<%BANNER%>
HIDE
 Front Cover
 Front Matter
 Title Page
 Copyright
 Dedication
 Frontispiece
 Table of Contents
 Preface
 Foreword
 The Lands and the People
 Historical Survey
 Peasant Revolution : Myths and...
 The Background to Revolution
 The Road to Violence
 The Birth of the Republic
 The Inyenzi at the Gates
 The Quest for Solidarity
 The Kingdom Reborn
 Burundi in Perspective : a Profile...
 Mwami-ship : Ethos and Structu...
 Ganwa Politics in Modern Guise...
 The Displacement of Conflict :...
 The Intervention of the Crown
 The Intrusion of External...
 The Dialectics of Succession
 The Army at the Helm
 Revolutionary Change and Natio...
 Chronology of the Kings of Rwanda...
 Genealogy of the Kings of Burundi...
 Reference
 Bibliography
 Index














Rwanda and Burundi
CITATION SEARCH THUMBNAILS PAGE TURNER PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00103151/00001
 Material Information
Title: Rwanda and Burundi
Series Title: Praeger library of African affairs
Physical Description: xi, 562 p. : illus., maps, ports. ; 23 cm.
Language: English
Creator: Lemarchand, René
Publisher: Praeger Publishers
Place of Publication: New York
New York
Publication Date: 1970
 Subjects
Subjects / Keywords: Politics and government -- Rwanda   ( lcsh )
Politics and government -- Burundi   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Spatial Coverage: Rwanda
Burundi
 Notes
Bibliography: Bibliography: p. 533-545.
 Record Information
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 00082954
lccn - 73077303
System ID: UF00103151:00001

Table of Contents
    Front Cover
        Page i
    Front Matter
        Page ii
    Title Page
        Page iii
    Copyright
        Page iv
    Dedication
        Page v
    Frontispiece
        Page vi
    Table of Contents
        Page vii
        Page viii
    Preface
        Page ix
        Page x
        Page xi
        Page xii
    Foreword
        Page xiii
        Page xiv
    The Lands and the People
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
    Historical Survey
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
    Peasant Revolution : Myths and Realities
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Images 114a
        Images 114b
        Page 115
        Page 116
        Page 117
    The Background to Revolution
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
    The Road to Violence
        Page 145
        Page 146
        Images 146a
        Images 146b
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
    The Birth of the Republic
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
    The Inyenzi at the Gates
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
        Page 205
        Page 206
        Page 207
        Page 208
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
        Page 221
        Page 222
        Page 223
        Page 224
        Page 225
        Page 226
        Page 227
    The Quest for Solidarity
        Page 228
        Page 229
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
        Page 240
        Page 241
        Page 242
        Page 243
        Page 244
        Page 245
        Page 246
        Page 247
        Page 248
        Page 249
        Page 250
        Page 251
        Page 252
        Page 253
        Page 254
        Page 255
        Page 256
        Page 257
        Page 258
        Page 259
        Page 260
        Page 261
        Page 262
        Page 263
    The Kingdom Reborn
        Page 264
        Page 265
        Page 266
        Page 267
        Page 268
        Page 269
        Page 270
        Page 271
        Page 272
        Page 273
        Page 274
        Page 275
        Page 276
        Page 277
        Page 278
        Page 279
        Page 280
        Page 281
        Page 282
        Page 283
        Page 284
        Page 285
        Page 286
        Page 287
        Page 288
    Burundi in Perspective : a Profile of Political Decomposition
        Page 289
        Page 290
        Page 291
        Page 292
        Page 293
        Page 294
        Page 295
        Page 296
        Page 297
        Page 298
        Page 299
        Page 300
    Mwami-ship : Ethos and Structure
        Page 301
        Page 302
        Page 303
        Page 304
        Page 305
        Page 306
        Images 306a
        Images 306b
        Page 307
        Page 308
        Page 309
        Page 310
        Page 311
        Page 312
        Page 313
        Page 314
        Page 315
        Page 316
        Page 317
        Page 318
        Page 319
        Page 320
        Page 321
        Page 322
        Page 323
    Ganwa Politics in Modern Guise : Bezi versus Batare
        Page 324
        Page 325
        Page 326
        Page 327
        Page 328
        Page 329
        Page 330
        Page 331
        Page 332
        Page 333
        Page 334
        Page 335
        Page 336
        Page 337
        Page 338
        Images 338a
        Images 338b
        Page 339
        Page 340
        Page 341
        Page 342
    The Displacement of Conflict : Hutu versus Tutsi
        Page 343
        Page 344
        Page 345
        Page 346
        Page 347
        Page 348
        Page 349
        Page 350
        Page 351
        Page 352
        Page 353
        Page 354
        Page 355
        Page 356
        Page 357
        Page 358
        Page 359
        Page 360
    The Intervention of the Crown
        Page 361
        Page 362
        Page 363
        Page 364
        Page 365
        Page 366
        Page 367
        Page 368
        Page 369
        Page 370
        Page 371
        Page 372
        Page 373
        Page 374
        Page 375
        Page 376
        Page 377
        Page 378
        Page 379
        Page 380
        Page 381
        Page 382
    The Intrusion of External Influences
        Page 383
        Page 384
        Page 385
        Page 386
        Page 387
        Page 388
        Page 389
        Page 390
        Page 391
        Page 392
        Page 393
        Page 394
        Page 395
        Page 396
        Page 397
        Page 398
        Page 399
        Page 400
        Page 401
    The Dialectics of Succession
        Page 402
        Page 403
        Page 404
        Page 405
        Page 406
        Page 407
        Page 408
        Page 409
        Page 410
        Page 411
        Page 412
        Page 413
        Page 414
        Page 415
        Page 416
        Page 417
        Page 418
        Page 419
        Page 420
        Page 421
        Page 422
        Page 423
        Page 424
        Page 425
        Page 426
        Page 427
        Page 428
        Page 429
        Page 430
        Page 431
        Page 432
        Page 433
        Page 434
        Page 435
    The Army at the Helm
        Page 436
        Page 437
        Page 438
        Page 439
        Page 440
        Page 441
        Page 442
        Page 443
        Page 444
        Page 445
        Page 446
        Page 447
        Page 448
        Page 449
        Page 450
        Page 451
        Page 452
        Page 453
        Page 454
        Page 455
        Page 456
        Page 457
        Page 458
        Page 459
        Page 460
        Page 461
        Page 462
        Page 463
        Page 464
        Page 465
        Page 466
        Page 467
        Page 468
    Revolutionary Change and Nation-Building
        Page 469
        Page 470
        Page 471
        Page 472
        Page 473
        Page 474
        Page 475
        Page 476
        Page 477
        Page 478
        Page 479
        Page 480
        Page 481
        Page 482
        Page 483
        Page 484
        Page 485
        Page 486
        Page 487
        Page 488
        Page 489
        Page 490
        Page 491
        Page 492
        Page 493
        Page 494
        Page 495
        Page 496
        Page 497
        Page 498
    Chronology of the Kings of Rwanda and Burundi
        Page 499
        Page 500
        Page 501
        Page 502
    Genealogy of the Kings of Burundi 1795-1966
        Page 503
        Page 504
        Page 505
        Page 506
        Page 507
        Page 508
    Reference
        Page 509
        Page 510
        Page 511
        Page 512
        Page 513
        Page 514
        Page 515
        Page 516
        Page 517
        Page 518
        Page 519
        Page 520
        Page 521
        Page 522
        Page 523
        Page 524
        Page 525
        Page 526
        Page 527
        Page 528
        Page 529
        Page 530
        Page 531
        Page 532
    Bibliography
        Page 533
        Page 534
        Page 535
        Page 536
        Page 537
        Page 538
        Page 539
        Page 540
        Page 541
        Page 542
        Page 543
        Page 544
        Page 545
        Page 546
    Index
        Page 547
        Page 548
        Page 549
        Page 550
        Page 551
        Page 552
        Page 553
        Page 554
        Page 555
        Page 556
        Page 557
        Page 558
        Page 559
        Page 560
        Page 561
        Page 562
Full Text






Rwanda and Burundi













PRAEGER LIBRARY OF AFRICAN AFFAIRS

The Praeger Library of African Affairs is intended to pro-
vide clear, authoritative, and objective information about
the historical, political, cultural, and economic background
of modern Africa. Individual countries and groupings of
countries will be dealt with as will general themes affecting
the whole continent and its relations with the rest of the
world. The library appears under the general editorship of
Colin Legum, with Philippe Decraene as consultant editor.

Already Published


T. A. BEETHAM
ALFRED G. GERTEINY
RICHARD GREENFIELD
RICHARD HALL
ALEX HEPPLE

JAMES R. HOOKER


GUY DE LUSIGNAN

HORACE MINER (ed.)
JOHN G. PIKE

WALTER SCHWARZ
RICHARD P. STEVENS


CLAUDE WAUTHIER


Christianity and the New Africa
Mauritania
Ethiopia: A New Political History
Zambia
South Africa: A Political and
Economic History
Black Revolutionary: George Pad-
more's Path from Communism to
Pan-Africanism
French-Speaking Africa Since
Independence
The City in Modern Africa
Malawi: A Political and Economic
History
Nigeria
Lesotho, Botswana, and Swaziland:
The Former High Commission Terri-
tories in Southern Africa
The Literature and Thought of
Modern Africa: A Survey








Rwanda and Burundi


RENE LEMARCHAND



















PRAEGER PUBLISHERS
New York Washington London








PRAEGER PUBLISHERS, Inc.
iii Fourth Avenue, New York, N. Y. 100ooo3, U.S.A.
5 Cromwell Place, London, S.W.7, England

Published in the United States of America in 1970
by Praeger Publishers, Inc.

Copyright 1970, in London, England,
by Rene Lemarchand

All rights reserved

Library of Congress Catalog Card Number: 73-77303


Printed in Great Britain















To My Father

In Memoriam
























0 BUFUNDU Nyanzao* .
Cyangugugu
S 'Butore

S.- h* Muhi

0 *\ ONgozi
BUYINZI
o *Bubanza BUYINZI
.Muramvya
S *B U R U
J Bujumbura
--
IGitog Ruyli
3 KILIMIRO


BUTUTSI
S Rutana
Bururi R n


BURAGANE

TANGANYIKA

Map x. Rwanda and Burundi: Regions and Towns


4?
./



'V


9W








Contents




Preface ix
A Note to the Reader xiii
Introduction i

Part One: Rwanda and Burundi: The Background
I The Lands and the People 13
2 Historical Survey 47

Part Two: Rwanda
3 The Peasant Revolution: Myths and Realities 93
4 The Background to Revolution 118
5 The Road to Violence 145
6 The Birth of the Republic 170
7 The Inyenzi at the Gates 197
8 The Quest for Solidarity 228
9 The Kingdom Reborn 264

Part Three: Burundi
o1 Burundi in Perspective: A Profile of Political
Decomposition 289
11 Mwami-ship: Ethos and Structure 301
12 Ganwa Politics in Modern Guise: Bezi versus Batare 324
13 The Displacement of Conflict: Hutu versus Tutsi 343
14 The Intervention of the Crown 361
15 The Intrusion of External Influences 383
16 The Dialectics of Succession 402
17 The Army at the Helm 436

Part Four: Rwanda and Burundi: Conclusion
18 Revolutionary Change and Nation-Building 469







Contents
Appendix I: Chronology of the Kings of Rwanda
and Burundi 499
Appendix II: Genealogy of the Kings of
Burundi, 1795-1966 503
Notes and References 509
Select Bibliography 533
Index 547


Maps
i Rwanda and Burundi: Regions and Towns vi
2 Rwanda: Major Inyenzi Attacks, March 1961-
November 1966 218
3 Burundi: Distribution of Ganwa Fiefs, circa 1916 320
4 Burundi: Batare-Controlled Chefferies, 1954 322


Illustrations Facing Page
i Mwami Yuhi Musinga of Rwanda (1896-1931)
and his uncle, the famous Kabare 114
2 Mwami Musinga and his court c. 1916 I15
3 Mutara Rudahigwa (1931-59)
4 Kigeri Ndahindurwa (1959-62)
5 The coup d'etat of Mwima, July 27, 1959 146
6 The Conseil du Pays of Rwanda, October 1959 147
7 Tutsi (Rwanda) 306
8 Tutsi-Hima (Burundi)
9 Hutu woman (Rwanda)
io Twa (Rwanda)
1i Hutu terrorists under surveillance 307
12 A Tutsi victim
13 Andr6 Muhirwa 338
14 Joseph Biroli
15 Leopold Bihumugani ('Biha')
16 Paul Mirerekano
17 Mwami Mwambutsa of Burundi (1916-66) 339
18 President Michel Micombero of Burundi








Preface


The purpose of this book is to acquaint the reader with the past and
recent history of those two states of Africa about which Western obser-
vers seem to know least. Apart from whatever merits the book may have
from the standpoint of political historiography, it is also meant as a trial
essay in the sociology of revolutionary change in contemporary Africa.
No attempt has been made, however, to match or supersede the theo-
retical studies already available on the general theme of revolution. This
is no more than a case study intended to suggest new perspectives and
foci of investigation, and, hopefully, new avenues for further research.
A personal note is perhaps not entirely out of place at this point.
My interest in Rwanda and Burundi goes back to the autumn of 1960,
when I first had occasion to visit what was then officially known as the
United Nations Trusteeship Territory of Ruanda-Urundi. My trip to
Rwanda, occurring at a particularly crucial juncture of the country's
recent evolution, gave me the opportunity to witness violence on a scale
which, by comparison, made the Congo look almost like a haven of
tranquillity. The reasons for writing this book are thus more than a
reflection of my own professional interest in the theme around which it is
constructed; they stem from a rather unique personal experience-
from what can only be described as a sense of cultural shock in the face
of wanton killings. Whether this has in any way prevented me from
keeping an appropriate distance from my subject is for the reader alone
to decide; but I probably would not have embarked upon this task had
I not been in a position to measure the appalling character of the events
I witnessed against the general indifference they seemed to evoke from
the outside world.
My greatest debt is to the many Barundi and Banyarwanda who
patiently shared with me their knowledge of the history and political
life of their respective countries, and sometimes their anxieties in the
face of a still uncertain future. For some, these premonitions proved
only too well-founded. It is with a sense of deeply-felt grief that I recall
the help and friendly hospitality that were once extended to me by the
late Patrice Mayondo, Paul Nibirantiza and Paul Mirerekano from
Burundi, and Prosper Bwanakweri from Rwanda. Their names, however,
represent but a fraction of the list of those to whom I am personally
indebted.






Rwanda and Burundi
A number of friends and colleagues have read portions of the manu-
script and kindly proffered their advice and criticisms. Special thanks
are owed to Professor Albert Trouwborst for his sustained interest
in my work, and in particular for his judicious comments and criticisms
of chapter i; to Professor Jan Vansina for patiently answering the many
questions put to him in the course of a lengthy correspondence; and to
Rachel Yeld for giving me the benefit of her first-hand experience of
the complexities of the Rwandese refugee problem. Professor Henri-
Philippe Cart's assistance during the last stages of the manuscript was
invaluable; without his kind co-operation, my efforts at penetrating
the arcana of contemporary Burundi politics would have been largely in
vain. To Professor Arnold Heidenheimer, who learned more about the
Burundi monarchy than he had initially bargained for, I also wish to
express my thanks for his valuable criticisms of chapter io. To another
friend and colleague at the University of Florida, Professor David
Chalmers, goes a similar debt for his comments on the Introduction and
Conclusion of this book.
My stay in Belgium would have been decidedly less pleasant, and my
research there far less fruitful, had it not been for the charming hospi-
tality of Denise and Jean-Pierre Derscheid. As much as their kindness
and generosity, their help in making available to me the papers of Jean-
Pierre's father, the late Jean-Marie Derscheid, deserves an acknowledge-
ment which no words can properly express. My sense of gratitude to
them, therefore, is also a reflection of the posthumous debt I-as well
as any other scholar for whom the names Rwanda and Burundi mean
anything-owe to Jean-Marie Derscheid, whose wide-ranging intellec-
tual interests and life-time devotion to Africa are amply reflected in the
collection of documents he has left to posterity, and on which part of
this work is based. At one time Professor of Colonial Law of the Institut
des Territoires d'Outre-Mer at Antwerp, Secretary-General of the Parc
Albert, co-founder of the Institut International pour la Protection de la
Nature, he combined the talents of the administrator, the meticulous-
ness of the scholar, the zest and stamina of the explorer, to which he
added, towards the end of his life, the gallantry of a resistantt de la
premiere heure'. Until his decapitation by the Gestapo at the Branden-
burg prison, on March 13, 1944, he was a leading figure of the Belgian
underground-and had been entrusted with, among other things, the
elaboration of a secret transmission code based on an adaptation of the
Bantu and Sudanese languages.
This study could not have been written without the generous financial
support of the Social Science Research Council, and of the African
Studies Center of the University of Florida; I am equally grateful for







Preface
the financial help I was able to secure through a Fullbright-Hays
Fellowship in the summer of 1966.
Portions of the research for this book were published earlier in article
form. For permission to use and reprint part of this material I wish
to express my thanks to the Editors of Africa Report, The Journal of
Modern African Studies, Civilisations, and Cahiers d'Etudes Africaines.
I am equally indebted to Oxford University Press for permission to
reproduce segments of my essay on Rwanda published in Robert I.
Rotberg and Ali Mazrui, eds., Traditions of Protest in Black Africa,
Oxford University Press, London and New York 1969. I wish to
express my thanks to the Centre Militaire d'Information et de Docu-
mentation pour l'Outre-Mer (CMIDOM) for permission to reproduce
photos I and 3; to Infor-Burundi for photos 2, 4, 14 and 17; to the
Ministry of Information of Rwanda for photos 7, 8, 9 and lo; to
Jean-Pierre Derscheid for photos 5 and 6. Photos i 12, 13, 15 and 16
are from my own collection.
Thanks are also owed to Jan Smith for her exemplary patience in
drawing and redrawing most of the maps which appear in this book;
to Professor Peter Nixdorff, and Ernst Wittig, for their help in translat-
ing various passages quoted from German works; to Ross Pye, of the
Pall Mall Press, for his meticulousness in giving final form to the
manuscript; last but not least is the debt I owe my wife for her graceful
resignation while much of my time was being spent, mentally and physic-
ally, in the hills of Central Africa. Bringing me back from time to time
to another level of reality was not the least of her contributions.
With all this help any error of fact or interpretation contained in this
book must clearly be my own and sole responsibility.


























A NOTE TO THE READER

The names of the countries discussed in this book are Rwanda and
Burundi. Their inhabitants are known as, respectively, Banyarwanda
and Barundi, and in the singular as Munyarwanda and Murundi.
The languages spoken are Kinyarwanda in Rwanda and Kirundi in
Burundi. The caste names are, in each country, Tutsi, Hutu and Twa.
Although the forms Batutsi, Bahutu and Batwa are also encountered,
the Bantu prefix 'ba-' has been dropped, in accordance with the prevail-
ing usage. Place names have been standardised in accordance with the
new appelations devised after independence, but to avoid possible con-
fusions (as in the case of a radically different spelling) the old place
names have sometimes been indicated in parentheses next to the new
ones.






















'There are general causes, whether moral or physical ...
which operate in every monarchy, to bring about its rise, its
duration and its fall. All accidents are subject to these causes,
and if the outcome of a single battle-i.e. a particular cause
-was the ruin of a state, there was a general cause which
decreed that that state was destined to perish through a single
battle. In short, the main impulse carries all the particular
accidents along with it.'
Montesquieu, Considerations
on the Causes of the Grandeur
and Decadence of the Romans.








Introduction





OF ALL the strange syllables recently added to the roster of newly-
independent African states few have a more esoteric connotation than
Rwanda and Burundi, or, to use the old terminology, Ruanda-Urundi.*
Except for a small circle of initiates these countries are still terra
incognita for the general public. If there is a touch of unreality about
their Lilliputian dimensions, the paucity of their economic resources
and the quaintness of some of their traditions, theirs is a history which
in its events and motivations is very much part of the real world.
It is the history of two semi-feudal societies caught up in the turmoil
of a swiftly changing environment, in which traditional bonds are sud-
denly ruptured or loosened to give birth to new social groupings and
values. It is the tale of two archaic kingdoms, both undergoing a drastic
alteration of their traditional structures and symbols of legitimacy, yet
each proceeding towards the goal of political modernisation at a different
pace and through different paths. Here, as elsewhere in Africa, the
familiar themes of accommodation and conflict, continuity and change,
provide the key leitmotivs.
That these themes are susceptible to major variations is nowhere
more apparent than in the contrasting patterns of development displayed
by these two states. What makes the analysis of their recent history both
interesting and rewarding'is that we are here dealing with societies which,
however similar their traditional political institutions and social struc-
tures, have responded in radically divergent ways to the challenge of
nation-building. And in both cases the developments have been rather
different from what one has come to expect of most African states.
In Rwanda the transition to independence was accompanied by one

In this study, the terms Rwanda and Burundi are used to refer both to the
traditional African kingdoms in existence before colonial rule and to the inde-
pendent states established on July I, 1962. Though at first an integral part of
German East Africa, after the First World War Rwanda and Burundi were en-
trusted to Belgium as a League of Nations Mandate, and, after the Second
World War, as a United Nations Trust Territory. The term Ruanda-Urundi has
been reserved for the territorial and administrative unit born of the amalgama-
tion of the two kingdoms under Belgian rule, in line with the official usage.
Except where quotations have maintained the old spelling, and where the his-
torical context of the discussion requires that it be preserved, the terms Rwanda
and Burundi have been substituted.







Rwanda and Burundi
of the most terrifying upheavals thus far recorded in the annals of
decolonisation, and by what, on the surface, seems a fundamental break
with the past. Hard on the heels of a peasant revolt in 1959 came the
abolition of the monarchy. The traditional ruling elites were forcibly
ejected from the seats of power, and in the process thousands of their
kinsmen were massacred or forced into exile. For months on end, swell-
ing waves of terrorism swept through the countryside, causing untold
casualties. From this shattering experience Rwanda emerged trans-
figured-openly committed to socialism, thoroughly imbued with a
sense of republican austerity, and, significantly, genuinely pro-Western.
To the West's 'liberating' impact, Rwanda's revolutionary elites have
responded with a gratitude matched only by their continuing aversion
for the ancien regime.
In Burundi, by contrast, passage to self-government was relatively
smooth. Although the monarchy has been abolished, its staying power
has nonetheless been remarkable by comparison with Rwanda.* Not
only was the timing of change different, but also the circumstances.
Even if violence did occur on some occasions, and in some places on a
substantial scale, the passing of the Burundi monarchy did not produce
anything comparable with the Rwandese chaos; more surprising still,
the final blow was not delivered by the representatives of the 'lower
orders', as in Rwanda, but by the army acting at the instigation of a
rather diverse group of youth leaders, university students and civil
servants, for the most part ethnically related to Burundi's traditional
ruling oligarchy. Although the extent to which the political style and
methods of the new regime differ from the old is not as yet fully
discernible, the general impression is one of far less drastic change
than in Rwanda.
In these contrasting patterns of development lies a unique oppor-
tunity for the study of revolution: unique in terms of the comparability
of the political units concerned, as both claim roughly the same popula-
tion figures, have similar geographical size and topographical features,
and have the same general type of social structure and colonial heritage;
unique, too, in terms of the paradoxical nature of the outcome. The pur-
pose of this book is to explain this paradox.
Part of the endeavour involves shedding light on a set of historical
The Rwandese monarchy was formally abolished in September z961, and the
Burundi monarchy in November 1966. Although the respite accorded to the
Burundi monarchy may not seem all that significant against the total span of its
previous history, the juncture at which it occurred deserves attention. That
Burundi managed to weather the crisis of independence while retaining the
outer shell of its monarchical institutions is in itself a significant index of its
greater stability in the face of social and political changes.






Introduction
events which have not yet received as much attention from social scien-
tists as they deserve. As one reflects on the volume of printed material
dealing with the concept of revolution in Africa, one cannot help being
astonished at the paucity of informed references to Rwanda. Yet
SRwanda provides one of the very rare examples of a genuine social
revolution accompanying the accession of an African state to independ-
ence, the only other example of its kind being Zanzibar.1 And it is the
only case of a large-scale, thorough-going transformation occurring
under the auspices, and indeed with the positive help and encourage-
ment, of a colonial power. These, presumably, are sufficient reasons for
giving Rwanda a more careful consideration than some might otherwise
be willing to concede.
The concern of this study, however, is not only the reality of a
specific revolutionary experience but what this reality tells us about pro-
cesses of social change in general. What are the relationships between
traditional social structures and their capacity to absorb modernising
influences? In what ways does social structure 'connect' with the politi-
cal system? Why is it that in some cases revolutionary change becomes
a sine qua non of political modernisation and not in others? If and when
revolutionary change overtakes a society how does a society reconstruct
itself? What are the concepts and categories which seem best suited to
carry the analytic burdens imposed upon them? To attempt definitive
answers to these questions would be presumptuous at this preliminary
stage; indeed, one wonders whether the answers can be formulated at
all, other than in the most tentative fashion. Even so, one must at least
try to delineate the general assumptions which underlie this study and
which in turn have dictated the choice of material for emphasis.

A NOTE ABOUT ASSUMPTIONS
That most of these assumptions should imply a partial rejection of
earlier ones about the nature of social change in contemporary Africa
is perhaps a reflection of the very special conditions encountered in
each state; but it also suggests possible shifts in the intellectual per-
spective from which most theories of social change have thus far pro-
ceeded. First and most obviously, the countries selected for analysis
provide rather poor illustrations of the popular contention that national
unity is the product of many centuries of shared historical experiences,
and hence that 'the prime condition for the building of [modern]
nations is that they have an opportunity to age in the wood'.2 If this
were so, how could one explain that neither Rwanda nor Burundi was
able to make significant headway toward the building of a modern,






Rwanda and Burundi
integrated, national community-except, in one case, through a major
political surgery, and, in the other, through the application of increas-
ingly stringent penalties on all 'dissident' elements? Clearly, the pro-
cesses of adaptation and innovation involved in nation-building may be
just as arduous in the case of 'historic' states like Rwanda and Burundi
as in the case of the newly-emergent national communities artificially
brought together under the aegis of Western imperialism. There are
many ways in which history and tradition may conspire to impede
otherwise feasible economic integration, to impose political institutions
ill-suited to meet the expectations and material needs of the masses, and
to perpetuate or intensify traditional cleavages among them, and all this
regardless of whether or not a state has been given an opportunity to
'age in the wood'. Nor is there much evidence to be gleaned from the
recent history of Rwanda and Burundi to support the contention that
a traditional system's capacity to innovate varies in direct proportion
to its degree of political centralisation. Or else how should one account
for the fact that the more centralised of the two, Rwanda, was the first
to succumb to the challenge of political modernisation?
Both states emerged from the recesses of the pre-colonial past
with reasonably stable boundaries, definable 'national' cultures, and
with political institutions whose legitimacy had long been established-
in short, possessing those very features which some have singled out as
the principal determinants of political modernisation.3 The other side
of the coin, however, is that the same historical factors which in the
past gave territorial unity to each state were also responsible for the
elements of disunity discernible in their social structures. The results
were poly-ethnic, hierarchically-organised societies, in which power
and influence were concentrated in the hands of a small oligarchy, and
wherever social and economic differences tended to coincide with ethnic
divisions the cohesiveness of the system was directly threatened. Thus
if the complex of events and forces that went into shaping their national
identities differed substantially from those which have conditioned the
building of most other African states, this in itself cannot be regarded as
an unmixed blessing.
Broadly speaking, the traditional social context of Rwanda and Burundi
is that of caste societies in which social change, or the absence of change,
takes place within and against the limits of separated social systems.*

Neglect of this cardinal feature is one reason why Ethel Albert's attempt to
investigate differences in the degree of 'receptivity to change' between Rwanda
and Burundi seems generally unsatisfactory. Unless one is prepared to recognize
at the outset that in each society change has operated selectively, depending on
both the sources and implications of change, and the cultural predispositions






Introduction
The crucial variable in this case lies not so much in the degree of cen-
tralisation or decentralisation of the traditional authority system as in
the degree of separateness which exists between different status groups.
Thus we may encounter a situation-as in Rwanda-where political
centralisation would normally favour the acceleration of change
throughout society, but where, in fact, the rigidity of the caste system
imposes severe limitations on how far down the social pyramid change
can be tolerated without at the same time endangering the legitimacy of
the political order. On the other hand, where the degree of social mobility
between various strata is such that the political system need not be
radically altered to accommodate social change-as initially happened
in Burundi-the survival of traditional institutions and symbols of
legitimacy is not immediately called into question.4 In other words,
whatever adaptiveness a political system may derive from its structural
properties ultimately depends on the degree of flexibility and openness
of its social structure.
When speaking of closed and open systems of stratification, one must
inevitably consider the ingredients which enter into the definition of
these situations, i.e. castes, classes and elites. In this study, the word
'caste' is used in the anthropological sense to refer to predominantly
endogamous, hierarchically-organised groups with specialised occupa-
tions. This is about as accurate a definition as can be devised for the
purpose of this discussion. Experience shows an amazing range of
variability in caste situations. Ideally, a caste system implies inequality
of status as well as the absence of vertical mobility between various
strata; yet the way in which the system is supposed to work and how it
actually works are two different things. Gerald Berreman's remark that
'there is a considerable variation in the characteristics of, and relation
among, the groups to which the term "caste" is applied' has particular
relevance to this discussion. Because of differences in the caste relation-
ships in Rwanda and Burundi, the word 'caste' is here applied cross-
culturally to refer to quite different sets of conditions. The varieties of
cultural and social pluralism it covers range from situations of relatively

of the constituent sub-units of each society, the socio-political realities of
Rwanda and Burundi are bound to remain something of an enigma. Instead of
asking ourselves to what extent Rwanda and Burundi, as distinct analytical
units, are amenable to change in general, as does Professor Albert, a more fruitful
way of approaching the question is to look at each society in terms of behavioral
and normative variations among castes as well as within each caste, taking into
account the nature of the changes confronting these different sections of each
society at any given time. See Ethel Albert, "Socio-political Organization and
Receptivity to Change: Some Differences between Ruanda and Urundi",
Southwestern Journal of Anthropology, Vol. xvi, 196o, pp. 46-74.







Rwanda and Burundi


high cultural integration-where different groups are held together by
the same language, political culture, religious and political symbols-
to the much more narrow definition offered by M. G. Smith, where
pluralism denotes institutional discreteness.6
The concept of 'class' is even more elusive in its connotations. Not
only is it susceptible to a wider range of correlations, but to very differ-
ent theoretical formulations. For the sake of clarity one may start with
the trivial but nonetheless essential postulate that 'class' and 'caste' are
both concerned with social differentiation, and that whereas one's
membership in a 'caste' usually depends on one's birth, a 'class' is
normally considered to be the outgrowth of social and economic trans-
formations that are typically modern in character. But this is a very
crude and approximate characterisation, and, depending on whether
one chooses to place the emphasis on the presumed rigidity of class
divisions or their fluidity, one can elaborate this definition in one of
two ways.
If one approaches the concept of class from a Marxist perspective to
refer to groups based on relations of productivity, the boundaries be-
tween 'caste' and 'class' are liable to become extremely hazy. Thus in
his more recent interpretation of Rwandese society Professor J. J.
Maquet does not hesitate to identify the caste system of Rwanda with a
class system, arguing in effect that the relations of dominance and sub-
ordination between castes were but a reflection of the uneven-share of
economic resources between different social 'classes'.7 In this sense
Maquet's thesis provides an important corrective to P. C. Lloyd's
contention that 'classes in the classic Marxist sense of property-owning
and non-owning groups exist neither in traditional or modern Africa'.8
A major difficulty about this line of analysis, however, is that it neces-
sarily conceals the social transformations that have taken place over time
within each caste and hence tends to exaggerate the degree of consistency
between 'class' and 'caste'; another is that it suggests an element of
conflict between castes which has not always been present in the tradi-
tional society. Despite the acuity of the social conflict of which Rwanda
became the scene, neither the timing nor the scale of conflict lends the
slightest credibility to the notion of a class struggle between 'haves' and
'have-nots', at least in the sense in which Marx might have used these
terms.
Another way of looking at the notion of class is to follow Weber's
footsteps and emphasise the possibility of vertical mobility between
different layers of society. Not only does this call attention to a major
difference between 'caste' and 'class', but it also helps to explain the
nature of the change which overtakes a society when the emergence of






Introduction
new reference groups and values brings about a reversal of traditional
statuses. In this sense the concept of class uncovers a major aspect of the
conflict between traditionalism and modernity in each country. Yet
it would be misleading to attribute to the emergent social structure of
Rwanda and Burundi the characteristics of a fully developed class
structure, for at least three reasons. Firstly, the degree of social mobility
from one social stratum to the other is as yet extremely limited, and
in some cases severely circumscribed by the persistence of caste divi-
sions. Furthermore, is it at all legitimate to speak of a class structure
in the absence of a substantial internal differentiation beyond the usual
cleavage between a small and relatively affluent elite of bureaucrats and
politicians on the one hand, and a vast rural stratum on the other?*
Finally, and to the extent that an embryonic class system can be said
to exist, its roots today are still primarily political and bureaucratic.
The result, then, is a heavily top-sided social pyramid, almost entirely
dominated by what Ren6 Dumont aptly refers to as 'a bourgeoisie of
civil servants'.9
This said, it is equally pertinent to note that in each country has arisen
a group of rural wage-earners or salaried workers which, for want
of a better term, may be said to represent an emergent middle-class.
In view of their low economic standing and continuing close ties with
their rural milieux, the expression 'rural proletariat' might better reflect
their socio-economic position and residential ties. Nonetheless, insofar
as they stood mid-way between the peasantry on the one hand, and the
traditional and modern educated elites on the other, they formed a kind
of 'middle sector' which, however narrow its base and ambivalent its
'class' consciousness, in both countries provided the crucial connecting
links between the emergent elites and the peasantry. Thus if the word
'class' has any relevance in the context of this discussion, rather than to
the traditional elites (whom Mosca might have called a 'ruling class'),
or the new elites of intellectuals and literati trained under the auspices
of the Church, it is to this embryonic category of rural sans-culottes
that the term can best be applied.
The term 'traditional elites' has been reserved to holders of traditional
offices whose claims to superiority and sense of corporateness were based
largely on birth, that is on their membership of an upper caste. While

Balandier suggests that it might be more accurate to abandon the notion of a
class structure for that of a class in process of formation: 'Seldom does contem-
porary political life [in Africa] reveal the existence of an established class struc-
ture, but appears, rather, as the instrument of a class in the process of being
born.' G. Balandier, "ProblBmatique des Classes Sociales en Afrique Noire",
Cahiers Internationaux de Sociologie, Vol. xxxvmi, 1965, p. 139.






Rwanda and Burundi


many of these elites were 'modern' in the sense of being westernised,
reasonably well-educated and affluent, their distinguishing characteristic
is that their claims to authority as well as the character of their offices
were primarily rooted in tradition. By 'modern elites', on the other hand,
is meant a group of people who may or may not occupy modern political
roles but who in any event claim as their main characteristic a relatively
high degree of education in relation to both the masses of the population
and the traditional elites.10
Like all generalisations, this scheme is inevitably limited by the
contextual specificities of each state. Hence certain qualifications must
be interposed. In Rwanda the percentage of incumbent elites whose
level of education meets the requirements of a 'modern' elite is relatively
small compared to Burundi, because of the timing of the revolution
and the long denial of educational opportunities until then faced by
members of the ethnic stratum from which the revolution derived its
impetus. Moreover, the distinction between 'traditional' and 'modern'
elites is complicated by the existence in northern Rwanda of a special
category of traditional authorities: special in that they all belonged
to the lower caste, i.e. to an ethnic stratum which, by virtue of its
revolutionary commitment, might otherwise be regarded as 'modern';
special, too, in that their traditional claims to authority were temporarily
suspended during the colonial period by the counter-claims of the upper-
caste traditional elites, and the support which the latter received from
the colonial authorities. This in turn underscores the need for adducing
yet another variable in explaining elite differences, based 6n inter-caste
variations. As we shall see, while in Rwanda inter-caste differences have
now largely lost their relevance insofar as intra-elite tensions are con-
cerned, in Burundi these differences remain the single most important
factor in an understanding of present and future developments.
The problem of defining general analytical categories inevitably raises
the question of how much emphasis should be given to the unique
features of these societies as against their common characteristics. In
the light of what has already been said, a convincing case could be made
for treating both countries within the same general analytical framework;
instead, however, each country has been treated separately in regard to
recent political developments. The justification for this approach lies
in the radically divergent patterns of evolution exhibited by each state.
In one case a violent, genuinely revolutionary change has taken place,
involving 'a drastic, sudden substitution of one group in charge of
the running of a territorial political entity for another group'."1 In
the other a process of evolution has occurred which, however close it
may have come to the threshold of revolution, focuses attention on very






Introduction
different forces, institutions and events. For some, this approach may
be regarded as proof of an unwarranted predilection for historicism.
The only reply to this is that historicism is not without certain virtues.
To the extent that it invites empirical observations from which more
general propositions can be derived, historicism may well be a precon-
dition to sound systematic analysis.
Because Rwanda and Burundi are among those countries of Africa
about which 'hard' data is extremely scarce, at least insofar as recent
developments are concerned, there has been heavy dependence for
information on interviews with officials and non-officials from each
country. From the very beginning, however, it became evident that such
information was frequently biased or incomplete. This was perhaps
inevitable in a cultural environment in which concealing or distorting
the truth are traditionally regarded as both a virtue and an art; but it
also reflects the degree to which political convictions and group loyalties,
when carried to an extreme, may inhibit the exchange of objective
information.
Whenever possible, therefore, an attempt has been made to cross-
check interviews by drawing informants from different ethnic strata.
A similar procedure has been employed to verify translations of the
vernacular press and documents written in Kinyarwanda and Kirundi.
Several administrators have kindly proffered their assistance in making
certain documents available, but they have requested that their names
should not be revealed. There is a similar obligation towards a number
of Africans and Europeans who allowed me to see the petitions they
wrote to the United Nations visiting missions. Sources have been cited
which have not been fully identified; references are made to interviews
with respondents whose names have not been disclosed. It is hoped,
however, that in spite of its obvious limitations and shortcomings this
book may cast a few rays of light on what, until not too long ago, was
still regarded as 'the darkest of Africa'.


















Part One

Rwanda and Burundi:
The Background











I. The Lands and the People


'A LAND of almost ideal beauty'; 'The Switzerland of Africa'-these
were the words used by early European travellers to describe the moun-
tain kingdoms that have since become the republics of Rwanda and
Burundi. Decades of contact with the West have done little to alter the
truth of these comments. For all the turmoil and political convulsions
suffered by each state, the physical landscape remains essentially what
it was when the first Europeans trudged up the rolling hills-a kind of
tropical Switzerland, whose geographical outline, shaped in the form of
a human heart, is an appropriate reminder of the countries' central
location on the map of Africa.

THE SETTING
Of approximately equal size and contiguous to each other, Rwanda and
Burundi are located in the Central African rift valley, slightly south of
the equator, in one of the highest-lying areas of the continent. They
form an elongated block of highlands of some 34,000 square miles (about
twice the size of Belgium), bounded on the east and west by the converg-
ing frontiers of Tanzania and the Congo, and in the north by Uganda.
Rwanda is separated from Burundi by the Akanyaru river, and in the
extreme east by the Kagera river valley, which, after a sharp northward
bend, becomes Rwanda's eastern border. Except for the lakeshore region
of Burundi, generally hot and humid throughout the year, the annual
average temperature for both countries fluctuates around 68F, with
only small seasonal variations. The average rainfall for the area varies
between 40 and 50 inches a year, but the volume and frequency of
precipitations varies markedly according to the season, and heavy rain-
falls are often succeeded by severe droughts.
There is bucolic charm as well as a touch of grandiose beauty in the
physical environment of both countries. East of Lake Kivu, and travers-
ing the entire region from north to south, surge the giant peaks of the
Congo-Nile crest, reaching a maximum of 14,ooo feet in the Virunga
chain. This great volcanic massif, covered with thick tropical woodlands,
merges into an undulating plateau with altitudes varying between 4,500
and 6,500oo feet. The typical landscape consists of hills and valleys scat-
tered with eucalyptus trees and banana groves, alternating with patches






Rwanda and Burundi
of luxuriant pasture. It is a fertile region, ideal for herding and the
cultivation of food crops. But it quickly shades off in the east into the
savannah zone, where the vegetation may range from vast stretches of
arid and treeless grassland to acacia scrublands and bamboo forests.
A similar type of savannah-like vegetation is found in the Imbo region,
a narrow strip of denuded territory which runs parallel to the Ruzizi
valley and along the eastern shore of Lake Tanganyika. If one compares
this environment with what the early German travellers saw in other
parts of East Africa, one can easily understand their enthusiastic
reaction upon discovering the area. 'It is a land flowing with milk and
honey', wrote the Duke of Mecklenburg in 1910, 'where the breeding
of cattle and bee culture flourish, and the cultivated soil bears rich crops
of fruit. A hilly country, thickly populated, full of beautiful scenery, and
possessing a climate incomparably fresh and healthy; a land of great
fertility, with watercourses which might be termed perennial streams;
a land which offers the brightest of prospects to the white settlers.'
Despite this surface impression of lushness and prosperity, Rwanda
and Burundi are among the poorest countries in the whole of Africa.
Partly because of the lack of any significant mineral resources, and partly
because of the absence in the past of incentives for development, their
economies have not progressed very far beyond the subsistence level.
The only important cash crop is coffee (of the 'arabica' variety), first
introduced by the Belgians in 1932 under a compulsory cultivation
system. Combined production for both countries increased from 10,000
metric tons in 1942 to 33,629 in 1961, but declined sharply after inde-
pendence. Although cotton growing has been tried with some success
in Burundi, in the Mosso region and along the Ruzizi valley, thus far it
accounts for only a small fraction of total cash exports. Secondary
commercial crops also include palm plantations along the shores of
Lake Tanganyika, and tobacco, barley and wheat in the regions of
higher altitude. Experiments in rice growing are at present being
conducted in the central region of Rwanda, under the supervision of
Nationalist Chinese agronomists; as yet, however, the results do not
seem so encouraging as to warrant large-scale production. The main
crops are beans, peas, sorghum, cassava, maize, and bananas for the
brewing of beer. While these are now grown in sufficient quantities to
satisfy the needs of domestic consumption in both countries, one needs
only to recall the famines of 1916 and 1943, which caused 50,000 and
36,000 deaths respectively, to realise the limits and vulnerability
of their economic resources. This, in turn, may help to explain why
Mecklenburg's dreams of white settlement never materialised, at least
on a substantial scale. An official report, written in 1920, stated: 'The






The Lands and the People
This state of affairs calls to mind a sequence of events of considerable
historical significance to an understanding of subsequent developments
in Burundi. Since the story has already been told elsewhere in greater
detail,7 the following brief summary will suffice. During the first half of
the nineteenth century, Mwami Ntare II Rugaamba (c. 1795-1852)
established his reputation as one of Burundi's most illustrious kings.
He owed much of his fame to his spectacular territorial conquests,
having successively incorporated into his domain the Bugufi region in
the east, parts of Buha and Buyogoma, and a substantial portion of
southern Rwanda. The spoils of victory went to his sons, however,
and in particular to Rwasha and Twarereye, who, after taking over the
administration of the new provinces, proceeded to assert their independ-
ence from the crown. This act of rebellion (for this is what it amounted
to) led to bitter conflicts between Ntare's sons and his successor on the
throne, Mwami Mwezi Kisabo (c. 1852-1908), culminating with the
death of Twarereye at the battle of Nkoondo, fought near the traditional
capital of Muramvya around 1860. The dynastic feuds between the
king and the princes went on unabated for many years, and by 1900
Mwezi Kisabo could claim effective control over only half his kingdom,
while the other half remained in the hands of Ntare's rebellious sons,
from then on known as the Batare. Thus what had initially begun as a
series of territorial accretions led to violent rivalries among the rep-
resentatives of different dynasties. By appointing his sons to rule as his
deputies over conquered territories, Ntare unwittingly sowed the seeds
of a bitter opposition among the princes of the blood, which in recent
times found expression in a resurgence of political antagonisms between
certain members of the Batare family and the descendants of King
Mwezi, the Bezi.
From this brief incursion into the past one can detect some significant
differences in the process of cultural amalgamation which have taken
place in each kingdom. By far the most important concerns the very
prominent status achieved by the princes of the blood, or ganwa, in the
political system of Burundi. Because of the special eminence conferred
upon them by the accidents of history, they became identified as a
separate ethnic group, whose prestige in society ranked far above that
of the ordinary. Tutsi. If, in addition, one remembers that there are
in Burundi two distinctive categories of Tutsi-the 'low-caste'

was attended by intrigue and conflict, disturbances stemmed from competition
among the maternal clans of the eligible princes-as happened at Rucuncu in
i896. See infra, chapter 2, p. 57-8. For an excellent discussion of the dynastic
implications of the Rucuncu coup, see M. d'Hertefelt and A. Coupez, La
Royautd Sacrte de I'Ancien Rwanda, Tervueren 1964, pp. 333-4.







Rwanda and Burundi
Tutsi-Hima and the 'upper-caste' Tutsi-Banyaruguru-the total picture
of society appears decidedly more variegated than in Rwanda.* This
greater variety of status groups, ranging from prince to commoner, is
one major reason why in the past Burundi society was relatively free of
racial tensions; just as the degrees of social distance within the Tutsi
stratum were at times far more perceptible than between Tutsi and
Hutu, the distance between them and the princely families was equally
if not more conspicuous.
Another factor of social cohesion associated with ganwa rule lies in
the very nature of the competitive relations that have developed among
them. Simmel's observation that 'conflict may also bring persons and
groups together which otherwise have nothing to do with each other's1
gives a clue to an understanding of the changing patterns of Hutu-Tutsi
relations created by princely rivalries. These rivalries caused the con-
testants, including the mwami, to seek the support of both Hutu and
Tutsi, and this could hardly have occurred in a situation of unadulterated
harmony among the ruling elites. Since neither the mwami nor the
ganwa could hope to exercise a monopoly of power, and since their
security depended ultimately on their ability to generate support from
below, they had to adopt a far more conciliatory attitude towards the
'lower orders' than might have been the case. In Rwanda, by contrast,
there was no need for the chiefs to pander to the masses because the army
was not only powerful but entirely loyal to the Tutsi cause. While in
Rwanda monarchical absolutism was a major determinant of the rigidity
of the caste structure, in Burundi the institutionalisation of rebellion
gave the social system a greater measure of internal cohesion at the local
or regional level.
Equally noteworthy are the chronological variations relating to the
different stages of Tutsi expansion. As one might expect, in both coun-
tries the most recently incorporated areas are also those where regional
Another factor contributing to the greater flexibility of Burundi's social
structure concerns the different rankings of social prestige attached to the various
patrilineages (imiryango) within each caste. Thus, within the Tutsi-Banyaruguru
(literally, 'the people from above'), by far the most prestigious ethno-cultural
segment within the Tutsi caste, the usual distinctions made by the Barundi are
between the very good families (imiryango myiza), those that are rather good
(imiryango myiza cane), neither good nor bad (imiryango si myiza si mibi) and bad
(imiryango mibi). No less than forty-four different patrilineages thus enter into
the Tutsi-Banyaruguru segment, each in turn falling into a specific social cate-
gory. Very much the same type of classification and terminology applies to the
Hutu. In this fashion lineage affiliations could substantially rectify and even
reverse the formal rank-ordering established through the caste system. For
further information, see J. Keuppens, L'Urundi Ancien et Moderne, Bujumbura
1956, mimeo.






The Lands and the People
sentiments are the most noticeable. In Burundi, for instance, the in-
habitants of the Mosso and of the northern parts of the Imbo and.
Mugamba regions still have a strong sense of regional consciousness.
But there is no equivalent in Burundi for the rugged individualism and
fierce sense of autonomy which characterise the Hutu populations of
northern Rwanda. Nor is there any precedent in Burundi for such a
late annexation as that of northern Rwanda. That this area was one where
the duration and intensity of contact between Hutu and Tutsi was
minimal helps to explain why even to this day it represents a distinctive
sub-culture within the broader context of Rwanda society.
Finally, differences in the patterns of Tutsi immigration have affected
the geographical distribution of ethnic groups in different ways. In
Burundi, the bulk of Tutsi elements are found, logically enough, in the
Bututsi region, where they make up between 80 and 85 per cent of the
local population. Most other areas have only a sprinkling of Tutsi.
About one-third of the country is inhabited by a mixture of Hutu and
Hima populations, with virtually no Tutsi. In Rwanda, however, Tutsi
elements were spread almost evenly throughout the country. The only
notable exception was the northern region (Ndorwa, Mutara, Mulera),
where the Tutsi never accounted for more than a tiny fraction of the
total population. In practice this meant that the caste structure of
Rwanda was fairly rigid throughout the area, while in Burundi local
variations dictated a much greater flexibility in the application of the
caste system. And just as in Rwanda the uniformity of the caste structure
invited a similar uniformity of political organisation, in Burundi
regional differences in the pattern of social stratification called for a
greater political diversification.
Whatever factors and circumstances have contributed to shape
Burundi's traditional social structure, evidences of inter-caste mobility
are undeniable. In Professor Albert's words: 'The life history of an
individual is limited but not fully determined by the formal structures
of the society. The dynamics of the social system make provision for
political and economic mobility by which all but the most wretched
Barundi know how to profit.'19 Applied to Rwanda, this characterisation
would amount to a travesty of socio-political realities. Seldom have the
threads of migrations, history and cultural evolution produced such
striking differences-and similarities-of social structure as between
Rwanda and Burundi, and it is from that perspective that one must
elucidate the divergences of structure and orientation apparent in their
traditional political systems.






Rwanda and Burundi


The Traditional Polities
That the kingdoms of Rwanda and Burundi differed from each other
in some essential ways was clearly sensed by early European observers.
Hans Meyer, for example, noted that 'under Mwezi Kisabo's predeces-
sors, and Kisabo himself, Urundi did not progress towards a united
state to the same extent as Rwanda. The kingdom had not yet created
such a supreme position of power over all the Tutsi chiefs as in Rwanda,
although Kisabo had concentrated his long and active life on this task
with great energy and much success.'20 The best summary of the relative
positions of the kings (bami) is found in a 1921 report of one of Belgium's
greatest colonial governors, Pierre Ryckmans:
From time immemorial the policy of the bami of Rwanda followed a
course of parcelling out the country to infinity, since large homo-
geneous provinces could easily become centres of resistance. Instead
of giving large domains to his favourites, the mwami preferred to grant
his vassals separate hills, distributed about the country. Within the
large fiefs, he carved out small dependencies where devoted persons
who owed him everything found excellent observation posts from
which to watch vassals who were very powerful, and thus to counter-
balance their influence. Besides, he encouraged rivalries by giving to
one vassal administration of Hutu lands, to the other, authority over
the Tutsi and the disposition of cattle. Finally he drained off the
power of the great lords by making of his capital a centre where
chiefs came to pay court ....
The situation was different in Urundi. Here as in Rwanda power
is a family affair. But in Urundi the family of the mwami possesses
a special and separate existence. It is united by strong bonds and
enjoys a special status. All its members, whatever the branch to
which they belong, have the generic name baganwa. In place of the
policy of splintering to infinity pursued by the bami of Rwanda,
those of Urundi sought on the contrary to reconstitute periodically
extended blocs to the profit of their sons. These, placed as often as
possible in distant provinces, were the strongest support of the
mwami, as much against enemies from outside as against possible
revolts of the princes of an older branch.21
One is reminded here of the classic distinction between 'pyramidal'
and 'hierarchical' patterns of authority. Burundi, with its highly de-
centralised organisation and its different levels of segmentation, was the
example par excellence of a pyramidal system. The political fragmenta-
tion engendered by ganwa rivalries produced a constellation of indepen-






The Lands and the People
current estimate is that in Rwanda a maximum of a dozen serious and
reasonably well-off farmers could settle down on the land without
causing prejudice to the natives.'2 By 1958, there were 1,218 bona fide
settlers in Rwanda and Burundi, of whom 550 lived in Bujumbura
(Burundi).3
Apart from the sheer paucity of natural resources, a major source of
economic stagnation lies in the perennial pressure of over-population on
the land. With a total of roughly 5 million people, almost evenly dis-
tributed between the two countries, the area has the highest population
densities in Africa: census figures for 1955 indicate an average density
of 227 per square mile in Rwanda and 185 in Burundi. At the present
rate of increase (3'3 per cent per annum), their populations are likely
to double in about thirty years. Over-population has led to a continued
exodus of African labourers to the neighboring territories: over 300,000
Banyarwanda seek employment in Uganda every year during the cotton-
picking season, and many of them are now permanently settled in and
around Kampala. But the flow of emigration has not been commensurate
with the growth of the population, and, while it may have provided
temporary relief to some saturated areas (especially in Rwanda), it did not
occur on a sufficient scale to alleviate the growing social and communal
tensions arising from the scarcity and unequal distribution of cultivable
areas.
A further strain on land resources is the high density of the cattle
population, numbering nearly 3 million in 1956. As in most
other pastoral societies of East Africa, the herding of cattle had more
than just an economic significance; the ownership of cows was not only
an important status symbol, but an essential ingredient in the traditional
socio-political systems of Rwanda and Burundi. Although their role in
contemporary society is no longer what it used to be, the sleek, lyre-
horned Ndanga cows are still an all too familiar sight. Their sheer
number tends to accelerate the process of erosion on deforested hill-
sides, having disastrous effects on soil productivity. Moreover, the
importance attached in the past to the ownership of cattle is one reason
why the adjustment of agricultural production to the requirements of a
fast-growing population has proved such a difficult task, and why, until
recently, the most fertile areas remained overstocked and overcropped.
The same factors which have tended to limit economic intercourse also
inhibited the growth of an economic infra-structure. Because of the
mountainous nature of the terrain and the absence of incentives for
large-scale industrial development, communication facilities are still
very inadequate, even by African standards. Railways are non-existent,
and until 1922 there were no roads for vehicles. The main axis of






Rwanda and Burundi
communication between the two countries (Bujumbura-Kigali-
Kakitumba) was not completed until 1931. At present there are about
5,400 miles of fair weather roads, but virtually no tarred roads. Despite
the improvements envisaged by the administering authorities after the
Second World War, an official publication lamented in 1959 that
'the building of arterial and feeder roads provided for in the Ten-Year
Plan [had] progressed very little'.4 Since then, neglect of regular road
maintenance work has, if anything, tended to further restrict access to
the interior.
With a population of 55,000, Bujumbura is both the capital city of
Burundi and the main port of entry from the south. The new port,
completed in 1960 at the cost of 149 million Belgian francs, has a handling
capacity of approximately 5,000 tons of merchandise annually. It is the
only trading centre of some importance, as well as the main processing
centre for coffee and cotton. Although trading and manufacturing
activities are no longer as prominent as they were before independence,
the bustling atmosphere of Bujumbura offers a striking contrast with
the air of puritan austerity that one encounters in Kigali, Rwanda's
capital city. With a population of 4,000 and only one asphalt road, Kigali
shares none of the characteristics of large, cosmopolitan towns. So
minimal is the rate of urbanisation in and around Kigali, so limited is
the incidence of commercial and industrial activities, that it is perhaps
better described as an overgrown village.
From these considerations emerges a significant and paradoxical re-
lationship between the rate of urbanisation and commercialization
in each country and the capacity of their respective political institutions
to withstand the impact of modernisation. However limited and periph-
eral in their incidence, that processes of social mobilisation were carried
out on a comparatively more intensive scale in Burundi provides an
interesting index of the greater pliancy of its monarchical institutions.
Nonetheless, the social environment has set definite limitations on
the 'staying power' of the Burundi monarchy. In the course of the
last few years Bujumbura has provided the setting for the emergence
of something very close to an urban 'mob', or at least of new elite
groups, eager to exert their influence and in time to challenge the
legitimacy of monarchic rule. This, coupled with the fact that Bujum-
bura happened to be the seat of the monarchy, of the government and
of parliament, has operated to focus processes of change on the capital
city, and to make it the epicentre of Burundi politics. That the fate of
the monarchy should have twice depended on the outcome of a power
play staged in Bujumbura is indeed symptomatic of the extent to which,
in recent times, national politics have become polarised around the






The Lands and the People
capital city. Nothing of this sort has yet happened in Rwanda. This does
not mean that there has not been a lack of balance in processes of social
change, only that this unevenness has had little to do with the rural-
urban split. There are no towns in Rwanda of a size comparable with
Bujumbura, and hence no scope for the kind of social and economic
forces which in Burundi have affected political processes and institu-
tions. Social change in Rwanda has been a primarily rural phenomenon.
Yet in each country the same ecological factors have tended to impede
the promotion of systematic change in the society as a whole. Real
villages do not exist. Today, as in the more distant past, the hill remains
the primary focus of political activity in the countryside. Beyond the
hill there is relatively little sense of unity among the rural communities;
even where caste solidarities are most in evidence, fragmentation and
parochialism are the rule rather than the exception. And in the absence of
adequate communications, the forbidding nature of the topography
raises further obstacles in the way of any large-scale political mobilisa-
tion. Thus one might be tempted to dismiss the significance of the eco-
logical factor as an independent variable and instead argue that in Rwanda
the traditional political institutions produced an awareness of the need
for change, as well as an orientation to change, which did not exist in
Burundi.
This is only partially true, however. In Rwanda as in Burundi change
has been extremely slow to penetrate into the fabric of the traditional
societies. This is why the Rwandese revolution has been so violent and
so untypical in many ways of what has happened elsewhere in Africa:
violent because the very slowness of the changes attempted through
constitutional reforms made it necessary at one point to substitute
intimidation for persuasion; untypical because of the peculiar circum-
stances under which violence was used, i.e. under the auspices or at
least with the tacit approval of the Belgian authorities. Yet neither
of these factors provides an adequate explanation for the contrasting
patterns of change, for these, in fact, reflect the radically different ways
in which change has been contained within the limits of their respective
traditional systems. It is to these differences that we must now direct
our attention.

THE INDIGENOUS SOCIETIES
The dominant impression conveyed by much of the literature dealing
with Rwanda and Burundi is that their societies were essentially feudal
in character. Certainly, to the extent it stresses the personal nature of
relationships among individuals, or, to use Marc Bloch's expression,






Rwanda and Burundi
'the ties of obedience and protection which bind man to man',5 the
feudal analogy draws attention to an important feature of their traditional
societies. By emphasising certain common characteristics, however, the
term 'feudalism' tends to obscure the crucial differences that have
developed over the years between these two kingdoms, and thus conveys
a degree of uniformity in the sphere of social and political relations which
did not always exist in either society.6 In attempting to reconstruct the
traditional political systems of Rwanda and Burundi the terminology
associated with European feudalism cannot serve as a substitute
for historical investigation, for only by reference to history can one
explain the structural and cultural variations which in recent years have
shaped the course of their political evolution.

Historical Perspectives
As in the case of many other African societies, the kingdoms of
Rwanda and Burundi developed their present territorial base partly
through conquest and partly through peaceful assimilation. The pattern
of expansion seems to have been the same throughout the interlacustrine
area: under the leadership of a royal clan, successive waves of nomadic
pastoralists spread their domination over the indigenous Bantu societies,
whose customs and traditions they gradually assimilated into their own
culture. This is admittedly an exceedingly simplistic view of historical
realities; the 'conquest-and-assimilation' theory of state formation has
been convincingly discredited by a number of historians, notably
Herbert Lewis and Jan Vansina. There is indeed every reason to
believe that the emergence of centralised state structures in both
countries must have occurred in part through the stimuli or conditioning
influences of pre-existing 'primary' kingdoms of Bantu origins. Non-
theless, the decisive expansion in the territorial scale of each kingdom
took place under the aegis of an 'alien' minority.
In both kingdoms the invading tribes were Tutsi or Hima pastoralists.
Although their origins are not firmly established, their physical features
suggest obvious ethnic affinities with the Galla tribes of southern
Ethiopia. Commenting on their proverbial tallness and graceful stature,
Mecklenburg observed: 'They possess that same graceful indolence in
gait which is peculiar to Oriental peoples, and their bronze-brown
skin reminds me of the inhabitants of the more hilly parts of northern
Africa. Unmistakable evidences of a foreign strain are betrayed in their
high foreheads, the curve of their nostrils, and the fine, oval shape of
their faces.'7 Dr Richard Kandt, the first German Resident in Rwanda,
was equally impressed by 'their gigantic stature, the sublimity of their
speech, the tasteful and unobtrusive way of their dress, their noble






The Lands and the People
traits and their quiet, penetrating, often even witty and irritating eyes'.8
As they drifted southward into the plateau area they came in contact
with the indigenous Hutu peasant populations. Generally short and
stocky, the Hutu share the physical characteristics of other Bantu tribes
of Central Africa. 'They are a medium-sized type of people', wrote
Mecklenburg, 'whose ungainly figures betoken hard toil, and who
patiently bow themselves in abject bondage to the later arrived yet
ruling race, the Tutsi.'9 Although their density in relation to the Tutsi
group varies greatly from one region to the next, they make up between
83 and 85 per cent of the total population in each country. If the exact
date of their arrival is impossible to determine, there is no question that
the first inhabitants to settle in the area were not the Hutu but the Twa,
a group of pygmoid forest dwellers who constitute less than one per
cent of the present population.
How the Tutsi minority managed to extend their hegemony over the
mass of the Hutu peasants is a question to which different people have
given different answers. For Hans Meyer, the German authority on
Burundi, the secret of Tutsi domination lay in their innate superiority-
in 'their superior intelligence, calmness, smartness, racial pride, solidar-
ity and political talent'.10 A more widely accepted explanation is that
the Tutsi used their cattle as a lever of economic power to subdue the
indigenous tribes; according to this view, the key to the whole situation
was a special form of cattle clientship, or cattle-contract, through which
the Tutsi oligarchy acquired sovereign political rights over their Hutu
clients. Historically, however, the situation appears to have been much
more complex. At some time in the remote past wandering bands of
Tutsi and Hima pastoralists infiltrated among the indigenous tribes,
with whom they established a symbiotic relationship. In some places
these intruders set themselves up as minor chiefs controlling a few
hills; elsewhere, relations between the two communities were essentially
of a commercial nature, involving the exchange of cattle for agricultural
products. The little that is known of this early period suggests that the
predominant form of political organisation was the clan or lineage group.
The transition from 'statelessness' to kingship was achieved by the
amalgamation of a few autonomous chieftaincies into a small nuclear
kingdom, under the leadership of a royal clan. In Rwanda this critical
step took place in the Bwanacambwe region (near Kigali) under the
reign of a Tutsi king named Ruganzu Bwimba, probably in the fifteenth
century." In Burundi the creation of the nuclear cell seems to have
taken place somewhat later, toward the middle of the seventeenth
century, and under the guidance of a Hima king who came from the
neighboring kingdom of Buha. According to one informant: 'The

'9
2-RAB *






Rwanda and Burundi
Hutu had taken over the country and divided it into several parts....
It was at that time that foreign kings entered the country. The first king
was Ntare, whose name means "rock".. . Ntare belonged to the same
family as the kings of the Ha country, of the Shuubi country, of the
Nyoro country, of the Toro country. He was a Muhinda of the clan of
the Bahinda. His mother was Inanjonaki. His father is unknown.'12
A final stage of development saw the gradual incorporation of the out-
lying areas into an expanding territorial unit. In Rwanda this process
began in the sixteenth century, with the absorption of what is today the
central region of Rwanda (Nduga-Marangara) into the nuclear kingdom.
With the accession of Ruganzu Ndori to power, in the seventeenth
century, a series of invasions was launched against formerly independent
Hutu communities which resulted in a further expansion of Rwanda's
boundaries, but the ultimate stage in this process of territorial accretion
was not completed until the latter half of the nineteenth century, under
the reign of Mwami Kigeri Rwabugiri, one of Rwanda's most prestigious
historical figures.
Rwabugiri has been credited with a spectacular series of conquests
and political reforms. In a lecture to the Universit6 Coloniale de Bel-
gique, in 1929, J. M. Derscheid described 'the essential characteristics
of Rwabugiri's policies' as follows:

To strengthen the military power of the nation; to rest the central
institutions of the realm firmly upon the support of agricultural
elements, hence of plebeians; to sap the power of the clans, of the
great feudatories and provincial chiefs at home, and at the same
time divert the energies of the Batutsi from internal matters to
military expeditions abroad.... I cannot think of a better comparison
of Rwabugiri's policies than with those pursued by Louis xi in
the domestic realm and Charles the Bold in the foreign realm.13
Mgr L6on Classe, one of Rwanda's first Apostolic Vicars, gave the
following interesting account of Rwabugiri's reign:
Rwabugiri was a conquering monarch, benevolent towards the masses,
ruthless towards the Batutsi. The masses loved him because anyone
could approach him and lay his claims and grievances before him.
The Batutsi feared him because of his utter disregard for human
life . Ceaselessly at war with his neighbours, Rwabugiri led the
Banyarwanda almost everywhere, providing them with unparalleled
opportunities to acquire abundant loot.... We had come a long way
from the time when the kings of Rwanda stood within the narrow
confines of the Nduga and Marangara regions.14 I







The Lands and the People
Where the conquered populations were already organised under a
dominant Tutsi lineage (as in the case of the Mubari, Bugesera and
Nduga regions), ethnic affinities provided a major integrative bond and
made for rapid assimilation. The case of the kingdom of Gisaka is the
only notable exception: according to A. d'Arianoff, no less than seven
expeditions were launched against the Tutsi chiefs of Gisaka between
1835 and 1852, before they were finally brought to heel by Rwogera's
warriors.15 But in those areas occupied by autonomous Hutu communi-
ties, the conquering tribes were faced with a very different situation.
Where these communities had already evolved a state system of their
own (as in the north*),1s or where natural obstacles hampered Tutsi
penetration, the extension of the Tutsi imperium proved very long and
difficult. Thus the effective annexation of the small Hutu kingdoms in
the northern and eastern 'marches' was not completed until the early
1920s, and would probably have taken even longer if it had not been
for the military assistance proffered first by the German and later by the
Belgian authorities.
The story of Burundi's quest for 'lebensraum' is remarkably similar.
From a small core area situated in the central region (Nkoma), King
Ntare Rushatsi (c.1675-1705) extended his rule over the Bututsi,
Kilimiro and Buyenzi regions. His successors saw their territorial
ambitions temporarily thwarted by the concommitant expansion of the
neighboring kingdoms of Bugesera in the east, Bula in the south-east
and Rwanda in the north. But in the first half of the nineteenth century
a new wave of conquests brought most of the peripheral areas into the
fold of the central kingdom; this was accomplished under the reign of
Mwami Ntare Rugaamba (c.1795-1852), who played for Burundi a role
somewhat similar to that played by Kigeri Rwabugiri for Rwanda.
Through his conquest of approximately half of Buha, the Buyogoma,
Ruyigi and Bugesera regions, Ntare Rugaamba expanded the original
boundaries of his kingdom on a wide scale, incorporating into his
domains sizable chunks of present-day Rwanda and Tanzania.
By then, however, the two kingdoms had spawned entirely different
The Hutu populations of northern Rwanda are also referred to as Kiga (or
Chiga) and are ethnically related to the Kiga of the Kigezi district of southern
Uganda. Unlike their Uganda kinsmen, however, described by Professor Edel
as possessing a 'basically anarchic structure', the Kiga of northern Rwanda
developed fairly centralised political structures, in which the key figure was a
'king' (muhinza). This process of political centralisation seems to have occurred
in response to the forays of invading Tutsi tribes, or in opposition to the existing
threat of a Tutsi centralised system in the south. (I asp grateful to Miss Rachel
Yeld for drawing my attention to this point.)







Rwanda and Burundi


types of political organisation. By 1900 Rwanda had achieved a remark-
able degree of centralisation. Except for the northern region, still await-
ing incorporation, all major administrative offices at the local level came
under the direct control of the king (mwami). As in Buganda, where the
authority of the clan heads (bataka) was gradually curtailed in favour of
chiefs appointed by the king (kabaka), the Rwanda kings consolidated
their rule by suppressing the autonomy of the local hereditary chiefs
and by replacing them with loyal retainers of Tutsi extraction. Although
we know virtually nothing of how and when this major structural trans-
formation occurred, it is difficult to see how it could have happened in
the absence of a strong military establishment The authority devolved
upon the army chiefs in the traditional political structure of Rwanda
bears testimony to the crucial role they must have played in bringing
about national unification.In Burundi, on the other hand, where the
military structure was conspicuously weak, the contest between the
monarchy and corporate descent groups resulted in far greater political
decentralisation. There was no parallel in Burundi for the centralised,
hierarchical pattern of authority found in Rwanda. Instead, power was
fragmented among relatively autonomous political units, each under the
authority of a prince. This authority was directly related to the rules of
royal succession, as the strength of a prince's claims ultimately depended
on the genealogical remoteness or proximity of the dynasty from which
he claimed descent. The dynastic names of the Burundi kings were
fixed by tradition as Ntare, Mwezi, Mutaga and Mwambutsa-in that
order-and their immediate descendants were known, respectively, as
Batare, Bezi, Bataga and Bambutsa. According to a tradition introduced
under the reign of Ntare II Rugaamba (c. 1795-1852), the descendants
of Ntare would hold office as ganwa until the reign of the second Ntare,
when a new generation of ganwa would come to power; likewise, the
offspring of Mwezi would remain in office until another Mwezi became
king, and so on. But these rules never became firmly institutionalized,
and thus the accession of each new king released new opportunities for
conflict. Not only were the incumbents disinclined to surrender their
authority to new aspirants, but their power was by that time so firmly
entrenched that they could easily face a trial of strength with the
monarchy.*
In Rwanda, as in Burundi, the kings assumed the dynastic names prescribed
by tradition-Cyilima, Kigeli, Mibambwe, Yuhi and Mutara-but the sons who
failed to become kings never acquired a position comparable to that of the
ganwa. The ruthless punishment dealt to potential challengers was enough to
deter even the most ambitious of the royal princes. While the heirs to the throne
were all selected from the same paternal clan (Banyiginya), their maternal clans
could be any one of four; thus, in those rare instances where royal succession






The Lands and the People
dent territorial units, each constituting 'a kingdom in miniature, owing
only a conditional allegiance to the larger unity of which it formed a
part'.22 In Rwanda, on the other hand, where political centralisation
was carried to an extreme, the mwami was the source and symbol of all
authority. He alone could confer legitimacy upon subordinate ranks.
Whereas the mwami of Burundi was little more than primus inter
pares in relation to the princes of the blood, the mwami of Rwanda
came as close to the image of an absolute ruler as any other African
monarch.
The despotic character of the Rwanda kingship was nowhere more
clearly evidenced than in the bureaucratisation of subordinate political
roles, and the precarious tenure of the occupants. The political system
was one in which a triple hierarchy of army chiefs, land chiefs and cattle
chiefs-all recruited from the dominant stratum-radiated from the
provincial capital to the provinces, and from the provinces to the
districts. Each province was entrusted to an army chief, and each
district to a land chief and a cattle chief who were responsible for the
collection of tithes in produce and cattle. Beneath this triumvirate
spread a vast network of subchiefs from whom tribute was exacted for
the higher chiefs and the kings. How much power the chiefs and
subchiefs claimed for themselves, and for how long, was entirely
dependent upon the mwami's grace. They were 'bureaucrats', as Lucy
Mair noted, 'in the sense that they did not claim their position by right
or inheritance or by virtue of any prior connexion with the area to which
they were appointed',23 but from the mwami's will.
By contrast Burundi tended to look at best like a loose aggregate of
semi-autonomous chiefdoms; at worst like a cluster of warring princi-
palities. Commenting on the state of affairs in the early days of German
rule, Hans Meyer observed: '[The king's] absolutism is only pretence
and the true rulers are the heads of the royal kin, the baganwa, who let
the royal puppet dance according to their wishes.'24 Even those ganwa
who paid formal allegiance to the king retained sovereign powers within
the limits of their domain. They were free to appoint their own sub-
chiefs, to raise their own armies in time of war, to exact tribute and
administer justice as they deemed fit. So circumscribed were the powers
of the mwami by princely prerogatives that one may indeed wonder
whether his writ was ever more authoritative or extensive than any of
the ganwa's. Below the ganwa, the administrative structure resembled
that of Rwanda, with power fragmented among relatively small terri-
torial units, each under the leadership of an appointed subchief; but
the pattern of recruitment was substantially different. In Burundi the
subchiefs were often recruited from the same clans but not necessarily






Rwanda and Burundi
from the same caste. Indeed, a remarkable feature of the traditional
system was the comparatively high proportion of Hutu chiefs who held
office in the royal domains (the so-called ivyivare). One administrative
report, for example, notes that 'there existed in the territoire of Rutuna
[prior to the arrival of the Belgians] several estates depending directly
on the mwami. These estates were administered by some Bahutu who
were not subject to the baganwa but acted like independent chiefs.'25
The same report goes on to recount how, after these Bahutu chiefs were
dismissed by the administration, in 1931, and their estates amalgamated
with those of a neighboring muganwa, the latter was unable to
command obedience from the incorporated local population. He, too,
had to be dismissed. That the king relied so heavily upon Hutu ele-
ments for administering the crown lands helps to explain the long-
lasting attachment of the Hutu peasantry to the cause of the Burundi
monarchy.
This tendency to extend the bases of political recruitment to different
ethnic strata also applied to a wide array of advisory roles for which
there was no equivalent in Rwanda. The officials in charge of perform-
ing these roles were found at each level of the political hierarchy and
were collectively known as bashigantahe. There were three categories of
bashigantahe: at the lowest level, were the so-called bashigantahe bo ku
mugina, entrusted with the task of settling disputes among families or
individuals on a hill; at the ganwa level, disputes for which local reme-
dies proved of no avail were handled by the bashigantahe bo ku nama;
still higher up in the hierarchy were the bashigantahe bo mu rulimbi,
attached to the royal court.26 Even though their functions were essen-
tially judicial, the bashigantahe could wield considerable political
influence. As one report explains: 'The mushigantahe is the natural
counsellor of his countrymen and of the traditional political authorities,
because of his wide knowledge of the country, his judgement, his fore-
sight and his sense of justice.' Whether by virtue of their training and
personal experience, or because of the nature of their office, they generally
enjoyed considerable esteem in society-'the mushigantahe of the
baganwa was always held in higher repute than an ordinary chief, and
those who were in the entourage of the mwami were more influential
than the ganwa themselves'.27 Equally significant in terms of inter-group
relations is that they were often selected on the basis of their own merits,
which meant that even a Hutu could qualify for office and thus achieve
higher status than many ordinary Tutsi.
In a sense, the bashigantahe formed the democratic core of Burundi
society, a built-in hierarchy of political roles which contained within
itself the making of a parliamentary body. That the deputies to the






The Lands and the People
National Assembly should have been called bashigantahe is not mere
coincidence. The political beliefs associated with this traditional class of
officials, and in particular the belief that it performed a legitimate and
useful function in society, expressed the deep affective commitment of
the Barundi to the notion of limited government. According to Andr6
Makarakiza, 'the dignity and office of bashigantahe are accessible to all
capable subjects regardless of their social background. Only the Batwa,
living on the margin of civil life, are excluded. The institution of the
bashigantahe . therefore insures the advantages of a democratic
government, despite the fact that the executive power is almost entirely
in the hands of the baganwa.'28
This is not to say that conciliar organs were unknown in Rwanda. The
more important chiefs of the realm, the so-called batware w'intebe (chiefs
of the stool), did act on occasion as a 'supreme council'. Yet the expres-
sion conveys an impression of formalism and efficiency which hardly
fits the contextual realities of traditional Rwanda. These 'supreme coun-
cillors' were in fact little more than a coterie of self-seeking sycophantic
courtiers. As client-chiefs they had no hereditary claims to authority.
Their tenure in office depended largely on the whims of the sovereign,
or, better, on their ability to turn his whims to their advantage.
What is one to make of these differences in the structure of authority
in each kingdom? The first point to note, somewhat in the nature of a
paradox, is that the very weakness of the Burundi monarchy has operated
to strengthen its legitimacy to an extent virtually unparalleled in Rwanda
during the years immediately preceding independence.
Unlike the situation which developed in Rwanda, where all attempts
at constitutional reform were generally interpreted by the crown as a
threat to its legitimacy, in Burundi the monarchy had already been
weakened to the point where the central issue no longer centred upon
whether or not the crown should be maintained as symbol of legitimacy
but on what particular group of princes should be allowed to gain power.
The issue of monarchic versus republican legitimacy did not intrude
upon the political scene until after independence, and, characteristically,
not until the monarchy had again shown tendencies towards royal
absolutism. Before then the key issue was the Bezi-Batare conflict.
Although the Bezi did use the symbol of the crown to fortify their claims
to authority, at no time before independence did the fate of the monarchy
become a major bone of contention between Bezi and Batare. It is not
only that the saliency of the Bezi-Batare rivalry tended to overshadow
the more fundamental issue of monarchic versus republican rule;
the constitutional issue had already been settled, for on one point at
least there seemed to be unanimous agreement among the contestants:






Rwanda and Burundi
the position of the mwami in the future political system should be that
of a constitutional monarch. Whatever opposition the crown might
have otherwise attracted unto itself was automatically deflected against
the ganwa or the parties with which they became identified, as it had
never become as closely associated with either chiefly rule or caste
supremacy as in Rwanda.
Thus there occurred in Burundi a phenomenon somewhat similar to
that which elsewhere in the world has enabled constitutional monarchies
to survive the surge of modern political competition, a phenomenon
which Edward Shils and Michael Young have described in the following
terms:
Whereas the lands where personal or absolute monarchy prevailed
were beset by revolution, countries of constitutional monarchy
became politically stable and orderly .... When protected from the
full blast of destructiveness by its very powerlessness royalty is able to
bask in the sunshine of an affection unadulterated by its opposite.
The institution of the constitutional monarchy is supported by one
of the mechanisms by which the mind defends itself from conflict,
namely, by the segregation of mutually antagonistic sentiments,
previously directed towards a single object, onto discrete and separate
objects.29
To the element of powerlessness must also be added the relative open-
ness of Burundi's caste structure. Indeed a major reason why the Burundi
monarchy was able to direct 'mutually antagonistic sentiments' to
institutions and agencies other than itself is that it never became as
closely associated with either chiefly rule or caste supremacy as in
Rwanda.
We have seen how in Burundi princely rivalries had the effect of
encouraging social cohesion between Hutu and Tutsi while in Rwanda
the absence of such rivalries tended to preserve or intensify the rigidity
of the caste structure. In Rwanda chiefly rule and caste supremacy were
both so intimately associated with the pure authority of the crown that
any statutory limitation on the privileges of the dominant caste other
than those decreed by the mwami implied a corresponding limitation
of the powers of the crown. This, however, is precisely what the system
could not tolerate. In contrast Burundi showed a much greater flexibility.
Tutsi hegemony never became a critical issue (at least, not until after
independence) for the simple reason that it had never been firmly
institutionalized. Moreover, whatever limitations were placed on the
powers of the chiefs were more likely than not to enhance, rather than
restrict, the authority of the crown. A last point, implicit in what has







The Lands and the People
already been said, is that the Burundi monarchy was relatively free of
ideological constraints, and hence all the more adaptable. Compared
to Rwanda, where ideological symbols and myths carefully specified
the limits beyond which the monarchy could not innovate without at the
same time endangering its own legitimacy, Burundi had very little in the
way of precisely articulated ideology. The climate of ideological vague-
ness surrounding the monarchy made it all the more receptive to
ideological change.
In view of the very different operative ideals shared by each society,
and of the different kinds of limitations they have placed upon institu-
tions, it may be useful at this point to investigate in somewhat greater
detail their respective political cultures, belief-systems and mythologies.

Ideologies and Norms*
All cultures are myth-sustained in that they derive their legitimacy
from a body of values and beliefs which tend to embellish or falsify
historical truth. But some more so than others. If there is any validity
to Malinowski's argument that 'the function of a myth is to strengthen
tradition and endow it with greater value and prestige, by tracing it
back to a higher, better, more supernatural reality of initial events',30
Rwanda is clearly the more cogent example.
What is particularly noteworthy about the social systems of Rwanda
and Burundi is not so much that they were both sustained by mythical
representations but that myth should have operated selectively, as if by
caprice tradition had given a differential value to some of the features
shared by each society. There was no equivalent in Burundi for the
wealth of traditional literary genres pertaining to the Rwandese mon-
archy, and even if such material did exist at one point it was obviously
not in the interest of the ganwa that it should be preserved. Conscious
as they were of the circumstances which brought them to power, the
ganwa naturally feared the verdict of history. For these reasons, as
Professor Vansina points out, Burundi was characteristically prejudiced

I am aware that much of what follows suffers from a 'functionalist' bias
insofar as primary stress is placed on the static, self-equilibrating aspect of
the political systems discussed-and of Rwanda in particular-at the expense
of the more dynamic, conflictive elements of their political culture. My only
excuse is that the limitations involved in this approach (R. Lemarchand, "Power
and Stratification in Rwanda: A Reconsideration", Cahiers d'Etudes Africains,
Vol. vi, 1966, pp. 592-6Io) are partially compensated by the advantages it
offers from the standpoint of comparative analysis. For a corrective to what
some might regard here as a pseudo-historical reconstruction, see, in addition
to the article cited above, R. Lemarchand, "Political Instability in Africa: The
case of Rwanda and Burundi", Civilisations, Vol. xvi, 1966, p. 310 ff.







Rwanda and Burundi
against history.* This is not to say that the field of oral traditions was
completely barren; a variety of folktales and legends filled the gaps of
history, of the same kind that one finds in Rwanda under the names of
umugani and igitekerezo. But in Rwanda these sources were supplemented
by the memory of court historians whose task was to hand down to
posterity the glorious traditions of the realm-not as history might have
it but, rather, as royal ordinance prescribed.
Not unnaturally, in Rwanda kingship was the focal point around
which mythical imagery clustered, as if to reinforce the exalted position
the monarchy already enjoyed in the political system. Three major
types of traditions aimed at magnifying the office of the king: the
ubwiru, the ubucurabwenge and the ibisigo. The last, transmitted by a
category of official bards known as abasizi, sought to re-enact the story
of the monarchy in a supernatural context, one in which the Rwanda
kings were inevitably cast in the mould of supermen. Something of the
same epic quality transpires from the ubucurabwenge-a collection of
traditions which preserved the genealogy of the bami--and from the
ubwiru-an elaborate body of ceremonial prescriptions, described by
Alexis Kagame as the 'esoteric code' of the Rwanda monarchy.31
Although the text of the ubwiru is no longer a secret, much of it remains
esoteric, at least to the uninitiated reader. Despite or because of its
obscurities, the ubwiru played a central role in the myth-system of
Rwanda.
The ubwiru was the ritual code of the monarchy, and the guardians
of the code, the biru, were its sole authoritative interpreters. But the
significance of the code extended far beyond the sphere of ritualised
knowledge. Since it also enshrined the testament of the departing king
and the choice of his successor, the interpreters of the code were also the
interpreters of the mwami's will, and in this capacity they tended to act
more like a constitutional court than a Delphic oracle. As Professor
Maquet observed: 'That traditional body was not unlike a constitution
in a modern state and the biru institution can be said to have had a role
similar to that of a supreme court judging whether a new rule is com-
patible with the fundamental charter of the country.'4AAlthough history
*See Jan Vansina, De la Tradition Orale: Essai de Methode Historique,
Tervueren 1961, passim. Some Belgian administrators admitted that this paucity
of reliable oral traditions imposed major handicaps on their work. As one of
them lamented, in 1935: 'It is extremely difficult to obtain any sort of reliable
information about the bami (of Burundi). Even those natives who gravitated
around Mwezi Kisabo ignore the feats of his predecessor. .. How much easier
would our work be, including genealogical research, if the court of Burundi
had anything like the ubucurabwenge of Rwanda I' Letter of Resident Oger
Coubeau to J. M. Derscheid; in the Derscheid Collection.






The Lands and the People
shows that their influence in the sphere of 'power politics' was fre-
quently overshadowed by the countervailing pressure of the great Tutsi
lineages, this does not mean that they had no political influence. Even
though they may not have had any control over a particular course of
events, they alone had the authority to sanction political change. In this
respect, perhaps no other institution did more to keep alive the 'myth'
of the monarchy, or, for that matter, to thwart any innovation likely to
endanger its primeval purity.
A content analysis of traditional sources shows the recurrence of
three dominant themes, which together formed the mythical axis of
Rwanda society. One such theme was that of a divinely ordained social
structure in which each individual was assigned a specific caste, and
each caste a specific rank. According to a dynastic poem entitled "The
Story of the Origins", the history of Rwanda begins with the reign of
Kigwa, who descended from heaven and sired three sons-GatWa,
Gahutu and Gatutsi. To choose his successor Kigwa decided to entrust
each of his sons with a pot of milk to watch over during the night. When
dawn came it turned out that Gatwa had drunk the milk; Gahutu had
gone to sleep and spilt his milk; only the watchful Gatutsi had stayed
up through the night to keep guard over his milk. To Kigwa this was
conclusive evidence that Gatutsi should be his successor and be forever
free of menial tasks. Gahutu was to be his serf. As for Gatwa, who
showed himself so utterly unreliable, his station in society was to be
that of a pariah. This myth, as Malinowski would put it, was for the
Rwandese 'neither a fictitious story, nor an account of a dead past, it
was a statement of a bigger reality still partially alive . through its
precedents, its law, its moral'.33 As such it provided a moral justification
for the maintenance of a system in which a tiny minority assumed the
status of a leisure class through the exploitation of the masses.
Next came the theme of royal omnipotence. The typical image of the
king conveyed through the lore was that of a Homeric figure. The king
was the incarnation of the deity (Imana), the embodiment of ancestral
virtues, and the source of all prosperity. In the words of a dynastic
poem dedicated to the memory of Mwami Mutara Rwogera (c. 18Io),
the popular view of the monarch was that of 'a faultless work of,art,
chiselled by chosen tools. . The nobility issued of the sacred groves
of Rwaniko. . The hero of manifold beauty whose decision cannot
be swayed, and whose memory will live forever in Rwanda.'34
Even though the king stood apartfrom Tutsi, Hutu and Twa, the
theme of kingship was inextricably bound up with the theme of Tutsi
supremacy. To rebel against the established order was no less sacrile-
gious than to rebel against the Mwami himself. According to a popular






Rwanda and Burundi
legend 'the King and the Tutsi [were] the heart of the country. Should
the Hutu chase them away, they would lose all they have and Imana
would punish them.'35 The fear of divine punishment, however real it
may have been, was not the only factor which helped the Tutsi to stay
in power; equally important was the popular notion that the Tutsi were
in fact a master race. From the accredited body of myths and super-
stitions the Tutsi emerged as Imana's elect, endowed with superior
military skill, extraordinary courage, great wealth and commensurate
intelligence. Thus, the 'official' history of Rwanda reads like a great
success story-like a saga of a few exceptional men performing remark-
able deeds. Moreover, Rwanda has no 'official' history prior to the arri-
val of the Tutsi, for in those dark ages apparently nothing seemed worth
recording. The significance of the transition to Tutsi rule is tersely
summed up in the opening sentence of a folktale of central Rwanda:
'Dead are the dogs and the rats, giving way to the cows and the
drum. .'36
As one compares the mythology of Rwanda with that of other inter-
lacustrine societies, one discovers that none of its constituent elements
was peculiar to Rwanda. The myth of origin of Bunyoro is strikingly
similar to that of Rwanda; Bunyoro and Ankole both possessed a myth-
conveyed scheme of values to explain the division of their societies into
distinctive social categories. As far as the principle of kingship is
concerned, one could easily match the omnipotence of the bami with
the despotic powers of the Kabaka of Buganda prior to 1900. But
Rwanda is unique in the sheer abundance of traditions purporting to
show the superiority of the Tutsi over the other castes, and in the
cumulative impact of these traditions on society as a whole.
To infer from the foregoing that Rwanda society was in a state of
permanent racial tension would be as far from the truth as to imagine
that Burundi was totally free from such tension. Professor Codere's
argument that the Tutsi of Rwanda maintained themselves in power
only through the application of naked force, and that the Hutu were
everywhere 'powerless', 'oppressed' and 'terrorized',37 overlooks a
basic aspect of stratificatory phenomena-namely, that they are almost
always rooted in a universally accepted code of values. IThis is where
Rwanda's mythology played such ja crucially important role in the
ordering of socio-political relations 'The caste structure of Rwanda was
based on a shared and 'culturally elaborated' image of society,"3 involv-
ing a combination of exclusiveness and reciprocity, of inequality and
solidarity. However instrumental the solidaristic features of the system
may have been in holding society together, one can hardly escape the
conclusion that it was the widespread adherence to what Maquet refers







The Lands and the People

to as the 'premise of inequality' which allowed the caste structure of
Rwanda to retain its stern rigidity over the years, until it could no longer
withstand the impact of egalitarianism.*
But if Rwanda was fundamentally averse to egalitarian influence,
this does not mean that it has always and unconditionally rejected
modernisation. As long as the principle of Tutsi supremacy was
maintained, modernisation could be easily tolerated. Although the
very centralised nature of the traditional polity tended to encourage
modernisation, its limits were very sharply delineated by the boundaries
of the caste system. Modernisation left off where caste distinctions
began, at the point where it threatened to negate the premise of
inequality.
In sum, although each society shared a certain consensus, certain
common understandings regarding the legitimacy of its social hierarchy,
the character and limits of this consensus varied widely from one state
to the other. Not only was the premise of inequality less prominent in
Burundi than in Rwanda; the lines of demarcation between groups were
drawn at different levels in each society. Whereas the main line of
cleavage in Rwanda was between Hutu and Tutsi, in Burundi the cru-
cially important distinction was between the princes, on the one hand,
and the Hutu and Tutsi on the other. The criteria of ranking, in other
words, did not involve ethnic differences as much as differences of
lineage and power.
Which brings us back to the main point of this discussion: whereas in
Rwanda a challenge to the operative norms of the system was really a
challenge to the entire socio-political structure, in Burundi there were

This is not meant to deny the part played by religious sanctions-in particular
by the cult of Ryangombwe and the Kubandwa sect, to which it gave rise-in
promoting inter-caste cohesion. In his recent work on Rwanda, Luc de Heusch
advances the hypothesis that the Kubandwa sect, by ritualistically reversing the
established hierarchy among castes, inaugurated at periodic intervals a kind of
fictitious egalitarian order which in turn served as a safety valve for accumulated
tensions and antagonisms among castes. 'The Kubandwa introduces a demo-
cratic religion which negates the divisions of the real society founded on the
ownership of cattle. . We do not accept the functional interpretation offered
by Maquet, according to which the Ryangombwe cult would only serve as an
additional factor of cohesion, uniting into a single belief system members of
different castes . The ritual of the Ryangombwe rested on a mystical and
radical negation of the established order': Luc de Heusch, Le Rwanda et la
Civilisation Interlacustre, Institut de Sociologie, Brussels 1966, p. 172. This
interpretation of the Kubandwa sect, as providing its adherents with a rite
d'inversion through which a temporary flight from reality could be achieved,
is challenged by Claudine Vidal in "Anthropologie et Histoire: le cas du Rwanda",
Cahiers Internationaux de Sociologie, Vol. XLII, 1967, pp. 143-57.






Rwanda and Burundi
relatively few ideals to challenge, save that of the monarchy, and even
then, as we have seen, the motives for contesting its existence were not
particularly compelling. Questioning the premise of inequality merely
meant questioning the legitimacy of the chiefs qua chiefs, that is of a
tiny minority of princely families; similarly, to call into doubt the legiti-
macy of the social hierarchy did not imply a threat to the privileges of a
particular caste, but to the prebends of particular individuals.

The Clientage System
In each kingdom the ties of clientship ran like a seamless web, linking
men in a relationship of mutual dependence. At the core of this relation-
ship lay an institution called buhake in Rwanda and bugabire in Burundi,
translated alternatively as 'cattle contract' or 'contract of pastoral
servitude'. But as Maquet points out, the Western conception of contract,
involving specific obligations on both parties, 'is not understood or
acted upon even by Banyarwanda who have long been familiar with
our culture'39-which, of course, is also a commentary on the difficulties
faced by Western observers when trying to elucidate the meaning of
clientage.
In essence, the clientage relationship was a highly personalised,
precarious relationship between a client and a patron, involving the
exchange of certain commodities and services. The procedure through
which this relationship was established is a familiar one to the student
of interlacustrine cultures. A client would commend himself to a more
wealthy patron by way of certain ritualised formulae-'always think of
me; make me rich; I ask you milk; etc.' If the patron accepted the cue,
the client was from then on entitled to certain privileges, usually in the
form of cattle and pasture land. But his rights were only those of usu-
fruct: he was entitled to the cow's milk and their calves, and that was
all; by the same token, the pasture rights enjoyed by the client did not
make him a land owner but only a tenant at the mercy of his lord. In
return for these privileges the client owed his patron whatever goods
and services had been agreed upon by the parties.
But clientship involved more than just an economic transaction be-
tween an inferior and a superior. It also involved a close personal
relationship, in some ways reminiscent of the ties of fealty which linked
the medieval lord to his vassal. The reciprocal bonds of loyalty between
client and patron meant that one became the other's 'man', just as in
feudal Europe the lord was the 'man' of the king, and the serf the 'man'
of his lord. In return for this act of homage the patron owed protection
to his client in every circumstance of life. In a society like Rwanda,
where centralisation of authority was carried to an extreme, the need for






The Lands and the People
protection was all the more deeply felt by the subordinate stratum, and so,
also, the need to establish reciprocal ties of obligation between inferior
and superior. 'Any superior is a protector', writes Maquet, 'and his
protection is of the same character as his authority: all-embracing and
limited only by his own convenience.'40 This was true not only of the
mwami towards the chiefs but of the latter towards the subchiefs. And
just as the precariousness of tenure associated with royal absolutism
often induced a sense of obligation on the part of the average office-
holder towards his superior, the vicissitudes of everyday life on the hills
made it mandatory for the average commoner to seek the protection of a
superior. In most cases this meant the protection of a Tutsi since by
definition the Tutsi were the prime holders of power and influence.
Very different was the situation in Burundi, if only because political
office was so deeply rooted in hereditary claims. Here the pattern of
reciprocity showed little or no coincidence with the political hierarchy,
and moreover, the obligations arising from the clientage relationship
fell evenly upon Hutu and Tutsi.
The roles of client and patron were not mutually exclusive. For
example, a man who had several clients could himself be the client of
a more wealthy patron, who in turn would be the client of an even
wealthier individual, and so forth. In this fashion, client-patron rela-
tionships formed a web of reciprocities embracing a wide segment of
the population. Its only limits were the mwami at the top, who was the
sole proprietor of cattle and land and hence the supreme patron, and,
at the other extreme, the ordinary Hutu peasant who was too poor
to impose his vassalage upon anyone but the members of his own
family.
One can easily imagine the difficulties that were liable to arise from
the juxtaposition of clientage and chieftaincy. Did allegiance to a patron
exonerate an individual from his obligations towards his chief? If not,
how were the conflicting claims of patron and chief reconciled? At
what point in the hierarchy would the authority of a chief supersede
the claims of a patron?
In Rwanda these questions were never posed in such clear-cut terms,
for chieftainship and clientage were but two faces of the same coin.
Political office was granted by the mwami or a chief as a 'privilege'
(amarembo) in return for an act of homage on the part of the recipient.
One finds in Max Weber's discussion of patrimonialism in medieval
Europe an obvious parallel with what happened in Rwanda: 'The
community was transformed into a stratum of aids to the rulers and
depended upon them for maintenance through the usufruct of land,
office fees, income in kind, salaries, and hence through prebends.






Rwanda and Burundi
The staff derived its legitimate power in greatly varying stages of appro-
priation, infeudation, conferment, and appointment. As a rule this
meant that princely prerogatives became patrimonial in nature.'41
Similarly, in traditional Rwanda chiefly positions became patrimoniall'
insofar as they were the prebends distributed by the king to retain
the loyalty of his 'men'.
There were some instances where a man owed a dual allegiance-to
his chief, for example, as well as to his patron, who might or might
not reside on the same hill. But this was the exception rather than the
rule. Since economic and political power tended to gravitate into the
same hands, the ties of clientship also tended to set the pattern of
political relations. As Pierre Gravel points out, 'buhake was identified
with the administrative system in that a chief was by definition a
person who was a vassal to some more powerful person, or to the
mwami himself, and he usually had received the hill as a fief, in which
case he was not likely to become the lord of all the notables of his hill,
and indirectly through them-or directly-that of the peasants of his
hill'.42 The average office holder would in effect combine a variety of
roles: as a patron he could always use his 'feudal' privileges to reinforce
his authority as chief, or vice-versa; as a client of a higher chief, how-
ever, he was also made aware of his obligations towards both his superiors
and his subordinates. It was this network of interlocking roles which gave
Rwanda a measure of cohesion and stability.
The political system of Rwanda was stable but not static. Since the
patron-client relationship could be terminated at any time by either
party if, for some reason or another, the arrangement proved unsatis-
factory, the system made allowance for occasional shifts in the balance
of forces. These transformations were especially noticeable at the hill
level, because the hill was the weakest link in the chain of command and
because at this level the structure of power was in some ways quite
different from what it was at the other echelons. Political competition
on the hill centred on three major institutions: the lineage, the chief-
tainship, and the nuclear feudal cluster (the patron and his clients).
As noted earlier, more often than not the chieftain was also the head
of the nuclear feudal cluster, and the combination of these two roles
evidently weakened the influence of the lineage. But in some cases the
strength of the local Hutu lineages was such that the Tutsi found it
expedient to absorb these meddlesome 'upstarts' into their own caste.
In a fascinating discussion of the power struggle that took place in
Remera, in eastern Rwanda, Gravel notes that 'the Hutu lineages which
have been in situ longest have acquired some sort of priority of rights
on the hill. Their members are respected and the heads of the lineages







The Lands and the People
have much influence on their neighbors, and have an important voice
in the local administration. ... The powerful lineages keep the power
of the chieftain in check. If, however, they become powerful enough to
threaten the chieftainship, they are absorbed into the upper caste.
Their Hutu origins are "forgotten".'43
This last point is central to an understanding of social stratification in
Rwanda. Besides showing the existence of opportunities for mobility
across caste lines, Gravel's findings suggest that the 'play for power'
was not confined to the Tutsi caste, nor bound to result in further Tutsi
oppression, despite some assertions to the contrary. Yet, the very fact
that a Hutu who successfully made his way up the social ladder should
ipso facto be assimilated into the Tutsi caste, and henceforth regarded
as a Tutsi, shows that as a group the Hutu were inevitably destined to
remain in an inferior position. A Tutsi could be both client and patron;
but a Hutu could only be a client.*
In Burundi the clientage system was separate from, and subsidiary to,
the political structure. It is true, of course, that office holders sometimes
used their prerogatives to build up bodies of loyal retainers whose
position resembled that of a client towards his lord. And there were
some instances in which a chief would receive an estate from the king
as a beneficee', as often happened in the case of those royal domains
that were scattered about the country like so many enclaves of royal
authority. But at the ganwa level at least political office was rooted in
hereditary claims, and no amount of patronage could possibly destroy
the local corporateness and autonomy resulting from hereditary
succession. Indeed, if any lesson can be drawn from the turbulent his-
tory of Burundi, it is that the mwami had no effective device at his dis-
posal to curb the excesses of rebellious princes. Recognition of hereditary
rights as the main source of authority carried some important conse-
quences. For one thing, the hierarchy of power and prestige associated
with political office was by no means co-terminous with the clientage
structure. While there were definite territorial limits to the scope of

It should be emphasised at this point that the clientship pattern just des-
cribed was confined to the central region of Rwanda. In the predominantly
Hutu areas of the north there developed a somewhat different kind of clientage,
based not on cattle but on a land-lease contract between the original owners of
the land (bakonde), for the most part of Hutu or Twa extraction, and the Hutu
lineages who opened it up for cultivation (bagererwa). According to this so-
called ubukonde system, the bagererwa offered a tribute in kind to the 'landlord'
in exchange for usufructuary rights over the land. It is both ironic and indicative
of the strength of indigenous Hutu traditions that this particular system of
land tenure, despite its obviously 'feudal' character, should still be practised
in republican Rwanda.






Rwanda and Burundi
political authority, the ties of vassalage paid no heed to such boundaries.
A man could be the client of a wealthy chief and yet evade all political
allegiance to him; conversely, within the limits of his jurisdiction a chief
exercised authority over a number of individuals who were not treated
as his clients but only as his subjects, although some could conceivably
be treated as both.
This in turn suggests that the dividing line between the economic and
political spheres was much more sharply delineated than in Rwanda.
In Rwanda a wealthy patron was by definition a powerful man, and
since a client's rights were largely dependent upon the solicitude of
his patron, it was obviously to his advantage to seek the protection of
the wealthy. Inferentially, the wealthier a patron the greater were the
chances that he might also be a chief. In Burundi, on the other hand,
whether a muganwa had a large or small retinue of clients had little
effect on his authority as a chief; by the same token, whether a patron
was more or less affluent did not substantially alter his bargaining posi-
tion vis-h-vis his chief, or at least not to the same extent as in Rwanda.
In practice, therefore, a client would usually turn to his chief, rather
than to his patron, to obtain redress for the wrongs that might have
been inflicted upon him. Even in the case of disputes between a client
and his lord, the chief remained the supreme arbiter.
Finally, because of the special context in which it was embedded, the
clientage system of Burundi operated within a fairly limited framework
of expectations. Thus the services expected from a client were not nearly
as burdensome as in Rwanda. 'As for the Hutu clients', one report
noted, 'their obligations towards their shebuja amount to precious little
-to pay court from time to time, to bring gifts of beer on even rarer
occasions, and that is all.'44 Moreover, relatively few Hutu were actually
involved in clientage relations. In the central region, for example, the
vast majority of the Hutu population lived in small, self-contained
communities that were virtually immune from contractual obligations.
If clientship served any purpose at all for the Hutu it was primarily as
an avenue for social mobility, rather than as an instrument of Tutsi
domination. Thus, if one were to generalise about the effects of client-
patron relationships on social stratification, one could argue that in
Burundi the bugabire tended to blur caste differences whereas in Rwanda
the buhake merely served to reinforce social and political inequalities
between Hutu and Tutsi.*
The bonds of subjection created by the buhake are nowhere better described
than in the following statement, by a group of Tutsi dvolues: 'The buhake
system is the means par excellence through which the Batutsi have managed to
maintain and safeguard their ascendency over the masses. The indefinite dura-







The Lands and the People
Here we touch upon one of the most paradoxical aspects of Rwanda's
recent political past. While there can be no question that the obligations
of clientage weighed heavily upon the masses, and that the solidaristic
features of the institution had been largely corrupted by the environ-
mental and administrative changes introduced by the Pax Belgica and
its legitimacy called into question by the more educated segments of the
Hutu elites, the clientage relationship served as the chief instrument
through which the Hutu leadership managed to enlist the support of the
peasantry. As we shall see, the key to this paradox lies in part in the
growing inability of the Tutsi patrons to properly fulfil their protective
roles as patrons, and in the persistence of a deeply felt need for such
protection-in particular for the psychic gratifications which presumably
arise from a relationship of dependency of the client-patron type-
among the masses. This double-edged aspect of the clientage system,
creating as it did a web of inter-caste solidarities as well as the condi-
tions of its rupture, is central to an understanding of the 'play for power'
in revolutionary Rwanda. This is also where the legacy of the traditional
system intrudes itself most forcefully upon the contemporary political
scene.

Traditional Patterns of Behaviour
The writings of early European visitors show a remarkable consensus
about the individual deportment of Hutu and Tutsi as well as about their
attitudes towards each other. They all seem to have been very forcefully
impressed by the extreme reserve of the Tutsi, which seemed so strange
when compared with the spontaneous effusions of other African tribes.
Of the Tutsi of Rwanda, Mecklenburg wrote that 'one received the
impression of being in the presence of an entirely different class of men,
who had nothing further in common with the "niggers" than their dark
complexion'.45 After his visit to Burundi Hans Meyer commented in a

tion of contractual ties it creates at each echelon of the hierarchy implies a
constant obligation to obey the dominant caste; through pure and simple
spoliation an instant remedy is found against the danger of over-rapid social
mobility or the emergence of competitive centres of power, while intrigue and
delation, both of which are encouraged by the system, maintain the omnipotence
of the powerful by fostering rivalries among the weak. Thus the system has
really become impractical and obsolete for all who have managed to evade the
constraints of the dominant caste. For these people-but not the masses!-
that is for those who are gainfully employed by Europeans, who live in townships,
in brief for a great many dvolu6s, the system is no longer acceptable.' See
Christophe Ruhara, Chrysologue Rwamasirabo, Gratien Sendanyoye, "Le
Buhake: Une coutume essentiellement munyarwanda", Bulletin de Jurisprudence
des Tribunaux Indiganes du Ruanda-Urundi, No. 3, May 1947, pp. 103-136.






Rwanda and Burundi
similar vein: 'The longer one has travelled in negro countries, and the
better one has got acquainted with the negro character, the more one is
impressed with the proud reserve of the Tutsi. There is no restless
curiosity, no noisy, partly fearful, partly good-hearted welcome, as
with most other negroes. The tall fellows stand still and relaxed, leaning
over their spears while watching the Europeans pass or approach, as if
this unusual sight did not impress them in the least.'46 But Meyer also
noted the reverse side of the picture, and in particular their laziness,
opportunism and dissimulation: 'The Tutsi never or only seldom says
what he thinks; one has to guess it. Lying is not only customary with
strangers but a permanent and deeply rooted defect.' He also noted that,
for all their mendacity, the Tutsi never concealed the fact that they
regarded themselves as the salt of the earth: 'The Tutsi consider them-
selves as the top of the creation from the standpoint of intelligence and
political genius.' Summarising the Tutsi philosophy of life, Meyer
concluded: 'To be rich and powerful and to enjoy life by doing nothing
is the symbol of all wisdom for the Tutsi, the ideal for which he strives
with utmost shrewdness and unscrupulousness.'47 By contrast the
Hutu seemed a singularly servile, boisterous and cowardly people,
whose sense of dignity and amour propre had been dulled almost to
extinction by centuries of bondage. Of the Hutu of Burundi, Meyer
wrote: 'Due to four centuries of terroristic rule, they have become
slaves in thinking and acting, though not so slave-like in character as
the Banyarwanda under their Hamitic despots.'48 If this last qualifier
sounds like an after-thought, subsequent observations show that this
was not Meyer's intention.
Contrasting the behaviour of the Banyarwanda with that of the
Barundi, he pointedly stressed the fact that 'in Burundi the Tutsi are
neither so pleasure-seeking, lazy, mendacious, violent and opportunist
as in Rwanda; nor are the Hutu so servile and hypocritical toward the
mighty and so impertinent toward the weak; nor is the king and his
court so addicted to idleness, wastefulness, intrigue, and so eager to
satisfy their depraved and cruel instincts'. He added that 'despite great
differences in status', in Burundi '... Tutsi and Hutu conduct friendly
social intercourse', and that 'the Hutu who is better off [economically]
considers himself socially on the same level as the ordinary Tutsi who
has no property'.49
Since these lines were written many observers have had occasion to
confirm the accuracy of this judgement. But how does it relate to con-
temporary patterns of behaviour? Two preliminary observations are in
order. Firstly, it would be profoundly misleading to ignore the fact that
in each society ethnic differences were associated with certain cultural






The Lands and the People
stereotypes. In Rwanda as in Burundi, Hutu and Tutsi were divided
into readily recognisable categories, definable in cultural terms; thus
in each state a genuine potential existed for a sharp polarisation of group
loyalties and identifications. Nonetheless the hierarchical ranking
attached to cultural differences was infinitely more pronounced in the
case of Rwanda. Thus the next important point to remember is that the
attitudinal variations described above were reducible not merely to
cultural discontinuities, nor even to disparities in the extent of inter-
caste mobility, but to the different moral scales-by which the Hutu as a
group were measured in each state. There was no equivalent in Burundi
for the moral and psychological inhibitions faced by the Hutu of Rwanda
in their effort to reverse the 'premise of inequality', no equivalent either
for the willing support and co-operation they once gave their masters,
Such inhibitions may conceivably provide a retrospective justification
for the stolid obduracy displayed by the dominant stratum while trying
to maintain its traditional supremacy; but they also offer a plausible
explanation for the psychological traumas, insecurities and retreats
generated among the masses of Rwanda by the advent of freedom.
The persistence of stereotyped conceptions of inferiority among t,
Hutu of Rwanda goes far in explaining their general reluctance to even
consider the possibility of changing the status quo-in short their long-
lasting political apathy. Since the Tutsi were culturally defined as highly
intelligent, refined and courageous, and the Hutu as dim-witted, gross
and cowardly, the corollary proposition was that the Tutsi were born
to rule and the Hutu to be ruled. And because many of the Hutu actually
saw themselves with the eyes of the Tutsi, they had understandably
little incentive to compete with their overlords. Hence the attitude of
sullen resignation which long characterized the Hutu of Rwanda, and
which gave currency to the stereotype that 'like almost all negro peop s
they have the natural desire to serve and be subjected to a strong and
leading hand'.50
While this 'natural desire to serve' did not last forever, submissiveness
and self-doubt remained the most enduring characteristics of Hutu
behaviour. Indeed, one of the most arduous tasks facing the Hutu
intelligentsia on the eve of independence was to break this habit of
passive obedience which even then continued to paralyse political
initiative. What made this task so difficult was not only that it violated
some basic cultural norms, but that the breaking of these norms released
tremendous psychological insecurities among the Hutu peasantry. As
they were suddenly asked to turn against the men and institutions
which for centuries had been their sole guarantee of security, many
Hutu felt hapless and bewildered. Even those who had nothing but






Rwanda and Burundi
genuine contempt and hatred for the old regime displayed an almost
pathological fear of being outsmarted at every turn by Tutsi, as if the
latter had been endowed by nature with superior gifts of shrewdness,
treachery and cunning. While there can be little question that this
stereotype lay at the root of all the insecurity and suspicion the Hutu
elites felt toward their former masters, the reverse image of the Hutu
as an inferior people, forever destined to be hewers of wood and drawers
of water, explains the intransigence of the conservative wing of the Tutsi
oligarchy in the months preceding independence. Commenting upon
this polarisation of attitudes, a prominent Hutu leader contrasted the
Rwanda revolution with the events of 1789 by pointing out: 'In France,
during the revolution, at least some noblemen saw the handwriting
on the wall. Danton, for example, became one of the staunchest support-
ers of the revolution. But in Rwanda not a single Tutsi ever committed
himself, by word or action, to our democratic ideal.'51
If insecurity bred suspicion, both led to aggression. The recent history
of Rwanda is punctuated with countless examples of bloodshed and
violence; but there are no precedents for the appalling brutality em-
ployed after independence by some Hutu officials. In late 1963 and
early 1964 thousands of innocent Tutsi were wantonly murdered in
what has been described as an act of genocide. Although the massacre
was in part the result of repeated provocations on the part of the exiled
Tutsi population, the scale and methods by which it was perpetrated
suggest that it can only be regarded as an extreme example of patho-
logical behaviour, as the blind reaction of a people traumatised by a deep
and lasting sense of inferiority. The heritage of humiliation created by
previous centuries of bondage, coupled with the frustration of trying to
adjust to new conditions while encountering new interference, provided
an ideal environment for aggressive behaviour.
All this is not meant to imply a situation of unadulterated harmony
among the castes of Burundi. Ethnic violence did occur in Burundi,
though never on a scale comparable to Rwanda. What was involved here,
moreover, was not so much an attempt on the part of a specific group to
reject or maintain a status of social inferiority as the expression of a
growing political competition between two culturally differentiated
segments of society. As one prominent ganwa observed, in 1957, 'in
Burundi social rank was determined by individual merit, regardless of
race, except for the Twa. ... Many ganwa gave their daughters in mar-
riage to Tutsi and Hutu alike.'52 Likewise, speaking on behalf of his
kinsmen, another famous scion of a princely family once confided to
Mgr Gorju: 'Do not be mistaken about our origins; our first ancestor
was a Hutu; we are all Bahutu.'53 From a Rwanda chief of similar rank






The Lands and the People
such a statement would have been inconceivable. Thus to associate the
occurrence of violence in Burundi with the outburst of a group whose
long-suppressed aspirations were bound sooner or later to lead to a
racial cataclysm would be profoundly misleading. Ethnic violence in
Burundi expressed a relatively low revolutionary quotient, and in any
event one that was so evenly shared by the protagonists as to make it
difficult at times to tell the insurgents from their opponents.
This said, one is nonetheless struck by the persistence in each state of
certain traditional patterns of behaviour. These are most clearly visible
in the mechanisms by which political clienteles are formed and cemen-
ted. In each state the building-up of a clientele remains the essential
prerequisite of influence. In each state the process involves a constant
probing of power relationships; it expresses itself in the judicious dis-
pensation of gifts to strategically placed individuals and the extension of
special favours in return for gifts; and it is generally governed by certain
tacit assumptions: that personal trust deserves a reward, either in terms
of political beneficee' or monetary gains, and that a higher reward may
well justify a change of heart from the recipient. Although they unfold
in different ethnic contexts and at different hierarchical levels, the
byzantine manoeuvrings involved in this kind of gamesmanship remain
the constant preoccupation of the aspiring and the ambitious.
It may be that by emphasising the differences discernible between
each state their similarities have been unconsciously glossed over. If
so it is worth reiterating that Rwanda and Burundi did share many
affinities, at least as great and as important as their differences. Both were
elitist, hierarchically-organised societies, in which power was concen-
trated at the top in the hands of a small oligarchy; both attached certain
physical and moral stereotypes to castes, and tended to associate quali-
ties of intelligence and resourcefulness with the upper strata; both
tended to equate power with wealth and wealth with cattle and land;
and in both societies the ties of clientage formed the basis of social
and/or political relations. Such being the case it would be naive to argue
that none of the problems that have arisen in Rwanda were ever likely to
be duplicated in Burundi. Nonetheless, that their initial patterns of de-
velopment after independence should have differed so strikingly from
each other has made it necessary for the purpose of this discussion to
play down their similarities and instead focus attention on their differ-
ences. Moreover, many of these differences were significantly enhanced
as a result of the policies and practices of the colonial powers. There is
scant evidence in support of the argument that the divergent paths to
independence followed by Rwanda and Burundi were necessarily fore-
ordained, as it were, by the structural and normative disparities of their






Rwanda and Burundi
traditional political systems. That these have had a bearing on the course
of subsequent developments is beyond question; but this occurred
largely as a consequence of European interpretations, or misinterpreta-
tions, and of the policies they have inspired. What these policies were,
and the effect they have had on the social structure of each society, is
what must now be examined.








2. Historical Survey


IN THE perspective of their recorded history, the period of foreign rule
experienced by Rwanda and Burundi was exceedingly brief. The
scramble had already engulfed most of Africa before the exploration of
the area even commenced. Subsequent efforts at penetration proceeded
so slowly as to prompt one scholar to remark, with perhaps only a touch
of exaggeration, that 'the beginning of the twentieth century saw the
passing of Ruanda-Urundi as the unexplored territory'.1 Yet, despite
the brevity of the colonial interlude, its impact was overwhelming. In
Rwanda it unleashed one of the most violent upheavals ever witnessed
by an African state at a similar stage of its evolution; in Burundi it
sowed the seeds of a racial conflict that may well prove equally devastat-
ing. It is one of the ironies of history, however, that the country which
has so far suffered the most radical transformation should also be the
one where traces of modernisation are least in evidence, and where the
support of traditional institutions was pursued with the greatest con-
sistency by successive administering authorities.
Although each society did not display the same degree of vulner-
ability to Western influences, the fact that both escaped the thrust of the
Arab intrusion made their initial contacts with the West all the more
shattering. At a time when the Zanzibar Arabs were already plying their
trade deep into the Congo, the people of Rwanda and Burundi continued
to live in a state of splendid isolation, owing as much to the natural
bulwark of swamps and mountains as to the fearsome reputation they
had earned among their neighbours. During his visit to Karagwe in
1876, Stanley was told by one Ahmed Ibrahim that the Banyarwanda
were 'a great people, but covetous, malignant, treacherous, and utterly
untrustworthy .... They have never yet allowed an Arab to trade in their
country, which proves them to be a bad lot.'2 Before long the Barundi
established a similar reputation for themselves: the first White Fathers
to set foot in Burundi were ruthlessly massacred at Rumonge in 1881.
Some ten years elapsed before the Austrian explorer Oskar Baumann
decided to venture into what later became known as Ruanda-Urundi-
this time with considerably more luck than his unfortunate predecessors.
In 1892 he crossed the Kagera into northern Burundi, and, after a quick
swerve through Rwanda, reached the Ruvubu river, which he mistook
for the source of the Nile. Baumann's was the first of a series of similar






Rwanda and Burundi
explorations by German officers, the most important of which, historic-
ally, was Count von Goetzen's expedition through Rwanda in 1894. It
was during his stay in Rwanda that von Goetzen met Mwami Rwabugiri
for the first time. The friendly welcome extended by Rwabugiri came
as a disappointment to von Goetzen, who apparently felt robbed of a
splendid opportunity to display his martial qualities; at the time a
lieutenant in the Second Royal Regiment of Ulans, von Goetzen's
comments were indeed worthy of his calling: 'Feeling strong and being
moderately equipped with weapons, we certainly would have liked to
cope once with a more serious enemy.'3 This opportunity would soon
present itself in Burundi, where the continuing struggle of the king
against rebellious princes confronted the Germans with a much more
difficult situation. The first encounter of Captain von Goetzen with
Mwami Mwezi Kisabo of Burundi took place in 1899; but the pacifica-
tion of the country by the Germans was not achieved until several years
later, at the cost of numerous military expeditions.
By then, however, von Goetzen occupied the more enviable position
of Governor-General of the East African Protectorate, in Dar es Salaam,
some six hundred miles away from the scene of operations. The impor-
tance of his new post was a fair measure of his previous accomplish-
ments. The results of his peregrinations across the continent appeared
in 1895, in his monumental Durch Afrika von Ost nach West, which
contained the first detailed account of Rwanda's geography. Eventually,
the dictum quoted in the preface of his book-'Initium Scientiae
Politicae Geographica'-found its justification in the founding of the
'MilitArstation' of Bujumbura in 1899. However, the newly-founded
military station was nothing more than a precarious outpost, and a
full decade had passed before the German government could make a
reasonable claim to effective control over the area.

THE GERMAN PHASE
Unlike the rest of German East Africa, where tribal dislocation offered
no other alternative but to rule through appointed local officials,
regardless of their traditional claims to authority, Ruanda-Urundi was
to be administered on the basis of indirect rule.4 This meant that 'by
degrees, and almost imperceptibly to the people ... the sultans [would]
eventually become nothing less than the executive instruments of the
Residents'. The attractions of this policy were practical and psycho-
logical. Apart from the fact that the centralised political systems of the
indigenous societies were admirably suited to indirect rule, any attempt
to displace the traditional rulers would probably have met with con-






Historical Survey
siderable resistance from the local populations; at all events, it would
have required a far greater number of European administrators than
was available at the time. This consideration takes on added weight
when one recalls that of all colonial administrations Germany's was
among the most grievously understaffed: as late as 1913, the whole of
German East Africa-a territory larger than Nigeria-was administered
by seventy European officials.6 Moreover, to the German officers on the
spot, virtually all of them of patrician origins, the claims of the Tutsi
aristocracy were no less sacrosanct than the privileges of the Junker
aristocracy in Bismarck's Germany. Their aristocratic leanings were
fully concordant with the postulate of indirect rule.
One of the earliest formulations of German policy is found in a
report of November 1905, by Captain von Grawert, then Resident of
Burundi: 'The ideal is: unqualified recognition of the authority of the
sultans from us, whether through taxes or other means, in a way that
will seem to them as little a burden as possible; this will link their inter-
ests with ours. This ideal will probably be realized more easily and earlier
in Ruanda, which is more tightly organised, than in Urundi, where we
must first re-establish the old authority of the sultan, which has generally
been weakened by wars with Europeans and other circumstances.'7 This
evaluation was based on more than mere conjecture however, for by
the time these lines were written von Grawert had already learned by
experience that this 'ideal' would not be realized with equal success in
each territory. While in Rwanda the implementation of indirect rule
came to reflect the logical forethought of von Grawert's formulation
almost to the letter, in Burundi German policies showed from the very
beginning a mixture of expediency and improvisation which in the end
led to surprising contradictions. The key to the situation lies in the
very different conditions then prevailing in each kingdom.

The Burundi Residency
When the Germans arrived in Burundi they found a situation
bordering on chaos. Internecine struggles between the ageing king,
Mwami Mwezi Kisabo, and the rebellious chiefs had reached the
point where power was divided among a host of princely factions,
with the king assuming the position of a 'potentate of limited power'.8
His very weak position at first prompted Kisabo to adopt a concilia-
tory attitude towards the Germans, but his promise of co-operation
later turned out to be nothing more than a facade. His real aim, as
far as can be determined, was to avoid a direct confrontation with
the German colonial troops so as to concentrate his energies against
his domestic foes.






Rwanda and Burundi
His principal challengers at the time were for the most part ganwa
of the Batare branch, the descendants of Mwami Ntare Rugaamba.
Mwezi Kisabo himself was one of Ntare's younger sons, but, as has
been noted already, his accession to the throne in 1860, so far from mut-
ing the claims of his elder brothers, resulted in a long yet inconclusive
fratricidal strife between Bezi and Batare-between Mwezi's supporters
and Ntare's marauding minions. The Batare were now firmly entrenched
on the periphery of the realm: Chiefs Kanugunu, Mbanzabugabo and
Busokoza controlled the eastern and north-eastern marches; Chief
Maconco held an important fief in the north;* but by far the most for-
midable of Kisabo's rivals was Chief Kilima, whose rule now extended
over a large chunk of territory in the north-west. And here and there on
the fringes of the realm stood a number of lesser chiefs whose authority
was exercised more or less independently of the crown.
The real identity of Kilima is somewhat obscure. According to lhis
own version, corroborated by the testimonies of his sons and subchiefs,
Kilima was the descendant of one of Ntare's sons named Nyanamus-
ango, who at one time had found refuge among the Bafulero in the Congo.
One of Nyanamusango's sons, Njitshi, married a certain Nabakile,
probably a Bafulero girl (said to have been 'presented' to him by the
king of Rwanda) who bore him a child named Kilima. In later years
Kilima concluded an alliance with a group of semi-independent chiefs
of the Ruzizi valley and with their assistance established his claims over
the north-western region of Burundi. According to one report: 'When
he learned that the Belgians had arrived, Kilima came down from the
high-lying mountain area between Kagobwami and the Ruzizi to meet
them. Since then his territory has stretched, in fact and by tacit agree-
ment, all the way to the Ruvubu.' At the time this report was written
(1918), Kilima was said to have been approximately sixty years old,
'with regular features, though slightly chubby'. His sons, Ruhabira,
Rusimbi, Kalibwami and Rwasha ('much less vulgar-looking than their
father'), were each given extensive tracts of land by the Belgian adminis-
tration; in 1918, Kilima claimed among his vassals Chiefs Gwinzo and
Kana, 'who boast about their Tutsi origins and look down upon (Chief)
Mugwaruso, whom they call "Muhutu"; they are themselves descend-
ants of Rukara, to whom King Ntare is said to have given land'.9

Maconco was the only rebellious chief who did not claim direct descent
from King Ntare: he is said to have been the son of a certain Kagaanza, maternal
uncle of King Mwezi. Kanugunu and his son, Mbanzabugabo, were the des-
cendants of King Ntare, through his son Ndivyariye. Busokoza was a cousin of
Kanugunu. For further information see the genealogical table in Appendix II,
pp. 504-8.






Historical Survey
According to Hans Meyer,* Kilima reconquered the northern region in
part with the help of his Congolese followers (whom he calls the 'Ban-
yambungu'), and in part by enlisting the support of local Hutu popula-
tions: 'When he was grown up, [Kilima] returned to his homeland, in
north-west Urundi, thanks to the support he had gained from the
Banyambungu. He quickly found followers among the Bahutu too, and
murdered all existing Batutsi there; that is why he is called "Batutsi-
killer".'"
The essence of the dilemma facing the German authorities was to
reconcile the conflicting claims of the king and the chiefs in a way that
would satisfy both. Yet by supporting the chiefs against the king they
risked the possibility of causing irreparable harm to the prestige of the
crown; and by throwing their weight behind the king they were bound
to antagonise the chiefs. In either case a trial of strength seemed
inevitable. But if the difficulties arising from these internal rivalries
were serious enough, the absence of co-ordination between the Govern-
ment-General in Dar es Salaam and the Residency in Bujumbura made
the situation even more intractable.
German policies in Burundi fall into three distinctive phases, each
reflecting a different assessment of how best to preserve the balance of
forces between the crown and the chiefs:
(i) From r899 to i903: a period of 'non-intervention', abruptly ter-
minated in 1903 by Captain von Beringe's military expedition against
Mwami Mwezi Kisabo and the recognition of Kilima and Maconco as
independent chiefs.
(ii) From 1903 to i908: a period of consolidation, marked by a partial
curtailment of the chiefs' independence, and culminating in 1905 with
the defeat and capture of Kilima.
(iii) From i9o8 to 19r5: a period of 'divide and rule', characterized by
an attempt to 'freeze' the status quo in such a way as to prevent the
crown from gaining a permanent ascendancy over the chiefs, and vice
versa.
From the outset, German policies revealed some fundamental dis-
agreements between Bujumbura and Dar es Salaam. For the German
Resident in Bujumbura, Captain von Beringe, the policy of non-inter-
vention repeatedly advocated by von Goetzen was self-defeating. In
the context of Burundi politics, argued von Beringe, indirect rule could
not be applied effectively until the king had made his submission to the
Residency, for only then would there be 'a basis on which to build in
Meyer's Die Barundi, largely corroborated by oral traditions, provides an
important clue to an understanding of the roots of anti-Tutsi sentiments cur-
rently encountered in northern Burundi; see infra, chapter 13.

51
3-RAB *






Rwanda and Burundi
Urundi an authority as strong and effective as the one in Ruanda'.11
Meanwhile the sporadic incursions of the king against his neighbours
would probably result in further anarchy. In a report of July 1902,
von Beringe described Kisabo as 'the sworn enemy of the Europeans'
and went on to request permission to launch a military expedition against
the Mwami. The object, he said, was not to depose Kisabo but to force
him once and for all to submit to the German authorities. Von Goetzen
withheld approval however, whereupon the Resident decided to act on
his own initiative. After securing reinforcements from Bismarcksburg
(Kasenga), von Beringe opened the hostilities against Mwezi Kisabo
in June 1903, and a few weeks later, on June 23, he triumphantly reported
to Dar es Salaam the story of his 'great success'. At long last Kisabo
had been brought to heel. In the course of the engagement two hundred
of Kisabo's men had lost their lives. In reparation Kisabo consented to
surrender 424 head of cattle; he was to open a road from Bujumbura to
Muyaga, and give all caravans free passage through his domain. More
important still, he was to recognize the independence of Chiefs Kilima
and Maconco, as well as the latter's claims over the royal fief of Mura-
mvya.12 As a result, 'the kingship, but, curiously enough, not the coun-
try, was divided among three claimants-Kilima in Bukeye, Maconco in
Imbuye, and Mwezi [Kisabo] in the Kiganda region'.13
Von Goetzen's reaction to the news was one of unmitigated fury.14
As one might have expected, von Beringe was 'called to other functions',
while his successor, Resident von Grawert, was asked to redress the
situation as best he could. More than ever von Goetzen was determined
to restore the authority of the crown; this, he thought, could best be
accomplished by gradual steps, with patience and diplomacy. Above
all, it was imperative to avoid 'brutal interventions'. However, this was
more easily said than done. In fact, von Goetzen's policy, while in
theory unexceptionable, ran foul of several unforeseen developments.
For one thing, by granting Kilima and Maconco their independence,
von Beringe had led some lesser chiefs, like Lussokossa in the north-
east and Lusengo in the Bugofi, to claim a similar status for themselves.
Before von Grawert even realized what was happening, the kingdom
was fragmented into half a dozen independent fiefdoms. Moreover, it
became increasingly clear that, as long as Kilima and Maconco insisted
upon enlarging their territorial holdings at the expense of the crown,
local resistance to these encroachments would result in further blood-
shed. As the situation threatened to deteriorate into complete chaos,
the intervention of the colonial troops became unavoidable. The most
brutal of such interventions was conducted against Kilima and Kanu-
gunu, in 1905. In October 1905 von Grawert penetrated Kilima's fief






Historical Survey
and destroyed every village in his path. Von Grawert handled this
task with professional conscience: 'The villages we occupied', he later
admitted, 'were all burned down; and to make sure the job was well
done I stayed in the region for a whole day.'15 Kanugunu's fief in the
north-east was subjected to similar treatment a few months later. In
1908 'a terribly bloody expedition', according to Pierre Ryckmans, 'was
launched against Chief Busokoza [Kanugunu's cousin], but came to
nought'.'1
The defeat of Kilima marked the beginning of a new course in Burundi
politics. Kisabo's most dangerous adversary was now in gaol; in May
1905 Maconco met his death in a suicidal attempt against von Grawert's
life. In October of the same year the Residency formally recognized
Mwezi Kisabo as the 'Sultan' of Burundi, with the assurance that 'as
long as he met the wishes of the Germans they would regard his enemies
as [their] enemies'.'7 Despite the continued resistance of Chiefs Mban-
zabugabo and Busokoza in the north-east, Burundi seemed well on its
way toward consolidation. The founding of the Burundi Residency, in
1906, signalled the restoration of Kisabo's authority and the beginning
of civilian administration8
But this was to be no more than a brief lull before the outbreak of new
turbulence. With the death of Kisabo in August 1908, Burundi politics
entered another phase of turmoil and confusion. Kisabo was succeeded
by his fifteen-year-old son, Mutaga. Though formally recognized as
the new Mwami of Burundi by the Residency, the adolescent king was
unable to check the centrifugal forces suddenly released by Kisabo's
death. In the north-east, Chiefs Lussokossa and Mbanzabugabo took
advantage of the situation to reassert their claims to independence;
simultaneously, many of the chiefs who had pledged allegiance to the
departed king now joined the ranks of the 'rebels'. Adding to the con-
fusion, Mutaga's own relatives wasted few opportunities to increase
their power at the expense of the crown. Among them were the regent
chiefs appointed by Kisabo on his deathbed to assist the new ruler-
Chiefs Ntarugera and Nduwumwe, two of Mutaga's elder brothers.
According to Hans Meyer 'these two and the Queen Mother [Ndiri-
kumutima]* took full advantage of their position to seize as much land
Ndirikumutima, one of the key figures in the entourage of the Mwami, was
later described by Ryckmans in these terms: 'An old woman, with all the
vices typical of her age, and endowed with a keen intelligence, she also displays
a stubborn obstinacy. Nothing can possibly pry her loose from her blissful
inertia. Once the wife of King Mwezi, who never cared much for contacts with
the whites, she represents with indomitable tenacity the spirit of the old autoc-
racy. She feels that the established order is the ideal and that everything we do
can only upset this order. Her life's sole and constant preoccupation is to plunder







Rwanda and Burundi
and property as was reasonably possible'.19 Eventually Ntarugera
became 'the greatest and richest man in Burundi, whose views [were]
listened to at Mutaga's court because he was feared'.20
By a curious coincidence, 1908 also marked a transition for the Ger-
man East African government. In that year von Rechenberg replaced
von Goetzen as Governor-General, and Resident von Grawert went
back to Germany on home leave. He was succeeded in Bujumbura by
Captain Fonck, whose diagnosis of the situation suggested a radically
different course of action from the one pursued by his predecessor. The
monarchy, said Fonck, was a pushover: under the circumstances the
only sensible policy for the Residency was to withdraw its support from
the Mwami and deal directly with the chiefs. Instead von Rechenberg
opted for a middle course, aiming at 'freezing' the status quo through
recognition of both the Mwami and the chiefs--of all chiefs. Accord-
ingly, Fonck was instructed to recognize, in addition to the Mwami,
three categories of chiefs: (i) those who were in fact independent, i.e.
Mbanzabugabo, Busokoza and Rwasha, Kilima's son; (ii) those who
had earlier recognized the authority of Mwami Mutaga and would
presumably continue to do so; and (iii) those who were 'more or less
independent'-a phrase covering a number of border-line cases to be
decided on the basis of their own merits.21
This policy, however, was difficult to reconcile with the rapid sequence
of events taking place in Burundi. Neither the king nor any of the chiefs
was willing to keep the static arrangement that von Rechenberg wished
to introduce. Nor was it always compatible with the attitude of the
German Residents: 'The Residents had little sympathy for the royal

other people's patrimony for the benefit of her sons. In her eyes the exercise of
power is reducible to taking property away from those who seem weak enough
to let her get away with it. ... She sees to it that the burden of our requisitions
falls upon the weak, so as to spare her favourites, her sons and the royal domain;
she plays host and gives moral support to rebellious elements. . And when she
senses that our patience is nearing the breaking point, she begins acting like a
poor old woman, sick and tired of the responsibilities of power. . .
Nduwume was said to be 'the most influential of her sons' and 'a worthy heir
to his mother's sentiments'; of Ntarugera, Ryckmans had a much higher
opinion: 'The administration could not find a more qualified regent during
Mwambutsa's minority than his great uncle Ntarugera. Once Mwezi's favourite
son, more knowledgeable than anybody else of the affairs of the realm, having
already exercised power over the region during his father's old age and at the
beginning of his brother's reign; . loved by all who feel threatened by
the ambitions of the ruling family (if only because he himself is the first to sense
the threat), this man [Ntarugera] will lend us the prestige of his authority because
he needs us in order to preserve it.' Observations sur le Rapport du ler Trimestre,
extract from a report by P. Ryckmans, 1918; in the Derscheid Collection.






Historical Survey
family; they tried to "divide and rule", to "play one chief off against .
the other" ',22 and although they frequently intervened on behalf of the/
chiefs, they did not make the faintest effort to restore the authority of
the crown. The standard argument advanced by one Resident after the
other was that the Mwami should first demonstrate the extent of his
authority by his capacity to resist aggression; only then could they decide
how much support should be given to the monarchy. That the Residency
in fact had no intention of helping the crown reassert its authority was
made abundantly clear by the release of Kilima in 1910: once regarded
as a dangerous 'rebel', Kilima was now welcome as a 'salutary counter-
weight to the disruptive influences of Mutaga'. The result was a further
weakening of the crown, a situation which Resident von Stegmann later
acknowledged with secret satisfaction-and no little exaggeration: 'The
Mwami has nothing to say except in his own village . His political
influence is non-existent; he exists because tradition says that he must;
but he is not the ruler of the country.'23 What von Stegmann neglected
to mention was that this state of affairs was the normal outcome of the
policy of 'divide and rule' consistently pursued by the Residency since
Kisabo's death.
Despite subsequent attempts to reverse the trend, the situation showed
few signs of improvement. For a while the transfer of the Residency-
from Usumbura to Kitega, in 1912, into the very heart of the country,
made it easier for the Germans to control the hinterland; it also brought
the administration closer to the court, and thus led to a more active
collaboration between the Resident and the Mwami. Mutaga was now
a grown man, and as his personal influence increased so did his willing-
ness to share in the responsibilities of government. But the terminal
years of the Burundi Residency also saw the recurrence of familiar
difficulties. The continued rivalry between the Queen Mother, Ndiriku-
mutima, and Ntarugera sapped the little that was left of Mutaga's
authority, until he himself became the victim of one of his nearest rela-
tives, in November 1915. The circumstances of Mutaga's death are
unclear, but according to Ryckman's version, which is the most reliable,
it probably came about as the result of a love affair between one of Muta-
ga's wives and his brother: 'Prince Bangura had become the lover of one
of his sisters! in law, married to King Mutaga. The latter eventually
became suspicious. He kept a close watch over his brother and one day
struck him in the chest with his spear. While trying to defend himself,
Bangura speared his assailant in the abdomen. Bangura preceded his
brother into the tomb by only a few days.'24
Burundi politics had reached their lowest ebb. In what seemed a
perfect re-enactment of the events following Kisabo's death, Mutaga





Rwanda and Burundi
was succeeded on the throne by his infant son, Mwambutsa, with
Ndirikumutima, Ntarugera and Nduwumwe in charge of the regency.
By the end of 1915, as the days of the German protectorate were quickly
drawing to an end, 'the situation in Burundi was perhaps worse than at
any time during the German era'.25
In his lucid commentary on German colonial rule in Burundi, Pierre
Ryckmans laid bare its essential shortcomings: 'On the eve of the
[First] World War the European administration was in a state of avowed
bankruptcy because it had worked toward the disintegration of a king-
dom whose traditions, mores and religion were unknown or ignored;
because in tolerating successful revolts, it had encouraged intrigues in-
stead of suppressing them; because each blow against the prestige of the
monarchy had rendered the white man more odious to the masses,
attached above all to the traditions of their divine monarchy.'2" In
pointing out the reasons for the unqualified failure of German policies
in Burundi, Ryckmans also hints at some of the more paradoxical
consequences of these policies. While the Residency undeniably en-
couraged the fragmentation of the realm into a mosaic of independent
fiefdoms, it also laid the foundation for a nationalist revival centred on
the crown. The dynastic implications of German policies were no less
important. As we shall see, much of the nationalist aura which in later
years surrounded the claims of the Bezi against the Batare can be traced
back to the resistance of their forefathers, and Mwezi Kisabo himself,
against the policy of favouritism pursued by the German Residents
towards the Bezi's arch-enemies, the Batare. In the minds of most
Barundi, the cause of national unity and monarchic legitimacy became
indissolubly linked with the cause of the Bezi.

The Rwanda Residency
The German record in Rwanda contrasts sharply with the erratic
course of their Burundi policies-a contrast perhaps over-emphasised by
Governor Schnee when he wrote, in 1913: 'The history of Urundi
since the German occupation is unfortunately not pleasant and contrasts
with the peaceful and pleasant state of things in Ruanda.'27 Those who
are familiar with the history of Rwanda know that Schnee's statement
cannot be taken too literally: only a year and a half before these lines
were written, the Rwanda Residency had launched an extremely brutal
expedition against a certain Ndungutse, until then one of the most
serious challengers of royal supremacy. But if German policies in
Rwanda were not always peaceful, violence was, in a sense, used more
'constructively' and hence less frequently than in Burundi. Instead of
dithering between the extremes of support and suppression of the,






Historical Survey
chiefs against the king, the administering authorities consistently and
successfully sought to bolster the authority of the king against his
rivals.
Unlike Burundi, Rwanda was a centralised state, in which the
Mwami was in fact and in theory the supreme ruler of the land. This
overwhelming concentration of power in the hands of a single individual
spared the Germans many of the difficulties they faced in Burundi.
Although the Mwami's authority did not go unchallenged, the con-
testants were neither so numerous nor so powerful as in Burundi;
there was no princely caste comparable to the ganwa, whose entrenched
rights and privileges might act as a counterpull to the powers of the
crown; and as long as the Mwami was willing to recognize the German
protectorate there was no need for the Residency to capitalise upon
internal divisions. Nor was the willingness of the court to co-operate
with the Germans entirely accidental. Given the circumstances of
Rwanda politics, the establishment of German rule was not without
certain reciprocal benefits. Just as the Mwami needed the support of the
Germans to enlarge the territorial bases of his authority, the Germans
needed the support of the Mwami to consolidate their protectorate
over the country and proceed with the tasks of administration. As it
turned out, the terms of the quid pro quo made it possible for the crown
to expand its hegemony far beyond the limits of its original jurisdiction.
It will be useful, for the sake of clarity, to take a brief glance at the
internal situation prevailing in Rwanda at the turn of the century-
a very unusual situation in many respects. Shortly after von Goetzen's
visit, in 1894, a period of factional strife set in, centred upon a disputed
succession to the throne and culminating in 1896 with the famous
coup of Rucuncu. In the annals of Rwandese history, Rucuncu marks
the beginning of a long civil war between the Bega and the Banyiginya
clans, in a way reminiscent of the protracted struggle between the House
of York and the House of Lancaster in fifteenth-century England.
The facts, briefly stated, are as follows. The death of Mwami Kigeri
Rwabugiri in 1895 had set the stage for a bitter struggle for the kingship
between the Bega and Banyiginya families. While being one of the most
prestigious of the several maternal clans from which a future Mwami
could be chosen, the Bega had fallen into disgrace under the reign of
Rwabugiri. Neither he nor his successor on the throne, Mibambwe
Rutalindwa, had any ties with the Bega, and although Rwabugiri had
consented that a Mwega named Kanyogera be appointed Queen Mother,
this concession apparently failed to satisfy the ambitions of her clan.
The Bega would not rest content until Kanyogera's own son, the future
Mwami Musinga, had taken Rutalindwa's place. In November 1896,







Rwanda and Burundi
at Rucuncu, the succession was finally settled to the advantage of the
Bega after a brief but bloody 'palace revolution' during which Ruta-
lindwa, his wife and three children lost their lives. Since the new
Mwami, Yuhi Musinga, was a minor at the time, his mother, Kanyogera,
acted as regent, together with his maternal uncles, Chiefs Kabare and
Ruhinankiko. With this triumvirate at the helm, the Bega then proceeded
to consolidate their hold over the country.28
As a first step towards this goal-and after the extermination of those
biru* who did not at once recognize the authority of the new Mwami-
the Banyiginya clan was systematically purged of its most influential
elements. At the same time every effort was made to 'fill the vacancies'
with trustworthy chiefs, most of them of Bega extraction. According to
Kagame, 'countless members of the defeated party were massacred on
the orders of Kabare and his acolytes ... and new chiefs were appointed
to fill the posts vacated by the death of the incumbents'.29 Yet the very
stringency of the repression led to the defection of some chiefs who had
initially cast their lot with usurpers, while others, whom Kagame refers
to as 'legitimists', i.e. pro-Banyiginya, sought refuge in the north and
the east.
With the spread of legitimist sentiment to the north, swells of unrest
began to penetrate into the region, culminating in 1911-12 with the
famous rebellion associated with the names of Muhumusa, Ndungutse
and Bassebya. Northern Rwanda, it must be remembered, was still more
or less independent from the crown. The 'annexation' of the region had
taken place a decade or so before the coup, under the reign of Mwami
Rwabugiri, and although Rwabugiri seems to have extracted a formal
pledge of allegiance from some of the local Kiga chiefs, their loyalty to
the monarchy was nonetheless highly questionable. Hardier and sturdier
than most ordinary Hutu, the Kiga people of northern Rwanda have
always been looked upon by their neighbours as a rebellious lot, fiercely
individualistic and contemptuous of established authority. It was among
these rugged mountaineers that one of Rwabugiri's wives, a presumably
full-blooded Munyiginya named Muhumusa, found refuge. She sought
to crystallise the support of a few local clans around a hard core of
legitimist chiefs, with a view to restoring her son Bilegeya to the throne
in Rwanda. The surprising thing about Muhumusa is not that she ul-
timately failed to carry out her plans but that she should have come so

The biru were the guardians of the esoteric code of the monarchy (ubwiru),
and thus it was especially important in order to establish the legitimacy of
Musinga that his claims be recognized by these officials. No matter how power-
less they may have been in deciding the outcome of the struggle, they alone could
give it a sanction of legitimacy. (See supra, chapter I, pp. 32-3.)







Historical Survey
near to realising them. 'Herself an outstanding personality', writes
J. M. Bessel, 'possessing great powers of leadership and organization,
and far more brains than probably any Mututsi woman before or since,
she was in intelligence quite up to the standards of her late husband.'30
Not only in intelligence but in ambition: in 19 1 she proclaimed herself
Queen of Ndorwa and promised her followers that she would soon liber-
ate the country from the yoke of the Europeans.
Although Muhumusa may have been the most prestigious representa-
tive of the legitimist faction, she was by no means the only one. Among
the Banyiginya chiefs who had sought asylum in the north, the most
famous was a certain Ndungutse, presumably Bilegeya's half-brother
through Muhumusa.* He and a Twa chief named Bassebya-described
by Father Dufays as 'a famous and fearsome brigand, rapacious and
sanguinary'32-instigated sporadic insurrections throughout the Rukiga
and in particular in the marshy country surrounding Lakes Bulera and
Luhondo, aided in this task by a Tutsi chief named Lukarra, better
known as the murderer of Father Loupias.32 After Muhumusa's capture
in September 1911, when she and a group of her followers were forced
to surrender to the British authorities of Bufumbiro (Uganda), Ndung-
utse emerged as the chief spokesman of the legitimist faction, insisting
time and again that Bilegeya was the sole legitimate heir to Rutalindwa.
He also claimed considerable authority in his own right. He came to be
viewed by the local populations as their saviour, as the prophet who
would restore peace to the country and free the labouring masses from
the servitude of the corv6e (ubuletwa), a servitude made all the more
intolerable after the additional tribute which the Bega had apparently
tried to extract from the northern populations.t Though himself a

Ndungutse's identity has been-and still is-a source of controversy among
historians. Some have claimed that he was the son of a princess from Ndorwa
named Nyiragumuhusa, sired by Mwami Rwabugiri; others have regarded
Bilegeya and Ndungutse as the same person, known first under the name of
Bilegeya and later as Ndungutse. The consensus of informed opinion, however,
is that Bilegeya and Ndungutse were two different persons. Though Ndungutse's
identity is still open to question, most historians seem to agree that Bilegeya
was indeed the son of Rwabugiri and Muhumusa.
t As Sandrart noted, in 1929, in his Rapport sur le territoire de Kigali, p. 45
(in the Derscheid Collection): 'The coming to power of the Bega led to a recru-
descence of corv6es and taxes; the new chiefs they had installed sought only to
enrich themselves at the expense of their people. Undoubtedly the Bahutu did
not voluntarily accept this somewhat burdensome fiscal regime. The most
telling proof of this is the enthusiastic response of the local populations to
Ndungutse's call in 1912, when he threatened to invade the [central] kingdom,
and the unanimous rallying of the peasantry to his cause at the mere mention of
suppressing the corvee.'






Rwanda and Burundi
Tutsi, Ndungutse's name became a symbol of anti-Tutsi sentiment, and
by implication of anti-European sentiment as well.
Ndungutse's appeal as a leader was closely linked to the emergence in
the north of the Nyabingi cult, a magico-religious cult which Muhu-
musa had used as a vehicle for propagating her ideas and solidifying
her support among the people of the Mulera, Bukonza and Buberuka
regions (often collectively referred to as the Kiga). Quite apart from its
original eschatology, the Nyabingi developed a strong political attraction
for the Kiga, which became apparent not only in Rwanda but in parts of
Uganda. 'It seems clear', notes Professor Edel in her discussion of the
Nyabingi movement in Kegezi (Uganda), 'that this force had become a
military arm, in rebellion against the constituted authority of the
attempted conquest by the Court of Rwanda.'33 More will be heard in a
subsequent chapter about the Nyabingi movement (see chapter 3,
pages 1oo-ioz); suffice it to note, for the time being, that it is against this
background of messianic activity that the roots of Ndungutse's popu-
larity must be understood, for not only did he pose as the safest ally of
the Kiga against the exactions of the ruling dynasty, but because of the
skill with which he managed to exploit the religious superstitions of his
followers a close organic relationship developed between certain segments
of the northern populations and the survivors of the Banyiginya clan.
As the movement became increasingly xenophobic in character and
aggressive in its methods, the 'hands-off' policy which the Residency had
heretofore advocated was no longer tenable. After the murder of Father
Loupias by Lukarra in 1910, the acting Resident, Gudovius, decided to
organise a punitive expedition in the north. The purpose was 'punish-
ment of the insubordinate districts and their peoples and chiefs by
causing the greatest possible damage until complete submission; other-
wise destruction of crops and settlements, and occupation of the theatre
of operations by chiefs appointed by the Resident who are faithful to
Musinga'.34 The expedition turned out to be an unqualified success on
both counts. In April 1912 Gudovius's troops attacked Ndungutse's
village, near Ruhengeri, killing about fifty defenders including Ndung-
utse himself. Military operations were later prosecuted by Lieutenant
Linde, who carried out his grim assignment to the letter: villages were
burned down, crops and settlements were destroyed, and all who resis-
ted were massacred. After the appointment of 'loyal' chiefs to rule over
the devastated area, Gudovius could boast that 'complete peace had
been restored to the country where Ndungutse and Bassebya and their
followers rebelled against Musinga'.35
Subsequent events demonstrated the precariousness of the Pax
Germanica in northern Rwanda. For many years after the Belgian







Historical Survey
'reprise' the northern region remained the scene of recurrent outbreaks
against the chiefs and the administration. However, Gudovius's expedi-
tion did consolidate the position of the monarchy in an area where it
had never been firmly established. In return the Germans 'could con-
tinue to count on the complete loyalty of Musinga, and could be assured
that he would try to fulfil every wish of the Resident'.36
More than anything else, the German Residents wished to use the
presence of European missionaries to educate the Tutsi chiefs and thus
convert them not only into good Christians but into efficient administra-
tors. This, however, is precisely where the Residency ran into difficulties.
At first the German Residents felt obligated to heed Musinga's request
that missionary activities be kept at a safe distance from the court,
lest the influence of Christianity weaken the authority of the crown. But
as they came to realise that the appeal of Christianity was strongest
among the Hutu, and discovered that the missionaries, on humanitarian
or religious grounds, endeavoured to restrain the abuses of the chiefs,
the Residents became understandably concerned over the possible reper-
cussions of missionary work. In their view 'missionary interference'
threatened the very basis of indirect rule, if only because the priests
sometimes meddled in 'the entirely internal disputes of the natives',
as Resident Kandt once put it.87 Furthermore, the predominance of
French elements among the Catholic clergy naturally held disquieting
implications for the German administrators, and in turn some of the
French missionaries became extremely suspicious of the Residents'
motives.* Apart from having led to a number of very ungainly incidents
between the Catholic Church and the secular administration, some of
which received abundant coverage in the writings of certain missionar-
ies,38 the foregoing circumstances also help to explain the more or less
systematic effort made by the German Residents to keep Catholic
missionary activities confined to the outer fringes of the realm-to those
And this not only on grounds of nationality but sometimes on cultural
and religious grounds. Commenting on Richard Kandt's Jewish origins,
Father A. Van Overschelde, gives this revealing portrayal of the German
Resident: 'Richard Kandt was a Jew, very intelligent, occasionally dabbling in
poetry, short, anaemic-looking, olive-complexioned. The bile, to which he
owed his complexion, did not run only under his skin: he was evil-minded.
His small stature, and perhaps also his ancestral habits, did not predispose him
to act in a straightforward manner. He excelled at tearing things down in the
dark, slyly, gingerly, like a cat.' A. Van Overschelde, Un Audacieux Pacifique:
Monseigneur L. P. Classe, Ap8tre du Ruanda, Grands Lacs, Namur 1948, p. 70.
For an excellent discussion of the relationships between the Church and the
European administration, both German and Belgian, see Alison Des Forges,
"Kings Without Crowns: The White Fathers in Ruanda", in Boston University
Papers in African History, Vol. in, Boston University Press, Boston 1967.






Rwanda and Burundi
regions where Tutsi rule was least stabilised. Except for the Save mis-
sion, located near Butare (formerly Astrida), the first Catholic missions
were established in 1900 in Nyundo and Zaza, both in the extreme
north. Because the missionaries were at times mistaken by the local
populations for the agents of the court, some of them (like Father
Loupias) became the target of the rebellious movements; at this point,
however, European solidarity prevailed over sectarian differences to
induce the Residency to take 'protective measures', usually in the
form of punitive expeditions. Thus, directly or indirectly, the missionary
presence became a major factor in the pacification of Rwanda. However,
if the circumstances of missionary penetration at first contributed to
strengthen both European and Tutsi over-rule in the north, in later
years the spread of Christianity among the Kiga served as a powerful
vector of revolutionary sentiment.
In sum, the impact of German rule upon the traditional political
structure of Rwanda was precisely the reverse of what happened in
Burundi. We have seen how, in Burundi, the policies of the Residents
tended to accentuate the trend toward fragmentation already present
in the traditional political organisation; how they favoured the emer-
gence of a pleiad of more or less independent fiefdoms and eventually
reduced the position of the mwami to that of a minor chief. In Rwanda,
on the other hand, every effort was made to strengthen and consolidate
the position of the crown. In either case punitive expeditions were the
chief instrument of German policy, but in Rwanda these were directed
against the mwami's opponents whereas in Burundi the mwami was
more often the victim than the beneficiary of German militarism. In
Rwanda the very success of indirect rule reinforced the absolutism of the
monarchy, and hence the hegemony of the ruling caste; in Burundi
the initial shortcomings of indirect rule enhanced the pluralistic bent of
the political system, and in the long run contributed to the softening
of caste antagonisms.
Since the entire period of German over-rule was so largely devoted to
the conduct of punitive expeditions, it is small wonder that relatively
little was accomplished in the realm of civil administration. True, an
effort was made to organise an administrative superstructure patterned
on the subdivision ('bezirken') already in existence in other parts of the
protectorate; separate Residencies were eventually established for each
kingdom; a rudimentary judicial system was implanted; and, beginning
in 1912, some partially fruitful attempts were made to collect taxes. In
fact, however, the administrative machinery set up by the Germans did
not amount to more than a few strategically located 'police posts'. The
very paucity of administrative personnel employed by the German






Historical Survey
government illustrates better than any lengthy enumeration how little
was done during this period: by 1914 the entire administrative-military
staff of the Rwanda Residency consisted of ten German nationals; the
Burundi Residency had only six civil authorities. Under the circum-
stances one is inclined to agree with Professor Louis that the successes of
German colonialism are more surprising than the failures.39 But the
failures cannot be overlooked. The geographical remoteness of Ruanda-
Urundi, the dearth of administrative personnel, along with the incurable
tendency of most Residents to resort to force rather than persuasion,
account for the more prominent shortcomings of German colonialism:
the absence of a viable administrative machinery at both the central and
local echelons, the crudeness of the judicial system, the inadequacy of
the economic infrastructure and the limited extent of the communica-
tion network. These in turn lend a measure of justification to Professor
Marzorati's statement that 'owing to its out-of-the-way position Ruanda-
Urundi had not received any methodical care until the arrival of the
Belgians. The Belgian government had therefore been obliged to work on
virgin soil.'40

THE BELGIAN MANDATE
The establishment of military rule through Ruanda-Urundi in 1916 was
not only a normal epilogue to Belgium's victories in East Africa but a
necessary condition for the realisation of its ultimate political objectives.
'One of the goals of our military effort in Africa', said the Belgian Minis-
ter of Colonies, Jules Renkin, in 1916, 'is to assure possession of German
territory for use as a pawn in negotiations. If, when the peace negotia-
tions open, changes in possession of African territories are envisaged,
the retention of this pawn would be favourable to Belgian interests from
every point of view.'41 Prior to the Peace Conference, the Belgians had
cast about for territorial concessions from Portugal, on the southern bank
of the Congo river, in exchange for the territory they had conquered
during the East African campaign. But their hopes failed to materialise.
To the chagrin of Belgian statesmen, who thought it rather meagre
compensation for their contribution to the war effort, the Milner-Orts
agreement of May 30, 1919, left Belgium with Ruanda-Urundi and gave
Britain the lion's share.
It was not until 1925, however, that the administrative status of the
mandated territory was finally settled. On August 21, 1925, the Belgian
government enacted a law providing for an administrative union between
their newly-acquired mandate and their Congo colony. Commenting
upon the practical implications of the merger, the Belgian representative







Rwanda and Burundi


to the Permanent Mandates Commission stated in October 1925:
'Ruanda-Urundi will take its place on a footing of the most complete
equality side by side with the four Congo provinces and will enjoy all
the benefits of the large measure of decentralisation possessed by those
provinces.... The Belgian government thought it good, in the interests
of the population of Ruanda-Urundi, not to double the already large
central services and the technical and medical services established at
Boma, but, thanks to the administrative union, to extend the working
of these services to the mandated territory.'42 Although Ruanda-
Urundi was now for all intents and purposes an appendage of the
Belgian colony, five or six more years would elapse before it could enjoy
the full benefits of a civil administration.* In the meantime the day-to-
day tasks of administration remained largely in the hands of the military
and at first did not extend very far beyond the immediate requirements
of peace and order.
There are several reasons for the incredibly slow pace at which Bel-
gium moved along the road to initiating administrative reforms. The
customs and institutions of the indigenous societies of Ruanda-Urundi
were unlike any found in the Congo; their social and political organisa-
tion seemed unusually, perhaps unnecessarily, strange to the Belgian
officers on the spot; the latter, moreover, by virtue of their training and
background showed little concern for the social and political problems
connected with the tasks of colonial administration. In addition-
and this is a point which Belgian officials repeatedly stressed before the
Permanent Mandates Commission-the administrative machinery
The first systematic attempt towards the introduction of a uniform system of
administration was made in 1929, when, at the request of the Minister of Colonies,
a series of general administrative inquiries was conducted in each of the adminis-
trative subdivisions (territoires) of Rwanda and Burundi. (These documents,
many of which can be secured through the Derscheid Collection, constitute one
of the most valuable sources of information for the student of 'native administra-
tion' in Ruanda-Urundi.) As a result of this investigation a set of general
instructions was issued by Vice-Governor General Voisin, in 1930, which
specified the goals of Belgian policies in Ruanda-Urundi in these terms:
I. To respect and reinforce native authority insofar as it is exercised in har-
mony with civilising directives.
2. To exercise a close check on possible abuses regarding customary tithes
presentationss] and compulsory labour [corvees].
3. To replace incapable chiefs with candidates designated with the accord of
the Mwami.
4. To regroup chiefdoms in such a way as to suppress the dispersion of fiefs
and make the administration easier and more efficient. The European personnel
must realise that without the collaboration of native authorities the occupying
power would be impotent and faced with anarchy.
See Historique et Chronologie du Ruanda, Kabgaye?, n.d., p. 25 ff.






Historical Survey
Germany had left behind was so rudimentary and inadequate that it
could serve only as a makeshift arrangement pending the introduction
of a new system. One Belgian spokesman carried the argument a step
further, intimating that the German record in Ruanda-Urundi did not
show a single creditable achievement, and that, consequently, Belgium
had been forced to make a completely fresh start: 'The Belgian mandate
had been set up in a country which had for practical purposes never
really come under European supervision-where there was but the
embryo of an administrative occupation and no European business
interests at all. It was the only mandated territory whose history had
commenced for all intents and purposes with the inauguration of the
mandate, and where the mandate experiment was not influenced by any
colonial past.'43
Nevertheless, the early years of the Belgian mandate bore the un-
mistakable imprint of the German legacy. As in the days of the German
protectorate the military was entrusted with a wide range of administra-
tive functions; and, like their German counterparts, the Belgian Resi-
dents were forced into a variety of roles, acting as trouble-shooters,
judges, counsellors and law-enforcing agents, and relying for assistance
on a mere handful of European deputies, 'whose sphere of action depen-
ded on the Resident and varied according to political circumstances'.44
While new regulations were eventually introduced to replace the German
legislation, until 1925 the fundamental law of Ruanda-Urundi was
German law (even though some Belgian officers candidly admitted
that they knew nothing of the German legal system). The same heavy
reliance of precedent can be seen in the early statements of Belgian
policy. 'Belgian policy', stated an official report of 1921, 'draws its
inspiration from the line of conduct followed earlier by the German
authorities: to insure peace and public order while maintaining the
existing balance between native groups.'46 One finds here an echo of the
principles earlier enunciated by Richard Kandt: 'Our political and
colonial interests require that we support the kings] and uphold the
extreme dependence of the great mass of the population. Considering
the nature of the country and the character of its people, this arrange-
ment can be reconciled with those humanitarian imperatives which
require the elimination of abuses of power and arbitrary rule over sub-
ject populations.'46
Not until 1925 did the Belgian version of indirect rule receive some-
thing approaching an official formulation. The core of the Belgian
doctrine is found in the 1925 Rapport sur l'Administration du Ruanda-
Urundi, in a passage which clearly reveals the authorship of Pierre
Ryckmans, Belgium's first Resident in Burundi:






Rwanda and Burundi
The co-operation of the kings constitutes an indispensable element
of progress and civilisation.... Without them the problem of govern-
ment would remain insoluble. There are among the chiefs some who
are incapable, imbecile, who will never gain authority. . There
are some who are irreducibly hostile and who will never accept
civilisation. . These chieftaincy crises are everywhere the great
stumbling block of native policies. To resolve them by dismissing a
bad chief and appoint in his stead one more amenable to European
influences is tantamount to substituting impotence for insubordina-
tion. Legitimacy is a moral factor of incalculable importance.47
These ideas were further elaborated upon in an article which appeared
in the Bulletin of the Soci&t6 Belge d'Etudes et d'Expansion, in 1925,
in which Ryckmans again emphatically stressed the importance of
legitimacy: 'Legitimacy is more powerful than violence. The only
smoothly functioning organ between us and the masses is the legitimate
chiefs. They alone, because they are legitimate, can induce acceptance of
necessary innovations.' While again conceding that some chiefs may
well turn out to be 'incapable', 'imbecile' or irreduciblyy hostile', he
also urged the greatest caution in the handling of these situations.
Removal from office was an extreme measure, to be adopted only in
the very last resort, after all other solutions had failed. In any event,
every effort should be made to respect the bami's authority and personal
prestige:

The presence of the king, the only one capable of conferring a legal,
customary investiture upon a candidate of our choice, makes it possible
for us to go forward without running the risk of being faced with a
fatal impasse, without having to make an impossible choice between a
rebellious legitimacy and an impotent submission.... It is therefore
not because of a pure love for tradition or local colour that we keep
the native kings. Let their powers be curtailed if necessary, but let
no one challenge their existence and outward prestige.48
In arguing his case Ryckmans had constantly in mind the example of
previous Belgian policies in the Congo, where the wholesale removal of
legitimate chiefs and their replacement by what he called 'chiefs of the
whites' had had disastrous consequences for the European administra-
tion as well as for the Africans. In contrast, he regarded the Belgian
position on Ruanda-Urundi as 'a privileged one'-'for we have kings'.
The kings would act as the prime legitimisers of Belgian colonial
policies and practices, or, as Ryckmans put it, 'as the familiar d6cor
which permits us to act in the wings without alarming the masses'.49






Historical Survey
Yet only a year after his plea, hundreds of Rwandese chiefs were dis-
missed from office and in some cases temporarily replaced by chiefs
of Hutu extraction. In 1931, Mwami Musinga of Rwanda was deposed
and replaced by his son, Mutara Rudahigwa. Meanwhile, in Burundi a
number of Batare chiefs-those very chiefs who only a few years earlier
had openly flouted the authority of the crown-were officially recognized
by the Belgian administration and granted a status of legitimacy similar
to the chiefs of the Bezi branch.
How should one account for such radical departures from the Ryck-
mans doctrine of legitimacy? First, by pointing to a fundamental flaw
in his reasoning, based as it was on the mistaken assumption that the
sanction of legitimacy expected of the kings would always be forth-
coming, almost automatically and regardless of the circumstances. In
evaluating the role of kingship-and chieftainship-he drew heavily
from his own experience and knowledge of the Burundi situation,
assuming in effect that the king and the chiefs formed two distinctive
political entities, which could be dealt with and manipulated more or
less independently of each other. While this may have been true of
Burundi, in Rwanda no such rigid dichotomy could be drawn, at least
as long as Musinga was in power. Moreover, Ryckmans greatly over-
estimated the area of compatibility between his own definition of legiti-
macy and the requirements of administrative efficiency, here again
basing his estimate almost exclusively on what he had learned in Burundi.
Thus it is not entirely by accident if the Ryckmans doctrine should
have worked out so much better in Burundi than in Rwanda, though not
always in the way that one might have imagined.
Before going any further it will be useful to distinguish at least three
levels at which Belgian policies have operated-the levels of kingship,
of chieftainship and of caste relations. In practice these different areas of
policy-making were not always clearly distinguishable, but for purposes
of analysis the institutions to which they refer must nevertheless be
treated as separate entities.
If, in Rwanda, the application of the Ryckmans doctrine was chal-
lenged from the outset by the all-encompassing, absolutist character of
the kingship, and the personal obduracy of Mwami Musinga, in Burundi,
where the kingship was already very weak and the office-holder still in
his teens, the results were decidedly more encouraging. As mentioned
earlier, the institution of kingship in Burundi was almost subsidiary to,
or at least on an equal footing with, that of ganwa-ship. Thus the really
important question was not so much whether or not to recognize the
authority of the incumbent king as to decide which of the two principal
dynasties-the Bezi or the Batare-should prevail over the other.






Rwanda and Burundi
Moreover, since Mwambutsa was as yet too young to assume the
responsibilities of kingship, a regent had to be chosen. Having learned
the lessons of the German experience, Ryckmans correctly saw that on
each of these counts political stability depended on reconciling the con-
flicting claims among the princes. For him this meant recognizing the
authority of the princes (Bezi and Batare) in their respective fiefs while
at the same time working towards a compromise at the centre. To
achieve this objective he resorted to a device rooted in tradition but
modified to suit the purposes of his policy; this was the Regency Coun-
cil, whose membership, after the death of Regent Ntarugera, in 1921,
was enlarged to include representatives of both the Bezi and Batare
dynasties. Ryckmans's initiative provided a basis for institutionalising
the interests of the princes in a way which at first led to compromise
and co-operation: 'The results obtained in Urundi are quite remarkable',
stated an official report in 1925; 'this kingdom, so profoundly divided at
the inception of Belgian occupation, has now regained its former boun-
daries. Five years of peaceful efforts have accomplished what six years
of warfare had failed to produce.'50 However, a closer look at the situa-
tion suggests a somewhat more guarded optimism. Though obviously
less apparent than in earlier times, the divisions between Bezi and Batare
were nonetheless real. By extending recognition to the Batare chiefs-
and admitting one of their representatives (Mbanzabugabo) to the
Regency Council in 1922-Ryckmans pursued a policy which most of
the Bezi viewed as contrary to their interests, as well as to the interests
of the crown. They felt, with justice, that the status quo enforced by the
Residency had been made 'legitimate' by the force of circumstances
(i.e., by the fiat of the administration and Mwambutsa's minority),
not by a royal ordinance from above. To revise this status quo and tilt
the balance of forces to their advantage became the main preoccupation
of the Bezi chiefs in subsequent years.
SIn Rwanda, the centralised character of the political system made
administrative control at the top relatively easy. On the other hand, the
imposition of administrative limitations upon the kingship was rendered
extremely difficult by the despotic nature of the office, the quasi-
pathological suspicion of the king towards Europeans in general, and the
maze of intrigue and double-dealings surrounding the court. As long
as this situation was allowed to persist, one Belgian administrator
contended, nothing but chaos would result.

The root of chaos lies in the existence of an autocratic, powerful
monarchy which, moreover, remains captive to the whims of a band
of sycophantic courtiers. The power of the crown has fluctuated back






Historical Survey
and forth in the hands of an ensnaring circle of favourites, and the
constant pressures arising from their sedulous courtship has brought
them lavish favours in return. Always close to the eyes and heart of
the sovereign, he who wished to hold on to what he had gained by
dint of attentions and vile flattery would take painful care to stay
exposed to the generous if inconstant rays of the monarchical sun.
Being absent from the court inevitably entailed all kinds of accusa-
tions and slanders, which in turn led to spoliations and often to the
absentee's physical liquidation. The result: a cabal of paramount
chiefs permanently stationed at the court who meanwhile found it
convenient to entrust the administration of their lands to their
delegates r'intendants concessionnaires'].51

The crisis came to a head in 1931, with the dismissal of Musinga and
the accession to the throne of his eighteen-year-old son, Rudahigwa,
thenceforth known as Mwami Mutara. The origins of the crisis, accord-
ing to one version, can be traced back to 1926; in that year 'Musinga
had attempted sodomy with sons of his nobles serving at the palace;
this caused a breach between him and the chiefs, for which he blamed
the White Fathers, saying they had exaggerated the facts'.52 Whether
there is any truth to these claims is difficult to say, but what is beyond
question is that Musinga's relations with the Catholic Church had never
been very cordial; according to Kagame, 'Musinga was bitterly opposed
to missionary enterprise, on the grounds that it undermined his
authority'.53 Interestingly, Musinga is reported to have said to one of his
biru that 'the Catholics were the most dangerous because they had
adopted the practice of making sacrificial offerings to God, unlike the
Protestants who recognized the authority of a supreme chief'. 4 While
openly distrustful of Catholic missionaries, Musinga had all along
adopted an attitude of passive resistance towards the administration, an
attitude which Professor Marzoratti attributed to the fact that 'at the
time of the occupation [Musinga] had been long subject to native
influence': 'It has therefore been difficult to change his mental outlook;
the limitation of power which the administration had imposed upon him
had contributed to his hostile attitude and latent surliness.'56
In support of Marzoratti's argument one might point to the much more
'co-operative' attitude later displayed by Mwami Mutara, and this in
spite of certain personality traits and inclinations clearly reminiscent of
his father's. At a time when Rudahigwa was still an ordinary chief,
in the Marangara province, a Belgian administrator gave the following
revealing portrayal of the future mwami: 'Very intelligent but is totally
lacking in character. Knavish and devious. Has been too long under the






Rwanda and Burundi
pernicious influence of the court. Cleverly conceals his true feelings
towards Europeans . Remains deeply attached to old traditions and
to a conception of power which claims for those who hold it all conceiv-
able rights over their subjects and no corresponding obligations and
responsibilities. While feigning indifference towards the missionaries
it is safe to say that Rudahigwa, without ever admitting it, entertains
considerable hostility towards them.'56 Although, in general, Rudahigwa
displayed no more sympathy towards Europeans than his predecessor,
his methods were very different. Where Musinga openly challenged
missionaries and administrators, Rudahigwa had a special talent for
working within the confines of the established superstructure, rather
than against it. He had come of age at a time when European rule was
already firmly established, and was better able to accommodate himself
to the norms of the new system. More important still, being himself the
product of mission schools, he shared with the up-and-coming genera-
tions of Tutsi elites the Western training and education which his
predecessor so conspicuously lacked.
Musinga had never been able or willing to work out a mutually
satisfactory relationship with the incipient 'class' of mission-educated
Tutsi chiefs and subchiefs, and the resulting tensions also played a part
in his dismissal. The new generations of chiefs had fully sensed the sig-
nificance of the social forces that lay behind the spread of Christianity.
They felt that the preservation of their traditional claims ultimately
depended upon their endorsement of the new creed. For Musinga,
however, the adoption of Christianity was seen as nothing short of a
betrayal of his authority. Christianity meant the desacralisation of
mwami-ship, and the relegation of the office-holder to a subordinate
position. He believed that, 'since he was no longer able to kill whom he
pleased, or even retain his followers in the traditional cult, he had lost
all his powers, and the missionaries were now more powerful than
himself'.57
Musinga's obduracy stemmed from his own unwillingness to accept
any sort of limitation that would alter either the sacredness of his office
or the structure of authority from which kingship derived its omni-
potence-a structure which, with its overwhelming concentration of
power at the top and multiple hierarchies of offices, leading to an
extraordinary atomisation of power at the base, was quite unsuited
to carry the burdens imposed upon it by the administration. The drastic
alterations eventually wrought into the system in the name of administra-
tive efficiency (and for which there has been no parallel in Burundi)"
made Musinga all the more suspicious of Belgian intentions.
It is at the chieftaincy level that the impact of administrative re-







Historical Survey
form had its most devastating effects. In Burundi, the issue of chief-
taincy was settled at an early date and, in the beginning at least, without
too much difficulty. As a corollary of the Ryckmans doctrine of legitim-
acy, the ganwa were given a relatively free hand to administer their
fiefs as they pleased; and, since chiefly rule was synonymous with
ganwa rule, this meant that the authority of the chiefs was never directly
threatened by the administration. Some chiefs in fact managed to
arrogate to themselves almost absolute powers in their provinces:
whether they paid or did not pay their taxes was entirely up to them.*
And, while the European administrators could generally tell the 'good'
chiefs from the 'bad' ones-a favourite dichotomy of Belgian officials-
the impression one gets is that, at least until 1929, the sphere of ganwa
politics was entirely beyond the ken of the administration. For many
years thereafter, 'non-interference' remained one of the guiding prin-
ciples of the Belgian administration.
In Rwanda, however, non-interference spelled protracted chaos.
There the triple hierarchy of political office at the local level confronted
the administration with a host of difficulties. The Belgians had no pre-
cise understanding of the functions ascribed to the land chief, the cattle
chief and the army chief; a rather illuminating, though not entirely
accurate, view of the system is found in the following administrative
report:
Chief Nduwumwe, Mwambutsa's uncle, is a case in point. According to
Pierre Ryckmans: 'Because of his close relationship with the king, he is virtually
exempt from corv6es. Not a single carrier, not a single load of food, not a single
head of cattle, not a single tax-payer, comes from his immense domain. His only
contacts with the administration concern the prestations asked by the "chef
de poste" of Kogamwami, which he takes great care in presenting as harmful to
the king's authority and incompatible with the dignity that is expected from the
"chef de poste" of Kitega.' Observations sur le Rapport Politique du ier Trimestre,
extract from a report by P. Ryckmans, 1918; in the Derscheid Collection.
Besides being the object of greater deference on the part of the Belgian
administrators, the very high rate of turnover among Belgian officials made it
extremely difficult for them to familiarise themselves with ganwa politics. Which,
incidentally, prompted one official to lament: 'The Governor has explained to
me that in the general interest and in the interest of colonial civil servants it
was advisable to see that territorial administrators and district commissioners
do not stay too long in office in their respective territories and districts. I am
flabbergasted by this theory which, I am given to understand, is currently applied
by the British in their territories, where the Governor has just spent a week. ...
This is, of course, the theory of the interchangeability of civil servants, which
tends to transform the administration into an impersonal, rigid organism, lacking
all drive and enthusiasm, and yielding negligible results. ... I would be less than
frank if I did not say that I feel rather disenchanted and that my enthusiasm
for the colony is waning.' Personal letter of Oger Coubeau to J. M. Derscheid,
July z2, 1933; in the Derscheid Collection.






Rwanda and Burundi
A greater or lesser number of hills falls under the jurisdiction of a
provincial chief, but the hills are in turn entrusted to say, 5, 8 or 15
subchiefs who keep him under their surveillance, always ready to
report to the king the slightest of his mistakes. Next to the provincial
chief, called the ubutaka chief, one finds another provincial chief
who, like the previous one, is entirely dependent upon the king but
whose authority concerns the cattle, the land and the Batutsi. He too
is subject to the surveillance and denunciations of his subordinates.
He is the umukenke chief. This duality of commands inevitably leads
to chaos. Finally there emerges a third fellow next to the others, the
ingabo chief. In theory he is an army chief; in fact he is merely another
patron who will also demand prestations and who, in return for gifts,
will act as an amicus curiae in the course of judicial proceedings and
as an intermediary to the king.58

Because the Belgian administration believed that the division of authority
would lead inevitably to conflict and anarchy, in the interests of adminis-
trative efficiency the Residency decided in 1926 to 'streamline' the struc-
ture of local government, and to replace this cumbersome trinity of
powers by the rule of a single chief.59
This measure struck at the very roots of Rwanda society. While it
did provide a temporary solution to the immediate problems, it is a
moot point whether the short-run benefits were worth the ultimate
risks it implied. For Kagame, it was the elimination of the army chiefs
from the traditional power structure, and the resultant abeyance of the
military code, which led to all the abuses associated with the buhake:
'The army was the basic social organisation which ensured to each
individual the enjoyment of his property, in return for certain obliga-
tions; it gave him the ready assistance of a public defender [avocat]
in the person of the army chief, who was obligated to defend him before
every tribunal, including the mwami's.'s0 If this picture seems somewhat
overdrawn, it is entirely plausible to assume that, by destroying the
pre-existing balance of forces on the hills, the 1926 reform prepared
the ground for the emergence of a more starkly authoritarian system,
centred on the rule of a single and virtually omnipotent chief. That
it dealt a telling blow to the prestige of Musinga is equally clear. Not
only did it deprive him of further opportunities to play off one chief
against another, but it also gave him a clear hint that from now on his
authority would depend upon the will of an alien power.
Far more significant historically-and infinitely more disquieting in
terms of the overall objectives of Belgian policies-was the concommit-
tant attempt made by the Residency to substitute Hutu chiefs and sub-






Historical Survey
chiefs for the Tutsi incumbents, a move apparently dictated by the resist-
ance of the more conservative chiefs to the 1926 reform.61 However,
the revolutionary implications of this initiative caused the greatest
misgivings among Catholic missionaries, some of whom did not hesitate
to voice their concern over 'the vacillation of the colonial authorities
with regard to the traditional hegemony of the well-born Tutsi'.62 In
1930 Mgr Classe issued a categorical warning to the administration
against any attempt to 'eliminate the Tutsi caste':
A revolution of that nature would lead the entire state directly into
anarchy and to bitter anti-European Communism. Far from further-
ing progress, it would nullify the government's action by depriving
it of auxiliaries who are, by birth, capable of understanding and fol-
lowing it. This is the view and the firm belief of all superiors of the
Ruanda mission, without exception. Generally speaking, we have no
chiefs who are better qualified, more intelligent, more active, more
capable of appreciating progress and more fully accepted by the
people than the Tutsi.3
In view of the radically different attitude adopted by the Catholic
Church after the Second World War, this statement has a peculiar
ring to it. For the time being, however, the Church posed as the
strongest advocate of Tutsi supremacy, largely on grounds of political
expediency. That due attention was paid to Mgr Classe's 'profession of
faith' was made abundantly clear by the subsequent direction of Belgian
policies. Not only were the Hutu chiefs and subchiefs all dismissed
from office and replaced by 'well-born Tutsi', but a positive effort was
made to preserve Tutsi hegemony in every walk of life.
The preservation-indeed the strengthening--of Tutsi supremacy
was achieved in three major ways, and in the following chronological
order: (i) by facilitating the territorial expansion of Tutsi political
hegemony; (ii) by a rigorous control over all educational opportunities;
and (iii) by the introduction of a judicial machinery designed to per-
petuate the subjection of the Hutu caste.
As a result of Belgian efforts to extend Tutsi domination to northern
Rwanda, a number of indigenous Hutu chiefs (bahinza) were summarily
removed from office in the early 1920s and replaced by Tutsi appointed
by the administration. This policy found its most systematic application
in the Ndorwa, Mutara and Mulera regions in the north (roughly
corresponding to Ndungutse's sphere of influence) and in the Busozo,
Bukinzi and Bushiru in the north-west. While this parachuting of Tutsi
chiefs into predominantly Hutu areas was but the continuation of a
trend initiated under the German protectorate, it is also true that in






Rwanda and Burundi
many places 'the Belgians helped to install the first Batutsi chiefs in the
country, the bahinza lacking competence to enforce the methods
advocated by the occupying authority'." Here as elsewhere in Rwanda
the Belgian authorities were led to perpetuate and systematise the poli-
cies inaugurated by their predecessors.
Similarly, just as the education of the Tutsi caste became a special
concern of the German Residents, on the grounds that they were the
natural auxiliaries of the protectorate, by the early 1930s and until well
after the Second World War, the consensus of opinion among Belgian
administrators was that the Tutsi should remain the sole recipients of
secular and missionary education. At an early date the government
schools of Nyanza, Ruhengeri, Gatsibu and Cyangugu became training
grounds exclusively for Tutsi (sons of chiefs as well as 'commoners')
who later served the administration in the capacity of 'secr6taires
indigenes' (i.e., interpreters, clerks, tax collectors, etc.). Many of these
educated Tutsi were later appointed chiefs and thus constituted the
embryo of a new category of functionaries which the administration
used as a counterweight to the apathy or resistance of the older genera-
tions. As one administrator put it:
The mass of the Ruanda chiefs and subchiefs has thus been infiltra-
ted [noyaut6e] by valuable elements, trained by us and influenced
by our methods and ideas. .... A sense of emulation has gradually
emerged among native leaders. . For those 'notables' whom we
found incapable or unwilling to accept our ideas we were thus able
to substitute some of our trainees. In this fashion the oppositional
mentality which Musinga himself had tried to foster among the nobil-
ity was kept in check. Thanks to the Nyanza school, we were able to
create an elite of intelligent chiefs, and, especially in the last few
years, to record genuine progress throughout the country.65
But only in 1929, with the creation of the Ecole des Freres de la Charit6
(better known as the Groupe Scolaire) in Astrida (now Butare), was a
special effort made to recruit students from among the sons of Tutsi
chiefs and to tailor the curriculum to the functions and skills expected of
a chief. In subsequent years the Groupe Scolaire became the grace-
giving institution through which the Tutsi elites managed to perpetuate
themselves in the seats of power, through which they gained the tech-
nical skills and training necessary for the preservation of their tradi-
tional claims to supremacy.
Although the Groupe Scolaire recruited students from both territor-
ies, the standards of admission were not nearly as restrictive in Burundi
as in Rwanda. As R. E. S. Tanner recently observed, 'in the develop-






Historical Survey
ment of local government in Urundi it has often been assumed that the
Belgian government educated the Tutsi .. as a deliberate policy of
maintaining the Tutsi-dominated status quo. There was in fact no such
deliberate preference, but the Tutsi saw early the advantages of
education, principally in terms of French and its political uses, and
purposefully pursued it.'66 Nonetheless, the inducements arising from
the pressure of the caste system were never quite as pronounced as in
Rwanda. The Tutsi of Burundi never felt the same psychological urge to
maintain their position of dominance, socially and politically, as their
Rwandese kinsmen. This, coupled with the more 'liberal' educational
policies of the Burundi Residency, helps to explain why the proportion
of Hutu students enrolled at the Groupe Scolaire during the 'thirties and
early 'forties was markedly higher for Burundi than for Rwanda.
It is also interesting to compare the enrolment figures for the early
government schools of Nyanza, in Rwanda, and Muramvya, in Burundi.
In 1925, the Nyanza Ecole pour Fils de Chefs had 349 students, all of
Tutsi origin; in 1928, the Muramvya Ecole pour Fils de Chefs had 177
students, of whom fifty were described'as 'sons of chiefs or belonging
to the high aristocracy', sixty-seven were Tutsi, fifty-three Hutu, one
mulatto, one Asian and five 'sons of soldiers'. The report from which
these figures are extracted goes on to note that 'the educational establish-
ment of Muramvya is far less frequented by the sons of paramount chiefs
than its counterpart in Rwanda. The influence of the court is far less
apparent in Burundi, and a good many chiefs prefer to send their sons
to the district schools.'67 As a group, the Tutsi of Rwanda were
relatively better educated than their counterparts in Burundi, which in
turn reinforced their sense of collective superiority vis-h-vis the Hutu
and gave further justification to the Belgian contention that 'the Tutsi
were the pick of the natives and [thus] should be retained in commanding
positions in the native social organisation'.68
In time some of the policies adopted in the Congo provided a new
pole of attraction for the testing and sorting out of native institutions.
This is best illustrated by the introduction of 'native tribunals' in 1936.
Despite the disastrous results of earlier experiments along these lines,
it was assumed that in the context of the mandated territory the native
tribunals would become the most effective instrument of indirect
rule. In the mind of the Belgian Resident the native court system would
provide the master key to every problem of native administration.
The native tribunals would act at one and the same time as 'a safeguard
of traditions and a brake upon their evolution', as 'a melting pot in which
past and present tendencies [would] coalesce', and as 'the means where-
by a progressive and progressist, yet slow and smooth, assimilation






Rwanda and Burundi
could be achieved'.69 In fact, these tribunals became the instruments
through which the ruling Tutsi oligarchy not only retained but abused
its privileges. Their function was not so much to dispense justice as to
legitimise abuses and wrong-doings. Since they were in every case
headed by Tutsi chiefs it is difficult to imagine how they could have
served a different purpose. Although the mwami's tribunal was intended
to serve as a court of appeal the long delays resulting from the accumula-
tion of pending litigations often amounted to a denial of justice. Thus,
with an average of only sixty cases handled each year, by 1949 the
mwami's tribunal was faced with a backlog of some 900 untried cases,
a situation described as 'clearly alarming'.70 If further evidence were
needed to dispel illusions about the true nature of the Rwandese court
system one could cite the following statement, by a former Belgian
official: 'The native tribunals never played a moderating role because
they were intimately linked to the political authorities. In many cases
these tribunals were the organs used by the Tutsi to give a semblance
of legality to their exactions . The only way to redress these injus-
tices was to seek the annulment of iniquitous decisions from the Parquet,
but the number of applications was so great that it was impossible to
examine each demand.'71
Although abuses were by no means unheard of in Burundi, justice was
never quite as grossly miscarried as in Rwanda. There were fewer
opportunities for abuses, as the customary obligations.arising from the
clientage system were less burdensome and encompassing. Moreover,
the mwami's tribunal acquired a reputation for efficiency and impar-
tiality which was never matched by its Rwanda counterpart. Last, the
traditional institutions of Burundi offered several alternative arenas for
the settlement of litigations, for which there was no equivalent in
Rwanda; particularly significant was the arbitrating role played by the
bashigantahe at each level of the political hierarchy.
From the foregoing observations emerge two major points of differ-
ence in the working and implications of indirect rule in Rwanda and
Burundi. In Rwanda the political monopoly of the Tutsi oligarchy was
identified with the retention of caste privileges to a far greater extent
than in Burundi; furthermore, in the context of Rwanda society the
institution of the monarchy was inextricably bound up with the actions,
values and instrumentalities employed by the Tutsi to maintain them-
selves in power. In Burundi, where administrative control was effected
largely through the heads of the princely families and only marginally
through the mwami, the crown was a minor element in the institutional
matrix of the country, and therefore never became the target of social
and economic grievances. One could even argue that, insofar as popular






Historical Survey
grievances were associated with the perpetuation of ganwa rule, the
opposition of certain princely families to the mwami's person served
merely to reinforce the legitimacy of the crown. In Rwanda, on the other
hand, political conflict expressed itself in the form of a violent clash
between the crown, which stood as the symbol of Tutsi hegemony, and
the egalitarian aspirations of the Westernised Hutu elites. The roots of
the conflict are to be found in part in the specific pattern of social
stratification and traditional institutions of Rwanda, and in the very
nature of the policies pursued by Belgium in the years preceding the
Second World War.
In spite of these differences, enough uniformity can be found in the
principles underlying the Belgian version of indirect rule to warrant a
brief comparison with the British model. The contrast between the
Belgian and the British conception of 'native administration' is nowhere
better illustrated than by Lord Lugard's definition of the role of the
chief in the British colonial context, and his later comments about the
chiefs of Ruanda-Urundi. In his classic work, The Dual Mandate,
Lugard wrote:
The essential feature of the system is that the native chiefs are
constituted as an integral part of the machinery of the administration.
There are not two sets of rulers, British and native, working either
separately or in co-operation, but a single government in which the
native chiefs have well defined duties and an acknowledged status
equally with the British official; their duties should never conflict and
should overlap as little as possible.7
When, some ten years later, Lord Lugard served as the accredited
British representative on the Permanent Mandates Commission, he was
visibly at a loss to reconcile the Belgian conception of the 'Dual Man-
date' with his own theoretical formulation. In his address to the Com-
mission, in December 1939, Lugard noted: 'At present the chiefs [of
Ruanda-Urundi] have no right to give any order that had not been
previously sanctioned by the Administration. They had no criminal
jurisdiction; their treasuries were under the direct control of the Admin-
istrator.' 'The present system', he said, 'did not seem to allow of any
personal responsibility for the chiefs.' When he asked the Belgian
representative, Halewyck, about the future native policy of Belgium,
he was told that 'the administration was confronted with a certain
number of excellent chiefs, some bad ones, and others who only reached
a mediocre or indifferent standard. If an attempt were made imme-
diately to so organise all-round indirect administration on the strength
of the experience gained with a few very good chiefs, serious






Rwanda and Burundi
disappointments would be in store.' Which in turn prompted Lord
Lugard to observe, with flawless logic, that 'the administration did not
see its way at present to entrust any personal responsibility to efficient
chiefs because there were also bad chiefs. .. The logical conclusion to
be drawn from the explanations given by the accredited representative
was that the grant of wider powers might be delayed indefinitely.'73
Regardless of the motives then actuating Belgian policies, the
abiding fact which emerges from the record is that the authority of the
chiefs suffered some crippling limitations from the omnivorous charac-
ter of the European administration. By comparison with the British
model, a former District Commissioner who served in Tanzania found
the Ruanda-Urundi administration 'both complicated and interfering',
further noting that 'while the British could not be said to have a
laissez-faire system Belgium administered a "rule of law" system,
imposed from above which involved no predisposing process of change
within the people to whom it was applied'.74 For most Belgian adminis-
trators all that was really needed for a satisfactory functioning of indirect
rule was 'to organise the outward prestige of the bami';75 that is the kind
of 'window-dressing' the Belgian representative to the Permanent
Mandates Commission had in mind when he stated, in 1929, that 'by
means of establishing a court and a guard of honour [for the bami],
the administration would increase their authority'. In support of his
views the Belgian spokesman cited the example of Uganda, 'where
much dignity had been conferred upon certain chiefs by organising
their prestige',76 but neglected to mention that in Uganda, as Lord
Lugard later emphasised, 'the chiefs and the subchiefs formed, in
effect, a native civil service under the Paramount's government'.77 In
Ruanda-Urundi, by contrast, the chiefs and subchiefs formed a corps
of native functionaries under the immediate and permanent supervision
of the European administration.
To conclude that the bami have always acted as puppets, as docile
'yes-men' in the hands of the coloniser, would be inaccurate. Depending
on the attitude of the Residents-and the political conjuncture-they
were occasionally deferred to over the appointment of chiefs and,
while they were made increasingly aware of the limits of their powers,
they also knew how to make the most of the opportunities offered by the
system to play one official off against another, how to conceal their real
intentions behind a facade of outward submissiveness, and how to put
into practice the kinyarwanda dictum that 'the Europeans are not clever'
(abazungo ntibaze ubwenge). But once this is said, it is equally plain that
the 'native authorities' of Ruanda-Urundi were by and large denied the
prerogatives and freedom the British version of indirect rule presup-






Historical Survey
posed. As we shall see, the constitutional reforms introduced after the
Second World War did little to alter this state of affairs.

FROM TRUSTEESHIP TO INDEPENDENCE
After the Second World War, the mandated territories, including
Ruanda-Urundi, became trusteeship territories under the United Nat-
ions. Belgium's commitment to the aims of the trusteeship (politically
far more significant than those stipulated under the mandates system)
implied a major departure from its previous policies. Whereas in 1931
the Belgian representative to the Permanent Mandates Commission
candidly admitted that 'the mandatory power's policy was not at all
directed at the abolition of the feudal system of Ruanda-Urundi, which,
with its hierarchy of chiefdoms and sub-chiefdoms, could quite well
be adapted to the government of the territory', but merely 'to regroup
the areas dependent on one chiefdom which had not been previously
united', after the Second World War the official viewpoint of the trust
authorities was that 'Belgian policy sought to bring to an end the feudal
regime', a major advance over the mulishly static posture of previous
years.78 Moreover, if the Trusteeship Council's visiting missions
undoubtedly played a part in hastening the political awakening of the
indigenous populations, the repeated criticisms voiced against Belgium
in the United Nations were equally instrumental in creating a climate
of world opinion which had a direct influence on the pace and direction
of its trust territory policy.
Nonetheless, constitutional reforms proceeded slowly and half-
heartedly, owing to the characteristic caution with which the Belgian
authorities approached the subject of political change, and, initially at
least, to the absence of modern forms of political self-expression in each
of the territories concerned. Only when confronted with what suddenly
appeared an irresistible popular pressure for change did the Belgians
actively and deliberately seek to synchronise constitutional reforms with
political change. At this late stage, however, synchronisation was no
longer possible. As in the Congo, Belgian policies in Ruanda-Urundi
were a classic example of 'too little and too late'.
The visit of the first UN mission to Ruanda-Urundi, in 1948, revealed
some familiar themes in the credo of Belgian colonial policy. Absolute
priority was to be given to the economic progress and moral uplift
of the indigenous populations; not until these preconditions were met
could one envisage a democratisation of existing political institutions.
In a personal statement submitted to the UN delegation, the Burundi
Resident, Robert Schmidt, explained the views of his government (the






Rwanda and Burundi
following are Schmidt's own words): 'In this process of changing the
whole political machinery [of Ruanda-Urundi], the degree of evolution,
the aspirations and faculty of assimilation of the people must be taken
into consideration. It would be harsh and unfair to render unhappy, or
in a state bordering on social anarchy, one or two generations by impos-
ing premature reforms by virtue of a political ideology or on the excuse
that we are hoping to bring happiness in this fashion to future genera-
tions.'79 The election of chiefs was envisaged as a conceivable yet distant
possibility, for it would 'require from the masses an understanding of
electoral procedures, and from the chiefs a moral preparation which
neither has yet attained'. Furthermore, said Schmidt, the setting up of a
democratic regime did not necessarily require the election of local
officials, a notion which, as he chose to phrase it, might have caused mild
consternation among some of his countrymen: 'Concerning the question
raised about electing chiefs and subchiefs, may I repeat a remark
that I made previously, I think, to one of you on this subject. Belgium,
as you will recognize, is an old democratic country. Very much so. Yet
our provincial burgomasters and governors (comparable to chiefs and
subchiefs here) are appointed by the king and not elected.'80 On the
question of 'Hutu emancipation' the Resident voiced similar reserva-
tions: 'It has been found that in many cases as soon as a man taken from
the people is given a position of trust, he generally misuses it, being very
liable to bribery and embezzlement. He is worse than any of the old
class of chiefs in corrupt practices detrimental to his brethren. This is
one of the things that has made us wary in bringing about too drastic
democratic reforms before the people are sufficiently educated to higher
standards and really understand what responsibility means and im-
plies.'81 The Resident's qualifying remarks that 'this of course does not
mean that we must or are ready to remain static' did not seem to augur a
more dramatic change of tempo than was then envisaged for the Congo;
in time, however, the pressures arising from Belgium's international
obligations prompted it to initiate political reforms in Ruanda-Urundi
long before similar steps were anticipated for its colony.
For our purpose it may be convenient to look at the constitutional
evolution of Ruanda-Urundi after the Second World War as falling
into two broad periods:

(i) From 1952 to 1959: a period of limited constitutional reforms,
resulting in the introduction of advisory councils at each level of the
administrative hierarchy.
(ii) From 1959 to 1962: a period of accelerated democratisation,
inaugurated by the government's declaration of November o1, 1959,







Historical Survey
which led in 1960 and 1961 to the establishment of popularly elected
organs of government both at the local and central levels, and ultimately
to the independence of each territory as a separate political entity.
From the radically different responses elicited by the more recent of
these constitutional transformations one can detect some equally striking
variations in the political cleavages and pressures operating in each
territory. In Rwanda, where the internal conflict between Hutu and
Tutsi tended to over-ride the conflict between the colonial society and
the coloniser, nationalism, as a cohesive force, has never been more than
an epiphenomenon; in Burundi, by contrast, where internal divisions
had not yet reached a comparable pitch of intensity, or a comparable
ethnic quotient, nationalist assertions came to the fore much more vigor-
ously and cohesively than in Rwanda.
Yet relatively little effort was made to accommodate the processes of
political transfer to these different patterns of change. Unlike British
policy, formulated in reaction to and through 'a process of inter-related
pressures',82 Belgian policy in Ruanda-Urundi was more in the nature
of a generalised response to a specific challenge, the challenge of the
Rwandese revolution. Just as in the early days of colonial rule Rwanda
had served as the model for German policies in Burundi, half a century
later it once again set the tone of Belgian policies in Burundi.
In 1952, for the first time, the decision was made to introduce a glim-
mer of democracy in the sphere of native administration. On July 14,
1952, a decree was issued providing for the establishment of representa-
tive organs at each level of the administrative pyramid: advisory coun-
cils were set up at the subchiefdom, chiefdom, district and territorial
levels, in the form of 'conseils de sous-chefferie', 'conseils de chefferie',
'conseils de territoire' and 'conseils sup6rieurs du pays' (csp).* But,
apart from the fact that the powers devolved upon the councils remained
strictly advisory, the complicated procedure of co-opting introduced by
Belgium cast immediate doubts on the value of the experiment. As
Professor Maquet pointed out, 'the system was only very moderately
elective and representative ... at each level there were unofficial mem-
bers but they constituted only a fraction of the council's membership
and were elected (i.e., co-opted) from among unofficial members of the
lower councils, which meant that the choice was very restricted'.83
*To avoid possible confusions, it is well to bear in mind that the French
term 'territoire' is by no means synonymous with 'territory'. In this study
'territory' refers to each of the countries that were amalgamated into the Trustee-
ship Territory of Ruanda-Urundi, i.e. what the Belgians referred to as 'pays'.
The 'territoires', in the official Belgian terminology, were administrative sub-
divisions roughly similar to the 'districts' in the British possessions.







Rwanda and Burundi
That the system was indeed 'only very moderately representative' was
made patently clear by the outcome of the 1953 elections. In Rwanda
52 per cent of the seats in the subchiefdom councils fell into Tutsi
hands, against 39 per cent in Burundi, with the Tutsi gaining an increas-
ingly larger representation at each ascending step in the administrative
ladder. Thus Tutsi controlled 90-6 per cent of the seats in Rwanda's
Conseil Superieur du Pays, as against 80o7 per cent in Burundi. Nor
was the situation markedly improved by the introduction in 1956 of
universal male suffrage for the election of the lower councils. In Rwanda,
the results of the 1956 elections showed a slight drop (7 per cent) in
Tutsi representation in the subchiefdom councils but further gains at
the top; in Burundi, the Tutsi consolidated their position at each level.*
Except for the subchiefdom councils, reflecting in each country a Hutu
majority, in the last analysis the composition of the higher councils
continued to show an overwhelming majority of Tutsi.
A study published in 195984 gives the following ethnic breakdown
of administrative offices in Ruanda-Urundi:

Ethnic Distribution
Tutsi Hutu Total
Offices Total % Total % Total %
I. Chiefs 81 98-8 I 1i2 82 1oo
II. Subchiefs 1050 95'5 50 4-5 11oo 1oo
II. Conseil Supdrieur du
Pays
1. Rwanda 31 94 2 6 33 1oo
2. Burundi 30 91 3 9 33 o00
IV. Conseils de Territoire
I. Rwanda 125 80o7 30 19"3 155 o00
2. Burundi 112 81z2 26 i8-8 138 ioo
v. Administrative Auxiliaries 284 67 122 33 406 Ioo

The hopes raised by these early constitutional reforms, and the bitter
disappointment caused by the subsequent realisation that these would
The statistics available for Burundi make no distinction between Tutsi,
Hima and ganwa; all three are apparently lumped together under the same
rubric as 'Tutsi'. If for no other reason one is led to suspect a serious distortion
in the tabulation of the results of the 1953 and 1956 elections made by J. J.
Maquet and M. d'Hertefelt in their otherwise excellent study, Elections en Socidtd
Feodale, ARSC, Brussels 1959, Vol. xxi, fasc. 2. In the absence of contrary
evidence other than what has been gleaned in the course of interviews, I had no
choice but to fall back on the data contained in the above study. The same
reservation applies to the figures in the table.






Historical Survey
hardly alter the privileged position of the ruling caste, were crucial
elements in the background of the Rwandese revolution. The electoral
processes were introduced at a time when the Hutu educated elites
constituted a very tiny minority, and when little fundamental change
had yet occurred in the traditional social structure. No parties in the
modern sense of the word had emerged in either Rwanda or Burundi.
In Maquet's own words, 'an elective system meant to give the people
a share in their own government had been introduced in a culture
founded on opposite premises, those of inequality, of the idea of born
rulers, of stratified society'.85 In these conditions it would be mislead-
ing to speak of a distortion of the vox populi through the electoral
system. More than anything else the popular vote expressed the con-
tinued attachment of the masses to a value system based on deference
towards the ruling caste. This fundamental fact, however, is precisely
what the nascent educated Hutu elites refused to accept. In their
minds, as Maquet remarked, 'the new institutions were understood in
the perspective of a democratic system of representation',86 one that
would presumably 'enthrone' the representatives of the Hutu majority;
instead, and to their utter dismay, they saw these institutions converted
into modern arenas for the expression of Tutsi supremacy. It is against
this background of disillusion and bitterness over the failure of constitu-
tional reforms to meet expected changes that one must seek the origins
of the Rwandese revolution.
It would be both an oversimplification and a travesty of facts to
infer from the foregoing that the Hutu of Burundi viewed the effects of
these reforms in their own territory with nothing but blissful content-
ment. The predominance of upper caste elements in the higher councils
raised considerable anxieties among certain educated Hutu. The
late Pierre Ngendadumwe (who twice held the prime ministership,
before his assassination in January 1965) expressed his concern that
'the bami and the chiefs' had emerged as 'the great beneficiaries of the
decree of July 1952', adding that 'the measures taken by assemblies
that are dominated by traditional elements are automatically suspect
and discredited by public opinion'. 'The present tragedy in Ruanda-
Urundi', wrote Ngendadumwe in 1959, 'does not consist only in the
fact of white colonisation but also in the paradox that in spite of rep-
resenting a minority the near totality of offices of chief, subchief and
judge are in the hands of the Tutsi'.s7 In spite of this and other evidences
of anti-Tutsi sentiment, the fact is, however, that the animus of the
educated Hutu of Burundi was at first primarily centred on the princely
families (i.e. the ganwa) rather than on the Tutsi as a group. What is
more, not only did the monarchy manage to escape the stigma of ethnic

83
4-RAB *






Rwanda and Burundi
favouritism; it emerged as the central focus of popular loyalties, and,
for a while at least, provided a major unifying bond for society as a whole.
The metropolitan Belgian government's declaration of November io,
1959, ushered in a new phase in the constitutional history of Ruanda-
Urundi, in a way characterized by the same mixture of blindness and
naivety as had already been revealed in the Congo-and would soon
become even more painfully apparent. The appointment of a parlia-
mentary study group groupe de travail) in April 1959, to investigate
the conditions under which a transfer of authority could safely be
accomplished; the issuance of a formal declaration, in November 1959,
concerning the character of the future political institutions of Ruanda-
Urundi; the subsequent decision to call a Round Table conference to
explore possibilities of agreement among the newly-created political
parties: all were indicative of how far the metropolitan authorities
leaned on the precedents established in the Congo in their effort to
transfer power in an orderly manner. And as in the case of an earlier
policy statement on the future of the Congo, the government's declara-
tion of November 1959 was quickly outstripped of its contents by the
pace of political developments in the trusteeship territory.
It is not only at the level of official policies that a pattern of interaction
can be discerned between the Congo and Ruanda-Urundi. Political
events within the Congo also had immediate repercussions on the atti-
tudes and expectations of local politicians in both Rwanda and Burundi.
Whereas in the mid-'fifties the more politically conscious of the Congo-
lese '6volubs' readily cited the example of Ruanda-Urundi as a justifica-
tion of their claims for constitutional advance, by early 1959 (and even
more so after the Brussels Round Table Conference of January 1960)
it was the turn of the Barundi and the Banyarwanda to take their cues
from the Congolese, now regarded as the more privileged. Just as there
were differences of orientation and ideology among Congolese leaders
and parties, the projection of these differences into the context of Ruanda-
Urundi has had a direct impact on the internal politics of each country.
By its declaration of November 1959 the Belgian government com-
mitted itself to a two-fold programme of political reform. First, to a
fundamental revamping of the local political structures of each territory.
The aim was to convert the subchiefdoms into communes, headed by a
burgomaster assisted by a popularly elected communal council. While
the communes were to form the basic political infrastructure, the chief-
doms would be transformed into purely administrative units. In each
country legislative powers were to be gradually devolved upon the
Conseils de Pays, with the mwami relegated to the position of a constitu-
tional head of state. Second, a positive effort was to be made to encour-






Historical Survey
age the creation of a broad political community, 'l'entit6 Ruanda-
Urundi', through 'judicious consultations and with the assistance of the
newly established Conseils de Pays'.88
By the time the statement was issued, much of its substance had
already been reduced to wishful thinking-at least so far as Rwanda was
concerned. As the first stage of the revolution got under way, there
seemedlittle prospect of anything but a protracted period of searing civil
strife. In any case, from then on the initiative lay almost entirely with the
local European administration. At first a Conseil Sp6cial Provisoire was
installed (in January 1960) to replace the Tutsi-dominated Conseil
Sup6rieur du Pays, in the hope that it would 'smooth the transition
between yesterday's autocratic system and tomorrow's democratic
regime'.89 By October 1960, however, a new set of provisional organs
was introduced, in the form of a forty-eight-member assembly and a
government, both appointed by the Residency after taking into account
the results of the recently-held communal elections. But their life-span
was to be even shorter than that of their predecessor. In January 1961
a Hutu-inspired and Belgian-assisted coup d'6tat was launched in
Gitarama, in central Rwanda, which led to the proclamation of a
republic and the installation of a new provisional government and an
assembly, both firmly under Hutu control. Despite the vehement
protestations raised in the United Nations, the Belgian administration
(as well as the metropolitan government) took the view that the authors
of the coup represented in effect the only legitimate provisional
government of Rwanda. With the local administration now acting as the
effective power underpinning the newly-established government, the
legislative elections of 1961 merely confirmed the de facto situation
engendered by the previous political upset.
The timing and forms of political transfer dictated by the Rwandese
situation were paralleled in Burundi by the organisation of communal
elections in September 1960, leading in January 1961 to the establish-
ment of a provisional government and a provisional council. Like its
Rwandese counterpart, this government was to serve 'as a purely transi-
tional organ, to be replaced by a permanent institution after the
[legislative] elections';90 and, as happened in Rwanda, the permanent
organs set up after these elections came to reflect a political complexion
quite unlike that of the interim institutions. But apart from the fact
that in Burundi political change was effected through the ballot box,
and not through a coup d'etat, the direction of change was very different.
The results of the legislative elections of mid-1961 legitimised the claims
of an ethnically-mixed, 'neo-traditionalist' party which, for the time
being at least, posed as the strongest supporter of the monarchy.






Rwanda and Burundi


In these circumstances one can better appreciate the difficulties
facing the United Nations as the question of the future of Ruanda-
Urundi came before the General Assembly's Fourth Committee, in
January 1962. Apart from the persistent threats of instability posed by
the sporadic incursions of armed bands of Tutsi refugees into Rwanda,
and the technical problems involved in the transfer of administrative
services heretofore shared with the Congo, the immediate preoccupation
of the United Nations was to devise a mutually acceptable formula to
keep Ruanda-Urundi a single political entity.
Hyphened together by an accident of history, yet lacking a central
institutional focus around which a common political consciousness could
be developed, there were ample grounds for questioning the prospects
of a durable union between the two states. Now that their recent political
evolution had drawn them further apart from each other, their unifica-
tion into a single independent state seemed even more improbable. It
was the attractiveness of the ideal of Pan-Africanism rather than the
immediate interest of the people of Ruanda-Urundi which led most of
the African delegates to support the eleven-point resolution endorsed
by the UN General Assembly on February 13, 1962, a resolution which
'reaffirmed the [General Assembly's] conviction that the best future of
Ruanda-Urundi lay in the emergence of a single state with economic
unity, common defence and external relations, without prejudice to the
internal autonomy of each entity'.91 Thus, under the terms of the same
resolution, the General Assembly entrusted to a five-member commis-
sion (better known as the Brooks Commission) the task of convening 'as
soon as possible, at Addis Ababa, a high-level conference ... with aview
to finding a mutually acceptable formula for the creation of the closest
form of political, economic, and administrative union, the role of the
commission being to endeavour to reconcile the two points of view of
the two Governments'.
The failure of the Addis Ababa conference to achieve its proposed
objective cannot be ascribed to a want of persuasiveness on the part of
its chairman, Miss Angie Brooks of Liberia. In presenting her case for
unity she reminded her audience that 'the balkanisation of Africa is
dangerous to the African cause of unity and solidarity', and that 'close
association between the two parts would enable the problems to be
tackled more effectively and would at least allow substantial budgetary
economies'. Certainly, much could be said for the argument that 'today
large economic units can better face the complex problems of develop-
ment and also better afford to take the independent economic and poli-
tical stand which alone is a serious guarantee of national independence'.92
The real question, however, was whether the prospects of economic