving assessment of the initial forest hamlets, they decided
to open beginning in summer 1963 a second phase of forestry worksites. These would follow the
same model for their setup and organization as the first phase. Whereas the first forty-one forestry
worksites were spread over twenty different departments, all but five of twenty-seventy opened in
the second phase were located in the four departments of the Provence Côte d’Azur region: AlpesMaritimes, Bouches-du-Rhône, Var, and Vaucluse (see table 4). Here, the harki men would work to
make the forests less prone to fires and serve as auxiliary firemen as during the previous five years
forty thousand acres of the Provencal forest had burned. According to Missoffe, this type of work
by the harkis—which he claimed other forestry workers refused to carry out—meant that funding
additional forestry worksites would be in the “intérêt économique national.”76
Department
Name of Forest
Hamlet
Number of Department
worksites
Name of Forest
Hamlet
Number of
worksites
Allier
Noyant-St-Hilaire
1
Juzet-d’Izaut*
1
Alpes-Maritimes
1
Magland*
1
Alpes-Maritimes
Alpes-Maritimes
Alpes-Maritimes
Breil-sur-Roya
(L’Oliveraie)**
L’Escarène**
Mouans-Sartoux**
Roquesteron**
HauteGaronne
Haute-Savoie
1
1.5
1.5
Hérault
Hérault
Hérault
1
2
3
Alpes-Maritimes
Ariège
Aude
Valbonne**
Montoulieu
Narbonne
2
1
1
Isère
Lozère
Lozère
Aude
Aude
Aude
1
1
2
Lozère
Lozère
Lozère
Aveyron
La Pradelle*
St-Martin des Puits*
Villeneuve-Minervois
(Pujol-de-Bosc)*
St-Rome de Cernon*
Avène-Truscas*
Lodève*
Saint Pons
(Plo de Maillac)*
Roybon*
Cassagnas*
Chadenet
(La Loubière)*
Chanac-Cultures*
Mende
Meyrueis*
1
Lozère
1
Aveyron
Basses-Alpes
Brusque*
Jausiers*
1
1
Lozère
PyrénéesOrientales
St-Etienne de
Valdonez*
Villefort*
Rivesaltes*
2
1
1
1
1
1
2
1
The government originally intended to open forty forestry worksites in the second phase, but difficulties finding
housing for the workers’ families forced officials to subsequently reduce the figure to thirty. CAC, 19920149/3/1,
Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans,” 2-3, May 14, 1963.
76
253
Basses-Alpes
Ongles
(St-Etienne des
Orgues)*
St-André des Alpes*
Sisteron*
1
Saône-et-Loire
1
1
Tarn
Tarn
Fuveau
(La Cité Brogilum)**
Jouques
(Le Logis d’Anne)**
La Ciotat
2
Tarn
1
Tarn
2
Var
2
Var
1
Var
1
1
2
1
1
Var
Var
Var
Var
Var
Côte d’Or
Dordogne
Drôme
La Roque
d’Anthéron
(La Baume)**
Chalvignac
(Aynes)*
La Tremblade*
Casamozza
Zonza
Baigneux-les-Juifs**
Is-sur-Tille
(Vernot-Saussy)**
Vanvey-sur-Ource*
Lanmary*
Beaurières*
1
1
1
Var
Var
Var
Drôme
Gard
Dié
La Grand’Combe
1
1
Var
Var
Gard
St. Sauveur du
Pourcil
(Villemagne)*
Mirande*
Montmorin*
Rosans**
3
1
1
1
Basses-Alpes
Basses-Alpes
Bouches-du-Rhône
Bouches-du-Rhône
Bouches-du-Rhône
Bouches-du-Rhône
Cantal
Charente-Maritime
Corse
Corse
Côte d’Or
Côte d’Or
Gers
Hautes-Alpes
Hautes-Alpes
Roussillon-enMorvan
(Glennes)*
Anglès**
Arfons
(Les Escudiers)**
Puycelci
(Grésigne)*
Vaour
1
Bormes
(Les Mimosas)*
Collobrières
(La Capelle)
1
Collobrières
(Le Capelude)*
Gonfaron**
La Londe**
Montmeyan**
Le Muy**
Néoules**
1
1
1
1
Vaucluse
Pignans*
Rians**
St-Maximin
(La Sainte Baume)**
St-Paul-en-Forêt**
St-Raphaël
(Aigue-Bonne)*
Apt**
Vaucluse
Vaucluse
Vaucluse
Total: 26
Cucuron**
Pertuis**
Sault**
Total: 72
1
1
1
Total: 89
1
1
1
1
1
2
2
1
2
1
1
1
1
Table 4: The complete list of the forest hamlets the French government opened in twenty-six departments with the
number of corresponding forestry worksites for each forest hamlet. Fourteen of the seventy-two forest hamlets were
double forest hamlets, housing the families of harki men assigned to two forestry worksites and two of them, Villemagne
and Saint Pons, were triple forest hamlets. The single forest hamlets had twenty-five to thirty-three houses, the double
forest hamlets contained fifty to fifty-six houses, and the Villemagne triple forest hamlet had seventy-eight houses. Some
of the double and triple forest hamlets, such as Valbonne and Saint-Pons, opened initially as smaller forest hamlets,
which were subsequently expanded. In like fashion, other double forest hamlets later became single forest hamlets, such
as Fuveau and la Roque d’Anthéron. The first phase of forestry worksites is marked with one asterisk and the second
phase is marked with two asterisks. Those that are unmarked were opened later. If the forest hamlet is also known by a
name different from that of the village de rattachement, this designation is given in parentheses.77
This table uses the names of French departments from the 1960s, before the Basses-Alpes department was renamed
Alpes de Haute-Provence in 1970 and the Corse department was split into two (Corse-du-Sud and Haute-Corse) in 1976.
77
254
The primary discrepancies between Missoffe’s letter, titled “Extending the forest hamlet
experiment,” and reports from local officials pertained to two of the integration policy’s five
prongs—schooling and interactions between the harkis, on one hand, and local officials and
populations on the other. Missoffe noted the benefits for the entire population’s integration if harki
children attended schools with other French children and affirmed that the harki children had
achieved an “intégration scolaire parfaite.”78 Several reports contradicted this claim. For example,
one submitted five days later by a Ministry of Agriculture official who had visited five forest hamlets
in the Midi-Pyrénées region revealed that while harki children in the two forest hamlets located close
to the nearest town (Juzet d’Izaut and Mirande) did attend municipal schools, those in the three
forest hamlets located farther away from the nearest town (Puycelsi, St-Rome-de-Cernon, and
Brusque) attended school on the campsite.79 Another report from the forestry commissioner in the
Bordeaux region revealed that harki children in the Lanmary (Dordogne) and La Tremblade
(Charente-Maritime) forest hamlets also attended school onsite at this time.80 The overall situation
was, nonetheless, an improvement over that in the Rivesaltes and Saint-Maurice-l’Ardoise camps,
where none of the children attended local schools.
The most glaring discrepancy—and the heart of the integration policy—was Missoffe’s
declaration that the forest hamlets had received an “excellent accueil des municipalités, des autorités
préfectorales et des populations locales.”81 However, accounts from local officials call into question
Appendix D includes an expanded table with information including the forest hamlets’ opening and closing dates, the
number of houses, and the sources from which the information was gathered.
78 CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…,” 2.
79 CAC, 19910097/40, M. Doniol, “Objet: Inspection des chantiers…,” 17.
80 CAC, 19910097/40 (dossier: Harkis – Correspondance 1963), Report from Forestry Expert J. Ducasse (Forest Service
of the Ministry of Agriculture), “Objet: Inspection des chantiers forestiers ‘harkis,’” 1, 3, Ig. nº 660, Apr. 11, 1963.
Children in forest hamlets opened subsequently (including Fuveau, Jouques, and la Roque d’Anthéron in the Bouchesdu-Rhône departments) did attend onsite schools at least for a brief period following the forest hamlet’s inauguration. A
folder marked “Scolarisation des harkis, rapatriés d’Algérie 1962-67” in ADBR, 131 W 377 contains multiple documents
with discussions about mobile classrooms and special teachers for these three forest hamlets.
81 CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…,”
1-2.
255
Missoffe’s wholesale claim. For example, in the same report cited above about the forest hamlets in
the Midi-Pyrénées, the Ministry of Agriculture official noted: “L’accueil des Harkis par les
populations locales a été variable selon les régions.” He explained that some of the mayors “ont
marqué un manifeste recul devant les intentions d’implantation de groupes dans leur commune,” but
this problem was often rectified if the prefect interceded. Moreover, officials in the departmental
repatriates office succeeded in assuaging certain concerns by reminding the mayors that the harkis,
who would have a salary, could be a boon for local businesses.82 Therefore, some of the local
officials who bore a great deal of the responsibility for facilitating the harki population’s
integration—per the government policy for the forest hamlets—were proving reluctant to welcome
these individuals into their communities. These local officials only acquiesced when those with
political power over them intervened or when they were reminded of potential economic benefits
for their constituents. During initial interactions between townspeople and the harkis this same
official disclosed some “minor incidents” (though he divulged no details). The manner in which
officials resolved the problems implied that the harkis were to blame: “Quelques éléments
perturbateurs, rares au demeurant, ont été renvoyés à leur camp d’origine et l’exemple ainsi
administré s’est révélé salutaire.” The report related that local populations subsequently felt
reassured owing to the “surveillance très étroite” exercised over the harki population by a network
of local services—gendarmes, social services, repatriates service, and forestry service. 83 This
statement first suggests the townspeople’s apprehension concerning having harki citizens reside near
them, an attitude indicating that they did not want the camp dwellers to be part of their community.
The statement also shows that local officials separated harki citizens from their neighbors,
demonstrating that encadrement, a core component of the government’s forest hamlet policy, led to
the differentiation—and not integration—of the two groups.
82
83
CAC, 19910097/40, M. Doniol, “Objet: Inspection des chantiers…,” 17.
Ibid., 18.
256
Nevertheless, Missoffe was not alone in touting the initial forest hamlets’ success because
they did resolve—in the short term—some of the major difficulties facing the harki population and
facilitated the type of control and discipline over the harkis that government officials believed was
imperative. One Ministry of Armies official even went as far as to call the forest hamlets the
“formule idéale de reclassement.”84
Officials agreed that this form of reclassement helped to address overcrowding in the
Rivesaltes and Saint-Maurice-l’Ardoise housing camps and allowed the government to move closer
to ending the “univers concentrationnaire.” As Missoffe claimed in his letter to Giscard d’Estaing:
“[L]a vie dans les camps… présente suffisamment de dangers et d’inconvénients de toutes sortes
pour que la nécessité continue d’apparaître de hâter le reclassement des anciens supplétifs.”85 From
December 3, 1962 to March 17, 1963, 35 percent of the harkis from the camps whom the
government rehoused were transferred to forest hamlets. The Ministry of Armies estimated that 45
percent of those whom the government would rehouse from mid-March to July 1, 1963 would be
sent to forest hamlets, including the second phase in the Provencal forest.86 It must be noted,
however, that in his letter Missoffe gave no indication of the harkis’ living conditions in the forest
hamlets, other than to mention that each one contained only twenty-five families and that a joint
Repatriates-Agriculture Commission had chosen new forest hamlets buildings “destinés à améliorer,
compte tenu de l’expérience acquise, la situation sanitaire des hameaux forestiers…”87
SHAT, 1K 744 (dossier: Point de la situation des harkis), Ministry of Armies, “Perspectives de décongestionnement
des Camps de Rivesaltes et de St-Maur l’Ardoise,” 2, Mar. 22, 1963.
85 CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…,” 1.
86 SHAT, 1K 744, Ministry of Armies, “Perspectives de décongestionnement…,” 1.
87 CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…,” 2.
Reports from local officials indicate that living conditions in some of the existing forest hamlets were insalubrious. For
example, the director of the Chanac-Cultures forest hamlet in the Lozère department had sent a complaint to Missoffe
the week before condemning the harkis’ living conditions as “très précaires.” The prefabricated houses contained two
bedrooms, each measuring three meters by three meters, there were no toilets or even outhouses, and the kitchens were
very small. AN, F1a 5137, Alfred Calvière to the Minister of Repatriates, “Objet: Mon affectation à Chanac-Cultures en
qualité du Chef de Hameau,” May 10, 1963.
84
257
The forest hamlets, moreover, offered solutions to many of the problems experienced with
other forms of reclassement. Police reports from across the country in late 1962 and early 1963
confirmed various threats from Algerians of FLN allegiance toward small groups of harkis placed in
factories, which also concerned potential employers.88 For example, a March 1963 letter from an
official in the Bouches-du-Rhône Prefecture to the minister of Repatriates related: “En effet,
l’emploi de petits groupes de harkis dans les grandes usines donne encore des résultats peu
encourageants… Dans les entreprises importantes, les migrants traditionnels continuent, comme par
le passé, à se regrouper par village ou par région d’origine et supportent mal des ‘étrangers’. Lorsque
ceux-ci sont, en plus, des harkis, les rapports deviennent franchement mauvais et même dangereux
pour ces derniers.”89 Therefore, Minister Missoffe argued that because the forest hamlets were
isolated, contained twenty-five harkis who worked among themselves, and were supervised by a
military officer and patrolled by local police, they assured the harkis greater security.90 The May 1963
report about the Midi-Pyrénées region’s forest hamlets cited above confirmed that warnings of
expulsion from France issued by local armed gendarmes to Algerian immigrants worked to stave off
any attacks on the harkis.91
The forest hamlets further addressed the lack of housing options, especially for harkis with
families, encountered even if the SFIM found them jobs. A report consolidating the findings of the
departmental reclassement commissions declared: “…pour l’ensemble des Départements, le leitmotiv
des Commissions de Reclassement est: possibilités d’emplois d’importance variable, mais pas de
CAC, 19920149/1/5a, Minister of State for Repatriates to the Prefect of…, “Objet: Reclassement des anciens
supplétifs musulmans,” 1, Nº 2.073 SFIM/MG, Dec. 7, 1962. Examples of these threats, such as those cited in
introduction to section two of this dissertation, can be found in the SCINA reports located in ADBR, 138 W 4-9. The
SFIM archives include a Sept. 1964 letter from a harki who believed that a factory manager refused to hire him because
he was a harki and Algerians of FLN allegiance were already working there. CAC, 19920149/5/11, Letter from Mr. G.
to the Minister of Repatriates, Sept. 7, 1964.
89 CAC, 19920149/3 (dossier: Travaux de la Commission départementale, Bouches-du-Rhône), IGAME for the 9th
Region (Prefecture of the Bouches-du-Rhône) to the Minister of Repatriates, “Objet: Reclassement des réfugiés
musulmans dans le département des Bouches-du-Rhône,” 2, Nº 03894, Mar. 16, 1963.
90 CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…,” 1.
91 CAC, 19910097/40, M. Doniol, “Objet: Inspection des chantiers…,” 17.
88
258
logements.”92 In several documents, national government officials exposed that the lack of technical
training that harkis, who predominantly originated from the Algerian countryside, possessed (“une
main d’œuvre dépourvue de qualification,” according to Missoffe) rendered it difficult for them to
secure jobs in the industrial sector.93 Moreover, some who were placed in such jobs had difficulties
adapting to this type of work; for example, 29 of the 233 harkis for whom the government found
mining jobs in Valenciennes in fall 1962 quit for this reason.94 In contrast, the type of work the
harkis performed on the forestry worksites was, in Missoffe’s words, “[un] travail parfaitement
adapté aux modes de vie et à la formation socio-professionnelle des intéressés.” 95 Finally, as
explained in the previous chapter, some of the harkis whom the government rehoused faced
difficulties adapting to colder climates and urban settings and subsequently returned to the
Rivesaltes camp. As the map in figure 26 shows, the forest hamlets and forestry worksites were
concentrated in rural southern France—sixty of the seventy-two forest hamlets were located in the
Corse, Languedoc-Roussillon, Midi-Pyrénées, and Provence-Côte d’Azur regions. This milder
climate and rural area arguably would not provoke as much “dépaysement” and therefore allowed
the harkis to “reprendre un genre de vie conforme à leurs traditions ancestrales,” as one Ministry of
Repatriates official put it.96
CAC, 19920149/2/13, SFIM, “Étude des rapports des Commissions Départementales de Reclassement,” 1, undated.
“L’ensemble des Départements” is underlined on the document.
93 CAC, 19920149/1/5a, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans,” 1,
Dec. 7, 1962.
94 ADBR, 138 W 5, SCINA, “Synthèse… journée du 13 novembre 1962,” 5, N˚ 1 790.
95 CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…,” 2.
A Feb. 1963 memo by the SFIM similarly explained, “Le milieu rural dans lequel [les harkis] seront fixés se trouve plus
conforme à leur ancien mode de vie et à leur mentalité et leur adaptation en sera donc plus rapide et plus sûre.” CAC,
19920149/3/2, Ministry of State for Repatriates (SFIM), “Reclassement des harkis…,” 1.
96 CAC, 19920149/3/2, Ministry of State for Repatriates (SFIM), “Reclassement des harkis…,” 1.
92
259
Figure 26: Number of forest hamlets per department. The data for this map comes from the chart in table 4.
97
http://drawmeagraph.com/view-13123442315QF
Finally, encadrement—the type of supervision that government officials insisted these former
colonial subjects needed—was a reason officials often cited for the forest hamlets and forestry
worksites’ success. A March 1963 Ministry of Agriculture internal document included among the
three principles that must be strictly respected in running the forestry worksites: “Maintenir l’unité et
la permanence de l’encadrement technique… Le bon ordre et l’efficacité des chantiers en
This map depicts the present-day French departments. Therefore, Corsica is shown as having two departments instead
of one, which was the case when the government built two forest hamlets on this island.
97
260
dépendent.”98 That same month, in letter to a Ministry of Finance official requesting funding to open
additional forest hamlets, Pérony included a chart that read at the top: “Il est indispensable d’encadrer
les harkis reclassés pendant la période d’adapation.” He then listed the types of officers and the funds
needed for this temporary surveillance.99 And Missoffe himself included “encadrement professionnel,
social et para-militaire bien adapté” in his list of the conditions contributing to the harkis’
“intégration durable et réelle” in the May 1963 letter.100
Despite government officials’ insistence on using encadrement as a central feature of their
integration policy for the forest hamlets, this practice isolated the harki population from other French
citizens and, therefore, impeded its integration into the surrounding community. Officials at all levels
maintained that encadrement was “essential” to keep order inside the worksites and hamlets so that
their inhabitants would not disturb the surrounding communities. When Minister of Armies Messmer
expressed his intention to withdraw soldiers assigned as camp directors at the end of July 1963,
Missoffe relayed Minister of Interior Roger Frey’s concerns: “Il pourrait en résulter alors des heurts
qui déborderaient le cadre du hameau et provoqueraient des inquiétudes au sein des populations
locales.” 101 Frey specified that this discipline must come from not only military officers, but
specifically those who had served in Algeria: “[S]i les harkis se plient facilement à une certaine
discipline militaire émanant de sous-officiers qui connaissent bien leur mentalité et sous les ordres
desquels ils ont servi longtemps en Algérie… par contre ils risquent de ne pas accepter aussi
CAC, 19920149/3/1, Ministry of Agriculture, “Projet de note transmis pour étude à Mr Ballu, ingénieur en chef des
eaux et forêts le vendredi 15 mars 1963,” 1.
99 CAC, 19920149/3/1, Yves Pérony to the Director of General Administration and Budget (Ministry of Finances),
“Objet: Hameaux de forestage. Création de 50 chantiers supplémentaires en 1963,” 3, Nº 667 SFIM/MR, Mar. 18, 1963.
100 CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…,”
1-2.
101 CAC, 19920149/3/2, Minister of Repatriates to the Minister of Armies, “Objet: Encadrement militaire des hameaux
forestiers ouverts pour les supplétifs,” 2, Nº 1088 SFIM/DIR, Apr. 24, 1963. Messmer’s letter to Missoffe can be found
in: CAC, 19920149/3/2, Minister of Armies to the Minister of Repatriates, “Objet: Cadres militaires mis à la disposition
des chantiers de forestage ouverts pour les ex-supplétifs,” Nº 07235 MA/CM, Feb. 18, 1963.
98
261
facilement l’autorité d’agents civils.”102 Frey’s opinions on this matter were particularly significant
because fifteen months later the Ministry of Repatriates would be dissolved and its remaining officials
who handled the harkis would report to him.
The conflict between integration and encadrement demonstrated the difficulties post-imperial
leaders confronted deciding whether the harkis should be included into or excluded from French
society. According to Frederick Cooper, struggles of this type (about “conjugat[ing] incorporation
and differentiation”) were a central feature to “thinking like an empire.”103 As he maintains, colonial
leaders disputed and continually shifted their strategies concerning “[w]here to find a balance
between the poles of incorporation (the empire’s claim that its subjects belonged within the empire)
and differentiation (the empire’s claim that different subjects should be governed differently).”104
While constantly underscoring the importance of the harkis’ integration in its directives regarding the
forest hamlets, the government’s simultaneous use of encadrement, which ineluctably isolated the
population under surveillance, showed a contradiction in the government’s conception of these
spaces. This inconsistent thinking bore a resemblance to “thinking like an empire,” except the
territory where these policies were implemented was in metropolitan France. This distinction is
important because, according to Cooper, the geographical distance between policymakers and the
agents who implemented its policies in the colonies diffused imperial power. This distance made
empires vulnerable to assertions of autonomy, resistance to central authority, and growth of circuits
that bypassed the imperial center.105 While the harkis could leave the forest hamlets at any time, the
fact that the locus of implementation had shifted to the (post-)imperial center limited the possibility
for resistances of all types by those who opted to stay put. Some of these individuals, already
hampered by unfamiliarity with the French territory and language, remained out of necessity—those
CAC, 19920149/3/2, Minister of Repatriates, “Objet: Encadrement militaire…,” 1.
Cooper, Colonialism in Question, 156.
104 Ibid.
105 Ibid., 200-01.
102
103
262
with limited job skills, those with large families, the sick, and the elderly. The harkis in these spaces
were treated as “different subjects” who were “governed differently.”
Encadrement—as conceived by national-level officials and implemented by local agents in the
forest hamlets and forestry worksites—perpetuated colonial structures in these locales through
continuities in the government’s choice of personnel and the methods of governance it advocated
and allowed. The employees that the Ministry of Agriculture hired for the worksites were “Agents
techniques” and “Agents de surveillance” repatriated from Algeria. According to one forestry
commissioner: “Leur connaissance de la langue arabe et des mœurs des Harkis leur donne l’autorité
requise par l’emploi d’une telle main d’œuvre qu’ils ont pratiquée et dirigée en Algérie.” 106 The
inspectors, camp directors, and social counselors that the Ministries of Repatriates and Armies
selected for the forest hamlets were often the same people who had similarly interacted with and
overseen the harkis when they were colonial subjects in Algeria.107 The employment application for
the position of social counselor targeted repatriates from Algeria by asking questions such as on what
date candidates returned to metropolitan France, whether they speak and/or write Arabic and Berber
(including whether they know Arab dialects), whether they prefer to work with Berber or Arab
families, and their “connaissances des Musulmans.”108 The completed dossiers for all three positions
show that almost all of the candidates had served in the military or worked for the Ministry of
Interior as social assistants for rural populations (ASSRA).109 The choice to employ agents with
experience overseeing French Muslim populations in Algeria was not unique to the harkis. For
CAC, 19910097/40, M. Doniol, “Objet: Inspection des chantiers…,” 20. The Ministry of Agriculture archives
contain torn out pages from the July 28, 1962 JORF with the names of agents who had been working in Algeria and
were transferred to various Forestry Commissions in metropolitan France. CAC, 19910097/40 (dossier: Harkis –
Correspondance 1962).
107 AN, F1a 5138, “Note sur la situation des monitrices d’action sociale,” Dec. 2, 1963. According to this document,
forty-eight of social counselors had been recruited by the Ministry of Armies Social Welfare Office, seven by the
Ministry of Health, and one by a private association.
108 Two slightly different applications were found in AN, F1a 5137: “Candidates aux postes de monitrices de promotion
sociale questionnaire” and “Recrutement de monitrices d’action sociale auprès des Français Musulmans refugiés.” The
names the government used for this position included: monitrice de promotion sociale, monitrice de préformation sociale, monitrice
d’action sociale, monitrice d’initiation sociale, and assistante socio-administrative.
109 These applications are located in AN, F1a 5137.
106
263
example, from 1952 to 1965 the Ministry of Interior put into place a corps of “conseillers techniques
pour les affaires musulmanes” (CTAM) in prefectures with a sizeable number of Algerian immigrants
in their departments. The role of these specialists in “Muslim affairs,” chosen for their previous
experience in Algeria, was to “faciliter l’organisation de l’assistance morale, matérielle et sociale à la
population musulmane… et l’adaptation des citoyens français musulmans en métropole et [de]
réaliser leur promotion sur le plan professionnel et social.”110
Practically speaking, choosing individuals who knew Arabic and Berber, were familiar with
Algerian ways of life, and had previously supervised Algerian workers facilitated communication with
the harki families. However, having recently migrated from a society where “French Muslims” were
inferior citizens necessarily influenced these agents’ mentalities and their resulting interactions with
the harkis.111 Their past had socialized them into a system of colonial domination. In a special issue of
the journal Politix, Françoise de Barros and Tom Charbit edited five articles, each of which focuses
on a post-imperial situation of continuities in personnel and practices of former colonial
administrators over Algerian populations in metropolitan France. The articles illustrate their thesis
that one specificity of the end of colonial Algeria was “l’arrivée en métropole de plus d’un million de
personnes qui emportent avec elles des manières d’être et de faire forgées dans une autre situation
que celle qu’elles vont désormais affronter.”112 It is not possible to quantify precisely how many of
the encadrement agents regarded the harki population under their supervision as inferior. However, the
fact that the government decided to employ former residents of Algeria to run the forest hamlets
fostered and prolonged the hierarchical society the harkis experienced in colonial Algeria and opened
Quoted in: Viet, La France immigrée, 180. For an analysis of the administrative practices of the CTAM, see Françoise de
Barros, “Contours d’un réseau administratif ‘algérien’ en construction d’une compétence en ‘affaires musulmanes’: Les
conseillers techniques pour les affaires musulmanes en métropole (1952-1965)” Politix 19, no. 76 (2006): 97-117 and
Arthur Grosjean, “L’action des conseillers techniques aux Affaires musulmanes. L’exemple du camp de Thol,” Matériaux
pour l’histoire de notre temps no. 92 (Oct-Dec 2008): 15-24.
111 In his article “L’histoire des mentalités” Philippe Ariès argues that the evolution of individuals’ mentalities is a slow
process that takes places over a “longue durée.” Ariès, “L’histoire des mentalités,” in La nouvelle histoire, ed. Jacques Le Goff
(Bruxelles: Éditions Complexes, 1988), 167-68.
112 Françoise de Barros and Tom Charbit, “La colonie rapatriée,” Politix 19, no. 76 (2006): 5.
110
264
the door to such treatment.113 For example, a harki residing in la Roque d’Anthéron forest hamlet
wrote a letter to President Pompidou in December 1968 in which he related harsh conduct by the
new forestry worksite director who had worked in Algeria. According to the harki, “[D]epuis l’arrivée
d’un nouveau chef de chantier M. BIRRITO des incidents continuels ont éclaté sur le chantier et cet
homme est vraiment très dur avec nous, nous traitant avec injustice, sans cesse il nous dit : ‘Si tu n’es
pas content tu n’a [sic] qu’à retourner en ALGERIE, ou bien je t’écrase.’”114 When these agents
repatriated from Algeria acted as though they were superior to the harkis and exhibited similar
domineering behavior toward them as they had in Algeria, this treatment distinguished the harkis
from their neighbors, which in turn impeded their integration. It must be underscored, nevertheless,
that while agents who had lived in Algeria brought with them colonial practices, the belief that the
harkis were second-class citizens was not exclusive to repatriated individuals. For example, in a May
1963 report about the functioning of the five worksites under his supervision the forestry
commissioner for the Midi-Pyrénées region critiqued the harkis for accomplishing only three-quarters
of the work of “ouvriers habituels” owing to their “prédisposition au ‘farniente’ et… leur esprit
revendicatif.” 115 The harkis’ questionable work habits and productivity would be cited in future
discussions concerning how long to keep open the forestry worksites among Ministry of Agriculture
officials.
In addition to continuities with the individuals who would encadrer the harki population,
national officials’ directives concerning the forest hamlets revealed continuities with colonial methods
of governance on both sides of the Mediterranean. The twin components of the government’s social
Government officials were, of course, aware of the dynamic between Europeans and indigenous populations in
Algeria. In a heated exchange during the May 4, 1962 Council of Ministers meeting about the disregard for the Evian
Accords by Muslim and European populations alike in Algeria, President de Gaulle offered his assessment of the present
relationship between the two groups: “Les Européens? Depuis cent trente ans, ils ont pris l’habitude de dominer les
musulmans, ils ont pensé que la France serait toujours là pour les protéger et assurer leur supériorité…” Quoted in
Peyrefitte, C’était de Gaulle, 139.
114 ADBR, 135 W 71 (dossier: Algérie 1965-71), letter from Mohamed B. to President Pompidou, Dec. 14, 1968.
115 CAC, 19910097/40, M. Doniol, “Objet: Inspection des chantiers…,” 20.
113
265
welfare policy specifically aimed at Algerian immigrant families residing in France during the post
Second World War era had been integration and surveillance.116 The SFIM addressed a memo to the
two government agents—a camp director and a social counselor—who lived on-site with the harkis,
with specific instructions of their roles, similar to these earlier social welfare policies. The camp
director, often a current or former military officer, was instructed to oversee the salubrity of the
forest hamlet and building upkeep; act as the interlocutor with town officials for individual harkis’
administrative matters (education, social welfare, pensions); aid and support individual population
members in their interactions with municipal services; protect them from possible threats from FLN
operative; keep order among the inhabitants; and “apporte[r] à l’assistante sociale le poids de son
autorité” should she encounter problems with residents. 117 The Ministry of Repatriates memo
concerning the social counselor described her primary function as “un travail socio-éducatif.” It listed
as her six tasks: train the women how to take care of their households; teach them about childcare;
offer language classes to women and older adolescents who were not enrolled in school; provide
basic first aid; ensure that the children attend school and become involved in extracurricular activities;
and aid the families with obtaining social services from state agencies.118 Whether a forest hamlet
housed the families of harkis who worked at one worksite (approximately twenty-five to thirty men)
or those who worked at two worksites (double this figure), the forest hamlet still had a maximum of
two government agents. (Later on, some forest hamlets would only have a social counselor owing to
budgetary reductions.)
Certain aspects of encadrement were helpful in aiding harki citizens adjust to life in France
owing to language barriers, unfamiliarity with how to run a household in France, and inexperience
Lyons, “Invisible Immigrants,” 4-7.
CAC, 19920149/3/1, Minister of State for Repatriates (SFIM), “Consignes du sous-officier détaché au chantier de
forestage.”
118 AN, F1a 5138, SFIM (Ministry of Repatriates), “Role et attributions de la Monitrice de Formation Sociale affectée sur
les hameaux forestiers.”
116
117
266
with French government agencies and laws. Moreover, the military officer’s presence would protect
them from possible attacks from other Algerians, which were only initially a concern. However, these
directives separated the harkis from their neighbors, sometimes instructed the social counselor and
camp director to govern by authoritarian means, and diminished the possibility for the harkis to
interact with local officials. This method of governance resembled in ways colonial France’s
“civilizing mission.” The memos further left much up to interpretation for carrying out the state’s
policies. Particularly given the forest hamlets’ isolation, which also meant limited oversight of its one
or two agents, the government’s conception of encadrement concentrated much power into the on-site
agents—responsible for the one hundred to two hundred residents who had difficulties speaking
French and were unfamiliar with French customs and culture.119 While the social counselors’ memo
emphasized that to facilitate the harkis’ integration, they should “help,” “accompany,” “teach,” and
“verify” rather than “do,” implementation would necessarily be variable based on individuals.120
Envisioning Permanent Forest Hamlets and Ending the Ministry of Repatriates’ Mission
While Minister of Repatriates Missoffe praised the “forest hamlet experiment” in his May
1963 letter to Giscard d’Estaing, he nonetheless cautioned, “…une des préoccupations essentielles
du Gouvernement [est que] tous les efforts doivent être accomplis pour éviter que se constitue une
population oisive s’installant définitivement dans une misère subventionnée.” 121 Yet, as table 5
illustrates the forest hamlets would continue beyond a temporary situation, leaving their
inhabitants—numbering nearly five thousand in December 1975, over thirteen years after the first
forest hamlet had opened as a temporary solution—to languish in a “misère subventionnée.” Soon
A March 22, 1963 report written by the Ministry of Armies revealed that the average population in the forest hamlets
for the initial 22 forestry worksites opened was 110 people. SHAT, 1K 744, Ministry of Armies, “Perspectives de
décongestionnement,” 3. This figure would increase markedly as the size of the harkis’ families grew.
120 AN, F1a 5138, SFIM, “Role et attributions….”
121 CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…,” 2.
119
267
after this time population data for the forest hamlets becomes unreliable, even though most
remained open at least until the early 1980s. 122 Therefore, in May 1963, the “forest hamlet
experiment” was far from over. This section analyzes why the Ministry of Repatriates’ initial vision
for the forest hamlets changed from a “temporary” to a “permanent” solution. It moreover
interprets what this evolving vision for the forest hamlets reveals about how government officials
regarded the status of the harki citizens residing in these spaces.
Date
Number
of forest
hamlets
Dec. 1962
Mar. 1963
May 1963
Oct. 1963
Nov. 1963
Feb. 1964
Jan. 1965
Jan. 1966
Jan. 1967
Jan. 1968
Jan. 1969
Jan. 1970
Jan. 1971
Jan. 1972
Jan. 1973
July 1973
Jan. 1974
Jan. 1975
Dec. 1975
10
20
24
40
----60
52
----------40
--36
31
31
29
Number
of forestry
worksites
Number
of families
Number
of workers
Number of
dwellings
Number of
inhabitants
11
22
28
--48
60
72
65
------45
--49
-----------
282
------------1,375
1,161
1,236
1,128
1,134
1,011
905
817
784
714
680
584
--------1,392
----1,438
------1,230
--1,026
--846
766
823
---
307
--733
1,054
----1,982
-------------------------
1,425
2,420
3,345
5,082
6,081
7,912
9,720
--7,546
7,519
7,344
7,463
6,795
6,343
5,997
5,877
5,391
5,275
4,645
Table 5: Forest hamlets and forestry worksites open from December 1962 through December 1975 with the number of
families, dwellings, workers, and inhabitants residing in the forest hamlets. Table cells marked “---” mean that the
information is unavailable. This table is reproduced in Appendix E, which provides the sources for this information.
An important modification to the government’s initial vision for the forest hamlets began to
emerge in early 1963. From the March 1962 Council of Ministers meeting when the government first
discussed housing the harkis in forest hamlets through Missoffe’s letter to Giscard d’Estaing in mid-
Charts with the number of forest hamlets and their population from the late 1970s onward show various anomalies.
For example, a January 1977 chart fails to include the Fuveau forest hamlet whose buildings had been reconstructed on
the same isolated site, yet does list the Roque d’Anthéron forest hamlet, which had closed on February 1, 1976. The
reasons for these discrepancies will be addressed in the next chapter. ACNMF, 16, “Hameaux Forestiers,” Jan. 10, 1977.
122
268
May 1963, government officials crafted a fairly consistent vision of these spaces—with one exception.
The duration of the “forest hamlet experiment” would change. A memo issued in late September
1962 by the SFIM branch of the Ministry of Repatriates summarizing its reclassement efforts explained
the set up for the forest hamlets: lightweight prefabricated houses, encadrement by soldiers and civilians
who had lived in Algeria, groups of twenty-five men working only among other harkis, and isolation
of the harki families’ living spaces. The memo also affirmed the temporary nature of the forest
hamlets: “Ces chantiers, sans avoir un caractère permanent, doivent procurer du travail pour une
durée de deux ans aux harkis, délai pendant lequel le Ministère de l’Agriculture se préoccupera de
trouver d’autres travaux pour les intéressés.”123 Therefore, initially the set-up of the forest hamlets
was planned with the intention of the forest hamlets being a temporary solution.
When the government’s vision for the duration of the forest hamlets would begin to change,
this modification would not precipitate a concomitant adjustment to its vision for their setup and
operation. As explained above, Ministries of Agriculture and Repatriates officials decided to open a
“second phase” of forest hamlets, most of which would be concentrated in the Provencal Forest,
where harki men would work toward fire prevention and be auxiliary firemen. On January 16, 1963,
Minister Pisani sent to the prefects of the four departments in the Provence Côte d’Azur region the
same memo with information about selecting the sites and how the forest hamlets would function as
he had issued in summer 1962, except without the sentence “La durée des travaux devra être au
minimum d’une année.” 124 No timetable was indicated for these new forest hamlets, but their
conception was otherwise exactly the same. A subsequent memo on February 23 from the Ministries
of Agriculture and Repatriates to departmental prefects summarized “les principes essentiels qui
président à l’installation et au fonctionnement de ces chantiers,” much like the three memos sent in
CAC, 19920149/1/5, SFIM, “Note sur les harkis et les moghaznis,” 4.
CAC, 19910097/40 (dossier: Harkis – Correspondance 1963), Minister of Agriculture Pisani to the prefects of the
Vaucluse, Var, Alpes-Maritimes, and Bouches-du-Rhône departments, Jan. 16, 1963.
123
124
269
fall 1962 by the Ministry of Repatriates. This February 1963 document retained the same elements of
the previous documents in terms of encadrement, selecting a site, the construction of houses, financing
the forest hamlets, and a sentence about integrating the harkis. Yet, unlike the fall 1962 documents, it
made no mention of the forest hamlets being temporary. In fact, it made no reference whatsoever to
the time frame.125 Missoffe’s May 1963 letter to Giscard d’Estaing asking to fund the second phase of
forest hamlets similarly omitted any reference to the temporality of forest hamlets.126 In what appears
to be a step in the opposite direction, in October 1963 SFIM Director Pérony explicitly referred to
the forest hamlets as a transitional step. As cited above, Pérony in a report from the Ministry of
Repatriates maintained that the forestry worksites aimed to reclasser members of the harki population
“dans un stade transitoire.”127 This suggests that Ministry of Repatriates officials, at least briefly,
vacillated in their thinking about incorporating or differentiating the harkis through their living spaces.
However, on January 7, 1964, Pérony, on behalf of Minister Missoffe, wrote to the Minister
of Agriculture a letter in which he explicitly signaled that Missoffe no longer viewed all the forest
hamlets as temporary. In so doing, he lessened the Ministry of Repatriate’s responsibility toward the
population. He asked the Forestry Commission director to place the forestry worksites into one of
two categories:
-…des chantiers qui doivent permettre un reclassement définitif des harkis et recevoir à
ce titre un complément d’équipement pour améliorer l’habitat et le mode de vie des familles
appelées à y vivre de nombreuses années.
-…des chantiers qui risquent d’être fermés en raison de la précarité et de l’importance
relative de travaux forestiers ou par suite de la sévérité des conditions de vie existantes ou de
la proximité d’emplois plus rémunérateurs dans un secteur voisin.128
CAC, 19920149/3/1, Ministry of Agriculture and Ministry of Repatriates, “Objet: Deuxième tranche de construction
de chantiers de forestage pour harkis,” Feb. 23, 1963.
126 CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…,” 2.
127 ACNMF, 15/13, Pérony, “Problèmes posés,” 1.
128 CAC, 19910097/40 (dossier: Harkis – Correspondance 1964), Minister of Repatriates to the Minister of Agriculture
(Director General of the Forestry Commission), “Objet: Avenir des hameaux et chantiers forestiers,” Nº 58 SFIM/MR,
Jan. 7, 1964.
125
270
This letter did not include the word “intégration,” or any synonym of it. Some of the forest hamlets
would now represent a definitive—and not temporary—reclassement for the harkis, and the only
change that Missoffe indicated making was to add supplementary equipment for the families. This
partial policy shift engendered another fundamental inconsistency with the government’s forest
hamlet policy: a permanent living situation in temporary living conditions. Continuing to house the
harkis in locations chosen for their isolation—the Ministry of Agriculture had specifically instructed
local officials to look for sites “en dehors des agglomérations”—prolonged these harki citizens’ exile
and their treatment by the French government as refugees, and not repatriates.129
Seven and a half months after Missoffe’s letter, on July 23, 1964, the Ministry of Repatriates
would cease to exist and the management of the harkis’ reclassement, still directed by Yves Pérony at
the SFIM, would subsequently fall under the purview of the Minister of Interior. At the June 10, 1964
Council of Ministers meeting, Missoffe proclaimed: “Les rapatriés, c’est fini. Il reste seulement le
règlement de quelques cas sociaux.”130 From these words alone, it is not completely certain that
“quelques cas sociaux” exclusively referred to the pieds-noirs. However, based on available source
material, Missoffe—like other Ministers and President de Gaulle—used “rapatriés” to denote
specifically the French repatriates in his previous contributions at Council of Ministers meetings.131
Further, de Gaulle and Pompidou’s responses to Missoffe’s statement that day focused only on the
pieds-noirs. Missoffe’s declaration begs the question of where and how the harkis fit into the
definition of repatriates. President de Gaulle retrospectively claimed in July 1964 that Algeria was
CAC, 19910097/40, Chief Forestry Expert Jolain, 1.
Quoted in Peyrefitte, C’était de Gaulle, 733.
131 See, for example, Peyrefitte, C’était de Gaulle, 409 and 733. According to Alain Peyrefitte, during the Feb. 13, 1963
meeting Missoffe delivered a report that exclusively treated French repatriates. De Gaulle reminded him: “Ne reniez pas
votre titre. Vous n’êtes pas ministre des pieds-noirs, vous êtes ministre des rapatriés” (733).
129
130
271
French “dans les fictions juridiques,” which was demonstrated in section one.132 Similarly, it appears
that the harkis were only repatriates “dans les fictions juridiques.”
When the Secretariat of Repatriates was created in May 1961, its function was to develop
and oversee the implementation of an “arrival” and “settling in” policy for repatriates. While its
efforts and resources overwhelmingly concentrated on the pieds-noirs, the responsibilities of this
Secretariat (and subsequently Ministry) included the reclassement of harki families. When de Gaulle
eliminated the Ministry of Repatriates in July 1964, harkis who had recently been released from
prison were still arriving from Algeria; the Red Cross estimated in January 1965 that 2,500 harkis
were imprisoned and 20,000 had been freed.133 And, as late as October 1969, 1,500 recently liberated
harkis migrated to France. 134 Moreover, in July 1964, 3,500 harkis and their family members
continued to reside at the Rivesaltes refugee camp, nearly 1,000 “unhouseables” lived at the Bias
arrival camp, and the forest hamlets—set up to be an intermediary stage to the harkis’ reclassement—
housed at least 8,500 people (see figure 27 and table 5).135 The Ministry of Repatriates would
therefore disappear with fourteen thousand of the harki population living in refugee conditions—
not “settled in”—and an indeterminate number in Algeria potentially to “arrive.” While Missoffe
was generally loath to include the harkis into the definition of “repatriates,” his Ministry nevertheless
was responsible for their reclassement. If the forest hamlets were classified as “temporaire” or
“transitoire” when Missoffe’s mandate ended, then his Ministry would not have been successful
integrating into French society the 8,500 plus harkis residing in these spaces and those in the
Quoted in Peyrefitte, C’était de Gaulle, 734. De Gaulle made this statement after the July 22, 1964 Council of Ministers
meeting, the final one before he eliminated the Ministry of Repatriates.
133 MAE, SEAA 155, “Le sort des harkis,” Feb. 12, 1966. This report predicted that not all of those released from prison
would choose to migrate to France.
134 CAC, 19880077/20, “Le Problème des Harkis,” 1, Mar. 12, 1974. This unsigned six-page report is located in the
Ministry of Interior archives.
135 The figure of at least 8,500 is estimated given the forest hamlet population of 7,912 on Feb. 1, 1964 and adding the
populations of the two forest hamlets that the author is certain opened in spring 1964, those in Fuveau and la Roque
d’Anthéron, each of which had initial populations of nearly 250 people. For the Bias camp population, see: Lanoizelez,
La CARA de Bias, 130. For the Rivesaltes camp population, see: CAC, 19920149/1/3.
132
272
Rivesaltes camp whose transfer was imminent. “[I]ntégrer les Français rapatriés dans les structures
économiques et sociales de la nation” was, after all, the goal of the December 1961 law for (all)
repatriates. Initiating a policy that would characterize some of the isolated forest hamlets as
“définitifs” allowed the French government to move closer to turning another page of the “Algerian
Affair.”136 Government institutions and officials would continue to distance themselves from—and
try to forget—reminders of its lost territory and empire.
Number of Forest Hamlet Inhabitants
10,000
9,000
X
8,000
7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
Figure 27: Number of forest hamlets inhabitants from December 15, 1962 to January 1, 1965. The “X” represents the
date that the Ministry of Repatriates was eliminated, July 23, 1964. Source: The data for this chart is from table 5.
Shifting Responsibilities and a Permanent Forest Hamlet Policy
Soon after the Ministry of Interior took over the Ministry of Repatriate’s responsibility for
organizing harkis’ reclassement and providing encadrement agents the forest hamlets in July 1964, Minister
of Interior Frey not only accelerated discussions about making the forest hamlets permanent but also
CAC, 19910097/40 (dossier: Harkis – Correspondance 1965), “Liste des hameaux de forestage ouverts le 1er janvier
1966.”
136
273
sought to rid his ministry of any role in managing the forest hamlets. Frey’s actions were the final
step toward ensuring that the forest hamlets would be permanent and the harkis residing there would
become “une population oisive s’installant définitivement dans une misère subventionnée,” of which
Missoffe had warned in May 1963 must be avoided.137 Making the forest hamlets permanent lessened
the work of the SFIM because these spaces now would be available to rehouse the newly-arrived
harkis who resided in the “Centre de Transit et de Transit de Lascours,” opened in April 1965 and
run by Frey’s ministry.138 As SFIM Director Pérony wrote in a January 1968 letter, the forest hamlets
were “notre principale source de reclassement.”139 Indeed, in order to close this center in 1970, the
government outlined a plan in early 1969 to place its remaining population of 370 people in the forest
hamlets in Provence and Corsica.140 This section examines the influence that the shift from temporary
to permanent forest hamlets, which ran counter to the initial goal of integrating the harki population,
had on the government’s policies for these spaces, and argues that the stagnations in government
policies influenced the harki population as much as the shifts. It further demonstrates that making
the forest hamlets permanent allowed government officials to distance themselves from the harki
population, whom the government increasingly viewed as immigrants in need of social welfare
assistance, and not repatriates in need of the indemnities and allocations to which the Boulin law and
subsequent legislation entitled them.
On March 12, 1965, Frey wrote to Pisani a letter that his cabinet director Jacques Aubert
signed on his behalf, and possibly wrote. The fact that these two Interior ministry officials had now
assumed the responsibility for the government’s reclassement policies for the harkis is significant. As
CAC, 19920149/3/1, Minister of State for Repatriates, “Objet: Reclassement des anciens supplétifs musulmans…,” 2.
AN, 5 AG 1/22, “Note concernant le service…,” May 6, 1966. In 1967 the Ministry of Social Affairs took over the
administration of this center.
139 CAC, 19910097/41 (dossier: Harkis – Correspondance 1968), letter from SFIM Director Yves Pérony to the Minister
of Economy and Finances (to the attention of the deputy cabinet director), 2, Nº 2215/SFM, Jan. 19, 1968.
140 CAC, 19910097/41 (dossier: Harkis – Correspondance 1969), Minister of Agriculture to Minister of State for Social
Affairs, “Objet: Emploi sur les chantiers forestiers du Ministère de l’Agriculture de 370 ex-harkis actuellement hébergés
au centre d’accueil et transit de Lascours (Gard),” F2/3 nº 00139, Feb. 3, 1969.
137
138
274
explained in chapter two, during the spring of 1962, both men had tried to deter the repatriation to
France of harkis threatened by violence. Aubert, who had led the police forces in Algeria for two
years and had recently returned to France to serve as the director of the National Police Force,
demonstrated little concern for the harkis’ safety in the face of FLN threats. In an April 25 letter to
Michel Massenet, he claimed that only a few hundred harkis would need to be repatriated and
maintained that the newly-formed Algerian government—and not the French Army—should curb
any violence that arose among “les familles politiques musulmanes.” Moreover, Aubert demonstrated
a supercilious attitude toward the harki population in the letter by calling them “‘buveurs de
soleil.’”141 For his part, Frey’s primary concern vis-à-vis the harkis in spring 1962 was not to advocate
creating a repatriation plan, even though FLN violence escalated in May and the government had no
procedure in place to provide the population with safe passage to France. Rather, in a succession of
telegrams and directives, he took a firm stance against the initiatives individuals took to repatriate
harkis, emphasizing that most of the people repatriating the harkis were politically dangerous.142
These men’s attitudes toward the harkis at this time demonstrated their desire to lessen the French
government’s responsibilities toward the population.
Similarly, in the 1965 letter to Pisani, Frey (and Aubert) proposed rethinking the forest
hamlets initial mission as a “formule transitoire,” which would lessen the Ministry of Interior’s
responsibility toward finding permanent jobs for the harkis. He argued that given the “indispensable”
nature of the harki workforce to the Forestry Commission, the fact that many of them were proving
to be good forestry workers, and their families seemed to be integrating little by little into their local
communities, it was logical to make the forest hamlets a definitive form of reclassement. Then came the
catch:
CAC 19910467/2, letter from Jacques Aubert to Michel Massenet, Apr. 25, 1962. See pages 88-89 of this dissertation.
SHAT, 1K 744 (dossier: Directives concernant les harkis), letter from Minister of the Interior Roger Frey to Minister
of Armies Pierre Messmer, 2, N° SN CAB 2924, May 15, 1962. See pages 109-10 of this dissertation.
141
142
275
Autant il était bien dans la vocation du Ministère des Rapatriés de prendre en charge,
pendant une période transitoire, ces travailleurs musulmans traumatisés par les évènements et
nullement adaptés à nos modes de vie et de pensée, et ceci pour vous permettre de les utiliser
valablement tout en leur donnant une formation technique suffisante, autant le Ministère de
l’Intérieur, lui, ne saurait, maintenant qu’une évolution certaine s’est manifestée dans le
comportement de ces populations et que vous avez fait de beaucoup de ces travailleurs des
ouvriers qualifiés de la Forêt, contracter, à l’égard de Français qu’on ne saurait traiter
beaucoup plus longtemps en citoyens mineurs, des obligations spéciales qui ne relèvent
nullement de ses attributions normales.143
Once again, similar to the view the Minister of Repatriates expressed one year before, making the
forest hamlets a permanent type of reclassement would release the Minister of Interior from any duties
toward the nearly ten thousand harkis residing in sixty forest hamlets at this time (see table 5). In this
rambling sentence he lauded simultaneously the harki population, whom he believed should no
longer be treated as lesser citizens (at least for his current purposes), for the evolution in its behaviors
and Ministry of Agriculture agents for transforming the harkis into trained forestry workers.144
Frey’s praise for these agents prefaced his ultimate proposal: to streamline the management of
the forestry worksites and the forest hamlets to fall solely on the Ministry of Agriculture. Labeling the
harkis as forestry workers, and good ones at that—and no longer repatriates—allowed Frey to
distance his Ministry from the population, for which the government charged him with finding
permanent jobs and housing. He wrote, “Je ne pense pas, par contre, qu’il soit ni rationnel ni
souhaitable, à partir du moment où l’aspect professionnel prend le pas sur l’aspect Rapatriés, de
confier à des Administrations différentes d’une part la direction des travaux forestiers, d’autre part la
gestion des hameau [sic] (que vos Services du Génie Rural connaissent d’ailleurs fort bien pour les
avoir construits).” Contradictorily, despite Frey’s desire to minimize the number of ministries
involved with overseeing the forest hamlet populations, as part of this new plan he offered to ask the
minister of Public Health and Population to take over the responsibility of providing social
CAC, 19910097/40 (dossier: Harkis Correspondance – 1965), Minister of Interior Frey (signed by Cabinet Director
Jacques Aubert) to Minister of Agriculture Pisani, “Objet: Reclassement de rapatriés musulmans sur des chantiers
forestiers,” 2, Nº 828/SFIM/DIR, Mar. 12, 1965.
144 Ibid.
143
276
counselors as he insisted that these “familles musulmanes” would still require social welfare assistance
“pendant de longues années.” As he argued, such responsibilities indeed fell under the mission of this
ministry, which was charged with helping other Muslim immigrant families adapt to life in
metropolitan France. Frey ended his letter with a threat: if the Minister of Agriculture did not accept
his proposal and the forest hamlets remained under his jurisdiction, “je me verrais alors dans
l’obligation de ne négliger aucune des autres formes de reclassement.”145 In other words, in one way
or another the harki population would end up in the forest hamlets.
Frey would wait three months for a reply. In the interim, he shot off another letter to Pisani
in which he added “ouverts par le Ministère de l’Agriculture” to the original subject line,
“Reclassement de rapatriés musulmans sur des chantiers forestiers,” seemingly to underscore this
ministry’s principal role in the affair.146 As demonstrated earlier in this chapter, the opening of the
forest hamlets and forestry worksites was a join effort between the Ministries of Agriculture and
Repatriates. When Pisani did respond, on June 16, he declined Frey’s proposal and presented a plan
to close all the forestry worksites, starting with those outside of the Provencal forest, by 1975. Pisani
argued that the work the harkis were doing did occupy not high enough priority to justify
permanently including them in the National Forests Office (ONF) budget. He added that he would
reconsider his decision for the forestry worksites in the Provence region where fires were a risk only
if Frey secured special financing that did not pull from the his ministry’s budget. Pisani did
nevertheless believe that it would be possible for his ministry to offer permanent jobs in the Alsace
and Moselle departments to approximately one hundred of the best harki workers, for whom he
would build houses, though not provide encadrement. Concerning Frey’s request to supply camp
Ibid., 3-4.
CAC, 19910097/40 (dossier: Harkis Correspondance – 1965), Minister of Interior Frey (signed by Cabinet Director
Jacques Aubert) to Minister of Agriculture Pisani, “Objet: Reclassement de rapatriés musulmans sur des chantiers
forestiers ouverts par le Ministère de l’Agriculture,” 1, Nº 1229/SFIM/DIR, Apr. 13, 1965. In this letter, Frey also wrote,
and underlined, “[J]’aimerais que vous me donniez néanmoins votre accord de principe sur l’adoption de cette nouvelle
formule.”
145
146
277
directors for the forest hamlets, Pisani maintained that owing to the “caractère social primitif” of
these spaces, his agents would not be qualified to oversee them. He suggested finding camp directors,
in addition to the social counselors, among employees of the Ministry of Public Health and
Population.147
Six weeks later Frey wrote a letter, again signed—and perhaps written—by his cabinet
director Aubert, in which he urged Pisani to reconsider his position and used Pisani’s previous
actions and words to try to sway him. He also revealed his opinion on the function of the forest
hamlets vis-à-vis the harkis’ integration. In this July 30, 1964 letter, Frey downplayed the Ministry of
Repatriates’ “strictement limité” role in the setup of the forestry worksites. He maintained that this
strictly limited role included its agents studying measures “susceptibles de faciliter et d’accélérer
l’intégration des musulmans dans leur nouveau milieu” and putting into place a structure of “léger
encadrement.” As Frey—who had not been involved in the process in 1962—explained it, the
integration of the “Muslims” was therefore an ancillary factor in opening the forest hamlets. This
statement indicates that Frey viewed the forest hamlets as foremost a place of work; the forest hamlet
policy for him was not an integration policy. Moreover, his choice of the word “musulmans,” a
commonly used term to refer to native Algerians’ nationality status during the colonial era, also
perpetuated the harkis’ inferior colonial classification. Frey reminded Pisani that the original worksites,
and consequently forest hamlets, were opened thanks to his “louable initiative” on grounds selected
by Ministry of Agriculture agents. Therefore, Frey argued that the former Ministry of Repatriates,
whose functions he had assumed, should not be held responsible if the forestry worksites were not
situated in priority work zones. Frey also attached a copy of Pisani’s January 16, 1963 memo to
prefects in the Provence region and referenced this document writing to Pisani: “[V]ous avez tenu
CAC, 19910097/40 (dossier: Harkis Correspondance – 1965), Minister of Agriculture Pisani (signed by Cabinet
Director Jean Vaudeville) to Minister of Interior Frey, “Objet: Reclassement de rapatriés musulmans sur des chantiers
forestiers ouverts par le Ministère de l’Agriculture,” 1-2, Nº 3630 EF/D4, June 16, 1965.
147
278
personnellement… à leur préciser que ces chantiers seraient ouverts à l’initiative et sous l’autorité des
Conservateurs des Eaux-et-Forêts.” Frey agreed with Pisani’s suggestion to progressively close the
twenty-nine forest hamlets located outside of Provence. However, he included sentences from
Pisani’s previous letters in which the Minister of Agriculture had praised the harkis’ fire prevention
work as “greatly beneficial” and Frey referenced the recent fires that had ravaged regions in the Var
and Alpes-Maritimes departments. He argued that due to these facts and that the harki labor would
be “difficult to replace” (because it would be hard to find laborers who would work for such low pay),
Pisani should make room in his own budget for the eight hundred harki workers on the twenty-three
forest hamlets located in Provence. Frey’s final proposal was for Pisani to use the money he intended
for building houses for one hundred permanent harki forestry workers in order to make
improvements on existing dwellings. He disagreed with Pisani’s proposal because “elle ‘écrèmerait’ les
meilleurs éléments des chantiers… [et] ces rapatriés musulmans livrés à eux-mêmes et sans
encadrement retournerait rapidement à leur oisiveté et à leur instabilité naturelles.”148 Whether these
are Frey or Aubert’s words, they demonstrate a condescending attitude toward even the “best” of the
harki population, with whose integration Frey seemed unconcerned.
After exchanges between officials in the Agriculture and Interior Ministries during the fall of
1965, in a December 2 letter Pisani ceded to Frey about making the Provence forestry worksites
permanent, but he firmly held that his Ministry would not assume the responsibility for running the
forest hamlets. He also presented Frey with the plan developed at an October 21 meeting of SFIM
and Forestry Commission officials to gradually reduce the forest hamlets located outside of Provence
CAC, 19910097/40 (dossier: Harkis Correspondance – 1965), Minister of Interior Frey (signed by Cabinet Director
Jacques Aubert) to Minister of Agriculture Pisani, “Objet: Reclassement de rapatriés musulmans sur des chantiers
forestiers,” 1-4, Nº 1508 CAB/IX, July 30, 1965.
148
279
over a six-year period and agreed to transfer the remaining construction funds in his budget to
enlarge and renovate houses in existing forest hamlets slated to become permanent.149
On January 13, 1966, Aubert presided over an interministerial meeting during which a
definitive plan emerged that officially changed the status of the “forest hamlet experiment” to
permanent.150 Instead of reducing the number of ministries involved with the forest hamlets to
streamline their management, this number grew to include the Ministry of Social Affairs. This
ministry, like many others in Fifth Republic France, was in a continual state of flux. Beyond the
fundamental contradiction in making permanent a solution that was intended to facilitate the harki
population’s integration in part because it was temporary, this plan did not accordingly modify the
temporary elements of the Ministry of Repatriates’ forest hamlet policy: encadrement, isolation, and
living and working conditions.
SFIM Director Pérony had written in March 1963: “Il est indispensable d’encadrer les harkis
reclassés pendant la période d’adapation.” 151 Frey viewed encadrement differently—as a means of
discipline and control, not as a transitional arrangement. As he expressed in his letters to Pisani, the
harki population would continue to require encadrement in the permanent iterations of the forest
hamlets owing to their “oisiveté et… instabilité naturelles.”152 Choosing to maintain this structure in
its original form ran counter to two of the three initial goals for having on-site encadrement agents. First,
the Ministry of Repatriates had tasked the camp directors with protecting the harkis from potential
violence by FLN agents. Reports in 1962 and the first half of 1963 revealed a handful of visits from
CAC, 19910097/40 (dossier: Harkis Correspondance – 1965), Minister of Agriculture Pisani (signed by Cabinet
Director Jean Vaudeville) to Minister of Interior Frey, “Objet: Reclassement de rapatriés musulmans sur des chantiers
forestiers ouverts par le Ministère de l’Agriculture,” 1-3, Nº 7423 EF/D4, Dec. 2, 1965. A summary of the Oct. 21
meeting can be found in: CAC 19910097/40 (dossier: Harkis Correspondance – 1965), “Réunion du 21 octobre 1965.”
150 CAC, 19910097/41 (dossier: Harkis – Correspondance 1966), SFIM, “Procès-verbal de la réunion du 13 janvier 1966
présidée par M. le Préfet Aubert, Directeur du Cabinet de M. le Ministre de l’Intérieur.”
151 CAC, 19920149/3/1, Yves Pérony to the Director of General Administration and Budget (Ministry of Finances),
“Objet: Hameaux de forestage. Création de 50 chantiers supplémentaires en 1963,” 3, Nº 667 SFIM/MR, Mar. 18, 1963.
152 CAC, 19910097/40 (dossier: Harkis Correspondance – 1965), Minister of Interior Frey (signed by Cabinet Director
Jacques Aubert) to Minister of Agriculture Pisani, “Objet: Reclassement de rapatriés musulmans sur des chantiers
forestiers,” 1-4, Nº 1508 CAB/IX, July 30, 1965.
149
280
Algerians to the forest hamlets. But in 1966, with the FLN now dissolved, potential violence had long
ceased to be an issue. Second, these agents were put in place to educate the harkis so as to ease their
initial transition to metropolitan French ways of life. This included acting as liaisons with local and
departmental government officials for administrative matters with the goal of the “helping,”
“teaching,” “accompanying,” and “verifying”—not “doing.” With the majority of the harkis having
arrived in France in 1962 and 1963, continuing to encadrer the harkis in the same fashion called into
question the length of this transitional period and set up a situation of dependency, precisely what
Pérony warned of in the epigraph to this chapter: “Il faut en effet éviter que ces groupes ethniques ne
se trouvent encore plus isolés en dépendant uniquement de structure trop spécialisée.”153 Often, the
educational function was obscured because the forest hamlets’ permanence enabled the agents to be a
crutch for the harki population, thereby lessening the former’s incentive to teach and the latter’s
necessity to learn.
The encadrement agents’ final role in Missoffe’s forest hamlet policy was to discipline the harkis
to assure the smooth functioning of these spaces. As Frey had desperately desired, his ministry would
no longer be responsible for the encadrement of the forest hamlets; instead, this task would be shifted
to a ministry whose responsibilities already included the oversight of immigrant populations. A
representative of the Ministry of Social Affairs (the newly-created successor to the Ministry of Public
Health and Population) confirmed at the January 1966 meeting that his ministry would provide
encadrement of the forest hamlets and, to this effect, had already put in a request to Minister of Finance
Michel Debré to allocate funds.154 This transfer from the Ministry of Interior to the Ministry of Social
ADV, 746 W 62, Yves Pérony to Departmental Prefects, 1.
On January 8, 1966, Prime Minister Pompidou reorganized his cabinet. While Frey remained at the helm of the
Ministry of Interior, the personnel changes that affected the harki population included Michel Debré, who had been
Prime Minister at the end of the Algerian War, as the new Minister of the Economy and Finance; Edgar Faure as the
new Minister of Agriculture (Pisani became the Minister of Facilities); and Jean-Marcel Jeanneney, the French
Ambassador in Algeria from July 1962 to January 1963, as the Minister of Social Affairs. Two of these new ministers had
played a role in decisions concerning the repatriation of harkis in 1962.
153
154
281
Affairs would officially take place on January 1, 1967.155 At the meeting, the Social Affairs official put
discipline at the center of the encadrement agents’ role by describing their function as “encaisser[er] les
loyers qui seraient demandés aux harkis et [être] chargés à la fois de la discipline et de la bonne tenue
dans ces locaux.” 156 He proposed assimilating the existing agents into its existing corps of
administrators responsible for overseeing single male immigrant workers in SONACOTRA hostels.157
According to this logic, the harkis in the forest hamlets would be treated like immigrants,
permanently. The Ministry of Social Affairs’ Department of Social Programs for Migrants
subsequently disseminated, in March 1969, a document titled “Règlement intérieur concernant
l’hébergement dans les hameaux forestiers,” which is reproduced in Appendix F. The language and
information in this text resembled the instructions for SONACOTRA foyers and HLM populated by
immigrants. It enumerated a list of conditions that would warrant the harkis’ expulsion from the
forest hamlet, including subjective measures such as “atteinte aux mœurs” and “incorrections ou
violences envers le personnel d’encadrement, envers d’autres hébergés ou des personnes du
voisinage.” While violence was measurable, determining what was impolite behavior and attacks on
accepted standards of behavior concentrated much power into the hands of the 1 or 2 agents
responsible for the encadrement of as many as 250 individuals and limited harki citizens’ freedom of
expression. This document included an added measure of control over the harki population, not
present for other immigrants in public housing. Only family heads with repatriate status could reside
in the forest hamlet; their children over the age of eighteen years old or those who were married were
no longer permitted to live on the grounds.158
AN, 5 AG 1/22, “Note concernant le service…,” May 6, 1966.
CAC, 19910097/41, SFIM, “Procès-verbal de la réunion du 13 janvier 1966…,” 2.
157 Ibid. For a discussion of SONACOTRA administrators, see Choukri Hmed, “‘Tenir ses hommes’: La gestion des
étrangers ‘isolés’ dans les foyers Sonacotra après la guerre d’Algérie” Politix 19, no. 76 (2006): 11-30.
158 CAC, 19910097/41 (dossier: Harkis – Correspondance 1969), Department of Social Programs for Migrants (Ministry
of Social Affairs, “Règlement intérieur concernant l’hébergement dans les hameaux forestiers,” Mar. 7, 1969.
155
156
282
Concerning the isolation element of the government’s original vision for the forest hamlets,
government officials at the 1966 meeting primarily based their decisions about which ones to keep
open on fire prevention. They did not factor into this determination the forest hamlets’ isolation
from their adjoining towns. The final government plan had three elements. First, twenty-five forest
hamlets, nine of which were double forest hamlets, were attributed a permanent status.159 Some of
these permanent forest hamlets included Bormes (Var), Collobrières (Var), and Saint-Pons (Hérault),
situated respectively eleven, fourteen, and ten kilometers from the nearest towns. Saint-Pons was one
of the six forest hamlets located outside Provence that Ministry of Agriculture officials decided
during the January 1966 meeting to preserve as a special exception to the general rule, owing to the
importance of the harkis’ work in the region.160 Isolation would become the greatest detriment to the
harki population’s integration. A 1974 report on the harkis by the Department of Social Programs for
Migrants concluded, “[L]a promotion sociale des familles est, en général moins avancés chez celles
vivant en hameaux forestiers que chez celles des villes. Cela provient d’une ‘ségrégation de fait’ dûe à
l’éloignement entre hameaux forestiers et localités de rattachement.”161
The second element of the government’s plan put forward at the January 1966 meeting was
to use the Ministry of Agriculture funds intended to finance renovating existing houses instead to
open three additional forest hamlets in Mediterranean regions particularly threatened by fire. The
procedure for selecting sites would once again require the forest hamlet inspectors to visit potential
sites and receive the accord of local officials.162 The instructions that the Bouches-du-Rhône prefect
presented the departmental director of Agriculture with in December 1967 revealed that the only
The following chart enumerates which forest hamlets would remain open: CAC 19910097/41 (dossier: Harkis
Correspondance – 1966), “Tableaux des hameaux de forestage maintenus à titre permanent.”
160 The other five forest hamlets outside of the Provence region that would be permanently kept open were AvesnesTrucas (Hérault), Lodève (Hérault), Pujol-de-Bosc (Aude), Rivesaltes (Pyrénées-Orientales), and St-Martin des Puits
(Aude). In exchange, the four forest hamlets in the Vaucluse, Jouques (Bouches-du-Rhône), and Rians (Var) would be
closed. CAC, 19910097/41, SFIM, “Procès-verbal de la réunion du 13 janvier 1966…,” 3.
161 CAC, 19910097/42 (dossier: Harkis – Correspondance 1974), Department of Social Programs for Migrants (Ministry
of Social Affairs), “Hameaux Forestiers: Situation au 1er janvier 1974,” 4, Jan. 16, 1974.
162 CAC, 19910097/41, SFIM, “Procès-verbal de la réunion du 13 janvier 1966…,” 2.
159
283
criteria given for choosing a site were its location in or near the zone designated in a July 1966 law as
“périmètres de protection et de reconstitution” and building the forest hamlet in close proximity to
the worksite. The memo did not mention integration or ensuring the forest hamlets were near a
neighboring town to facilitate the harkis’ interactions with their surrounding community.163 In the end,
despite difficulties faced because municipalities were unwilling to provide land, five additional forest
hamlets opened between 1967 and 1973, two of which (Die in the Drôme department and
Collobrières) reopened forest hamlets that had closed on different sites.164
The plan’s final component was to close twenty-eight forest hamlets by the end of 1971, at a
rhythm of four to six per year. Harkis who were unable to find housing or jobs by the time a forest
hamlet shut would be transferred to the permanent ones.165 This situation led one self-described
“Française tout ordinaire” living near the Anglès forest hamlet whose inhabitants were slated to be
moved to a newly-opened forest hamlet in Corsica to liken the transfer of older harkis who could
find other jobs to a deportation. In a 1971 letter to the Secretary of State for Health and Social
Security, this resident of the Tarn department defended these men who were protesting their transfer:
“Ils sont des hommes et ne veulent pas être considérés comme des déportés…”166 One of the forest
hamlet inspectors for the social monitors explained the flaw in this policy as losing sight of the harkis’
ADBR, 125 W 173, Jean La Porte (Bouches-du-Rhône Prefect) “Objet: Regroupement des familles de Harkis dans la
Région Provence, côte d’Azur – Corse,” 1, Dec. 4, 1967. The law in question was “Loi nº 66-505 relative aux mesueres
de protection et de constitution à prendre dans les massifs forestiers particulièrement exposés aux incendies dans la
région méditerranéenne du 12 juillet 1966.”
164 The three other new forest hamlets included: La Ciotat (Bouches-du-Rhône, a double forest hamlet), Zonza (Corse, a
double forest hamlet), and La Grand’Combe (Gard). The Die forest hamlet replaced the Beaurières forest hamlet, which
had closed in 1964 and the Collobrières forest hamlet moved from “Le Capelude” to “La Capelle,” which is closer to
town. On the reticence of municipalities in the Bouches-du-Rhône to allow forest hamlets built on their land see: ADBR,
125 W 173, A. Astier (Departmental Director of Agriculture), “Objet: Réunion du Conseil d’Administration du
21.12.1970, Implantation d’un chantier de forestage dans la commune de Gardanne ‘Ferme Martin,’” Dec. 17, 1970.
165 CAC, 19910097/41, SFIM, “Procès-verbal de la réunion du 13 janvier 1966…,” 3 and CAC, 19910097/41 (dossier:
Harkis – Correspondance 1966), Ministers of Interior and Agriculture to the Minister of Economy and Finance, “Objet:
Reclassement des anciens harkis employés sur les chantiers forestiers ouverts à cet effet par le Ministère de l’Agriculture,”
2, Mar. 18, 1966. The following chart specifies the dates when each forest hamlets was slated to close: CAC
19910097/41 (dossier: Harkis Correspondance – 1966), “Plan de résorption des hameaux de forestage harkis établi au
cours de la réunion du 13.1.1966.”
166 CAC, 19910097/42 (dossier: Harkis Correspondance – 1971), Jeanne Cauquil to Secretary of State for Health and
Social Security Dienesch, 1, June 7, 1971.
163
284
humanity, much like Minister of Interior Michel Poniatowski signaled in the epigraph to this chapter.
The inspector wrote: “Les règles de la technocratie veulent que les hommes et les situations soient
interchangeables à volonté, et toute dérogation est la preuve de la mauvaise volonté évidente des
intéressés.”167 The closures would nonetheless proceed more slowly than anticipated. By the end of
1971 forty-two forest hamlets existed instead of the twenty-seven anticipated; for example, the Apt
forest hamlet in the Vaucluse, which was supposed to close in 1968 remained open until 1979.
According to a 1985 report published by a division of the new Secretary of State for Repatriates, this
delay was due in large part to difficulties the harkis encountered finding jobs and housing, particularly
as the population aged and their families grew larger.168 While serving as an impediment to the harki
population’s integration from the start, isolation, when it was temporary would necessarily inhibit its
integration only for a short while. Under the plan developed in 1966, the harki population’s isolation
would be permanent.
Finally, despite the modifications to the forest hamlet policy explained above, the
government’s plan in 1966 did not consider harki citizens’ quality of life. When some of the forest
hamlets became a permanent form of reclassement, others were slated to close as much as five years
later, and new ones were to be constructed, the government did not simultaneously opt to improve
the harki population’s housing and working conditions. When government officials initially set up
the forest hamlets as a temporary solution in emergency conditions, they ordered the construction of
small lightweight prefabricated houses so as to open them as quickly as possible. The choice at the
January 1966 meeting to use construction funds transferred from the Ministry of Agriculture budget
to open new forest hamlets instead of enlarging and making improvements on existing houses was a
crucial one demonstrating a disregard for the harkis’ humanity. The same 1974 report cited above
Heinis, “L’insertion des Français-Musulmans,” 101.
ADBR, 1451 W 115, Office National à l’Action Sociale Educative et Culturelle pour les Rapatriés, “Français
Musulmans Rapatriés Information. 1962-1981: La Politique d’Acceuil,” 3, Apr. 18, 1985.
167
168
285
revealed that due to a “très forte natalité” the average size of a harki family living in each two- or
three-room forest hamlet dwelling, which measured approximately forty square meters, was nearly
eight people.169 Moreover, the forest hamlets that opened after the meeting would follow the original
prescriptions for construction. The summary of an August 1970 meeting in Paris between
government officials and a delegation of twelve harkis exposed: “[L]e Génie Rural venait d’édifier le
hameau de Collobrières (Var) selon les normes anciennes (préfabriqué, logements trop petits). Il en
était de même à Zonza (Corse) où un hameau était en cours de construction.”170 Indeed, the Zonza
forest hamlet was the subject of much documentation concerning difficult living conditions,
including its segregated colonial-like conception, as one ONF official noted: “Il comporte 54
logements (50 pour les familles des travailleurs – 1 pour le Chef de hameau et 3 pour l’encadrement
forestier). Il est conçu sur le type casbah, ce qui et probablement une erreur pour des familles que
l’on souhaite intégrer dans un mode de vie correspondant à celui des travailleurs français.”171 The
government did not always treat accounts of poor living conditions in the forest hamlets as urgent
concerns. For example, during a March 1972 interministerial meeting concerning the forest hamlets,
the subject of the Bormes forest hamlet in the Var department arose, as both the mayor of the town
and a representative of the Ministry of Social Affairs submitted reports detailing its insalubrity and
hygiene problems. The meeting’s participants concluded that given no viable solution had yet
surfaced, the director general of the Forestry Commission would soon examine the site for himself,
CAC, 19910097/42, Department of Social Programs for Migrants, “Hameaux forestiers…,” 4, Jan. 16, 1974.
CAC, 19910097/41 (dossier: Harkis – Correspondance 1970), Service d’Assistance Technique, “Compte rendu de la
réception d’une délégation d’anciens supplétifs de l’Armée Française par Monsieur le Ministre de l’Intérieur le jeudi 13
août 1970, à 12 heures,” 2.
171 CAC, 19910097/42 (dossier: Harkis – Correspondance 1972), J. Ducasse (Forestry Expert, ONF), “Objet: Problèmes
soulevés par l’installation d’un groupe de ex-harkis à Zonza (Corse),” 2, Ig. nº 33/1972, July 17, 1972. The Ministry of
Agriculture archives contain three entire cartons of documents relating to the Zonza forest hamlet: CAC, 19910097/4446.
169
170
286
but this was not a rush: “Au demeurant, cette affaire ne présente aucun caractère d’urgence et une
solution ne saurait être envisagée avant 1973.”172
Besides neglecting the harki population’s difficult living situation, the January 1966 plan did
nothing to address the men’s poor working conditions. Like harki soldiers during the Algerian War
for Independence, the harkis who worked for the ONF were day workers. This status meant that
their salaries were small and they were rarely paid if they did not work. This included during
inclement weather (which was out of their control), until a 1969 convention mandated that the
harkis be paid for a half-day of work, and if they had on the job injuries, which were frequent
occurrences given the nature of their work.173 Most harkis on the Fuveau forest hamlet, for example,
earned twenty francs per day from 1964 to 1969 (the equivalent of twenty-four to twenty-seven
euros in 2011 when adjusted for inflation) and worked between twenty-three and twenty-seven days
per month in 1967. Beginning in 1969, the Fuveau harkis were paid by the hour; in 1971, most
earned four francs per hour (the equivalent of four euros in 2011 when adjusted for inflation) and
worked between 175 and 200 hours per month. 174 However, the harkis’ pay situation and the
number of days they worked varied from worksite to worksite. Not until an interministerial order on
December 8, 1975 did the government mandate uniform working conditions for all the forestry
worksites.175 Even once the harkis were salaried they still faced challenges regarding their pay owing
to the special status the government created for them. For example, when the minimum wage
(SMIC) changed in 1976, the harkis did not benefit from the increase immediately because the
Minister of Agriculture needed to approve any salary modifications, yet did not do so for almost one
CAC, 19910097/42 (dossier: Harkis – Correspondance 1972), Ministry of Agriculture, “Compte-rendu de la réunion
interministérielle du Mercredi 8 Mars 1972 à 10 heures,” 3, Mar. 9, 1972.
173 AONFBR, Ministry of Agriculture Department of Forestry, “Convention générale relative à l’emploi d’anciens harkis
par l’Office National des Forêts 1969,” 5. Before this time, some forestry commissioners opted to pay the harkis when
there was inclement weather, but this policy was neither uniform nor consistent.
174
This information comes from the following cartons at the AONFBR: Harkis Salaires 1964-65, Harkis Salaires 1966,
FSIRAN Salaires 1967, Harkis Salaires 1968, Harkis Salaires 1969-1970, and Harkis Salaires 1971-1972.
175 AONFBR, “L’Evolution de l’organisation et des missions des chantiers d’anciens harkis,” 3.
172
287
year. 176 While some families wanted to remain in the forest hamlets, the work situation the
government set up for them—low salaries and difficulty getting days off work to search for new
jobs and housing—added another obstacle to moving their families out of the forest hamlets.
Conclusion: Failed Integration Policy for the Harki “Repatriates”
A May 1976 report offered the following assessment about the three remaining forest hamlets
in the Bouches-du-Rhône department:
[L]es problèmes proviennent surtout de l’isolement et des mauvaises conditions de
logement. Tous ces hameaux se trouvent à plusieurs kilomètres des villes les plus proches. Il
[a] fallu créer des écoles spéciales pour les enfants. En particulier des maternelles, CM 1 et
CM 2, les villages n’ayant pas la possibilité d’absorber tous les enfants.
Les femmes vivent entre elles comme dans leurs douars d’origine et n’ont pas ou peu de
contacts avec les ménagères européennes.
Quant aux hommes, presque tous d’origine agricole ou forestière, ils n’ont aucune
spécialité permettant leur reclassement dans une profession valable.177
Nearly fourteen years after the first forest hamlets opened with the goal of serving as a temporary
step toward the harki population’s integration into French society, 28 forest hamlets housed 534
families with a total of 4,255 people.178 These 534 families represented 2.67 percent of the 20,000
harki families repatriated to France during the 1960s. As the report above contends, harki men
worked in isolated conditions among only other harkis and their wives had little contact with
“Europeans” outside of the forest hamlets, thereby signaling that the social counselors and local
officials had failed in their roles to integrate the adult residents into their surrounding communities.
Integration was also problematic for their children. Contrary to the Ministry of Repatriates’ inclusion
of the tenet that harki children must be socialized into the French educational system into its
integration policy, some of children attended special schools because those in the nearby small
CAC, 19910097/48, Minister of Agriculture, “Gestion des anciens harkis par l’Office National des Forêts (O.N.F.),”
16-17, Jan. 24, 1977.
177 ADBR, 135 W 51, Louis Sessa (B.I.A.C. Director) and Didier Cultiaux, “Rapport sur la Situation des FrançaisMusulmans, installés dans le department des BOUCHES DU RHONE,” 3, May 18, 1976.
178 ACNMF, 16, “Hameaux Forestiers au 1er juillet 1976.”
176
288
villages could not accommodate the growing forest hamlet populations. When French government
officials decided to radically alter the forest hamlet experiment by making these spaces permanent in
1966, they did not make concomitant changes to the other elements of the forest hamlet policy to
align with the Ministry of Repatriates’ initial discourse of integrating the harkis into their communities.
From the beginning, the government’s forest hamlet policy was flawed, given the inherent
contradiction between integration, on one hand, and isolation and encadrement, on the other, and the
extreme reliance on local officials to facilitate the harkis’ integration. However, by rendering
permanent the forest hamlets and maintaining their temporary conditions while continuing to use
them as the principal type of reclassement, the government made harki citizens’ integration into their
surrounding communities virtually unobtainable.
The manner in which the national government created policies for and administered the
forest hamlets further decreased the possibility for these spaces to succeed. In April 1974 Anne
Heinis, a social welfare inspector for the SFIM, delivered a report in which she proposed modifying
the existing structures that regulated the forest hamlets and the two cités d’accueil. Going beyond the
evident problems posed by these harki populations’ isolation and dependence on the social
counselors, she enumerated specifically how the organization of the housing system contributed to
these harki citizens’ failure to integrate.179 First, despite Frey’s call for streamlining the management
of the forest hamlets in 1966—in part so it would not involve his ministry—these responsibilities
were subsequently spread among several different ministries: Agriculture, Health, Interior, Labor, and
Social Affairs in their various iterations.180 Heinis argued that for the “insertion”—a term that was
commonly used in place of “intégration” at the time—of the harki community to take place, all its
ACNMF, 3/7, Anne Heinis (Regional Inspector for French Muslims), “Note à l’attention de Monsieur le Président
Parodi,” 1, Apr. 12, 1974.
180 All these ministries were in a continuous state of flux at the end of the 1960s in the wake of the young Fifth
Republic’s crises, which led its founder President de Gaulle to precipitously step down in April 1969, one year after the
events of May 1968.
179
289
social welfare problems must be addressed by a competent person in one of the ministries.181 Second,
Heinis maintained that the organization of the SFIM failed to take into account the “very variable”
nature of the local forest hamlet situations when addressing their problems. The government’s setup
of these spaces, whose integration depended on the one or two state agents living with the harki
citizens and the support of local officials, assured variable results and unique problems. Finally, the
existing structure lacked effective means for communication of information between the social
counselors in the individual forest hamlets and the national government. This situation prevented the
social counselors from receiving access to all the information they needed to pass onto the harkis and
the national government officials from knowing what the forest hamlets were truly like. Heinis
suggested that the representative of the national government mentioned above make periodic visits to
the forest hamlets to be able to address whatever problems arise with a conceptualization of the
actual places in mind.182
Putting aside the policy itself briefly, one must question whether the secretary and ministers
of Repatriates factored integrating the harki citizens residing in the forest hamlets into the economic
and social structures of the French nation—the mission given to them by the December 1961 law for
all repatriates—into their decisions about these spaces. Robert Boulin was named Secretary of State
for Repatriates in May 1961 and launched the first phase of forest hamlets, envisioning these spaces
as isolated venues with encadrement structures resembling those used in colonial Algeria, at the same
time as he created and oversaw wide-reaching policies to integrate the French repatriates into society.
Alain Peyrefitte spent two and a half months as Minister of Repatriates from September to
November 1962, during which time it appears that he did not author any documents concerning the
ACNMF, 3/7, Anne Heinis, “Note à l’attention…,” 1, Apr. 12, 1974. Abdelmalek Sayad argues that the various terms
that government officials have used over time and in different contexts—adaption, assimilation, adaptation, insertion, and
intégration—all essentially describe the same social reality and the same sociological process. Sayad, “Qu’est-ce que
l’intégration?” Hommes & Migrations, no. 1182 (Dec. 1984): 9. While intégration focuses on an individual, insertion focuses
on a community. On the usage of assimilation, insertion, and intégration by government officials and the variations in the
terms’ definitions, see Weil, La France et ses étrangers, 368-70.
182 ACNMF, 3/7, Anne Heinis, “Note à l’attention…,” 1.
181
290
harkis in the forest hamlets. François Missoffe was the final Minister of Repatriates, who was in
office when over fifty of the forest hamlets opened and initiated discussions during the waning
months of his ministry about making permanent what he had referred to, just months prior, as “the
forest hamlet experiment.” Missoffe’s early assessments of the forest hamlets’ success in and
possibility for integrating the harki population into their communities were inaccurate and
contradicted the reports of local agents who had made site visits. Yet, Missoffe opted to push for a
policy of expanding the forest hamlets and making them permanent as a way to fulfill his ministry’s
goal. This was in spite of the population of nearly 10,000 in the forest hamlet at the time when he
stepped down—representing approximately 10 percent of the harki population repatriated to
France—not being integrated into France’s social and economic structures and the setup of the forest
hamlets impeding any possibility for future integration. The reclassement of repatriates from Algeria
was indeed an apartheid system.
As time went on, government officials increasingly regarded the harki population in the forest
hamlets like they did other Algerian immigrants with social welfare problems—and not like
repatriates or citizens. Given the flaws in the forest hamlet policy analyzed above and numerous
negative assessments made officials who visited the sites in the 1970s, it is evident the government
did not attempt to complete its mission to integrate these repatriates into France’s social and
economic structures. In fact, until 1973, the government still was opening new forest hamlets, with
over a decade of failed integration results. On July 15, 1970, the government passed a law authorizing
indemnity payments to French citizens who had lost property and material goods in its former
overseas possessions.183 As will be discussed in the next chapter, this legislation was rarely available to
the harkis given the documentation required. When the focus of the government’s integration
JORF, “Loi nº 70-632 du 15 juillet 1970…” The August 5, 1970 decree specified that the scope of the law applied to
property and material goods in Algeria. Subsequent decrees extended the law to repatriates from Morocco and Tunisia
(April 21, 1971), Indochina (January 29, 1973), and Guinea (March 13, 1975). Scioldo-Zürcher, Devenir métropolitain, 343.
183
291
policies for French repatriates shifted from jobs and housing to compensation—a crucial turning
point—the government had not yet enacted a viable housing integration policy for the harki
population residing in the forest hamlets—or elsewhere. 184 Compensation for the harkis’ lost
possessions and land was still decades away.
Once again for the harki population with the forest hamlets—as with the government’s
nationality, repatriation, and camp policies—decisions made in the months that followed the Algerian
War for Independence marginalized these former soldiers and their families from French society and
continued to influence them for decades. In 1962, the Ministry of Repatriates set up a different
service for the harkis’ reclassement, the SFIM, which soon privileged the forest hamlets as its preferred
location for rehousing the harki population. For as long as each forest hamlet remained open, it
resembled a microcosm of colonial Algeria. Each of the forest hamlets had similar physical structures
and governance and offers a consistent story of failed integration for its citizens. While the
government’s forest hamlet policy underwent several significant modifications from 1962 to 1966,
after this time, no national policies concerning the forest hamlets would be enacted until a 1975
decision to eradicate the twenty-nine remaining forest hamlets, which will be analyzed in the next
chapter. The stagnations in government policies were equally significant for the harkis’ failed
integration as were the policy shifts.
With an understanding of the creation of and modifications to the government’s forest
hamlet policy, the next chapter analyzes the setup and functioning of the Fuveau forest hamlet,
which was opened on April 15, 1964 as part of the Ministry of Repatriates’ “second phase” and still
houses thirty-two harki families today. This case study, which is placed within the departmental and
national contexts, will give texture to the specific shortcomings of the government’s forest hamlets
policy that led to a failure to integrate the harki families into their community.
184
Scioldo-Zürcher, Devenir métropolitain, 331.
292
CHAPTER 5
Immigrants? Isolation in Rural Southern France
“Enfin, quitter la tente, avoir un toit, un travail,
un motif de vivre!”
“Ici, c’est notre pays!”
“Que désirer encore? Que les voisins nous
considèrent comme des amis. Nous ferons tout pour
que ce bon voisinage soit réel.”1
-Testimonies of harkis arriving at the Fuveau
forest hamlet on April 15, 1964
Monsieur le Préfet,
Si on peut trouver de l’argent pour “intégrer”
d’inintégrables immigrés, il est temps d’en trouver
pour aider enfin les Français Harkis.
Pour cela, il faut démanteler le camp de
harkis de FUVEAU.2
-Petition addressed to the Bouches-du-Rhône
Prefect by a resident of Fuveau affiliated with the
National Front party in July 1993
In 1964 the 2,600-person village of Fuveau, located 35 kilometers northeast of Marseille in
the Bouches-du-Rhône department, absorbed a new migrant population.3 On April 15 two hundred
harki population members arrived by convoy from the Rivesaltes camp to a forest hamlet atop
brown coalfields, known as the Fuveau basin. Their arrival to this eleventh-century Provencal village
signified the appearance of a type of immigrant unique to its modern history. Foremost, the harkis
were not labor migrants, like those driven by the job opportunities in Fuveau’s coalmines during the
nineteenth and first half of the twentieth centuries. The foreign coalminers most often were single,
Jacques Vautier, “50 harkis et leurs familles ont trouvé logis, emplois et école dans la région de Fuveau,” Le Mériodional,
Apr. 16, 1964.
2 Archives personnelles de Marie-Josephe Haupt, Fuveau town councilwoman from 1989 to 1995 (hereafter APMH),
Petition from Loïc de Marion (National Front) to the Bouches-du-Rhône Prefect, July 1993. On the original document
the second sentence is in boldface and larger type size than the first sentence.
3 The figure of 2,600 inhabitants is estimated using census data from 1962 that lists Fuveau’s population at 2,528 people.
“Fuveau – Notice communale,” Des villages de Cassini aux communes d’aujourd’hui,
<http://cassini.ehess.fr/cassini/fr/html/fiche.php?select_resultat=14947#>.
1
293
whereas all but a handful of the harkis who resided in Fuveau were accompanied by their families.
When these former soldiers and their families moved into their new homes, many were already
French citizens, having benefitted from the expedited process explained in chapter two. This status
was unlike that of Fuveau’s foreign coalminers; many were transient workers, sensitive to the health
of the mining industry and national economy, and did not seek French nationality.4 For those who
did, their residency preceded their becoming French citizens (sometimes by as much as ten years), as
French nationality legislation since 1889 has been based the principle of socializing immigrants into
society before they can become French citizens.5 With the exception of a small number of Algerians
who arrived after the Second World War, Fuveau’s earlier migrants hailed from European countries:
Italy, Spain, Poland, Czechoslovakia, and Yugoslavia. While these foreigners tended to live among
themselves, often in abandoned houses in the neighborhood behind the town church (see figure 28),
and made little contact with their neighbors, they nevertheless remained in close proximity to native
Fuvelians at home and in the coalmines.6 State agents did not supervise the foreign coalminers, who
sought out social services themselves. In contrast, the harkis worked only with other harkis in the
Provencal forest. Exclusively harki families, the camp director, and the social counselor inhabited
the forest hamlet, situated three kilometers southwest of the town center. The harki population in
Fuveau necessitated a much greater intervention than local officials were used to with previous
migrants.
Michel Colon, Fuveau autrefois: Le temps de mineurs paysans, 1809-1962 (Marseille: Presses de l’Imprimerie Caractère, 2001),
53 and 138.
5 The three principal laws were passed in 1889, 1927, and 1945; the latter is known as the Nationality Code and is still in
effect today. The laws required, respectively, ten, three, and five years of residence in France before granting nationality
to a foreigner. On this legislation, see Weil, Qu’est-ce qu’un Français?, 60-61, 77-78, and 150-53.
6 Colon, Fuveau autrefois, 53 and 148.
4
294
Figure 28: The town of Fuveau with the spire of the Saint-Michel church in the foreground and the Sainte-Victoire
mountain in the background. Source: Photograph by the author, November 2007.
The Fuveau forest hamlet adhered to the prescriptions laid out by the Ministry of Repatriates
in 1962 and 1963. The site was situated in an isolated forest clearing on the outskirts of a small town.
As a double forest hamlet—home to workers on two adjacent forestry worksites—it consisted of
lightweight prefabricated houses for the families of fifty-two harki men. Their children, after an
initial period of attending school on the camp grounds, were most often educated in local schools. A
regional inspector, camp director, and social counselor—recruited among agents who had lived in
Algeria—assured the harki citizens’ encadrement. Government memos instructed them to work with
local officials to help the population integrate into the Fuvelian community, teach the harkis and
their family members about metropolitan ways of life, and ensure that they did not disturb their
neighbors. In short, the setup of the Fuveau forest hamlet isolated its residents from their neighbors,
sheltered them in temporary living conditions, provided an encadrement structure, and required the
reliance on local officials to help integrate the harki population. However, as attests the document
295
cited in the second epigraph—written nearly thirty years after the first harki families arrived—the
setup and functioning of the Fuveau camp, still located on the same isolated plot today, had failed to
integrate them into the community. This situation was in spite of a series of national government
policy changes in 1975 to address the “harki problem,” including the decision to eliminate all forest
hamlets by the end of 1979.
The three testimonies of the harkis migrating to the Fuveau forest hamlet from the
Rivesaltes camp in 1964, collected by an Arabic-speaking journalist for the Mériodional regional
newspaper, bring out salient points to consider while analyzing the Fuvelian harki population’s
isolation. First, while the first harkis who arrived in Fuveau had left the Rivesaltes camp, they had
not left behind camp life. In many ways, the Fuveau forest hamlet was a microcosm of this camp,
retaining the key features of “internment camps” (whether for refugees or prisoners), as laid out by
Denis Peschanski. The Fuveau forest hamlet was composed of a “société hétéroclite.” It, like the
seventy-one other forest hamlets across France, was set up in emergency conditions as a temporary
solution. The buildings were poorly- and hurriedly-constructed. Its occupants were under military
surveillance, at least initially, and a system of encadrement that marginalized them from their
surrounding community. While being located in the municipality of Fuveau and having positive
economic benefits for commerce, local officials viewed the camp and its residents as incidental to
municipal affairs. Finally, forest hamlet residents attempted in later years to subvert the authority of
the camp.7 Nevertheless, while consistently exhibiting these elements of internment camps, the
Fuveau forest hamlet challenges the traditional conception of a refugee camp because of how long
the state opted to keep it open in these temporary conditions. Individuals made a choice about how
long to stay, though this decision was mitigated by factors such as language, large families, mental or
physical illness, a lack of professional skills, and money since the families paid no rent while residing
7
Peschanski, La France des camps, infra.
296
in the prefabricated buildings. This latter element, in part owing to the meager wages the harkis
earned and the fact that as unsalaried workers for the first decade they had limited social benefits, set
up a situation of state handouts that was difficult to leave. Five of the first fifty workers and their
families made the Fuveau forest hamlet their permanent home, with four still alive in August 2010
and one having died of natural causes in 1989.8
Second, the testimonies demonstrate that migrating to France placed the harkis and their
families into a liminal space between two nations and two cultures. Given the harki population’s
unique exile, “notre pays,” as one of the harkis wanted to claim in his testimony, was an unattainable
goal. As Abdelmalek Sayad argued, “…immigrer c’est immigrer avec son histoire (l’immigration
étant elle-même partie intégrante de cette histoire), avec ses traditions, ses manières de vivre, de
sentir, d’agir et de penser, avec sa langue, sa religion ainsi que toutes les autres structures sociales,
politiques, mentales de sa société, structures caractéristiques de la personne et solidairement de la
société…, bref sa culture.”9 The first three chapters of this dissertation explored why the harkis
decided to leave Algeria and the conditions upon their arrival to the transit and housing camps. As
their lives in Algeria became increasingly distant while they settled into Fuveau, we must continue to
recognize the harki population as both emigrants from Algeria and “immigrants” to France. More
specifically, any analysis of their situation in France—social, economic, demographic, cultural, and
political aspects—must take into account the lives in Algeria that FLN violence forced them to give
up. Interviews that I conducted in Fuveau and those published in a collection by harki daughter
Dalila Kerchouche demonstrate an ambiguous relationship with Algeria and with France. In regard
to the former, older harki population members’ nostalgia for the land and their family members was
often juxtaposed with memories stained with fear and death. With respect to the latter, they were
grateful for having been rescued from violence and perhaps death, yet irresolute and angry as to why
8
9
These men are: Aissa B., Lazreg C., Mohamed ben Ahmed M., Mohamed ben Milou M., and Abdelkader Z.
Sayad, La double absence, 18.
297
they were parked in camps.10 During an interview conducted with two harkis in 2007, both men
repeatedly asked, seemingly rhetorically, “What have we done wrong to France?”11
Finally, while one must consider harki citizens’ status as emigrants from Algeria and
immigrants to France, at the same time their Algerian culture did not lead officials to deem the
forest hamlet integration policy a failure. These officials’ analyses derived from the insular nature in
which the forest hamlets forced their residents to live. The harki population’s physical distance from
town and lack of opportunities to interact with their neighbors resulted in the formation of a
separate community that limited the possibility for the “bon voisinage,” to which one of the harkis
arriving in 1964 referred. Integration is not possible if groups do not interact; in this case, there is
social exclusion or marginalization.
Before moving on, an important distinction must be made concerning two different usages
of “integration,” which are sometimes errantly applied interchangeably and can render the word
ambiguous. Integration—a term that has evolved since its appearance in the context of populations
during the mid-twentieth century—is both a process that sociologists measure and a type of policy
that government officials define, develop, and ultimately evaluate. The sociological meaning today
connotes the difficult to quantify process of measuring the adoption by a minority group (often
immigrants) of socially normative behaviors, those of the dominant, and their political participation
in society.12 The political meaning—which this study uses—is a public course of action aimed at
fostering social cohesion among individuals within a society; through a set of policies, including
K. family, interview; Abdelkader B. and Abdelkader Z., interview with the author, Dec. 19, 2007, Fuveau; and B.
family, Interview with the author, June 14, 2008, Fuveau. Stéphan Gladieu and Dalila Kerchouche, Destins de Harkis: Aux
racines d’un exil (Paris: Autrement 2003), interviews with Mme Kerchaoui (100-01) and Mme Zendji-Choucha (105-06).
11 Abdelkader B. and Abdelkader Z., interview.
12 On sociologists’ varying definitions of integration and the evolution of this term, see Dominique Schnapper, Qu’est-ce
que l’intégration? (Paris: Gallimard, 2007), 11-25. Schnapper highlights that, like any process, the group’s integration in one
domain can happen more quickly than its integration in another domain and sometimes reversals occur (24). For an
earlier analysis of integration, see Sayad, “Qu’est-ce que l’intégration?,” 8-14.
10
298
attending republican schools, individuals will integrate one by one into French society.13 Today,
France’s Haut Conseil à l’Intégration maintains that an integration policy requires a “reciprocal
effort” from immigrants and native French citizens; the latter must be open to immigrants’ diversity,
which enriches society.14 However, the Ministry of Repatriates forest hamlet integration policy for
the harki population in 1962 aimed to divest its members of their diversity and have them adopt
French ways of life. Most importantly, it isolated them from their surrounding communities.
This chapter first analyzes the setup of the Fuveau forest hamlet by town and departmental
officials. It next traces the first decade of harki citizens’ residence in Fuveau, focusing particularly on
how government’s forest hamlet policy and its implementation by encadrement agent and municipal
officials isolated them from town residents. This analysis demonstrates that the Ministry of
Repatriates’ initial vision for the forest hamlets and the subsequent amendments to the forest hamlet
policy initiated by the Ministry of Interior in 1966 did not facilitate the Fuvelian harki citizens’
integration into their surrounding community. Instead, they have remained isolated from other
Fuvelians for nearly five decades.
Fuveau’s “Foreigners”: From Miners to Harkis
While the first exploration of Fuveau’s rich coalfields took place during the mid-seventeenth
century, not until an 1809 decree by Napoleon allowing for industrial exploitation of coal across
France did mining begin to take a preeminent place in the Fuvelian economy. 15 By the midnineteenth century miners displaced farmers as the primary profession in Fuveau, and by 1881 the
former outnumbered the latter by four to one.16 This shift, at the height of the Industrial Revolution,
Weil, La France et ses étrangers, 369.
“Mots de l’intégration,” Haut Conseil à l’Intégration, http://www.hci.gouv.fr/rubrique.php3?id_rubrique=19#I
(accessed October 7, 2011).
15 Colon, Fuveau autrefois, 7. This imperial decree was made official by the Apr. 21, 1810 “Loi concernant les Mines, les
Minières et les Carrières.” Reprinted in Colon, Fuveau autrefois, 39.
16 Ibid., 45-46.
13
14
299
was attributable to the need for coal to fuel the region’s expanding train system and the steam heat
used in the soap, tile, and brick factories in neighboring Marseille.17 As a result of the jobs the
mining industry created, a significant population of transient foreigners from neighboring
countries—particularly Italy—migrated to Fuveau between the mid-nineteenth and the midtwentieth century. For example, in 1931 foreigners from seven European countries accounted for 23
percent of Fuveau’s miners.18 Reflecting national trends after the Second World War, Fuveau’s
economy underwent structural changes, which influenced the use of its land and consequently the
composition of its population. In 1946 the French government nationalized many of its industries
and France’s new service sector spawned a wealth of jobs in public service. Miners still retained an
important presence in the local economy over the next decade during France’s reconstruction, as
mining and related professions (electricians, blacksmiths, and surveyors) employed 34 percent of
Fuvelian workers in 1954.19 However, after this time, given France’s decreasing reliance on coal for
energy production, no new positions were available in the mining industry. This signaled the end to
over a century of European laborers migrating to mine the Fuveau basin.
The last vestige of Fuveau’s mining industry was the land sitting atop its coalfields that had
fallen into disuse. The diversification of professions led the regional mining company, Houillères de
Provence, to create two industrial parks on its former fields to generate jobs for its employees in the
burgeoning service sector. The Rousset-Peynier park three kilometers east of Fuveau’s Saint-Michel
church (depicted in figure 28) was inaugurated in March 1961 while the Gardanne-Palun park eight
kilometers west of Fuveau’s center opened in August 1963.20 Despite these ventures, the Houillères
de Provence still possessed unused land in this town whose population numbered 2,528 in 1962.21
Ibid., 40.
Ibid., 137-38.
19 Ibid., 137.
20 Ibid., 66.
21 “Fuveau – Notice communale,” Des villages de Cassini aux communes d’aujourd’hui,
<http://cassini.ehess.fr/cassini/fr/html/fiche.php?select_resultat=14947#>.
17
18
300
The sharp decline in the mining industry in the early 1960s left expanses of terrain on the fringes of
Fuveau at the same moment when the Ministry of Repatriates was reaching out to mayors of rural
towns in the Provencal forest to provide land for constructing forest hamlets.
Figure 29: Satellite image of
Fuveau in 2011. The red
diamond marks Fuveau’s town
hall and the blue one the forest
hamlet where the harki
population resided. Source:
Google maps.
On March 26, 1963, the Secretary General of the Bouches-du-Rhône Prefecture held a
meeting to discuss the possibility of opening forestry worksites in the department. He tasked a fiveperson commission including local representatives of the Agriculture, Armies, and Repatriates
Ministries with visiting twelve potential sites, in hopes of finding six that would work. Under
pressure from the Minister of Repatriates, who intended to open the forest hamlets that summer,
these men began canvassing two days later. Their first stop was a five-acre parcel of land belonging
301
to the Houillères de Provence situated in Fuveau’s “quartier Madame Saint André.” To access the
plot, Fuveau’s Socialist Mayor Alexandre Philip, also the president of the union of Bouches-duRhône mayors, accompanied the men from the town hall (the red diamond in figure 29). They
traveled southwest for three kilometers until they reached a small forest clearing, the blue diamond
on figure 29, located on a steep ascent of the two-lane State Highway 96. As shown by the map in
figure 30, which dates from 1974 when Fuveau’s population living outside the forest hamlet was five
hundred more than that when the harkis arrived in 1964, this sylvan area was isolated from the
town’s other houses. After exploring the ground above the former coalmines, Mayor Philip gave the
men his verbal authorization to construct a forest hamlet.22 Demonstrating the project’s urgency,
government officials returned twenty times over the next three months to continue to survey the site
with public works officials and representatives of the Schroth building company.23 This company,
based in the Haut-Rhin department, was able to submit a bid on April 27, a process usually slowed
down by the acquisition of building permits.24 Keeping with the goal of the forest hamlets as a
temporary solution, SFIM Director Pérony had specified to the Secretary General of the Bouchesdu-Rhône Prefecture two days earlier that building permits were not necessary “en raison du
caractère provisoire des logements.”25
The other municipalities included: Peynier, Boulbon, Aurons, Rognes, Gèmenos, Coulin, La Roque d’Anthéron,
Roquefort-la-Bédoule, Carry-le-Rouet, Sausset-les-Pins, and Meyreuil. ADBR, 137 W 460, Prefect’s cabinet, “Note
concernant le reclassement des anciens supplétifs musulmans dans le département des Bouches-du-Rhône,” Apr. 4, 1963.
23 ADV, 746 W 68, D. Thomas (Chief Agricultural Engineer) to Mr. Nourrit, “Objet: Logement des harkis, Frais de
déplacement,” G.R. Nº 2809, June 25, 1963.
24 ADV, 746 W 68, L. Torrion (Director of the Agricultural Engineering Division, Marseille), “Ordre de service nº 1:
Logement de harkis. Chantier de Fuveau,” G.R. 3148, July 24, 1963.
25 ADV, 746 W 68, L. Torrion to Agricultural Engineer Collignon, “Objet: Relogement des harkis. Permis de construire,”
G.R. Nº 1913, Apr. 29, 1963.
22
302
Figure 30: Map of Fuveau in 1974 when its population, excluding its 230 harkis and their family members,
numbered approximately 3,100. This figure represents an increase of approximately 500 people over the town’s
population before the harkis arrived in 1964.26 The town is located at the top of the picture and the forest hamlet is in
the center surrounded by a circle. Source: Archives de la Société Anonyme d’HLM de Marseille (hereafter
ASANHLMM), 54/1, SANHLMM, “Commune de Fuveau (13), Reconstruction du village de harkis,” Oct. 16, 1974.
On May 6, 1963, immediately after the canvas was completed in the Bouches-du-Rhône and
the three other departments comprising the Provence region, Pérony led a meeting concerning the
“opération forêt provençal” at the Ministry of Repatriates. With the goal of opening the forest
hamlets before the end of the summer, a group of national and departmental officials approved the
construction of three forest hamlets in the Bouches-du-Rhône and sixteen in the rest of the region.
Fuveau’s population statistics for 1964 and 1974 were approximated using data from the EHESS Cassini project.
Fuveau had 2,528 residents in 1962, 3,028 residents in 1968, and 3,348 residents in 1975. “Fuveau – Notice communale,”
Des villages de Cassini aux communes d’aujourd’hui,
<http://cassini.ehess.fr/cassini/fr/html/fiche.php?select_resultat=14947#>. The number of harkis residing in the
town in 1974 is from table 6.
26
303
None would meet this goal. The first forest hamlet in the “opération forêt provençale” opened in
late September 1963, eight more opened before the end of the year, eight additional ones opened
from February to August 1964, one in 1965, and two (La Turbie in the Var and La Bastide in the
Vaucluse) never opened.27 Those in the Bouches-du-Rhône included a double forest hamlet in
Fuveau, a double forest hamlet in la Roque d’Anthéron bordering the Vaucluse department, and one
forestry worksite located twelve kilometers from the town of Jouques in the Durance Valley.28 In a
document addressed to the four companies that officials approved during the meeting to build the
forest hamlets, the Ministry of Repatriates stressed the necessity for them to move quickly by listing
the project’s goal as “[l]oger dans les plus brefs délais, un certain nombre de familles d’anciens
supplétifs musulmans.” 29 The government agents at the May 6 meeting had decided that all
construction on the sites—including installing sewage systems and electricity, widening access roads,
and erecting pre-fabricated buildings—must be completed within seventy-five days of breaking
ground.30 This goal subsequently posed a number of challenges, particularly given the isolated locales
chosen for the projects, and equally demonstrated that Ministry of Repatriates wanted hastilyconstructed, not well-constructed, homes for the harkis.
ADV, 746 W 63 (dossier: Harkis Var bâtiments 1963), L. Nourrit, “Note à M. Cullignon,” 3, May 13, 1963. This list
determined at the May 6 meeting did not include the following forest hamlets that would subsequently become a part of
the “Operation Provencal Forest”: Pertuis in the Vaucluse and St-Paul-en-Forêt in the Var. However, St Paul-en-Forêt
was listed as a “probable” site, along with five others that would never open. See Appendix D for the opening dates of
the forest hamlets.
28 Jouques was unlike any other forest hamlet in France. Here, the forestry workers and their families resided in the
“Logis d’Anne” housing complex, constructed by the SONACOTRA, alongside the families of eighty-five harkis
employed nearby in various unskilled laborer positions. This site had previously housed workers at the nearby EDF
hydraulic station. For a study of the “Logis d’Anne” in Jouques, see Marwan Abi Samra and François-Jérôme Finas,
“Regroupement et dispersion: Rélégation, réseaux et territoires des Français-Musulmans,” Study Commissioned by the
Caisse Nationale d’Allocations Familiales, 1985.
29 ADV, 746 W 63 (dossier: Schroth), Ministry of Repatriates, “Devis-programme relatif à la Construction et à la mise en
place de bâtiments préfabriqués pour anciens supplétifs musulmans,” 1. In the May 6 meeting, Pérony made a point of
emphasizing that, in contrast to policy for the first forest hamlets opened, the harkis should not participate in the
construction process since the Ministry of Repatriates was unable to provide appropriate encadrement for this purpose.
ADV, 746 W 63 (dossier: Harkis Var bâtiments 1963), L. Nourrit, “Note à M. Cullignon,” 2, May 13, 1963.
30 ADV, 746 W 63 (dossier: Harkis Var bâtiments 1963), L. Nourrit, “Note à Messieurs les Ingénieurs en Chef du Génie
Rural, Objet: Logement des Harkis – Exécution des travaux de V.R.D.,” 3, May 9, 1963.
27
304
Over the course of four meetings held from May 6 to 9, Ministries of Repatriates and
Agriculture officials finalized uniform instructions to relay to the construction companies
concerning the size, materials, and layout of the structures in the forest hamlets.31 The exterior and
interior of the buildings in each forest hamlet would be identical, no matter their location. Of the
608 residences to erect in the Provence-Côte d’Azur region, all but 36 (those at the Roquestron
forest hamlet in the Alpes-Maritimes department), would follow the model of “bâtiments A,” each
of which would contain four residences measuring 37.5 square meters, as depicted in figure 31, and
“bâtiments A bis,” with three additional residences of the same size and a meeting room.32 The
Fuveau double forest hamlet, built by the Schroth Company (whose bid the Bouches-du-Rhône
Prefect approved on June 16),33 would contain twelve “bâtiments A” and one “bâtiment A bis” for
the harki population. 34 The social counselor and the camp director would share an additional
prefabricated building, which was located on a higher plain than the harki citizens’ residences in
Fuveau. Per Ministry of Repatriates instructions, the exterior of the buildings was to be constructed
with materials that were “démontable[s] et récuperable[s]” so that they would be available for later
use.35 This instruction signaled that the government was already thinking about reusing the buildings
from what were intended to be temporary residences. As visible on figure 31, the interior of each
apartment had two bedrooms; a living room with a kitchen sink and a woodstove in the corner for
cooking and heating; and a shower room with a small changing area. The Turkish-style toilets
adjacent to the shower room were only accessible by a door, with no lock, on the outside of the
A six-page summary of the meetings are located in ADV, 746 W 63 (dossier: Harkis Var bâtiments 1963), L. Nourrit,
“Note à M. Cullignon,” May 13, 1963.
32 Ibid., 3.
33 ADV, 746 W 68, L. Torrion, “Ordre de service nº 1: Logement de harkis…”
34 ADV, 746 W 63, L. Nourrit, “Note à M. Cullignon,” 2-3.
35 ADV, 746 W 63 (dossier: Schroth), Ministry of Repatriates, “Devis descriptif Nº 50/63, Construction et mise en place
de Logements Préfabriqués pour HARKIS, Bâtiments démontables et récupérables,” June 13, 1963.
31
305
building.36 To put in perspective the insufficient size of the harkis’ residences in the forest hamlets,
those at the Rivesaltes camp were only 17.5 to 12.5 square meters smaller; the overcrowded
conditions in Rivesaltes were blamed, at least initially, for the spread of tuberculosis. Moreover, the
longer that the forest hamlet in Fuveau remained open, the larger the harki families grew. By 1975,
on the eve of a project to reconstruct the barracks, the average harki family in Fuveau had over
seven people (see table 6) and one family had thirteen children.37
Figure 31: Standardized architectural drawing of the “type A” forest hamlet buildings that the Ministry of Repatriates
provided to building companies. Source: ADV, 746 W 63 (dossier: Harkis Var bâtiments 1963).
The director of the Agricultural Engineering Service in Marseille authorized the Schroth
Company to break ground for the Fuveau forest hamlet on August 1, 1963 with the expectation that
The precise prescriptions for the interior of the house, including information such as the ceiling height, the thickness
of walls, wattage of lamps, and the size of pipes can be found in the following document distributed to building
companies: ADV, 746 W 67, Ministry of Repatriates, “Logements Préfabriqués pour Harkis.”
37 ASRBR, social welfare file of Ali H., nº B99100.
36
306
all work would be completed seventy-five days later, on October 15.38 In anticipation of its opening,
in late August the departmental forest hamlet inspector, Captain Cheneau (an active duty military
officer), hired a former SAS and Army captain to be the camp director and his wife to be the social
counselor.39 However, the project would take six extra months to complete, principally owing to
difficulties the building company encountered with the land, which was both rocky and uneven since
the plot was located on a steep hill over coalmines. For each change to the original contract, the
company had to contact the director of the Agricultural Engineering Service in Marseille, who either
approved the changes himself or asked for authorization from the Ministry of Repatriates. For
example, in early August the builders sought permission from Marseille to add another base under
the buildings because they recognized that the ground would be too unleveled to support the
houses.40 With additions of this type the Prefect of the Bouches-du-Rhône, who was responsible for
directly paying the construction company, needed to request additional funds from the Ministry of
Repatriates, which retarded the process. At the outset, based on discussions from the May 6 meeting
cited above, the Ministry of Repatriates had budgeted 12,000 francs to construct and furnish each
residence and 2,000 francs per residence for “Travaux de voirie et réseaux divers” (VRD).41 VRD
included breaking the ground before construction could start; equipping the residences with
electricity, a sewage system, and drinking water; leveling the ground; and adding stairs, railings, and
footpaths.42 Initially, given the 52 harki residences, the Ministry of Repatriates had provided the
ADV, 746 W 68, L. Torrion, “Ordre de service nº 1: Logement de harkis…”
ADBR (Aix-en-Provence), 98 W 325, Letter from the Bouches-du-Rhône prefect’s office director to the mayor of
Fuveau, “Objet: Hameaux de forestage employant des harkis,” Aug. 26, 1963. Captain Cheneau’s first name was not
listed on this document.
40 ADV, 746 W 68, L. Torrion (Director of the Agricultural Engineering Division, Marseille), “Ordre de service nº 2:
Logement de harkis. Chantier de Fuveau,” G.R. 3308, Aug. 9, 1963.
41 ADV, 746 W 63, L. Nourrit, “Note à M. Cullignon,” 4.
42 For a list of the VRD work done at the Fuveau forest hamlet, see: ADV, 746 W 68, Bouches-du-Rhône Department
Agricultural Engineering Service, “Hameau de forestage de Fuveau – Travaux de voirie et réseaux divers devis estimatif,”
Apr. 4, 1964.
38
39
307
Bouches-du-Rhône Prefect with 110,000 francs for VRD in Fuveau.43 However, the final budget,
submitted on April 4, 1964, indicates that this figure had more than doubled to 255,000 francs. Most
of the additional expenses went toward “terrassements en terrain rocheux” and the construction of a
water purification system for the forest hamlet, separate from the one that served the rest of
Fuveau.44
Under pressure from the Ministry of Repatriates to find transitional homes for the harkis,
departmental officials’ haste to procure land for the forest hamlets often led to a situation whereby
this terrain was not immediately (or ever) suitable for construction. These representatives were at the
mercy of those few mayors willing, for whatever reason, to allow a forest hamlet to be constructed
on the edge of their towns. They, therefore, had little choice with respect to the grounds. For
example, in an April 1964 report the director of the Agricultural Engineering Service in Marseille
addressed safety concerns that SFIM Director Pérony had observed in a visit to the nearby Roque
d’Anthéron forest hamlet. The edge of the some of the buildings abutted a cliff overlooking a twolane road, which could prove particularly dangerous for children. Recommending that the Bouchesdu-Rhône prefect ask for funds to erect a lightweight wall around the area, the director wrote: “Ces
quelques précautions permettront d’utiliser un terrain mis volontiers à la disposition du Ministère
des Rapatriés par la Commune de la Roque d’Anthéron, une des rares du Département des Bouchesdu-Rhône à y avoir consenti.”45 Indeed, departmental officials had noted exactly one year earlier,
“…la situation dans le département est difficile. (Peu de forêts – ambiance politique peu
favorable.)”46 Given the lack of options for locales to construct forest hamlets in the Bouches-duRhône department and the mandate from national officials to open them as quickly as possible,
ADV, 746 W 68, L. Torrion (Director of the Agricultural Engineering Division, Marseille) to the Bouches-du-Rhône
Prefect, “Objet: Chantiers de logements de harkis. Crédits V.R.D.,” 2, G.R. 123, Jan. 16, 1964.
44 Ibid., 2 and ADV, 746 W 68, Bouches-du-Rhône Department Agricultural Engineering Service, “Hameau de forestage
de Fuveau,” 3, Apr. 4, 1964.
45 Ibid., 1-2.
46 ADBR, 137 W 460, Prefect’s cabinet, “Note concernant le reclassement…,” Apr. 4, 1963.
43
308
construction companies and departmental officials, some of who had little interest in the projects,
had to make do with the land that mayors offered.
The First Decade in Fuveau: Isolation and Encadrement
On the morning of April 15, 1964, “le camp de harkis de Fuveau” opened on the literal and
symbolic fringes of the 2,600-person village. Fifty harkis and their families, totaling nearly two
hundred people, traveled overnight by military busses from the Rivesaltes camp to their new
homes.47 Described by one newspaper reporter as “pas de luxe, mais [ayant] (tout) le nécessaire,” the
furnishings included twin beds lined up next to each other in the two bedrooms and a table,
sideboard, chairs, and a large cupboard in the living room/kitchen.48 On this sunny day, over thirty
local officials—including a representative of the Red Cross, the secretary general of the departmental
veterans office, and the entire town council—held an official ceremony to greet the families (see
figure 32). 49 The newly appointed regional forestry worksite inspector, Captain Emile Bouleau,
presided over a flag-raising ceremony. Bouleau possessed “[une] connaissance approfondie des
milieux musulmans,” a criterion specified in the Ministry of Repatriates’ memo for selecting forestry
worksite inspectors.50 He had served in the police forces in the Sahara region of Algeria from 1949
to 1962 (including commanding 659 officers and auxiliary police officers for 3 years) and
subsequently acted as the deputy director of the SFIM office at the Saint-Maurice-l’Ardoise camp.51
An article in Le Méridional cites 50 harkis and 250 total people, while one in Le Provençal offers the figure of 48 harkis
and 220 total people. Jacques Vautier, “50 harkis et leurs familles ont trouvé logis, emplois et école dans la région de
Fuveau,” Le Mériodional, Apr. 16, 1964 and “Un village de harkis inauguré près de Fuveau,” Le Provençal, Apr. 16, 1964.
However, a document in the ACNMF archives from May 21, 1964 lists the population as 196 (see table 6) and payroll
records at the ONF cite 50 workers who began employment on the Fuveau forestry worksites on April 15, 1964.
48 Vautier, “50 harkis et leurs familles…,” Le Mériodional, Apr. 16, 1964.
49 Another photograph printed in Le Provençal on Apr. 16, 1964 showed that over thirty personalities attended the
ceremony.
50 ADV, 768 W 62, Minister of Repatriates, 2, Dec. 19, 1962.
51 CAC, 19970146/2 (dossier: Emile Bouleau), Battalion Chief Mourrey (Director of Saharan Affairs in the Oasis
Department), “Fiche concernant le Capitaine du Cadre de Réserve: BOULEAU, Emile,” June 20, 1962, and CAC,
19970146/2 (dossier: Emile Bouleau), Yves Pérony, “Note de Service pour M. Majorel,” Nº 716/SFIM/DIR.MG., Mar.
22, 1963.
47
309
Next, the camp director, Augustin Bernabeu, and the social counselor, Mireille Darmon, both of
who had recently returned from Algeria, directed the harkis to their new homes.
Figure 32: Désiré Henry of the Bouches-du-Rhône Departmental Council amidst other officials welcoming the harkis
and their families to the sylvan surroundings in Fuveau. Source: Le Mériodional, Apr. 16, 1964.
As discussed in the previous chapter, national officials viewed encadrement as the key to the
forest hamlets’ efficacy ever since the earliest discussions about this form of collective reclassement. In
evaluations during spring 1963, shortly after the first forest hamlets opened, both national and local
officials touted encadrement as a reason for their success. Yet, in practice, the national government’s
policies concerning encadrement agents served to isolate forest hamlet harki populations from their
surrounding communities and impede their integration. According to Ministry of Repatriates
directives, these agents were to facilitate interactions with local and departmental officials and teach
the harki population about metropolitan ways of life to promote their integration. However,
prescribing that the forest hamlets should be physically isolated from neighboring small towns (not
to mention cities where social services were more readily available) and simultaneously concentrating
310
the responsibility for ensuring the harkis’ integration into the hands of one or two agents opened the
door to policy failures.
As was the case around the country, the camp director and the social counselor in Fuveau
exacerbated the harki population’s isolation. The Fuveau forest hamlet was situated thirty-five
kilometers from the repatriates service and the branch of the Service des Français Musulmans (SFM),
the successor to the SFIM, in the Marseille Prefecture. It also was upwards of fifteen kilometers
from the forest hamlet and social welfare inspectors’ bureau in Jouques, Aix-en-Provence, and
Montpellier (the offices’ location changed several times over the first decade). Given the infrequent
onsite presence of officials representing these agencies, the structure of the Fuveau forest hamlet’s
governance entrusted all the power for running the day-to-day operations to two people for its first
three years and then to one person for the next twenty years. The lack of oversight made it difficult
to verify that these individuals were carrying out their jobs in a way that was beneficial for the
Fuveau harki citizens’ integration and fostered a situation of clientelism. Through an examination of
the Fuveau forest hamlet’s encadrement personnel, this section analyzes how the government’s forest
hamlet policies set up these agents for failure, particularly when it came to helping the harki
population escape their isolation from the Fuvelian community.
Two different camp directors ran the Fuveau forest hamlet was run by from April 1964 until
August 1967, when this position was eliminated. The same social counselor remained from the time
the forest hamlet opened until her retirement in 1985, after which she continued to reside on the
grounds until 1991. The camp direction of the Fuveau forest hamlet got off to an inauspicious start.
The husband and wife hired in August 1963 to be the camp director and social counselor were fired
on January 29, 1964, before the forest hamlet opened, owing to a dispute with regional forest hamlet
inspector Bouleau and negative performance evaluations in their interim positions as camp director
311
and social counselor in the Glennes forest hamlet.52 The SFIM worked quickly to find a new camp
director. Pérony vetted the next candidate by inquiring with his previous military superiors and the
prefect of the Var department, where the applicant currently resided, about his morality, conduct,
and political leanings (the latter to determine whether he had links to the OAS).53
Following favorable responses from these officials, on February 13, 1964, Pérony named as
the new camp director Augustin Bernabeu, who had been born in Algeria and served for twenty-six
years as a gendarme in the countryside outside of Algiers before fleeing to France at the end of the
war.54 However, six months after the Fuveau forest hamlet opened, Bouleau wrote a letter to Pérony
in which he explained that Bernabeu’s attitude and behaviors, present for some time, created a
difficult atmosphere detrimental to the harki population. He further exposed that the relations
between Mayor Philip and Bernabeu “ne sont pas ceux que l’on serait en droit d’attendre,” which
was a crucial relationship to facilitate the harki population’s integration into the Fuvelian community,
and forwarded the mayor’s request to transfer Bernabeu to a different forest hamlet. Moreover,
Bouleau revealed that the camp director did not get along well with the supervisors of the forestry
worksites and he was at loggerheads with the social counselor, Mireille Darmon.55 Indeed, the
regional director of social counselors noted in a letter to Pérony that Bernabeu only allowed
Darmon to tend to teaching women about domestic tasks (and not the other socio-administrative
tasks her role specified) and regularly insulted Darmon in front of the “Muslims,” which gave them
CAC, 19970146/6 (dossier: Capitaine Roger Vincent), letter from Emile Bouleau to Yves Pérony, “Au sujet de M.
Vincent, Chef de hameau forestier de Fuveau (Bouches-du-Rhône),” Jan. 10, 1964 and CAC, 19970146/6 (dossier:
Capitaine Roger Vincent), Yves Pérony, “Note,” Jan. 29, 1964.
53 CAC, 19970146/1 (dossier: Augustin Bernabeu), letter from Minister of Repatriates Missoffe to the prefect of the Var
department, “Objet: Demande de renseignements sur un candidat à un emploi,” nº 293 SFIM/MG, Jan. 21, 1964.
54 CAC, 19970146/1 (dossier: Augustin Bernabeu), letter from Augustin Bernabeu to Minister of Repatriates Missoffe,
“Objet: Demande d’emploi offerte à d’anciens sous-officiers pour assurer la gestion de hameaux,” 1, June 4, 1963 and
CAC, 19970146/1 (dossier: Augustin Bernabeu), letter from Yves Pérony to Augustin Bernabeu, nº 658 SFIM/MG, Feb.
13, 1964. Bernabeu had solicited this position eight months prior after having seen an announcement for Army veterans
and noted in a letter to Missoffe that thanks to his military service in rural Algeria, “[J]’ai pu acquérir l’expérience des
populations musulmanes.”
55 CAC, 19970146/1 (dossier: Augustin Bernabeu), letter from Emile Bouleau to Yves Pérony, “Objet: Monsieur
Bernabeu, Chef du Hameau Forestier de Harkis de Fuveau,” nº 591/INSP./C., Oct. 17, 1964.
52
312
“une raison supplémentaire de ne pas accorder à leur monitrice le respect et la déférence
indispensable à son action.”56 Interestingly, Bouleau made no mention in his letter of Bernabeu’s
comportment toward the harki citizens, though the camp director’s combative nature suggests that
he likely was not respectful in his interactions with these individuals either. Pérony decided to relieve
Bernabeu of his position as of November 15, 1964.57 This example with Bernabeu was not atypical;
one only has to look as far as the first camp director and social counselor hired for Fuveau who
received negative evaluations for their comportment while running the Glennes forest hamlet. These
situations demonstrate the potential for problems arising out of the setup of the forest hamlets:
power concentrated into the hands of one or two encadrement agents and the lack of oversight to
promptly handle such problems.
Fuveau’s next camp director, Georges Pouron, who assumed his position on November 27,
1964, also was a career military man. Though he was born in France, from 1949 to 1962 he lived in
Madagascar and French Equatorial Africa (AEF), where he served as a gendarme.58 Although no
documentation is available concerning Pouron’s job performance or interactions with the forest
hamlet residents and government officials, his dismissal was significant. It demonstrates that after
these spaces became permanent the national government did not allot to the forest hamlets the
necessary resources to ensure their initial goal to integrate the harki residents into the surrounding
community. Immediately after the Ministry of Social Affairs took over the responsibility of
supplying and funding encadrement agents for the forest hamlets from the Ministry of Interior on
January 1, 1967, it downsized this corps. As a result, Pouron’s position was one of those eliminated.
He received notice that he would be dismissed on August 1, 1967 due to the cuts to the Ministry of
CAC, 19970146/1 (dossier: Augustin Bernabeu), letter from Madame Giravalli to Yves Pérony, “Objet: Difficultés
rencontrées par Madame Darmon,” Oct. 24, 1964.
57 CAC, 19970146/1 (dossier: Augustin Bernabeu), letter from Yves Pérony to Augustin Bernabeu, nº 3870/SFIM/MG,
Oct. 22, 1964.
58 CAC, 19970146/5 (dossier: Georges Pouron), “Notice Individuelle: Georges Pouron.”
56
313
Social Affairs budget, which created another obstacle to integrating the Fuvelian harki population.59
Even though relying heavily on the camp director and social counselor to facilitate the harki
population’s integration was arguably a flawed policy from the outset, it nevertheless was the
primary means that the Ministry of Repatriates provided to integrate the adult forest hamlet
residents into their communities. The SFIM encouraged local officials to play an active part in the
harki citizens’ integration. However, these officials were not employees of the ministries responsible
for the population’s encadrement whose job descriptions included this role. As explained in the
previous chapter, the Ministry of Repatriates listed among the camp director’s responsibilities acting
as an interlocutor with departmental and local officials for individual harkis’ administrative matters
(birth certificates and pensions, for example) and aiding these individuals when they interacted with
municipal and departmental services.60 Pérony had insisted in December 1962 that the best method
to assure the “refugees’” integration and avoid their exclusive reliance on a specialized structure was
to incorporate them into existing social services.61 The camp director’s assistance toward this end—
such as accompanying them to meetings and offering French language classes—was a vital part of
this plan. Eliminating the camp director in Fuveau meant taking away one of the two individuals
responsible for facilitating interactions between the Fuvelian harki population, which numbered 220
people in January 1967, and local government officials.62 In an October 1974 report the Deputy
Secretary General of the Bouches-du-Rhône Office of Liaison and Advancement of Migrants,
Didier Cultiaux, strongly critiqued the situation in the Fuveau and la Roque d’Anthéron forest
hamlets. Referring to this personnel as “très insuffisant en nombre et en qualité,” Cultiaux called the
CAC, 19970146/5 (dossier: Georges Pouron), Letter from the Director of the General Administration for Personnel
and Budget (Ministry of Finances) to the Director of Population and Migrations (Ministry of Social Affairs), June 16,
1967.
60 CAC, 19920149/3/1, Minister of State for Repatriates (SFIM), “Consignes du sous-officier détaché au chantier de
forestage.”
61 ADV, 746 W 62, Minister of State for Repatriates, 1, Dec. 19, 1962.
62 ACNMF, 32/1, Captain Sessa (ATOM-CNMF, Antenne de Marseille), “Étude sur les activités de: L’Antenne de
Marseille et le Centre Social de la Cité ‘Les Tilleuls,’” 3, Jan. 13, 1967.
59
314
choice made by Parisian administrators to eliminate the camp directors in order to cut costs
“deplorable,” because it left only one “isolated” social counselor.63
Cultiaux’s report also included a critique of the living arrangements of the Fuveau and la
Roque d’Anthéron forest hamlets’ social counselors: “[I]l ne faudrait pas que les monitrices sociales
habitent les hameaux… Ce sont elles qui s’assimilent aux musulmans et elles n’ont aucune
autorité.”64 Mireille Darmon was born near Blida, Algeria in 1924 and left her home country at the
end of the war. She began her employment on the Fuveau forest hamlet in February 1964 and lived
there until 1991, when children of harkis reportedly chased her off for the years of oppression she
subjected their parents to, including her failure to give them the tools imperative for their
integration.65 The lack of authority to which Cultiaux refers in 1974 to a large extent stemmed from
the awakening of consciousness by older harki children, now fluent in French through their
education, of Darmon’s negligence. As one daughter of a harki who grew up in the camp related in
an interview: “Elle ne les a fait pas participer à la vie française. Plutôt elle les a aidé à les reculer, à les
cacher pour qu’en fait ils ne s’en sortent pas et qu’ils soient toujours sous sa coupe…”66 While the
camp director was a part of the government’s policies to integrate the harki population, the July
1965 Ministry of Interior report previously cited explains that the social counselor was the “keystone”
of its policies:
Il fallait éviter à la fois qu[e les familles de harkis] rebutent leur voisinage européen et se
replient sur elles-mêmes. Par ailleurs, les adultes presque tous illettrés ne pouvaient
accomplir eux-mêmes aucune des formalités administratives quasi-quotidienne dans des
familles souvent nombreuses. Il a donc été nécessaire d’instaurer un dispositif qui permette
de suivre de très près les implantations de familles musulmanes, du moins quand elles étaient
To remedy this problem, Cultiaux recommended adding an onsite camp director for each forest hamlet and an
itinerant deputy interdepartmental forest hamlet inspector who would circulate among the Bouches-du-Rhône and
Vaucluse forest hamlets. ACNMF, 32/2, Didier Cultiaux to the Bouches-du-Rhône Prefect, “Objet: Situation des
anciens harkis dans le département des Bouches-du-Rhône,” 4-6, Nº 677, Oct. 30, 1974.
64 Ibid., 6.
65 Family K., interview.
66 Ibid.
63
315
groupées. La clef de voûte de toute l’organisation de tutelle sociale est constituée par un
réseau de 90 Monitrices sociales…”67
Mireille Darmon encapsulates the flaws inherent in the importance of encadrement to the
government’s forest hamlet integration policy. This policy instituted a system of dependence on one
person (or two people) fluent in Arabic and/or Berber, whose French linguistic competence and
western social behaviors created a power dynamic open to corruption. At the same time, the
encadrement agents were the sole means the policy provided for teaching the adult harki population
French language and customs so that these individuals would be able to function independently. The
French language simultaneously was source of and legitimized their power over the harki citizens
under their authority. When the agents did not fulfill their socio-educative mission the harki
population became mired in a situation of dependence on those whose responsibilities included
filling out paperwork for government assistance, translating mail from government agencies,
interacting with schoolteachers and medical personnel, and distributing clothes from charitable
organizations.68 The Fuveau forest hamlet’s isolation and the lack of personnel in place to properly
oversee that Darmon was indeed teaching the harkis and their wives French perpetuated and
exacerbated this situation of dependence. At the same time, during Fuveau’s first decade the
government did not enact policies giving financial support to hire language teachers who could aid
the social counselor in this function. A 1977 report by the ONF noted that illiteracy still posed a
grave population for many of the harkis on forestry worksites throughout France. For thirty weeks
between May 1976 and May 1977 state educational agencies offered the men in the Bouches-duRhône and Vaucluse departments four hours of French classes per week, demonstrating that neither
AN, 5 AG 1/22, Minister of the Interior, “Rapport sur l’application…,” 122, July 26, 1965. This report was issued
after the Ministry of Interior had assumed the Ministry of Repatriates’ functions concerning the forest hamlets and
before it would pass them on to the Ministry of Social Affairs.
68 AN, F1a 5138, SFIM, “Role et attributions….”
67
316
Darmon, nor anyone else had fulfilled this role.69 Moreover, a 2002 report concerning the Fuveau
harki population’s living conditions noted that only three of the twenty-two harkis and their wives
interviewed were able to express themselves in French without a translator.70 The Fuveau harki
citizens’ lack of French language skills, in part the responsibility of Darmon, added another obstacle
to leaving the forest hamlet along with being assured a job and free housing while residing there.
At the Fuveau forest hamlet, a clientelistic system emerged in which Darmon profited from
the adult harki population’s incomprehension of the French language and social welfare system.
Several interviewees offered examples of such situations. Darmon was in charge of distributing
bottles of gas for cooking and heating each family’s home. According to one harki, while each family
had the right to five bottles per month, Darmon would sometimes hold these back from harkis she
disfavored, which created jealousies among them.71 Another interviewee revealed that she would
make the harkis do chores in order to receive the bags of clothes that the Red Cross donated for
their families.72 This same person revealed that following her sister’s grave accident in which she was
struck by a car near the forest hamlet entrance (the camp was located off a two-lane highway) and
entered into a coma Darmon instructed her father to sign a paper, explaining that the document was
so that social services would not take his daughter away. The family later learned that he ceded his
right to sue the driver.73
The Fuvelian harkis’ dependence on social counselor Darmon to read their mail and fill out
forms sometimes impeded their access to benefits to which they were entitled given their postimperial classification as “harkis” by the French government. This financial assistance was crucial to
helping the harki families become more independent. Records from the departmental veterans’
CAC, 19910097/48, Regional Director of the ONF for the Mediterranean Region to the Director of the ONF,
“Objet: Gestion des anciens harkis par l’Office National des Forêts,” 71-72, nº 168 GR/JD/CR, Apr. 28, 1977.
70 Archives Communales de Fuveau (hereafter ACF), Arts et Management de la Ville, “Étude sur le devenir du site de la
cité Brogilum à Fuveau,” 20, Oct. 2002.
71 Abdelkader Z., interview.
72 Family K., interview.
73 Ibid.
69
317
office (ONAC) in Marseille demonstrate that Darmon helped twenty-five harkis fill out paperwork
and procure the proper documentation from other state agencies for veterans’ cards, once they
became eligible for this status following a December 1974 law. 74 Nevertheless, there is scant
evidence of Darmon fulfilling her obligation of helping the families to fill out the paperwork to seek
social welfare services to which they were entitled at this office and the Bouches-du-Rhône
Repatriates Service. Of the twenty-five files examined at the former office and the twenty social
welfare files consulted at the latter, representing thirty-three harkis, Darmon only helped two
individuals seek financial assistance, one in 1977 and the other in 1981.75
Cultiaux’s October 1974 report about the harki population in the Bouches-du-Rhône
department noted that harki citizens residing in Marseille, including those in the Tilleuls cité urbaine
(an urban collective reclassement housing project), were in a stronger financial position than harkis
outside of the capital. The proximity of social welfare offices and the personnel in place to inform
the harkis of the benefits to which they were entitled facilitated their access to these benefits.76
Indeed, consulting sixteen Tilleuls harkis’ files in the ONAC office in Marseille and thirty-two
Tilleuls harkis’ files in the Bouches-du-Rhône Repatriate Service proves that these urban harkis
accessed social welfare services sooner (as early as 1966) and far more often than the Fuveau harkis
while Darmon was the social counselor. Cultiaux even went as far to declare that the non-Marseillais
harkis residing outside the forest hamlets, even if few had succeeded in integrating, were better off
than those in the forest hamlets. He reported that local governments, one potential source of
JORF, “Loi nº 7401944 du 9 décembre 1974 donnant vocation à la qualité de combattant aux personnes ayant
participé aux opérations effectuées en Afrique du Nord, entre le 1er janvier 1952 et le 2 juillet 1962,” Dec. 10, 1974,
12284. The records for the twenty-five men who requested veteran’s cards while residing at the Fuveau forest hamlet are
located in the Bouches-du-Rhône Departmental Veterans Office in Marseille (AONACBR).
75 The social welfare files are part of the department’s repatriates’ service archives (ASRBR), located in a storage space at
the Bouches-du-Rhône Prefecture in Marseille. The two instances where Darmon filled out paperwork for Fuveau harkis
are: ASRBR social welfare file of Ben Abdallah R., Madame Darmon to the Director of the Bouches-du-Rhône
Repatriates Service, “Objet: Subvention d’installation,” Jan. 26, 1977 and AONACBR, individual file of Larbi S., Form
for financial assistance filled out by Mireille Darmon, Jan. 14, 1981.
76 ACNMF, 32/2, Didier Cultiaux, “Objet: Situation des anciens harkis…,” 2.
74
318
financial assistance, were more or less accommodating and repatriates associations (he did not
indicate whether these were harki, pied-noir, or joint associations) helped the harkis to seek financial
assistance.77 A better financial situation, if the harkis were informed about their rights to social
welfare benefits, made leaving the forest hamlets possible, should a family desire. This demonstrates
another way that the set up of the forest hamlets’ encadrement structure exacerbated the Fuveau harki
population’s isolation.
The First Decade in Fuveau: Isolation and Integration
According to a report in one of the local newspapers, Le Provençal, during the April 15, 1964
ceremony at the Fuveau forest hamlet Mayor Philip predicted the harki population’s “rapide
intégration dans la commune.”78 This prediction would be incorrect in part owing to the failure of
local officials, particularly Mayor Philip (who would be in office until March 1977), to enact policies
to help integrate the harki residents into their community or treat them as such. An article the same
day in the other local newspaper, Le Mériodional, noted the similarity between the Fuveau forest
hamlet and the life the harkis left behind by sharing an observation of one of the schoolteachers
who had just migrated from Sétif: “Tout à fait un village d’Algérie!” Indeed, the forest hamlet
resembled a microcosm of colonial Algeria, yet it was missing the traditional social structures the
harki population knew in Algeria. The reporter Jacques Vautier, who referred to himself as
“arabisant” (leaving one to wonder whether his familiarity with Arab cultures was acquired from
having lived in North Africa), collected the following testimonies from the harkis as they made their
way toward their new homes, as depicted in figure 33:
-Ici, c’est notre pays!
-Nous ne demandons qu’à travailler, à assurer la vie de nos familles et de nos enfants.
77
78
Ibid., 3.
“Un village de harkis…,” Le Provençal, Apr. 16, 1964.
319
-Nous nous plairons ici, car il y a du soleil.
-La campagne est sauvage, elle est belle, nous serons à l’aise.
-Enfin, quitter la tente, avoir un toit, un travail, un motif de vivre!
-Ces drapeaux qui nous accueillent sont des signes de joie; l’avenir est moins sombre,
moins triste.
-Que désirer encore? Que les voisins nous considèrent comme des amis. Nous ferons
tout pour que ce bon voisinage soit réel.79
Figure 33: A harki and his family walking to their new home in the Fuveau forest hamlet. Pictured in the background are
several of the 150 square meter buildings, each of which housed 4 harki families. As this image illustrates, the terrain was
still uneven and rocky. Source: Le Provençal, Apr. 16, 1964.
Despite this positive outlook, the isolated setup of the Fuveau forest hamlet would challenge
the harki population’s integration into the Fuvelian community. Fuveau’s harki citizens indeed
became a separate and separated community, living a refugee-like existence and treated as
79
Vautier, “50 harkis et leurs familles…,” Le Mériodional, Apr. 16, 1964.
320
undesirable immigrants by many around them. The Fuveau forest hamlet, and by extension the
other seventy-one ones organized in the same way, were spaces that isolated their harki
populations—and did not promote integration. As mentioned in the previous chapter, by the mid1970s government officials evaluated the forest hamlet integration policy as a failure, both the policy
in itself and its implementation. The national government’s initial vision for the forest hamlets in
1962 set them up to be unsuccessful. Its subsequent decision in 1966 to change the status of twentyfive of the forest hamlets, including Fuveau, from “temporary” to “permanent,” negated a central
feature of the government’s integration policy for the forest hamlets. Finally, as will be treated in the
next chapter, the choice in the mid-1970s to rebuild the houses on the same site and keep the harkis
under the encadrement of a social counselor, employed over a decade ago to help them transition into
life in metropolitan life, thwarted the remaining possibility for the integration policy to succeed.
In the testimonies quoted above, the harkis expressed relief that they had left the ways of life
imposed on them at the Rivesaltes camp (though the tents were gone by January 1963). They
welcomed a new life on this bucolic landscape that reminded them of the homeland they were
forced to recently flee. And the last interviewee hoped to become integrated among the town’s other
residents. Yet the harki population’s isolation, which was threefold—exile from their established
communities and ways of life, few interactions with local and departmental government officials,
and distance and differentiation from other Fuvelian residents—challenged their integration into
Fuveau’s community.
Departmental and local officials grouped the families in the Fuveau forest hamlet together
artificially, which demonstrated little regard for the harki population’s ethnic, linguistic, and regional
differences. The diversity of populations and the transient nature of the forest hamlets resulted in
the harki population’s social disintegration (désintégration sociale) and isolation from customary ways of
321
life, including their hierarchical societies comprised of family units.80 The social counselor and/or
the camp director took the place of what had once been the role of elders. Regulations instituted by
the Ministries of Repatriates and Agriculture stipulated that once a harki was no longer employed on
a forestry worksite he and his family must give up their residence in the forest hamlet immediately.
While in some cases families resisted this rule, payroll records from the ONF nevertheless provide
an accurate depiction of the composition the population in the Fuveau forest hamlet.81 According to
these records, a total of 124 harkis worked on the 2 worksites which, accounting for a few instances
of adult children forestry workers, translated into roughly the same number of families residing in
the Fuveau forest hamlet from 1964 until today.82 Cross-referencing these names with civil status
documents at the Fuveau town hall, social welfare documents at the Bouches-du-Rhône Prefecture,
and military records from the departmental veterans office reveals the birthplace of sixty-four of the
harki men, including nineteen of forest hamlet’s original inhabitants.83 There was a wide geographical
distribution across Algeria focused in rural areas. With European settlers concentrated on the
Algerian coast and in cities, ways of life in many of these rural and mountainous regions remained
relatively undisrupted until the Algerian War for Independence when the French Army recruited
In Sociologie de l’Algérie Pierre Bourdieu describes the various types of social disintegration that took place in colonial
Algeria owing to disruption caused by the imposition of French governing, economic, and social structures. Pierre
Bourdieu, Sociologie de l’Algérie (Paris: Presses universitaires de France, 1963 [1958]), 105-26.
81 The ONF was created in January 1966 under the authority of the Ministry of Agriculture to oversee the upkeep of and
policies concerning France’s state-owned forests. The archives for the Departmental division of the ONF in the
Bouches-du-Rhône department, including payroll records (the harkis were paid every two weeks), are located in its Aixen-Provence office (AONFBR).
82 There is one unique case of a non-ONF worker living in the forest hamlet. The widow of a harki who was captured in
1962 and died five years later as a result of injuries suffered during his incarceration migrated to France in 1986 and
moved into the Fuveau forest hamlet where her cousin resided. ASRBR, social welfare file of Taouès C.
83 With some Fuveau residents having served in multiple capacities during the Algerian War for Independence, these
records show seventeen harkis, six moghaznis (responsible for protecting SAS members), two GMS (rural police forces),
six conscripts or volunteers in the regular army, and two men described as agricultural workers for the French Army. In
addition to fighting in the Algerian War for Independence, one man was a tirailleur algérien during the interwar period,
one served in a peacekeeping regiment of tirailleurs algériens in Germany after the Second World War, three fought in the
Second World War, and two fought in the Indochinese War. As this information suggests, the ages of the men employed
on the Fuveau worksites varied; the oldest worker was born in 1907 and the youngest was born in 1946. The birthdates
for these men can be found on their individual pay slips in the AONFBDR.
80
322
rural Algerians for their knowledge of the countryside.84 Living in such close proximity in Fuveau—
more than two hundred individuals together on a five-acre plot—unsettled their various traditional
social and cultural structures.
The flow of harkis into and out of the Fuveau forest hamlet during its first decade—
transient during the first four years, then increasingly stable—was another example of disrupting
traditional structures and challenged their integration. These population movements leading some
harki families to reside in the Fuveau forest hamlet for a brief amount of time, as advocated in the
government’s initial policies, added another obstacle to harki population members’ integration into
the surrounding community. When the forest hamlets were originally envisioned as a two-year
solution, the Ministry of Repatriates viewed such movements as a sign of progress that workers had
found housing and jobs in the private sector.85 Yet the Ministry of Repatriates’ vision for the forest
hamlet experiment shifted from temporary to permanent in early 1964, before Fuveau opened. The
movements to and from the Fuveau forest hamlet during the 1960s most often meant that families
were transferred to or from other forest hamlets—they did not symbolize the population’s
integration. As mentioned above, 50 harkis began working at the Fuveau forestry worksites on the
day after their April 15, 1964 arrival.86 ONF payroll records show that two workers left after two and
a half months (one left a forwarding address in the Moselle department) and five other workers
would move from Fuveau by the end of 1964.87 Following the national level policies enacted in
January 1966 referenced in chapter 4, most new arrivals to the Fuveau forest hamlet transferred
Rivet, Le Maghreb à l’épreuve, 294-95.
CAC, 19910097/40, M. Doniol, “Objet: Inspection des chantiers…,” 16.
86 This figure does not include one of the fifty harkis who arrived from Rivesaltes on April 15 and immediately was
hospitalized in the Psychiatric Center in nearby Aix-en-Provence. He and his family returned to the Rivesaltes camp a
little over one month later. CAC, 19920149/5/11, Emile Bouleau to Yves Pérony, “Objet: Famille F.,” Nº 288/Insp.,
May 27, 1964. The additional worker appears to be the son of a harki, who disappeared from payroll records after May
14, 1964. It is unclear whether the eighteen-year-old (Abdelkader ben Mohamed H.) left the forest hamlet since a man
with the same last name whose name was Mohamed (ben is an Arabic term used in names to indicate “the son of”) and
was thirty-four years older, Mohamed ben Abdelkader H., remained employed by the ONF in Fuveau until June 27,
1967. AONFBR, Fiches de paie, Fuveau.
87 AONFBR, Fiches de paie, Fuveau.
84
85
323
from other forest hamlets, particularly those that were slated for closure. By the end of April 1968,
the Fuveau forestry worksites had employed a total of 102 harki men, which translated into roughly
the same number families residing in the forest hamlet during its first 4 years. After this time,
population movements slowed down significantly. A handful of those who moved to Fuveau in the
late 1960s came from the “Centre de transit de Lascours” in the Gard. Until 1970, this site was a
reception center for the harkis, some of whom had been incarcerated by the FLN, who migrated
from Algeria after the Rivesaltes camp closed. According to available records from ONAC, four
men who were placed in the Fuveau forest hamlet passed through Lascours, each after having been
prisoners of the FLN for approximately five years.88 The last two harkis to begin working at the
Fuveau forest hamlet transferred from the recently-closed St. Hilaire forest hamlet in the Allier
Department on August 1, 1973.89
Besides their regional diversity, Fuveau’s harkis were ethnically and linguistically
heterogeneous. As the social counselor Darmon noted in a July 1972 report, the Fuveau forest
hamlet included families of Arab, Kabyle, and Chaouia descent.90 Information gathered from a
variety of sources exposes the ethnicity of 45 of Fuveau’s 124 harkis: 23 Arabs, 9 Kabyles, and 13
Chaouias. Among its original 50 inhabitants, these sources reveal 10 Arabs, 3 Kabyles, and 2
Chaouias. This ethnic diversity, which in turn meant linguistic differences (both Arabic and
Tamazight plus regional dialects within these languages), posed problems for the harki population
during its initial years. One Fuveau harki who was among the forest hamlet’s first inhabitants
explained in an interview that those who were not conversant in both Berber and Arabic were at a
AONACBR, dossiers for Mohamed (Honoré) B., Areski B., Mohamed C., and Smaïl R.
AONFBR, Fiches de paie, Fuveau. The two men were Mohamed C. and Daoudi K.
90 CAC, 19870256/7, Mireille Darmon, “Hameau de Fuveau. Situation du hameau,” 1, July 1972. The Chaouias are of
Berber descent and tend to live in the Aurès mountains in northeastern Algeria just south of Constantine. Kabyles, the
largest Berber group in Algeria, and Chaouias, the second-largest group, speak dialects of the Berber language
(Tamazight).
88
89
324
disadvantage.91 This challenge was in addition to their struggle to comprehend the French language.
Beyond linguistic obstacles, ingrained ethnic rivalries from Algeria sometimes contributed to a
strained atmosphere among the population and, on occasion, sparked physical fights. As Regional
Forest Hamlet Inspector Louis Sessa wrote in 1969 to the President of the Departmental Council
explaining a fight between a Kabyle harki and a group of Arab harkis on Fuveau’s forestry worksites,
“[I]l existe une animosité latente entre les arabes et les kabyles.”92 Indeed during the previous year,
fights between Arab and Kabyle harkis had resulted in the transfer of two harki families to other
forest hamlets.93 The social counselor’s report in mid-1972, by which time the population had
stabilized, noted that the situation was calmer: “Il règne une bonne ambiance et une grande
solidarité.”94 Darmon’s positive evaluation nevertheless must be weighed against the fact that any
problems she reported in the forest hamlet would have ultimately been a reflection of her inability to
control the harki population and therefore a self-indictment of her job performance.
Fuveau was not the only ethnically diverse forest hamlet; for example, Jouques and la Roque
d’Anthéron, both in the Bouches-du-Rhône department, counted Arabs and Kabyles among its
forest hamlet residents. 95 Yet, Fuveau’s ethnic heterogeneity contrasted with some other forest
hamlets, such as the four in the Vaucluse. Three of the Vaucluse forest hamlets—Apt, Cucuron, and
Pertuis—were constituted of mostly Arab families with a few Chaouias from the Saint-Mauricel’Ardoise camp and its fourth, Sault, contained exclusively Kabyle families. In fact, in a letter to
SFIM director Pérony the mayor of Sault specifically requested Kabyles, which necessitated that
Forest Hamlet Inspector Emile Bouleau travel to the Rivesaltes camp to select them since there
Abdelkader Z., Interview with the author, Dec. 19, 2007, Fuveau.
ADBR, 135 W 51, Inspector Louis Sessa to Louis Philibert, “Objet: Harkis de Fuveau,” Jan. 10, 1969.
93 AONFBR, dossier: Fiches individuelles Fuveau 1968, “Compte rendu de modification d’effectif, chantier de Fuveau,”
June 6, 1968, and “Compte rendu de modification d’effectif, chantier de Fuveau,” Aug. 23, 1968.
94 CAC, 19870256/7, Darmon, “Hameau de Fuveau.”
95 A report written by Colonel Schoen of the CNMF after a July 1964 visit to the Marseille region noted: “Jouques…
Incidents entre Kabyles (forestiers) et Arabes (autres emplois). Que sera-ce dans un an ou deux[?]” ACNMF, 32/1,
Colonel Schoen, “Visite à Marseille,” 2, July 22, 1964. Indeed Jouques’ twenty-five original families included eighteen of
Kabyle descent and seven of Arab descent. Abi Samra and Finas, “Regroupement et Dispersion,” 96.
91
92
325
were only five Kabyle families remaining in the nearby Saint-Maurice-l’Ardoise camp. 96 Mayor
Charles Martel justified his preference for families of Kabyle origins in his letter to Pérony by his
misguided belief in “the Kabyle myth,” the superiority of Kabyles to Arabs: “Cette assimilation [des
harkis] au point de vue ethnique ne présente pas de gros inconvénients puisque les kabyles de race
berbère, appartiennent, comme les Provençaux, au même fond racial méditerranéen….” This sixpage letter moreover demonstrated that the mayor had given much thought to the place of the
harkis in his town and the setup of their living situation.97 The absence of documentary materials
makes it difficult to determine precisely why ethnicity was not a factor for the government officials
who assigned harki families from the Rivesaltes camp to the Fuveau forest hamlet initially. However,
this situation suggests departmental and local government officials’ lack of understanding of—or
perhaps interest in—the harki population.
The availability of isolated land in the forest on the outskirts of Fuveau, due to the mining
industry’s decline, fit into the Ministry of Repatriates’ prescription for choosing a site. It is
nevertheless unclear whether Mayor Philip’s position as the president of the union of Bouches-duRhône mayors exerted pressure on him to offer his village as a potential site or whether he took a
genuine interest in the harki population. What is clear from the proceedings of Fuveau’s monthly
town council meetings, containing minutes from its open and closed sessions, is that the mayor
never brought up the topic of establishing a forest hamlet in this open forum. Therefore, it is
impossible to gauge the effect of his department-wide office on this decision, and more important,
the opinion of other local officials. By not publicly discussing this decision, whose implications
included increasing Fuveau’s population by nearly 10 percent in one day, Mayor Philip set a
precedent for excluding the harkis from municipal affairs and the municipality itself. In fact, the first
Moumen, Les Français musulmans en Vaucluse, 84-85.
CAC, 19920149/3/2, Letter from Charles Martel to Yves Pérony, 1, May 3, 1963. On the Kabyle Myth, see Patricia M.
E. Lorcin, Imperial Identities. Stereotyping, Prejudice and Race in Colonial Algeria (New York: I.B. Tauris & Co Ltd, 1995).
96
97
326
time that the harkis were discussed in Fuveau’s town council meetings was on April 24, 1972—eight
years after the forest hamlet opened.98 Before this time, any discussions that took place about the
harkis—of which there exists no documentary evidence in Fuveau’s municipal archives—were
conducted off the official record. This suggests that the harkis’ affairs did not concern the local
government in the same way as those of other town residents.
A statement issued by town council in January 1972 demonstrates the distance that the local
government inserted between itself and Fuveau’s harki citizens. It also shows that discussions
concerning harki citizens did indeed take place among at least some members of the town council
(the participation of all members is unknown) outside of official meetings. Following a heated
debate among Fuvelians over whether to continue to educate harki children at the forest hamlet
after a new elementary school had opened at the other end of the municipality, the town council
published the following statement in the January 1, 1972 edition of Le Provençal: “Le Conseil
municipal rappelle… que l’administration qui régit la bonne marche du hameau forestier (Narkis)
[sic] incombe exclusivement au ministère des Affaires sociales…” 99 No official record of this
discussion exists in the register of town council meeting proceedings. Moreover, the Fuveau
government’s first public statement concerning the forest hamlet residents made clear its position:
harki citizens were external to the municipality’s affairs.
This exclusion of the harkis from municipal affairs was not the case in all towns. In contrast,
when the mayor of la Roque d’Anthéron was deciding whether to open a forest hamlet in his
municipality, he convoked an emergency session of the town council on March 3, 1963 with the
agenda item: “Installation de Harkis.” After reading aloud an official request from the departmental
prefect to open a double forest hamlet in the town with a description of how the forest hamlet
Fuveau, Register of Town Council Meeting Proceedings, “Nº 72-32 – Ecoles pour les enfants de Harkis,” Apr. 24,
1972.
99 “Fuveau: Avis de la mairie,” Le Provençal, Jan. 1, 1972. It is unclear whether the newspaper or the town council
misspelled “harkis.”
98
327
would function, Mayor Paul Onoratini opened the floor to his colleagues. They unanimously
accepted the prefect’s proposal, offering, among other reasons: “Considérant… que l’union et
l’amour réciproque des hommes du globe est nécessaire et qu’il appartient à tous d’aider au
maximum nos frères qui souffrent de quelques races qu’il appartient et que la France se doit d’aider
les Harkis, qui, au péril de leur vie ont choisi notre pays et notre civilisation.”100 Topics relating to
the harkis—including schooling for the children and the construction of the buildings—would be
on la Roque d’Anthéron town council’s agenda five additional times before “Le hameau de la Baume”
opened in May 1964, and then six more times before the end of the decade. The difference between
these two mayors’ initial vision of the forest hamlet’s relationship to their municipality—Captain
Emile Bouleau noted in December 1965, “Le hameau de la Roque, grâce à Monsieur Onoratini, est
le plus complet et mieux organisé de toute la région”—had future implications for the harkis.101 As
will be explained below, when the national government enacted a policy requiring municipalities to
take part in rebuilding houses in the early 1970s, the mayor of la Roque d’Anthéron opted to move
the families closer to town to help their integration, while the mayor of Fuveau decided to build new
houses on the same isolated site.
While harki citizens’ isolation from Fuvelians was both physical and cultural, officials would
later judge their physical isolation to be the greater impediment to the success of the forest hamlet
integration policy. The national government’s conception for the forest hamlets’ setup and
functioning combined with Mayor Philip and departmental officials’ choice to situate Fuveau forest
hamlet on a secluded plot three kilometers from town created, from the outset, their physical
isolation. The tone that the mayor and his town council subsequently set by treating of the harkis as
outsiders to municipal affairs exacerbated this physical isolation and likely had a negative effect on
La Roque d’Anthéron, Register of Town Council Meeting Proceedings, “Séance extraordinaire du 3 mars 1963.”
AN, F1a 5142, Inspecteur Bouleau, “Fiche de renseignements sur les personnes qui ont aidé les harkis – M. le maire
Onoratini,” Dec. 3, 1965.
100
101
328
Fuvelians’ perception of the isolated population beyond any existing “racial” prejudices some
individuals’ harbored toward Algerians. The harkis’ different “culture,” as defined by Sayad in the
introduction, already set them apart and isolated them from other town residents. When the harki
population arrived, few of them spoke French. They did not share a common history with other
Fuvelians. The two group’s religions and traditions were different. Their social customs and attitudes,
such as those concerning marriage and family, varied. Yet, these cultural differences did not in
themselves preclude the harki population’s integration. Instead, departmental officials evaluating the
forest hamlet integration policy in the early and mid-1970s deemed the population’s physical
isolation from their surrounding community the greatest factor impeding its success. The director of
the Bouches-du-Rhône Prefecture’s SFM office wrote in April 1973 concerning the relationship
between “Europeans” and the harkis living in the forest hamlets: “[L]es hameaux, …par leur
situation géographique gênent considérablement les contacts avec les européens. Seuls les jeunes
cherchent à s’évader du milieu familial et tribal pour se lier avec des jeunes amis européens des
villages voisins.”102 Eighteen months later, the deputy secretary general of the Bouches-du-Rhône
Office of Liaison and Advancement of Migrants succinctly summarized the situation of the
department’s four forest hamlets: “[L]es problèmes résultent surtout de leur isolement…”103
According to members of the harki population and social counselor Darmon, townspeople
did not visit the Fuveau forest hamlet, with the exception of occasional itinerant merchants. At the
same time, the forest hamlet’s location coupled with a lack of means to easily travel into town (few
harkis had cars until the early 1970s and no buses linked the forest hamlet to the town), limited the
harki population’s ability to seek out interactions.104 Town hall records show that nine harkis chose
to remarry their wives in a French ceremony, ostensibly as an attempt to integrate into society.
ACNMF, 32/2, Captain Louis Sessa (Director of the SFM office in the Bouches-du-Rhône department and Marseille),
“Rapport Annuel 1972 concernant les Rapatriés Français-Musulmans,” 5, Apr. 11, 1973.
103 ACNMF, 32/2, Cultiaux, “Situation des anciens harkis,” 3.
104 CAC, 19870256/7, Darmon, “Hameau de Fuveau.”
102
329
Among the eighteen official witnesses for these marriages, in eleven cases this person was a Fuvelian
harki, in six instances either the camp director or social counselor was a witness, and in one case the
town hall secretary filled in.105 The Fuvelian harki community remained apart from other town
residents. Given that the social counselor’s job was to act on the adults’ behalf in administrative
affairs, these individuals had little necessity to leave their isolated living space. Departmental and
local officials, responsible for giving the harki citizens the same access to social services as any other
residents, according to December 1962 instructions from the Minister of Repatriates, did not enact
policies that would facilitate such interactions and help the harki population to integrate into their
community.106
The only regular contact between harki adults and townspeople during the first decade
consisted of women walking down highway 96 into town for food provisions. The few sources
concerning townspeople’s attitudes toward the adult harkis during their first decade in Fuveau
present a fairly consistent picture of indifference, unwelcomeness, and distance. In an interview
conducted in July 2008, a harki wife, Ferhat K., and her daughter Aïcha who grew up in the camp
related that during school breaks and for trips to the grocery store in the late afternoons, the women
relied on their children to act as interlocutors with vendors. But they were on their own for the
farmer’s market held twice a week in the mornings. Aïcha described the harki wives’ interactions
with townspeople: “Avec les anciens il [n’]y avait pas de problèmes. Ils arrivaient enfin à se faire
comprendre. Il y avait des rejets bien sur avec certaines personnes, mais en groupe, non…” Ferhat
then proudly interjected in Arabic (which her daughter translated), “On est allées avec nos
vêtements traditionnels à Fuveau et personne ne nous a embêtées.” The October 1974 report by the
Bouches-du-Rhône Prefecture cited above offers the following explanation for this common type of
behavior by residents of its four towns with forest hamlets: “Pour les Français de souche, et en dépit
105
106
Fuveau town hall, “Actes de Mariage.”
ADV, 746 W 62, Minister of State for Repatriates, 1, Dec. 19, 1962.
330
des services rendus, [les harkis] sont toujours considérés comme des ‘Arabes’ et ne sont l’objet
d’aucune bienveillance particulière.”107 Aïcha added that while the parents did not encounter many
problems, the harki children (especially the boys), on the other hand, were increasingly negatively
perceived by townspeople.108 Indeed, explicit tensions between Fuvelians and the harki population,
particularly the children, became pronounced in the late 1970s.
Wives of harkis residing in the Fuveau forest hamlet were particularly affected by the
physical and cultural isolation, which was compounded by their native culture discouraging women
to leave the private space. While their children attended school and their husbands worked, they
stayed in the forest hamlet performing household chores and taking care of small children. Given
their immobility aside from when they went to the market without their husbands or children, they
had little opportunity to learn about French culture and language, except from the social counselor.
As explained above, the national government’s setup of the forest hamlets enabled this system of
dependence and the lack of oversight meant that no means were in place to ensure that the social
counselor was performing her duties of teaching French language. Only a few harki wives residing in
Fuveau ever learned how to speak French; instead they relied initially upon the social counselor and
subsequently their children to communicate with the outside world. One harki wife recounted in
2003 about the situation in a published testimony: “Je suis l’une des rares femmes de harkis à parler
français. Les autres ne peuvent rien faire sans leurs enfants ni faire les courses, ni téléphoner, ni lire
une lettre, ni sortir du camp, ni remplir le moindre papier administratif.”109
Government officials’ vision for the harkis’ jobs isolated them from the their neighbors
because they worked long hours on remote sites exclusively with other harkis. From Monday to
Saturday mornings, trucks picked up the two teams of harkis dressed in military fatigues at the
ACNMF, 32/2, Cultiaux, “Situation des anciens harkis,” 2, Oct. 30, 1974.
K. Family, interview.
109 The testimony of Mrs. Zendji-Choucha was published in: Gladieu and Kerchouche, Destins de Harkis, 106.
107
108
331
Fuveau forest hamlet at seven o’clock. Their work was physically taxing with workplace accidents a
regular occurrence, to which personnel files at the ONF attest. One harki, Mokhfi L., even perished
in 1993 of a heart attack just after he left the worksite (this followed three serious accidents over the
previous decade).110 A July 1975 Ministry of Agriculture document listed ten out of Fuveau’s thirtytwo workers as “handicapped.”111 According to payroll records from the 1960s, the men worked ten
hour days between twenty-three and twenty-seven days per month, which left them little time to
interact with townspeople.112 The teams were dropped off each morning at a worksite located within
a fifteen to twenty kilometer circumference of Fuveau, an area comprising 72,000 hectares. Annual
reports in 1967 and from 1969 to 1972 showed that each year the men worked on the outskirts of
approximately twenty different municipalities in addition to Fuveau. Their tasks primarily consisted
of preventing forest fires by clearing brushwood from the ground, building firebreaks, and creating
and maintaining paths.113 The only non-harkis whom the men came into contact with were the
forestry agents who were responsible for their encadrement. In addition to the truck driver, a pied-noir
worksite director (chef de chantier) who spoke Arabic supervised each team’s work and handled
discipline problems. According to the 1964 local newspaper articles, one of the directors was an
“agent technique” and the other was an “agent de surveillance,” however both were listed as “chefs
de chantier” in ONF payroll records.114 During the summer months, the harkis also served as first
responders to forest fires in eleven municipalities including and surrounding Fuveau. 115 As the
General Director of the Forestry Commission claimed, they were “pompiers auxiliaries,” a situation
See the following personnel files in the AONFBR for examples of workplace accidents: Ali H., Benaissa L., Mokhfi
L., Mokhtar R., Benabdellah R., Salah S., and Saïd T.
111 CAC, 19910097/48, Ministry of Agriculture, “Travailleurs Ex Harkis, Recensement au 1er Juillet 1975,” 1.
112 AONFBR, Fiches de paie.
113 ADBR, 125 W 173, National Forests Office, Bouches-du-Rhône department, “Chantier de forestage de Fuveau.
Programme des travaux prévus pour l’exercice 1967, description du secteur.” Similar documents for 1969, 1970, 1971,
and 1972 are also located in this carton.
114 “Un village de harkis…,” Le Provençal, Apr. 16, 1964 and Vautier, “50 harkis et leurs familles,” Le Méridional, Apr. 16,
1964.
115 CAC, 19910097/40 (dossier: Harkis – Correspondance 1964), “Procès-verbal de la réunion tenue le 24 mars 1964, à
la 38ème Conservation des Eaux et Forêts à AIX EN PROVENCE en vue d’étudier les conditions de la participation
des harkis à la lutte contre les incendies de forêts dans les Bouches-du-Rhône.”
110
332
analogous to the harkis’ role during the Algerian War for Independence when they served as
“troupes auxiliaires” in the French Army.116
The set up of the Fuveau forest hamlet gave the children of harkis the most opportunities to
interact with those outside of the camp. Initially, they attended school on site in two mobile
classrooms provided by the Ministry of Repatriates with two teachers who had previously taught in
Algeria (one was Algerian and the other French).117 The classes primarily focused on their French
language skills in order to prepare them to attend local establishments. Beginning in the fall of 1965,
the younger children attended Ecole primaire d’Emile Loubet 2.5 kilometers away in Fuveau and the
middle school aged children travelled to a nearby town with other Fuvelian children to attend the
Collège de Gréasque. The linguistic transition was particularly difficult for older children, some of
who were put into classes with younger children. One harki son, Bouziane, related that because he
was thirteen when he finished CM2 (the equivalent of fifth grade in the United States), he was too
old to start the middle school in Gréasque. Along with several other children in Fuveau, Bouziane
was sent to the Centre éducatif d’Ongles (Basses-Alpes), opened in March 1965 on the site of a
former forest hamlet with the purpose of providing older harki children remedial and vocational
education classes in construction.118 Younger children struggled in school as well, owing in part to
the French language not being spoken at home. Data gathered in 1972 for the harki children living
in the thirty-six forest hamlets, seventeen cités urbaines, and two camps revealed that 68.7 percent of
them were held back at least one elementary school grade and they were double as likely as their
CAC, 19910097/40 (dossier: Harkis – Correspondance 1964), Director General of the Forestry Commission to the
Forestry Commissioner in Nice, “Objet: Emploi des ex-harkis affectés à la defense des forêts provençales contre
l’incendie,” 1, EF/D4 Nº 7142, Oct. 13, 1964.
117 Vautier, “50 harkis et leurs familles,” Le Méridional, Apr. 16, 1964. ADBR, 131 W 377, Communication téléphonique
de M. Aymeric (Inspecteur primaire), “Situation au 30 mai 1964.”
118 Bouziane K., interview with the author, Fuveau, July 23, 2008. AN, 5 AG 1/22, Ministry of Interior, “Rapport sur
l’application,” 124, July 1, 1965.
116
333
“French” peers to be behind children of the same age.119 In this report social counselor Darmon
classified the Fuveau harki children’s overall situation as “niveau scolaire bas.”120
Until early 1971 other than the few children at the center in Ongles, Fuveau’s harki children
attended local schools, which allowed them to come into regular contact with the French language
and other French citizens. However, a classroom for young children was opened on the site once
again in 1971, but only remained briefly in part given the intervention of the town council. Despite
having claimed in a public communication published in the January 1, 1972 edition of Le Provençal
that the forest hamlet lie outside of its jurisdiction, during its April 24, 1972 meeting, the councilmen
voted unanimously not to keep this classroom open. They cited three reasons: the lack of space in
the forest hamlet given the growing families, the importance of integrating harki children into local
schools, and the possibility of creating a nursery school for young children in the Barque
neighborhood. The previous year a daycare center had opened in la Barque, located at the other end
of Fuveau from the forest hamlet. The town council envisioned creating a school, however,
according to Ministry of Education rules, opening a nursery school necessitated forty-five children,
which was more than resided in la Barque. The town council reasoned: “C’est pour cette raison que
nous espérons que les jeunes enfants du hameau forestier pourraient amener la fréquentation de la
nouvelle maternelle à l’effectif requis.” These words and its conclusion, the first mention of
integration in the minutes, indicates that finding enough children to attend the nursery school was
the most important factor in the town council’s decision: “Conclusion: Avoir un effectif suffisant
pour faire reconnaître la classe maternelle pour les 2 motifs suivants: A) les enfants dont les parents
résident à la Barque ou dans son périmètre immédiat. B) Les enfants du hameau forestier dont le
transport est régulièrement assuré par notre service de ramassage scolaire. A partir de cette
119
120
ACNMF, 6, Servier, “Enquête sur les Musulmans Français,” Part 2, 3, 1972.
Ibid., Part 2, 31.
334
réalisation nous sommes convaincus que l’intégration se fera parfaitement.”121 Integration was an
afterthought for the harki children who would have to travel six kilometers to the school.
Date
May 1964122
Nov.
1964123
Dec. 1965124
Jan. 1967125
Oct. 1967126
Jan. 1972127
Jan. 1973128
May 1974129
Jan. 1975130
May 1976131
Number of
harki males
50
47 (1 unmarried)
Number of
wives/ widows
--46
Number of
descendants
--99
Number
of families
50
48
Total number
of inhabitants
196
193
49 (5 unmarried)
--39
35
33
31
32
31
--------29
28
-----
--------168
160
-----
49
------33
31
32
31
223
220
--236
230
219
230
240
Table 6: Population of the Fuveau forest hamlet from 1964 to 1976.
The reason for the school debate was the high birthrate in the forest hamlet, which far
outpaced that of Fuvelians and set the harki population apart from their neighbors. According to
data from INSEE, the average Fuvelian household, including those residing in the forest hamlet, in
1968 and 1975 had three people.132 In comparison, as table 6 illustrates, the average forest hamlet
household had five people in 1965 and over seven people in 1975. An examination of family records
Fuveau, Register of Town Council Meeting Proceedings, “Nº 72-32 – Ecoles pour les enfants de Harkis,” Apr. 24,
1972.
122 ACNMF 32/1, unsigned document, “Bouches-du-Rhône,” May 21, 1964.
123 AN, F1a 5142, Inspector Emile Bouleau, “Recensement numérique des anciens harkis et supplétifs musulmans
refugiés dans le département des Bouches-du-Rhône,” Dec. 9, 1964. These figures can also be found in multiple
documents in ACNMF, 32/1.
124 ACNMF, 32/1, CNMF, “Département des Bouches-du-Rhône, Effectifs et répartition des Musulmans Français (au
31 Décembre 1965).”
125 ACNMF, 32/1, Captain Sessa, “Étude sur les activités…” 3.
126 AONFBR, “1967 Salaires FSIRAN Chantiers de La Ciotat, Jouques, Fuveau,” District Chief Antoine Casorla, “Objet:
Convention générale d’emploi des anciens harkis,” Jan. 24, 1968.
127 ACNMF, 6, Servier, “Enquête sur les Musulmans Français,” Part 2, 31.
128 ACNMF, 32/2, Sessa, “Rapport Annuel 1972…,” Annexe 1.
129 CAC, 19910097/42 (dossier: Harkis – Correspondance 1974), Réunion Interministérielle du 30 mai 1974.
130 ACNMF, 32/2, Louis Sessa (chargé de l’Antenne de Marseille du SFM), “Rapport d’Activités pour 1973 et 1974,”
Annexe 1, Jan. 1, 1975.
131 ADBR, 135 W 41, Louis Sessa (Director of BIAC in Marseille), “Rapport sur la situation des Français Musulmans
installés dans le département des Bouches-du-Rhône,” Map, May, 18, 1976.
132 INSEE, Fuveau (13040 – Commune): Chiffres clés, “FAM G1M – Evolution de la taille de ménages,”
<http://www.insee.fr/fr/bases-de-donnees/default.asp?page=statistiques-locales.htm.>.
121
335
at the Fuveau town hall and the Marseille repatriates service reveals that eighteen of the thirty-one
harki families who remained permanently in the forest hamlet had eight or more children, including
one family with eighteen children. A large number of children made it difficult for these families to
find suitable housing alternatives outside of the forest hamlet. Yet their living conditions in the
prefabricated lightweight houses had rapidly degraded by the mid-1970s, as was the case in forest
hamlets and other collective reclassement situations across France. This deterioration and the failed
integration of these individuals who had been isolated from view for over a decade soon became
visible to the French public. The next chapter shifts back to the national level to analyze the
government and harki population’s responses to the precarious situation of France’s harki citizens.
336
CHAPTER 6
Refugees, Repatriates, and Immigrants. The “Harki Problem”
-LE PROBLEME DES HARKISLes problèmes posés en France par l’accueil,
l’installation et le reclassement des algériens
musulmans ayant servi en ALGERIE aux côtés de
l’Armée ou de l’Administration Française en qualité
de harkis, supplétifs, etc … restent toujours
douloureux et sans solution.1
-Service des Français Musulmans (SFM), Ministry
of Interior, March 12, 1974
After over a decade of silence and inaction at the national level regarding the harki
population’s difficulties integrating into the French community, the government agenda in the mid1970s began to include what some documents referred to as the “le problème des harkis,” including
that in the epigraph, which bore this phrase as its title. As many sources indicate, one reason—or
perhaps the principal reason—that government officials demonstrated interest in these citizens’
problems was that they remained an untapped voting bloc. A report written by the director of the
SFM office in the Bouches-du-Rhône department in 1973 noted: “Le motif de l’intérêt porté à nos
ex-supplétifs bien souvent oubliés même par ceux qui les ont utilisés et aimés: les Elections
législatives [de 1972]. Il fallait trouver des électeurs faciles à manier, simples, aimant les promesses,
respectueux des traditions, marqués par leur passé militaire… et on a redécouvert ‘Les Harkis.’”2
After “rediscovering” France’s harki citizens, most promises that candidates and officials at all levels
made to them would, however, go unfulfilled.
The concrete issues of particular concern for the population included job training for men
and children, scholastic performance among the youth, access to indemnity payments, and the living
1
2
CAC, 19880077/20, SFM (Ministry of Interior), “Le problème des harkis,” Mar. 12, 1974.
ACNMF, 32/2, Sessa, “Rapport Annuel 1972…,” 1.
337
conditions of the population “regroupée” in the cités d’accueil, cités urbaines, and forest hamlets. At this
time, only 10 percent of the harki population resided in these spaces—labeled “camps” and “ghettos”
by leaders of the newly emerging harki associations and appearing in numerous press articles.3
Nevertheless, in the mid-1970s, thanks in large part to the actions and words of leaders of harki
associations, which at times linked the harkis’ plight to that of Jews during the Second World War,
the camps significantly emerged as the lasting symbol of the entire harki population’s failed
integration and marginalization.
A more abstract—and contradictory—issue consistently plagued the entire harki population:
its troubled relationship with French citizenship. Despite their service to France during the Algerian
War for Independence and the fact that almost all harki repatriates chose to become French citizens,
government officials and the French population often confused them and their family members for
Algerian immigrants. Following the suggestion from a report issued by a working group to study the
challenges that the harkis faced, Prime Minister Pierre Messmer issued a letter in June 1973 to the
members of his cabinet, regional prefects, and departmental prefects underscoring that they must be
treated as French citizens and given access to accompanying benefits. Referencing the nationality
procedure that most harkis had chosen to undergo upon their arrival in France, Messmer wrote:
“Pour mettre fin à une situation contraire à la législation [du 21 juillet 1962] et, de surcroît,
inacceptable au plan humain, je vous demande de bien vouloir rappeler à toutes les administrations
et services relevant de votre autorité que les rapatriés français musulmans jouissent de la plénitude
des droits attachés à la possession de la qualité de citoyen français et ne doivent, par conséquent,
faire l’objet d’aucune mesure discriminatoire qui pourrait les mettre dans une situation différente de
Jean Servier estimated that the total harki population on Jan. 1, 1972 comprised 162,000 individuals, 16,000 of which
were “regroupés” in collective reclassement housing. ACNMF, 6, Servier, “Enquête sur les Français Musulmans,” Part 1, 1
and Part 2, 1.
3
338
celle de leurs concitoyens d’origine métropolitaine.”4 Nine months later, the president of the CNMF
informed the Minister of Labor that Messmer’s memo had not made a significant difference, for
harkis still faced discrimination from government agents and potential employers: “Des
informations recueillies en divers points de notre territoire me donnent à penser que l’esprit de cette
circulaire n’a pas encore été entendu par tous les agents en contact avec cette catégorie de nos
compatriotes.”5
Analogous to how the decisions concerning the harkis’ repatriation, nationality, and housing
that government officials made immediately after the war had enduring negative consequences for
the population, those made in the mid-1970s, which in effect enacted few changes to previous
policies and continued to treat the harki population as inferior repatriates to the pieds-noirs, would
have repercussions for decades. This chapter first places the challenges the harki population faced
owing to its muddled status—the “harki problem”—into national and transnational political
contexts. It then gives an overview of the national-level policy discussions concerning the
population that began in 1970. It examines harki association leaders and harki children’s increasingly
violent demonstrations, during which they used the camps as a vehicle for their protests and
essentialized harki citizens’ plight by laying their focus on the spaces, commonly associated with
refugees. It next looks at the effect of these protests on public opinion and government policies,
including the government’s revision to its integration policies. Despite the heterogeneity of
individual experiences, harki citizens’ de jure status as “Muslim repatriates” and de facto treatment
as Algerian immigrants (and sometimes refugees) by government officials and other French citizens
translated into a similar exile from French society and a common “oubli” by the government. The
chapter concludes by analyzing how the resulting national-level policy changes influenced the
SHAT, 3R 79/7, Pierre Messmer to the Ministers and Secretaries of State and the Regional and Departmental Prefects,
“Objet: Situation des Musulmans français originaires d’Algérie ayant opté pour la nationalité française après
l’indépendance de l’Algérie,” 2, nº 8582/SG, June 26, 1973.
5 ACNMF, 3/7, letter from Alexandre Parodi to the Minister of Labor, Employment, and Population, Mar. 14, 1974.
4
339
Fuveau forest hamlet and returns to the discussion appearing throughout this section concerning the
harki population’s unique relationship to France as “harki citizens.”
Reevaluating Integration Policies
In the early 1970s the entire harki population faced two particular challenges that
underscored their differences, one relative to “French” repatriates and the other to Algerian
immigrants. First, the government had passed a law in July 1970 to financially compensate “French
citizens” who had lost property and material goods in its former overseas possessions.6 While these
benefits were technically available to the harkis, Social Affairs Minister Edgar Faure noted almost
three years after its passage that the law had “very rarely” been applied to them.7 The nature of
group landownership in rural Algeria—family members often owned parcels together—and the fact
that many fled the violence in Algeria in haste without important papers made it very challenging for
the harkis to provide the proper documentation to benefit from the legislation. Moreover, family
members kept possession of the harkis’ land after they fled to France, thereby countering the
argument that they had lost their property. 8 As previously demonstrated, the harki population
already lagged behind the pieds-noirs in terms of government aid for jobs and housing under the
1961 Boulin law. Now the government created a new gap between the two groups of repatriates,
which only would be fully addressed with a law in 2005.
The second challenge the harki population encountered during the early 1970s with respect
to their status was that government officials and other French citizens often mistook harkis for
Algerian immigrants. As explained in chapter four, harki citizens’ portfolio had shifted from the
JORF, “Loi nº 70-632 du 15 juillet 1970…”
CAC, 19910097/42 (dossier: Harkis – Correspondance 1973), letter from Minister Edgar Faure to Prime Minister
Pierre Messmer, 1, Mar. 8, 1973.
8 Conseil Economique et Social, “La Situation sociale des enfants de Harkis,” prepared by Hafida Chabi (Paris: JORF
Avis et Rapports du CES, 2007), 44-45.
6
7
340
Ministry of Repatriates to the Ministry of the Interior—responsible for internal affairs and named in
opposition to the Ministry of Foreign Affairs—and finally to the Ministry of Social Affairs (which
subsequently became the Ministry of Labor, Jobs, and Populations), and specifically the Office of
Population and Migrations. This office primarily dealt with immigrants. The confusion between the
harkis and other North African immigrants was particularly difficult in light of the increase in racism
toward foreigners during the early 1970s and a rupture in the political relationship between France
and Algeria. Following numerous racist incidents in France during summer 1973, including clashes
at a public meeting of the far right party Ordre Nouveau encouraging French citizens to resist
against “l’immigration sauvage,” in September the Algerian government, citing this rise in racism,
stopped all emigration to France.9 Mistaken by many for Algerian immigrants, not fully integrated
into repatriate status, and with some dwelling in refugee conditions in camps the harki population
was entangled in a unique situation.
On August 13, 1970, Minister of Interior Raymond Marcellin held a meeting with
representatives of the ministries of Interior, Agriculture, and Labor and a delegation of twelve harkis
concerning harki citizens’ “insertion” in the French community and the specific problems they faced.
Five of these men resided in forest hamlets, two in cités urbaines, two in the Bias and Saint-Mauricel’Ardoise cités d’accueil, and three outside state-funded housing (known as isolés) in the Paris region.
Marcellin opened the meeting by assuring the harkis that their service in the French Army was not
forgotten, to which this gathering of “concerned” high-level officials attested. Following the twohour discussion, more than half of which focused on the situation in the forest hamlets, the Ministry
of Interior issued a list of eleven resolutions. These included providing job training in professions
other than the saturated building sector for harki children, making a uniform policy for forestry
workers’ salaries, and undertaking a study to find housing in the Parisian suburbs for the isolés. Atop
9
Weil, La France et ses étrangers, 101.
341
the list of resolutions sat the “desserrement” of the forest hamlets. At the start of the meeting, the
harkis from the Gonfaron (Var) and Mouans-Sartoux (Alpes-Maritimes) forest hamlets
communicated that the lightweight prefabricated dwellings (37.5 square meters in size) had become
too small for their growing families. Consequently, they asked to be rehoused in larger permanent
structures. One harki stated his preference for remaining on the same site, whereas another
preferred moving to an apartment in town so as to be integrated into the community. The officials
concluded that a case-by-case study must be completed in each forest hamlet to determine how to
proceed. Yet, no policy changes were implemented to address these harkis’ concerns.10
The pressure on the government to take action in support of the harki population primarily
emanated from two sources, both of which underscored that harki citizens suffered from their
unique status. First, in 1971 the CNMF financed a sociological study undertaken by University of
Montpellier sociologist Jean Servier to examine the population’s situation. With the support of the
Director of Population and Migrations, Michel Massenet, Servier and his researchers carried out an
ambitious project to collect data from the thirty-six forest hamlets, seventeen cités urbaines, and two
cités d’accueil.11 This data was complemented by the results of a lengthy questionnaire for the entire
harki population that Massenet’s office sent to each departmental Prefecture. The list of twentyeight questions for the families included information about past and present residences, nationality,
allocations received, relations with French and Muslim neighbors, and the children’s schooling and
All the information in this paragraph comes from: CAC, 19910097/41 (dossier: Harkis – Correspondance 1970),
“Compte rendu de la réception d’une délégation d’anciens supplétifs de l’Armée Française par Monsieur le Ministre de
l’Intérieur le jeudi 13 août 1970, à 12 heures,” 1-7. Earlier that morning, the nine government officials had convened to
prepare for the meeting, the summary of which is located in: CAC, 19910097/41 (dossier: Harkis – Correspondance
1970), “Compte rendu de la réunion qu’a tenue le jeudi 13 août 1970, à 11 heures, en son cabinet, Monsieur Marcellin,
Ministre de l’Intérieur.”
11 Servier and his researchers, including Anne Heinis, distributed questionnaires to each of these spaces either by mail or
by conducting onsite visits. ACNMF, 6, Servier, “Enquête sur les Musulmans Français,” Part 2, 1. As explained in
chapter 2, in spring 1962 Massenet had chaired a Commission whose findings advocated that the government
immediately institute a repatriation plan for the harkis, a recommendation that was not followed.
10
342
employment.12 Servier was invited to the first meeting of the Barbeau group to share the findings of
his 114-page study.13
The results attested to the failure of the Boulin law and subsequent policies to successfully
integrate harki citizens’ into French society vis-à-vis housing and employment, with the camps
visibly demonstrating the government’s shortfalls. Beyond the concrete problems of housing and
employment, the study highlighted that the harki population’s unique status as Arab and Berber—
and not Caucasian—repatriates complicated their integration into French and Algerian communities
alike. As Servier wrote, some—though not all—of those who arrived in France from a higher social
standing and with education provided examples of successful integration in the French community.
However, those from a humble background composing the majority of the population, most often
illiterate peasants, felt marginalized. He explained: “Ils occupent en France des emplois très
modestes, sont désorientés et déracinés, rejetés par les Algériens qui les considèrent comme des
traitres à la cause algérienne, rejetés également par les Français qui les assimilent aux migrants
algériens, les humiliant ainsi profondément.”14 They were of Algerian origin, but they were not
Algerian “immigrants” who chose to come to France for work opportunities; they had been forced
from Algeria by their choice to fight for France during the Algerian War for Independence. More
importantly, they were French citizens—like “French repatriates”—but they were not always treated
as such since they were easily confused by virtue of their appearance and given the government
offices for immigrant populations to which they were assigned.
The second source of pressure on the government to take concrete actions to ameliorate the
harki population’s situation was external—associations. Even before the first harkis arrived in
The accompanying memo instructed prefects to distribute the questionnaire to every fifth harki family on its
alphabetical record so as not to skew the results in one direction or the other. CAC, 19870256/7, Office of Population
and Migrations (Ministry of Labor, Jobs, and Population), “Objet: Enquête sur la situation des anciens supplétifs,”
PSM.3/nº 5184, Dec. 10, 1970.
13 CAC, 19910097/42, Barbeau Commission, “Rapport sur la situation…,” 1.
13 CAC, 19910097/42, Faure to Messmer, 2.
14 ACNMF, 6, Servier, “Enquête sur les Musulmans Français,” Part 2, 17.
12
343
France associations were created to help them, notably the CNMF. Officially founded in January
1963 by French citizens who had fought alongside “Muslim French” soldiers during the war, the
association listed as its goals in the January 5, 1963 edition of the JORF: defense of moral and
material interests of the Muslim French; giving aid to the government as it welcomes, helps to settle
in, and finds jobs for Muslim French families in France; and coordinating and leading private
initiatives with the same objectives. In 1964 the government began subsidizing the association’s
efforts, which included selecting and overseeing social counselors in collective reclassement housing
and distributing funds to specific harki causes.15 Yet, a 1974 CNMF memo underscored, while
tireless in efforts to aid harki citizens receive the benefits they were legally owed and defend their
interests generally (evident by the thirty-four cartons of archival documents catalogued in 2008), the
CNMF was an association for harkis, not led by harkis.16 Its two presidents were Vice-President of
the Council of State Alexandre Parodi (1963-1979) and a wealthy French banker, André Worsmer
(1979-2007), who during his obligatory military service from 1959 to 1960 had commanded a harka
and, in his words, “Je suis devenu attaché à mes hommes.”17
Beginning in 1971 harkis and Muslim Algerian elites—many of whose material conditions
they in part owed to the aid of the CNMF—created their own associations. Several of these leaders
believed that the associations led by “Frenchmen” were too paternalistic and failed to represent
harki citizens’ interests. The more successfully integrated harkis now focused their attention on
recognition of their contributions to the French Army and indemnity payments, like those the
French repatriates benefitted from under the July 1970 law.18 Meanwhile the CNMF, for example,
concentrated on social and educational policies, particularly aiding the numerous harki children who
ACNMF, 3/2, “Comité National pour les Musulmans Français,” 1.
ACNMF, 16, “Note sur les ‘Harkis,’” 2, Feb. 1974. Mohand Hamoumou details the history of and challenges that
associations for and by harkis faced in: Hamoumou, Et ils sont devenus harkis, 297-315. On the tensions between harki
children and the CNMF, see Wormser, Pour l’honneur des harkis, 75-80.
17 André Wormser, Interview with the author, Paris, July 11, 2003. Wormser’s posthumously published memoirs include
a history of the association: Wormser, Pour l’honneur des harkis.
18 Hamoumou, Et ils sont devenus harkis, 299.
15
16
344
performed poorly in school.19 While 140 harki associations were created between 1971 and 1983,
which was a sign of mobilization, the associations’ lack of unity (in part due to the large number)
diminished their success. 20 Ahmed Djebbour—the leader of the Front National des Rapatriés
Français de Confession Islamique (FNRFCI), a confederation of forty associations from regions
stretching from le Nord to Provence created in October 1972—quickly became the most visible
voice of the harki population to the public and the one most listened to by government officials.21
French Algeria proponent Jean-Marie Le Pen influenced Djebbour, a former pro-French Algeria
deputy for the department of Algiers. Le Pen had recently established a far right political party, the
Front National, from which Djebbour borrowed the beginning of his association’s name.22 The end
of the name, “de confession islamique,” was Djebbour’s attempt to distance harki citizens from
other “Français Musulmans,” particularly given the tensions between France and Algeria at the
time.23 Referring the harkis as “sous-rapatriés” given their special nationality process, the difficulties
they had accessing indemnities, and the fact that many were still “parked” in inhuman living
conditions in camps, Djebbour maintained as his objective educating the public about the
population’s plight. During an October 1973 press conference, excerpts of which were reprinted in
Le Figaro and Le Monde, Djebbour stated, “Nous voulons attirer l’attention de l’opinion publique
indifférente et mal informée sur les anomalies inadmissibles qu’une administration fait peser sur les
rapatriés musulmans.”24
In the context of the Servier report and these new associations placing pressure on the
government, on October 27, 1972, Pierre Messmer—the newly appointed Prime Minister—called
ACNMF, 20, “Le Comité National pour les Musulmans Français (1962-1999),” 2.
Hamoumou, Et ils sont devenus harkis, 297. Le Mouvement d’Assistance et de Défense des Rapatriés Musulmans
d’Afrique du Nord (MADRMAN), founded on August 1, 1971, is recognized as the first harki association led by a
“Français musulman.” However, its president, Ahmed Kaberseli was a Muslim elite, not a harki.
21 Hamoumou, Et ils sont devenus harkis, 300-01.
22 CAC, 19880077/20, Ministry of Interior, “Le problème des harkis,” 5.
23 SPDP, dossier: Harkis, J.C. Guillebau, “Les harkis oubliés par l’histoire. I. Onze ans après…,” Le Monde, July 3, 1973.
24 SPDP, dossier: Harkis, Robert Begou, “Les rapatriés musulmans: ‘Une législation discriminatoire nous est appliquée,”
Le Figaro, Oct. 6-7, 1973.
19
20
345
on Minister of Social Affairs Faure to appoint an interministerial working group to study the
problems facing harki citizens and propose measures to facilitate their integration.25 Government
officials increasingly recognized that the government had failed in its integration policies for the
harki population, particularly visible in the remaining collective reclassement living situations, and
realized that the harkis had difficulties accessing the July 1970 law. Having lost their possessions in
Algeria and not being compensated with indemnity payments in France left many harki families,
with a birthrate that government officials described as “extrêmement elevé,” in dire financial
straits.26
The “Barbeau Group,” named after its president Michel Barbeau (Director of Population
and Migrations at the Ministry of Social Affairs), was comprised of representatives from eleven
different ministries and Alexandre Parodi, the president of the CNMF.27 The commission met four
times between December 1972 and January 1973 before submitting a report in March 1973. In the
letter accompanying this document Faure wrote to Messmer that one of the group’s principal
findings concerned “la qualité de ‘Français’” of the harkis, despite their French nationality. Faure
explained:
La méconnaissance de cette citoyenneté est trop souvent la source de regrettables
malentendus de la part des populations environnantes et se traduit dans les rapports des
intéressés avec les organismes et services publics par un sentiment de frustration, étant
donné l’option par laquelle ils se sont volontairement engagés dans la communauté française.
La reconnaissance de la nationalité française devrait donc leur permettre de recourir au
premier chef à l’intervention pleine et entière des services publics de droit commun.
Toutefois, compte tenu de leur insertion sociale, une action complémentaire ainsi que des
mesures nouvelles de caractère spécifique s’avèrent nécessaires dans bien des domaines.28
CAC, 19910097/42 (dossier: Harkis – Correspondance 1973), “Rapport sur la situation des Français Musulmans
rapatriés d’Algérie,” 1, Mar. 8, 1973.
26 Ibid., 3.
27 The government representatives at the meetings were from Prime Minister’s office and the ministries of National
Defense; Social Affairs; Interior; Economy and Finances; National Education; Planning, Housing, and Tourism;
Agriculture and Rural Development; Public Health; Veterans and War Victims; and Youth, Sports, and Activities. The
list of names is available in the Appendix to the Barbeau Group’s report: CAC, 19910097/42, “Rapport sur la
situation…”
28 CAC, 19910097/42, Faure to Messmer, 2, Mar. 8, 1973.
25
346
In short, through the presentations made by Servier, who attended the first meeting, and committee
members’ own observations (including those of Parodi, whose association had been actively
involved in aiding the harki population for a decade), the Barbeau group recognized the
shortcomings of government integration policies for the population and its suffering due to
questions concerning their French citizenship.
On May 22, 1973, the deputy director of Prime Minister Messmer’s cabinet presided over an
interministerial meeting to discuss this report and the list of solutions proposed to address “the harki
problem.” 29 The topics covered included, among others, housing, recognition, and their social
integration into welfare services. In terms of collective housing situations, particularly the forest
hamlets, the Barbeau group favored “le brassage des populations afin de favoriser l’intégration des
musulmans français à la communauté française.” Toward this end, it recommended giving two
hundred thousand francs immediately to expand the size of the forest hamlet residences and
building fifty HLM apartments in 1973 and two hundred per year for the next five years to rehouse
the harkis in the cités d’accueil and forest hamlets.30 The government approved allocating 150,000
francs for improvements on the forest hamlets and erecting 50 HLM residences in 1973, but it
deferred the decision concerning construction of 200 HLM residences per year. 31 This goal of
constructing enough HLM apartments to close the forest hamlets and two cités d’accueil by 1978
would not be met; at the end of 1979 nineteen forest hamlets and one of the cités d’accueil still
remained.
The government representatives at the meetings were from Prime Minister’s office and the ministries of Justice;
Interior; Armies; Economy and Finances; National Education; Planning, Housing, and Tourism; Agriculture and Rural
Development; Public Health; and Veterans and War Victims. The list of names is available on the last page of the
minutes: CAC, 19910097/42 (dossier: Harkis – Correspondance 1973), Secretariat General of the Government, “Relevé
des décisions de la réunion interministérielle du 22 mai 1973 consacrée à l’examen des suites à donner au rapport fait par
le ministère du travail, de l’emploi et de la population sur la situation des Français musulmans rapatriés d’Algérie,” June 7,
1973.
30 CAC, 19910097/42, “Rapport sur la situation…,” 3-4.
31 CAC, 19910097/42, Secretariat General of the Government, “Relevé des décisions…,” 1-2.
29
347
To address some of the moral problems the harkis and their families faced because the
French government had not officially recognized their contributions during the Algerian War for
Independence, the Barbeau Group recommended honoring them with the “Titre de Reconnaissance
de la Nation,” as had been given to all French soldiers who had spent more than three months
fighting in Algeria. This title would also give them access to material benefits from the office for
veterans (ONAC).32 At the May 23 meeting, the minister of Veterans explained that the harkis (and
other auxiliary forces—moghazni policemen, for example) were not officially classified as soldiers
during the war and therefore ineligible for government recognition.33 Even though they fought on
the battlefield their status as “auxiliary soldiers” rendered them civilian combatants during the war,
which Parodi wrote in a subsequent memo to the minister of Veterans made them feel as though
they were not being treated as “Français à part entière.”34 The government therefore did not approve
the Barbeau Group’s proposal. Instead it asked the minister of Veterans in concert with other
ministries to create a special status to recognize the harkis and other auxiliary forces’ efforts during
the war.35 These auxiliary soldiers would not receive the government’s official recognition until a
1994 law, a source of consternation for the entire harki population. Finally, concerning harki citizens’
difficult “insertion sociale” and the confusion between them and other Algerian immigrants, the
Barbeau Group suggested placing special counselors in each departmental prefecture who could
direct the harkis to the proper offices that handled claims for French citizens. It also advocated
adding itinerant social counselors in areas with large harki populations (notably those surrounding
forest hamlets and cités urbaines) who could help them access social welfare benefits. 36 The
government authorized these suggestions, though it left the implementation up to a variety of
CAC, 19910097/42, “Rapport sur la situation…,” 6.
CAC, 19910097/42, Secretariat General of the Government, “Relevé des décisions…,” 2.
34 ACNMF, 3/11, Alexandre Parodi to Minister of Veterans André Bord, “Titre de Reconnaisance de la Nation,”
undated.
35 CAC, 19910097/42, Secretariat General of the Government, “Relevé des décisions…,” 3.
36 CAC, 19910097/42, “Rapport sur la situation…,” 8.
32
33
348
officials.37 Additional counselors would not be added to the prefectures until after a wave of protests
from the harki population in 1975 and a new round of interministerial meetings.
In the months following the May 1973 interministerial meeting, Djebbour began to escalate
his rhetoric concerning harki citizens’ mistreatment by the government. He honed in particularly on
the camps—cités d’accueil, cités urbaines, and forest hamlets—as symbols of failed integration and
misery and even made parallels with the encamped harkis’ situation and that of Jews during the
Second World War. After the police refused to allow former harki soldiers participate in a July 14,
1973 procession at the Arc de Triomphe alongside other veterans of the Algerian War for
Independence, Le Monde quoted Djebbour calling on the government to close “camps de
concentration où sont entassés de très nombreux harkis.”38 At this time approximately 16,000 of the
180,000 harki population members resided in camps, however, the impression Djebbour gave
gravely exaggerated harki citizens’ situations—both the number residing in camps and the actual
living conditions—by inserting images of the Holocaust. 39 These images came at a time when
memories of the Second World War emerged into the public space following decades of
repression.40 Djebbour even addressed a letter to French parliamentary members demanding the
closure of the “camps d’internement,” his designation for the forest hamlets, claiming the support of
650,000 French Muslims. 41 With his fiery words finding their place into France’s most read
newspapers and to French politicians, albeit infrequently, Djebbour increasingly condemned the
CAC, 19910097/42, Secretariat General of the Government, “Relevé des décisions…,” 4.
Djebbour also criticized French government officials for recent negotiations with their Algerian homologues, whom
he linked to the massacre of sixty thousand harkis following the war: “Que les fantômes de soixante mille cadavres
soient présents aux yeux de nos gouvernements lorsqu’ils doivent négocier avec nos assassins d’hier.” SPDP, dossier:
Harkis, “Les harkis et la fête nationale,” Le Monde, July 20, 1973.
39 ACNMF, 16, CNMF, “Note sur les ‘Harkis,’” 3.
40 According to Henry Rousso, 1971-74 marked “the return of repressed memories,” the third phase in what he terms
“the Vichy syndrome.” The first phase (1944-54) was “grief and contradictions” and the second (1954-1971) “the
resistancialist myth” during which the French population repressed its memories of the war. Rousso, Le syndrôme de Vichy
de 1944 à nos jours 2nd ed. (Paris: Seuil, 1990), 19.
41 CAC, 19880077/20, SFM, “Le problème des harkis,” 2.
37
38
349
government’s inaction to dismantle the camps and “[reconnaître] dans les faits de leur dignité de
citoyens français.” 42
Government officials believed that the turning point for Djebbour’s call for violence was
Arab coalition forces’ successful surprise attacks on Israel during the October 1973 Arab-Israeli war,
which gave his cause inspiration to fight for the harki population’s rights as French citizens in their
country. 43 In November and December 1973, Djebbours’s press conferences and meetings,
according to these officials, “visèrent… à dramatiser la situation et à pousser les harkis à des actes
extrêmes.”44 Djebbour accelerated his rhetoric about the camps and the harki population’s French
citizenship. In March 1974, the Catholic newspaper La Croix published an article, which quoted him
using “camps de la honte” as a moniker for the cités d’accueil and labeling the forest hamlets “bagnes”
and “Cayenne transplanté en France” (a reference to the forced labor penal colony in French
Guyana).45
Protests and Revising Integration Policies
At this time, a more militant figure emerged who rallied harkis and particularly their children
to publicly protest, initially through non-violent actions and then by armed violence. These acts
forced government officials’ hands. M’hamed Laradji, a native Algerian who fought as a “regular”
soldier in the French Army (not in the auxiliary forces), had been the leader of FNRFCI in
Normandy before creating his own association, the Confédération des Français musulmans rapatriés
d’Algérie et leurs amis (CFMRAA), which became a rival to the FNRFCI.46 Laradji’s difficulties
accessing indemnity payments from the French government to compensate for the land his family
SPDP, dossier: Harkis, “Le sort des anciens harkis,” Le Monde, Jan. 15, 1974.
While the Arab coalition was unable to hold its initial gains, these successful strikes nevertheless vindicated the Arab
World’s embarrassing losses during the 1967 Six-Day War.
44 CAC, 19980077/20, SFM, “Le problème des harkis,” 3.
45 SPDP, dossier: Harkis, Anne Gallois, “Les harkis en France: déracinés et oubliés,” La Croix, Mar. 23, 1974.
46 CAC, 19980077/20, Ministry of Interior “Le problème des harkis,” 4.
42
43
350
lost in Algeria led him to undertake a fight against the government for the mistreatment of the entire
French Muslim repatriate population, but especially the harkis.47 Close to former OAS members, as
was Djebbour, he began a hunger strike in a church in Evreux on January 1, 1974 “pour prendre
conscience au gouvernement français que les harkis sont des citoyens à part entière.”48 His actions
sparked continuous hunger strikes, primarily in places of worship, across France and even one in
Geneva on the anniversary of the Evian Accords in March 1975. Laradji’s words and actions
attracted government officials’ attention, in part owing to the press coverage he received, which
sometimes essentialized the harki population’s situation, particularly relative to living in the camps.
For example, on October 12, 1974, at the height of a month-long hunger strike at La Madeleine
church in Paris undertaken by eight members of Laradji’s association, Le Monde published an article
titled “Pour obtenir leurs droits de Français à part entière.” While the journalist quoted Laradji
several times, he wrote the following in his own words (ostensibly inflected by those of Laradji): “La
plus grande partie des Français musulmans rapatriés d’Algérie—ils représentent actuellement une
population de près de cinq cent mille personnes—vivent en marge de la société, dans des camps…
rejetés par leur patrie d’adoption, la France…. Pour la totalité, même pour ceux qui ont été reclassés
professionnellement, notamment par les anciens des SAS, c’est, de toute manière le ‘ghetto.’”49 As
cited above, in 1974 approximately 16,000 of the 180,000 harki population members resided in
camps—not “la plus grande partie des Français musulmans rapatriés d’Algérie.”
Given such press articles and even occasional coverage on the evening news of these words
and actions, the harkis’ citizenship and “ghetto” living conditions became an unavoidable political
issue, which the Bouches-du-Rhône SFM director called the “goût du jour.” 50 In particular,
politicians outside of newly-elected President Valéry Giscard d’Estaing’s Rassemblement pour la
SPDP, dossier: Harkis, Guy Le Bolzer, “Qui est Laradji l’homme qui se bat pour les harkis?,” Le Figaro, Aug. 14, 1975.
CAC, 19980077/20, SFM, “Le problème des harkis,” 4.
49 SPDP, dossier: Harkis, Leo Palacio, “Pour obtenir leurs droits de Français à part entière,” Le Monde, Oct. 12, 1974.
50 ACNMF, 32/2, Sessa, “Rapport d’Activités pour 1973 et 1974,” 1, Jan. 1, 1975.
47
48
351
France (RPF) party latched onto “the harki problem.” The President, through allegiances he had
formed with harki associations such as the FNRFCI and the CFMRAA, had received many votes
from harki citizens in the May 1974 elections. On October 13-14, 1974, Le Monde published a press
release by three Centre démocrate deputies with their suggestions for how those in power should act
in the face of the failure of successive administrations to integrate the harki population: “Les
douloureux problèmes que pose l’intégration de ceux qui ont manifesté courageusement leur fidélité
à la France ne sont pas toujours réglés. Ils sont particulièrement complexes et nécessiteraient la
création d’une délégation confiée à un haut fonctionnaire, qui aurait l’autorité nécessaire pour agir
comme médiateur auprès des différentes administrations concernées.” 51 Moreover, deputies of
various political allegiances during an October 16 National Assembly session condemned the
government’s inaction toward this mistreated group. For example,
Député Edouard Frédéric-Dupont (Républicain indépendant): Le gouvernement doit faire
un effort pour éviter qu’ils [les Français musulmans] soient parqués dans des ghettos et pour
améliorer leur condition matérielle…
Député Georges Frêche (Parti socialiste de la France): Une délégation de notre parti est allée
à la Madeleine leur apporter notre soutien. Mais nous avons décidé de ne pas poser de
question car l’heure n’est plus à la désignation des commissions: les harkis veulent la justice,
la dignité et la solidarité nationale et nous regrettons que le conseil des ministres n’en ait pas
traité ce matin.52
Under political pressure, Giscard d’Estaing’s cabinet director went to the Madeleine church in order
to “confirmer l’intérêt personnel que le président porte pour les Français-Musulmans.” However,
SPDP, dossier: Harkis, “L’un des huit harkis de l’église de la Madeleine est dans un état très grave,” Le Monde, Oct. 1314, 1974.
52 SPDP, dossier: Harkis, Pierre Chaumeil, “Justice pour les Harkis!,” Aspects de la France, Oct. 24, 1974. Frêche’s
statement of support for the harkis in 1974 contrasts with the critical words he uttered during a February 11, 2006
ceremony in homage to a former pied-noir leader. Frêche, President of the Regional Council for Languedoc-Roussillon,
condemned the participation of harkis in a protest concerning the controversial February 23, 2005 law sponsored by the
UMP party (a descendant of the Gaullist RPF party): “[Les gaullistes] ont massacré les vôtres en Algérie et encore vous
allez leur lécher les bottes! (...) Vous êtes des sous-hommes.” While Frêche maintained that his words were pointed
toward one particular harki, the leader of the AJIR association for the Hérault department, his polemical reference was
widely interpreted as aimed toward the entire community. The designation “sous-hommes” sparked an uproar in the
harki community and beyond, which led to Frêche’s censure from the Socialist party. These events in 1974 and 2005
demonstrate the politicization of the harki population. “Georges Frêche: ‘Je visais un homme, pas la communauté harki,’”
Associated Press, Feb. 13, 2006 and “Les propos de Georges Frêches sur les harkis déclenchent un tollé,” La Croix, Feb. 14,
2006.
51
352
this rhetoric was not backed up by policy changes. The hunger strikes continued and soon the
“camps de concentration,” as Djebbour referred to them, became the loci for the harki population’s
protests.
On April 22, 1975, when the prefect of the Lot-et-Garonne department arrived at the Bias
Camp for a meeting to discuss the problems at the camp he was greeted by a sign that read
“Bienvenue au camp de concentration” stretched across its entrance.53 On May 7, Laradji and the
vice president of CFMRAA and a former OAS militant, André Christophe, brought arms to the Bias
camp and encouraged the youth to rise up against the government’s segregation. For two weeks they
occupied its administrative and school buildings, blocked access to the camp entrance, and handed
out to passing motorists tracts that explained their situation and demanded that the government
close “les camps de la honte.” The night of May 14-15, harki youth set fire to the onsite school and
the next day between 100 and 150 protesters took siege on the camp offices, breaking windows and
attacking the policemen guarding this space.54
In the interim, on May 13, 1975, Prime Minister Jacques Chirac created a “Permanent
Interministerial Commission to study the problems of the French of Islamic origin repatriated from
North Africa” under the presidency of Minister of Interior Michel Poniatowski. In addition to
government representatives, its members included the presidents of the three most prominent harki
associations: Laradji representing the CFMRAA, the new president of the FNRFCI Abdellatif Khiari,
and Ahmed Kaberseli of the MADRMAN.55 As Poniatowski explained in a circular to departmental
and regional prefects two months later, the Commission’s objective was to “hâter l’intégration
harmonieuse de nos compatriotes originaires d’Afrique du Nord.” He listed as the commission’s
Lanoizelez, “La CARA de Bias,” 84-85. Since October 30, 1973, quarterly meetings to discuss problems residents
faced had taken place at the Bias Camp and the neighboring Sainte-Livrade Camp, which housed Indochinese refugees.
Minutes of these meetings are located in: CAC, 19870256/7.
54 Ibid., 85.
55 CAC, 19910097/48, Minister of State for the Interior to the Regional and Departmental Prefects, Circulaire nº 75 340,
“Objet: Création d’une Commission Interministérielle pour les problèmes,” 4, July 4, 1975.
53
353
priority the progressive elimination of the cités d’accueil and forest hamlets “afin de mettre un terme à
une ségrégration fâcheuse.” 56 The establishment of this commission, one in a long line of
government delegations to “study” the problems in the harki community, did not stop the protests.
In fact, the social movements subsequently spread and escalated.
M’Hamed Laradji arrived at the Saint-Maurice-l’Ardoise camp on May 18, 1975 with
Christophe and Jean-Pierre Nicolas, a second lieutenant during the Algerian War for Independence.
They urged the camp’s youth to follow the actions of its Bias brethren.57 One harki son, who was
twenty-two years old at the time, recalled in a 2011 documentary the words that Laradji pronounced
at the camp’s entrance on May 19: “Jeunes, révoltez-vous! Vous êtes pas venus en France pour
rester dans les ghettos.”58 From there, thirty young men and women (aged twelve to twenty-five)
stormed camp director François Langlet’s office with the three leaders and, according to Langlet, the
camp residents “hurlent, vocifèrent, me menacent, [et] paraissent déchainés.”59 For over a month, as
Langlet chronicled in minute detail, the youth—spurred on by Laradji, Christophe, and Nicolas—
destroyed and pillaged the camp administrative offices, cut off the electricity to the camp staff’s
apartments, protested outside the town hall, and, as depicted in figure 34, closed the camp’s gates to
block anyone from entering and leaving.60 The protests culminated on June 19 when four armed,
masked sons of harkis took hostage Langlet, who had moved his office to the Saint-Laurent-desArbres town hall. If the police intervened, they threatened to detonate dynamite, which one armed
assailant was photographed holding on the balcony of the town hall (see figure 35). The harki sons
informed the mayor that they would release Langlet only if members of the Interministerial
Ibid.
CAC, 19870256/7, François Langlet (Director of the Saint-Maurice-l’Ardoise cité d’accueil), Handwritten letter,
“Chronologie des faits,” 2, June 11, 1975.
58 Hocine, le combat d’une vie, dir. Jean-Claude Honnorat, http://vimeo.com/28820197 (Sept. 2011).
59 CAC, 19870256/7, Langlet, “Chronologie des faits,” 1.
60 Ibid. For updates on the situation at Bias and Saint-Maurice-l’Ardoise by a civil servant in the Population and
Migrations office of the Ministry of Labor see, CAC, 19870256/7, “Note à l’attention de Monsieur le Directeur de la
Population et des Migrations,” May 22, 1975, May 26, 1975, May 27, 1975, and May 29, 1975.
56
57
354
Commission traveled from Paris to meet with them.61 Twenty-eight hours later, they retreated from
this condition and left the town hall in a car driven by Laradji with assurances of immunity from
prosecution.62
Figure 34: Children of harkis blocking the entrance to the Saint-Maurice-l’Ardoise camp.
Source: Agence France-Presse archives, June 2, 1975.
The harki sons also took two secretaries hostage, but released them immediately. CAC, 19870256/7, Prefect of the
Gard department to the Ministers of Interior and Labor and the Director of Population and Migrations (Labor), Urgent
Telegram, “Objet: Prise d’otages par des ressortissants de la Cité de Saint Maurice l Ardoise,” nº 841, June 19, 1975.
62 Jacques-M. Bourget and Christian Durandet, “L’Affaire des fils de harkis,” L’Aurore, June 21-22, 1975.
61
355
Figure 35: Masked harki sons bearing
rifles and dynamite on the balcony of
the Saint-Laurent-des-Arbres town
hall. Inside, two other harki sons held
camp director François Langlet
hostage. Source: Jacques-M. Bourget
and Christian Durandet, “L’Affaire
des fils de harkis,” L’Aurore, June 2122, 1975.
Though July remained comparatively calm at the camps, in early August a new wave of
violence erupted. These events once again thrust harki citizens’ plight into the news with numerous
articles appearing in periodicals across the political spectrum, such as “Harkis: quinze ans de honte”
(Le Point), “Les Français musulmans attendant des mesures concrètes d’intégration” (Le Figaro), and
“La Révolte des harkis” (Le Nouvel Observateur).63 This situation even garnered international attention
with an article in the International Herald Tribune titled “The Harkis Who Live as Pariahs in France”
and another in The Times of London about the Kradaoui Affair.64 On July 31, Algerian officials
forbade a seven year-old harki child, Borzani Kradaoui, from returning to France from Algeria
where he was visiting family with his mother. His father had not taken the trip since the Algerian
SPDP, dossier: Harkis, Jules Roy, “Harkis: quinze ans de honte,” Le Point, Aug. 18, 1975, 17; Laurent Greilsamer, “Les
Français musulmans attendant des mesures concrètes d’intégration,” Le Figaro, Aug. 9, 1975; and Guy Sitbon, “La
révolte des harkis,” Le Nouvel Observateur, Aug. 25, 1975.
64 SPDP, dossier: Harkis, Jack Monet, “The Harkis Who Live as Pariahs in France,” International Herald Tribune, Aug. 12,
1975 and Richard Wigg, “Algerian workers seized by sons of former Muslim soldiers for the return of boy to France,”
The Times, Aug. 8, 1975.
63
356
government forbade former harki soldiers from entering the country. At the same time, Algerian
customs law stipulated that a child was not permitted to leave unless given paternal consent, which
Borzani’s father was not present to do. In retaliation, on August 6 ten harki children from the SaintMaurice-l’Ardoise camp went to a nearby factory where they wounded two Algerian immigrant
laborers and took four others back with them to the camp as their prisoners. Two days later,
following Borzani’s safe return from Oran, the harki children released their hostages.65 As in June,
the offenders struck a deal not to be prosecuted for their actions. However, officials did not offer
Laradji this same opportunity and he was subsequently imprisoned for several months.66 Absent
their leader, additional hostage situations nevertheless raged throughout month; the image in figure
36 depicts riot police on the Bias camp’s perimeter following a hostage situation. Meanwhile, less
violent protests spread to forest hamlets near the Saint-Maurice-l’Ardoise camp, including Apt,
Cucuron, Jouques, and Pertuis.67
SPDP, dossier: Harkis, “Arabes contre Arabes. À qui la faute?,” Libération, Aug. 8, 1975 and Pierre Dumas,
“L’escalade,” L’Aurore, Aug. 18, 1975.
66 Crapanzano, The Harkis, 148.
67 Moumen, Les Français musulmans en Vaucluse, 154-55 and ADBR, PHI 420/285, “Les harkis ‘forestiers’ d’Apt et du
Lubéron en grève,” Le Provençal, Aug. 14, 1975.
65
357
Figure 36: Riot police outside of the Bias camp on August 18, 1975, after harki children took a leader of the Algerian
immigrant association ADAF hostage. Source: Agence France-Presse archives.
Before examining government policy changes during the “été de révolte,” it must be noted
that the choice of Algerian immigrant hostages in the “Kradaoui affair” appeared to serve two
purposes, which underscore the harki population’s difficult relationship with French citizenship.
First, as the numerous local and national news outlets that reported this incident maintained,
imprisoning Algerian nationals would impel the Algerian government to engage in a hostage
exchange. Second, this decision also seemed to have a symbolic meaning, as the title of a front-page
article in the August 8, 1975 edition of Libération evokes: “Arabes contre Arabes. À qui la faute?”
The harki youth, struggling to find their identity, wanted to differentiate themselves from other
Algerian immigrants and underline their unique situation: rejected by most Algerians who believed
their fathers were traitors and many in the French community who were racist toward them, like any
358
other Arab youth.68 As a June 1975 Paris-Match article titled “Les Harkis dans le ghetto” quoted one
Saint Maurice l’Ardoise resident saying: “‘Mon père, dit Belkacem, 23 ans, sans travail, mon père
était un crétin. Il a choisi la France. Regardez où on en est. Tout le monde nous rejette et nous traite
de bougnoules…’”69 These articles were a harbinger of the difficulties that government policies,
enacted thirteen years after the harki population was “parked” in camps, would have addressing its
citizenship.
The same day that the harki sons took the Algerian immigrants hostage, August 6, Minister
of Interior Poniatowski presented on the TF1 evening news a list of “solutions” that the
government proposed to address the “problems et difficulties” that the harki population faced. The
series of measures decided on during a Council of Ministers meeting that day included indemnity
payments to harkis imprisoned in Algeria after the war, consideration of time in captivity for harkis’
salaries and pensions, a “a very large effort” in job training for harki children to help them
“completely integrate into the nation,” and the elimination of the Saint-Maurice-l’Ardoise and Bias
camps by the end of 1976. The Minister of Interior also revealed that the Permanent Interministerial
Commission would study further several other measures relating to housing and employment.70
Following its September 24 meeting, Poniatowski sent a circular to the regional and departmental
prefects to which he attached the eleven measures that his commission approved.71 In addition to
those decided by the Council of Ministers, the commission proposed financial assistance for harki
families who left the camps and forest hamlets; a permanent professional status for forestry workers
to address inconsistencies in salary and working conditions; financial incentives to towns that hired
On the complex struggle of harki children, particularly those who grew up in camps, to find their identity, see:
Crapanzano, The Harkis and Crapanzano, “The Dead but Living Father, the Living but Dead Father,” in The Dead
Father: A Psychoanalytic Inquiry, ed. L. J. Kalinich and S. W. Taylor (London: Routledge, 2009), 163-173.
69 CAC, 19870256/7, Philippe Ganier-Raymond, “Les Harkis dans le ghetto,” 37, Paris Match, June 14, 1975.
70 “20 heures,” TF1 (Aug. 6, 1975), accessed by <www.ina.fr>. The decisions are also summarized in: CAC,
19870444/15, D. Bourkoba (Mission Interministérielle pour les Français Musulmans), “L’action du gouvernment en
faveur des Français Musulmans,” 5-6, Oct. 18, 1979.
71 CAC, 19910097/48, Minister of State for the Interior Poniatowski to the Regional and Departmental Prefects,
Circulaire nº 75-486, “Objet: Mesures prises en faveur des français raptriés d’origine musulmane,” 1, Sept. 26, 1975.
68
359
harkis and provided housing for their family; the creation of bureaux d’information, d’aide
administrative et de conseils pour les Français musulmans (BIAC) in areas with a sizeable population
of isolés; and a large-scale effort to information harki citizens of these measures.72 Poniatowski
underscored that since most of these items related to rehousing, the prefects had an important
responsibility to follow through with local HLM offices and local officials to ensure the
implementation of these policies.73 During its October 1, 1975 meeting, the Council of Ministers
approved these measures and followed up with a promise to eliminate the remaining twenty-nine
forest hamlets by the end of 1979.74 At this point, much of the onus to implement these policies was
devolved to local officials, without much oversight from the national government. Poniatowski sent
another circular to the regional and departmental prefects seven months later decrying the “rythme
trop lent” of rehousing the harki population in camps and forest hamlets (already prescribed by
earlier instructions listed above). HLM companies had only constructed fifty of the five-hundred
apartments they were supposed to complete by the end of 1975. Moreover, based on reports he had
received of harkis being turned away from rental properties, he contended, “Il est également
anormal que nos compatriotes musulmans soient l’objet de discriminations, de la part d’organismes
constructeurs qui, sous des prétextes divers, leur refusent parfois accès à la propriété.”75
Despite the increased attention that national level officials accorded to harki citizens in the
mid-1970s and modifications to government integration policies, the harki population would still
remain plagued by their unique status that challenged their French citizenship and decisions
government officials made when they arrived in France, notably their placement in camps. This
CAC, 19910097/48, “Réunion de la Commission Interministérielle Permanente pour les problèmes des Français
Musulmans Rapatriés d’Afrique du Nord – mercredi 24 septembre 1975.”
73 CAC, 19910097/48, Poniatowski, “Objet: Mesures prises en faveur…,” 2, Sept. 26, 1975.
74 CAC 19870444/15, R. Saint-Jean (Office of Social Welfare for French of North African Origin), “Note pour
Monsieur Bourokba, Chef de la Mission Interministérielle pour les Français Musulmans. Annexe 1: Résorption des
hameaux de forestage,” 1, Dec. 6, 1979.
75 CAC, 19910097/48, Minister of State for the Interior Poniatowski to the Regional and Departmental Prefects,
Circulaire nº 76-239, “Objet: Mesures prises en faveur des français raptriés d’origine musulmane,” 1, Apr. 29, 1976.
72
360
unique status—difficulties accessing benefits they legally had a right to as repatriates and treatment
as Algerian immigrants despite their French nationality—affected the entire harki population,
whether or not they resided in camps or forest hamlets. Nevertheless, with the publicly visible
protests, camps and forest hamlets became the symbol of the population’s failed integration and the
government’s mistreatment of the “harki problem.” While the Saint-Maurice-l’Ardoise camp closed
in October 1976, the Bias camp would remain open until 2000, as the mayor of the town refused to
allow the construction of buildings in which the families would be rehoused on municipal land.76
Moreover, by the end of 1979, only ten of the twenty-nine forest hamlets had been eliminated.77
This chapter returns to the story of the Fuveau forest hamlet, specifically its “elimination,” which is
symbolic of the government’s failed integration policies for its harki citizens.
The Failure of Revised Integration Policies in Fuveau
Trente et un nouveaux logements destinés à des familles françaises musulmanes ont été
inaugurés hier matin à Fuveau…
Ces nouveaux logements sont destinés à remplacer l’ancien hameau de forestage se
trouvant à proximité, où les intéressés étaient logés dans des conditions précaires…
Cette opération découle de mesures décidées par le Conseil des ministres du 6 août 1975,
tendant à améliorer la situation de cette catégorie des Français. [C’est] l’intégration en milieu
ouvert…voulue par le gouvernement….78
On August 10, 1976 the Fuveau “hameau de forestage” was rebaptized “la Cité Brogilum,”
the Latin word for “breuil,” signifying an enclosed wooded space serving as a hunting preserve or
dense vegetation in a field.79 However, the reconstruction of the forest hamlet buildings (and the
new name, which became interchangeable with the old one) did nothing to address the Fuveau harki
citizens’ integration, the goal of the 1975 policy decisions by national officials. The forest hamlet’s
location and the encadrement structure—the two greatest challenges to integration—remained the
Lanoizelez, “La CARA de Bias,” 100.
CAC 19870444/15, R. Saint-Jean “Résorption des hameaux de forestage,” 1, Dec. 6, 1979.
78 Michel Proust, “Inauguration de trente et un logements pour des Français musulmans,” Le Provençal, Aug. 11, 1976.
79 For the definition of “breuil,” the French word for “brogilum,” see: <http://www.cnrtl.fr/etymologie/breuil>.
76
77
361
same. The excerpt cited above of an article published on August 11 in Le Provençal points toward
several ways that the government’s 1975 policy changes were unsuccessful in Fuveau: living
conditions, isolation, and French citizenship. Some of these failures also reflect the tension between
national policies and local initiatives as well as the ineffectiveness of the national government to
ensure the harki population’s integration.
Figure 37: The fourteen original barracks, two barracks constructed in 1971, and an ONF hangar for work supplies are
depicted by the white rectangles on the left side of the image. The six rows of new houses, totaling thirty-one residences,
are labeled A-F and appear to the right of the older buildings. The social counselor’s new house is the gray box amidst
the barracks. Source: ASANHLMM, 54/1.
First, as the article indicates, the thirty-one houses (plus one for social counselor Darmon),
lined up in six rows of four or five, were to replace the lightweight prefabricated houses whose living
conditions had become unsafe (see figure 37). While the new residences—named “type ‘Alger’” by
the architectural firm—were bigger and permanent structures, they nevertheless did not take into
account the large size of the harki families and contained several construction faults.80 With the
exception of two buildings constructed in 1971 to address demographic concerns (the average family
ASANHLMM, 54/2, Jules Viaux & Ses Fils to the President of the SANHLMM, “Objet: Village Harkis,” 1, Feb. 5,
1975.
80
362
in 1972 comprised almost seven individuals),81 the other flimsy original residences were hastily built
in 1964 to last for only a short time. Each apartment originally contained four single beds in two
bedrooms. Therefore, as families moved away from the forest hamlet, other families took over their
37.5 square meter apartments. One family appropriated an entire building of barracks, where it
remained until 1997 before it was evicted. The patriarch, Ali H.—who served in the French Army
during the Second World War, the Indochinese War, and as a harki during the Algerian War for
Independence, after which he was an FLN prisoner for a month—is pictured wearing his military
medals in front of the dilapidated structure in figure 38.82 He refused to move his family to one of
the new residences for two reasons, one relating the problems that led to government discussions
and harki population protests in the mid-1970s and the other to the buildings that local officials had
commissioned. On one hand, he would be unable to pay the rent because, as a non-salaried worker
for the ONF, he received neither his salary nor worker’s compensation benefits for twenty months
following an on the job accident,83 and he had been unable to access indemnity payments he was
owed as a former harki soldier.84 On the other, the size of the new buildings could not accommodate
his family containing fifteen children aged one to twenty-two in the summer of 1976.85 While the
new houses were larger, they were insufficient for the fourteen families with upwards of eight
children in 1975. As table 7 demonstrates, two-thirds of the houses had three or less bedrooms and
only one-tenth contained five or six bedrooms.
ADBR, 135 W 41, Louis Sessa (Director of BIAC in Marseille), “Rapport sur la situation des Français Musulmans
installés dans le département des Bouches-du-Rhône,” Map, May, 18, 1976 and Fuveau, Register of Town Council
Meeting Proceedings, “Nº 72-32 – Ecoles pour les enfants de Harkis,” Apr. 24, 1972.
82 AONACBR, individual file of Ali H., “Demande de Carte du Combattant,” Annexe 1, Aug. 14, 1973.
83 AONFBR, individual file of Ali H., Forestry Expert Raymond Bois to Mr. Figura (Mutualité Sociale Agricole), “Objet:
Situation de M. H. Ali,” 1, Nº 3408 RB/AR, Nov. 2, 1976.
84 ASRBR, social welfare file of Ali H.
85 Ibid.
81
363
Number of bedrooms
1
2
3
4
5
6
Number of houses
3
8
9
8
2
1
Size of house
53m2
64m2
85m2
95m2
115m2
127m2
Table 7: Size and number of the thirty-one new houses in Fuveau.
Source: ACF, SANHLMM, “Fuveau Cité Brogilum. Vente Logements HLM,” 3, 1988.
Figure 38: Ali H. (right), whose family included nineteen children, pictured in front of one of the original forest hamlet
buildings in 1996 with a former GMS soldier who resides in Fuveau. The H. family arrived in Fuveau on September 1,
1966 from the Sault forest hamlet in the Vaucluse and remained until August 1997, when authorities forcibly evicted it.
Photograph used with the permission of Saïd Merabte, president of the regional Association Justice Information
Réparation pour les harkis (AJIR).
The other defects with the new residences, which had been built by apprentices and not
professional workers, resulted from the “lamentable design of the site” and “poor quality” of
construction, as a 2002 report conducted by an independent firm indicated. The design faults
included a lack of privacy owing to the houses being clustered together (see figure 39), an absence of
outdoor gathering spaces, and no playgrounds. Rubble from the barracks demolished in 1976 still
364
littered the ground and posed safety concerns for the children. The report also noted the houses had
been constructed almost directly on the ground and contained poorly-insulated walls and roofs,
which caused cracks in the walls and continual dampness and mildew inside the houses.86 These
problems manifested themselves almost immediately. Following a rainstorm less than three weeks
after the Cité Brogilum opened, given the lack of gutters and poor construction, some of the houses
flooded, which required firemen to vacuum out the water.87 Moreover, the unpaved driveways and
grounds became mud puddles, blocking all cars from entering the forest hamlet and the unstable
ground created a dangerous situation for the residents, particularly the children. As the floors had
been constructed almost directly on the land mud had even seeped up into some residences.88 These
initial difficult living conditions were compounded by the fact that the residences did not contain
heat for the first year. The local construction and management company now in charge of the forest
hamlet, the Société Anonyme Nouvelle d’HLM de Marseille (SANHLMM), decided to place the
same oil-fired stoves that had heated the 37.5 square meter barracks into the dining room of their
new homes. However, following the family’s complaints as the temperatures dropped that winter,
the company soon realized that this method of heating was grossly insufficient for the new
residences measuring double or triple the size of the old ones. It therefore installed radiators and
boilers in summer 1977.89 While the Fuvelian harki citizens’ living conditions did improve, the poor
construction and design of the buildings did not markedly improve their situation.
ACF, Arts et Management de la Ville, “Étude sur le devenir…,” 16, Oct. 2002.
ASANHLMM, 54/1, “Les harkis de Fuveau en colère: notre village est inondé,” Le Provençal, date not indicated.
88 ASANHLMM, 54/1, “Vivre à la française, oui!... mais au sec,” Le Provençal, date not indicated. Both articles are from
the same day.
89 ASANHLMM, 54/2, SANHLMM, “Fuveau – Cité Brogilum. Construction du hameau de forestage à Fuveau,” 1-2,
Sept. 4, 1980.
86
87
365
Figure 39: Row of four reconstructed residences at the Fuveau forest hamlet. The houses’ proximity and poor sound
insulation did not afford the families much privacy. Photograph taken by the author in December 2007.
The second indication of the ineffectiveness of the government’s revised integration policies
for harki citizens is that the reconstruction of the Fuveau forest hamlet was not a result of the
October 1975 Council of Ministers decision to eliminate the twenty-nine remaining forest hamlets
by the end of 1979. National policies did not influence the reconstruction of the Fuveau forest
hamlet. Local officials had decided in late 1973 to keep the forest hamlet on the same isolated site
and even though construction did not begin until one month after the Council of Minister’s August
1975 meeting, they did not modify their plans following the national government’s policy changes.90
This decision countered the national government’s primary objective of integrating the “populations
ASANHLMM, 54/2, Director of the SANHLMM to the Bouches-du-Rhône Departmental Director for Facilities,
“Objet: Ville de Fuveau. Relogement des anciennes familles de harkis. Difficultés rencontrées dans le démarrage de
travaux,” 1-2, Aug. 25, 1975.
90
366
regroupées” into the French community and did not appear to be “l’intégration en milieu
ouvert…voulue par le gouvernement,” as the August 1976 Le Provençal article cited above claimed.
On November 26, 1973 Mayor Philip, a representative of the Houillères de Provence,
architects, the interdepartmental forest hamlet inspector, and representatives from the departmental
prefect’s office met at the forest hamlet with the goal of “[l’]établissement du programme de
construction.”91 The following week during Fuveau’s town council meeting, the mayor read aloud a
letter from the departmental prefect concerning rehousing the harkis in “logements ‘en dur.’” Town
council members subsequently deliberated and decided that the town would buy the land from the
Houillères de Provence and “céder une parcelle pour permettre la construction d’une trentaine de
logements.”92 Instead, the SANHLMM ended up purchasing the land in September 1975 just before
it undertook the construction project. 93 The departmental director for facilities explained in a
January 1974 letter to the president of the SANHLMM the reason for rebuilding the forest hamlet
on the same site as: “…l’entretien de la forêt justifie leur présence et leur logement sur place.”94
Departmental and Fuvealian officials, moreover, made this decision without consulting the
harki citizens about whether they wished to remain three kilometers from town and simultaneously
ignored the suggestions of departmental SFM director, Louis Sessa. In an April 1973 report,
referencing the recent recommendations of the Barbeau group to facilitate the harki population’s
integration, Sessa maintained that it was possible for the harkis to remain forestry workers and live
closer to town: “Tout en conservant leur emploi d’agent forestier aux chefs de familles, il est
envisagé de poursuivre l’effort d’intégration des familles déjà amorcée en leur faveur et qui deviendra
ASANHLMM, 54/1, Director General to the Bouches-du-Rhône prefect, “Objet: Commune de Fuveau. Logements
‘en dur’ des migrants employés aux travaux de forestage. Réunion du 26 Novembre 1973 à 15 Heures sur le Terrain,” 2,
Nov. 15, 1973.
92 Fuveau, Register of Town Council Meeting Proceedings, “Nº 73-92 – Logement des Harkis,” Dec. 5, 1973.
93 ASANHLMM, 54/1, “Vente par les Houillères de Bassin du Centre et du Midi à la Société Anonyme Nouvelle
d’HLM de Marseille,” 3, Sept. 19, 1975.
94 ASANHLMM, 54/2, Bouches-du-Rhône departmental director of facilities to the President of the SANHLMM,
“Objet: Commune de Fuveau. Logement des Migrants employés aux travaux de forestage,” 1, Jan. 29, 1974.
91
367
plus concrète lorsqu’elles seront relogées en HLM et brassées avec des familles françaises bien
assimilées.”95 A few pages later, he noted that many of the harkis in the department’s forest hamlets
did want to be integrated: “Interrogés sur leur avenir, ces travailleurs acceptent d’envisager leur
départ des hameaux pour des cités où ils seront brassés avec des populations de souche.”96 The
mayor of la Roque d’Anthéron chose integration for the harki families in his village by moving them,
between January 1975 and February 1976, from the forest hamlet to an HLM located in town.97 All
of these men were able to conserve their employment as forestry workers. In contrast, in Fuveau
(and in Jouques) local officials chose to keep their harki citizens segregated.
In the midst of nationwide demonstrations by harki children in summer 1991, those in
Fuveau protested their isolation. The town hall issued a statement with an explanation of this
decision by the mayor in 1975, Max Guiguet: “[A] l’époque le secrétariat aux rapatriés était favorable
à l’éclatement des hameaux de harkis pour une meilleure insertion dans les structures d’habitat local.
Cette solution n’a pas abouti sous la pression de l’ONF pour des commodités de transport du
personnel forestier.”98 Among the documentation in archives from the ONF, the SANHLMM,
Fuveau, and the Bouches-du-Rhône department, there is no evidence of this explanation given at
the time. Moreover, the harkis in la Roque d’Anthéron (where municipal officials had consistently
taken more actions to help them and include them in its community) employed by the same ONF
office as the Fuveau harkis moved into town and kept their employment.
Therefore, the national government discourse of turning “milieux fermés” into “milieux
ouverts” was accompanied by Fuvelian officials’ paternalistic measures that continued to segregate
the population of “French Muslims” from other French citizens and keep them under the encadrement
of a social counselor. At the same time, national government officials interpreted the reconstruction
ACNMF, 32/2, Sessa, “Rapport Annuel 1972…,” 2.
Ibid., 4.
97 ADBR, 135 W 51, “Les Français-Musulmans. Rapport d’activités du B.I.A.C.,” 1, May 18, 1976.
98 ADBR, 135 W 51, “Communiqué de la Municipalité de Fuveau,” July 1, 1991.
95
96
368
project in Fuveau as the elimination of the forest hamlet. A document in the CNMF archives that
lists the twenty-seven forest hamlets open on January 10, 1977 does not include Fuveau, even
though the harki families still resided on the same isolated terrain.99 A July 1976 document noted this
situation in the margin with the handwritten words: “pavillons reconstruits sur place – milieu
ouvert.”100 Indeed, another chart from August 1978 includes Fuveau’s harkis in the category of
“nombre de travailleurs ONF vivant hors hameaux.” 101 These documents reveal that rebuilding the
houses on the same isolated terrain and keeping the same encadrement structure in place constituted
for the government closing the Fuveau forest hamlet. Yet, the reconstruction of new buildings
would do nothing to address their integration as the second epigraph to the previous chapter, a
petition written in July 1993 by a Fuvelian resident affiliated with the National Front party, alleged:
“Si on peut trouver de l’argent pour ‘intégrer’ d’inintégrables immigrés, il est temps d’en trouver
pour aider enfin les Français Harkis. Pour cela, il faut démanteler le camp de harkis de
FUVEAU.”102
Conclusion: Refugees, Repatriates, Immigrants?
In the August 11, 1976 newspaper article cited above, titled “Inauguration de trente et un
logements pour des Français musulmans,” the journalist referred to the harki population as “familles
françaises musulmanes” and “cette catégorie des Français.”103 No amount of reminders to public
officials or private citizens from national politicians would suffice to distance the harki population
from either their inferior nationality status in colonial Algeria (Français musulmans), possible
ACNMF, 16, “Hameaux Forestiers,” Jan. 10, 1977. This chart includes la Roque d’Anthéron, which was closed on
February 1, 1976.
100 ACNMF, 16, “Hameaux Forestiers au 1er juillet 1976.”
101 ACNMF, 16, “État des Hameaux Forestiers août 1978.” This chart also puts la Roque d’Anthéron’s harkis into the
category of “nombre de travailleurs ONF vivant hors hameaux,” a true statement in this case.
102 APMH, Petition from Loïc de Marion (National Front) to the Bouches-du-Rhône Prefect, July 1993. On the original
document the second sentence is in boldface and larger type size than the first sentence.
103 Proust, “Inauguration de trente et un logements…”
99
369
confusion with Algerian immigrants, or their distinctiveness from other French citizens. Moreover
“Français musulmans rapatriés” would become in the early 1980s the government’s preferred term
for harki population members. They were legally French citizens, but rarely considered or treated as
such. The exiled harkis and their families—whether or not they ever resided in camps—had a
unique status amalgamating refugees, repatriates, and immigrants. They were harki citizens.
Harkis and their families arrived in France from Algeria owing to their well-founded fear of
persecution by FLN members for the “political opinion” the men had taken during the war. The
harkis’ situation falls under the definition of “refugee” by the 1951 Geneva Convention—except this
statute requires refugees to migrate to a country where they do not hold nationality. The harkis’
exceptional nationality provisions—they could become French citizens by a pro forma procedure—
simultaneously negated the possibility for them to claim refugee status and differentiated them from
French repatriates. Nevertheless, the French government treated them as refugees by placing over
half in camps upon their arrival, which it did not do with the French repatriates, many of who fled
to France in similar refugee circumstances. This treatment upon their arrival was reinforced by
government discourse as official documents at this time continually labeled the harki population as
refugees. This categorization of the harki population as refugees soon after the war while a
significant number resided in the Rivesaltes and Saint-Maurice-l’Ardoise camps perhaps is not very
surprising. However, a document written by a Ministry of Armies official in late 1966—by which
time over 53,000 harkis had acquired French nationality (see table 3 in chapter two)—offered an
assessment of “la situation des réfugiés musulmans.” The author broke the harki citizens into three
categories. One-third, many of who had been officers, had assimilated well into life in metropolitan
France. They were among the first to arrive in France and few anticipated returning to Algeria. He
labeled the second category (approximately 40 to 50 percent) as “les instables et les hésitants.” Most
of these harkis and their families left Algeria owing to their inability to sustain themselves
370
economically after the war and not the FLN violence. Therefore, he argued, they came to France for
the financial benefits associated with repatriate status and would return to Algeria once the situation
had stabilized. He also noted their extreme similarity to “simples migrants du travail” and believed
that some could make some progress toward assimilation. The final category (15 to 25 percent)
included harki citizens whom he deemed “ne justifient guère les efforts consentis en leur faveur.”
Argumentative, nomadic, completely inadaptable, and “allergiques à tout travail,” he predicted that
they would eventually be forced to return to Algeria by a French government act.104
At the same time, harki citizens consistently faced difficulties fitting into their de jure status
of repatriates because state legislation for “rapatriés” focused on those of European descent, as
rehousing policies following the 1961 Boulin law and the 1970 indemnity law confirm. For the SFIM
liaison with department prefectures at the Rivesaltes camp, the former was “une sorte ‘d’apartheid’
dûe à une seule catégorie de citoyens.”105 For harki association leader Djebbour, the entire situation
created a class of “sous-rapatriés.” 106 The harkis were rarely considered “repatriates” and
encountered delays and roadblocks to accessing their rights as repatriates.
Finally, prevalent public perceptions of visible differences of their Arab or Berber bodies
regularly led to the harki population’s confusion with Algerian immigrants. These differences, a
construction, set them apart from “French” citizens. Because the harkis fell under the definition of
immigrant, a person who moves to a foreign country, and were often treated as such they
concomitantly faced the prejudices inherent to this classification. Algerian immigrants in France are
often perceived as racialized others.
SHAT, 10T 549/3a, EMAT 2e bureau (Immigration algérienne en France), “Situation de la migration algérienne en
métropole à la fin de l’année 1966,” 4.
105 CAC, 19970146/2, Robert Bourgat, “Objet: Situation des ‘Harki’ recasés…,” 2.
106 SPDP, dossier: Harkis, Robert Begou, “Les rapatriés musulmans: ‘Une législation discriminatoire nous est appliquée,”
Le Figaro, Oct. 6-7, 1973.
104
371
The public actions and words of harki associations in the 1970s simultaneously illuminated
the unique position of the entire harki population—harki citizens—and essentialized their situation
by focusing on the camps, which were a symbol of their failed integration and refugees. The BourgLastic and Larzac transit camps that sheltered harki families in tents for several months in 1962, the
Rivesaltes and Saint-Maurice-l’Ardoise housing camps open for twenty-seven and fourteen months
respectively, the Saint-Maurice-l’Ardoise and Bias cités d’accueil containing “unhouseable” members of
the harki population for over a decade, the seventy-two forest hamlets like Fuveau, and the seventyone cités urbaines became harkis citizens’ lieux de mémoire. These spaces represent the government’s
failure to enact an effective integration policy for their inhabitants—in 1962, in 1975, and afterward.
The conclusion of this dissertation will show the significance of these local camp situations for the
national legislative and memorial initiatives to commemorate and properly indemnify harki citizens,
which did not begin until the middle of the 1990s.
372
CONCLUSION
Harki Citizens. Remembering the “oubliés de l’Histoire”
Dans cette loi [de 1994], c’est la première fois où
la République reconnaît… les sacrifices des forces
supplétives et assimilés, des prisonniers. On essaie à la
fois de reconnaître les sacrifices et d’aider
matériellement… Par contre, parfois
l’administration—certains maires, certains offices
HLM—n’étaient [pas] très actifs… et il n’y a pas
beaucoup [de harkis] qui en profit[ait].1
- Former Minister delegate in charge of
Repatriates Roger Romani, March 2008
La France, en quittant le sol algérien, n’a pas su
empêcher [les massacres de harkis], c’est vrai. Elle n’a
pas pu sauver ses enfants.2
- President Jacques Chirac, Journée nationale
d’hommage aux harkis, September 25, 2001
Je regrette que [le débat sur l’article quatre de la
loi de 2005] ait masqué 80 percent de la teneur de
cette loi. Il y avait des choses extraordinaires et très
importantes. C’est la meilleure loi—et ce n’est pas
parce que c’est moi qui l’ai portée, je vous le dis
franchement—c’est la meilleure loi qu’on a pu faire
pour les harkis et leurs enfants depuis toujours.3
- Minister of State for Veterans Hamlaoui
Mekachera, February 2007
Beginning in the mid-1990s, over three decades after the end of the Algerian War for
Independence, the French government enacted three measures to commemorate and compensate its
harki citizens: two laws (1994 and 2005) and a national day in their honor (2001). In discussions
surrounding these initiatives, derivatives of “oubli” and the term “oubliés de l’Histoire” were
ubiquitous in many government officials’ discourse to describe the harki population’s relationships
Roger Romani, Interview with the author, Mar. 30, 2008, Luxembourg Palace, Paris.
“Chirac: ‘les massacres (de harkis) commis en 1962 doivent être reconnus,’” Agence France-Presse, Sept. 25, 2001.
3 Hamlaoui Mekachera, Interview with the author, Feb. 15, 2007, Ministry of Defense, Paris.
1
2
373
to French government and society since the ceasefire. Some officials recognized that France’s harki
citizens—whether through their “sacrifices” for France in Algeria or their isolation in France—had
not simply been passively forgotten. The French government willfully and actively forgot them.
Despite all three measures’ laudable attempts to remember and pay homage to France’s harki
citizens, each had shortcomings. Government officials failed to ensure that harki citizens would be
able to receive the concrete benefits made available by the 1994 law, to which the words of its
author, Minister delegate in charge of Repatriates Roger Romani, in the epigraph attest.4 The portion
of President Jacques Chirac’s speech at the inaugural National Day in Homage to the Harkis in 2001
cited in the epigraph provided a revisionist version of the harkis’ history. The French government in
1962 was indeed aware of the massacres that took place in Algeria, which documents cited in
chapter two confirm, but opted not to intervene for President de Gaulle’s fear of reigniting the war
and his belief that after the ceasefire “[l]a France ne doit plus avoir aucune responsabilité dans le
maintien de l’ordre.”5 Finally, measures in the 2005 law, which had been inserted under pressure
from the pied-noir community, minimized the French government’s commemoration of its harki
citizens, which its sponsor, Minister of State Mekachera regretted (see above). The legislation that he
had intended to principally recognize the harki population became best known for its controversial
fourth article advocating that scholastic programs “reconnaissent en particulier le rôle positif de la
présence française outre-mer.”6
The government enacted these three initiatives in favor of France’s harki citizens in part as a
reaction to protests by harki children over previous government’s policies toward them, which will
JORF, “Loi nº 94-488 du 11 juin 1994 relative aux rapatriés anciens membres des formations supplétives et assimilés ou
victimes de la captivité en Algérie,” June 14, 1994, 8567.
5 Quoted in Peyrefitte, C’était de Gaulle, 151.
6 JORF, “Loi n°2005-158 du 23 février 2005…” On the public debate that ensued after the law was passed concerning
how the French government should memorialize past discriminatory actions or persecutions, see Romain Bertrand,
Mémoires d’empire, la controverse autour du “fait” colonial (Broissieux: Éditions du Croquant, 2006); Isabelle Flahault (ed.),
“L’État et les mémoires,” special issue of Regards sur l’actualité no. 325 (Nov. 2006); Jean-Philippe Mathy, Melancholy
Politics: Loss, Mourning, and Memory in Late Modern France (University Park: The Pennsylvania State University Press, 2011),
154-66; and Patrick Weil, “Politique de la mémoire: l’interdit de la commémoration,” Esprit no. 332 (Feb. 2007): 124-42.
4
374
be treated in the next section. They also unfolded within a more general context relative to the past
and present relationships between France and Algeria as well as the French government’s
commemoration of its history. Beginning in the 1980s French society increasingly focused on the
notion of commemoration. Pierre Nora’s Lieux de mémoire, published in four volumes beginning in
1984, turned the public’s gaze toward these lieux—places, events, peoples, leaders, symbols, books,
ideas, and traditions. Nora contended that lieux de mémoire “exist because there are no longer any
milieux de mémoire, settings in which memory is a real part of everyday experience.”7 Events such as
the exorbitant Goude parade for the bicentennial of the French Revolution attested to the
government’s fervor for commemorating certain, mostly glorifying, episodes of France’s past. At the
same time, in the early 1990s, plural memories of the Algerian War for Independence emerged into
the public space, thanks in part to documentaries airing on public television that exposed the
controversies and intricacies of the war. Peter Batty’s La guerre d’Algérie was broadcast in five
episodes in August and September 1990, while Benjamin Stora’s four-part documentary, Les années
algériennes, appeared in September and October 1991. Stora explained his project as a visual history,
which presented “la confrontation et la circulation des différentes mémoires.”8 According to a poll
carried out in 1992, 83 percent of French youths aged 17 to 30 learned about the Algerian War for
Independence from watching television, substantiating the rationale for Stora’s documentary.9 In
addition to the increasing visibility in France of memories of this war, the Second Algerian War
began raging on the other side of the Mediterranean in spring 1991. In both nations, it evoked
memories of the 1954-62 war—notably those of violence—and attested to the failures of Algeria’s
Pierre Nora, “General Introduction: Between Memory and History,” in Realms of Memory, vol. 1, trans. Arthur
Goldhammer (New York: Columbia University Press, 1996), 1.
8 Dimitri Nicolaïdis, “Entretien de Benjamin Stora,” in Oublier nos crimes: L’amnésie nationale, une spécificité française? (Paris:
Autrement, 1994), 209.
9 Branche, La guerre d’Algérie, 93.
7
375
postcolonial governments after the French government briskly cut ties with its former
departments.10
In the mid-1990s the French government began to confront more controversial—“present
time”—events in its history first through a momentous speech and then through legislation.11 As
explained in the introduction, on July 16, 1995, President Chirac symbolically chose to dedicate his
first public allocution after his election to acknowledging the role that the French State had played in
the persecution of Jews and victims of the Third Reich’s persecution of minorities. By admitting the
French government’s complicity, most notoriously in the Vél d’Hiv roundup, Chirac simultaneously
commemorated the victims and their families and acknowledged the responsibility of France’s
government at the time. 12 Four years later, the French government addressed the other recent
“conflict” whose plural memories circulated in society. On October 18, 1999, the Parliament passed
a law that changed the name of what official documents had until then called “les opérations
effectués en Afrique du Nord,” “les événements d’Algérie,” and “les opérations de maintien de
l’ordre” to “la guerre d’Algérie.”13 Such euphemisms had stifled the plural memories of the war and
minimized the enormous, continuing effects of France’s colonization of Algeria on political, social,
and cultural institutions in both nations. In May 2001 the Parliament passed the Taubira law, which
recognized that the slave trade and slavery—in which France took part from the fifteenth century
through the abolition of slavery in 1848—constituted a crime against humanity.14 Therefore, the
On the 1990s Algerian war, see James Le Sueur, Between Terror and Democracy: Algeria since 1989 (New York: Zed Books,
2010).
11 L’histoire du temps présent has several definitions. Its chronological definition situates this historical period as beginning
with the Second World War. Its thematic definition (preferred by Henry Rousso, the former director of the Institut de
l’Histoire du Temp Présent) focuses on events whose actors are still living (“un passé qui n’est pas mort”). Gérard
Noiriel, “Pour une autre histoire du temps présent,” in Les origines républicaines de Vichy (Paris: Hachette, 1999), 10.
12 Many scholars argue that the policies of the 1940-44 French State presented a rupture with those of previous and
future republican governments. See, for example, Weil, Qu’est-ce qu’un Français?, 97-134. For an opposing viewpoint, see
Noiriel, Les origines républicaines de Vichy.
13 JORF, “Loi nº 99-882 du 18 octobre 1999…”
14 JORF, “Loi nº 2001-434 du 21 mai 2001 tendant à la reconnaissance de la traite et de l’esclavage en tant que crime
contre l’humanité,” May 23, 2001, 8175. Eleven years prior, Parliament had passed the Gayssot law making crimes
against humanity, as defined by the Nuremburg Statutes, punishable by law. JORF, “Loi nº 90-615 du 13 juillet 1990,
10
376
French government had begun to adopt legislation that sought to address its role in past
discriminatory acts and persecutions and demonstrated an evolving relationship with its colonial
past.15
This final chapter first examines the specific context leading to the elaboration of the three
commemorative initiatives for France’s harki citizens. In summer 1991, harki children carried out
widespread protests over the failure of three decades of policies intended to help their families
integrate into French society. Next, it investigates in more detail the 1994 law, the 2001 day of
recognition, and the 2005 law, continuing to place these measures within the French government’s
awakened commemorative consciousness and events specific to Franco-Algerian postcolonial
relations. Finally, this chapter analyzes the government’s commemorative policies toward the harki
citizens alongside previous policies concerning their nationality, repatriation, housing, and
integration. I demonstrate that the harki population’s position between France and Algeria as well as
their history spanning the colonial and postcolonial eras fostered a unique citizenship that has
affected not only harki citizens themselves but also contemporary French society as a whole.
From 1975 to 1991: Protests, Status Quo, and More Protests
Sixteen years after harki children protested the living conditions of the families who
remained in camps and the failed integration of many of France’s harki citizens, in summer 1991
demonstrations by harki children and grandchildren erupted. These actions once again targeted the
failure of government policies intended to help harki citizens’ integrate into French society and
tendant à réprimer tout acte raciste, antisémite ou xenophobe,” July 14, 1990, 8333. On these memorial laws and the
January 2001 law recognizing the Armenian genocide, see Flahault (ed.), “L’État et les mémoires” and Weil, “Politique
de la mémoire.”
15 The laws cited in the previous footnote and the 2005 law also sparked questions about the role memorial initiatives by
the government in writing history. During the controversy over article four of the 2005 law, President Chirac was quoted
as saying: “Dans la République, il n’y a pas d’histoire officielle… Ce n’est pas à la loi d’écrire l’histoire. L’écriture de
l’Histoire, c’est l’affaire des historiens.” Béatrice Gurrey, “Mémoire coloniale: Jacques Chirac temporise,” Le Monde, Dec.
10, 2005.
377
decried the lack of recognition by the French government of their families’ sacrifices. The protests
occurred within the general contexts of government actions to commemorate certain events in
French history and the new wave of violence in Algeria. In comparison with the 1975
demonstrations, which were localized, those in 1991 took place in more venues across France and
attracted increased media coverage.16 The remaining camps—Fuveau, Jouques, Mas-Thibert, and
Bias—and other towns with large concentrations of harki citizens showcased to the public that harki
families still lived in camps and ghetto conditions. Although the number of families residing in these
spaces was very small, media outlets’ images of camps became a signifier of the harki population to
the public and evoked in the population itself memories of the camps in which the government
placed them upon their arrival in France. At this time when French citizens identified an array of
places, events, peoples, leaders, symbols, books, ideas, and traditions as manifestations and
reminders of their past, the camps became harki citizens’ lieux de mémoire in France.
To help the aging harkis and their wives integrate into French society, government officials
at the national and local levels needed to ensure implementation of the policies proposed by the
Council of Ministers in 1975. Though some officials believed that it was already too late for harkis
and their wives to integrate into French society, this opportunity during the late 1970s was, in many
ways, the last while men were still part of the active work force. However, the government failed to
follow through on its policies and promises. This inaction during the late 1970s and 1980s proved
detrimental to not only the older generation of harkis but also their children. Many of them were
born in the 1960s and were coming of age in a country whose president had tried to pass laws to
encourage North African immigrants to return home in the late 1970s. Moreover, the increasingly
popular far-right National Front party heightened tensions toward North Africans through its antiimmigrant platform in the early 1980s. While these policies by President Giscard d’Estaing and
16
1991 is the first year that Le Monde included the term “harki” as one of the key words in its annual index.
378
sentiments by Jean-Marie Le Pen were not aimed toward harki citizens (in fact, Le Pen invited harkis
into his party), the backlash nevertheless influenced them. By virtue of their Algerian origins they
were indistinguishable from other North African immigrants in France, whom Le Pen, Giscard
d’Estaing, and other politicians blamed for the unemployment crisis that had plagued society since
the mid 1970s.17
After the 1975 summer of protests, the French government put in place Bureaux
d’information, d’aide administrative et de conseils pour les Français musulmans (BIAC) across
France and promised the closure of the forest hamlets and camps by the end of 1979. The latter
would not happen and the BIAC were eliminated in June 1981 after the government discovered that
these offices had misused funds intended to help harki families. Soon after his election, President
François Mitterrand (the first socialist president since the Popular Front) appointed a Secretary of
State for Repatriates, who created in May 1982 the Office national à l’action sociale, éducative et
culturelle (ONASEC) with offices in every regional prefecture across France. These offices “vise[nt]
à donner aux Français musulmans rapatriés les moyens économiques correspondant à l’exercice de
leur citoyenneté: emploi, éducation, logement, formation, expression culturelle.” 18 Yet, the
ONASEC (like the BIAC) failed in its mission to promote harki citizens’ integration owing to
electoral politics and ineffective policies. 19 Soon after his appointment by right-leaning Prime
Minister Jacques Chirac in 1986 (the first instance of a President and Prime Minister from different
political parties during the Fifth Republic), a new Secretary of State for Repatriates opened an
investigation into this institution instituted by his predecessor. The ensuing report determined that
over half of the money intended for the harki population went to the functioning of the ONASEC
On the “lois de retour,” see Weil, La France et ses étrangers, 144-92. On the rise of the National Front, see Françoise
Gaspard, Une petite ville en France (Paris: Gallimard, 1990).
18 JORF, Assemblée Nationale Débats Parlementaires (hereafter ANDP), Gérard Bapt (rapporteur spécial chargé des
rapatriés), Nov. 13, 1984.
19 For an analysis of how the political situation between the socialists and the National Front party in the early 1980s
influenced policies of Mitterrand’s government, see Choi, “The Muslim Veteran…,” 35-39.
17
379
offices.20 Once again, harki citizens became caught up in partisan politics and did not receive the aid
from the French government intended to address their integration difficulties. These failures let to a
second summer of protests.
In October 1990 isolated protests occurred at the Bias camp, which, despite promises from
Prime Minister Chirac in 1975 that it would be eliminated by the end of 1976, still remained open.
As a result of the events at the Bias camp, Prime Minister Michel Rocard established the Mission de
Réflexion sur les Harkis in December. This commission, unlike its predecessors, was composed
primarily of harkis and Algerian elites who had fought in the French Army (twelve out of fourteen
members). In May 1991 the commission released a report, which opened by admitting the failures of
previous government policies and commissions that the government had put in place: “Près de
trente ans après son rapatriement en France dans des conditions particulièrement hostiles, la
communauté rapatriée d’origine nord-africaine n’a pas réussi, pour la grande majorité de ses
membres, son intégration au sein de la société française. Les politiques suivies depuis de nombreuses
années n’ont pas permis d’atteindre cet objectif.” 21 After listing thirty proposals ranging from
concrete aid such as educational scholarships to the creation of “conditions d’une véritable
citoyenneté,” the report closed by warning the government that it must address the harki
population’s needs immediately to avoid a volatile situation: “Cette attente est demeurée longtemps
patiente et silencieuse, mais la mission est convaincue qu’elle ne le restera pas durablement. Une
nouvelle déception provoquera fatalement une explosion.”22 The Commission’s prediction came true
just one month later.
SPDP, dossier: Harkis, Annette Levy-Willard, “Les drôles de fenêtres par où s’est envolé l’argent des harkis,” Libération,
June 24, 1986.
21 ACNMF, 20, “Rapport à Madame le Premier Ministre de la Mission de Réflexion sur la communauté rapatriée
d’origine nord-africaine,” 1, May 1991.
22 Ibid., 42.
20
380
In June 1991, violence erupted by the second and now the third generation of harkis, who
cried out against what they believed had been thirty years of empty promises by the French
government. The revolts began on June 25 when two sons of harkis, who lived in la cité des Oliviers
in Narbonne near the Bias camp, and two other non-harki youths, were jailed for throwing Molotov
cocktails that ignited storefronts and cars. A few hours after their arrests, sixty sons of harkis set
siege to the city demanding the release of their friends and new governmental measures to benefit
France’s harki citizens. The next day, Minister of Social Affairs Jean-Louis Bianco promised, “‘dans
les 48 heures, les crédits nécessaires seront débloqués pour les bourses scolaires et surtout
l’emploi.’”23 Despite the immediate governmental response and promises, violence spread around
the country with incidents such as blocked highways, barricading camp entrances, and setting fire to
cars. On July 2, the youths who had been arrested on June 25 were denied bail, which sparked new
manifestations of violence and appeals to the public. The violence, in particular, grabbed the
attention of the media, which honed in on actions by male harki children (few females participated
in the protests). Newspapers that covered these strikes frequently employed words such as “révolte,”
“ghetto,” “agitation,” and “banlieue,” images associated with other Arab immigrants. Article titles
about the harki sons’ actions encompassed: “Harkis: attention danger!,”24 “Narbonne, le cri des fils
de harkis,”25 and “Violences chez les harkis.”26
Muller, Le silence des harkis, 137.
SPDP, dossier: Harkis, Florence Mahe, “Harkis: attention danger!,” Quotidien de Paris, July 9, 1991.
25 SPDP, dossier: Harkis, Christian Goutorbe, “Narbonne: le cri des fils de harkis,” Le Figaro, June 25, 1991.
26 SPDP, dossier: Harkis, “Violences chez les harkis,” Le Monde, July 6, 1991.
23
24
381
Figure 40: Protest at the Fuveau forest hamlet, July 1, 1991. Source: Agence France-Presse archives.
The harki children not only wanted to change their present situation, but during these
protests they also became increasingly focused on having their families’ history recognized by the
French state and the public. To combat and justify the violent images proliferating in the press (see
figure 40), harki children in Fuveau passed out the following tract to motorists on the adjacent
National Highway 96 and at tollbooths on the A8 interstate explaining their cause:
AVIS À LA POPULATION
Les harkis de Fuveau comme dans toute la France sont en lutte.
Depuis plusieurs jours les chaines de télévision et l’ensemble des médias nous présentent
comme des voyous et des terroristes.
En réalité qui sommes nous?
Des Français qui veulent tout simplement être vraiment reconnu en tant que tel.
C’est à dire :
-Avoir le droit à un emploi pour vivre décemment,
382
-Avoir le droit à un logement et non être parqués dans des ‘réserves à indiens’
-Que nos parents soilent [sic] indemnisés de ce qu’ils ont perdu après avoir servi de ‘chair à
canon’.
Nous vous posons la question:
Est-ce être des terroristes que de demander des droits aussi élémentaires?
Depuis trentes [sic] ans les gouvernements nous ont fait des prommesses [sic] (qu’ils ont
renouvellé [sic] lors de chaque campagne électorale).
Aujoud’hui [sic] nous ne voulons plus de paroles mais nous exigeons du gouvernement des
actes!
À vous de juger, nous avons besoin de votre soutien.27
This tract, similar to others distributed by harki citizens around France, summarized their core
frustration: the government did not treat them as full French citizens.
Some protests by harki citizens took a more peaceful approach, while still insisting on their
second-class citizenship. On July 13, between hundreds, perhaps one thousand, former harki
soldiers and their families led a march from la Place de la Bastille to la Place de la République in
Paris wearing badges such as “Mon pays, la France?” and “Où est la France?” (see figure 41). The
next day, Bastille Day, newly-inaugurated center-left Prime Minister Edith Cresson unveiled a
twenty-five point plan in favor of the harkis with the title “Pour une véritable citoyenneté.”
Although the French government in part recognized the problems faced by the harki population, the
measures it took ignored many of the suggestions from the Mission de Réflexion sur les Harkis.
According to Le Monde journalist, Alain Rollat, the new measures “ne retient qu’un tiers en laissant
apparemment de côté celles qui étaient jugées les plus importantes par les principaux intéressés.”28 In
particular, Cresson did not call for a parliamentary debate, which would have constituted a symbolic
way to mark the official recognition of the services rendered by the harkis to the French state. She
did not authorize to the former harki soldiers the indemnity payments that they had been denied for
decades. Finally, the Prime Minister did not apply suggestions that would have helped to prevent the
27
28
ADBR, 1693 W 232, “Avis à la population,” July 1991.
Alain Rollat, “Déphasages,” Le Monde, July 15, 1991.
383
daily discriminations that children and grandchildren of harkis faced each day.29 Sociologist Laurent
Muller, who worked with the harki population during his military service in the late 1980s, concurred
with Rollat’s opinion that the government ignored the recommendations that could help the harki
citizens most in need: “En résumé, toutes les mesures spécifiques aux harkis, apparemment
accessibles au plus grand nombre, ne sont en fait réservées qu’à une minorité composée de
personnes déterminées, sachant composer avec les arcanes d’une administration bien
contraignante.”30 For example, only eighteen people qualified for a fund that had been set up for the
most poverty-stricken harkis of the first generation. Moreover, these harkis received between five
hundred and one thousand francs each.31 In short, Cresson’s government implemented neither the
measures recommended by the commission that would recognize the harkis’ contributions to the
Algerian War for Independence nor those that would provide concrete assistance to the most
destitute members of the population. These protests, nevertheless, did bring the government’s
attention to its harki citizens’ plight and opened a parliamentary debate about recognizing the harkis’
sacrifices.
Ibid.
Muller, Le silence des harkis, 160.
31 Ibid.
29
30
384
Figure 41: Peaceful march led by harkis and their families on July 13, 1991. Source: Agence France-Presse archives.
Harki Citizens and Commemoration
Soon after rightwing Prime Minister Edouard Balladur named Roger Romani Minister
delegate in charge of repatriates in 1993, he created a working group to address the problems of the
“Français musulmans rapatriés,” as government officials now referred to harki citizens. Like many
other ministers before him, Romani critiqued previous administrations for the population’s failed
integration: “L’accueil de la communauté Français musulman d’Algérie s’est effectuée dans de telles
conditions d’impropriété que, 30 ans après, elle n’a toujours pas trouvé sa place au sein de la
communauté nationale, comme l’ont montré les graves incidents de l’été 1991.”32 As a result of the
working group’s findings, Romani sponsored a law, which the Parliament passed on June 11, 1994.
The first article of “la loi dite Romani” affirmed the gratitude of the French Republic toward the
CAC, 19980027/3, “Note sur le chiffrage des mesures attendues à l’issue du Groupe de Travail sur les FMR,” Feb. 18.
1994.
32
385
harkis: “La République française témoigne sa reconnaissance envers les rapatriés anciens membres
des formations supplétives et assimilés ou victimes de la captivité en Algérie pour les sacrifices qu’ils
ont consentis.”33 In addition to officially recognizing former auxiliary soldiers, some of whom had
been prisoners of war, the legislation allocated a total of 2.5 billion francs to assist France’s harki
citizens. 34 While the harki community welcomed this law and some received partial indemnity
payments, the government took as long as three years to enact legislation allowing harki citizens to
receive the money intended to compensate them for previous financial measures for repatriates that
they had been unable to access.35 Moreover, Romani admitted in a 2008 interview with the author, as
cited in the epigraph, that the law’s financial benefits were not as far reaching as he intended.
With the passage of these financial measures, many harki citizens wanted public recognition
for their acts and sacrifices during the war. This goal became a more tangible possibility in 1999,
when the French government recognized the Algerian War as a war by changing the official lexicon
to “la guerre d’Algérie.” Many harki citizens’ chief demands for the government evolved from jobs,
indemnity payments, educational scholarships, and recognition in 1991 to public commemoration,
free circulation to Algeria, and the French government taking responsibility for “abandoning”
former auxiliary soldiers after the March 1962 ceasefire in the late 1990s. With many harkis having
perished and those remaining in their sixties, seventies, and eighties, it was too late for them to
integrate into French society.
Calls for commemoration of the harki population intensified after Algerian President
Bouteflika’s June 2000 state visit to France during which he refused to discuss with President Chirac
JORF, “Loi nº 94-488 du 11 juin 1994…,” 8567.
ACNMF, 20, Ministre délégué aux relations avec le Sénat, chargé de rapatriés, “Mise en œuvre du plan d’action en
faveur des anciens membres de formations supplétives et de leurs familles. Loi nº 94-488 du 11 juin 1994,” 4.
35 One of the principal texts to apply the law was passed in 1997: JORF, “Décret nº 97-677 du 31 mai 1997 modifiant le
décret nº 94-648 du 22 juillet 1994 portant application de la loi nº 94-488 du 11 juin 1994 relative aux rapatriés anciens
membres des formations supplétives et assimilés ou victimes de la captivité en Algérie,” June 1, 1997, 8758. In his
memoirs, former CNMF president André Wormser wrote about the 1994 law: “Notable avancée, donc, que cette loi.
Mais, je déplorais la lenteur de ses applications.” Wormser, Pour l’honneur des harkis, 109.
33
34
386
the possibility of harkis visiting Algeria and referred to them as “collabos.” The harki community
was infuriated by these remarks, despite President Chirac’s denunciation (albeit a month later). Some
believed that the French government should have been more insistent with the Algerian government
to ensure that harkis would be able to visit family and friends in Algeria whom they had not seen
since the end of the war. Following the incident, the French government was under constant
pressure by not only harki associations but also members of government, the press, and academics
to officially recognize the harkis’ contributions to the Algerian War for Independence. This pressure
unfolded in the political context of the historicization of the relationship between France and
Algeria and of increased public interest in the war. In particular, one of its more controversial
aspects, torture, which FLN members accused the harkis of carrying out, garnered attention. In June
2000, Le Monde ran an article detailing the torture and rape of a female FLN-supporter, Louisette
Ighilahriz, by French soldiers. 36 In May 2001, French general Paul Aussaresses released his
unapologetic memoirs recounting the torture he committed and ordered during the war.37 Finally,
this same year, historian Raphaëlle Branche published her dissertation, La torture et l’armée pendant la
guerre d’Algérie, which coincided with, and arguably sparked, an increased interest for the war.38
On September 25, 2001, the French government extended to the harkis its formal
recognition for their services during the Algerian War for Independence. President Jacques Chirac
presided over a ceremony for the Journée nationale d’hommage aux Harkis held in the courtyard of
the Hôtel national des Invalides, the symbolic site of the French Army. Soldiers placed a plaque in
the main courtyard of the Invalides that reiterated article one of the June 11, 1994 law. Similar
plaques were unveiled in twenty-seven cities and towns across the country with large harki
Florence Beaugé, “Torturée par l’armée française en Algérie, ‘Lila’ recherche l’homme qui l’a sauvée,” Le Monde, June
19, 2000.
37 Paul Aussaresses, Services spéciaux: Algérie 1955-1957 (Paris: Perrin, 2001).
38 Raphaëlle Branche, La torture et l’armée pendant la guerre d’Algérie (Paris: Gallimard, 2001).
36
387
populations. President Chirac also individually acknowledged one hundred harkis with medals of
honor. In his speech, Chirac recognized a “debt of honor” to the harkis for their service by stating:
Les anciens des forces supplétives, les harkis et leurs familles ont été les victimes d’une
terrible tragédie. Les massacres commis en 1962, frappant les militaires comme les civils, les
femmes comme les enfants, laisseront pour toujours l’empreinte de la barbarie. Ils doivent
être reconnus…. La France, en quittant le sol algérien, n’a pas su les empêcher, c’est vrai.
Elle n’a pas pu sauver ses enfants. Qu’elles soient tombées avant ou après le cessez-le-feu,
nous devons à toutes les victimes l’hommage du souvenir. Oublier une partie d’entre elles, ce
serait les trahir toutes. Aux combattants, à ces hommes, à ces femmes, j’exprime la
reconnaissance de la Nation. La République ne laissera pas l’injure raviver les douleurs du
passé. Elle ne laissera pas l’abandon s’ajouter au sacrifice, l’oubli recouvrir la mort et la
souffrance.39
This act of commemoration by the French government was a victory for its harki citizens.
Chirac’s speech represented the culmination of a four-decade battle to receive acknowledgement for
their actions. Although they were pleased with such recognition, members of the harki community
exposed three complaints about Chirac’s discourse. First, some protested it was a rewriting, not a
recognition, of history since he did not admit any responsibility of the French government for their
fate.40 Specifically, they were infuriated by Chirac’s statement, “La France, en quittant le sol algérien,
n’a pas su les [les massacres] empêcher, c’est vrai. Elle n’a pas pu sauver ses enfants.” Chirac asserted
that France could not save the harkis and their family members from the massacres. However, as
chapter two of this dissertation demonstrated, the French government was aware of the massacres
and yet did nothing to stop them. Further, some in the harki community argued that the French
government gave orders to disarm its harki soldiers, “abandoned” them, tried to prevent their
repatriation to France, and did not intercede with the new Algerian government to prevent the
massacres. This was the history that harki citizens wanted the French government to take
responsibility for. Second, harki citizens had hoped that the President would have asked for
forgiveness for the instructions of the Debré government, by threat of sanctions, not to allow
39
40
“Chirac: ‘les massacres (de harkis) commis en 1962 doivent être reconnus,’” Agence France-Presse, Sept. 25, 2001.
John Lichfield, “Chirac recognises ‘debt of honour’ to Algerians,” The Independent, Sept. 26, 2001.
388
returning French soldiers from Algeria to bring harkis back to France to save them. 41 Lastly,
members of the harki population had wanted Chirac to make, and follow through on, promises to
insist that the Algerian government allow former harki soldiers to visit Algeria.
To continue to address the compensation and commemoration of France’s harki citizens,
Secretary of State for Veterans Hamlaoui Mekachera—an Algerian who had served as a career
French Army officer—delivered a speech before the National Assembly on December 2, 2003 to
open a debate about allocating indemnity payments to all repatriates to France. While the scope of
the legislation would be broad, Mekachera’s intention was to focus particularly on France’s harki
citizens, a category of repatriates whom he claimed “a été trop longtemps oubliée par l’histoire.”42
The Secretary of State described the harkis’ history, which spanned the colonial and postcolonial
periods and geographically stretched from Algeria to France, in these terms: “Leur engagement total
au service de la France, la tragédie immense qu’ils ont vécue, leur abandon et leur détresse morale
ont été longtemps ignorés de la communauté nationale. Les familles ont souffert de l’isolement et
des conditions de vie qui leur ont été réservées à leur arrivée sur leur nouvelle terre d’accueil. Leur
insertion dans le tissu économique et social n’en a été que plus difficile.”43 According to Mekachera,
French society was unaware of the harkis’ tragic history—before and after decolonization and in
Algeria as well as France. The new law he sponsored aimed to rectify this situation.
On March 10, 2004, Minister of Defense Michelle Alliot-Marie presented the first draft of
the law signed by Prime Minister Jean-Pierre Raffarin to the National Assembly. The first article
expressed the Nation’s gratitude to all women and men who participated in France’s colonial
ventures in Algeria, Morocco, Tunisia, and all other former colonies. The next three articles included
financial measures to repay the harkis and their widows for lost goods in Algeria and their sacrifices
Philippe Bernard, “Harkis: l’aveu du président,” Le Monde, Oct. 2, 2001.
JORF, ANDP, “2e Séance du 2 décembre 2003,” Dec. 3, 2003, 11520.
43 Ibid.
41
42
389
in France and the final two articles concerned indemnities for all repatriates.44 This law reflected
Mekachera’s contention that the law was created to primarily benefit the harkis.
Nevertheless, when the law returned from Senate and National Assembly committees in
June 2004, articles two through four became articles six, seven, and nine, thereby moving these
measures for the harkis down. Instead, the Senate and the National Assembly inserted three articles
concerning all repatriates and an article to benefit the harkis specifically. The transcript of the June
11 National Assembly debate reveals pressure from the pieds-noirs to include more measures to
recognize them. Article two, added by the Senate, bestowed the title of “combattants morts pour la
France en Afrique du Nord” that had been reserved for soldiers to civilian victims of both the
Algerian War for Independence, notably those who perished after the March 19 ceasefire, and the
much less violent independence struggles in Morocco and Tunisia. The Senate also added article five,
which made it illegal to call someone a “harki” as an insult (as a synonym for traitor) and to deny
that harkis and other auxiliary forces had been tortured and massacred after the ceasefire. The
National Assembly added articles three and four. Article three established a “Fondation pour la
mémoire de la guerre d’Algérie, des combats du Maroc et de Tunisie,” which would constitute an
oral history repository for repatriates. Finally, article four stated: “Les programmes de recherche
universitaire accordent à l’histoire de la présence française outre-mer, notamment en Afrique du
Nord, la place qu’elle mérite. Les programmes scolaires reconnaissent en particulier le rôle positif de
la présence française outre-mer, notamment en Afrique du Nord, et accordent à l’histoire et aux
sacrifices des combattants de l’armée française issus de ces territoires la place éminente à laquelle ils
ont droit.”45 These amendments were incorporated into the final version of the law passed by
Parliament on February 23, 2005. The fourth article immediately became a point of public debate
JORF, “Projet de loi portant reconnaissance de la Nation et contribution nationale en faveur des Français rapatriés,”
Mar. 10, 2004.
45 JORF, “Projet de loi portant reconnaissance de la Nation et contribution nationale en faveur des Français rapatriés,”
June 11, 2004. The final version of the law retained all of these measures.
44
390
since the government was asking educators to emphasize the positive influences of colonialism,
when there were many negative aspects. This article advocated that descriptions of the harkis and
other native soldiers in the French Army be included in French textbooks. However, the phrase “le
rôle positif de la présence française outre-mer,” changed the meaning of this sentence. The debate
over this controversial article lasted almost ten months before this sentence was repealed from the
law. Thanks to this legislation, harkis or their widows received indemnity payments that had been
delayed over forty years. Yet the debate sparked by the article that legislators inserted under pressure
from the pied-noir community overshadowed the French government’s commemoration of its harki
citizens.
From Algeria to France, From the Colonial to the Postcolonial
On September 25, 2009, I stood among some fifty townspeople, local officials, and members
of the harki population, including three former soldiers, for the ninth annual Journée nationale
d’hommage aux harkis in Fuveau. I watched the solemn twelve-minute ceremony unfold. A veteran
wearing his military uniform adorned with medals performed the call to arms while holding a
tricolor flag with “À nous le souvenir. À eux l’immortalité” embroidered in gold thread. Fuveau’s
mayor and deputy mayor, two Bouches-du-Rhône departmental officials, and a former harki soldier
and a former GMS soldier placed at the base of the monument three sprays of flowers girded with
ribbons reading “Hommage Aux Soldats Harkis De France.”
During the second half of the brief ceremony, I listened while an official from the
departmental veterans office in Marseille and Mayor Jean Bonfillon delivered short speeches, which
the national ONAC office in Paris had prepared for officials across France. After welcoming
everyone to this symbolic place in front of the monument aux morts, the departmental representative
read the phrase that the government used to announce the opening of Cité Nationale de l’Histoire
391
de l’Immigration in Paris: “Leur histoire est notre histoire.” He continued by offering the thanks of
the entire nation for the harkis’ faithful service and sacrifices, including their exile from their native
Algeria and from “une métropole qui ne les attendait pas.” He concluded by noting the failures—of
the government and the population—to include the harkis and their families into French society:
“Ils ont mérité le soutien de la communauté nationale.” 46 Mayor Bonfillon’s speech began by
acknowledging the repression of the Algerian War for Independence from the French government’s
“official memory,” evidenced by the use of terms such as “les événements d’Algérie.” His words
recognized the ongoing effects of the war’s violence, which had left marks of psychological and
moral trauma on both societies—particularly on individuals, such as the harkis. His discourse
attested to precisely how the French government had failed its harki citizens, which this day in their
honor sought to reverse:
Nous savons tous, que trop longtemps la France a baissé les bras devant l’obligation
contractée à l’égard des harkis. Elle n’a pas su reconnaître les sacrifices de leurs biens, de
leurs terres, de leurs droits, et de leur sécurité, parfois même de leur vie… Au moment où les
harkis s’en remettaient à elle, elle les a conduits par les chemins de l’oubli dans les camps de
transit, les hameaux forestiers, et les cités urbaines. Depuis 2001, la journée nationale qui leur
est dédiée interdit cette démission de la mémoire. Elle célèbre, ce que nous faisons
aujourd’hui, leur fidélité et leur bravoure. Elle est à honorer, cette communauté, à faire
connaître, et comprendre la dette que la France lui conserve.47
The mayor explained that the purpose of this ceremony was in part pedagogical: to heighten the
awareness of their concitoyens français to the “vérité” of the harkis’ history, as written above, so that it
can be properly transmitted to future generations. In so doing, the harki community and the national
community could move forward together to build a stronger bond. He closed by affirming Fuveau
and the entire nation’s recognition of the harkis.
These men’s speeches, with texts provided by the ONAC, underscore that harki citizens’
history comprises decades of social exclusion counterbalanced with recent efforts by the
46
47
Representative of the ONAC departmental office in Marseille, public address, Fuveau, Sept. 25, 2009.
Mayor Jean Bonfillon, public address, Fuveau, Sept. 25, 2009.
392
government to include them into “official” memory and history. Harki citizens’ history is more
nuanced than the description given by these two speeches. Government discourse and policies since
1962 demonstrated its officials’ difficulties in negotiating between poles of incorporation (harki
citizens belonged in French society) and differentiation (treating them as a different type of citizen
who necessitated different treatment), which had been one of the central features of imperial
governance.48 Yet, unlike in the colonies—and even Algeria where natives were accorded secondclass French citizenship—the harkis and their family members were not subjects. After undergoing
the special nationality procedure open exclusively to Algerian natives, they became French citizens.
However, as this dissertation has demonstrated, the government did not treat them as full French
citizens. They were—and are—“harki citizens.”
The harki population’s exile from both nations defines the uniqueness of their situation.
Harki citizens’ exile from Algeria resulted from their choice to fight for France during the Algerian
War for Independence. The end result of government policies toward the harki population since the
first harkis and their family members arrived in the metropole in spring 1962 indeed exiled them
from French society, as the ONAC official’s speech maintained. The mayor’s speech noted that for
“trop longtemps la France a baissé les bras devant l’obligation contractée à l’égard des harkis.” The
most notable example of a government policy that marginalized the harki population from French
society was the choice to place harkis and their families requesting housing assistance upon their
“repatriation” to France into refugee-like camps like Bourg-Lastic, Larzac, Saint-Maurice-l’Ardoise,
and Rivesaltes. They arrived “dans une métropole qui ne les attendait pas,” as the ONAC official
pointed out. Mayor Bonfillon’s words correctly assessed that “[la France] les a conduits par les
chemins de l’oubli” because these initial camps opened the door to future camp living situations.
Given the post-Second World War housing crisis, the Ministry of Repatriates’ priority of rehousing
48
Cooper, Colonialism in Question, 154.
393
pieds-noirs, and the government’s vision of the harki population as unready for life in metropolitan
France, the SFIM opted to rehouse 25 percent of those already in camps in cités urbaines, forest
hamlets, and other camps for the “unhouseables.” All of these spaces pushed harki citizens to the
fringes of French society—sometimes the literal geographic frontiers of France, as in the case of the
Rivesaltes camp on the Spanish border—because they lived exclusively with other harki families and
were often treated differently by the government than other French citizens. Reminiscent of
previous state-run housing for foreign migrants, the French government agents whose responsibility
was to oversee, or encadrer, the residents had a double objective: surveillance and education. Former
civilian and military agents, many of whom had served in Algeria, carried out this encadrement, which
resembled practices in the colony and metropole during the imperial era. The government’s housing
policies cast its harki citizens as second-class citizens, as they had been in Algeria.
Minister of Repatriates François Missoffe’s discourse initially maintained that all of these
housing options were intended to be a transitional step toward integrating harki citizens into French
society—a discourse of inclusion. However, in July 1964 when President de Gaulle eliminated the
Ministry of Repatriates—whose job included the reclassement of the harki population—4,500
members of the harki population lived in camps, 8,500 had moved to the forest hamlets, and the cités
urbaines contained approximately 1,500 apartments.49 Missoffe began to distance himself from the
description of the forest hamlets as “temporary” as early as May 1963. Soon after he stepped down
and the Ministry of Interior took over his functions, the isolated forest hamlets became a
“permanent” solution, which challenged these harki citizens’ belonging in French society and
continued to treat them like refugees.
Subsequent government policies (notably those enacted in 1975 and 1991 following protests
by children of harkis) sought to correct the failure of previous ones to integrate France’s harki
For the number of cités urbaines apartments, see CAC, 19920149/1/4, “Programme de Construction pour les ‘Harkis:
Situation au 1er janvier 1965,” 3. See page 271 for the number of residents in the forest hamlets and camps.
49
394
citizens and treat them like full French citizens. These new policies, such as the 1975 Council of
Ministers plan to eliminate the remaining camps and forest hamlets, promised to make significant
changes to address the harki population’s difficult relationship with French citizenship. However,
both a lack of follow through by national government officials and varied implementation of these
policies at the local level translated into this goal remaining unmet. At these moments of protest, the
remaining camps became the symbol of the government’s “oubli” of its population of harki citizens.
The harki community’s inextricable link to the so-called “événements d’Algérie” was another
reason for their marginalization from French society. The French government did not begin to enact
legislation to specifically compensate the harkis for their lost land and material goods and officially
remember them for their role during the Algerian War for Independence until the mid-1990s. These
acts to include the harkis into its policies for other veterans and repatriates coincided with France’s
evolving relationship with its colonial past. Mayor Bonfillon explained that for a long time France
“n’a pas su reconnaître les sacrifices de leurs biens, de leurs terres, de leurs droits, et de leur sécurité,
parfois même de leur vie.” The French government arguably did “know” how to commemorate and
compensate its harki citizens, but chose not to. The 1961 and 1970 laws for all repatriates were open
to the harki population. Nevertheless, these individuals regularly encountered difficulties accessing
these benefits. The government was aware of this situation yet did little to remedy it. Though more
harki citizens were able to access benefits following the 1994 law, these benefits, as its sponsor
Roger Romani admits, failed to fully compensate them. The 2005 law succeeded in financially
compensating France’s harki citizens. However, the commemoration of harki soldiers and other
auxiliary forces became mired in the debate over the law’s fourth article, which advocated that
schoolteachers emphasize the positive effects of France’s colonial empire. The original version of
this law, which Minister Mekachera intended foremost to recognize France’s harki citizens, did not
contain this sentence. Instead, former pro-French Algeria deputies and others influenced by the
395
vocal pied-noir lobby inserted this phrase that was bound to strike a controversy in a French society
whose colonial past still provoked bitter disputes. The harki citizens lost again in this debate over
France’s colonial ventures, just as the Franco-French war in 1962 had obscured the harkis’ interests.
“Leur histoire est notre histoire” was a telling choice for the third sentence of the ONAC
representative’s speech because this phrase was used to announce the opening of the Cité Nationale
de l’Histoire de l’Immigration in 2007. On a symbolic level, this expression linked France’s harki
citizens to the controversy surrounding this museum for immigrant history. Eight scholars on its
scientific steering committee collectively and publicly resigned in May 2007 to protest the creation of
a Minister of Immigration, Integration, National Identity, and Co-development by newly elected
President Nicolas Sarkozy. They argued in an article published in Le Monde that the association of
immigration with national identity inscribes immigrants as a problem for France, even those who
become French citizens.50 Moreover, “leur histoire est notre histoire” underscores the difficulties the
French government continues to have categorizing the harki population. This expression seemingly
attempted to incorporate the history of harkis citizens into French national history while
simultaneously confirming a gap between “our” history and “their” history, which has been
relegated to the periphery, like that of France’s immigrants. The marginalization of immigration
history from French national history was symptomatic of how the French government has treated its
immigrant populations. France’s newly arrived laborers (and their families) often resided in
precarious conditions in boarding houses, cités du transit, shantytowns, HLM buildings, and
SONACOTRA hostels—with some remaining in these spaces for decades. State-run housing for
immigrants, like that for the harki population, included encadrement agents whose function was to
Marie-Claude Blanc-Chaléard, Geneviève Dreyfus-Armand, Nancy Green, Gérard Noiriel, Patrick Simon, Vincent
Viet, Marie-Christine Volovitch-Tavarès, and Patrick Weil, “Immigration et identité nationale: une association
inacceptable,” Le Monde, May 21, 2007. Neither President Sarkozy, nor any member of his cabinet (including the new
Minister of Immigration) was present for the museum opening in September 2007, as is customary when new national
museums are inaugurated in France.
50
396
educate and police them. While the French government’s administration of harki citizens and that of
immigrants display many similarities, the harkis were not immigrants. Even though the government
sometimes treated them as refugees, they were not legally accorded refugee status. They were
repatriates, even though they did not receive all of the rights and benefits associated with this status.
Perhaps the biggest obstacle to the harkis being considered as French citizens by the
government and the public was their inescapable link to Algeria and Algerian ethnicity. Many fled
Algeria without their identity papers and almost all became French citizens, yet the government
continued to refer to them by labels that linked them to their colonial Algerian status as inferior
citizens. At the same time, and in seeming contradiction, government documents, pamphlets, and
the name of integration policies from the early 1970s onward continually reminded national and
local government agents responsible for implementing policies that the harkis and their families were
citizens and should be treated as such. The June 1973 letter that Prime Minister Pierre Messmer
wrote to the members of his cabinet, regional prefects, and departmental prefects underscoring that
the harkis must be treated as French citizens contained the subject line: “Situation des Musulmans
français.”51 In 1984 the Secretary of State for Repatriates published a pamphlet “Vaincre l’oubli,”
which encompassed articles such as “Devenir citoyen à part entière de sa ville ou de son village.”
This pamphlet explained to local officials specific measures they could take to integrate the
“Français musulmans rapatriés” in their communities.52 Finally, the plan that the government created
in response to harki children’s protests in 1991 was named “Pour une véritable citoyenneté” and was
intended for the “communauté rapatriée d’origine nord-africaine,” whom Minister of Social Affairs
and Integration Jean-Louis Bianco claimed “reste trop souvent en marge de la communauté
51
52
SHAT, 3R 79/7, Pierre Messmer, “Objet: Situation des Musulmans français,” 2, June 26, 1973.
ADBR, 1451 W 115, Secretary of State for Repatriates, “Vaincre l’oubli,” Dec. 1984.
397
nationale.”53 The harkis citizens’ past in Algeria and that in France influenced how the French
government treated them at the end of the colonial period and throughout the postcolonial era. The
postcolonial treatment of the harki population as inferior citizens has also marked French citizens
and indeed French society today.
Figure 42: Aïssa B. and Abdelkader B. in front of le monument aux morts in Fuveau on September 25, 2009.
Photograph taken by the author.
The ceremony in Fuveau on September 25, 2009 represented a gap between the government
officials’ words and actions. As I observed the events, my attention was continually drawn to the
three of the remaining eleven vieux who had traveled from the isolated camp on the town’s outskirts
Archives communales de la Roque d’Anthéron, Minister of Social Affairs and Integration Jean-Louis Bianco to
departmental prefects, “Pour une veritable citoyenneté,” 1, July 17, 1991.
53
398
where harki families had resided over the last forty-five years. They stood with somber expressions
on their faces separated from the rest of the crowd. The ceremony took place on a Friday morning
at eleven o’clock, which made it difficult for anyone who worked and school-aged children to attend
(in previous years it had been held in the early evening). Indeed, part of the pedagogical lesson that
the mayor had listed as the ceremony’s primary objective was lost. While I listened to the mayor’s
speech I waited for him to personally recognize the service and sacrifices of Mr. B., Mr. B., and Mr.
Z. These men stood about fifteen meters away from him. But Mayor Bonfillon continued to read
from the speech issued by the ONAC. He rarely looked up from the paper and never met the gaze
of these three soldiers and citizens. After the ceremony, three townspeople—two wearing military
uniforms and a former town councilman assigned to helping the harki population—shook the men’s
hands and exchanged pleasantries with them. Since I had interviewed two of these men and knew
the other from my dozen trips to the forest hamlet, I asked if I could take their picture in front the
monument (see figure 42). Their solemn faces showed unease, even at the ceremony intended to
recognize them. The crowd dispersed; two of the harkis walked to the town hall for refreshments,
the third decided to go home. I stood with these two men and the former town councilman for the
second part of this ceremony in their honor, communicating with them as best as possible, given
their broken French. They smiled as they looked around the room. Few people approached us. The
mayor did not come over to thank these men to whom France owed a debt, of honor and
understanding, according to the words in his speech. He did, however, make sure to greet me after
the men left, demonstrating this French citizen and government official’s unease in the presence of
harki citizens. After fifteen minutes, the twenty attendees at the reception dwindled. The men
thanked me, another outsider, for coming. They walked out unnoticed, without recognition from
their concitoyens français.
399
APPENDIX A
Interview with Monsieur le Chancelier de l’Institut Pierre Messmer*
Paris, L’Institut de France, le 5 juillet 2004, 11h00
Pierre Messmer (PM): Et tout de suite je dois vous dire que je suis pas le meilleur interlocuteur sur
ce sujet-là parce que j’étais ministre des Armées à l’époque et ma compétence en ce qui concerne les
rapatriés d’Algérie quel qui soit, sauf naturellement les militaires, s’arrêtaient à leur retour en France.
Et c’était une autre administration qui s’occupait d’eux. C’était le ministère des Rapatriés qui était
alors dirigé par un homme jeune, dynamique, et qui vient de mourir qui est mort il y a très peu de
temps et qui avait sous sa responsabilité l’ensemble des actions en faveur des rapatriés. C’était
Monsieur Missoffe. Alors, ce qui de mon côté peut vous intéresser c’est ce qui touche à l’effectif des
harkis qui ont été rapatriés et de leurs familles aussi puis qu’on a rapatrié leurs familles avec eux. Sur
ce point là, vous trouverez, vous l’avez peut-être déjà trouvé, les chiffres dans un livre que j’ai écrit.
Jeannette Miller (JM): Les Blancs s’en vont.
PM: Les Blancs s’en vont ou un chapitre que vous l’avez vu est consacré aux harkis. Ce problème du
nombre des harkis rentrés en France, enfin refugiés si je peux dire en France, est un problème qui
m’a beaucoup intéressé par conséquent. Je me suis efforcé d’avoir le maximum d’information
sérieux sur la question et je crois que les chiffres que j’ai donnés dans ce livre-là ne sont pas
contestés. Par les historiens, ils sont plus contestés maintenant parce que vous avez dû voir il y a des
regroupements qui ont été faits à l’occasion des recensements par la suite. Donc il y a pas de réelle
contestation sur ces chiffres. Ce qu’il faut ajouter et ce qui peut vous intéresser par pour votre travail
c’est que ces harkis et leurs familles naturellement ce sont des ruraux, sont des paysans. Ce sont pas
des gens de ville, pas du tout. Et cela s’explique par le fait que les harkas à l’origine de la guerre
d’Algérie avaient été constitués pour la défense des villages, pour l’auto-défense en quelque sorte.
Par la suite, il y a eu des harkas qui sont devenus plus actifs, mais c’était essentiellement de la défense
du village et éventuellement des villages environnants. Donc ces harkis ce sont des ruraux. Beaucoup
d’entre eux parlent même pas le français. Ils comprennent, ils comprennent quelques mots, mais ils
parlent pas le français. Et par conséquent lorsqu’ils vont arriver en France avec leur famille ils sont
vraiment très dépaysés. Ils sont très dépaysés parce que la France n’est plus même dans les années
soixante n’est plus un pays rural. Il y a en effet beaucoup de campagne en France, mais on peut pas
on peut pas donner de propriété rurale aux harkis parce qu’il y a des propriétaires. La France n’est
pas un pays que l’on peut coloniser et puis d’autre part même les ruraux français dans les années 60
sont des hommes, des femmes qui sont déjà très évolués qui servent du machinisme largement. Ils
ont des tracteurs, ils ont toute sorte des machines. Les harkis sont incapables de servir de tout cela,
totalement incapables. Alors, ils vont être regroupés parce qu’on veut pas les disperser, on veut qu’ils
Pierre Messmer (1916-2007) had a lengthy military and public service career in the French colonies and metropolitan
France. Upon his 1937 graduation from the elite Ecole nationale de la France d’outre-mer (ENFOMA), he extended his
two-year mandatory military service into eight years in the French Army. Following the Second World War, he returned
to his chosen vocation as a colonial administrator first in Indochina (1946 to 1948) and then in sub-Saharan Africa (1949
to 1959), including a two-year stint as the last Governor-General of French West Africa (AOF). In January 1960, he
executed a brief tour of duty in Petite Kabylie as a lieutenant colonel in the eighth regiment of navy parachutists before
being summoned to Paris to serve in President Charles de Gaulle’s cabinet. He subsequently held the following
positions: Minister of Armies (1960-69), Minister of State for Overseas Departments and Territories (1971-72), and
Prime Minister (1972-74). Messmer was elected to the Académie Française in 1999, and served as Chancellor of the
Institut de France from 1999 until 2006.
*
400
restent un peu entre eux. On les regroupait dans des camps d’abord, puis ensuite dans des espèces
de villages de harkis surtout dans le Midi, dans le sud de la France où on les emploie des travaux
forestiers des choses comme cela. Mais, la vérité est que à de rares exceptions près, ils vont pas
s’intégrer. Pour leurs enfants, c’est autre chose. Ah, les enfants, c’est toute autre chose. Les enfants
eux vont à l’école, parlent français, et vont s’intégrer progressivement dans la société française
JM: Mais, est-ce qu’il y a eu des problèmes pour les enfants d’aller à l’école parce que j’ai lu dans
d’autres livres et je sais que les livres qui existent sont un peu…
PM: Oui, oui. Il y a eu des problèmes, c’est tout à fait vrai. Il y a des problèmes parce que le français
n’était pas leur langue maternelle, c’est une langue étrangère pour eux. Quand ils rentraient chez eux,
ils parlaient l’arabe ou le berbère avec leurs parents. Mais ils ont tous parlé le français, tous. Et par
conséquent, ils ont beaucoup plus facilement trouvé des emplois et ils se sont beaucoup mieux
intégrés que leurs parents dans la société française. Il y a pas de doute. Massivement, ils sont mieux
intégrés, mais il y a des exceptions dans les deux sens: il y en a très brillants et il y en a au contraire
ceux qui sont des échecs, mais dans la masse, ils se sont intégrés. Avec quelquefois des difficultés
particulières parce que les autres Algériens les considéraient un peu comme des adversaires, le
gouvernement algérien encore aujourd’hui considère les harkis comme des traitres, c’est ce qu’a dit
en tout cas le président Bouteflika, ce qui évidemment ne facilite pas les relations. Il y a eu en effet
une transition difficile, une transition entre ses parents qui ne pouvaient pas s’intégrer ou qui
s’intégraient très mal et les enfants qui, eux, n’avaient pas le choix, ils étaient obligés de s’intégrer. Il
y a eu une période très, très difficile, c’est vrai. A mon avis, elle est passée, elle est passée parce que
les plus jeunes ce qui sont nés, en fait les plus âgés plutôt, qui sont nés à la rigueur en France. Ils ont
quarante ans maintenant, ils ont des enfants et les plus âgés sont déjà à la retraite. C’est d’ailleurs un
bon moment pour faire une étude historique sur ce sujet-là. Alors si vous avez des questions, allez-y.
JM: Oui, oui, oui. Ce que je voudrais faire est d’étudier les lois de l’Etat français à l’égard des harkis
et comment cette politique a changé, pourquoi, à quels moments, etc. et pour faire cela il faut
d’abord voir la guerre d’Algérie, surtout le début des accords d’Evian en voyant qu’est-ce qui était dit
sur les harkis parce qu’il me semble que de Gaulle…
PM: Alors, écoutez, si vous souhaitez pour la partie guerre d’Algérie et retour d’Algérie vous avez
des archives qui sont au Service historique de l’armée et si vous souhaitez les consulter, moi je vous
donnerai une autorisation de le faire. Il faut que vous preniez en contact avec la direction du service
historique, c’est à Vincennes. Mais si vous voulez consulter ces archives, toutes les archives sont
déposées au service historique et je vous donnerai l’autorisation pour les consulter. Vous pouvez
voir tous que vous voudriez là.
JM: Alors, si c’est OK, je voudrais vous poser quelques questions. Alors, est-ce que, peut-être si
c’est dans les archives, excusez-moi, je peux regarder là, mais est-ce que général de Gaulle avait un
discours à l’égard des harkis pendant la conférence d’Evian parce que je sais que vous avez posé la
question des harkis, de Gaulle il a répondu: « Vous allez rallumer la guerre d’Algérie. »
PM: (Des pauses) La réponse est simple. C’est que de Gaulle veut sortir d’Algérie, veut sortir. Et
aller chercher les harkis là où ils se retrouvent, comme je vous disais dans les villages, c’est chaque
fois une opération militaire et ce que de Gaulle pense c’est que ces opérations militaires risquent de
rallumer la guerre d’Algérie. Et ca, on veut pas. Par conséquent, il est logique avec lui-même sur
401
l’Algérie. On peut reprocher à de G d’avoir été très dur. Mais il est d’une logique impeccable. Il y a
pas de contradictions dans sa politique. Alors évidemment il fait pas de sentiment.
JM: Mais vous, vous avez…
PM: Moi, j’étais plus jeune, alors j’avais encore un peu de sentiment (rires).
JM: Je vois. Pendant que vous étiez ministre des Armées je sais que ce que j’ai lu est que les harkis
étaient dans les camps et c’étaient les militaires qui ont dirigé les camps. Donc, la direction des
camps, c’était dans votre domaine comme ministre des Armées?
PM: Les harkis?
JM: Oui.
PM: Alors, c’est assez compliqué parce que vous le savez sans doute: on a réuni sous le nom de
harkis des activités qui étaient assez différentes. Il y avait les harkis proprement dit. Donc, qu’est-ce
que c’est qu’un harki, un vrai harki? C’est un supplétif et alors oui, c’est un supplétif de l’armée.
Mais il y a d’autres groupes de moghaznis de différentes activités qui ne sont pas des supplétifs de
l’armée, mais des supplétifs de l’administration. Et qui, eux, n’avaient pas ce statut de militaire. En
fait, on les a tous traité de la même façon, c’est-à-dire on les assimile en harkis. Mais, il faut savoir
que les harkis eux sont des supplétifs et donc dépend de l’armée, mais si je peux dire il y a des faux
harkis, c’est-à-dire que dans un village, par exemple, l’officier qui chargeait de la responsabilité de ce
village embauchera comme harki des gens qui serviront pour être plantés devant un bureau, pour
être cycliste, pour porter des puits. Ce sont des faux harkis en réalité, mais on les traite tous de la
même façon. Je n’avais pas sur la situation des harkis le même optimisme que la plupart des
militaires parce que j’avais vu ce qui s’était passé en Indochine, où les supplétifs avaient été ensuite
massacrés par le Viet Minh.
JM: Combien, vous savez?
PM: Et ça je savais. C’est pourquoi j’étais très, très réticent en ce qui concerne l’engagement des
harkis et je me suis toujours efforcé à partir du jour où je suis devenu ministre de faire diminuer leur
nombre. D’ailleurs vous verrez justement dans les statistiques que je donne le nombre des harkis ont
diminué. Et c’est moi qui veux que le nombre de harkis diminue parce que je sais qu’il y aura un
drame au moment où nous quitterons Algérie.
JM: Et vous avez dit qu’à partir de 1960 il est devenu plus difficile à attirer des harkis.
PM: Oui.
JM: et donc c’était les deux choses ou pourquoi c’était plus difficile? Parce qu’ils s’est [sic] rendu
compte que la guerre va terminer ou on a perdu?
PM: Parce que tout le monde commence à comprendre que général de Gaulle veut mettre fin à la
guerre. Alors, à partir de ce moment-là, chacun devient prudent. Les villageois berbères évitent de
trop s’engager avec nous et les officiers comprennent qu’il vaut mieux de ne pas engager de
nouveaux harkis, ce qui fait que les effectifs diminuent.
402
JM: Pour revenir à la guerre d’Indochine, pourquoi est-ce qu’il n’y avait pas de rapatriement comme
avec les harkis? Parce qu’il me semble que beaucoup de Vietnamiens, de Cambodgiens, et Laotiens
de l’armée française, la vaste majorité sont restés en Indochine. Pourquoi est-ce que la France a évité
ce rapatriement?
PM: Il faut d’abord savoir que les choses se sont pas passées de tout de la même façon. Quand la
France a quitté l’Indochine, l’Indochine était encore coupée en deux – nord et sud. Dans le sud
restait un gouvernement qui a très volontiers accueilli tous les gens du nord qui voulaient et qui était
du nord et venait du sud. Il y en a beaucoup sauf pour les populations de montagne qui sont dans le
nord de Vietnam et dans le nord de Laos qui voulaient pas quitter. Mais les Vietnamiens eux-mêmes,
en particulier les Vietnamiens du nord qui étaient catholiques, ont presque tous quitté le Vietnam du
nord après le départ des Français et sont installés au sud. Et le sud est resté encore, je dirais, libre
plus longtemps après puis qu’il l’était toujours que les Américains ont commencé leurs interventions
en revenant le gouvernement d’abord dans le sud et puis ensuite en renvoyant une véritable force.
C’est-à-dire, le problème de la répression en question s’est posé après le départ américain, pas après
le départ français, il s’est posé après le départ américain et à ce moment-là le Viet Minh a crée des
camps de concentration et il a mis beaucoup de gens dans les camps de concentration, mais ça se
passe après le départ des Américains.
JM: Oui.
PM: Mais nous avions vu quand même ce qui s’était passé au nord de Vietnam et on pouvait pas
oublier, on pouvait pas penser que les choses se dérouleraient plus facilement pour les supplétifs en
Algérie qu’avait tourné au Vietnam, au nord du Vietnam. On pouvait pas le penser et c’est toujours
la même chose on va s’apercevoir la même chose en Iraq.
JM: Oui, c’est triste.
PM: Les supplétifs sont encore mieux usés.
JM: C’est un autre travail à faire.
PM: Ah mais, c’est un constant historique cela. A un certain nombre, l’histoire se répète jamais mais
je crois que c’est ceux qui disaient l’histoire se répète pas mais lui arrive de balbutier. L’affaire de
supplétifs malheureusement empire.
JM: Pour reparler un peu du rapatriement des harkis, il semble que la politique de rapatriement ait
provoqué une rupture entre vous et le ministre Joxe. Maintes fois, vous exprimez que vous avez
refusé de tenir compte des déclarations du gouvernement.
PM: Oui, c’est-à-dire, oui. Il y a eu une tension. La tension entre nous s’est expliquée par que, pour
une double raison: l’une qui était rapidement réglée qui tient au fait que Joxe a eu la maladresse de
menacer les officiers de sanctions, de certaines sanctions. Alors moi, je pouvais pas accepter cela
parce que il n’y a que le ministre des Armées qui a le droit de sanctionner les militaires. Je pouvais
accepter qu’un autre ministre menace les militaires de sanctions, ce n’était pas possible. Et alors
c’était à ces règles sur cette affaire, mais elle s’est réglée très vite parce que Joxe a compris qu’il avait
été maladroit et qu’il ne pouvait pas espérer que général de Gaulle lui donnerait raison évidemment
403
et par conséquent, c’est réglée très vite. Mais, il y a eu une autre difficulté entre nous qui a duré
beaucoup plus longtemps. C’est que j’insistais naturellement pour que Joxe exige le respect des
accords et qui avait été passé entre le FLN et nous. Et que nous respections nous mais que le FLN
ne respectait pas toujours, en particulier, en ce qui concerne les harkis. Et Joxe qui d’une part avait le
souci de ne pas briser les accords, le même que de Gaulle avait le souci de ne pas commencer la
guerre, disons Joxe temporisait, voir insister pour qui proteste et même qui menace. Et lui se
contentait de, si je peux dire, d’intervention « polie » et ça, ça créait entre nous une tension assez
forte pendant quelque mois et Joxe savait bien que sur ce sujet c’était lui qui respectait la politique de
général de Gaulle beaucoup plus que moi et par conséquent…
JM: A l’égard des harkis?
PM: Oui. Sa politique vraiment était dans l’esprit qu’avait été celle de général de Gaulle. Et par
conséquent, il acceptait, euh, je ne dirais pas de fermer les yeux car il ne fermait pas les yeux, il était
trop intelligent et trop compétent pour cela. Mais il acceptait de tolérer des excès que moi je ne
tolérais pas. C’était cela, le fond du débat entre nous.
JM: Mais j’ai lu dans quelques livres et peut-être ce n’est pas le cas qu’il y avait des bateaux qui ont
été renvoyés à Alger. Non?
PM: Ce n’est pas vrai. Non. Ce n’est pas vrai. Ca était dit beaucoup à l’époque et écrit, il y a pas un
seul exemple de contrôler, de retour de harkis qui viennent en France et qu’on a envoyé en Algérie.
Il n’y a pas un exemple. On en a beaucoup parlé, oui. On en a beaucoup parlé parce qu’il y a des
exemples en revanche de harkis qui avec l’accord de leurs officiers ont embarqué clandestinement.
Alors, ça oui. Oui, les embarquements clandestins, la réponse est oui. Il y en a eu, on peut pas dire
combien exactement, mais il y en a eu pas mal. Mais le retour, non. Vous savez, à l’époque il y avait
une polémique terrible en France. Il faut pas oublier que c’est l’époque où il y a encore des actions
précises de l’OAS. L’OAS est encore active à l’époque. L’OAS a essayé d’assassiner le général de
Gaulle. Alors, naturellement les hommes OAS faisaient cette propagande anti-gaulliste, mais
maintenant vraiment on a les éléments qui permettent avec le recul de voir ce qui s’est passé
vraiment – la réponse est non. Il y a eu beaucoup d’embarquements clandestins, mais il n’y a pas eu
de retours.
JM: Pendant que vous étiez Premier ministre, je sais qu’il y avait des grandes manifestations qui ont
commencé à partir de novembre 1974, mais c’était après votre démission. Mais en fait il y avait de
petites manifestations avant cela à partir de 72. Est-ce que votre gouvernement a intervenu dans ces
manifestations? Etiez-vous au courant de ce qui s’était passé …
PM: Vous pensez à quelles manifestations?
JM: En fait il n’y a pas … j’ai lu seulement qu’à partir de 72 il y avait de petites manifestations, pas
les grèves de faim.
PM: De harkis?
JM: Oui.
PM: Les petites manifestations de harkis, oui.
404
JM: Et votre gouvernement s’occupait de …
PM: Oui. Il y a eu, mais des manifestations, qui étaient pas des manifestations très importantes, qui
étaient… ces manifestations sont provoquées par le fait que je vous avais dit, les harkis n’arrivaient
pas à s’intégrer. Et alors, ça crée de plus en plus de tensions dans leur propre famille et avec leurs
enfants qui eux en souffrent beaucoup plus que leurs parents. C’est qu’aussi que des tensions avec le
voisinage. Il y a des camps qui se trouvent à proximité de petites villes ou gros villages, et les
relations sont pas bonnes. Et alors, tout ça crée en effet un mauvais climat et c’est ce qui s’explique
ces manifestations. Jusqu’en 1973-74, le gouvernement de 74 on veut pas dire que c’étaient des
grosses manifestations. C’étaient des manifestations de mauvaise humeur, souvent d’ailleurs justifiées,
souvent justifiées parce que par exemple nous n’avions pas fait tous que nous aurions dû faire pour
le logement. Disons après leur retour beaucoup était encore très mal logés dans des baraques. A
partir de ce moment-là un effort était fait, mais ces manifestations étaient très limitées en le nombre
de gens qui participait. Mais traduisait en état d’esprit, certainement.
JM: Mais ces manifestations, est-ce qu’elles ont provoqué une politique par votre gouvernement, ou
c’est après les grandes manifestations de 74?
PM: Moi, je crois beaucoup de ces manifestations était provoqué par l’action des associations
rapatriées. L’Association rapatriée d’Indochine, qui était uniquement une association française
d’ailleurs, était hostile au gouvernement, naturellement très hostile à de Gaulle, mais aussi très hostile
avec les hommes comme moi, bien entendu. Et elles ont trouvé que c’était qui ne s’occupait des
harkis, mais elles ont trouvé que c’était un terrain sur lequel on pouvait émouvoir l’opinion publique,
ce qui est vrai. Et alors, à partir de ce moment-là, les associations rapatriées ont commencé à faire
beaucoup de bruit autour des harkis. Je dis pas qu’ils avaient tort, mais pour elles c’était surtout
l’occasion de manifester leur opposition au gouvernement.
JM: Donc il n’y a pas un changement de politique …
PM: Mais, ils ont certainement suscité beaucoup de manifestations.
JM: Oui. Et ça n’a pas eu un effet sur votre politique envers les rapatriés?
PM: Non, non.
JM: C’était après les grandes manifestations?
PM: Non, sur ce point-là, moi j’ai jamais changé de politique. L’évolution commence après moi
quand M. Giscard d’Estaing est élu président de la République. Pour des raisons électorales, dans sa
campagne électorale, il avait recherché les suffrages des rapatriés. Alors ce qui lui était facile puisqu’il
se présentait comme un homme qui était relativement éloigné du général de Gaulle, même si ça
n’était pas tout à fait vrai. Dans sa campagne électorale c’est ce qu’il a voulu faire. Et, par conséquent,
il a été amené à prendre des engagements vis-à-vis d’associations de rapatriés. Et ensuite pendant
qu’il était Président de la République, il leur a accordé des indemnités.
405
JM: En 91 qui étaient la prochaine fois qu’il y a eu de grande manifestations, qu’est-ce que vous avez
pensé a provoqué le changement parce qu’après cela il y avait une commission qui a été créée pour
traiter cette question, c’était pour les raisons politiques ou…
PM: Oui, vous avez raison. Ce sont ces manifestations qui ont amené le gouvernement d’époque à
créer cette commission et à, disons, accorder un certain nombre de montants supplémentaires. C’est
la manifestation qui a causé cela, cette évolution, c’est vrai.
JM: Et donc, sans des manifestations il n’y a pas vraiment un changement politique depuis les 40
ans…
PM: Non.
JM: Et à votre avis, pourquoi est-ce que le gouvernement français a décidé en 2001 de reconnaître
officiellement les harkis et leur service à la France pendant la guerre? La reconnaissance de 2001.
PM: Ah, oui. Oui, cette reconnaissance qui est, à mon avis, très justifiée. Cette reconnaissance
correspond surtout à une manifestation je dirais morale beaucoup plus qu’à une réalité parce que
cette reconnaissance ne correspond pas aux faits qui se sont passés. Mais, je crois que c’est une
façon de leur rendre, de leur rendre… je dirais de leur rendre leur honneur. Et c’est d’autant plus
utile que le gouvernement algérien qui est vis-à-vis de ces hommes par des hommes du FLN le
gouvernement algérien continue de les traiter comme je vous ai dit. Et pour répondre à ce
comportement le gouvernement algérien c’était très bien que la France manifeste publiquement que
les harkis avaient été des hommes qui avaient rendu en Algérie de bon service à l’armée française et à
la France. Et c’est surtout important de point de vue moral parce que pour les hommes et je dirais
pour ces harkis, ce qui survivent encore, pour leurs enfants aussi, l’honneur ça signifie quelque chose.
Ce que l’on peut dire simplement c’est que la guerre d’Algérie et ses suites créent encore des
débats passionnés en France. Ce n’est pas un évènement historique que l’on regarde avec une
certaine distance comme par exemple la guerre d’Indochine. Maintenant la guerre d’Indochine est
considérée avec je dirais un grand intérêt, mais sans véritable passion. On l’a vu à l’occasion de la
célébration du cinquantième anniversaire de Diên Biên Phu qui était très largement commémoré en
France avec beaucoup de dignité d’ailleurs, mais sans passion. Au contraire, je dirais avec une
ouverture vers le Vietnam, une ouverture naturelle pas artificielle de tout. La guerre d’Algérie, on
n’est pas là.
JM: Mais surtout la politique officielle du gouvernement parce que Chirac a pris la décision de la
reconnaître…
PM: Je dirais que dans la politique des gouvernements français il y a une volonté de dépasser la
guerre d’Algérie. Mais, moi, je vous parlais de l’opinion publique. L’opinion publique française est
encore sensible à la guerre d’Algérie, même si elle ne l’est plus à la guerre d’Indochine.
JM: Benjamin Stora a rédigé un livre en mars qui s’appelle La guerre d’Algérie: Fin de l’amnésie. Est-ce
que vous croyez qu’il y ait une fin de l’amnésie à l’égard de la guerre d’Algérie.
PM: Je crois pas encore. Pas encore. Je vous ai dit oui pour la guerre d’Indochine, mais je dis non
pour la guerre d’Algérie.
406
APPENDIX B
“Annexes relatant les principales exactions connues”
SHAT 1K 744 (Dossier: Situation des harkis en Algérie)
Michel de Brébisson, Général de Corps d’Armée, Commandant Supérieur des FAFA
A
M. le Ministre des Armées (Cabinet)
Le 17 août 1962
Objet: Situation des ex-harkis en Algérie depuis le cessez-le-feu
ANNEXE 1
22˚ CA Philippeville
Date
16 juillet
Lieu
Guelma
21 juillet
22 juillet
23 juillet
Mondovi
Medjez-SFA
(commune
de Duvivier)
Morsott
23 juillet
Mondovi
27 juillet
Sedrata
27 juillet
Mansourah
28 juillet
Cherka
Nature de l’exaction
-Manifestation de la population musulmane contre les ex-supplétifs.
-8 ex-harkis blessés – hospitalisés.
-Massacre de 4 ex-harkis par la population encadrée par l’ALN.
-Les membres de l’ALN se saisissent de ex-harkis et leur font subir
diverses tortures sur la place publique.
-Des ex-harkis sont arrêtés par l’ALN.
-Des sévices sont exerces a leur encontre. Enchaines et exposés au
soleil, ils sont astreints a des travaux pénibles (creusement de tranchées).
-4 ex-harkis et deux femmes sont lynchés a mort lors d’une
manifestation populaire.
-Les locaux de l’ancienne brigade de Gendarmerie sont transformés en
prison. Les musulmans ayant servi la Fr y sont incarcérés. Trois fois par
jour ils sont soumis à des sévices. Attachés aux arbres de la cour de la
caserne ils sont livrés à la foule qui les tourmente à son gré. Une femme
musulmane également a été soumise à ce traitement. Elle a été libérée
méconnaissable, le visage ensanglanté et tuméfié par les coups.
-Un ancien militaire de la police de Sedrata, connu sous le nom de xxx,
est attelé avec le harnais d’un âne à un tonneau de 200 litres monté sur
roues, il assure ainsi une corvée d’eau sous les quolibets et les coups de
la population.
-Dans les locaux de l’ancienne SAS occupée par l’ALN sont internés 25
musulmans ayant travaillé avec la France. Ils sont soumis à des
interrogatoires et à des sévices. Certains auraient été exécutés après un
jugement sommaire.
-Cinq ex-supplétifs sont enlevés. Les corps de deux sont retrouvés.
-Dans la nuit du 22 au 28 cinq autres ex-supplétifs se sont refugies
auprès de l’autorité militaire de OUEL EL AMEL. Molestés par la
population, ils portent sur le visage de nombreuses ecchymoses.
407
Fin juillet
Biskra
2 aout
4 aout
Biskra
Lambèse
6 aout
Bordj’r’dir
-Au cours d’une déclaration à la population le Colonel CHABANI Chef
de la Willaya VI déclare que tous les anciens harkis, GMS et autres
supplétifs de l’Armée Française, traitres à l’ALN, étaient condamnés a
mort.
-Cinq ex-harkis sont enlevés entre le 1er juillet et le 2 aout.
-Environ 70 prisonniers ex-harkis, moghaznis et civils musulmans ayant
coopéré avec la France sont actuellement internés à la Centrale de
Lambèse et gardés par des militaires de l’ALN. Ils sont maltraités et
astreints à des travaux de force. Leur état de santé parait déficient et ils
semblent sous-alimentés. Certains portent des traces de sévices.
-l’ALN assassine un sergent en retraite répondant au nom de xxx. Le
corps mutilé est exposé à quelques centaines de mètres du village.
ANNEXE 2
23˚ CA Alger
Date
Lieu
11 juillet
Aumale
11 juillet
13 juillet
17 juillet
18 juillet
21 juillet
22 juillet
23 juillet
23 juillet
23 juillet
23 juillet
Nature de l’exaction
-Enlèvement d’un Moghazni, son épouse et ses deux enfants. Relâché
le 12, l’épouse a déclaré que son mari avait été exécuté.
Trolard-Taza -Un ex-harki échappé à une exécution en simulant la mort. Il rapporte
que 6 de ces camarades ont été tués lors de cette exécution.
Tablat
-Un tirailleur emprisonné par l’ALN à Tablat s’évade le 13 juillet après
avoir subi des sévices. Il déclare avoir vu 3 ex-harkis abattus et 2
tirailleurs du 5˚ RT torturés.
Kerba
-Un ancien harki enlevé par l’ALN s’enfuit et se réfugie auprès de
l’Armée Française. Il rend compte de l’assassinat par l’ALN de plusieurs
ex-harkis.
Lodi
-Deux anciens harkis demandent protection à la CCS du 6˚ RI. L’un
deux est hospitalisé à la suite de sévices dont il a été l’objet de la part de
l’ALN.
Loverdo
-Arrestations par l’ALN de 15 anciens harkis.
Michelet
-Le 22 juillet 62, jour de marché dans les Illilten les combattants de
l’ALN auraient exécuté plusieurs personnes.
-Les nommés xxx, originaires de Zoubga, commune de Illilten, excaporal Chef GMS 78 et xxx ex-Moghazni à la SAS de Iferounene
auraient été sauvagement torturés et tués a coups de hache devant toute
la population contrainte d’assister à ces exécutions.
-D’autres exécutions aussi sauvages auraient eu lieu dans la région de
Tamazirt.
Oued Aissi
-2 anciens Moghaznis demandent protection aux Forces Armées
Françaises. Le premier s’est évadé de Moukda où il devait être égorgé.
Le second s’est évadé de l’ancienne SAS de Grand Remblais où se
trouveraient une vingtaine d’autres anciens supplétifs.
El Esnam
- Enlèvement de cinq musulmans, ex-harkis.
Bouira
- Enlèvement de quatre ex-harkis.
Zurich
-Défilé de la population pour narguer et insulter 10 ex-harkis ou
membres des groupes d’auto-défense condamnés aux travaux forcés.
408
24 juillet
Nord Dra el
Mizan
Nuit du
24 au 25
juillet
25 juillet
25 juillet
Desaix
25 juillet
26 juillet
26 juillet
27 juillet
28 juillet
30 juillet
31 juillet
Fin juillet
-Dans la région de Bou Marni (Nord de Dra el Mizan) l’ALN arrête tous
les anciens harkis, moghaznis et musulmans des auto-défenses.
(Ce groupe est estimé à 200 ou 250 personnes).
Ils seraient astreints à des travaux de terrassement et maltraités par leurs
gardiens.
Après le travail, ils seraient présentés à la population qui se livre alors à
toutes sortes de sévices à leur égard.
-Trois ex-moghaznis de l’ancienne SAS sont égorgés.
Meurad
-Arrestation de 30 ex-moghaznis.
Dra el Mizan -L’ajutant Si Mohamed, responsable de l’ALN des zones autonomes de
Dra el Mizan, Tizi Reniff et Bogni confirme les enlèvements de tous les
ex-harkis de Bou-Mahni (une trentaine environ).
(Renseignement obtenu par le Commandant de la Brigade de
Gendarmerie de Dra el Mizan au cours d’une liaison de Si Mohamed.)
Palestro
-25 ex-harkis sur 40 détenus depuis vingt jours au poste de Sidi
Abdelkader sont égorgés et enterrés dans le lit de l’Oued Borja, après
avoir été torturés.
Palestro
-Un ex-harki est enlevé par l’ALN et interné dans un camp (ancienne
SAS de Khal Kal) où se trouvent déjà une trentaine d’anciens harkis.
Après s’être évadé et réfugié auprès de l’Armée Française il rapporte que
les détenus de ce camp sont soumis à des sévices et que certains d’entre
eux sont exécutés d’une manière atroce.
Khalkal
-Le Sergent Amrane, de l’ancienne commande de chasse de Palestro est
(Palestro)
égorgé, puis son corps est découpé en morceaux.
-30 anciens harkis, après avoir été torturés, sont abattus au pistoletmitrailleur.
Rivet
-Un ancien sous-officier du 1/117˚ RI enlevé par l’ALN réussit à
s’évader et rejoint son ancien corps. Il rapporte l’existence d’un camp.
d’internement dans le douar Zaiane où les anciens harkis sont soumis à
des travaux pénibles, subissent des mauvais traitements et où ont lieu
des exécutions.
Lodi
-3 ex-harkis sont arrêtés, conduits sur un piton (ancien poste de
Dragons) et exécutés sommairement au pistolet-mitrailleur.
-10 autres supplétifs ont été fusillés au même endroit et les corps laissés
sur place.
Borely-la-10 anciens harkis arrosés d’essence sont brûlés vifs.
Sapie
Beni-19 ex-harkis appartenant pour la plupart au 2/10˚ R.A.M. sont égorgés
Boudouane
et donnés en pâture aux chacals dans la région de Béni-Boudouane et
Betaya
Betaya.
Teniet el
-Exécution de 20 ex-harkis.
Hadd
409
ANNEXE 3
24˚ CA Oran
Date
Lieu
16 juillet
Bouguirat
16 juillet
Lamorcière
16 juillet
Abdellys
18 juillet
Région AinTemouchent
18 juillet
Lamorcière
22 juillet
Sebdou
24 juillet
Lamorcière
2 aout
4 aout
Deriec
Raghou (10
km sudouest de
Beni Saf)
Nature de l’exaction
-Un ex-harki est égorgé.
-Disparition d’une centaine d’anciens harkis.
-Cinquante huit anciens harkis sont arrêtés par l’ALN
-Environ 200 anciens militaires musulmans ayant servi dans l’Armée Fr
sont enlevés par l’ALN
-Une centaine d’ex-harkis sont enlevés par l’ALN
-Cinquante musulmans ex-supplétifs disparaissent.
-Cinq musulmans ex-harkis ont été exécutés par l’ALN au cours des
dernières 48 heures.
-Trente ex-harkis sont arrêtés et incarcérés par l’ALN.
-Une vingtaine de prisonniers, en majorité ex-harkis, sont détenus par
l’ALN.
410
APPENDIX C
Audition du nommé AA
SHAT, 1H 1793/1
CSFA Française en Algérie
23e Corps d’Armée
Etat-Major 2º Bureau
Cap Matifou
le 22 aout 1962
Nº 2246/23ºCA/B2
Transmis à M. le Général de Corps d’Armée, Commandant Supérieure des Forces Armée Française
en Algérie – EMI-2º Bureau
A titre de compte-rendu
Par Le Général de Brigade LEVE Commandant Pvt le 23º Corps d’Armée
****
A S.P. 87.193 le 16.8.62
20º Division d’Infanterie Groupement E
Nº 1230/GE/2/S
Audition du nommé AA,1 venu se refugier le 13 aout 1962 auprès des Forces d’Armée à Lodi
1. Renseignements d’identité :
AA
Né en 1931 aux Ouled Messaoud
Demeurant à Lodi
Ex-harki à la harka du 2/6º RI durant trois ans
2. Déclarations de l’intéressé
Le 12 aout, vers 21 heures, veille du Mouloud, j’ai été accoste dans une rue de Lodi, à
proximité de la Mairie, par trois membres de l’ALN armés qui m’ont intimé l’ordre de les suivre
jusqu’au bureau FLN de la localité. J’ai été conduit dans une maison à plusieurs pièces que les
militaires français utilisaient auparavant comme magasin à vivre et à habillement. Là, j’ai été enfermé
dans les W.C. Apres être resté une demi-heure à cet endroit, mon neveu, MH, ex-harki de la 5º Cie
du 2/6º RI, y a été également amené. Pour ma part, j’ai tenté de m’évader de cette cellule improvisée
mais la lucarne était trop étroite pour réaliser mon projet. Le lendemain, vers 7 h, nous avons été
amenés tous deux dans un garage contigu à la dite maison, ou se traouvaient (sic) déjà deux autres
anciens harkis de Lodi : BB et un individu dont je ne connais pas le nom avec exactitude. Tous
quatre, nous avons subi à tour de rôle un interrogatoire et des supplices qui n’ont cessé que vers 11
heures 30. Ce genre d’enquête a été mené par un responsable F.L.N. des renseignements, ME, par
1
All names of harkis and FLN members are replaced with their initials.
411
un chef militaire se faisant appeler B et par 6 ou 7 djounouds ou policiers. Nous avons été
copieusement battus a coups de bâton et de manche de pioche et ns avons endure le “supplice d’eau”
qui consistait à nous tenir la bouche ouverte à l’aide d’un bâton et à nous gonfler d’eau au moyen
d’un tuyau d’arrosage jusqu’à perdre la haleine. L’interrogatoire se résumait à des questions se
rapportant à notre situation de famille et à nos activités passées en tant que harkis:
Pourquoi n’as-tu pas répondu à notre appel lorsque nous t’avons demandé de nous fournir
des armes et des munitions et pourquoi n’as-tu pas déserté?
-Avoue-nous combien tu as tué de moudjahidines et de civils!
-Tu as participé à la démolition de mechtas; à combien ?
-Combien de femmes as-tu violées ?
A la question: “Combien d’enfants as-tu?” il nous était répliqué dès la réponse donnée: “Eh bien, ce
soir ce seront des orphelins, mais avant, il faut que vous mourriez a ‘petit feu.’”
Le harki dont je ne connais pas le nom a reconnu, sous la torture, avoir tué 6 hommes et une femme
lors des manifestations du 1º Novembre a MEDEA. Il a été emmené vers 11 heures peut-être, je
suppose exécuté. Nous ne sommes donc restés que trois prisonniers dans le garage.
Vers 11 heures 30, après le départ des deux chefs ALN charges de l’enquête, les policiers nous ont à
nouveau battus puis nous ont lié les poignets et les pieds avec du fil électrique. Un bon moment
passé, nos gardiens se sont endormis, pendant que nous ne pouvions leur échapper dans l’état ou
nous nous trouvions.
Surmontant mon accablement, j’ai alors remarqué la présence d’un couteau sur une petite table de
bois se trouvant près de moi. En me soulevant, j’ai réussi à saisir le couteau avec les dents et à le
faire glisser doucement à terre. Puis, tournant légèrement sur moi-même j’ai pu le prendre dans une
main et trancher les liens qui emprisonnaient mes poignets et mes pieds. La porte du garage étant
restée entrebâillée, j’ai fui rapidement à travers les rues, désertés a cette heure de l’après-midi, et je
me suis refugié auprès des militaires français.
J’ai hésité à délivrer mes compagnons d’infortune car je craignais, d’une part, qu’ils ne soient pas, par
peur, disposés à me suivre et, d’autre part, que l’un des gardiens se réveille.
Note B 2 GROUPEMENT :
L’action dite « d’épuration » entreprise par l’A.L.N. à l’encontre des anciens serviteurs des
FO s’est généralisée depuis le 1º Juillet et de poursuit.
Le seul reproche « impardonnable » qui leur est adressé tient dans le fait qu’ils ont servi la
France et ont combattu « leurs frères musulmans. »
Les photographies et les photocopies du certificat médical, jointes en annexe, se passent de
commentaires.
L’intervention tentée par les Forces Armées pour récupérer les deux autres prisonniers s’est
révélée négative. Les ravisseurs ont dû se rendre compte rapidement de l’évasion [de le harki] et ont
fait disparaître immédiatement les intéressés.
Signé: Le Colonel DE NADAILLAC Commandant le Groupement E
Destinatires : -M. Le Gal Cdt la 20º DI (EM 2B)
-Cdt 1/6º R.I.
412
APPENDIX D
List of Forest Hamlets in France
Department Name of
Forest Hamlet
Number of Number
worksites
of houses
Type of
houses
Date
Opened
Date
Closed
Allier
Noyant-St-Hilaire
1
30
11/6/63
1973
AlpesMaritimes
AlpesMaritimes
AlpesMaritimes
AlpesMaritimes
AlpesMaritimes
Ariège
Breil-sur-Roya
(L’Oliveraie)**
L’Escarène**
1
28
Existing houses
formerly used
for Indochinese
Repatriates
Prefabricated
1
28
Prefabricated
3/9/64
1/79
MouansSartoux**
Roquesteron**
1.5
40
Prefabricated
2/22/64
1980
1.5
38
Prefabricated
2/1/64
1971
Valbonne**
2
28
Prefabricated
Montoulieu
1
33
5/20/64
Aude
Aude
Narbonne
La Pradelle*
1
1
31
Aude
St-Martin des
Puits*
VilleneuveMinervois
(Pujol-de-Bosc)*
St-Rome de
Cernon*
Brusque*
Jausiers*
1
28
28 Prefabricated
5 Restored
SONACOTRA
12 Prefabricated
19 Restored
Prefabricated
2
52
2 Prefabricated
50 Restored
1
25
Prefabricated
1
1
25
32
Prefabricated
Prefabricated
Ongles
(St-Etienne des
Orgues)*
St-André des
Alpes*
Sisteron*
Fuveau
(La Cité
Brogilum)**
Jouques
(Le Logis
d’Anne)**
La Ciotat
1
32
Prefabricated
1/17/63
and
2/27/63
12/17/19
62
1/4/1963
10/6/196
2
9/7/62
1
32
Prefabricated
10/11/62
1973
1
2
28
52
Prefabricated
Prefabricated
11/1/63
4/15/64
1973
Still
open
1
28
SONACOTRA
10/1/63
Prefabricated
1/67
Aude
Aveyron
Aveyron
Basses-Alpes
Basses-Alpes
Basses-Alpes
Basses-Alpes
Bouches-duRhône
Bouches-duRhône
Bouches-duRhône
2
413
12/7/63
9/8/62
12/10/63
6/30/66
6/30/65
9/65
2/15/65
(1980)
Bouches-duRhône
La Roque
d’Anthéron
(La Baume)**
Chalvignac
(Aynes)*
La Tremblade*
2
52
Prefabricated
5/30/64
2/1/76
1
25
Prefabricated
7/1/63
1973
1
25
2/6/63
12/31/6
4
Casamozza
Zonza
1
2
28
50
23 Prefabricated
2 Restored
Prefabricated
Prefabricated
1/16/64
8/15/71
1981
1
28
Prefabricated
12/63
1971
1
28
Prefabricated
1
25
Prefabricated
2/27/63
1
1
1
1
3
25
30
Prefabricated
Prefabricated
Prefabricated
Restored
Restored
3/9/63
9/20/62
10/1/65
10/1/64
Gers
Baigneux-lesJuifs**
Is-sur-Tille
(Vernot-Saussy)**
Vanvey-surOurce*
Lanmary*
Beaurières*
Die
La Grand’Combe
St. Sauveur du
Pourcil
(Villemagne)*
Mirande*
1
27
1973
11/15/62
and
3/30/63
9/9/62
1971
Hautes-Alpes
Montmorin*
1
27
5/8/63
7/8/65
Hautes-Alpes
HauteGaronne
Haute-Savoie
Hérault
Hérault
Rosans**
Juzet-d’Izaut*
1
1
28
25
2/27/64
2/15/63
2/24/65
Magland*
Avènes-Truscas*
Lodève*
1
1
2
24
27
60
Hérault
Saint-Pons
(Plo de Maillac)*
3
53 +
Prefabricated
Prefabricated
Prefabricated
(SONACOTRA
)
Prefabricated
Isère
Roybon*
2
53
Prefabricated
Lozère
Lozère
Cassagnas*
Chadenet
(La Loubière)*
Chanac-Cultures*
Mende
Meyrueis*
St-Etienne de
Valdonez*
Villefort*
1
1
28
27
1
1
1
1
2
Cantal
CharenteMaritime
Corse (Haute)
Corse (du
Sud)
Côte d’Or
Côte d’Or
Côte d’Or
Dordogne
Drôme
Drôme
Gard
Gard
Lozère
Lozère
Lozère
Lozère
Lozère
25
78
23 Prefabricated
4 Restored
25 Prefabricated
2 Restored
Prefabricated
Restored
1971
8/11/63
5/14/63
8/4/64
1970
at least
72
Prefabricated
Prefabricated
5/4/63
and
6/10/63
12/15/62
and
12/11/63
9/26/63
3/23/63
29
40
28
28
Prefabricated
Prefabricated
Prefabricated
Prefabricated
4/23/63
1967
3/19/63
3/2/63
1967
56
Prefabricated
7/6/63
414
1972
1967
3/1/65
11/65
10/1/65
PyrénéesOrientales
Saône-etLoire
Rivesaltes*
1
25
Prefabricated
12/13/62
1
28
Prefabricated
7/12/63
1
1
28
28
Prefabricated
Prefabricated
7/29/64
1/8/64
1971
1971
1
28
Prefabricated
5/1/63
1971
1
1
28
29
Prefabricated
Prefabricated
10/14/64
2/26/63
9/1/79
Prefabricated
1970
Var
Roussillon-enMorvan
(Glennes)*
Anglès**
Arfons
(Les Escudiers)**
Puycelci
(Grésigne)*
Vaour
Bormes
(Les Mimosas)*
Collobrières
(La Capelle)
Collobrières
(Le Capelude)*
Gonfaron**
Var
Var
Var
Var
Var
Var
Var
Tarn
Tarn
Tarn
Tarn
Var
Var
Var
Var
Var
Vaucluse
Vaucluse
Vaucluse
Vaucluse
Total: 26
1
1976
1
29
Prefabricated
2/14/63
1970
2
52
Prefabricated
12/17/63
10/15/7
9
La Londe**
Montmeyan**
Le Muy**
2
1
2
Prefabricated
Prefabricated
Prefabricated
12/21/63
8/5/64
8/22/64
Néoules**
Pignans*
Rians**
St-Maximin
(La Sainte
Baume)**
St-Paul-enForêt**
St-Raphaël
(Aigue-Bonne)*
Apt**
Cucuron**
Pertuis**
Sault**
Total: 72
1
1
1
1
56
28
28 / 2nd
opened in
66 or 67
28
29
28
28
Prefabricated
Prefabricated
Prefabricated
Prefabricated
7/4/64
3/21/63
11/25/63
9/25/63
1
28
Prefabricated
9/7/64
1
29
Prefabricated
3/2/63
1
1
1
1
Total: 89
28
28
28
28
Prefabricated
Prefabricated
Prefabricated
Prefabricated
11/28/63
11/28/63
11/23/63
11/25/63
9/1/79
(FIN 79)
late 1973
8/1/79
6/30/79
late 1973
12/1/66
*First phase forestry worksites.
**Second phase forestry worksites
If the name of the forest hamlet differs from the name of a town, this is given in parentheses.
The information in this table was compiled from documents located in:
ACNMF, 3/2; ADV 746 W 68; CAC, 19910097/40 (dossier: Harkis – Correspondance 1963); CAC 19910097/40
(dossier: Harkis – Correspondance 1965); CAC, 19910097/41 (dossier: Harkis – Correspondance 1966); CAC,
19910097/41 (dossier: Harkis – Correspondance 1968); CAC, 19910097/42 (dossier: Harkis – Correspondance 1971);
CAC, 19910097/42 (dossier: Harkis – Correspondance 1972); CAC, 19910097/42 (dossier: Harkis – Correspondance
1974); CAC, 19910097/48; CAC 19920149/1/4; CAC 19920149/1/16; CAC, 19920149/2/8; and CAC, 19920149/3/1.
415
APPENDIX E
Population of Forest Hamlets
Date
Dec. 19621
Mar. 19632
May 19633
Oct. 19634
Nov. 19635
Feb. 19646
Jan. 19657
Jan. 19668
Jan. 19679
Jan. 196810
Jan. 196911
Jan. 197012
Jan. 197113
Jan. 197214
Jan. 197315
July 197316
Jan. 197417
Jan. 197518
Dec. 1975
Number of
forest
hamlets
10
20
24
40
----60
52
----------40
--36
31
31
29
Number of
forestry
worksites
11
22
28
--48
60
72
65
------45
--49
-----------
Number of
families
282
------------1,375
1,161
1,236
1,128
1,134
1,011
905
817
784
714
680
584
Number of
workers
--------1,392
----1,438
------1,230
--1,026
--846
766
823
---
Number of
dwellings
Number of
inhabitants
307
--733
1,054
----1,982
-------------------------
1,425
2,420
3,345
5,082
6,081
7,912
9,720
--7,546
7,519
7,344
7,463
6,795
6,343
5,997
5,877
5,391
5,275
4,645
CAC, 19920149/3/1, “Hameaux de forestage où l’installation des familles est effective à la date du 15 décembre 1962.”
SHAT, 1K 744 (dossier: Point de la Situation des Harkis, janvier et juin 1963), Minister of Armies Messmer, “Perspectives de
décongestionnement des Camps de Rivesaltes et de St-Maurice l’Ardoise,” Mar. 22, 1963.
3 CAC, 19910097/40 (dossier: Harkis – Correspondance 1963), Ministry of Agriculture, “Situation des chantiers ouverts au 15 mai
1963.”
4 CAC, 19920149/1/13, J. Besson (Ministry of Repatriates), “Note pour Monsieur Perony,” 3 and 5, Oct. 9, 1963.
5 AN, F1a 5142, SFIM, “Ex-supplétifs musulmans passés par les camps et reclassés par secteurs d’activité à la date du 15 novembre
1963.”
6 CNMF, 3/2, “Situation des Rapatriés musulmans au 1er février 1964,” 1, Feb. 10, 1964.
7 For the number of forest hamlets and forestry worksites, see CAC, 19910149/1/4, “Programme de construction pour les ‘Harkis: II.
Chantiers de forestage,’” Jan. 1, 1965. For the number of inhabitants, see CAC 19920149/1/5, “Logement des ex-supplétifs
musulmans rapatriés.”
8 CAC, 19910097/40 (dossier: Harkis – Correspondance 1965), “Liste des hameaux de forestage ouverts le 1er janvier 1966.”
9 CAC 19910097/42 (dossier: Harkis – Correspondance 1974), Sous-Direction des Programmes Sociaux en faveur des Migrants,
“Hameaux forestiers: Situation au 1er janvier 1974,” 2.
10 Ibid.
11 Ibid.
12 Ibid for the number of families and number of inhabitants. CAC, 19910097/42 (dossier: Harkis – Correspondance 1971), M.
Bergogne (L’Ingénieur Général du Génie Rural, des Eaux et des Forêts de Nice) à Monsieur le Chef du Service des Forêts, “Objet:
Repliement des harkis vers la région méditerranéenne,” 3, June 29, 1971 for the number of worksites and workers.
13 CAC 19910097/42 (dossier: Harkis – Correspondance 1974), Sous-Direction des Programmes Sociaux en faveur des Migrants,
“Hameaux forestiers: Situation au 1er janvier 1974,” 2.
14 Ibid for the number of families and number of inhabitants. CAC, 19910097/42 (dossier: Harkis – Correspondance 1972),
“Chantiers de travailleurs ex-harkis. Effectifs au 31.1.72” for the number of forest hamlets, worksites, and workers.
15 CAC 19910097/42 (dossier: Harkis – Correspondance 1974), Sous-Direction des Programmes Sociaux en faveur des Migrants,
“Hameaux forestiers: Situation au 1er janvier 1974,” 2.
16 CAC, 19910097/42 (dossier: Harkis – Correspondance 1973), Sous-Direction des Programmes Sociaux en faveur des Migrants,
“Note concernant les Français Musulmans Rapatriés,” 4-5, Oct. 1, 1973.
17 CAC 19910097/42 (dossier: Harkis – Correspondance 1974), Sous-Direction des Programmes Sociaux en faveur des Migrants,
“Hameaux forestiers: Situation au 1er janvier 1974,” 1.
18 CAC, 19870444/15, “Situation des hameaux forestiers,” May 1, 1979.
1
2
416
APPENDIX F
“Règlement intérieur concernant l’hébergement dans les hameaux forestiers”
Source: CAC, 19910097/41 (dossier: Harkis – Correspondance 1969).
417
BIBLIOGRAPHY
ARCHIVAL SOURCES
Archives Communales de Fuveau (ACF)
Documents relating to the harki population, not classified in any order, were consulted with the
special permission of Mayor Jean Bonfillon. Since there are no carton numbers, I just list the
bibliographic information for each document that I cite.
Régistres des délibérations du Conseil Municipal, janvier 1963-décembre 2007.
Archives Communales de la Roque d’Anthéron (ACRA)
Documents were consulted with the special permission of Mayor Jean-Louis Turcan.
Régistres des délibérations du Conseil Municipal, janvier 1963-décembre 2007.
Archives de la Société Nouvelle d’HLM de Marseille (ASANHLMM)
Cartons were consulted with the special permission of the Director General of the SANHLMM,
Frédéric Bultel.
54/1, 54/2
Archives de l’Office National des Forêts, Service Départemental des Bouches-du-Rhône, Aix-enProvence (AONFBR)
Cartons and dossiers were consulted with the special permission of the departmental director of the
ONF.
Harkis salaires 1964-1976
Principaux textes concernant les ouvriers FSIRAN
Personnel files for 15 harkis who worked on the Fuveau forestry worksites
Archives Départementales des Bouches-du-Rhône, Aix-en-Provence (ADBR-Aix)
Préfecture/Sous-préfecture d’Aix-en-Provence
98 W 325
Archives Départementales des Bouches-du-Rhône, Marseille (ADBR)
Cartons marked with an asterisk were consulted with the special permission of the director of the
Archives de France.
Conseil général/Documentation
1488 W 30
Fonds Albert Payan (chef de sécurité du port de Marseille)
22 J 9
418
Inspection académique
*131 W 377
Préfecture
12 O 2357
Préfecture/Affaires musulmanes
*138 W 3, *138 W 4, *138 W 5, *138 W 6, *138 W 7, *138 W 8, *138 W 9
Préfecture/Cabinet
135 W 51*, 135 W 398*, *137 W 3, *137 W 4, *137 W 426, *137 W 427, *137 W 460,
*137 W 543, 137 W 661, *1476 W 10, *1513 W 18, *1693 W 96, *1693 W 232
Préfecture/Droits des femmes
*1451 W 115
Préfecture/Protection civile
125 W 173
Préfecture/Suivi des collectivités et des organismes locaux
*1692 W 1-2
Archives Départementales des Pyrénées-Orientales, Perpignan (ADPO)
Cartons marked with an asterisk were consulted with the special permission of the director of the
Archives de France.
Périodiques
1111 Per 73 (L’Indépendant)
Préfecture Cabinet
*104 W 6, *104 W 24, *1419 W 109
Préfecture Cabinet, Service des Rapatriés
*1789 W 7, 1789 W 8, 1789 W 9
Archives Départementales du Var, Draguignan (ADV)
Direction Départementale de l’Agriculture Draguignan
746 W 62, 746 W 63, 746 W 64, 746 W 67, 746 W 68, 937 W 75
Direction de la Réglementation Préfecture du Var
1160 W 51
419
Archives du Comité National pour les Musulmans Français, Paris (ACNMF)
All cartons were consulted in May-June 2008 at the Association Génériques (34 rue des Citeaux,
75012 Paris) with the special permission of Mr. Marcel Wormser. The carton numbers below
correspond to the cataloguing completed by the Association Génériques archivist.
3/2, 3/7, 3/11, 4/3, 4/4, 5/7, 6, 11/8, 15/13, 16, 19, 20, 25/15, 25/17, 28, 32/1, 32/2,
32/3
Archives du Ministère des Affaires Etrangères, Paris (MAE)
Secrétariat d’Etat aux Affaires Algériennes
97, 119, 125, 126, 137, 152, 155, 162
Archives du Service Départemental de l’Office national des Anciens Combattants et Victimes de
Guerre des Bouches-du-Rhône, Marseille (AONACBR)
Dossiers consulted with the special permission of the director general of ONAC.
Individual dossiers for thirty-eight harkis residing in Fuveau and Marseille
Archives du Service des Rapatriés des Bouches-du-Rhône, Marseille (ASRBR)
Cartons and dossiers consulted at the Marseille Prefecture with the special permission of the Subprefect of the Bouches-du-Rhône department.
Actions en faveur des RONA, Commission Département 89-90
Rapatriés d’Origine Nord Africaine: Cité les Tilleuls Marseille
Social welfare files for fifty-two harkis residing in Fuveau and Marseille
Archives Nationales, Paris (AN)
All cartons were consulted with the special permission of the director of the Archives de France.
5 AG 1 Series: Papiers des Chefs d’Etat, Charles de Gaulle
5 AG 1/22
5 AG 2 Series: Papiers des Chefs d’Etat, Georges Pompidou
5 AG 2/102, 5 AG 2/1157
F1a Series: Le Service des Affaires musulmanes et de l’Action Sociale (SAMAS) devenu le Service de
Liaison et Promotion des Migrants (SLPM)
F1a 5017, F1a 5053, F1a 5059, F1a 5125, F1a 5129, F1a 5137, F1a 5138, F1a 5139, F1a 5140,
F1a 5141, F1a 5142
Centre des Archives Contemporaines, Fontainebleau (CAC)
Cartons marked with an asterisk were consulted with the special permission of the director of the
Archives de France. Cartons itemized in the CAC archival inventories under two ministries are
listed here under the first alphabetical appearance.
Ministère des Affaires Sociales
19870256/7, *19920149/1, *19920149/2, 19920149/3, 19920149/4, 19920149/5
420
Ministère de l’Agriculture
*19889376/7, 19910097/40, 19910097/41, 19910097/42, 19910097/43, 19910097/47,
19910097/48
Ministère des Anciens Combattants
19980331/1, 19980331/2, 19980331/4, 19980331/6, 19980331/11, 19980331/12
Ministère de l’Education Nationale
19770641/11
Ministère de l’Intérieur
19770097/35, 19880077/20, *19910467/1, *19910467/2
Premier Ministre
*19910281/3, *19960121/30, *19980027/1, *19980027/2, *19980027/3, *19980027/4
Ministère de la Santé
*19870444/15
Services Communs à la Santé et au Travail
*19970146/1, *19970146/2, *19970146/3, *19970146/4, *19970146/5, *19970146/6
Ministère du Travail
19770391/3, 19770391/8, 19770391/9, 19770391/10
Service Historique de l’Armée de Terre, Vincennes (SHAT)
Cartons marked with an asterisk were consulted with the special permission of the Service
Historique de la Défense (formerly SHAT). The dossiers in Series 1K 744 were consulted with the
special permission of former Minister of Armies and Prime Minister Pierre Messmer.
1H Series: Archives de l’Algérie, 1945-1967
1H 1170, 1H 1260/1, 1H 1260/2, *1H 1260/3, *1H 1260/4, 1H 1393/3, 1H 1393/4, 1H
1397/1, *1H 1397/7, 1H 1397/8, *1H 1397/9, *1H 1793/1, *1H 1793/2, 1H 1794/2, *1H
2028/5, 1H 2090/3, 1H 2090/4, 1H 2467/3bis, 1H 2467/6, *1H 2745/1, 1H 2799/5, 1H
2799/6, 1H 2801/3, *1H 2964/8, *1H 2982/1, 1H 2996/3, 1H 2996/8, 1H 3017/4, *1H
3077/2, *1H 3090/2, *1H 3155/5, *1H 3229/6, *1H 3517/3, *1H 3616/1, *1H 3616/3,
*1H 4194/1, *1H 4214/1, *1H 4400/3, 1H 4769/8
1K 744 Series: Fonds privé de Pierre Messmer
Dossiers: Accueil à St. Maurice l’Ardoise, Rivesaltes et Bias, Conseils interministériels,
Directives concernant les harkis, Point de la situation des harkis (janvier et juin 1963),
Premières mesures d’accueil en France (Larzac et Bourg-Lastic), Situation des harkis en
Algérie, Suggestions en vue du “recasement” des Harkis
R Series: Cabinet du Ministre de la Défense et organismes rattachés, 1945-1969
*1R 41/2, *1R 336/6, *1R 336/8, *1R 337/1, *1R 366/16, *1R 367/7, *3R 79/7, *9R 450/6,
*9R 450/7, *21R 179/2, *23R 16/2
421
T Series: Etat-Major de l’Armée de Terre et organismes rattachés
7T 249/1, 7T 249/6, 7T 253/1, 7T 253/3, 7T 253/4, 7T 253/5, 7T 257/1, *10T 549/3,
*10T 561/3, 13T 286, 14T 91/3, 14T 91/4, 14T 91/5, 14T 92, 19T 257/2
LIBRARY COLLECTIONS
La Bibliothèque de Sciences Politiques, Paris
Dossiers de Presse (SPDP) “Harkis,” 1962-2005
Aspects de la France
L’Aurore
Le Canard Enchaîné
Carrefour
Combat
La Croix
Les Echos
L’Evènement du Jeudi
L’Express
Le Figaro
Figaro Magazine
France Observateur
L’Humanité
International Herald Tribune
La Lettre de la Nation
Libération
Lutte Ouvrière
Le Matin (Suisse)
Le Matin de Paris
Le Monde
Nation
National Hebdo
New York Herald Tribune
Le Nouveau Journal
Le Nouvel Observateur
Paris Presse
Politique Hebdo
Politis
Le Populaire
Quotidien de Paris
Reforme
Rivarol
Rouge
Témoignage Chrétien
The Times (UK)
Tribune Socialiste
L’Unité
422
Journaux Officiels de la République Française (JORF)
Microfilm collection, 1961-2006
Journaux Officiels de la République Française Débats Parlementaires (JORFDP)
Microfilm collection, 1965-2005
La Bibliothèque Méjanes, Aix-en-Provence
Le Méridional, April 1964-September 2001
Le Provençal, April 1964-September 2001
REPORTS
Abi Samra, Marwan, and François-Jérôme Finas. “Regroupement et dispersion: Rélégation, reseaux
et territoires des Français-Musulmans.” Paris: Caisse nationale d’allocations familiales, 1985.
Conseil Economique et Social. “Problèmes posés par le rapatriement des réfugiés d’Algérie.”
Prepared by Robert de Vernejoul. Paris: JORF Avis et Rapports du CES, 1963.
Conseil Economique et Social. “La Situation sociale des enfants de Harkis.” Prepared by Hafida
Chabi. Paris: JORF Avis et Rapports du CES, 2007.
Diefenbacher, Michel (chargé de mission sur les rapatriés). “Parachever l’effort de solidarité
nationale envers les rapatriés, promouvoir l’œuvre collective de la France outre-mer.” Paris:
Premier Ministre, 2003.
Office National des Anciens Combattants et Victimes de Guerre. “Rapport d’Activité de l’ONAC,
2005.” Paris: ONAC, 2006.
Servier, Jean. “Enquête sur les Musulmans Français.” Paris: Comité National pour les Musulmans
Français, 1972.
SECONDARY SOURCES
Ageron, Charles-Robert. “Le drame des harkis en 1962.” Vingtième siècle. Revue d’histoire, no. 42 (June
1994): 3-6.
———. “Le ‘Drame des harkis’: Mémoire ou histoire?” Vingtième siècle. Revue d’histoire, no. 68
(December 2000): 3-15.
———. “Les supplétifs algériens dans l’armée française pendant la guerre d’Algérie.” Vingtième siècle.
Revue d’histoire, no. 48 (December 1995): 3-20.
———. Modern Algeria: A History from 1830 to the Present. Edited by Michael Brett. Translated by
Michael Brett. Trenton, NJ: African World Press, 1991.
Aldrich, Robert. Greater France: A History of French Overseas Expansion. London: MacMillan Press, 1996.
Azéma, Jean-Pierre, Jean-Pierre Rioux, and Henry Rousso. “Les guerres Franco-Françaises.”
Vingtième siècle. Revue d’histoire, no. 5 (March 1985): 3-5.
Balibar, Etienne. “Algérie, France: Une ou deux nations?” Lignes 30 (February 1997): 7-21.
De Barros, Françoise. “Contours d’un réseau administratif ‘algérien’ en construction d’une
compétence en ‘affaires musulmanes’: Les conseillers techniques pour les affaires
musulmanes en métropole (1952-1965).” Politix 19, no. 76 (2006): 97-117.
Bernardot, Marc. Camps d’étrangers. Bellecombe-en-Bauges, France: Editions du Croquant, 2008.
———. “Etre interné au Larzac: La politique d’assignation à résidence surveillée durant la guerre
d’Algérie (1958-1962).” Politix 24, no. 69 (2004): 39-61.
423
Besnaci-Lancou, Fatima. Fille de harki. Le bouleversant témoignage d’une enfant de la guerre d’Algérie. Paris:
Editions de l’Atelier, 2003.
———. Nos mères, paroles blessées. Une autre histoire des harkis. Lunay, France: Zellige, 2006.
Besnaci-Lancou, Fatima, and Gilles Manceron. “Introduction: En finir avec toutes les légendes.” In
Les harkis dans la décolonisation et après, edited by Fatima Besnaci-Lancou and Gilles Manceron,
13-32. Paris: Editions de l’Atelier, 2008.
Blanchard, Emmanuel. La police parisienne et les Algériens (1944-1962). Paris: Nouveau Monde éditions,
2011.
Blanchard, Pascal, Nicolas Bancel, and Sandrine Lemaire, eds. La fracture coloniale: La société française au
prisme de l’héritage colonial. Paris: La Découverte, 2005.
Blévis, Laure. “La citoyenneté française au miroir de la colonisation: Etude des demandes de
naturalisation des ‘sujets francais’ en Algérie coloniale.” Genèses, no. 53 (December 2003): 2547.
Boitel, Anne. Le camp de Rivesaltes, 1941-1942: Du centre d’hébergement au “Drancy de la zone libre”.
Perpignan: Presses Universitaires de Perignan/Mare Nostrum, 2001.
Boualam, Bachaga. Les harkis au service de la France. Paris: Editions France-Empire, 1963.
———. Mon pays... la France! Paris: Editions France-Empire, 1962.
Bourdieu, Pierre. Sociologie de l’Algérie. Paris: Presses Universitaires de France, 2010.
Branche, Raphaëlle. La Guerre d’Algérie: une histoire apaisée? Paris: Seuil, 2005.
———. La torture et l’armée pendant la guerre d’Algérie. Paris: Gallimard, 2001.
Brubaker, Rogers. Citizenship and Nationhood in France and Germany. Cambridge, MA: Harvard
University Press, 1998.
Byrnes, Melissa. “French Like Us? Municipal Policies and North African Migrants in the Parisian
Banlieues, 1945-1975.” PhD diss., Georgetown University, 2008.
Charbit, Tom. Les harkis. Paris: La Découverte, 2006.
———. “Un petit monde colonial en métropole. Le camp de harkis de Saint-Maurice-l’Ardoise
(1962-1976).” Politix 19, no. 76 (2006): 31-52.
Charbit, Tom, and Françoise De Barros. “La colonie rapatriée.” Politix 19, no. 76 (2006): 3-7.
Choi, Sung. “The Muslim Veteran in Postcolonial France: The Politics of the Integration of Harkis
After 1962.” French Politics, Culture & Society 29, no. 1 (Spring 2011): 24-45.
Cohen, William. “The Algerian War, the French State and Official Memory.” Historical Reflections 28,
no. 2 (2002): 219-39.
———. “The Harkis, History and Memory.” In Algeria and France, 1800–2000: Identity, Memory, and
Nostalgia, edited by Patricia Lorcin, 164-80. Syracuse, NY: Syracuse University Press, 2006.
Colon, Michel. Fuveau autrefois: Le temps des mineurs paysans, 1809-1962. Marseille: L’Imprimerie
caractère, 2001.
Conklin, Alice. A Mission to Civilize: The Republican Idea of Empire in France and West Africa, 1895-1930.
Stanford, CA: Stanford University Press, 1997.
Connelly, Matthew. A Diplomatic Revolution: Algeria’s Fight for Independence and the Origin of the Post-Cold
War Era. New York: Oxford University Press, 2002.
Cooper, Frederick. Colonialism in Question: Theory, Knowledge, History. Berkeley: University of California
Press, 2005.
———. “Decolonizing Situations: The Rise, Fall, and Rise of Colonial Studies, 1951-2001.” French
Politics, Culture & Society 20, no. 2 (Summer 2002): 47-76.
Cooper, Frederick, and Ann Laura Stoler, eds. Tensions of Empire: Colonial Cultures in a Bourgeois World.
Berkeley: University of California Press, 1997.
424
Crapanzano, Vincent. “The Dead but Living Father, the Living but Dead Father.” In The Dead
Father: A Psychoanalytic Inquiry, edited by L. J. Kalinich and S. W. Taylor, 163-73. London:
Routledge, 2009.
———. The Harkis: The Wound that Never Heals. Chicago, IL: The University of Chicago Press, 2011.
Eldridge, Claire. “‘We’ve never had a voice’: memory construction and the children of the
harkis (1962–1991).” French History 23, no. 1 (2009): 88-107.
Fabbiano, Guilia. “Les harkis du Bachaga Boualam. Des Beni-Boudouanes à Mas-Thibert.” In Les
harkis dans la décolonisation et après, edited by Fatima Besnaci-Lancou and Gilles Manceron,
113-124. Paris: Editions de l’Atelier, 2008.
Faivre, Maurice. Les combattants musulmans de la guerre d’Algérie: Des soldats sacrifiés. Paris: L’Harmattan,
1995.
Fogarty, Richard S. Race and War in France: Colonial Subjects in the French Army, 1914-1918. Baltimore,
MD: Johns Hopkins University Press, 2008.
Foucault, Michel. Security, Territory, Population. Lectures at the Collège de France, 1977-1978. Edited by
Michel Senellart. Translated by Graham Burchell. New York: Palgrave Macmillan, 2007.
———. “Society Must Be Defended.” Lectures at the Collège de France, 1975-1976. Translated by David
Macey. New York: Picador, 2003.
———. Surveiller et punir. Naissance de la prison. Paris: Gallimard, 1975.
———. The Birth of Biopolitics. Lectures at the Collège de France, 1978-1979. Edited by Michel Senellart.
Translated by Graham Burchell. New York: Palgrave Macmillan, 2008.
De Gaulle, Charles. Mémoires d’espoir: L’Effort, 1962-…, tome II. Paris: Plon, 1971.
Gladieu, Stéphan, and Dalila Kerchouche. Destins de Harkis: Aux racines d’un exil. Paris: Autrement,
2003.
Golsan, Richard. Vichy’s Afterlife: History and Counterhistory in Postwar France. Lincoln: University of
Nebraska Press, 2000.
Grynberg, Anne. Les Camps de la honte. Les internés juifs des camps français, 1939-1944. Paris: La
Découverte, 1991.
Hamoumou, Mohand. Et ils sont devenus harkis. Paris: Fayard, 1993.
Harbi, Mohammed, and Gilbert Meynier. Le FLN: Documents et Histoire, 1954-1962. Paris: Fayard,
2004.
Haroun, Ali. La 7e Wilaya: La guerre du FLN en France 1954-1962. Paris: Seuil, 1986.
Hautreux, François-Xavier. “L’armée française et les supplétifs ‘Français musulmans’ pendant la
guerre d’Algérie (1954-1962): Expérience et enjeux.” PhD diss., Université de Paris X Nanterre, 2009.
———. “L’engagement des harkis (1954-1962). Essai de périodisation.” Vingtième siècle. Revue
d’histoire 90 (June 2006): 33-46.
Heinis, Anne. “L’Insertion des Français Musulmans: Etude faite sur les populations regroupées dans
le Midi de la France et les centres de ex-harkis”. Doctorat de Spécialité 3°cycle, Université
Paul Valéry de Montpellier, 1977.
Hmed, Choukri. “Loger les étrangers ‘isolés’ en France. Socio-histoire d’une institution d’Etat: la
Sonacotra (1956-2006).” Université de Paris-I, Panthéon-La Sorbonne, 2006.
———. “‘Tenir ses hommes’: La gestion des étrangers ‘isolés’ dans les foyers Sonacotra après la
guerre d’Algérie.” Politix 19, no. 76 (2006): 11-30.
Hoffmann, Stanley. “Collaborationism in France during World War II.” The Journal of Modern History
40, no. 3 (September 1968): 375-395.
House, Jim, and Neil MacMaster. Paris 1961: Algerians, State Terror, and Memory. Oxford: Oxford
University Press, 2006.
Jordi, Jean-Jacques. De l’exode à l’exil: Rapatriés et Pieds-Noirs en France. Paris: L’Harmattan, 1993.
425
Jordi, Jean-Jacques, and Mohand Hamoumou. Les harkis, une mémoire enfouie. Paris: Autrement, 1999.
Judt, Tony. “Epilogue. From the House of the Dead: An Essay on Modern European Memory.” In
Postwar: A History of Europe Since 1945, 813-31. New York: Penguin Books, 2005.
Kateb, Kamel. Européens, “indigènes” et juifs en Algérie (1830-1962). Représentations et réalités des populations.
Paris: Éditions de l’Institut national d’études démographiques, 2001.
Kerchouche, Dalila. Mon père, ce Harki. Paris: Seuil, 2003.
Langelier, Elise. La situation juridique des Harkis (1962-2007). Paris: Librairie Juridique de Référence,
2009.
Laurens, Sylvain. “La noblesse d’Etat à l’épreuve de ‘l’Algérie’ et de l’après 1962.” Politix 76, no. 4
(2006): 75-96.
Lebourg, Nicolas. “Histoire générale du Camp de Rivesaltes.” Rivesaltes, 2007.
Lewis, Mary Dewhurst. The Boundaries of the Republic: Migrant Rights and the Limits of Universalism in
France, 1918-1940. Stanford, CA: Stanford University Press, 2007.
Loomba, Ania. Colonialism/Postcolonialism. New York: Routledge, 1998.
Lorcin, Patricia M. E. Imperial Identities: Stereotyping, Prejudice and Race in Colonial Algeria. New York: I.
B. Tauris, 1999.
Lorcin, Patricia, ed. Algeria and France 1800-2000: Identity, Memory, Nostalgia. Syracuse, NY: Syracuse
University Press, 2006.
Lyons, Amelia. “Invisible Immigrants: Algerian Families and the French Welfare State in the Era of
Decolonization (1947-1974).” PhD diss., University of California, Irvine, 2004.
MacMaster, Neil. Colonial Migrants and Racism: Algerians in France, 1900-62. New York: St. Martin’s
Press, 1997.
Mamdani, Mahmood. Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism.
Princeton, NJ: Princeton University Press, 1996.
Mann, Gregory. Native Sons: West African Veterans and France in the Twentieth Century. Durham, NC:
Duke University Press, 2006.
Marrus, Michael R. The Unwanted: European Refugees from the First World War through the Cold War. 2nd
ed. Philadelphia, PA: Temple University Press, 2002.
Marsh, Kate. “Territorial Loss and the Construction of French Colonial Identities, 1763-1962.” In
France’s lost empires: fragmentation, nostalgia, and la fracture colonial, edited by Kate Marsh and
Nicola Frith, 1-13. Lanham, MD: Lexington Books, 2011.
Martinez, Gabriel. “Enquête psycho-pathologique dans un groupe d’adolescents et d’adolescentes en
milieu harki”. Doctorat en Médecine, Université d’Aix-Marseille II, 1984.
Mathy, Jean-Philippe. Melancholy Politics: Loss, Mourning, and Memory in Late Modern France. University
Park: The Pennsylvania State University Press, 2011.
Méliani, Abd-El-Aziz. Le drame des harkis. Paris: Perrin, 2001.
Messina, Anthony M., and Gallya Lahav, eds. The Migration Reader: Exploring Politics and Policies.
Boulder, CO: Lynne Rienner Publishers, 2006.
Messmer, Pierre. Après tant de batailles... Paris: Albin Michel, 1992.
———. Les Blancs s’en vont: Récits de décolonisation. Paris: Albin Michel, 1998.
———. Ma part de France: Entretiens avec Philippe de Saint Robert. Paris: François-Xavier de Guibert,
2003.
Mettay, Joel. L’Archipel du mépris: Histoire du camp de Rivesaltes de 1939 à nos jours. Canet: Editions
Trabucaire, 2001.
Miège, Jean-Louis, and Colette Dubois, eds. L’Europe retrouvée. Les migrations de la décolonisation. Paris:
L’Harmattan, 1994.
Morelle, Chantal. “Les pouvoirs publics français et le rapatriement des harkis en 1961-1962.”
Vingtième siècle. Revue d’histoire 83 (September 2004): 109-119.
426
Moumen, Abderahmen. “Du camp de transit à la cité d’accueil, Saint-Maurice-l’Ardoise: 1962-1976.”
In Les harkis dans la décolonisation et après, edited by Fatima Besnaci-Lancou and Gilles
Manceron. Paris: Editions de l’Atelier, 2008. 131-45.
———. Les Francais musulmans en Vaucluse 1962-1991. Installation et difficultés d’intégration d’une
communauté de rapatriés d’Algérie. Paris: L’Harmattan, 2003.
Muller, Laurent. Le silence des harkis. Paris: L’Harmattan, 1999.
Mung, Emmanuel Ma, Kamel Dorai, Frantz Loyer, and Marie-Antoinette Hily. “Bilan des travaux
sur la circulation migratoire.” Migrations Etudes 84 (December 1998): 1-3.
Nicolaïdis, Dimitri. Oublier nos crimes: L’amnésie nationale, une spécificité française? Paris: Autrement, 1994.
Noiriel, Gérard. Etat, nation et immigration: Vers une histoire du pouvoir. Socio-histoires,; Paris: Belin,
2001.
———. Réfugiés et sans-papiers: La république face au droit d’asile, XIXe-XXe siècle. Paris: Hachette, 1998.
Nora, Pierre. “General Introduction: Between Memory and History.” In Realms of Memory, Volume 1.
Paris: Gallimard, 1984. 1-20.
Pervillé, Guy. “Comment appeler les habitants de l’Algérie avant la définition légale d’une nationalité
algérienne?” Cahiers de la Méditerranée 54 (June 1997): 55-60.
Peschanski, Denis. La France des camps: L’internement, 1939-1946. Paris: Gallimard, 2002.
Peyrefitte, Alain. C’était de Gaulle. Paris: Gallimard, 2002.
Raybaud, Antoine. “Deuil sans travail, travail sans deuil: La France a-t-elle une mémoire coloniale?”
Dédale, no. 5-6 (Spring 1997): 87-104.
Razac, Olivier. Histoire politique du barbelé: La prairie, la tranchée, le camp. Paris: Fabrique, 2000.
Rivet, Daniel. Le Maghreb à l’épreuve de la colonisation. Paris: Hachette, 2002.
Rosanvallon, Pierre. Le sacre du citoyen: Histoire du suffrage universel en France. Paris: Gallimard, 1992.
Rosenberg, Clifford. Policing Paris: The Origins of Modern Immigration Control Between the Wars. Ithaca,
NY: Cornell University Press, 2006.
Rousset, David. L’univers concentrationnaire. Paris: Editions du Pavois, 1946.
Rousso, Henry. “L’Epuration en France: Une histoire inachevée.” Vingtième siècle. Revue d’histoire, no.
33 (March 1992).
———. Le syndrome de Vichy: De 1944 à nos jours. Paris: Seuil, 1987.
Ruedy, John. Modern Algeria: The Origins and Development of a Nation. Bloomington: Indiana University
Press, 1992.
Saada, Emmanuelle. “Abdelmalek Sayad and the Double Absence: Toward a Total Sociology of
Immigration.” French Politics, Culture & Society 18, no. 1 (2000): 28-47.
———. Les enfants de la colonie: Les métis de l’Empire français entre sujétion et citoyenneté. Paris: La
Découverte, 2007.
———. “The Empire of Law: Dignity, Prestige, and Domination in the ‘Colonial Situation’.” French
Politics, Culture & Society 20, no. 2 (2002): 98-120.
———. “Un racisme de l’expansion. Les discriminations raciales au regard des situations coloniales.”
In De la question sociale à la question raciale? Représenter la société française, edited by Didier Fassin
and Eric Fassin, 63-79. 2nd ed. Paris: La Découverte, 2009.
Sayad, Abdelmalek. La double absence: Des illusions de l’émigré aux souffrances de l’immigré. Paris: Seuil, 1999.
———. “Les trois âges de l’émigration algérienne en France.” Actes de la recherche en sciences sociales, no.
15 (June 1977): 59-79.
———. “Qu’est-ce que l’intégration?” Hommes & Migrations, no. 1182 (December 1984): 8-14.
Sayad, Abdelmalek, Jean-Jacques Jordi, and Emile Temime. Migrance: Histoire des migrations à Marseille.
Vol. 4. Aix-en-Provence: Edisud, 1991.
Schnapper, Dominique. Qu’est-ce que l’intégration? Paris: Gallimard, 2007.
427
Scioldo-Zürcher, Yann. Devenir métropolitain: Politique d’intégration et parcours de rapatriés d’Algérie en
métropole (1954-2005). Paris: Editions EHESS, 2010.
Shepard, Todd. The Invention of Decolonization: The Algerian War and the Remaking of France. Ithaca, NY:
Cornell University Press, 2005.
Smith, Andrea L., ed. Europe’s Invisible Migrants: Consequences of the Colonists’ Return. Amsterdam:
Amsterdam University Press, 2003.
Spire, Alexis. Etrangers à la carte: L’administration de l’immigration en France (1945-1975). Paris: Grasset,
2005.
———. “Semblables et pourtants différents. La citoyenneté paradoxales des ‘Francais musulmans
d’Algérie’ en métropole.” Genèses, no. 53 (December 2003): 48-68.
Stoler, Ann Laura. “Colonial Aphasia: Race and Disabled Histories in France.” Public Culture 23, no.
1 (2011): 121-56.
Stora, Benjamin. Histoire de la guerre d’Algérie (1954-1962). New ed. Paris: La Découverte, 2001.
———. La gangrène et l’oubli: La mémoire de la guerre d’Algérie. 2nd ed. Paris: La Découverte, 1998.
Stovall, Tyler. “Colour-blind France? Colonial Workers during the First World War.” Race and Class
35, no. 2 (1993): 35-55.
Taguieff, Pierre-André, and Patrick Weil. “Immigration, fait national et citoyenneté.” Esprit (May
1990): 87-102.
Van Thao, Trinh. “Le retour des rapatriés d’Indochine. L’expérience des Centres d’accueil (19541960).” In Marseille et le choc des décolonisations, edited by Jean-Jacques Jordi and Emile Temime,
29-38. Aix-en-Provence: Edisud, 1996.
Thénault, Sylvie. Histoire de la guerre d’indépendence algérienne. Paris: Flammarion, 2005.
———. “Massacre des harkis ou massacres de harkis? Que sait-on?” In Les harkis dans la
décolonisation et après, edited by Fatima Besnaci-Lancou and Gilles Manceron, 81-91. Paris:
Editions de l’Atelier, 2008.
———. “Personnel et internés dans les camps français de la guerre d’Algérie: Entre stéréoypes
coloniaux et combat pour l’indépendance.” Politix, no. 69 (2004): 63-81.
———. Une drôle de justice: Les magistrats dans la guerre d’Algérie. Paris: La Découverte, 2001.
Todorov, Tzvetan. Nous et les autres: La réflexion sur la diversité humaine. Paris: Seuil, 1989.
Vers la paix en Algérie: Les négociations d’Evian dans les archives diplomatiques françaises (15 janvier 1961 – 29
juin 1962). Brussels: Bruylant, 2003.
Viet, Vincent. La France immigrée. Construction d’une politique 1914-1997. Paris: Fayard, 1998.
Weil, Patrick. “Histoire et mémoire des discriminations en matière de nationalité française: Le cas de
Juifs et des Musulmans d’origine algérienne.” Vingtième siècle. Revue d’histoire 84 (December
2004): 5-22.
———. La France et ses étrangers: L’aventure d’une politique de l’immigration, 1938-1991. New ed. Paris:
Gallimard, 2004.
———. “Politique de la mémoire: l’interdit de la commémoration.” Esprit no. 332 (February 2007):
124-42.
———. Qu’est-ce qu’un Francais? Histoire de la nationalité francaise depuis la Révolution. Paris: Grasset, 2002.
Wilder, Gary. The French Imperial Nation-State: Negritude and Colonial Humanism between the Two World
Wars. Chicago, IL: University of Chicago Press, 2005.
Wormser, André. Pour l’honneur des harkis, 1 an de combats, 45 années de lutte. Marseille: Editions Sillages,
2009.
428
VITA
JEANNETTE E. MILLER
[email protected]
EDUCATION
PhD. The Pennsylvania State University, University Park, PA, Department of French and Francophone
Studies, Dissertation Defended on December 12, 2011.
MA. New York University, New York, NY, Institute of French Studies, 2003.
BA with Distinction. University of Virginia, Charlottesville, VA, French/Foreign Affairs, 1998.
PUBLICATIONS
Accepted
“Post-Imperial Citizenship and the Reciprocity of the National and the Local: Writing a History of the
Harki Population.” Revue Européenne des Migrations Internationales 29, no. 1 (Forthcoming Spring 2013).
15 pp.
Accepted Pending Revisions
“A Camp for Foreigners and ‘Aliens’: Harki Citizens’ Exile at the Rivesaltes Camp (1962-1964).” French
Politics, Culture & Society (Forthcoming). 28 pp.
TEACHING EXPERIENCE
Goucher College, Visiting Instructor/Assistant Professor of French, Department of Modern Languages,
Literatures & Cultures, 2010-present.
The Pennsylvania State University, Teaching Assistant, Department of French and Francophone Studies,
2003-2010.
Institute for American Universities, Visiting Instructor, Aix-en-Provence, France, Summer 2008.
SELECTED AWARDS AND FELLOWSHIPS
Camargo Foundation Residential Fellowship, Cassis, France, Fall 2009.
Africana Research Center Dissertation Fellowship, The Pennsylvania State University, College of Liberal
Arts, 2008-2009.
Bourse Jeanne Marandon, Société des professeurs français et francophones d’Amérique, 2007-2008.
Fulbright Fellowship to France, United States Department of State, 2006-2007.
Florence Gould Foundation Pre-Dissertation Fellowship, Council for European Studies, 2004.
Graduate Scholar Award, The Pennsylvania State University, College of Liberal Arts, 2003-2010.
GSAS Fellowship, New York University, Institute of French Studies, 2002-2003.