Egbe Lembá

Lemba, 1650-1930
A Drum of Affliction in
Africa and the New World
Volume 11
Critical Studies on
Black Life and Culture
Advisory Editor
Professor Charles T. Davis, Chairman
Afro-American Studies, Yale University
Assistant Advisory Editor

Professor Henry-Louis Gates
Afro-American Studies, Yale University
John M. Janzen
Lemba, 1650-1930
Garland Publishing, Inc. N e w York and London

1982
® 1982 JohnM. Janzen
All rights reserved
Library of Congress Cataloging in Publication Data

Janzen, John M.
Lemba, 1650-1930.
(Critical studies on Black life and culture ; v. 11)
Bibliography: p.
Includes index.
1. Lemba (Cult) I. Title. II. Series.
BL2470.C6J36 967 80-9044
ISBN 0-8240-9306-2 AACR2
Printed on acid-free, 250-year-life paper

Manufactured in the United States of America
Contents
Plates ix
Figures xi
Preface xiii
Chapter 1: Introduction to Lemba

Lemba: Historic Equatorial African Drum of Affliction 3
Issues in the Study of Lemba of Theoretical Significance 7
Part I: The Public Setting of Lemba: Management of Society

and its Resources 27
Chapter 2: Economic, Social, and Political History
of the Lemba Region 27
Environment, Resource Use, Adaptation, Trade
and Currency 25
Social Structure 36
Political Systems of the Lemba Region 45
The Coastal Kingdoms 45
The Teke Federation 58
The Nsundi Systems 61
Life Between the Kingdoms: Market Laws
and Lemba's Marriage Politics 70
9
Chapter 3: Biographical Sketches of Lemba s Demise 81
Introduction 81
Accepting Colonial Authority:
Ngambula, Mampuya, Sobisa 83
Populist Religion and the Missions: Katula 86
The Common Man's Dilemma: Ndibu 87
Seeking Redemption in the Colonial Setting 90
Part II: The Rituals of Lemba: Management of Reality 93

Introduction 95
v
VJ CONTENTS
Chapter 4: The Northern (Kamba, Yaa) Variant of Lemba 105

Introduction to the Sources 105
The Lemba Seance Near Mboko Nsongo
(Among the Kamba) 106
Text 1 106
Lemba in Indo (Among the Yaa) 125
Text 2 125
The Expressive Domains and their Codes 130
The Spatial and Temporal Distribution of Events 130
Economic and Exchange Structures in the Rite 135
9
Social Organization: Lemba s Ceremonial Kinship 138
Sacred Medicines 140
Verbal Categories of Ritual Action 143
The Lyrical and its Message 146
Chapter 5: The Eastern (Lari) Variant of Lemba 157

The Lemba Seance at Madzia (Among the Lari) 158
Text3 158
The Expressive Domains and their Codes 163
The Spatial and Temporal Distribution of Events 163
Economic and Exchange Structures 166
Kinship and Social Organization 169
Medicine of Lemba 172
The Lyrical 179
Text 4 183
Chapter 6: The South-Central (Nsundi, Bwende) Variant

of Lemba 187
The Lemba Seance in Nseke-Mbanza 187
The Expressive Domains of Lemba 189
Spatial-Temporal Distribution of Events 189
Medicine, Meaning, and the Categories
of Ritual Action 195
The Lyrical: The Many Dual Faces of Mahungu 200
Text 5 202
Text 6 207
Text 7 210
Text 8 223
CONTENTS VJJ
Chapter 7: The Western (Yombe, Woyo, Vili) Variant

of Lemba 233
The Lemba Seance in Eastern Mayombe 234
Text 9 234
Text 10 236
The Expressive Domains 243
Space and Time: From Portable to
Fixed Ritual Symbolism 243
Economic and Exchange Structures 245
Verbal Categories, Ritual Action,
and Lemba Names 247
Medicine and Art 251
The Lyrical: Moni-Mambu Trickster
and the Origins of Lemba 258
Text 11 259
Text 12 267
Chapter 8: Lemba in the New World 273
Introduction 273
The Petro Complex of Deities in a Voodoo Loa Service 280
Lemba-Petro in the Cul-de-Sac Valley, Haiti 286
Part III: Structure and Movement in Lemba Therapeutics 293

Introduction 295
Chapter 9: Modes of Lemba Thought 297
Anchoring in Myth, and the Problem of History 297
Successional and Relational Modes of Authority 299
Text 13 302
Conjunctive and Disjunctive Mediation 304
Humanization and Naturalization of the Discourse 307
Structure, Choice, and Medicine 310
Chapter 10: The Ideology of Lemba Therapeutics 315
The Affliction and its Etiology 315
The Therapy 318
The Virtuous Society 319
Chapter 11: Conclusion: What Were the Questions? 323
Viü CONTENTS
Notes 331
Sources 349
Index 365
Plates
(Plates 1-4 follow p. xv; plates 5 - 2 0 follow p. 257.)
1. Loango, Mid-Seventeenth Century

2. Market Scene from Lemba Region:
General View of Market
3. Market Scene from Lemba Region: Entrance
4. Lemba Chest from Ngoyo
5. Lemba Chest from Nkangu, Mayombe: Lid Top
6. Lemba Chest from Nkangu: Contents
7. Lemba Chest from Nkangu: Contents of Raphia Bag
8. Copper Lemba Bracelet from Loango
9. Brass Lemba Bracelet from Loango
10. Wooden Model for Lemba Bracelet Imprint:
Multi-Petal Motif
11. Wooden Model for Bracelet Imprint: Cowrie Motif
12. Copper Lemba Bracelet,
Multiple-Petal and Orator's-Charm Motifs
13. Lemba Tin Bracelet, Mayombe
14. Ngoma-Lemba Drum, Mayombe
15. Miniature Drum Necklace, Loango
16. Miniature Drum Pendants in Bone, Wood, Metal
17. Mukonzi a Lemba Drum, Mayombe
18. Lemba Couple Seated on Trunk, Vili
19. Lemba Couple, Back View
20. Loango Ivory
ix
Figures
1. Regional Distribution of Lemba, based on Ethnographie,

Textual, and Art-Historical Sources 2
2. Political and Economic Map of Lemba Region,
1600-1930 5
3. Nsundi, Seventeenth through Nineteenth Centuries 62
4. Market Areas and Villages of Kivunda Area,
Luozi Territory 73
5. Locations in Village Setting of Lemba Initiation
Markers andSignificant Sites 132
6. Spatial and Temporal Organization of Events in
Northern Lemba Variant, after Kwamba 133
7. Exchange Structure of Lernba Séance 136
8. Spatial/Temporal Organization in Eastern Lemba,
after Kimbembe 164
after Stenström 164
10. Spatial/Temporal Organization in Eastern jLeraôa,
after Malonga 165
after Andersson 165
12. Exchange Structure, Eastern Lemba Séance 167
13. Eastern Lemba Tukobe and Ingredients 176
14. Lemba's Dominant Metaphor, Eastern Variant 178
15. Metaphoric Association of Domains, Eastern Variant 185
16. Spatial/Temporal Organization, South-Central Lemba,
after Fukiau 191
17. Metaphoric Association of Domains,
South-Central Variant 201
18. Spatial/Temporal Organization, Western Variant,
after Babutidi 244
19. Exchange Structure, Western Variant 246
20. Tukobe of Coastal Lemba Variants,
after Dapper and Bittremieux 252
21. Tukobe of Mayombe Lemba Variants,
after B ittremieux and B àbutidi 25 3
xi
XJJ FIGURES
22. Dominant Metaphor, Western Variant 256

23. Correspondences of African and Catholic Deities and
Saints in Bahia and Recife, Brazil, after Bastide 276
24. Temporal/Spatial Organization of Haitian loa Service,
after Herskovits 281
25. Categories of Ritual Transformation in Kongo 299
26. Charter Myth Structures of Several Lemba Variants
and Other Political-Religious Orders 301
Preface
Friends and colleagues alike have encouraged me to declare openly

that I have written a book about a seventeenth-century "cure for
capitalism," created by insightful Congo coast people who perceived
that the great trade was destroying their society. I have tried to avoid
the faddish language of contemporary social analysis, both of the
value-neutral school and of the critical Marxists, and have instead
tried to marshal the evidence needed to show how one segment of
coastal African society tried, in its own idiom—that of the ngoma,
the "drum of affliction"—imaginatively to cope with a force that
transformed African society, like many others, from what it must have
been in earlier centuries into what it was when colonial flag-planters
claimed the coastal Congo region in the late nineteenth century. We
see, in accounts like those of Cabinda story teller Solomo Nitu, how
the wealth to be earned at the coast is characterized in images of a
siren woman who lures a young man to her, exacting from him the
terrible price of forever giving up relations with his father. How the
coastal culture came to terms with this force, over three centuries, is
the story told in this book.
As I have worked on different manifestations of Lemba —the myth
narratives, the initiation reports, the regional economic and political
history—I have gradually perceived the scale of the system—truly a
"world system"—to whichLemba" $> adherents were responding,
namely the expanding imperium of European trade and influence. In
North Africa, more specifically in Egypt, there are examples of
seventeenth-century Islamic brotherhoods which organized mer
chants and businessmen into tight, ethically-conservative groups to
protect their interests in an enhanced economic status while, at the
same time, preserving their Islamic social and religious identity. In
North American Indian societies there are similar examples. The
eastern Calumet system organized trading relationships prior to the
French usurpation of this system, and on the Plains the Hako
ceremonial trading society of the Pawnee brought trade relations under
an umbrella of ritual kinship adoptions, again, seemingly for the
protection of trade interests and also for the protection of the social
fabric. This is the larger, world-wide, context of Lemba, beyond the
"world system" of recent scholarship.
xiii
XJV PREFACE
A work such as this, begun more than a decade ago when I pursued
my own curiosity in a north-Kongo field study, also becomes a
scholarly pilgrimage. It offers the opportunity—a danger to clarity,
perhaps—of adding paradigm upon paradigm to a subject matter
requiring explanation from several sides. The pilgrimage's steps will
be evident in the orientations taken in the book's three sections: first, a
political-economic study of a region; second, a symbolist-struc
turalist study of myths and rituals; third, a humanistic interpretation
of Kongo's therapeutic effectiveness, especially that of Lemba.
I owe my scholarly journey in part to generous benefactors who
have permitted travel and study resources. My original field studies in
Equatorial Africa were sponsored by the Foreign Area Fellowship
Program and the Social Science Research Council (United States).
The Canada Council provided funds for a summer's exploration in
1971 in Sweden where I discovered ( for myself) the Congo catechists'
notebooks of the Laman collection which became the primary texts
for the reconstruction of the Lemba inauguration rites. I am also
indebted to the University of Kansas Graduate Research Fund for the
means to microfilm these notebooks and to prepare them for further
analysis and publication here and elsewhere.
I am deeply appreciative to the Alexander von Humboldt Stiftung
of the Federal Republic of Germany for granting me a research
fellowship in 1977 to pursue work in Central-European archival and
museum sources pertaining to Lemba, especially those of the
German Loango Expedition of the 1870's found in Berlin, and those
of numerous B elgian collectors found in the Royal African Museum in
Tervuren, Belgium.
I am also indebted to the National Endowment for the Humanities
for a six-month research fellowship in spring 1978, which permitted
further work on the Lemba manuscript and the exploration of
9
Lemba s New-World distribution.
I wish to thank staff members of museums and archives who have
helped identify collections pertaining to Lemba, especially Drs.
Maesen and Van der Gelwe of the Central Africa Museum of
Tervuren, Belgium; Drs. Zwerneman and Lohse of the Hamburg
Völkerkunde Museum; Dr. J.F. Thiel and others of the Anthropos
Institute of St Augustine, Germany; Dr. Krieger of the Berlin-
Dahlem Museum für Völkerkunde; Dr. Koloss of the Linden
Museum, Stuttgart; Dr. H. Witte of the Afrika Museum, Berg en Dal,
The Netherlands; Mme. N'diaye of the Musée de l'Homme, Paris;
staff members of the Göteborg and Stockholm Ethnographic
Museums in Sweden; R. Widman of the Svenska Missionsforbundet,
Preface xv
Stockholm and Lidingo; and the personnel of the Jesuit Institute of

Bonn. I am also grateful to Eugenia Herbert, who looked for Lemba-
related articles at the British Museum, and Stanley Yoder, who
examined collections of the National Museum of Zaire in this
connection. This museum research has been extensive but not
exhaustive, so I expect other Lemba-relzted objects to be identified
once this book appears.
I am grateful to several talented persons for assisting me with the
translation of enigmatic KiKongo songs found in the Lemba inaugura
tions. I mention especially Milembamani, Wamba dia Wamba, and
Fukiau, all of whom cherish their language and know it far better than
an outside student like myself ever will. Responsibility for lingering
errors is mine.
I acknowledge those who read earlier versions of the manuscript
and offered corrections and revisions, in particular Jan Vansina,
Victor Turner, Wyatt MacGaffey, and Robert F. Thompson, as well
as several anonymous readers.
I must acknowledge my wife Reinhild's continued encouragement
and criticism and the gracious forbearance of my children, Bernd,
Gesine, and Marike, during the years of my preoccupation with
"Lemia-kinship-medical-systems" as they once succinctly parodied
it.
Lastly, I acknowledge an intellectual indebtedness to my teacher,
the late Professor Lloyd A. Fallers of the University of Chicago, who
led me to see anthropology as a discipline broad enough to allow one
to perceive the unexpected and to interpret it without destroying i t
J M J
Heubuden
January 1981
Plate 1. Loango, mid-seventeenth century, at the time Lemba is first recorded as a medicine of the
king and the nobility. This print, taken from the 1670 French edition of 0 . Dapper, Description de
VAfnque, bears inscriptions of the original 1668 Dutch edition, indicating the following details: (a)
king's palace; (b) wives' compound; (c) crier's tower;
(d) royal wine house; (e) royal dining house; (f) public audience court; (g) royal garden; (h) wives'
garden; (i,k) two fetishes; (1) road criminals are taken to capital punishment. (Courtesy of Depart
ment of Special Collections, Spencer Research Library, University of Kansas.)
Plate 4. Lemba medicine chest (n'kobe Lemba) from N'goyo bearing charac
teristic petal motif. Collected by L. Bittremieux prior to 1933. (Musée royale de
l'Afrique centrale, Tervuren, 35191.) Contents depicted in Figure 20.
Lemba, 1650-1930
Figure 1
Distribution of Lemba, based on territorially-designated ethnographic, textual,
and artifactual sources (see key to numbers)
Mpangu
Key
E = Ethnography B-D = Berlin-Dahlem Museum für

T = Text Völkerkunde
A = Artifact G E M = Göteborg Ethnographie Museum
LMS = Linden Museum Stuttgart
LR M = Leiden Rijksmuseum
MAC = Musée d'Afrique Central (Tervuren)
S E M = Stockholm Ethnographie Museum
1. (E) Dapper, 1670, p. 536 19. (E/T) Konda, "Mamundi" near 36. (A) Hammar, "BaKuta-BaTeke," GEM
2. (E) Bastian, 1874, pp. 170-3 Kinkenge, ca. 1918 1968.11.208, rattle, and
3. (A) Loango Expedition, 1875, B-D III C 20. (E/T) Babutidi, "Mamundi" near 1968.11.241, Lemba drum
347, engraved copper bracelet Kinkenge, ca. 1918 37. (A) Andersson, "BaKuta-BaYaka,"
4. (A) Visser, n.d., B-D III C 8136, 21. (E/T) Kwamba, "Mongo-Luala" near "Ntele," GEM 1 9 3 8 . 3 1 . 1 1 - 1 2 ,
"BaVili," balsa wood bracelet Kingoyi, ca. 1918 Lemba bracelets
model template 22. (A) Hammar, Kingoyi, GEM '38. (A) Maesen, "Tsanga," 1954, MAC
5. (A) Visser, n.d., B-D III C 1381 Oa-f, 68.11.241, drum, ca. 1910 53.74.1332, Lemba bracelet
"BaVili," clay mold for bracelets 23. (E/T) Kionga, Kingoyi, ca. 1 9 1 6 - 1 8 39. (A) Laman, "Mayombe," SEM
6. (A) Bastian, 1874, B-D Catalogue § 24. (E/T) Kimbembe, "Madzia,"ca. 1918 1919.1.445, drum
372, Lemba pipe 25. (E/T) Lunungu, "BaYaka," Indo, ca. 40. (A) Laman, "Mayombe-Sundi," SEM
7. (A) Bastian, 1875, B-D III C 4 2 3 , 1918 1919.1.437, drum
engraved copper bracelet 26. (A) Visser, "Kayo," 1904, B-D III C 41. (A) Anon., "BaKuta-BaYaka," "Ntele,"
8. (A) Visser, 1900, B-D III C 13871, 13744, Lemba sack charm GEM 1 9 3 8 . 3 1 . 1 1 - 1 2 , Lemba
Lemba sack charm 27. (E/T) Stenstrom, 1969, pp. 3 7 - 5 7 ; bracelets
9. (A) Hammar, 1910, GEM 68.11.241, Andersson, 1953, fig. 22, Musana 42. (A) Maesen, "Tsanga," 1954, MAC
"Nganda,"drum 28. (E/T) Fukiau, "Nseke-Mbanza," 1969, 54.74.1335, Lemba bracelet
10. (E) Dennett, 1907, pp. 1 1 , 8 9 , 9 1 , pp. 4 1 - 5 6 , figs. 3 7 - 3 9 4 3 . (E/A) Bittremieux, "Mayombe," 1937,
133 29. (A) Visser, "BaVili," 1904, B-D III C MAC 3 7 9 7 2 , Lemba n'*ooe
11. (E) GOssfeldt, Falkenstein, Pechuel- 18921, drum 44. (E/A) Bittremieux, "Kangu," 1937,
Loesche, 1879, p. 71 30. (A) Visser, "Luema River," 1905, L M S MAC 4 3 0 4 0 , Lemba n'kobe
12. (E/A) Bastian, 1874, B-D 38363, Lemba necklace charm 4 5 . (A) Bittremieux, "BaWoyo," 1933,
13. (E) Cuvelier, 1946, p. 326 31. (A) Anon., "Boma," L R M 1032/136, MAC35191,35192,2n'Jtabe
casting mold for bracelet Lemba
14. (E) Deleval, 1912
32. (A) Anon., "Banana," LRM 1032/53, 46. (T/E) Tastevin, Cabinda, 1935, pp.
15. (E/A) L a m a n . S E M 1919.1.583,
1032/59, 1032/60, Lemba 105-111. 191-7, 257-73
"Mukimbungu" Lemba n'kobe
bracelets 47. (A/T/E) Vaz, 1969, p p . l l 6 f . , 320f..
16. (A) Hammar, 1906, GEM 68.11.171,
33. (A) Maesen, "Tshela," 1954, MAC 413f.,420f., pot lids
"Babwende-BaYombe," at
53.74.1333, Lemba bracelet 48. (E) Bittremieux, 1925, "Kangu"
"Manianga-Mayombe," drum
34. (A) Lagergren, "Kitona," n.d., G E M 4 9 . (T) Nitu, "Masala, Luadi river, Zala
17. (A) Loango Expedition, 1875, B-D III C
1966.15.3, Lemba bracelet area, Cabinda," 1961, MS.
710b, brass bracelet
35. (A) Andersson, "Kinzaba," 50. (T/E) Malonga, "Lari," 1958, pp.
18. (A) Visser, 1901, B-D III C 13743,
"BaKamba," n.d., GEM 45-9, 51-61
"Kayo," Lemba sack charm
1938.28.16a, Lemba drum
Chapter 1
Introduction to Lemba
Lemba: Historic Equatorial African Drum of Affliction
Lemba, a major historic cult of healing, trade, and marriage relations,

came into being in the seventeenth century in a triangular region
extending from the Atlantic coast to Malebo Pool between today's
cities of Kinshasa and Brazzaville, and from the Congo (Zaire) River
northward to the Kwilu-Niari River valley (figure 1). Mpumbu
market, situated on a hilltop overlooking Malebo Pool, was the
western end point of the vast riveraih trade network which covered the
Congo Basin. From Mpumbu westward, rapids on the Congo River
required trade to the coast to follow land routes. Until railroads were
built from Malebo Pool to Pointe Noire and Matadi three centuries
later, caravans of porters brought ivory, copper, slaves, and other
products to the coast, and European wares such as cloth, guns, liquor,
and beads back inland over the same routes. Lemba controlled this
trade on the north bank, keeping the routes open, regulating local
markets, and assuring that the massive international trade did not
destroy local communities. In the mid-eighteenth century, for ex
ample, fifteen thousand slaves were being shipped annually from the
ports of Loango, Cabinda, and Malemba. These slaves were drawn
from inland societies which tended to see the trade, despite its
economic advantages, as disruptive. Conflicts of interest between the
trade and social order may explain why Lemba—a word meaning "to
calm" (lembikisa)—took the form of a therapeutic association, a
"drum of affliction" (ngoma or nkonko).
Lemba's illness is described in a variety of ways: as possession by
Lemba's ancestors, common in drums as a mode of affliction; as any
illness affecting head, heart, abdomen, and sides, that is, the key
organs of the person; as difficulty breathing, a typical witchcraft
symptom; as miraculous recovery from a deadly disease, and more.
However this erratic list of Lemba symptoms and causes tells less
about it than does identification of the types of individuals afflicted
and directed toward membership in Lemba. They normally were the
region's elite, prominent healers, chiefs, and judges, especially those
3
4 I N T R O D U C T I O N TO LEMBA
engaged in mercantile work. Their ability to succeed in commerce and

their aspirations to wealth made them vulnerable to the envy of their
kinsmen, thus in a sense marginal in the society and "sick" with the
Lemba affliction.
In typical drum of affliction manner, Lemba doctor-priests took the
"sufferer" in hand and administered the initial purification. If he
could muster further sponsorship, he would undergo full therapeutic
initiation before the priests and priestesses of the locality. A
ceremonial Lemba marriage with his leading wife, a lavish feast for
Lemba priests and priestesses as well as for the public, and extensive
instructions made up part of the initiatory ritual, before the priestly
couple were qualified to themselves perform Lemba's therapy.
Given such characteristics among many more to be explained in
this book, it is not surprising that Lemba is considered by those who
saw it in full bloom as an extraordinary institution, the most important
of the consecrated medicines (min'kisi) among those that achieved
corporate status. It is described as having been "a medicine of the
village"; "a medicine of the family and its perpetuation"; "fertility
medicine"; "the sacred medicine of governing" (Lemba i n'kisi
wangyaadila) according to one clan head; "the government of
multiplication and reproduction" (luyaalu lua niekisa), by a former
Lemba wife; and "sacred medicine integrating people, villages and
markets" (n'kisi wabundisa bantu, mavata ye mazandu), by a
contemporary merchant and clan head who wrote a local history. In
effect, although couched in the mold of a drum of affliction, Lemba
was the governing order in a region much of which had no centralized
institutions.
The Lemba region was bordered by prominent historical states—
centralized kingdoms—which predate the international trade that
began from the Congo coast in the sixteenth century (figure 2). South
of the Zaire/Congo River stretched the great Kongo kingdom with its
capital at Mbanza Kongo. Its northern province, Nsundi, extended
from the southern bank of the river across to the north, although
probably not as far as some early maps indicate. Smaller kingdoms
and chiefdoms such as Vungu, Mbanza Mwembe, and Mwembe
Nsundi existed along the river and northward into what became the
Lemba region. On the coast were situated the famous old states of
Loango, Kakongo, and Ngoyo, well documented by five centuries of
European explorers, traders, and missionaries. To the north and east
of the Lemba region could be found the great Teke (or Tio) complex of
federated lords around their king Makoko of Mbe.
Introduction to Lemba 5
Figure 2
Political and economic map of Lemba region, 1600-1930 (Based on maps in
Vansina's The Tio)
The distinctiveness of the Lemba system of government compared

to that of the surrounding kingdoms can be illustrated by briefly noting
different approaches to justice, the use of force, and the structure of
economic resources. In the kingdoms, judicial order was maintained
by courts of the principal chiefs or kings, which became courts of
appeal in specific cases. The instrument for enforcing order at the
disposal of some of the kingdoms included a hastily assembled army,
most fully developed in the riverain polities, to defend the capital
when it was under attack and to exert control over recalcitrant
subordinate chiefs. A third expression of state order in the kingdoms
was the levying of tribute by the center upon subordinate levels. At the
zenith of each of these kingdoms tribute networks were widespread.
Finally, the kingdoms were characterized by the symbolism of sacred
and semidivine kingship, which was passed out along the tribute
routes to extend the cloak of royal identity to outlying areas of the
kingdom.
9
Under Lemba s influence, justice, the exercise of force, and
economics were institutionalized in "horizontal" mechanisms, most
9
important of which was, beside Lemba s ceremonial exchange
system, the market and trade network. Local markets, sponsored and
owned by adjacent villages, were organized into cycles, or weeks, of
four days. A code of market laws prescribed the details of trade,
justice, capital punishment, and peacemaking. Although markets
existed in the kingdoms, these market laws took on greater prom
inence in the acephalous regions of the Lower Zaire. Feuds between
local groups could not, as in the kingdoms, be arbitrated by a central
regent or his army. Conflict had to be absorbed by communities
adjoining those of the antagonists through mechanisms spelled out in
the market laws. The vertical exchange of tribute for symbols of
kingship found its acephalous equivalent in the way Lemba's
protection of the markets acted, in effect, as a consecrated commodity
to be exchanged against significant goods. Lemba's priests enhanced
regional trade by maintaining peace in the marketplace. The effect of
Lemba's high initiation fees and its encouragement of redistribution
and consumption of locally-produced and trade-produced goods in
the seances, stimulated exchange. The regional network of kinship
ties established in the special Lemba marriages strengthened alli
ances between prominent local clans. By making prominent judges,
clan heads, and reputable healers, along with their principal wives,
priests and priestesses of Lemba, a regional integration was achieved
as well suited, perhaps even better suited, to the conditions of the
international trade as that which existed in the localized chiefdoms
and kingdoms.
Although Lemba disappeared from the scene as an active institu
tion in the first decades of this century in the face of Belgian, French,
and Portuguese colonial rule, mission activity, European commercial
hegemony, and mechanized transport, its three-century presence
across the north-bank Lower Congo/Zaire region, and its move with
slave-emigrants to the New World where it still survives, make it not
only one of the longest-enduring Bantu drums of affliction on record
but one which has not heretofore received its due attention in scholar
ship. Its significance is comparable to that of major West-African
cults such as Poro and Sande on the Guinea coast, Ikenga, the cult
of the hand in Eastern Nigeria, Bwiti in Gabon, Nzila in Western
Zambia, as well as Beni-Ngoma and the Kalela Dance in East
Africa, and the Copper Belt, or the Isangoma diviners association in
South Africa.
Going beyond the concerns of historical social research in Africa, a
study of a major therapeutic society has a special modern appeal
arising from our postindustrial concerns with the nature of organiza-
tion in handling chronic afflictions not dealt with by biomedical

professionals. At a time when noncentralized therapies of the
"anonymous" variety are scoring successes in such areas as alcohol
ism, drug addiction, child abuse, cardiac rehabilitation, and a myriad
of other cultural illnesses, research into alternative modes of concep
tualizing and organizing health is in order. Planners in Africa itself
may find the study of a highly developed example of a drum of
affliction helpful in drafting present health care policies toward
traditional therapies, or in the development of secular therapies based
upon indigenous genres of care and organization. The present study
will, we hope, illuminate the implications of a radically broadened
definition of health, and of illness, going beyond the strictly personal,
physical, realm of modern medicine.
Issues in the Study of Lemba of Theoretical Significance
The study of a major historical drum of affliction of Equatorial Africa

abounds with issues of theoretical significance. Perhaps the most
pressing need is to bring this discussion into closer congruence with
sophisticated indigenous theories on the subject. Scholarly analyses
have frequently made distinctions where none exist in indigenous
thought, or have failed to take note of important cues in indigenous
expressions. This problem is particularly acute in the interpretation of
the corporate, therapeutic institutions which controlled major social
and economic resources, thus combining functions normally kept
discrete in Western thought. A more accurate understanding is
required of the complex of cults, shrines, movements, and political
structures in relation to therapeutics as a type of social process. The
historical perspective of this study implies a concern for the rise,
persistence, and decline of particular institutions, that is, the "waxing
and waning" characteristic of cultic and therapeutic institutions in
Bantu Africa. Scholars are divided over whether cults, drums, and
shrines are to be seen as merely reactive barometers of social
processes, or as creative efforts actually to construct social orders.
The fundamental question behind both views, and the one which must
be asked before embarking on this venture, is: what, in effect, are the
independent and dependent variables of Central-African social and
cultural change? And, necessarily, what theories are best suited for
answering this question?
The present volume brings several analytic perspectives to the task

of understanding Lemba and institutions like it. One perspective,
developed in Part I, reconstructs the societies of the Lemba region in
terms of political and economic structures over the past three
centuries, roughly the period of Lemba? s rise, its domination, and its
demise. This is essentially a history of corporate institutions and the
control of public resources—people, food production, symbols of
prestige, strategic locations, and key political roles. Although con
ventional history rarely views therapeutics together with corporate
politics, this juxtaposition is necessary to an understanding of the role
of consecrated medicines and their clientele in the public life of the
seventeenth- to the early twentieth-century Western Congo.
Another perspective, developed in Part II, uses a more precise,
closer-range, analytic language for the understanding of rituals and
myths of Lemba initiatory and therapeutic seances. This perspective
draws from structuralism, symbolic anthropology, and semiotics in an
effort to order and draw maximum understanding from the diverse and
uneven sources available for a project such as this: fieldwork notes,
historic indigenous accounts, colonial archives, travelers accounts,
and museum collections. (A fuller discussion of the particular theories
used here may be found in the introduction to Part II.)
A further perspective, developed in Part III, uses the combined
analytic tools of structuralism, symbolic anthropology, and semiotics
to grasp self-conscious ideas of illness causation and classification in
Lemba thought, the nature of therapies, and the rather explicitLemia
image of the healthy—or virtuous—society.
These approaches move sequentially from an historical analysis of
the organization of public resources, to a closer-range study of
symbols and meanings in ceremonial events, to an understanding of a
self-conscious Lemba therapeutics found among a limited segment
of the populace—the elite faced with problems of wealth and power in
a redistributive, community-based society. All of these approaches
are needed to account for the forces, consequences, and internal views
of social change.
Understandably, there has not been general agreement among
scholars regarding the character and purpose of African therapeutic
cults. Indeed, the history of scholarship on them in the Lower Congo
and in Africa as a whole reads much like a history of social science in
the twentieth century, reflecting many trends and perspectives.
Because many of these perspectives continue to maintain credence in
one circle or another of scholarship, a detailed review of their
applications to Lower-Congo and Equatorial-African societies seems

warranted.
During the early colonial years African therapeutic cults aroused
lively scholarly attention in Europe under the rubric of "secret
societies," with the Lower Congo being in the forefront. Lemba,
however, due perhaps to its practitioners' extreme skill in eluding
investigation or to the relative absence of masks and exotic displays in
its ritual, rarely figured in the discussions. A summary of the views of
Belgian ethnologist Edward DeJonghe from the turn of the century
until World War II offers an understanding of prevailing scholarly
1
approaches to secret societies.
Psychological interpretations of secret societies were led by
2
Heinrich Schurtz whose Altersklassen und Mannerbunde (1902)
postulated that secret societies constitute a synthesis between puberty
rites and adult social groupings, in turn reflections of the two dominant
instincts of human life: a natural sexual instinct governing puberty
rites, and an artificial instinct (or need) for gregariousness leading to
age grades and other ties across local communities. Accordingly,
women, possessing an abundance of the sexual instinct, belonged in
the home with children, whereas men, possessing an abundance of the
social instinct, belonged in public life. Despite their patent Victorian
ethnocentricity, these theories were influential, finding their way into
the work of Meyer, Meinhof, Weule, Lasch, Freud (see his Totem
1
and Taboo), and Wundt, whose Folk Psychology* had a strong
5
influence on Makinowski. DeJonghe recognized the wide appeal of
the psychological approach, but doubted the applicability of such a
universal explanation to an institutional type whose forms varied so
extremely from place to place.
Religious studies of secret societies were heavily influenced by the
work of James Frazer, including his well-known Golden Bough (1890)
6
and Totemism and Exogamy (1910ff.). In these works Frazer
linked rites of puberty and rituals of death and resurrection—whether
simulated or real—to the social process of assimilating a youth to his
clan or totemic group. Although Lower-Congo ritual symbolism used
death and resurrection symbolism in initiations to secret societies,
5
DeJonghe rejected the full application of Frazer s totemic theories on
the basis that totemism was not worldwide, hence could not be applied
naively everywhere, especially where it was barely represented as in
the Lower Congo. Of more validity in the study of secret societies
DeJonghe found the work of Durkheim's sociological school and the
new field of European folklore, especially Van Gennep'sLes rites de
7
passage (1909). Addressing the same issues as Frazer, Van Gennep
postulated his now well-known stages of ritual not only in initiation to
special associations but in most social transitions, groups, and
activities. Separation from the old status or role, a threshold,
"liminal" stage of transition, and reincorporation in a new position
marked all status changes, however varied the form. Van Gennep's
research model was praised for its insight, but judged too subtle for
work with most sources available to scholars of the time. Although
Van Gennep himself used examples from the Loango coast in his
famous book, full-scale application of his analytic ideas to Central-
African ritual would need to wait decades until Victor Turner would
use them in studies of Ndembu drums of affliction, a subject taken up
later in this chapter.
The culture-historical tradition of research under the inspiration of
Leo Frobenius and Fritz Grabener was DeJonghe's own theoretical
preference for the study of secret societies. Great praise had been
lavished on Frobenius' magnificently illustrated work Masks and
8
Secret Societies (1898). Masks from Loango and Cabinda figured
prominently in the volume, and although Lemba was ignored—
because it had developed few masks?—other Kongo societies such as
the puberty initiation activities of Khimba, and a parallel society,
Ndembo, were featured. Frobenius had given West-African secret
societies his attention under the rubric of "culture areas" and"culture
complexes." Masking in this part of the world had a Malaysian origin,
he believed, with close associations to sun and nature worship and
other elements of the African worldview. Central to this worldview
was its "manism," whereby ancestors over the course of time
underwent a gradual "spiritization" (Vergeisterung), in the process
becoming deities, abstract ideas, masks, and cults. Although the
ancestor cult of a particular social group might be central to this local
world view, other levels of collective spirit symbolism could be
interpreted as having grown from it. Masks represented the effort to
consolidate disparate elements in spiritization, whence arose two-,
three-, and four-faced masks throughout the secret-society culture
area of Africa. This notion had a direct impact on many Africanists,
including Karl Laman whose interpretation of Kongo consecrated
medicines is that they are the representation of ancestors whose
9
individual identities have been lost in time.
According to DeJonghe, the foregoing theoretical approaches to
Africa's and especially the Lower Congo's secret societies were all
either too reductionist, prone to explain all in terms of individual
needs and instincts, or too nominalist, forcing data into abstract

categories. He advocated an inductive approach based on the
ethnographic material itself. In this vein he recommended the use of
Fritz Grabener's special application of Frobenius' culture-area
method, finding ideal Grabener's study of Oceania's "patriarchial
totemic cycle," a complex of elements including the puberty rite,
death and resurrection symbolism, the boomerang, lunar mythology,
10
the breaking out of teeth, and so on. Just as analysis of these
elements as parts of a complex had clarified Oceanic cultural history,
so the institutional complex of Africa's secret societies would be
clarified by a similar analysis.
Because African cultural complexes were, however, mixed to
gether in a tangle of cross-cutting influences, noted DeJonghe, it
would be necessary to begin with inductive studies, shorn of precon
ceived ideas, in the comparison of peoples and their characteristic
complexes and in the geographic distributions of the latter. DeJonghe
identified twelve elements which, taken as a complex, characterized
Lower-Congo secret societies. These were: the extension in space and
time of a name for the society; a normative age for adepts; criteria and
procedures for admission of adepts; duration of ordeals; place of
ordeals relative to other significant locations; the entrance ceremony;
special names for the cult and its various features; deformations or
special identifying decorations donned by participants such as tattoos
and costumes; instructions given the adepts; prescriptions and
prohibitions; exit ceremonies; and activities related to, but outside,
11
the direct initiation. Khimba Longo,Kimpasi,
9 zndNdembo were
analyzed in terms of this grid; Lemba was again overlooked.
In the conclusion of a lecture given in 1923 to the Semaine
d'Ethnologie Religieuse DeJonghe departed from his well-reasoned,
academic survey to give a more visceral summation of African secret
societies. They are a "threat to the colonial order," he began.
Although they are " a very old type of organization, new versions arise
under your eyes in opposition to the European." Although they are
often "tribal in their boundaries, they can become intertribal," like the
notorious Aniota leopard cult which utilized violence on its oppo
nents. DeJonghe characterized the institution "from Senegal to
Congo" as a "secret religious fraternity whose eponymous tutelary
spirit, often a forest spirit, appears in the village' speaking? with a bull
roarer, not out of love of music but with an aim of tyrannizing and
dominating. They are successful in convincing the populace, for
everywhere there has been resistance to colonial government. They
12 INTRODUCTION TO LEMBA
axe, in short, a significant political force, regulating justice, policing

markets, establishing the price of merchandise, issuing boycotts, even
killing to get their wishes fulfilled. Their front is extremely varied;
some resemble castes, others cults to certain spirits. Yet others are
grafted onto initiations; others pursue social and political aims."
Concluded DeJonghe, "it is no exaggeration to say that secret
societies are one of the largest and most serious infections (plaies) on
12
Africa!"
DeJonghe's impassioned remarks on secret societies, despite their
partisan procolonial sound, were better than received scientific
theories in one important sense. They acknowledged the substantial
political character of these institutions. Unlike kingdoms and chief-
doms of Africa which had been early acknowledged by colonialists to
be political entities and had been incorporated where possible into
the colonial structure, the horizontally-organized "drums" were not
understood in political terms at all, and even when they were, they
could not easily be accommodated. Therefore, in the years sur
rounding DeJonghe's 1923 lecture, which was itself two years after
the massive disturbances accompanying the appearance of the Kongo
prophet Kimbangu, secret societies of the Congo coast, including
Lemba, were banned for resisting colonial "progress," and for
diverting economic and labor resources away from colonial goals into
lavish feasts and resistance. With this redefinition of the "secret
society" from object of general scholarly curiosity to mechanism for
resistance to colonialism, went the gradual decline of" secret society"
13
as a research paradigm. And, although writers like DeJonghe, Van
14 15
Wing, and Bittremieux continued to find and describe new secret
societies in the twenties and thirties, for all intents and purposes this
paradigm had been exhausted in serious ethnological scholarship by
World War II.
Social research in Africa since World War II has inserted
institutions such as Lemba into a more diversified rubric of move
ments, social and political organizations, cults, and ritual idioms,
especially the therapeutic. Among the most important of the postwar
scholars in this field has been Victor Turner whose fieldwork in the
fifties on the ritual process among the Ndembu of Zambia has set the
pace for a generation of scholarship. In particular, Turner has defined
precisely the nature of drums of affliction—based on the Bantu
cognate term ngoma—or, as they are more widely known, "cults of
affliction." Turner's portrayal of the cult of affliction begins with a
cultural interpretation of misfortune (for example, bodily disorders,
accidents) in terms of domination by a specific nonhuman agent, and
proceeds with an attempt to remove the source of misfortune by

bringing the sufferer into closer association with a cult group
16
venerating that specific agent. Turner's carefully conducted work
among the Ndembu revealed a range of such drums devoted to
specific problem idioms in areas such as hunting, reproduction, and
sickness, as well as in growing dislocations resulting from colonialism
and labor migration.
Turner's work on the ritual processes giving rise to drums of
affliction stresses the presence of a system of thought about mis
fortune and its treatment which may well be as widespread as the
Bantu-speaking societies of Central and Southern Africa. A number
of excellent field studies have identified common elements of this
system, from Equatorial Africa to East Africa, and from the rain
17
forests to the mountains of the south.
The causal premises behind health and disease in this "Bantu"
cosmology trace all of life to a central source of power, often named
God or some spirit immanent in nature. This power is mediated by
middle-range spirits and consectrated human priests or visionary
prophets who maintain contact with, or derive inspiration from, the
source of power and life. Misfortune, including disease, is any
condition whether social, personal, physical, or mystical which
falls short of the ordered universe of life—in other words, chaos.
Balance between the universe's elements is a subordinate theme, a j
is purity.
A crucial cosmological notion, therefore, is the distinction drawn in
many Bantu societies between "naturally-caused" (God-caused)
diseases or misfortunes and those attributed to "human cause." The
18
former misfortunes "just happen" or are "in the order of things," as,
for example, in the death of a very old person or in an affliction with
readily recognized symptoms and signs which respond to treatment
as expected. A widespread range of treatments such as plant
preparations, massages, and manipulative techniques are appropriate
for afflictions of this type, as are nowadays techniques of modern
biomedicine practiced by Western-trained doctors and nurses in
hospitals and dispensaries.
In contrast to natural misfortunes and diseases are those caused by
chaos in the human world or in the relationship of humans to their
environment. An individual may bring disease and suffering upon
himself by disregarding social etiquette, ignoring good eating habits,
or by turning his back on kinsmen, elders, ancestors, and spirits. An
aura of ritual pollution frequently accompanies sickness by "human
cause," requiring the sufferer and his fellows to seek ritual purification
through sacrifices and confessions so as to achieve reintegration with

the good graces of society. Most human-caused affliction in Bantu
thought is attributed to the evil intentions of others, or situations of
contradiction in which persons are at odds or cross-purposes with one
another, as, for example, in the struggle to distribute land equitably
from a limited estate at a time when the dependent population is
increasing or in launching an enterprise for profit in the face of a strong
ethic of the redistribution of goods. Such situations are believed to
incur the ill will or envy of others and to lead directly to the breakdown
of health in a person, to visible physical sickness, or even to the
person's death. This belief in mystically channelled ill wishing
operates to reinforce the morality of social redistribution and loyalty
to family and kin.
Bantu therapeutic systems follow from these assumptions about the
nature of the world and the causes of misfortune, articulating
techniques—empirical, social, symbolical—and their specialized
experts. Not all medicines are highly specialized. Many are house
hold techniques practiced by parents on children, or by anyone on
himself. Yet in areas of life where there is crisis, transition, danger,
recurrent accident, high responsibility, or a focus on core social
values, consecrated medicines appear, complete with origin charters,
exact codes for their use, and the dangers of their misuse. These
consecrated medicines may be techniques, chemotherapeutic treat
ments, behavioral procedures, or highly magical and esoteric affairs.
The emergence of a consecrated medicine in an area of life probably
derives from the perception that the technique so consecrated, or the
ingredient, is powerful, effective, and in need of legitimate control. It
is said that "a medicine that can kill, can also heal," and therefore
must be carefully used and authorized, the same as in other
19
therapeutic traditions.
A Bantu therapeutic system is commonly, then, organized into a
series of specialist types (for example, nganga, pi. banganga) each of
which has knowledge in one or more specialized and consecrated
20
medicines. Among the Kongo, classically, the nganga mbuki (from
buka, to heal) is the local doctor who uses herbs; the nganga n 'kisi is
a more advanced or powerful doctor competent in numerous conse
crated medicines, especially those pertaining to anger and magically
caused afflictions; the nganga ngombo (named for the basket he
uses) is a diviner. The diviner's specialty is the sorting out of particular
details of cases brought to him or her and the offering of an expert
diagnosis of the affliction or misfortune's cause, whether it be of
natural or human cause, or another, spirit-related, cause. Because of

its integral place in the analysis and interpretation of human
misfortune in Bantu society, divination has responded sensitively to
social change. Particular perspectives in divination have come and
gone; inspirational diviners and prophet-seers, even mission-trained
pastors and priests, have taken on the role in the past century. And
where the divination specialist has disappeared as a consecrated role
specialty, kinsmen of sufferers do this work in the form of a "therapy
21
managing group." The institution of divination—analysis and
diagnosis—has permitted Bantu therapeutics to generate new conse
crated medicines, like surface transformations, without violating or
changing basic premises.
In most local Bantu therapeutic systems a few of the consecrated
medicines take on the public stature of corporate drums of affliction.
As Turner has noted for the Ndembu, sickness is seen as a sacred
calling, manifested in the form of a possession. In a consecrated
medicine that becomes a drum, those afflicted and initiated to the cure
or stabilized in their relationship to the sickness are the best suited to
become doctors of the ailment. It is a form of religious immunization
such that as the disease is conquered, the possessing spirit is placated;
the ordeal purifies and energizes the sufferer, placing him in debt to
society so that he is henceforth expected to devote his newly acquired
gift of mediumship in a specialized idiom to the service of others.
Reference in the foregoing sentences to social change, the divina-
tory "reading" of current events, the interpretation of new kinds of
misfortune, and the generation of new consecrated medicines with the
creation of major drums of affliction that wax and then wane call to
mind the fact that any standardized picture of Bantu therapeutics is
only as good as its depiction of local variance and historical
vicissitude. The perspective must be historical. The waxing and
waning phenomenon of consecrated medicines and therapeutic move
ments, linked as they are to a close reading of misfortune in
individuals' lives, can be used to understand the nature of social
instability, the sources of stress and chronic affliction in a society.
Read longitudinally, that is historically, these taxonomies, as they
unfold, become barometers of the local human condition. (In Chapter
2 such an historical reading will be done of consecrated medicines in
Loango over three centuries to detail the background against which
Lemba emerged and endured, Loango being the only region of
22
Equatorial Africa known to have this depth of documentation. )
Read latitudinally, that is synchronically at a given moment, medical
taxonomies indicate the contours of the system of dealing with

adversity and the lines of shifting orientation. Such a slice of medical
taxonomy is offered here from the Lemba region to situate the
theoretical discussion more fully in evidence pertinent to the subject
of the book.
In about 1900 a Lower-Congo/Zaire writer sketched the then
current array of consecrated medicines (min'kisi) in the north-bank
region, noting those that were of major status, that is "those drummed
up with ngoma." The inventory included such areas as "personal
physical growth," "a child's upbringing," "spirit children and how to
deal with them," "twinship and the parenting of twins," "headache,"
"purification with the cupping horn," numerous approaches to
"divination," "origin, residence, identity," a variety of women's
troubles, and at least five drums of affliction: Kinkita, devoted to
"clan leadership," Bunzi and Matinu for "chiefship," Mbola for
"water spirits," Kkondi for "judicial affairs," andLemba for "order
23
in markets and public sites." A static taxonomy such as this may be
studied in terms of analytical questions that indicate points of change.
Thus, the inventory of consecrated medicines may indicate a change
of particular items within consistent categories like divination, or it
may indicate the emergence of new categories and new medicines. In
line with this, the categories which contain major public medicines—
drums, cults—may demonstrate either a waxing and waning of
particular organizations within them, or the emergence of innovative
categories and public organizations.
Considerable scholarship in recent years has gone into the discern
ment in Central- and Equatorial-African cult history of the rise of
large-scale, regional units in conjunction with historical forces such as
trade networks, colonial empires, labor migration, and the modern
24
nation-state. A crucial analytical distinction has been drawn, once
more by Turner, between areas of ritual organization that represent
local "ancestral" identities and those which represent "earth"
25
identities. The former are maintained by political rituals and
divisive polarities such as lineage segmentation, whereas the latter are
maintained by fertility rituals and nonsectarian religious specialists.
Often political conquerers are aligned in opposition or in contrast to
indigenous priests. Citing examples from the Shilluk of the Sudan, the
Tallensi of Ghana, and the Shona of Zimbabwe, thus from three
widely differing cultures across Africa, Turner suggests that this
distinction between "exclusive" sectional ritual identities and "inclu
sive" ritual identities creates a system of shrines or medicines which is
characterized by overlapping and interpenetrating fields of ritual

relations: the first type of identity representing power divisions and
classificatory distinctions, including factional conflict, and giving rise
to homicide; the second type representing ritual bonds between
groups, common ideals, and values, such as are needed in the
purification of the land by earth priests in the event of fratricidal
26
murder in some societies. Although the former ritual levels are not
displaced, they are overlaid, as it were, with universalistic ideas of
ethics and morality such as brotherhood, purity, love, the "kinship of
all people," and related notions found in world religions.
The important distinction drawn by Turner, and others, between
"exclusive" and "inclusive"ritualorders may readily be discerned in
the taxonomy of sacred medicines and drums from Lower Zaire given
above. Those pertaining to chiefship (Matinu, (Ma)Bunzi) and to
clan leadership (Kinkita) are exclusive, whereas those pertaining to
water spirits (Mbola), judicial affairs (N'kondi) and order in markets
and public places (Lemba) are inclusive. Particularly N'kondi, the
well-known nail-and-blade charm used in the regulation of feuds
between local groups, and Lemba were the most explicit articulations
of regionalism.
More work has been done recently on "regional cults" as reported
in a volume of this same title recording the 1976 Association of Social
Anthropology (ASA) conference. Regional cults are defined as cults
"in the middle" between the "parochial cult of the little community"
and the world religion with its universal form. The central places of
these regional cults are shrines in towns and villages, at crossroads or
even in the wild, "where great populations from various communities
or their representatives come to supplicate, sacrifice, or simply make
27
pilgrimage." Such cults have a topography of their own, con
ceptually defined by the people themselves and marked apart from
other features of cultural landscapes and ritual activities. This group
of scholars emphasizes the fixed central shrine as a key feature in
regional cults. Contributions from the Islamic north to the Bantu
southeast of Africa are used to illustrate this. Van Binsbergen, writing
on regional drums of affliction in Western Zambia, emphasizes as
criteria the specific healing idiom, representation in congregations
over a vast area, overarching organizational ties binding local
chapters into a region, further identified by controlled information and
communication within the organizational structure. Nzila and Bi-
tuma, the major Western Zambian regional cults of affliction, have
28
central shrines and headquarters for a centralized staff. Schof-
feleers' study of the Mbona cult of Malawi also emphasizes these

29
points. .
The criteria of regional cults developed by the ASA group,
which are largely built upon earlier work by Turner, suggest that
Lower-Zaire historical developments in this area may be some
what anomalous in one respect. The cults or ritual orders most
readily expressing general values and identities, namely Mbola,
Nkondi, and Lemba, lack centralized shrines. Of those expressing
exclusive values, only Bunzi has (or had) a central shrine (in
Ngoyo kingdom near the Congo River at the Atlantic Coast) and
thus meets the criteria for a regional cult. As more detailed
accounts in Chapter 2 will show, already by the sixteenth century
Bunzi's priests were consecrating the kings of Loango, Kakongo,
and Ngoyo on the coast. In time, apparently, Bunzi spawned
secondary "MaBunzi" throughout the Mayombe, with their sub
sidiary territorial shrines. This may have been the result of a
regional consciousness. But what is even clearer about cultic
changes in the sixteenth and seventeenth centuries is the emer
gence of movable shrines such as those kept by Lemba priests,
their tukobe. Comparable developments characterize the rise of
movable shrines in the hands of Tio lords, vis-á-vis the fixed earth
30
shrine of the Tio king to earth spirit N'kwe Mbali.
I mention these details only to suggest that the fixed shrine center,
and a centralized organization with a recognizable leader, may not be
as instrumental to the creation of a regional consciousness as the ASA
31
group suggests. In some cases, such cultic centers as Bunzi may
actually lose out at the expense of other, moving, ritual varieties of the
kind that Lemba characterizes. Bunzi will receive fuller comment in
later chapters, for it is apparent that it consecrated not only the three
coastal kingdoms but a number of horizontal cults such as Lemba in
the eighteenth and nineteenth centuries.
The characterization of regional cults is no more important in
scholarship, however, than the question of what forces bring them
about. How are the transformations to be explained? As suggested
earlier, analytic distinctions such as that drawn by Turner between
inclusive and exclusive ritual orders and by the ASA group between
cults with regional organization and those with localized small-scale
organization are made so that the student may obtain a grasp of the
phenomenon of social change. As Werbner, in his introduction to the
ASA volume, points out, much of the work on African history in the
therapeutic and religious arena is grounded in the correspondence
theory of religion, propounded by Durkheim, and before him
32
Robertson-Smith. The theory states that religious consciousness,
cult forms, and the deities worshipped or evoked are a function of the
social system, its contours, and its changes. The emergence of
regional cults, and the waxing and waning of drums of affliction,
religious movements, and shrines and therapeutic systems, are
readily explained in terms of social change. Thus, Van Binsbergen,
writing about the three major cults of affliction in Western Zambia—
Nzila, Bituma, Moya—states that the political and economic
changes of recent decades constitute the "motor" behind religious
transformation. The cults and the way they draw differently from the
populace express "the emergence of interlocal structures and move
ments of peoples, due to precolonial and colonial state formation,
raiding, long distance trade, labour migration, all of which calls for
new religious and social forms to legitimate new structures and meet
33
existential and interpretive needs of people."
In a comparable line of explanation, Van Velzen offers an explicitly
materialist interpretation of the emergence of the Gaan Tata, high-
God, cult among Bush Negro and Maroon tribes of Surinam and
French Guiana, who from 1880 to 1920 and decreasingly thereafter
controlled therivertraffic in connection with the exploitation of gold
34
in the interior. The cult emerged in several locations along the
transportation network and then coalesced into a regional cult with a
central shrine and a hierarchic structure in its heyday in the 1890's,
monopolizing river transport and maintaining high wages and freight
prices for the Bush Negro and Maroon workers. The resemblances
of Gaan Tata to Lemba are so striking that I shall develop in fuller
detail Van Velzen's approach.
Van Velzen calls the independent variable in his analysis of
35
regional cults the "alteration in the mode of production." In the case
of Gaan Tata this was the introduction into Bush Negro and Maroon
society of a massive new resource from the outside—the discovery of
gold, and the consequent need of transport by outside prospecting and
mining interests—resulting in a sudden influx of wealth and setting off
a polarization of the society. The centrifugal effects of this polari
zation created a powerful need to allay anxiety about witchcraft on the
part of those benefitting from the wealth. Simultaneously, there
emerged an increased vigilance for a stonger morality, a more
generalized, even universal, ethic, supported by anew and centralized
high-God cosmology at odds with the traditional one. Van Velzen
argues that the universalizing ethic in Gaan Tata was an outcome of
the cult's oracle to protect the entrepreneurs from kin envy, to uphold
their reputation, and to establish antiwitchcraft standards of social
life. Thirty to forty percent of goods received were channeled to

kinsmen; another substantial portion was dumped at Gaan Tata's
shrines. Only a minimal portion was "used" by an elite. In effect, the
cult managed to thwart the more exaggerated forms of elitism.
How Lemba resolved this problem will be a major discussion in the
present work.
In some writing on cults, drums, and therapeutics, there is an
undertone of criticism toward the correspondence theory, which
tends, it is argued, to represent these ritual orders as tails that are
wagged by the dog of the socioeconomic order. Such explanation, it is
asserted, often neglects the generic qualities of heightened concern
with ethics and morality; the development of sophisticated concepts
of redemption such as grace, purity, and balance; the methods of
sacrifice, prayer, and devotion before deities worshipped; or, above
all, the manner in which individuals seek redemption in the midst of
chaos, suffering, and evil. This criticism echoes a point made by
Burridge about the explanation of millenarian movements in the
36 37
social sciences, or by Geertz on the general study of religion.
Turner cleverly addresses these issues without abandoning the
socioeconomic correspondence theory altogether by locating most
religious activity in the "liminal" mode of human life, that apparently
universal aspiration to escape, periodically, the structures of social
life with their deadening formalism and routine.
But surely, the challenge in theory-building is to offer a sound basis
for interpretation of intrinsic characteristics of regional cults, cults or
drums of affliction, consecrated medicines, independent churches,
and related rituals; and not to pass them off as mere dependent
variables—mirrors—of socioeconomic forces, or unpredictable,
inverse occurrences in the cracks of society's structures. Put another
way, the theoretical question most acutely in need of attention is: do
these institutions creatively shape society? Or are they a reaction to
it? What is the most sophisticated way of measuring the mixture of the
two possibilities?
Lemba offers new and important information about the subject
under discussion. It provides evidence from the study of a more
substantial therapeutic order than has been documented heretofore of
the way cults can act as a creative impulse to shape a region's society.
Certainly it reacted in some senses to the external pressures of the
international trade. But several alternative responses would have
been possible, among them doing nothing. Ranger argues that
choosing the institutional form of the drum of affliction made the
Tntroduction to Lemba 21
response a creative, positive effort in the formation of a "new

society." Just as kinship has been seen in some social theories as a
metaphor for the construction of the social order, so here, and in other
drums of affliction, sickness and healing provided a metaphor for the
38
construction of a new social order. The concept of a new society
ordered in terms of cults of affliction is frequently idiomatic and
partial, dealing with such areas as twinship, hunting, purity, witch
craft eradication, or fecundity. But potentially, therapeutics may be
the metaphor serving to facilitate consolidation of substantial re
sources, material and human, and to aid long-term re-ordering of
institutions of redress, economic redistribution, and ideological
change.
Theoretical discussion of the conditions giving rise to new cults
should, therefore, compare ritual and institutional phenomena across
a broad spectrum, including the formation of states. Once this is done
it is immediately apparent that the various theories of cult emergence
are the same ones used in accounting for the rise of traditional states,
at least in Africa. Most are "correspondence" theories, such as that
which makes economic surpluses a precondition of state emergence,
or that which links the rise of the centralized state to an expanding
regional sphere of activity in trade, warfare, or the movement of
peoples. I cannot here survey in any detail the voluminous literature
on state formation skillfully handled or reviewed by such scholars as
39 40 41 42
Vansina, Goody, Miller, de Heusch, and others. My concern is
rather with the apparent lack of discussion between those who study
state formation and those who study cult cycles, and the failure of
these persons to address the question of why such similar precondi
tions as both groups often consider give rise in the one case to a
"state," and in the other to a "cult."
A theoretical framework is called for which bridges organizational
gradations from centralized to decentralized; from an institution with
multiple idioms to one with a specific idiom; from collegial forms of
authority to serially-occupied offices of solitary authority, and so on.
Much writing on "states" is couched in assumptions of nineteenth-
century teleology by which less centralized social forms become
transformed, through economic, social, or ideational impetus, into
centralized forms. The present work will interpret the divergent
institutions such as centralized shrines (Bunzi, N'kwe Mbali),
kingdoms (Loango, Kakongo, Ngoyo, Tio), movable shrines or major
drums (Lemba, the Tio nkobi), local horizontal specialists'
medicines (N'kondi feud resolution techniques), and other institu-
43 44
tions in terms of corporation theory as developed by Maine, Smith,
45
and for Central Africa by MacGaffey. Corporate theory acknowl
edges authority and power of both centralized and decentralized
polities in terms of corporateness: that is, a presumptive perpetual
aggregate with a unique identity; having determinate social boun
daries and membership; possessing the autonomy, organization, and
agreed upon procedures to regulate exclusive collective affairs. In
other words, a corporate group forms a social structure around a set of
diverse issues. InLemba, this would include the maintenance of order
in market places, regulating trade over long distances, establishing
marriage alliances between local lineages, and healing the personality
disorders of the "marginal" mercantile elite.
Corporate theory goes on to elaborate the measure of a group's
corporate strength by the manner in which it develops leadership roles
(or commissions, consecrated leadership roles); how this authority is
delegated, and administratively coordinated; how a constitution
making explicit the premises and understandings of the corporation is
articulated. In Lemba, this would include the adaptation of regional
cosmologies to unique Lemba values, and the development of codes
of behavior, purity laws, and levels of morality for the Lemba
membership and for public society under its influence.
Corporate theory makes a further distinction between the corpora
tion sole and the corporation aggregate, the former a leadership role
or commission standing for the group in which a series of individual
officeholders move through an office in succession; the latter an office
occupied by multiple officeholders simultaneously. The distinction,
in theory, readily clarifies the nuances of difference between a
centralized kingdom, chiefdom, or shrine, and a network-like aggre
gate of figures in a major drum of affliction such as Lemba. That this is
a distinction of degree and not of kind might be suggested from the fact
that drum symbolism, denoting consecrated leadership, exists across
the set of political types.
Corporate theory, furthermore, offers a theoretical avenue for an
understanding of the phenomenon of "waxing and waning" of
kingdoms, drums of affliction, and shrines, as well as the transforma
tion of the one into the other. Corporations sole may be based upon
the consolidation of corporations aggregate. One suspects that this is
the way kings emerged in some African states. It is clear however that
transformations can go the other way. The proliferation of BunzVs
"daughters" in the cults Lemba, Pfemba, and Lunga (see figure 26
below) and into chiefdoms inland represents a case in point.
Corporate theory, finally, distinguishes between corporate groups,

discussed in the foregoing lines, and corporate categories. These latter
lack the power and authority of the corporate group, because they lack
the organization, internal hierarchy, and command of resources.
However, a corporate category meets criteria of presumptive per
petuity, determinate social boundaries, identity, and membership. In
the Lower Congo, for example, a matrilineal clan is a corporate
category because of the putative descent of its members from a
common ancestor. It is distinguished from the local clan, or set of
related lineages, in that these have internal organization, leadership,
common affairs, property relations, and a command of resources. The
distinction between corporate category and corporate group would
also, at the level of theory, explain the difference between adepts of a
simple consecrated medicine whose only common experience would
be their individual interaction with a given medicine and its priests,
and adepts of a powerful corporate medicine, a public "drum of
affliction," whose common affairs, command of resources, and
coordination would give them greater power and authority in public
affairs.
Once spelled out in this wise, it is apparent that the waxing and
waning of kingdoms, centralized shrines, diffuse cults, drums of
affliction, ephemeral movements, and consecrated medicines can
best be accounted for in terms of a unified theory of differential
corporateness. This I attempt in the next two chapters for the Lemba
region: the Malebo Pool (Mpumbu)-Cabinda-Loango triangle north
of the Congo River, from the seventeenth to the early twentieth
centuries.
Parti
The Public Setting of Lemba :

Management of Society
and Its Resources
"Les auteurs o n t . . . des notions très vagues sur les populations de la

rive droite du Congo inférieur."—J. VAN W I N G , Études BaKongo,
(1959)
Chapter 2
Economic, Social, and

Political History of the
Lemba Region
Environment, Resource Use, Adaptation,

Trade and Currency
The Lemba region is defined by natural boundaries, the Zaire and
Kuilu-Nyari River systems on the south and north, respectively, and
by the ocean on the west. These river valleys are, however, quite
different one from the other. The Kwilu-Nyari, which begins in the
area and flows through a fertile alluvial plain, offers excellent
possibility for cultivation. The Zaire valley, by contrast, is the end-
point of the great and mighty river system of Central Africa. It is
marked by its deeply eroded canyon and dangerous cataracts, making
it hostile in places to boat travel and useless for cultivation. Fishing is
the only resource common to both rivers. The Atlantic coast, by
comparison, offers some possibility for fishing, and was in the past an
important site for the gleaning of salt. Tributaries of the Zaire and the
Kwilu-Nyari, and rivers such as the Loango flowing directly into the
ocean, offer these same possibilities for fishing, and are in part
navigable.
Between these major rivers mountains rise to a level of 750 meters
in the region of Mboko Nsongo, containing deposits of iron, copper,
and lead in some quantity. These minerals have been instrumental in
the wider region for centuries as material for the tools of cultivation,
hunting, warfare, and trade.
The terrain eastward of the mountains levels off into the vast, sandy
Teke Plateau. Westward toward the ocean, a deep rain forest covers
the Mayombe. The central area is covered with a mixed vegetation,
and rain forests are present where slash and burn agriculture has not
been followed by secondary scrub woods and savanna grassland. The
soil of the region is characteristically tropical. Where forest exists, the
lush tree growth protects the delicate balance of the elements. Where
27
28 T H E PUBLIC SETTING
clearings have been cut out of the forest, mixed cropping produces an
initial rich harvest, but soon the devastation of soil leaching because
of high rainfall reduces the terrain to less productive and barren
grassland or even rock-like laterite.
Until recently hunting was an important subsistence activity in the
entire region. Its important status is attested to by the knowledge of
the net, trap, and stalking techniques found in adjoining regions such as
1
the Cameroons where hunting still survives as a primary activity. Of
edible animals, elephants, buffalo, a variety of antelope, and smaller
animals as well as birds of many kinds constituted the source of food.
This status is still reflected in the symbols of Lemba. Archaeological
research in the region has documented the existence of a variety of
prehistoric lithic industries and of ancient hunting traditions dating
2
back for millennia. Knowledge of metalworking had been introduced
by the middle of the first millennium A.D. Metalworking, according to
north-bank oral traditions, was known by the "Teke" peoples who
gradually gave it to the "Kongo" peoples. Iron arrow tips and knives,
as well as spears, were used widely in intensified hunting. By the
eighteenth century, however, guns were introduced in massive quanti
ties on the coast, replacing the bow and arrow and lance as main
hunting instruments. These guns were rapidly incorporated into the
regional smithing industry, whose artisans soon reworked old barrels,
made bullets of lead, and manufactured gunpowder locally.
Trade had a devastating effect on some types of wildlife. Several
decades after the beginning of ivory trade in the seventeenth
century—before the gun—delivery of ivory to the Loango coast
declined as a result of an annual elephant kill of between 3000 and
3
4000 animals, a rate which nearly depleted the herds. Not until the
late nineteenth century, following the end of the international slave
trade, were ivory tusks to become important again. At that time
elephants were again hunted—this time with a gun—nearly to
4
extinction in the region. By the mid-twentieth century only a few
small elephant herds still remained in the Mayombe forest and in
swampy lands along the Zaire River. The gun had a similar effect on
other wildlife, so that less and less did the highly romanticized life of
the hunter, important for the male image, correspond to the actual role
of hunting in the economy.
The tradition of cultivation remained that of the hoe, ax, adze, and
bush knife from the time ironworking was introduced in the region.
Malaysian crops—yams, bananas—entered the region and with
African crops—possibly millet, the palm (Elaeis guineenis), various
cucurbits, and fruit trees (Canarium schweinfurthii, for example)—
History of the Lemba Region 29
constituted the staple foods until the seventeenth century. Portu

guese trade with Brazil beginning even earlier than the seventeenth
century brought manioc, maize, and beans into the diet, exerting a
significant change in cultivation patterns. Manioc, in particular,
introduced a heavier use of fields with the resulting tendency to clear
wide areas of forest. By the late nineteenth century, that is after
several centuries of manioc cultivation, the central and northern
region of the Lower Congo/Zaire had been largely deforested, and
what was left was falling year by year before the ax and hoe. Dupont,
who made these observations as a foot traveler in 1887, also
suggested that the firewood need of the metalworking industry of the
region from Mboko Nsongo to Mindouli had contributed its share to
5
the extensive depletion of the forest. Later, around 1885-1900,
colonial governments began to levy agricultural taxes to support
caravans, porters, and soldiers, a practice that placed further pressure
upon the land. Export possibilities in the late nineteenth century for
palm oil, palm nuts, ground nuts, and rubber gum (from 1869)
introduced cash cropping, which became the major burden upon the
population and the environment. We will return to the nature of trade
at the end of this section.
The foregoing paragraphs have reviewed natural resources avail
able in the Lemba region, and the technology used to exploit rivers,
soil, metal, wildlife, domestic crops, and trees. In a region as
diversified as this, trade is a logical human activity for the redistri
bution of localized resources. For example, salt, gathered at the
ocean, was traded for metal from the highlands. Active trade of this
kind existed long before European merchants arrived in the sixteenth
century with their goods. But as coastal trade intensified, it not only
stimulated this internal trade, it created a unique coastal trade. The
resulting mixed internal and external commercial system had the
effect of thrusting into sharper juxtaposition the value equivalency of
nearly all cultural objects, from food, drink, animals, clothing, metals,
to the human being.
Local markets, tied to the four-day week, fed into the long-distance
trade routes and their markets, key among which were Mpumbu at the
pool, controlled by Tio lords; Manianga on the Zaire; the copper
markets from Mboko Nsongo to Mindouli; San Salvador, the capital
of the Kongo kingdom; and the ports of Loango Bay, Cabinda, and
Malemba and Ambriz on the coast (see figure 2 above). Long
distance caravans of up to 500 persons carried thousands of tons
hundreds of kilometers in this trade.
The extremely complex question of resource value in the Lemba
30 THE PUBLIC SETTING
area has to be limited in this study to the period from approximately

1850 to 1920, which commercially means the transition from full
slavetrading to so-called legitimate trade, followed by the early
colonial era. Within this scope I am primarily concerned with
illuminating the value of the units used in Lemba inaugurals and
traded by Lemba adherents in markets of the region.
An 1885 market list from the mining area of Mboko-Nsongo, in the
heart of Lemba area, offers a glimpse of the "internal" trade of the
6
region. On sale were:
a large antelope, gun powder

cut into pieces copper wire
tripe lead ingots
rats on sticks goat skins
smoked fish raphia cloth
chicken mats
yams baskets
sweet potatoes egg plant
manioc roots corn
manioc bread pepper
peas or beans bananas
cabbage ground nuts
green legumes palm nuts
mushrooms sugar cane
red earth tobacco leaves
salt calabashes of palm wine
In another market at Ngoyo, upstream from Manianga, in 1877, local

7
items (left column) were exchanged for foreign goods (right column).
bananas cloth
pineapples beads
guavas wire (brass?) (mitako)
limes guns
onions powder
fish crockery
casava bread (chikwanga)
groundnuts
palm butter
earthenware pots
baskets
nets
These items fill the major categories of resources in the Lemba

region. Much of the list consists of locally-grown food, Malaysian and
American. These items, perishable and thus not suited for long
distance trade, nevertheless constitute the mainstay of prestations
used in local social transactions. Palm wine is the smallest unit of this
exchange system, being measured either in terms of the "cup"—what
one man can drink—or in terms of the calabash or the demijohn. The
"cup" is an etiquette measure, used to open a transaction or ritual, to
lodge a request, or to close a deal. It is an important sign in prayers and
purification rites. The "chicken" is a larger unit in this system of
exchange, used for interpersonal gifts as rewards or tokens of
appreciation, as well as for sacrifices. Naturally, these uses imply the
consumption of chicken, making it both a nutritional and a trans
actional resource. The "pig" is the largest unit in this social currency.
It is not mentioned in the Mboko Nsongo market list, but from all
evidence including the Lemba inauguration reported from the area,
pigs were common and greatly appreciated. Pigs were reserved for
major rituals in which their exchange constituted a sign of reconcilia
tion or obligation met between clans. The pig was considered (and still
is) the appropriate unit of recompense for human blood shed in a
quarrel.
These three units—the "cup" (mbungu), the "chicken" (n'susu),
and the "pig" (n'gulu)—should be considered abstract values in
regional society, because they are frequently substituted for in
concrete exchanges by other items or by currency of another sort.
Palm wine is preferred in the "cup," but other natural wines, or
European-type liqueurs in bottles, will do, as will cash. In a 1915
Lemba initiation in the Mboko Nsongo region, for example, mention
8
is made of "pig of five baskets of raphia cloth." In local parlance, the
authentic items are identified as "wet" cups, and "blood" chickens or
pigs, while substitution in cloth or money constitutes a "dry" cup,
chicken, or pig. Goats and sheep may sometimes be substituted for
pigs; however, wild game is not entered into this exchange system.
In the Mboko Nsongo market list several categories of items may be
singled out as important in the legitimate trade following the slave
trade, most of which continue to have a local use value. Foodstuffs
include especially groundnuts and palmnuts and oil. Other items are
long-standing trade items, and can be understood in terms of their
historic uses. "Red earth for body paint" was a highly popular
cosmetic but already a major export item in the seventeenth century
for use in European textile dyes. Salt, which probably came from the
coast, was cherished in seasoning and preservation. Remaining items
relate to metalworking (copper wire, lead, gunpowder), and cloth

"money" (raphia, baskets).
Mboko Nsongo constituted a strategic point in the mining region
that extended to Mindouli. Copper, iron, and lead were the main ores
mined, smelted, and sold to blacksmiths, jewelers, and traders. The
copper trade around Mboko Nsongo alone constituted a major heavy
industry. Dupont describes the largest of three mines in 1885,
Songho, as a cut 250 meters long into a hillside, more than 10 meters
deep in places. Such mines were owned by surrounding village
communities and worked by local inhabitants. A special copper
9
market once a week attracted buyers from all directions. Mboko
Nsongo's special "red" copper may have reached as far into the
10
interior as the Alima River upstream on the Congo/Zaire, and
certainly in trade to Loango and Kongo coasts, where shipments as
early as the seventeenth century are reported. For example, two ships
sailed from Loango for Brazil in 1641 carrying 39,613 pounds of
11
copper. A trading agent on the coast estimated that the Loango
coast—that is, its interior—could supply up to forty tons annually of
12
the metal. The most stable use over the centuries of copper from
these mines, however, was in currency and jewelry. The ngiele copper
rodjs described variously as being as thick as a little fmgeTand one
inch long, or half an inch in diameter and three inches long, bent at the
13
ends. The rod constituted one of the region's main currencies until
the 1880's when the European brass wire, the mitako, took its place in
connection with colonial trade. The ngiele resembles .the copper
bracelet used for jewelry and for ritual in several orders of authority,
especially Lemba. Lemba's bracelets, poured into sand or clay
molds in the ateliers of Vili and Yombe smith-jewelers, represented the
artistic climax of this copperwork. Beyond Lemba's> borders, some
lords of Tio wore such Vili-made copper bracelets with reliefs on
14
them. Even after imported European mitako brass money had
replaced the copper ngiele as regional currency, the bracelets
continued to be manufactured for Lemba initiation. Production of
copper still continued in 1885 when Dupont visited Mboko Nsongo.
Metals such as iron and lead were used for other specific purposes:
iron for hoes, axes and adzes, knives and gun mechanisms; lead,
sometimes for bracelets, but usually for gunshot.
The value of these metals, in trade or commercial power, varied
over the centuries. I have found no exchange equivalents for the
Manianga copper ngiele, although Vansina reports an exchange of
one hundred ngiele (no doubt the smaller type) for one ibuunu of cloth
among the Tio, and one hundred nzi shells for one ngieleP
The mitako introduced by European traders had by 1880 become

the colonial currency in the Mpumbu region at the east of the Lemba
area. However, this unit, like the ngiele, soon experienced an
inflationary spiral, reducing its value. On July 1, 1910, the Belgian
colony introduced the Congo franc (5 francs = 1 mpata) as the
currency of obligatory taxes; it also appeared in Lemba.
Cloth currencies seem over the centuries of coastal trade to have
held a more stable valuation than metals. Raphia cloth, made from
two types of palms, was a major exchange item in the sixteenth and
seventeenth centuries when coastal trade began. Its smallest unit was
the libongo (mbongo) or the makute, about the size of a large
handkerchief, made on local looms. These raphia pieces were sewn
together into a short loincloth, constituting what some coastal traders
16
viewed as the "fathom"—one and one-half to two yards. Raphia
cloth continued to be woven in the twentieth century, and as Lemba
initiation texts demonstrate, it continued to be used longest as
ceremonial currency. However, very early on in the coastal trade
European and Indian cloth came to be highly appreciated, both for its
practical and stylish purposes in apparel and as a medium of
exchange; in effect a major exchange item which, like raphia, tended
not to lose its value as quickly as metal currencies. One type of cloth
might replace another, but throughout the centuries units of cloth
appeared in exchange for ivory, metal, slaves, guns, camwood, and
within African society, for ritual purposes. Coastal trade established
cloth as a common denominator of raphia, woven in the interior and
used in internal transactions, and of the European unit known,
depending on the language as the pièce, pégà, long, and cortado: a
cloth approximately six yards in length—the same as the piece of
cloth sold in the Kinshasa market in 1977!
Before correlating the various resources, currencies, and cloth units
so as to glean an estimate ofLemba's worth in these terms during the
period from 1850 to 1920, we must consider the impact of slavery on
the local society and economy. Internal slavery no doubt existed in
the seventeenth century, but in terms of our twentieth-century
understanding of the phenomenon, the internal slave status was both
more humane, and more varied, than the international slave status.
Bond slaves, debt slaves, war prisoners, hereditary but well-placed
slaves, all had their place in society. Jural slave status did not always
necessarily mean de facto low status. In Lemba we see how slaves
often belonged to the ruling elite. None of this, however, makes the
coastal slave trade less despicable.
International trade was by 1600 exporting copper, ivory, red
camwood, and raphia cloth. With the diminution of ivory, and the
increase in demand for slaves on American plantations, Loango coast
rulers and trading officers—the mafouk—began by 1650 to supply
slaves to their Dutch, Portuguese and English trading partners.
Traffic that began as a trickle had by 1680 reached 4000 persons per
1 7
year from Ngoy o, C abinda, and Loango. B y l 7 5 0 i n C abinda alone,
5000 to 6000 slaves were being exported annually; by the 1780's the
three ports of Malemba, Cabinda, and Loango Bay were processing
18
15,000 slaves annually. At the beginning of the eighteenth century a
slave was worth from ten to fifteen pieces of cloth or 350 makoute in
19
raphia. An increasingly favored item traded for slaves was the gun,
obsolete castoffs from European arsenals. During the increase of the
Congo coast slave trade in the early eighteenth century, a slave could
be had for about eleven to twelve guns, each worth around thirty
20
makoute, or half a piece of cloth. At its peak in the mid-to-late
eighteenth century, the slave trade on the north-bank ports annually
imported up to 50,000 guns and an assortment of other items against
21
roughly 15,000 slaves.
The volume of all trade declined during the late eighteenth century
because of growing opposition to the slave trade, increasing police
action on the high seas against it, and internally because of a growing
scarcity of slaves and an attendant rise in the price. This increase has
been calculated in terms of the pièce of cloth per slave for C abinda
22
and Malemba.
1702-13 10 to 15 pieces
1750-60 20 to 30 pieces
1770 30 pieces
1770's 4 0 + pieces
1780's 54 pieces + 10 pieces brokerage fee
The volume of the slave trade continued to decline in the nineteenth
century and had ended completely by 1870. "Legitimate" trade now
took its place, with the internal suppliers providing ivory once more,
gum copal, palm oil, nuts, groundnuts, pepper, malachite, and baobab
bark, in exchange for tobacco, rum and other liquors, cloth, gun
powder, guns, and assorted manufactured goods.
Lemba's role in the trade will be discussed more fully in the next
section of this chapter. Here the commercial value of certain aspects
of Lemba may be situated in respect to the foregoing review of trade
and exchange of resources. An argument has sometimes been made
for the specificity of domains of exchange in traditional African
economies; a given commodity or resource is said to be specific to

23
certain types of social and commercial transactions. Such a
characteristic is minimal if not altogether missing in the Lemba
economy; wherever exchange is recorded, equivalents between items
are also given. We have seen that in the late-nineteenth-century
marketplaces of Mboko Nsongo and Ngoyo near Manianga, both
subsistence and trade items were on hand. The same holds true for the
exchanges within Lemba, ostensively for "ritual" purposes. The
mechanism of equivalence in all cases is the "wet wine/dry wine,"
"blood chicken/dry chicken" comparison, permitting a specific
domain of transaction defined functionally by a given item—chicken
or wine—to substitute for another item—cash.
The question has also been raised as to why, given this equivalency
mechanism, no single currency emerged. The answer may be that any
single currency is vulnerable to inflation, as demonstrated by the
copper ngiele in long-distance trade, and in the brass wire mitako in
the early colonial period. Furthermore, a given abstract and general
currency tends to be only as stable as the political order of which it is
ultimately a value symbol. There is greater security in the simul
taneous use of several currencies or value symbols deriving from
objects which meet widespread human needs, as do cloth, wine,
chickens, and pigs. Thus, Lemba transactions, like those in the
history of much of the trade, tend to occur in combinations (bundles)
of items.
For example, a Lemba inaugural in the Mboko-Nsongo/Kingoyi/
"Bakamba" region around 1910(analyzed at length in Chapter 4) cost
the neophyte priest and his clan twenty-five pigs, numerous chickens,
palm wine, and manioc loaves. Since pigs are hardly mentioned in the
international trade literature—for reasons having to do with the
transportation of pigs—it is necessary to convert this cost to items
already mentioned. The author of the report has given "equiva
lencies" of the payment of dues: " a pig of 5 ntete baskets, raphia
cloth"; elsewhere in payment of his defilement, the priest collects "10
pigs or 1 person." The Lemba inauguration thus cost approximately
twenty-five pigs, plus wine and chickens, and so forth; or 125 ntete
baskets raphia at one pig/five baskets; or two and one-half persons, at
ten pigs/one person. One person (slave) thus equalled fifty ntete
baskets raphia in the Mboko Nsongo region at the turn of the century.
A report on Lemba from Yaa to the north gives as two persons
the fee for the profanation of the Lemba priest (Text 2.23 below), a
figure comparable to that in Mboko Nsongo. Lemba's worth may be
compared to yet other types of transactions at the turn of the century.

At the Pool, where Vansina has extensively reported on equiva
lencies, a person varied in value from 200 mitako to upward of 1200,
averaging around 500 to 600 mitako}* The only items which
exceeded slaves in price were boats, at around 1000 mitako, and huge
quantities of ivory (say 2,500 pounds for 12,500 mitako, thus five
mitako/pound). Interns of these equivalencies, the Lemba inaugural
would have cost anywhere from 1000 to 2000 mitako, at 1885 rates
of exchange; that would be the price of a big boat, or 200 to 300
pounds of ivory.
Of course by 1910 the mitako had totally disappeared having been
replaced by Belgian and French currency. Thereafter Lemba inau
gurals begin to be paid in this currency. One Lemba inaugural in
eastern Mayombe from the first decade of this century cost the
neophyte and his supporters wine, chickens, manioc, and other
banquet items, as well as two mpidi baskets raphia, two nkwala mats,
and 565 francs, the equivalent of several years trading or wage labor.
The bulk of the payment had here been converted to the new currency,
although the inaugurated office was a continuation of the old.
However measured, Lemba may be seen as a significant institution
in the region from the Pool to the ocean. I have tried to relate this
significance quantitatively to Lemba's inaugural fees, where such
data are given. This measure of value may shift from subsistence
resources such as chickens, wine, pigs, goats, etc., to labor power in
slaves, to clothing and currency in raphia cloth and colonial "money."
These quantitative measures do not however indicate just how
1
Lemba 's qualitatively defined rights worked. I hope to show in
subsequent parts of this book that Lemba arose with the copper and
ivory "boom" of the seventeenth century, and remained a structuring
institution for the era of heavy coastal trade that was to continue for
several centuries. To the east this seventeenth-century development
gave rise to the Tio order of lords, symbolized by the nkobi shrine
basket, and in most places heavy copper bracelets worn by the lords
and priests. Lemba priests and Tio lords converted their prowess in
trade into social influence, becoming a mercantile elite which con
trolled—"calmed"—markets and trade and organized caravans.
Social Structure
Three centuries of trade with the coast established caravan routes

across the north-bank territory; pulled out of their home communities
hundreds of thousands of slaves and drove them over these routes to

waiting ships; created floating "broker" colonies, tariff walls, inva
sions, and wars. Yet most ethnologies of the twentieth century
describe the region's people and their organization in terms of
territorially discrete tribes bounded by firm borders and identified
25
with single names.
To gain an historically accurate picture of the movements of people
in the past 300 years in this region, it must be assumed that at various
times the relatively quick wealth that was to be found in trade and
resource exploitation unleashed great forces for change; the seven
teenth-century "boom" in Manianga copper and inland ivory brought
its "rushers" and "brokers," left its victims, and generally turned
things topsy-turvy, and the eighteenth-century slave trade and the
nineteenth-century legitimate trade on the coast had similar effects.
The ethnic maps mentioned above tell us very little about wandering
bands who came northward across the Zaire possibly toward the
copper region, occupied the mines for a time, and settled there. Nor do
such maps tell the story of Vili traders who came eastward from the
coast, also seeking to capitalize on mining and trading and eventually
settling in commercial enclaves up and down the coast and throughout
the Lower Zaire where trade was the foremost public activity. Not
only the Vili were thus involved in trade. Traders and caravans moved
regularly across the entire area. At Manianga market, persons were
noted in the 1880's from Ngoyo and Kakongo on the coast, Ntombo
Mataka, Ngombi, Ilembi, Kingoma, Kilanga, Kinzore, Suki, Nguru;
Mebelo, Zinga, and Nzabi upriver,all coming into contact with those
from Ndunga, Mbu, Bakongo, and Bassesse (a Tio group specialized
26
in making raphia cloth). These part-time traders could easily
become full-time traders, as had the Pombieros at the Pool, or they
could control trade as did the Yombe chiefs. All are to be compared to
the Hausa, Lebanese, and Portuguese colonies of West and Equa
torial Africa, the Indians of East Africa, or the Chinese of Indonesia.
In all cases these ethnically identified groups acquired an identity
linked with their trading profession.
An inverse process occurred too, in which populations of diverse
origin "melted" into a single new identity defined by a locale or
territory. By the nineteenth century the coastal" Vili," for example,
included slave remnants from Teke, Yombe, Kongo, Songe,
Mondongo, Bayangela, Babongo, Bayansi, Mboma, Sundi, Mbamba,
27
and Mbete. Also, many south-bank populations fled en masse
north-ward from 1885 on, to escape colonial labor caravan
28
recruitment.
The Sundi invasion of the north bank of the Zaire from the
seventeenth century on, and the way it has been recorded, is the most
annoying example of confusion in the literature. Ethnographers such
as Laman (edited by Lagercrantz), Soret, and Murdock have iden
tified the central-north-bank " Sundi" and have shown spearheads of
Sundi occupation extending northeastward into Teke country, north
west into Kamba and Kunyi country, and westward into Yombe
29
country. Ethnographers, like some Sundi informants, suggest that
the regions were "uninhabited" in the seventeenth century and that
the Sundi moved in without resistance to create their homogeneous
culture. Closer probing and circumstantial evidence indicate the
contrary. "Sundi" or "Kongo" communities often subjugated
"Teke" mining camps, obliging the smiths and craftsmen to work
under their hegemony. AtNsundi-Masiki(Lutete,BaKongo) south of
Mindouli, these Nsundi immigrants moved alongside other groups
such as the Kimbanga with extremely deep local genealogies and old
cemeteries, and an affinity to Teke clans (Kimbanga = Imbaaw). The
Sundi thus situated themselves among pre-existing populations,
intermarried with them, and declared their Kongo-oriented hege
mony over them.
The most serious difficulty in the studies positing a homogeneous
ethnic group in Lower Zaire is their distortion of the organizational
principles of the north-bank societies. Scholars who identify an
indigenous equivalent for "tribe" or "ethnie" usually use the term
mvila (also luvila, pi. tuvila). Deleval, for example, does this, but
then wonders why so many people in the north" Yombe" are ignorant
30
of their "tribe." This query suggests that he may have been dealing
with slaves who have no legal genealogical identity, and that like
many other scholars he misunderstood the character of the mvila.
The mvila—given such names as Nsundi, Manianga, Bwende,
Kuimba, Yombe, etc.—should be defined as an exogamous,
matrilineal descent category traced back to a putative common
ancestor, with emphasis on its categorial nature. The mvila is not
primarily an organized group; it is rather an element in a social
worldview within which nine or twelve exogamous and therefore
intermarrying tuvila are posited to have existed "in the beginning,"
and from which various bifurcations and branches may be traced.
The list of particular names included in this set varies from version to
version and from region to region; along the Zaire it tends to link with
a royal, Kongo origin legend (tuuka Kongo dia Ntootila). In one
area in the Mayombe it has to do with the westward movement of
31
Manianga refugees in a great famine. The persistence of these
origin legends which integrate numerous tuvila into one scheme, and
the dejure exogamic principle defining a single mvila, suggest quite
clearly that this unit cannot be a homogeneous territorial "tribe" as
the culture-area ethnographers would have it. It is a matrilineal
descent category, a clan.
The dikanda (pi. makanda) by contrast is the local organization
which carries the same name. For example, one may be a member of
the Nsundi clan because one's mother was Nsundi; one lives in or is a
member of a Nsundi dikanda. These "local" clan sections may
ordinarily be found in three closely proximate settlements, each
inhabited also with other clan sections of other extended clans. While
the extended clan is the category of origin, and is exogamic only in
principle, the dikanda bearing the same name is the organized
exogamic group. The internal communities of exogamic and cor
porate local clan sections (dikanda), led by either a single head
(mfumu nsi) or a type of committee executive, help each other in
defense of their common land, in assembling alliance prestations, and
in warfare where they are prohibited from killing one another
"because they are of the same blood." These local clan sections are
thus genealogically defined through memorized records used in time
of need such as land defense, identification of a person's rights, and a
variety of other situations.
Because these local clan sections are genealogically chartered,
incorporated, and land-based, in time they experience natural seg
mentary growth. Junior, senior, and middle "houses" (nzo) emerge,
each expecting its parcel of land in the local clan estate and its place
"around the fire" in the men's house, that is, expecting to participate
in collective clan affairs. Under certain circumstances the "house"
32
may become the effective exogamous unit.
Within the "house" are found lineages (mwelo-nzo, "door of the
house"). These are residential clusters in settlements. They are also
the effective familial units within which decisions regarding personal
life and production are made. Rey has emphasized the former
strategic importance of this unit among the Kunyi in organizing
slavery and has devised the term "lineage mode of production" to
33
characterize it. It is the effective unit through which the "natural"
matrilineal unit augments its productive and reproductive potential.
As MacGaffey has pointed out, a woman's reproductive capacity
34
cannot be transferred by rules of matrilineality alone; there must be
some form of nonmatrilineal recruitment such as slavery or clientage
to augment the unit, to form a new estate out of the old one over time,
35
to use Gluckman's phrase.
The major alternative principle of recruitment is, of course,

patrifiliality (kise, from se, father), which in native thought contrasts
to matrilineality (kingudi, from mother), the normative principle for
the formation of the clan, local section, house, and lineage. Most
persons can identify their matrilineage and clan, knowing who their
mothers are. However, patrifilial identity is the distinct mark of a
freeman (mfumu dikanda), one whose relations to his "four corners
of character" (ndambu ziiya za muntu) are legitimated by his own,
his father's (his kise), his mother's father's, and his father's father's
lines.
Patrifiliality and matrilineality are rooted in the ideology of
descent which, while varying some across the territory being sur
veyed, generally regards matrilineality (kingudi) as a type of con
tainer for the individual, conveying membership in a group; whereas
patrifiliality is thought of as the spiritual nurturance or identity,
strengthened in the fetal stage by the semen of the father, which turns
into blood creating the personality. Among the Vili and Yombe this
paternal substance has been called xina, a name and a prohi-
hibition; inland it does not bear this name, but its contents are identical.
Normative naming practices hyphenate a man's name to his father's.
Filiation is thus, strictly speaking, bilateral in that a person traces his
links if possible to all ascending groups, both patrifilial and matri
lineal. However, descent is best characterized as double unilineal
according to which principle complementary rights and links come
togther into the individual. Local and general clan membership are
always gained consanguineally, from the mother. Spiritual identity,
personality, and perhaps the name are gained agnaticaliy from the
father.
The leading premise which flows from the foregoing beliefs is that
an individual, indeed human society itself, must contain both
principles, complementarily, to be complete. This theme is repeated
at many levels of organization such as kinship terminology; it is the
basis ofjudicial decisions on incest and marriage; it is present in much
ritual, such as the Lemba inaugural; and it is articulated in legends.
Kinship terminology distinguishes the two ideas in the first
ascending generation of one's immediate parentage by calling one's
father and his classificatory brothers "se" (pi. mase), or "tata," and
calling the father's uterine and classificatory sisters "female fathers"
(se dia nkento). Conversely, mother and her classificatory sisters are
termed "mother" (ngudi), and her brothers, uterine and classifi
catory, "male mothers" (ngudi nkazi). Thus the male/female,
patrifilial/matrilineal dichotomy is incorporated into the immediate

organization of kin. The alternative ascending generation is, however,
again labelled uniformly as "nkaka" uterine and classificatory
grandparent on both sides. One's own generation terms are also
lumped together with reference to siblings, both uterine and classifi
catory brothers ("mpangi") and sisters ("nkazi" or "busi").
The first generation descending, is, like the first ascending,
distinguished according to one's own embodiment of the matrilineal
and patrifilial principles. A male ego speaks of the children within his
own local clan or lineage as "bana bankazi" reckoned by the female
principle; a female ego calls her own children "bana" but generalizes
within her own clan to speak of "bana bankazi." By contrast, a
person speaking of the first descending generation of children born to a
local clan's or lineage's male members would call them "patrifilial
children" (bana bambuta), children born to the clan or lineage,
applying the complementary patrifilial principle (kise or kimbuta).
The role of the patrifilial children (bana bambuta) needs to be
emphasized in connection with the principle of complementarity and
of primary-group formation. In a given community, they are natural
allies in that, having grown up in the midst of their fathers, they share
common spiritual paternal substance (kise). They act as priests and
arbiters on behalf of their fathers, conducting burials and mediating all
contacts between the living and the ancestors of their fathers'
matrilineage, as well as intervening in time of crisis—for example,
segmentation rituals to guarantee peace. Finally, they act as political
supporters and priests in creating and inaugurating consecrated
offices of authority. Their role is crucial in the Lemba rites.
The ability of the local clan thus to retain its children, either as allies
scattered throughout adjacent communities, or as continuing resi
dents in their birthplace, depends on prowess in providing them with a
livelihood in land, trade, or other form of support and opportunity.
Normally, a free male child (mwana) goes to live with his mother's
brothers at puberty; a girl (also mwana) when she marries moves to
reside with her husband. Should the husband die, or they divorce, she
would be free, with her children, to move to her clan home with her
brothers and maternal uncles. Hereditary slaves or others lacking a
means of livelihood in their home clan could prevail on their fathers to
offer them a temporary opportunity; the fathers, in turn, could attract
their children with such generosity in the hope of creating a following.
Such people were often called "people wealth" (bantu mbongo) or
"children of the village" (bana bavata). In the precolonial period a
42 THE P U B L I C S E T T I N G
large clientele could be assembled in short order by the establishment

of a landed estate, proper marriage policies, and the retention of both
one's clan children (bankazi) and one's patrifilial children (bam
buta), thereby making the village or town the dominant political unit
(mbanza). There is some evidence that Lemba concentrated on
strategic alliances between clans and their patrifilial children.
Marriage within the local clan is regarded as incestuous, both for
spiritual and practical political reasons. However, marriage could be
used effectively to build up the clan's following and to keep its own
people together by retaining slaves and dependent clans, and by
marrying and reproducing extensively with these local groups. Power
was proportional to the number of people that could be organized into
such a local unit—the unit of production—or into a cluster of several
under one head. But the major drawback of such an approach to
organization and influence was that it isolated the community of
intermarrying clans and lineages, rendering it vulnerable to en
croachments from its neighbors, and cutting it off from trade and
exchange partners. In other words, although this was and is an
attractive possibility for the short-run acquisition of influence and
power, it is beset with dangers.
The solution to this dilemma was marriage with influential clans
and lineages in neighboring settlements, thereby creating alliances
which could be used to maintain peace and commerce. A survey of the
north-bank region reveals two basic approaches to political organi
zation through marriage: the endogenous approach, in which one
marries within the local polity and settlement; and the exogenous
approach, in which one marries between polities and settlements. The
former tended to be used as a strategy to centralize power, wealth, and
offices of authority; the latter, as a strategy to create regionally
interlocking networks. In the former matrilateral cross-cousin mar
riage or a classificatory equivalent prevailed, in the latter, patrilateral
cross-cousin marriage or its classificatory equivalent. Both could be
found together in specific communities, with the one or the other
receiving conscious preference.
The significance of this differential structuring of marriage types
has been worked out at great length in the social-anthropological
36
literature. The salient points for our purposes are the following. If
the matrilateral cross-cousin marriage form is not carried out as a so-
called "circulating connubium," it becomes an asymmetrical rela
tionship between unequal groups in which the offspring's position
may be controlled. In the region with which we are concerned, a
lineage or clan that can exercise influence over its female children and
get them to marry its women's sons succeeds in short-circuiting the
bride payment normally made to fathers and benefits politically as
"fathers" or patrons of the offspring. This is above all true if such
daughters are members of client groups, residing with and bound to
their fathers' group.
In the patrilateral cross-cousin marriage, by contrast, women-
giving over several generations becomes a reciprocal act between two
exogamous groups which, because the bride's father is not also the
mother's brother of the groom, tends to be a significant exchange of
wealth each time an alliance is extended. Furthermore, this exchange,
because reciprocal, enhances the status of the wife-giver at each
juncture in the relationship, since goods are received in compensation
for or as a gauge against the services and well-being of the bride. In
practice this type of marriage relationship is maintained between
landed free clan sections, or between lords and their dependents who,
although in a status of jural subservience, in fact command sufficient
people and wealth that they receive the recognition of status equals.
Alliance ideology recognized the integral contributions of the
patrilateral cross-cousin marriage to the high ideal of "blood reci
procity" (mvutudulu a menga), the return of descent substance to its
point of origin in a marriage transaction. The son who married his
father's sister's daughter—or a classificatory equivalent—did this. In
the central Lemba region such a person was pointedly named
Masamba, one who crosses over, returns, or clears the way (sambila,
samba) between two groups.
Kinship terminology in the central Lemba region reflects the
preferred status of such a return-blood marriage. Same-generational
patrilateral kinsmen are frequently called "grandparents" (nkaka),
whereas their matrilateral counterparts are termed "grandchildren"
(ntekolo). One's "fathers" (mase) are of a higher status: to marry into
their group or to receive a wife from them is to marry "up."
Furthermore, one is thereby balancing the credit sheet of the alliance
relationship or even establishing a mutually enhancing exchange
relationship which will stand well in the subsequent generatioa
This discussion of the structural principles of north-bank social
organization—probably of all local societies of the Lemba region-
has established a basis for demonstrating how explicit policies and
historical forces created alternative patterns of public order.
Throughout the north-bank region there were localized estates or
domains known as nsi, or tzi (pi. zitsi) whose chief, mfumu nsi was
y
a lineal descendent (or proxy) of the dominant clan section of the

locality. The nsi was land based, andrightto its estate was claimed by
asserting prior arrival and attachment to the lineal group possessing
the area's main cemetery grove (makulu). Chiefship of the domain
was consecrated to ancestor spirits buried in the grove, or to BiKinda,
Simbi, earth, water, or other fixed spirits at that location. The domain
frequently bore the same name as the chief. Where such domain chiefs
consolidated their power, their identity could take the place of
37
inhabitants' clan identities. Under such circumstances the clan
(luvila) of the chief of the domain (nsi) could, drawing on the power of
the cult to the local spirit, gain prominence throughout an area and
take on the appearance of a homogeneous cultural zone which
ethnographers have mistakenly called "tribes"—Sundi, Mbenza,
Mpudi a Zinga, for example.
The emergence of such a dominant clan domain created another
problem, with its own unique solution. If a ruling clan became all
pervasive in an area, absorbing lineages and children to itself, with
whom did its members marry, given the clan endogamy prohibition?
One solution (the endogenous approach) has already been illustrated,
namely the mating of nobles or aristocrats with their clients,
dependents, and slaves, that is their patrifilial children. The difficulty
in this approach is that it isolates the endogamous group in an instable
asymmetric marriage system. Another solution often chosen was for
the dominant clan to segment into two separate entities through
sacrifice of a pig, a ritual act intended to "sever" the unified blood.
The ease with which a symbolic solution to a structural problem was
found indicates the ideological nature of descent blood as an
expression of corporate strength. Nsundi enclaves on the north bank
are known to have taken this solution to the above dilemma.
In addition to these solutions of relatively greater centralization and
relatively lesser centralization, there is a third alternative which
became theLemba organizational model. WhereLemba achieved its
greatest structural consistency, on the periphery of and between
kingdoms and aristocratic domains (zinsi), symmetrical marriage
relationships were maintained between members of modest clan
estates, as well as with their influential dependents, clients, and
slaves. The marriage form sanctioned by Lemba was the symmetrical
exchange between relatively powerful local clans. Only a "reciprocal
blood" marriage of the type that would most readily have been the
patrifilial cross-cousin marriage would have fit this expectation. Such
a marriage stimulated exchange of moveable goods—an expected
goal in a mercantile society; it wove a fabric of peaceful ties between

communities, thus overcoming isolationistic tendencies; and it cre
ated a nonterritorial estate which emphasized movement, trade,
influence, knowledge, and the maintenance of horizontal relation
ships across the dividing lines of local societies. It is fitting that the
name of such an estate should have been "peace," Lemba, and the
self-image of its rituals, therapy.
Political Systems of the Lemba Region

Thus far in this chapter I have sketched the natural resources of the
Lemba region and the way subsistence exploitation as well as trade in
these resources tied the various sectors of production into a complex
economy. The great influx of trade in the seventeenth century brought
this regional economy into integral involvement with the international
economy of the maritime nations. I then sketched the social organiza
tion of the region, noting that this trade—especially the slave trade—
introduced major upheavals, migrations, and power shifts into the
populace, so that to speak of organization one must note movements
of people. The common social structure of the region provided the
terms of localized variations, particularly in marriage practices and
offices of authority.
In this section I shall elaborate more fully on the distinct
institutional forms that emerged within the common social structure:
first, the coastal kingdoms of Loango, KaKongo, Ngoyo, Vungu, and
the Yombe chiefdoms; second, the Teke federation in the east; third,
the Nsundi enclaves which play a prominent role in this region's
history; finally, I shall give fullest attention to "life between the
kingdoms," where Lemba took on its most sophisticated form as
"government medicine."
THE COASTAL KINGDOMS
The literature offers several alternative origins of the coastal king

doms of Loango, KaKongo, Ngoyo, and Vungu: either origin from a
common northern parent kingdom, Ngunu, or secession from the
Kongo kingdom, or separation from one another. Despite this
ambiguity over origins, the literature is clear that these kingdoms
predate the coastal European trade and that they have similar
38
historical structure.
The internal structure of these kingdoms rested on the local chief of

a given area and the resources and spirits of the earth. The chiefs paid
tribute to the intermediary level of authority, usually the provincial
governors, who paid it to the ruler himself, the MaLoango, the
MaKongo, and the MaNgoyo. Each king maintained control over
coastal trade within his kingdom, organized through a minister of
trade (mafouk), for each of the three major ports, Loango Bay in
Loango, Malemba in KaKongo, and Cabinda in Ngoyo. The interna
tional trade system was therefore controlled through the same
centralizing structure as the internal tribute system. In Loango, where
the provincial centralizing structure is best described and most likely
to have been significant because of the local seat of Loango as a
kingdom, the provincial governorship was an integral part of the
maintenance of the overall structure. In the seventeenth century,
Loango was ruled by a matrilineally related series of brothers or
cousins, princes of the Kondi clan, who acceded to the throne after
having successively held the provincial governorships. These prov
inces, Cane (Caye), Bukkameale, Dingy, and Kesok, were ruled by
the governors called respectively Manicaye, Manibock, Manisalag,
and Manicabango. On the death of the MaLoango, office holders of
the provinces would shift to another province, Manicaye becoming
the new MaLoango, Manibock becoming Manicaye, Manisalag
becoming Manibock, Manicabango becoming Manisalag, and a new
39
Manicabango being appointed from the ruling dynasty's princes.
Whether KaKongo and Ngoyo developed this system to the same
40
extent as Loango is not clear. In Kongo, where the candidate to the
throne was also known as ManiKai, the leading dynasty had broken
into two factions by the seventeenth century, suggesting that succes
41
sion was as often from father to son as from brother to brother. By
the eighteenth century—thus by the time of accelerated coastal
trade—this condition, which Europeans found detracted from royal
power, had appeared also in Loango. Patrilateral succession sug
gested that power was shared by two intermarrying dynastic lines
within an enduring state structure.
In Loango, as in the other coastal kingdoms, the king was spoken of
as the "supreme lawgiver and protector of the people, the ultimate
sanction of all administrative activities on a local and national level"
who held a semi-divine status—a unique relationship with Nzambi
the Creator—holding the title of Ntotela, supreme ruler, Ntinu,
42
supreme judge. Rituals of first fruits, rainmaking, and eating and
History of the Lemba Region 4 7
drinking in isolation were attributed to him. The lighting of a central

state flame maintained throughout his rule was part of the MaLoango's
inauguration; it was distributed to the provinces, offering a source of
fire to the people's hearths. Despite its prominence over Ngoyo and
KaKongo and the divine trappings of its king, Loango was dependent
on its southern neighbor for the ritual observance of the oracle to
Bunzi, earth deity, whose main shrine was located in Ngoyo territory.
In the late nineteenth century the relationship between the three rulers
was seen in Loango as that of husband Loango (nunni) to wife
KaKongo (mokassi), with Ngoyo as priest (itomma), terms that crop
43
up to describe many ceremonial relationships including Lemba. In
Ngoyo, on the other hand, the three kingdoms of the coast were seen
as having equal status in Bunzi's pantheon, with the first rulers being
brothers born to the union of Bunzi, a female deity, with MeMbinda,
44
the first human. In all instances there existed a complementarity
between cult and capital, priest and king, similar to that which existed
in chiefdoms of the region (the nsi) and kingdoms, all of whom either
participated in the shrine of Bunzi or some equivalent earth or water or
place deity. The king's power, as reflected by these several cultic
arrangements of Nansi and the earth shrine Bunzi, was shared with a
female and priestly element. In the structure of the kingdoms the
queen mother and princesses of the royal clan held prominent,
legislative authority, helping represent the interests of the royal clan
and themselves being the ground in which were born future leaders. In
Loango this female figure was called the Makunda, either queen
mother or sister to the MaLoango who protected interests of women
throughout the kingdom and sometimes took over ruling clan interests
45
during the interregnum.
In addition to the female dimension of power embodied in the
Makunda role and the Bunzi shrine, alternative power was held by the
cadaver priest (Nganga Mvumbi), an interregnum figure who took
over the kingdom and ruled until the new king could be inaugurated. In
Loango, where this system is best described, the Nganga Mvumbi
became relatively permanent in the eighteenth and nineteenth cen
turies because of the inability of the royal clan to consolidate its power
and to stage the very costly inauguration of a king, part of which was
the lavish burial of the incumbent. By the end of the nineteenth
century in Loango, cadaver priests were already succeeding each
other, and the interregnum of MaLoango had gone on for decades!
Corresponding to this disintegration of kingship, trade control along
the coast became dispersed among many brokers (mafouk) whose

brokerage role had spread quite thin, each controlling only a small
section of the total trade.
While the kingdoms on the coast are described as having progres
sively lost their power, falling to forces of disorganization from the
eighteenth century on, the opposite can be said for the chiefdoms of
the Yombe interior. Although these chiefdoms continued to pay token
tribute to coastal kings, their own fortunes had risen through trade
along the caravan routes into the interior. What few accounts exist of
caravans suggest that great wealth was to be made from taxes levied
on caravan organizers. Although eighteenth-century and earlier
sources may not have been aware of the political structure of trade in
the interior, late nineteenth-century accounts describe the existence
of a tariff wall, made of poles and bamboo, between the Guena
(Loeme?) and Kwilu Rivers at the boundary between the coastal Vili
in the kingdom and the forest of Mayombe, allowing passage at only
three points where brokers gathered taxes on all merchandise
46
imported or exported. Veistroffer observed in 1880 that the "king of
47
mayombe" had reduced this to a single passage. Others described
further tariff walls and gates to the east of the Mayombe in the
48
direction of Kunyi territory. German merchant Robert Visser
relates that the Mayombe trade barriers on the Loango were broken
through by force as late as 1885 by the steamer Pollux.***
These Yombe who ruled as either hereditary clans or nineteenth-
century "nouveau riches" created by control of the trade were
legitimated by the same general religious concepts and symbols as the
coastal kingdoms. In addition to the genealogical legends of the
50 51
division of society into eight, nine, or twelve original and many
52
more de facto nsi, earth deity Bunzi was central to these symbols.
Further charms were collected by Yombe chiefs as they were created
53
and became available, such as the nkisi tsi cult. The chief of Tseke
Banza in Mayombe, for example, who had amassed a fortune on the
slave trade, retained the right of investiture over subchiefs in his mini-
kingdom for the price of two slaves, in exchange for all sorts of "fetish
54
powers."
Historians of coastal political entities become very uncertain about
the late eighteenth century; they are prone to complain, as does
Martin, that losses of political coherence occur for reasons that the
literature does not document well: "By 1770, the Mayombe hinter
land was a collection of small independent chieftancies. When and
55
why this happened is not clear from the sources." Closer attention to
goal in a mercantile society; it wove a fabric of peaceful ties between

communities, thus overcoming isolationistic tendencies; and it cre
ated a nonterritorial estate which emphasized movement, trade,
influence, knowledge, and the maintenance of horizontal relation
ships across the dividing lines of local societies. It is fitting that the
name of such an estate should have been "peace," Lemba, and the
self-image of its rituals, therapy.
Political Systems of the Lemba Region

Thus far in this chapter I have sketched the natural resources of the
Lemba region and the way subsistence exploitation as well as trade in
these resources tied the various sectors of production into a complex
economy. The great influx of trade in the seventeenth century brought
this regional economy into integral involvement with the international
economy of the maritime nations. I then sketched the social organiza
tion of the region, noting that this trade—especially the slave trade—
introduced major upheavals, migrations, and power shifts into the
populace, so that to speak of organization one must note movements
of people. The common social structure of the region provided the
terms of localized variations, particularly in marriage practices and
offices of authority.
In this section I shall elaborate more fully on the distinct
institutional forms that emerged within the common social structure:
first, the coastal kingdoms of Loango, KaKongo, Ngoyo, Vungu, and
the Yombe chiefdoms; second, the Teke federation in the east; third,
the Nsundi enclaves which play a prominent role in this region's
history; finally, I shall give fullest attention to "life between the
kingdoms," where Lemba took on its most sophisticated form as
"government medicine."
THE COASTAL KINGDOMS
The literature offers several alternative origins of the coastal king

doms of Loango, KaKongo, Ngoyo, and Vungu: either origin from a
common northern parent kingdom, Ngunu, or secession from the
Kongo kingdom, or separation from one another. Despite this
ambiguity over origins, the literature is clear that these kingdoms
predate the coastal European trade and that they have similar
38
historical structure.
The internal structure of these kingdoms rested on the local chief of

a given area and the resources and spirits of the earth. The chiefs paid
tribute to the intermediary level of authority, usually the provincial
governors, who paid it to the ruler himself, the MaLoango, the
MaKongo, and the MaNgoyo. Each king maintained control over
coastal trade within his kingdom, organized through a minister of
trade (mafouk), for each of the three major ports, Loango Bay in
Loango, Malemba in KaKongo, and Cabinda in Ngoyo. The interna
tional trade system was therefore controlled through the same
centralizing structure as the internal tribute system. In Loango, where
the provincial centralizing structure is best described and most likely
to have been significant because of the local seat of Loango as a
kingdom, the provincial governorship was an integral part of the
maintenance of the overall structure. In the seventeenth century,
Loango was ruled by a matrilineally related series of brothers or
cousins, princes of the Kondi clan, who acceded to the throne after
having successively held the provincial governorships. These prov
inces, Cane (Caye), Bukkameale, Dingy, and Kesok, were ruled by
the governors called respectively Manicaye, Manibock, Manisalag,
and Manicabango. On the death of the MaLoango, office holders of
the provinces would shift to another province, Manicaye becoming
the new MaLoango, Manibock becoming Manicaye, Manisalag
becoming Manibock, Manicabango becoming Manisalag, and a new
39
Manicabango being appointed from the ruling dynasty's princes.
Whether KaKongo and Ngoyo developed this system to the same
40
extent as Loango is not clear. In Kongo, where the candidate to the
throne was also known as ManiKai, the leading dynasty had broken
into two factions by the seventeenth century, suggesting that succes
41
sion was as often from father to son as from brother to brother. By
the eighteenth century—thus by the time of accelerated coastal
trade—this condition, which Europeans found detracted from royal
power, had appeared also in Loango. Patrilateral succession sug
gested that power was shared by two intermarrying dynastic lines
within an enduring state structure.
In Loango, as in the other coastal kingdoms, the king was spoken of
as the "supreme lawgiver and protector of the people, the ultimate
sanction of all administrative activities on a local and national level"
who held a semi-divine status—a unique relationship with Nzambi
the Creator—holding the title of Ntotela, supreme ruler, Ntinu,
42
supreme judge. Rituals of first fruits, rainmaking, and eating and
drinking in isolation were attributed to him. The lighting of a central

state flame maintained throughout his rule was part of the MaLoango's
inauguration; it was distributed to the provinces, offering a source of
fire to the people's hearths. Despite its prominence over Ngoyo and
KaKongo and the divine trappings of its king, Loango was dependent
on its southern neighbor for the ritual observance of the oracle to
Bunzi, earth deity, whose main shrine was located in Ngoyo territory.
In the late nineteenth century the relationship between the three rulers
was seen in Loango as that of husband Loango (nunni) to wife
KaKongo (mokassi), with Ngoyo as priest (itomma), terms that crop
43
up to describe many ceremonial relationships including Lemba. In
Ngoyo, on the other hand, the three kingdoms of the coast were seen
as having equal status in Bunzi's pantheon, with the first rulers being
brothers born to the union of Bunzi, a female deity, with MeMbinda,
44
the first human. In all instances there existed a complementarity
between cult and capital, priest and king, similar to that which existed
in chiefdoms of the region (the nsi) and kingdoms, all of whom either
participated in the shrine of Bunzi or some equivalent earth or water or
place deity. The king's power, as reflected by these several cultic
arrangements of Nansi and the earth shrine Bunzi, was shared with a
female and priestly element. In the structure of the kingdoms the
queen mother and princesses of the royal clan held prominent,
legislative authority, helping represent the interests of the royal clan
and themselves being the ground in which were born future leaders. In
Loango this female figure was called the Makunda, either queen
mother or sister to the MaLoango who protected interests of women
throughout the kingdom and sometimes took over ruling clan interests
45
during the interregnum.
In addition to the female dimension of power embodied in the
Makunda role and the Bunzi shrine, alternative power was held by the
cadaver priest (Nganga Mvumbi), an interregnum figure who took
over the kingdom and ruled until the new king could be inaugurated. In
Loango, where this system is best described, the Nganga Mvumbi
became relatively permanent in the eighteenth and nineteenth cen
turies because of the inability of the royal clan to consolidate its power
and to stage the very costly inauguration of a king, part of which was
the lavish burial of the incumbent. By the end of the nineteenth
century in Loango, cadaver priests were already succeeding each
other, and the interregnum of MaLoango had gone on for decades!
Corresponding to this disintegration of kingship, trade control along
48 T H E P U B L I C SETTING
the coast became dispersed among many brokers (mafouk) whose

brokerage role had spread quite thin, each controlling only a small
section of the total trade.
While the kingdoms on the coast are described as having progres
sively lost their power, falling to forces of disorganization from the
eighteenth century on, the opposite can be said for the chiefdoms of
the Yombe interior. Although these chiefdoms continued to pay token
tribute to coastal kings, their own fortunes had risen through trade
along the caravan routes into the interior. What few accounts exist of
caravans suggest that great wealth was to be made from taxes levied
on caravan organizers. Although eighteenth-century and earlier
sources may not have been aware of the political structure of trade in
the interior, late nineteenth-century accounts describe the existence
of a tariff wall, made of poles and bamboo, between the Guena
(Loeme?) and Kwilu Rivers at the boundary between the coastal Vili
in the kingdom and the forest of Mayombe, allowing passage at only
three points where brokers gathered taxes on all merchandise
46
imported or exported. Veistroffer observed in 1880 that the "king of
47
mayombe" had reduced this to a single passage. Others described
further tariff walls and gates to the east of the Mayombe in the
48
direction of Kunyi territory. German merchant Robert Visser
relates that the Mayombe trade barriers on the Loango were broken
49
through by force as late as 1885 by the steamer Pollux.
These Yombe who ruled as either hereditary clans or nineteenth-
century "nouveau riches" created by control of the trade were
legitimated by the same general religious concepts and symbols as the
coastal kingdoms. In addition to the genealogical legends of the
50 51
division of society into eight, nine, or twelve original and many
52
more de facto nsi, earth deity Bunzi was central to these symbols.
Further charms were collected by Yombe chiefs as they were created
53
and became available, such as the nkisi tsi c u l t The chief of Tseke
Banza in Mayombe, for example, who had amassed a fortune on the
slave trade, retained the right of investiture over subchiefs in his mini-
kingdom for the price of two slaves, in exchange for all sorts of "fetish
54
powers."
Historians of coastal political entities become very uncertain about
the late eighteenth century; they are prone to complain, as does
Martin, that losses of political coherence occur for reasons that the
literature does not document well: "By 1770, the Mayombe hinter
land was a collection of small independent chieftancies. When and
55
why this happened is not clear from the sources." Closer attention to
the social history of the region in terms of control of this trade and the
structuring of society into a network of interconnected marriage and
trading partners gives a fuller picture of the situation. It also allows us
to indicate the role of Lemba as a legitimating symbol and social
institution.
Already by the seventeenth and eighteenth centuries, observers
were suggesting that the coastal kingdoms, with their matrilineal
dynasties, often broke up into contesting factions and, permitting
patrifilial succession, constituted rather fragile state systems by
comparison to European states of the time. The kings married
polygynously, usually with commoner and client women. Through
sheer numbers these women and their children (bana) provided the
rulers with political support and prestige. Noblemen married simi
larly, although on a smaller scale, only sparingly marrying noble
women. These latter, by contrast, married monogamously, usually
with a client or slave male, and sometimes they did not marry at all,
retaining a series of male concubines. Noblewomen were closely
guarded, since they were the mothers of potential rulers, and the noble
male progeny had to be restricted as narrowly as possible to avoid
succession feuds in the royal clan. The queen mother herself, who
could choose her spouse or lover(s), was closely guarded by her
56
brothers and uncles.
This account, referring mainly to Loango, falls well within the
picture of social structure drawn up earlier. But it is a variation well on
the "endogenous" end, in which noble or aristocratic matrilineages
protectively restrict their consanguineal offspring (from their women),
while prodigiously producing "people power" (mbongo-bantu) from
their males.
Hyacinthe de Bologne, a very observant early-eighteenth-century
Capucin, describes in vivid detail the behavior of the princes of
Sonyo, among the Solongo, farther to the south but still within the
region that was becomingZe/nôa area. Father Hyacinthe noted that it
was respectable for the groom to pay a bride price to the bride's family
after she had borne a child. This payment obligated him to her for his
lifetime, with the sanction of both their families. However, the princes
were in the habit of doing something additional which Hyacinthe
regarded as "coutume vraiment païenne!" They would contract a
bride payment with one woman, and then go off and have children by
another, leaving the first union unconsummated. He had difficulty
understanding this bizarre form of marriage—"ce mariage disor
donné"—which occurred primarily among the prominent people of
50 THE P U B L I C S E T T I N G
57
the land, rarely among the commoners, and never among slaves.
What was also bad, in Hyacinthe's view, was that the fifteen or so
"seigneurs" of Sonyo who paraded around as legitimate heirs to the
throne-—when there should have been only one—preferred to marry
in this manner with their close relatives, even their "sisters."
Hyacinthe advised priests to beware of the lords of Sonyo who would
try to trick the church into sanctioning such a marriage not allowed by
ecclesiastical law. For example, D o n Compte Barreta de Sylva
wanted to marry Dame Lucie Barrett (sic), his sister-in-law. He told
the priests that he had paid the bride price for one Marie Nquemque.
But when the time came for the ceremony in church, Dame Lucie was
brought in disguised as Marie Nquemque, whereupon the priest
announced he would excommunicate anyone who tried "marriage
58
fraud," and had the charade stopped. Most of the 400 marriages that
Hyacinthe had performed were among commoners. The nobility and
royalty desired to marry their close kin in the church, but would rarely
consummate these relationships.
The picture that emerges of alliance patterns on the coast in the
early eighteenth century is one in which the landowning freemen,
nobility and royalty, marry close relatives who happen to fall within
the Roman church's restricted list, presumably between cousins.
Hyacinthe is not clear as to the degree of cousin, but the church
forbids first cousins, parallel and cross. These cousin marriages are
those in which the bride payment is used, per contractum, but they
are the marriages which often are not consummated. On the other
hand, these princes frequently have several "concubines" and other
wives, with whom they have many children. Among commoners and
slaves contractual marriages are consummated.
Hyacinthe reserved his most severe judgement, however, for an
ancillary "pagan" rite performed in connection with the noble or royal
marriages, contracted but frequently leftunconsummated. The wife is
obliged to "marry with the devil," and then she enjoys superin
tendence over the other wives of her husband. Only her sons succeed
59
to paternal inheritance. This combination of a marriage between
lineages of high status, for apparently political alliance reasons, with
ritual as well as economic sanctioning, and in which sons of the
woman accede to their father's positions, would describe exactly the
60
set of practices known as Lemba.
In due course the princes of Sonyo, Ngoyo, KaKongo, and Loango
succumbed to the threats and promises of the European missionaries
and began to have children within monogamous unions between noble
and royal clans. The direct consequence of this action was to increase
significantly the number of legitimate heirs to the thrones, thereby
virtually guaranteeing the disintegration of the matrilineal royal
succession process. If Hyacinthe's "marriage with the devil" on the
part of noble and royal women was indeed Lemba, it would have
offered an alternative alliance structure to the narrow royal house
hold, with one queen mother, a king, her brother, and only one or two
contending princes. It would have constituted a sanctioned alliance
between several prominent exogamous clans.
Evidence that we are dealing with Lemba here comes from further
sources. A nearly identical marriage structure is reported among the
late-nineteenth-century and early-twentieth-century Kavati (nsi)
domain of Madula, in the north Yombe, inland from Loango.
Noblewomen rarely marry. It is their custom to keep a series of male
concubines until they tire of such an arrangement. Having borne only
a few children, they live out their later years in their "home" lineage
settlements with their brothers. They take on chiefly roles, and if
especially winsome, make diplomatic calls to vassal chiefs to collect
tribute. Those few noblewomen who marry formally are exchanged in
marriage with their counterparts in neighboring domains and villages
to stabilize the two groups' relationships. Noblemen marry polygy-
nously, mostly with slave women. Commoners and slaves marry, as
they can, with an exchange of the bride price. But the formal noble
61
marriages between domains and villages are usually made in Lemba.
These accounts all suggest that Lemba emerged within the broad
ening power base of coastal populations in the seventeenth and
eighteenth centuries when the coastal kingdoms were experiencing
their initial trade upsurge with European merchants.
In this setting, Lemba was one of the large corporate sacred
medicine cults (min'kisi) of the region. It is helpful to describe this
ritual context briefly. In Loango an excellent historical record of
medicines provides a "barometer" of change from the seventeenth to
the twentieth centuries. The earliest direct reference to Lemba is
Dapper's in the 1660's, where it was seen in the royal court.
Malemba is a moquissie of great significance, worthy to be

seen as sacred by the king, for whose bodily health it serves: for,
as long as he does not become ill, he is surely kept well by it. It
consists of a small four-cornered mat one-and-a-half feet large,
with a string at the top, on which are hung several small
calabashes, cuttlebones, feathers, dry shells, iron bits, bones,
and such things, all colored with tukula-red which represents in

this connection something special. To celebrate it a boy takes a
small drum and strikes it in a distinctive way, accompanied by
the resonance of rattling shell whistles. To this is added spitting
of consecrated kola, and annointment with tukula and con
secrated water out of a small pot with an asperge onto the
moquissie, then onto the body of the nganga, then that of the
king, to the accompaniment of singing or chanting appropriate
to the occasion. Even the nobility who are in attendance receive
for their efforts annointment of their bodies of a few red stripes
(of tukula), receiving thereby the honor of Lemba, which they
then carry away and hang in its place, its small pot, its asperge
62
and satchel.
Other major public medicines noted by Dapper in seventeenth-

century Loango included Tiriko, a shrine located in nearby Boarie,
with four anthropomorphic pillars to hold up the roof. The priest of
Tiriko, accompanied by a boy, daily prayed for the health of the king
and the well-being of the land, as well as the growth of crops, the luck
of merchants and fishermen.
Boessi-Batta was another n'kisi of major importance focused on
bringing into one's homestead objects acquired in long-distance trade,
thus especially pertinent to merchants. It consisted of several parts: a
large lion-skin sack filled with all sorts of shells, iron bits, herbs, tree
bark, feathers, ore, resin, roots, seeds, rags, fishbones, claws, horns,
teeth, hair and nails of albinos and other "unnatural" creatures
(ndondos); to this satchel were added two calabashes covered with
shells (Schnackenhornern, simbos) and topped with a bush of
feathers, decorated with iron hooks, and colored with tukula wood
(red). Atop this calabash a mouth-like orifice had been carved, into
which was poured wine to activate the n 'kisi. The whole set of objects,
satchel and calabashes, was placed atop a table-like construction
outside the door of the house. When they would go out to fetch
merchandise, even at a distance, adherents to Boessi-Batta would
take the moveable part along with them, even though it weighed ten to
twelve pounds. The n 'kisi appears to have paralleled the movement of
articles of trade. On returning with new wares, the priest would draw
lines on his body and incant in arisingtone of voice as he unpacked the
n'kisi ingredients. Presently his eyes would exorbit, and he would
become possessed by the spirit of Tiriko. Drinking a liquid to calm
himself, he would declare the wish of Boessi-Batta, the fee, and
63
measure to be taken or other min'kisi to be used.
Another n'kisi of seventeenth-century Loango was Kikokoo, an

anthropomorphic wooden shrine standing in the seaside village of
Kinga, at the side of a large cemetery common to the region, perhaps
to Loango Bay itself. Kikokoo protected the dead against witches
(doojes [ndokil]) who in their nocturnal craft would drag off the souls
of the dead to slavery and forced labor. Kikokoo also was to assure
the arrival of ships with fish and merchandise.
N'kisi Bomba was celebrated, noted Dapper, in a special feast
associated with the coming out of the Khimba initiates dressed in a
head garb of feathers and skirts of palm raphia, and carrying a red and
white hand rattle. He noted their frenzied drumming, their mock-mad
behavior. This is the final public phase of the Khimba initiation which
according to later observers included adoration of Mbumba Luangu,
the rainbow serpent.
Other min'kisi reported by Dapper included Makongo, consisting
of rattles, drums, small sacks, and red fish hooks; Mimi, & small
house shrine in a banana grove with a throne holding a basket of
objects including a "paternoster" of seashells and a wooden statue of
Father Masako (?); Kossi, a sack of white snail shells, filled with
white clay, used in rites of crawling between one anothers' legs,
eating, washing, donning bracelets and bands of protection against
lightning thunder, and sickness; Kimaje, a pile of potsherds on which
priests deposit old ragged caps and other worn out ritual parapher
nalia and dedicate their new replacements, so as to assure the new
moon and new year's coming, as well as protection on the seas;
Injami, a shrine found in a village near Loango, represented in a huge
statue in a house; Kitauba, a huge wooden gong used in swearing the
oath or sending sickness to another; Bans a, another statue covered
with red powder; Pongo, a "wooden" calabash or container covered
with shells (simbos) and filled with many carved symbols, used in
"black magic"; and Moanzi, a pot partially buried in the ground
between dedicated trees, holding an arrow and a string on which hung
green leaves, and whose adherents wore copper armrings and avoided
64
eating kola.
This seventeenth-century inventory of min'kisi sharpens the im
pression we already have of Loango. The kingdom is still intact, as
repeated reference to the well-being of the king suggests. Many
shrines and medicines combine the well-being of the king with
plentiful harvests, good fishing expeditions, abundant trade goods,
and control of rain, a common attribute of centralized African power.
Lemba is referred to as a"big"«'£/.s7, and in many respects Dapper's
account resembles later accounts, including the ritual elements and
the emphasis on a gentry. This is however the only account of Lemba

that includes a king in the rite.
There is some indication, in the symbolism of the min'kisi Kikokoo
and Boesi-Batta that the kingdom is beset by serious problems
(
resulting from trade, including the slave trade. Kikokoo' s charac
teristics are laced with slave-trade attributes, the theft of souls, their
forced work, and the coming in of trade goods. Boessi-Batta appears
even more sharply to express the concern for protecting the inner
world of the household and homestead from the danger of things
brought from afar. However this sense of protection at the house door
is ambiguously charged with attractiveness of these same goods. The
importance of Boessi-Batta in seventeenth-century Loango is indi
cated by the fact that it was exported with African slaves to Haiti,
where, as shall be seen in Chapter 8, significant symbols of the
contrast of the household with the beyond remain in the Bosu shrine of
voodoo ceremonialism. Objectively, the seventeenth century saw the
control of trade by the kings of coastal kingdoms, through their
brokers, slipping from them into the hands of a mercantile elite. This
elite, on the coast as well as inland, was mostly synonymous with the
Lemba priesthood. The shifting base of power growing from control of
the trade was also reflected in the structure of shrines. Generally,
fixed territorial shrines such as Tiriko, Kikokoo, Mimi, Kimaje,
and Bunzi did not get transported across the ocean in the rites and
lives of slaves, as did some of the "portable" shrines or medicines.
However, the fixed shrines such as Bunzi moved beyond the orbit of
royal families and their priests, and came to be found in various
chiefdoms (zinsi) of the coast and Mayombe, even inland as far as
Kibunzi, in the form of what later observers called "pseudo-2?wnzf"
shrines and cults.
By the mid-nineteenth century, for which better sources are
available on Loango's min 'kisi system, a picture emerges that is quite
different from Dapper's seventeenth-century account. The fate of the
centralized, earth-cult inspired kingdoms on the coast is such that
little mention is made of the king, let alone his well-being. Trade,
adjudication, thief- and witch-finding medicine is widespread, and
there is a serious concern for the fertility of women. The German
65
Loango expedition of the 1870's describes this setting.
Bunzi is considered an important palaver oracle, as well as a
rainmaker, relating to the wind that brings rain. Pechuel-Loesche
however considers this male version a pseudo-Bunzi, an unortho
66
dox variant of the traditional earth shrine in Ngoyo.
Gombiri, a female n'kisi, which protects Loango Bay and locates

witches and murders, has replaced the now extinct Kikokoo, which it
is noted served the Loango king as his major shrine. Mansi, men
tioned by D apper, is a protective shrine on the ocean coast which rose
to great popularity as an oracle in the 1840's and 50's but disappeared
in the 1860's after its main altar was demolished by an unknown
person, animal, or storm. Ngombo, the widespread divination oracle,
67
is used in Loango in the mid-nineteenth century.
These sources list a long series of min'kisi concerned with
adjudicatory and retaliatory functions, no doubt necessitated by the
collapse of formal appeal courts in the kingdom. Tschimpuku, a
woven bag, Mpusu, a four-cornered basket with tightly-fitting lid, is
companion-piece withMalasi, a two-headed hippo-shaped sculpture;
Mboyo-zu-Mambi, a pot lid resting on three legs, and Mpangu, a
wooden block wrapped tightly with a chain—these are all juridical
min'kisi found in Loango during the mid-nineteenth century, concen
trating their songs, techniques, and symbolism upon measures that
were thought to bring clarity and justice to the increasingly tangled
social relations present in the port city. Mboyo was said to have been a
direct successor of the now extinct n 'kisiMaramba, mentioned in the
68
sixteenth century by English sailor Battel. Openly aggressive
techniques of private self-defense and assertion are concentrated in
69
min'kisi described by Bastian. Simbuka can "kill with a quick
strike"; Kunja lames, Kanga ikanga creates a headache in its victim,
causing him to run off wildly into the wilderness; Mabiala
Mandembe (Mapanje), sometimes given human form, drives its
70
victim, especially thieves, mad.
Against these methods of active aggression, there were many
protective min'kisi. Mandombe, embodied in an iron chain, pro
tected its devotee in war and fighting; Imba, a bracelet with a shell
affixed to it, protected its wearer from drawing blood in a fight or
palaver. These latter, as well as the foregoing, are individualized
min'kisi. But they suggest that collective institutions were unable to
satisfy, or render security. Related to the theme of seeking security is
Njambe (Injami of Dapper's seventeenth-century account). Using
the medicine to drive away sleep and achieve ecstasy, an individual
could take white seeds from nganga Njambe and to the accompani
ment of musicians reach possession (sulo umbuiti).
Possibly because of the pervasive insecurity and the history of
slavery on the Congo coast, medicine in this period turned toward the
security of children, that is, fertility. Whereas in the seventeenth
century generalized medicine had included fertility of crops and

women, along with happy trading, now there is a proliferation of
medicines for pregnancy and childbirth. Mpemba (alsopfemba,
umpembe) became very popular and consisted of one or several
treatment centers drawing scores of women seeking advice and
entrance to the rituals of Mpemba. Pechuel-Loesche thinks these
shrines were derived from a famous midwife in Loango who had
71
special techniques or power. The movement gave rise to the
celebrated Mpemba statues of women holding their breasts, and
72
sometimes a child on their laps. This n'kisi was off-limits to men,
and its activities were carried out only in a moon-lit night. Mbinda
supported marriages and healed women's problems; it too was strictly
a woman's affair, carried out in the moonlight, with the women shaven
and naked. Men, hair, tobacco smoke, liquor, and water were taboo to
its adepts. Sasi, a drink administered by a female nganga, was created
for pregnant women in childbirth, and for newly born children. Kulo-
Malonga stopped excessive menstrual bleeding; Bitungu cured
sterility in women; Dembacani and Cuango-Malimbi cured impo
73
tence in men.
A variety of particular "secular" treatments could be added to
these to suggest that Loango medicine in the middle of the nineteenth
century resembled the Occidental concept of medicine more closely
than the foregoing public medicine in certain respects. There were
treatments for stomach ache, both simple and serious. Mpodi the
cupping horn accompanied by skin incisions was used for all kinds of
complaints. Bone setting was done in the case of fractures by the
Lunga doctor and splints of bamboo or some other stiff bandage-like
material were applied to support the break. Infections and swelling
were handled by skin punctures with a knife, upon which was applied
powder of the kola nut and other seeds. There were snake-bite
remedies and many more herbal and manipulative treatments.
The largest category of public medicine evidencing growth, along
with adjudication, aggression and fertility medicine, is that relating to
trade and entrepreneurial undertaking. Mangossu is the lord of trade,
travel, marriage, or any enterprise. The priests of this famous oracle
were well recognized, but had a hard time finding a permanent home
for the shrine since its spirit was a restless wanderer. Tschivuku, a
man's n'kisi, was embodied in a woven ball kept in a rack-like shrine
hut in a village, to assure successful trading. Its observation consisted,
among other things, of the men returned from their trading journeys
playing a sort of kickball with the n'kisi, laughing and carrying on,
74
while the women remained out of sight. Mpinda was a large bust,
three-fourths human size, an n'kisi of the land, who protected river

trade. Pechuel-Loesche mentions an unnamed trading n'kisi con
sisting of a red trunk kept in a "factory"—trading warehouse—
75
containing all sorts of medicinal objects (see plates 19, 20).
Lemba fits well into this taxonomy of nineteenth-century medicinal
"growth" and development. Gussfeldt and Bastain of the German
Loango expedition illustrate Lemba as combining fertility, marriage,
adjudication, and trade control in a unique synthesis that became the
wide regional system to which this book is devoted.
Farther back from the village, surrounded by a papyrus fence,
there is a hut called the Lemba house. It is accessible only by a
couple which is associated with theLemba fetish. All others are
forbidden entrance. This consecrated house is used to store
precious things, much as the Parthenon in Athens was a state
76
treasurehouse.
The closest marriage relationship a man may have with a
woman is in Lemba. She becomes his nkaziLemba. She wears
the Lemba charm. The bracelet (malungu) is an expensive
affair, and the charm requires a decorated house, thus only the
wealthy can afford it. Between the doors of the well-locked
Lemba house, which constitutes the treasure-house of its
possessor, are planted two trees, a baobab for the man, and a
cottonwood (mafuma) for the woman.... the key of the house
is given the wife with the Lemba ring, and her alone. This ring,
worn on the right arm, is consecrated together with a roundbox
(ludu Lemba). The second wife has a house too, and wears a
small medicine satchel on her arm
When husband and wife meet for marriage in theLemba house,
the man dons a large copper ring, and the woman a smaller one,
the charm being consecrated with a spittle of wine. Whoever is
intitiated into the secrets of Lemba can as "Tata Lemba" pass
on the order to a "mwana Lemba", and when the latter, the
Lemba child, becomes ill, he must come to his spiritual Father
Lemba to be healed through his medicine. In a Lemba marriage
husband and wife are obligated to mutually trust one another
and confess all shortcomings to one another, lest they become
77
sick when eating together.
Not a trace of royalty is evident. The widespread existence of Lemba
copper bracelets and drums in museum collections from the coast and
from Mayombe dating from this time, as well as the accounts of
broadening wealth in coastal societies following the complete col

lapse of royalty, give this account credence.
In the coastal societies, then, Lemba emerged in the seventeenth
century in connection with copper trade and consolidated its strength
in the eighteenth century. As royal clans found themselves over
whelmed by various political and economic problems, a wealthy
mercantile class emerged to control the interior trade and the various
posts that were needed to administer it. Traditional rules of exogamy
and exchange ordered relationships between these clans, partially
related to domainal nsi estates, partially linked to trade. Lemba, as
both rite and organization, legitimated this new structure. Later
chapters will show how it adapted conventional religious symbols to
its own purpose, and developed a pervasive and unique ideology of
healing relating to its concept of a stateless political order.
THE T E K E F E D E R A T I O N
The Tio king, Makoko, like those of Loango, Kakongo, Ngoyo, and
Vungu, was ordained by a powerful, permanent spirit linked to a
territorial domain. This was the spirit Nkwe Mbali, thought to be as
old as Tio kingship itself. There are no good historical indicators of the
date of origin of the Tio kingdom, although legends in the wider
region—from Kongo to the coast of Loango—suggest its origin well
before the fifteenth century. The ideology of the local authority rooted
to an estate and ordained in a territorial or place-specific spirit is
widespread in the Equatorial African region, as it was on the coast.
Vansina believes that Tio kingship arose out of the acknowledgement
of mystical superiority of one of the local authorities—"squires"—
followed by a modest tribute payment in exchange for insignia of
legitimation. The king thus became a primus inter pares among the
regional domainal squires. This mystical—ideological—origin theory
of the state among the Tio is defended by Vansina because of the low
population densities of the region, which would rule out conquest as
an explanation, and because of the structural similarity of the role of
78
local squire to that of the king.
The Tio kingdom was already an ancient and important presence
by the sixteenth century. Variously named Anzicans, the Tege or
Teke, or the kingdom of Macoco, this political entity is one of the best
documented of Central Africa in the literature of travelers and
chroniclers. Its territory covered extensive river trade routes north of
the Congo/Zaire, and along its northern tributaries. Its miners and
smiths developed advanced techniques of metalworking. Although
Vansina doubts any integral association of the kingdom's formation to

the introduction of metal-working technology, royal symbolism
closely associates the kingship with smithing. Royal anvils, a sacred
fire, the royal smith, and the second title of the kingdom, ngandzuunu,
"owner of anvils," gave the king an aura of mastery over this
important technology. Teke miners and smiths worked the metals of
the Mboko Nsongo and Mindouli region before the Kongo and Sundi
invaders took over. Indeed, Teke forgers were often captured to teach
smithing to the new Kongo masters.
In the mid-seventeenth century a major reform was introduced in
Tio political structure related to emergence and spread of auton
omous lords. The implications for the present study are considerable,
since these lords possessed many characteristics in common with
the Lemba priests, including their major insignium the nkobi
charm box. The nkobi lords ruled over the squires, collected tribute
from them, and were recognized by them as nkani, mediators of
tribute to higher-up lords, and ultimately the king. About twenty
lower-level nkobi lords became the nkani of the local squires. A few
of these lords, and several major lords, constituted thefirst-ranknkobi
79
lords of greatest prominence.
A nkobi lord's autonomy was indicated firstly by his freedom of
movement although he collected tribute and drew his spiritual
strength from local squires. Each nkobi was rooted in a local nkira
spirit and its local shrine. But the nkobi lords could travel; they had
their own histories. Each nkobi had its own name, its master title. For
example, Impaw nkobi was held by the lord Ngempaw, whose nkira
was Ngaalito which resided in the forest of Kongo. Such lore
constituted the foundation charter of the nkobi. The nkobi were
graded according to their importance, and a lord's rank varied
accordingly. This expressed the competition between major lords and
made the system vulnerable to personal political ambition and
historical opportunities. The major nkobi were in theory twelve in
number, although actual lists varied. One reason for this was that the
famous nkobi had "mothered" children which contained some of their
ingredients. This mechanism of regeneration of the nkobi had the
80
effect of extending the reward and the tribute system.
The ambivalence of the lords' relationship to the king was
expressed in two, clearly diverging, sets of myths of nkobi origin. The
5
one set hold that one day a huge nkobe appeared at the king s capital,
Mbe, after people from all over had waited for a long time for it. The
king then distributed the twelve major nkobi to his deserving depen-
81
dents praising his nkira earth spirit Nza, the creator himself. This
may be called the royal, centralizing, myth of nkobi authority. A
second set of myths represents a more autonomous legitimation for
the nobility. Accordingly, the nkobi came from Lord Ngia in Imbwe
near Abala region, or even farther away from a great chief at Mpiina
Ntsa on the Ntsaa plain. After he had brought the twelve nkobi, a war
broke out near Mbe, the Tio capital, and in the peace settlement that
followed it was agreed that the twelve tukobe should be divided
between the lords of Ntsaa and Mbe. In this version the nkobi give a
legitimate authority which does not derive from the king but which can
be acquired by persons of wealth. Its antagonistic character to
kingship is well expressed; it is a mythical way of saying that power
82
flows from one's following and the number of guns one owns.
Vansina suggests that the nkobi was a mystical force behind the
lords' authority, a glorification of competition and power struggles,
and an assertion of authority with regard to the kingdom and its
ideologies. The nkobi lords were major judges, they controlled the
trade, they were polygynists with many wives. Their courts were well
built, even palisaded; their paths were cleared; their fields were big.
Despite this clear autonomy, a new lord had to be confirmed in his title
by the king, from whom he received a brass collar. The nkobi was
somewhat independent of the king's approval. The king's inaugural
incorporated formal recognition of the lords, in that at this occasion
the rank and the role of the lords would be spelled out. But the real
administration of the kingdom was carried out by the lords, who acted
as judges, controlling trade and tribute, while the king granted ritual
83
authority. It is, says Vansina, as if the Tio kingdom had really two
constitutions: one a very old one based on kingship and the domains;
the other somewhat more recent (seventeenth century) based on the
84
ranking of the nkobi and their lords.
Two developments related to the nkobi reform of the seventeenth
century bear special attention in a study of Lemba. The first is the role
of trade in the rise of the Teke lords; the second is the similarity of the
nkobi of the Teke lords with nkobe Lemba. N o doubt the great river
trade that linked Central Africa to the Atlantic coast provided the
basis of the lord's growing autonomy. In his review of the Teke role in
controlling the ivory, raphia, and slave trade, Vansina notes that
"Teke" came to be known as a category of slave in Colombia as early
as 15 60, and later in the seventeenth century as a special class of slave
in Brazil, the "Ansiku." Of particular importance is the Teke control
of trade at the giant market of Mpumbu at Malebo Pool. Historical
texts show strong centralized kingship up to the middle of the

seventeenth century; the king is in charge of provinces and functions
such as trade. Ngobila is the special administrator who controls the
river. After 1700, however, references to the kingdom become scarce.
Now the chiefs of the plains, especially Mboo, acquire legitimacy
through the nkobi ideology, reflecting a rise in wealth from the influx
of goods brought from Loango over the Tio-Laadi route and later a
85
similar influx over Nkemi and Ntsei.
Vansina also interprets the invention of the political nkobi to
reflect the growing affluence of traders linked with metallurgy in the
Abala area along the Alima river to the north of the Tio kingdom as
well as in the Mindouli area. The nkobi were not given to the lords for
political recognition of dependents so much as they were bought by
them as "super-charms." Only wealthy lords could afford them, thus
demonstrating their ties with the wealth of mining, forging, and
86
trade. To what extent this also explains Lemba may be determined
by comparing the ingredients of both nkobe traditions. The similarity
is striking. The Tio lords' nkobi contains kaolin, camwood, three
other varieties of red obtained from stones, several types of shells,
87
indeed shells from Lower Congo, also metal objects such as
hammers, anvils, and imported manillas called ntsaa, war bracelets
and anklets fashioned in Nsundi and Loango, necklaces of beads and
cowries. Around the nkobi shrine are found miniature bells, mvaba
signs of authority, anvils, and full-sized bracelets, red copper ngiele, a
ring of the ntsii chief, or ordinary ngiele and brass rings with relief
worked at Loango, carrying the onia broom emblem of justice (see plate
88
12). These objects are in many cases identical to the contents of
nkobe Lemba, as fuller analysis in later chapters will reveal. For the
Tio lords they reflected unu (power) and were all trade objects,
expressing power gained from prowess in trade. Lemba was then a
Western extension of the Tio nkobi system. The differences were that
the Tio lords were attached loosely to a king, an nkira domain source,
and the nkobi were hereditary. But in their function and much of their
insignia they were identical to Lemba. Their rise in the mid-
seventeenth century corresponds to the rise of an independent
mercantile elite in the coastal kingdoms.
T H E N S U N D I SYSTEMS
Nsundi (Sundi, Soondy, Masundi) in ethnographical and historical

literature describes societies of both banks of the Zaire River, from the
Matadi/Vungu area in the west all the way to the Pool in the east. The
term also labels the important northern province of the old Kongo
kingdom. It describes as well chiefdoms that spread north of the river
during the seventeenth and eighteenth centuries, perhaps even earlier.
Finally the term is the name of an important clan of the region. Each of
these functions or entities needs to be examined here to understand
the political organization on the southern periphery of the Lemba
area, and its relationship to the coastal kingdoms and the Tio (see
figure 3).
Nsundi, as the northern-most of the four original provinces of the
Kongo kingdom, dates back at least to the fourteenth century. Some
maps depicting this area suggest that the Nsundi province took in
Figure 3
Seventeenth-century Nsundi districts and their capitals, and eighteenth- and
nineteenth-century Nsundi strongholds north of Congo/Zaire River (indicated
by names within dotted-line boundaries). District names, mostly south of river,
based on visits by Montesarchio (1650), Marcellin d'Atri (1697), Luc de
Catalanossetta( 1697-99), andMatheo de Anguiano(1706), as recorded in J.
Cuvelier, L'Ancien royaume de Congo, 1946, pp. 341-50 and 362b (map).
Eighteenth- and nineteenth-century north-bank sites based on author's field-
work and colonial archival sources, Luozi.
major areas north of the Zaire River, including the coastal kingdoms
89
of Loango, KaKongo, Ngoyo, and Vungu. Although the Kongo king
did cite himself as ruler of these kingdoms in praise epithets and may
have received gifts from them, it is unlikely that this represented
significant hegemony over the north-bank region. By the early
sixteenth century, when Europeans had begun to frequent the Congo
coast for purposes of trade, diplomacy, and missionary work, the
region north of the river under Kongo (Nsundi) control is depicted as
a small area directly north of the Nsundi capital—no more than
Dondo Mazinga and Nsanga districts—between Vungu and Tio
90
territory.
As the northern province of the Kongo kingdom, Nsundi was ruled
by the Mani-Nsundi (MaNsundi) who was appointed by the Kongo
king. Until the late sixteenth century, the MaNsundi was frequently
the eldest son of the Kongo king, appointed to rule this key province
before acceding to the Kongo throne. Nsundi's central place in the
kingdom derived from several historical and economic factors. First,
the original conquest of Nimi a Lukeni in the thirteenth century
probably came from Vungu and Nsundi, so the Kongo king had
natural allies there. Further, once trade between Mpumbu at the Pool
and the coastal ports such as Ambriz had developed, Nsundi was
strategically situated for the control of caravan trade routes running
right through Mbanza Nsundi and Mbanza Kongo.
Nsundi's integral place in the early Kongo kingdom is evident from
the well-known events and personalities involved in early contact
with Portugal in the late fifteenth and early sixteenth centuries.
Nzinga a Nkuwu was ruler of Kongo in 1491 when Portuguese
explorer Diogo Cao visited his court at Mbanza Kongo. At this time
Nzinga a Nkuwu's son Nzinga Mpemba ruled at Nsundi. When
Nzinga a Nkuwu died in 1506, another brother Mpanzu a Kitima
contended with Nzinga Mpemba for the throne. As is known to readers
of Kongo history, Nzinga Mpemba had embraced the new religion of
Christianity, whereas Mpanzu supported the autochthonous spirits.
In the battle between the two brothers, the Christian Nzinga Mpemba
killed the "pagan" Mpanzu. Nzinga Mpemba was inaugurated as
Affonso I, whose long rule represents in many ways a high point in
Kongo history. Trade with Portugal flourished, at least at first;
diplomatic relations were opened with Lisbon; ambassadors and
students were sent from Kongo to Europe; missionaries opened
schools; and craftsmen came to construct churches in the capitals of
Kongo. However, already during his reign Affonso witnessed the
91
development of the slave trade and intrigue against his authority.
Following Affonso's death the Kongo kingdom was further paralyzed

by the take-over of Portuguese mercantile and political interests, by
insurrections of vassals, and by the invasion of warriors who not only
sacked the capital but also drove through northern and southern
Kongo territories, establishing small but fearsome military enclaves.
One of these military colonies was created on the north bank near or in
the Mindouli/Mboko Nsongo mining fields, ostensibly for the control
of trade and minerals which were becoming important in the coastal
trade of the late sixteenth century.
The fluctuating stability of the Kongo throne after the death of
Affonso affected affairs within the provinces; for example, the extent
to which the king could dictate policies in the provincial governor's
(mani's) appointments, the choice of hereditary princes, and the
extent of tribute to be collected. Nsundi continued to be fairly stable
as is well documented by resident missionaries like Father Jerome de
Montesarchio. In theory the Kongo king was to be chosen by the
electors of the royal family. He in turn selected the provincial
governors from among candidates presented by local provincial clans.
The king could also in theory nominate governors of the provincial
districts, although in Nsundi this right was assumed by the provincial
governor (Mani-Nsundi). Chiefs of the provincial districts were
92
elected by matrilineal relatives of the incumbent. In Nsundi there
were at least twelve such districts whose names appear consistently
over several centuries and still today as names of modem villages,
towns, or clan estates: (from east to west) Binza, Kinsitu, Boenze,
Nsanga (north of the river), Wembo (just south of Mbanza Nsundi),
Sevo, Kwilu, Kasi, Ngombe a Kinu, and Mpemba Kasi, Mazinga
(north of the river), Ntanda Kongo, Kilemfu, Lemba, and Nkanga
93
Nsundi (see figure 3 above). Each district had its Mani, representing
the ruling local lineages. Each in theory paid tribute to the provincial
Mani-Nsundi who represented the King. A centralized, hierarchic
structure existed of the sort seen in the coastal kingdoms and among
the Tio, although the scale here appears to have been greater. With the
growing influence of the Portuguese, such titles as "duke" and
"marquis" appeared in respect to the local offices and landed estates,
as well as "fidalgos" for nonterritorial posts. These are little more
than overlay on pre-existing office titles. The "duque" of Nsundi,
given permission to nominate "marquises," appointed the marquis of
Nsanga, Kifuma, Zimba, Ngombe, Mazinga, etc., officers already
known to be mani of the various districts of Nsundi.
After the decline of influence of the Kongo kingdom in the late

seventeenth century, trade shifted from just south of the river to the
northern area and to the far southern area of Cassange and Loanda.
Glimpses of Nsundi provincial structure thereafter suggest a con
tinuing centralized polity, with the autonomous n'tinu MaNsundi
being inaugurated with the regalia and the insignia of the leopard skin
like neighboring groups, required to battle for his rights to the throne,
to kill a matrilineal kinsman, and to pay his own consecration fees
instead of receiving the nomination from Kongo. In the inauguration
of the last MaNsundi, NaMenta, in the early nineteenth century, no
mention is made of the Kongo king, or of any influence from Mbanza
94
Kongo. All is left to the nkazi a nsi, the "wife-clan" Mpanzu,
customarily designated as the inaugurators and wife-givers of the
Nsundi king-designate. Mpanzu "kidnapped" NaMenta early in his
life to prepare him for his role. When he reached puberty they
castrated him not only to keep him from fornicating with unmarried
women but also to guarantee his symbolic "otherness." After
collecting the coronation tax from Nsundi, Mpanzu and the candidate
waged war with the other contenders to establish their right to occupy
the mbanza. Having routed their opponents, they proceeded to
decorate their candidate-king with the ingredients of the royal
medicine, kiyaazi (from yaala, to rule), including chalk to anoint
him and leopard skin upon which to place him.
Then they placed a rod over NaMenta's shoulders, set a

leopard-skin diadem on his brow, a necklace of leopard's teeth
round his neck, a plaited cap (mpu) on his head, and a loin-cloth
95
about his loins.
A free-born woman of the Kimpanzu clan was brought before him and
designated n'kazi or nkama, queen. Because of the king's sterility,
she went to bed with the king's brother. This structure of the
nonconsummated royal marriage resembles that in Loango and
Sonyo nobility.
Because of widespread poverty in Nsundi by the early nineteenth
century, the last Nsundi n'tinu was not given proper burial on his
death in 1835 either by the combined supporters of the Mpanzu (the
coronators [n'kazi]), by the children and grandchildren (mayaala)
of the Nsundi, or by the Nsundi themselves. As in Loango, however,
other types of offices and practices maintained the skeleton of hier
archic tribute and title exchange for a time, even though the royal
office of king (n'tinu) had now lapsed. In a sense the kingship had
already lapsed much earlier, becoming a mere expression of symbolic
96
authority, the king being no more than a charm of power. Elsewhere,
closer to the north-bank trade which continued throughout the
nineteenth century, the symbols of Nsundi authority and the titles that
were conveyed took on new substance and even expanded.
In the region of the Manianga market along the trade route in the old
districts of Sevo and Nsanga, the Ns aku clan carried on the practice of
granting designation of the Mpu title of chiefship to local authorities
in exchange for tribute in slaves, pigs, and cloth. It had been an ancient
duty of an Nsaku (Nsaku Lau, Nsaku ne Vunda) autochthonous
priestly class," senior of the Kong clans," to ordain the Kongo king and
97
to counsel him in his decisions. Now in Nsundi, an Nsaku ruler
named Ntotila Fuanda, born to his mother Matele in Kinkuzu, in the
Sevo district south of the river, had received the Mpu from Ntete,
governing at Kinkuzu, as had Kata Mandala at Dunga (also in Sevo)
and Makita at Ngombe Matadi (just north of old Mbanza Nsundi).
Fuanda crossed the river northward, establishing his rule at "Kim-
banza," where he bestowed the Mpu upon subject chiefs in exchange
for tribute in slaves, pigs, and cloth which he forwarded annually to
Ntete in the south (see figure 3 above). The clans receiving the Mpu
98
from Fuanda north of the river were Nsundi and Kingoyo. As in
several other instances the titles of chiefly office are bestowed upon a
candidate by office-holders or priests of another clan. Here, the
Nsaku priestly clan carries out this function.
The fundamental feature of such a structure of authority is the
mutual complementarity of two exogamous clans, the acknowledge
ment of one as "royal" (kimfumu) providing the candidate, the other
as providing the queen (kinkazi) and possibly the sacred emblems of
authority (kiyaazi). A variation is that the one exogamous clan
becomes "priestly," the other "secular." The regent must in this case
be "detached" from his kin group through a ritual murder of a
matrilineal kin, his castration, or ritual acts such as a series of taboos.
Finally, there is always reference in Kongo chiefly and kingly power
to the "children" (baand) and "grandchildren" (batekolo), who
collect the offspring of males of the ruling house (the mayaala) who,
themselves unable to accede to the royal throne, have a strong vested
interest in creating and maintaining their "fathers" in authority. These
are the elements of Western Congo social structure out of which
centralized structures of authority have repeatedly been erected. The
exchange of titles for tribute between the mayaala and nkazi on the
one hand, and the mfumu on the other, assures the perpetuation of the
system. But in order for the complementarity to work, the flow of
tribute goods must be maintained. Where this depended on trade, it is
apparent that fluctuations in trade set off crises in the structure of local
authority.
The authority system that developed among the Nsundi clans
northwest of Manianga perhaps as early as the late seventeenth
century and early eighteenth century followed the structures outlined
here. Historical and legendary accounts refer to Mwembe Nsundi as
the capital of a complex of chiefdoms extending from the Zaire River
northward into the copper mining area of Mboko-Nsongo, westward
into Mayombe, and eastward as far as Mindouli." Mwembe Nsundi's
location is specified further as being in the watershed area between the
Luala and the Luangu Rivers, which is the point where the northern
Mpumbu to Loango caravan route crossed the copper-mining region.
The presence in legendary history and in contemporary place-
names in the general Eastern Mayombe-Manianga region of the name
Mwembe, and documentary evidence of several clusters of chiefdoms
spawned by invading Nsundi clans in the seventeenth century and
later, make appropriate its designation as a "Mwembe system."
Recurrence of the name "Mbanza Nsundi" suggests that the north-
bank Nsundi societies are the result of cultural migration in which
entire place-name and authority structures were transported from a
home region recreating new settlements from blueprints of the old,
analogous to North American place-names such as New York, New
Berlin, New Orleans, New Mexico, New Hampshire, etc. Mbanza
Nsundi (Nsundi Center, chief ship) was used in the settlements
created by those who rushed to the copper mines of Manianga and the
control of lucrative trade along the highland caravan route from
Mpumbu to Loango. Archival materials help us to describe one such
Nsundi enclave around Mbanza Mwembe (see figure 3 above).
At Mbanza Nsundi (north of the river, not to be confused with

ancient Nsundi provincial capital) the Nsundi clan [in ca. 1700]
divided and scattered to find a new land. One group settled in
Mbanza Mwembe, attaching to its name Nsundi the term
"Mwembe" to differentiate itself from the first group, with
which it then married. At Mbanza Mwembe the clan prospered,
adding numerous settlements at Ngombe, Ngombe-Masaka,
Kombe, Diada, and others. Thus installed, the clan organized
itself separately in the tradition from which it came. A chief was
68 THE P U B L I C SETTING
named to reside at Mbanza Mwembe; the villages put at their

heads chiefs (mfumu bwala) or judges (nzonzi) to settle
100
disputes.
Instead of intermarrying with local autochthonous clans—whom the
Nsundi oral historians sometimes allege were not there, the land
having been "empty" when they came—they saw fit to divide the
leading clan into two exogamous halves, permanently severing the
"blood" through the sacrifice of a pig and a common feast. On the
model of pre-existing Nsundi polities, they devised a system of local
chiefs and judges. In due course the royal model could be recreated
around a ruler and other basic elements, just as had been done in
Kimbanza by Ntotila Fuanda.
Nkombo Mafwana, chief of Mbanza Mwembe, who in the
meantime found that the land of that region was not suited for
him, left to found a new settlement at Ntombo. There he called a
meeting of his chiefs and had himself named Lulendo (power)
chief, with authority over the entire clan. He received the
investiture from Mayombolo, chief at Mbanza Nsundi, and was
succeeded in the Lulendo office by Mangovo Kabi, Nsango
Lumba, Makai ma Diengila, Ngoma Ngonde, and Mazanza,
the last of whom had just been inaugurated when the Europeans
101
arrived.
The inauguration ceremonies followed appointment typically by
election of the incumbent Lulendo chiefs brother, or maternal
nephew within the royal house, and ratification by all the clans.
In his new Lulendo enclosure of trees, the Lulendo chief
underwent isolation. N o member of his clan could see him, only
his wives and the "children" (band). A priest was attached to
the investiture to teach him his prohibitions (longo). After the
royal charm (Bueno nkisi) had been brought by the chief of
Mbanzi Nsundi, the sword of authority (Mbele a lulendo)
would be brought before the new chief, seated on his leopard
skin. Two slaves who served as executioners (bayala mabangu,
"rulers of the sword") would bring two of the new chiefs
matrilineal nephews. For the first and last time in his life he
would take the knife and cut off his kinsmen's heads, bathe
himself in their blood. He would then receive Bueno. Never
again could the chief touch his sword; it was kept by his guards
102
and used for the execution of criminals.
The ritual murder of a kinsman indicated the intention to set the

candidate apart from his lineage as well as the creation of an order of
law and administration on a scale larger than any of the local clan
estates. In the particular polity under discussion (Mwembe Ntombo),
sources suggest that the Lulendo's duties centered on settling feuds
through the intervention of his soldiers and holding an appeal court
over which he was the supreme judge. His soldiers consisted mainly of
"bana" bom to slave-women and therefore totally dependent on him
for their well-being. Sources say nothing of tribute, but it is hard to
imagine a centralized polity with an "army" and a "court" lacking
any type of tribute. It is known that Lulendo's powers were held in
check by the children (band) and dependents (bamayaala) who not
only ratified his election from the ruling clan but also saw to it that he
would not "take the clans with him" by dying of disease in old age;
they choked him to death when he became too old to rule effectively.
In the sources on this Lulendo chiefship in the southem-most of the
Mwembe Nsundi enclaves there is mention of a Kiasi chief akin to the
autochthonous Kinsako priest of Kongo and Nsundi, or the Mpu
priest, or the Bunzi priest in Ngoyo, a mediator of the local spiritual
forces needed to legitimate the invading, conquering, military office.
In Mwembe, the Kiasi bore the title of Ma Muene (lord); he also
levied market tribute. Like the Lulendo chief, he was inaugurated by a
priest from Mbanza Nsundi.
When the seclusion was over the masses entered the Kiasi's
enclosure to find him sitting on a leopard skin laid over elephant
tusks. The chief of Mbanza Nsundi would give him his
regulations of office, and trace in mpemba chalk on his body: a
transversal line on the forehead, on the nose, and a line on each
leg. Then he would be given the miasi emblems of office, a collar
of leopard teeth, and bracelets of iron for each leg and arm, as
well as earth from the burial grounds of the previous Lulendo
103
and Kiasi chiefs.
These Nsundi political systems were thus a series of loosely
affiliated small conquest states or chiefdoms radiating out from a
number of centers such as Mwembe Nsundi in the copper- and iron-
mining area around Mboko Nsongo, situated on the highland
Mpumbu to Loango trade route. The emigrating enclave that estab
lished itself as a small state sought legitimacy via a link back to its own
traditional source (Mbanza Nsundi) or via the benediction by the
autochthonous spirits of the earth (Bunzi, Bikinda, Mpu, Kiasi, for
70 T H E P U B L I C SETTING
example). In all instances there seems however to be a concern for the

establishment of a royalist model of governance. In this sense the
Nsundi invaders differ from many other north-bank political tradi
tions. Lemba constitutes the other major transformation of the basic
structure of public authority legitimated by the autochthonous
priesthood of the earth.
LIFE B E T W E E N T H E K I N G D O M S : MARKET LAWS A N D LEMBA'S

M A R R I A G E POLITICS
Between the kingdoms on the coast and the Teke federation, and
where no Nsundi chief had asserted his local hegemony, there existed
an acephalous political system. We may contrast it to the centralized
model of the polity found in foregoing discussion by noting that there
was no office created by either a ritual murder of a kinsman or through
appointment by another central figure. There was generally no tribute
payment to another chief. N o grand judge could be found in the form
of the N'tinu who acted as appeal court. There were no standing
armies nor loyal dependents like the Lulendo chiefs soldiers in
Mwembe Ntombo to impose order. A colonial report in 1938
described the political system in the same negative terms as those
used in ethnology of the time:
As always, the "northern" region has shown itself more
resistant Not one chef médaillé has succeeded in imposing
himself in these chiefdoms whose populations traditionally, for
that matter, never submitted to the authority of one man. Well
before our occupation the indigenous people lived in a sort of
anarchy, not recognizing any authority other than the hardly
effective family chief holding authority more often religious than
104
political.
A positive description of this political system would need to include
the following characteristics: (1) several types of local chiefs,
including the lineage or clan chiefs (mfumu dikanda, mfumu nsi) and
the various types of judges (nzonzi); (2) several ceremonially-rich
min'kisi which recognized authority, including Mpu, N'kondi, and
Lemba; (3) the so-called "market laws" (min'siku mia zandu), & set
of prohibitions and practices that spelled out the rules of peaceable
trading, and regulated the measures of adjudication and punishment
for their violation. These dealt with most public affairs such as trade,
hunting, and marriage, and were maintained by councils of judges

(zinzonzi).
The local chiefs were attached either to descent communities or
residential communities and markets. These two types of polities or
organizational principles interacted to provide a dynamic balance of
power and a coverage of administrative function. The lowest level of
lineage chief was the head of the mwelo "door" of the house. His role
was not really titled unless as judge. Such an "elder" assured the well-
being of his dependents: to provide them with land, trees, and other
rights, and to see that they were properly married and cared for. The
role of the "clan chief (mfumu dikanda) was a more inclusive
version of this; his duties as well covered the subsistence of kin
dependents and their proper marriage relationships. His was the role
of the men's hut chiefship where clansmen and their dependents
gathered daily for eating and palavering. The clan chief had also to
assure that the ceremonial duties of the local clan section were taken
care of by the patrifilial children and "grandchildren," as priests of
their "fathers."
The "village chiefs" distinguished between the principle of landed
estates and the populace on it. The assumption was that one or two
clans were landowners, claimants of the ancestral cemetery grove, thus
the presumed autochthonous inhabitants of the immediate area.
Several titles, reflecting regional characteristics and historical dif
ferences, belong here. All gave the titleholder the recognized acknowl
edgement of being a freeman. Perhaps the most widespread title for
the role of head of such a landed estate was mfumu nsi (landed estate
chief), a term in widespread use in the Mayombe. The term was also
known in the Manianga region where the office holder wore a twisted
copper bracelet that passed within the matrilineal line of succession.
Such clans as Mazinga, Nsundi, Kikwimba, Bwende, Kimbanga, and
others were noted mfumu zinsi of this region. In some areas the title
mfumu yetila or yetisi was used, but it connotated chiefship of the
village.
Comparable to the estate chief is the "crowned" chief mfumu mpu.
Some informants in the North Manianga said it was identical to the
domainal chiefship, reflecting freeman status in a land-holding local
clan. The Mpu chief, like the others of this category, was inaugurated
into office and supported by the collective "children" bom to the
clan's men. Through the isolation of the Mpu candidate from his
matrilineal kinsmen, the patrifilial children were able to generalize the
loyalties of their fathers, to broaden allegiances so that fractional

differences within the matrilineal clan of their fathers would not
destroy them as a landed, corporate group. Mpu-ship was often
allocated to a person who was "sick." Structurally, the Mpu was
situated in the free clan so that his consecration to ceremonial office
was complementary with the structural positions of power elsewhere
in the clan. The Mpu could be used as an appointed title in a
centralized kingdom such as Kimbanza at the northern fringes of
Nsundi, but in a neighboring acephalous setting, it would be generated
from "below" to thwart political fragmentation.
Thus far the political structure of theLemba region, as described, is
not distinctive from low-level organization elsewhere in the general
Kongo region. However, examination of the more inclusive levels of
political structure reveals a striking contrast to the kings and
prominent chiefs found elsewhere. The "laws of the market"
(minsiku mia nzandu) and supporting roles, functions, and cere
monials, reflect a unique political system. The market system in
north-bank society operated around the four-day week (see figure 4).
Each market, located on an open plain between several villages, was
"ruled" by a market committee of chiefs or judges from these
surrounding communities. While the notion of a sovereign was absent
in this system of acephalous governance of markets, the functions of
large-scale government were very evident The market was regarded
as a "court," with the right to impose capital punishment. The market
committee and its constituent communities constituted a "market
area," which could combine with neighboring market committees to
solve large-scale disputes. Often village or clan chiefs would have
tenure in two markets for more effective regulation of disputes. The
"laws of the market" thus regulated both trade in the markets as well
as the fabric of society that kept the markets open and viable. Munzele
has listed the major market laws from the Kivunda area:
1) Theft is prohibited on all paths.
2) Fighting is prohibited.
3) Beginning a feud in any village or market district is
prohibited on pain of paying for it in persons.
4) A person may not enter the market area ov sit there before
/
the market heads have given the signal.

5) Kimbanzia grass, because it is sacred, may not be picked
or removed in any way from the market; it stands for the health
and prosperity of the clan, the benediction from God, the
ancestors, and the clan.
Figure 4
Market areas and villages of the Kivunda region, Luozi (the Manianga);
corresponds to Nsundi Masiki region in figure 3
NKQYI
\ V / Kikangaya ^ O
Kimoanda I * — — ' / «\ v
/ v. ^ / Kikieka M P , A
K N ^x.
NKONSOy / . ^
^ ( Kiniangi
~ V \ J
' Kikungu n v j ^ m
f N
v v Kintwala y '
1
\ _ y /
I Kimbanga I - /
* NKQYI * ' ^* - ^ \
V i I Kimbedi
\ « i NKONSO 1
w \ Kimbaku / \ Buyala M
1
^ / * Nsundi Nsundi /
B a K O n g 0 l U T e t e %
X - ' , ' Ì n U
MPIKA \ ^ ' - * ^
\ •
\
Kimpondo / • *"nioi \ /
t / Masangi NKILA / ' K l ¥ U n d a
Kimbwala N k a k a J
_ \ Nseke
/ \ x ' ^ " ^ ^ X N Mbanza
^ v
_ - ^ / King,la N ^ N
Kintadi , ^ ^ — — ( \
' LEGEND i \
Ngoyo — Village \ NKQYI \

NKILA— marketplace ^ ^
) — market area
\
10 kilometers \
1
' * i l l \
6) Only the market committee has the right to invoke the

death penalty on a person.
7) A severe punishment falls upon any individual or clan
who kills or harms a palm wine tapper, for palm wine is the
ingredient of social intercourse.
8) A judge (nzonzi) or a messenger carrying the judge's staff
may not be harmed or seized, on pain of being punished in the
105
market place.
The market chiefs would meet in the market center prior to trading to
"open the market" while contingents of men, women, and children
from surrounding villages waited at their respective entrances (see
plates 2, 3). This committee determined what would be sold or
exchanged, whether criminals—murderers, recidivist thieves, adul
terers, or violators of the market laws—in any of the "entrance"
groups (mafula) deserved being "planted in the market," a euphem
ism for the method of capital punishment whereby the victim would be
mercifully given great quantities of wine, seated in a hole, and buried
before a sharpened bamboo stake was driven into his head by his
possessor or close superior. The market council might also decide that
hostilities between groups were so great as to require canceling the
market on that day. If and when trading was opened, it was done in the
center of the market by delegate groups from the "entrances."
Women and children were rarely permitted access to the center, for
fear that they would be seized as pawns or hostages in on-going feuds,
or as debt payment. Following a peaceful market, men would gather
under a shelter to drink and talk.
Where feuds broke out between participants in a market group—
over land, women, hunting accidents—the absence of central judicial
institutions became evident immediately. Even the fabric of warfare
reflected the lateral alliances rooted in the market structure and in
marriage ties and descent, as well as in the organization of the landed
clan villages with their clients and slaves. All of these social domains
structured feuding and peacemaking, as the following example of the
"War of Kidiba," fought in ca. 1880, in the Nsundi Lutete area,
illustrates. (For villages mentioned refer to figure 4.)
The "War of K I D I B A " between the people of Kisiasia and the

people of Kimbaku took place on account of M F W E M O - M A Y -
AMBI of the Mazinga-Kingila clan and his wife M A N G E Y E -
KIDIBA of Kikwimba of Kimbaka (MFWEMO-MAYAMBI was a
slave at Kisiasia.) KIDIBA ran away from her husband; she didn't
care to live with him any longer, even though they had two chil
dren, a boy, B A N U N D A - D I A T A , and a girl, KILENGO. When she
ran from her husband she went to live with another man named
M U N S W A N G A L A who lived at Masangi, of the Mazinga clan of
chief KIODI. This angered M F W E M O - M A Y A M B I greatly, so he
gathered the people of Kisiasia so that they could go and seize a
hostage (bwila nkole) at Kimbaku: a woman in replacement.
Meanwhile KIDIBA had left Masangi to go to her lineage home at
Kimbaku. When the people of Kisiasia saw that it was
MFWEMO'S wife, they seized her and brought her to Kisiasia to

her husband. But MFWEMO'S anger was not satisfied: "We must
go to war (nwana muzingu) with her uncles (bankazi) at
Kimbaku." Our elders (here in Kisiasia) told him though "there
is no need to go to war over your wife. Let us be satisfied that she
is in our hands again." But MFWEMO would hear nothing of it.
"We must go to. war over the matter," he insisted. The elders
then told him, "we cannot follow you in this matter; if you must
fight, it is your affair."
Then the warriors (makes a, young men) of Kisiasia agreed to
go to war as MFWEMO wished. They went near Kimbaku to wait
for the lineage headman (mfumu dikanda) of the woman
KIDIBA to warn him with a gun salvo (sika bizongo). The people
of Kimbaku replied, "Never mind, we have warriors also and
they know how to fight too." And so those from Kimbaku fought
against those from Kisiasia.
Then other villages entered the conflict: Kintwala and
Kimwanda were brought into the fight on the side of Kisiasia,
and Masangi and Kumbi on the side of Kimbaku. MAZALA and
M U N A N U of Kintwala, of the Nsundi clan of BINKITA, were killed
and hurt with a ball in the eye respectively, MANAKA of
Kimwanda was killed too. When the people of Kintwala and
Kimwanda saw that they had each lost a man, they went to
Kimbaku and killed N Z U Z I , a sister of chief ZIONA of the
Kindamba clan, who was in the same house with BAYEKULA of
the Kisiasia. She [Bayekula] was unharmed.
Now, seeing all the killing, the elders andjudges (bambuta vo
nzonzi) sought to call it to a halt, and sit to talk out the matter
(zonza nsamu). They agreed that KIDIBA was in the wrong for
having left her husband. She was charged with payment of " one
corpse" (futa mvumbi), that is seven people. Further, they
judged that MUNSWANGALA of Masangi was also at fault for
having taken a married woman from her husband. He too must
"pay a corpse," or seven people (to the clan victimized), plus
seven cloths (mbongo). MFWEMO was judged for having started
the war, and charged with paying "one corpse and a half in
people (of his lineage), BILONGO, MAKWENIA, MIHAMBANTU,
KIHENI, and he himself MFWEMO-MAYAMBI paid for the corpses
killed in the war. They further paid their debt as follows: out of
MFWEMO'S lineage, KUKINGA went to MALEWO of Kisiasia
where he became a slave of the Kimbanga clan; TEMBOSO, a
woman, was sold to DIAKOKA at Kintwala; MAKTJNDU, a male
was sold to M U N D E L E at Kimbedi. Further the members of the

Mazinga line (MFWEMO'S) sold the land they had bought
(kudisa ntoto) to the Kimbanga clan of Kisiasia, to allow
MFWEMO to pay his debt to Kimwanda for the man they had lost
106
on his account.
Seizing a hostage was a favorite manner of taking justice into one's

own hands. Yet it was not one's direct antagonist from whom a
hostage was taken, rather from another stronger community which,
when informed of the strategic purpose of the kidnapping, could
impose settlement terms on the original antagonist, its status sub
ordinate. As in contemporary international terrorism, this method
could also result in the embroilment of a dozen villages in many
market units, and in numerous deaths. The laws of the market had
extremely harsh recompense terms for the guilty, once the passion of
the young warriors had subsided and the cool rhetoric of the judges
had taken over. Not merely the lex talionis of an eye for an eye was
invoked, but seven persons for every human casualty in the war. Thus
Mfwemo's entire clan and its land were dispersed to pay for the three
deaths inflicted in the war.
Another war in the same region in 1912 was provoked by an
impatient hunter who had lost his dog to an errant ball from a fellow
hunter's gun. In the settlement the aggrieved hunter shot at (but
missed) the judge. The council ofjudges fined him eight pigs and other
articles, to be paid to the market-area groups participating in the
judgement, as directed by the laws of the market.
The judicial structure of both cases reflects the absence of a
hierarchic appeal court. The conflict is absorbed through lateral
alliances within which the judges from the very groups involved, and
their neighboring judges, effectively impose upon the antagonist and
their patrons radical and binding punishments. Given the strategic
role ofjudges, it is not surprising to see in these regions, in connection
with the market-law system, elaborate rhetorical usages, songs,
proverbs, call-and-response sanctions, and rituals of conflict arbitra
tion. The category of charm called N'kondi existed at the level of the
clan, village, and market, as a ritual contract between persons or
groups who vowed not to make war, or to seize hostages from each
other's populace. Such N'kondi were canine or anthropomorphic
wooden figures into which blades, screws, and iron nails or hardwood
pegs were driven as the oath was sworn never to engage in mutual
hostilities, or following hostilities, to arouse N'kondfs m%tr{rr\funyia
= ferocity = wedge) against the antagonist, thus sanctioning them to

hold to their promises. Sacrifices of animals, usually goats, were
made as the N'kondi priest drove "blood wedges" intoN'kondi, and
evoked promises from the two partners:
Between your village and my village we have an accord, you
may not seize hostages from us, and we will not seize hostages
from you. If one of our people does something to the other group,
107
we will meet to talk and will not fight
The other major charm of public order, beside N'kondi, was of
course Lemba. It is described by Munzele as an n'kisi integrating
people, villages, and markets (Lemba i n'kisi wabundisa bantu,
mavata ye mazandu). It permitted persons from one market area to
travel and trade in another market area. One who carried Lemba* %
insignia—nkobe, bracelet, staff—conveyed the information that he
9
had received Lemba s instructions, and was therefore trustworthy.
He must be received in the market, for he was a person of peace, a
carrier of light (n'nati a mpemba). Lemba priests and priestesses
were the prominent citizens. In the above narration of the War of
Kidiba, the narrator's father, Nsundi-Mukila of Kintwala, freeman
from Nsundi and a well-known judge, and his wife Kibuni, freeman of
Kimbanga (Kisiasia), were initiated intoLemba. Likewise, Mundele-
Mbenza, who shared in Mfwemo's "spoils," was a Lemba priest as
was one of his wives. Mundele-Mbenza was a slave who because of
his recognized intelligence and integrity had been elected by his
masters to represent their interests; he was a prominent judge and
Lemba priest.
The promotion of a slave figure to judgeship and Lemba priesthood
reflects the crucial role of marriage politics as another, major, factor in
the maintenance of alliances between clans, villages, and market
areas in the "region between kingdoms." The landed estate system
that extended from the Loango coast into Mayombe and eastward
into Teke country found local villages comprising a major matrilineal
clan with its mfumu nsi or Mpu. Client and slave lineages (clan
fragments) usually comprised a type of hierarchy of exogamous sub-
communities, closely intermarried The dominant clan of the domain
extended land and women to its client males, but needing their
support in every way and receiving from them women and labor and
political support. Most of the client lineages derived from women or
girls who had been seized in feuds or traded for debts and then had
married a male of the dominant clan. They usually came from neigh-
boring free clans. In the Manianga region dominant lines married

closely into their clientage, particularly during the first few genera
tions of their habitation. Men of the leading clan, during the nineteenth
century, could have from five to ten wives, mostly of slave origin. In
order to assure plentiful progeny in the dominant line, its women were
married to client men, thus keeping them at home with their children.
As has been seen in the review of coastal polities such as Loango
and Mayombe, the women of dominant (noble or royal) clans had a
sharply restricted progeny, thereby reducing greatly the number of
heirs. In the acephalous political context, priorities were reversed.
Preference was given in dominant nsi clans to marriage of their
women with numerous client and neighboring free clans. Often elderly
women of these lines would have had three or four consecutive
husbands—serial polyandry—to maximize their effectiveness as al
liance creators whose progeny belonged to the dominant group. A
given dominant clan would thus maintain from five to ten key
reciprocal alliance ties with neighboring clans to offset the tendency to
become an isolated endogenously-marrying polity. Alliances be
tween free clan sections and between a dominant landed clan and its
more powerful, populous, client lineages were the alliances most often
initiated, both husband and wife, into Lemba.
For example, in the Kimbanga village of Kisiasia, where I conduc
ted fieldwork in the 1960's and 1970's, about six Lemba households
had existed in 1920 at the end of Lemba: two representing Kimbanga
freemen married polygynously to client women, two more represent
ing Kimbanga women married to men of sizeable client clans. Several
other Kimbanga women were married in Lemba to freemen of
neighboring clans. These interclan alliances constituted traditions of
permanent "blood reciprocity" (jnvutudulu a menga) in which a son
by marrying his father's sister's daughter or some classificatory
equivalent returned to his father's clan the hereditary substance his
clan had received when his father married his mother. In one instance
Lemba slave priests perpetuated the spiritual tradition of their fathers
where they had had no female progeny. Lemba was thus a symbol as
well as a structure assuring the continuation of a descent line and
corporate community through the correct exchange of women.
Lemba represented the patrilateral flow between clans of an on-going
alliance, binding clan to clan and extending the requisite network
across the countryside to assure effective commerce. In the absence of
any other centralizing structures, Lemba assumed the functional
equivalence of the state.
In summary, it is possible to see underlying similarities in the

Lemba region's political systems and the way they were expressed in
sacred symbols. On the one hand there were everywhere polities
linked to landed estates, many small-scale such as the domains of the
Teke, of the Central highlands, and the Mayombe. A few such as the
Tio king Makoko, the coastal kingdoms, and some of the Nsundi
chiefdoms were larger in scale. But these territorial, landed polities
were invariably legitimated by localized earth or water spirits such as
Kwe Mbali of the Makoko and Bikanda or Bunzi farther westward.
These were autochthonous spirits related to specific places, either
waterfalls or tracts of land, as were the estates they symbolized. On
the other hand there were the "moveable estates" linked to justice,
trade, or conquest. The Tio lords and the Lemba priestly couples with
their moveable shrines were consecrated to a variety of spirits, some
of which derived from localized shrines such as the Tio local shrines
or the western Bunzi; other symbols were legitimated in nonterritorial
spirits and forces such as Bunzi's daughter Lusunzi, the trickster
Mani-Mambu, the androgynous demigod Mahungu. All contributed
to the estate of Lemba, created and sustained by exchange, alliance,
and trade.
Chapter 3
Biographical Sketches of
Lemba's Demise
Introduction
An institution as widespread in scale and as deep in historical dura

tion as Lemba cannot be made homogeneous by ethnological and
historical reconstruction. Change, internal tension, regional varia
tion, and the institution's disappearance must be accounted for. This
chapter introduces biographies of Lemba priests and aspirants to
show how the drastic changes that occurred at the hand of the Congo
Free State and as a result of Belgian, French, and Portuguese
colonialism of the late nineteenth and early twentieth centuries
confronted individuals with alternatives leading to the fragmentation
of Lemba and its ultimate disappearance from the cultural and
political landscape of Lower Zaire.
The caravan routes that Lemba supported and controlled were
used extensively to move the machinery and baggage of early
colonialism from the coast to the Pool From ca. 1885 the demand for
porters drawn from local populations along the routes was accom
panied with threat, seizure of family hostages, and meager payment.
By 1887 the exigencies of the southern caravan route along the Zaire
River had through death or desertion depleted the supply of porters.
Many of the villages in the Ngombe area had fled north across the
1
river. More widespread labor recruitment became the practice in the
Free State backed up by " in-kind" food taxes levied to support the
armies and caravans. Labor recruitment for portage and for the
construction of the Matadi-Leopoldville railroad became so heavy
that numerous revolts erupted, and the loss in lives was high. Chinese
laborers had to be brought in to supplement the African labor force.
By 1896-7 portage along the northern route from Loango to
Brazzaville reached the limits of its labor potential carrying amission
of ninety tons overland (3000 loads at sixty pounds [thirty kilograms]
per porter). Here too there was a revolt, the " Sundi revolt," which was
put down repressively. Completion of the Matadi-Leopoldville rail-
81
road by 1900 permitted the abandonment of portage on both banks,

much to the relief of local African populations, reeling from epidemic
diseases, high death losses, and famines from overtaxation on the time
and resources for food production. Portage seems to have continued
on a sporadic basis in the French colony until 1915 when the auto
route was completed from Loango to Brazzaville. The Congo-Ocean
2
railroad to Pointe Noire was completed in the late twenties.
Although the cessation of colonial portage alleviated the heavy
burden of labor recruitment, its total substitution by rail and auto-
9
route transportation had the long-term effect of replacing Lembo s
economic function of three centuries, the revenue from which had
9
generated Lemba s ceremonial fund. Shorn of this traditional base,
Lemba could only sustain its economic viability by cutting deeply into
resources now claimed by the colonial taxes and corvée labor
demands. Disenchantment with Lemba grew within the populace.
One old Kivunda observer noted that Lemba priests and priestesses
only met now to fatten themselves on pork. Colonial authorities
perceived the competition for resources; thus French authorities
barmedLemba in the early thirties for "exploitation and extortion of
3
fees" destined for the colony's tax coffers.
Other dimensions of the colonial intrusion affected Lemba ad
versely by dividing its ranks into competing interest groups. There
were those who accepted the colonial chiefship and allowed them
selves to be drawn into the taxation system. Those precolonial chiefs
who refused, sending forward instead vassals, deputies, and even
slaves to receive the medals, benefits, and supports offered by the
Europeans, found their authority overturned by these one-time sub
4
ordinates. Colonial collaboration was now necessary to preserve
one's influence, although it destroyed one's authority through taxa
tion abuse.
Others abandoned Lemba to become mission catechists. There
were those who might once have been drawn into Lemba, but for
reasons having to do with the clash of old ceremonial demands and
new colonial taxes, perceived Lemba as having become too expen
sive, although still the preferred ritual solution to their ills. Christian
ity offered a way out of the impasse.
Finally, there were those who acknowledged the apparent bank
ruptcy of the old rituals, but refused to play along with either the
colony or the missions. Like Kimbangu, the Kongo prophet, they
forged new African rituals, bypassing the foreign priests and pastors
Biographical Sketches 83
to create an African response to challenges of colonialism and

competition of a strange world religion thrust upon them.
Accepting Colonial Authority: Ngarnbula,

Mampuya, Sobisa
All three of these Lemba priests accepted colonial authority. Two

(Ngarnbula and Luvuangu-Mampuya) became medalled "mpalata"
chiefs under the Free State/Belgian colonial system. Another
(Sobisa) was named chief by his clan, although still very young,
because it was thought he could better cope with the intricacies and
foibles of colonial officials. In all cases the colonial government
imposed difficult choices on individual Lemba adherents.
5
Ngarnbula's life story, told by Bittremieux in the twenties, reflects
these pressures on a successful leader in the old system. Ngarnbula
grew up a freeman of the Vinda clan in the Nkangu region of the
Mayombe. After his childhood and adolescence in his father's village
in the late decades of the nineteenth century, he entered the N'kimba
initiation school, where his N'kimba name Tu Masungi was added to
his childhood name N'lele Mbutu. For his recognized leadership he
was chosen to be a local N'kimba master.
After his happy youth a series of crises appeared around him and in
his life. A drought so severe occurred that the Lukula River near his
home dried up. This, in addition to the hardships of colonial labor, led
to famine. A s if drought and famine were not enough, locust plagues
descended on the area. Most significant for his personal future,
however, was an experience that grew out of a court case involving his
clan.
Ngarnbula remembered attending the palaver in the court of the
great Yombe chief Mabwaka of Mazamba, in Kangu region. Many
mfumu nsi chiefs came with double-gong musicians, staff carriers and
special advocates, favorite wives and their servants. Nobility were
seated on mats or leopard skins shaded by umbrellas. The debate was
over livestock, women, and other issues compounded into a major
confrontation between two clans. Speakers for the two sides would
take turns, arguing eloquently and waving their arms about. Ngarn
bula remembered particularly the musical chime resounding from the
bracelets of the Lemba priests. After the settlement was reached and
payment made in small livestock and wine, food was distributed,
ending the adjudication on a festive note. Ngarnbula remembered the

delicious mutton, and how he felt fortunate to be at so auspicious an
occasion in the court of the great chief Mabwaka, with so many
Lemba priests displaying their engraved copper bracelets while
toasting the health of their wives.
But above all Ngarnbula remembered his great consternation when
after the feast he was told that in secret bargaining he had been given to
chief Mabwaka as a pawn in payment of his clan's debt settlement. He
felt stigmatized, suddenly having been turned into property and a
trade good. However he remained in Mabwaka's court and was well
taken care of.
The next memorable event of Ngambula's life was Mabwaka's
death. Succession deliberations by the clan counselors failed to
identify a competent brother or nephew, so they turned to an outsider,
and favor fell on Ngarnbula. According to the custom of the land, he
could not become the MaKangu, chief in the Kangu dynasty; he had to
be made mmvuanzi chief, whose own direct line is nullified. In an act
of "chiefly incest" he took anpfumu Vinda wife from his own family,
thus severing his own line of descent. He was now a member of no
lineage, in order to be a chief over many. The people of Mabwaka
accordingly consecrated him as priest of the earth cult Mbenza.
Never again could he take k'oze (a dark fruit) in his hands, or climb a
palm tree. On an Nsona day he was placed on a dais (vunda) with his
chiefly staff in his hand; four palms were planted around him; and a
palm canopy erected over him. He was placed on a leopard skin and
anointed with white chalk (pezo) and red powder (ngunzi), and given
a fly whisk, while a healer danced about him (banguka). He was given
the new name of Mboma Simbi, designating his new status as an
Nkita spirit devotee—perhaps evenaniVfc/ta spirit, notes Bittremieux.
Amidst much dancing, drumming, and drinking, and made up with red
and white stripes all over his body, he was brought from seclusion to
his palace (m 'bongi). He was now called the MaMboma, and his staff
accompanied him everywhere.
The next important phase of Ngambula's life began with his head
wife's request to become his Lemba partner. Initiation to Lemba
would be expensive, but he agreed. The consecration was organized,
and many Lemba priests and priestesses were assembled. Part of the
ritual was held in the village, part in the bush, characteristic for
Lemba. The Lemba shrine consisted of the n'kobe (to be elaborated
in Chapter 7), a pair of small drums, and the copper medicine
bracelets worn by the priest and priestess. Sacred Lemba trees were
planted in the grove behind the house: mfuma the silk cotton tree,
lubota, nkumbi, and kuaku, as well as other plants. Based on the
proverb menga ma tsusu, simba: kuambula (chicken's blood, hold it
and let it go), he received his Lemba name, Ngarnbula.
Ngarnbula was now N'kimba master, priest of Mbenza, the
MaMboma chief, and priest of Lemba, all the roles of authority to
which a turn-of-the-century Yombe could aspire. In the eyes of
Belgian colonial officials Ngarnbula met the criteria of a nobleman. In
the Free State and early Belgian Congo tradition of autocratic
authority, Ngarnbula received from the colonial administration yet
one more title, Mfumu Palata, the "medalled chiefship," linking him
to the colonial state, making him eligible for protection, and obligating
him to recruit laborers, collect taxes, and maintain order. Ngarnbula
regarded the colonial chiefship as an enhancement of his authority.
But the difference with this title was that the tribute collected in its
name passed on to the central colonial coffers for the maintenance of a
distant state. Where the tax was "in kind" (food, for example) it was
used to feed soldiers and porters. But it was neither circulated in the
market nor redistributed to the local populace in the form of a lavish
feast.
Across Lemba territory, individuals like Ngarnbula struggled with
this issue of collaboration with the colonial government. In Manianga,
to the east, two other Lemba priests, Luvuangu-Mampuya of the
Kingoyi clan in Kimata and Sobisa of the Kimbanga clan in Kisiasia,
were typical of those who cooperated. Luvuangu-Mampuya had
already received the colonial chiefship when he participated, with his
wife, in the last regional Lemba inauguration at Nseke-Mbanza in
1919. He represented the type of figure who in due course becomes a
chef de groupement, an administrative level created beyond tradi
tional chiefdoms, to consolidate small-scale domains and clan
chiefships of north-bank segmentary society. The creation of a
tribunal with jurisdiction over the groupement lent this level of
colonial administration a certain legitimacy in the eyes of the popu
lace, and thus some sense of coming to terms with the new government
Meanwhile, Luvuangu-Mampuya retained his Lemba insignia, in
sisting as late as 1965 that he was still nganga Lemba and that
therefore he could not talk of his secrets. He would die with his Lemba
adherence, his ritual objects going with him to the grave.
The Kimbanga of Kisiasia, as has been pointed out in the previous
chapter, had several Lemba priests in their midst at the turn of the
century when the Free State's agents appeared. Nzuzi Pierre, clan-
section head, suggested in 1970 that the thinking in 1915-20 on

cooperating with the Belgians was that they would hardly disappear,
so it was necessary to deal with them as resourcefully as possible.
When taxes were first levied in 1910-15, a woman was delegated to
organize this work since women had the groundnuts needed for the
African soldiers. At the same time it became necessary to name a new
clan-section chief, mfumu dikanda. Skipping over many older men
who were eligible, clan counselors elected youthful Sobisa. Their
reasoning was that since he knew how to read he would better
understand the problems of the new era. But he was not the only one
who could read. Their decision was also based on criteria exercised in
initiating him mio Lemba. The idesALemba man had courage, rhetor
ical ability, and sound judgement. Thus even though Lemba's public
functions were usurped by the colonial state, Lemba's leadership
criteria continued to be exercised. Still, the divisive forces of the new
9
order drew Lemba s initiates farther and farther apart as individuals
accommodated themselves to new roles.
Populist Religion and the Missions: Katula

Ngarnbula discarded his ancestral medicines and charms in 1917, two
years after he met Scheutist Fathers at Kangu mission, because he
had a dream in which the white-robed Catholic priests surrounding
him persuaded him to take up the "truth, the word of God." Despite
the strength of the dream, he kept Yds Lemba name and identity, and
postponed baptism until he was on his deathbed some years later.
Another Lemba priest who converted to Christianity, but unlike
Ngarnbula became an active evangelist and teacher of the new faith,
was Katula Davidi of Nseke-Mbanza in the eastern Manianga.
Katula with his young wife had participated in the last Lemba
inauguration of the region in 1919 along with Luvuangu-Mampuya
and his wife. Whereas Luvuangu-Mampuya had stubbornly held to
Lemba while joining the colonial administration, Katula was per
suaded by the forceful preaching of Swedish missionaries to give up
his Lemba insignia. He only kept the hollow n 'kobe box as "souvenir^'
after discarding its contents. Reminiscing to me in 1965, he noted that
Lemba's membership criteria had been particularly important in
maintaining a high level of leadership in precolonial society. What
attracted him to Christianity, and particularly to Protestantism, was
the possibility with a few years schooling of becoming leader of a local
congregation and teacher in the local school, with literacy the key to
interpretation of the Scriptures.
Although Christianity attracted many for this reason, it was in its
way divisive. By the twenties, mission societies had divided up the
entire Congo, and Protestant and Catholic missions competed every
where for African souls. A friendly missionary priest or pastor was a
powerful ally in dealing with the foreign colonial government. The
tendency for Africans to identify with and support "their mission"
and "their missionaries" led to the alignment of the foreign Catholic/
Protestant schism with pre-existing African divisions and distinc
tions, precipitating numerous religious feuds.
Protestant populism particularly undermined the priestly class of
Lemba. Although both Catholicism and Protestantism—indeed,
Lemba as well—publicly maintained the ideology that their way was
for all, Protestant missions were more prone than other religious
persuasions in Congo to make everyone a "priest." Early converts
were often slaves ransomed and brought up by the missions. Early
translation and wide dissemination of KiKongo Bibles amplified this
effect. Catholics handled access to religious truth more cautiously,
insisting on properly consecrated priestly roles and graded Bible-
story books and catechisms. Both types of Christianity drew many
converts from the ranks of those who had aspired to the powerful
traditional medicines and cults, but who for a variety of social or
economic reasons had failed.
The Common Man's Dilemma: Ndibu
Ndibu, who at the turn of the century lived in Kingoyi, east of Mboko
Nsongo, felt called to joinLemba, the highest ranked n 'kisi. However
even with the patronage of his clan he could not afford it, given the
gradual shift of resources from kinship and ceremonial expenditures
to colonial taxes. A feeling of blockage led him to Christian baptism.
Ndibu's autobiography, written for Swedish missionary Laman,
6
describes beautifully the situation of a common man.
To begin with, Ndibu's marital status was complicated. In youth
with the help of his father he chose a girl as his bride. Since she was too
young to take immediately, he married an older woman first, again on
suggestion of his father. The older woman died soon after. At the same
time his father died, leaving behind two wives. Before the burial could
be held, his father's estate and the fate of the two women and their
children needed to be decided. Ndibu hoped to receive some money

5
from his father s estate, but was informed that not only had his father
left no money, but that the slave-women and children would become
his. Ndibu reluctantly accepted this proposition. His clansmen
comforted him that women would be better than money, which would
be spent leaving him with nothing. A woman could be of help and
solace.
The date for his father's burial feast was set. Mats, cloth, pieces,
and other supplies were gathered, and the tomb prepared. Many
friends came. Great quantities of gunpowder were consumed in salvos
to make it an "honorable burial." Several elephant tusk horn
ensembles performed to stimulate the dancing. The festival so
strengthened Ndibu that he was able that night to go to bed with one of
the women he inherited, and, as anticipated, she became pregnant.
Ndibu, on returning from drawing palm wine one day, found his
neighbors gathered in his court. They announced to him that he had
just become a father of twins, a girl and a boy. Under pressure from his
neighbors he called the priest of Funza to initiate him, his wife, and
their newborns to the twin cult of Funza. In the MbokoNsongo region
Funza is held to be a spirit in the bakisi class, the origin of all min'kisi.
Nzambi is the invisible source of unity in the universe; Funza is
Nzambi's visible and material side, patron of twins and the source of
the multiplicity of powers. Parents of twins when initiated become not
only priests of Funza but high priests of all the min'kisi, since their
twin children are special spirit-children.
All this was awesome to Ndibu, but he went along with the
initiation rituals. Each step of the initiation involved persons from the
family and elements of the natural universe, both being sacralized
with a song. The priest pronounced instructions for raising twins, such
as not to treat them unequally and always to remember that they were
spirit emissaries. And he stressed that Ndibu and his wife were the
parents of twins, as the song for crossroads indicated:
I am the source of Nsimba;
I am the source of Nzuzi;
I have given birth to Nsimba;
I have given birth to Nzuzi—
Easily, easily,
Like the cricket, like the grasshopper.
Ndibu took his responsibility very seriously, doing all he must to abide
by the codes of twin parenthood in Funza. But then disaster hit,
leaving him fearful and disillusioned. He wrote:
Look what happened. When the children began to grow up and

to stand, one of them took sick with a swollen stomach and
suddenly died. Oh! how sad. The little girl had been such a
beautiful child. But we prepared it for burial in a grass mat, and
dug a grave for it at the crossroads. As it was the female child
that died, we mounted a mukuta basket, took Lemba-Lemba
plants, and Nsanga-dinkonde and a Lubota tree and planted
them on the child's tomb. Then we prepared a mortuary statue,
assembled the medicines, and told the remaining child not to
become angry because of the death of its twin Nzuzi. In doing
this we hoped that the bakisi would leave our land alone, since
they might have thought that the prepared mukisi was con
veying a sign to seize parents of twins somewhere else. We also
distributed a token of palm wine But then in 1912 the
mother of the twins herself died. In that same year I decided to
be baptized.
In spelling out his motivation for baptism to Christianity, Ndibu
expresses less a religious conviction for his actions than an existential
desire to clarify his muddled life situation, to find meaning in his
sadness and relief from fear of bakisi that haunt parents of twins.
When one gives birth to twins, one makes association with all
the bakisi, and one should not fear a mukisi When the bakisi
call to go prepare them n'kisi medicine, you go. The instruction
is to join the bakisi of the earth with those of the water.
Only two of the bakisi refrain from killing parents of twins:
Lemba and Lumani. These alone. Lumani's priest has ade
quate drink and food and would never seek out parents of twins.
Lemba's owner, even if he were angry, would never harass a
parent of twins. But to get this privilege, parents of twins must
pay the extravagant sum of five or six pigs. Were this not the
case, people would initiate only to Lemba. Especially parents of
twins, for Lemba and Lumani are the only bakisi that do not
harass parents of twins.
Even after baptism, Ndibu considers Lemba to be the ideal ritual

solution for parents of twins. He blamed his lowly economic situation
for his inability to make it into Lemba.
There were no doubt many more individuals like Ndibu who saw
themselves alienated from the sources of power and protection offered
in the traditional religious system. Populistic Christianity offered by
the missionaries, backed up by some political protection and the hope
of education, held forth a solution to many in Ndibu's rank. Others

found this a new type of alienation and did something about it by
participating in the prophet movement that broke out in the Lower
Zaire in 1921.
Seeking Redemption in the Colonial Setting
Despite the drain on indigenous labor and resources of portage,

railroad construction, and colonial taxes, and despite the gradual
undermining of indigenous chiefship, the negative rhetoric of
missionaries toward indigenous religion in Lower Zaire, for nearly a
generation after the beginning of colonialism Lemba chapters con
tinued to control exchange circuits, to play a role in peacekeeping,
and to hold initiatory seances. In the thinking of the populace and
Lemba priests with whom I spoke, the sudden appearance of African
prophets, more than anything else, destroyed Lemba.
Luvuangu-Mampuya and Katula both walked to Nkamba and
Kiese near the Zaire River to see for themselves what Kimbangu and
the other prophets were doing. Min'kisi were being discarded en
masse. What impressed people most was that whereas the old
magicians and healers had only been able to heal, the new prophets
could "raise the dead." Although few Lemba priests joined the
prophets, many functions of the old, ceremonial organizations and of
chiefship were adopted by the prophetic communities.
Who were the prophets of 1921, and why did they appear just then?
This subject has prompted voluminous resarch; it must suffice here to
7
review a few of the explanations I have developed elsewhere. Of the
new roles of leadership that developed in Belgian colonialism, the
catechists who worked with European missionaries probably came
the closest to acceptance in the knowledge and power of the new
order. Their authority centered on access to the Bible, the sacred
source of European authority (smn'kisi, in effect), knowledge of
which, with the attendant skill of literacy, opened up new vistas.
Despite their access to the source of European religion, it had become
apparent to some catechists by the first decades of the century that
their authority merely served European colonial superiority. Espe
cially Protestant catechists, to whom full spiritual power had been
promised, were liable to growing disenchantment. Complaints began
to be heard about not being able to study beyond the second or third
grade, to study medicine in Europe, or to earn the same salary for
services rendered as the Europeans. The growing gap between

expectations and realizations produced myths of secret or hidden
knowledge kept from the Africans by the missionaries.
Kimbangu and his apostles, drawn from the ranks of disenchanted
catechists or catechist- aspirants, were perceived to have been granted
the true power of the Holy Spirit and the Bible, bypassing the
mediation of European missionaries. Even though Kimbangu and his
main followers were arrested after a few months of public ministry and
exiled for life, their revolution destroyed what remained of the
authority of the old political-religious institutions like Lemba. By
taking upon them some of the trappings of authority such as the staff
(nkawa, mvwala), and the redness, whiteness, ecstatic possession,
the power of purification, and the power to heal and raise the dead, the
prophets had laid claim to the symbols of the paralyzed traditional
offices of authority. A new era had begun.
Partii
The Rituals of Lemba:

Management of Reality
"Who can bring a clean thing out of an unclean? There is not one."
—Job in Job 14:4
"Those who suffer He rescues through suffering and teaches them by
the discipline of affliction."—Elihu in Job 36:15
Introduction
The present section examines Lemba's rituals and their role in

Lemba's influence over society and its material resource. The
authority of Lemba's adherents rested on the right of the most pure to
levy fines for moral transgressions. The high level of rhetorical skill
and influence of Lemba priests and priestesses created an effective
governing order over a network of markets, alliances, and trade
routes. The Lemba priesthood thus usually coincided with the bases
of economic and political power in the wider society and with the
possession of crucial knowledge. Songs, puns, legends, and rites
indicate that the authority wielded by Lemba was rooted in an
ideology of esoteric and mystical sources such as ancestors and
spirits. Practitioners of Lemba thus consolidated its governing
capability and coordinated public resources by manipulating ritual
symbols effectively. This is why a closer examination of Lemba
rituals, the task of this part of the book, is important for an
understanding of Lemba's overall impact on seventeenth- to early
twentieth-century Congo coast society.
The major source of evidence on theLemba rituals is a set of largely
unpublished indigenous texts, with supporting evidence coming from
ethnographies and the mute records of museum artifacts. These
textual, ethnographic, and artifactual—art historical—sources will be
presented in the form of four Lower-Congo, regional variants and a
composite profile of Lemba in the New World. Chapter 4 presents the
northern variant among the Kamba and the Yaa people living to the
right and the left of the Kwilu-Niari River valley in today's Republic
ofCongo. Chapter5 presents the eastern variant among the Lari north
of today's capital of the Congo, Brazzaville. Chapter 6 depicts the
central variant of Lemba, among the so-called Bwende and Nsundi
peoples of the Manianga region of Lower Zaire. Chapter 7 provides
the western variant from the inland Yombe and the coastal Vili and
Woyo, the former living largely in Zaire, the latter in Cabinda and
Congo. Chapter 8 presents the New-World extension of Lemba with
material from Brazil and Haiti. Unique characteristics and problems
in the source materials will be discussed at the opening of each
chapter; a more general methodological and theoretical critique of the
issues in analyzing rituals needs to be explored first.
95
96 RITUALS
One issue concerns the varied nature of the evidence and the
random manner in which it was collected. A set of old ethnological
questions may be asked: Are the variations spurious or significant?
What are their underlying determinants? Are the varied song texts,
ritual acts, and symbolic forms and combinations "free variations"
which reflect individual creativity within a latitude of more structured
limits? Or are they due to structural variations in social, economic,
possibly even ecological zones described in foregoing chapters on the
Lemba region?
The four regional variants are in part grounded in a quite self-
conscious differentiation made by the indigenous writers between
"schools" or"styles" ofLemba ritual. The author of Text 1 (northern
variant) speaks of a distinctive "Kamba" style which is contrasted to
a "N'tini a Mongo" style with which he is more familiar. Similar
allusions are made in the indigenous accounts of" Yaa" and "Lari"
Lemba rites which have, in the eyes of the authors, distinctive
features. These "ethnic" diacritica will be used where appropriate,
but cannot however be systematically applied. In fact they seem to
be used by authors outside their home areas, and these latter, such as
"Nsundi" and "Bwende," do not appear in the indigenous accounts,
nor do "Yombe," "Vili," or "Woyo." It is possible that these ethnic
designata, which I have critiqued in Chapter 2, were and are intro
duced by Europeans and are not part of the indigenous culture at all.
Or if they are, they pertain to distinctive Lemba zones or stylistic
characteristics.
Some stylistic variations not consciously announced by the indig
enous writers are however significant in terms of structural variations
in the societies of the Lemba region. For example, as will be
explained in more detail later, coastal Vili and Woyo Lemba shrines
reflect a less elaborate initiation rite and a less complex nkobe than is
found farther eastward; this is combined however with an extensive
backyard grove and sometimes a fixed "house" shrine. Such a gradual
transition from the portable nkobe to the backyard shrine may be
correlated with a gradual transition in local political structure from the
market governing committee without chiefs, historically, to the
presence of prominent local chiefs and even kings nearer the coast.
The western variant of Lemba seems warranted then because of the
unique set of symbols correlating with a specialized type of political
structure.
Other variations of importance in understanding social change are
not of a regional nature. All writers describing Lemba initiations
Introduction 97
meticulously detail the type and quantity of economic object ex

changed. Thus it is evident how the economic levels of the initiations
vary from the opulence of the Kamba area near copper mines and rich
agricultural lands to the relative poverty of the Lari area on the sandy
Teke plateau, and from dues paid in traditional goods to the use of
colonial tax currency in the Western variant Changes introduced
with colonial currency are not to be explained as mere regional
variations, they are part of the eroding exchange economy of the early
colonial period in which Lemba had by 1930 been destroyed.
Another important variation in Lemba ritual which is not regional
is the differential reference to widespread patron spirits such as earth
goddess Bunzi, dualist demigod Mahungu, trickster Moni-Mambu,
and numerous ancestral figures. It is not easy to account for
alternating uses of these spirits in Lemba since they exist throughout
much of the region in most local pantheons. Differential reference to
the one or the other may reflect a "sectarian" preference, or the
subordination of one figure to another.
To adequately capture variations bound both by regional con
siderations and those spanning regions or those having no particular
regional articulation requires an analytical approach which can
combine both local, concrete and more abstract issues. For this
reason the chapters of this section will each begin with a portrayal of
local rites based on textual accounts, but will then in the latter part of
each chapter pick up themes which may figure more widely such as
the social structure of clan alliance, the logic in Lemba medicines,
Lemba naming, or the characteristics of Lemba? s patron spirits.
Related to the variation of Lemba'% rituals is the issue of the
assessment of type and quality of data, and the determination of which
is the best analytic framework for the uneven assemblage of lyrics,
medicinal recipes, lists of rules, etiological myths, historical and
ethnographic interpretations, and artifacts. In other words, which
theoretical model best bridges all the textual, ethnographic, and art
historical (artifactual) data? An initial determination of types of
available data and their locations of origin was made in Figure 1 (see
also the essay on sources, below). Sources were divided into
"artifactual" objects derived from authentic Lemba settings
and events; "etiological texts" derived from authentic Lemba events
and explanations in KiKongo; and "ethnographies," that is attempts
to describe Lemba and to relate it to its social and physical environ
ment Despite the high quality of these sources, they are of varied
scope. The ethnographies of Lemba initiations range from those
98 RITUALS
which are event-specific, probably even based in a few instances on

eye-witness or participation, to those which describe norms or
customs or even ideals. Among the lyrics, some appear to be highly
original and individualized, whereas others, or parts of all of them,
appear to be standardized phrases which occur widely. In the case of
the artifacts, some are crude whereas others reflect great craftsman
ship. It is necessary to exercise a critique of quality which delineates
the type of data, its inner form as intended by the actor or as
interpreted by the analyst, as well as its quality as an aesthetic object
or performance.
These concerns for assessing the data and analyzing the various
kinds of evidence ofLemba ritual will be met through the use, in each
of the following chapters, of a limited number of expressive domains
drawn out of textual and artifactual evidence. These expressive
domains will for the moment be defined as areas or modes of behavior
with a high degree of consistency in form and meaning and a
considerable specificity in the vehicle used, because they are based
upon a cultural consensus or are inherent in the sensory capacity of all
actors. Expressive domains to be used in following chapters are: (a)
the spatial and temporal distribution of events in the seances, that is
the formal events structured by an apparent sense of the sacred; (b)
the exchanges of goods and symbols in these events, tied to the local
economy of subsistence production as well as the regional and
worldwide commercial economy; (c) the social organizational idioms
used in Lemba, often based on kinship, with special emphasis on
modes of achieving public order; (d) the sacramental objects
(min'kisi) composed for the seance and given meaning in the context
of the culture's classification system; (e) verbal categories of ritual
process as found in indigenous exegeses and descriptions of the
institution; and (f) lyrical scores of songs and etiological myths of
Lemba. This set of domains is not exhaustive. It excludes, for
example, dance and instrumentation, both known to have been
primary in Lemba but unavailable for analysis because of Lemba:*s
extinct status. Nor is this set of domains necessarily composed in the
only manner possible. Colors could have been dealt with separately,
1
as they have been by several authors. However, colors are often part
of a larger ritual code and have here been subsumed under the sacra
mental objects. The rationale for the present choice of expressive
domains lies both in the demand for a consistent analytic format in the
face of randomly composed data, and in the theoretical understanding
of how aritesuch asLemba's initiation generates meaning, affect, and
Introduction 99
social control. It will become clear that what I call expressive domains
provides Lemba actors with a set of related yet autonomous vehicles
of expression to deal with difficult issues and formidable contra
dictions. I shall illustrate this shortly.
The approach taken rests on several lines of analysis in the study of
symbolism, metaphor, and communication, and it is appropriate to
review these approaches very briefly. Proponents of the so-called
symbolist approach to the study of expressive behavior have empha
sized the many-layered "strands of meaning" in symbols and their
2
expression of "deep meaning" or "dominant" themes. In this view, a
culture's basic values and themes, as well as its major institutional
profiles, can be gleaned best through the study of ritual symbols
3 4
performed in "total events" such as cockfights, electoral campaigns,
5 6
divination seances, festivals, pilgrimages, healing and initiation
7
ordeals, and religious movements, to name a few. Criticism has been
leveled at this approach for allowing arbitrariness and mere des
cription to substitute for explanation of symbolic behavior. To
counter this criticism, some proponents of the approach have
emphasized that expressive symbols are always rooted in materialis
tic, sensorily observable areas such as the physiological, emotional,
8
affective, and economic realms of human experience. It is a strength
of this approach that, indeed, the sensory and emotional can be
combined with the cognitive realms of human life. These are
appropriate emphases for a study ofLemba which for three centuries
constituted a major north-bank institution and provided the context
for dominant symbols and social ideologies. The fact that these
symbols focused on illness and healing confirms the importance of a
theoretical view rooting symbols in materialistic considerations on
the one hand and in ideologies on the other.
Another scholarly tradition in the study of expressive domains has
emphasized the structure of metaphor. In this view the linkages and
associations between expressive domains are made the primary focus
in analysis. Modes of expression such as sculpture, literature, myth,
music, masking and the like are considered to each have their
inherent, autonomous characteristics. Analysis looks at the com
bined, "orchestrated" media, and the structure by which com
binations of media are joined. This line of analysis yields "root
9 10
metaphors" and "deep structures" which articulate common
principles across all expressive media. Although these metaphors and
structures resemble the deep or dominant symbols of the foregoing
approach, there is here a greater emphasis on understanding the
100 RITUALS
manner by which a metaphor achieves "movement," the way in which

one domain is associated with another or with the total social context
11
to achieve evocative and emotional power. In the study of Lemba 's
expressiveness this approach is of considerable help. While the
distinctive contribution of each mode of expression—be it song,
proverb, medicine, Active kinship, gift exchange, choreography—
must be understood in its own terms, all must be related to the
dominant metaphoric statement about Lemba alliances, economic
ties, and healing. There is in Lemba ritual an orchestrated quality by
which each "instrument"—each expressive domain—plays its
unique music but contributes to the overall harmony of the major
themes which are more than the sum of their parts.
The overall meaning of many levels of expression in a concerted
performance is grasped best by a third approach which combines the
sensitivities of the symbolists for the focal symbol and of the
structuralists for the relationship of expressive levels in metaphor.
12
This approach has been called "semiotics" by some, and "com
13
munications theory" by others; its varied proponents work in such a
diversity of ways, however, as to render the use of these labels almost
meaningless. Nevertheless, the analysis of multiple levels or modes of
expression as approached by these scholars proceeds in a helpful way,
for present purposes, by establishing a set of redundant messages
along numerous expressive channels that point toward "meta-
messages" or signification of greater generality than a single level or
mode, thereby addressing the condition of the whole culture and its
14
relationship to the natural world. In this approach the dichotomy
culture/nature, so much a tender nerve of contention in the whole field
of performance analysis, disappears before the set of messages that
range along the entire spectrum from genetic to kinesic to reflexive-
muscular to verbal to behavioral to the self-conscious dramatic. In
other words, there is no break between the cultural and the natural;
there are only numerous idioms or vehicles of performance and
communication, each with their sensory modes and their intangible
signification. Also, the analyst is not held to be a human deity who
possesses "etic" objectivity in the study of indigenous "emic"
subjectivity. Rather, each mode or domain of expression has its
"code" which permits the actor to work towards performance and
excellence, and which helps the analyst to achieve understanding.
Such codes do not, of course, fall ready made onto the researcher's
notebook. He must construct them if they are not evident in the
performance. In the analysis of performance in anthropology, folk-
Introduction 101
lore, drama, and criticism, and in social and behavioral sciences

generally, emphasis in deciphering codes has been placed either on
cultural genres with their own logic and meaning or upon universal
sensory-perceptive capacities. In the present study the codes of
expressive domains will provide insight into mostly cultural and
social rules and principles, although there is in some areas such as
reproduction and spatial/temporal ordering of events, as well as in the
therapeutic framework of the entire initiation, evidence of noncultural
constraints.
The basic units of analysis of Lemba's expressiveness in the
present study will then be (1) the domains—space and time in the
initiation rite's layout, exchange of goods and symbols, social
organization, medicine and the therapeutic, verbal categories of ritual
action, and the lyrical—(2) the domains' codes, that is the regularity
in terms of which variations make singular sense, and (3) the rela
tionships between domains, in particular in whole situations with a
clear context Some of the codes will be derived from the persistent
and widespread interpretations given in native exegesis to such
symbols as the Lemba drum or bracelet, or explanations derived from
verbal categories such as handa, "to initiate or consecrate a
medicine," and handa n'kisi, "to enter a cure in the consecration of
medicine." Other codes will be derived from recurring types of events
or acts at particular places in the space-time order of the seance, or
from the regularity of certain kinds of exchanges between leading
figures such as father and child, husband and wife, or the initiated and
the noninitiated.
Once such order has been determined within a domain, and its
explanatory code established, it will be possible to perceive the
metaphoric links between domains. For example, while the exchange
of goods, gestures, and words builds up a fabric of reciprocal
obligation between the neophyte, his kin, and Lemba, the lyrical
domain characterizes in song and declaration the heroes of the past or
spirits in whose name events and persons receive legitimation. Thus,
as the neophyte's kin receive an offering of pork stew and manioc
bread, the neophyte priest and his wife (or wives) receive from the
Lemba priesthood their symbols of authority; the metaphor (or
shifter) is the statement that "gods' food" is the medicine of the new
Lemba couple. The exchange of human food against offerings of trade
goods occurs contrapuntally, as it were, to the distribution of
medicines. The priestly couple's political authority is established
amongst the patrifilial children and the local public, while they rise in
102 RITUALS
the Lemba order. AU this activity in the social and economic domains
sets the stage for the reconstruction of reality in the lyrical domain.
Culture heroes such as earth goddess Bunzi, trickster Tsimona-
Mambu, androgynous demigod Mahungu, to name a few, span the
gap between a distinct Lemba consciousness and the conventional
religious culture of the region. In conventional narrations these heroes
are often entangled in dilemmas of human life, showing scenes well
known to students of African oral literature. Human tragedy is
frequently transformed into animal parody in which monstrous
animals, ghosts, or familiars take up the plot as if they are dancing out
a cleverly concealed psychoanalytic transference. In Lemba's etio
logical narratives, by contrast, these heroes avoid such traps and
entanglements, taking lengthy narrative bypasses to reach what is
heralded as "the Lemba solution." In effect, Lemba's ideology
attempts to resolve complications enacted in the conventional narra
tives. The problematic, thus blocked out of Lemba's lyrical domain
through possible resolution, is however brought back into considera
tion in other expressive domains such as the spatial ordering of séance
events, the ceremonial exchanges, and the composition of medicines
where action and object are better able to articulate contradictions
and to mediate symbolic resolution. Thus the methodology of
expressive domains permits us actually to see the alignment of
alternatives in a culture with the intent of bypassing implacable
problems.
The opemngLemba initiation variant from the Kamba (Chapter 4)
will offer the context in which to establish the analytic technique's
usefulness. This reference text is the longest and most detailed as well
as internally the most consistent of all accounts of aLemba rite at our
disposal. A short discussion ofLemba from the Yaa to the north of the
Kamba stands in sharp contrast to it.
The eastern variant, based on a number of texts and accounts
(Chapter 5), tests the method's capacity for identifying reportorial
errors and gaps, as contrasted to simple ethnographic variation. An
important element in this test is the use of the notion of "code" to
explain each domain. Additional variants of the rite will allow further
testing. For example, in the domain of spatial and temporal ordering
of séance events, it becomes clear that an inner logic requires
mediation of the village with the world beyond: the bush, forest, river,
and cemetery. This rhythm between the inner household world and
the outer world of "powers" is found everywhere, even in Haiti in the
New World. Accounts which omit it may be considered erroneous or
Introduction 103
partial. Elaboration of a domain's code thus lends the analysis its

most effective criticism of the quality of textual material.
The central variant (Chapter 6) permits development of the
analytic issue of the Lemba etiological text in relationship to non-
Lemba myths in the conventional Kongo-language oral literature.
Mahungu, the hero to whomLemba's origin is attributed in the central
region, is compared to other Mahungu myths.
The western variant (Chapter 7) extends the analysis of hero
mythology—the lyrical expressive domain—to another common
figure, trickster Tsimona-Mambu. Common structural features in
Mahungu and Moni-Mambu myths within and withoutLemba permit
generalizations about the "ideological" manipulation of conscious
ness in Lemba, from the "tragic" endings of the conventional
narratives on these figures to the "heroic" or "resolution" endings in
Lemba narratives. Because the vehicle of Lemba resolution of social
contradiction is the Lemba medicine, the drum and the bracelet, and
because of the high concentration of collected Lemba artifacts in the
western region, full analysis of the expressive domain of "medicine"
is deferred until the chapter on the western variant.
The New-World variant, concluding with Price-Mars' account of a
Lemba-Fetro seance in Haiti (Chapter 8), offers a final test of the
validity of codes discovered in the expressive domains. There is at
least a 150-year divergence between New-World and Old-World
Lemba expressiveness. Not only do the Lower-Congo ritual codes
and variants assist in the identification of authentic Lemba elements
in the Haitian variant, but also the converse is true. This latter
identification, which reflects an earlier level of Lemba, suggests the
presence of "proto-Lemba" structures, symbols, and metaphors at
the basis of both African and Afro-Caribbean culture. These features
exhibit a greater emphasis on fertility than on the trade and
mercantilism so prevalent in the late-nineteenth- and early-twentieth-
century Congo manifestations of Lemba.
Chapter 4
The Northern (Kamba, Yaa)

Variant of Lemba
"That which was a 'stitch' of pain has become the path to the
priesthood."—Kwamba, Text 1.94
Introduction to the Sources

The northern account of Lemba is based on two textual sources; the
1
first from "Kamba" country, by Kwamba Elie, an excellently
detailed rendering of an initiation and hearing, with insightful
comments on Lemba? § organization and its theory of power; the
2
second from "Yaa" country, by Lunungu Moise, a briefer report of
the rituals but with more extensive discussion of the pollution and
purification theories in Lemba. The two authors were members of the
group of sixty catechists who at the beginning of this century formed
the teaching staff of Swedish missionary-linguist-ethnologist Karl
Laman. At their workposts throughout north-bank Kongo they
collected responses to Laman's ethnographic questionnaire. The
Lemba texts are a small portion of the overall catechists' corpus of
3
23,000 pages.
Very little is known about the catechists' ethnographic instructions,
although the list of questions around which Laman's well-known
Uppsala series The Kongo I-IV was organized indicates that all texts
on Lemba come from answers in the section on "min'kisi" conse
crated medicines. It is also apparent that neither Kwamba nor
Lunungu are in their home territories, since in prefatory remarks not
translated here they distinguish Lemba as they are witnessing it with
Lemba "at home," or they state that it is just the same as " at home."
Nevertheless, their incorporation of technical "native" terms is
noteworthy, suggesting that they had no difficulty with the language.
105
106 RITUALS
The translation I offer is as literal as possible. Further exegesis and

commentary follow in the second part of the chapter in keeping with
the method of "expressive domains" outlined earlier.
The Lemba Seance Near Mboko Nsongo

(Among the Kamba)
Text 1
RECEIVING THE LEMBA C U R E FOR LEMBA ILLNESS
(1) Vo muntu una handa Lemba, buna una teka baka kubela
kwa ntulu ye kihemi mu ngudi ntulu sangama.
If a person would receive Lemba, first he must get a chest
cough, the stitch, or breathe with difficulty.
[(7) Nkianguna Lemba wazensila mu nitu andi mu mubedisa.]
For Lemba manifests itself in his body by making it ill.
(2) Buna yandi una fidisa mwana ye nsusu ye malamu kwa
nganga Lemba.
Then he must send a boy with chicken and wine to the Lemba
priest
(3) Nganga una vitula mpolo yena mu nsaba ye fidisa kwa
mubedo.
The priest will mix Mpolo [earth] into a small pot and send it to
the sufferer.
(4) Mubedo bu kanwini Mpolo Lemba bio kahodidi buna weka
Mwana ma Lemba ye nganga weka Tata ma Lemba.
When the sufferer has drunk the Mpolo Lemba and improves
he then becomes a Lemba Child and the priest a, Lemba
Father.
(5) Buna una fila nsusu zole kwa Mwana ma Lemba kasukula
mu nwa Mpolo kampodisila ye banzila mpe lumbu kiantula
Lemba mu nitu.
Then [the Father] sends two chickens to the Lemba Child
whose mouth was cleansed with the Mpolo, and instructs him
in considering a day for putting Lemba into his body [for
initiation].
The Northern Variant 107
(6) Kadi wonso wanwa Mpolo una handa; yandi kidi kabiala
mu mbongo.
For whoever drinks Mpolo will be initiated; he must be in
command of fortunes.
(8) Idiodio Mwana ma Lemba una kubama mu keba ngulu
zazingi, nsusu, mbizi bia ndia biankaka bialumbu kia
mpandulu mboko fila ntumwa kwa Tata ma Lemba kiza
kahandisa mwana.
Therefore the Lemba Child will begin assembling many pigs,
chickens, and other edible animals for the day of initiation.
Then he sends a messenger to the Lemba Father that he should
come initiate his Child.
T H E FIRST G A T H E R I N G OF THE PRIESTS
(9) Yandi Tata ma Lemba una bokila nganga zankaka bende

nandi mu handisa Lemba kwa mwana wambikisi.
The Lemba father then calls the other priests to help him
initiate to Lemba the named Child.
(10) Bu bizi bakatumisa buna una sola makambu manata kwa
mwana ma Lemba bonso buzolele Tata ma Lemba kwa
yandi mu mpandulu, mbo nanguna biabionsono bivwilu mu
nkisi wa Lemba ye nata.
When the invited come then he elects a delegate to send to the
Lemba Child expressing to him the Lemba Father's wishes
regarding the initiation; all who possess Lemba come and
bring it with them.
(11) Bu bizi tula kwa Mwana ma Lemba buna Tata una sika
ngoma ye mikonzi ku fula dia bula.
When those who will initiate the Lemba Child come, then the
Father sounds ngoma and mikonzi [drums] at the village
entrance.
(12) Ye makambu una kwenda kwa Mwana ma Lemba mu ta vo:
The messenger [from the priests' ranks] goes to the Lemba
Child and tells him:
(13) Bonga malamu ma Tat'aku ma Lemba ye ma mimbanda
wabakotisa ku lumbu Iwaku.
Fetch the wine for your Lemba Father and his wives and bring
it to them in your enclosure.
108 RITUALS
(14) Buna yandi una nanguna biabio ye hana kwa Tata ma

Lemba, mbo kotidi mu ngudi hata dia Mwana.
Then the Child will hold all up and present it to the Lemba
Father as he enters the court of the Child's village.
E N T R A N C E INTO C H I L D ' S C O U R T Y A R D ;
INSTRUCTIONS OF MPOLO C U R E
(15) Bu kakotidi una tangununa ebu:

When [the Lemba Father] has entered, he sings this song:
(16) Ko-ko-ko? Ko!
Wanunga Lemba wanunga? E—Lemba!
Will you gain Lembal Yes—Lembal
(17) Kiazinga Lemba, Kuyusaul Lemba
What Lemba gives, Lemba takes away;
Kiasa ntangu, Kuyusula ntangu.
What the sun gives, The sun takes away.
(18) Ndozi binunu kalotwa,
You may have dreamed of ancestors,
Milunga mianganga miakwangi kalotwa,
The copper bracelets of priests he has dreamed.
(19) Ngulu zanusinguku
Pigs to assure
Kima kiabwa mu mamba
That the thing [causing illness] may fall
into the water,
Katwizidi
That's why we've come.
(20) Ko-ko-ko? Ko!
Wanunga ngeye wanunga ? E—Lemba!
Will you win it? Yes—Lembal
Mboko una bonga nkunga ena:
Then [the Lemba Father] takes this song:
(21) Dibedi vwamva,
That which was the fawn,
Diankabi lolo.
Is today the adult nkabi antelope.
(22) Dibweni mulengo ma Lemba? -A-ma Lemba!

Have you encountered
difficulties with Lemba? A-a-ma Lembal
(23) Mbe kifwanga kiaseha
Oh! It is that which dies, which laughs
na bambwa.
with the dogs.
(24) Nieka tadulu dia kanda
Become guardian of
ma Lemba. A-a-ma Lembal
the Lemba clan. A-a-ma Lembal
(25) Bana Lemba kukandila
Children of Lemba, protect them;
Ku zumbu kinzinga;
And their forests and fields;
Bu kiahanda ma Lemba. A-a-ma Lemba!
Those who initiate to Lemba. A-a-ma Lembal
(26) Mbota tatu bungidi wo
Three mbota sticks you broke
Handa nasumika lusaba
Initiate, and I apply the {lusaba) medicine pot
ma Lemba. A-a-ma Lemba!
of Lemba. A-a-ma Lembal
(27) Nabonga mandala minatula
I took palm arches and installed them,
ye mandala;
And palm branch arches;
Mbwangi mu lusaba Iwa Lemba.
9
And put copper bracelets in Lemba s (lusaba) pot.
(28) Nayika Tata ma Lemba
I became a Lemba Father
ma Lemba! A-a-ma Lemba!
of Lembal A-a-ma Lembal
( 2 9 ) Tata ma Lemba bu keti sa bobo buna ha kimosiye Mwana
Lemba keti longuka ye banganga bumosi beti sa bu bed
sumika lusaba luanzensila makaya.
110 RITUALS
As the Lemba Father sings this he is together with the Lemba

Child who is learningfromthe priests when to apply the lusaba
pot and to cut plants into it.
(30) Mpimpa yoyo bana kina nkununu yanene nate ye bwisi bu
kiedi.
This same night all dance a big Nkununu dance that ends at the
break of day.
(31) Banganga bu bana swaka mwana ma Lemba, mboyambula
kina.
When the priests have washed the Lemba Child, the dancing
ends.
G O I N G TO A D I S T A N T P L A I N
(32) Lwaka mpila mosi, kwenda nseke nda mu vwota nsokia vo

mbongo zina mana mu moko maku;
Then they go to a distant plain to don nsokia grass skirts so
money will [not] be depleted in their hands;
(33) kansi butuku vwikidi mulunga wankwangi mu koko bubu
weka nganga yantela mu Lemba.
but that access to the copper bracelets will be forthcoming to
him who would become a priest in Lemba.
(34) Nkiangunu wonso una kuta kimpela buna ukusumuni mu
diodio Lemba kana kuhanina mbongo ye una mana yandi.
Whoever desecrates him will spoil Lemba because of this, and
must be charged a fine thereby furthering the neophyte in his
initiation.
NEOPHYTE'S D U E S PRESENTED; INSTRUCTIONS CONCERNING

PROFANATION; L E M B A " M A R R I A G E "
(35) Bu bavutukidi ha hata, buna makambu una kwenda ye

mwana ma Lemba ku lutengo mu mukamba biayenda na
Lemba vo muntu swakulu bonso ena: ngulu yamimbanda,
ngulu tanu zanganga zanatwa kwa Tata ma Lemba; ngulu
tatu zanene zambudulu ngungu ye Tat'aku ma Lemba.
When they return to the village, the priests' delegate takes the
Lemba Child aside and informs him that those initiating to
Lemba owe the following at this point a pig for the priests'
wives; five pigs for the Lemba priests—which are brought to
the Lemba Father—three pigs, large ones, to the Lemba

Father for the drumming of the mukonzi.
(36) Buna una hana ngulu zazonso bonso butudi makambu;
mboko honda ngulu ye mwangisa menga mu muzumba
kinzungidila madiedie kansi ngulu tatu zambulu nkungu
zina kebo zamoyo nateye nganga zazo zenzi, mboko honda
zo.
Then [the Child] gives all the pigs, as the delegate instructed
Then they kill the pigs and spread the blood in a circular trench.
But the three pigs of drumming the songs are kept alive until all
priests have spoken, then they are killed.
(37) Tata ma Lemba bu kahondidizo, buna una bonga menga ye
zunga ha bunsaba.
When the Lemba Father has killed them, he takes the blood of
an nsaba pot measure
(38) Mboko tula ku nima nzo ye ha mwelo.
and puts it behind the house and on the door.
(39) Mboko bana kwenda ye Mwana ma Lemba ha nima nzo mu
kaba ngulu zozo ye kunkamba vo: tala mpala nkute kimpela.
Then they go with the Lemba Child behind his house to
distribute these pigs and to instruct him in guarding
against strangers who would disrespect him or profane him.
(40) Buna una tomba mono yakubela makambu mu kufutisila dio
kidi watala bonso nganga Lemba kasilanga nkasu mu dia
mbongo za Lemba.
Then [instructs Ids Lemba Father] you will call me to show
you how to demand the fine, so you can see how the Lemba
9
priest exhibits vigor in collecting Lemba s money.
(42) Mukento bu kena mu mamba vo bakala una kulumuka ye
lembo kunkamba katomboka mu mamba kaluta, buna una
futa.
If a woman is at a stream, and a man descends and speaks to
her, and ascends together with her, he must pay.
(41) Nga ti ndonga yamwana buna lomba malamu kaka.
[For this instruction] a single "wine" is request from the ranks
of the neophyte.
(43) Mbo bana nwa malamu koko nima nzo ye mimbanda
miatata ye mwana ye sika minkonzi ye nkunga ena:
112 RITUALS
Then they drink this wine behind the house together with the
wives of Father and Child, and drum minkonzi for this song:
(44) Tata na Mwana basunda ko,
Father and Child will succeed in all,
of Lemba. A-a-of Lembal
(45) Kibedi kununika
He who perseveres
Malenge kieka,
Becomes successful before
Singu diatubisa Bunzi
The curse thrown by Bunzi
of Lemba. A-a-of Lembal
(46) Dibe kanga diesama
That which was a barren plot
kiekimpwaka ma Lemba. A-a-ma Lemba!
has become fertile—by Lemba. A-a-by Lembal
(47) Ndozi bata buyakwama
Dreams of Lemba Fathers
when they tortured me. A- a-of Lembal
(48) Ko-ko? Ko!
(49) Tata ma Lemba nwa malamu mankunga za Lemba.
The Lemba Father drinks the wine for the Lemba songs.
(50) Nata mikole mu hembo;
Wear the band over the shoulder,
(51) Mwana wabaka mbongo zalunga;
The Child found sufficient funds;
(52) Dieka bobombo,
And later he will gain
wiza wambindula.
access to Lemba.
(53)Nkundidi ku nsi,
I supplicated the earth,
Nkundidi ku zulu.
I supplicated the sky.
(54) Wanunga Lemba wanunga?
Will you gain Lemba?
Ko-ko bwe Lemba?
Ko-ko what is it with Lemba?
Tata ma Lemba vo:
The Lemba Father replies:
(55) Kubedi bunganga
That which was difficult
(56) Matondo ya ma Lemba, ma Lemba,
In gratitude from Lemba
(57) Weka kungwamu mutumbi mabinda
Has become a source of healing power, that
Nyiaka bana, ti nionzi zakula;
Heals the children,
makes nionzi fish grow big.
Bihambi vunza tolo ma Lemba.
Those who wait get a portion in Lemba.
(58) Mboko bonga milunga ye sakumuna mio.
Then [the Father] takes the bracelets and blesses them.
(59) Lwika mimbanda mia Mwana ma Lemba mu nsoko umosi
f
umosi ye hambana, bu kameni bieka Mwana ngang andi.
When he has consecrated his neophyte priest, he takes Child's
wives and each receives a copper bracelet one by one.
NEOPHYTE'S FIRST PROFANEMENT
(60) Lumbu biankaka bu biahioka, buna watubwa kimpela kwa

mpala mu diba, yandi Mwana ma Lemba buna ntumbu
tumisa Tata ma Lemba kiza kansonga bonso kana lomba
mbongo zanzensidi mu kimpela kwa Lemba.
Sometime later when he goes out and another person drops
something from a palmtree on him, the Lemba Child quickly
114 RITUALS
sends for his Lemba F ather to show how he must exact the fine
of profanement.
(61) Tata ma Lemba bu kizidi, una sika ye yimbila vo:
When the Lemba Father comes he drums up and sings this:
(62) Ko-ko-ko? Ko!
(63) Yebedi mfwenta yeka
He who was lazy, has become
mbukuni i mbumba
industrious—
mvamba ma Lemba! A-a-ma Lemba!
paradox of Lembal A-a-ofLembal
(64) Mukonzi nasika kwandi
The nkonzi drum is sounding
u-nkembo ma Lemba. A-a-ma Lemba!
for the festival of Lemba. A-a-ofLembal
(65) Kayika biyadi mfuba nzamba
He has become a harvester
of the nzamba field,
muna nzo ami kwandi
in my house
mubuyangi. A-a-ma Lemba!
in happiness. A-arma Lembal
(66) Mbo una lomba malamu mamunungusu Lemba kwa muntu
wa ta kimpela kwa Mwan'andi ma Lemba ye ngulu ya
nkonko Lemba ye buta kwa nkonko Lemba ye ngulu ya
binganini bia Lemba ye ngulu ya mimbanda.
Then [Father Lemba] requests the" wine of achieving Lemba"
from the one who profaned his Lemba Child, and a pig for
the nkonko Lemba drummer and elder of the nkonko, and a
pig for the Lemba host, and a pig for the priestesses.
(67) Mbokofuta ngudi nzonza mu muntu vo kumi dia ngulu ye
ngulu yanswekiya Tata ma Lemba buna muzita Lemba una
mona ye tambudila sungama.
Then [the debtor] pays the chief speaker a bondsperson or ten
pigs and to the Lemba Father a " secret pig" so that the Child's
Lemba statue will see and respond well.
(68) Mboko mwana ma Lemba una sumba ngulu yinajwa mitete
mitanu mu yandi mbongo yambundukila.
And the Lemba child pays a pig of five ntete baskets of raphia
cloth for his redemption.
(69) Bu kasumbidi ngulu tumisa Tata ma Lemba ye nganga
zankaka biza bamanisa Lemba ye yokila wo kuni ngudi
nkobeye kaka nzo mu mbiekolo yazimunina mu mpandulu
Lemba.
When he has the pig ready he summons his Lemba Father and
the other priests to come complete Lemba, dry the wood for
the Lemba box, and "secure" the house for the final conse
cration of the Lemba initiation.
ASSEMBLY OF PRIESTS FOR CLOSING RITE
(70) Nganga zazonsono bu zizidi bana bikula vo:

When all the priests are on hand they proclaim:
(71) Ko-ko-ko! Ko!
Wanunga Lemba wanunga? A—Lemba!
Will you gain Lemba? Yes—Lembal
(72) Dibe disu diatia
That which was a spark in the fire
Dieka nkama ye makumi
Has become a hundred and ten
ma Lemba! A-a-ma Lemba, wansila bo!
of Lembal A-a-ma Lemba, doitlike that!
(73) Wahanda Lemba, sika mukonzi,
If you would receive Lemba,
You must beat the mukonzi drum;
Wasa na mbongo ma Lemba. A-a-ma Lemba.
And pay Lemba's wealth. A-a-its wealth.
Wangula mu yoko mbau.
Cleansed in the fire.
(74) Mboko Mwana ma Lemba bongidi ngulu tatu, bahondidi
zankielolo.
116 RITUALS
Then the Lemba Child takes three pigs and they are
slaughtered.
(75) Mpimpa yamvimba yimbila nkunga vo:
All night long this song is sung:
(76) Wahanda Lemba
You who have received Lemba
Sunga tolo nge twangembo
Be alert like the bat
Muna bubu ma Lemba.
In the night of Lemba.
(77) A-a meka ngie maluboko-yongo,
Be evasive like the night-jar,
Mbo wayene ma Lemba.
So you will see Lemba.
(78) Mukento wakikundi nakutula
Put away your mistresses
Muniku nungu ma Lemba.
So you can keep access to Lemba.
(79) Mboko mu nsuka bana honda ngulu yina yaneneye kwanga
mankonde mamingi, biobio bina lamba kwa bala babutu mu
kanda diodio diaweti handa Lemba.
Then in the morning the big pig is killed, and lots of manioc
bread and plaintains are cooked by the patrifilial children of the
clan receiving Lemba.
(80) Bala bu beti sala bobo, buna banganga benzi nseke nda.
While the "children" are preparing the meal, the priests go to a
distant plain.
(81) Bu bena кипа bana ta kwa Mwana ma Lemba vo: bu tuna
vutuka mu bula, buna ngeye una leka mumbanda wa Tat'aku
ma Lemba, yeyandi una leka n'kento wawaku mpe.
As they go they say to the Lemba Child: When we return to
the village, you shall lay with the priestess-wife of your Lemba
Father, and he shall lay with your wife too.
(82) Buna mu diodio una kia meso mu lomba mbongo kwa muntu
utubidi kimpela ye ta sa mina mia Lemba bonso miena mu
ngolo kwa nganga.
In doing this your eyes will be opened to ways of requesting
goods from those who profane and transgress the laws of

Lemba that give the priest his power.
(83) Bu bameni kumutela bobo, mboko tota makaya mama:
mumbwangu-mbwangu, mundanda-nzila, ndimba, minkwi-
sa-mianseke, nuila-mwindu, mulolo. Momo mana kanga ha
lubongo mboko vutuka.
After this has been stated, the following plants are collected
and mixed: mumbwangu-mbwangu, mundanda-nzila, ndim
ba, minkwisa-mianseke, nuila-mwindu, mulolo. These are
wrapped in a raphia cloth and then they return.
(84) Bu bavutukidi ku nseke nda viokakana ku mpemba mu
bonga mukuyu wanduka mu nkobe mu bedi bakulu bandi,
buna bonga kanga (tadi dia butoto) ha bulu (zindiama).
On their return from the far plain they pass by a burial place to
fetch a wise n'kuyu spirit from amongst his ancestors by taking
a clod of earth from the grave; it is put in the Lemba box.
(85) Bu batukidi ku bikinda bobo, buna bana kwiza mu kaka nzo
mandala ye mandala—mbwangi
When they have returned from the cemetery they enclose the
house with many palm branch arches.
(86) Mboko bana ledika lukaya lumosi-lumosi mu momo
matukidi mau ku nseke nda momo mandala maba, ye mboko
fuka milele miamingi kunzunga ye kotisa mumbanda mia
Tata ma Lemba ye mia Mwana ma Lemba.
They take one from each plant from the plain and attach them
to the palm branch arches, and wrap many cloths around the
"lodge"; a wife of the Lemba Father and one of the Lemba Son
are brought into the lodge.
(87) Batuta makaya momo matukidi ku nseke nda mu matadi.
They then take the plants brought from the plain and mash
them together on a rock.
(88) Bu bameni tuta makaya momo buna Tata ma Lemba ye
Mwana ma Lemba ntumbu kota nganga zankaka zina sika
mikonzi ye minkunga ye makinu ku mbazi.
After the leaves are mashed Father and Son Lemba enter the
lodge quickly while the other priests are drumming mikonzi
and singing and dancing outside in the courtyard.
(89) Buna bau bole bana solana bakento baluku umosi umosi
mbo lekana bo.
118 RITUALS
The Father and Son have each chosen the most beautiful wife,
and have intercourse with them.
(90) Bu bameni ngiela yoyo bana kubasona namanga-manga,
kidi nganga zankaka zazaya vo Tata ma Lemba ye Mwana
ma Lemba makiolani.
When they have done this they make designs on the women so
the other priests will know that the Lemba Father and Son
have had coitus with them.
(91) Buna bana bonga bidiu bilembi bala kilambu ye kwenda ku
nima nzo mu kaba kio ye handa Mpolo Lemba.
They take the food cooked by the patrifilial children—the
kilambu meal—and go behind the house to distribute it and to
compose Mpolo Lemba medicine.
(92) Buna banganga bana kaba kilambu kiokio kwa bala batuka
mu kanda diodio babonsono.
The priests distribute the kilambu meal amongst all the patri
filial children of the [initiate's] clan.
(93) Kilambu bu kilungidi, buna bana sika mukonzi ena:
When the meal has sufficed, the "Children" drum this
mukonzi:
(94) Kuka-kuka [sound of drum]
Kibe kihemi
That which was a "stitch" pain
Lweka lukula banda manganga he.
Has become the path to the priesthood.
(95) A-a kieka kimakisa
A-a he has caused to rise
Ntangu ma Lemba.
The sun of Lemba.
(96) Kimjwila kiami kidba
My death has occurred
Na Tata ma Lemba
In the Lemba Father,
Ku mukula moyo ma Lemba.
Now there is life in Lemba.
(97) Mboko bonga makaya momo mabatutidi mu vula kia milele
ye hesona kanga dibe ku ndiamu soba hanionso mu bulu ku

nima nzo mu mamba ye malamu, mbo tula mu nsaba,
Then they take the plants pounded together in the village, take
them from the raphia cloth, and take grave earth, and mix all
this together with water and wine in an nsaba pot which is
(98) bunayeka Mpolo Lembayilenda Jidusu kwa wonso bwilulu
kwa Lemba, ye nsaba yoyo kukangila mu lutumbu koko
nima nzo ntangu zozo.
inserted in a hole behind the house. This prepared Mpolo
Lemba can then be sent to whoever is struck by Lemba.
(99) Diaka bana bonga nkobeye tula kanga diodio momo buna
dieka mukuyu.
Again they take up the box and put the sack in it thus binding
in the ancestor spirit.
(100) Tula tukula ye kolwa biamakaya mabedi ku nseke nda
matuta mimbanda mu ngudi nkobe;
Tukula red and the bundle of plants from the plain are
mashed by the Lemba wives and put in the center of the box;
(101) ye mfumbu zabundwa kwa nganga zazonsono zazi:
milengi, nsala zamoko ye nsalu, nkongi, mikenzi mianlele
miankulu mianzo muntu-muntu.
also a package of the following from each priest hair, finger
nails and toenails, eyebrows and pieces of old cloth from each
one's house.
(102) Mbo mimbanda mia Mwana ma Lemba mina kanga mizita
mu lukongolo Iwankobe mina kala dimbu kwa mulumi au
vo mumbanda sumukini mu miau, buna una tola mizita
kani miena bonso bumi a tudidi
Then the wives of the Lemba Child tie the statues together in
the circular box as a sign to their husband. If the wives profane
themselves the husband can inspect the statues to see if they
are still in place.
(103) Kansi vo muzita watengama buna wasumunu.
If a statue is leaning on the side, this indicates profanation.
(104) Mboko bana nanguna nkobe Lemba ye kwenda yo kotisa
mu ngudi nzo a yulu Iwa mbatu ye mbo nianika mikonzi
Iwandu ha mwelo nzo.
120 RITUALS
Then the Lemba box is ready and is placed inside the house
on a shelf; the nkonzi drum is placed over the door of the
house.
THE POWER OF LEMBA IN HEALING
(105) Lulendo Iwa Lemba Iwena mu bakisa muntu kubela mbo

mpe kumpodisa, kidika baka mbongo muyandibu kameni
kuntula mu nitu.
The power of Lemba manifests itself in seizing with sickness
an individual, also with healing him and taking his money
after it has been put into his body.
(106) Bilongo bina sadulwa mu mbukulu a mbwidulu a Lemba i
bina kaka bibasoka mu mbungu: bieka nkumbu Mpolo
Lemba.
The medicine used in healing those struck by Lemba is only
that which has been put in the cup: consecrated, it is called
"Mpolo Lemba."
(107) Nkiangunu nganga una nianguna nsaba ye katula bina
findambu ye nwika.
Therefore the priest will open the nsaba pot and remove a
small portion and give it to the patient to drink.
(108) Bu keti katula Mpolo, buna una tangumuna mpe ena:
When he removes the Mpolo, he will chant this:
(109) Kitukidi mu ntangu
That which comes from the sun
Kukatula ntangu;
The sun takes away;
Kitukidi mu ngonda
What comes from the moon
Kukatula ngonda.
The moon takes away.
(110) Tata ma Lemba
Father Lemba,
Wambuta, wandela.
He gendered me, he raised me.
(111) Zangata mansi
Praise the earth,
Wazangata mayulu.
Praise the sky.
(112) Kadi nayikidi
For I am enhanced.
Nyenda nseke-nseke
I have gone far.
Nanseke watwawalula
From far I brought it back.
(113) Ubajimba mu ndonga
Search in the ranks
Yabala babutu
Of the patrifilial children
Mu luvila lualu.
Of your clan.
(114) Ko-ko-ko? Ko!
Wanunga Lemba, wanunga? E—Lemba!
Will you gain Lemba? Yes—Lembal
(115) Yambula mumbedo waniaka,
Let go of the sufferer so he may be healed,
Kidi wakotisa mwanzole mbongo
He will bring the goods accordingly,
Fiwansakila nabangang'aku.
Thereby offering a gift to your priests.
cue) Mbo kana bonga nkulaye ndimba, kusa mu usoso zameso

ye kila mamoni mpe mu moko, kidi vumi kia kinganga
kaluta monika mu yandi.
Then he takes nkula and ndimba red and rubs it around his
eyes and in lines along his arms so that the respect due the
priesthood will be made manifest in him.
THE LAWS OF LEMBA
(117) Nga mina mia Lemba miena kwa nganga kaka ye

mimbanda miandl
The laws of Lemba apply only to the priest and his wives.
(118) Lemba una katuka mavumba, katuka ngolo tina ye lembo
kala diaka.
122 RITUALS —
But Lemba may lose its scent and lose its strength per
manently.
(119) Imu diambu diamizita miatulwa mu lukongolo Iwa nkobe
bukahandu vo kiokio kina kaka dimbu kiampandila ma
nitu kwa Lemba vo sumunu kwa muntu.
The figure-statues placed in the circular box at initiation are a
sign of the presence of Lemba in case of transgression of the
laws by the person.
(120) Ikiangunu, nganga kana sweka muntu umusumuni ye
lembo mu dia mbongo ko.
For this reason the priest cannot ignore the profanation and
refuse to take the fine payment due him.
(121) Vo mumbanda una sweka, buna mizita mitengamani
Lemba sumukina, tinini, weka kizekele, kifwa mbombo ye
makutu.
If the Lemba wife tries to hide a sin, the statues will
indicate Lemba's profanation, its loss of scent, and its
deafness.
(122) Diela dimeni mvu-mvu yikatuki mu yandi weka kifwanga.
Its spirit will have left it forever and the one associated with it
will die.
(123) Lemba uhandungu kwa bakala dimosi mboko bieka
mimbanda miandi mu keba mina mina bakila ulumi'au
mbongo mu Lemba.
Lemba is received by a single man and his wives are
consecrated to keep each of the laws so that their offspring
may be preserved in Lemba.
(124) Nga yandi wabiekwa Lemba kwa ngudi a nganga bu
kabwilu kwa Lemba kibeni kidikalunga mu kanda dia
mase mandi mu baka mbongo nkiangunu una wo zaba
nsadulu ye mbukulu ye mpandusulu ye mpandulu ye
bilongo kwa ngudi nganga mina tuka miomio miamio-
nsono.
He who is consecrated in Lemba by the high priest, when he
is possessed by Lemba he will go request money in his
father's clan in order to learn healing and initiating and the
initiation and the medicine before the appropriate high priest,
whence all this comes.
(125) Buna Mwana ma Lemba zebi bonso bwena nzila mu yandi
Lemba.
In this wise the Lemba Child will know the way of Lemba,
and its laws.
(126) Nganga Lemba kana mona mpene vo yanga nkento
nganga kol
A Lemba priest may not see a nude woman nor lay with
the wife of another.
(127) Kana dia mbongo zabalembolo kuntela kimpela ko!
He may not take fines other than those arising out of
profanation.
(128) Mutima ngulu kana dia Tata ma Lemba bu kakidimoyo
ko, kansi ufwiti kala wa Tata ma Lemba wampandisa.
The heart of the pig may not be eaten by any Lemba
Father, it must be the Lemba Father who initiated the
neophyte.
(129) Kalenda dia ntoba mpe.
He may not eat ntoba manioc stew either.
(130) Kalendi dia biabiomba ko.
He may not eat anything unclean.
(131) Kana lomba mbongo za Lemba ye ka lembolo nganga
ko.
N o one may ask for Lemba" s goods without being a
Lemba priest
(132) Vo sumukini kina kimosi mu biobio, buna una tumisa
nganga zankaka biza kaka nzo ye honda ngulu mu
udiangolokolo kwa Lemba.
If any of these laws are transgressed, then that one will call
other priests to erect a lodge and kill a pig to renew Lemba.
(133) Buna nganga veedidi ye vutukidi kala bonso kana ntete
waduna wakondwa bunkuta kwa Lemba diaka bu
kabanza vo una hondwa.
Thus the priest is purified and returns to the former state, not
fearing the wrath of Lemba as before thinking he would be
killed.
(134) Nganga Lemba ka sadilanga kinganga mu kwenda buki
yonso ntangu kazolele ko, kansi ubakilanga mbongo vo bete
dia malamu dinsotonini muntu wankaka, buna didizo kwa
yandi, vo titi kianguba, vo titi kia yaka, vo fikanga.
The Lemba priest does not use his priestly office to go heal
any time he wishes, rather he collects fines only if wine is
124 RITUALS
spilled on him by another—then he charges for it—or a

peanut shell, or a bit of manioc, or some earth.
(135) Ye wena nsisi bu ketizieta mu mavata malembolo Lemba.
He is greatly feared as he travels in villages without Lemba.
(136) Mukawa kaka kana nata ye nlunga wankwangi mu koko;
nkuutu ye nkalu yansekola malamu bina natwa kwa kilezi
kitalanga vo kololo ye bansumuni, buna yandi una kamba
vo nanguna mamba.
He will carry only a staff and a copper bracelet on his arm; the
satchel and a calabash for tapping wine will be carried by a
youth who watches that he is not made drunk or profaned, in
which case a fine will be imposed.
(137) Tata ma Lemba wena diswasani ye Mwana ma Lemba,
kadi Mwana ma Lemba kaka mu diedila bantu katubukidi
dio kansi Tata ma Lemba wena kundu ha ntandu a kingan-
ga kiandi.
The Lemba Father is different from the Lemba Child, for the
Lemba Child can use his intelligence on people only to heal,
whereas the Lemba Father has kundu power surpassing that
of the Child.
(138) Mu diodio Mwana ma Lemba una lembana kala detila ye
Tata ma Lemba mu vumi.
For this reason the Lemba Child will be unable to surpass his
Lemba Father in honor.
(139) Kadi Mwana lenda tina vo si kafukisa kanda diandimu dia
ye ngudi a nganga.
The Child can leave his clan if it is threatened with decimation
and "eat" with the high priest.
(140) Lemba kalendi zimbakana kanda diateka kala ngang'andi
ko nkiangunu kana bwila muntu vo katumbulu banzila mu
t
hingana mu kala nganga ko, mfumu a Lemba bu kafwa.

Lemba will not decimate a clan which already has one of its
priests, and it will not possess a person who has thoughts of
being a priest when a Lemba chief dies.
(141) Ikuma habedi nganga Lemba, nganga una hingana hoho,
kidi mbongo zasensisa Lemba mu kanda kazidi katuka mu
kanda diodio ye zimbala.
Therefore Lemba's priesthood is inherited with the wealth
given to Lemba. Lemba is then in that clan and it will not be
extinguished.
Lemba in Indo (Among the Yaa)
Text 2
THE LEMBA PURIFICATION
(1) Mpila ankaka a nkwedolo i mu Lemba, mbiekolo a bakento

ku nkisiaLemba mu lembwayanga. Mu nsiyayi mpe mwena
nkisi wau nate bwabu.
Another way of marrying is in Lemba, by consecrating women
who wish to be pure in nkisi Lemba. The nkisi continues in
this country till today.
(2) Buna tuka nsi ankulu, buna fu kiaki kiena mu mbiekolo yayi
ya ngiobosolo kwa banganga zihendi Lemba.
Since ancient times there is this custom to consecrate through
cleansing from priests initiated to Lemba.
(3) Vo bakala dizolele, vo bwabu nkento ami ka una vutuka
yanga ko, buna buketi mona beti handa Lemba bahaika ha
bwala, kadi bonso zeyi Lemba diena finhandila ku mfinda,
kansi mambu mamakeke, buna bakwiza mo maninga ha
bwala.
If the male wishes it, and if "now my wife desires to remain
pure," then it is considered necessary to compose Lemba
outside the village, for Lemba is composed partly in the forest,
and partly also composed on the return to the village.
(4) Ikuma Lemba buditotokele ha bwala. Mu handila, buna nga
nkento una wawana ye banganga vo bana sukula nkento
andi ye yandi beni, kidi kana vutuka yanga ko.
Lemba is brought out first in the village. To initiate, the woman
and her husband meet with priests to cleanse the woman and
himself, so there is no return to adultery.
(5) Nsukulu yena bonso bwabu. Banganga bana kanga luko-
ngolo (is. v. nkangulu yavandumuka) buna bana sukulu ye
zinganga bana kota nsi a lukongolo, mboki banganga bana
kubasikula mu mbwasulu (mbwangulu) anlangu mu bisafi.
The purification is like this. The priests create a circle (them
selves standing in line) for cleansing and the priests enter the
circle, then theyrinsethem off with a water-soaked sprinkler of
palm fronds.
126 RITUALS
(6) Tula nlangu mu longa mboki bondika saji biobio mu nlangu,

mboki banda mu nitu zau, nate tesobukilungidi bahaikidi
kwau.
They put water in a basin then they dip the frond in water, and
strike it on their bodies until it is sufficient, then they come out.
THE PADDLES OF PURIFICATION
(7) Mboki bobo basukulu, buna nganga una kuvana fxbaya

nkento ye bakala. Vo bakala wena ntalu a bakento baya,
buna babo bajwiti sukula ye yandi, kidi bakento bandi
kabana vutuka yanga ko. Wanga nganga uvana ntalu
bimbaya-mbaya bonso bwena ntalu bakento. Kadi mu
bimbaya-mbaya biobio momo mwena dimbu kianzaila vo
bwabu nkento umeni yanga.
When they have all been washed, the priest gives a small
wooden paddle to the woman and to the man. If the man has
several wives, then each one must be washed with him so they
will not lapse into adultery. The priest will distribute as many
of the small paddles as there are women. The meaning of these
paddles is that they are a sign that the woman will no longer
practice adultery.
(8) Lumbu kibabiekongo, vo sukulungu ikiokio, nganga
ubumbikanga makasu ha kati diabaya bonso bwabu mboki
vwidi nkento bu kamwani ti makasu momo masumukini ye
lomba buna ntumbu zaya vo nkento yengi.
On the day of the consecration after the washing, a priest ties a
s o
vine around the middle of the paddle y^T\ that the
owner of the woman when he sees
will know the woman has been violated.
(9) Buna vwidi nkento ntumbu tumisa zinganga mu kwiza mona
kwa tata (is.v., samanuka bakala dileleyandi). Ibobo bwena
bila kiabaya.
The woman's owner will quickly call the priests to come be seen
of the father (i.e., expose to the male her cloth). This is the
nature of the paddle.
(10) Bakala diayanga nkento wabieko ku Lembafwitifuta ntalu
beni: muntu mosi ye bibulu ye mbongo kwa vwidi nkento.
The man who has violated a woman consecrated in Lemba
pays a stiff fine: one person, animals, and money to the
woman's owner.
(11) Mboki yandi kibeni vwidi nkento Jwiti vana nkombo mosi
yibana dia zinganga. Mboki diodio bu dimeni buna nganga
una vutuka tula vofula diaka makasu mamona va baya.
Then the owner himself of the woman must pay one goat kid to
the priests. This is for the purpose of the priest returning to put a
new vine around the paddle.
(12) Ibobo bwena dimbu kiabaya diakento. Nga mu baya dia-
bakala, buna kadiena mu diambu diayanga ko, kansi mu
kundu diandi, kidi kana kwe lokingi bantu na mpimpa ko,
kadi wenda ku loka, buna fwa kaka.
This is the nature of the woman's paddle. The man's paddle is
not for indicating adultery, but on account of his mystical
power, so he will not curse people in the night; for whoever is
cursed, dies.
(13) Ikuma vo yandi kibeni bakalafwidi, buna mu baya diodio
dina ta kimbangi kadi vo bakento bandi banhondele bu bas-
wekele nsamu wu balekele ye bakala diankaka, kani vo
yandi yandi kibeni.
If the man himself dies, this wooden insignium will be a witness
as to whether it is his wives who killed him for hiding their
affairs with other men, or whether he himself was responsible.
(14) Bu bameni tala mabaya moso namabakento bandi, vo
nkento. Bu batedivo mabakento mandivo diankento ka disa-
mukini ko, kansi diayandi beni disamukiniye lombapinda,
buna kabalendi vutuka kwamisa nkento ko, vo ngeye didi
toko diaku, kansi buna kundi diandi dindidi.
They then investigate the paddles of his wives, or his wife. If
they see that his wives or wife did not commit profanation, but
he himself profaned and called for evil, they will not come back
and harass the woman alleging "you ate your man," for that
would mean she had "eaten" her companion.
PROHIBITIONS OF LEMBA
(15) NkentobukamenisukulukuLemba, kalendivutukasimbwa

kwa babakala ko. Ulenda kunsebuna nkasi andiye mwana
nkasi buko ye tat'andi.
A woman who has been cleansed in Lemba, may not be
touched by men, except of course she may be near her maternal
uncle, her nephew, and her father.
128 RITUALS
(16) Nga mu babakala bakimi, bunafuta kaka mu simba kuba-

simbidl
But when men show themselves off and they touch her, they
must pay.
(17) Nlele audi kasimba ko, nga bunafuta.
If they touch her clothes, they pay.
(18) Wasumbuka nti ukadieti, futa.
If they step over a log where she has walked, they pay.
(19) Kuhiole, vo kwena kini kiandi wahiokila ko, futa.
If they linger and do not withdraw where she dances, they pay.
(20) Kulendi kuntela mbila ko, nga futa.
If they stare at her for any reason, they pay.
(21) Kulendi kuntuba loko ko, nga bunafuta.
No injury may be spoken against her, or they will pay.
(22) Vo muntu weti bula (teta) nkandi kantimbudila sula futa
kaka.
If a person is breaking palm nuts and one ricochets and hits her,
that person shall pay.
(23) Mu nsi ankuluyakwaku (Indo) buna mfutuluyaluta kala mu
bantu bole bole mu futa, muntu ndiena sumuni nkento wasu-
kulu ku Lemba. Bantu nkatu, buna futa mbongo.
In earlier times in this region of Indo, payment of two people
was the main way of handling the profanation of a woman
cleansed in Lemba. If no person was available for payment,
money was used.
THE COMPOSITION OF N'KISI LEMBA
(24) Mpandulu a Lemba kizeyi ko vo yena diswasana ye

mpanduluyeyaBakongo. Kadimina miankaka, buna mpila
mosikwandi(is. v., ku betihandila muntu wankaka ka lendi
ko kwenda, mboki mfutulu ambongo zazingi zingi, buna
mpila mosi, kadi kundambu nsi eto buna difutulwa mbongo
zazingi mu handa).
I do not know whether the initiation to Lemba is different from
that of the Bakongo. There are some rules that are the same
(e.g., during initiation another person may not pass by, he will
be taxed much money; also the same as in our country, the ini
tiation itself costs much money).
(25) Mboki mu zingi, buna mpila mosi, kadi kwaku, buna zingila
beni bazingulanga.
With the ceremony, it is the same; here they simply dance
around together.
(26) Lemba mu nsi yayi, mbaduku, buna i nsukudulu.
Lemba in this country, initially, consisted of the cleansing.
(27) Mboki bukazole handa nkati handa, buna ukwendanga ku
mjinda ye nganga mu handa nkati nsalu. Kadi mpandulu a
Lemba yena bonso nkisi miankaka miaminene, A. v., teba
longo kimosi mboki yimbila nkunga. Mboki, longuka
mbukulu ye longuka ndwenga zankaka mu kiloka.
When they want to initiate to it, they simply go to the forest
and the priest does the composing. For the composition of
Lemba is like any other large nkisi: that is, teaching a rule,
then singing a song. Then, learning therapeutics and tech
niques for saying spells.
(28) Kadi wonso ukwendanga ku beti handila Lemba vo muntu
una kwenda koko, buna ntumbu kunhonda mu kiloka.
Whoever strays into the area where they are composing
Lemba, he will be killed by spells.
(29) Vo ka bazolele kunhonda, buna ntumbu kunsa zengi kiami-
onso-mionso nsamu tambudila kwandi, b.v., kina, tambu-
didi kwandi mu kinda, yimbila wayimbila kwandi, mboki
vova monso-monso mambu.
If they do not want to kill him, they create some special effects
to acknowledge him, such as dancing, receiving him into the
cemetery, singing, or speaking about any affair whatsoever.
(30) Mpila nkadulu yibikwangwa Mayingi, ka lauka nadede ko.
Nga kijwani kialauka. Nga lenda kwenda konso-konso mpe.
This treatment produces a state called Mayingi; if s not exactly
madness, although it resembles it, and a person can go any
which way from it
(31) Nga Lemba diandambu yayi balenda dio handila kwandi ha
bwala, kansi mu nkubu a nsusu bakwendanga ku mfinda
mu handa.
Lemba in this country can be composed in the village, but usu
ally at the first cock's crow they go to the forest to compose it.
(32) Nga bu beti handa ha bwala, bahanda beti handa kwau;
130 RITUALS
mboki bankaka mu dia ye nwa malafu mau ye kina kwau,

kadi mpandulu a Lemba yena lumbu biankungi.
When they compose it in the village, they just do it among
themselves, while others drink their wine and dance their
dances, for the initiation to Lemba is a matter of days of festi
vity.
(33) Nga nlunga wakinganga kalendi vwikulwa ha bwala ko, ku
mfinda; ye nkobe a Lemba yena bilongo ye mfusa ye betota
biankaka, buna kayinde lekulwa kuhendi dio ha bwala ko,
kansi ku mfinda bana dio tundila ha moko mandi
The bracelet of the priesthood may not be put on in the village,
it must be the forest; the Lemba chest bearing medicines,
chalk, and other assorted items are thought not to be validly
composed in the village but in the forest. In the forest the brace
lets are placed on [the priest's] arms.
(34) Mbokibateka kumbufanga benibonso bubazolele. Mbokibu
kana vutuka ku bwala, buna ku nsi a nlele kana vaikila ku
bwala. Ibobo bwena mankaka mampandulu a Lemba mu nsi
yayi.
They first dine mightily according to their desires. When they
return to the village, they are shrouded under a cloth; they
come out from under it in the village. This is how Lemba is
composed in this country.
The Expressive Domains and Their Codes
THE SPATIAL AND TEMPORAL DISTRIBUTION OF EVENTS
This domain is quite explicit in Kwamba's account of a Lemba

initiation. Turner and other scholars of Central African ritual have
noted the participants' tendency to express their ritual actions-
spatially. In North Kongo the reason for this is obvious, for the village
constitutes a cosmological space and an ideological unit lending
symbolic structure to all rituals held in its confines.
Kwamba's Lemba seance begins as the priests gather at the village
entrance (elsewhere it is a crossroads). Both places are auspicious
sites of ritual attention, for the burial of twins, or protective medicines,
the exchange of gifts, or the rendering of sacrifices. The village
entrance is the main opening to the "social body," and stands in
isomorphic relation to bodily openings. Thus when, in the village
entrance (fula dia bula), the Lemba novice receives medicines of

Lemba, "it enters his body" (Text 1.5), and after gifts are exchanged,
the priests "enter the center of the village" (1.14). Sixty years later
lemba-lemba plants used in the " opening" (bonzo) part of medicine in
Kongo are planted at the entrance to villages, markets, and near
houses associated with twins. The spatial zones demarcated in the
Lemba rite are clas'sificatory analogies to other aspects of the ritual
cosmology such as the human body, social relationships, plants, and
special words and names.
The village courtyard (ngudi hata, "mother," source, center; also
nkanu, court) is the site of several key phases of the rite, such as initial
instructions to the neophyte, an all-night dance, a ritual cleansing,
part of the Lemba "marriage" as well as its "consummation." The
North-Kongo village courtyard is a public place, even when a ritual
event occurs there. It is also thought of as a "male" place; the men's
house (mbongi) is found either in its center or at one end. Villages of
the region usually consist of two parallel and facing rows of houses. If
the village is on a ridge, as is often the case in this hilly region, the
parallel rows are straight. Otherwise, they may bulge to form a circle.
"Behind" the houses are cooking hearths and kitchen huts; this is
"female" space. Beyond that are gardens, domestic fruit trees, trash
pits, and drying racks, and then the bush or forest, considered "wild."
All these zones—entrance, courtyard, house, hearth, garden, b u s h -
become important m Lemba (see figures 5 and 6).
After the first large meeting in the courtyard on the first day, and the
dance of the night (1.15-31), the activities move out to a distant
savanna clearing (nseke nda). It is unlikely that Kwamba observed
these activities, since he devotes only a few lines (1.32-4) to what
must have taken much of the morning of the second day. Also his
explanation of the purpose of the savanna phase is mostly interpreta
tion, rather than observation. It is understandable that he should stay
away, if not a Lemba devotee. Leinba rhetoric stated that a profane
person who wandered near these sacred savanna or bush areas would
be beaten, fined, forcibly initiated, or even killed. Other accounts
suggest that what occurred within the Lemba group was the first
instructions—as Kwamba intimates—on how to act like a Lemba
priest. The first phase of the initiation was thus evenly divided
between village and savanna clearing, a clear spatial ceremonial
rhythm.
Following the priests' return from the savanna on the second day—
the celebrants having stayed up all night—the neophyte couple
132 RITUALS
Figure 5
Spatial zones of Lemba ritual
as described in Text 1 by Kwamba
ku-nimanzo, " b e h i n d " house,

h e a r t h , kitchen hut, food &
place of m e d i c i n e prepared here
hearth T
/ mpemba, \
f "the white"; *
zindiamu, tombs;
I bikinda, cemetery I
\ shrine of
receives its copper bracelets in a ceremony in the courtyard. The

neophyte or his patrons have paid the dues in pigs, which are now
slaughtered, and their blood is poured in a circular trench at the door
of the neophyte's house, approximately where the lodge will be built
later. Sometime later priests and the neophyte couple move to the
kitchen area behind the house, where more pigs are slaughtered. The
couple is instructed in the ways of guarding against defilement, and
the Lemba priestesses sing their song. The neophyte couple puts on
their newly consecrated bracelets, and they, as a couple, are blessed.
Pigs' blood from an nsaba pot is spread on the door of the house facing
The Northern Variant 13 3
Figure 6
Spatial and temporal organization of events
in Kwamba's account of Lemba inauguration.
Numbers refer to lines in Text 1.
night
instructions,
blessing of
Y "children"
prepare
distribution
of feast &
feast (79)
couple, women's medicines
song ( 3 7 - 5 9 )
medicines^
mjxedj87i (91-104)
instructions, dues paid, charging fine, ,
constructing dance Father/son
dance, all night (30) pigs killed, lodge ( 6 0 - 9 ) vigil wife
(14-29) bracelets opening of 1(75-8)1 exchange
cutting plants (29) w a s n i n g ( 3 1 ) given ( 3 5 - 6 ) second phase ( 7 0 - 3 ) (88-90)
entrance priests
(11-13) wearing of collect
grass skirt, plants
instructions (80-3)
(32-34)
priests
collect
tomb earth
(84-5)
the hearth. The slaughter of so many animals suggests a distribution or

feast at this point The first half of the rite is now completed.
The rite has moved full cycle, from village to savanna and back to
village. Within the village, the complementary opposition of court
yard and hearth has also been established, and in connection with the
latter opposition, that of male and female has been made isomorphic.
The "marriage" ceremony, the exchange of pigs for the bracelets, and
further slaughter of pigs for the consecration of the bracelets suggest a
sort of step-by-step movement through the ritual zones, a sacrifice or
exchange at each crucial place. From this first phase of the rite the
analyst must conclude that ritual here is anything but amorphous
antistructure. If anything, it is superstructure, in which all available
codes are made available for the creation of a cultural, cosmological
statement. This is further in evidence when Kwamba notes that the
priest, having been taught his prohibitions and consecrated, goes out
for an indeterminate period of time to be "profaned" at some
undetermined spot Daily life is profane, its placement is hardly
possible. Only when the neophyte priest is profaned—by a verbal
assault, a drop of spilled wine, etc.—and notifies his Lemba father is
the initiation rite reconvened at a determinate location and time.
The second phase of Lemba, concerned with restoring the new
priest's purity and showing him how to maintain it, is in broad lines
134 RITUALS
like the first phase in that the same spatial zones of courtyard, kitchen,
savanna are again covered. Several additional zones and groups are
added in the second phase. The patrifilial children—the neophyte's
matriclan's male members' children, collectively—occupy the site
behind the house where they take charge of the feast preparations, the
usual role of patrifilial children. Simultaneously, in the courtyard
before the neophyte's house, a palm arch (mandala) lodge is erected.
During this preparatory work, the Lemba priests march off with the
neophyte priest to fetch plants from the savanna and earth from a
tomb for the neophyte's shrine box (n'kobe Lemba). Kwamba is not
explicit about which cemetery is visited, again suggesting that he was
not present However, Kongoritualstructure would allow it only to be
his father's clan's cemetery, where he would be a patrifilial child
" priest," where he would be mediator, and where he would receive his
spiritual identity.
On the priests' return from savanna and cemetery, the front and
back of the neophyte's house become sites of accentuated symbolic
charging. Whereas previously these zones had been given "male" and
"female" charging, they now are made the poles of a series of
exchanges between the Lemba father and son, and the two clans they
represent, as well as between the son's clan and his patrifilial
children. Lemba presides over this double layer of patrifilial relation
ships centering on the new priest. First the Lemba father and son
"exchange" wives in the lodge before the house, that is, they have
coitus with them. (The symbolism of the semen of this exchange
coitus becomes clear in other regional variants, especially the eastern
variant dealt with in the next chapter.) The neophyte's patrifilial
children, meanwhile, are preparing the feast of real human food, while
the Lemba priests, acting a parallel role with regard to the medicines,
Active ritual "cooks" so to speak, are preparing for the neophyte
couple their Active "food," that is, their medicines. The two levels of
food symbolism are striking. Ordinary human food is gottenfromthe
perpetrator of profanement; sacred "food"-medicine is gotten from
the savanna and the ancestors. Similarly, the two levels may be seen
in the distribution. The medicines are prepared in the sacred
"kitchen" of Lemba, in an nsaba pot in a hole in the ground
somewhere in the hearth area. At the same time, the feast of human
food, prepared by the neophyte's children, is distributed by Lemba to
these same children, dependents and supporters of the neophyte's
clan, and to the general public.
The spatial and temporal coordinates of the rite have shown the
central role of the village/savanna, courtyard/kitchen dichotomies,
and the day/night/day temporal rhythm of the sessions. Within these

ritual axes, other social and cultural signifiers such as male/female,
father/son, and food/medicine are manipulated to achieve a full
statement of enormous condensation and power.
ECONOMIC AND EXCHANGE STRUCTURES IN THE RITE
The exchange of material goods such as manioc, pigs, copper brace

lets, and the like—items used in the regional trade—should be con
sidered under the same rubric as the "symbolic" exchanges such as
father and son's coitus with each others' wives. All goods, services,
and symbols exchange hands within one and the same social matrix.
There are not two economies, but rather a single social structure in
which numerous media get exchanged to create or maintain state
ments of value. In clearly orchestrated progression these media follow
exchanges betweenLemba son and father, Lemba and the public, the
Lemba son and his patrifilial children, and the like (see figure 7).
The first phase of the exchange, during the first meeting of a
day/night/day, appears to be a fairly straightforward exchange of
material goods from the neophyte-sufferer to the Lemba father for
ceremonial goods such as medicine, instructions, baths. The second
phase entails, as has just been shown, several levels of exchange of
various types of "food," including medicine as a sort of mystical food,
requiring a more nuanced reading of the ceremonial economy of
Lemba.
It is important to recognize that the neophyte does not pay more
than a small portion of the overall amount of goods circulated and
consumed, and that in the first phase in connection with receipt of the
medicine and the bracelets. The greatest portion, that which opens the
second phase, is paid by the "perpetrator of profanation" who comes
9
up with at least fifteen pigs, a considerable sum indicating Lemba s
influence in the Mboko Nsongo region even as late as 1915. It is,
furthermore, important to recognize that this sum is not consumed by
9
Lemba s priests in an exploitative fashion, but is redistributed
through Lemba, to the neophyte priest's patrifilial children and the
general public (1.92). As important as the volume of consumption
appears, in such a seance, it cannot be explained in terms of an easing
of the burden of growing biomass in an ecological niche, as New
4
Guinea "pigs for the ancestors" feasts have been explained. Lemba
exists in an open-market situation of considerable volume in trade, as
Chapter 2 has already shown. What then is the function of the
profanement tax and the redistribution so clearly described in
Kwamba's account?
Figure 7
Exchange structure of Lemba seance:
goods, services, symbols, persons
Receiving Lemba Cure
chicken and
neophyte wine Lemba Father
neophyte,
sufferer mpolo medicine Lemba Father
sufferer
become Lemba • 2 chickens Lemba Father
Son
Entrance of Lemba Priests into Neophyte's Courtyard
Lemba Father
neophyte - — wine and other
gifts
Lemba Father's
wives
Father's Entrance, Ali-night Dance, Instructions, Ritual Washing
instructions re
Lusaba pot, mpolo
neophyte Lemba priests
Gathering of Lemba on Savanna Hilltop
neophyte instructions Lemba priests
Dues Presented, Instructions, Dedication of Bracelets
a pig Lemba priestesses
neophyte
5 pigs • Lemba priests
3 pigs Lemba Father

(distributed behind
house)
neophyte instructions on - Lemba priests

collecting or Father
profanation fee
neophyte
consecrated bracelets - L e m b a Father
and wives
Figure 7 (cont'd)
First Profanation, Préparation for Conclusion to Séance
Lemba F a t h e r s
Lemba d r u m m e r ,
Lemba chief
perpetrator drummer
of profanation
L e m b a master %
of ceremonies
Lemba priestesses.
N
Lemba Fathers
^Chief orator.,
neophyte a "pig" of 5 raphia baskets • Lemba [?]
I preparation of L e m b a box \+—p, Lemba priests .

construction of lodge
patrifilial meat and other food items

children of of feast, and preparation
Lemba neophyte of feast
Closing Events of Séance
goods for feast

patrifilial _ prepared by
children children
plants from plain,

ancestral clods 1 - L e m b a priests-
Father's wife
neophyte . L e m b a Father
Son's (neophyte's) wife \~*** r
Kilambu feast . Lemba priests .

patrifilial
children
others?
plants, ancestral
neophyte L e m b a priests
spirit (patron),
become and
basket, d r u m
Lemba priest patrifilial
" c h i l d r e n , wives
mpolo medicine
138 RITUALS
Lemba's economic function here would be the stimulation of the

exchange structure over a local, perhaps "market," area, and the
generation or renewal of authority symbols within that context The
exchange network brought into play in this account is the same as
exists in a major marriage, funeral, or chiefly inauguration, with the
difference that Lemba's priestly couples from a wider area partici
pate. From a local region, perhaps a market area or two, these
prominent couples, themselves linking key clans, function as a local
elite network; therefore their ties as couples, and as a collective elite,
are renewed. The neophyte and his spouses are brought into this
collectivity during the ceremony. Of equal importance is the group of
the neophyte's patrifilial children. A s a collectivity they represent
dozens of localized groups, clans, who have potential reciprocal
relations to him and toLemba. The patrifilial children of the neophyte
represent a redundant set of ties already present to a degree inLemba.
Thus the fabric of regional alliances is charged, and given special
merit.
Where there is exchange, there is social cohesion. Distribution
generates authority. Thus the connection of the feast to the neophyte's
medicines is not hard to understand. The medicine, his nkobe as
symbol of authority, is a paraphrase of the exchange and the cluster of
obligations it establishes with the patrifilial children. In stimulating
distribution to the children, who represent, it must be emphasized,
dozens of neighboring clans, Lemba is merely using conventional
alliance-building mechanisms available in North-Kongo society. But
this alliance network is activated by the levying of a fine from some
profaner "out there." ThusLemôa's authority is made to extend over
the profane world, creating a fan of wider legitimacy than merely
those of marriage and patrifilial children. Lemba broadens the scope
of gifts a "father" can use for the beneficence of his "children," who
are also his supporters and allies.
SOCIAL ORGANIZATION! LEMBA'S CEREMONIAL KINSHIP
Since the overall social structural setting of Lemba has been

discussed in Chapter 2, what may be emphasized here are the unique
characteristics of the present variant. Little is said of the organization
of Lemba as a translocal entity, although it may be assumed that
someone was the master of ceremonies, the convener, and the
manager. Other variants develop these features. Kwamba's account
is most insightful in explaining how conventional kinship ties of father
to child and husband to wife were specifically shifted by Lemba to

maximize their translocal political integration.
The Father/Son (se/mwana) dyad in Lemba is defined as a
mystical hierarchy. The Lemba Father has greater kundu power than
his Lemba Son (1.137), and, reciprocally, the Son must give his
Father greater respect (vumi) (1.138). The son may heal, once he is
inaugurated, whereas the father may heal and govern, destroy
witches, and use his kundu mystical power to destroy individuals for
the well-being of a collectivity. But this hierarchy is a relative or
situational matter, since every son is also a father, as is the Lemba
neophyte to his patrifilial children. In certain conditions two indivi
duals and clans will be reciprocally patrifilial, to each other "father"
and " son," a situation that elicits much good humor and the ambiguity
out of which profound African rituals are made. It is, in the Lemba
seance as described by Kwamba, the structural logic behind the
convergence of ritual roles in the kitchen: the neophyte's children
prepare the feast; the Lemba priests, the medicines for the neophyte
couple. To understand the social structural possibility of this mutual
patrifiliality of two clans, closer attention must be paid to the Lemba
marriage, and its unique qualities.
Kwamba and other writers on Lemba all emphasize the extreme
morality of the Lemba household. Deleval in the Kavati-Mayombe,
Bastian in Loango, Lunungu among the Bayaka to the north, all
contrast the Lemba household's enduring permanence to the insta
bility of other types of households. The rules binding the Lemba
couple to a rigid code of purity, in Kwamba's account, are striking.
Lemba's laws apply specifically to priests and priestesses (1.117),
not to just anyone. A Lemba priest may not see a nude woman or
f
commit adultery (1.126). The figurines placed in the n kobe are a sign
of Lemba'$ power in the case of profanation (1.111). Further, the
stated purpose of receiving Lemba is so their offspring may be
"preserved in Lemba" (1.123). According to this set of prohibitions
and qualifiers, the Lemba marriage is a secondary, "super" marriage
singling out certain alliances between clans for special attention. A
clue as to which alliances so qualify for attention, which have this high
priority, is given in the presence of the two-tiered set of father/child
dyads in the Lemba seance: The Lemba Son and his initiatory
relationship to his Lemba Father, and this same son's relationship as
father to his patrifilial children. There is here a concern for the
maintenance of an unending patrifilial chain such as exists in the
permanent and enduring alliance between two clans in which every
140 RITUALS
generation's son marries back into father's clan, or in which there is

the mutual patrifiliality spoken of above. The emphasis on the patri
filial chain is present in the convergence of roles in the kitchen, where
Lemba representatives and the patrifilial children of the neophyte
each "cook": the latter, feasts; and the former, medicines. Elsewhere
we will see these two categories of preparations blending together
even more. But the kingpin of the system, as it were, is the singling out
from the neophyte's polygynous household one wife for special
attention—one who is "separated" (1.59)—presumably one who has
the right alliance for Lemba's emphasis and the ritual exchange of
wives between the Lemba father and son (1.89). This exaggerated
9
act—Lemba s form of "royal incest"—highlights the great preference
in the Lemba worldview for patrifilial exchange, colloquially known
as "blood reciprocity." The specific form of marriage that most
effectively embodies this preference is the patrilateral cross-cousin
marriage, already described in Chapter 2. Linking isolated villages
and markets, reinforcing the patrifilial chain across more than two
generations, it establishes a permanent relationship of complimentary
reciprocity between exogamous groups. The father/son wife ex
change commemorates the convergence of the two classes of relation
ships that allow a segmentary society to function effectively: (1) the
father/son tie as the model of hierarchic relations; and (2) the
extension of this relationship to the corporate level in which two
exogamous groups of equal status exchange women, that is the
husband/wife relationship.
SACRED MEDICINES
The Lemba medicines, called min'kisi because they are consecrated,

consist of the bracelets worn by the priest and priestess(es), the drum
or drums that are used in the couple's inauguration to Lemba and kept
on their house wall, perhaps as an identifier, and the large n'kobe, a
cylindrical bark box approximately forty centimeters tall, within
which is kept a variety of symbolic substances collected during the
initiation ceremony and used later either for purification or for the
Lemba priest's own future healing activities. The ingredients of these
"medicines" are collected in much the same way as the exchange
items discussed above—food, livestock, drink—and passed along the
same social channels. The two sets of objects are distinguished by the
fact that the one—the exchange items—is assembled, distributed and
consumed, whereas the other—the medicines—is assembled and
kept, as a sacred "documentation" of the ceremony. Exchange items
are for the most part in the charge of the patrifilial children, resulting in
a huge feast; the medicines are in the charge of priests, following a
course roughly epiphenomenal of the exchange. Both, therefore,
stand for human relationships. The food brings people together, the
medicines remind them of their common lot in roles and groups.
The difficulty with the analysis of the medicine objects is that their
meaning at the time of ritual use was filled with multiple connotations.
By contrast, verbal accounts of 1915 may be approximately under
stood in light of current KiKongo usage, which has not changed
beyond recognition. Social structure and exchange patterns, simi
larly, have remained quite constant. But the medicines of the early
twentieth century have gone out of fashion, and few if any individuals
can give reliable exegetical account of their use sixty years ago. Thus,
although Kwamba faithfully relates the collection of ingredients for
the medicines of Lemba in 1915, he hardly hints at their meaning to
the initiands.
The analytic strategy to be employed on the medicines of Lemba
will be to compare slightly diverging accounts with the help of what
exegetical and contextual meaning can be gotten. As in the other
expressive domains, additional variants will allow insight into the
cultural structure that gave rise to the particular cases. Kwamba's
account of Lemba medicine will be given only a contextual examin
ation here; further interpretation will be possible in light of other
variants examined later in the work, especially in Chapter 7 on the
western variant.
The basic Lemba medicine is the mpolo-Lemba a small pot of
9
chalk used by the Lemba priest to treat or purify the Lemba sufferer.
When the treatment is considered effective, a father/child relation
ship is opened that becomes the Lemba patrifilial relationship. One of
the ingredients of the n 'kobe Lemba box is the neophyte's own mpolo
medicine. It should be noted that Lemba healing or initiatory seances
are accompanied by drumming (for example, 1.11) or singing and
dancing. As in most Kongo ceremony, these expressions transport the
ritual action to another plane: drums are the voice of ancestors; the
songs articulate the medicine's significance (for instance, Kwamba's
mpolo song 1.108-15).
The copper bracelets, mentioned repeatedly in songs early in the
ceremony's first phase (1.18, 33), are given to the neophyte couple at
the end of the first phase (1.58). These bracelets, made everywhere
especially for the individual priests and priestesses, when donned give
them their first public insignia of their new rank. The bracelets seem to
142 RITUALS
be the culminating symbol of the first phase of the initiation, and

ceremonies leading up to the actual blessing and donning of the copper
arm rings mention them as the artifact for either showing moral purity
(1.32-4, 1.135-6) or of exercising authority in collecting fines. The
bracelets are also used to differentiate the Lemba wife or wives from
other wives, in a polygynous household (1.59). The copper bracelets
constructed of red ore from Mboko Nsongo and Mindouli are not only
beautiful in color, but also their resonant tone, when striking other
metals, contributes much to the aesthetic appeal they had and to the
collective dances reported everywhere in Lemba. A dancing group of
Lemba priests and priestesses would have sounded like a chorus of
voices combining soft chimes and small double-tone drums. At some
point early in the initiatory seance—Kwamba does not specify it—the
neophyte couple receives its small nkonzi drum. This too is an object
of prestige and beauty; it is displayed over the door of their house.
However the major part of Lemba medicine is the n'kobe box and
its ingredients. The ingredients are collected throughout the second
phase of the ceremony, that is after the neophyte has been "profaned"
and is being led into full membership. Five sorts of savanna plants and
earth from an ancestors tomb are brought into the ritual events of
phase two of the rite. In a raphia cloth bag they are hung on the lodge
during the "wife exchange," then they are taken to the "kitchen"
beyond the house, prepared with mortar and pestle, and mixed in a pot
(1.83, 86, 87, 97) with other ingredients such as tukula red (100).
The apparent contextual meaning of these plants is their movement
through or across all the zones of Lemba ritual attention: savanna,
courtyard, hearth. In much North-Kongo ritual plant use, however,
significance is lent the particular plant through a pun on its name, such
that a verb fitting its name gives it an action appropriate to the ritual
situation. Thus mundanda nzila, "path follower," might be used
because the symbolism of paths is appropriate. But unfortunately,
Kwamba's text is silent about such plant-name meanings, and they
cannot usually be reconstructed because they are not pancultural in
Kongo territory. They are like proverbs which have an opaque
meaning until the particular intention is known or explained. The
meaning of tomb earth and tukula bark powder in this context is not as
likely to vary from the conventional understanding. Tomb earth, tobe
(when mixed with wine), conveys purity and spiritual continuity.
Tukula, a red ochre, conveys transition and mediation with the
beyond or across social roles. Water, palmwine, tomb earth, and
tukula go into the neophyte's new mpolo medicine pot, stored in the
n'kobe. The satchel of plants as well enters the n'kobe, perhaps as a

memento of the ceremony of initiation. Another satchel filled with
hair, eyebrows, nail parings, and old clothing rags of the Lemba
brotherhood and sisterhood also enters the n 'kobe (1.101), perhaps as
a sign of collective identity in which the neophyte couple shares.
Finally, the statues of the wives, bound together and inserted also into
the n'kobe, are given the significance of a sort of moral watch or
reflector (1.102). The bundle if upright denotes purity, if toppled over
denotes profanement These bear some relationship to the pro
hibitions binding on the Lemba wives not to commit adultery.
Much is left unsaid in Kwamba's treatment of the medicines. Later
chapters will deal more fully with this part of Lemba.
VERBAL CATEGORIES OF RITUAL ACTION
The two texts reveal a clear and consistent language of ritual action,
both defining in standard Kongo cultural vocabulary the manner of
charging and using symbols, as well as indicating the structure of
ritual space which results from the manipulation of charged ritual
symbols. This vocabulary varies somewhat from account to account,
and region to region, but, as in the verbal punning on plants, the
significance of patterns remains consistent.
The key verbal category of ritual action in Lemba is handa,
meaning variously to initiate, to compose, or to identify a person with
the consecrated medicine designated by his treatment or by the
diviner's judgment. Handa appears in phrases such as "handa
Lemba" (1.1; 2.2-3), that is to " compose Lemba" as well as in the
form of initiating someone to Lemba, handisa (1.9). This distinction
of composing the medicine and initiating the neophyte also is carried
over into the nominative forms of mpandulu (1.69,1.124; 2.24) and
mpandusulu (1.124). Another major category of ritual action which
appears in the first two texts is bieka, to consecrate, as applied to
either persons taking up special commissions or ritual objects being
activated (1.69,1.106,1.123; 2.1-2). Another well-known category
of ritual action appearing in these accounts is that of loka, the
utterance of power words, spells, curses, and the like, for both
beneficial and malevolent purposes (2.12,2.27-8). Less well-known
categories of ritual action include the following:" securing, protecting
the house with palm frond arches," kaka nzo ye mándala (1.69,
1.132); "tying a knot," either in a literal sense around the ritual object
representing the purity of the Lemba priestesses (bumba makasu,
144 RITUALS
tula makasu, 2.8), or in a less direct sense of appropriating power to

obtain something for Lemba (sila nkasu, 1.40), or even in the
abstract sense of the Lemba community surrounding the neophyte
couple during their ritual bath to "absorb" them into their purity
(kanga lukongolo, 2.5). A comparable ritual action is the process of
"opening the eyes" of the neophyte to give him greater effectiveness in
his mission (kia meso, 1.82).
These general ritual acts, labelled by standard phrases, may be
9
called acts of e&ectingLemba s— and Kongo ritual's—•more nuanced
states of being, again both positive and negative. In the course of the
descriptions of Lemba by Kwamba and Lunungu, terms are used to
identify the positive ritual actions oipurification the end goal of which
is the achievement of power. Alternatively, through profanement
power may be lost, as well as through the violation of clearly-defined
prohibitions. This structure of ritual space, made of both actions and
states, permits a host of particular incidents and experiences to be
given meaning under one rubric: Lemba control of reality.
Positive ritual action clusters around the healing and purifying of
the neophyte and his wife or wives. The initial mpolo medicine " sucks
out" (hola, 1.5, 1.105) impurities, the dirt of his disease, from his
body, and "heals" (mbukulu, 1.124) the neophyte. Throughout
North Kongo the cupping horn, n'kisi mpodi, exercises this function
in a less collective manner. Later in the initiation the neophyte and his
wife or wives are "bathed" (swaka, 1.30; sukula, 2.4-6, 2.15;
ngiobosolo, from yobila, 2.2), another ritual practice with wide
spread therapeutic connotations in Kongo medicine. Later when the
neophyte couple experiences profanement, payment of the fine ren
ders them "redeemed" (bundikila, 1.68). the experimental profana
tion and the related repurification described by Kwamba, brings the
neophyte again to the purity of his original initiation (1.133). This
striving after the ideal of purity (vedisa 1.133)
9 implies a state of the
pure (veela, 1.133). The Lemba priest is taught how to seek the
contrast between purity and its opposite; purity opens the path to
power, whereas profanation is the loss of power.
Profanement terms are numerous in these Lemba accounts, begin
ning with the general termsumukini, orsamukini (1.34; 2.14) which
Christian missionaries translated as "sin" (masumu). FortheLemba
wife profanement is defined in terms of sexual congress with other
men, or their inclinations to approach her with sexual or flirtatious
insinuations (yanga nkento wabieka ku Lemba, 2.10, literally,
"have congress with a Lemba-consecrated woman"). The wife's
profanation is believed to be signalled by the "loosening" of the vine

tied around the "purity paddle" (2.8) or by the upsetting or undoing of
the statues in the n'kobe shrine representing the wives or the couple
(1.102-3, 1.119). The husband's profanation is more inclined to
come as a result of the misuse of his ritual powers for private gain
(1.127) or of his mystical powers to harm others (lomba pinda, 2.10,
literally, "call up darkness"; lokinga bantu na mpimpa, 2.12,
literally, "curse people in the night"), whether intentionally or
unintentionally through his superior mystical powers (kundu, 2.12).
Profanement is not something the priest can ignore or disregard. It
may strike him unawares and destroy his relationship to his wives and
their clans (1.122) or even kill them (1.22; 2.13), and, in any case,
unless it is attended to through propitiatory sacrifices and payments, it
diminishes the Lemba couple's power (1.118; 2.13). Similarly, the
priest and priestess must continually be on guard against inadvertent
profanement by those around them who "throw at them something or
speak foul of them" (ta kimpela, 1.60, 1.134, 1.13). Their life is
described as a struggle to maintain purity in the face of profaning
situations; their goal is to strive for ascetic power released through
purity.
Explicit rules binding on the Lemba couple articulate the boun
daries of the moral-legal and ethical code, just as the visible
mementos—purity paddles, statues in the shrine—express profane
ment. Such concrete embodiments are necessary where the thing
expressed is as abstract as purity and profanement. The figures "by
their life" indicate the extent of profanement of the priestly couple; so
the rules (mina, 1.117-31; 2.15-23, also prohibitions, laws, codes)
sketch ethical constraints governing numerous public relationships in
the Lemba order and across the society at large. Numerous rules
govern male-female relationships, the backbone of the Lemba-
buttressed network of the region (1.126; 2.15-22). Others regulate
the governing or "taxing" function of Lemba, forbidding anyone
other than an authorized Lemba priest to take Lemba fines (1.131)
and conversely forbidding the Lemba priest from taking money other
than that associated with his Lemba prerogatives (1.127). Some
rules govern strictly internal Lemba relationships and postures, such
as those forbidding the eating of unclean food (1.129-30), and those,
again often articulated by food prescriptions, defining the father-son
relationship (1.128). The purpose of the rules, suggests Kwamba, is
to give the priest and priestess their power, to define their prerogative
in society.
146 RITUALS
The power-terms in Lemba theory are perhaps more abstract than

the other sectors of the ritual structure. They are not embedded in
concrete action, but used to spell out a quality, an idea, or a state that
links the subject to mystical sources. Thus, in delineating laws
Lemba's power (here ngolo, 1.82) shows a social referent. In afflic
tion and healing (here lulendo, 1.105) it has a mystical and physical
referent. In the use of cosmetic decorations such as red earth color it
takes on an aesthetic referent, making manifest the priest's " glory and
prestige" (yumi, 1.116). Redness also shows an element of the
mystical in that it refers to the threshold between the visible world and
the other world of invisible spiritual powers. Finally, the power of
Lemba is defined in distinguishing the mystical authority (kundu,
1.37;2.12) and the prestige (yumi, 1.138) of the Lemba Father and
Son, the former always possessing more than the latter, in the unend
ing chain that makes every Son a F ather, with his own neophyte- Son.
The verbal categories enunciated here portray the social nature of
power in Lemba, and by extension much of the Congo Basin. The
terms are dynamically related so that a transformational process is
apparent, moving the subject Purification and power are positive;
profanement and prohibition, negative. Purification and profanement
are categories of action, whereas power and prohibition are attributes
or qualities, states. The rites of purification remove the neophyte from
the ranks of the common, profane, and "open" him and his wife to
initiation. Purity is the ritual action requisite to power. The right to
levy fines, learned in conjunction with the "experimental" profana
tion, tests the neophyte's ability to seek purity. But the power is
concretized best in the porhibitions. Violation of prohibitions put the
priest(ess) into contact with the polluted, profane world, weakening
the priest(ess) and making it necessary to go through purification ritu
als again to regain strength and power.
THE LYRICAL AND ITS MESSAGE
The "lyrical" combines verbal narrative with musical performance.

Nowhere better than in its songs does Lemba* s style or sense of
special reality emerge more effectively than in the initiatory seances.
Kwamba's account gives songs by the Lemba Father to several
initiatory groups, by the wives of Lemba Father and Son, by the
Lemba priests, by the patrifilial children on the day of the feast, and
by the neophyte Lemba priest after he begins his practice. The lyrical
domain has a strikingly consistent structure, which facilitates
analysis. Each song is announced or "framed" by being "drummed

up" with the small hand-held, two-tone drum (n'konko,n'konzi): the
call "Ko-ko-ko!" (sound of drum), which is answered by "ko." A
phrase call such as "Will you gain LembaV, and the reply, "E-
LembaV introduces the lyrics proper and closes them. The body of
the song consists of a series of couplets that may be repeated in sets of
two or three for hours on end. Kwamba's version gives these songs
only once, but in performance they must have been repeated each
twenty to thirty times over. For example, the song sung "all night" is
only two phrases long. There appear not to have been set Lemba
songs across the entire region; rather, local variants develop com
parable themes such as the neophyte's illness, his progress through
the initiation, Lemba's strength, the joy of the patrifilial children, and
so on. Common images and symbols such as the bracelets, the role of
certain natural objects used in the medicines, recur throughout the
area.
The lyrical couplets which are developed between the opening and
closing frames invariably combine images in one of several ways. One
common association creates the substitutable referent in an action
setting, as in this phrase:
What Lemba gives, Lemba takes away.
What the sun gives, the sun takes away. (1.17)
Lemba's action is defined by substituting the sun in its place. The
sun's action of rising and setting, perhaps emitting or controlling light
and heat, is made comparable to Lemba's action, perhaps that of
creating and controlling power. The couplet may also state a sort of
"transitive" association, in which the substitution of terms suggests a
fusion or identity of the two objects, as in this phrase:
The neophyte dreamed of ancestors,
The neophyte dreamed of copper bracelets. (1.16)
Ancestors and copper bracelets are brought into a relationship of
identity, or identity representation. A third type of lyrical combina
tion may focus upon some process or observation in nature suggesting
a transformation, as in this phrase:
That which was the fawn,
Has become the grown antelope. (1.21)
The transformation of fawn to grown antelope refers here to some
process within Lemba.
148 RITUALS
A s already suggested in the introduction to this part of the book,

these lyrics may be studied both as symbol and metaphor. Symbols
are seen in the nouns—neophyte, ancestor, bracelet, sun, fawn/
antelope—as well as in the verbal actions—dreamed, give/take.
Symbols are distinguished as perceptible to the senses at one level,
and at another level conceptually or ideologically charged. In the
Lemba lyrics natural and human objects, things, are repeatedly
associated with ideological and moral assertions, values, and expres
sions. Lyrical metaphors emerge then in the associations made within
and across expressive domains. It is possible to speak of both verbal
and nonverbal metaphors, and of multimedia metaphors. Ritual
symbols are always polysemic, with multiple, necessarily contra
dictory, meanings. For example, the cluster of terms, objects, and
connotations surrounding "Lemba" are understood both as sickness
and health, both bane and blessing. Multiple and contradictory
semantic structures such as this serve the purpose of allowing
movement, or the possibility of movement Structures of metaphoric
relations identify members of a class, they link similars and oppose
differences, and they range units on scales of ascending value. Perhaps
the most common metaphoric structure in Lemba expression is that
which takes elements from the hierarchically arranged series of beings
"animal/human/supernatural," and associates with these elements a
contrastive set of "human type A" and "human type B." A series of
allusions or expressive metaphors links the two series, establishing a
matrix of movement, most often of the neophyte and his household.
Some analysts of narrative genres have insisted that reality-
defining lyrics such as these in Lemba be called "myth." In the
Lemba setting I would reserve this term for tales told about the heroes
and their relationship to humans. The distinction of such tales and
lyrics is made by Lemba adherents. Whereas the lyrics are an
important part of the ceremony, in concert with drumming, the tales
are not told in the Lemba setting. However, as shall become clear in
subsequent chapters, heroes and human figures mentioned in the
ceremonial lyrics of Lemba may also be the figures of the tales.
Therefore, an analytic method is required which can move from the
one "genre" to the other.
The present analysis, which covers both lyrics and tales, can
benefit from a simple rewriting procedure which identifies the
principal nouns, verbs, and objects of narrative sentences—the
elements of verbal metaphors—and arranges them so as to facilitate
the recognition of "movement," such as transitive, causative,
reflexive, or gradual, long-term, and permanent syntagmatic trans

5
formation of a figure, be it the legendary hero or the human neophyte.
TheLemba Father's opening song in the courtyard may be rewritten
in the following manner to sketch metaphoric relationships.
NOUN VERB OBJECT
(1.17) iLemba) (gives)

X
(sun > (takes)
(ancestors )
(1-18) neophyte dreamed of
(copper bracelets)
(1.19) [payment of] X's falling into
supports water
pigs
The first couplet, as already suggested above, associates Lemba and

the sun with the verbs giving and taking. The KiKongo phrase
indicates that an object is involved whose identity is not clear in the
context of the couplet alone. The broader context of the Lemba
initiation hints that the object (X) is the force or spirit in Lemba, the
attacking, harassing illness, the malaise. Later songs will show that
this "hidden referent" is being"moved." The third phrase of the song
(1.19) states the condition of this movement If the Lemba child pays
or sacrifices pigs, X may be neutralized or removed into water. Other
Kongo n'kisi phrase the hold of an illness upon the sufferer in this
way. Thus here the first couplet defines the general context of
9
Lemba s order by relating it to the sun, and both to a hidden object
The neophyte's dreams are acknowledged, in which ancestors are
represented in copper bracelets. The conditions of release and
fulfillment are stated in clear causal terms: pay pigs, X will disappear.
In his second song, which is sung through the night of the first phase
of the ceremony (see figure 13 below), the Lemba Father introduces
further images of movement, both those drawn from nature—the fawn
which becomes the grown antelope—and from his own initiation.
(1.21) fawn becomes grown antelope

(1.22) Lemba causes difficulties
(1.23) jX which dies) l a u g h ( s ) a t neophyte?
1
dogs *
(1.24) X /guards i /Lemba "clan"
(1.25) (protects' \ children
) property
' novices
150 RITUALS
(1.26) lusaba pot,

palm arches
(1.27) Lemba Father copper
bracelets
(1.28) as novice Lemba priest
At least three levels of movement are suggested in this song. The fawn
becoming a grown antelope is the most obvious natural symbol on
which all else is hung. From lines 22-25 Lemba or the hidden referent
X moves from causer of difficulties to protector and guardian of
LemZxz-associated clans, children, property, and novices. Dogs have
an interesting relationship to the spirit world, in that they are able to
see invisible spirits, and dogs are often taken as signs of mediators,
pointers to the beyond, as their role in hunting hidden animals strongly
suggests. A third movement occurs in lines 26-28 where the Lemba
Father recalls his own progression as a novice and the pertinent ritual
objects—the lusaba pot's medicine (mpolo), the arches of the lodge,
the bracelets—thereby sketching for his own Lemba Son the move
ment anticipated in the latter's ceremony. This song, especially lines
22-25, and part of the previous song, especially line 17, introduce a
pattern in the way Lemba expression handles ambiguity and contrast.
The passages speak of a given force or power causing both positive
and negative effects: giving and taking away, causing difficulty and
protecting. They thus demonstrate the concern with shifting the force
of X from threat to beneficial ally. In the song of the Lemba
priestesses—the mimbanda Lemba—sung behind the house in the
kitchen hearth area at the end of the first phase as the bracelets are
consecrated and the Lemba couple blessed, this hidden referent is
identified as Bunzi.
The song of the priestesses is more a rejoicing over father and son's
perseverance in their quest than an agonized introspection of the son-
neophyte's dreams.
(1.44) (father ) will succeed in Lemba

t son )
(1.45) he who (perseveres against> Bunzi's
(is successful with) curse
(1.46) field (barren)
in < fertile I
(1.47) dreams persisted

about
Lemba
Fathers
(1.50) child wears shoulder
strap
[of Lemba
n'kobe]
(1.51-2) child found ) money \
(1.53)
will gain )
supplicate
Ì access to >
Lemba )
i earth )
<sky \
In accordance with the interpretation introduced above about the tight

juxtaposition of ambiguity with spiritual force, here Bunzi, earth
deity, simbi spirit or land, territory, and water spirit, is related to
barrenness and fertility. The overriding theme of the song however is
not Bunzi but the beauty of the patrifilial continuity of father and son,
and unending dreams by sons or children about fathers inLemba. The
neophyte has gained access to Lemba. He is already pictured as
wearing the n'kobe.
The presiding Lemba father replies, while consecrating bracelets
and blessing the couple, in lines that define iheLemba transformation
in yet more ways.
(1.55-57a) was > ( i difficult j

'has become) I source of healing j
power
Lemba (heals > (children j
< causes to grow) (ni
nionzi fish )
(1.57b) those get portion of
who Lemba
wait
Here, as in other songs above, the Lemba Father spells out virtues of
Lemba members. Lemba's power helps the novice to heal, to cause
growth and prosperity. Essential personal virtues of the Lemba priest
and priestess are patience, perseverance, persistence, clarity.
152 RITUALS
The power ofLemba and the virtues of the novice-priest are lauded
in even more spectacular terms when the test profanement has
occurred smdLemba priests and priestesses reconvene for the second
phase of the initiation. The Lemba Father opens the sence (1.63 ff)
and the priests and priestesses reiterate, in an evening and all night
celebration (1.70 ff), the virtues and characteristics of Lemba
adherents.
(1.63) the lazy one has become I industrious j

(1.65) ] harvester in
( "nzamba" field!
(1.64) nkonzi drum assembles festival of Lemba
The Lemba Father praises his Son for industriousness. The "harvest"
undoubtedly has reference to the amount of tax in ceremonial goods
levied on the perpetrator of the profanement Nzamba may refer to a
type of harvest net or a type of grass of which the net or basket is made,
but this "field" is almost certainly allegorical, referring to the
operations of the Lemba priest in society, his ability to rule and levy
morality fines, to be a noble person and an effective one. In later texts
(8.3) references criticizing laziness and lauding industriousness and
work will appear in similar rubrics.
The Lemba brotherhood's evening and night songs define the
virtues of the Lemba adherent in more exotic ways. The mood of the
following songs is clearly festive:
(1.72) spark has become " a hundred and ten"

(a raging fire)
(1.73) receiving beating mukonzi \
Lemba requires j paying money >
cleansing in fire )
(1.76) Lemba are alert
recipients
(like) bats
(1.77) (like) (can"see") evasive
night-jar (insightful)
(1.78) "faithful"
(no mistresses)
Lemba witnesses with whom I spoke told of huge fires at these final
Lemba initiatory festivals. To define the quality of Lemba producti
vity as a "spark" is quite natural, one which jumps from "one" to
"hundred and ten." The image is a powerful one suggesting offspring,
political effectiveness, enrichment, and the like. A s insightful are the
further character traits of Lemba recipients, alertness, evasiveness,
insightfulness, faithfulness. Creatures of the "night of Lemba" are
drawn into the metaphor-making process. N o analysis can recreate
the expressive content of these verbal terms. An entire night of singing
and dancing around a fire, in the company of regional leaders, with the
promise of effective recruitment to leadership and a banquet on the
following day, hints at the background of these song texts. The bat is
lightning fast, able to veer and dart instantly. The night-jar, another
nocturnal creature, is able to catch insects on the wing by darting
about rapidly. These character traits of prowess in the midst of social
and political life define the Lemba adherent. Then, they are summed
up as "seeing" in the dark of life just as the nocturnal bat and bird
" see" in the night Such virtues of the public realm are combined with
marital fidelity.
Kwamba gives no songs for the morning afterwards. It is very likely
that an esoteric Lemba lyrical tradition existed for the portions of the
ritual in the savanna clearing and even in the long walk to the cemetery
to collect tomb earth. The song of the patrifilial children of the
neophyte (1.94-96) is brief but poignant It is sung in connection with
the combined feast distribution and preparation of medicines in the
kitchen area. It reflects the "movement" of the other songs in its
acknowledgment of the neophyte's early obstacle-ridden status
outside of Lemba and his new status as successful priestly mediator.
9
(1.94) X (was i ("stitch" of pain j

(has become) (path to the priesthood)
(1.95) X caused dawn of sun of
Lemba
(1.96) novice (died > (in Lemba F ather)
(rose to life> (in Lemba >
Several powerful classes of metaphors are combined here to chart the

ritual movement of the neophyte. Once more the ambivalence of a
domain—pain/priesthood, life/death— is in juxtaposition with the
hidden or unnamed referent This power is transformed. In the most
154 RITUALS
characteristic symbolism for an African drum of affliction, the force of

the pain becomes the force of priesthood. Allusions to the "sun of
Lemba" combine outer cosmic references, where the sun circles the
earth, with the inner cosmology in which the individual is born and
dies. Thus, a multiple metaphor of cosmological proportions emerges
such that pain-becoming-priesthood is like the sun's rising, which in
turn alludes to the metaphysical rebirth of the Lemba neophyte
through death in his patron-father. The patrifilial children in Lemba
would be analogous to the chorus of a Greek drama, defining the
cosmic connections and proportions of the protagonist's life course.
These motifs reappear in other variants even more forcefully.
A final lyric in Kwamba's account is the new priest's chant used in
his own healing seances (1.109-115). It brings Lemba'$ ritual full
circle, demonstrating sufferer turned healer, his analysis of himself as
sufferer now projected upon a new sufferer. It opens with the creation
of a ritualized context in the natural and social universe, following
which the force causing the suffering is addressed.
(1.109) (Sun > (grants )

(Moon) (takes away)
I
(1.110) Lemba Father bore \
gendered > me
raised )
(1.111) Praise (earth)
(sky )
(1.112) (am enhanced >
(have gone far)
(1.113) Search patrifilial j
children's J
! ranks
(1.115) X release sufferer
(Sufferer) (will bring ) gift
(Priests > (will receive)
The ritual function of the songs is to encourage the sufferer-

neophyte through the tortuous path from illness to health, from
conflict to clarity of purpose, from being victim of an attacking spirit to
its beneficiary. Each song suggests at least one type of encour
agement, from a variety of sources—his Lemba Father, the priests
and priestesses together, the Lemba wives, the patrifilial children—
moving him from negative to positive ritual status. The father is

concerned with the terms of the movement from harassed sufferer to
one who is protected by Lemba, in touch with its powers. The priests
reflect this in their song, but emphasize the character traits of the
complete Lemba adherent The Lemba wives mention the transition
from barrenness to fertility as a result of effective patrifilial continuity.
The patrifilial children of the neophyte recognize movement from
illness to the priesthood, from "death" in the Lemba Father to "life"
in Lemba. A host of specific natural symbols are related to these ritual
conditions, symbols such as the fawn becoming a grown antelope, the
sun and moon's setting and rising, dogs laughing, fish growing,
harvest, fire sparking, fire purifying, bats being evasive, night-jars
being alert, and the like. By "rooting" the mystical themes in these
very familiar and concrete natural movements, the deeper truths of
Lemba emerge, easy to understand, sharp and plain. Most difficult of
all seems to be the articulation of the invisible mystical force behind the
ambiguity of the sufferer's condition. Mostly hidden, or buried, the
unnamed referent crops up most frequently when the descriptive
signifiers in the lyrics are contradictory and contrasting. This may be
an attempt to clarify ambiguity through sharper contrast, even
exaggerated contrast. Or the contradictions of life are stated clearly so
as to push the neophyte toward the positive pole of ritual experience,
helping him to control his own reality more effectively. In Kwamba's
account the mystical referent varies from the vague unnamed "X'\ to
simply "Lemba" to the "curse of Bunzi" identified by Lemba
priestesses. In variants to follow pre-existing figures are used much
more frequently.
The metaphoric structure of lyrical expressiveness may be fol
lowed in two other directions, into the ritual objects such as bracelets
and medicines, or into the structure of social-role relations lived in by
the Lemba Father, Son, wives, patrifilial children, the clans, and
Lemba itself. Further chapters shall explore these possibilities.
Chapter 5
The Eastern (Lari) Variant

of Lemba
"Their women gather to beat drums and clang bracelets in great

celebration."—Kimbembe, Text 3.25
Lemba's eastern variant is represented by four extensive accounts

and several different types of commentary, ranging from a turn-of-the-
century catechist's observations to the work of several missionary
ethnographers and an account and analysis by a modern Brazzaville
African writer of the region.
Kimbembe, the catechist, offers a description of a Lemba seance
dating from about 1915. It will be given in full and used as the point of
reference for the Lari variants of Lemba} Kimbembe's description
concentrates on the public portion of Lemba.
Stenstrom's account (1969) is a synoptic reconstruction based on
2
several sources gathered in 1929 in the Madzia region. The esoteric
portion of the seance is better represented here but since it is not
interpreted to any degree it presents serious problems for use.
Andersson's account (1953) is based on information given by a
3
hunter at Madzia and catechists at Musana in about 1930-31. Like
the account by Stenstrom, this is a synoptic reconstruction, collating
several disparate versions. Madzia in Lari country is fifty kilometers
distant from Musana in Nsundi and Kongo country. Internal analysis
is required to identify regional sources in the data.
Malonga's account of Lemba (1958) is also synoptic in that it
speaks for the entire "Lari" region and people, all of Lemba and,
indeed, the whole ancestral past, which is regrettably being drowned
4
out by Western European influences according to the author.
Malonga's description is internally coherent, however, reflecting a
157
158 RITUALS
skillful synthesis of the main lines of emphasis, gleaned from one

locality or related localities. In apologetic style, Malonga calls
Lemba not only the "biggest n'kisi of the past," an assertion made
once by Fulbert Youlou, but also "one of the most important
sociopolitical organizations of the Lari." In his essay he endeavors to
articulate the ideas and customs of this major institution, insofar as
9
this is possible given Lemba s extremely effective policy of secrecy
over esoteric aspects of the rite.
Malonga's lucid analysis of Lemba as a conscious political
institution of the traditional elite (bitomi), who understood the
manipulation of symbols of healing, contrasts markedly with the
Europeans' focus on the mythic origins of Lemba. Questions of how
Lemba's adherents understood their role will be taken up here and in
Chapter 10.
The Lemba Seance at Madzia (Among the Lari)
Text 3
TAKING BONZO MEDICINE
(1) Lemba diyokidi mikisi miakaka mu bunene ye mu ntalu.

Lemba is bigger and costlier than all other mikisi.
(2) Kadi nti muntu bakidi yuku-yuku diankokila lumbu kasidi
lumbu buna bana kwenda kwa tata dia Lemba mu nwana
bonzo.
When a person catches evening fever a day is set to approach a
Lemba Father to drink the bonzo medicine.
(3) Bu kemusila bonzo tala yasidi buna basidi laki nti diankoyi
bemusana (buka).
When the bonzo has been given then they need to set a time to
complete the healing.
THE FIRST DAY OF INITIATION
(4) Lumbu cina bucitula buna bankwezi zandi bana twala

malavu.
On the day it is done [the initiates'] in-laws bring wine.
The Eastern Variant 159
(5) Bantu bana bakila nkuni zazingi zazingi.

Other folks bring lots and lots of firewood.
(6) Tata dia Lembayandi katuvila monika vata, ka ku fula kana
yakala nateyemwana ma Lemba una tambika nsusuzolekwa
matwala malemba (wele landa nganga).
The Lemba F ather may not see the village; he will stay at the en
trance till the Lemba Child brings two chickens to the masters of
ceremonies who fetch the priests.
(7) Matwala bu kakwe twadi biabio buna tata dia lemba kulu-
mukini ku vata.
When the masters of ceremonies have arranged all then the
Lemba Father descends to the village.
(8) Basabo nitizabizi yidika nabanganga nandl
They are told what will happen; the priests may begin.
(9) Bu meni yidika tiya nibikula bekabikula nti
When the fire is going the call goes out;
(10) Widi Widi
D o you hear? We hear!
Tala Tala
Tala
Look. Look.
Mona Mona
Mona
D o you see? We see!
Kia nunga? Lemba!
Lemba!
What will he earn? Lembal
Lembal
Banganga Ko! Ko! IKo!
Priests. [sound of <drums, bracelets]
(11) Widi? Widi!
Widi!
E lutundulu nui'n'wandi
The lutundulu ant, his mouth
Kadila wele kindula.
Devours the sweet fruit ravenously.
(12) Widi? Widi!
E nsibisiulu,
O the nsibizi rat,
160 RITUALS
Mudimba kadila wele kindula.

In the low places he roots ravenously.
(13) Widi? Widil
Do you hear? We hear!
E nsibisiulu
O the nsibizi rat,
Mudimba kadila wele sengola.
He devours the very land that is his grainery.
(14) Widi? Widi!
Yoka vindumuki kwaku
Go! Rout out here.
Kani vakanga sina kubwanina.
I'll find you on a termite mound.
(15) Widi? Widi!
Mwini Malamu mu mukonzi
Sir Malamu in the nkonzi drum
Tata bwanga wa nyieni mo
Father is drumming, he leads me with it
Yenda lundi mafunda
To go store up treasures
Tata bwanga wa kidi moyo!
Father drums, he is alive!
Na nso kuzimbakana Buna bamubikudidi.
What was forgotten, Shall be remembered.
(16) Widi? Widi!
Mwikwa wamukazi
The pubes of the mukazi wife,
Kadi kasidi sonyia.
It causes embarrassment.
(17) Bu bamana sa bo, mimukunga una wantete, ubabadikilaka
babatikidi nti: tata na mwana bakina kwau—E yaya e-e!
When they have sung the opening song, Father and Child
dance together— O yeah!
SORTING THE MEDICINES

(18) Bu basa bo bakielolo na teye mbazipadi bu baso dia ngulu
zole benzi kwau, kadi biau bio ni bilongo kwau bizi mu
tangisa ye ngudi a Lemba ku nkula (ntwala) biakidi
When they have done this, they wait till the cock's crow at
dawn, when they eat two pigs. For it is their medicine which
they count out into the core of Lemba where they put nkula
red.
(19) Masuku bu mavioka basidi malaki (bilumbu) nti ta nwatu
masuku makemana (makaminisa)Lemba matudidi(lweke).
When the leaves have been brought out into the festive place
then Lemba has been assembled (has arrived).
PRESENTING THE PATRIFILIAL CHILDREN AND LEMBA WIVES
(20) Buna suku (lumbu) dina buditudidi nkwezi main matatu ye

mupata.
When day breaks, the in-laws [of the neophyte] give three palm
wine portions and 15 francs (5 mupata).
(21) Nkumbu biaubio biabionso bibakutanga bibikwanga mpunza
za Lemba.
Collectively, the name of all these bringing something is
"Mpunza za Lemba"—the paternal gift of Lemba.
(22) Basidi bo bena mu ndambuyo bizikina kwauyekielolo bwisi
bwamvimba.
Those who are there dance by themselves and wait for the
others.
(23) Banganga bubizidibanayidika tiya kufula, mboko matwala
malemba una nata nsusu nana ye mutete wabikwanga ye
malu matanu.
When the priests have come they build a fire in the entrance;
the masters of ceremonies bring nine chickens andfivebaskets
of manioc bread.
(24) Bu bameni zo dia ku fula, buna bana kwiza ku bula.
When they have eaten in the entrance, they move into the
village.
(25) Bakento bau bana kubabwana ye sika mikonzi ye betisa
mulunga mu nkembo wawingi.
Their women gather and beat drums and clang the bracelets in
much celebrating.
162 RITUALS
(26) Buna bau bubasa bo bizi vwanda muntu mankulu andi

zakala.
When they are through, the old people sit down.
(27) Bu bani yidika tiya, buna bantu bantete bana tekila kina.
When they have lit the fire, the first people begin to dance.
(28) Tata Malemba mboko mwana malemba ye bakaka mbo
sibalanda mwe kina.
The Lemba Father, then the Lemba Child, and all the others
follow in dancing.
(29) Kukina na mukento andi bu ke bonga kito kiansusu
umudikidi buna yandi wo vwendi.
Each dances out with his wife holding a chicken which he then
extends to the one who receives it.
(30) Nsusuyantete buyikuba basididibala bamfumbuyebasodidi
mpe bambanda.
The first chicken is extended to the representative of the
patrifilial children (bala bamfumbu); the neophyte's chosen
wives are also there,
(31) Balayemimbandabubameni solobo, bunamwiniwamvimba
kuseva pele nga nsusu kana futa kwa bangudi zanganga.
f
When the children and the Lemba wives have been presented,
all day they dare not laugh, or else a chicken is paid to the high
priests.
CONSTRUCTING THE PALM-ARCH LODGE;

THE "DEATH" AND "RESURRECTION" OF LEMBA NEOPHYTE
(32) Ntangu nsinza buyitula buna bayidikidi bala bamfumbu,

mwana nzo wamandala kunima nzoye bambanda ku mwela
nzo.
In the afternoon the patrifilial children construct for the Child a
palm-frond arch lodge behind [his] house and for his wives one
before the house.
(33) Ntangu malengola beteka mamba ye tedika mfumba ye bu
betedika zo, konso kinzu kiteka bila batedi mulolo ye
balundidi zo nate ye fuku diakaka.
At the appropriate time they fetch water and boil the big
kettles; every pot is filled so they cook for everyone; the
portions are stored till the next morning.
(34) Fuku dio bu dibwidi buna bakubudi yandi una vanda

Lemba, ye banganga bavovanga nti fwidi-jwidi.
When morning falls, then they strike the one initiating to
Lemba, and the priests speak about that he has died.
(35) Bu basa bo, nifula (fwekisa) beka kubafula, ye bubafulukidi
nimilolo beka ta ye kubula mfula.
Then, they blow [in his ear] with wind, and when they have
resurrected him they raise a shout of joy and fire a gun salute.
THE KILAMBU MEAL AND THE DONNING OF BRACELETS
(36) Buna badidikilambuye bwisi bubukia beso vonda ngulu iya.

Then they eat the Kilambu meal, and have four pigs slaugh
tered.
(37) Ye bu bekoma mimbanda milunga benzi kwau ka tata
malemba una sala lumbu kimosi nate ye beso dia ngulu
yamivambanu ye mwana malemba, mbo bavambani
And when they then have fitted the wives' bracelets the Lemba
Father remains another day to eat the pork and then he takes
leave of his Lemba Child.
The Expressive Domains and their Codes
THE SPATIAL AND TEMPORAL DISTRIBUTION OF EVENTS
All four accounts of the eastern (Lari) variant indicate it lasts three
days at least and two nights (figures 8-11). Kimbembe and Malonga
describe it as if there is a pause sometime after the taking of the initial
bonzo medicine, during which resources are collected for the full
ceremony, and the neophyte receives intensive instruction. Malonga
says this may take six months to a year, Stenstrom's account also
suggests such a time lapse. Only where the neophyte and his
supporters dispose of all necessary resources at the outset may the
seance occur in one staging, as Andersson's account suggests.
All four accounts suggest, further, the common feature of a rhythm
between ceremonial location in the village and outside the village, in
the savanna or near a stream. As was clear in Kwamba's account of
the northern variant, the rhythm is indicative of the relationship of
1
Lemba 's public and esoteric worlds. The eastern variant under
consideration is especially clear in demonstrating how these two
164 RITUALS
Figure 8
Spatial and temporal distribution of events
in Kimbembe's account of eastern Lemba inauguration.
Numbers refer to lines in Text 2.
night night day
— "tin day I pause I
—I j- construction
of lodge
in kitchen "death" & Kilambu
area, preparation "resurrection" feast;
presenting of feast ( 3 2 - 3 ) of neophyte closing
all night patrifilial construction (34-5)
dance around & wives'
c n i k i r e n of lodge bracelets
fire ( 7 - 1 7 ) o
w i v e s t
before (36-7)
Lemba, dance neophyte's
(24-31) house(32)
neophyte opening Lemba priests
given ¿0020 wine by & wives e a t &
medicine in-laws mix medicine
(2-3) (4-5) (18-23)
[collection
of medicine?]
Figure 9
in Stenstrom's account of eastern Lemba inauguration
(Stenstrom, 1969)
• day { day night 1

day night
neophyte Lemba
all "marriage"
night "resurrected"
by Lemba in
dance burning house (p. 54)
neophyte with "mbondi" father;
given medicine L b a e m
dance with Lemba 8
bonzo scraping j W v e s
father (p. 5 1 ) father's 2=1
(p. 38) CP- >4 0

(p. 40) departure
dance
ordeal n'kobe
of prepared
Lemba wives (p. 54)
wives of ( . 52)
p
neophyte
instructed (p. 52)
gathering instructions neophyte neophyte
opening of priests to neophyte "killed" priest's
feast of & slaughter (p. 4 9 ) & bracelet
Lemba of animals donning of closed
priests for feast with neophyte's (p. 51)
(p. 38) neophyte (p. 48) bracelet
(p. 5 0 )
The E astern Variant 165
Figure 10
in Malonga's account of eastern Lemba inauguration
(Malonga, 1958)
(cock's
..time- day j night jI M S 1
afternoon 1 night ' crow) day evening
l
1
public 1
i meeting of regional 1
consummation preparing public l
dance, Lemba adherents & of Lemba of n'kobe feast &
neophyte public, dance and marriage in Lemba & distribution
village
presents recitation of Lemba lodge; "death" contents by Lemba

space
"children" w i s d o m by neophyte; & "resurrection" (p. 60) (p. 60)

& wives ordeal of neophyte's of neophyte
{pp. 5 7 - 8 ) wives (p. 5 9 ) (p. 60)
1
nsandzou secret
entrance
preliminary ritual
consecration meal of
. (p. 5 7 )
[bonzo?] Lemba (p. 60)
[collection
of
medicines?]
1 a
hillto
savar
purificatory
bath, shaving
of pubic
stream
hair (p. 6 0 )
Figure 11
in Andersson's account of eastern Lemba inauguration
(Andersson, 1953)
day night day I night day
"marriage
in house dance of
opening "resurrection" or lodge, [feast?] neophyte
dances of neophyte, instructions, priest &
bracelet n'kobe separation
donned prepared with father
anointment
ritual
of neophyte meal,
[bonzo?] animals
slaughtered
ritual "death" of instructions

slaughter neophyte to neophyte's
of animals, wives
meal [collection of
medicines?]
166 RITUALS
spheres—public/esoteric, village/nonvillage—come to be inverted

near the close of the initiation, so that the public space around the
neophyte's house, hitherto the scene of conventional dances and
feasting, becomes the location for the Lemba marriage and its special
ritual qualities. Of the four accounts, only Malonga's has the clear
spatial-temporal organization and rhythm present in Kwamba's
account in the foregoing chapter, with the ceremony divided into two
phases, each touching savanna and village, and within the village
dividing activities clearly between courtyard and kitchen.
The effect of such rhythmic integration of public and esoteric
spheres, so common in Equatorial and Central-African ritual organi
zation, is to charge or load symbols by juxtaposing contrasting poles
of meaning: savanna/village, day/night, courtyard/hearth, and so on.
The accounts show the movement of the ceremony, for example, from
the entrance or hilltop area at the outset into the village square for a
public dance, corresponding as well with the contextual movement
from daytime to night, light to darkness. Also, in the ritual "death"
and "resurrection," Andersson and Stenstrom describe this process
moving from the esoteric setting of the savanna hilltop to the public
setting of the village square. Several of the accounts also recognize the
court/kitchen opposition in the marriage, as was the case in Kwam
ba's northern variant. Kwamba indicated that the conventional
meaning of court was its public, male significance, and that of the
kitchen its semiprivate, female significance, but that this was inverted
when the courtyard lodge became the site of the "wife exchange," and
the kitchen the site of the preparation of medicines. In several of the
eastern variants, this same inversion occurs: the court becoming the
site of consummation of the Lemba marriage, the kitchen the site
9
where Lemba s ritual food, or medicine, is prepared. To understand
the implications of this spatial rhythm and the inversion of conven
tional spatial and sex-role definitions, we must examine other
expressive domains.
ECONOMIC AND EXCHANGE STRUCTURES
Only Kimbembe and Stenstrom offer systematic information on the

nature and quantity of exchange goods in the eastern variant of
9
Lemba s inauguration. A glance at figure 12, which summarizes this
information, suggests that these quantities are only a fourth of those
reported by Kwamba in the Mboko Nsongo region: six pigs are
consumed, instead of the twenty-six in Kwamba's account The final
grand Kilambu feast, emphasized by Kwamba, and Malonga in
The E astern Variant 16 7
Figure 12
Exchange structures of eastern Lemba séance
Kimbembe
Taking Bonzo Medicine Taking Mabonzo
Lemba
neophyte -I bonzo neophyte
Father
First Day of Initiation Opening Feast of Lemba Priests
neophyte's
wine& neophyte
in-laws & Lemba
firewood and/or
friends priests on
family in
hilltop
Lemba village
neophyte -) 2 chickens • master of
ceremonies
Sorting of Medicines In Village, Sorting Medicines

L e m b a master of
2 P8
neophyte -H 'ÏZ1 ~ Lemba ceremonies
medicines """
priestesses
neophyte's palmwine in Nga Malamu

chief fire (Lemba
spirit)
Presenting Patrifilial Children Gathering of Priests on Hilltop

& Lemba Wives
1 large pig
1 goat
3 palmwine
5 chickens
portions
neophyte's 2 baskets manioc bread Lemba
(demijons?)
neophyte's village salt, pepper ' priests
in-laws\ cauldron
machete, water
15 francs
firewood
9 chickens
5 baskets manioc bread
scraps, i.e.,
entrails, bits
of liver,
neophyte heart, fat,
Lemba-lemba
leaves, chalk
Feast Scraps Buried Near Neophyte's House
neophyte's bracelet [?] Instructions to Neophyte
neophyte
168 RITUALS
Figure 12 (cont'd)
Kimbembe (con't) Stenström (con't)
Death & Resurrection of Neophytes;

Father and Son Dance
Bracelet Closed
Neophyte's Wives Instructed
neophyte's calabashes
wives of
chalk paste
N'kobi Chest Prepared
neophyte's n'kobi[7]
Construction of Palm-Arches for Neophyte Priest's Marriage to His

Marriage; Death & Resurrection Wives in House
neophyte's marriage
patnfilial arches
children before &
behind
house
Kilambu feast [?]

preparation
for Kilambu
meal
- \ 4 pigs
Kilambu Meal & Closing of

Wives' Bracelets
patrifilial
children Kilambu
& feast
public
neophyte's
wives
Lemba Father's Departure
eastern reports, is totally missing from Stenstrom's account. Al

though this might be an ethnographic oversight, it might also indicate
hard times, the absence of a ceremonial fund in the early colonial
period due to heavy taxation, the decline or absence of trade goods or
other sources of local wealth. There is no question in the four accounts
of Lemba? § high status in the Lari region, thus ceremonial funds
would not have gone to another n'kisi. The proximity of the Lari
region to the colonial capital ofBrazzaville suggests that by 1915-30,
the time of the reports under consideration, Lemba was at the very
final stages of existence as a public institution, still highly esteemed
but economically being choked out of existence. By 1930, suggests
Stenstrom, Lemba was decreed illegal by French colonial powers
because it was consuming too many resources.
The overall structure of exchange is comparable to the northern
1
account of Lemba 's inauguration. The sufferer approaches siLemba
priest seeking a "cure," and when the bonzo has been administered,
the priest becomes the sufferer'sLemba Father. After theLemba Son
has assembled the requisite goods and patronage, he and his
kinsmen—of his own clan and village, as well as his wives' clansmen
—then the regional Lemba adherents are notified. Numerous ex
changes occur between Lemba and these groups identified with the
neophyte, with Lemba presenting the n'kobe box and the bracelets,
drums, and other paraphernalia he and his wife or wives will keep. At
the close of the séance the neophyte's lineage's patrifilial children
prepare the Kilambu feast, which is distributed to the local public in
attendance. Lemba's secrets, wisdom, and medicinal mementos are
given the neophyte's household. One important difference in the
exchange pattern here is the absence of a massive infusion of goods
from a "profanation" tax.
KINSHIP AND SOCIAL ORGANIZATION
Throughout the accounts a clear status opposition is seen between the

neophyte, called "sufferer," "sick man," "deacon being initiated,"
"Lemba child," and the "healer," "senior priest," and "Lemba
Father." There is some disagreement on the exact nature of the tie
between these two figures. Malonga states that an individual prefers
to be initiated to Lemba by his true father, or if his father is deceased,
by his father's brother, hisparrain nature!. A man is thus encouraged:
"Initiate yourself to Lemba while your father is still alive, ohrichone,
5
so that you won't lament him when he is no longer there." Indicative
of the emphasis on patrifiliality was the phrase, mouana kou mbouti
170 RITUALS
ko, kou ntombole mpemba'ko, "if the child is not engendered by me,
he does not receive the white chalk," that is, the blessing. Or, from the
son's view, mpemba tata mpeni moyo, "whiteness given by the
6
father, gave me life."
Cognatic ties, on the other hand, were recognized in the Lemba
system only in the combination of groups in the public part of the
ceremony: the neophyte's matrilineage, the matrilineages of his wives
(his affines), and the fragments of matrilineages present in his own
clan's patrifilial children. Some of these groups played special roles in
the Lemba ceremony. For example, the neophyte's clan's patrifilial
children (the bala bamfumbu) became the pages, cooks, and partial
priests for the inauguration. His in-laws, collectively called thepunza
dia Lemba, or the clans behind the co-wives in the Lemba household,
brought wine and firewood, according to one account The Lemba
inauguration or renewal brought together these many descent groups,
stimulating interlineage and village exchanges.
The idea of marriage, which links these various descent groups into
an extensive network and maintains it over time, is then at the heart of
Lemba as an institution. Lemba's marriage code, reflecting the
importance of the affinal and patrifilial ties to extensive networks, is
spelled out in rules given the neophyte priest and his wife or wives. He
is warned that if he commits adultery he will be stricken by Lemba\
illness. He must live in accord with his wives, preferably eating with
them—a custom not common in Central Africa. The Lemba wife or
wives are instructed not to have extramarital affairs and to resist the
advances of other men as strongly as they resist insults during
initiation (Andersson, Stenstrom's account) at the hands of estab
7
lished Lemba wives, During the height of the Lemba "marriage"
the wives' faithfulness is tested.
The neophyte priest and his favorite wife are brought to the palm-
branch lodge before his house. There, under the eyes of the patrifilial
children, the Lemba marriage is "consummated." At the instant
Malonga terms "psychologically right" the neophyte's genitalia are
struck so that he momentarily loses consciousness. Lemba theory has
it that the quicker he revives, the more faithful are his wives. If he fails
to revive quickly, his wives will be called to confess their infidelity.
During his unconsciousness sperm is taken and mixed with the wife or
wives' pubic hair for a preparation of a powerful symbol in the
8
memento, the mizita figurines.
The quality of the Lemba marital relationship is also suggested by
the structure of participation of the punza dia Lemba group, the
TheJE astern Variant 171
cowives of Lemba, in exchanges. Their payment to the Lemba

ceremony constitutes a sort of reverse bride payment—perhaps a
ritual dowry—uncommon in conventional Central-African exchange,
which tends to equalize obligations between the neophyte's and his
wives' clans, making thepunza copartners in the Lemba household.
The patrifilial children (bala bamfumbu) like the descent groups of
cowives (punza dia Lemba) constituted a network of several
communities supporting the neophyte. Those present at the inaugura
tion, at least in eastern accounts, appear to be selected representatives
of the entire group of the neophyte's clan's patrifilial children. As in
general Kongo ceremonial life to the present, the patrifilial children
presided over the kitchen behind the neophyte's house, served in the
dance, and acted as priest during the "death" and "resurrection" of
the neophyte, as they preside over all transitions between the two
worlds of the living and the dead in their fathers' clan.
In an important sense the patrifilial children are the source of their
fathers' power. A s has been seen in the northern variant and to a
degree in the eastern account (see figures 7 and 12), the patrifilial
children receive and redistribute the final feast as the neophyte
receives, from Lemba priests, the symbols of authority and power.
The patrifilial children offer their loyalty and support, thus becoming
the continuation of the line that Lemba father and son have
established. The three-tiered polity that results is mirrored in the
eastern Lemba etiology myth in which Nga Malamu, Kuba, and
Magungu are the succession of priests to have originated the idea of
continuous Lemba insight. The final section on lyrics in this chapter
returns to this matter.
The patrifilial alliance, extending a political network over space
and continuity over time, broke the isolation of landed matrilineages
as effectively as some of the chiefly traditions of the Lower Zaire.
Malonga's, of the four eastern accounts of Lemba, gives extensive
insight into just how Lemba governed on a day-to-day basis. Lemba
was holder of truth, the source of laws. Its matwala, like major
ministers of the regional seances, initiations, and renewals, were
9
responsible for the maintenance of Lemba s control of regional
resources, both the materials that flowed through the markets, as well
as the knowledge held by the various specialists, such as smiths and
healers. The matwala often went out under disguise to collect formu
las and techniques, both beneficent and malific, from magicians.
Lemba appropriated all knowledge in the society to itself; only in this
manner was it able to presume to combat witchcraft, that is, antisocial
172 RITUALS
powers and persons. The neophyte's socialization into the deep

understanding of society occurred during the six months of the year
9
after his initial medicine, and before his inauguration.
In order to maintain such a monopoly of resources Lemba needed
to assure the recruitment of the "brightest and best" men and women
to its ranks, those who constituted a natural basis of power in society.
Reluctant leaders were harassed through mystical threats or pro
mised great wealth and power. Despite the persistent rhetoric of
illness and healing, there was no specific "Lemba illness," states
Malonga. The idiom of illness was used to speak about social
disorder, and specific episodes of suffering were used to advantage to
extendLemitf's network among the qualified. Malonga makes it very
clear that Lemba maintained a highly conscious public image,
manipulated behind a front held in common secrecy by priests and
priestesses. TheregionalLemfia elder, ihemunkukunyungu, held the
"high priesthood" as a sort of senior, honorary title. The chief
treasurer and minister, the matwala, was a position of power. He,
along with the other banganga Lemba and their mimbanda wives,
constituted the local elite of every region. They were the links in the
continuous network of market trade and peace-keeping.
The explicit ideology which used the idiom of illness and healing to
maintain public order was in the eastern variant embodied in the role
dichotomy of the bitomi, the pure and initiated, and the bihinga, the
profane. (Tomi derives from toma, to excel, be superior or pure;
hinga from to beg, supplicate.) It is important to understand this
rhetoric of purity in Lemba not as an entrenched power of a contin
uous elite but as a mode of ritual in a segmentary society. All sources
emphasize that Lemba membership could not flow within the
matrilineage. Offices could only be perpetuated by extending the
patrifilial link, which in turn stimulated exchange. Priests and
priestesses were buried, when they died, with their bracelets and
paraphernalia. Everywhere Lemba? s wealth returned to the localized
village supporters in the form of the lavish feast distributed by the
patrifilial children of the neophyte. Lemba was thus a stimulation of
exchange between exogamous groups, rather than a hoarding asso
ciation within an endogamous, hereditary elite.
MEDICINE OF LEMBA
It should be apparent that the expressive domain "medicine" is the

nonverbal articulation of social structure, the exchange of goods and
gestures, words and songs. The clay, hair, fingernails, plants, and bits
of wood designated as "medicine" (bilongo, contained in the n'kisi)

contain little intrinsic meaning, but they are imbued with meaning
through words or songs uttered in their collection, mixing, and the
social context in which they are used. A type of verbal punning is
widespread in Kongo cultures whereby words spoken in the seance
correspond to the plants received by the sufferer—thus, for example,
"lemba, lemba" (peace, peace), describes the action of the plant
lemba-lemba (Brillantaisia patula T. Anders) given in the oral
potion. A s I have argued elsewhere, this type of verbal composition
principle need not exclude an understanding of the chemotherapeutic
effect of plants, especially where such medicines are taken orally or
10
applied to the skin. Nor should the possibility be excluded that a few
of the ingredients, particularly the bodily fluids and exuvia—-hair,
semen, feces, blood, fingernails—carry a powerful connotational,
even iconic, impact. But even here the precise meaning of the bodily
object is given in word, song, or social context. This is very well
illustrated in the eastern variant's exchange between "food" and
"medicine," already hinted at in the analysis of the northern variant.
At the outset of the ceremony the neophyte and his supporters give
food and drink (chicken, wine) to Lemba in exchange for bonzo
medicine (chalk, etc.). At the second stage, food is again offered the
Lemba priesthood by the neophyte, and medicines are sent back in
exchange. At a later stage the Lemba priests are presented with an
even larger quantity of food, a lavish feast, for which the neophyte
receives medicine, but this time mixed with scraps of food from the
priests' meal, specifically the chicken "giblets" of heart and liver. The
bracelets for the neophyte and his spouse(s) are prepared during the
episode of death and resurrection, and further medicines are devel
oped as the bracelets are consecrated. During the consummation of
the Lemba marriage, the groom's semen and the bride's shaven pubic
hair are mixed into a medicine. The final stage constitutes a complete
inversion of the opening, in terms of the exchange of food and
medicine. Lemba, through the beneficence of the patrifilial children of
the neophyte, distributes the Kilambu feast to a general public. The
Lemba priests and priestesses, now including the neophyte and his
wives, partake of a medicinal "feast," made up of food bits from the
Kilambu, but heavily enriched with "medicine" such as semen and
pubic hair from the marriage ceremony and dried excrements of
Lemba adherents.
Commenting on this final meal of the Lemba pure, Malonga notes
that "the symbol of this rite, revolting to the profane, is a key to the
indissolubility of the Lemba order. It consecrates that feeling of
174 RITUALS
fraternity uniting all the pure (bitomi). They must consider them
11
selves as constituting one body whose trust they may not betray." In
terms of the code that is followed, this exchange and mixing of
medicine and food is expressive of an abstract view on purity and
profanity. Lemba's purity is defined by its exposure to the neophyte's
impure social context, his "illness." Profane filth collected around the
neophyte is first absorbed by mixing chalk (luvemba, whiteness) and
lemba-lemba herbs (calmness) with it, then by having the Lemba
priests and priestesses eat the neophyte couple's sexuality, impurity,
dirt, excrements, and the like. As Mary Douglas has so correctly
observed, ritual power is generated by the absorption to the sacred of
12
dirt, the ambiguous, death, sin, and the impure.
All this, and more, is borne in the permanent "documentary"
symbolism of the n'kobe and its contents and in the drum and
bracelets, which, taken together, are expected to conserve the values
of exchange, well-being, purity, peace, and political loyalty lauded in
songs, speeches, rules, and instructions. In the eastern variant, as in
9
the others, Lemba s permanent medicine is composed in two phases:
the first in connection with the initial bonzo (elsewhere called mpolo),
the second in connection with the novice priest's own n'kobe. Each
will be dealt with briefly here. Chapter 7 will consider medicines in
fuller, and comparative detail.
Stenstrom's account offers the ingredients of the Lemba high
priest's bonzo (pi. mabonzo), or opening medicine, intended to clear
the way for the new status which is to follow.
Lemba-[lemba] herb
Nsangu dia dinkondo (seeds) (Ocimum bazilicum), with
pungent odor or aroma
Makala manzo mbongi (sic), "charcoal from the hearth of
menstruating woman's house" [This cannot be correct, for
mbongi is the "men's house."]
Mpemba, white chalk
Lutundu, herb with red fruit
Mansunsu (Ocium arborences), an aromatic plant
Nkukidila niajnba, leaves, twigs, silt, etc., thrown onto river
bank during flood
Muyitu, ashes of herbs and leaves from sweat bath
Ngasi zasombo, small palm nuts cracked by teeth
Nsala zankuka, feathers of the nkuka bird (Turacus persa.)
13
Nsangi, a small fish
These items are recorded to suggest that they are wrapped in the skins
of two antelopes, the small red kinkululu and the mbambi water
antelope. However, it is not clear which ingredients are associated
with which skin, nor what the verbal definers are of the articles in the
context used. It cannot be assumed that the meaning of such items is
consistent in northern Kongo cultures, thus an interpretation or
dictionary meaning given in one setting cannot be carried over to
another. Stenstrom has conflated several informants' lists of ingredi
ents, and it is impossible to analyze them effectively. Some of the
same ingredients appear in the second phase of medicinal composi
tion for the nkobe, in which context their lyrical definition is
fortunately given (see figure 13).
N'kobe ingredients begin to be collected in the "scraping" cere
mony at the time the Lemba couple and their pages and patrifilial
children are presented to Lemba officiants early in the first phase (see
Stenstrom's account, figure 9; Kimbembe, figure 8, Text 3.18-19).
These ingredients, in Stenstrom's account, are placed onto two skins,
of the nkumbi rat and the musimba wildcat respectively, and after
they are tied shut they are hung temporarily on the neophyte's house
until the end of the initiation. As they are scraped, the plants or
substances are "charged" with verbal meaning by the priest. For
example, the nsangi fish is cut up to the singing of the following
phrase:
Nsangi, nsangi mu mbanda eke e
Nsangi fish, nsangi fish, be mixed—cause to
jump, dance—the mbanda wife, eh, eh.
As the lufumbu vine is scraped, the priest sings:
14
A muti ambumbu, tata walembo kina, ka ulamo ko
Oh mbumbu tree, father doesn't dance, he won't dance.
The objects are so to speak given life and meaning in the context of the
performance. This situational attribution of meaning may vary greatly
from one locale or initiation to another, and from one set of medicines
to the next, for all but a few of the classic Kongo or Central-African
ritual symbols, which seem consistently to retain their underlying
meaning from context to context, region to region, even decade to
decade.
Thus, mpemba chalk, present in all bonzo openings, symbolizes
"whiteness" and is a sign of purity, correctness, loyalty, innocence,
and truth, synonymous virtues in Lemba. Lemba's cure, as many
176 RITUALS
Figure 13
Eastern Lemba tukobe and ingredients
others, contains mpemba chalk because of the ready contrast it offers

15
with states of sickness and impurity. Thus, in a rather predictable
interpretation as they add chalk to the medicines, the Lemba priest
sings
mpemba batata mpedi moyo e e
With mpemba the fathers gave me life;
mpemba tata mpedi moyo e e
16
mpemba, father gave me life.
Another pervasive classic ingredient is lukula or tukula, red
powder of the tukula tree (Pterocarpus cabrae [elsoyauxii]), which is
usually used in conjunction with mpemba chalk to designate contrast
ing but related qualities. In the eastern Lemba accounts, mpemba in
the nkobe denotes life or power in the neophyte priest's life, and it is
kept in the skin of an nkumbi rat (see figure 14, Stenstrdm and
Malonga). Tukula red, kept in the skin of the musimba wildcat,
17
signifies the benediction of the ancestors upon the Lemba couple.
The juxtaposition of mpemba and tukula in nearly allLemba tukobe,
in connection with virtues of the Lemba priest and priestess respec
tively, suggests that there is here a basic metaphor. Indeed, it
reappears in all regional accounts ofLemba from the Teke-influenced
Lari in the eastern variant, to the Mayombe and coastal accounts. The
elements of the metaphor are drawn from what is known to be
widespread in Equatorial-African cosmology and thus characteristic
of a widespread ritualization process. The sex roles of the Lemba
male and female are first of all defined by whiteness and redness, the
first denoting purity, contact with ancestors, virtue, the second denot
ing power, fusion or mediation, sometimes danger. The sense of these
substance and color symbols is made crystal clear by their containers
in the n'kobe. The male "whiteness" is placed in the skin of the
nkumbi rat, a striped rodent whose underground burrowing skills are
widely lauded. He stands for contact with the underground domain of
mpemba, the land of the dead. Female "redness" is contained, in
Stenstrom's account, in the spotted skin of the musimba cat, a tree
animal, whose characteristics thus suggest skilled human interaction,
especially in the realm of relations between the sexes and in fertility.
The fragment of the Lemba metaphor might be sketched as in figure
14.
Malonga's exegesis of other n'kobe ingredients explicates some of
the other common ritual items found mLemba. The fibers of the palm
nut—variation may be palm nut, palm branch—convey the wealth of
178 RITUALS
Figure 14
Sketch of Lemba's dominant metaphor,
based on eastern variant
sexual roles substance animal cosmological

and and order space
identities color
male whiteness n'kumbi burrows underground

mpemba rat
redness musimba runs on ground,

female tukula cat climbs trees
the palm in Kongo societies: oil, wine, cloth ( = money) are all derived
from it. Inclusion of a bit of the palm draws the neophyte into
18
association with the productive realm of society. Ashes are the
symbol of the hearth that every respected man must possess. The
Lemba priest has a sacred obligation not to become isolated, without
hearth or household, wives and children. Charcoal is the "allegory of
discretion," denoting that what is hidden in obscurity is difficult to
uncover. Discretion is a virtue of every Lemba person, every butomi.
Charcoal is a sign of the ability of magicians to render invisible from
19
profane eyes and minds their intentions and methods. Menstrual
blood associates Lemba with the moment of a woman's greatest
fertility and reproductive potential. Menstrual symbolism comple
ments the male element in Lemba, advocating multiplication and
productivity, life.
These medicinal ingredients (bilongo) are attached, by words and
connotations, to the active norms and ideals of the Lemba couple and
Lemba as a wider social institution. Some of them also gain their
meaning out of the functional context in which they are used in
productive life. Others, as suggested earlier, may have a direct iconic
derivation having to do with the body of the initiate or of the corporate
body of Lemba. Bodily clippings, exuvia, fluids, and parts find their
use in expressive efforts to represent unity of individuals. Thus the
mizita figurines present in all accounts (see figure 13) contain ashes of
the wives' pubic hair and the husband's semen, mementos of the
transcendent "moment of truth" in the consummation of the marriage
when the groom is knocked unconscious ("dead") so he may be raised
("resurrected") to the new unity in Lemba. The mizita may have
individually colored beads to represent each wife, but their contents

denote permanent transcendence of individuality in the new life of
Lemba.
The unity of the larger corporate body of Lemba is represented in
the nail-parings and hair (elsewhere, rags from old clothes) clipped
from each priest and priestess and placed into the miniature ngoma
drum or mavungu horn (figure 13) that are inserted into the n'kobe.
Whether these drums are called ngoma, nkonzi, or nkonko, they are
called the voice of the ancestors or spirits, representing the corporate
collectivity. Thus in one song(Text 3.15), Lemba's legendary founder
Nga Malamu is addressed:
Mwini Malamu mu mukonzi
Sir Malamu, in the n'konzi drum
Tata bwanga wa nyieni mo

Father is drumming, he leads me by it
In another, Kuba:
The ngoma goes mbwel mbwel
Don't you hear it?
Oh Kuba!
20
I move to it!
N o less than the bones or hairs of martyred heroes of Western
Christianity inspire the church in shrines, or ancestral chiefs' bones
are revered in the ancestor baskets of clans in Central Africa, these
bundles of fingernails, hairs, etc., of the Lemba community iconize
the disparate individuals into one communal body, joined with distant
9
heroic ancestors. But Lemba s icon, it must be recalled, is not
hereditary but must be renewed by initiation of new couples.
THE LYRICAL
The lyric sense was already present in the previous section, defining
or sharpening the meaning of inchoate medicinal ingredients or other
ritual objects. In this section I want to illustrate the full form of the
lyrical in one important song and the etiological myth ofNga Malamu,
Kuba, and Magungu present in a fragmentary way in most of the
eastern variants of Lemba. It is now apparent that the ritual objects
such as clanging bracelets, throbbing drums, burrowing rodents,
semen and pubic hair ashes, food and the like lend social interaction a
180 RITUALS
sensory immediacy of great power. On these objects, as we have seen

with such cases as the miniature drums, the bracelets, and other
articles, metaphors of esoteric reality are constructed that embody
less tangible but no less real statements about the human and
supernatural world. In the Lemba setting, social relations are often
linked with the realm of spirits and deities, and both are concretized in
ceremonial objects.
This type of linkage of the human, material, and spiritual realms is
evident in a favorite eastern Lemba song type about two types of
animals, one of whom is an effective, the other an ineffective burrower
21
for food. Kimbembe (Text 3.11-16) and Malonga offer comparable
texts.
KIMBEMBE MALONGA
A\ Lou nkouma-nkouma tafe,

lou na loua koumini ngoudi
a nganga.
Oh, my son, the truth resides in
the words of the chief priest.
E lutundulu nuinwandi, Ma tondélé nkoumbi, na

Kadila wele kindula. mayelo ma tembélé.
O the lutundulu ant, his mouth Whatever causes nkumbi rat to
devours the sweet fruit rejoice, makes his
moustache dance (twitch).
E nsibisiulu, mudimba kadila

wele kindula.
But oh the nsibizi rat, in the
deep places he devours with
ravenous appetite.
E nsibisiulu, mudimba kadila A\ ntsibizizoba, dimba ka dila

wele sengola. meni sengola.
Oh the nsibizi, he even devours But the stupid ntsibizi, he
his own granary! burrows up the ground that
is his granary!
Yoka vindumuki kwaku El Yenda, vindoumouki

kani vakanga si na ku kou'akou!
bwanina Kani ghakanga tsia na ku
Go! Roll up! I'll catch you bouanina.
on a termite mound. Go! Roll up! I'll catch you
on a termite mound.
Mwini Malamu mu mukonzi Al Bou kouenda bou kouenda,

Tata bwanga wa nyieni mo. yenda yenda na nkouma na
Sir Malamu in the nkonzi ngou'andi a diambou.
drum, On your journeys, see that you
Father beckons, he leads me are accompanied by the
by it. " word of truth" (proverb and
message).
Yenda lundi mafunda

Tata bwanga wa kidi moyo.
Go, store up treasures
Father beckons, he lives!
Na nso kuzimbakana
Buna Immubikudidil
What was hidden,
Shall be revealed!
Mwikwa wamukazi Taba dia nkazi

Kadi kasidi sonyia. Ka sidi bemba.
The Lemba wife's pubes, Il s'est certainement frotté à la
Cause embarrassment 'garniture' de sa femme eu
venant.
Didi nkoukou hou koukoum'

anga.
Il a mangé le coucou qui bégaie.
The texts characterize sets of natural creatures: Kimbembe, the

lutundulu ant and the nsibizi rodent; Malonga, the nkumbi rat and
the nsibizi. All are burrowing, digging, or wiggling creatures. As
suggested earlier, nkumbi, with striped skin and joyful twitching
whiskers, appears in many Lemba contexts, his burrowing highly
182 RITUALS
lauded, his skin taken as a container for the white chalk symbolizing
the Lemba neophyte's mystical and rhetorical abilities.
The songs also speak of the Lemba Father and Son, the high
priest, Lemba's founder Nga Malamu, and the mukazi Lemba wife.
The Lemba Father is depicted as an animator, reviver of the Son, who
puts him in touch with the powers, secrets, and wealth of Lemba. Not
only are mystical links intended here but also the power of words,
skillfully used as an access to power. Thus, Sir Malamu speaks to
the Lemba Child through father's drumming on rikonzi. Similarly,
the neophyte priest is told to take with him on his journeys the essence
of truth, "the word"—probably special proverbs or oratorical tech
niques. The songs move from the natural object concretizing truth, to
the less tangible qualities defining access to Lemba's powers.
There is in these songs, as in many of the couplets studied in the
northern variant, a sense of the hidden referent, probably given in
the Lemba context some esoteric meaning. Here, as in the songs of
the previous chapter, the hidden referent—"that which was hidden,
shall be revealed" (Kimbembe)—occurs in the face of a contradictory
or dichotomous situation. At one level this contradiction is visible in
the set of creatures, one of whom is successful, or skilled, the other of
whom is unsuccessful, stupid. In the first song by Kimbembe,
lutundulu reaches his sweet fruit, whereas nsibizi destroys his food
source and gets caught. In the second song, nkumbi is successful,
nsibizi stupid. Nsibizi is caught on a termite hill, itself full of pores
and tunnels, as he rolls himself up in a ball, thinking he has protection.
The successful creatures are the Lemba priests who can reveal
secrets, know and use rhetorical words to great effect. They can
handle truth. The unsuccessful "diggers" are the profane, lacking
truth and power.
There seems to be a further sexual-political signification in the
songs relating the burrowing animals to the nkazi-wife. The "suc
cessful" digger understands fertility, the secrets of reproduction, and
replenishment of the earth, whereas the "unsuccessful" digger
destroys, like the metaphoric nsibizi, the very source of his food.
Instead of children and followers, he effects destruction and loss of
human resources. He is politically inept.
The movement of the Lemba neophyte and his wives into touch
with these truths appears to have not only a practical consequence in
their political and economic effectiveness, but a metaphysical con
notation, formulated in the eastern region by the idea of a succession
of Lemba's spiritual substance (kitswa) from one generation to the

next. Many of the references in the songs and incantations, as well as
the mizita figures and the mavungu horn in the Lemba chest (figure
13), refer to Text 4, the eastern etiological myth of Lemba, which
follows.
Text 4
According to the version from Madzia, the creator of Lemba

was a man called Nga Malamu—Nga designating the status of
chief. A s a man, Nga Malamu composed the first n 'kisi Lemba
with the inspiration of a protecting spirit (kitswa) of a deceased
kinsman. When, in his turn, he died, he became the patron spirit
(kitswa) of Lemba. Now Nga Malamu is considered to be a
supernatural being, according to his name, immortal, one who
endures, who is and subsists. Nga Malamu it is who gives force
to Lemba. Kuba, who followed him, fell ill. Nga Malamu healed
him, and he became the first priest of Nga Malamu. The third
founder and protector of Lemba is Magungu. These three per
sonages act in Lemba and by Lemba. Lemba is so old that
generations have been born and have died without anyone recall
ing a time when Lemba did not exist According to some, the
22
founder may have been Yaya Mwaya.
The figures of Nga Malamu, Kuba, and Magungu occur repeatedly in

the ceremony in songs, alongside the creatures and ritual objects such
as nkumbi rat and whiteness. Their characteristics thus come to be
defined more exactly. A few examples suffice to convey this integra
tion of etiological heroes, medicinal objects, and social roles of
Lemba. Thus, in one account of the savanna ceremony early in the
first phase, the priests and priestesses sing "Tremble before
23
Mavungu."
Mbokidi makondo
Kneel, spy
Ka mbakidi akina ko
I have been unable to dance
Wauleno Mavungu
Tremble before Mavungu.
184 RITUALS
Later, during the scraping of medicines ceremony in the village

24
square, the song "Nzambi-God passes" is sung.
Widi, widi
Hear, Hear
Mpemba tata mpedi moyo
Mpemba chalk, father gave me life
Lembolo mpemba, kadi nganga ko
Without chalk, one isn't a priest
Mpemba nkima yikwenda nami
Chalk is what accompanies me
Bandama kwaku
So bow down
Wadi Nzambi kayoka
For God is passing
Nzambi watele vunguteno
God has said: murmur hum, hum
Nsunsu ka kuba ngumbi kwa kudi
Chicken to Kuba, partridge
Nkonso kidi ngolo
Whoever would be strong
Kidi ntedi Kuba diamwana Lemba.
Then Kuba became the Lemba Child.
This song recalls the transfer of the kitswa protecting spirit from Nga
Malamu to Kuba, the second priest Ritual substances such as white
chalk and auspicious sacrifice objects such as chickens and partridges
are mentioned. Nzambi's passing is an indicator of awesome sacred
power. A little later in the all-night dance the participants sing "Kuba
in Ngoma Drum," already cited above. Following a song of gratitude
to the nkumbi rodent, successful striped burrower with joyful quiver
ing whiskers, another song is sung on the savanna hilltop lauding
25
Lemba with its unique humming tone: "God has said Hum."
Tuidi bamintiobo
We are like worms
Ga tutambukila kweto
Here we will burst
Nzambi wasidi, vunguteno
God has said, hum, hum
Tuidi bambwetete
We are like the stars
Ka tuwaka lutangu ko
We are without number
Wauleno Magungu!
Glory to Magungu.
A t once an alliteration of the peculiar Lemba low humming "grum
ble" and the name of the third kitswa of Lemba, the song expresses
satisfaction with the banquet of pork, in terms befitting a well-fed
Lemba priest.
The foregoing textual evidence from diverse sources can be brought
together in a single metaphoric structure to depict the movement
present in the eastern variant of Lemba. Social roles, medicines, and
supernatural heroes or deities are joined to create a semantic fabric
within which direction is apparent. Frequently it is the descriptive
traits of the medicines—nkumbi rats, white chalk, lemba-lemba
plants— that do the defining, with the help of short couplets or phrases
uttered during their use. The effect of this ritual action is however to
move the power of the Lemba Father into the Lemba Son and on to
his wives and patrifilial children, in the same terms as kitswa, the
spiritual substance in the etiological myth, moves from Nzambi to
Nga Malamu, to Kuba, and to Mavungu (figure 15).
Figure 15
Metaphoric association of domains
as depicted in eastern variant of Lemba
Nzambi
Munkukunyungu
& Matwala
n'konzi
Lemba Father ngoma Kuba
Lemba wives
n'kumbi skin
mpemba chalk
Lemba Son — Lemba ancestors
wives musimba skin — •—• ' ' ^
tukula
bracelets
patrifilial
children
mediating movement of
hierarchic order kitswa in
of Lemba roles medicines
etiological myth
Chapter 6
The South-Central (Nsundi,

Bwende) Variant of Lemba
The Lemba Seance in Nseke-Mbanza

9
The south-central variant of Lemba s inauguration ceremony is
based on an account by Fukiau, recognized MuKongo writer and
educator. Because this account, representative of much of the
Manianga region, has been published in Fukiau's Kongo cosmogony
(Nza Kongo)} I need not offer it here but shall proceed with an
analysis of the expressive domains related in it. A few introductory
comments are in order about Fukiau's Lemba account and the condi
tions of its compilation, because it has come under serious attack by a
2
fellow Maniangan author, Batsikama ba Mampuya. Batsikama
charges Fukiau with having invented a personalistic view of Kongo
cosmology and perpetrated it for self-aggrandizement. It should be
noted that B atsikama originates not far from where Fukiau is at home,
and there is more to this attack than its scholarly content. Without
wishing to join in or rehash the debate, I might note that Fukiau could
have averted such charges by acknowledging his considerable debt
to Lemba priest Katula of Nseke Mbanza. Later in this chapter I shall
return to a few of Batsikama's specific criticisms. Here I should clarify
the conditions under which Katula's account of Lemba was gathered.
Early in my fieldwork in 1965, Fukiau and I spent an entire day
with Katula. Katula was well aware of the stringency with which
some Lemba priests kept the secrets of Lemba decades after its
demise as an active institution. Although he shared with them a great
admiration for it, he believed the secrets deserved to be told for
posterity. A s an educator (see Chapter 3), he frequently spoke of the
ancestral past to his students. As speaker (nzonzi) for his clan, the
Kikwimba of Nseke-Mbanza, he utilized his remarkable oratorical
skills in a way that reflected his Lemba training. When Fukiau and I
suggested a thorough coverage oiLemba, Katula willingly consented
to meet with us.
187
188 RITUALS
The "interview" was held, at Katula's request, out in the high

savanna bush and at the outskirts of Nseke-Mbanza where the last
seance of the region had occurred in 1919. Although Katula felt
unsure speaking about Lemba in general, he related with ease and
remarkable memory the seance of 1919. Fukiau's account reflects
this specificity. My own KiKongo was quite rudimentary at that time,
those parts of the interview pertaining to esoteric features of Lemba
songs and symbols passed me by, but Fukiau took extensive notes of
them all, as his published account shows. Answers to some of my
questions also appear in Fukiau's account. Fukiau's published
account of the Nseke-Mbanza Lemba inauguration may thus well be
the most exact and authentic account on all of Lemba. What errors
exist in the account are distortions having to do with extraordinary
detail of some aspects alongside mere hastily-covered generalities of
others.
The spatial-temporal distribution of events in the seance is accurate
because of Katula's in situ presence where the 1919 event occurred.
The logic of the stages, presented by Fukiau with KiKongo glosses,
reflects this precision. Similarly, the visual details of ritual space are
exact. Katula sketched for us the precise details of various Lemba
altars, symbols, dances and ceremonious actions as carried out in
1919. He told us what had been esoteric knowledge, beyond the range
of observation by the uninitiated (bihinga). The significance to our
analysis of some of these details will become apparent in Chapter 8
on Lemba in Haiti, where they reappear, slightly altered, in the
seance reported in the thirties of this century by Price-Mars.
Fukiau's account of Lemba gives mythic hero Mahungu a central
position, not only in the Lemba etiology legend, but in the "Kongo
cosmogony." The role of Mahungu in relation to Lemba varies from
region to region, just as Mahungu's identity is variable, around the
persistent motif of duality. In the foregoing chapter, Mahungu was
third priest of Lemba, represented in a hunting horn. In Fukiau's
(Katula's) account of Lemba, Mahungu becomes the central androg
ynous deity, source of Lemba. The present chapter examines at
length this figure Mahungu, hero of complementarity, both as repre
sented in Lemba etiologies and beyond Lemba. What in foregoing
chapters was analyzed as a "hidden referent" of Lemba causation, in
this chapter becomes the study, at closer range, of the composition of
a deity. In addition to the Mahungu text provided by Fukiau (Text 4),
three further Mahungu tales from a Nsundi enclave in Cabinda, told
The South-Central Variant 189
by Nitu (Texts 5, 6, 7), offer materials for this analysis of a mythic

3
hero reaching across Lemba territory from east to west.
Detailed materials on the figure of Mahungu make up for gaps in
information on Lemba:'s social and economic context, material
already studied in the northern and eastern variants.
The Expressive Domains of Lemba
SPATIAL-TEMPORAL DISTRIBUTION OF EVENTS
Fukiau distinguishes the sites of stages of the Lemba initiation that

dealt with the neophyte as still profane (bihinga) from those sites and
4
stages that saw him either partially initiated or fully initiated. The
first phase, the "tying of knees" (mbundulu a makoto), refers to the
neophyte's attachment to a Lemba priest, his first consecration. It
takes place at a crossroads (mpambu a nzila), an auspicious location
suggestive of the change occurring in the life of the neophyte. As
elsewhere, the reasons for entering Lemba that are given by Fukiau
are diverse. The neophyte may be afflicted, or there may be a Lemba
position in a clan that needs to be filled. After sacrificing a chicken and
some wine, over a fire, the Lemba Father "jumps" (dumund) the
child three times in a common form of blessing, and he is consecrated
(biekwa).
When the neophyte has acquired a pig, the local Lemba adherents
are contacted, and the next phase of the initiation may occur. It is
called ku sinda, or the "rubbing of the bracelets" (kusa n'lunga) and
the "holding up to the sky of the staffs" of the priests, signs which
suggest that the Lemba brotherhood is cohesive and together, that the
new priest will join their ranks. The pig is slaughtered and eaten in a
festive setting, not far from the savanna lodge site.
The initiation moves directly into the next phase of instructions at
the lodge of Lemba (ku londe) in a clearing in the bush. Here the
neophyte priest is given the laws of Lemba, such as prohibitions
against pollution, through eating, acting, or being demeaned in spe
cific ways. The consecration and instruction of the neophyte's wife or
wives is described separately by Fukiau. It appears to have occurred
parallel to the neophyte priest's consecration and instruction.
Fukiau's account suggests the scheduling of the Lemba inaugura
tion in two major phases, separated by a pause of indeterminate length
190 RITUALS
(see figure 16.) This corresponds to the broad lines of the inaugural
reported by other observers in foregoing chapters. One eyewitness of
Lemba in the north Manianga suggested that the entire Lemba
initiation took up about two four-day weeks after the neophyte had
been given the initial blessings. Nkila and Nkoyi, according to this
witness, were "Lemba days" and auspicious points of beginning for
the first full cycle. One may speculate that the two day/night/day
cycles comprising the inauguration each opened on the same day of
the week, Nkila or Nkoyi, leading the first time into a limited-scale
event with only Lemba's priests and priestesses present, and moving
from there into a full-scale cycle with all affines and patrifilial children
present.
In any event, the second cycle opens with double sets of feasts. At
the savanna hilltop lodge, a banquet is served the Lemba priests (and
priestesses?), whereas the neophytes receive crude "unsalted" (lack
ing in meat) "feasts," as a type of ordeal to test their patience.
Meanwhile, in the village, another feast has opened with dancing and
eating. This double feast suggests the food dualism of other accounts,
in which conventional food is mixed at times with medicinal food or
has a closely structured relationship to it. While the village festival is
under way, the priests take the neophyte(s) to the confrontation with
the Lemba ancestors down at a stream, via other phases of this
mystical penetration at the savanna lodge (londe). This corresponds
to the ritual "death" of the neophyte in other accounts, as well as to
the visit to the cemetery to fetch earth from a tomb. This process of
direct encounter with the forces of mpemba, the beyond or the dead,
was the terrible high point of the initiation, surrounded by much awe
and fright. The neophyte needed to be especially consecrated at the
savanna lodge prior to the "descent to Lemba" (nkulumukunu a
Lemba).
The "descent to Lemba" is situated in Fukiau's account within the
terms of his Kongo cosmology. In this view, the village and the beyond
are opposing poles of a cosmic opposition, the former being the realm
of humans, the latter the realm of ancestors, spirits, andNzambi-God.
The former is symbolized by the color "black" and is represented with
charcoal; the latter is "white" (ku-mpemba) and is represented with
luvemba-chaik. This Kongo world is shaped like two inverted disks,
suggesting a type of terrestrial knoll floating on a cosmic water called
Kalunga, within or beyond which is the realm of mpemba. To estab
lish contact with this source of power, a human being must know the
role of the priest. The priest, in turn, must know how to relate to the
Figure 16
Spatial and temporal organization of events in Fukiau's account of
Lemba inauguration (Fukiau, 1969)
savanna hilltop space
stream lodge, ku nearby

londe lodge, ku cross
ku-mpemba village
sinda road
pause, t o
f i n d pigs
CT _=
3 c
2 °"
CP cr
. 2co. a"
crq
[night?]
r- e» 3*
c: cd
;tru
ÉM
3 D) W o
5'
3
p
indeter
minate
I pause
[all night night

dance?]
T3 O (B 2 0 1
^ C ° 3. O. I i s t ss:
^ 3 ? (tao 3 ff
3 o
37=
Q) a>
3.3
0) c r
I
r— J - =3 ?k"a-s>
32 co o cp
T> =3 =j o „ c» re Q» 3 n »
-Co- a. £
J * go<5 Ol
192 RITUALS
ancestors who are in turn in contact with God(Nzambi), ultimate and

unitary power. The gradual initiation moves from mundane black
ness, through red transition, to terrifying mpemba-brillimce. Prior to
their encounter with the white the neophytes are consecrated. After
their encounter they must take an oath and again consecrate them
selves at the altar of Lemba (dikenge, yowa). Finally, their nkobe
and bracelet and drum represent mementos of their encounter.
Fukiau's description of the neophytes' encounter with mpemba
and the following events merits closer examination because he offers
exact drawings—given by Katula—of the ritual spaces, the chore
ography, of these events. In the "encounter with the white" the neo
phytes find themselves face to face with a masked figure on a pole,
decorated in white, red, and black. Surrounding them are the Lemba
priests, now wearing similar masks. The priests are playing their
instruments—probably nkonzi drums, the "ancestors' voices"—and
5
singing such lyrics as the following:
01 Tula!
O, Look!
Tala matebo!
Look at the shades!
O n'kuyu!
O demon!
E Mpungu-tulendo!
God Almighty!
Banganga, ka tuswama ko e?
The priests won't be afraid.
Bakulu ku mpemba!
Ancestors in the White!
At the propitious moment loud salvos are fired from guns nearby and a
priest in the outside circle pulls a string animating the masked central
figure, all of which has the effect on the psychologically-prepared
neophytes of terrorizing them so that they want to flee.
Trembling with fright, they are brought to a nearby site called
Konzi dia Lemba (the "drum" of Lemba) where they are told what
has just befallen them. It is a classic initiation sequence of mystery
and fright followed by clarification. A Lemba song they have heard
before, in earlier stages of the ceremony, is sung to explain the
6
episode:
Ntondele kwami
In gratitude
Na mayedo ma nkumbi
The whiskers of nkumbi rat
MetembiVe
Wiggle, twitch
Wanga mbiVe
Be attentive to the call
Mahungu e
Of Mahungu
Wanga
Be attentive.
Further clarifications are given, how this experience is like the life of
the Lemba priest and that he can expect situations like that which he
has been through. Then he swears an oath of loyalty to Lemba and its
secrets.
Priests and neophytes now move to the savanna hilltop lodge of
Lemba, but before they enter it they circle it thrice, symbolizing the
"life of man" (luzingu lua muntu, zingu = circle), singing the song
comparing the endless circle of life to this circle. The Kongo cos
mology is expressed clearly in this: the path of the sun around the
earth being analogous to the path of life from birth to death and then to
rebirth. One might note that this episode is situated in the same
moment of the inauguration as the "death" and "resurrection" events
in other accounts.
At the lodge a ritual space is prepared, called dikenga or diyowa,
which replicates the dichotomy this world/the beyond with that of
neophyte/priest. The term dikenga suggests a place of circling, the
center of a circling motion, or the hill around which one circles, here
representing the life-cycle process referred to above. Diyowa sug
gests yoba, to bathe or anoint, and is here as in other Kongo ritual
definitions indicative of the boundary between this world and "the
white." The cosmogram suggested in the combined features of
dikenga, place of the life cycle, and diyowa, place of anointment to
the beyond, commemorates the encounter with the terrifying mask in
a more serene and detached manner. The encounter at the stream was
the genuine confrontation with power, now the neophytes are shown
how to relate ceremonially to the abstract reality of power and the
beyond. The priests enter the dikenga circle on one side, the neophyte
194 RITUALS
on the other. On the priests' side is planted the "tree of life," and
alongside the central pit or cross-like trench at the circle's intersection
are further symbols of mediation with the beyond, a small jar or pot of
palm wine and a sack (or sacks) of mpemba chalk and tukula red.
The procession of Lemba now divides into three groups. The
Lemba Father presiding over the seance enters the cosmogram on the
side of the tree; the neophyte(s) on the opposite side; all other
adherents surround the cosmogram in a circle. Father and son face
each other across the cross-like trench, the son holding his staff with
two hands, the father with one, overthe diyowa, sign of swearing. The
father, before the witnesses surrounding him, swears to his son that if
he has any doubts about the uprightness of his father-teacher, that
Lemba may punish him. The father puts together a packet of nine
sticks (vua) constituting "men," and nine more representing "women"
(mizita)—the Lemba wife will receive these later as well. Now the
Lemba Son recites a similar oath to his father before the surrounding
witnesses, as the father pours wine into the diyowa, and passes the
knife thrice under the neophyte's throat. With the two packets of
7
sticks in his hand, the neophyte swears:
Nge tata
My father
Va lukongolo
Before this circle
Tutelamane
We stand
Ngatu yakuwila
If I hear
Maniungu-niu ngu
Rumors, gossip
Ngatu n'samu wambi
or evil reports
Vo lufwa
May death come.
The neophyte now kneels and rubs his mouth in the kitoba mud of
palm wine and dirt in the cross trench and passes the knife thrice under
his throat. After another affirmation or call of the Lemba Father to
the twitching whiskers of Nkumbi and the completeness of Mahungu,
the neophyte is marked with white clay and red powder. Then the
group eats and drinks at nkonzi and returns to the village. There, in
connection with the final feast that Fukiau mentions but does not
describe, the bracelets and the nkobe Lemba are given the neophyte
Lemba couple(s).
MEDICINE, MEANING, AND THE CATEGORIES OF RITUAL ACTION
The Fukiau/Katula account of Lemba's inaugural in the north-

Manianga region does not give specific ethnographic data on the
exchange of goods and symbols in the rite, nor does it delineate very
well the social roles of all participating in the rite. Since these features
have been dealt with in previous variants, it is appropriate here to
analyze in more careful detail the medicines of Lemba and the impli
cations of their composition in ritual acts. Ritual has appropriately
been defined as the manipulation of symbols. In the Lemba cere
mony depicted by Fukiau, the ritual symbols that are used in the
seance reappear in the ingredients of the nkobe, as synecdochic
representations of the processes out of which they arose. Fukiau tends
to explain this process in terms of the "Kongo cosmogony" which
constitutes perhaps the dominant metaphor of ritual action. That
figure, the opposition of the profane human world to the sacred world
of the beyond (mpemba) and its powers, offers a generally correct
explanation of the rite. The neophyte is introduced to the priesthood
of controlling powers in society. He learns, in Lemba, to manipulate
the signs of power, and understands its effects in rhetoric, ritual, and
political and economic action. Some of the supporting significations
need to be amplified, however, because they suggest a rich and
nuanced symbolic understanding in north-bank Kongo culture.
Fukiau's list of ingredients in the Lemba nkobe is based on
Katula's own nkobe, as he related it to us. In most instances these
ingredients define both the ritual actions of Lemba as well as the ideal
characteristics of the Lemba priest and priestess.
The small pot of mpolo medicine included in the nkobe was usually
a mixture of whitish ash from the hearth and luvemba chalk. It consti
tuted a type of lime which retained qualities of cohesiveness and stuck
to the body in ritual settings, and it also had uses as a substitute for
salt. In Lemba the mpolo medicine was used as an initial purifica
tion, and at the end it entered the neophyte's nkobe for his own
treating. It was thus representative of his own status transformation. It
symbolized power (wisa, lulendo), sanctity (n'longo), and clarity
196 RITUALS
(kid). Whenever the Lemba priest appeared wearing or using luvemba

or mpolo medicine, he was understood as a bringer and wielder of
justice, consecration, and victory over obscurity and confusion.
The use of tukula red powder, and its representation in the nkobe in
a small satchel, set forth a cluster of qualities comparable to, yet
contrasting with, those in "whiteness." Whereas whiteness repre
sented absolute sanctity, redness was seen as another type of sanctity
which people feared. The Lemba priest had powers of life and death,
and he was a judge who could perceive the inner nature of things and
events. In many ritual settings both white and red were used, sug
gestive of the duality of power in society: its legitimacy on the one
hand, its fearfulness on the other.
Salt was included in the nkobe as a practical preservative. But as
such it also signified the Lemba priest's use of "righteous indig
nation" (sinsu kia nganzi) at the right moment. Indiscriminate anger
is seen as destructive in society. Precise, well-aimed anger is a valuable
rhetorical tool in the conservation of society. Salt is contrasted, in
rhetorical theory, to oil, a substance for smoothing over.
Nsaku-nsaku leaves and bark (Cyperus articulatus L.=
Cyperaceae) when mashed produce a pleasant incense-like aroma.
The Lemba priest, in like manner, should be pleasant, agreeable, a
person desirous of peaceful relations with others.
As in many Central-African rituals, Fukiau's Lemba séance has
further plant symbols. There is the "tree of Lemba" planted within
the circle of rikonzi. This tree is a banana sapling which represents
the renewal of life, the rebirth of the neophyte, and the source of
human multiplication. Perhaps the most important in terms of its
inclusion in the nkobe is the palm tree (ba—Elaeis guineensis).
Throughout the ceremony palm wine is present as an offering, as a
drink both refreshing and slightly intoxicating when turned, and as a
symbol of transcendence. Called nsamba, from samba, to "cross
over," "transcend," another derivation of which is sambila, prayer,
palm wine is used in the rikonzi ceremony to prepare the toba mud in
the cross-like trench for the oath of Lemba before the ancestors. Palm
wine is the pan-Congo liquid of prayer and sacrifice. The palm is a
central feature in several of the etiological myths later in this chapter.
Several parts of the palm enter the nkobe to signify this all-purpose
resource of the land.
The palm nut (sombo) in the nkobe represents the multiple attrac
tions of people to one with resources. At one level there is the recog
nition that palm trees are resources vital to the survival of humans.
The fibers give cloth; the wine is refreshing and nourishing; the
clusters of nuts give oil; the nuts may—by the time of Lemba's
political zenith—be traded for goods and money. At another level, the
palm nut is recognized because it attracts ants and termites. Thus, in
Fukiau's interpretation, it represents healing among men, the oppo
site of illness which destroys human association.
For related reasons, the palm fiber, raphia, is also included in the
nkobe. It symbolizes the Lemba priest's authority to travel across
local territories others might find hostile and dangerous. A frond of
raphia tied to the staff was supposed to indicate the Lemba priesf s
power, his legitimate right to trade and travel. It must be recalled that
throughout the history of the coastal trade the raphia cloth or ptece
constituted the main currency (mbongo = money).
The cowrie shell (lusungwa) was either inserted in the nkobe, or
mounted atop the nkobe* s lid (see plate 5), or engraved upon a copper
Lemba bracelet (plate 7). Fukiau suggests that its significance was
that of reminding the Lemba leaders of the importance of people,
individuals, and the origin of all persons in childhood. Thus the
prominent should not abuse the weak and the infant.
A final essential ingredient of the n'kobe was the double statue of
Mahungu, whose male facet, Lumbu, was tied to or joined with the
female facet, Nzita. In other variants the figurines of the wives are
called mizita. Sometimes they are combined with figures or emblems
of the etiological deities such as Kuba or Nga Malamu. Here the
representation of mystical origin and femaleness are joined into one
object, Mahungu, whom Fukiau interprets as primarily a representa
tion of the complementarity of opposing forces in the universe—
male/female, strong/weak, violent/peaceful, and so on—required to
maintain an effective, stable social order.
Although all the foregoing objects are by their presence in the nkobe
given a symbolic prominence and cohesiveness, it is possible in the
lyrics of the ritual to identify a dynamic metaphoric movement behind
their composition. That is, they are not just isolated symbols with a
given meaning but moving elements in a coherent universe of
domains, like that which was identified in other variants already
examined (see figure 15).
Songs introduced in the opening stages of the "Binding of Knees"
(the first anointment or medicine) and repeated throughout the sub
sequent stages of the ceremony amplify this universe of domains. The
first song outlines a relationship between humans, ancestors, and
8
Mahungu.
198 RITUALS
Kubele bakulu
In the time of the ancestors
Kubele bantu
In the time of humans
Nsdngo na nsdngo
Scores of copper bracelets
Se zibetana
Chimed
Tuaniungufeno kweto
So let us do our humming
(distinctive Lemba hum)
Kubde bakulu
In the time of ancestors
Kubele bantu
In the time of humans
E Mahungu e!
Oh Mahungu!
Nge' bahungila'
You preserved them
Badianga
So they could eat.
The structure of the song is one of double couplets in which an initial
phrase is repeated in each couplet and followed by different actions:
copper bracelets chiming and Mahungu preserving life. As already
suggested in the analysis of the songs of the northern variant, such
substitutable referents constitute the shifting or moving action in a
metaphor. We could then rewrite the song in the following manner.
(bracelets) (chimed ) (in time of) (ancestors)

(Mahungu) \ preserved j (life of j j humans j
Bracelets chiming would stand in a relationship of metaphoric move

ment to the action of Mahungu preserving the society of humans. The
verb hungila, describing the action of Mahungu, suggests verbal
punning. The verb designating the unique low throaty hum of Lemba
singing—nunga or hunga—ties the pun and the metaphor together
another way. It is likely that Lemba priests had developed other
esoteric meanings for these suggestive formations.
The opening rite contains two further songs, one defining Mahungu
in relation to the nkumbi rat, the other seeming to link the entire ritual
9
to the "passing of God." The text of the first song follows:
Ntondele kwami
I am grateful
Na mayedo ma nkumbi
That nkumbi's whiskers
Me tembiVe
twitch
Wanga mbiVe
Listen to the call of
Mahungu e
Mahungu
Kuwanga ko / buna bwasisa bakulu beto
Don't you hear, what our ancestors left us?
Kuwanga ko e
Don't you hear?
Wanga mbiVe
Listen to the call of
Mahungu el
Mahungu
Wanga
Listen.
This song could be rewritten as follows:
nkumbi's twitch
whiskers
(Lemba) rejoices
ancestors have told us
Mahungu calls
The implied meaning would be that nkumbi's whiskers stand to
Lemba in the same relationship as the ancestors to Mahungu, perhaps
suggesting that nkumbi is a sign or referent of the spiritual power of
the ancestors and Mahungu. The opening episode ends with the other
10
song "God passes."
Bandama
Bow down
200 RITUALS
Nzambi kayoka
God passes
Nima na mbyo
On the back as the abdomen
Luse na mbyo
On the face as the abdomen.
Fukiau's interpretation of this song, chanted as the neophyte is
smeared with luvemba white and tukula red, is that God is conse
crating the neophyte, but that God is so almighty that no ordinary
mortal can see Him with his own eyes. To " see" the power of God, the
neophyte must bow down and cover himself "with fear" lest he die.
A metaphoric statement emerges like that in the eastern (Lari)
variant (figure 15). A hierarchy of powers ascends from humans to
ancestors, to Mahungu, and to God. Within the human world a clear
dichotomy is made between the Lemba priests and priestesses on the
one hand, and the neophyte couple on the other. Mediating ritual
objects such as masks, the nkumbi rat and its twitching whiskers, the
objects of the nkobe such as luvemba chalk and tukula red, and the
bracelets are poised between the two (figure 17).
These songs and ritual objects tell us which powers Lemba sought
to organize and control. In the next section the myth-making process
itself is examined, in terms of a number of Mahungu texts, the first
used by Katula to explain the orgin of Lemba; the others from a non-
Lemba source which clarifies the nature of Mahungu as mythic figure
in the rigion with which we are concerned.
THE LYRICAL: THE MANY D U A L F A C E S OF M A H U N G U
In ritual representations, lyrics, and myths throughout the Lemba

region Mahungu's identity varies from that of third priest of Lemba
represented in a hunting horn n'kisi (eastern variant), to a set of
hunter-brothers, to the hunter and the hunted, to healer and sufferer,
to androgyne (Fukiau's account), husband and wife, and father and
son. Whatever else this mythic figure may be, it is definitely a creature
embodying duality and complementarity.
Efforts to identify Mahungu through etymological exercises have
not been very successful. First of all, across the region the presence of
the varying/h/, /g/, /v/, /b/ phoneme creates enormous problems of a
purely formal sort. Furthermore, in the lyrical context of word use
new meanings are constantly being created, as is clear from the songs
Figure 17
Metaphoric association of domains
in south-central variant of Lemba
Nzambi-God
Mahungu
Lemba
priests & i
Ancestors
priestesses
Humans
Lemba
neophyte
couple
the profane
hierarchic order mediating hierarchy

of Lemba roles ritual of
objects mystical
powers
just analyzed. Thus, Mahungu is said, in Fukiau's account of the

Lemba ritual, to preserve (hungila) the Lemba order, while a pun or
rhyme is made in the same song on the murmuring hum of the priests
11
(tuaniungut'eno). Phonological analysis has usually followed the
same method in order to ascribe a word's meaning as the method
native speakers and singers use to invent new meanings. But such
analysis can at best lead to circularity and confusion. Thus, to explain
Mahungu, Laman uses the verb hungula, an exchange which sets in
motion a ritual or relation, bringing two persons or groups together.
But there is, for Laman, another derivation, vunga (syn. wunga),
which may mean to blow or cause noise, as air blowing through
12
trees. It is this latter understanding on which Fukiau has focused, in
which terms Mahungu becomes the double being whose spirit either
"blows softly" as a creative force or "blows violently" as a
destructive force. Mahungu, he concludes, is a being who thus
13
embodies contradictory and opposing forces. Batsikama invokes the
term hungu (syn. wungu or vungu), force or power, thus Mahungu is
he who holds force or power. For Batsikama, Mahungu (Mavungu) is
the ancient priest who consecrated the rulers of the kingdom of Vungu
14
which once held hegemony over north-bank societies.
In ritual, lyrical, and mythical contexts, all of the foregoing
meanings of Mahungu appear, therefore it is this contextual usage
which must be further analyzed. First, Fukiau's Lemba-relaied
202 RITUALS
15
Mahungu text will be studied; then, three somewhat less closely
linked to Lemba will be examined.
Text 5
(1) Ntangu yayo yavioka, Mahungu wakala mu kiese kialunga.

In times past, Mahungu was in complete happiness.
(2) Kakala ye mpasi nkutu ko; wakondwa mpisulu mu keto;
wakondwa nadede mpisulu mu nzinunu; wakala walunga
mu mamonsono.
He had no pain; he knew no jealousy; he had no knowledge o f
hatred; he was complete in every respect.
(3) Wakondwa konso banza mu luzingu; wakondwa banza mu
nsatu zazo zena zakangama mu kimuntu.
His life caused him no cares ("thoughts"); he had no concerns
for the needs in life known to man.
9
(4) Mu kinzungidila kiakundwa kwa Mahungu mwamena n ti
wabikwa vo "mutie-mpungu" vo "ba-dia-Nzambi".
In the surroundings where Mahungu lived there grew a t r e e
called Mpungu's tree or God's palm.
(5) Nti wau wajaka wabsusu beni mu nsemono ya muntu; mu wau
mwavaika mpila ngolo zakala ye lendo beni ye zalenda
vambisa nsemono ya muntu eto wantete.
This tree was greatly feared in the human creatures; from it there
emanated a special force; it had special power in the division o f
our first created man.
(6) Lukufiyen'tiwowo ka vafinama konso lekwa kiamoyo nkutu
ko, kadi buna i nsisi kwa kiau.
Close to this tree no living thing dared to approach; for it was an
aweful thing for them.
(7) Mu kuma kia mpeveyalebakana (ya nkento)yena mu muntu,
muntu wazina moyo mu finama ye n'ti wowo, "ba-dia-
Nzambi, " mu mona ye bakula mankaka mu wau.
Compelled by the feebler spirit (female) in man, man desired t o
approach this tree, God's palm, in order to see and understand
more of it
(8) Mahungu wasindusulwa kwa ngindu zozo zalebakana mu

zungana kinzungidila ye n'tL
Mahungu was driven by these weaker thoughts to circle around
the tree.
(9) Kansi mboki, bu kameni lungisa nzunga yantete ya n'ti
wam'vimba, wapamuka kadi wamona vo kasidi diaka mosi
ko kansi bole mu mbonokono zole zanswaswani:
y nkentoye
bakala.
When he had made the first full circle of the tree he was
horrified, for he saw that he was no longer one, but two beings
of different natures: female and male.
(10) Wonga ye kiadi kiababwila bu kabalendi kala diaka se
umosi ko.
Fear and sorrow came over them when they could no longer be
one.
(11) Batalasana napii, bayindalala.
They regarded one another pensively, and reflected.
(12) Lumbu (bakala) wakitala ye banzila ndambu andi ya
kikento yatina ye bobo Muzita (n'kento) wakisuna mpe ye
tomba ndambu andi ya kibakala yatina mu yandi.
Lumbu (male) saw and thought about his female half that had
left him and Muzita (female) longed also and sought her male
half that had left her.
(13) Bu bameni yindula, bavovasana mu zungulula ba-dia-
Nzambi mu lusunga lutalane ye Iwantete.
After reflecting on this they decided to return around God's
tree in the opposite direction.
(14) Nzunga yoyo bu yimeni, basikila bole kaka bonso bateka
kala: n'kento ye bakala.
This circle completed, they remained two beings: female and
male.
(15) Tuka ntangu yoyo bau bole batombasana: n'kento watomba
bakala kad'i ndambu andi; ye, bakala n'kento.
Since that time woman and man seek each other; woman seeks
man because he is half of her, and man seeks woman.
(16) Nzinunu zakota mu muntu, mu ntombasani yoyo, mpasi vo
bana vutuka kala va mosi se umosi, mu nkadulu au yantete
yan'longo.
204 RITUALS
In this original mutual longing for each other, they tried to

return to be one, the being they had first been in purity.
(17) Muntu wakitula mu lujimpu Iwanene mu solula mvutu
yifwanene kwa nsamu wowo wa mvambunununu a muntu
wantete mu bantu bole banswaswani.
Man set out on a great search to find an adequate reply to this
puzzle of the division of the first man into two different persons.
(18) Muntu wayenda kwanda ye mabanza mandi ye, va
zimunina, wabwa mu banza dia nkwedolo: bundana landila
ngwawani a tuzolo tole tuanswaswani
Man sought far in his thoughts, and finally came up with the
solution of marriage: the union of two separate beings in
mutual trust.
(19) Landila nkabununu a Mahungu mu nsemono zole za-
swasana, nkento ye bakala, nkento ye bakala bayika
kinsona mpe.
Following the separation of Mahungu into two creatures of
distinct natures, man and woman felt alone.
(20) Yenge ye luvuvamu biatina mu bau.
Joy and hope had eluded them.
(21) Bau bole batombasana ye bavwa n'kinzia nsalasani va kati
kwau.
The two sought each other and had need of mutual help
between them.
(22) Bakala, mu kibakala kiandi, wakivanga kesa ye kinwani va
kati kwa semwa biabio; n'kento wayalwa kwa wonga ye
lebakana.
The man, in his "maleness" was inclined toward violence and
aggressiveness amongst all things created; the woman was
overcome by fear and weakness.
(23) Ndieu, nkento, wavwa n'kinzi a lusanusu Iwa bakala bu ka
kalenda zinga mu kinsona ko va kati kwa mavanga mansisi
ma nsemono yayonsono.
Thus the woman had need of the man to be able to live without
fear amidst the terrifying acts of all creatures.
(24) Wanama mu tambi bia bakala mu kum'vana Iwaka
lusadusu; bau bole bakunda va kimosi: bakwelasana.
It is a trait of the man to give her this help; the two lived
together: they married.
(25) Mu kuma kia tezo kia lusadusu Iwanata muntu-muntu mu
mbundani yoyo.
This is the measure of help each brought the other in their
union.
(26) N'kento wayika n'sadisi kaka kwa bakala diandi.
The woman became the helper of her man.
(27) Mu kuma kia nsemono au yakala diswasani, bau bole
bazayana ye mboki, n'kento wayaka (wabaka ntunda) ye
buta.
Because of their creation according to difference, the two knew
each other and the woman conceived and gave birth.
(28) Mu bila kiokio, n'kento wabika bakala diandi n'lumi ye
bakala wabika n'kento audi m'buti, kadiyand'i "m'buti-a-
m'fuma, ye mindimba" (mbuti a bakento ye babakala).
For this reason the woman called her husband genitor and the
man called his wife génitrice, for she is the "mother of the
mfuma and mindimba trees" (females and males).
(29) Mu nkalasani yayi, nkwedolo, muntu wamona nkièvo vo
watungulula nkadulu andi yantete yan'longo yavila tuka
nzungununu a ba-dia-Nzambi vo mutie-Mpunguye, muyau,
nkwedolo, muntu wasolula nzila yanayaki-yaki mu niekisa
n'kun'andi mpe.
In the state of marriage, it was as if man had recreated the
original sacred condition that he had lost in the encirclement of
God's palm or the Mpungu tree; in marriage, man discovered
the easiest way to multiply his family.
(30) Muntu wavisa vo nkwedolo i nzila yaluta mbote mu niekisa
kanda diandi.
Man understood that marriage is the best way to multiply the
numbers of his clan.
(31) Imubilakiokio, mun'kungamiandi, wabadikasevilanzaya
bibulu yikondolo nkwedolo.
For this reason, in his song, man began to mock the world of
animals lacking marriage.
206 RITUALS
(32) Ya nsusu na kakwela? / kakwela ko / Muntu na muntu /

kwela kwandi / kakwela ko.
Does the Chicken marry? No, it doesn't marry, each the other.
Does it marry? No, it doesn't.
(33) Ya mbulu na wan'kwela? /Kakwelwa ko/wonso ka wonso/
kakwelwa / kakwelwa kwandi/ kakwelwa ko.
Does the jackal marry? No, it doesn't marry. Anyone can
marry it, it doesn't marry, it doesn't marry.
(34) Na-nsinsina'wansompa?Kasompwa ko/konso ka konso/
sompa kwandi/ kasompwa ko.
The wildcat, who married it? It doesn't marry, anyone, yes
anyone can marry it, it doesn't marry.
(35) Na-nsesi na'wan'yula? / Kayulwa ko / wonso ka wonso /
kayula kwandi / kayulwa ko.
The little red antelope, who asked for it? It hasn't been asked.
Anyone, yes anyone can ask it, it never gets asked.
Although Fukiau uses this text to explain the origin of man and
human society, its immediate referent in the Lemba ceremony is
the n'konzi oath enactment during which Lemba "sons" follow their
"fathers" around the tree to seal their initiation to the mysteries. A
comparable rite is held for the neophyte wives. In addition, the bound
figurines of Lumbu and N'zita depict the male and female parts of
Mahungu, and the complementarity of sexes in the Lemba marriage.
The myth introduces an important further dimension into the
relationship of father to children, and male to female, that of the
unfolding of complementarity from androgyny. As most Kongo
myths, this one is constructed in three stages: (1) an original condition
of homogeneous authority, self-sufficiency, purity; (2) a lapse of time,
passage over space, or some differentiation such as the dispersal of
clans or here, "dispersal" of the sexes; and (3) an actual con
temporaneous condition, human awareness of present flaws in view of
past perfection, some attempt at solution. Fukiau argues that the aim
of the myth's representation by two bound figures in the nkobe is to
seize the moment of greatest tension and dynamic strength in their
relationship, the moment of greatest complementary opposition
between male and female, strength and weakness, creativity and
destruction. At this moment the "powers are bound" (ngolo
zabunduswd), synthetically restoring the vision of Stage 1, the
original perfection, of the myth.
In ritual contexts, these principles are clearly expressed in the

rotational pattern of dancing. There is no agreement over the direction
of Stages 1, 2, or 3, but if clockwise dancing is used to depict the
separation of powers or qualities from an original unity, then
counterclockwise dancing will indicate the attempt to reintegrate
these powers or qualities. In contemporary Kongo ritual both
directions are used. Healing prophets dance around their sufferers one
way to "undo the illness," then the other way to restore health.
Funerary ceremonies dance one way in the village, then the opposite
way in the cemetery.
Dance directions and stages in narrative have the identical purpose
of articulating the relationship of unity to diversity, the processes of
endogeny ("within one") to exogeny ("between two"), and in Lemba
this was seen with regard to marriage. But at a philosophical level the
problem was more general, addressed to the relationship of the one to
the many, of peace to violence, of strength to weakness, of social order
to disorder, and similar issues. This approach to cosmogony is
characteristically Bantu. Unitary power must somehow be related to,
and preserved in, diversity and multiplicity. In the next Mahungu text,
the problem of endogeny and exogeny is depicted with reference to the
culture of food: who eats whom, and what this has to do with social
order.
I have already suggested that Mahungu, an important north-Kongo
deity of duality, existed in local cosmologies across the entire Lemba
area. Three texts from eastern Cabinda represent this wider corpus.
Text 6
(1) Va kala muntu evo bantu bwadi; kifu kiau kwe vondanga
bantu bangana badi muntsi.
There was once a man or two people whose custom it was to kill
other inhabitants of the land.
(2) Bu bameni ku vonda, ku sasa muntu wowo; batunga bianga
biodi.
When they had killed someone, they cut him into pieces, and
divided
(3) Bu batunga bianga bibiodi bu ba meni kubasa muntu
kukaba kutula kikuku va kianga kinka dedi.
208 RITUALS
the meat into the two baskets they had made, and dried it over
the hearth fire.
(4) Buna babasalanga pila yoyo bayolukanga bayolukanga:
"nanie? nanie? wo? nani e?"
A s they would be treating their victims thusly, passers-by
would call: "Who is there? Who?"
(5) "O miMavungu eManga nsitu, muna rn manga diambu ngie
kuviokila mu dikubu."
"Ah, it is Mavungu the ogre of the forest. To avoid trouble pass
by in the trees beside the road."
(6) Batu Banka "e e yisa, nkwenda kwama ko yayu."
Some would say, "I'll not go to the side."
(7) Yuwa ti mi Mavungu ukituka buyoba kumanga kwaku, ku
viokila mu dikubu.
The answer would come, "Listen, it is Mavungu; you're crazy
to come here, pass by the side."
(8) Mi yisa nkwenda kwama ko, evo mi yisa ku kwenda kwama
ko.
And again they would say, "I'll not bother to go to the side."
(9) Di bakala be dio dibeki diela bu vioki tsikwandi. Ba salapila
yoyo.
Wise men would pass by to the side. And so it continued.
(10) Buna muntu nka bwesa vioka. Bu bayolukanga kuna "Nani
nkolukanga kokwe?"
Then another person came along, and he too called out as he
approached: "To whom am I calling out there?"
(11) Ti o mi Mavungu.
It is Mavungu.
(12) Titika biyobwe. Nkwe yendanga kuna.
He was foolish, and went there.
(13) Nkwe sa basika vana, ka bianga bibiodi, bisa nsimbidila.
When he arrived there, he saw two baskets; he was grabbed,
(14) Kwitsa va kala va befu yinu bonso.
and told: "come stay with us here together." (He did)
(15) Buna bavingilafwatinikilumbu kisa kwila ko. Muntu waka
vana.
They waited a little and one day they saw another person
arriving.
(16) Bembi mu ntubanga vo "O mi Mavungu, utitika biyoba
manga diambu vioka mu kubu.
And they said, "we are Mavungu, do not be foolish. To avoid
trouble, pass by the side.
(17) Buna befu tu kedikwitu vave nandibefuyau e. Tu nvondanga
bantu babe.
We are ourselves here, we are who we are. We kill people.
(18) Buna kiange kiki kiame kiange kiki kiandi
This is my basket, that is his."
(19) Buna bianga bibiodi ngie nzitu ti e yimweni; a buna nandi ti
tidi kwenda ti e e tubantu vava.
Then they saw that their two baskets were empty; and he said
they would need to get some more people.
(20) Befu yinu ti befu tu nvondanga bangana, buti ngie wa
n'nanguka kwaku wa nkwenda ku bwala; mi yibesa ku
kamba kwaku ko ti mi Mavungu ukituka buyoba viokila mu
dikubwe.
We will go by ourselves to kill some more, you may go to the
village [for other food], but don't tell about us, that here is
Mavungu who warns people to pass by the side.
(21) Buna bakala vana babakanga bababwadi.
[When they heard] they went and seized them both.
(22) Ah! Befu na kubalanga ti bantu banvinha ko be mu nvon
danga bantu.
Ah—we are not madmen to have lived here killing other people.
(23) Bababwila babanata bababwadi bayenda peleso.
They fell upon them, and carried them both to prison.
(24) Pila mweka. Pila mweka befu yinu bana ba Nzambi ba nka.
And so it happened. So it is with us children of God on earth.
The violent antisocial side of Mahungu, merely alluded to in

16
Texts 4 and 5, here becomes the focus of the tale. BaKongo
storytellers frequently generate such antiheroes to illustrate a social
norm of which the negative figure is the inversion. Not having heard of
210 RITUALS
Lévi-Strauss's concept of the "symmetrical and inverse" possibility

of myth, they call these antiheroes "diabolos" referring to the apt
metaphor of the photographic negative, which reveals the mirror
image inverted in every way. Real witches, less comfortable to speak
about, have the same qualities. The tale of Mavungu the ogre of the
forest was undoubtedly meant to entertain as well as teach, and to
frighten just a bit. Nitu's narration has numerous ellipses in it, as if
everyone present knew the story well enough to fit the pieces together
around the gaps.
There is no historical evidence thatBaKongo were cannibalistic;
however such tales as this confused missionaries and travelers.
BaKongo did, and still do, use the figure of speech "eating" to speak
about consumption of all kinds, whether it be food, medicine, money,
or another's physical and psychic energy. An overabundant con
sumption, or an inappropriate consumption, is defined as witchcraft.
In terms of this very abstract understanding of "eating," the present
tale defines propriety by illustrating its negative inversion. Mavungu
lives in the forest; human beings live in villages. Mavungu eats the
flesh of his own kind, and only that; humans eat flesh of animals and a
variety of other foods. It is not clear from the text whether there is one
or two Mavungus, and if the latter, whether the traveler who joins
Mavungu perhaps becomes the second. It is evident, however, that
this traveling middle figure mediates the civilized inhabitants of the
village and the cannibalistic ogre of the forest
In this text, as in the foregoing, the two Mavungus (Mahungus) who
are otherwise so dissimilar have in common their initial endogyny—in
sexual identity, in "eating" their own kind—and the way they are
transformed into figures of exogeny—sexual differentiation and mar
riage, and prohibited cannibalism, the tacit enactment of which would
17
be eating other species and foods. In the next myth text, even more
complex dualities will be heaped upon Mahungu.
Text 7
( 1 ) Ba tunga bwala. Bu ba tunga bwala ba buta bana, bau badi vo

bwala.
They built a village, and when they had built their village, they
had children.
(2) Buna bu ba buta bana, unka u kamba a mwayi:
Then, a certain one said to his fellow:
(3) "Bwabu mi buta bwisi mbote ko, pana dionga diakupasiye

lotsa bibulu du nsitu buna unvana dionga diandi."
"My weapon is inadequate, give me your spear to go hunt wild
animals in the fields."
(4) Bu ka uenda u mona zinzau beti dia; ubonga dionga ubanda
mu nzau.
He went out and saw some elephants eating. He took the spear,
and hit one.
(5) Nzau unata dionga.
The elephant carried away the spear (lodged in its side.)
(6) Kunlanda menga kulanda menga, we bwa mu nlangu u
n'neni wenzi kwandi.
He followed the trail of blood until it disappeared in a large
river.
(7) Ah! Pila kiadi dionga di mwayi di ma zimbala.
Oh, what sadness, the spear of his fellow was lost.
(8) Buna uyitsa kwidi nzadiandi: "Mwayi, dionga dicimbidi; ti
tola mwayi ti dionga diama disi mbakana kwe?"
Then he returned and told the owner, "brother, the spear is
lost! How will I find it?"
(9) "Diambu ve komVaku yiyi kwela ngie ku kwela komb'ami,
mi ku kwela kombaku."
"Do not worry. I'll take your sister in marriage [as payment],
and you will marry my sister, and we shall be intermarried."
(10) "Ka diambu ve komb'aku yiyi kwela, ki lendi kuwa ko!'*
"I cannot hear of marrying your sister!"
(11) "Ka diambu evo ku botuka, ka botuka kwandi nate mi yi
m'mona dionga diama."
"Either that, or I shall leave, unless my spear is found."
(12) Buna kiadi kibwa kwidi nzadaindi nandi nzadVama ma
tsutika mu dionga, ulambalala mu bwilu.
Then great sadness came over his brother, the one who had lost
the spear, and he went to sleep for the night.
(13) Buna ulotu ndosi.
He dreamed of a voice saying,
212 RITUALS
(14) Baka ngengeyoyo kileko kiyolukanga kio buna kwenda mu

nzila kuvulanga ngenge bantanguninanga kuvulanga kileko,
f
"Yayi nzila yenda kwami?" Buna m mona na kanvwena
"swi, swi" wakayenda.
"Take this hunting bell ngenge and go on a journey asking it to
answer questions:'' Shall I take this road?' And if you see that it
says 'swi, swi' do not take the road.
(15) Kansi buna m'mona kunkuvula, "Yayi nzila be mi yenda
kwami e?" buna nandisika tuba "nge, nge, nge, nge"yenda
mu nzila yoyo.
But if you ask 'shall I take this road?' and it speaks 'nge, nge,
nge, nge' then take the road.
(16) Buna bu ka yenda nzila yoyo utumamana bonso bu nkamba
ndosi
If you follow these instructions, you will see your dream
fulfilled."
(17) Buna bu ka yenda mu nzila umona divambu.
Then he took to the path, and soon came to a fork.
(18) "A ngenge, yayi nzila yenda kwami e?" "Nge, nge, nge, nge,
nge, nge."
"Oh ngenge, shall I take this path here?" "Nge, nge, nge."
(19) Buna ka diata, ka diata kwe sadilanga pila yoyo zinzila
ziazio uvuka kwandi.
Then he walked and walked, approaching all paths he came
to in this wise.
(20) We bata nlangu un'neni bwatu bwidi vovo, "A a ngenge, yi
kandama mu bwatwe?" "Nge, nge, nge, nge."
At long last he came to the great river, and saw a boat. "Oh
ngenge shall I launch out in the boat?" "Nge, nge, nge."
(21) "A ngenge yenda ku mongo nlangwe?" "Swi, swi, swi."
"Oh ngenge shall I go upstream?" "Swi, swi, swi."
(22) "A ngenge, kwidi yenda? A ngenga yenda ku wanda
nlangwe?" "Nge, nge, nge, nge."
"Oh ngenge, which way? Shall I go downstream?" "Nge, nge,
nge."
(23) Buna nandibu ka yenda, buna ubonga ntievo dilemo uvwila.
Buna uyimbilanga "Mavungu palabanda, e Mavungwe
palabanda, e Mavungu palabanda, e Mavungu palabanda,
e Mavungu palabanda.
Then he took the oar as he departed, singing "Mavungu going

downstream, Mavungu going downstream, downstream,
downstream... ."
(24) Buna bwatu bweti diata ngolo buna bu ka diata, ubaka va
kielo kinka batu bapwedi bisanzu bidi vana.
Then he was moving along rapidly in the boat, when he came
upon a dilapidated house with many people in it.
(25) "Mavungu!" Mavungu "nhinga!" "0 witsa mu diyamba
yitsa kwaku yitsa tapuka. Kanga bwatu, witsa mwa kwaku
diyamba."
"Mavungu" he heard. "What?" "Come over here to smoke
some hemp with us. Tie up your boat and come smoke hemp."
(26) "Ve minu a, a ngenge yenda ye nwa diyambe?" "Swi, swi,
swi."
"Oh ngenge, he asked, shall I stop to smoke hemp?" "Swi,
-swi, swi."
(27) "Yenda ku wanda nlangu?" "Nge, nge, nge, nge, nge."
"Then shall I continue downstream?" "Nge, nge, nge."
(28) Buna udiata usalapila yoyo. Buna udiata, uyimbila Iwimbu
Iwandi "Mavungu palabanda."
Then he continued in this manner, singing as he moved along
his song "Mavungu going downstream."
(29) Buna udiata ke basika va dikabu di ba kieto bandumba,
buna "ti una vioka vava kulendi baka ndumba evo nkietu u
ulenda sakana yaku ko?"
Presently, as he moved along, he came upon a band of young
women who called out," can't you cross over here and take one
of us to amuse yourself?"
(30) Buna nandi uyuvula ngenga. "Yenda kwama kune?" Swi,
swi, swi.
Then he asked ngenge, "Shall I go over there?" "Swi, swi,
swi."
(31) A ngenge, "Yenda ku wanda nlangu?" "Nge, nge, nge, nge"
And again, "Shall I continue downstream?" "Nge, nge, nge."
(32) "Te u manga mu mavanga mamo ma bantonta u manga."
"Not until you reject everything offered you."
(33) Buna uyenda umona va tapukila nzau na menga.
214 RITUALS _____
Finally he saw, as he went, the place where the elephant

trailing drops of blood had gone ashore.
(34) "A ngenga yi tapuka vavava?" "Nge, nge, nge, nge. "Buna u
yenda vana.
"Oh ngenge, shall I go ashore here?" "Nge, nge." So he
pulled ashore.
(35) Ngengeyandi u yolukanga "nge, nge, "pasi ba zaba ti nandi
nganga: "nge, nge" "a mboti tata, mboti, mboti, mboti "
There ngenge sounded ahead of them so that others should
think him to be a healer. "Nge, nge" "Oh, greetings, sir,
greetings."
(36) "A ngie widi nganga?" "Nhinga a ko." "Kwidi bedi ko yi
bedi, katsi mi nganga yineni beniyidi.
"Oh, are you a healer?" "Yes indeed. A very famous one, a
very famous healer indeed."
(37) Buna o fefufumu bwala baka bazebi ko muntu wowo vo
nandi u ba na dionga diandi. Uba na dionga di ka tsuma
nzau ka banzei kwau ko, buna "bejwé mutu widi vava boba
yenzi ku ku die banvengi pasi kuné.
The village chief did not know that this was the man attached to
the spear that had injured the elephant and so he told him,
"With us here is a person who was injured and he is in great
pain."
(38) Buna ti ngie widi nganga ti ulenda botula nsongu wou, beju si
tu kufuta.
"If you are a healer you could remove his suffering. We would
pay you."
(39) Ye nandi mboti kwandi A ngenga, "yi lenda kwami buka
nsongu wou?" Ngenge, "Nge, nge,
"All right," he accepted. Asking ngenge, "May I help this
sufferer?" Ngenge said, "nge, nge"
(40) "Ti mu kwa nkisi ama bu ka ntubanga ti o a benu bwabu mi
y a mbuka nsongu wou. "
"In listening to my instrument, I can tell you that I shall heal
this sufferer."
(41) Buna a ngenga, "tibapenibidiayilenda dia kwami e? Pasi yi
buka nsongo?" "Nge, nge, nge, nge."
Then he asked ngenge, "What they pay me for healing, may I
take it for myself?" "Nge, nge, nge."
(42) Uyoluka. Buna landila diodio nandi u yenda mu nzo u tala

nsongo wou ti "e benwe bonso mwa ku u lu wetisia?" "Ah!E
mame e mame, kadikwica, kadikwitsa. TiIwitsitsisikafwa.
Then he entered the house to look at the patient, asking
"Which one of you in there is the sufferer?" "O mame,
O mame, come, come quickly, or I shall die," he heard.
(43) E a tata sala kwaku. Buna nandiuyenda uye kota ku nzo. Na
nzau yoyo u mona va kotila dionga u bonga bilongo bonso
(datolo) ou(nganga) u tula vanapasi dionga dilebila. Buna
ubwata pila yayi ngolo.
To the father he said "wait," and he entered the house. There
he saw the elephant injured by the spear; he took medicine and
like a healer put it on the spear wound to diminish the pain, and
he pulled hard on the spear.
(44) "E mame, e mame, e mame" dionga di vodikanga divodi-
kanga. Buna nandi u tuba a beno pananu kwala bwabubu
mambu mameni.
"O mame mame," he cried as the spear slowly came out.
"Now," he said, "find a wrapping for the wound, the case will
be over."
(45) We twala kwala ka luzibula twala vava divudu ka, luzibula.
Buna nandi banvana kwala kuna nzo u zinga pasi u zinga
dionga dio diaka monika kwidi bantu.
And they brought him a wrapping which he wrapped around
the wound, instructing them not to open it. He also wrapped the
spear so it would not be seen by the people.
(46) A a "ngenga zibika mesu mau."Buna ngenga u zibika mesu
ma babo. Buna nandi u nata dionga diodio kwe sweka kuna
tsomo ku ka yizila muna bwatu kwe tula, buna u vutuka.
"Ngenge close their eyes," he said. Then ngenge closed their
eyes, and he secretly carried the spear over to the boat and laid
it there, then he returned to them.
(47) "Bwabu muefuti" "E befu twa kufute. Bonga tsanga kitebi be
yo.
"Now you can pay me." "Yes, we will pay you. Take this
banana sapling.
(48) Kwe vataya muna kilumbu bonga tsanga kite beyo kwe vata
ya muna kilumbu ku mena muna kilumbu kubuta."
On the same day as it is planted, it will bear fruit, on the very
same day."
216 RITUALS
(49) Buna banvana mikailu minkaka u yitsa kwandi. Buna ka

mana ku Iwaka vovo u vuka pe pila most Bantumisa mu
ndambu na ka zola kwenda ko.
They gave him other gifts also to honor him. Then he returned
in the same manner as he had come, but he had been so honored
he hardly wanted to leave.
(50) Buna ulwaka kwidi nzadi'andi, a a polo ti ndamba mwayi,
"nzadi, twala dionga diama."
When he arrived at the place of his brother to whom he had a
debt, this one asked "brother, where's my spear?"
(51) "Ti dionga disiko di sa bakana ko." do, "ka tuna tuban
zipolo ko."
"I didn't bring the spear" he feigned. "We cannot discuss the
debt," the other replied.
(52) Buna nadi "mwayi bonga dionga diaku." Buna u tuma
nkietu'andi twala di kieta, vata tsanga yoyo.
So he said, "brother, take your spear." He also sent his wife to
bring kola nuts and the banana sapling.
(53) Tsangayo ba tsimbukuluyima kumbu menayima buta. "Ah!
ah!"
They planted the sapling, and in one day it produced ripe
bananas.
(54) Nzadi, "siala mboti mi yima kwenda ku diangala ku ku
kangala. Siala mboti ukeba bwala. Keba na cikomba ciaku
mi yi ma kwenda ki diangala."
"Brother," said the one, "I am taking a walk to the forest.
Farewell, guard the village, and watch over your sister, I'm off
to the forest."
(55) Buna nandi bu kayenda ku diangala, mwana mamatoto kite
kio kibenga: "mamatoto, mamatoto!" udila, udila, udila.
So when he had left for the forest, a child in the village began to
cry for banana stew: "Banana stew, banana stew!"
(56) A ukamba n'nuni'andi yfe bongila mwana toto. Mwana
wandilanga ngwandi katsi nandi unata kitebi
The man instructed the baby's guardian to bring it banana
stew. The child cried for its mother but was brought bananas.
(57) Nandi dibakala dio u bitala kulubuka bonga ngie veka e ngie
dibakala bemweni tidi kwaku kuvonda mwana bonga toto,
buna uyenda.
He looked at the bananas before him, but she scolded him,

"You alone are aiming to harm the child by taking those
bananas."
(58) "E ka diambu ko totu eyi kite ki nzadVami, kiziko mambu."
y
Buna udokula toto.

"It doesn't matter," he said, "these bananas belong to my
brother," and he tore one off.
(59) Bu kamana kudokula toto mwana udia, bu kamana kudia
toto batsimbukula kitebi vie vie, ki ma tina.
When he had done so, the child ate of it. But as the child ate, the
banana bunch disappeared completely.
(60) Bu kitina nandi ku diangala kukayenda utuba. A bwau
bwabu a yi tidi kwenda. Diambu dia kavana bwala mwayi
nzadi u yi bikici diambu diaka vana.
When this happened he went to the forest to tell the owner,
who, upon hearing of the disappearance of the bananas,
quickly returned.
(61) Buna uyitsa. Bu ka yitsa, utala kite mwayi: "kamba kitebi
konso ku kiyenzi Tiwitsa kamba ko a mwayi pila kiadi mu
kite kiama Kamba kite kiama konso ku ki yenzi."
And he came. When he arrived, he looked at his brother,
"show me the banana tree, and tell me how the bunch
disappeared. Oh what sadness over my lost bananas."
(62) Eamwayi, ndenvukilaminu, mwan'akukazienandiubedila
lukuji ku fwa. Buna miyi lokuni toto yiveni mwan'aku kazi
mi yisa yi dia kwama kove, vangi mwana. Kansi nandi udidi.
Buna minu bu yiveni udidi buna kite kitinitsi.
"Believe me, brother, your child was in agony, and close to
dying of hunger. So I took the bananas and gave him some. I
myself ate none, he alone. So truly, when I gave him the
bananas, and he ate, the entire bunch disappeared."
(63) Mwayi, diambupe tu ntuba va kite kiama yitidi. Twala kite
kiama ka diambu ko mwan'aku evo mwan'ami kansi kite
kiama yitidi.
"Brother, it does us no good to discuss the bananas, just bring
me the missing bunch. N o matter if it was your child or mine,
just return the bunch."
(64) Buna ununguka kiambu diodio diyitsa kwidi nandi. "A
Mavungu, yi Mavungu, a Mavungu e." Bwabu ngie
makiekulu mu kitebi." "Nhinga."
218 RITUALS
Then he slept on the matter and someone spoke to him

"Mavungu," "I Mavungu?" "You Mavungu. Now you shall
find your bananas." "Yes."
(65) "Bonga ngenga ye vatakana, ye vanga bonso buvanga
nzad'aku. Buna uyenda utakana bwatu bobwau. Uyuvula
kwandi keti va nzila. Uyuvula ngenga "yenda mu ndambu
yo? Swi, swi" bokwa bwau.
"Take ngenge and do as your brother did." So he went out
toward the boat. He asked which path to take. He asked
ngenge "shall I go this way?" "Swi, swi" reject it.
(66) Ye tula mu bwatu a ngenga "Yi kandama mu bwatu?" "Nge,
nge, nge." Ka kandama mu bwatu lumbu lo kwa Iwau, "E
Mavungo e palabanda, E Mavungwe palabanda." Buna
uyenda.
Shall I take this way in the boat?" And ngenge, "Nge, nge,
nge." So he took off in the boat singing "Mavungu going
downstream, going downstream." He heard a voice:
(67) "A Mavungu, Mavungu eyiza nwa dyamba. "Buna utapuka
uye nwa dyamba. Bu kamana kunwa dyamba usia nkaka
ubasisa mwisi.
"Mavungu, Mavungu, come over here to smoke hemp." And
he crossed over to smoke hemp, soon to be obscured in a haze
of smoke.
(68) Kansi kazimbukulu we kavana bwala, vana bwala bu ka
fuma.
He lost his way, and before he realized what had happened, he
was in the place from whence he had come. In the village they
shouted,
(69) "0 mwayi ma kwizal" "Ti e e nkieto e nandi kadi nata
kitebi." "Mwayi ma kwiza e e"; "kitebi kiami?" 'A mwayi, yi
sa ki baka ko!"
"Oh, brother is back. Woman, come carry the bananas."
"Brother has come!" "What, my banana bunch?" " Oh,
brother, I didn't find them."
(70) Ulembakana mu diambu bakutumisa mu mavanga mapwedi
na kwe tapukanga. Buna ulembakana kubaka kitebi kiokio
mu diambu di buvulu bwandi ki bila nkisi wowo ulenda
vanga kwidi wonso-wonso ukuntumamanga.
He failed because he let himself be distracted by the many
seductions along the way; because of his own foolishness; and
because he did not follow the magic of ngenge which could

have guided him wherever he needed it.
(71) Kansi nandi bu ka manga tumamana nkisi wowo ye mpila
yoyo buna te bafundana na veka dikanda kiandi Ba kana
kivika kwidi mutu wowo.
Because he refused to be guided by this magic, they accused
him even within his own clan, and they abandoned him to
slavery.
Of the several substructures that are woven into the foregoing

Mahungu myth, the following will be examined in detail: (1) the
relationship between two brothers, who are Mavungu's two sides; (2)
the opposition of hunting and cultivation; (3) the interaction of the
human and the nonhuman (animal) realms; and (4) the distinction
between successful and failed mediation.
Relationships between the two brothers are initially couched in
terms of equality and familiarity; the one calls the other mwayi, fellow,
brother, confidant, kin (7.2). As their relationship becomes one of
greater distance, they become nzadi to each other, a term which while
still denoting "brother" can also refer to in-law, or someone of a
collateral group, of the other part of a settlement or social unit (7.8).
By the end, this relationship has been transformed into one of status
differentiation, the one having become a successful healer and medi
ator, the other a slave, muntu mu kivika (7.71). The myth directly
addresses the issue of social unity and difference, the one and the
many.
Loss of a spear (7.6) is the narrative element introduced to discuss
the differentiation in society that results from ambiguous relations of
technique, subsistence, and ownership. If the spear cannot be
returned to a brother, what is its equivalent The original spear owner
proposes that a sister given in marriage would suffice. The spear loser
rejects the proposition as "unthinkable," whereupon the first brother
threatens to leave the village—to segment. The reason for the refusal
of a sister exchange on the part of the spear loser is not made explicit,
but the lines of the dilemma are sufficiently like real life in Kongo
society that it may be inferred. To accept a sister in exchange for a lost
spear would confuse the lines of descent, ownership, and marriage. At
one level it would be incestuous since the social unit of shared and
borrowed tools is the descent group, the matrilineage, within which
marriage is prohibited. The classic Kongo social-structural question
220 RITUALS
is raised, of how an undivided, descent community of goods, can be

divided. The well-known ritual of sacrificing a pig to "kill the descent
of one blood" (yonda ngulu a luvila) is often invoked as the solution
to the problem. In actual practice, it is rarely resorted to, but is said to
have been used in the past where dominant single-lineage communi
ties like the Nsundi invaders needed to set up polities of their own
kind. A minimal exchange duality could be created in this wise. In the
myth, the spear loser is thus faced with several options, each with
clear consequences. If he finds and returns the spear, he maintains
descent ties with his brother, or if he ignores the spear and exchanges a
sister, he must break descent ties. The metaphor of goods and blood is
clear, and may be graphically presented:
return maintain give sister break descent

#
spear ' descent ties " in marriage (blood) ties
The group whose members borrow from one another—the matri-

lineage—seeks its spouses beyond its own confines, but to make this
the group of marriage exchange (by severing blood ties) would render
the unit of exogamous alliance very limiting. The rejection by
Mavungu (the spear loser) of the tight exchange model is an implicit
affirmation of the wider exchange model. This appears to entail
another consequence in his actions. Having rejected alliance with his
brother, yet not wishing this latter to segment, he resorts to a mystical
solution to solve his problem: find the spear and establish contact with
"the world beyond." Although there is no evidence that this myth was
related ioLemba, structurally it does so relate by negating the isolated
endogamous community, and establishing a preference for effective
mystical mediation. It is at this point that hunting symbolism becomes
significant.
Mavungu (the spear loser) is cast as a hunter and successful mys
tical mediator; the other Mavungu, whom we may call the banana
bunch loser, is put into the role of the cultivator and the ineffec
tual mediator. One specific object relates hunting to effective mys
tical mediation. Ngenge, the bell hung under the belly of the
hunting dog, leads the hunter through tall grass and forest to the game.
African dogs do not bark. Their bells announce their presence to the
hunter while not frightening away the game. This perception of the
bell that smells out game is lifted and used as a metaphor for the
effective diviner and mystical mediator. In the present myth ngenge is
thus the near-living instrument of effective magic. Metaphors pile up
in short order as Mavungu, spear loser but effective mediator, sets out
on his journey to track the elephant. At one level the quest is a simple
hunter's pursuit of a wounded animal. But the "path" of this pursuit is
a wide river, suggesting a mystical journey. At another level the quest
is indeed a dream, in which ngenge the hunting bell, by deft analogy,
takes the figure of Mavungu (spear loser) on a mystical course to a
correct solution of his problem with his brother back home.
There are two conclusions to the myth: one having to do with the
consequences of effective mystical mediation, the other having to do
with the outcome of relations between the brothers Mavungu. In the
first, Mavungu (spear loser), as successful mediator, becomes a
renowned healer with an ability to "hear" magical messages. Ngenge
leads him to his goal, the "wounded elephant," who, transported to a
mystical plain, becomes a human sufferer whom he cures. Hunting is
a wide-spread metaphor for healing in Kongo culture, in which the
hunter relates to animals (the hunted) approximately as the healer
relates not to the sufferer but to the illness. The healer seeks out the
affliction and finds it, in order to eradicate it. The healer is like the
hunter in another sense. He goes out into the wilds of nature and
brings back the plants with which he treats, just as the hunter goes out
to bring back food for humans. In Kongo thinking, wildness possesses
power and is the source of man's strength. Thus it is understood that
Mavungu (spear loser turned healer) brings the banana tree back from
the wild for his home and for his brother. Hunter and healer straddle
two sides of the continuum from domestic to wild. The hunter begins
5
in the village, but his field of action is in the wild. The healer s origin is
outside in the wilds, but his field of operation is domestic society. In
the present myth, the hunter chases the animal from the domestic field
back into the wild, whereas the healer brings back plants from the
wild. The two domains, the domestic and the wild, inversions one of
the other, are mediated by ngenge. Again, although there is no
mention of this text as Lemia-related, the central ritual function of a
small bell is appropriate for Lemba. Numerous references from
Dapper's 1668 account, several from the western variant, and that
from Haiti describe such a bell. In the present text, this central
mediating role may be sketched in the manner on page 222. In struc
turalist analysis, there are several columns of oppositions which are
mediated by a middle operator, or which operate upon each other.
Ngenge, the magical voice of the bell which moves between the village
and the wilds, transforms the hunter into the healer, who heals the
wound he has inflicted. His spear is transformed into an object of
222 RITUALS
medicinal attention, for which he is rewarded a palm sapling. Finally,

he is given power over mediation, the power of treating illness in
society and ruling. That such a structure or set of procedures exists in
Kongo thinking, is clear from examining the herbaria of banganga.
There is, for example, "corpse hunter" (munkula mvumbi), SL small
tree growing at the edge of the forest, used to treat side pain and
cramps. As the hunter goes into the wilds to kill in order to live, so the
healer, with his origin outside of society and his mystical roots
elsewhere, chases death from the domestic scene.
illness in "
society
medicine
healer
cures
wound
palm
sapling ngenge
hunter
inflicts
spear J wound.
game in
wilds
It is of course the different attitude of the two brothers Mavungu

toward ngenge that determines their relationship to each other, and
their status in society. Mavungu (spear loser turned healer) becomes
the model of the priest, faithful to his" voice," and the recipient of high
status. Mavungu (banana bunch loser) is depicted as the least disci
plined person around, who is seduced away from the "straight and
narrow" for a shallow "high" of hemp smoking. There may be a
commentary here on some tendencies in Lemba. In the artifactual
9
record of Lemba s western variant there are numerous pipes identi
fied either as "Lemba pipes" or "hemp pipes," some nicely carved,
others large water-filled gourd pipes.
The composite picture of Mavungu in this text suggests something
of the culture hero, bringer of messages from the beyond and of the
banana tree to man. The dualistic play on two sides to every issue or
question dominates. The text is set up as a double confrontation with
the same challenge: to find the lost item—spear or banana tree—
received from the other. The fact that both protagonists carry the same
name, Mavungu, emphatically drives across this point of minimal
differences in similarity, that is complementarity. Even the conclu

sion is suggestive of this: whereas Mavungu (spear loser turned
healer) is lauded, Mavungu (banana bunch loser) is ridiculed and
made a slave.
18
In the next text, Mavungu is again a two-in-one figure, father/son,
for whom the complementarity of two-within-one becomes a tragic
dilemma.
Text 8
(1) Mwana unka dizina diandi Mavungo ma. Mavungu kadi

sola; bana be, kadi salanga.
There was once a child named Mavungu. Of all the children, he
worked the least.
(2) A pila mwana, be u ngie kadi sola; nandi kadi sala kisalu
kiandi kuvola kitimba kalendi sala ko.
Oh! What a manner of child that he would not work. His only
work was to suck on his pipe. Regular work he would not do.
(3) Ngundi ye disia baluta kuzola bana bobo basalanga buna
nandi ukalanga kwandi va bwala.
Mother and Father loved the children who worked, but
Mavungu was usually alone in the village.
(4) Buna mu kilumbu kinka ukota ku nzo ununguka, bu ka
nunguka: umona nkieto u kitoko beni uyitsa kwidi nandi.
One day when Mavungu went into the house to sleep he
dreamed: In his dream he saw a beautiful woman speaking to
him.
(5) "A ngie Mavungo, yuwa mayintuba, minu kivana meni."
"Oh Mavungu, listen to what I tell you."
(6) "Kubonga mbedi kulengula mani; buna nkwitsa va tsitsi nti
nhoka yo bantanguninanga ti mboma umweni. Mboma
kutin'ani boma ko. Kupasula mboma yoyo si umona diambu
dilenda kukusadisa, dilenda kukuvana moyo, w mutu
unvwama."
"Take a knife and sharpen it. Then come to the foot of the tree
indicated and you will see a python. Do not run from it in fear.
Cut open the python and you will see something which can help
you, give you life, and make you rich."
224 RITUALS
(7) Buna bu kakotuka yimeni, Mavungu ubonga mwa mbedi

ulengula, bantu bayituka benil "O Mavungu, ti ka salanga
ko. A bu bwidi kanlengudila mbedi?"
So Mavungu took a knife; the people were greatly surprised.
"Oh, Mavungu the one who never works! How will he sharpen
a knife?"
(8) Ulengula, ulengula, uyenda ku diala ku nsitu wo unkamba
ndozi. Uyenda va tsitsi nti, umona mboma uyilama vana.
He sharpened and sharpened, and went to the forest shown in
the dream, and to thefieldindicated, where he saw at the foot of
a tree a python.
(9) Ukimba monho, usimba mboma, upasuna, ka zimbukula
nkietu 0 be kwe nkamba ndozi ma kitoko beni.
He seized the python, and cut it open as instructed. There he
uncovered the very beautiful woman who had spoken in the
dream.
(10) "A Mavungo, ngie ngie mbadi, mbakidi kibwa kimbote ti O
yizebi nduko kuna."
"O Mavungu, partner, I've a good place, come with me there,"
she said.
(11) Buna bu bayenda kuna nkieto wowo uyimbila: "Ye ditete a
Mavungo minu bu ngina vanga monso mangina vanga mu
ngieu.
Then he followed her there, and as they went she sang: "You
cannot believe what I can do for you.
(12) Bidi, bidi te lufwa Iwaku, siaku bonso bwa kwitsa bela pila
disia diaku kumona (na)pe befu evo beyandi, kina kilumbu
mho mu m'monana beyandi e kianvutuka, Mavungu Ma
Kata Mbamba, bonso bu u teka kala va mwelo nzo siaku. Yu
wutswe?"
Only you must promise that until your death, your father will
not see you, and you will not see him. On the day that you
should meet with him, you will return to your prior state,
MaKata Mbamba, a nobody, as you were before. Under
stand?"
(13) "Ti yuwutsu kwama."
"I understand."
(14) Buna bu bayenda va kibwa kiokio kimboti beni kikineni
kimula uyimbila: "Mavungu makuvu utomba Iwangu
lukubia-dila; Iwangu mbi ka mbiadila Iwalu." Buna

katsimbukulu beti sola.
As they proceeded to the village the woman sang to conjure up
water to reside by: "Mavungu wants to live by the water,
let there be water." And they found what they sought
(15) Uyimbilamulumbulolokazimbukulatsinzo. Uyimbulamu
lumbu lolo katsimbukula va kakibanga kitoko beni kia
t
kavana. Buna Mavungu umona nhenzi.

She then sang a song, and they discovered a vast piece of land.
She sang a song for every beautiful thing they came upon, and
Mavungu was very happy.
(16) Buna kayimbila lumbu waka ku nlangu makumbiye bisalu
binkaka bikwizilanga mu tangu mindela binkwiza mo
dikumbi kwiza kota ku nzo unvwika, zisapatu uvula
mivwatu miandi mi kabela mu zipasi evo mu kiadi unvwika
mivwatu mikitoko.
Then she sang another song and steamers appeared on the
water in great hosts, with white men in them; one came near the
house and they entered the house, giving him new shoes and
replacing his old clothes with new beautiful clothes.
(17) Buna Mavungu uyangdlala beni ye dinka buvulu buwombo
bu bayandi zingindu.
Then Mavungu rejoiced greatly at his own good fortune,
imagining his own folks' great awe in knowing of it.
(18) Buna zingulu zio ziyendanga ku bwala bu tat'andi kwe
dianga mayaka mangana. Buna siandi utuba, "a ngie n tela,
yenda we tala zingulu ziozio zindianga mayaka."
One day Mavungu's pigs strayed to his father's village where
they ate much manioc out of the field. His father spoke to a
hunter, "go and find these pigs that are eating all the manioc."
(19) Buna n'tela wowo uyenda u bakuka bakuka. Uyenda,
ubasika va bwala bubuneni.
And the hunter found where they had rutted the field. He pur
sued them on and on until he discovered a large town.
(20) Ubaka boma. "A pila kiadi minu, ntela yinkwitsanga tangu
katsio mu nsitu wau, buna bwala bubunene kwe bu
basikidi?"
He was seized with fright. "How curious! A hunter comes into
this forest regularly, yet suddenly there is a vast town here.
226 RITUALS .
How has that come about?"

(21) Kansi nandi n'tela utuba, "A amika kina Una kileko ko, kiyi
kambu monanga mu meso ko."
But the hunter thought, "I cannot flee from something just be
cause I haven't seen it before."
(22) Uyenda kuna bwala beni uyuvula, "Nanibeka bwala bubu?"
0 Mavungo" "Mavungu mbi?" "O Mavungu Ma Kata
Mbamba." "A mwana tata uzimbala tama kotso nandi
makwitsa vanga ma pila yiyi."
So he went into the village and asked "Whose town is this?"
"Why Mavungu's." "Which Mavungu?" "Mavungu Ma Kata
Mbamba." "Then this is the child that father lost long ago.
How has he come to this present position?"
(23) Buna u ye basika kwidi nandi unata nandi bu ka ye mona
siandi uvana kio.
Then Mavungu brought gifts to send to his father in the hunter's
hands.
(24) "Kedika Kedika Mavungu mweni." Ti nandi, "Yi mweni,
tata, yimweni tata, yi m'mweni mu mesu mami."
"Truly, truly I saw Mavungu. I saw him, father, I saw him
father with my own eyes."
(25) Buna landila Mavungu ukubika nzila mu kwe mona
mwanandi naveka ubonga ndusi.
Then Mavungu (the elder) prepared to journey to see his child
whom he called "Ndusi" (homonym).
(26) Bu kayenda mu nzila buna nkietu wo utala a Mavungu:
"Kadi kwenda kwidi siaku, kansi kwenda buna nkietu wo
ukubika binuanunu. Uvanga bobo.
As he was going the woman warned Mavungu (the younger),
"When your father comes, do not meet him. Stay with me, for
1 am preparing the weapons to battle against him." She did this.
(27) Mu kilumbu kinka, Mavungu uyindula: "A bwe tulendi
simbidila nsamu wo ko bwabubu miyaka kisina yibeki.
Bileko bipwedi, tata kalendi kubimona ko?"
Mavungu thought to himself on that day, "How can this be
handled? I have received many riches. But is my father not
even allowed to see them?"
(28) A mi didi "Mboti Tata!" kamona ku sa ntsibikila ku nzo
utola nzo mu ngolo ziandi.
When he arrived and greeted them the woman closed up the

house with all her strength to keep Mavungu inside.
(29) Nkieto uyenda kukunuana kadizaba tinnuniama widi mu
kwiza.
She went out tofightwith him knowing it was his father coming.
(30) Buna uyenda "a u tata?" "Tia u Mavungu?" "A mboti tata,
mboti!" Nkietu waka mu kudila mu diambu di muina
ukavana.
But he came and greeted his father, and his father greeted him.
The woman began to weep on account of the prohibition she
had given.
(31) Buna bu kadila pila yoyo wa vutuka. A Mavungu, buna
nandi nkieto wowo uzola kikwezi kiadi beni, yi unkanina.
When he heard her cry thus, Mavungu returned to her. He
thought, "if she feels this way, and likes her father-in-law, I
might compromise her rule."
(32) Ubanza vo bika yivana ndambu lusadusu pasi kasadila mu
Jitangu-tangu. Kasi evo kadi yizebi kwa ti bisalu bia ntina.
He thought," let me go ahead and give some of my wealth to my
father," thinking it would not disappear.
(33) Buna uvana siandi ndambu lusadusu; basimbana mu mioko:
"O yonso tangu kwiza pasi tumonana."
Then he gave his father some gifts, and they embraced: "Any
time you may come so we may see each other."
(34) Ye bu basiala kumbusa a dibakala bu mengi kwa mambu
mami: "Siuvutuka, Mavungu maKataMbamba, siuvutuka
va lukalu Iwaku lu uteka ba."
When this happened, the wife disapproved in these words:
"Mavungu ma Kata Mbamba, you will revert to what you were
before."
(35) Dibakala diodio udila, udila udila. Kalendi kunlenvukila
t
ko. Buna samba bo bwau Mavungu makalu wutomba

Iwangu lukabiala biabio bi bavana bwala bimana kutina.
The man (Mavungu) wept and wept, but she would not forgive
him. All the wealth he had sought as well as the town suddenly
disappeared.
(36) Mavungu udila udila. Bu kadila kazimbukula nsitu uma boe
kwelakana bo bwau bu ubela.
228 RITUALS
Mavungu cried and cried. As he wept he found himself sud

denly in the field where they had "married."
(37) Buna nkietu mpe waka mu kudila mu diambu dikiadi buketi
bundimina. Buna udila pila yoyo.
The woman was also now sorrowful because of the impos
sibility of their union.
(38) Udedakana Mavungu uyesingama va nti va katwama
singamanga.
Her farewells were as fond as his by the tree where they had
met.
(39) Bunazimbambazivutakana bana mbiyayoyikala mu nandi
yivutakana mu nandi.
And all of the misery of his former state returned to Mavungu.
(40) Buna mu kiadi ki pila yoyo Mavungu ukituka bonso bu
f
kateka kala. Buna mu kiadi kipila yoyo ufwila vana.

And with such sadness, Mavungu changed back to what he had
formerly been, and in sorrow there he died.
The prohibition imposed on Mavungu (the younger) that he must

reject forever his father to have a good life with a wife, creates a tragic
dilemma for him. He must choose between accepting his father, at the
cost of remaining politically impotent, or going with the woman to
have material rewards but without being able to exchange them with
his father. In this conceptualization of Kongo society the central
juridical corporate institution, the matrilineage, is left unmentioned.
Rather, the society is conceived in terms of the relationships an
individual may voluntarily emphasize: patrilaterality and alliance.
This text, like the preceding, seems to exaggerate the mutual exclusiv
ity of the alternatives, and in so doing actually demonstrates the
importance of integrating the two in a workable view of human
society. It would appear that one of the main characteristics of these
Mavungu myths is their internal analysis of society's various com
plementarities: male and female, age-different siblings, modes of sub
sistence, different aspects of kinship such as patrifiliality and alliance.
Every case of exaggerated or disjunctive complementarity contains
its own solution, its unique form of mediation. Above I formulated this
in terms of endogenous and exogenous process, or attempted solu
tions within one and achieved solutions between two. In the Lemba
origin myth about Mahungu (Text 5), endogyny was expressed in the
primal androgynous state of the hero, and exogeny was created by
first exaggerating sexual difference, then by resolving it in the comple

mentarity of marriage. In Text 6 about Mavungu the forest ogre,
endogyny was expressed in the character of a radically antisocial hero
who lived in the forest and ate only human flesh, and exogeny was
emphasized in the arrest of the cannibalistic ("species endogastro-
nomic") Mavungu and his incorporation into a noncannibalistic
("species exogastronomic"), law-abiding community. Text 7 about
Mavungu (spear loser turned healer) and Mavungu (banana bunch
loser turned slave) demonstrated several themes of endogyny becom
ing exogeny. The two brothers were initially of one status. With the
loss of the spear, internal exchange of wives (sisters) and segmenta
tion were proposed, but rejected. Original endogynous solutions to
property, alliance, and descent were shown to be unacceptable.
Exogenous solutions stated in the outcomes of the myth were status
differentiation between the two brothers, and the differential access to
mystical mediation resulting from differential subsistence roles of
hunting, cultivation, and healing. Also, the desirability was under
lined of mediation between the human and the "other-human" world
of animals and spirits.
In Text 8, about Mavungu father/son, the exogenous solution is left
unmentioned, or one might say it is mentioned by the exaggeration of
endogenous alternatives—remaining with father and being weak and
lazy, or leaving the identity of father and benefiting from the rewards
of an alliance with the " other": in this case, trade at the oceanside and
great political success, as seen in the grand city. The implicit exogen
ous solution is not hard to uncover. If Mavungu's "death'' is due to the
mutual exclusivity of patrifiliality and alliance, then "life" results
from the complementarity of these two types of relations. This latter
is, of course, the Lemba solution to the excessive endogyny of
isolated settlements, marrying internally, thereby weakening trade
and peace networks across the countryside. Any marriage "out," but
particularly a "return blood" marriage such as the patrilateral cross-
cousin marriage, accomplishes the feat of structuring patrifilial con
tinuity together with alliance.
Numerous literary works have taken up this problem. Malonga's
novel M'pfoumou ma Mazono is the story of a woman who is forced
to flee her husband, a prominent chief and her cousin, after an illicit
19
affair with her husband's slave. In a vision her maternal grand
mother tells her that the child she carries is her husband's. This
comforts her greatly, but she dares not yet return and becomes a
renegade, taking up life in the wild. There her son is born, and with his
230 RITUALS
mother grows up in a cave by a lake. He teaches himself all the arts of

human survival, especially hunting. This life in the depths of the wild
is interrupted by two hunters from the husband-father's community,
who as runaway slaves find the cave. The son, meanwhile, has
become a strong and capable youth and invites them to join the forest
settlement. This "wild" society of renegades grows into a vast city
based on principles of egalitarian democracy instead of status dif
ferentiation, including slavery. Ultimately the healer-magician of the
group, himself a runaway slave, divines the necessity to re-establish
contact with civil society. Under the leadership of the youthful son,
they move their mysterious community into the midst of other villages
and establish a market to trade with them. At first threatened to the
point of challenging war to regain lost slaves, the established society
selects as its negotiator the husband-father. Father and son recognize
one another, which of course permits father and mother to be recon
ciled as well, restoring peace to the region. Father conveys upon son a
legitimate political office in the form of an inauguration to a ritual
order (Lembal). Each now has his own independent community, and
they continue as allies in trade and war, closely intermarried.
The theme of the wilderness settlement recurs in mythology and
lore, as well as ritual. Sometimes the inhabitants of such a settlement
are ghost and spirits (matembo, minkuyu, bisimbi), at other times they
represent a type of human community, different from but potentially
related to the human community as a whole. Invariably these wilder
ness settlements are sources of contrast and of renewal, and powerful
sources of wealth and danger, simultaneously. They embody periph-
erality and marginality, the power that comes from liminality and
contrast. Often the mystery community represents an inversion of
human society (Mavungu as the forest ogre) or an exaggeration of
contradictions within the human society (Mavungu's city of wealth by
the sea, Text 8). There is thus the theme of mediation achieved or
mediation failed which runs through all accounts. One of the tech
niques of mediation in myth, ritual, and institution-building is to
devise mediatory symbols such as ngenge, rivers, journeys, burrow
ing nkumbi rats, tukula red, and mpemba white, as well as com
posite mythic-religious figures such as Mahungu and the python-
woman in the tree. These mediatory symbols and figures vary from
the more-or-less human to the more-or-less natural (nonhuman).
Sometimes, as in the case of the python-woman, this variance occurs
in the midst of the text. The tendency to disguise or convert a hero into
a natural or animal object, to remove him/her from the human realm,
seems to correspond to the degree to which a social contradiction is

consciously acknowledged. Thus, in Text 7 (Mavungu as the two
brothers) and Text 8 (Mavungu father/son) the hunter is a significant
mediator who stands in close relation to the natural world. Moving
across the human/nature opposition, he is able to illustrate the dif
ferences as well as the interdependencies of the two realms. Mahungu
as hero of complementary opposition often expresses contradictions
in political and social life. The most interesting case of mediation
failed is that in which the wife is couched in the form of a python ( Text
8). Mediation achieved is portrayed in Text 5, and to a certain extent
in Texts 6 and 7.
Insofar, then, as renewal, restoration, or regeneration of society is
achieved the mediating symbols are of a humanizing quality in the
myths. But insofar as these harmonious states are not achieved, the
mediatory symbols tend to become naturalized. The importance of
this issue for theories of myth analysis is great. It moves analysis
beyond the point where all myth functions as a " logical model capable
20
of overcoming a contradiction" to the realizaton that societies are
able to perceive the tragic dimensions of these contradictions and
to understand that they cannot be overcome. Or, if they can be
overcome, just what institutional alternatives must be followed, and
which must be avoided because their consequences are tragic. These
issues will be taken up in subsequent chapters, as will be that of how
recognition of social contradiction relates to the medicines of healing
in Lemba.
Chapter 7
The Western (Yombe, Woyo,

Vili) Variant of Lemba
The western variant of Lemba is based on a great quantity and a wide

diversity of sources because of centuries of coastal contact with
Europeans. However, despite the extensive work by such ethnog
raphers as Bittremieux and the Loango expedition, only the Laman
catechists have described in any detail Lemba's rites in this region.
The two texts which follow, from eastern Mayombe, describe respec
tively the modes of recruitment to Lemba through "dream, curse or
1 2
possession" (Text 9) and the initiatory seance (Text 10). An etio
logical myth from the forest region of the Mayombe near Kangu (text
11) features Kongo trickster Tsimona-Mambu as culture hero,
bringer of Lemba, providing a basis for the analysis of the lyrical
domain in the western variant. Trickster myths from outside Lemba
(Text 12) offer contrastive material of the same range as that pre
sented on Mahungu in the previous chapter.
The western variant demonstrates major points of differentiation
between the Yombe forest and the coastal kingdoms and peoples.
Although the eastern Yombe versions of Lemba resemble areas
already studied, the coastal accounts change in an important sense.
Art-historical expressions of Lemba—bracelets, drums, shrines-
become more ornate. Among the coastal Vili, and possibly inland
among some Yombe, Bembe, and Kunyi, the portable nkobe dis
appears in favor of a fixed shrine-house in the back yard "pantheon."
Charm-jewelry appears in the form of miniature drums, figurines,
shrine doors, necklaces, and elaborate bracelets in cast or engraved
copper and brass. These objects permit a fuller analysis of the expres
sive domain of consecrated medicine.
233
234 RITUALS
The Lemba Seance in Eastern Mayombe
Text 9
R E C R U I T M E N T TO L E M B A
(1) Lemba i nkisi wena mu nkonko yijwanene bonso nlunga.

Lemba is an nkisi in the form of an nkonko drum or a bracelet
(2) Nkisi wowo ubanzwa vo wambaki mbongo zazingi zikalenda
tudulwa kwa bakulu.
It is thought of this nkisi that it requires greath wealth to get the
ancestors to bestow it.
(3) Yandi wavangulwa mu ntinta yampemba isokwanga mu
mpandulu andi
It is enacted with white chalk, given during its consecration.
(4) Mboki, longo (bongo?) biandi biambukila biena muna mpe.
Also, a marriage ceremony constitutes part of it.
(5) Yandi Lemba ulotuswa ndozi mu bakulu.
Lemba itself is presented in a dream from the ancestors.
(6) Yandi ubikwanga vo nkisi a bakulu. Yandi ukwendanga ku
mpemba ye ku bwala.
It is called an nkisi of the ancestors; it mediates the land of the
dead and the village.
(7) Ikuma kakomwanga mu ndotolo a ndozi ye mu mpinunu a
mafina ye mu mbwanunu a minkuyu.
Thus it is activated by dreaming, by nightmares of suffocating
by a curse, or through spirit possession.
(8) Mboki mpandulu vo nkebolo a Lemba zena zazingi beni.
For this reason initiations and initiands to Lemba are many.
(9) Bankaka bakebanga Lemba mu nlunga a koko.
Some keep Lemba in the bracelet on their arm.
(10) Nlunga una mpe sadulwanga mbatu bonso nkonko a Lemba.
A bracelet is sometimes used like the nkonko Lemba drum.
(11) Vo bakondolo nkonko, buna i nlunga balenda komina nloko
mwankaka.
If they lack the drum, they use the bracelet to strike certain
spells.
The Western Variant 235
(12) Balutidi sadila nlunga wanzongo mu tula mu koko.

Copper bracelets are mostly used, worn on the arm.
(13) Mboki nlunga wowo ukalanga na nganga Lemba; vo Jwidi
mpe ka ulendi katulwa ko.
So the bracelet of the priest remains with him; if he dies it may
not be removed.
(14) Mboki kansimbi muntu wankaka ko walembwa dio vanda.
Thus [Lemba] will not seize another person if it is not trans
mitted to another initiand.
(15) Nga vo una simba mu diba, bunafwanane mu futa vo nsusu
mosi kwa nganga Lemba.
Even if one is seized, it suffices to pay a chicken to a Lemba
priest.
(16) Ndiena wavanda Lemba diandi usanga vuvu vo bakulu
bandi babana bana fwa ku mpemba balenda kwiza kuntu-
dila mbongo.
The one who composes his own Lemba, hopes that his ances
tors in Mpemba will provide him with funds [for the initiation].
(17) Yandi Lemba lenda heha mbongo zena kwa bakulu ye zena
kwa bamoyo mu diambu dialenda kiandi kiena kwandi mu
twadisa mbongo kwa mfumu andikadibu batombulanga wo,
buna bahehanga mu mafula makwizila mbongo mpasi bantu
bakaka mambu Lemba basinduka mo mana bakamana
mbongo za Lemba.
Lemba is capable of exchanging wealth of the ancestors with
the living so that it may be at their disposal, providing wealth to
a person's lord so he may receive it; thus they exchange in the
entrances whence comes wealth, and those who are in Lemba
with their problems receive Lemba 's wealth.
(18) Nkisi una mpe wena ye nkazi a Lemba.
The nkisi also has a Lemba wife.
(19) Ukotanga wo yandi i kundi ku nima nkisi wowo.
She enters it .to be the "friend" behind the nkisi.
(20) Vo muntu una bwana nkuyu kansi ndinga yuzikamani buna
nganga Lemba una bonga lutete Iwansudia vo Iwa tende.
If a person is possesssed by a spirit but his speech is blocked,
the Lemba priest will take a lutete gourd seed.
236 RITUALS
(21) Vo yandi mbevo una lo tota, buna ngudi a nganga una

kunkamba vo: "Yoya, monso mamweni, samuna."
If the sufferer cracks it open [with his teeth], the chief priest
tells him: "Whatever you see, tell it"
(22) Buna mbevo i ntumbu badika vova vo kakedi vova ko, buna
yandi una samuna makamweni
Then the sufferer immediately begins to speak if he has not
already and tells all he has seen.
(23) Vo nkuyu kamweni vo unkembi mambu, buna yandi una
ntumbu samuna ye zaikisa yayonsono.
If a spirit has appeared or spoken, he will reveal it and make all
known.
(24) Bobo i salu bia Lemba.
This then is Lemba's purpose.
(25) Mboki nkisi wowo ukembwanga mu nkela.
This nkisi is kept in a box.
(26) Nkela yoyo ka ilendi talu wankaka mu ngudi ko.
N o one [beside the owner] may look into it.
(27) Mboki nkisi wowo ukebwanga ku vinga kia mu ngudi a nzo,
kuna vinga kiokio ka kikoti muntu ko.
It is guarded in a special room in the interior of the house where
no one may enter.
(28) Buna una kubwa, nga vo ka bwa ko, buna una tekwa mu
mbongo za Lemba.
Should someone do this, he would be sold for the benefit
ofLemba.
(29) Fisidi Lemba nga diena mwamu. Konso diodio disadul-
wanga kwa babingi.
Perhaps this is all about Lemba. Many use it in this way.
(30) Ka diafwidi ko nate ye bwabu.
It has not died out till this day.
Text 10
PRESENTING MWEMO-A-LEMBA MEDICINE
(21) Fu kia zinganga za Lemba, vo bamweni muntu wena

kimvwama, buna i ntumbu kunkamba vo nwa mwemo a
Lemba, Ls.v., makaya ma Lemba-lemba.
It is a custom of Lemba priests, when they see a person of

means, to tell him he must drink Mwemo-a-Lemba, that is,
leaves of the lemba-lemba plant.
(22) Kansi vo muntu vumbidi vumu, buna si bavela munsumbiye
si bavela wa mu nkumba a Lemba.
But if a person has a swollen stomach, they give him
munsumbi leaves.
LEMBA CHILD BARGAINS FEE FOR COMPOSING NKISI
(23) Mwana nganga i ntumbu bonga mpataye kwe bwanisa ngudi

nganga kuna fula.
The neophyte priest next gives his high priest five francs at the
village entrance.
(24) Zinganga zazonsono zavanda Lemba mu tini kiansi kiokio,
bu buwilu nsamu vo Lemba si dwandusu ku kingandi, buna
bana kwina, kidi babaka zimbongo mu mpandusulu au.
All priests who have been consecrated to Lemba in the sur
rounding area, when they hear of the Lemba affair, they make
their way hither in order to enrich themselves by their initiatory
expertise.
(25) Ngudi a nganga si katambula kumi evo kumiye mpata tanu.
The chief priest will receive 50-75 francs (10-15 mpata).
(26) Mboki zinganga zankaka zazono kani 30 vo 40 si zasola
mwala (nzonzi) au mu kwe kubalombila zimbongo.
Thirty or forty other priests will send their representatives
forward to request payments of money.
(27) Buna mwala si kateka tambula kani mpata mole vo tatu.
The representatives will receive 10-15 francs (2-3 mpata)
each.
(28) Mboki si katangunanga nganga vo: ndieu ebu kadilanga,
mboki mwana nganga si kavana ndiana ntalu yayi kata-
mbulanga ye yandi i ntumbu futa bonso buvovele nzonzi nate
ye babonsono bamenifuta.
The priests will be told: the one who has requested Lemba,
as Lemba child will be told the fee for receiving it, and he will
pay whatever the spokesman says, until all have been paid.
(29) Mboki i ntumbu vandisa nkisi.
Then his nkisi will be composed.
238 RITUALS
(30) Nkama Lemba ufutwanga ngulu ye mpata tatu, minkwala

zole mpidi zole zambongo.
y
The chief priest's wife is paid a pig and fifteen francs, two
nkwala mats, and two mpidi baskets full of raphia cloth.
COMPOSING NKISI LEMBA
(1) Lemba i nkisi watudulwa mu mwila a tola.

Lemba is an nkisi which is put in a cylinder.
(2) Mu ngudi a mwila wowo mwasokwa mafutu mole mami-
nkanda miabulu, nkumbu a futu diabakala: "Nsasa
Lemba."
Inside it are placed two sacks of animal skin, the name of the
male bag being: "Nsasa Lemba."
(3) Va diau vatulwa bilongo.
[In] On it are placed medicines.
(4) Singa kiakala mu mbu ye mpemba biau biatoma kangwa va
futu diankanda mbala.
String is wound tightly around it to better contain the mpemba
chalk in this bag of mbala antelope skin.
(5) Futu diankaka dia nkanda nkumbi divwilu kwa nkama
Lemba; va diau vena bilongo: n'nanga ye tukula.
The other sack, made of the nkumbi [rat's?] skin, is the
Lemba wife's; on [in] are put medicines: cauries and tukula
red.
(6) Futu diodio dibikwanga nkumbu "Mpemba Lemba."
This sack is called "Mpemba Lemba."
(7) Lemba biekwa ku Nsona.
Lemba is consecrated on Nsona day.
(8) Nsuka lumbu mpaikulu au kumbazi, bakala sikateka bonga
Nsasa Lemba ye bonga mpemba ye teka tula mampemba
mandi vafutu diodio; mboki sonika mpemba mu mpenga ye
mu mvamba miamoko.
Next morning they go to the square; the male [initiand] first
takes the Nsasa Lemba bag and puts chalk onto the bag's skin;
then he inscribes chalk on his temples and his hands.
(9) Mboki nkento mpe si kakutula funda ye sonika tukula mu
mpenga ye mu moko.
Then the woman [initiand] also removes tukula red from her
bag and inscribes it upon her temples and hands.
(10) Mboki bau bole ntumbu vaika.
Then the two of them come out.
(11) . . . buna kabalenda zieta vo sala mu lumbu kiokio ko, kansi
si bavundila kio kaka.
. . . they may not walk or work on this day but must sit the day
out quietly.
(12) Lembama kiokio, bau bana baka kimbevo vo kijfwa.
Failing to obey this, they may take sick or die.
(13) Mafutu matulwa mu nti.
The bags are hung in a tree.
(14) Bilongo biankaka batudulwa mu ngudi a mwilu: dingongo,
makayi kwa Lembe.
Other medicines are placed into the cylinder, dingongo nuts
and Lemba herbs [calmants?].
(15) Lekwa kiokio i nsuki zatebwa ku ntu a ndieu wavanda
Lemba ye zakangwa va nsi a nkanda nsesi ye nkaka ye
nkanda a kubu, wakangwa mpe zinsuki zamwana nganga ye
za ngudi a nganga.
[Another] thing is hair from the head of he who has com
posed Lemba; it is tied into skins of nsesi antelope, pangolin,
and kubu antelope; also in it are hair of the neophyte priest and
the chief priest.
(16) Nkanda wowo batambulanga wo mu lumbu kina kiteki
mana vanda nkisi wowo.
This skin [?] is brought along on the day when they have
completed composing the nkisi.
(17) Mafunda momo miatatu miabikwa "minkunda."
These three bags are called "the abode."
(18) Batulanga mpe bikengi.
Bikengi water plants are also put in it.
(19) Nkisi wowo wena mpe ye funda dibikwanga "kikundu dia
Lemba": va diau batulwa bilongo biampila mu mpila—
mweba, ntutu, cizika, munsumbi-nsumbi, nkuku-nona, ye
nionzo ye makaya manlolo.
240 RITUALS
The nkisi also has a satchel called"Lemba power" into which

is put a variety of medicines—mweba, ntutu bark, cizika,
munsumbi-nsumbi, nkuku-nona, nionzo and nlolo leaves.
(20) Nkisi wowo wasadulwa mu mayela ma mpila i.s.v., vumu,
ntima, ntu, lubanzi.
This nkisi is used for a variety of illnesses such as those
affecting abdomen, heart, head, side.
THE MARRIAGE CEREMONY BEHIND THE HOUSE
(31) [illegible]... buna i ntumbu kunata ku mbusa nzoye nkento

andi.
[the initiand and rite objects] are carried behind the house with
his wife.
(32) Ngeye n'nuni Lemba, keba nkazi aku Lemba bwambote; si
bakamba mpe kwa nkento, ngeye nkama Lemba, kebannuni
aku Lemba bwambote.
[The priest intones] "You Lemba husband, guard your
Lemba wife well"; they advise the wife, "You Lemba wife,
keep your Lemba husband well."
(33) Mu nzo a Lemba ka mulendi kota muntu wankaka ko.
"No other person may enter the Lemba house.
(34) Vo umweni muntu dieti nkwalu yiwena, buna kamba nuni
aku evo umweni mpe muntu viokele mpe evo diata mu Iwi
nkulu Iwa nzo aku, buna kamba nuni aku.
If you see another person do so, tell your husband; or if you see
someone leaving or walking near your house, tell your
husband."
WHEN THE PROFANATION OCCURS
(35) Buna vo muntu si kasumuna mina miomio dia Lemba, buna

si ka futa mvika; kansi mu ntanguyayilenda futa kani mpata
4 evo bonso buzolele nganga.
When a person violates the prescriptions of Lemba, he must
pay a fine; however nowadays [ca. 1915] it may be 20 francs or
whatever the priest decides.
(36) Bu beti wo vandisa, buna nganga ye nkento andi mboki

mwana ngo ye nkento andi mpe; buna babakala si bateka
vwanda vo ntandu a buku biole bianti.
When it is composed, the priest and his wife, then the "leopard
child" and his wife [ are present]; the men first sit on the bark of
two trees.
(37) Mboki bankaka bana kubafuta ku nsi a vunga napi ye
zinganga zankaka zed sika zinkonko e zindungu ye beti
zimbila nkunga wau:
The others come to pay them remain silently beneath a blanket
while other priests beat nkonko and ndungu drums, singing
this song:
(38) Wavanda lemba, kusuka kwamoyo.
He consecrates to Lemba, let him cleanse his life.
(39) Lemba diami dia bumpati bwa nganga
My Lemba enclosure is the glory of the priesthood.
Mayamona mu Lemba ndiadi
Behold this Lembal
Sukula ko!
Cleanse it!
(40) Bu bameni diodio, i ntumbu vaika ye bakento ntumbu kwe
vinganga ye diatasana minlembo ye simbasana moko ku nsi
a vunga ye minkunga ye zinkonko mina tamana bonso
busilulu babakala nateye kani lokula biole, mboki vaika.
When they have done this, [the priests] leave and the women
present "walk" over one another with fingers and hold hands
under the blanket while singing and drumming the nkonko
drum like the men did it, until the two are cleansed, then they
come out
FINAL RITE IN THE FOREST
(41) Landila bilumbu biole tata nganga kunata mwana nganga

ku mjinda.
Two days later the Lemba father takes the Lemba child to the
forest.
(42) Kuna mjinda bavwandila va mfuma ye makuku, bisoma
mole va nsi bau biekanga—is. v., Mpemba Lemba ye Nsasa
Lemba.
242 RITUALS
There in the forest they sit upon an mfuma cotton tree and a set
of termite mounds respectively, the ground beneath them
consecrated as "Mpemba Lemba" and "Nsasa Lemba"
(43) Mbangudulu a mambu momo i vo bakala evo nkento andi
wena sita buna si kabuta mu diambu diakameni vanda nkisi
ko.
The meaning of this is: if the male or the female is sterile, they
may give birth because they composed this nkisi.
(44) Mbangudulu a makuku i vo bana sibutwa.
The meaning of the termite mounds is that children will be
born.
(45) Bu si bakola nyo a nonia ye kuntentika yo va mbata a ntu
andi, binonia biobiobubetikunzanzalayekuntatika ku ntu,
buna kalendi yaula vo nikuka nkutu ko.
When they take one with termites in it and place it atop the
head when the termites begin to crawl out and bite the head,
they must not cry out or squirm.
(46) Nga vo si kanikuka vo kubula binonia, buna i mabuta
kakubudi, si bana kumbika vo ndoki.
If they squirm or slap the termites their own offspring are struck
and they will be called witches.
(47) Bonso bwena ntalu a binzulu biobio bieti kunzanzala, i bobo
buna kala ntalu a nkun'andi mpe.
As is the number of these termites [ ants] so shall be the number
of their offspring.
(48) Mwana wantete vo bana buta vo wankento, una bikwa
"Mpemba Lemba;" vo bakala, "Nsasa Lemba" bonso
bwabiekwa bisama biabiole zinkumbu.
If the first child born is a girl, it will be called "Mpemba
Lemba"; if a male, "Nsasa Lemba" corresponding to the
names of the consecrated signs.
(49) Landila diodio, si banika ngunzi ye mpemba mboki kwe
biosonikingi mu nitu mwana nganga yamvimba matona-
matona mampembe ye mambwaki.
Following this, they grind a mixture of ngunzi-rsd and
mpemba-chslk and trace the whole body of the neophyte priest
with white and red spots.
The Expressive Domains

SPACE AND TIME:""FROM PORTABLE TO FIXED RITUAL SYMBOLISM
Most characteristic of the western variant's spatial-temporal domain

is the transformation of the symbolism of ceremonial ritual into the
symbolism of permanent architecture and garden (figure 18). The
richness of both rite and n'kobe diminish. The rite described by
Babutidi (Text 10) lacks the elaborate rhythmic flow between village,
savanna, forest, and stream of the northern and south-central vari
ants. The western-Kongo landscape is largely forested, and villages
and towns and fields cut and burned out of the forest constitute the
only clearings. Lemba's ceremonial rhythm spans this simple dichot
omy of forest and village clearing. Across the Mayombe, n'kobe and
reference to a "house"-shrine are both present in varying degrees of
elaboration. In Babutidi's account, for example, from the far eastern
Mayombe, the n'kobe is as complete as farther east and north,
although one of the satchels within it is called the Lemba "house"
(minkunda, 10.17), and although there is this reference to a "house"
(10.33), the final ceremonies are held in the forest. Farther west, on
the coast and in Loango especially, the n'kobe disappears entirely.
The Lemba shrine becomes a fixed installation at the intersection of
the forest and village, an elaborate "kitchen" behind the hearth, a
garden grove of trees, filled with other objects.
Ethnographies from the Mayombe and Loango describe this
Lemba "house" and details of the fixed installation. In the Kangu
area of Mayombe, Bittremieux described the Lemba couple's yard as
a wooded grove (bikulu bia Lemba) of wild shrubs and trees trans
planted to the domestic area between house and shrine, and sur
rounded by a fence. The trees included the mfuma cotton tree (JErio-
dendron anfractuosum), nkumbi (Lannea Welwitshii), lubota
(Milletia), kuaku (Oncoba sp), nsanga-nsanga (Ricinodendron
africanum); smaller plants included lolo kitseke (Annona Senega-
lensis), mvuila tseke (planted together, perhaps suggesting their
association with the human world ( = savanna, tseke) in contrast to
the ancestral world ( = forest), ditambi-tambi (in association with
the lubota tree). By themselves in a small enclosure behind the shrine
house were mutsanga-lavu and dilemba-lemba {Brillantaisia
alata). The whole arrangement suggests a floral cosmogram fixed to a
spatial location much as the n'kobe eastward contains a miniaturized
cosmogram of plants collected from the zones of savanna, forest,
On О
/ О О
н >>
cl -° О
CO
L V
• 1
a-s
+-» -o О
UH Ö
asdei — & R
: -S О- !
Лвр-г 3 0) -о =
1
s * i!3
S со оо
äi-S
pa^pddsun э ^ ц а о э и ]o ^иэшэщэр
г-, со
SS
EU О
4-1
СО Г -
го
JìSss
иS О
О
¿ é мо m Л,
S ra - °
OA О 3 - Г
5 г-
E ? Xi оо
ë §! oboi»
3 «
-.0
СО
W 0 PQ #
С "й . *
и Д (S
-tí «I О \
ü о E
S ° *
О Я ,2
5
ОО С ОО*
• й *С PQ С С С В> ^
и, Q
SS ÖL S ttS
О <5
2 ^
8^ <u . У
w
S ° ra ~°
tí
* о
i
aequi-пц ozi; esnq-m¡
св
о, pjeAjjnoo цдеэц Я 'S I
aoeds
village, water, and cemetery. The purpose of bringing all these plants
from the wild to be represented in the shrine grove, states Bittremieux,
is to assure that the Lemba couple will be genuinely "other," truly
3
changed (baluka), permanently liminal.
The Lemba shrine-house is described by several writers. In the
Mayombe Bittremieux reports that on its gable there is a sculpture
made of woven nsoni grass of the ndimba snake (Psammophis
4
sibilans, or Leptodira Duchesnii), and a representation of the lolo
and mvuila plants. Loango coast expedition writers report that the
Lemba house in that area is situated in the forest well away from the
houses, surrounded by a papyrus fence, and is used as a treasure
house by the Lemba couple, secluded, locked with fanciful door
5
locks. Its decor is such that only the wealthy can afford it. Between
the two front doors, one for the Lemba husband, the other for the
Lemba wife, are planted respectively a baobab tree and a mfuma silk
6
cotton tree. Surprisingly little of this fixed architecture has survived,
either in museum holdings of sculpture or in photographs and
drawings.
More is known of the shrine context of the Lemba house. Lemba,
where it appeared, was evidently the major shrine installation in a
pantheon of n 'kisi figures and objects. In the back-yard shrine-garden
of Yombe chief Ngambula, whose life and Lemba priesthood was
depicted in Chapter 3, Lemba's shrine was flanked by other
min'kisi, including: Simbu, in annzungu pot, deity of time; a small
ndubi statue with a mirror in its stomach (kundu); a lukatu statue in
a pot, also with a mirror in its midsection; Mbudila, an n'kisi figure
related to time and to the reign of the Mamboma chief; and a variety of
others called Mambinda, Maluangu, Mangaka, and Nzola, either
7
represented by a pot, a statue, or another object
The transformation of Lemba'% symbolism in the western region
corresponds, then, to a general shift from moveable to fixed estates;
from hunting and shifting cultivation to fixed landed domains; from
segmentary polities to kingdoms and chiefdoms; from caravans tra
versing vast stretches over trails to endpoints, ports, and commerce at
the coast.
ECONOMIC AND EXCHANGE STRUCTURES
A brief perusal of the exchanges inBabutidi's account (Text 10, figure

19) suggests an important difference from exchanges in other
accounts already examined. Here the main payments are made in the
currency of the late Free State and early Belgian colony. Only pay-
246 RITUALS
Figure 19
Exchange structure of Lemba account,
after Babutidi (Text 10)
Neophyte Takes Mwemo Medicine
mpata ( 5 f r . ) chief Lemba

Lemba child, mwemo priest
"sufferer"
Bargaining the Fee, Presenting Neophyte
mpata 1 0 - 1 5 ( 5 0 - 7 5 fr.) [ • chief priest
mpata 2 3 ( 1 0 - 1 5 fr. e a c h , 3 0 - 4 0 L e m b a priests

= 3 0 0 - 4 5 0 fr.)
L e m b a child
promise to c o m p o s e
L e m b a for neophyte
1 pig
mpata 3 ( 1 5 fr.)
2 nkwala mats wife of chief priest
2 mpidi baskets of
mbongo raphia cloths
opening
feast?
Composing Medicine & Marriage

n'kisi in
cylindrical
container Lemba priests
L e m b a neophyte
(nkela)
L e m b a couple marriage Lemba priests

& bracelets?
Purification Ceremony
moata 4 ( 2 0 fr.)
singing & d r u m m i n g L e m b a priests &
to c o u p l e under priestesses
Lemba blanket-shroud
couple
Closing Ceremony in Forest

officiating of
L e m b a couple ~« ceremony, Lemba father
anointing red & w h i t e
ments to the chief priest's wife are in the traditional currency of

mats, mbongo raphia cloth, and pigs.
Following the Free State's introduction of brass wires in the
eighties to compete with indigenous copper wire, the mpata (=five
Belgian francs, ten United States cents) was introduced in 1910 by the
Belgian Congo concurrently with the imposition of the head tax in this
currency. With the creation of railroads on the south bank of the
Congo River and in the Mayombe forests, European commercial
interests were more effectively able to reach the sizeable Lower-
Congo, north-bank productive resources. The old ability to mount
caravans was revived briefly as individuals, lineages, and local chiefs
portered palm nuts, ground nuts, beans, tobacco, even sheep and pigs,
to these commercial outposts in exchange for colonial money and
manufactured goods. Direct equivalences are not easy to establish
between pigs, francs, raphia, and Lemba, as Chapter 2 has already
suggested. It may be noted, however, that Babutidi's account com
piles about 5 80 francs, one pig, and two baskets of raphia cloth as hard
items of cost. In terms of trade figures of ca. 1928, this would have
required bringing something like fifty to seventy man-loads of palm
nuts to the nearest purchasing station, at thirty kilograms per person.
Or, in terms of workers' incomes in the towns in 1925-30, about forty
months of rations for one person (at ten to twenty francs per month),
8
and two years salary (forty to sixty francs per month). In other words,
only someone who possessed the patronage of his lineage or who was
a merchant of means could afford Lemba?$ initiation. It is surprising
that anyone could still afford this type of ceremonial expenditure in an
era of the head tax, corvee labor, and the emergence of wage labor and
the exchange of wages for manufactured goods.
The level of expenditure, and the involvement of thirty or forty
Lemba priests in the initiation, plus their wives, suggest something of
the commitment to the institution in the 1910-15 era in eastern
Mayombe, and the extent of Lemba's authority even this late in the
colonial era
VERBAL CATEGORIES, RITUAL ACTION, AND LEMBA NAMES
There is something eternal about Kongo ritual. It is continually

extracting from the events of daily life and human society the salient
features which pertain to lasting perspectives, values, and categories
of Kongo culture. This process is far more enduring than a particular
individual, chiefdom, lineage, kingdom, drum of affliction or religious
movement. But these latter are situated in terms of the enduring. In the
248 RITUALS
western coastal region there are schemas in the historical record which
relate the various kingdoms and ritual associations, and individuals
within them. Thus, in the Loango of 1875, according to Bastian,
someone suggested that Loango, Kakongo, and Ngoyo were like hus
band (nnuni), wife (n'kazi), and priestly mediator (itomwa=one
who is sent) to each other: a complementarity of roles. Ngoyo was the
home territory of the prestigious Bunzi shrine, not only the ritual basis
of the Ngoyo kingdom but also that of Kakongo and Loango as well.
In actual practice this meant little more than that the northern king
doms sent emissaries to Bunzi during important decisions and transi
9
tions, to consult В unzi's oracle. Throughout the region another prin
ciple of spiritual "rooting" existed in the notion of the xina or zina
"name" that flowed from a person's father to himself or herself, and
back again to the father's lineage in the next generation. In noble
marriages this principle was very important, and a person's status
depended on his having a "father" and a "name" (zina). Names and
role terms, therefore, reflected a classificatory scheme deriving either
from the values of kinship, class, or polity. Lemba was no exception,
and its relationship to other social processes and schemes is the
subject of this section.
Konda and Babutidi, like Kwamba in his description of the north
ern variant, offer very clearly labeled categories of ritual action in
Lemba, and the names received during the inaugurals by Lemba
priests reflect these categories closely. The ritual process is described
in terms of a few central concepts, beginning with the term кота, "to
strike, augment, obligate, assemble, or constrain." The term has
achieved some notoriety in Kongo studies from its use in connection
with nail or wedge charms and the driving in of the wedges (кота
nsonso, кота m'funyia). Konda uses it in a broader sense to refer to
the awakening of a force which may be either negative and injurious,
or positive and redemptive. Lemba is "aroused" (komwa) through
dreaming of a. Lemba ancestor, having a nightmare that one is suffo
cating (being bewitched), or being possessed by a spirit (9.7). A
person may also have his "speech blocked" (9.20). All are symptoms
whose etiology may lead to the recommendation that the afflicted
should take the Lemba medicine and be considered for initiation.
Both Konda and Babutidi speak of the priest in Lemba experiencing
further negative ritual action in his defilement (sumuna Lemba, 9.27;
sumuna mina dia Lemba, 10.28).
Positive ritual action to overcome these negative states includes the
expected prescriptions to drink the Lemba medicine (nwa mwemo a
Lemba, 10.21), consecrate Lemba (10.7), and initiate to the nkisi

(vanda, 10.19, 23, 32). There are also idiomatic terms of positive
ritual action such as "cracking the lutete seed," done by the presiding
priest as a sign to the " afflicted" that he may now loosen his words and
tell all he has seen in his dream or nightmare, in particular the identity
of the person or spirit afflicting him. The basic term of ritual action,
koma, is used with the verb loka, to speak or cast a spell; thus koma
nloko. Whereas many Lemba priests and couples use their nkonko
drum to create a spell, some do this by rubbing or spitting palm wine
on their copper Lemba bracelets (Konda, text 9.9-12).
Possibly the most indicative phrase in the western variant depicting
Lemba's ritual action is this: "Lemba is capable of exchanging wealth
5
of the ancestors with that of the living so it may be at the latters
disposal... so they exchange in the entrances whence wealth comes,
and those who are involved with Lemba with their problems receive
Lemba's wealth" (Yandi Lemba lenda heha mbongo zena kwa
bakulu ye zena kwa bamoyo mu diambu dialenda kiandi kiena
kwandi mu twadisa mbongo ... buna bahehanga mu mafula
makwizila mbongo mpasibantu bakaka mambu Lemba basinduka
mo mana bakamana mbongo za Lemba, 9.17). The key terms here
are heha and fula. The first, also veva or veeva, means to blow,
clean, softly vibrate in the breeze, flutter, as well as to be bewitched or
have a spell cast over one. In its substantive form the term becomes
mpeve, which Bible translators took to express spirit, whence mpeve
a nlongo, Holy Spirit. The second term, fula, denotes to blow, spit, or
channel a substance, or a path, route, and the wind following these
paths. One may fudila n'kisi, arouse an n'kisi by spitting on it, or
fula nzonza, arouse a quarrel (zonza, rhetoric, speaking). Mafula
are entrances to villages, markets, cemeteries. The verb fula is used
in the eastern variant, as well as in other accounts of Lemba in the
Mayombe, to refer to the ritual resurrection of the neophyte. The
adept, having "died" (fwa ngambu) and seen spirits, is awakened
(fulukidi) and given to drink from the tsasa pot by the Lemba
priestess. In the east, the priests sing the song "Resurrect the Child"
(kimfula, fudila mwana) as an invocation to Nga Malamu while one
of the priests blows into the neophyte's ear. Related to the same
cognate is mbondo fula, the broom ofjustice and oratory, depicted on
some Lemba bracelets (plate 12). The text (9.17) suggests that
Lemba is able to arouse the ancestors in a type of sacrificial trade so
as to enhance the wealth of the living. This process occurs most
auspiciously in the "entrances" (mafula) which is, indeed, where
250 RITUALS
much of the actual exchanging is done in markets (see plate 3). It is

this term fula, to raise, blow, bring into being, to "spiritize," which
becomes the key word in Konda's concept of '' spirit of mercantilism''
or "capitalism" (heha mbongo, arouse the spirit of wealth; or
hehanga mu mafula makwizila mbongo, arouse the spirit in the
entrances whence comes wealth).
The names of Lemba priests reflect preoccupations with the range
of ritual actions already suggested: oratorical skill, grasping or seeing
mystical power, enjoying full ritual purity, and increasing offspring as
well as wealth. The names are chosen to characterize the new person,
resurrected to Lemba and its prevailing values. But they also uncan
nily portray persons grappling with personal problems. Bittremieux
10
cites the following. Mvuza, "jabberer," is the name of one whose
words now flowfreely, recalling the affliction of the Lemba sufferer
with "blocked words" (9.20). Makunduku, "he masters powers,"
suggests a preoccupation with the problem of power. The full name is
drawn from the phrase makunduku va mbata vwa, "on our death
God alone is almighty," recognizing human mortality. Ngambula(cf.
Chapter 3) means "washer-ofF, from menga ma tsusu, simba:
kwambula, "the blood of chickens; hold it: release it," that is, one
who is not afraid to penetrate to the core of a serious social issue, and
deal with it, but who then is able to extract himself from it. Nyambi-
vana, "God-gives," is used for someone who has accumulated the
requisite wealth to enter Lemba while still a youth.
Thus far these names reflect serene spiritual Lemba values. Others
reflect economic values of acquisitiveness. Mvu, "year," is the name
of a priest who says "a whole year I have slaved to be able to afford
Lemba's inauguration." Yindindi, "thinking," refers to one who was
always thinking how he might make his fortunes cover Lemba's
costs. Valata, "scraping and scrounging," referred to one who con
sidered himself having to scrape, scrounge, and struggle to afford a
Lemba marriage. Phila-mose, "same-as," is the name of one who
says to himself, "Now I stand on the same footing as the old men who
had to collect the necessary sum over many years." Mueba is the
name of the chief priest, after the mueba tree (Irvingia Barteri),
referring to the notion that "everyone gets his just desserts—that is,
fruits." Sabu, "crossing-over," refers to one who tsabukidiLemba,
crossed over into Lemba, one who "had his money working." Or
Vandu-vandu, "charm-activator," from Dilemba diedi kuiza
vandulu zimbongo, "my Lemba, when it's turned on will enrich me
(in the funds and children)." These names bring out the variety of
values emphasized in Lemba, and their rooting in broader categories

of ritual action.
It remains, however, to explain the link between this spirit of
acquisitiveness and mercantilism so prevalent in the foregoing names
and ritual acts, and the symptomatology and etiology of the Lemba
affliction. Konda suggests that the affliction is brought on by some
thing the sufferer may have seen in a dream, a nightmare, or posses
sion by a spirit (9.20-22). Babutidi speaks of it as something that may
affect abdomen, heart, head, and side (10.20). Further consideration
of this issue will be deferred till Chapters 9 and 10. It is clear,
however, that this Lemba "spirit of mercantilism" provoked consid
erable antagonism in a society in which the ethic of redistribution is
deeply engrained.
MEDICINE AND ART
In previous sections on Lemba medicine the contents of the n'kobe

have been examined in terms of their function. Here an interpretation
will be undertaken of all material objects created specifically under
the rubric of Lemba medicine: the nkobe (or western equivalents),
the fixed shrine, the bracelet, the drum, the rattle or bell, pipes, and
statues. In each of these mediums, individual objects—wood, plants,
metal—are shaped, named, and brought together in associative cate
gories. They are thus given a signification. Sometimes these objects
may contain bodily elements of persons involved—the neophyte's
semen, hair, nails—and thus have an added, iconic signification. In
other cases they may bring together statements of symbolism known
in the world of conventional meaning, such as the relationship of men
to women with their characteristics. A complex ritual system such
as Lemba, however, takes these iconic and symbolic statements and
reshapes or reassociates them for its own purposes. This process,
which has been shown to use and construct metaphor, can be analyzed
to gain special understanding of the institution's major concepts and
values. The metaphoric process uses parallel series of expressions to
make a composite statement about society and the universe, about the
relationship of resources to power, and about the living to the ances
tors from whom all power flows. Because Lemba expressiveness
reaches out to make statements of these social, natural, cosmic
wholes, aesthetic criteria are implied. That is, elements are abstracted
from the whole in order to represent it. For example, the western
Lemba bracelets depict a man and a woman joined by a flower-petal
252 RITUALS
motif or a cowrie shell (plates 8-12). Miniature drums worn on

necklaces or sculpted on pottery lids depict the drums of Lemba. In
many cases these miniatures are worn as jewelry and used as charms
(see plates 15, 16, 18). Mythic heroes, represented in the nkobe as
shakers or rattles, are often hung from the priests' belts during dances
dedicated to the deity or hero.
The most common statement of signification in the Lemba n'kobe
Figure 20
Tukobe of the Western Lemba region:
Loango and Ngoyo (Cabinda)
basket filled
w i t h s a n d y clay
c h e w e d kola surrounding
p o w d e r of r e d w o o d w o o d e n statue,
holy water. barely protruding
at t o p
"small pot"
In mat, "red"
cowrie s h e l l
bits of bottles"
shells
feathers
trade b e a d s -
("tubes of
clay")
bones
small bells
claws
loose in nkobe,
rags
pipe resin drops
rattles w i t h feat
shell
seed
"drum played
by boy w i t h
hand"
L o a n g o Coast, after N g o y o (Woyo),

Dapper, 1670, pp. 336-7 MAC 35191,1933, collected
by Bittremieux (Plate 4)
and shrine is the relationship between male and female. This rela
tionship refers not only to the reproductive process in which children
are born, but also it takes the male-female union as a model for all
complementarity between social groups, for example, neighboring
lineages. In the Yombe variants of the Lemba n'kobe (figure 21),
male and female complementarity is couched in a number of meta-
phoric contrasts summed up in the names Tsasa-Lemba SLndPfemba-
Lemba respectively.
Western Mayombe, Kangu, Eastern Mayombe, Mamundi,

MAC 43040,1937, collected after Babutidi (Text 10)
by Bittremieux (Plates 5-7)
254 RITUALS
The iconography ofTsasa and Pfemba is rather complex and

needs further clarification. Firstly, Pfemba is not the same as
mpemba (chalk, whiteness, clarity), despite Babutidi's use of the term
(10.6; 10.42; 10.48) and scholarly attempts to explain the apparent
11
contradiction of "red" symbolism being named "white." Pfemba,
the "female" satchel, is indeed "red," as Bittremieux's research into
12
the Mayombe nkobe he collected (figure 21) indicates. The Pfemba
packet in Lemba is closely associated with the well-known Pfemba
(also Phemba) maternity figures of the Mayombe and the coast,
figures which reveal often a reddish hue of tukula anointment. These
figures, which attracted collectors' attention in the last century and
are therefore widespread in European museums, are in turn related to
an n'kisi cult which had to do with women's problems in general and
midwifery in particular. Bastian and Pechuel-Loesch, who visited a
Pfemba center in the 1870's, recorded a legend about the well-known
13
midwife who founded the movement
Ethnographic research by Lehuard has identified the pose of these
children in arms and laps of Pfemba maternity figures as dead child
14
ren being mourned by their mothers. One possible interpretation of
this puzzling posture, one which I propose, is that these children are
immaculate-conception, spirit children akin to simbi children "taken"
by the spirits. In this way they reconcile, or mediate at an abstract
level, the domestic role of childbearing with the notion of female
rulership. In Chapter 2, I suggested that one could find many
variations on the theme of fertility—augmenting local kin groups—
and leadership transcending kinship to create an overarching political
order. Ancient Loango and Sonyo noble marriages served a purely
political purpose, as was seen in Chapter 2, and were commonly not
even consummated In Lemba, where fertility of the local clan
section was combined with political alliances, Pfemba "redness" in
the "female" section of the nkobe offered a unique alternative
resolution of this problem of fertility and leadership, of the exclusivity
of local kinship and the need to create broader alliances.
This male-female relationship as articulated in whiteness and red
ness, respectively, is extended further by the relationship of the tree-
climbing mbala civet cat to the burrowing rodent nkumbi, in whose
skins Tsasa-Lemba and Pfemba-Lemba are kept. This metaphor is
nearly identical to that of the eastern variant (Chapter 5) except that
there the nkumbi skin contained the male "white" ingredients, and
the musimba, another tree-climbing cat, contained the female ele
ment. However, eastern sources are not reliable enough to make
anything of this symbolic inversion.
9
The third satchel in Babutidi s depiction of the n'kobe (figure 21)
extends several domains of the above metaphor—social roles and
small animals—to the patrifilial line of Lemba. Hair from the heads of
the Lemba chief priest, from the presiding Lemba father, and the
neophyte, along with leaves of lemba-lemba and dingongo beans,
are placed in a container of the skins of nsesi and kubu antelopes
and nkaka, the pangolin. The correlation of these skins to the three
roles is not given; however, their stereotypes are suggestive. Nsesi is
slight and clever, swift and evasive, and is depicted in fables as able to
outwit larger animals. Kubu is a marsh antelope. Pangolin is
recognized not only for its anomalous character, its scales and its
mammalian features such as suckling its young, but also for its long
tongue which can penetrate any cavity in a termite mound. Indications
are that the metaphor linking Tsasa and Pfemba to mbala and
nkumbi skins also relates the three figures of patrifilial continuity—
from mystical purity to profanity—to the three animals whose
characteristics extend from water to land, from inner to outer qualities.
Combined, the three satchels in animal skin situate the key social
roles of Lemba through marriage and patrifiliality in terms of natural
attributes, animal and cosmological, as shown in figure 22. The cos-
mological "coordinates" appear to be plotted on two axes: one vertical,
linking trees to underground burrows, the other Unking the water, sign
of mystical communion, with land and the outside with the inside of an
anthill or termite mound.
The column of plants links the above-mentioned three satchels,
known as the " domestic abode of Lemba'' to the fourth satchel in the
n'kobe, known as the "power of Lemba." Whereas "abode" is con
tained in wild animals' skins, "power" is contained in a domestic
plant, raphia. Whereas "abode's" plants are semidomesticated, wild
plants growing in the village, "power's" plants, many in number, grow
wild. Subtle contrastive principles of inversion appear to be at work
here. Whereas the cosmological spacing and character traits of the
animals explained their use in defining social roles in Lemba, the use
of raphia, and the contents of the satchel called "power," define
another attribute in Lemba, its adherent's skill or ability in capturing
the extensive, outside, wild, and public forces. The plants listed in
Babutidi's "power" satchel are like those collected in Kwamba's
northern variant during the priests' final trek onto the savanna and to
the cemetery. In Fukiau's account raphia is included because it stands
for the Lemba priest's ability to journey far without being interrupted.
The cognate satchel in another Mayombe n'kobe (see figure 21, col
lected by Bittremieux) contains many trade goods, including iron slag,
256 RITUALS
Figure 22
Dominant metaphor of correspondences,
Western variant of Lemba
author's name sexual roles substance animal cosmologie plants
analytic and social and order space
category identities color
"Tsasa" male white mbala aboveground,

"clear" cat tree-level
mpemba
"Pfemba" female red n'kumbi burrows

tukula rat underground
domestic
abode
high priest kubu water lemba-

antelope lemba
Lemba Father n'kaka termite di-

pangolin mound ngongo
bean
Lemba child nsesi forest

antelope
public savanna raphia

power palm
a nail, a key, a ring, trade beads, and shells. Dapper's account of

Lemba from seventeenth-century Loango has European goods (trade
beads, a bottle, a bell) and shells bespeaking the coast laid upon a
"mat," surely a raphia mat. Running throughout the contents of
the tukobe is this theme, then, of the power that comes from knowing
and controlling the outside, the realm beyond the domestic.
In most tukobe this concept "beyond the domestic" is linked to the
ancestors, either in the form of a cultic hero's effigy in the nkobe in the
form of a small "nzita" figure used on a dance rattle, or as an mbinda
figure hung on the belt, or as a piece of ntobe tomb earth. The inclu
sion of savanna or forest plants in this concept's representation, as in
the satchel "power of Lemba" above, would be consonant with a
broader, cosmological statement linking ancestral and natural
powers.
The Lemba drums, bracelets, rattles, and statues are the instru
ments with which this power is captured and used. Numerous texts
speak of the ancestors or mythic heroes being "in" the drums, or of
them "speaking" through the drums, or of Lemba being "drummed
up" by the Lemba father or chief priest The drum of Lemba—
whether nkonzi, nkonko or ngoma—wsiS frequently copied in
9
miniature, the cavity being filled with plant substance (the "powers"
above) to retain or conjure a spell (ndokolo). Such a process of
representing a representation, or metaphorizing a metaphor, may
permissibly be termed fetishization. The miniature drums of Lemba

were either hung on statues, or carried about on one's body as a charm
(see plates 13-16). On Woyo pot-lids illustrating proverbs, Lemba
15
was represented by the drum motif.
Bracelets, like drums, contained the power of Lemba. The Lemba
bracelets were sometimes wrapped in raphia fronds to capture the
16
palm fiber's connotation of movement beyond the domestic realm.
Elsewhere the bracelets were regarded as substitutes for the drums,
for, as Konda suggests (9.9-11), the wearer could spit palm wine on
the bracelet, rub it, or simply wear it in public to demonstrate his real
social status and power. Motifs appearing on Lemba bracelets indi
cate specific applications of power. In the western variant, male and
female images representing the Lemba household are very common.
The male figure, usually seated, is accompanied by one, two, or three
female figures. The male figure is holding a staff, whereas the female
figures occasionally have in their hands what appears to be a pestle or
pipe. Between the figures, and on bracelets without human figures,
other motifs make their appearance.
One of the most common of these decorative motifs on the bracelets
f
is the cowrie shell. It may also appear atop the n kobe, or atop one of
the interior containers of the n'kobe, or within a satchel. Its general
meaning is that of cherished continuity in the populace, that is the
precious quality of children, and the importance of fertility. It thus
defines, on the bracelets, the relationship of the male to female figures
also represented there.
In a cognate position to the cowrie, many bracelets display a flower
or "spider-web" motif, with four, six, eight, or ten "threads." The
exact meaning of this motif remains unclear. It resembles Cameroon
spider masks which have the same motif atop the human head. The
spider occurs as mystical mediator in Kongo legends, including the
trickster Lemba origin myth to be examined shortly. The motif may
also represent the diyowa of Lemba, the place of purification and
clarity in the face of the powers of the beyond. On one bracelet the
motif is presented as a four-petaled flower or wheel and as a cross.
Other bracelets juxtapose the flower/wheel "cosmogram" with a
motif resembling a bow tie, but which is the rhetorician's charm
mbondo fula, worn or held during intensive palaver sessions by the
professional advocate (nzonzi) (plate 12). Mbondo fula appears on
pot-lid proverbs along with the Lemba drum and other musical
instruments, to acclaim the virtues of authority in the household and
17
of persuasiveness in public life. In light of the importance of rhe-
Plate 5. Lei nba medicine chest lid from Mayombe, with evidence of pointed zin±
shells, cow *ie shell stand (center), and smith's bellow motif. Collected by L. B
tremieux pr or to 1937. (MRAC 43040.) Contents depicted in Figure 21.
Plate 6. Contents of Lemba chest (in Plate 5), identified further in Figure 21.
Plate 7. "Powers of Lemba" raphia bag contents from Lemba chest (Plates 5
and 6), including ironworking articles, locally manufactured and traded;
trade beads, shells, horns, and seed pods. (MRAC 43040.) See Figure 21.
Plate 8. Detail of cast copper Lemba bracelet from Loango coast, collected
by Adolph Bastian, ca. 1870. Museum für Völkerkunde, Berlin-Dahlem III
C 347. Drawing of full bracelet with three figures and multi-petal motif
from A. Bastian, Deutsche Expedition an der Loango Küste, Jena, 1874.
Inside cover plate.
Plate 9. Brass bracelet with cowrie motif, often denot
ing fertility, possibly in this case in connection with
Lemba, Kunya. (Museum für Völkerkunde, Berlin-
Dahlem, III C 6524. Donated by Robert Visser, 1896.)
Plate 10. Wooden model for Lemba bracelet
imprint in casting sand or clay, with multi-
petal motif over human figure; Vili. (Mu
seum für Völkerkunde, Berlin-Dahlem, III
C 8136. Donated by Robert Visser, n.d.
[1890's].)
Plate 11. Cowrie motif on opposite side of wooden
bracelet model pictured in Plate 10. (Museum für
Völkerkunde, Berlin-Dahlem, III C 8136. Do
nated by Robert Visser.)
Plate 12. Copper Lemba bracelet with floral motif (top)
and orator's charm mbondo-fula (bottom). (MRAC
24985, collected by J. Renkin, 1909.)
Plate 13. Brass Lemba bracelet with four human figures. (MRAC
54.60.2, collector unknown.)
Plate 14. Ngoma-Lemba dram, Mayombe. (MRAC 34778, donated by L.
Bittremieux, n.d.)
Plate 15. Miniature charm Lemba drum necklace, probably Loango coast.
(Linden Museum, Stuttgart, 38363, donated by Robert Visser, 1905.)
Plate 17. Mukonzi a Lemba drum, Mayombe. (Statens Etnografiska Mu
seum, Stockholm, 1919.1.437, donated by Karl Laman.)
Plate 18. Couple seated on trunk (in wood, 14 cm.), Landana, Cabinda.
(Afrika Museum Berg en Dal, 177, n.d.) That this is a depiction of a Lemba
couple is suggested by the figures' matching bracelets, female figure's
miniature drum necklace pendant, the posture of closeness, and the caps
with characteristic "multi-petal" motif on male's cap (see Plate 19).
Plate 19. Couple seated on trunk, Landana, Cabinda (rear view). (Afrika
Museum Berg en Dal, 177, n.d.) Multi-petal motif on top of cap of male
figure contains spiral recalling similar motifs on bracelets—above heads—
and on Lemba chest lids. Although obscure, this design is thought by some
to denote idea of "navel of head," i.e., thefontanelle, or "fount of wisdom"
from beyond. This would explain why many Lemba figures, like contempo
rary Kongo prophets, wear caps, or protect hair.
Plate 20. This undated (40 cm.) Loango ivory ((A) front view, (B) rear view) from
the Julian and Grace Rymar collection depicts a series of couples and poly-
gynous household groups in poses and costumes characteristic of a nineteenth-
century Loango gathering, possibly including Lemba seances. As shown in the
detailed view on the next 2 pages, the top couple, arms embracing, suggests the
closeness of a Lemba couple. He is smoking a pipe, recalling references to
Lemba pipes (B-D III C 13873 and B-D Katalog No. 372, Appendix) and to
smoking in the Lemba ceremony (see Brazil Lemba song, Chapter 8, opening).
She is holding her breast in a Kongo gesture of benediction. The tops of their
caps bear spiral designs similar to those of the couple depicted in Plates 19 and
20. Other figures and scenes in the panels of the ivory are less clear, but depict
possible ceremonial meaning as well. The top scene, which suggests defloration,
childbirth, and an emphasis On fertility, stands in interesting inverse relationship
to the bottom scene, in which the accent is the top of the head. Presumably the
top figure is female, the bottom male. The two depictions match the iconic
duality in Lemba symbolism on the integral complementarity of fertility, sym
bolized often by the cowrie shell on bracelets and the medicine chest, and on
power or clairvoyance, symbolized by the multi-petal motif on bracelets, the
medicine chest, and atop caps. However, a fuller interpretation of this rich piece
awaits the systematic study of several centuries of Loango ivory carving and
related art.
258 RITUALS
torical skill in public elsewhere in Lemba''s self-image and of the

equal importance of a flourishing household represented in the cow
rie, it is likely that the flower/spider/wheel motif, which in one or two
cases is a "cross" of four spokes, is a unique Lemba cosmogram
embodying the complex values of Lemba at a more abstract level than
is given in any of the texts at my disposal. In the Ngoyo region it
appears in red on the lid of the nkobe (see plate 4). Another Lemba
n'kobe lid (figure 21, plate 5) restates comparable values with the
arrangement of four-pointed, long zinga shells surrounded a center,
interspersed with blacksmith's bellows and some other round object.
Again, these are the objects of the "powers of Lemba" inBabutidi's
account, which link the Lemba adherent with the forces of nature, the
beyond, trade, and public politics.
THE LYRICAL: MONI-MAMBU TRICKSTER AND THE ORIGINS OF LEMBA
Trickster, represented across Kongo society, is variously called

"Seer of troubles" (Moni-Mambu, Tsimona-Mambu), "Trouble
maker" (Mumboni-a-Mpasi), quarreller, or some other term describ
ing his problem-making nature in most secular tales. He is also called
visionary and healer where legends of him are related to a cultic
context, as in the Lemba etiology myth to follow, where he is cast as
the culture hero who goes to God the father to seek a solution to his
wives' troubles, encounters many difficulties and trials, but eventu
ally receives his father's recognition, and brings back Lemba.
If Mahungu's principal characteristic was the inherent dualistic
nature of a social role, Moni-Mambu's is the lack of a sense of what is
right at the right time in social discourse. In most legendary settings,
his character is in the process of developing. Usually his civil, moral,
and even physical sensibilities are incomplete. In episode after epi
sode he is caught up in social intricacies and ambiguities he does not
understand, confused by semantic nuances too subtle for his experi
ence. It is as if he has learned the phonemes of human interaction but
has not heard of syntax. He is like Mahungu in one sense. In some
narratives Moni-Mambu develops into a figure of mediation and
wisdom, yet in other narratives he is parodied as unsuccessful media
tor, indeed, as a tragic figure who gets caught on the dilemmas of his
own tricks. Whereas both tendencies are developed in a single
Mahungu narrative, they are separate in the trickster narratives. In the
first of the following texts, trickster brings Lemba.
Text 11
(1) There once was a man with two wives, each with a food
prohibition: one could eat no lizard, the other no partridge.
Their husband brought home the prohibited meat and, after
discussing whether they should eat it, they gave in to tempta
tion and prepared the food, not knowing what would be their
due. They ate and suffered swollen stomachs and cramps.
(2) To find a cure for his wives the man went to Fly. He told Fly of
his wives' food prohibitions, and asked him to "smell out" the
cause of the swelling. Fly "smelled" and danced about, but
could not divine the cause. The man, dissatisfied, got his
money back. Then he went to another nganga diviner, Night-
Swallow, who researched the cause and told him he would
need to go see God the Father with his problem.
THE JOURNEY
(3) Taking his nkutu travel sack, the man set out to see God the
Father. Just outside his village he stubbed his toe on a small
stump which spoke out at him, "Go on, Tsimona-Mambu, pay
attention, this is a mysterious n'kisi affair."
(4) Tsimona-Mambu walked on until he came to an abandoned
village with only one house left in it. As he approached, the
keyhole in the door spoke out at him, "Go on but pay attention,
for this is a mysterious n'kisi affair."
(5) H e met a woman working her groundnut patch. She inquired
where he was going, and he replied, "To God the Father, with
the blessing of the nganga" She invited him to eat a snack,
offering him potatoes with her child, who was resting under a
shelter. Meanwhile she returned to her work. After a time she
called to ask how he liked the potatoes—and where was her
child? He had eaten it, he retorted, as she had told him to do.
"Tsimona-Mambu has eaten my child! The murderer! Come
and seize and bind him," she called out to the menfolk.
Tsimona-Mambu was taken before the judge, but because he
had not been aware of what he was doing, he was freed.
(6) Again free, Tsimona-Mambu walked on until he came to a
stream, surprised to see the water flowing upstream. Dis
5
believing, he exclaimed, "That can tbe! You will run dry!" The
260 RITUALS
stream answered, "Don't worry, and go on. This is a

mysterious n'kisi affair."
(7) Tsimona-Mambu next saw a young man tapping palm. To his
amazement, the man had left his bones on the ground and was
climbing the tree flesh only. When Tsimona-Mambu asked
him the meaning of all this, he retorted, "Go on, it's a
mysterious n'kisi affair."
(8) He walked on again and encountered a man with a gun across
his shoulder. "Where are you going?" the man greeted him.
Tsimona-Mambu explained about his wives and the doctor's
blessing. The man interrupted him, "Can you shoot?"
Tsimona-Mambu replied that he could. "Take my gun then
and go hunt monkeys." Tsimona-Mambu returned to the palm,
where the youth told him to " stay hidden, for the monkeys will
soon come for nuts." As soon as there was a rustle Tsimona-
Mambu shot out, dropping the game to the ground. He made a
net of palm fronds, and dragged his catch to the nearest village.
Curious people gathered. Alas, it was their headman. In anger
the people seized Tsimona-Mambu and brought him to the
judge. But because he had not known better, and because he
had used the hunter's gun, Tsimona-Mambu was freed.
ASSEMBLING THE SATCHEL CONTENTS
(9) Tsimona-Mambu encountered a wood-borer working on a log.

He called out, "Tsimona-Mambu, where are you going?" He
told him "To see God the Father." The beetle requested to be
taken along, so Tsimona-Mambu put him in his nkutu bag.
(10) They came to another village with only an old cripple at home.
"Where are you going?" this latter asked. "To God the
Father" was the reply. The cripple asked to be taken along, but
Tsimona-Mambu refused because of his many sores, and left.
Before long the path thinned and became a wild animal track
Returning, he asked the cripple where the road could be found.
The old man ribbed him, saying "Aren't you sorry you didn't
take me along?" So Tsimona-Mambu packed him into his bag,
and was off, the cripple directing.
(11) They walked a while until they came to a giant spider spinning
her house. "Good-day, Tsimona-Mambu" she spoke out.
"Where are you going?" He told of his two wives and how he
was going to see God the Father. "Can I come along?" the
spider requested. Reluctantly, because of her big crooked legs,

he put her in his bag and continued.
(12) A big wasp flew down to meet them.'' Where are you going?'' it
inquired. "Can I come along?" He told her where he was going,
but refused at first to let her come along because he felt she
would make an angry nuisance of herself. But she insisted, so
he put her in his bag and moved on.
(13) Shortly they came upon the summer wind, which blows in the
hottest month of the year. " Yeka, yeka, yeka" it was singing.
"Tsimona-Mambu, where are you going? I can reward you for
letting me come along by drying up everything." Tsimona-
Mambu replied, "You who scorches the whole earth, how can I
take you?" "In your bag, of course." So he packed up the hot
wind and moved on.
(14) As soon as he had gone from there he was met by another gust
of wind. "Brrrrr, Tsimona-Mambu," it blustered, "where are
you going?" "The wives are sick and I am going to God the
Father," answered Tsimona-Mambu. The wind wanted along.
"What? A storm in a satchel?" But in the end Tsimona-
Mambu packed it too in his bag and went on.
(15) Soon he came across a big kumbi bush rat who greeted him
with "Where to Tsimona-Mambu?" "Why to God the
Father." "Can I come along?" Tsimona-Mambu agreed, put
the kumbi rat in his satchel, and walked on.
(16) Eventually he came to a village and encountered the headman
in the square. "Ho, Tsimona-Mambu, what's new?" And
Tsimona-Mambu again told his story about the two wives, how
they were sick, and that he was going to God the Father." C an
you shoot well?" interrupted the headman. "Of course," said
Tsimona-Mambu. So the headman gave Tsimona-Mambu his
gun and told him to go stand guard at the banana grove
overnight and shoot the first creature which would come along
to steal bananas. Early the next day along came the headman's
wife to get her supply of bananas. Tsimona-Mambu, believing
her to be a beast out to steal bananas, shot her. Making a quick
holder of banana leaves, he laid his prey in it and came to the
headman proudly showing it off. People gathered. Shocked,
the headman exclaimed, "Why, you've killed my first wife, my
beloved MaMbanda!" He called the young men to bind
Tsimona-Mambu hand and foot, but to leave his mouth free.
262 RITUALS
Then he took him to the judge. When he had told his story, how
the headman had given him the gun, he was declared innocent
and released.
(17) Then Tsimona-Mambu came to the edge of the Loango River,
and hesitated. How would he get from the river bank to God?
An idea came to him. Tsik'utuk'utu, he said to his satchel,
"Are you still there?" "Yes," it answered. "Say, is the spider
still along?" "Certainly." "Well then, bring her out" And in
one-two-three she had spun a thread across the river, a bridge
from earth to heaven.
(18) Tsimona-Mambu mounted the bridge, higher and ever higher,
until he walked right into heaven. A s soon as they had arrived
there, the small cripple gave him these instructions. "When we
come to the middle of a village, you will see a big man sitting on
a stool. Don't greet him. But you'll see another, sitting on a mat
on the ground. That is God the Father, whom you will greet."
Tsimona-Mambu did just as told, and passed the big man and
came before God the Father. With three hand-claps—mue-
mue-mue—he greeted God and began to tell Him his
problems.
THE TRIALS OF FATHER'S RECOGNITION
(19) "O Father God! I have two wives. One is not supposed to eat
lizard, the other no partridge. Both have eaten what is for
bidden, and have swollen stomachs. I went to the doctors for
advice, and was told to come to You. That's why I've come."
(20) God spoke no word. Tsimona-Mambu implored him, "O God
Father, hear my problem, for I am your son." But God denied
this, saying "I shall believe that you are my son if green banana
stems carry ripe bananas, and if black palm nuts turn red. Only
then shall I know that I bore you."
(21) Fortunately Tsimona-Mambu's bag knew what to do, and
suggested in a whisper to him that everyone should first sleep.
The bananas and palm nuts would be ripe by morning. God
agreed. During the night Wasp went out to sting up the banana
stems so that the bananas ripened hurriedly, and the palm nuts
similarly so they would color.
(22) Then next morning Tsimona-Mambu told God's overseer to go
and inspect the bananas and palms. To his amazement they
were ripe. He notified God and asked, "Do you believe now
that Tsimona-Mambu is your son?" God did not yet believe it,
demanding that before he would recognize him, Tsimona-
Mambu would need to gather a large package of the strongest
forest vines in one night. Again Tsimona-Mambu consulted his
bag, and through it, advised all to first sleep. When morning
came the bag called the gust of wind which blew the vines
together. The overseer saw this and implored God to recognize
his son, but God refused to recognize Tsimona-Mambu.
(23) To recognize him as His son, Tsimona-Mambu would have t o
burn down the moist wood in the just-cleared field. Only then
would God acknowledge that he had engendered him. Once
again the satchel advised that all should sleep before the test. In
the morning God's overseer took Tsimona-Mambu out to the
field and invited him to prove himself by setting fire to and
burning the decaying vegetation. Tsimona-Mambu called into
his satchel for the hot summer wind; it came, blasted and
withered the clearing—у ayа, у ay a, yeka, yeka, yeka—and all
dried up so Tsimona-Mambu could easily burn it down to the
ground.
(24) The overseer implored God to consider TsimonaMambu's
problem so that he might return home, but the great Lord
insisted, "I engendered him not!" Tsimona-Mambu must first
drink a calabash of palm wine while hanging from the palm tree
on the back slope. So with God's overseer Tsimona-Mambu
found the tree. He climbed it, found the calabash, and began t o
drink. A s he drank the wine—па kiu, klukkluk—the overseer
addressed the tree, "fly up—кипа tsala kayeka—into the
highest heights so that Tsimona-Mambu dies from it." And the
palm tree jerked itself loose and sprouted higher and higher into
the sky with Tsimona-Mambu hanging on his life belt from the
tree's crown. High up in the air Tsimona-Mambu asked his bag
what he should do, certain he would die. The bag said, "pro
nounce the spell that the overseer sinks into the ground so far
that he too dies." So as Tsimona-Mambu spoke the overseer
began to sink into the ground, deeper and deeper to his neck.
"Help, Father God!" he cried. And to Tsimona-Mambu he
said, "Surely you are God's son." As he spoke the palm tree
settled down to where it had been before. And Tsimona-
Mambu permitted the overseer to crawl up out of the ground,
264 RITUALS
saying to him, "Let's go together because we are equal

in supernatural strength."
(25) Again they were before God the Father. Again the overseer
implored God to consider Tsimona-Mambu's affair and
acknowledge him as His son. But still God refused, stating he
would only recognize Tsimona-Mambu if he would first show
himself capable of felling the mfuma cottonwood tree growing
on the back slope with one blow of the axe. Only then would he
recognize Tsimona-Mambu as his son.
(26) Tsimona-Mambu asked the bag what to do now, and received
the word that everyone must first sleep. This they did,
including God. Then Tsimona-Mambu called the wood-borer
out of the bag and told him to go to the mfuma tree and hollow
out the inside so that only a thin ring remained. When morning
dawned Tsimona-Mambu asked the overseer to show him the
tree. The wood-borer had left a mark on the tree to show him
where to let the axe fall. Tsimona-Mambu took the axe and
with one blow brought the tree crashing down.
(27) The overseer once more imploredGod to"Take up the affair of
Tsimona-Mambu so he can return home. Admit that you are
his Father." And God finally showed his pity, saying "Go set
him something to eat in the house; he has worked hard."
Tsimona-Mambu entered, sat down on the bed, and ate what
had been put before him. But after he was in the house God told
his overseer to lock the door so that Tsimona-Mambu could
not leave, and would surely die inside.
(28) When Tsimona-Mambu realized that the door had been locked,
he consulted his bag. He was told to have the kumbi rat dig a
tunnel under the wall of the house. Several days passed and
Tsimona-Mambu became hungry. He instructed the rat to
bring him food by dragging it under the wall through the tunnel.
In this way he got a fat she-goat, a wild beast, and another
goat.
(29) After a long time God ordered the overseer to open the house
and remove Tsimona-Mambu's body. When the overseer
opened the door, and Tsimona-Mambu jumped out alive, the
overseer ran back to God saying that Tsimona-Mambu was
alive. He implored God, "Can't you see that he's your son?
Take up his affair now." And thus Tsimona-Mambu at last
won his audience before God the Father.
THE LEMBA MEDICINE REVEALED AND TAKEN HOME
(30) God spoke to Tsimona-Mambu: "Bring me a small pot covered

with a palm kete-kete leaf. Also bring the two copperringsthat
are in the house. The one ring you are to put on your wife's arm,
the other on your own. These are the arm-rings of DiLemba.
(31) Tsimona-Mambu brought all the things God had instructed
and then God questioned him: "Do you have the fruit of the
eggplant and the dilemba-lemba plant in the region where you
live?" Tsimona-Mambu said he did. "Then," said God, "take
all these things and use them to heal your wives. At the
outskirts of the village, pick some egg-plant fruit and dilemba-
lemba. Crush them in your hand into a pulp. Take the leaves
from the pot, and squeeze the juice of the two plants into it and
give the juice to both wives to drink. Their bellies will surely go
down and they will be cured and will be able to recount what
they have seen with their own eyes. With the two bracelets, you
shall do as I said: Put one on the arm of your wife, and the other
on your own arm. Go in Peace, and do as I have told you."
(32) As soon as Tsimona-Mambu came to the edge of God's village
he instructed the giant spider to crawl out of the satchel and
spin a web down to earth. On his way back he set off all the
creatures he had brought along. When all had been let off, he
came near to his village and went immediately to see his wives.
Their stomachs were still swollen— dio yukuku! dio yila
yila!—wp to their necks. He went to the village outskirts and
picked the lemba-lemba leaves and the egg-plant fruit, and
prepared the medicine in the small pot as God had instructed.
When the wives had drunk it, both stood up and were healed.
(33) To one of his wives he gave the copper bracelet and placed it on
her arm, calling her his NankaziLemba. The other ring he put
on his own arm. Then he told his wives everything he had heard
from God the Father. They were amazed, saying that this is
indeed an important enough cure that people the whole world
over ought to be inaugurated to it, for it came directly from God
the Father.
n
(34) Since then, people in this land are initiated to Lemba.
The manner in which an array of objects, personalities, and situa

tions confronts the hero on his journey suggests that if he is careful
order will emerge from the confusion of mundane life, as the divina-
266 RITUALS
tion suggested. He encounters warnings, traps, and offers of assis

tance. The dichotomy of fly, the unsuccessful daytime seer, and night-
swallow, the successful nighttime seer, indicates the type of cosmo-
logical order that will emerge if he follows the right path. This initial
divination scene (11.2) picks up the same imagery as songs elsewhere
in Lemba's initiation of the "sight" of the night-jar and the bat (1.76-
7).
The warnings about dabbling with mystical powers come from
auspicious objects and characters. Keyholes are passages, and aban
doned villages are cemeteries inhabited by transitory ancestor-type
figures. Streams, especially those that run the wrong way, are mysti
cal paths, and palm wine is the embodiment of communion with
invisible agents, and palmwine tappers their mediators. These warn
ings are a prelude to the true mediation that will follow; they, there
fore, suggest the first "true speech" the hero can trust.
The traps come in the form of encounters with conventional human
beings making perfectly reasonable requests of the hero. However, in
each case the hero takes the instruction wrongly—in fact, literally—
thereby ignoring the more important moral order. So, Moni-Mambu,
asked to "eat groundnuts with the child," eats groundnuts and the
child (11.5). Asked to shoot the monkey robbing palmnuts, he shoots
the headman instead (11.8). Asked to guard the banana tree in the
village, he shoots the chief s favorite wife(11.16).In each case he has
heard only the shallow meaning of the instruction, and in carrying out
the instruction literally he has violated basic moral precepts which
also happen to be Lemba-rzlatzA. precepts: the respect for offspring,
strong moral authority that controls collection of palm tree products
(recall the palm symbolism in the powers of the outside), and order
within the village under the tutelage of a first (Lemba) wife.
The offers of assistance increase as he moves along, and in due
course he has assembled in his satchel a host of helpers, the bilongo
ingredients, for his n'kisi bag. These helpers are, of course, those
symbols of mediation that assist him in gaining recognition of God the
father. The woodborer, the first to be placed in the satchel and next to
the last to be used, relates the vertical, untouched tree to the fallen,
horizontal tree, used in domestic arts and crafts. The cripple, found in
an abandoned village, shows Moni-Mambu the "path" to God his
father, just as later, when he is drawn from the satchel, he shows the
hero his true father. The spider's web from "this shore to that," from
earth to heaven, links the visible and invisible worlds. A common
symbol of mediation, the spider or the spider* s web may be repre-
sented in the "flower/wheel" motif discussed above (see figure 20;

plates 4, 8,10,12). Wasp's sting mediates raw (black) palmnuts and
ripe (red) palmnuts. Hot wind of the dry season allows the fields and
savannas to be burned, preparing the way for the fertile growth of
crops to follow. The whirlwind brings a bundle of forest vines out and
makes them accessible for human cultural order. Finally, kumbi rat,
predictably, opens up the closed and choked, permitting the flow of
food, patrifilial recognition, and life itself.
The ingredients of the satchel thus mediate a series of oppositions
or rhythms ranging from the seasons, to the processes of human
culture, to the intricacies of human interaction, especially patrifili-
ality and husband/wife alliance. It is also apparent that the processes
of differentiation and mediation are being worked out. In society one
must learn to differentiate between people and animals, between
major and secondary wives, and so on. As well, there must be effective
mediation of differences, as the satchel's contents point out In the
case of Moni-Mambu the ambiguity of social life—its traps—is over
come by learning to differentiate and then to mediate. In the next text,
failure to learn life's underlying true categories is illustrated by the
same trickster. Although Moni-Mambu discovers the rules of life and
interaction, he uses them to destroy instead of to construct.
Text 12
(1) Eleven women went fishing and paired off two by two, leaving
one out She was joined by a stranger (antebo spirit), with
whom she fished ten pools. When they had finished, they began
dividing the fish under a palm tree, atop of which was a man.
Every time the fish were counted, they had a different number.
The stranger called fellow spirits to witness the count, and
eventually many spirits had congregated. These spirits became
very hostile to the woman, and would have killed her, but the
palmwine tapper atop the tree dropped his calabash on the
chief spirif s head and frightened them away. He descended,
married the woman, and one of their many children was Moni-
Mambu.
(2) On his way home from market one day Moni-Mambu came
through a village of two brothers, sons of the same mother, who
had never quarrelled. Moni-Mambu bathed in the river nearby
and saw the nets of younger brother, a fisherman. At night he
268 RITUALS
exchanged the nets with the calabashes of the older brother on

a palm tree. In the morning each blamed the other, cursing,
"your mother!" Moni-Mambu mocked them and left.
(3) Moni-Mambu was offered "some peanuts with the children"
by a woman harvesting a field. He made a fire, and roasted her
children until their heads burst. Accused before the court, he
was vindicated because the woman had told him to "eat
peanuts with the children."
(4) Moni-Mambu arrived in a new village, saw no one, but heard
pestles pounding mortars. The women at the outskirts of the
village called him over to "pound them their mortars." He
accepted, went and struck two dead with the pestle. A third
escaped, seeking help. Moni-Mambu, bound and brought
before the court, was accused but vindicated because he
obeyed the women's instructions.
(5) A chief decided to hunt a savanna using fire. After clearing the
air of witchcraft and activating his medicines, the men slept
Next morning arms and powder were distributed. Moni-Mambu
was given a gun and told to shoot" anything that moves." When
the hunt began he followed his instructions, killing snakes,
lizards, one small girl, four hunting dogs, fifteen hunters, and the
chiefs wife. The chief threatened to kill him on the spot, but
Moni-Mambu begged to be drowned in a river, not killed in a
village. On route to the river he escaped.
(6) During a rainstorm Moni-Mambu found shelter in a cavity of a
large tree, and discovered it was a leopard's den. The leopard
arrived with a goat for her young to eat, whereupon Moni-
Mambu attached a wooden hunting bell to the animal's tail.
Believing herself hunted, the leopard fled and died of fright.
Moni-Mambu killed her cub and took the goat for himself.
(7) The chief arranged for Moni-Mambu, who had little motivation,
to be married. After having been married a while, his wife
complained that she needed more than just manioc to feed him
and sent him hunting. He shot an antelope which, as he grabbed
it, fled leaving its skin in his hands. He pursued it to a river, but it
escaped upstream while he went downstream. After a time he
reached the village of his parents-in-law, who gave him to eat.
His nose was dripping into the food, so he took it off and laid it
down on a leaf, assuring it that he would pick it up later. A dog
came along and stole it He threw a firebrand at the dog, which
hit his in-laws' house, burning it to the ground. Going to the

riverside to cut logs with which to build a new house, Moni-
Mambu lost the axe in the water on the second tree. Diving into
the water to fetch the axe, he was seized by a crocodile. Hawk
(Na Yimbi), curious as to the identity of Crocodile's victim,
asked to see it. Thrice Crocodile opened its mouth so the bird
could see Moni-Mambu's face, and the third time the bird seized
Moni-Mambu thus rescuing him.
(8) After Moni-Mambu left the village of his parents-in-law, he met
a young girl bathing in a side channel of a river, repeatedly
dipping her head into the water. She told Moni-Mambu that
bathing like this caused her no shame, even if he were her father-
in-law. Hearing this, he cut off her head and put it in his bag.
Presently he encountered a second girl eating some viands with
pepper. He asked her to share them, but she retorted that she had
none left. So he asked her to go fetch more. While she was away
he ate the meat in the bowl and put in its stead the head of the first
girl. When the second girl returned and saw the head in her bowl,
she screamed in horror. Men came running with their weapons,
Moni-Mambu told them how the ancients had eaten human
heads. They believed him, and killed the girl.
(9) Moni-Mambu, walking through an abandoned village, saw a
human head drying on a pole and asked it how it had died. Its
reply:" sickness and what not all else." When he came home and
related the men in the men's hut this, they called him a liar. He
challenged them to examine the head for themselves. If his story
were false, they could kill him. They went, but when they asked
the head the cause of its death, it remained silent. Returning,
they threatened to kill Moni-Mambu immediately. The chief
calmed them and convened a court. Moni-Mambu was charged
as guilty and condemned to death. When he asked for water "to
calm his heart" the chief refused, saying he had no heart; he was
ordered speared to death. The dogs licked his blood and his
corpse was thrown out on the savanna where witches are
19
thrown. The people rejoiced the death of Moni-Mambu.
The present text differs quite remarkably from the foregoing one in
the character of the mediations enacted. Tsimona-Mambu bringer of
Lemba medicine initially committed a number of errors, but then
went on to become an effective mediator in not only cosmological but
also human oppositions such as his relation to his father, his relation
270 RITUALS
to his wives, and theirs to each other. In the present text, Moni-
Mambu, the cynical farceur, commits errors, is initially forgiven, but
ultimately judged guilty of having willfully designed situations of mis
understanding, conflict, and violence, and he is condemned to the
death of a witch.
In both texts generous use is made of the incongruous, the ambigu
ous, and the outright contradictory to stage episodes in which the hero
may be cast as either an effective or ineffective mediator. In both texts
episodes are related of the trickster's misunderstanding of an instruc
tion: he understands a phrase literally without understanding its
underlying intention. He is invited to "eat groundnuts [alternately
potatoes, 11.5] with the children" (12.3), and he eats the children too.
He is invited to join women working with mortars and pestles, to
"strike them their mortars" (12.4), and he kills two women. He is
invited to watch for a monkey in a palm tree, and he shoots the chief
(11.8). He is invited to stand guard of a banana tree, and he shoots the
chiefs wife (11.16). He is invited to join a hunt and to "shoot every
thing that moves" (12.5), whereupon he shoots dogs, hunters, chil
dren, and so on. These passages of "rhetorical ambiguity" obviously
make for excellent and anticipated storytelling devices to bring an audi
ence into a responsive mood. They belong to the genre of the African
20
dilemma tale represented across the entire continent Their incor
poration in the trickster tale has the effect, however, of probing under
the fabric of harmonious daily life to the difficulties of social dis
course.
At the level of actions, as differentiated from words, the trickster
figure commits deeds that are in sharp violation of the moral code of
the society. He kills esteemed figures such as chiefs, first wives,
headmen. But he also hunts the hunters, kills innocent children, con
fuses species when it comes to eating meats, approaches girls and
women while they are bathing, and so on. Moni-Mambu thus not only
ignores the moral codes of society, he exposes their vulnerable points.
Every imaginable kind of scandal erupts in connection with the trick
ster of Text 12: brothers who had never quarreled are led to curse each
other; his parents-in-law's house is burned down; he falsely induces a
charge of cannibalism upon a poor girl, causing her death; in another
episode not related in Text 12 (but see note 19), he causes an inter-
village feud, in which an entire village is burned to the ground.
Resolution of rhetorical and structural ambiguity in these texts may
follow one of several paths. The dilemma may be left to follow its own
course, with wildly unpredictable consequences—for example, the
hunter being hunted, seized by an animal (12.7). This is good for

laughs, and a talented storyteller can go on and on with this type of
narrative. At a more serious level, ambiguity and contradiction may
be resolved in a number of ways. Structural interpretations may be
offered of solutions to the difficulty. Thus, the ambiguity of status-
equal brothers or wives will be resolved by establishing a priority of
one over the other; failure to meet one's father will be resolved by
reconciliation with the father, and, as in some Mahungu texts,
"endogenous" social structures that are conducive to incest, matri-
lineage isolation, and warfare are transformed into "exogenous"
structures through appropriate alliances, emphases on trade, and
brotherhoods such as Lemba. As is clear, these conjunctive myths
are favorites for use in cult-related circumstances. Disjunctive or
"tragic" myths such as Mavungu with python-woman-in-tree (Text
8) and Moni-Mambu trickster-turned-witch (Text 12) suggest a third
type of resolution, already mentioned in the conclusion of the previ
ous chapter.
Such disjunctive myths hinge on the hero's conscious decision, at
some point midway through the narrative, to embark on a particular
course of action regardless of the consequences. In Mavungu and the
python-woman-in-tree the hero follows the woman and riches know
ing that this will negatively condition his future relations with his
father. In Moni-Mambu trickster-turned-witch the hero is held
responsible for having consciously plotted to kill, deceive, and
destroy. In other words, these narratives are studies of the implica
tions and consequences of consciously-made decisions in the context
of alternative choices. They focus on those decisions which lead to the
hero's tragic death, and by implication the alternative choices which
might lead to life.
Studies of trickster cycles have depicted their heroes as moving in
the development of the cycle from a primordial or aboriginal phase,
lacking conscience, morality, and intelligence, to a stage of full
conscience, morality, intelligence, and civility. Radin's Winnebago
trickster very clearly portrays these developmental features, and
much of the hero's activity is cast midway between these poles. He
can assemble or disassemble culture, integrate and disintegrate,
21
transcend or particularize, depending on the intent of the narrator.
Makarius' writing on the relationship of mythological trickster figures
to clowns emphasizes the characteristic of "inverseness" as espe
cially germane to tricksters. Even more evident than their represen
tation of nonconscious, primordial, puppet-like movement in society's
272 RITUALS
conventions is their contrariness, their violation of rules and pro

hibitions. The purpose of this, she argues, is to call attention to the
importance of the very rules being violated. The trickster is usually
alone in his behavior, he is an outcast. By committing the forbidden—
incest, cannibalism, wrong speech—he is taking on an expiatory role.
"It is necessary that he should be conceived as 'the other,' in opposi
22
tion to the group, even though he acts on their behalf."
Certainly Kongo mythic heroes accessible to Lemba practitioners
utilized both traits of the trickster Moni-Mambu: his development
toward consciousness and effective mediation with which they identi
fied themselves, and the negativeness of wrong choices in the anti-
trickster who meets death. Other characteristics that have elsewhere
been given the trickster, such as the embodiment of opposites (male-
female, elder-junior, father-son) are in north-Kongo myths given to
Mahungu, who, like the trickster, expresses both conjunctive as well
as disjunctive narrative possibilities. The implications of these two
modes of resolution will be explored in Part III of the book after
examining Lemba in the N e w World.
Chapter 8
Lemba in the New World
"It is undeniable that the survivals outnumber the variations."

—R. Bastide, African Religions of Brazil (1978)
"The term petro is not used in the north and northwest of Haiti where
these spirits go under the n^mt Lemba."—A. Métraux, Voodoo in
Haiti (1959)
Introduction
It has been abundantly established that the massive Atlantic slave
trade over several centuries transplanted many features of north-bank
Kongo society into Brazil, Haiti, Jamaica, Cuba, the United States,
and elsewhere in the New World. Priorities given to the study of
Lemba in its African setting permit only a cursory review of the
general historical and cultural framework of this cultural migration.
This chapter's chief aim is to present and evaluate evidence at hand
on Lemba and related ritual features in the New World.
Scholarly awareness of Lemba? § presence in the New World has
been scanty. A few examples suffice to show the sketchy or marginal
status attributed to it and to related Kongo or Bantu elements.
1
Brazilian historian Arthur Ramos published in 1934 a number of
songs to Bantu deities, including this:
Lemba, O Lemba.
Lembá de canabura
Zambiapongo no
coporolá
Lembá Lembá de lei ö Lemba of the law, law
salei Senör Lemba priest
Sinhö Lembá enganga Already we have dug the
Já furamo sé sé hole
Lembá engangajáfumo Lemba's priest has
Carole. already smoked.
273
274 RITUALS
The song's meaning is enigmatic without any contextual explan

ation. In any event, Ramos used it to illustrate the presence of Zambi-
a-Pongo, the Bantu high God, rather than Lemba. Drawn from a
Candomblé in Bahia, Brazil, this song and others including those to
Zambi, Kalunga, Kalundu, and other Bantu cult or deity names
illustrated the persistence of elements, however poorly understood, of
2
Bantu religion.
Herskovits, who worked in West Africa and in the New World,
spoke of the dearth of understanding of Congo, or Bantu, influence in
the N e w World.
Despite the multitude of designations for the great numbers of
gods that must have been worshiped by the various tribes from
which came the slave population, few deities except those from
the central region have present-day devotees on this side of the
Atlantic. Zambi, Simbi, Bumba, Lemba, who are worshiped in
the Congo, are exceptions to this rule . . . yet we know enough
about the gods of peoples outside this "core" to be struck by the
paucity of correspondences to them found in the New World,
especially when this is compared to the wealth of carry-overs of
3
Ashanti, Dahomean, and Yoruban supernatural beings.
Herskovits acknowledged that further research might reveal survivals
or connections between elements hitherto unrecognized in a Bantu, or
Congo, complex comparable to those from West Africa. Individual
features of this Congo complex in New-World African religion are at
times so little recognized, however, that they are lumped in with West-
African materials. Thus Bastide, in his otherwise remarkable work,
4
places Lemba, in passing, in the Dahomean pantheon, not recog
nizing it as the same entity listed, misspelled, in another reference to
work by Brazilian colleague C arniero on B antu influences in Brazilian
5
religion. .
Despite these occasional errors, Bantu and Congo influence is
widely acknowledged in the New-World African religion and culture, as
a category comparable to influences from Yoruba (Nagó), Dahomey
(Gégé), and Ashanti. In Brazilian scholarship Yoruban and
Dahomean influences are commonly distinguished from Congo and
Angola influences. In Haiti, Dahomean and Congo (or Guinea) are
recognized as distinct influences. Commonly West-African influ
ences are seen by scholars as dominant—Yoruba in Brazil, Dahomey
in Haiti—whereas the Congo or Bantu influence is seen as having
adapted to, or subordinated itself within, the West-African system of
Lemba in the New World 275
ritual symbols and deities. However there is reason to suspect the

finality of such scholarly judgment, since Congo and Bantu influences
have not until recently been the subject of thorough scholarly inquiry
to match the extensive fieldwork of a Herskovits in the Old and the
6
New World on Dahomey, or the in-depth reconstruction of a single
tradition in the New World comparable to Bastide's work on Yoruba
7
(Nago) in Brazil. Thompson's current work comparing art and dance
styles in Kongo with those of the Kongo-influenced portions of the New
8 9
World, replicating similar work on the Yoruba, is an important step
toward understanding the Atlantic continuity of Kongo artistic cul
ture, including Lemba.
In Brazil scholars such as Carniero and Bastide have, however,
identified correspondences between Bantu and West-African ritual
systems in order to establish the path by which a dominant symbolic
system might absorb, or supplant, a subordinate one. If, as Bastide
suggests, the Nago (Yoruba) rites exercised a sort of hegemony over
those from other African "nationalities," and the Catholic deities and
saints over all African ritual symbols, then it is possible to identify
distinctive Bantu and Congo elements in their appropriate places. In
other words, assimilation has not been haphazard, but has followed
rules of correspondence (figure 23). These correspondences between
Congo (Kongo) and Angola inkisses, Dahomean (Gege) voduns,
Yoruba (Nago) orixds, and Catholic saints and deities, are drawn
largely from court registers of African cults in Brazilian cities of
Recife, Pernambuco, and Bahia. They suggest that, as Bastide says,
Bantu deities retain their existence in ecstatic "language" songs in
which they are named Zambi, Bumba, Lemba, and so on, while the
rites are basically the same across the nationalities. Even so, there has
been a tendency for inkisses and voduns to be submerged in orixds,
and these in Catholic deities and saints. Cultic movements such as
Candomble, Macumba, and most recently Umbanda which cut
across these nationalities have a tendency to rearrange and submerge
autonomous elements even further.
The list of Congo deities or inkisses in figure 23 bears a certain
resemblance to comparable inventories of inkisses from the seven
teenth- and eighteenth-century Congo, in particular that given by
Dapper (see Chapter 2), for the Loango coast. Zambi is not
mentioned by Dapper but is widespread in other early references.
However, correspondences between Pongo and Zambiapongo,
Malemba or Lemba and Lomba (sp?), Bomba or Bombo and
Bombongira, Kossi and Incossi, Makongo and Quincongo, and
О
о аз
E
CL
о
iо5E
>>
О
Ug"О 0+¿0со Dû Л
( со v.
-о
ПhE-
С ч°- «C
п
SZ
cd S? E
er
О х: _j "5яз о
"•i E"го Ö
Ela Li сга < E с га •- С «D ( -£-¡ со
Л °3
О £ 40
о
00саЗ E "О ocdocu
СЛ (От о L
U 1 j a l e
00• - Л* So .¿о.сиr -ососо rare сосо га ^ ч-ö £ _EX
со о со
<ü о ^ zi"а.
< оо _,j j о
—
p
--оо <
-
û о о
—~
Q
- •~
о.
оо с
.
-g оо о 00I^1OQ.
со со
KB o
S 5' 3 <¿
— 0) со со со•Е IS Hсо 2 " o 1
с:ico Q
о ш с со
4Ö с>и сcеu . со < 0?О о О

Hol
CN со
On
Oч,
О
T<D
3
СО РE -сс
.>-. .<о P
co ^=_l
P o
PG
Q ссо E
О E
О „со
-оо оо
со
со
з 2
E .s
-a
C
O
§
со
•g 8
"5
чоз
1
Ucd
-а осиз <
«.cо" Xco>
с -
¿
< .с — со <л .2 .qcEo 3rä.fS
сJ5
-E - -Ûфco "О co
< Mcgo -g £ £
F ccuo —S— •О
i° I
fsl
cu
g
cu
с о S «E
8,
co ^ -5 -О -O •2 .E E
E E S S
ё
о
U
possibly Pansa and Panda suggest on the basis of name similarity

alone an important source of continuity from early Congo and later
10
Brazilian ritual culture. As important as these similarities are, the
names of spirits or medicines (min'kisi) in the Dapper list include
several which do not appear in Brazil, nor anywhere else in the New
World. Of these we know that Thirico, Kikokoo, Injami, and
Bunzi were local or regional earth shrines, and as a set they do not
appear to have been reconstituted in the New World. Further evi
dence may be noted of the continuity of a Congo system. Zambi and
Olorun, the Nago high god, do not become transformed into a
particular Catholic deity or saint, and generally Zambi is not made
the object of a particular cult, nor does it become a possessing deity
(true in Haiti as well). This suggests a transfer of the Bantu idea of a
distant or ungraspable ultimate power. The Christian God the Father
is rather equated with Oxala, Olissassa, and Lemba, and He appears
in the context of a dyadic relationship with his son Jesus, or in the
Trinity, or in the Holy Family of Joseph, the Virgin Mary, and the
Christ child. This equation of Lemba with Oxala [Orisanla] the Nago
(Yoruba) god of the sky and deity of procreation and creativity, and
with the cluster of Christian features of the Trinity, the Holy Family,
and the God(Father)-Christ (Son) relationship—all features charac
teristic of Lemba in preceding Kongo variants—indicates another
important conceptual carry-over to New-World Bantu religion.
In Haiti the African cultural and religious influences were, as
already noted, similarly divided into West-African and Bantu cycles
of worship. Both in ritual observance and in scholarship the dominant
West-African influence is acknowledged to be that of Dahomey, as
witnessed by the cultic vocabulary of spirits (loa), cult complexes
(youdou), cult leadership (houngan, mambo), cult locations (hounsi),
and so on, although even here there is the possibility of Kongo influ
ence (for example, mambo from Moni-Mambu, or mambuk, healer).
Dahomean deities are collectively known as Rada (from the slaving
port town Arada, itself named after Allada), whereas those of Congo
and Bantu origins have recently been identified as strongly reappear
11
ing in the Petro cycle of deities. This dichotomy needs to be seen, in
Haiti, against the backdrop of Haitian history.
Haitian history is a running record of impulses for independence on
the part of black immigrants to the island, who by 1600 had inherited
12
an agricultural economy from the by then extinct Taino Indians.
Spain gradually ceded its dominance of the island to the French, who
by the eighteenth century had developed a plantation economy
278 RITUALS
around sugar, which required massive regular injections of slave

labor. During this century 15,000 slaves a year were being shipped
from the ports of Cabinda, Malemba, and Loango. Slaving ships
would run a triangular route: to Africa for slaves, then to Santo
Domingo where these were traded for sugar which was shipped to
Bordeaux and Nantes. In the late eighteenth century this society based
upon slaves, sugar and slave-owners was hit with undercurrents of the
French Revolution, which did not leave the blacks untouched. The
oppression of the slave plantations combined with exclusion from any
participation in the emancipation offered by revolutionary ideas
touched off a massive slave revolt in 1791 that turned into a decade-
long war of liberation. Legendary history recounts how one Boukman,
a giant of a man and a religious priest, led the partisans in the taking of
an oath near Mourne Rouge, and how an old Negress sacrificed a
black pig, catching the blood in a gourd and passing it around for each
to drink as they swore loyalty to Boukman. The plantations were set to
13
the torch and the oppressors killed. Warfare continued until all
Europeans had been driven out, and in 1804 Haiti was pronounced an
independent republic. Toussaint L'Ouverture became its founder,
and Dessalines, Christophe, and Petion, his lieutenants in the revolt,
became his ministers, and his successors.
Although Haiti was now independent, its early governments were
hardly less repressive than had been the colonial predecessor.
Dessalines created a type of serfdom and was assassinated.
Christophe followed with similar laws to the effect that cultivators had
to stay rooted to the soil—the plantations—where they worked, while
a titled aristocracy was allowed to emerge. Petion, however, instituted
some liberalization, such as individual peasant holdings. He is also
known for the construction of the Citadel, a fortress of Old-World
castle proportions, as a beleaguered free black state's last defense
from its potential enemies. It was never used. These founding fathers
remain legendary heroes in Haitian public and religious rituals,
including as will be seen Lemba-Petro.
This link of early Haitian history to Lemba is tenuous, but what
historical details are known are very suggestive, principally put in
perspective through Jean Price-Mars' reading of the work of Moreau
14
de Saint-Mery, a late eighteenth-century French writer. Saint-
Mery collected his observations in the late colonial era and published
in 1797, before Haitian independence. Already then, at the beginning
of the rebellion, the distinction was apparent between Voodoo and
Petro (or Pedro) complexes in African-Haitian dance. The orgin of
the "Don Pedro" dance, accordingto Saint-Mery, went back to 1768

when a Spanish black from Petit-Goave, by the name of Don Pedro,
introduced a dance analogous to the Voodoo rites but of a far greater
rapidity and level of excitation. To reach even greater levels of
excitement and pitch, devotees of "Don Pedro" spiked their rum with
gunpowder. The effect, in Price-Mars' summary of Saint-Mery's
observations, was "electrifying," producing a "convulsive drunken
ness" which, heightened by the singing and the rhythm of the drums,
15
evoked a collective crisis. Writing again in 1803, explicitly about
Haitian dance, Saint-Mery reiterated the stylistic contrast of "Don
Pedro" to "Voodoo," noting that the former frequently touched off
public disorder. This and the banning of the dance by authorities,
however ineffective in their efforts, suggest that the Don Pedro dance
may have had an important role in sustaining resistance to French
16
authority.
The grounds for equating this historic Don Pedro dance with
today's Petro complex of rites, and by inference the Congo influence
in Haitian religion and culture, are several, according to Price-Mars.
First, there is its contrast to Voodoo in Saint-Mery's eighteenth- and
nineteenth-century account, and there is the linguistic transformation
from the Spanish Don Pedro to the Haitian-French Creole "Jean
Petro." Second, Price-Mars notes the concentration of the Petro
complex of dances in the west and south of Haiti, corresponding to
Saint-Mery's reference to the Don Pedro dance in the west of Saint
Dominique. And a third indication of this dance's continuity is the
recurrent harassment it has received from government, both colonial
and independent. Indeed, Dessalines, when he was Toussaint
L'Ouverture's Inspector of Culture, is known to have had fifty wor
shippers of a religious dance bayonetted in the Plain of Cul-de-Sac,
17
the general region where Don Pedro was concentrated.
The Petro complex of deities (loa) in Haitian religion, acknowl
edged to be a reflection of the Congo-Guinea influence, is held apart
from the unique Lemba-Petro rite witnessed by Price-Mars in the
Cul-de-Sac valley in the 1930's. Of all Haitian rites Lemba-Petro is
the only one, in Price-Mars' words, which is celebrated secretly, deep
in the forest, rather than in the village or town near the hounfor
Voodoo temple. Price-Mars goes so far as to state that Petro and
Lemba are synonymous, and that the ancient Don Pedro was prob
18
ably some version of the African rite of Lemba. However, this is not
very illuminating, since his only source on Lemba in Africa is
Chatelain's brief reference to a household fertility medicine in
280 Rituals
19
Angola. Métraux, who calls Lemba an African "tribe," suggests
that it takes the place of Pétro in the north and northwest of Haiti as a
category of divinities and rites equal to and identical with the Pétro
20
complex. Thompson is the first to identify Lemba's rightful source
in connection with Haitian religion as the north-Kongo healing cult by
21
that same name.
Before describing the autonomous Lemba-Pétro rite of Haiti, it is
appropriate to describe the more commonly known Pétro segment
of loa gods and services in well-known Voodoo settings.
The Pétro Complex of Deities in a Voodoo Lao Service
In the valley adjoining the Cul-de-Sac where the Lemba-Pétro rite

was reported, near the town of Mirebalais, Herskovits in the 1930's
carefully documented a loa service he belived to be typical for the
22
region. This section will identify features in this rite that are impor
tant in the comparison of Voodoo with Congo, especially Lemba,
ritual structure and intent.
Resemblance of the loa service to Lemba, in strictly choreo
graphic—spatial and temporal—ordering, is striking. The Haitian rite
is constructed around two major phases, separated by an interval. The
first such day/night/day phase is devoted to the Pétro rite, the second
to the Rada rite and its deities (see figure 24).
The family putting on the service had experienced numerous deaths
and illnesses and, as customary, it had to celebrate once in a decade or
so a major ceremony for the loa spirits. The first phase was observed
around three altars: the first, a table called the "trône," which is
covered with white cloth and a canopy, holds ,a variety of plant
offerings, crucifixes, holy water, candles, oil, and other food offerings,
as well as chromolithographie images of saints; the second, a table
devoted to the loa named Simbi, bears various cooked and uncooked
foods; the third, and most important, is a table devoted to the loa
Bosu, bearing food and drink, nearby which are tied offerings of
sacrificial animals including a large black boar, two small black pigs, a
black goat, a black turkey, a guinea hen, two white pigeons, and
several chickens.
In the early evening the family putting on the service, and the
officiants, held the "action de grâce," a Catholic-like ritual led by the
prêt savanne. Holding a candle in his hand, the leader directed the
family in "desultory" singing from time to time pouring water from a
service in
shrine
procession
the meal
sacrif ices to Rada loa S?

ON
possession by Rada loa
describing vévés to Rada loa >

o
o 00
C3
pause ,g
Tf q
^^
« .ti
a wo
3
the meal j 2
CO
Id
sacrifices to Petro /oa^ Pu
possession by Petro loa §

describing veves (paths of o
gods) on ground g
H
U CO
CO v .
= no
282 RITUALS
container on the trône into a small depression directly before the altar.
All knelt for the Pater Noster, the Credo, the Ave Maria, and for a
long adoration and benediction.
The European music of the action de grâce gave way then to
African rhythm and drumming, at which time the hungan took over
the officiating from the prêt' savanne. The hungan ordered each
family member to tell the loa spirits which troubles had befallen them
and what favors were desired. His assistant then took six candles and
proceeded, in a counterclockwise direction, to trace in the meadow
near the village a large circle, fixing four candles at cardinal points
representing the four corners of the earth. In addition, one was for
the Maif 'Bitation (Master of the Habitation) and another for Maif
Source (Master of the Stream). The family members, holding can
dles, placed their individual candles near those already emplanted in
the meadow, forming several clusters in the overall shape of a cosmo-
gram, dotting the hillside in the deepening dusk.
After all had returned to the altar, the hungan began to "trace
ververs" on the ground. ( Ververs—usually called vévés—are lines or
paths for the spirits to follow, often in the Port-au-Prince area drawn
in highly decorative patterns with flour or other powders.) Initially he
traced a white line in corn meal from the first table for Bosu to the
second table and tree for Simbi, on to the trône where subsidiary
motifs were added in gray and black. The vévé paths anticipated the
coming of the gods up to the human world where they would be
recognized, fed, placated, and then "restrained" or put away" for
another decade.
At this point in the ritual, sacrificial animals and birds were brought
to their respective altars for washing and presenting to the gods. The
hungan became possessed by Simbi, announced his "fee" for
performing the rite, and began a long invocation to the various Pétro
loa, while the participants waited for the animals to "recognize," that
is begin eating, their appropriate food. Thus, the red and black cock
was offered grains for Legba, the black fowl grain for Gede, the red
fowl grains for Ogun, the white pigeon grains for Damballa, the fowl
with ruffled plumage grains for Congo Zandor, the small pigs grains
for ' Ti Kita Démembré, the boar grain for Bosu, the goat leaves from
the sacred tree for Erzilie Gé Rouge, and so on. The animals, later to
be sacrificed to their appropriate loa, were to take to their food in the
same way as the loa were to possess members of the family and
receive their sacrificial meals, members of the family were "crossed"
with animals by the hungan to encourage the gods to come to them.
This action was repeated before the Bosu altar.
As the singing became louder and more rhythmical, with the drums
leading the pace, the deities finally responded to the calls, "mount
ing," possessing, the family members one by one, each enacting the
characteristic ecstatic behaviors, for example, jumping into thefirefor
Legba. Herskovits notes that the devotees moved back and forth
between the three altars—the Catholic trône, the altar to Bosu, and
that to Simbi—as the loa possessed family members before each
appropriate altar, thus pattern and structure were evident despite the
appearance of utter chaos in the outcries and diverse agitated behav
iors of possession. One of the several songs sung during the possession
service lists and to some extent characterized the deities who
appeared during this Pétro phase.
Bosu found something on the ground,

D o not pick it up.
Bosu goes about,
He found something on the ground,
Do not pick it up.
You will come to harm.
If you pick it up,
You will anger Bosu.
Simbi en Deux Eaux,
I do for you,
Do for me;
Help, O Congo,
Congo Zandor.
Simbi Kita Kita, work for us,
Today for us,
Tomorrow for others.
Erzilie, O Lemba,
I ask, where is Erzilie,
The food is ready.
Gedenibo, behind the cross, Gede.
Before Baron, Gede,
Behind the cross, Gede,
Today I am troubled,
Gedebi, call Gede, this Gede,
Today I am troubled.
Cease to sweep, sprinkle, hoe.
23
I am troubled, Baron Samedi.
At about midnight, when all the animals to be offered had been
washed and prepared, and all possession had ceased, the time had
come for the sacrifices. The hungan, with knife in hand, decapitated
or punctured the jugular vein of the appropriate animal for each loa,
while an assistant collected some of the blood in a small calabash.
Thus killed, the offerings were taken elsewhere for dressing and
cooking. The loa which had been troubling the family were then " sent
away," that is, buried symbolically with an offering of meat. Thus
handled were Erzilie Gé Rouge (Red Eyes), Bosu Trois Comes
(Three Horns), Ti Kita Démembré, and 'Ti Jean Pié Sèche. Those
"restrained" were Lemba and Simbi djo (de Veau, of the water),
Lemba under the mango tree where the altar to Bosu had been, Simbi
djo where the trône had stood. Each was buried with its offering and
an iron cross or other object to "preoccupy it" for a period from seven
to seventeen years. The loa sent away were told to go back to Guinea
or Africa and not bother the family. Those restrained were dealt with
more favorably, the family needing their help. The idea was clearly
conveyed that these Pétro deities were responsible for the family's
misfortunes, and that they should stay away now that they had been
placated. The feast was intended for them, at least rhetorically,
although the human community finished that which the gods had left.
Gods' leftovers are human sustenance. This meal persisted into day
light, with final ceremonies.
The Rada phase was held a few days later, and approximated the
same structure, opening with the action de grâce, and continuing with
the tracing of vèvës, possession by Rada loa such as Damballah
Wedo, Legba, Ogun, Ashade, and so on, the sacrifice at midnight, and
the meal.
The issue of to what extent Haitian loa services reflect African
continuities may be discussed in terms of similarities of structure
which reveal underlying patterns or motifs, and similarities of content
which reveal direct terminological or semantic continuities. Although
the dominant Dahomean influence is evident in the vocabulary of
Haitian Voodoo, the role names of officiants, and many liturgical
symbols, the resemblance of the temporal structure with Old-World
Lemba variants suggests that Congo-Bantu continuities must be given
serious consideration. In addition, the tracing of ritual spaces in
the Pétro phase of the Mirebalais loa service offers some striking
resemblances to several Lemba variants and other Kongo rituals (for
example, the diyowa in N'khimba, Munkukusa). In the staking out of
the cardinal points with candles, the hungan used a common, perhaps
world-wide, motif. However, in circling this space in a counterclock
wise direction and then dividing it into two, one half representing the
domestic realm (governed by "lord of the house," Maif 'Habitation),
the other half the realm of the wilds, the deep, of water (governed by
"lord of the deep," Maif Source), he was tracing a cosmogram the
way it is done in many Kongo contexts. The opposition of the human
realm and that of the beyond (whether the world of spirits, wild nature,
or trade routes) serves as the paradigm for ritual action wherever one
must come to terms with the interaction of the visible and experiential
with the invisible and intangible. The cyclic movement turns this
opposition into a statement about time in the life of man and society,
denoting rhythmic cycles such as life and death, coming and going,
and perhaps, as in this loa service, the awakening of the gods and the
return of the same to their appropriate resting places.
Going beyond the general to particular continuities of the Kongo-
Haiti axis is more difficult; however, some interpretations seem war
ranted. The two altars erected to "Pétro" deities, Bosu and Simbi,
suggest this opposition of the domestic and the beyond so common
in Lemba and other Kongo ritual settings. Simbi spirits are well
known in Kongo to indwell water courses, pools, springs, rocks,
forests, ravines, and other passages to the nether world of mystical
power, a characteristic which makes them suited for the elaboration of
public authority symbols. Kita, also present in the Pétro cluster of loa
spirits, is a type of Kongo simbi. Bosu represents a problem in inter
pretation, not the least of which is because of his well-known origin in
the Fongbe (Dahomean) pantheon as a three-horned, hunch-backed
monster. Despite this clear origin, corroborated both by historical
research and inHaitianfolkthought,Bosuis"puf' into the Petro, that
is Lemba and Kongo, class of loa, to represent forces at work in the
24
domestic, human society, realm.
The two deities who are "restrained," namely Lemba and Simbi
djo, appear to replicate this complementarity between the realm of the
wilds and that of the community. Both oppositions span the tensions
inherent in the life of Haitian peasants, on the one hand uniting and yet
on the other dividing the human world and the supernatural, the New
World and ancestral Africa, as well as the older span between the
trade routes, markets, and home. These fundamental tensions, ritu-
ally integrated into one system of rhythms and contrasts, regenerate
the community.
286 RITUALS
Lemba-Pétro in the Cul-de-Sac Valley, Haiti
The Pétro complex, in general, has much in common with Kongo

religion, and its structure is shared with Lemba. To find specific
elements and meanings which correspond to Lemba, however, it is
necessary to look at Lemba-Pétro, the secretive rite which Price-
25
Mars witnessed, in part, in the Cul-de-Sac Valley in the 1930's. He
strayed onto the service one afternoon while on a hunting expedition,
and because he was recognized by the celebrants as their physician he
was invited to stay. The service was already under way, deep in the
protective forests.
At the foot of a tree Price-Mars noted two altars. The first was a
table upon which had been erected a bamboo elevation with a niche in
it In this niche there were painted figures representing the Sacred
Heart of Jesus, the Infant Jesus, the Holy Virgin, St Joseph, and
images of other Catholic saints. This partial list suggests correspon
dences to Lemba similar to those in Brazil. On the table were placed a
bell, a castor oil lamp, a jug, a bottle of kola, and several plates of
candy and cake. Under the table, extended on the ground, was a white
towel, on which had been placed four forged iron crosses, plates of
grilled corn, bread, small cakes, almond sugar, and two bottles of kola.
The second altar was a table covered with red and blue cloth.
Before it had been dug a large, deep hole, in which had been placed
two carved pieces of mahogany wood about a meter long, each
wrapped in the middle with a band of Thai silk, and that wrapped with
crimson cloth.
The officiating hungan took one piece of wood, looked at it for a
time, turned it and held it in his hands, then chanted these words:
Bordé, bordé
coci, manga
va atiloca sorci
macorni, au, vati.
Then he took a hammer and nails and drove the nails into the two
pieces of wood in rhythmic alternation to the points of this chant sung
by the participants:
Kol M'tendé kol
M'apé cloué bois!
Ko! M'tendé Kol
M'apé cloué bois!
Then the hungan took a ball of strong cord and measured out five
meters, twice. Each piece of string was taken by a robust man at each
end and held taut with the two crossing each other. Then, with the help
of assistants, the hungan tied the pieces of mahogany wood, clothed
in crimson, with nails in them, onto the cross-like form of cord. As
26
each knot was tied with great force, he intoned this refrain:
Assure! Assure!
N'ap'assure point la!
Hi! Hi!
Nou prale mare loa We are going to tie down
Petro! the petro loa, Hi! Hi!
Hi! Hi! Jean-Petro, chain which
Jean Petro! Chainne qui is a chain; he has
chainne broken it as if it were a
Li casse li rope.
Qui dirait corde! Hi! Hi!
Hi! Hi!
N'ap mare n'ap mare
Loa Petro
Hi! Hi!
This continued until the cords were entirely knotted around the pieces
of wood. This package together with another carved mahogany cross
and two crosses forged of iron were placed at the bottom of the hole, at
the foot of the altar. They were then buried with a deep layer of earth.
(This ceremony probably represented the restraining of the loa Jean
27
Petro, within the Lemba framework. Price-Mars remains silent as
to its meaning.)
Meanwhile darkness had fallen. The hungan sounded his bell and
the congregation formed two rows and walked a distance westward
along a path until they came to another site where another deep pit had
been dug. The hungan asked for complete silence. Raising his index
finger over the pit, he spoke this prayer:
Earth, holy earth,

We are a few of your children
among the best in these parts
gathered to ask forgiveness
for offenses to you, which
we may have involuntarily committed.
288 RITUALS
Especially in the name of

all the members of the family,
present or absent,
I implore their absolution
from all errors committed toward you.
In their name
I offer you drink and food.
Deign to accept these humble offerings.
We wanted to bring you
lavish presents; however,
our means are less and less adequate.
We are but poor folk
who beg your mercy.
O Earth, our mother and patron!
Jean Pétro, Toussaint-L'Ouverture,
Rigaud, Dessalines, Christophe, Pétion,
You know that we are not of this land.
Our ancestors, to all of us come from Africa,
Be favorable to our requests to you.
Jean Pétro, we have done all to please you.
Save us from the calamities which beset us.
Save us from fever, smallpox, leprosy,
Tuberculosis, automobile accidents.
Accept these libations which we make
In your name. Amen.
And the participants replied, "amen." Then the hungan took five
female animals, a goat, a chicken, a guinea fowl, a turkey hen, and a
dove, which he decapitated one after another and held over the pit so
their blood flowed copiously into it, after which he threw their bodies
into the pit as well. Then, with the help of an assistant, he poured water
into a basin and from that into the pit, asking each member of the
family to repeat this gesture after him. From another container he also
took liquid, and did the same, again asking the family to repeat it after
him. The same action followed with a bowl of diverse grains which
were thrown into the large pit in honor of the earth and the ancestors.
A s the hungan pronounced final incantations—inaudible to Price-
Mars—and prepared to return to the initial ritual site, a man in the
congregation suddenly let out a loud cry and began gesticulating about
with such petulance that the assistants cleared the area around.
Possessed by a Petro god—Price-Mars does not say which—he

cried:
Brisé! Broken!
Je suis brisé! I am broken!
C'est moi brisé! Itisi, broken!
This flood of words, the agitation, rage and convulsiveness, according
to Price-Mars, indicated the Pétro god's recognition, in possessing a
member of the family, of their penitence over violation of its ordi
nances. The hungan failed to calm the man by speaking to him and
finally tied his hands with a cord, whereupon he quieted down some
but continued to tremble and plaintively mutter "Je suis brisé" as the
congregation, again in two rows, filed back to the initial ritual site
following the hungan who was murmuring pronouncements of for
giveness.
Having returned to the site of the two altars, the hungan untied the
man who had been possessed, and proceeded with the next phase of
the rite, the final sacrifice. A sow, the last animal at the site, was
brought to the foot of the altar, and held fast by the assistants. The
hungan, while phrasing a mostly unintelligible (to Price-Mars)
incantation to "criminal Pétro" deftly stuck his knife into the sow's
jugular vein, catching the blood as it throbbed out in a white vase. The
vase of blood was placed on the altar and mixed with another
beverage. Then, continuing to chant, the hungan gathered the family
group of men, women, and children, now dressed in white and wearing
red bracelets, around the altar in a half-circle. Taking the vase, he
dipped two fingers of his right hand into the blood and spread it onto
the lips of each person around him, continuing the chant about "Petro
criminal." Price-Mars observes that during this "blood communion"
numerous participants shivered, momentarily losing their composure
in brief attacks of ecstasy or deep sighs of emotion, witness of their
"collective agony."
With this, the main part of the rite was over, and Price-Mars left the
scene. It was now after midnight. The devotees would continue with a
nocturnal meal.
It is apparent that Price-Mars has not witnessed the entire Lemba-
Pétro ritual. For example, no action is described before the altar
bearing Catholic images. What occurred there may have been com
parable to the action de grace in the loa service recounted earlier, it
may, however, relate more directly to a Lemba perception of familial
roles as seen in correspondences of Bantu divinities to Catholic saints
290 RITUALS
and gods. The symbolism of food for the gods relates it, as well as all of
Haitian Voodoo, to African ritual. The theme is highly developed
in Lemba, for example in the eastern variant, where the priests
receive food offerings from the neophyte, in exchange for medicines.
A s the rite progresses, the return medicine gift contains increasing
doses of food. By the close, Lemba is issuing banquets to the general
public, and the neophyte priestly couple, with senior priests, are
receiving and eating banquets of symbols and sacred "dirt." It is as if
the gods, whose food is medicine, and those who mediate the rela
tionship between gods and men, participate in negotiation: offering
tokens to the gods who ultimately return generosity and well-being to
the human community. To be sure, this feature is not unique to
Lemba, but the imagery of "priming the pump" of well-being, of the
earth's fertility and the ancestors' beneficence, is prominent in
Lemba, both in the Old and the New World. It is perhaps more
prominent in the Haitian Lemba-Pétro rite, subordinated in the
African variants by the commercial motif.
A further theme that is prominent in Kongo religion, which was
noted in the loa service and present in Lemba-Pétro as well, is the
complementarity of powers. The wrapped mahogany sticks and the
cords which are used in "binding" Pétro recall not only the tracing of
the cosmogram of Kongo ritual, but also a specific feature of Lemba
in which similar sticks are brought together to represent the creative
complementarity of male and female. In the south-central variant, as
the neophyte stands before the cruciform diyowa trench preparatory
to the oath-taking, each priest (vua) and priestess (mizita) is repre
sented in such a stick. These are held together by the neophyte as he
swears his oath of loyalty to YnsLemba father and toLemba. Later the
sticks are bound in "couples," as in the Haitian séance, and placed in
the n'kobe. Male and female together represent Mahungu, the andro
gynous demigod of complementarity. It is as if the diyowa cruciform
trench of Kongo ritual combines the idea of a hole in the ground with
the cross-form represented momentarily by the cord and the sticks.
The oath that is sworn, in the south-central variant, by the neophyte
who submerges his face in the ancestral mud of the trench while
holding the sticks is, in the Haitian variant, projected into the binding
up of the two sticks of mahogany.
The language of the Haitian rite, the resounding "ko!" of the nails
being driven into hard wood, makes it clearly Kongo in origin. This
"strike word" is, of course, an abbreviation of koma, a term which
denotes the act of driving nails or wedges into a ritual object, or of
spitting on a bracelet. It also is used to describe the abstract idea of

awakening in a person the mystical force of the spirits. Konda, for
example, uses it to speak of the effect on the Lemba neophyte of a
dream of ancestors, or a nightmare (Text 8.7). The burial of Petro,
thus "bound," before the altar, may be a unique Haitian procedure,
picked up from Voodoo rites. However, it may, in the Lemba-Petro
context, indicate the survival of a form of activating and " recharging"
9
a mystical force that is older than Lemba s adoption of the moveable
shrine-n'kobe, most certainly related to long-distance commerce.
Lemba's continuity is perfectly clear in the final ritual episode
recounted by Price-Mars, in which the sow is sacrificed and the brace
let-bearing congregants gather around in a half-circle for the final
"communion." It is a greatly simplified rendering of the elaborate rite
of the northern (Kamba) variant in which the Lemba priest receives
many pigs for the instruction of the neophyte's profanation (1.35-5 9).
The pigs' blood is collected in an nsaba pot and taken behind the
house to the ceremonial "kitchen," where the Lemba neophyte
couple is given its code of behavior and blessed by the Lemba father
as the bracelets are issued. The blood is spread on the doorposts of
their house.
How can one speak of continuity, though, when there is, in the
Haitian rite, no village with courtyard and hearth, no house with door
posts? Lemba-Petro has totally adapted the ritual rhythm of the
African rite—the movement from courtyard to hearth,fromthe village
to the beyond—to the hidden forest setting, befitting its status since
the eighteenth century as a persecuted sect Amazingly, however, the
rhythm of spatial movement has been retained; thus, the site of the two
altars is the village, by analogy. The two altars may serve the court
yard/hearth opposition, although details are missing. The sacrifice of
birds before the large earth pit, the site of possession by Petro, repli
cates the esoteric phase of the African variants. What is important is
that the significance of the ritual rhythm has survived, over several
centuries in two traditions diverging from a common point, in mutu
ally intelligible ritual gestures.
Retentions and continuity alone do not assure vitality. Perhaps the
opposite. Thus some variations are as indicative of Lemba" s survival
as are the continuities. In the African variations Lemba adherents
invoked a variety of gods, appropriate according to regional myth
ological systems and structures of authority. In the east, a Teke-
related chief spirit, Nga Malamu, was addressed; elsewhere it was
Kuba, Mahungu, or Moni-Mambu; on the coast and some inland
292 RITUALS
regions it was earth god Bunzi. What could be more fitting, in Haiti,
than to address Jean Pétro, the "criminal" of the dominant system,
and to build on that beginning with Haiti's founding heroes Toussaint
L'Ouverture, Rigaud, Desslaines, Christophe, and Pétion.
Part III
Structure and Movement

in Lemba Therapeutics
"Myth shields us from music while at the same time giving music its
maximum freedom. In exchange, music endows the tragic myth with a
convincing metaphysical significance, which the unsupported word
and image could never achieve, and, moreover, assures the spectator
of a supreme delight—though the way passes through annihilation and
negation, so that he is made to feel that the very womb of things speaks
audibly to him."—Friedrich Nietzsche, The Birth of Tragedy
293
Introduction
Lemba sources studied so far reveal a strong preoccupation with

conscious thought and deliberative action. The Lemba adherent is
admonished to keep secrets, be vigilant, argue effectively, prevail in
public debate, trade skillfully, and maintain a strong household and a
peaceful community. Legendary figures and ritual symbols reflect
these virtues; antiheroes define them negatively by demonstrating
the sad consequences of erroneous choices.
The general content of these ideas ofLemba well-being mdLemba
illness may now be spelled out more fully. We have analyzed the
expressive domains to understand the Lemba ritual setting, and we
have clarified the codes of these domains to determine the logic of the
expressions. Now we can attempt the difficult task of examining
Lemba's self-perception. In other words, is it possible to identify the
perspectives of Lemba priests and priestesses themselves?
It would seem possible to reconstruct such aLemba consciousness,
given the emphasis on clarity of thought and skillful rhetoric. How
ever, there is a problem. Paradoxically, Lemba''s origin and ritual is
often elaborated in terms of apparently mystifying creations such as
rodents, earth colors, plants, and mythic heroes. Yet these seem to be
interwoven with allusions to the familiar ground of daily experiences
such as family life, collecting food, and relations between clans. There
is thus an obvious tendency to align mundane experience with more
abstract principles from myth, cosmology, and social structure. The
manner in which the two levels of reality are combined reveals a selec
tivity of choice of alternatives, a conspicuous manipulation of bits and
pieces from the received lore of myth and cosmology to reinforce a
particular set of alternatives. The moment this occurs in an institu
5
tion s or society's culture, it has created an ideology for itself. Why
this occurs, the context in which it occurs, and why such an ideology
must take recourse to a multitude of images from the natural world will
be examined in this section.
Chapter 9 identifies the mechanisms of thought which lie behind the
construction of Lemba's unique world view, the process by which it
anchors social forms with specific ideals, selecting particular types of
authority, defining moral order through specific heroes and anti-
heroes. And it goes on to explain how Lemba elaborates conjunctive
295
296 STRUCTURE
and disjunctive modes of mediation within relationships and situa

tions, and shifts disjunctive mediation from the verbal realm of dis
course to the nonverbal, a process which is the key to understanding
Lemba's "illness" and "therapy."
Chapter 10 brings this structural preference of choices—Lemba's
ideology—together into a theory of order and meaningful existence in
a segmentary society caught up in the great trade of coastal Africa
from the seventeenth century to the twentieth.
Chapter 9
Modes of Lemba Thought
Anchoring in Myth, and the Problem of History
Each Lemba variant has reflected a concern for relating its main con
cepts, medicines, and ritual symbols to a legitimate source, be it
culture hero Moni-Mambu, ancestors,firstpriest and "founder" Nga-
Malamu, or someone's Lemba Father. Kerenyi and Jung call this
1
feature an etiological myth's "anchorage."
Kongo min 'kisi reflect an anchoring structure in their charters. The
n 'kisi is revealed to an individual or community, and through its priest
it is handed on, extended(vandisa), or received (tumiswa). Ann'kisi
"apostle" is, therefore, a sent one {ritumwa). Chiefly commissions
function in like manner, except that they are exclusive corporations
whose officeholders succeed one another, and modern Kongo pro
phets tend to be consecrated by a comparable structure. All conse
crated roles and functions in the society, of whatever generality or
2
specificity, share a common anchoring structure.
The structure of a mythic charter, whether of an n'kisi, a ruling
dynasty, a prophetic or clan genealogy, has three stages of reference,
reflecting a temporal, spatial, social, and cosmological sequence:
first, the original "anchoring" figure; second, the mediator, last, the
human here-and-now. An attempt is made to relate the fragmentary,
multiple, dispersed, or confused state of present human society to the
unitary, idealized, and orderly condition "in the beginning." The
charter myth's sequence of names, priests, officeholders, or places
inhabited, and conditions met and dealt with, thus refer to the structure
of society and the universe, or to recollected human events and indi
3
viduals. Frequently Kongo etiology myths have been confused with
history, or with pure legend, because of the parallel (metonymic)
series of persons, roles, legendary figures, animals, colors, or cosmo
logical categories which inhabit them. An effort to resolve this
problem of history in myth is pertinent to understanding Lemba,
whose lyrics contain, as the last section shows, a combined view of
human and nonhuman structures.
297
298 STRUCTURE
The historical school of ethnology has a clear answer to this prob

lem, one which numerous analysts of Kongo and Lower-Zaire culture
have adopted. Frobenius' theory of "spiritization" (Vergeisterung) of
ancestors holds that masks, cult deities, mytjiic heroes such as Nga
Malamu, perhaps even Mavungu, were manistic creations standing
4
for the long-deceased in the time-dimmed memories of the living.
According to this theory, the ancestors' individual identities became
blended with sun, plant, and animal identities in the process of
working out cultural conflicts, whence the frequency of masks with
two, three, and four faces throughout the secret-society region of West
and Equatorial Africa. This view of African ritual symbols and art
was very common at an earlier period in European ethnology. Laman
adopted it in regard to Kongo min'kisi, arguing that one could justi
fiably abandon the min'kisi without doing damage to a postulated
fundamental Kongo religion of which they were merely a manistic
5
degradation. Bittremieux used the theory to suggest that Lemba was
6
a more recent institution than earth-shrine Bunzi. Recently Thiel has
analyzed regional Zaire and Kasai-basin religion in terms of the
gradual spiritization of human figures into middle-range deities who
are situated in a religious cosmology alongside or beneath true "high
7
God" figures.
The problem with this approach is the partiality introduced by the
historical bias. Some of the figures of Lemba charters may very well
have been historic persons of prominence, such as an eponymous
Teke chief Nga Malamu, or Haitian national heroes Jean Petro,
ToussaintL'Ouverture, Rigaud, Dessalines, Christophe, andPetion
in the New-World Lemba variant. Yet at the other end of the con
tinuum there are those charters whose figures are typological male/
female, androgynous beings, or some other characterization of a
social role or relationship. One variant of Lemba called Lemba-
Nsongi states outrightly its goal as assuring the continuity of the pat-
rifilial blessing over three generations, from fathers to children and
grandchildren, thereby closing the circuit of exchanges so important
8
to social harmony. In this latter example a rather abstracted function
is attributed to the n'kisi, one of succession as well as relationship, of
diachronic continuity as well as synchronic coherence.
A view of human thought which clarifies this problem in myth
analysis allows models of society and human experience to be devised
by either (1) personalizing the individual who has an experience or (2)
emphasizing the relational situation in which he may be involved.
History is enhanced by personalizing, individualizing, an experience,
Modes of Lemba Thought 299
but at theriskof exposing the individual to social threats and conflicts.

Even so, Lemba initiation, like other initiations, gave an intiate anew
personal name, often one expressing his accomplishments and
anxieties. Emphasis on the relational, by contrast, enhances the
social and religious categories that lend meaning to an experience.
In Lemba thought, which is characteristic of Central-African
throught, the relational emphasis in ritual invokes the various
domains of the natural and cosmological world, often it would seem
simply for the purpose of connecting the individual experience to all
the rest of the universe, to give it a totalizing framework. Another
important reason for naturalizing a ritual situation is to deal more
readily with those uncomfortable, ugly human emotions associated
with anger, witchcraft, envy, competition, and power, which cannot
well be diagnosed by isolating individual, named persons.
The next several sections will show how, and at which moments,
these possibilities of metaphoric transference and movement are
invoked. Figure 25 illustrates some of the features of the Kongo ritual
metaphor vocabulary.
Successional and Relational Modes of Authority
The successional and relational aspects of Kongo ritual charters may

be manipulated to express authority modes ranging from hierarchic to
egalitarian, as well as ranging from the autonomy of a person or cor
porate unit to its complementary involvement with other units. Clan
genealogies that express extensive complementarity of local and
regional lineages frequently show three houses descending from a
common source. Exclusive genealogies tend to show a direct lineal
descent from an original founder. A similar structural transformation
is possible mn'kisi charters, inchxdmgLemba. Often charters of indi
vidualistic medicines and commissions are inserted in a more inclu
sive charter of a collective nature. The individualistic charters are
fragments of the larger ones. It is in the major min'kisi and related
commissions of chiefship, kingship, and prophet movements that
transformations from centralized and hierarchic to decentralized and
complementary may be found. For example, in a set of texts by
Laman's catechists derived from the Kamba region around Mboko
Nsongo, the same set of deities and mediators—Funza, Mpulu-
Bunzi, Mahambu-Lukabu, and so on—are ranged in one instance as a
successional hierarchy through which all power passes until in
300 STRUCTURE
9
Mpulu-Bunzi it is dispersed to "all other min'kisi" But in the other
variant, this dispersal is made from Funza, the god of twins, to three
mediator gods, Mahambu-Lukabu, Mpulu-Bunzi, and Mabiala ma
Ndembe, from whom power is distributed to all other min'kisi (figure
10
26, a and b).
Figure 25
Categories of ritual transformation in Kongo
naturalization humanization
persons
social roles
& groups
plants
deities
shades &
animals
cosmo logical
colors
categories
"anchored"
Primordial
originator
- r
.
1 V
mediator
f
|
"here-and-now"
Human
I A .A.Ar L L
r
.
Figure 26
Charter myths of several Lemba variants
and other minkisi and kingdoms
Funza
I
Mahambu-Lukabu Mahambu- Mabiala-
Mpulu-Bunzi Lukabu ma Ndembe
a\\\
all other mm kisi
(a)
/im
BaKamba
Nzambi Nzambi Nzambi

Nga M a l a m u Mahungu Tsimona-
I / \ Mambu\
Kuba Nlumi = Mbuti
I
Magungu two wives
(c) Text 4 (d) Text 5 ( e ) T e x t 11
Balari Manianga Mayombe
N z a m b i a Pongo
Lemba
(f) Brazilian
Candomblé
J e a n Petro
I
Toussaint L'Ouverture
I
Rigaud
Dessa lines
I
Christophe
Petion KaBuanga MaLoango
I
Lemba
Buanga
(g) Haitian (Pfemba) KaKongo Loango
Lunga
Lemba-Petro
(h) Text 13
Cabinda/Ngoyo
302 STRUCTURE
The relational mode of authority is in Lemba and other charters

often expressed as a marriage, out of which flow further qualities or
heroic figures, emphasizing the complementarity of social and cosmic
units. A series of original incestuous sibling/couples inhabits these
myths, as is characteristic of many Bantu etiology myths. The most
inclusive of these myths that includes Lemba comes from Cabinda,
as told by the last court historian of Ngoyo.
Text 13
(1) The earth, sky, ocean and Nzadi river have always existed, and
all that moves, moves in this framework. At the beginning of life,
in Yalala Songo, a lake at the foot of the second cataract
upstream on the Nzadi, there appeared the heads of the three
god: Kuiti-Kuiti, most powerful of all, creator of all, master of
the world, also named Nzambi Mpu Ngu, and Kisi-a-Nsi, earth
god; at his right side appeared Bati Randa, also called Kunda
Bala, ruler of animals with tails and their creator, who is in
charge of rain and water; the third deity was M'boze, chieftess of
prayer. At first they were one, but as they grew up they
separated. The three gods take any form, preferably orphidic.
Kuiti-Kuiti and his sister/wife take the shape of huge male and
female serpents named N'Tyama and LuBendo. N'Tyama is the
"stallion" of Kuiti-Kuiti, LuBendo the "mare" of M'Boze.
Kuiti-Kuiti and M'Boze never separate, nor N'Tyama and
LuBendo, their masters, twins who form whirlpools in the
Nzadi.
(2) Kuiti-Kuiti went out with Bati Landa leaving behind M'Boze his
wife with her son Kanga. When he returned he saw to his
consternation that M'Boze was pregnant. He accused his son,
who said his mother had enticed him. Kuiti-Kuiti killed them
both in rage. While she was being murdered a deep fog settled
over the world, and she bore a girl named Bunzi. Later Kuiti-
Kuiti regretted his act of murder, and resurrected his sister/wife
and son. Then he gave the new goddess Bunzi in marriage to
Kanga, telling him to go away with her and never return. Kanga
went to live at Nto on the right bank of the Nzadi; Bunzi
upstream at Ne M'lao, Tchi-Sinda, near Banana.
(3) At Nto, Kanga had a large temple, six meters long and three
wide, surrounded with fromagiere trees. On a throne there was a
cane of m 'bota wood (which also could be found in the temple of
Lusunzi), ornamented with a buffalo tail and an ivory figure of

Kanga. Kanga is invoked when rain fails to come because of
sexual violations in the land.
(4) Kanga and Bunzi are the main gods of Ngoyo, their devotees
giving pre-eminence to one or the other. They preside with their
daughter, the mermaid LuSunzi, at the crowning of the
MaNgoyo, king of Ngoyo. Bunzi, not wanting to be dependent
on her father and mother, M'Boze and Kuiti-Kuiti, went out to
N e M'Binda, first man, and father of the triplet kings
MaLoango, MaKongo, and MaNgoyo, receiving from him
white chalk with which to consecrate the kings.
(5) Among the BaVili, Bunzi, tranquil, retiring, and moderately
interested in affairs of the soil, is called mother of mermaid Tchi
Kambizi, who is extremely violent like a soldier with arms,
ready to fight to protect the authorities. She favors her friends,
offering them money; she destroys her enemies. At Loango the
tempests are attributed to her. At Ngoyo she is called LuSunzi
(N'sunzi), who lives in a river of her name, the Tchi Buanga,
matron of education.
(6) LuSunzi is a female with two faces, one white, the other black.
Her body is similarly colored. From a distance, she is the color
of fire. Nzambi Mpungu sent her to be married to Ma N'kakala.
She is the legislator of men, who gives the laws of LuSunzi. At
midnight she protests excesses of those who have her laws,
telling her priests in each village to condemn or absolve them.
She appears in dreams, revealing the reasons for her actions.
She permits kisi-ba-si to represent themselves as living or dead;
those bakisi who are her lieutenants, the njimbe, each take on
the name of their genie, such as Bunzi, Tchi Sinda, Kanga.
(Thus, the MaNgoyo became Njimbi Kanga a Nto.)
(7) LuSunzi created the cults: Bingo (Buanga) to distinguish and
classify families; Lemba, to prot^_hqiuehplds; Kalunga, to
confess sins between married spouses and to absolve them. She
also instituted medicine doctors, diviners, talismen, funerary
statues, to cure illness and reveal the hidden, and to protect
against witches.
(8) Progeny are asked from min 'kisi Lemba, Kalunga, and Bingo
(Mua Bua Nga). Lemba is the spirit of peace, as its name
indicates. In Lemba, Nzambi is asked, "Nzambi Lemba, give
us fecundity." He guards us, and must be respected and
11
obeyed.
304 STRUCTURE
A number of modes of relationship are used to build this cosmological

picture of complementary kingdoms and cults derived from Bunzi.
Patron founder of the kingdoms is the first human MeMbinda, and of
the cults, mermaid Lusunzi. Male and female are used here, as at
every stage of the myth, to bridge contrasts. At first, in the myth's
progression, these pairs are endogenous, incestuous, whereas in the
later stages of the myth the differentiation grows, and exogenous
pairs, even triads, are recognized. Other contrasting pairs of figures
are noteworthy. Peace and violence appear to be brought together in
related terms. Bunzi, who is calm, gives birth to LuSunzi, the violent
one. Bunzi is an earth shrine of very general application. LuSunzi
possesses men and provides laws with which to rule, and to maintain
peace, whence the nameLemba, "calm," "peaceful." The myth also
combines the successional mode of authority in the sequential pairs of
sibling/couples and male/female dyads, while at the same time estab
lishing sets of triple public structures in an intricate relational pattern
of authority (for example, the cults to each other, the kingdoms to each
other, the cults to the kingdoms). The Cabinda origin myth also
addresses the problem of history as it was formulated earlier in this
chapter. Human history, as distinct from the history of the gods and
autochtonous creatures, begins with MeMbinda, father of the king
doms. As in most Kongo origin myths, this one clearly differentiates
series of beings: humans, ancestors, nonhuman creatures, plants,
cosmological domains, and the like, and relates them according to
several modes of association and succession. Whatever order emerges
is not erroneous history, or the history of humans forgotten, but the
resolution of human and cosmic dilemmas through analogy and
metaphor, by transference to parallel domains. This chapter builds up
the hypothesis that humanization and naturalization in myth and
ritual are directly related to the perception of these dilemmas. What
needs to be determined yet is under what conditions, and in what
situations, humanization or naturalization of the discourse occurs.
Conjunctive and Disjunctive Mediation

An important feature of the lyrical in Lemba is the manner in which
songs and origin myths diagnose characteristic human dilemmas and
then offer the "Lemba solution." In foregoing chapters it has been
suggested that heroes such as Mahungu and Tsimona-Mambu become,
in these Lemba-related narrations (for example, Texts 5, 11), the
masters of conjunctive mediation. Outside of Lemba, these heroes
and others become involved in disjunctive or failed mediation. These

concepts, loosely borrowed from structural analysis, need to be more
12
sharply defined for purposes of the present study.
The lyrics introduced above are replete with common problems of
historical and present life in north-Kongo society. Some of these have
to do with attitudes, rhetorical or surface problems, and confusions.
There is the concern for barrenness (Text 1.46), for harassing spirits
(1.21-5), pain(1.94), dreams, nightmares, and possession (9.8), for
failure to return a borrowed tool that has been lost (7.5ff.), and other
problems of daily life. Some of the trickster tales focus on another
type of problem of daily life, that of misunderstandings, the con
sequences of which can be disastrous. Yet other facets of these same
lyrical passages should be called structural analyses of deeper socie
tal dilemmas, such as the consequences of wrong or inadequate
political alliances in society (Texts 7-8, 11), the importance of
following the counsel of ancestors and spirits, and being at peace with
the supernatural (Texts 5, 11), and the great desirability of open
relations between fathers and children (Texts 5, 8, 11).
A considerable degree of attention is given to formulating the
negative side of these dilemmas, to narrating the confusing words and
leading the character of the story through wrong action; or of
developing at great length the antihero who embodies the negative
values and traits. Some of this negativity, as the dilemmas which are
embodied in the hero, is simply an inversion of the commonly under
stood social and cultural norm. Frequently the trickster Moni-
Mambu is put into this role. Mahungu's characteristics differ from
this, in that here he (it) is often cast as a dichotomized character type,
whose two parts develop the polar implications of a dilemma, or the
two roles in a relationship. Sometimes both parts will be "false"
solutions (for example, Text 8, life with the python woman excludes
life with father, and vice versa). At other times one will be victorious
and the other defeated (for example, Text 7, one brother a slave, the
other a master). There are here then a number of ways of formulating
the negative, implying related ways of reaching solutions.
Some of the exaggerated negativity, particularly that having to do
with misunderstood verbal instructions, has,to do with a play on
"comic" possibilities, ludicrous developments of horrors and errors
setting the hero off from real life but reminding listeners of the dangers
of ambiguous situations. The early Moni-Mambu, supposedly
unaware of social syntax, is typically cast in these situations.
As soon as the trickster, or some other lyrical figure, becomes con
scious of the syntax—the structure— and begins to manipulate it or be
306 STRUCTURE
manipulated by it, the dilemmas are more serious and take on the aura
of deeply heroic, religious, and tragic motifs. The conclusion of the
myths, their denouement, may take several turns, conjunctive or dis
junctive: bringing together the odds, resolving the dilemma, over
coming the contradiction; or being destroyed by the dilemmas, con
tradictions, and odds.
The figure who is cast as antihero, inverse of norms, may be rein
corporated into society, redeemed so to speak. Mahungu as ogre-of-
the-forest (Text 6) meets this kind of end when he is arrested and
brought into civil society. Moni-Manbu as bringer of Lemba medi
cine is brought back into normal healthy life by his effective maneuver
ing of the satchel—and n'kisi—and his recognition by his father, God
(Text 11). The conjunctive resolution of the dichotomous hero, such
as Mahungu, may be seen in the reintegration of the two parts into one
complementary whole (for example, male and female in marriage,
Text 4, or perhaps the two brothers working out a complementarity as
hunter and cultivator, superior and inferior mediator, in Text 7).
The disjunctive conclusion of these lyrical accounts makes the anti-
hero, the inverse of the norm, into a victim who is killed by the forces
of authority in society. Moni-Mambu in Text 12 picks up all the sins
imaginable in Kongo society, from innocently misunderstanding in
structions to consciously flaunting rules and conventions such as
relating wrongly with his in-laws, to confusing codes of food, and to
causing death and destruction around him. Perhaps he could be called
13
a scapegoat figure, but he is more the hero of tragedy in that his fate
and character change when he becomes conscious of the contra
14
dictions in society. In this he is different from the Greek tragic hero
who is destroyed despite his ignorance of cosmic and social laws.
The Kongo tragic hero is vindicated for his errors until he becomes
aware of the laws, then he is punished for continuing mistakes that
destroy others.
The disjunctive conclusion is somewhat different in the case of the
dichotomous hero who embodies alternative possibilities in a
dilemma. This type of hero, personified by Mahungu in the present
study and in Lemba, seems always drawn beyond singular charac
terization to the dramatization of a relationship. His failure to develop
such a relationship within his complementary facets (male/female,
elder/junior), results in his destruction and death because of the
paralysis arising from incompletion, as was true of Mavungu in Text 8
who went with the python woman and rejected his father. The hero of
these types of disjunctive conclusion myths is closer to the Greek
tragic hero. Mavungu of Text 8 is portrayed as if irrevocably bound t o

his decision, made at a time when he was a youth bored by life under
his father's tutelage, unaware of the significance of adult exchange
with his father. The tragic hero here may be characterized b y his
entanglement in circumstances and structures that lead him to a con
frontation with disparate and irreconcilable alternatives and thereby
to his own destruction.
It would be wrong and oversimplified to conclude here that
Lemba? s therapeutic concept was, then, the fabrication of an image of
society and personalities in which all dilemmas are somehow over
come through "the Lemba way" of doing things. It has been sug
gested that rising self-confidence characteristic of a bourgeoisie
15
engenders a positive self-image, without an element of the tragic.
The elite status of Lemba might account for the absence of tragedy in
the origin myths. But to stop here would be to ignore the significance
of Lemba's medicines and their ritual function, the subject to which I
turn next.
Humanization and Naturalization of the Discourse
Bernard Dadié, the well-known African novelist, has stated openly

what many observers of African cosmology and thought have sus
pected, namely that African fables about human-like animals are
16
indeed commentaries about the human scene. The material of the
present study suggests that this process of substituting animals, o r
other dramatis personae, for real living human beings is a rather o l d
practice. It is not just a literary convention for the purpose of amusing
an audience but an integral necessity of life in a society where it i s
dangerous to speak openly of anger, conflict, power, and the other
dilemmas of social structure. My explanation of Lemba medicines
and their ritual function, deferred until here because of the complexity
of the issue, is related to the naturalization of expression in t h o s e
situations and at those moments when it is impossible to handle the
discourse of human interaction by direct, verbal means. Moreover, I
shall hypothesize that naturalization of the discourse between persons
permits feelings and phenomena intrinsically muddled, contradictory,
and many-sided to gain a more coherent understanding within a c o m
munity. Lemba's various modes of expressiveness, the domains
studied in parallel, reflect a tendency to move the conscious, verbal
308_JIRUCTURE
narratives toward a positive, conjunctive resolution of characteristic

dilemmas of social life—for example, patrilateral ties reinforced,
marriage alliances strengthened, market trade encouraged, and so
on—all the while the naturalized rituals of medicine "draw off" the
dirty, contradictory, polluted, and evil aspects of the neophyte's
illness.
Both the songs and the origin myths of Lemba demonstrate a certain
tendency toward naturalization of imagery in those settings where
negative ritual statuses are considered. The antelope's growth from
fawn to adult is invoked as an appropriate analogy for the transition of
the Lemba neophyte from harassed candidate to full and secure
member (Text 1.21-5). Barrenness is spoken of in terms of a "field's
barrenness," which will be made fertile (1.46). The transition from
ritual death to resurrection in Lemba is described by referring to the
movement of the sun (1.92-6). The incompetent orator is compared
to the nsibizi rodent, whereas the effective orator, the Lemba advo
cate, is likened to nkumbi rat (Text 3). The palm of God is the natural
object at the center of the separation of Mahungu male and female
(Text 5), necessitating their efforts at thought and mediation. And, of
course, the satchel of helpers is assembled by Tsimona-Mambu
during his quest for relief for his wives and is unpacked and used
during the various tests set up by a father reluctant to give his recog
nition (Text 11). In Lemba, however, the lyrical domain usually
retains as its point of reference the humans who are in Lemba—
father, son, wives, priests, priestesses—and their virtues. Passages of
naturalization in the lyrical scores are fairly specific analogical
metaphors to amplify the character traits of a specific affliction or
adverse condition. This is possible, as suggested, because the full
content of adverse emotions and perceptions is diverted to the
medicinal ingredients, the nonverbal domains of expression.
In non-Lemba myths the naturalization of disjunctive passages is
more pronounced. Semihuman characters such as the python-woman
(mermaid?) make their appearance, or individuals become changed to
animals or birds midway through the story. In the tale of Mavungu-
son's marriage that will exclude exchanging with his father, the clue
that this will be an impossible and tragic marriage is the woman's half-
god or god-disguised identity. Mavungu-father's attempted mediation
is signalled by stray pigs wandering from Mavungu-son's village,
pursued by hunters, signifying an unconscious effort to accomplish
the forbidden (Text 8 ) .
The transformation of the hunted elephant with a spear in its side

into a human sufferer (Text 7) represents a confirmation of the prin
ciple by negative example: humanization of the discourse. The suc
cessful mediator, Mavungu-spear-loser (elder), attempts to reconcile
contrasting domains and follows his mystical guide obediently, there
by bringing the interaction of the two main characters to a human
level. Mavungu-unsuccessful-mediator, his brother, again demon
strates the case of naturalization, in that his dilemma of the lost
banana tree is not resolved but is transformed into a "bad trip" of
hemp smoking, further transposing the discourse away from the open
human level in the face of an impasse. The hunting scene with Moni-
Mambu, tragic trickster, (12.7) illustrates the potential of naturali
zation in the face of disjunctive mediation. Attempting marriage but
finding his wife hungry for meat, Moni-Mambu misses his game, des
troys his in-laws' house, drops the axe into a river, and gets eaten by
crocodile. The scene shifts to a drama between crocodile and hawk,
the original hunter having become "hunted" game.
Much of this material deals with patrifilial and alliance relations
because of such an emphasis in Lemba. However myths may be
found in Kongo oral literature to reflect the matrilineage viewpoint
There, even more than in the present texts, naturalization of discourse
imagery is used to mediate dilemmas. The Kuba-Ntu cycle, for
example, describes the fate of a woman and her daughters and son
when they discover that their oft-absent merchant father is really a
17
witch wishing to sell them. Kuba-Ntu the son by mistake breaks his
father's powder box and is hidden in a tree by the mother. To avoid
being trapped by the father, now disguised as a hunter (or in league
with a hunter), the son transforms himself into a brightly-spotted
guinea fowl which, on the day the father is selling the daughters at the
market, drops feathers or bright beads to distract everyone and
snatches his sisters away to safety. Another variant depicts the father,
turned crocodile, seizing the daughters while they are fishing and
keeping them in a cavern nearby where they are finally discovered by
their bird-brother. In all variants the son succeeds in taking his mother
and sisters to the safety of their maternal uncles' home, from where
they initiate a legal suit against the father. In some variants he is
accused of witchcraft and killed.
Of course, symbolic naturalization of dilemma situations is wide
spread in Central-African thought and verbal act, as well as ritual. If
the foregoing examples represent attempts to discuss varieties of
310 STRUCTURE
human experience, then it is true, as Dadie says, that fable-like trans

formations are made up in many instances of human relationships too
volatile to discuss openly yet which require commentary. Also, belief
in animals as witch's familiars, as in the father-crocodile example
above, may prompt the choice of a particular animal. The transfor
mation of discourse from explicitly human to veiled animal, like
Bateson's understanding of redundancy in a noisy or clogged com
munications channel, permits discourse to continue even though
troubled by conflict, anxiety, fear, jealousy, or other negative
18
emotions.
Naturalization of the discourse may also result from the ritual
operator's or storyteller's desire to describe a human situation more
cogently by invoking an image drawn from the world around him and
known to everyone. For example, to contrast the fumbling nsibizi
with the clever nkumbi and to identify Lemba's rejoicing with the
twitch of nkumbi's whiskers charges the expression with nuances
more vivid and lively than is possible in abstractions. The art of
speaking in Kongo requires this sort of metaphor building. Moreover,
ritual is based on the amplification of nonverbal metaphor. In Lemba,
the music of chiming bracelets and throbbing drums as backdrop for
the lyrics, as well as the totalizing statement of the medicines drawn
from natural and human domains, together construct a universe of
meaning that can readily objectify inchoate feelings of anxiety or ill.
As Nietzsche noted in "The Birth of Tragedy," music—the non
verbal—could express the objectivity of the will itself "and therefore
represent the metaphysical of everything physical in the world," far
better than "unsupported word or image." Music created the tragic
myth because it alone could penetrate to the real depths of a person's
experience, "through annihilation and negation, so that he is made to
19
feel that the very womb of things speaks audibly to him."
Structure, Choice, and Medicine
Given, then, that naturalization of the discourse occurs in dilemmas

of human experience and is used by metaphor-makers to focus or
amplify their messages, is it possible to identify precisely the situa
tions in which this naturalization occurs? Levi-Strauss's early
understanding of the logic of mythic discourse suggested that it
provided a manner of resolving ja cultural contradiction through
metaphoric "templates" each slightly different from the previous
20
one. This characteristic of myth would explain the transfer of the
father-son conflict over control of the daughters (sisters) in the Kuba-
Ntu cycle to the plane of the bird-crocodile drama within natural
cosmological parameters. Lévi-Strauss did not at first find the context
of such metaphoric mediations within social life; rather, he found it in
the oppositions created by the "savage"—human—mind. He was
criticized for this apparent confusion of "structural oppositions" and
"contradictions." Burridge, for example, has defined a contradiction
as alternative, goal-oriented activities or processes which effectively
21
and simultaneously negate each other. "Life" and "death" are
alternatives, meaning either to be alive or to be dead. Later Lévi-
Strauss revised the notion of mythic mediation to include the "sym
22
bolic," the "imaginary," and the "real," suggesting that some
oppositions might stem from structures of classification such as earth/
sky, birdj/reptiles, and the like, but that others might be based upon
contradictions within the structures of society and cultural values. For
Burridge, true contradictions could only occur in dramatic situations
in recognizable historic events. That is, they must be based upon
alternatives between which an actor must choose. Thus Burridge
injects into the analysis of myth structure the concrete terms of given
historic moments as well as the structure of ideology.
Such moments of choice in the face of alternatives constitute in
Kongo society the context of metaphorization of the discourse.
Judicial palaver (nsamu) may well be the prototypical context for
this. When antagonists become uncomfortably alienated from one
another, a third party or two spokesmen take up their causes. In the
negotiations that ensue, they invoke songs, proverbs, and other rhe
torical devices. Up to a point these techniques examine evidence and
probe possibilities for conciliation. The texts examined in this work
show moments like this, as for example when the trickster (in Texts 11
and 12) is judged for having killed. A chief orjudge decides the matter,
and the decision of guilt or innocence is final. However, it is in the
portrayal of irreconcilable alternatives such as that which the python-
woman presents Mavungu (Text 8) that we find naturalized
metaphors.
Examination of Lemba lyrics suggests that recourse to naturalized
metaphors may also have to do with the psychological process of
drawing a sufferer/neophyte out of an intractable personal dilemma.
A framework for analysis of this process is lent by Lacan's inter
23
pretation of psychoanalytic processes as applied to language. The
frequency in Lemba-rtlated lyrics of dreams, visions, nightmares,
312 STRUCTURE
and possession (see Texts 1.18,1.45, 1.115, 7.13-16, 8.4-6, 9.5-7)

indicates the presence of a mode of image association Lacan would
call "the sliding of the signified under the signifier," in which a hidden
24
message is postulated because of the mystification of the subject. In
Kwamba's report (Text 1) on the Lemba initiation, repeated refer
ence is made to a hidden referent or signified force such as that which
is "given and taken away" by Lemba, that which "falls in the water
when the payments are made," or that which " dies and laughs with the
dogs." These lyrics, like many types of divination, float an image
which will "catch" the subject's perception of his condition.
Another psychodynamic process may be identified in the Lemba
attempt to construct a metaphor of parallel expressive domains once
an appropriate image is discovered for the affliction or its cause. Often
one domain or point of reference in the metaphor is the hidden force or
referent in the life of the neophyte, and the second point, to which the
first is associated, is drawn from nature, resulting in such constructs
as, "What Lemba given, Lemba takes away; what the sun given, the
sun takes away." Signifiers which stand in a relationship of near
similarity are "piled up" until their common point of reference, the
hidden referent, is made clear. This symbolic process is sometimes
25
called condensation. Its role in the psychodynamics of Lemba
healing appears to be that of attaching the presumed affliction to a set
of symbolic anchors so that it maybe more effectively "moved" in the
subject's life.
A third type of process in the psychological effect of the metaphor
seen in Lemba is the "displacement" of signifier to signified, which
Lacan speaks of as the "veering off of meaning" from one object or
person to another, comparable to the action in grammar of the transi
26
tive verb in getting the noun to influence the object. Lemba lyrics
such as "that which was the fawn, is today the grown antelope"
(1.21), or "that which was barren . . . has become fertile" (1.46), or
"that which was a 'stitch' of pain, has become the path to the
priesthood" (1.94), take one quality or entity and transform it into a
related, different member of the same class. The series of naturalizing
allusions to spirits, plants, animals, and cosmological spaces serve to
bring movement into the neophyte's or the neophyte couple's status.
Evidence from other drums of affliction suggests that this is the way
they too use ritual metaphor.
In transferring the image of conflict, contradiction, and evil "down"
to natural metaphors, Lemba was able to emphasize a lyrical, verbal
consciousness of positive mediation. The conscious image of positive
mediation was clothed through the ideal roles of the Lemba Father,
the culture-hero trickster, or the androgynous hero who invented

marriage. Simultaneously the natural metaphors were made to draw
away from the neophyte couple all the dirt, evil, pollution, and contra
diction. This activity was absorbed by Lemba, and apotheosized
through identification with a mythic hero or ritual object like the
nkobe. Personal and social contradictions once contained in this
manner became "powers" (makundu) in the rule of Lemba and its
image of an ideal society.
It should by now be apparent how Lemba healed, and why it was
heuristically advantageous to adopt a model of analysis which identi
fied distinctive expressive domains, each with its own codes and
characteristics of discourse. This image of an ideal society and a
therapeutic ideology may be spelled out even more explicitly, drawing
from pronouncements by Lemba adherents, early ethnographers^
and recent interpreters of this major drum of the Congo region.
Chapter 10
The Ideology of Lemba

Therapeutics
In this chapter I shall consider the evidence for an explicit Lemba

theory of therapeutics. In the previous chapter it becomes clear how
Lemba ritual and thought manipulated symbolic frameworks to put
forward a coherent lyrical or verbal picture of human society—the
Lemba model—transferring downward into nonverbal symbols the
negative and contradictory where it became sacralized and powerful
"medicine." The task of this chapter is to engage in a somewhat
hypothetical conversation with Lemba priests and priestesses over
the conscious mission they had. Texts presented in this work and
other sources give localized and historically specific versions of this
Lemba consciousness. Interpretations since Lemba's demise by
scholars, both African and non-African, are almost as localized and
time-specific. It is the role of the historian and the ethnologist to take
all such evidence and extrapolate from it a synthesized conscious
ness, theory, or ideology from which to make sense of the institution.
The Affliction and its Etiology
Probably all Lemba adherents would agree that Lemba was a "drum
of affliction." It was spoken of as an n'kisi whose rituals were
drummed up with its unique hand-held instrument (ngoma or
nkonko). In earlier chapters, the indigenous theory of drums of afflic
tion was spelled out in terms of public, corporate, sacred medicines. In
one region of the Lower Congo/Zaire, such drums were devoted to
clan leadership, chiefship, water spirits, judicial affairs, and order in
markets and public sites (the case of Lemba)}
A more difficult to understand aspect of Lemba therapeutics has
been the conception of the affliction it was intended to treat. Particular
symptoms designated as the "Lemba illness" vary greatly. Thus one
finds a host of physical symptoms mentioned such as "evening fever"
(Text 3.2), "chest cough, stitch, or breathing with difficulty," or other
315
316 STRUCTURE
2
respiratory ailments (1.1), sterility of self or spouse, "swollen
stomach" (10.22), and the like. Other accounts offer psychosocial
afflictions such as "spouse's infidelity" (Í0.3i-4;1.105ff.), dreams
or nightmares of Lemba ancestors or authorities (1.18;9.7), hallucina
tions and outright possession by "Lemba spirits" (9.7,9.20). Several
writers attribute to Lemba panacean claims of universal healing. The
European version of this notion attributes to Lemba even the ability
3
to heal the "incurably ill"! The African version of this is that Lemba
deals with all afflictions of the abdomen, head, heart, and sides, that is
to say, the whole person as defined by Kongo thought (10.10). Speci
fic afflictions such as these no doubt originate in personal accounts of
individuals having been treated by Lemba, but they convey far too
particularistic and individualistic a view of Lemba? $ orientation.
They originate in second- and third-hand accounts quite removed
from Lemba? s therapeutic consciousness and distantly removed from
any sense of Lemba* s ideology of healing.
Closer to Lemba, and even among Lemba adherents, one still
finds particularistic notions of a "Lemba illness," such as one priest's
view that Lemba dealt with persons who had experienced a miracu
4
lous cure from an incurable disease and was a type of votive offering
by the neophyte to the power that had cured him.
Kongo authors of the past twenty years who have sought to clarify
this aspect of the Lemba affliction have tended to argue that there was
no single, specific Lemba illness. Ngoma, for example, in his disser
tation on Kongo initiation, suggests that a variety of illnesses, indeed
any illness, could precipitate the curative stages entailed in & Lemba
5
initiation. The duration of the cure would be determined by the
wealth of the candidate. Malonga and Mampuya agree with Ngoma
that there was no specific Lemba affliction. However they shift the
perspective of the question around to make the entire therapeutic
symbolism of the initiation an artifact of Lemba? s explicit approach
to social control. In Malonga's view the manifestation of one or
another affliction was a shrewd feigning of illness proposed by the
Lemba Father. With the help of a diviner treatment or pretreatment
might induce a skin rash or other symptoms which the full therapy
then pretended to relieve. Lemba afflictions were thus "iatrogenic"
6
elements in the maintenance of Lemba's public posture. Mampuya
is harshly critical of Lemba, suggesting that this sickness induced in
the neophyte, whether through psychical manipulation or through
outright mystical threat, served only the " antisocial" ends of coercion
and exploitation. Malonga, however, respectfully calls this aspect a
The Ideology of Lemba Therapeutics 317
feature of the major "social institution"—the government—of the

region, echoing Munzele's view that Lemba integrated "markets,
7
villages, and people," or echoing that of a Lemba wife that Lemba
was "a medicine of government," or that of a clan head that it was the
"government of multiplication and reproduction."
Although the scholarly views of Ngoma, Malonga, and Mampuya
thus reflect an understanding of Lemba's function which corresponds
to the views of others close to Lemba, they still fall short of compre
hension of the nature of Lemba's ideology of affliction and therapy.
All three scholars, as well as Fukiau whose work on Lemba has been
reviewed extensively in Chapter 6, portray Lemba as it was in the
early colonial era: an institution losing its control of resources and
influence. Only the primary texts in the catechists' accounts, written
early in the twentieth century, offer a glimpse of what must have been
an earlier, more characteristic picture of Lemba therapeutics.
Konda, in Text 8 (Chapter 7), offers a remarkable picture of the
Lemba affliction. Lemba may be "aroused" (komwa) through
dreams of Lemba ancestors, nightmares of suffocation, or outright
possession. These modes of affliction are sometimes accompanied, he
notes, by blockage of speech requiring therapeutic "loosening," or
"opening." Bittremieux's depiction of Lemba priests as reflected in
their names suggests that they were persons driven, even obsessed,
8
with success in trade, influence, and public prestige. Konda, of all
writers, comes closest to relating the affliction to this reputation of
Lemba adherents and aspirants for seeking wealth and influence. He
suggests that the force emanating from the ancestors to make one
wealthy and influential will also make one sick unless one is somehow
protected. The therapy for this affliction is appropriately first a con
fession by the sufferer of blocked speech revealing what he has seen in
dreams and nightmares, followed by the prescribed rituals of marriage
and massive redistribution of goods to his patrifilial children and other
bystanders.
This view of Lemba's affliction corresponds to what is known of
the historic surroundings of Lemba adherents. They were wealthy (or
wealth acquiring) and influential merchants, judges, healers, diviners,
and chiefs. Frequently they were at the head of slave segments of
clans. Their alliances to other clans, consecrated in Lemba, consti
tuted a network of sociopolitical relationships across a vast region, in
particular where the inland routes of the international trade required
some form of social control. Such wealth and influence as this brought
introduced strong currents of envy and jealousy into the small-scale
318 STRUCTURE
clan communities, touching off in the minds of these wealthy and

influential persons such symptoms as dreams, nightmares, fears of
sorcery attack, sterility, death, and a host of other specific symptoms.
It is thus not wrong to identify specific symptoms such as cough,
respiratory troubles, headaches, and other psychic afflictions as
"Lemba afflictions," if they are seen as deriving from the vulnerable
condition of the influential in a society with a strong egalitarian ethic.
The Lemba affliction is seen then as a psychic or physical symptom
of vulnerability and marginality among the influential in a society with
a normative ideology emphasizing egalitarian authority and redistri-
butive economics.
The Therapy
Lemba's therapy followed from such a conceptualization of the

Lemba affliction. The Lemba sufferer, having perceived either his
" calling" to gain wealth or his marginality and vulnerability before his
kinsmen or the loss of his productive base in his natural clan commu
nity and household, wanted to gain adherence to the corporate body
which could protect him. A variety of tests were administered to such
a candidate. Was he really a capable, influential person? Then let him
pay huge fees or find the patronage to do it for him. Did he have
strategic alliances with other clans? Then let him identify the wife or
wives who with their clans would undergo the Lemba marriage sanc
tifying this alliance. Was he torn apart by fear of sorcery directed
against him and his offspring? Then he must take medicine of purifica
tion to find a new integration above and beyond the divisive forces that
threatened to destroy him. Let him dedicate his household(s) to
Lemba so that they could continue to have children and survive into
the next generation. Was the candidate indulging in antipublic, prima-
donna rhetoric with his speaking skills and his talents for curing or
trading? Then he must by all means be "drafted" or even harassed
into Lemba circles so his influence would not tear asunder the
regional social fabric.
Lemba initiatory therapy moved the influential but vulnerable
sufferer into a course of greater self-awareness and social responsi
bility. The symbolism of the Lemba initiation came to grips with
massive contradictions, working them out in the aura of Lemba's
greater coherence as expressed in the etiology myths. For example,
the envy, greed, and destruction seen to accompany irresponsibly
held wealth was neutralized in the Lemba medicine, while the

generosity of the feast served notice that the neophyte was after all
concerned with his dependents' welfare. In several local variants, as
has been shown, the neutralizing, purifying power of white chalk
(luvemba) was combined with food and excrements and sometimes
the body parts—hair, nails—of the Lemba members. As the neo
phyte moved into the protective sancity of Lemba and was made to
distribute massive amounts of his wealth so that others might feast, his
own food consisted mainly of chalk, excrements, and token bits of real
food.
At the same time as such social contradictions were transferred to
nonverbal ritual objects in food, medicine, and the permanent memen
tos of the n'kobe shrine, mythic heroes in the etiological narratives—
trickster, androgynous demigod, earth deity, ancestor—were made to
iron out all these dilemmas and contradictions, thereby constructing a
conscious, positive ideological unity. A s with most public ideologies,
this perfected version of the institution stands at odds with the actual
lives of individuals.
At Lemba's end in the colonial era, with a strong lingering memory
of wealth and influence but with few remaining goods to distribute,
9
Lemba's "lies and exploitation" as Mampuya calls them may have
well been very far from the institution's earlier self-image. W.H.
Bentley, a pioneer British missionary, wrote that in "the 1860's
elembe was a word that filled people with fear. The cry 'elembe edio'
(this is Lemba) would stop a caravan in its path and make it submit to
capture, robbery, and death without a struggle. . . . It is difficult to
understand why the word had such an effect, but many who knew the
'elembe period' as it was called have told the same story. It was
shockingly abused and made the means of much violence and
10
robbery." But here too there is evidence of second- and third-hand
anecdotal information about Lemba in its latter phases, succumbing
to the disintegrative influences of the late, coastal slave trade.
The Virtuous Society

It is important to press the inquiry of a conscious therapeutic ideology
of Lemba further, to the point of asking what may have been the
Lemba ideology of health? That is, what was Lemba's "theory" of
social order, of the good life, and of the ideal person? A major institu
tion which prevailed over so vast a region for three centuries leaves its
320 STRUCTURE
mark on public consciousness. It must have a notion of the virtuous

society. This notion is made up of a specific model of society—a
conscious set of alternatives—in this case arising from an acephalous,
segmentary society in which problems of trade and authority needed
to be dealt with.
It must have been an item of Lemba theory to identify the natural
base of power in the society—the talented orator, the skilled merchant,
the able healer, and so on—and to bring such persons together into
Lemba's ranks. Many informants tell of Lemba's character criteria of
clarity, gracefulness, integrity, and the like. Malonga notes that
Lemba arrogated to itself all knowledge in Lari society through tech
11
niques of spying and information sharing. On this basis it was able to
maintain order in a society lacking centralized institutions of law,
police, and taxation. Its governmental methods were derived from this
need to keep abreast of the real basis of legitimacy. For this reason
there is recurring mention of the membership of slaves in Lemba. In
contrast, a hereditary aristocracy would have had to maintain control
by means of institutional centralization.
Because of the integral role of trade in the regional economy,
Lemba's self-concept included an elaborate code of market and trade-
route behavior. The marketplaces became the nodes of economic
interaction, as well as the sites for the execution of criminals. Lemba
priests, as trade and market police, were said to enjoy immunity from
tariffs and seizure of the entire region. Economic life and public order
were thus conceptually joined.
Emphasis on trade and regional social order merged in another
feature of Lemba, the initiation marriage. The importance of alli
ances between prominent clans has been frequently mentioned. It
may be noted here that marriage and the protection of the Lemba
household were central in the ritual symbolism and in explicit prohibi
tions binding priest and priestess. The importance of strong alliances
was grounded in symbols of purity in the Lemba marriage. This is so
striking that observers of the highly moral tone of Lemba marriages
have suggested influence from eighteenth-century Christian mission
aries. However, similar marital conservatism among coastal nobility
and royalty in the sixteenth and seventeenth centuries makes it alto
gether probable that this emphasis was internal and indigenous, grow
ing out of the keystone role of marriages to the social order. In some
places in the Nsundi andBwende regions (the Manianga), the empha
sis on alliances bears a very specific ideological label, the "reciprocal
blood" marriage. Usually embodied in a patrilateral, cross-cousin
marriage, this reciprocation in an alliance requires a male to return, in

his marriage, the descent substance received from his father's lineage.
An alliance of the reciprocating type combines the emphasis on mar
riage with that on patrifilial continuity. This special understanding of
social order, often found in communities of the Lemba region, thus
possessed not only a territorial component but also a long-term,
temporal component.
This temporal continuity contained in the diachronic structure of
reciprocal marriages between exogamous clans is also the model of
the religious hierophany. Just as the father-child dyad embodies the
Lemba master-neophyte relationship of continuing spiritual power
(kundu or kitswa), so it serves as the model of relationships between
major spiritual beings and lesser, and between these and medicines
via the human priests who convey their power to human society.
Patrifiliality in Lemba was the relational metaphor of all ascending
and descending pairs in the religious hierophany.
There is no doubt, therefore, that the Lemba therapeutic ideology
was a model for a fully-formed social state. That it was realized to a
great degree in the form of a drum of affliction surprises only those
who have failed to take Central-African society on its own terms.
Chapter 11
Conclusion: What Were the

Questions?
The immediate aim of this work has been to elucidate the emergence,
duration, and decline—the life cycle—of a major drum of affliction in
the hope of better understanding this important genre of African
therapeutic institution. Because the dmm(ngoma) combines features
often kept discrete in Western institutional perception—healing,
politics, commerce, marriage—it has been necessary to show how
these features articulated in an original manner. Once this was done, a
second aim could be addressed: whether the drum of affliction in
general, and Lemba in particular, is a reflection of the socioeconomic
order or whether, in striving to resolve certain dilemmas and contra
dictions in that order, it actually shapes society in a unique way. In
other words, is the drum of affliction an independent or a dependent
variable of social change?
The first part of the work consisted in a reconstruction of the
economic, political, and social history of the Malebo Pool-Cabinda-
Loango triangle. This exercise, a regional history which needed to be
written, showed the interdependent character of the well-known states
of the area—Loango, Kakongo, Ngoyo, Nsundi, Kongo, Tio—as well
as the less-well-known cults and shrines. In analytic language this
history of the Lemba region demonstrated the interlinked corporate
structures of local landed estates, the large-scale polities, and the
movable corporate estates. The same interdependence characterized
the relationship of the local economic base to the international trade.
In other words, local, regional, and international economies were
definitely found to intersect one with the other, just as was true of the
"economy" of power symbols and medicines. In all of this Lemba,
the major transcending institution of the region from the seventeenth
century to the twentieth century, apparently played the key role of
creating a ceremonial context for the economy of trade to mesh with
the economy of agricultural production, and a context in which to
generate medicinal symbolism to assuage the lives of those at the
intersection of the two economies.
323
324 STRUCTURE
The second part of the work was devoted more pointedly to Lemba
rituals and to the study of their variations along the lines of differential
corporate structures of the region. Going beyond a good measure of
free variation in Lemba's ritual style, there was evidence of a corre
spondence between the rites of Lemba as a portable medicine shrine
and the long-distance overland trade, on the one hand, and between
Lemba's fixed shrines on the Loango coast and the endpoint of the
trade, on the other hand, where the brokerage role of a sedentary com
mercial elite in touch with European traders was a determinant.
At another level the Lemba rituals of therapy functioned to address
a concern for the protection of Lemba members and their households
from the envy of others and for relief from symptoms of such envy and
the social precariousness accompanying it, for example, dreams,
nightmares, possessions, and a range of physical symptoms such as
"stitch in the side." Lemba's therapeutic functions went beyond the
individual and the household to the society at large where an effort
was made at restructuring social relations. Ceremonial goods were
distributed to the neophyte priest's dependents; stable alliances were
created and legitimated; a new reality was forged through hero narra
tives which resolved some otherwise implacable dilemmas of the
prevailing culture. These narratives deserve closer attention here.
The significance of the resolution narratives of Lemba lies not so
much, however, in particular outcomes as in the way the mode of
narrative used in Lemba therapy contrasts with other types of prob
lem-resolution efforts, particularly those offered by the creation, or
attempted creation, of new polities, especially centralized state-like
regimes. There is widespread evidence of experimentation in the
creation of institutions, especially coastal institutions, in the centuries
surrounding Lemba, institutions that vary from centralized mini-
states to alliance networks and ritual movements. In other words it is
apparent that Lemba might have been displaced or replaced by the
creation of another kind of institution. Thus it is significant that
inhabitants of the region made a selective choice for the kind of public
order that emerged, that, instead of imposing a new order to deal with
the coastal trade which resembled a state, they developed a solution to
the challenge of trade which emphasized the redefinition of reality in
therapeutic terms. There are important implications in this for the
writing of intellectual history, very much a concern among Afri-
1
canists.
It is important to explore beyond historical "solutions" offered by
institutions and their ideologies to discover the problems or dilemmas
Conclusion 325
they were intended to resolve, and to discern the way a society

imagined alternatives open to itself and the consequences of such
alternatives if taken. A brief allusion to a comparative setting will
emphasize the approach I have tried to follow. In a recent work on the
origin of the "political" in ancient Greece, Christian Meier suggests
that the recognition of alternative courses of action in the Athenian
polity was greatly enhanced by the public arena of the theaters,
especially the tragic theaters. Public funds were expended on play
wrights who were invited annually to produce new dramas. They often
took story material from Greek mythology and legends which they
then reshaped into commentary on contemporary (fifth century B.C.)
life in Athens and the surrounding region. Thus arose a keen apprecia
tion for tragedy, which spelled out the consequences of courses of
action around the family, the polity, and the individual in relationship
2
to the gods and the forces of fate.
The task of the third part of this present work has been to identify
comparable settings or idioms in which alternatives, contradictions,
and ambivalences were raised and dealt with, an analytic course
which is imperative if one wants to explain the emergence of a distinc
tive institution such as a major drum of affliction, or for that matter of a
state or any other institution that alters the texture of society. The
questions raised by Part III suggest that one can identify traces,
vestiges, of the perceived purpose of Lemba. Text 8, for example,
shows Mavungu moving through the alternatives open to any indi
vidual coming to terms with the coastal trade, a situation which
certainly provoked debate and questioning. Mavungu is lured to the
"city at the coast" (8.16) where there are "whites" and "great
wealth" in the "ships." However the same text goes on to show how
Mavungu believed he needed to exchange his newly-earned wealth
with his father and other kin. The dilemma of accumulation versus
exchange is however couched in terms of conventional alliance versus
identity, thus pitting marriage and affluence against patrifilial rela
tions. This "tragic" exclusion of one of two idealized goals is
accounted for in terms of the seductive lure of the "Mammy-Water"
like female embodied in the serpent, but the conscious formulation of
alternatives and consequences in connection with the trade is clear.
Other narratives (Text 7, for example) speak of social cleavages in
society or the dangers of ambiguous language (Texts 9-10). The point
at issue here is that we can best compare attempted resolutions of
these dilemmas, contradictions, and alternative courses of action in
terms of the backdrop of implicit and explicit formulations of their
326 STRUCTURE
extreme, that is tragic, consequences. It is safe to assume that without

a cultural awareness of such alternatives no solution of any kind
would be attempted.
The therapeutic rites and associated etiological narratives record in
much greater detail the attempted solutions of various dilemmas,
contradictions, and alternative courses of action. I have shown that
with the help of separate "expressive domains" an ideology of health
is promoted at the level of verbal narrative. This is, as I have shown,
more than a denial or sublimation of the contradictory. As close study
of the "medicinal" code shows, a great deal of symbolic tension is
present between the dilemma-negating narrative and the other levels
where an all-out program is involved in harnessing, capturing, or
containing the power of the "illness." Through a paradigmatic shift
from the "human" to the "natural" the power of the illness is trans
formed into the power of the priesthood, recalling recent biofeedback
experiments designed to give a patient a visual concept of his affliction
so as to gain mental, emotional, and, one hopes, physiological control
over it.
Whether a given drum of affliction in the Central-African setting is
then a dependent or an independent variable of social change and all
its forces of contradiction can only be answered once the questions
which become the grounds for attempted solutions are elucidated,
including a variety of rituals that pour answers into the questions as
well as a variety of efforts to impose new orders of institutions such as
the centralized state. In Greek antiquity such questions had to do with
the freedom of the individual in the face of impersonal fate. In Kongo
during at least part of the era of the great trade such questions had to
do with the accumulation of wealth and influence in a society of
egalitarian expectations, the protection of the individual engaged in
commerce, his relationship to his kin, and the creation of a regional
institution alongside local clans and dependents. Greek society
thought out its alternatives and their consequences in the tragic
theater and acted on them through a form of participatory democracy.
Kongo society thought out its alternatives in legends, dreams, and in
affliction etiologies and acted on them in a therapeutic-alliance-
trading association. There is no doubt thai Lemba shaped the public
order, although by no means all medicines did so or may be expected
to.
Notes,
Sources,
Index
Notes
Chapter 1
1. De Jonghe wrote widely on African religion and ethnology, and his
views on "secret societies" and ethnological theory are found condensed in
the following works: Les sociétés secrètes au Bas-Congo (Bruxelles, 1907);
"Les sociétés secrètes en Afrique," Semaine d'Ethnologie Religieuse Ser. 3
( 1923); "Formations récents de sociétés secrètes au Congo Belge," Africa 9,
no. 1 (1936): 56-63. Unless otherwise indicated, the review of works is
drawn from De Jonghe's 1923 article.
2. H. Schurtz, Altersklassen und Männerbünde (Berlin, 1902).
3. S. Freud, Totem and Taboo (New York, 1950). See especially the
introduction for explicit mention of this influence.
4. W. Wundt, Völkerpsychologie, 10 vols. (Leipzig, 1910-20).
5. A. Kuper, Anthropologists and Anthropology: The British School
1922-72 (New York, 1973), p. 24.
6. J. Frazer, The Golden Bough, 2 vols. (London, 1890); Totemism
and Exogamy, 4 vols. (London, 1910).
7. A. Van Gennep, Les rites de passage (Paris, 1909).
8. L. Frobenius, Die Masken und Geheimbünde (Halle, 1898).
9. K. Laman, The Kongo, III (Uppsala, 1962), p. 67.
10. F. Gräbener, "Kulturkreise und Kulturschichten in Ozeanien,"
Zeitschrift für Ethnologie 37 (1905): 84-90.
11. De Jonghe, Les sociétés secrètes au Bas-Congo.
12. De Jonghe, "Les sociétés secrètes en Afrique."
13. De Jonghe, "Formations récents de sociétés secrètes."
14. J. Van Wing, Etudes BaKongo (Bruxelles, 1959), pp. 426-508.
15. L. Bittremieux, La société secrète des Bakhimba au Mayombe
(Bruxelles, 1936).
16. V. Turner, Drums of Affliction (Oxford, 1968), p. 15.
17. Certainly E.E. Evans-Pritchard's classic Witchcraft, Oracles, and
Magic among the Azande (Oxford, 1937) set the tone for recognition of a
system of explanation of misfortune and the means of dealing with it Recent
regional studies in Bantu-speaking Africa that illuminate the general lines of
this system include the following: from Tanzania, M.L. Swantz, Ritual and
Symbol in Transitional Zaramo Society (Uppsala, 1970); from Uganda, J.
Orley, "African Medical Taxonomy," Journal ofthe Anthropological Soci-
ety of Oxford 1, no. 3 (1970): 137-150; from Zimbabwe, GX. Chavunduka,
Interaction of Folk and Scientific Beliefs in Shona Medical Practices
(London, 1972); from South Africa, H. Ngubane, Body and Mind in Zulu
Medicine (London, 1977); from Western Zambia, G. Prins, "Disease at the
331
332 NOTES
Crossroads: Towards a History of Therapeutics in BuLozi since 1876,"

Social Science and Medicine 13B (1979): 285-315, and W.M.J. Van
Binsbergen, "Regional and Non-Regional Cults of Affliction in Western
Zambia," in R P . Werbner, ed., Regional Cults (London, 1977), pp. 141—
175; from Western Congo (Zaire), J.M. Janzen, The Quest for Therapy in
Lower Zaire (Berkeley, 1978), and Janzen, "Ideologies and Institutions in
the Precolonial History of Equatorial African Therapeutic Systems," Social
Science and Medicine 13B (1979): 317-326. For an essay that seeks to
describe the historical subcontinental system, see W. DeCraemer, J.
Vansina, and R. Fox, "Religious Movements in Central Africa," Compara-
tive Studies of Society and History 18 (1976): 458-475.
18. I am drawing these phrases from my own work, but they appear in
most of the above-mentioned publications, frequently using the very same
cognates from region to region.
19. The phrase is drawn from Y. Kusikila, Lufwa evo Kimongi e?
(Kumba, 1966), published in English translation in J.M. Janzen and W.
MacGaffey, eds., Anthology of Kongo Religion (Lawrence, 1974), pp. 48-
55, although the idea is widespread in Bantu Africa.
20. These terms are drawn from Kongo, although the verbal stem of
n ganga appears across the entire Bantu-speaking region, as do other terms
associated with the therapeutic system under consideration. M. Guthrie's
Comparative Bantu, 4 vols. (Hants, 1967) is an important reference for
tracing the distribution of verbal concepts connected with this system.
21. Janzen, The Quest for Therapy, 1978.
22. Prins, "Disease at the Crossroads," has demonstrated the effective
ness of this approach in reconstructing intellectual history. Long-term sur
viving values and categories, amidst changes, indicate for him the "core" of a
culture. He has used therapeutics as such a barometer of Lozi culture in the
long term in his historical study Hidden Hippopotamus (Cambridge, 1980).
23. Nsemi, "Min'kisi: Sacred Medicines" in Janzen and MacGaffey,
Anthology of Kongo Religion, pp. 34-38.
24. This is especially true of the work in Werbner, Regional Cults, and is
5
exemplified in Werbner s introduction, pp. ix-xxxviii.
25. Ibid. Werbner bases his analytical perspective on regional cults in
large measure on V. Turner, "Pilgrimages as Social Process," Chapter 5 in
his Dramas, Fields, and Metaphors (Ithaca, 1974), p. 185.
26. Turner, "Pilgrimage as Social Process," pp. 185ff.
27. Werbner, Regional Cults, p. ix.
28. VanBinsbergen,"RegionalandNon-RegionalCults,"pp. 141-175.
29. J.M. Schoffeleers, "Cult Idioms and the Dialectics of a Region" in
Werbner, Regional Cults, pp.- 219-239.
30. J. Vansina, The Tio Kingdom of the Middle Congo, 1880-1892
(Oxford, 1973), pp. 221-243.
31. K. Garbett's "Disparate Regional Cults and a Unitary Ritual Field
in Zimbabwe" in Werbner, Regional Cults, pp. 55-92, is an exception to
Notes 333
this. Garbett, in his analysis of the Mutota cult of Zimbabwe, develops a clear
picture of several cults interpenetrating in a single region. Alongside the
centralized, hierarchic ancestor cults he finds other nonhierarchic and ter
ritorially undefined cults (p. 58).
32. Werbner, Regional Cults, pp. xvii-xxii.
33. Van Binsbergen, "Regional and Non-Regional Cults," p. 144.
34. B.T. Van Velzen, "Bush Negro Regional Cults: A Materialist
Explanation," in Werbner, Regional Cults, pp. 93-116.
35. Ibid., p. 94.
36. K. Burridge, New Heaven, New Earth (Toronto, 1969), says that
millenarian activities provide a test case in social analysis for the joining of
statements valid for both participants and investigator. "Beyond their intrin
sic human interest.. . millenarian activities constitute an acute theoretical
challenge. They invite a statement through which particular actions and
rationalizations may be given a more general validity" (p. 2).
37. C. Geertz, in "Religion as a Cultural System," in M. Banton, ed.,
Anthropological Approaches to the Study of Religion (New York, 1966),
cites Santay ana to the effect that" any attempt to speak without speaking any
particular language is not more hopeless than the attempt to have a religion
that shall be no religion in particular.... Thus every living and healthy
religion has a marked idiosyncracy" (p. 1).
38. T.O. Ranger, "Healing and Society in Colonial Southern Africa."
Unpublished MS, 1978.
39. J. Vansina has developed models of state formation specific to the
Tio and Kuba kingdoms in his The Tio Kingdom, and in The Children of
Woot: A History of the Kuba Peoples (Madison, 1978).
40. J. Goody, in his Technology, Tradition, and the State in Africa
(London, 1971), develops an analysis of state formation for Africa. He
acknowledges the importance of Southall's concept of the" segmentary state"
in which central and local powers have equal weight, a condition that has
often arisen as larger empires or states disintegrate (pp. 9-10). For reasons of
technological small scale in food production (the use of the hoe rather than
horse- or ox-drawn plow), Goody rejects a "feudal stage" in state formation
for most of West and Central Africa, arguing instead for a variety of historical
types: the hereditary structuring of ritual powers; the ability to attract and
keep a following (privileged descent groups); conquest; diffusion of the insti
tution and idea of a state; the emergence of a central state from a nucleus in
lineages, age sets, cult associations, and other institutions in acephalous
society; in opposition to slave raids; or the need to move trade goods across
long distances occupied by peoples lacking chiefs (pp. 12-18). Goody would
then concur, perhaps, that it is difficult, even unnecessary, to make a sharp
distinction between " state" and " cult," and that either can fulfill the functions
of centralized or regional institutions.
41. J. Miller, in Ms Kings and Kinsmen: Early Mbundu States in
Angola (Oxford, 1976), has reviewed models of state formation in Central
334 NOTES
Africa, arguing that generally "state" institutions are created in response to

the need and desirability of contact between unrelated lineages. Relationships
created are by definition political. They may come about in one or several of
the following ways: to control scarce but valuable resources such as salt, iron
and copper ore, or trade advantages; military or strategic advantage; as
innovations capable of attracting manpower, for example, the Mbundu and
Imbangala kilombo war camp, a converted initiation camp used to overcome
particularistic loyalties of descent groups; ideological innovation relating to
large-scale institutional integration; the assistance of outside allies; the con
trol of commercial monopolies such as those taken over by so-called "broker
states" like Angola, between Kasanji and Portugal and Brazil; agricultural
surpluses, a necessary but not sufficient factor in state formation; techno
logical superiority; and individual genius. Most of these factors could be cited
as elements in the emergence of Lemba, suggesting the pitfalls of reading into
a combination of them any teleological tendency to political consolidation.
42. L. de Heusch, in "Structures de réciprocité et structures de subor
dination," in his Pourquoi l'épouser? (Paris, 1971), has studied the Nyimi
(kingship) institution among Bushong peoples—Kuba, Lele—and has sug
gested that centralized "statehood" is but a variation of several possibilities
found throughout the cluster of Bushong peoples. This leads him, in the
explanation of state formation, to emphasize a rather willful historical event
such as the murder of a kinsman, the act of royal incest, or similar break with
the past, to destroy the "structure of reciprocity" having existed in a society
and to establish in its place a "structure of subordination." This sheds much
light on transformations of authority symbolism in the Lemba region and
related centralized systems. Extensive experimentation occurred to reach
appropriate combinations of "subordination" and "reciprocity."
43. H. Maine, Ancient Law (London, 1917).
44. M.G. Smith, Corporations and Society (London, 1974).
45. W. MacGaffey, "The Religious Commissions of the BaKongo,"
Man, n.s. 5, no. 1 (1970): 27-38; "Corporation Theory and Political
Change," in S. Newman, ed, Small States and Segmented Societies (New
York, 1976), pp. 121-138; "Economic and Social Dimensions of Kongo
Slavery," in S. Meirs and I. Kopytoff, eds., Slavery in Africa: An Historical
and Anthropological Perspective (Madison, 1977), pp. 235-257.
Chapter 2
1. H. Koch, Magie et chasse au Cameroun (Paris, 1968).
2. C. Droux and H. Kelly, "Récherches préhistoriques dans la région
de Boko-Songho et à Pointe-Noire (Moyen-Congo)," Journal de la société
des Africanistes 9 (1939): 71-87.
3. P. Martin, The External Trade of the Loango Coast, 1576-1870
(Oxford, 1972), p. 57.
4. Globus (1876), p. 307. Ambriz alone exported 185 tons of ivory in
1874, at 150 pounds per tusk, which represented 1295 elephants that year.
Notes 335
5. E. Dupont, Lettres sur le Congo (Paris, 1889), p. 340.

6. Ibid., p. 345.
7. H.M. Stanley, Through the Dark Continent (London, 1877), p.
182.
8. Kwamba, Text 1 (Chapter 4, below), line 68.
9. Dupont, Lettres sur le Congo, pp. 338-339.
10. J. Vansina, The Tio Kingdom of the Middle Congo, 1880-1892
(London, 1973), p. 326.
11. Martin, The External Trade, p. 57.
12. Ibid., p. 59.
13. Vansina, The Tio Kingdom, p. 282.
14. Ibid., p. 326.
15. Ibid., p. 287.
17. Ibid., p. 94.
18. Ibid., p. 86.
19. Ibid., pp. 113, 107.
20. Ibid., p. 107.
2 1 . Vansina, The Tio Kingdom, p. 448.
22. Martin, The External Trade, pp. 113-114, 132. See also W.
Randies, L'ancien royaume du Congo (Paris, 1968), p. 203, for similar
figures.
23. Vansina, The Tio Kingdom, p. 304; P. Bohannan, "The Impact of
Money on an African Subsistence Economy," Journal of Economic History
19, no. 4 (1959): 491-503.
24. Vansina, The Tio Kingdom, pp. 288-292.
25. M. Soret, Les Kongo Nord-Occidentaux (Paris, 1959); G.P.
Murdock, Africa, Its People and theirCulture History (New'York, 1959); J.
Maes and O. Boone, Les peuplades du Congo belge (Bruxelles, 1935).
26. Stanley, Through the Dark Continent, p. 183.
28. Dupont, Lettres sur le Congo, pp. 330-340.
29. The intellectual source of this simplistic notion of homogeneous
ethnicity is the "culture-area" {Kulturkreis) school inspired by Gräbener,
Wissler, Frobenius, and perhaps Murdock. Lacking in understanding of the
dynamics of social organization, the adherents of this approach map out
cultural units on the basis of features such as language, dialect, or material
culture affixing them to a given territory. By definition, a culture area defined
by its most "characteristic" elements or complex of elements is therefore
"pure." An improvement in the Africanist literature is the work of Baumann
and Westermann of the so-called "historical" school. In their maps and
ethnographies, one can note the superimposition of one identity over another,
the accretion within a population of successive historical trends and polities.
See H. Baumann and D. Westermann, Les peuples et civilisations de
VAfrique (Paris, 1967).
336 NOTES
30. H. Deleval, "Les tribus Kavati du Mayombe," La Revue Congo-

laise 3 (1912): 34.
31. E. de Cleene, "Les chefs indigènes au Mayombe," Africa 8, no. 1
(1935): 63-75; Deleval, "Les tribus Kavati," p. 33; L. Bittremieux, "Un
vieux chef," Missions en Chine, au Congo, et aux Philippines 6 (1909):
121-123.
32. Vansina, The Tio Kingdom, on the Tio ndzo.
3 3. P.-P. Rey, " L'esclavage lignager chez les tsangui, les punu et les kuni
du Congo-Brazzaville," in C. Meillassoux, éd., L'esclavage en Afrique
précoloniale (Paris, 1975), pp. 509-528. Rey's approach to defining the
levels of Kunyi organization concentrates on the large territorial estate
(mukuna) comparable to the nsi. However, because he is interested mainly
in the visible unit of production, the lineage (diku), ignoring other levels of
categorial classification, he is led to make the statement that the clan
(ifumba) may span ethnic boundaries of such peoples as the Kunyi, Tsangui,
and Punu which are therefore presumably territorial and discrete. This
approach stops short of solving problems inherent in the view of one ethnic
unit seen as territorially discrete.
34. W. MacGaffey, "Economic and Social Dimensions of Kongo Slav
ery," in S. Meirs andl. Kopytoff, eds., Slavery in Africa: An Historical and
Anthropological Perspective (Madison, 1977), p. 243.
35. M. Gluckman, The Ideas in Barotse Jurisprudence (New Haven,
1965), pp. 88-90.
36. Foremost is C. Levi-Strauss's classic, Les structures élémentaires
de la parentée (Paris, 1949).
37. De Cleene, "Les chefs indigènes," pp. 63-75; Deleval, "Les tribus
Kavati," pp. 34-36.
38. See for example Martin, The External Trade, pp. 3-5, who reviews
the sources.
39. A. Battel, "The Strange Adventures of Andrew Battel," in J.
Pinkerton, ed.,^4 General Collection of the Best and Most Interesting
Voyages and Travels in all Parts of the World (London, 1814), p. 331.
40. J. Cuvelier, "Traditions Congolaises," Congo 10, no. 1 (1931):
203.
41. Ibid.
42. Battel, quoted in Martin, The External Trade, p. 19.
43. A. Bastían, Die Deutsche Expedition an der Loango-Kuste, 2 vols.
(Jena, 1874), p. 238.
44. C. Tastevin, "Idées religieuses des indigènes de l'enclave de
Cabinda," Etudes missionaires (1935), pp. 105-111, 191-197, 257-273.
45. Martin, The External Trade, pp. 24-25, 162.
46. Bastian, Die Deutsche Expedition, p. 327.
47. F. Hagenbucher-Sacripanti, Les fondaments spirituels du pouvoir
au royaume de Loango (Paris, 1973), p. 21.
Notes 337
49. R Visser, "Der Kaufmann unter Wilden (Entwicklung des Handels

am Kongo)," Jahresbericht des Naturwissenschaftlichen Vereins zu Kre-
feld (1907-8), pp. 59-60.
50. Delevai, "Les tribus Kavati," p. 33.
51. De Cleene, "Les chefs indigènes."
52. L. Bittremieux, Mayombish Idioticon (Gent, 1923), map.
53. De Cleene, "Les chefs indigènes"; A. Doutreloux, L'ombre des
fétiches (Louvain, 1967), pp. 176-208.
54. Bittremieux, "Un vieux chef."
56. A. Proyart, "Histoire de Loango, Kakongo, et autres royaumes
d'Afrique," in Pinkerton, ,4 General Collection of... Voyages, p. 579.
57. Hyacinthe de Bologne, La pratique missionaire des PP. Capucins
Italiens dans les royaumes de Congo, Angola et contrées adjacents
(Louvain, 1931), pp. 102-103.
58. Ibid., pp. 109-110.
59. Ibid., p. 118.
60. Suggested by L. Bittremieux in his Société secrète des Bakhimba
(Bruxelles, 1936), p. 187.
61. Deleval, "Les tribus Kavati " p. 147.
62. O. Dapper, Umständliche undEigenliche Beschreibung von Afrika
(Amsterdam, 1670), p. 536.
63. Ibid., pp. 534-535.
64. Ibid., pp. 535-537.
65. In addition to Bastian, Die Deutsche Expedition, the Loango expe
dition was published in P. Güssfeldt, J. Falkenstein, and E. Pechuel-
Loesche, Die Loango Expedition 1873-6 (Leipzig, 1879), and in E.
Pechuel-Loesche, Volkskunde von Loango (Stuttgart, 1907).
66. Pechuel-Loesche, Volkskunde von Loango, p. 380.
67. Ibid., pp. 383-384.
68. Ibid., p. 382.
69. Bastian, Die Deutsche Expedition, pp. 162-163.
70. Ibid., p. 163; Pechuel-Loesche, Volkskunde von Loango, p. 381.
72. R Lehuard, Les phemba du Mayombe (Paris, 1977).
73. Bastian, Die Deutsche Expedition, p. 164.
75. Ibid., p. 377.
76. Güssfeldt, et al., Die Loango Expedition, p. 71.
77. Bastian, Die Deutsche Expedition, pp. 170-173.
78. Vansina, The Tio Kingdom, pp. 439-443.
79. Ibid., pp. 323-324.
80. Ibid., p. 330.
81. Ibid., p. 331.
82. Ibid., p. 332.
338 NOTES
83. Ibid., p. 396.

84. Ibid., p. 384.
85. Ibid., p. 456.
86. Ibid.
87. Ibid., p. 324.
88. Ibid., p. 326.
89. J. Van Wing, Etudes Bakongo (Bruxelles, 1959), Map I.
90. Ibid.; J. Vansina, Kingdoms of the Savanna (Madison, 1965), p.
236; J. Cuvelier, L'ancien royaume de Congo (Bruxelles, 1946), Map I.
91. Vansina, Kingdoms of the Savanna, Chapter 3.
92. Montesarchio, in Cuvelier, L'ancien royaume de Congo, pp. 298-
299.
93. Ibid., Map I, and Montesarchio's list, p. 301.
94. K. Laman, The Kongo II (Uppsala, 1957), pp. 140-142.
95. Ibid.
96. W. MacGaffey, Custom and Government in the Lower Congo
(Berkeley, 1970).
97. Cuvelier, L'ancien royaume de Congo, p. 304.
98. Vercraeye, Dossier No. 129, Luozi Territorial Archives, 31 Aug.
1932.
99. Bittremieux, Mayombish Idioticon; K. Laman, The Kongo I
(Uppsala, 1953), Chapter 1; Maillet, Luozi Territorial Archives, n.d.
100. Maillet, Luozi Territorial Archives.
101. Ibid.
102. Ibid.
103. Ibid.
104. Rapport annuel Affaires Indigènes et Main d'Oeuvre, "Etat d'esprit
de la population," Luozi, 1938.
105. J. Munzele, Bakulu Beto ye Diela Diau (Kumba, 1965), p. 12.
106. Pierre Nzuzi of the Kimbanga clan of Kisiasia, Luozi Territory,
narrated this to me in 1969.
107. Munzele, Bakulu Beto, p. 19.
Chapter 3
1. E. Dupont, Lettres sur le Congo (Paris, 1889), pp. 330-340.
2. C. Coquery-Vidrovitch, Le Congo au temps des grandes com-
pagnies concessionnaires, 1898-1930 (Paris, 1972), p. 187.
3. O. Stenstrôm, "The Lemba Cuit," Ethnos (1969), pp. 1-4.
4. Reports, Luozi Territorial Archives, 1930.
5. L. Bittremieux, Van Een Ouden Blinden Hoofdman (Antwerpen,
1925).
6. J. Ndibu, Notebook 345, Laman Collection, Svenska Missionsfor-
bundet, Lidingô, n.d.
Notes 339
7. For example the introduction to J.M. Janzen and W. MacGafFey,

Anthology of Kongo Religion (Lawrence, 1974); and Janzen, "Deep
Thought: Structure and Intention in Kongo Prophetism, 1910-21," Social
Research 46, no. 1 (1979): 106-139.
Introduction to Part II
1. V.W. Turner, "Colour Classification in Ndembu Ritual," in M.
Banton, ed., Anthropological Approaches to the Study of Religion (New
York, 1966), pp. 47-84; A. Jacobson-Widding, Red-White-Black as a
Mode of Thought (Uppsala, 1979).
2. For present purposes, V.W. Turner's Drums of Affliction (Oxford,
1968) is the leading example of this approach in the area of African thera
peutics. More generally, C. Geertz's work on religion as a symbol system
characterizes it, as for example in his "Religion as a Cultural System" in
Banton, Anthropological Approaches to... Religion, pp. 1-46. This ap
proach is given historical depth and critical justification in J.L. Dolgin, D.S.
Kemnitzer, andD.M. Schneider, eds., Symbolic Anthropology (New York,
1977).
3. C. Geertz, "Deep Play: Notes on the Balinese Cockfight," Dae-
dalus 101, no. 1 (1972): 1-38.
4. C. Geertz, The Social History of an Indonesian Town (Cambridge,
1965), pp. 153-202.
5. V.W. Turner, Revelation and Divination in Ndembu Ritual (Ith
aca, 1975).
6. V.W. Turner, Dramas, Fields, and Metaphors: Symbolic Action in
Human Society (Ithaca, 1974).
7. See Turner, Drums of Affliction, and accounts of the Isoma and
Wubwang'u rituals among the Ndembu in Turner, The Ritual Process
(Chicago, 1969).
8. Turner, "Symbols in African Ritual," in Dolgin, et al., Symbolic
Anthropology, pp. 183-194.
9. Exemplified in R.P. Armstrong, The Affecting Presence (Urbana,
1971); but see also J. Fernandez, "Persuasions and Performances: Of the
Beast in Every Body . . . and the Metaphors of Everyman," Daedalus 101,
no. 1 (1972): 39-60.
10. I am thinking primarily of the work of C. Lévi-Strauss, La pensée
sauvage (Paris, 1962); Mythologiques, Z-JTF (Paris, 1964-71); and that of
P. Maranda and E. Kòngàs-Maranda, eds., Structural Analysis of Oral
Tradition (Philadelphia, 1971).
11. D. Sperber, Rethinking Symbolism (Cambridge, 1974); also
Fernandez, "Persuasions and Performances."
12. Primarily I have in mind the semiotics of R Barthes as formulated
in Système de la mode (Paris, 1967), or his literary essays such as S/Z
340 NOTES
(Paris, 1970), with its emphasis on levels or codes, and the relationship
between these codes, although I am aware of the theoretical sophistication of
C.S. Peirce's much earlier work.
13. G. Bateson, Steps to an Ecology ofMind (New York, 1972), and E.
Leach, Culture and Communication: The Logic by which Symbols are
Co nnected (Cambridge, 1976), are two examples of the kind of work I have in
mind.
14. An issue raised by most of the above writers and resolved or
formulated in a great diversity of ways too involved for further discussion in
this study.
Chapter 4
1. Text 1 by E. Kwamba is taken from Notebooks 142 and 143, dated
about 1910, of the Catechists' Cahiers section of Karl Laman's collection in
the Svenska Missionsforbundet Archives (SMF), Lidingo, Sweden. Parts
of the full text presented here have appeared in English in Laman, The Kongo
III (Uppsala, 1962), pp. 113-116, and in J.M. Janzen and W. MacGaffey,
An Anthology of Kongo Religion (Lawrence, 1974), pp. 97-102.
2. Text 2 by M. Lunungu is drawn from Notebook 181 of the Laman
Collection, SMF, Lidingo, and may be dated at about 1915.
3. I have elsewhere discussed the origin and fate of this extensive
KiKongo corpus: see my'' Laman's Kongo Ethnography: Observations on
Sources, Methodology, and Theory," Africa 42, no. 4 (1972): 316-328.
When Laman returned to Sweden in 1919 with the notebooks, they provided
him with the basis in idioms and vocabulary to produce his masterful
Dictionnaire KiKongo-Francais (Bruxelles, 1936). He also translated (into
Swedish) passages on Kongo custom using categories of his original ques
tionnaire, selecting from the notebooks what he regarded as most represen
tative and best written. Posthumously, Laman's Swedish text was translated
into English, under the direction of S. Lagerkrantz, and published in the
Studia Ethnographica Uppsaliensia series as The Kongo I-IV (Uppsala,
1953, 1957, 1962, 1968). Fortunately scholars of Kongo and Central-
African studies have had access to these English sources although they have
been difficult to work with because of the lack of reference in them to place,
context, and authorship. Also, some of the materials having gone through
double translation have lost their original meaning or have become very
elliptical. The reader may wish to compare Kwamba's original text on
Lemba given here with the version offered in Laman, The Kongo III, pp.
113-116.
Lunungu's description of Lemba (Text 2) has not been published anywhere
to my knowledge.
4. R. Rappaport, Pigs for the Ancestors (New Haven, 1968).
5. For such a rewriting, used as partial model in this work, see A.
Dundes, E. Leach, P. Maranda, andD. Maybury-Lewis, "An Experiment:
Suggestions and Queries from the Desk, with a Reply from the Ethnogra-
Notes 341
pher," in P. Maranda andE. Köngäs-Maranda, eds., Structural Analysis of

Oral Tradition (Philadelphia, 1971), pp. 292-324.
Chapter 5
1. Kimbembe, Notebook 80, Laman Collection, Svenska Missions-
forbundet(SMF), Lidingö, ca. 1915.
2. O. Stenström, "The Lemba Cult," Ethnos (1969), pp. 37-57.
3. E. Andersson, Contribution à l'ethnographie desKuta / (Uppsala,
1953), pp. 247-249.
4. J. Malonga, "La sorcellerie et l'ordre du Lemba chez les Lari,"
Liaison 62 (1958): 45-49; 63 (1958): 51-61.
5. Ibid., p. 53.
6. Ibid.
7. Ibid., p. 59.
8. Ibid., p. 60.
9. Ibid., p. 54.
10. J.M. Janzen, The Quest for Therapy in Lower Zaire (Berkeley,
1978), pp. 196-203.
11. Malonga, "La sorcellerie et l'ordre," p. 61.
12. M. Douglas, Purity and Danger (New York, 1966).
13. Stenström, "The Lemba Cuit," p. 38.
14. Ibid., p. 41.
15. Malonga, "La sorcellerie et l'ordre," pp. 54, 61; also, V. Turner,
"Colour Classification in Ndembu Ritual," in M. Banton, ed., Anthropolo-
gical Approaches to the Study of Religion (New York, 1966), pp. 47-84;
Fukiau, N'kongoyeNza/Cosmogonie-Kongo(Kinshasa., 1969), pp. 35-39.
16. Stenström, "The Lemba Cult," p. 41.
18. Ibid., p. 55.
19. Ibid., p. 56.
20. Stenström, "The Lemba Cult," p. 44.
21. Malonga, "La sorcellerie et l'ordre," pp. 59-60.
22. Andersson, Contribution, pp. 247-248, based on notes taken 1930-
31 at Madzia from Milonga Jean and Muyakidi, at Musana from Masamba
Philippe and Ngusa Ruben, a Lali. It is noteworthy that this origin account
traces Lemba to a founder with the "Nga" title, similar to most of the
Teke nkobi holders.
23. Stenström, "The Lemba Cult"
24. Ibid.
25. Ibid.
Chapter 6
1. Fukiau, N'kongo ye Nza/Cosmogonie-Kongo (Kinshasa, 1969).
2. Batsikama ba Mampuya, "A propos de i a cosmogonie Kongo,' "
Cultures au Zaire et en Afrique 4 (1974): 239-264.
342 NOTES
3. Solomo Nitu, noted storyteller and catechist of Cabinda, narrated

these Mavungu texts (5, 6, 7) to Charles Harvey in 1961 at the age of 75
years. I am indebted to Harvey for recording Nitu's narrative treasure as well
as for the generosity of allowing me to use them here. Nitu lived in the village
of Masala, near the Portuguese administrative post of Luadi, also near the
mission of M'boku, on the banks of the Mbika River, in the so-called Zala
area, in tall forest country.
4. Fukiau, N'kongo ye Nza, pp. 41-56.
5. Ibid., p. 47, my translation.
6. Ibid., p. 46.
7. Ibid., p. 50.
8. Ibid., p. 43.
9. Ibid.
10. Ibid.
11. Ibid., pp. 43, 49.
12. K. Laman, Dictionnaire KiKongo-Francais (Bruxelles, 1936).
13. Fukiau, Nkongo ye Nza, pp. 18-21.
14. Batsikama, "A propos... Kongo," pp. 250-252.
15. Fukiau, N'kongo ye Nza, pp. 19-21.
16. Nitu, Text XXIIIa, MS.
17. Nitu, Text XXXIX, MS.
18. Nitu, Text XLIV, MS.
19. J. Malonga, La legènde de M'pfoumou ma Mazono (Paris, 1954).
20. C. Lévi-Strauss, "The Structural Study of Myth," Journal of Ameri-
can Folklore 67 (1955): 428-444.
Chapter 7
1. J. Konda, Notebook 119, Laman Collection, Svenska Missionsfor-
bundet(SMF), Lidingö, ca. 1915.
2. T. Babutidi, Notebook 16, Laman Collection, SMF, Lidingö, ca.
1915. This account of a Lemba initiation is identified as having occurred at
Mamundi, "westward of Kinkenge." I have been unable to locate this site
exactly, but Kinkenge is at the boundary of the Yombe area, therefore I have
identified it as Eastern Mayombe. Parts of this text have been published in K.
Laman, The Kongo III (Uppsala, 1962), p. 116.
3. L. Bittremieux, Mayombsche Namen (Leuwen, 1934), pp. 41-42.
4. Ibid., p. 41.
5. Paul Güssfeldt, Julius Falkenstein, and Eduard Pechuel-Loesche,
Die Loango Expedition, 1873-6 (Leipzig, 1879), p. 71.
6. A. Bastian, Die Deutsche Expedition an der Loango-Küste, 2 vols.
(Jena, 1874), pp. 167, 169.
7. L. Bittremieux, Van Een Ouden Blinden Hoofdman (Antwerpen,
1925), pp. 31-34.
8. Based on Belgian Colonial Ministry economic reports on the Bas-
Notes 343
Congo, in Congo (1925), II, pp. 651-676; (1928), p. 691; (1929), II, pp.
714-726; (1932), II, pp. 293-313.
9. Bastian, Die Deutschen Expedition, vol. I, p. 258.
10. Bittremieux, Mayombsche Namen, pp. 42-44.
11. A. Jacobson-Widding, in hex Red-White-Black as a Mode of
Thought (Uppsala, 1979), pp. 250-251, makes much of the discrepancy
betweenthe"white"nameMpemba Lemba andthe"red" tukula ingredients
in Babutidi's report, from which she is working in Laman's Kongo ethnog
raphy, possibly without realizing it. She relates this to her theory of the need
to reconcile what she speaks of as the "inner" with the " outer" man, and with
the "matrilineal principle of inheritance of male characteristics." Unfortu
nately, this elaborate theory is based on the perpetuation of a simple phonetic
error by Babutidi who, not being a.Lemba initiate, confused/?/èmZ>a or
phemba for mpemba. The same mistake was made by Lehuard, and before
him by Maes, and before him by Bastian and Pechuel-Loesche. Bittremieux's
research ( see note 12 below) and circumstantial evidence in Lehuard's mono
graph and in Pechuel-Loesche's account of n'kisi Phemba permit a correct
interpretation.
12. Bittremieux's original research on the distinction oîpfemba (or
phemba) and mpemba is reported in unpublished letters written in 1939 to
the then director of the Musée d'Afrique Central, at Tervuren, J. Maes, from
Kangu where he had collected two nkobe Lemba, including the one sketched
in figure 21 and pictured in plates 5-7.
13. E. Pechuel-Loesch, Volkskunde von Loango (Stuttgart, 1907), p.
385.
14. R. Lehuard, Les phemba du mayombe (Paris, 1977), is an excellent
presentation of this sculptural genre, except for the confusion of mpemba
with phemba, which I have discussed at greater length in my review of this
monograph m African Arts 11, no. 2 (1978): 88-89.
15. J.M. Vaz, Filosofia Tradicional dos Cabindas (Lisboa, 1969).
Figures 30, 103, 140, 142, 143, 160, 187-A, 199, 216, 230, and 242
show Lemba as a drum in the company of other min 'kisi, includingMbondo-
Fula, Mbonzo, Mikono, N'kobe-Ibingu, Nkwangi, Koko.
16. Lehuard, Les phemba du mayombe, p. 87.
17. Vaz, Filosofia Tradicional, fig. 187. See also L. Bittremieux,
Symbolisme in de Negerkunst (Brussel, 1937), Object 112, for a discussion
on Mbondo-Fula.
18. W. Dionga, narrator of "Tsimona-Mambu, de Wonderziener of de
Oorsprong van het huwelijk bij Dilemba, naar een Mayombsche légende," in
L. Bittremieux, Congo 2 (1926): 398-404; 551-6. English translation
of this tale appears in J.M. Janzen and W. MacGaffey, An Anthology of
Kongo Religion (Lawrence, 1974), pp. 102-106.
19. This summary is based on J. Van Wing and C. Schôller, "Les
aventures merveilleuses de Moni-Mambu le querelleur," in their Légendes
des BaKongo orientaux (Louvain, 1940), pp. 11-44, which is a heavily
344 NOTES
edited Kongo narrative for school children based on an earlier text published
in KiKongo, circulated in ca. 1935, itself transcribed from an unknown
narrator. I have not been able to find the KiKongo original, but it was retold
by A.-R. Bolamba as "La légende de Moni-Mambu chez les BaKongo,"
in Arts et metiers indigènes 9 ( 193 8): 17-19.1 have added Bolamba's episode
8 to the Van Wing and Schöller version since it was apparently in the original
KiKongo. Other accounts, much shorter, were given by Jules Benga and
Pierre Ndakivangi, as "Mumboni-a-Mpasi, celui qui avait beaucoup de
palabres," Arts et metier indigènes 9(1938): 20-21.
20. For a general analysis of the genre and an anthology of examples, see
W. Bascom, ed., African Dilemma Tales (The Hague, 1975). Locally, in
North Kongo, they are called ngana zakindembikisa, a collection of which is
available in J. Bahelele, Kinzonzi ye ntekolo'andi Makundu (Matadi,
1961), pp. 36-38.
21. P. Radin, The Trickster: A Study in American Indian Mythology
(New York, 1956).
22. L. Makarius, "Ritual Clowns and Symbolic Behavior," Diogenes 69
(1970): 67; also, R.D. Pelton, The Trickster in West Africa (Berkeley,
1980), offers a similar but more extensive analysis of the trickster in four
West-African traditions.
Chapter 8
1. A. Ramos, O Negro Brasileiro (Rio de Janeiro, 1934), p. 106,
partial translation by Roberto Fiorillo. The reference to the priest's smoking
recalls Lemba pipes from Loango and reference to smoking in the Mavungu
legends (Texts 7-8).
2. Ibid., Chapter 4, "Os cultos de procedencia Bantu," pp. 99-135.
3. M. Herskovits, The Myth of the Negro Past (Boston, 1958), p. 84.
4. R. Bastide, African Religions of Brazil (Baltimore, 1978), p. 196.
5. E. Carneiro, Negros Bantus (Rio de Janeiro, 1937).
6. In addition to Herskovits, Myth of the Negro Past, see his Daho-
mey: An Ancient West African Kingdom (New York, 1938); and Life in a
Haitian Village (New York, 1937).
7. R. Bastide, LecandomblédeBahia(riteNagô)(TheH2ig\ie, 1958);
and Bastide, African Religions of Brazil.
8. R F . Thompson, "Transatlantic African Art Traditions," MSS 1975,
1977; and Thompson, "The Flash of the Spirit: Haiti's Africanizing Vodun
Art," in Ute Stebich, ed., Haitian Art (Brooklyn, 1978), pp. 26-37.
9. R.F. Thompson, Black Gods and Kings (Los Angeles, 1971).
10. Bastide, African Religions of Brazil, p. 195; O. Dapper,
Umständliche und Eigenliche Beschreibung von Afrika (Amsterdam,
1670), pp. 534-537.
11. Thompson, "The Flash of Spirit," p. 26.
12. A. Métraux, Black Peasants and Voodoo (New York, 1960).
13. Ibid., p. 14.
Notes 345
14. J. Price-Mars, "Lemba-Pétro, un cult secret," Revue de la société

haïtienne d'histoire, de géographie, et de géologie 9, no. 28 (1938): 12-31.
15. Ibid., pp. 18-19.
16. Ibid., p. 19.
17. Ibid., pp. 19-20.
18. Ibid.
19. H. Chatelein, Folk-tales ofAngola (Boston and New York, 1894),
p. 11.
20. A. Métraux, Voodoo in Haiti (New York, 1959), p. 86.
21. Thompson, "The Flash of Spirit," pp. 26-27.
22. Herskovits, Life in a Haitian Village, pp. 154-176.
23. Ibid., pp. 164-165.
24. It would be tempting, both on grounds of terminological similarity
and of functional resemblance, to suggest an association between Bosu of the
Haitian rite and Boessi-Batta of the Loango coast and Cabinda cosmologies
and rites (see Chapter 2). The latter included components of the domestic
realm and the wild or the beyond, as represented in the trade. There is also in
Boessi-Bata's ritual, as reported in Dapper, Beschreibung von Afrika, pp.
534-537, a divinatory process which identifies the subject's needs through
the ecstasy of the officiating priest. This same dualism of the domestic realm
and of the beyond is present in nineteenth-century Cabinda figures M'boze,
female domestic personage, and Beti-Randa, male, keeper of the wilds, the
two being mediated by Kuiti-Kuiti (also known as Nzambi-Mpungu), as
recorded by C. Tastevin, "Idées religieuses des indigènes de l'enclave de
Cabinda," Etudes missionaires 3 (1935), no. 1, 105-111; no. 2, 191-197.
However, if Bosu is to be identified with Boessi-Bata or M'boze it would
mean either some resemblance between the Fongbe(Dahomean) and Kongo
deities, or a conjuncture in Haitian thought of the two. What is striking is
Bosu's presence in the Pétro (or Congo) set of deities in the Haitian rite.
25. Price-Mars, "Lemba-Pétro," pp. 12-31.
26. This translation of Price-Mars's 1938 text is published in Métraux,
Voodoo in Haiti, p. 307.
27. An interpretation given by Métraux, Voodoo in Haiti, p. 307.
Chapter 9
1. C. Kerenyi,"Prolegomena," inC. Jung andC. Kerenyi, eds.,Essays
on a Science of Mythology (Princeton, 1969), pp. 1-24.
2. W. MacGaffey, "The Religious Commissions of the BaKongo,"
Man n.s. 5, no. 1 (1970): 27-38.
3. A more extended discussion of this process in Kongo culture
structure is found in J.M. Janzen and W. MacGaffey, Anthology of Kongo
Religion (Lawrence, 1974), pp. 87-89.
4. L. Frobenius, Die Masken und Geheimbünden (Halle, 1898), Part
II.
346 NOTES
5. K. Laman, The Kongo III (Uppsala, 1962), p. 67.

6. L. Bittremieux, La société sécrète des Bakhimba (Bruxelles, 1936),
p. 67.
7. J.-F. Thiel, Ahnen, Geister, Höchste Wesen (St Augustin, 1977).
8. E. Lutete, Notebook 229, Laman Collection, Svenska Missions-
9. Ibid.
10. Ibid.
11. C. Tastevin, "Idées religieuses des indigènes de l'enclave de
Cabinda," Etudes missionaires 3 (1935), no. 1, 105-111; no. 2,191-197;
no. 3, 257-273. For a shortened version of the Woyo origin myth see J.
Knappert, "The Water Spirits," in his Myths and Legends of the Congo
(Nairobi and London, 1971), pp. 138-139.
12. Especially the distinction between "opposition" and "contradic
tion" as indicated by K. Burridge, "Lévi-Strauss and Myth," in E. Leach,
ed., The Structural Study of Myth (London, 1967), pp. 91-118.
13. In the sense understood by R. Girard, La violence et le sacré (Paris,
1972).
14. G. Boas, "The Evolution of the Tragic Hero," in R. W. Corrigan, ed.,
Tragedy: Vision and Form (San Francisco, 1965), pp. 117-131.
15. Corrigan, Tragedy, p. ix.
16. B. Dadié, personal communications, 1971.
17. "L'histoire de Kuba-Ntu" in J. Van Wing and C. Schöller, eds.,
Légendes des BaKongo orientaux (Louvain, 1940), pp. 53-60.
18. G. Bateson, Steps to an Ecology of Mind (New York, 1972).
19. F. Nietzsche, "The Birth of Tragedy," in Corrigan, Tragedy, p. 448.
20. C. Lévi-Strauss, "The Structural Study of Myth," Journal of
American Folklore 68 (1955): 428-444.
21. Burridge, "Lévi-Strauss and Myth."
22. C. Lévi-Strauss, "Finale," in his L'Homme Nu (Paris, 1971), pp.
559-621.
23. J. Lacan, "The Insistence of the Letter in the Unconscious," in R
and F. DeGeorge, eds., The Structuralists: From Marx to Lévi-Strauss
(Garden City, 1972), pp. 287-323.
24. Lacan' s point of reference here is Freud' s notion of Einstellung from
the "Interpretation of Dreams," a term which has also been translated as
"distortion" or "misrepresentation."
25. Freud's Verdichtung, a bringing together of several signifiers.
26. Freud's Verschiebung.
Chapter 10
1. I. Nsemi, "Min'kisi: Sacred Medicines," in J.M. Janzen and W.

MacGaffey, eds., Anthology ofKongo Religion (Lawrence, 1974), pp. 34-
38.
Notes 347
2. E. Andersson, L*Ethnographie desKuta (Uppsala, 1953), pp. 247-

249; A. Bastian, Die Deutsche Expedition an der Loango-Küste (Jena,
1874); H. Châtelain, Folktales of Angola (Boston and New York, 1894), p.
11.
3. O. Stenstrom, "The Lemba Cuit," Ethnos (1969), p. 37.
4. ZablonMakunza,Kinganga village, inDe la Kenge,LuoziTerritory.
5. F. Ngoma, L'initiation BaKongo et sa signification (Lubumbashi,
1963), p. 120.
6. J. Malonga, "La sorcellerie et Tordre du 'Lemba' chez les Lan,"
Liaison 63 (1958): 53-54; Batsikama ba Mampuya, "A propos de 'La
cosmogonie Kongo,' " Cultures au Zaire et en Afrique 4 (1974): 257.
7. J. Munzele, Bakulu beto yeDiela diau (Kumba, 1965), p. 19.
8. L. Bittremieux, Mayombsche Namen (Leuwen, 1934), pp. 42-44.
9. Batsikama ba Mampuya, "A propos . . . Kongo," p. 257.
10. W.H. Bentley, Pioneering on the Congo (London, 1900), p. 290.
Chapter 11
1. J. Vansina, in his Children ofWoot: A History ofthe Kuba Peoples
(Madison, 1978), p. 209, offers a negative assessment of the possibility of
retrieving an intellectual history for much of precolonial Africa. For a more
optimistic appraisal of the situation and an attempt at an intellectual history of
a precolonial setting through the use of therapeutic central values in Lozi
history, see G. Prins, Hidden Hippopotamus (Cambridge, 1980), pp. 150-
157.
2. C. Meier, Die Entstehung des Politischen bei den Griechen (Frank
furt, 1980), pp. 154-158.
Sources
The Reconstruction of Lemba

During fieldwork campaigns in the Manianga (Territory of Luozi) in 1964-
1966 and 1969 I began to perceive Lemba's historic importance to the
region. I pursued some inquiries into the context of its political activities and
the role of its priests and priestesses in marketing, conflict resolution, and
other judicial and economic activities. Those Lemba priests and priestesses
initiated before 1920 who were willing to speak to me and whose memories
were lucid were of inestimable help in reconstructing the local nature of
Lemba in the Kivunda district. Of these, special mention must be made of
Katula Davidi of Nseke Mbanza. Nzuzi Pierre of Kisiasia and Munzele
Yacobi of Kumbi although too young to have been initiated were both very
helpful in explaining Lemba's role in their communities. Fukiau was a
valuable guide in technical KiKongo discussions with these people.
Published and unpublished accounts from various specialized sources have
been used to reconstruct Lower-Congo society and Lemba's role within it I
have combed many historical chronicles and ethnographies for explicit or
circumstantial details on Lemba. Mention of the term "lemba" does not
always mean Lemba the drum of affliction, since the name is used in a
number of towns and villages. As far as I can determine, there is no historic
connection between Lemba the drum of affliction and the use of lemba as the
title for a chiefly order in Bandundu and Kasai societies, although lemba
does appear as a Western Bantu cognate in Guthrie's Proto-Bantu word list
Fortunately, Lemba is not a clan name in Lower-Congo societies. First
mention of Lemba as n'kisi appears in print in Dapper's 1668 Beschreibung
of Loango. Thereafter allusions appear in reports by traders, missionaries,
and other travellers and eventually in writings by Kongo authors. These have
been consulted in numerous specialized libraries and archives. Such writings
as those by the missionaries Hyacinthe de Bolonge and Montesarchio and
other sources published by the Catholic church I have located mainly in the
Jesuit Institute of Bonn and in the Anthropos Institute of St Augustine. The
German Loango coast expedition of the 1870's organized and carried out by
Adolph Bastian offers the first professional ethnographies of Congo-Basin
societies, including some perceptive passages on Lemba. These are all in
published works, but unfortunately I was not able to locate any unpublished
documents of the Loango expedition, except for some pertaining to material
culture collections in the Berlin-Dahlem Museum of Ethnography, which will
be detailed below. Neither the Catholic missionaries nor the Loango expedi
tion personnel link Lemba to trade. Personal correspondence of the
merchant Robert Visser in the Linden Museum, Stuttgart, offers some light on
this trade, although Lemba is not mentioned explicitly by him. On other
349
350 SOURCES
aspects of Lemba, missionaries, both Catholic and Protestant, of the turn of

the century and a bit later offer exceptional material on Lemba and its setting.
Taste vin' s work in C abinda I located in the Jesuit Institute of Bonn, and much
of Bittremieux's unpublished work from the Kangu district in Mayombe is
available in the Musée d'Afrique Central at Tervuren. Swedish missionaries'
work and correspondence were consulted either in the Svenska Missions-
forbundet in Lidingô, in the Stockholm Ethnographic Museum, or in the
Gôteburg Ethnographic Museum. These include Laman, Hammar, Westlind,
Bôrrison, Andersson and Stenstrom, all of whom worked in Nsundi, Kamba,
and Lali regions, and who offer excellent although often anecdotal informa
tion on Lemba.
A source of comparable importance in the reconstruction of the setting
of Lemba is the local colonial archival network, particularly those docu
ments resulting from detailed inquiries made in the early decades of this
century by colonial agents seeking to understand the region's history and
basis of legitimacy so as to achieve fuller penetration of the colonial adminis
tration into African society. Although the Luozi Territorial Archives and
offices were burned to the ground by partisans during the interval from Free
State rule to Belgian colonial rule in 1907-1910, later surveys made by
territorial agents and administrators such as Maillet, Vercraeye, and Dele val
are available and were consulted in Luozi, giving accounts, based on inter
views, of inaugurations of precolonial chiefs, clan migrations, and village
compositions. Little mention is given Lemba in these archives. What they
convey is a contempt for "northern" anarchy in which there are no big chiefs.
Repeated efforts were made to identify or create such chiefs, in the process of
which the segmentary social system and its horizontal mechanisms of integra
tion were generally overlooked and misunderstood. One exception to this is
the work of Deleval, whose writings both published and in the archives reflect
remarkable insight. Known locally as mundele si-si-si, "Whiteman Yes-yes-
yes," after the petulant manner in which he answered questions, Deleval's
materials go well beyond the standards of administrative requirements for
information and constitute very helpful scientific accounts.
The most exciting unpublished written sources on Lemba are accounts of
initiation séances and exegetical remarks prepared by Kongo catechists for
Swedish missionary Karl Laman from ca. 1905-1919, before Lemba's
demise. Local accounts culled from the 430 notebooks, used by Laman to
prepare his KiKongo-French dictionary and the four volumes of Kongo
ethnography, provide much of the primary material for the regional variants
of Lemba in Chapters 4-8. These notebooks by largely unheralded Kongo
writers are in the Laman Collection in the Svenska Missionsforbundet
archives in Lidingô, and have been microfilmed for easier access.
The tradition of Kongo writing begun with the catechists continues,
offering further accounts of Lemba by modern writers such as Malonga Jean,
Brazzaville novelist and essayist, Ngoma Ferdinand, Zairian sociologist, and
Mampuya and Fukiau, both active writers. All four have been seriously
Sources 351
concerned with Lemba, the major institution of their recent culture history.
We may add to this list Haiti's well-known ethnologist-physician, Jean Price-
Mars, who has written the only in-depth account, however brief, of Lemba in
the N^w World. These efforts by indigenous writers to describe and interpret
Lemba"s significance provide the base for my own résumé of a conscious
Lemba ideology of therapeutics in the ngoma, drum-of-affliction tradition in
the final part of the book.
A final, and very important, source on Lemba is artifactual. Many of the
aforementioned missionaries, traders, travellers, and colonial officials gath
ered cultural objects and deposited them in African and European museums.
Rarely have Lemba objects been labelled and displayed for what they are.
However in several research trips to Central-European museums I discov
ered extensive holdings of Lemba objects, revealing a mute record of
9
Lemba s historic existence, its geographic distribution, and its integral role in
Congo-Basin social and cultural history. Particular museums consulted are
listed in the inventory of museum objects below.
In sum, I have used all possible sources for the reconstruction of Lemba
and its context. Figure 1 records the geographical distribution of those which
are identified with a specific location. The following lists of sources identify
other nonlocalized objects and documents.
Scholarly References
Andersson, Effraim. Contribution à l'ethnographie des Kuta I. Uppsala:
Studia Ethnographica Upsalensia, VI, 1953
Armstrong, Robert P. The Affecting Presence. Urbana, Chicago, and Lon
don: University of Illinois Press, 1971.
Bahelele, Jacques. Kinzonziye ntekolo andiMakundu. Matadi: Imprimerie
de l'Eglise Evangelique du Manianga et Matadi, 1961.
Banton, Michael, ed. Anthropological Approaches to the Study of Religion.
New York: Praeger, 1966.
Barthes, Roland. "Eléments de sémiologie." Communications 4 (1964):
91-135.
— . Système de la mode. Paris: Editions du Seuil, 1967.
. S/Z. Paris: Editions du Seuil, 1970.
Bascom, William, ed. African Dilemma Tales. The Hague: Mouton, 1975.
Bastian, Adolph. Die Deutsche Expedition an der Loango Küste. 2 vols.
Jena: Costenoble, 1874.
Bastide, Roger. Le candomblé de Bahia (rite Nagô). The Hague: Mouton,
1958.
. African Religions of Brazil. Baltimore: The Johns Hopkins
University Press, 1978.
Bateson, Gregory. Steps to an Ecology of Mind. New York: Ballantine,
1972.
352 SOURCES
Batsikama ba Mampuya "A propos de 'la cosmogonie Kongo.' " Cultures

au Zaire et en Afrique 4 (1974): 239-264.
Battel, Andrew. "The Strange Adventures of Andrew Battel." In ,4 General
Collection of the Best and Most Interesting Voyages and Travels in all
Parts of the World, edited by John Pinkerton. London: Kimber and
Conrad, 1814.
Baumann, Hermann, and Westermann, Dietrich. Les peuples et civilisations
de l'Afrique. Paris: Payot, 1967.
Benga, Jules, and NdaKivangi, Pierre. "Mumboni-a-Mpasi, celui qui avait
beaucoup de palabres," Arts et métiers indigènes 9 (1938): 20-21.
Bentley, William H. Pioneering on the Congo. 2 vols. London: Religious
Tract Society, 1900.
Bittremieux, Leon. "Un vieux chef." Missions en Chine, au Congo, et aux
Philippines 6 (1909): 121-123.
. Van E en Ouden Blinden Hoofdman. Antwerpen: Oude Beurs,
1925.
. "Tsimona-Mambu, le visionaire; ou l'origine du mariage chez
dilemba." Congo 2 (1926): 398-404; 551-561.
. Mayombsch Idioticon. 3 vols. Vols. 1, 2, Gent Erasmus, 1923;
Vol. 3, Brüssel: Essorial, 1927.
. Mayombsch Namen. Leuven: H.H. Harten, 1934.
. La société sécrète des Bakhimba. Bruxelles: Institut Royal
Colonial Beige, 1936.
. Symbolisme in de Negerkunst. Brüssel: Vromant, 1937.
Boas, George. "The Evolution of the Tragic Hero." In Tragedy: Vision and
Form, edited by Robert W. Corrigan, pp. 117-131. San Francisco:
Chandler, 1965.
Bohannan, Paul. "The Impact of Money on an African Subsistence Econ
omy." Journal of Economic History 19, no. 4 (1959): 491-503.
Bolamba, André Roger. "La légende de Moni-Mambu chez les BaKongo."
Arts et métiers indigènes 9 (1938): 17-19.
Burridge, Kenelm O. "Lévi-Strauss and Myth." In The Structural Study of
Myth and Totemism, edited by Edmund Leach, pp. 91-118. London:
Tavistock Publications, 1967.
. New Heaven, New Earth: A Study of Millenarian Activities.
Toronto: Copp Clark, 1969.
Cameiro, Edison. Negros Bantus. Rio de Janeiro: Civilizaçao Brasileira,
1937.
Chatelein, Heli. Folk-tales of Angola. Boston and New York: G.E. Stechert
and Co., 1894.
Chavunduka, G L . "Interaction of Folk and Scientific Beliefs in Shona
Medical Practices." Ph.D. dissertation, University of London, 1972.
Congo. Belgian Colonial Ministry Economic Reports onBas-Congo: ( 1925),
pp. 651-676; (1928), p. 691; (1929), pp. 714-726; (1932), pp. 293-313.
Sources 353
Coquery-Vidrovitch, Cathrine. Le Congo au temps des grandes compagnies

concessionnaires 1898-1930. Paris: Mouton, 1972.
Corrigan, Robert W., ed. Tragedy: Vision and Form. San Francisco:
Chandler, 1965.
Cuvelier, J. "Traditions congolaises/' Congo no. 2 (1930): 469-487; no. 1
(1931): 193-208.
. L'ancien Royaume de Congo. Bruxelles: Desclée de Brouwer,
1946.
Dadié, Bernard. Le pagne noir. Paris: Présence africaine, 1955.
Dapper, Olfert. Umständliche und Eigenliche Beschreibung von Afrika.
Amsterdam: Jacob von Meurs, 1670.
De Cleene, N. "Les chefs indigènes auMayombe." Africa 8, no. 1 (1935):
63-75.
DeCraemer, Willy; Vansina, Jan; and Fox, René. "Religious Movements in
Central Africa." Comparative Studies of Society and History 18(1976):
458+.
DeGeorge, Richard and Fernande, eds. The Structuralists: From Marx to
Lévi-Strauss. Garden City: Anchor Books, 1972.
De Heusch, Luc. Pourquoi Vépouser? Paris: Gallimard, 1971.
DeJonghe, Edouard. "Les sociétés secrètes au Bas-Congo." In Revue des
questionnes scientifiques. Bruxelles: Joseph Polleums, 1907.
. "Les sociétés secrètes en Afrique." Semaine d'ethnologie reli-
gieuse (1923), Ser. 3.
. "Formations récentes de sociétés secrètes au Congo Belge."
Africa 9 (1935): 56-63.
Deleval, Henri. "Les tribus Kavati du Mayombe." La Revue Congolaise 3
(1912): 32-40, 103-113, 170-186, 253-264.
Dionga, William. "Tsimona-Mambu, de Wonderziener of de Oorsprong van
het huwelijk bij Dilemba, naar een Mayombsche legende." Congo 2
(1926): 398-404, 551-556. (See also Leon Bittremieux, "Tsimona-
Mambu, le visionaire," above.)
Dolgin, Janet L.; Kemnitzer, David S.; and Schneider, David M., eds.
Symbolic Anthropology: A Reader in the Study of Symbols and Mean-
ings. New York: Columbia University Press, 1977.
Douglas, Mary. Purity and Danger. New York Praeger, 1966.
Doutreloux, Albert. L'ombre des fétiches: société et culture y ombe. Louvain:
Nauwelaerts, 1967.
Droux, C , and Kelly, H. "Recherches préhistoriques dans la région de
Boko-Songho et à Pointe-Noire (Moyen-Congo)." Journal de la société
des Africanistes 9 (1939): 71-87.
Dundes, Alan; Leach, Edmund; Maranda, Pierre; and Maybury-Lewis,
David. "An Experiment: Suggestions and Queries from the Desk, with a
Reply from the Ethnographer." In Structural Analysis of Oral Tradition,
edited by Pierre Maranda and Elie Köngäs-Maranda. Philadelphia: Uni
versity of Pennsylvania Press, 1971.
354 SOURCES
Dupont, Edouard. Lettres sur le Congo. Paris: C. Reinwald, 1889.

Evans-Pritchard, Edward E. Witchcraft, Oracles, and Magic among the
Azande. Oxford: Oxford University Press, 1937.
Fernandez, James. "Persuasions and Performances: Of the Beast in Every
Body... and the Metaphors of Everyman," Daedalus 101 (1972): 39-
60.
Frazer, James. The Golden Bough. 2 vols. London: Macmillan, 1890.
. Totemism and Exogamy. 4 vols. London: Macmillan, 1910-37.
Freud, Sigmund. Totem and Taboo. New York: W.W. Norton, 1950
(1913).
Frobenius, Leo. Die Masken und Geheimbünde. Nova Acta Abhandlung
der Kaiserliche Leopold-Carolinische Deutschen Akademie der Natur
forscher, vol. 74, no. 1. Halle: 1898.
Fukiau kia Bunseki-Lumanisa. N'kongo ye Nz'a/Cosmogonie-Kongo.
Kinshasa: Office National de la Recherche et de Développement, 1969.
Garbett, Kingsley. "Disparate Regional Cults and a Unitary Ritual Field in
Zimbabwe."In Regional Cults, edited by Richard Werbner, pp. 141-173.
New York and London: Academic Press, 1977.
Geertz, Clifford. The Social History of an Indonesian Town. Cambridge,
Mass.: M.I.T. Press, 1965.
. "Religion as a Cultural System." In Anthropological Approaches
to the Study of Religion, edited by Michael Banton. New York: Praeger,
1966.
. "Deep Play: Notes on the Balinese Cockfight" Daedalus 101
(1972): 1-38.
Girard, René. La violence et le sacré. Paris: Grasset, 1972.
Globus (1876), p. 307.
Gluckman, Max. Ideas in Barotse Jurisprudence. New Haven: Yale Uni
versity Press, 1965.
Goody, Jack. Technology, Tradition, and the State in Africa. London,
Ibadan and Oxford: Oxford University Press, 1971.
Gräbner, Fritz. "Kulturkreise und Kulturschichten in Ozeanien." Zeit
schrift für Ethnologie 37 (1905): 84-90.
Güssfeldt, Paul; Falkenstein, Julius; and Pechuel-Loesche, Eduard Die
Loango-Expedition: 1873-6. Leipzig: Verlag v. Paul Frohberg, 1879.
Guthrie, Malcolm. Comparative Bantu. 4 vols. Hants: Gregg Press, 1967.
Hagenbucher-Sacripanti, F. Les fondaments spirituels du pouvoir au roy
aume de Loango. L'office de la recherche scientifiques et technique outre
mer, no. 67. Paris: 1973.
Herskovits, Melville. Life in a Haitian Village. New York: Knopf, 1937.
. Dahomey: An Ancient West African Kingdom. 2 vols. New York:
Augustin, 1938.
. The Myth of the Negro Past. Boston: Beacon Press, 1958 (1941).
Sources 355
Hyacinthe de Bologne. La pratique missionaire des PP. Capucins Italiens

dans les royaumes de Congo, Angola et contrées adjacents. Louvain:
Editions de l'Aucam, 1931 (1747).
Jacobson-Widding, Anita. Red-White-Black as a Mode of Thought: A
Study ofTriadic Classification by Colours in the Ritual Symbolism and
Cognitive Thought ofthe Peoples ofthe Lower Congo. Acta Universitatis
Upsaliensis (Uppsala Studies in Cultural Anthropology, no. 1). Uppsala:
1979.
Janzen, John M. "Laman's Kongo Ethnography: Observations on Sources,
Methodology and Theory." Africa 42 (1972): 316-328.
. The Quest for Therapy in Lower Zaire. Berkeley and London:
University of California Press, 1978.
. Review of Les phemba du Mayombe by Raoul Lehuard. African
Arts 11, no. 2(1978): 88-89.
. "Ideologies and Institutions in the Precolonial History of Equa
torial African Therapeutic Systems." Social Science and Medicine 13B,
no. 4(1979): 317-326.
. "Deep Thought Structure and Intention in Kongo Prophetism,
1910-1921." Social Research 46, no. 1 (1979): 106-139.
, and MacGaffey, Wyatt^/i Anthology of Kongo Religion: Pri
mary Sources from Lower Zaire. University of Kansas Publications in
Anthropology, no. 5. Lawrence: 1974.
Jung, Carl, and Kerenyi, Carl. Essays on a Science of Mythology. Bollingen
Series 22. Princeton: Princeton University Press, 1969.
Kerenyi, Carl. "Prolegomena." In Essays on a Science of Mythology, by
Carl Jung and Carl Kerenyi, pp. 1-24. Princeton: Princeton University
Press, 1969.
Knappert, Jan. Myths and Legends of the Congo. Nairobi and London:
Heinemann Educational Books, 1971.
Koch, Henri. Magie et chasse au Cameroun. Paris: Editions Berger-
Levrault, 1968.
Kuper, Adam. Anthropologists and Anthropology: The British School
1922-72. New York: Pica Press, 1972.
Kusikila, Yoswe. Lujwa evo Kimongi e? Kumba: Académie Congolaise,
1966.
Kwamba, Elie. "A Lemba Initiation." In An Anthology of Kongo Religion,
edited by John M. Janzen and Wyatt MacGaffey, pp. 97-102. University
of Kansas Publications in Anthropology, no. 5. Lawrence: 1974.
Lacan, Jacques. "The Insistence of the Letter in the Unconscious." In The
Structuralists, edited by Richard and Fernande DeGeorge, pp. 287-323.
Garden City: Anchor, 1972.
Laman, KarlE. Dictionnaire Kikongo-Français. Bruxelles: Institut Royale
Coloniale Belge, 1936.
356 SOURCES
. The Kongo I. Studia Ethnographica Upsaliensia, no. 4. Uppsala:

1953.
. The Kongo II. Studia Ethnographica Upsaliensia, no. 8. Uppsala:
1957.
. The Kongo III. Studia Ethnographica Upsaliensia, no. 13.
Uppsala: 1962.
. The Kongo IV. Studia Ethnographica Upsaliensia, no. 16.
Uppsala: 1968.
Leach, Edmund. Culture and Communication: The Logic by which Sym-
bols are Connected. Cambridge: Cambridge University Press, 1976.
Lehuard, Raoul. Les phemba du mayombe. Paris: Collection Arts d'Afrique
Noire, 1977.
Lévi-Strauss, Claude. Les structures élémentaires de la parentée. Paris:
Presses Universitaires de France, 1949.
. "The Structural Study of Myth." Journal of American Folklore
67 (1955): 428-444.
. La pensée sauvage. Paris: Pion, 1962.
. Le cru et le cuit: Mythologiques I. Paris: Pion, 1964.
. Du miel aux cendres: Mythologiques II. Paris: Pion, 1966.
. L'origine des manières de table: Mythologiques III. Paris: Pion,
1968.
. L'homme nu: Mythologiques IV. Paris: Pion, 1971.
MacGaffey, Wyatt Custom and Government in the Lower Congo. Berke
ley, Los Angeles, and London: University of California Press, 1970.
. "The Religious Commissions of the BaKongo." Man n.s. 5
(1970): 27-38.
. "Corporation Theory and Political Change." In Small States and
Segmented Societies, edited by Stephanie Newman. New York Praeger,
1976.
. "Economic and Social Dimensions of Kongo Slavery." In
Slavery in Africa, edited by Suzanne Meirs and Igor Kopytoff. Madison:
University of Wisconsin Press, 1977.
Maes, Joseph, and Boone, Olga. Les peuplades du Congo belge. Bruxelles:
Imprimerie Veuve Monnom, 1935.
Maine, Henry. Ancient Law. London: Dent, 1960.
Makarius, Laura. "Ritual Clowns and Symbolic Behavior." Diogenes 69
(1970); 44-73.
Malonga, Jean. La legènde deM'pfoumou ma Mazono. Paris: Editions Afri
caines, 1954.
. "La sorcellerie et l'ordre du 'Lemba' chez les Lari." Liaison
62 (1958): 45-49, 51-61.
Maranda, Pierre, and Köngäs-Maranda, Elie, eds. Structural Analysis of
Oral Tradition. Philadelphia: University of Pennsylvania Press, 1971.
Martin, Phyllis. The External Trade of the Loango Coast 1576-1870.
Oxford: Oxford University Press, 1972.
Sources 357
Meier, Christian. Die Entstehung des Politischen bei den Griechen. Frank
furt am Main: Suhrkamp Verlag, 1980.
Meillassoux, Claude, ed. L'esclavage en Afriqueprécoloniale. Paris: Mas-
pero, 1975.
Meirs, Suzanne, and KopytofF, Igor, eds. Slavery in Africa: An Historical
and Anthropological Perspective. Madison: University of Wisconsin
Press, 1977.
Métraux, Alfred. Voodoo in Haiti. New York: Oxford University Press,
1959.
. Black Peasants and Voodoo. New York: Universe Books, 1960.
Miller, Joseph. Kings and Kinsmen: Early Mbundu States in Angola.
Oxford: Oxford University Press, 1976.
Munzele, Jacques. Bakulu beto ye Diela diau. Kumba: Académie Con-
golaise, 1965.
Murdock, George P. Africa, Its People and their Culture History. New
York: McGraw-Hill, 1959.
Newman, Stephanie, ed. Small States and Segmented Societies. NewYork:
Praeger, 1976.
Ngoma, Ferdinand. Vinitiation bakongo et sa signification. Lubumbashi:
Centre d'Etude des Problèmes Sociaux Indigenes, 1963.
Ngubane, Harriet Body and Mind in Zulu Medicine. New York and
London: Academic Press, 1977.
Nietzsche, Friedrich. "The Birth of Tragedy." In Tragedy: Vision and
Form, edited by Robert W. Corrigan. San Francisco: Chandler, 1965.
Nsemi, Isaac. "Min'kisi: Sacred Medicines." In An Anthology of Kongo
Religion, edited by John M. Janzen and Wyatt MacGaffey, pp. 34-38.
University of Kansas Publications in Anthropology, no. 5. Lawrence:
1974.
Orley, Joha "African Medical Taxonomy." Journal ofthe Anthropological
Society of Oxford 1, no. 3 (1970): 137-150.
Pechuel-Loesche, Eduard. Volkskunde von Loango. Stuttgart: Streckerand
Schroeder, 1907.
Pelton, Robert D. The Trickster in West Africa: A Study of Mythic Irony
and Sacred Delight. Berkeley, Los Angeles, and London: University of
California Press, 1980.
Pinkerton, John, ed. A General Collection of the Best and Most Interesting
Voyages and Travels in All Parts of the World. London: Kimber and
Conrad, 1814.
Price-Mars, Jean. "Lemba-Pétro, un eulte secret" Revue de la société
d'histoire et de géographie dHaiti 9, no. 28 (1938): 12-31.
Prins, Gwya "Disease at the Crossroads: Towards a History of Thera
peutics in Bulozi since 1876." Social Science and Medicine 13B (1979):
285-316.
358 SOURCES
. The Hidden Hippopotamus: Reappraisal in African History, the

Early Colonial Experience in Western Zambia. Cambridge: Cambridge
Proyart, A. "Histoire de Loango, Kakongo et autres royaumes d'Afrique."
In A General Collection ofthe Best and Most Interesting Voyages, edited
by John Pinkerton. London: Kimber and Conrad, 1814.
Radin, Paul. The Trickster: A Study in American Indian Mythology. New
York: Bell Publishing Co., 1961.
Ramos, Arthur. O Negro Brasileiro. Rio de Janeiro: Civilizacäo Bra-
sileria, 1934.
Ranger, Terence O. Dance and Society in Eastern Africa: The BeniNgoma.
London: Heinemann, 1975.
. "Healing and Society in Colonial Southern Africa." Unpublished
MS, 1978.
Rappaport, Roy A. Pigs for the Ancestors: Ritual in the Ecology of a New
Guinea People. New Haven and London: Yale University Press, 1968.
Rey, Pierre-Philippe. "L'esclavage lignager chez les tsangui, les punu et les
kuni du Congo-Brazzaville." In L'esclavage en Afrique précoloniale,
edited by Claude Meillassoux, pp. 509-528. Paris: Maspero, 1975.
Schoffeleers, Matthew. "Cult Idioms and the Dialectics of a Region." In
Regional Cults, edited by Richard Werbner, pp. 219-239. New York and
London: Academic Press, 1977.
Schultz, Heinrich. Altersklassen und Männerbünde. Berlin: G. Reimer,
1902.
Smith, Michael G. Corporations and Society. London: Duckworth, 1974.
Soret, Marcel. Les Kongo Nord-Occidentaux. Paris: Presses Universitaires
de France, 1959,
Sperber, Dan. Rethinking Symbolism. Cambridge: Cambridge University
Press, 1975.
Stanley, Henry M. Through the Dark Continent. 2 vols. New York: Harper,
1879.
Stebich, Ute, ed. Haitian Art. Brooklyn: The Brooklyn Museum, 1978.
Stenström, Oscar. "The Lemba Cult" Ethnos (1969), pp. 37-57.
Swantz, Marja-Liisa. Ritual and Symbol in Transitional Zaramo Society.
Uppsala: Gleerup, 1970.
Tastevin, C. "Idées religieuses des indigènes de l'enclave de Cabinda."
Etudes missionnaires 3 (1935): 105-111, 191-197, 257-273.
Thiel, Josef Franz. Ahnen, Geister, Höchste Wesen. Studia InstitutiAnthro-
pos, no. 26. St Augustine: 1977.
Thompson, Robert F arris. Black Gods and Kings. Los Angeles: University
of California Press, 1971.
. "Transatlantic African Art Traditions." Unpublished MSS,
1975, 1977.
. "The Flash of the Spirit: Haiti's Africanizing Vodun Art" In
Haitian Art, edited by Ute Stebich, pp. 26-37. Brooklyn: The Brooklyn
Museum, 1978.
Sources 359
Turner, Victor W. "Colour Classification in Ndembu Ritual." In Anthro-

pologicalApproaches to the Study of Religion, edited by Michael Banton,
pp. 47-84. New York: Praeger, 1966.
. Drums of Affliction: A Study of Religious Processes among the
Ndembu of Zambia. Oxford: Clarendon Press, 1968.
. The Ritual Process. Chicago: Aldine, 1969.
. Dramas, Fields and Metaphors: Symbolic Action in Human
Society. Ithaca: Cornell University Press, 1974.
. Revelation and Divination in Ndembu Ritual. Ithaca: Cornell
. "Symbols in African Ritual." In Symbolic Anthropology, edited
by Janet L. Dolgin, et al. New York: Columbia University Press, 1977.
Van Binsbergen, Wim. "Regional and Non-Regional Cults of Affliction in
Western Zambia." In Regional Cults, edited by Richard Werbner, pp.
141-173. New York and London: Academic Press, 1977.
Van Gennep, Arnold. Les rites de passage. Paris: Emile Nourry, 1909.
Vansina, Jan. Kingdoms ofthe Savanna. Madison: University of Wisconsin
Press, 1965.
. The Tio Kingdom of the Middle Congo, 1880-1892. Oxford:
Oxford University Press, 1973.
. The Children of Woot: Essays in Kuba History. Madison:
University of Wisconsin Press, 1978.
Van Velzen, Bonno Thoden. "Bush Negro Regional Cults: A Materialist
Explanation." In Regional Cults, edited by Richard Werbner, pp. 93-
116. New York and London: Academic Press, 1977.
Van Wing, Joseph. Etudes BaKongo. Bruxelles: DescléedeBrouwer, 1959.
, and Schöller, Claude. Legendes des BaKongo orientaux. Lou
vain: Aucam, 1940.
Vaz, José Martins. Filosofia Tradicional dos Cabindas. Lisboa: Agencia-
Geral do Ultramer, 1969.
Visser, Robert "Der Kaufmann unter Wilden (Entwicklung des Handels am
Kongo)." Jahresbericht des Naturwissenschaftlichen Vereins zu Krefeld
(1907-8), pp. 59-60.
Werbner, Richard, ed. Regional Cults. New York and London: Academic
Press, 1977.
Wundt, Wilhelm. Völkerpsychologie. 10 vols. Leipzig, 1911-20. Vols. 1-2
published by W. Engelmann; vols. 3-10 published by Alfred Kröner.
Archival Sources
AIMO. "Etat d'esprit de la population." Rapport annuel affaires indigenes
et main d'oeuvre. Luozi Territorial Archives, 1936.
Babutidi, Timotio. Notebook 16. Laman Collection, Svenska Missions-
Bittremieux, Leon. Correspondence with Musée d'Afrique Central
(Tervuren) from Kangu Mission, Mayombe. 2 August 1939.
360 SOURCES
Congo beige. Reports. Luozi Territorial Archives, 1930.

Kimbembe. Notebook 80. Laman Collection, SMF, Lidingo, n.d.
Konda, Jean. Notebook 119. Laman Collection, SMF, Lidingo, n.d.
Kwamba, Elie. Notebooks 142-3. Laman Collection, SMF, Lidingo, 1915.
Lunungu, Moise. Notebook 181. Laman Collection, SMF, Lidingo, 1917.
Lutete, Esaie. Notebook 229. Laman Collection, SMF, Lidingo, 1915.
Maillet, Robert. Report. Luozi Territorial Archives, n.d.
Ndibu, Joseph. Notebook 345. Laman Collection, SMF, Lidingo, n.d.
Nitu, Solomo. Cabinda Legends Transcriptions. Recorded by Charles
Harvey, 1961.
Vercraeye, J. Dossier No. 129. Luozi Territorial Archives, 31 August 1932.
Visser, Robert Correspondence. Linden Museum, Stuttgart.
Museum Collections
ABBREVIATIONS OF MUSEUMS
AMBG Afrika Museum, B erg en D al, The Netherlands

B-D Berlin-Dahlem Museum fur Völkerkunde
GEM Göteborg Ethnographic Museum, Sweden
KM Kimpese Museum, Lower Zaire
LMS Linden Museum, Stuttgart, West Germany
LRM Leiden Rijks Museum, The Netherlands
MAC Musée d'Afrique Central, Tervuren, Belgium
SEM Stockholm Ethnographic Museum, Sweden
LEMBA DRUMS
SEM 1907.8.34 "Konko Lemba" from Kimbunu. don. J.

Lindstrom, 1907.
SEM 1907.8.38 "Nkonko a Lemba" from Mbamba,
Lower Congo, don. J. Lindstrom, 1907.
SEM 1919.1.437 "Mukonzi a Lemba" from Mayombe,
don. K. Laman 1919 (No. 40, fig. 1).
SEM 1919.1.445 "Nkonko a Lemba" from Sundi in
Ludima, don. Laman, 1919 (No. 39, fig. 1).
GEM 68.11.230 "Nkonko Lemba".
GEM 68.11.241 "Nkonko Lemba" fromNganda, Kingoyi,
don. J. Hammar, 1910 (No. 9, fig. 1).
B-D III C 18921 "coco Lemba", fromBavili, don. R Visser,
1904 (No. 29, fig. 1).
MAC 34778 "Ngoma Lemba".
MAC 34774-34782 "Zingoma zaDilemba", don. L.
Bittremieux.
MAC 35342 miniature bone Lemba drum amulet.
MAC 43695 miniature wood Lemba drum amulet.
Sources 361
MAC 69.59.498 miniature wood Lemba drum amulet from

Mayumba, don. MJ. Sena, 1946.
MAC 53.74.1323 miniature metal Lemba drum amulet
MAC 35052 Lemba miniature drum, don. B ittremieux.
LMS 38363 Necklace strung with teeth, miniature
Lemba drum as main ornament, don. R.
Visser, 1905.
LEMBA BRACELETS
B-D III C 347 Copper Lemba bracelet, Tschicambo,

Loango, don. Loango Expedition, 1875.
B-D III C 710b Brass Woman's Lemba bracelet, 8 cm.,
don. Loango Expedition, 1875.
B-D III C 6524 Brass bracelet "Bilunge Chilenge,"
BaKunya, don. R Visser, 1896.
B-D III C 8136 C Wood bracelet form for casting, don.
Visser.
B-D III C 13810a-f Clay forms for casting Lemba bracelet,
don. R Visser, 1901 (no longer
existent).
LRM 1032/136 Casting mould of clay for a bracelet,
(I.C.E.Inv.No.) Boma(No. 31, fig. 1).
LRM 1032/60 Red copper engraved bracelet, Banana
(I.C.E.Inv.No.) (No. 32, fig. 1).
LRM 1032/59 Yellow copper (brass) engraved
(I.C.E.Inv.No.) bracelet, Banana.
LRM 1034/53 Red copper engraved bracelet, Banana.
(I.C.E.Inv.No.)
GEM 66.15.3 Bronze "Nsongo a Lemba," Kitoma,
Atlantic coast, don. J. Lagergren, 1966.
5 cm. (No. 34, fig. 1).
GEM 1938.31.12 "Mulunga Lemba," Indo Bayaka,
Bakuta, 7 cm. don. Andersson (No. 37,
fig. 1).
GEM 1938.31.11 "Mulunga Lemba," Ntele, Bayaka,
Bakuta, 10 cm.
MAC 35021 Lemba bracelet, don. Bittremieux.
MAC 21372 Red copper bracelet, don. S.G. Arnold,
1917. 9 mm.
MAC 54.2.10 Lead bracelet, purch. Walschot, 1954.
MAC 43587 Red copper, traces of ngula (sic), Lemba
bracelet, Mayombe, don. M. Sena,
1946.
362 SOURCES
MAC 43578 Copper Lemba bracelet, Mayombe,

don. Seha, 1946, 9 cm.
MAC 43584 Yellow copper Lemba bracelet, don.
Seha, 1946, 8 cm.
MAC 24985 Copper bracelet, don. Renkin, 1909, 8
cm.
MAC 23275 Copper bracelet, don. Michel, 1919, 9
cm.
MAC 75.51.21 Lemba bracelet, lead or brass.
MAC (recent Lemba bracelet, don. J. Walschot
acquisition)
MAC 54.60.2 Lemba bracelet, brass, 10 cm.
MAC 53.74.1312 Copper bracelet "Lunga milemba,"
Tshela, Sundi tribe, Zubu, don. A.
Maesen, 1954.
MAC 53.74.1339 Copper "lunga Lemba," Tshela,
Mayumbe, Kai ku Tsanga, don. A.
Maesen, 1954.
MAC 53.74.1313 "Lunga miLemba," Tshela, Sundi/
,Fubu, made in Mama Tsanga village,
sector Tsanga Nord, don. Maesen,
1954.
MAC 53.74.1324 Copper bracelet for Lemba wife, right
arm, Tshela, Sundi tribe, Kangu-Lufu,
don. A. Maesen, 6 cm.
MAC 53.74.1333 Copper bracelet, "Lunga Lemba,"
Tshela, Mayumbe, Kai ku Tsanga, don.
Maesen, 9 cm.
MAC 53.74.1332 Copper "lunga Lemba," Mayumbe, Kai
ku Tsanga, clan Makaba, don. A.
Maesen, 9 cm.
MAC 54.74.1335 Copper "lunga Lemba," Mayumbe, Kai
ku Tsanga, don. Maesen, 8 cm.
AMBG 417.50 bracelet.
AMBG 415 I 536 Three copper bracelets, one unworked,
(inv. no. 29-536) of Vili manufacture, two with incised
snakes, ordered from Europe.
B-D III C 423 Bracelet, Loango Expedition 1875,
Bastian.
B-D III C 6535 Copper Lemba bracelet, don. Visser,
1896 (missing since WWII from
collection).
Sources 363
LEMBA MEDICINE CONTAINERS
B-D III C Rafia Lemba charm sack ("Sackfetisch

13871 Lembe"), Caio, Loango Coast, don. R.
Visser, 1900 (no longer in collection
since WWII).
B-D III C "Lembe Sackfetisch", Caio, Loango
13743 Coast, don. Visser, 1901 (no longer in
collection).
B-D III C "Lembe Sackfetisch", Caio, Loango
13744 Coast, don. Visser, 1901 (no longer in
collection).
MAC 43040 "N'kobe Lemba," Kangu, Mayombe,
don. Bittremieux.
MAC 37972 "Khobe Lemba," don. Bittremieux,
1937.
MAC 35191 "Khobe Lemba," Woyo, don.
Bittremieux, 1933.
MAC 35192 "Khobe Lemba," Woyo, don.
Bittremieux, 1933.
SEM 1919.1.583 "Nkisi Lemba," Nsundi, Mukimbungu,
don. Laman, 1919.
KM (no identif. Lemba n'kobe (no documentation),
number)
MISCELLANEOUS LEMBA OBJECTS
MAC 34766 Smoking pipe head.

B-D III C Lemba pipe ("N'timbeliambe"), doa
13873 Visser, 1900.
B-D Katalog Lemba opium pipe ("Hanfrauche
372 Lemba"), don. Bastian, 1874 (missing
from collection).
AMBG 177 Probably Lemba couple, statue,
Landana, Cabinda.
MAC 35176 Lemba anthropomorphic statue
("Ndubi Lemba"), with left arm raised
and croissant-drum hanging from same
elbow.
GEM 68.11.208 Lemba dance rattles, Basundi, Nganda,
don. J. Hammar, 1906.
Index
Abala (region), 61 Babutidi (Kongo catechist-

Affonso (Kongo king, also called ethnographer), 243-4,
Nzinga a Mpemba), 63-4 248-51,253,255,258, 342,
Alexander von Humboldt 343
(foundation), vi Bade (Fongbe-Brazilian vodun),
Auma (river, region), 32, 61 176
Aluvais (Angola-Brazilian Bahelele,J.,344
inkissi), 276 Bahia (Brazilian city), 274-6
Ambriz (port town), 2, 29, 63 Bakongo (village), 37
America(n), food plants in Africa, See also Kongo (people)
29, 31; extensions of Lemba Bana bambuta (patrifilial
to, 273-92 children); see Mwana
Andersson, E., 2,157-85passim, Banana (city), 2
341 Bansa (Loango nkisi), 53
Angola, 276, 281 Banton, M., 339, 341
Angoro (Angola-Brazilian Bantu (language, cultural group of
inkissi), 276 peoples), 12-15;
Angoromea (Congo-Brazilian characteristic therapeutics,
inkissi), 276 12-15; influences in Brazil,
Aniota (Congo "leopard" cult), 274-7; waxing and waning of
11 cultic and therapeutic
Anne, St., in Afro-Brazilian institutions among, 7
religion, 276 Barbara, St., in Afro-Brazilian
Ansiku (people; Teke), 60 religion, 276
Anthony, St., in Afro-Brazilian Baron Samedi (Haitian loa), 283
religion, 276 Barthes,R., 339-40
Arada (also Alada; Dahomean Bartholomew, St., in Afro-
town), 277 Brazilian religion, 276
Armstrong, R.P.,339 Bascom, W., 344
Ashade (Haitian "Rada" loa), 284 Bassesse (village), 37
Ashanti (people), influences in Bastian, A., 2, 57, 244,336, 337,
New World, 273-7 342, 346, PI. 8
Atlantic (ocean, coast), 2, 3,5,18, Bateson,G.,310,340, 346
27; trade across, 273-5 Batshikama ba Mampuya, 187,
Auge (Fongbe-Brazilian vodun), 201,341-2, 347
276 Battel, A., 55, 336
Ayoani (Fongbe-Brazilian vodun), Baumann, H., 335
276 Belgium (colonialism in Congo),
Aziri (Fongbe-Brazilian vodun), 6, 33,81,83,85-6,90,245,
276 247,342-3
365
366 INDEX
Bembe (people, region), 2, 233 Bosu (Haitian "Petro" loa), 54,

Benedict, St., in Afro-Brazilian 281-3, 285
religion, 276 Bosu Trois Comes (Haitian loa),
Benga, J., 344 284
Beni-Ngoma (drum association), Boukman (hero of Haitian
6 revolution), 278
Beti-Landa (Woyo mythological Brazil, 29, 32, 60, 95, 273-7
figure), 301-3, 345 Brazzaville, 2-3, 82, 95,157
See also Kunda-Bala Bueno (nkisi of chiefship), 68
Bible, translation of, and Kongo Bumba (Loango, Atlantic coast
literacy, 86-7; access to, and nkisi, also Bomba), 53, 274,
power of prophets, 90-1; 277
translation of "Holy Spirit" See also Mbumba Luangu;
in, 249 Khimba
See also Christianity Bunzi (earth spirit, shrine, in
Bikinda (ancestral, earth spirit), Ngoyo; cult of affliction),
44, 69, 79 47769,79,97,248f277,
Bingo (Woyo nkisi), 301, 303; see 291-2, 302-303; as afflicting
KaBuanga; Pfemba spirit,4*2,150-1; as central
Binza (Nsundi district, town), 62, shrine of regional cult, 21;
64 and chiefship, 15,17,18;
Bittremieux, L. (Belgian proliferation of, 22, 54
missionary-ethnographer), 2, Burridge, K., 20, 311,333,346
12,83,233-45,255,298, Bush Negroes (Guyana,
336-8, 342-3, 346, Pis. 4, 5, 6, Surinam), 19
14 See also Maroon; Gaan Tata
Bituma (Zambian cult of Bwende (clan, ethnic label), 2, 38
affliction), 17, 19 Bwiti (Fang cult, Gabon), 6
Bitungu (Loango nkisi), 56
Boas, G., 346 Cabinda (port town, region), 2-3,
Boenza~(Nsundi district), 62, 64 5, 10, 29, 34, 46, 252,278,
Boessi-Bata (Loango nkisi), 52-3, 300, Pis. 18, 19
345 Cajanja (Angola-Brazilian
Bohannan, P., 335 inkissi), 276
Bolamba, A.R.,344 Calumet (ceremonial trading
Boma (city), 2 system), v
Bombongira (Congo-Brazilian Cameroons, 28,257
inkissi), 276-7 Candomble (Afro-Brazilian cult
Bomfin, Savior of, in Afro- form), 274-5, 301
Brazilian religion, 276 Capitalism, arousing spirit of
Bongo (people), 37 (heha mbongo), 250-1;
Boone, O., 335 Lemba as cure for, v
Bordeaux (city), 278 Carniero, E., 274-5, 344
Index 367
Cassange (southern savanna 81, 85, 245, 247; influences in

kingdom), 65 New World culture, 273-7;
Cassunbenca (Angola-Brazilian Lower Congo region, 9-10,
inkissi), 276 15, 23, 25, 95, 247; Republic
Catechists, authors of of, 95; river, 18,32
ethnographic notebooks, 87, See also Nzadi; Zaire
105, 157,233,340 Congo Zandor (Haitian "Pétro"
Catende (Congo- and Angola- loa), 282-3
Brazilian inkissi), 276 Coquery-Vidrovitch, C , 338
Chatelein, H., 280-1,345, 347 Corporation theory, presented,
Chavunduka,G.,331 22-3 ; in explanation of
Chef de groupement (Belgian diversity of Western
colonial title), 85 Equatorial African
Chef médaillé (colonial institutions, 21; in relation to
chiefship), 70, 85 social and political systems
Chinese, 37, 81 of Lemba region, 45-79
Christ, in Afro-Brazilian religion, passim. x
276 Correspondence theory, 18-19;

Christianity, 82, 86-7; conversion criticized, 20
to, 87, 89; deities and saints Corrigan, R., 346
in New World Afro- Cuango-Malimba (Loango nkisi),
American religion, 273-7; in 56
historic Kongo kingdom, 63; Cuba, 274
suspected influence in Culture area theory, 11
Lemba ethics, 320; as taught Cuvelier, J.,62,336, 338
by prophet Kimbangu, 91
Christophe (Haitian president),
278,288,292,298 Dadie,B.,307,346
Clan, 38-41; local section, 39, 44 Dahomey, influences in New
See also Dikanda; Mvila World, 273-7
Codes, of expressive domaines, Damballah Wedo (Haitian "Rada"
100-1; identification and loo), 282, 284
deciphering of, 100-1 ; use in Dandalunda (Angola-Brazilian
disparate data criticism and inkissi), 276
evaluation, 103 Dapper, O., 2, 51-5, 251-2, 256,
See also Expressive domains 275, 277, 337, 345, PI. 1
Colonialisme, 81-91,169 De Cleene, E.,336,337
Colombia, Teke slaves De Craemer, W.,332
distinguished in, 60 DeGeorge, R. and F.,346
Congo (river and basin, Belgian De Heusch,L.,21,334
colony, republic), 3-4,5,9, DeJonghe, E., 9-12passim, 331
10, 29, 34,46, 95, 252, 278, Deleval,H.,38,336, 337
300, Pis. 18, 19; Free State, Dembacani (Loango nkisi), 56
368 INDEX
Dennet, J., 2 Elegba (Fongbe-Brazilian voduri),

Dessalines (Haitian president), 276
278,288, 292,298 See also Legba
Devil, in Afro-Brazilian religion, English, 34
276; alleged marriage with, Erzilie (Haitian "Petro" loa), 283
51 Erzilie Ge Rouge (Haitian "Petro"
Dikanda (lineage, clan section), loa), 282, 284
39; Evans-Pritchard, E., 331
See also Kanda Expedit, St., in Afro-Brazilian
Diogo Cao (Portuguese explorer), religion, 276
63 Expressive domains, as analytic
Dionga, W., 343 method in study of
Dolgin,J.,339 therapeutic ritual, 102, 313; of
Dondo Mazinga (Nsundi district), color, 98; defined, 98-9;
63 exchange of goods and
See also Mazinga symbols, 98,135-8,166-9,
Don Pedro (Haitian legendary 245-7; hypothesis of
hero, lod) See Jean Petro, transference between, and
Lemba-Petro Lemba therapeutics, 304-5;
Douglas, M., 174,341 of the lyrical, 98,146-55,
Doutreloux, A., 337 179-85, 200-31, 258-72; of
Droux,C, 334 medicine, 98,140-3,172-9,
Drum, drumming, 52,107, 111, 195-200; of medicine and art,
114-15,118,137,157,179,192, 251-58; redundancy of, 100,
234,252,279, 282, Pis. 14-17; 310; social organization as,
of affliction, v, 3,12-14, 20, 98,138-40,169-72; space and
315; casting spell with, 241, time in ritual, 98,130-5,
249, 256-7; miniaturized as 163-6, 189-95, 243-5, 280,
charm, 252,256, pi. 16; as 291; of verbal categories of
non-verbal therapeutic ritual action, 98,143-6,
metaphor, 310; rhythm as 247-51
coded expressive domain, 98; See also Codes; Metaphor
in Haitian loa possession, Exu (Yoruba-Brazilian orisha),
238; waxing and waning of 276
drums of affliction, 7,14
See also Nkonzi; Nkonko; Falkenstein, J., 2, 337
Ngoma Fallers, L., vii
Duke (title in Nsundi), 64 Fernandez, J., 339
Dundes, A., 340 Fidalgos (Nsundi appointments),
Dunga (Nsundi town), 66 64
Dupont, E.,29, 32, 335,338 Fongbe (people, Dahomean,
Durkheim, E.,9,19 Republic of Benin), 285
Dutch, 34 France, French, 6, 64, 81-2,169,
Index 369
272; revolution in Haiti, 278 George, St., in Afro-Brazilian

Francis, St., in Afro-Brazilian religion, 276
religion, 276 Girard,R.,346
Frazer, J.,9,331 Ghana, 16
Freud, S., 9, 331, 346 Gluckman,M.,38,336
Frobenius, L., 10,11, 298, 331, God, 60, 87,185,192,201-202,
335, 345 274, 276, 284, 301-302; in
Fuanda (Nsundi ruler), 66, 68 centralized cult, 19; as
Fukiau (Kongo writer), vii, 2, creator, 46; as Father, in
187-201,202-206,317,341, source of trickster's Lemba
342 medicine, 258-67; in Holy
Fula (to blow, arouse, channel); Family, 176-7; word of, 86; in
fudila mwana, arouse, raise a Lemba lyrics, 184
child, 249; fudila nkisi, See also Nzambi
arouse an nkisi, 249; fula Gombiri (Loango nkisi), 55
beka kubafula,fulikidi, to Goody, J., 21, 333
resurrect, 163, 249; fula dia Graebener, F , 10,11,331
bula, village entrance, 107, Gussfeldt, P., 2, 57, 244,337,342
132,161,237,249;/«/0</fa Guinea, coast, 6
zandu, market entrance, PL Gun (Fongbe-Brazilian vodun),
3; he ha mu mafula, arouse, 276
stir up in the entrances [of Guthrie, M., 332
market, village], 235, 249-50; Guyana, 19
mafula mia zandu, market
trading groups, 74; mbondo Hagenbucher-Sacripanti, F , 336
fula, oration charm, 249, 257, Haiti, 95,102-3,188,273-92
PL 12 passim.
Funza (twinship cult), 88-9, Hako (Pawnee ceremonial trading
300-301 association), v
Hammar, J., 2
Gaan Tata (Guyana regional cult), Handa, vanda (to initiate to or
19-20 compose nkisi), 101,106-107,
Gabon, 6 115,122,125,128,234, 241-2,
Gabriel, St., in Afro-Brazilian 297; in legitimating beliefs,
religion, 276 297; logic of associations in,
Gaetano, St., in Afro-Brazilian 297-9, 311-13; in trickster
religion, 276 legend, 259-65; as verbal
Gangala (people, also Bayangala), category of ritual action,
2, 37 143-4
Garbett, K., 332 See also Medicine, Nkisi,
Gedenibo (Haitian "Petro" loa), Nkobe, Plants
283 Harvey, C , 342
Geertz,C.,20,333,339 Hausa (people), 37
370 INDEX
Heha, veva (to arouse, infuse Islamic brotherhoods, Egypt, v

with spirit), 235, 249-50
Herbert, E., vii Jacobson-Widding, A., 343
Herskovits, M., 274-5, 281-5, Jamaica, 273
344, 345 Janzen, J., 332,339, 340, 341, 343,
Historical school, and manist 345,346
explanation of religious Jean Petro (legendary Haitian
change,10, 298 hero), 277-92 passim, 298,
Holy family, in Afro-American 301
religion, 276,286 See also Don Pedro; Petro-
Holy Spirit, 249, 276 Lemba
Hyacinthe de Bologne, 49-50, 337 Jerome, St., in Afro-Brazilian
religion, 276
Ideology, and choice of John, St., in Afro-Brazilian
alternatives, 295-6; beyond religion, 276
ideology, to explore John the Baptist, St., in Afro-
dilemmas, 324-5; myth Brazilian religion, 276
structure and, 311; of Lemba Jung, C , 297, 345
therapeutics, 315-21;
transference of dilemmas KaBuanga (Woyo nkisi), 301
and, 311-17 See also Bingo; Pfemba
Ikenga (cult of the hand), 6 Kaiala (Angola-Brazilian inkissi),
Imba (Loango nkisi), 55 276
Imbaw (Teke clan), 38 Kakongo (kingdom), 4-5,18, 20,
See also Kimbanga 37,45-58passim, 248
Immaculate Conception, Our Kalela Dance (ngoma
Lady of, in Afro-Brazilian association), 6
religion, 276 Kalundu (Bantu-Brazilian cult
Impaw (Teke nkobi), 59 name), 274
Incossi Mucumbe (Congo- Kalunga (Woyo nkisi, attribute of
Brazilian inkissi), 276-7 Nzambi-God), 274, 301-303
See also Kossi; Roche Kamba (people, region), 2, 35, 38,
Mucumbe 95-6,102,105-55passim, 291,
Indian (East African people), 37 300
Indo (town, mission), 128 Kambaranguanje (Congo-
Indonesia, 37 Brazilian inkissi), 276
Injami (Loango inkissi), 53, 55, Kambizi (Woyo nkisi spirit),
277 301-303
See also Njambe See also Lusunzi
Iroko-loko (Yoruba-Brazilian Kanda (lineage, clan section; also
orisha), 276 dikanda, singular, makanda,
See also Loko, Ossain plural) 39; Lemba as
Isangoma (diviners' order among preserver of, 124
Nguni-speaking peoples), 6 See also.Mvila
Index 371
Kanga (Woyo spirit), 301-303 Kinkenge (village, mission), 2

Kanga ikanga (Loango nkisi), 55 Kinkita (cult, drum of affliction),
Kangu (Mayombe mission, 16
region, chiefdom), 2, 83-4, See also Nkita; Simbi
86, 233,243, 253, Pis. 5, 6 Kinkuzu (Nsundi capital), 66
Kasai (river, province), 298 Kinshasa (village, city), 2-3, 33
Kasi (Nsundi district),62, 64 Kinsitu (Nsundi district), 62, 64
Kata Mandala (Nsundi chief), 66 Kinzaba (village), 2
Katula, D. (Lemba priest, Kinzore (village), 37
catechist), 86-7, 90,187-200 Kise (paternity), 40
Kavati (Mayombe district, See also Se
people), 51 Kisi (spirit of nkisi; sing., mukisi,
Kayo (village), 2
nkisi; pi., bakisi), 89
Kelly, H., 2, 334
Kemnitzer, D.S.,339 See also Nkisi; Medicine
Kenge (people, ethnic label), 2 Kisiasia (village), 73-8passim.,
Kerenyi, C , 297, 345 86
Khimba (youth initiation school), Kita (cult, drum of affliction), see
10-11,53,83,284 Kinkita; Nkita
Kiasi (chiefly title), 68-9 Kitauba (Loango nkisi), 53
Kibuco (Angola-Brazilian inkissi), Kitona (village), 2
276 Kitswa (patron spirit of Lemba),
Kidiba, war of, 74-6 183,185
Kiese (village), 90 Kiyaazi (Nsundi royal medicine,
Kiessubangango (Angola- rulership), 65
Brazilian inkissi), 276 Knappert, J., 346
Kifuna (village), 64 Koch, E., 334
Kikokoo (Loango nkisi), 53-5, Koloss, H., vi
277 Konda, J. (Kongo catechist-
Kilanga (village), 37 ethnographer), 2, 248-51,
Kilemfu (Nsundi district), 62, 64 257,291,317,342
Kimaje (Loango nkisi), 53-4 Köngäs-Maranda, E., 339, 341
Kimata (village), 86 Kongo (kingdom, place, ethnic
Kimbanga (clan), 38, 77-8, 85 label, people), 2,10,12,14,
See also Imbaw 37,58, 62-70passim., 82,
Kimbangu, S. (Kongo prophet), 157, 248, 257; forest in Teke
12,82-3,90-1 country, 59; historic cults of
Kimbembe (Kongo catechist- affliction and medicines in,
ethnographer), 157-85 15; kingdom and people, 5,
passim, 341 28-9,45,62-5
Kimpasi (youth initiation school), 11 See also Congo; Nzadi; Zaire
Kingoma (village), 37 Kopytoff, I., 45, 334, 336
Kingoyi, Kingoyo (village, Kossi (Loango nkisi) 53, 277
mission, clan), 66, 85 See also Incossi Mucumbe
372 INDEX
Kuba (legendary Lemba priest), Lebanese, 37

176,179-85 passim., 291 Legba (Fongbe-Brazilian vodun),
Kuba-Ntu (mythical figure), 309, 276,282-3, 284
311 Lehuard,R.,254,337, 343
Kuimba, Kikuimba (clan), 38, Lemba, etymological origin of
74-5 word, lembikisa, to calm,
Kuiti-Kuiti (Woyo mythical cool, purify, 3, 36; female
figure, Nzambi Mpungu), element of, Pfemba-Lemba,
301-303 253, 343; in Haiti, Pétro-
See also Nzambi Lemba, 273, 278; historical
Kulo-Malonga (Loango nkisi), 56 distribution, 3-7, 51-2;
Kunda-Bala (Woyo mythic Lemba laws, 122-23, 127-8,
figure), 301-303 242, 320; Lemba names,
Kundu (mystical power, 250—1 ; Lemba sickness, 3-4,
substance, pi. makundu), 106,158, 234, 315-18; male
124,146; in nkobe, 253, 256; element, Tsasa-Lemba, 253;
rules against abuses of, 127 nature of regional variations,
Kunja (Loango nkisi), 55 96-8; as purifying, calming
Kuper, A., 331 medicine, bonzo-Lemba, 158,
Kusikila, Y.,332 167-8, m/?o/oLemba,
Kuta (ethnic label), 2 106-107,120,141,150,
Kwamba E. (Kongo catechist- mwemo-Lemba, 236-7
ethnographer), 2,104-24 Lemba-Nsongi (Kongo nkisi), 298
passim., 130-55, 166,248, Leopoldville (city), 81-2
255,312, 340 Lévi-Strauss, C , 210, 311, 336,
Kwilu (Nsundi district), 62, 64 339, 342, 346
Kwilu-Niari (river, valley), 2-3, Lineage, 39; Lemba as preserver
27-46, 95 of, 124; mode of production,
39
Laadi, Lari (people, region), 2, See also Clan; Dikanda
157-8 Lisbon, 63
Lacan, J., 312, 346 Loa (Haitian voodoo deities), 277,
Lagercrantz, S., 38, 340 281-92 passim.
Lagergren, J., 2 Loanda (city), 65
Lali (people, region), 2 Loango (kingdom, port town,
Laman, K.,2,10,38, 87,105, river), 2-5,10,18, 21, 27, 29,
200-1, 298, 331,338, 340, 342, 32, 34, 45-58passim., 61, 65,
346, PI. 17 67,82,139,243,248, 252,
Lari, Laadi (people, region), 2, 278, Pis. 1,8,15,20
157-8 See also MaLoango
Lasch, R., 9 Loango Expedition, German, vi,
Lawrence, St., in Afro-Brazilian 2, 54-8
religion, 276 See also Bastian, A.
Leach, E., 340, 346 Lohse, W., vi
Index 373
Loka (to bewitch, utter strong derivative from Bunzi), 18,

words, spell), 127,129,143, 22,54
249, 256, 286 See also Bunzi
Loko (Fongbe-Brazilian vodun), Mabwaka (Yombe chief), 83-4
276 MacGaffey, W., vii, 21, 38, 332,
Lomba (Congo-Brazilian inkissi), 334,336,338,340,343,345,
276 346
See also Lemba Macumba (Afro-Brazilian cult
Lombarengenga (Angola- form), 275
Brazilian inkissi), 276 See also Incossi Mucumbe,
See also Lemba Roche Mucumbe
Longo (youth initiation rite), 11 Madula (Mayombe district), 51
Lower Congo, as region of study, Madzia (village), 2,157-63
7-23passim., 29, 37-8, passim.
See also Congo; Kongo; Zaire Maes, J., 335, 343
Luadi (river), 2 Maesen, A., vi, 2
Luala (river), 67 Mafouk (Loango trade minister)
Luanga (river), 67 34, 46, 48
Lubendo (Woyo mythical figure), Mahambu-Lukabu (Kongo nkisi),
301-303 300-301
Luc de Catalanossetta, 62 Mahungu (mythic figure, also
Luema (river), 2, 48 Mavungu, Magungu), 78, 97,
Lulendo (chiefly title, nkisi), 103,176,179-85 passim.,
68-70 188-9,193-4, 197-251
Lumani (Kongo nkisi), 89 pass/™., 291, 298,301, 304,
Lunga (Woyo nkisi cult), 22, 56, 306-309,311
301-2 Maillet,R.,338
See also Kalunga Maine, H., 21, 334
Lunungu, M. (Kongo catechist- MaKangu (Yombe chiefly title),
ethnographer), 2,105,125-30, 84
144,340 See also Ngambula
Luozi (town, territorial capital), Makariuse., 271-2, 344
338,360 Makita (Nsundi chief), 66
Lusunzi (Woyo mythical figure), Makoko (Tio king), 58-61
78, 301-304 passim., 19
Lutete, E. (Kongo catechist- MaKongo (ruler of KaKongo),
ethnographer), 346 46
Luvuangu-Mampuya (Lemba Makongo (Kongo nkisi), 53, 277
priest), 83, 85-6, 90 Makunda (Queen mother,
Loango), 47, 301
Mabiala ma Ndembe (Kongo, Makunza, Z. (Lemba priest), 347
Loango nkisi, also Mapanje), Malasi (Loango nkisi), 55
55, 301 Malawi, 18
MaBunzi (shrines, minkisi, Malaysia, 10, 28, 31
374 INDEX
Malebo Pool (in Congo/Zaire Malemba, 3,6, 29,46;

river), 2-3, 36-7, 61 Manianga, 6, 29; Mboko
Malemba (port town, capital), Nsongo, 29, 30; Mpumbu, 3,
2-3,5,29,34,46, 278 29; Ngoyo, 31; San Salvador,
Malinowski, B., 9 29; slave markets, 3, 3 3 ^
MaLoango (ruler of Loango), See also Trade
46-7, 301 Maroons (Surinam and Guyana
Malonga, J., 2,164-85, 229-30, Afro-Americans), 19
341, 342, 347 See also Bush Negroes; Gaan
Mambinda (Woyo, Yombe Tata
mythical figure, also Marquis (Nsundi title), 64
Membinda), 47, 247, 301-302 Marriage, basic social structural
Mampuya, see Batshikamaba issues of, 41-4,138-140,
Mampuya; Luvuangu- 170-171, 318; exogenous and
Mampuya endogenous approaches to,
Mamundi (village), 2, 253 42; Lemba marriage in
Mandombe (Loango nkisi), 55 Loango, 57-8; in north
Mangaka (Yombe nkisi shrine), Kongo, 78; in Yombe, 84,
247 240; in Kamba, 113; in Yaa,
Mangossu (Loango nkisi), 56 125-30; in myth, 202-6, 220,
MaNgoyo (ruler of Woyo, Ngoyo 223-8; among nobles in
kingdom), 46, 301 Sonyo, 49-51; in Kavati, 51;
Mam (title), 46, 64 reciprocating "return blood"
Manianga (market, region, clan), (mvutudulu a menga)
28-30, 35, 37-8, 62, 66-7, marriage, 43, 78,140; Roman
passim., $5-6, 95, Church's interdiction of close
187-8, 301 alliances, 49-50
Mansi (Loango nkisi), 55 Masala (village), 2
MaNsundi (Nsundi ruler), 63 Masamba (role name), 43
Maramba (Loango nkisi), 55 Matadi (port town) 2, 3, 62, 81-2
Maramba, P., 339, 340, 341 Matamba (Angola-Brazilian
Marcellind'Atri, 62 inkissi), 276
Martin, P., 48, 334, 335, 336, 337 Matele (Nsundi chief), 66
Market, 17, 230, 249-50,320, Pis. Materialist explanations, v; in rise
2-3; Ambriz, 29; area, 73-7, of regional cults, 18-19; of
138;Cabinda,3,6, 29,46; symbolic expressions, 99
calming of, 36; cycles, See also Expressive domains
weeks, 6, 29; entrance to Matheo de Anguna, 62
ifula dia zandu) 74, 235, PI. 3; Matinu (title), see Ntinu
laws, 3-4, 5-6, 12, 69-70, Mavungu (mythic figure), See
72-5; laws and feud Mahungu
resolution, 74—7, and Lemba, Mayakidi, 341
77, and N'kondi, 76-7; Maybury-Lewis, D., 340
Loango, 3, 6, 29, 46; Mayombe (region), 2,18, 27-8,
Index 375
36-8,45-58passim., 67, 79, punning, 172—3; emergence of

83,139,177, 234, 243-7, 249, new medicines, 14; as
253, 301, Pis. 5, 6,14,17 expressive domain, 172-3,
Mayumba (town), PI. 16 140-3,195-9,251-8; in
Mazamba (village), 83 relation to art, 251-8 and
Mazinga (clan), 62, 64, 74-5 iconicity, 172-3; taxonomies
Mbamba (people, ethnic label), 37 of, and social change, 14
Mbanza (generic capital city, seat See also Expressive domains;
of authority), Mbanza Nkisi
Kongo, 4-5, 63-5; Mbanza Meier, C , 325, 347
Mwembe, 4, 62, 67-9; Meillassoux, C , 336
Mbanza Nsundi, 5, 62-9 Meinhof, C.,9
passim. Meirs,S.,334, 336
Mbe (Tio capital), 58-61 Menga (blood), in descent and
Mbenza (clan, governing nkisi), marriage, 43, 78
44, 84-5 Mercantilism, and Lemba elite,
Mbete (people), 37 36; arousing spirit of, 248-51
Mbinda (Loango nkisi), 56 See also Capitalism
See also Mambinda Metaphor, study of, 99-100;
Mboko Nsongo (mining area, association of domains,
village, market), 2, 5,27, 99-100,178,185,201,256,
30-33,35,59,62,64,67,87, 304; of goods and blood, 220;
106-24,168, 300 metaphorization of
See also Market discourse, 311; movement in,
Mbola (Kongo nkisi), 15,17-18 100, 299, 308;ritualas
Mboma (people), 2, 37; amplification of non-verbal
MaMboma (title), 84, 247 metaphor, 310; root, 99; and
See also Ngambula shifting of subject across
Mbona (regional shrine), 18 domains, 299-300, 304;
Mbongo (raphia cloth, money), templates in myth analysis,
32-6; in Lemba initiations, 310-11; therapeutics as
135-8,168-9,245-7 metaphor for new social
See also Trade order, 21
Mboo (Teke chief), 61 See also Code; Expressive
Mboyo-zu-Mambi (Loango nkisi), domains; Semiotics; Symbols
55 Metraux, A., 273, 344-5
M'boze (Woyo mythic figure), Meyer, 9
301-302 Mfumu (chief), 39-40; dikanda
Mbu (village), 37 (lineage chief, freeman), 40,
Mbudila (Yombe nkisi), 247 71, 75, 86; kimfumu
Mbumba Luangu (Yombe, Woyo (chiefship), 66-7; mfumu
rainbow serpent), 53 bwala (village chief), 68; mpu
See also Bumba, Khimba (local clan chief, insignia),
Medicine, defined by verbal 70, 77; nsi (domainal chief),
376 INDEX
39, 43-4, 77,83;pa/ata mythic figure, nkisi), 300-301

(colonial chief, medallion, See also Bunzi; MaBunzi
also chef médaillé), S5;yetila Mpumbu (market, town), 2-3, 5,
(local, village chief), 71 29,33,60-1,63,67
Mfwemo, 74—5 Mpusu (Loango nkisi), 55
See also Kidiba, war of Munkukusa (Kongo purificatory
Milembamani, B., vii movement), 285
Miller, J., 21, 332 Munzele, J., 72-3,77, 317, 338,
Milonga, J., 341 347
Mimi (Loango nkisi), 53-4 Murdock,G.,38,335
Mindouli (mines, city), 5, 29, 32, Musana (village), 2, 157
38, 59, 64, 67 Music and verbal myth, 293
Mirebalais (Haitian town), 281-5 Mutacalombo (Congo-Brazilian
Misfortune, in Bantu thought, and inkissi), 276
distinction of "natural" and Mvila, Luvila (clan), 38-9;
"human" cause, 13 becoming dominant regional
Moanzi (Loango nkisi), 53 identity, 44
Mongo Luala (village), 2 Mwana (child, pi. bana), 41-2
Moni-Mambu (trickster in Kongo See also Bana bambuta; Se
mythology), 78, 97,103,233, Mwembe Nsundi (village), 4-5,
258-72passim., 277, 291, 62-8
297, 301, 306, 309
Montesarchio, J., 62, 64, 338 Na Menta (Nsundi king), 65
Moreau de St. Méry, 278-9 Nansi (Woyo shrine, ruler's
Mourne Rouge (town), 278 charm), 47
Moya (Zambian cult of affliction), See also Bunzi
19 Nantes, 278
Mpangu (people, ethnic label), 2, N'daiye, vi
55 Ndakivangi, P., 344
Mpanzu (clan), 65 Ndembo (Kongo, Yombe
Mpanzu a Kitima, 63 initiation rite), 10-11
Mpemba (as midwifery cult), 56 Ndembu (people, Zambia), 12-13;
See also Pfemba; Buanga; drums of affliction, 12-13
Bingo Ndibu,J.,87-90,338
Mpemba Kasi (Nsundi district), Ndibu (people, ethnic label), 2
62,64 Ndongo (people, ethnic label, also
Mpinda (Loango nkisi), 56 Mondongo), 37
Mpodi (Loango, Kongo nkisi), 56 Ndunga (village), 37
Mpu (chiefly title, rulership nkisi), Newman, S., 334
66,69,70,71-2,77 Nga Malamu (legendary Lemba
See also Mfumu priest, founder), 160,167,176,
Mpudi a Zinga (clan), 44 179-85passim., 249, 274, 291,
Mpulu-Bunzi (Woyo, Yombe 297-8
Index 377
Ngambula (Lemba priest, chief), Njambe (Loango nkisi), 55

83-5,247,250 Nkamba (village, center of
Nganda (village, mission), 2 Kimbanguist shrine), 90
Nganga (doctor, priest, magician, Nkanga Nsundi (Nsundi district),
pi. banganga), in Bantu 62,64
therapeutic systems, 14-15; N'khimba, see Khimba
depicted in myth, 214, 259; in Nkira (spirit, Teke domainal
Lemba initiations, and role squire's insignium), 59-61
definitions, 106-30 passim., Nkisi (Kongo consecrated
158-63 passim., 234-42 medicine, spirit; pi. min'kisi),
passim.; rules against abuses 4; array of, in Lower Congo,
by nganga Lemba, 127 in 1900,15; in Brazil
Ngobila (Teke administrator of (inkisses), 273-7; code of
Zaire river traffic), 61 composition of, 101;
Ngoma, R, 316-17, 347 composition of Lemba as,
Ngoma (drum as instrument, as 172-3,140-3,195-9, 251-8;
association), v, 107,179,185, Lemba as major, 4; in
257; Lemba, PL 14; Loango history, 51-8
medicines drummed up with, See also Expressive domains;
15, 315 Handa; Medicine; Plants
See also Drum; Nkonko; Nkita (cult, drum of affliction), 15,
Nkonzi; Sika 17,84 '
Ngombe a Kinu (Nsundi district), See also Kinkita; Simbi
62,64 Nkobe (generic Kongo term for
Ngombe Matadi (Nsundi town), cylindrical medicine chest as
66 used in Lemba), 140-3,173-6,
Ngombi (village), 37 195-201, 251-5; Pis. 4-7;
Ngombo (divination technique, composition of in trickster
nganga specialty), 14, 55 narrative, 259-66; relation to
Ngoyo (kingdom), 4-5,18, 20, 34, ritual, 195-201, to fixed
37, 45-58passim., 248, 252, shrine, 243-5
301-302 See also Medicine; Nkisi
See also Woyo, Cabinda
Nkobi (Tio administrative reform,
Ngoyo (market and village on
ruling ideology, charm
Zaire river), 30,35,54
chest), 58-61 passim.
Ngubane,H.,331
Ngunu (historic kingdom), 45 N'kondi (Kongo, Yombe nkisi),
Nguru (village), 37 15,17,18,21,70,76-7
Ngusa, R., 341 Nkonko (drum), 3,114,147,179,
Nietsche, R, 293, 310, 346 234, 257
Nigeria, 6 Nkonzi (drum) in Lemba, 107, 111,
Nimi a Lukeni (Kongo king), 63 114,117-19,142,147, 152,179,
Nitu,S.,v, 207-31,342 182,185,192, 257
378 INDEX
N'kwe Mbali (earth spirit patron Pongo, Nzambi a Mpungu,

ofTio king), 18,21,56,79 274, 276, 301-302
Nsaku (Kongo clan, also Nsaku See also Nza; God; Kuiti-Kuiti
Lau, Nsaku ne Vunda), 66 Nzila (Zambian cult of affliction),
Nsanga (Nsundi district), 62-4, 66 6,17,19
Nseke Mbanza (village), 2, 73, Nzinga a Mpemba (Nsundi,
85-7, 187-200 Kongo king), 63
Nsemi, I. (Kongo catechist- See also Affonso
ethnographer), 332, 334 Nzinga a Nkuwu (Kongo king),
Nsi (domain, local district, also 63
ntsii (Teke), tzi (Yombe)), 44, Nzola (Yombe nkisi), 247
51, 54, 58, 61; mfumu nsi Nzonzi (speaker, advocate), 70
(domainal chief), 44,70-1,77; Nzuzi,P.,86, 338
nkisi nsi (domainal ruler's
charm), 48; nkazi a nsi Obessem (Yoruba-Brazilian
(counterpart clan to royalty), oris ha), 276
65 Oceania, 11
Nsundi (clan, ethnic label, Ode (Yoruba-Brazilian oris ha),
province, chiefdom), 2,4-5, 276
37,38,44-5,61-70, 73,77, Ogun (Yoruba-Brazilian oris ha),
79, 95,157,187-200passim. 276, 282, 284
Ntanda Kongo (Nsundi district), Oia (Yoruba-Brazilian oris ha),
62,64 276
Ntete (Nsundi chief), 66 Olissassa (Yoruba-Brazilian
Ntele (village), 2 oris ha), 276-7
Ntini a Mongo (ceremonial style, Olorun (Yoruba-Brazilian oris ha),
after region), 96 276-7
Ntinu (supreme judge, king), 46, Omolü (Yoruba-Brazilian orisha),
65, 70; in Nsundi, 65 276
Ntotila (king), 46, 66 Orishas (deities of Yoruba
Ntsaa (plain in Teke area), 60 religion), influence in New
Nunvurucomabuva (Congo- World, 273-7
Brazilian inkissi), 276 Orley, J., 331
Nza (Tio supreme creator), 60 Ossain (Yoruba-Brazilian orisha),
See also Nzambi 276
Nzabi (village), 37 Our Lady, in Afro-Brazilian
Nzambi (Western Bantu Supreme religion; of Carmel; of
Being, God, Creator), 87, Compassion; of the
192, 301-302; in Brazil, 276; Immaculate Conception; of
as creator, 46; in Lemba the Immaculate Conception
lyrics, 184-5,192, 201; of the Beach; of Lourdes and
Nzambi Mpungu Tulendo, Candlemas; of Navigators; of
192; as source of Lemba Pleasures; the Rosary; of
medicine, 258-67; Zambi a Sorrows; 276
Index 379
Oxalä (Yoruba-Brazilian orisha), use of, on tombs, 89

276-7 See also Medicine; Nkisi
Oxössi (Yoruba-Brazilian orisha), Pointe Noire (city), 2-3
276 Pombieros (tra3ers)73^^
Oxun-Mare (Yoruba-Brazilian Pongo (Loango, Kongo nkisi), 53,
orisha), 276 275, 277
Poro (ceremonial society), 6
Panda (Congo-Brazilian inkissi), Portugal, Portuguese, 6, 34, 37,
276-7 63-4, 81
Pansa (Loango, Kongo nkisi), 277 Price-Mars, J., 103,188,278,
Paul, St., in Afro-Brazilian 286-92, 345
religion, 276 Prins,G.,331,332,347
Pechuel-Loesche, E., 2, 56, 337, Proyart, A., 337
343 Psychological, explanations of
Peirce, C , 340 secret societies, 9; processes
Pelton, R.,344 in Lemba healing, 311-13
Pernambuco (city), 276 Purification, purity, as
Peter, St., in Afro-Brazilian administered with mpolo-
religion, 276 Lemba, 106, bcrco-Lemba,
Petion (Haitian president), 278-9, 158, mwmo-Lemba, 236-7,
288, 292, 298 and in food-medicine
Petit-Goäve (town), 279 mixture, 173-4; as concept of
Petro (category of Haitian voodoo redemption, 17,20; with
loa linked to Kongo and cupping horn, 15; as indicated
Lemba), 273, 277-92 passim by Lemba paddles, 126-7; of
See also Jean Petro; Rada Lemba couple, 125-6, 240-1;
Pfemba (Loango, Woyo nkisi, of Lemba neophyte,
midwife's group, part of following profanation, 110-15,
Lemba nkobe), 22, 56, 253, 240-1; logic of, in Lemba,
301,343 319; as ritual action, 144;
See also Bingo; KaBuanga following sickness of
Pinkerton, J.,336, 337 "human" cause, 13-14
Plants, depicted in myth, 202,
217-18, 265; in food crops Quincongo (Congo-Brazilian
originating in Africa, 28-9, inkissi), 276-7
America, 29, 31, Malaysia,
28, 31; in Lemba, 57,109,117, Rada (category of Haitian voodoo
119,137,143,173-6,196, 237, loa linked to Fongbe origins),
239-40, 252-3, 265; liminality 277,281,284-5
of, 245; in markets, 30, 72; in See also Arada
metaphoric correspondence Radin,P.,271, 344
with other elements, 255-6; Railroad, Matadi to Leopoldville,
planted in shrines, 243-5; and 81; completion of and
ritual transformations, 300; Lemba's demise, 81-2
380 INDEX
Ramos, A., 273, 344 185, 256; in myth, 223-30,

Randies, W., 335 259-65
Ranger, T.,20, 333 See also Kise; Mwana
Rappaport, R., 340 Sebastian, St., in Afro-Brazilian
Recife (city), 276 religion, 276
Renkin, PL 12 Secret societies, category of
Rey, P.P., 39, 336 scholarship, 9-12; as political
Rigaud (Haitian president), 288, force, 12; as threat to
292,298 European order, 11
Ritual, analyzed in use of Seha, M.,P1.16
"expressive domains" and Semiotics, in study of multiple
"codes," 98-101; categories expressive modes, 100
of ritual action in Lemba, See also Expressive domains
143-6,195-201,248-51; Sevo (Nsundi district), 62, 64
defined, 95; power generated Shangö (Yoruba-Brazilian
in, 174; as understood in orishä), 276
symbolist school, 99, by Shilluk (people), 16
structuralists, 99-100, in Shona (people), 16
semiotics, 100; Simbi (water, nature spirits), 44,
transformations in Kongo 274, 282-3, 285
ritual, 297-300 Simbi djo (Haitian "Petro" loa),
See also Expressive domains; 284
Codes Simbi en Deux Eaux (Haitian
Robertson-Smith, 19 "Petro" loa), 284
Roch, St., in Afro-Brazilian Simbi Kita Kita (Haitian "Petro"
religion, 276 loa), 283
Roche Mucumbe (Angola- Simbu (Yombe nkisi shrine), 247
Brazilian inkissi), 276 Simbuka (Loango nkisi), 55
Rymer, J. and G., PI. 20 Slave, slavery, capture as
hostages in feud, 74-6;
San Salvador (capital city), 29 importance of, in "lineage
See also Mbanza Kongo mode of production," 39; in
Sande (ceremonial society), 6 Lemba membership, 77, 320;
Santo Domingo, 278 in payment of Lemba dues,
Sasi (Loango nkisi), 56 114,128; portrayed in myths,
Schneider, D., 339 219; status indicated by
Schoffeleers,M., 17-18,332 ignorance of clan identity, 38;
Schöller, C , 343-4, 346 in trade and exchange, 3, 28,
Schurtz, H.,9,331 31,33-4,84,274,319
Se (father), in kinship, 40, 43; in Smith, M.G., 21, 334
Lemba patrifiliality within Sobisa (Lemba priest), 83, 85-6
healer-sufferer (Se-Mwana) Sobo, Sogbo (Fongbe-Brazilian
relationship, 106-24 passim, vodun), 276
139-40,153-4,159-63,169, Social control, and legitimacy,
Index 381
320; in Lemba governance, symbolism, 243-5; of food,

3-4; in market laws of 134, color, 254, hunting, 220,
acephalous area, 70-9; medicine, 324, menstruation,
needed in face of 178, power, 324;
international trade, 3; and manipulation of, to define
therapy of iatrogenic ritual space, 143; polysemy,
diseases, 316-17 contradictory strands of
Sociocultural change, and meaning in, 99,148,166;
changing modes of sensory and cognitive
production, 19; and combined in study of, 99,148;
divinatory interpretation of symbolist approach and
misfortunes, 14; independent criticism of, 99
and dependent variables in
study of, 7, 8,19-20, 96, 326 Taino (historic Indians of Haiti),
Solongo (region, people), 49-50 277
Songe (people), 37 Tallensi (people), 16
Songho (copper mine), 32 Tastevin,C.,336, 345,346
Sonyo (region, people), 49-51, 65 Teke (people, region), 2,5, 21,
Soret, M., 38, 335 27-8, 32, 37-8,45,58-61,79,
South Africa, 6 97, 177
Southall, A.,333 See also Tio
Spain, 277 Tervuren (Central Africa
Sperber, D., 339 Museum), vi, 2, Pis. 4-6,
Stanley, H.,335 12-15
Stebich, U.,344 Theories, in ethnology and
Stenstrom, O., 2,157-85, 338, history, 7-23; an explicit
341,347 Lemba theory of
Structuralism, vi, 8, 99, 305; therapeutics and social order,
opposition and contradiction 315-21; in study of ritual,
in, 311 98-101
Sudan, 16 See also Ideology
Suki (village), 37 Therapeutics, 318-21; Bantu
Sundi, see Nsundi therapeutics and generation
Surinam, 19 of new medicines, 14-15;
Svenska Missionsforbundet, Pis. Lemba ideology of; Kongo
2,3,17 therapeutics as system,
Swantz, M., 331 14-15; as metaphor of social
Swedish, 86 formation, 20-1; as self-
Symbols, symbolism: adaptation conscious social process,
of conventional symbolism in 7-8, and corporate politics, 8;
Lemba, 251, and variance of structure and movement in
symbolic forms, 96; Lemba therapeutics,
documentary symbolism in 293-313; and taxonomies of
nkobe, 174; fixed ritual illnesses, medicines, 14-16
382 INDEX
Thiel, J., vi, 298, 346 Vansina, J., vii, 5, 21, 32, 36,
Thiriko (Loango nkisi), 52, 54, 277 58-61 passim., 332, 333, 335,
Thompson, R., vii, 275, 344, 345 336,337,338,347
Ti Jean Pie Sèche (Haitian Van Velzen,B., 19-20, 333
"Pétro" loa), 284 Van Wing, J., 12, 331, 338, 342
T i Kita Démembré (Haitian Vaz, J.,2, 343
"Pétro" loa), 282, 284 Veistroffer, 48
Tio (kingdom), 21, 32, 36, 37, Vercraeye, J., 338
58-63 passim., 79 Vili (people), 2, 32, 37,40
Toussaint F Ouverture (Haitian Vinda (clan), 83-4
president), 278, 288, 292, 298 Visser, R., 2, 48, 336, Pis. 9-11,15
Trade, 19, 21,29-30, 36,53, PI. 7; Voduns (Fongbe spirits, deities),
beads, 252-3, PI. 7; influence in New World,
ceremonial trade societies, 273-7
v-vi; copper trade, 32, 58; Voodoo (Haitian religious form,
goods as nkisi, 52; legitimate also Vodun, Voudou), 273-92
trade, 3,29-31; medicines, passim.
56-7, Pis. 18,19; networks Vungu (kingdom, region), 4-5, 45,
weakened by endogenous 61-3
marriage, 229, renewed by
exogenous marriage, 6; Wamba dia Wamba, vii
routes, 3, 5,58; slave, 3, 28, Wembo (Nsundi district), 62, 64
31, 319; units of exchange and Werbner, R., 18-19, 331, 332, 333
currency in, 30-35 Westermann, D., 335
See also Market Weule, 9
Tsanga (village), 2 Widman, R., vi
Tschivuku (Loango nkisi), 56 Winnebago (people), 271
Tseke Banza (Mayombe town), 48 Wissler, C.,335
Tshela (town), 2 Witte, H., vi
Tshimona-Mambu (trickster), see Woyo (people, Western Lemba
Moni-Mambu region), 2, 247-58 passim.
Tshimpuku (Loango nkisi), 55 See also Ngoyo
Turner, V., vii, 10,12-13,15-16, Wundt, W.,9, 331
18,20,130,331,332,339,341
Yaa (people, ethnic label), 2, 35
Umbanda (Afro-Brazilian cult Yalala Songo (legendary lake in
form), 275 Congo, Nzadi, river), 302
United States, Kongo culture in, Yansan (Yoruba-Brazilian
273 orisha), 276
Yansi (people), 37
Van Binsbergen, W., 17,19, 332, Yemanja (Yoruba-Brazilian
333 orisha), 276
Van der Gelwe, vi Yoder, S., vii
Van Gennep, A.,9-10, 331 Yombe (people of Western Lemba
Index 383
region), 2, 37-8,40,83, Zambi (Western Bantu high God),

233-58 passim. 274; a mpongo, 276-7
See also Mayombe See also Nzambi
Yoruba, influences in New World, Zambia, 6,17,19
273-7 Zimba, (Nsundi district), 64
Zimbabwe, 16
Zaire (river, country, from Nzadi, Zinga (village), 37
great river), 2-3, 4, 27-8, 37, Zulu (people), 2
58, 62-3 Zwerneman, J., vi

Egbe Lembá

Uploaded by

Copyright:

Available Formats

Egbe Lembá

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Egbe Lembá

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

What regions and peoples are discussed in the book?

What regions and peoples are discussed in the book?

Lemba, 1650-1930

Assistant Advisory Editor

Garland Publishing, Inc. N e w York and London

Library of Congress Cataloging in Publication Data

Printed on acid-free, 250-year-life paper

Chapter 1: Introduction to Lemba

Part I: The Public Setting of Lemba: Management of Society

Part II: The Rituals of Lemba: Management of Reality 93

Chapter 4: The Northern (Kamba, Yaa) Variant of Lemba 105

Chapter 5: The Eastern (Lari) Variant of Lemba 157

Chapter 6: The South-Central (Nsundi, Bwende) Variant

Chapter 7: The Western (Yombe, Woyo, Vili) Variant

Part III: Structure and Movement in Lemba Therapeutics 293

1. Loango, Mid-Seventeenth Century

1. Regional Distribution of Lemba, based on Ethnographie,

22. Dominant Metaphor, Western Variant 256

Friends and colleagues alike have encouraged me to declare openly

Stockholm and Lidingo; and the personnel of the Jesuit Institute of

E = Ethnography B-D = Berlin-Dahlem Museum für

Lemba, a major historic cult of healing, trade, and marriage relations,

engaged in mercantile work. Their ability to succeed in commerce and

The distinctiveness of the Lemba system of government compared

tion in handling chronic afflictions not dealt with by biomedical

Issues in the Study of Lemba of Theoretical Significance

The study of a major historical drum of affliction of Equatorial Africa

The present volume brings several analytic perspectives to the task

applications to Lower-Congo and Equatorial-African societies seems

needs and instincts, or too nominalist, forcing data into abstract

axe, in short, a significant political force, regulating justice, policing

proceeds with an attempt to remove the source of misfortune by

through sacrifices and confessions so as to achieve reintegration with

natural or human cause, or another, spirit-related, cause. Because of

taxonomies indicate the contours of the system of dealing with

characterized by overlapping and interpenetrating fields of ritual

feleers' study of the Mbona cult of Malawi also emphasizes these

life. Thirty to forty percent of goods received were channeled to

response a creative, positive effort in the formation of a "new

Corporate theory, finally, distinguishes between corporate groups,

The Public Setting of Lemba :

"Les auteurs o n t . . . des notions très vagues sur les populations de la

Economic, Social, and

Environment, Resource Use, Adaptation,

constituted the staple foods until the seventeenth century. Portu­

area has to be limited in this study to the period from approximately

a large antelope, gun powder

In another market at Ngoyo, upstream from Manianga, in 1877, local

These items fill the major categories of resources in the Lemba

relate to metalworking (copper wire, lead, gunpowder), and cloth

The mitako introduced by European traders had by 1880 become

economies; a given commodity or resource is said to be specific to

compared to yet other types of transactions at the turn of the century.

Three centuries of trade with the coast established caravan routes

hundreds of thousands of slaves and drove them over these routes to

The major alternative principle of recruitment is, of course,

patrifilial/matrilineal dichotomy is incorporated into the immediate

large clientele could be assembled in short order by the establishment

constituted the staple foods until the seventeenth century. Portu

The literature offers several alternative origins of the coastal king

The literature offers several alternative origins of the coastal king

broadening wealth in coastal societies following the complete col

An institution as widespread in scale and as deep in historical dura

meticulously detail the type and quantity of economic object ex