f
VANGUARD BO
(Pty.). Ltd.
28, Joubert Streel
.fdfl Jl^f-^
THE NUER
OXFORD UNIVERSITY PRESS
AMEN HOUSE, E.C. 4
London Edinburgh Glasgow New York
Toronto Melbourne Capetown Bombay
Calcutta Madras
HUMPHREY MILFORD
PUBLISHER TO THE UNIVERSITY
t? h-p:)^'
I
m
r
Section of homestead and kraal (Eastern Gaajok)
THE NUER
A DESCRIPTION OF
THE MODES OF LIVELIHOOD AND
POLITICAL INSTITUTIONS OF
A NILOTIC PEOPLE
BY
E. E. EVANS-PRITCHARD
M.A. (OXON.), PH.D. (LONDON)
RESEARCH LECTURER IN AFRICAN SOCIOLOGY AT THE
UNIVERSITY OF OXFORD
SOMETIME PROFESSOR OF SOCIOLOGY AT THE
EGYPTIAN UNIVERSITY, CAIRO
OXFORD
AT THE CLARENDON PRESS
1940
PRINTED IN GREAT BRITAIN
TO
THE STAFF OF THE
AMERICAN MISSION
AT NASSER
'Ah, the land of the rusthng of wings, which
is beyond the rivers of Ethiopia : that sendeth
ambassadors by the sea, even in vessels of
papyrus upon the waters, (saying) Go ye swift
messengers, to a nation tall and smooth, to
a people terrible from their beginning onward ;
a nation that meteth out and treadeth down,
whose land the rivers divide.'
{The Holy Bible (Revised Version),
Isaiah xviii. 1-2.)
PREFACE
My study of the Nuer was undertaken at the request of, and
was mainly financed by, the Government of the Anglo-Egyptian
Sudan, which also contributed generously towards the publica-
tion of its results. Part of the inquiry was carried out as a
Leverhulme Research Fellow. To the Sudan Government and
to the Leverhulme Research Fellowships Committee I make
grateful acknowledgements.
I owe Professor and Mrs. C. G. Seligman a great debt for
their friendship during the last fifteen years. Without their
backing and encouragement this book might not have been
written. Moreover, although they made no investigations
among the Nuer, their brilliant researches among other Nilotic
peoples, particularly the Shilluk and Dinka, laid the foundations
of all future studies in these regions. '
I thank all those in the Sudan, at Khartoum and in Nuerland,
who have given me hospitality and assistance ; Sir John Maffey,
then Governor-General; Sir Harold MacMichael, then Civil
Secretary ; Mr. and Mrs. S. Hillelson ; Mr. C. A. Willis, Mr. A. G.
Pawson, Mr. M. W. Parr, and Mr. E. G. Coryton, who were in turn
Governors of the Upper Nile Province; Mr. P. Coriat, Capt.
A. H. A. Alban, Capt. H. A. Romilly, Mr. J. F. Tierney, the
late Mr. L. F. Hamer, Mr. B. J. Chatterton, Mr. B. A. Lewis,
and Mr. F. D. Corfield, all of whom were at one time Com-
missioners of Nuer Districts. To Mr. F. D. Corfield, amico el
condiscipulo meo, I am especially grateful for the interest he has
shown in my work and for his generosity in allowing me to use
many of his fine photographs.
I thank also the staff of the American Mission at Nasser, of
the Congregation of Verona at Yoahnyang, and of the Church
Missionary Society at Ler. I wish to make particular acknow-
ledgement to the staff of the American Mission, especially to
Miss B. Soule, who unreservedly placed their home, their time,
and their knowledge at my disposal. I dedicate this book to
them not only as an expression of personal gratitude, but also
as a tribute to their devoted service to the Nuer.
' Pagan Tribes of the Nilotic Sudan, by C. G. and B. Z, Seligman, 1932.
viii PREFACE
My warmest thanks are further rendered to the many Nuer
who made me their guest and befriended me. Rather than
speak of individuals, I express my general respect for this brave
and gentle people.
The following friends and colleagues have read this book and
have given me valuable criticism and advice: Professor C. G.
Seligman, Professor A. R. Radcliffe-Brown, whose influence on
the theoretical side of my work will be obvious to any student
of anthropology, Dr. M. Fortes, and Dr. H. M. Gluckman. I
owe a special debt to Dr. Fortes. My ideas about the aims
and methods of Social Anthropology have been influenced by
the many talks we have had on the subject during several years
of comradeship and, since in such a relationship it is not easy to
state what one has taken and given, I acknowledge unreservedly
that I have been greatly stimulated by our discussions.
Professor Seligman has pointed out to me, in reading the
proofs, that my use of 'horticulture' and 'horticultural' is
unusual. I did not intend to depart from conventional usage.
However, I did not feel justified in altering these words to 'agri-
culture' and 'agricultural' throughout the book in the present
difficulties of publication. Readers who prefer 'agriculture'
and 'agricultural' can make the substitution for themselves.
A considerable part of the facts related in this book have been
previously recorded, chiefly in Sudan Notes and Records and
Africa, and I thank the editors of these journals and the editor
of Custom is King for permission to republish them. I am
indebted also to the editors and printers of both journals, to
George Routledge & Sons, Ltd., and to Messrs. Hutchinson &
Co., for the use of photographic blocks.
Several friends have lent me photographs, sketch-maps, and
diagrams. These are acknowledged in the list of plates and
figures, but I desire to record expressly my gratitude to Mr. F. D.
Corfield, Dr. H. E. Hurst, Director of the Physical Depart-
ment of the Egyptian Government, Mr. B. A. Lewis, Mr. C. L.
Armstrong, the staff of the American Mission, Nasser, the late
Mr. L. F. Hamer, Dr. E. S. Crispin, and Yuzbashi Talib Ismail.
My thanks are due to Mr. W. R. Key for his many secretarial
services in the preparation of this volume.
E. E. E.-P.
January 1940
CONTENTS
LIST OF PLATES
LIST OF MAPS AND TEXT FIGURES
INTRODUCTORY
I. INTEREST IN CATTLE
^. OECOLOGY
\pi. TIME AND SPACE
IV. THE POLITICAL SYSTEM
V. THE LINEAGE SYSTEM
VI. THE AGE-SET SYSTEM. SUMMARY
INDEX ....
XI
I
i6
51
94
139
192
249
267
LIST OF PLATES
Section of homestead and kraal . . Frontispiece
I. Youth .... facing page 8
II. Girl in kraal [Coy field) . . . .16
III. Milking a restless cow . . . .22
IV. Ox . . . . . . .30
V. Girl milking . . . . . .36
VI. Typical savannah . . . . .54
VII. a. Homesteads on mound, b. Homesteads on mound . 62
VIII. Sandy ridge . . . . . .68
IX. Harpoon-fishing from canoe (Corfield) . . .72
X. Harpoon-fishing in shallows (Corfield) . . .78
XI. a. Savannahin the dry season. 5. Clearing millet garden 84
XII. Millet garden in October . . . .90
XIII. Girl in millet garden ..... 104
XIV. August shower . . . . .120
XV. a. Windscreen, b. Well . . . -134
XVI. Air- view of villages {Royal Air Force) . . 146
XVII. Boy collecting dung-fuel .... 156
XVIII. Building a cattle byre {Corfield) . . . 166
XIX. a. Cattle camp. b. Typical swampy depression . 176
XX. Cattle travelling . . . . .186
XXI. a. Cattle grazing on ridge, b. Cattle camp . . 200
XXII. a. Spearing fish from dam {Hamer). b. Harpoon-
fishing in lake ..... 210
XXIII. a. Section of camp kraal, b. Sobat river in the dry
season . . . . . .216
XXIV. A leopard-skin chief {Corfield) . . . 222
XXV. a. and b. Ngundeng's pyramid {Crispin) . . 228
XXVI. a. Youth, b. Youth {Corfield) . . .236
XXVII. Initiation of boys {American Mission) . . 250
XXVIII. Three Nuer men {a. and c. Talib Ismail) . . 256
XXIX. Man {Nasser Post) . . . . .262
Plates VII, VIII, XV, XXI, and XXVI are from blocks in Sudan Notes
and Records; Plates XXV and XXVIII from C. G. and B. Z. Seligman,
Pagan Tribes of the Nilotic Sudan (George Routledge & Sons, Ltd.) ;
and Plate XXIII from Custom is King (Messrs. Hutchinson & Co.).
Plate XVI is from a print supplied by the Physical Department of the
Egyptian Government.
MAPS AND TEXT-FIGURES
Map. Approximate area occupied by the Nuer . P<ig^ xii
Map. The Nuer and neighbouring peoples . . facing page 4
Map. Distribution of the larger Nuer tribes . . pcig^ 8
Churning gourd . . . . . .24
Gourd for storing cheese . . . . .27
Bags . . . . . . -29
Ox-bell and collar . . . . . -32
5. Stuffed calf's head . . . . . .34
6. Calf's weaning ring . . . . . • 35
7. Mud figures of oxen . . . . -39
8. Colour distributions . . . . .42
9. Colour distributions . . . . .43
Temperature and Rainfall estimates for Nuerland . . 52
Rise and fall of the Sobat river . . . -53
Sketch-map. Seasonal movements of the Lou tribe . . 56
Sketch-map. Seasonal movements of the Jikany tribes (after
Mr. C. L. Armstrong). . . . . • 5^
Sketch-map. Seasonal movements of the Zeraf tribes (after
Mr. B. A. Lewis) . . . . . .60
0. Instruments for attracting fish . . . -71
1. Horn and ebony spears . . . . .115
Map. Tribal distribution about i860 (after V. A. Malte-Brun) 129
Baked-mud grindstone . . . . .170
13. Calf's bell-necklace . . . . . .187
Sketch-map. The Eastern Jikany tribes (after Mr. C. L. Arm-
strong) . . . . . . .233
14. Buffalo-horn spoons . . . . -237
15. Leather flail ...... 246
Figs, i, 5, 6, 7, 10, and 15 are from drawings of specimens which form
part of the author's collection in the Pitt-Rivers Museum, Oxford, and
Figs. 2, 3, 4, II, 12, 13, and 14 are from drawings which form part of
the author's collection in the University Museum of Archaeology and
Ethnology, Cambridge.
12
'SfitV'tUINEA--
Map showing approximate area {shaded) occupied by the Nuer.
INTRODUCTORY
I
From 1840, when Werne, Arnaud, and Thibaut made their ill-
assorted voyage, to 1881, when the successful revolt of the
Mahdi Muhammad Ahmed closed the Sudan to further explora-
tion, several travellers penetrated Nuerland by one or other of
the three great rivers that traverse it : the Bahr el Jebel (with
the Bahr el Zeraf), the Bahr el Ghazal, and the Sobat. I have
not been able to make much use of their writings, however,
for their contact with the Nuer was slight and the impressions
they recorded were superficial, and sometimes spurious. The
most accurate and the least pretentious account is by the
Savoyard elephant-hunter Jules Poncet, who spent several
years on the borders of Nuerland.'
A later source of information about the Nuer are the Sudan
Intelligence Reports which run from the reconquest of the Sudan
in 1899 to the present day, their ethnological value decreasing
in recent years. In the first two decades after the reconquest
there are a few reports by military officers which contain
interesting, and often shrewd, observations.^ The publication of
Sudan Notes and Records, commencing in 1918, provided a new
medium for recording observations on the customs of the
peoples of the Anglo-Egyptian Sudan, and several political
officers contributed papers on the Nuer. Two of these officers
were killed in the performance of their duty. Major C. H. Sti-
gand by the Aliab Dinka in 1919 and Captain V. H. Fergusson
by the Nuong Nuer in 1927. In the same journal appeared the
' Some of the writings from which I have derived information are
Ferdinand Weme, Expedition zur Entdeckung der Quellen des Weissen Nil
(1840-1), 1848; Hadji-Abd-el-Hamid Bey (C. L. du Couret), Voyage au Pays
des Niam-Niams ou Hommes a Queue, 1854; Brun-Rollet, Le Nil Blanc et le
Soudan, 1855; G. Lejean, Bulletin de la Socieid de Geographie, Paris, i860;
Jules Poncet, Le Fleuve Blanc [Extrait des Nouvelles Annales de Voyages,
1863-4) >■ Mr. and Mrs. J. Petherick, Travels in Central Africa, 1869 ; Ernst Marno,
Reisen im Gebiete des blauen und weissen Nil, ini egyptischen Sudan und den
angrenzenden Negerldndern, in den Jahren i86g bis 1873, iSy^. Others are
mentioned later, particularly on pp. 126-7 ^i^d 134.
^ These reports were used by Lieut. -Colonel Count Gleichen in his compila-
tion: The Anglo-Egyptian Sudan, 2 vols., 1905.
4507 B
2 INTRODUCTORY
first attempt, by Mr. H. C. Jackson, to write a comprehensive
account of the Nuer, and great credit is due to him for the man-
ner in which, in spite of serious obstacles, he carried it out.'
After I had begun my researches a book by Miss Ray Huff-
man, of the American Mission, and some papers by Father J. P.
Crazzolara, of the Congregation of Verona, were pubhshed.^
Although my own contributions to various journals are re-
printed, in a condensed form, in this book, or will be reprinted
in a subsequent volume, I allude to them here so that the reader
may have a complete bibliography. I have omitted much detail
that appeared in these articles. 3
Lists of a few Nuer words were compiled by Brun-RoUet and
Marno. More detailed vocabularies have been written by Major
Stigand and Miss Huffman, and grammars by Professor Wester-
mann and Father Crazzolara, Professor Westermann's paper
contains also some ethnological material.'*
' Major C. H. Stigand, 'Warrior Classes of the Nuers', S. N. & R., 1918,
pp. 1 16-18, and 'The Story of Kir and the White Spear ', ibid., 1919. PP- 224-6;
Capt. V. H. Fergusson, 'The Nuong Nuer', ibid., 1921, pp. 146-55, and
'Nuer Beast Tales', ibid., 1924, pp. 105-12; H. C. Jackson, 'The Nuer of the
Upper Nile Province', ibid., 1923, pp. 59-107 and 123-89 (this paper was
reprinted as a book under the same title by El Hadara Printing Press, Khar-
toum, no date, and contained a terminal essay of 23 pages by P. Coriat on
'The Gaweir Nuers').
^ Ray Huffman, Nuer Customs and Folk-lore, 1931, 105 pp. ; Father J. P.
Crazzolara, 'Die Gar-Zeremonie bei den Nuer', Africa, 1932. PP- 28-39,
and 'Die Bedeutung des Rindes bei den Nuer', ibid., 1934. PP- 300-20.
3 E. E. Evans-Pritchard. 'The Nuer. Tribe and Clan', S. N. &■ R., 1933,
pp. 1-53, 1934, pp. 1-57, and 1935, pp. 37-87 ; ' The Nuer, Age-Sets ', ibid., 1936,
pp. 233-69; 'Economic Life of the Nuer', ibid., I937. PP- 209-45, and 1938.
pp. 31-77; 'Customs Relating to Twins among the Nilotic Nuer', Uganda
Journal, 1936, pp. 230-8; 'Daily Life of the Nuer in Dry Season Camps',
Custom is King, A Collection of Essays in Honour of R. R. Marett, 1936,
pp. 291-9; 'Some Aspects of Marriage and the Family among the Nuer',
Zeitschrift fiir vergleichende Rechtswissenschaft, 1938, pp. 306-92 ; 'Nuer Time-
Reckoning', Africa, 1939, pp. 189-216. The chapter on the Nuer (Chap. VI)
in Pagan Tribes of the Nilotic Sudan, by Prof. C. G. and Mrs. B. Z. Seligman,
1932, was compiled from my notebooks.
♦ Brun-Rollet, ' Vokabularien der Dinka-, Nuehr- und Schilluk-Sprachen ',
Petermann's Mittheilungen, Erg. II. 1862-3, pp. 25-30 ; Marno, ' Kleine Vocabu-
larien der Fungi-, Tabi-, Bertat- und Nuehr-Sprache ', Reisen im Gebiete des blauen
und weissen Nil, 1874, pp. 481-95 ; Professor Diedrich Westermann, 'The Nuer
Language', Mitteilungen des Seminars fiir Orientalische Sprachen, 1912, pp. 84-
141; Major C. H. Stigand. A Nuer-English Vocabulary, 1923, 33 pp.; Ray
Huffman, Nuer-English Dictionary, 1929, 63 pp., and English-Nuer Dictionary,
1931. 80 pp.; Father J. P. Crazzolara, Outlines of a Nuer Grammar, 1933,
218 pp.
INTRODUCTORY 3
II
I describe in this volume the ways in which a Nilotic people
obtain their livelihood, and their political institutions. The
information I collected about their domestic life will be pub-
lished in a second volume.
The Nuer/ who call themselves Nath (sing, ran), are round
about 200,000 souls and live in the swamps and open savannah
that stretch on both sides of the Nile south of its junction with
the Sobat and Bahr el Ghazal, and on both banks of these two
tributaries. They are tall, long-limbed, and narrow-headed,
as may be seen in the illustrations. Culturally they are similar
to the Dinka, and the two peoples together form a subdivision
of the Nilotic group, which occupies part of an East-African
culture-area the characteristics and extent of which are at
present ill-defined. A second Nilotic subdivision comprises the
Shilluk and various peoples who speak languages similar to
Shilluk (Luo, Anuak, Lango, &c.). Probably these Shilluk-
speaking peoples are all more alike to one another than any one
of them is to the Shilluk, though little is yet known about most
of them. A tentative classification may be thus presented :
EAST AFRICAN CULTURE TYPE
I I I
Other cultures Nilo-Hamites Nilotes
Shilluk-Luo group Nuer-Dinka group
I I
I III
Shilluk Shilluk-speaking Nuer Dinka
peoples
Nuer and Dinka are too much alike physically and their
languages and customs are too similar for any doubt to arise
about their common origin, though the history of their diver-
gence is unknown. The problem is complicated: for example,
the Atwot, to the west of the Nile, appear to be a Nuer tribe who
have adopted many Dinka habits,^ while the Jikany tribes of
' The word 'Nuer' is sanctioned by a century of usage. It is probably of
Dinka origin. I use it in singular and plural, speaking of 'a Nuer man'
and of 'the Nuer people'.
^ Poncet, op. cit., p. 54. They appear as Atot in the map on p. 129.
4 INTRODUCTORY
Nuerland are said to be of Dinka origin. Moreover, tliere has
been continuous contact between the two peoples that has
resulted in much miscegenation and cultural borrowing. Both
peoples recognize their common origin.
When we possess more information about some of the Shilluk-
speaking peoples it will be possible to state what are the defining
characters of Nilotic^ulture and sqcjal structure. At present
such a classification is exceedingly difficult and I postpone the
attempt, devoting this book to a p lain account of the Nuer a nd
neglecting the many obvious comparisons that might be made
with other Nilotic peoples.
■ Political "institutions are its main theme, but they cannot be
unde^tood wtffiout^taking into account environment and modes
of livelihood. I therefore_devote_the_earli.eii_£ar^ book
tcr a'des j^riptioj of th*^ rountry-4n- M4vi'^h th^ NufxJvv^~^Tifr?>f
how they obtainthe_^ecessit4es.£jliife^^ It will be seeri^^haLtKe
N uer political system is c oniistent with4bStf -<:)enalogy.
The groups chiefly dealt with in the later part of the book are
the people, the tribe and its segments, the clan and its lineages,
and the age-sets. Each of these groups is, or forms part of, a
segmentary system, by reference to which it is defined, and,
consequently the status of its members, when acting as such
towards one another and to outsiders, is undifferentiated.
These statements will be elucidated in the course of our inquiry.
We first^e.scribeJhejriterrelation of territorial segmentswithin
a territorial, or political, systenTand^TheiTthe relation of other
social systems to this system. What we understand by pohtical
structure will be evident as we proceed, but we may state as an
initial definition that we refer to relations within a territorial
system between groups of persons who live in spatially well-de-
fined areas and are conscious of their identity and exclusiveness.
Only in the smallest of these communities are their members
in constant contact with one another. We distinguish these
political groups from local groups of a different kind, namely
domestic groups, the family, the household, and the joint family,
which are not, and do not form part of, segmentary systems,
and in which the status of members in respect to each other and
to outsiders is differentiated. SociaHiesin domestic groups-are
primarily of a kinship order ,..and corporate life is normal.
i5
to
i5
iO
OINKA NUER SHILLUK
The Nuer and neighbouring peoples
AN u A K
INTRODUCTORY 5
The Nuer political system includes all the peoples with whom
they come into contactr^^5r 'people^^e'TTieaTrairpersons who
speaTc the same language and have, in other respects, tEe~same
culture, and consider themselves to be distinct from TiKe' aggre-
gates. The Nuer, the Shilluk, and the Anuak each occupy a
continuous territory, but a people may be distributed in widely
separate areas, e.g. the Dinka. When a people is, like the
Shilluk, politically centralized, we may speak of a 'nation '. The
Nuer and Dinka, on the other hand, are divided into a number
of tribes which have no common organization or central adminis-
tration and these peoples may be said to be, politically, a con-
geries of tribes, which sometimes form loose federations. ^Ihe^
Nuer differentiate those tribes which live in the homeland to the
west of the Nile from those which have migrated to the east of it.
We find it convenient to make the same distinction and to speak
of the Western Nuer and the Eastern Nuer. The Eastern Nuer
may be further divided, for descriptive purposes, into those
tribes which live near the Zeraf river and those which live to
north and south of the Sobat river.
The largest political segment among the Nuer is the tribe.
There is no larger group who, besides recognizing themselves as
a distinct local community, affirm their obligation to combine
in warfare against outsiders and acknowledge the rights of their
members to compensation for injury. A tribe is divided into
a number of territorial segments and these are more than
mere geographical divisions, for the members of each consider
themselves to be distinct communities and sometimes act as
such. We call the largest tribal segments 'primary sections ', the
segments of a primary section 'secondary sections', and the
segments of a secondary section 'tertiary sections'. A tertiary
tribal section consists of a number of villages which are the
smallest political units of Nuerland. A village is made up of
domestic groups, occupying hamlets, homesteads, and huts.
We discuss the institution of the feud and the part played in
it by the leopard-skin chief in relation to the political system.
The word 'chief may be a misleading designation, but it is
sufficiently vague to be retained in the absence of a more suit-
able English word. He is a sacred person without political
authority. Indeed, the Nuer have no government, and their
6 INTRODUCTORY
state mightjbejiescxibed as an ordered anarchy. Likewise they
lack law, if we understand by this term judgements delivered by
an independent and impartial authority which has, also, power
to enforce its decisions. There are signs that certain changes
were taking place in this respect, and at the end of the chapter
on the political system we describe the emergence of prophets,
persons in whom dwell the spirits of Sky-gods, and we suggest
that in them we may perceive the beginnings of political
development. Leopard-skin chiefs and prophets are the only
ritual specialists who, in our opinion, have any political im-
portance.
After an examination of the political structure_we^ describe
the lineage system and discuss the relation between the two.
Nuer lineages are agnatic, i.e. they consist of persons who trace
their descent exclusively through males to a common ancestor.
The clan is the largest group of lineages which is definable by
reference to rules of exogamy, though agnatic relationship is
recognized between several clans. A clan is segmented into
lineages, which are diverging branches of descent from a common
ancestor. We call the largest segments into which a clan is
divided its 'maximal lineages', the segments of a maximal
lineage its 'major lineages', the segments of a major lineage its
'minor lineages', and the segments of a minor lineage its 'mini-
mal lineages'. The minimal lineage is the one to which a man
' usually refers when asked what is his lineage. A lineage is thus
a group of agnates, dead or alive, between whom kinship can be
traced genealogically, and a clan is an exogamous system of
lineages. These lineage groups differ from political groups in
that the relationship of their members to one another is based
,^ on descent and not on residence, for lineages are dispersed and
do not compose exclusive local communities, and, also, in that
lineage values often operate in a different range of situations
from political values.
After discussing the lineage system in its relation to terri-
torial segmentation we describe briefly the age-set system. The
1 adult male population falls into stratified groups based on age,
[and we call these groups 'age-sets'. The members of each set
become such by initiation and they remain in it till death. The
sets do not form a cycle, but a progressive system, the junior set
INTRODUCTORY 7
passing through positions of relative seniority till it becomes the
senior set, after which its members die and the set becomes a
memory, since its name does not recur. The only significant age-
grades are those of boyhood and manhood, so that once a lad has
been initiated into a set he remains in the same age-grade for the
rest of his life. There are no grades of warriors and elders such
as are found in other parts of East Africa. Though the sets are
conscious of their social identity they have no corporate func-
tions. The members of a set may act jointly in a small locality,
but the whole group never co-operates exclusively in any activity.
Nevertheless^ the system is organized tribally and each tribe is
stratified according to age independently of other tribes, though
adjacent tribes may co-ordinate their age-sets.
The Nuer, like all other peoples, are also socially differentiated
according to sex. This dichotomy has a very limited, and nega-
tive, significance for the structural relations which form the
subject of this book. Its importance is domestic rather than
political and little attention is paid to it. The Nuer cannot be
said to be stratified into classes. Within a tribe there is slight
differentiation of status between members of a dominant clan,
Nuer of other clans, and Dinka who have been incorporated into
the tribe, but, except perhaps on the periphery of Nuer expan-
sion eastwards, this constitutes distinction of categories rather
than of ranks.
Such, briefly, is the plan of this book and such are the mean-
ings we attach to the words most frequently used to describe
the groups discussed in it. We hope in the course of our inquiry
to refine these definitions. The inquiry is directed to two ends :
to describe the life of the Nuer, and to lay bare some of the prinj
ciples of their social structure. We have endeavoured to give
as concise an account of their life as possible, believing that a
short book is of greater value to the student and administrator
than a long one, and, omitting much material, we have recorded
only what is significant for the limited subject of discussion.
Ill
When the Government of the Anglo-Egyptian Sudan asked
me to make a study of the Nuer I accepted after hesitation and
with misgivings. I was anxious to complete my study of the
PLATE I
Youth (Eastern Gaajok) fastening giraffe-hair necklace on friend
INTRODUCTORY 9
Azande before embarking on a new task. I also knew that a
study of the Nuer would be extremely difficult . Their country
and character are alike intractable and what little I had pre-
viously seen of them convinced me that I would fail to establish
friendly relations with them.
I have always considered, and still consider, that an adequate
sociological study of the Nuer was impossible in the circum-
stances in which most of my work was done. The reader must
judge what I have accomplished. I would ask him not to judge
too harshly, for if my account is sometimes scanty and uneven,
I would urge that the investigation was carried out in adverse
circumstances ; that Nuer social organization is simple and their
culture bare ; and that what I describe is almost entirely based
on direct observation and is not augmented by copious notes
taken down from regular informants, of whom, indeed, I had
none. I^unlike most readers, know the Nuer, and must judge
my work more severely than they^ and I can say that if this
book reveals many insufficiencies I am amazed that it has^ver
appeared at all. A man must judge his labours by the obstacles
he has overcome and the hardships he has endured, and by these
standards I am not ashamed of the results.
It may interest readers if I give them a short description of the
.conditions in which I pursued my studies, for they will then be
better able to decide which statements are likely to be based on
sound observation and which to be less well-grounded.
I arrived in Nuerland early in 1930. Stormy weather pre-
vented my luggage from joining me at Marseilles, and owing to
errors, for which I was not responsible, my food stores were not
forwarded from Malakal and my Zande servants were not
instructed to meet me. I proceeded to Nuerland (Leek country)
with my tent, some equipment, and a few stores bought at
Malakal, and two servants, an Atwot and a Bellanda, picked
up hastily at the same place.
When I landed at Yoahnyang' on the Bahr el Ghazal the
Catholic missionaries there showed me much kindness. I waited
' I take this early opportunity to inform readers that I have not spelt Nuer
names and other words with phonetic consistency. I raise no objection, there-
fore, to any one speUing them differently. Generally I have given the nomina-
tive form, but an occasional genitive has crept into the text, diagrams, and
maps.
4507 C
10 INTRODUCTORY
for nine days on the river bank for the carriers I had been pro-
mised. By the tenth day only four of them had arrived and if
it had not been for the assistance of an Arab merchant, who
recruited some local women, I might have been delayed for an
indefinite period.
On the following morning I set out for the neighbouring vil-
lage of Pakur, where my carriers dropped tent and stores in the
centre of a treeless plain, near some homesteads, and refused to
bear them to the shade about half a mile further. Next day was
devoted to erecting my tent and trying to persuade the Nuer,
through my Atwot servant who spoke Nuer and some Arabic,
to remove my abode to the vicinity of shade and water, which
they refused to do. Fortunately a youth, Nhial, who has since
been my constant companion in Nuerland, attachedTiTmselfto~
rne^nd after twelve days persuaded his kinsmen to carry my
goods to the edge of the forest where they lived.
My servants, who, like most natives of the Southern Sudan,
were terrified of the Nuer, had by this time become so scared
that after several sleepless and apprehensive nights they bolted
to the river to await the next steamer to Malakal, and I was left
alone with Nhial. During this time the local Nuer would not
lend a hand to assist me in anything and they only visited me to
ask for tobacco, expressing displeasure when it was denied them.
When I shot game to feed myself and my Zande servants, who
had at last arrived, they took the animals and ate them in the
bush, answering my remonstrances with the rejoinder that
since the beasts had been killed on their land they had a right
to them.
My main difficulty at this early stage was inability to converse
freely with the Nuer. I had no interpreter. None of the Nuer
spoke Arabic. There was no adequate grammar of the language
and, apart from three short Nuer-English vocabularies, no
dictionary. Consequently the whole of my first and a large part
of my second expedition were taken up with trying to master
the language sufficiently to make inquiries through it, and only
those who have tried to learn a very difficult tongue without the
aid of an interpreter and adequate literary guidance will fully
appreciate the magnitude of the task.
After leaving Leek country I went with Nhial and my two
INTRODUCTORY ii
Zande servants to Lou country. We motored to Muot dit with
the intention of residing by the side of its lake, but found it
entirely deserted, for it was too early for the annual concentra-
tion there. When some Nuer were found they refused to divulge
the whereabouts of nearby camps and it was with considerable
difficulty that we located one. We pitched our tents there alrxcT
when the campers retired on Muot dit we accompanied them.
My days at Muot dit were happy and remunerative. I made
friends with many Nuer youths who endeavoured to teach me
their language and to show me that if I was a stranger they did
not regard me as an obnoxious one. Every day I spent hours
fishing with these lads in the lake and conversing with them in
my tent. I began to feel my confidence returning and would
have remained at Muot dit had the political situation been more
favourable. A Government force surrounded our camp one
morning at sunrise, searched for two prophets who had been
leaders in a recent revolt, took hostages, and threatened to take
many more if the prophets were not handed over. I felt that I
was in an equivocal position, since such incidents might recur,
and shortly afterwards returned to my home in Zandeland,
having accomplished only three and a half months' work among
the Nuer. , " -—-.-. •
It would at any time have been difficult to do research among
the Nuer, and at the period of my visit they were unusually_
hostile, for their recent defeat by Government forces and the
measures taken to ensure their final submission had occasioned
deep resentment. Nuer have often remarked to me, 'You raid
us, yet you say we cannot raid the Dinka'; 'You overcame us
with firearms and we had only spears. If we had had firearms
we would have routed you'; and so forth. When I entered a
cattle camp it was not only as a stranger but as an enemy, and
they seldom tried to conceal their disgust at my presence, refus-'
ing to answer my greetings and even turning away when I
addressed them.
At the end of my 1930 visit to Nuerland I had learnt a little
of the language but had the scantiest notes of their customs. J[n_
the^ drx^eason of 1931 I returned to make a fresh attempt,
going first for a fortnight to the American Mission at NasserT
where I was generously assisted by the American and Nuer staff.
12 INTRODUCTORY
and then to cattle camps on the Nyanding river — an unfortu-
nate choice, for the Nuer there were more hostile than those I
had hitherto encountered and the conditions were harsher than
any I had previously experienced. The water was scanty and
foul, the cattle were dying of rinderpest, and the camps swarmed
with flies. The Nuer would not carry my stores and equipment,
and as I had only two donkeys, one of them lame, it was im-
possible to move. Eventually I managed to obtain a lorry and
extricate myself, but not before experiencing the Nuer in his
most paralysing mood. As every effort was made to prevent me
from entering the cattle camps and it was seldom that I had
visitors I was almost entirely cut off from communication with
the people. My attempts to prosecute inquiries were persistently
obstructed.
Nuer are expert at sabotaging an inquiry and until one has
I resided with them for some weeks they steadfastly stultify all
efforts to elicit the simplest facts and to elucidate the most
innocent practices. I have obtained in Zandeland more in-
formation in a few days than I obtained in Nuerland in as many
weeks. After a while the people were prepared to visit me in my
tent, to smoke my tobacco, and even to joke and make small
talk, but they were unwilling either to receive me in their
windscreens or to discuss serious matters. Questions about
customs were blocked by a technique I can commend to natives
who are inconvenienced by the curiosity of ethnologists. The
following specimen of Nuer methods is the commencement of
a conversation on the Nyanding river, on a subject which
admits of some obscurity but, with willingness to co-operate,
can soon be elucidated.
I : Who are you ?
Cuol: A man.
/ : What is your name ?
Cuol : Do you want to know my name ?
/: Yes.
Cuol : You want to know my name ?
/ : Yes, you have come to visit me in my tent and I would like to
know who you are.
CmoZ: All right. I am Cuol. What is your name ?
/ : My name is Pritchard.
Cuol : What is your father's name ?
INTRODUCTORY 13
/ : My father's name is also Pritchard.
Cuol: No, that cannot be true. You cannot have the same name
as your father.
I: It is the name of my Hneage. What is the name of yourHneage?
Cuol : Do you want to know the name of my hneage ?
7: Yes.
Cuol : What will you do with it if I tell you ? Will you take it to
your country ?
/: I don't want to do anything with it. I just want to know it
since I am living at your camp.
Cuol: Oh well, we are Lou.
/: I did not ask you the name of your tribe. I know that. I am
asking you the name of your lineage.
Cuol : Why do you want to know the name of my lineage ?
/: I don't want to know it.
Cuol : Then why do you ask me for it ? Give me some tobacco.
I defy the most patient ethnologist to make headway against
this kind of opposition. One is just driven crazy by it. Indeed,
after a few weeks of associating solely with Nuer one displays, if
the pun be allowed, the most evident symptoms of 'Nuerosis'. ^
From the Nyanding I moved, still without having made any
real progress, to a cattle camp at Yakwac on the Sobat river
where I pitched my tent a few yards from the windscreens.
Here I remained, save for a short interval spent at the American
Mission, for over three months — till the commencement of the
rains. After the usual initial difficulties I at last began to feel
myself a member of a community and to be accepted as such,
especially when I had acquired a few cattle. When the campers
at Yakwac returned to their inland village I had no means of
accompanying them and intended to visit Leek country again.
A severe attack of malaria sent me to Malakal hospital, and
thence to England, instead. Five and a half months' work was
accomplished on this second expedition.
During the tenure of a subsequent appointment in Egypt I
published in Sudan Notes and Records essays which form the
basis of this book, for I had not expected to have a further
opportunity to visit the Nuer. However, in 1935 I was granted
a two years' research fellowship by the Leverhulme trustees to
make an intensive study of the Pagan Galla of Ethiopia. As
delay was caused by diplomatic chicanery I spent two and a half
14 INTRODUCTORY
months on the Sudan-Ethiopian frontier making a survey of
the Eastern Anuak, and when, at last, I entered Ethiopia the im-
minence of Itahan invasion compelled me to jettison my Galla
studies and enabled me to advance my investigation of the Nuer,
during a further seven weeks' residence in their country, by
revising earlier notes and by collecting more material. I visited
the Nuer who live on the Pibor river, renewed my acquaintance
with friends of the Nasser Mission and at Yakwac, and spent
about a month among the Eastern Jikany at the mouth of the
Nyanding.
In 1936, after making a survey of the Nilotic Luo of Kenya,
I spent a final seven weeks in Nuerland, visiting that part of it
which hes to the west of the Nile, especially the Karlual section
i of the Leek tribe. My total residence among the Nuer was thus
.; about a year. I do not consider a year adequate time in which
to make a sociological study of a people, especially of a difficult
people in adverse circumstances, but serious sickness on both
the 1935 and 1936 expeditions closed investigations prematurely.
Besides physical discomfort at all times, suspicion and obsti-
nate resistance encountered in the early stages of research,
absence of interpreter, lack of adequate grammar and dictionary,
and failure to procure the usual informants, there developed a
J further difficulty as the inquiry proceeded. As I became more
friendly with the Nuer and more at home in their language they
visited me from early morning till late at night, and hardly a
moment of the day passed without men, women, or boys in my
- tent. As soon as I began to discuss a custom with one man
another would interrupt the conversation in pursuance of some
affair of his own or by an exchange of pleasantries and jokes. The
men came at milking-time and some of them remained till mid-
day. Then the girls, who had just finished dairy-work, arrived
and insisted on attention. Married women were less frequent
visitors, but boys were generally under the awning of my tent if
grown-ups were not present to drive them away. These endless
visits entailed constant badinage and interruption and, although
they offered opportunity for improving my knowledge of the
Nuer language, imposed a severe strain. Nevertheless, if one
chooses to reside in a Nuer camp one must submit to Nuer
L custom, and they are persistent and tireless visitors. The chief
INTRODUCTORY 15
privation was the publicity to which all my actions were exposed,
and it was long before I became hardened, though never entirely
insensitive, to performing the most intimate operations before an
audience or in full view of the camp.
Since my tent was always in the midst of homesteads or
windscreens and my inquiries had to be conducted in public,
I was seldom able to hold confidential conversations and never
succeeded in training informants capable of dictating texts and
giving detailed descriptions and commentaries. This failure
was compensated for by the intimacy I was compelled to estab-
lish with the Nuer. As I could not use the easier and shorter
method of working through regular informants I had to fall back
on direct observation of, and participation in, the everyday life
of the people. From the door of my tent I could see what was
happening in camp or village and every moment was spent in
Nuer company. Information was thus gathered in particles,
each Nuer I met being used as a source of knowledge, and not,
as it were, in chunks supplied by selected and trained infor-
mants. Because I had to live in such close contact with the Nuer
I knew them more intimately than the Azande, about whom I
am able to write a much more detailed account. Azande would
not allow me to live as one of themselves ; Nuer would not allow
me to live otherwise. Among Azande I was compelled to live
outside the community; among Nuer I was compelled to be
a member of it, Azande treated me as a superior ; Nuer as an
equal.
I do not make far-reaching claims. I believe that I have
understood the chief values of the Nuer and am able to present
a true outline of their social structure, but I regard, and have
designed, this volume as a contribution to the ethnology of a
particular area rather than as a detailed sociological study, and
I shall be content if it is accepted as such. There is much that
I did not see or inquire into and therefore plenty of opportunity
for others to make investigations in the same field and among
neighbouring peoples. I hope they will do so and that one day
we may have a fairly complete record of Nilotic social systems.
"'"-^'^^
CHAPTER I
INTEREST IN CATTLE
I
A PEOPLE whose material culture is as simple as that of the
Nuer are highly dependent on their environment. They are
pre-eminently pastoral, though they grow more millet and maize
than is commonly supposed. Some tribes cultivate more and
some less, according to conditions of soil and surface water and
their wealth in cattle, but all alike regard horticulture as toil
forced on them by poverty of stock, for at heart they are herds-
men, and the only labour in which they delight is care of cattle.
They not only depend on cattle for many of life's necessities but
they have the herdsman's outlook on the world. Cattle are their
dearest possession and they gladly risk their lives to defend
their herds or to pillage those of their neighbours. Most of their
social activities concern cattle and cherchez la vache is the best
advice that can be given to those who desire to understand Nuer
behaviour.!
The attitude of Nuer towards, and their relations with, neigh-
bouring peoples are influenced by their love of cattle and their
desire to acquire them. They have profound contempt for
peoples with few or no cattle, like the Anuak, while their wars
against Dinka tribes have been directed to seizure of cattle and
control of pastures. Each Nuer tribe and tribal section has its
own pastures and water-supplies, and political fission is closely
related to distribution of these natural resources, ownership of
which is generally expressed in terms of clans and lineages. Dis-
putes between tribal sections are very often about cattle, and
cattle are the compensation for loss of life and limb that is so
frequently their outcome. Leopard-skin chiefs and prophets
are arbiters in questions in which cattle are the issue, or ritual
agents in situations demanding sacrifice of ox or ram. Another
ritual specialist is the wutghok, the Man of the Cattle. Likewise,
in speaking of age-sets and age-grades we find ourselves des-
' Nuer interest in their cattle has been emphasized by early travellers in
their country. Vide Marno, op. cit.. p. 343 ; Werne, op. cit., p. 439 ; du
Couret, op. cit., p. 82.
PLATE II
Girl in kraal (Eastern Gaajok)
INTEREST IN CATTLE 17
cribing the relations of men to their cattle, for the change from
boyhood to manhood is most clearly marked by a corresponding
change in those relations at initiation.
Small local groups pasture their cattle in common and jointly
defend their homes and herds. Their solidarity is most evident in
the dry season when they live in a circle of windscreens around
a common kraal, but it can also be seen in their wet season
isolation. A single family or household cannot protect and herd
their cattle alone and the cohesion of territorial groups must be
considered in the light of this fact.
The network of kinship ties which links members of local com-
munities is brought about by the operation of exogamous
rules, often stated in terms of cattle. The union of marriage is
brought about by payment of cattle and every phase of the
ritual is marked by their transference or slaughter. The legal
status of the partners and of their children is defined by cattle-
rights and obligations.
Cattle are owned by families. While the head of the house-
hold is alive he has full rights of disposal over the herd, though
his wives have rights of use in the cows and his sons own some of
the oxen. As each son, in order of seniority, reaches the age of
marriage he marries with cows from the herd. The next son will
have to wait till the herd has reached its earlier strength before
he can marry in his turn. When the head of the household dies
the herd still remains the centre of family life and Nuer strongly
deprecate breaking it up, at any rate till all the sons have mar-
ried, for it is a common herd in which all have equal rights.
When the sons are married they and their wives and children
generally live in adjacent homesteads. In the early part of the
dry season one sees a joint family of this kind living in a circle
of windscreens around a common kraal, and in the big camps
formed later in the year one finds them occupying a distinct
section in the lines of windscreens. The bond of cattle between
brothers is continued long after each has a home and children of
his own, for when a daughter of any one of them is married the
others receive a large portion of her bride- wealth. Her grand-
parents, maternal uncles, paternal and maternal aunts, and
even more distant relatives, also receive a portion. Kinship is
customarily defined by reference to these payments, being most
4507 p
i8 INTEREST IN CATTLE
clearly pointed at marriage, when movements of cattle from
kraal to kraal are equivalent to lines on a genealogical chart.
It is also emphasized by division of sacrificial meat among
agnatic and cognatic relatives.
The importance of cattle in Nuer life and thought is further
exemplified in personal names. Men are frequently addressed
by names that refer to the form and colour of their favourite
oxen, and women take names from oxen and from the cows they
milk. Even small boys call one another by ox-names when
playing together in the pastures, a child usually taking his name
from the bull-calf of the cow he and his mother milk. Often a
man receives an ox-name or cow-name at birth. Sometimes the
name of a man which is handed down to posterity is his ox-name
and not his birth-name. Hence a Nuer genealogy may sound
like an inventory of a kraal. The linguistic identification of a
man vdth his favourite ox cannot fail to affect his attitude to the
j beast, and to Europeans the custom is the most striking evidence
L of the pastoral mentality of the Nuer.
Since cattle are a Nuer's most cherished possession, being an
essential food-supply and the most important social asset, itjs
easy to understand why they play a foremost part-in ntuah^^
man establishes contact with the ghosts and spirits iHrough his
cattle. If one is able to obtain the history of each cow in a
kraal, one obtains at the same time not only an account of all
the kinship links and affinities of the owners but also of all their
mystical connexions. Cows are dedicated to the spirits of the
lineages of the owner and of his wife and to any personal spirit
that has at some time possessed either of them. Other beasts
are dedicated to ghosts of the dead. By rubbing ashes along the
back of a cow or ox one may get into touch with the spirit or
ghost associated with it and ask it for assistance. Another way
of communicating with the dead and with spirits is by sacrifice,
and no Nuer ceremony is complete without the sacrifice of a
ram, he-goat, or ox.
We have seen in a brief survey of some Nuer institutions and
customs that most of their social behaviour directly concerns
their cattle. A fuller study of their culture would show every-
where the same dominant interest in cattle, e.g. in their folk-
r lore. They are always talking about their beasts. I used some-
INTEREST IN CATTLE 19
times to despair that I never discussed anything with the young
men but hve stock and girls, and even the subject of girls led
inevitably to that of cattle. Start on whatever subject I would,
and approach it from whatever angle, we would soon be speak-
ing of cows and oxen, heifers and steers, rams and sheep, he-
goats and she-goats, calves and lambs and kids. I have already
indicated that this obsession — for such it seems to an outsider —
is due not only to the great economic value of cattle but also to
the fact that they are links in numerous social relationships.
Nuer tend to define all social processes and relationships in
terms of cattle. Their social idiom is a bovine idiom. *
Consequently he who lives among Nuer and wishes to under-
stand their social life must first master a vocabulary referring to
cattle and to the life of the herds. Such complicated discussions
as those which take place in negotiations of marriage, in ritual
situations, and in legal disputes can only be followed when one
understands the difficult cattle-terminology of colours, ages,
sexes, and so forth.
Important though horticultural and piscatorial pursuits are
in Nuer economy, pastoral pursuits take precedence because
cattle not only have nutritive utility but have a general social
value in other respects. I have mentioned a few situations in
which this value is manifested, but have not recorded every
role of cattle in Nuer culture, for they are significant in many
social processes, including some I have mentioned, which lie
outside the limited scope of this book. It seemed necessary to
give an introductory sketch on these lines in order that the
reader might understand that Nuer devotion to the herdsman's
art is inspired by a range of interests far wider than simple need
for food, and why cattle are a dominant value in their lives. We
shall ask later how this value is related to environmental con-
ditions and how far the two, taken together, help us to explain
some characteristics of Nuer political structure.
II
Before the present century Nuer were far richer in cattle than
they are now and it is probable that they cultivated less millet.
Their stock has been impaired by repeated outbreaks of rinder-
pest, which still decimate the herds. It was probably more
20 INTEREST IN CATTLE
destructive in the past than now, though the attacks I witnessed
were severe; but in the past the wariike Nuer could always
restore their losses by raiding Dinka. All Nuer agree that in the
last generation their herds were more considerable and that the
payments of bride-wealth and blood-wealth were forty, and
sometimes fifty to sixty, head of cattle, whereas to-day the
kinsmen of a bride do not expect to receive more than twenty to
thirty. At the present time I would say, on a general impression,
that the Nuer are far richer in stock than the Shilluk, but not so
prosperous as the more favoured of the Dinka tribes.
It was difficult to make a census of cattle, even in a small
area, and Nuer would certainly have regarded such an attempt
with repugnance. On the few estimates made I would reckon
an average of ten head of cattle and five goats and sheep to the
byre. A byre of the ordinary size cannot hold more than a dozen
or so adult kine. As there are some eight persons to a byre the
cattle probably do not greatly exceed the human population.
Cows predominate and probably compose about two-thirds of
the herds. Many plates in this book show the appearance of
Nuer cattle. Nuer say that a very large hump shows Beir origin
and that very long horns are evidence of Dinka stock.
Some tribes are richer in cattle than others. Lou country is
considered especially suitable for raising stock and is renowned
for its large herds. The Eastern Jikany were once very rich in
cattle, but their herds are still recovering from losses in epi-
demics that forced the people to cultivate more extensively.
Cattle are everywhere evenly distributed. Hardly any one is
entirely without them, and no one is very rich. Although
cattle are a form of wealth that can be accumulated, a man
never possesses many more beasts than his byre will hold,
because as soon as his herd is large enough he, or one of his
family, marries. The herd is thereby reduced to two or three
beasts and the next few years are spent in repairing its losses.
Every household goes through these alternating periods of
poverty and comparative wealth. Marriages and epidemics
prevent accumulation of cattle and no disparity in wealth
offends the democratic sentiment of the people.
When we come to examine the Nuer political system we shall
keep in mind that tiU recent years they have probably been
INTEREST IN CATTLE 21
more exclusively pastoral, and more nomadic, than at the
present time, and that the dwindling of their herds may partly ,
explain their persistent aggressiveness. J
III
Although cattle have many uses they are chiefly useful for
the milk they provide. Milk and millet (sorghum) are the staple
foods of the Nuer. In some parts of their country, especially
among the Lou, the millet supply seldom lasts the whole year,
and when it is exhausted people are dependent on milk and
fish. At such times a family may be sustained by the milk of a
single cow. In all parts the millet crop is uncertain and more or
less severe famines are frequent, during which people rely on
fish, wild roots, fruits, and seeds, but mainly on the milk of their
herds. Even when millet is plentiful it is seldom eaten alone,
for without milk, whey, or liquid cheese, Nuer find it stodgy,
unpalatable, and, especially for children, indigestible. They
regard milk as essential for children, believing that they cannot
be well and happy without it, and the needs of children are
always the first to be satisfied even if, as happens in times of
privation, their elders have to deny themselves. In Nuer eyes
the happiest state is that in which a family possesses several
lactating cows, for then the children are well-nourished and
there is a surplus that can be devoted to cheese-making and to
assisting kinsmen and entertaining guests. A household can
generally obtain milk for its little children because a kinsman
will lend them a lactating cow, or give them part of its milk, if
they do not possess one. This kinship obligation is acknowledged
by all and is generously fulfilled, because it is recognized that
the needs of children are the concern of neighbours and rela-
tives, and not of the parents alone. Occasionally, however,
after an epidemic or, to a lesser degree, after two or three
youths of the group have married, an entire hamlet, or even
a whole village may experience scarcity. Sometimes, also,
shortage is caused by a tendency for the cows of a village to
cease lactating at about the same time.
Nuer value their cows according to the amount of milk they
give and they know the merits of each in this respect. The
calves of a good milch cow are more highly prized than the
22 INTEREST IN CATTLE
calves of a cow that gives little milk. A cow is never to them
just a cow, but is always a good cow or a bad cow, and a Nuer
who is owed a cow will not accept in payment of his debt one
that does not meet with his approval. If you ask a Nuer in a
cattle camp which are the best and worst cows in the herd he
can tell you at once. In judging their points he pays little
attention to those aesthetic qualities which please him in an
ox, especially fatness, colour, and shape of horns, but he selects
those which indicate a good milch cow: a broad loose back,
prominent haunch bones, large milk-veins, and a much-
wrinkled milk-bag. In judging the age of a cow he notes the
depth of the trenches which run on either side of its rump
towards the tail, the number and sharpness of its teeth, the
firmness of its gait, and the number of rings on its horns. Nuer
cows have the familiar angular and thin-fieshed characteristics
of dairy stock.
Milking is performed twice daily by women, girls, and uninitiated
boys. Men are forbidden to milk cows unless, as on journeys or
war expeditions, there are no women or boys present. The milker
squats by the cow and milks a single teat at a time into the narrow
mouth of a bottle-necked gourd balanced on her thighs (see Plates
III and IV). She milks with thumb and first finger but, the other
fingers being closed, the teat is to some extent pressed against the
whole hand. It is both a squeezing and a pulling motion. The
gourd is kept in position by the downward stroke of the hands
which press it against the thighs. When a pot, or a gourd with a
wider mouth, is used it is held between the knees and the milker
squeezes two teats at a time. Occasionally one sees two girls milking
a cow, one at either side. If a cow is restless a man may hold it still
by putting his hand in its mouth and gripping its muzzle, and if it
kicks, a noose is placed round its hind legs and they are pulled
together (see Plate III). I was told that sometimes they ring the
nose of a cow that is habitually restless during milking.
The process of milking is as follows. The calf is loosened and
with its tethering-cord round its neck runs at once to its dam and
begins violently butting her udder. This starts the flow of milk,
and Nuer hold that if the calf were not first to suck the cow would
hold up its milk. They do not pat the udder with the hand unless
the calf is dead, for this is considered bad for the cow. When the
calf has sucked a little it is dragged away, resisting stubbornly,
and tethered to its dam's peg, where it rubs against her forelegs
and she licks it. The girl now mQks the first milking, known as
PLATE III
.Mill<in,i; a r^siless cow (Loiij
INTEREST IN CATTLE 23
the wic. When the teats become soft and empty the calf is again
loosened and the process is repeated. The second milking is called
tip indit, the greater tip. As a rule there are only two milkings,
but if it is a very good milch cow at the height of her lactation
period the calf may once more be loosened and a third milking,
called tip intot, the lesser tip, be taken. When the girl has finished
milking she wipes her thighs and the milk-gourd with the cow's
tail and loosens the calf to finish off what milk is left. The first
milking takes longer time and produces more milk than the
second, and the second more than the third. The morning yield
is greater than the evening yield.
A series of measurements suggest that four to five pints a
day may be regarded as a general average for Nuer cows
during their lactation period, which lasts, on an average, about
seven months. It must be remembered, however, that this is
an estimate of the yield for human consumption. The calf gets
its share before, during, and after the milking. It is possible,
moreover, that, as Nuer declare, some cows hold up their milk
for their calves, since the calves often suck for several minutes
after milking before their dams refuse them by kicking them or
moving so that they cannot reach their udders. Sometimes a
small boy drags the calf away and milks the udders himself,
licking the milk off his hands, or shares the teats with the calf,
but as a rule the calf gets the remainder of the milk. The total
yield may, therefore, be as high as seven to nine pints a day and
it appears to be far richer than milk given by English cows. It
is not surprising that the yield is small, because Nuer cows
receive no artificial feeding, succulent pasturage is often diffi-
cult to obtain, and they have to endure great hardship. It
must, moreover, be emphasized that whereas English dairy
farmers require only milk, Nuer herdsmen require milk and
also wish to preserve every calf. Human needs have to be
subordinated to the needs of the calves, which are the first
consideration if the herd is to be perpetuated.
Milk is consumed in various ways. Fresh milk is drunk, especially
by children, and is also consumed with millet-porridge. Fresh milk
is chiefly drunk by adults in the heat of the dry season when a
refreshing draught is most appreciated and food is scarce. Some
milk is put aside, where it soon, very rapidly in hot weather, sours
and thickens, in which condition it is relished. Nuer like to have
24 INTEREST IN CATTLE
a gourd of sour milk always at hand in case visitors come. Part of
the daily yield is kept for making cheese, and if there are several
cows in lactation one may be reserved for this purpose. Milk for
churning is drawn into a different gourd to that used for drinking
milk. It is then transferred to a churning gourd (see Fig. i), in
which it stands for several hours, and as churning gourds are not
cleaned, unless they smell bad, the acids which remain from the
previous churning curdle the milk. After standing it is churned
Fig. I. Churning gourd.
by a woman, or girl, who sits on the ground with her legs stretched
in front of her, and, raising the gourd, brings it down with a jerk
on her thighs where she rocks it a few times before repeating her
actions : a simple but lengthy way of churning. A small quantity
of water is poured into the gourd when the curds are beginning to
form to make them set well and to increase the quantity of whey,
and some ox's urine may be added to give them consistency. When
they have formed, the woman pours the milk into a cup-shaped
gourd and scoops them out with a mussel shell into another gourd
vessel which is hung up in a hut. The whey, mixed with fresh
milk, is mainly drunk by women and boys. Every day they add
to the supply of curds and now and again stir some ox's urine with
them to prevent them from going bad. They may add to the supply
for several weeks before the final boiling over a quick fire, which
INTEREST IN CATTLE 25
turns the curds, lieth in bor, into solid deep yellow cheese, lieth in
car. After boiling for a time the liquid is poured into a gourd and
the oil on top is removed, to be used as a flavouring for porridge.
The cheese is suspended in a net from the roof of a hut in a round
gourd, a piece of the shell of which has been cut out so that cords
run through it and it acts as a sliding lid (see Fig. 2), and, if air
is excluded by a coating of cattle dung, it will keep in good condi-
tion for months. Milk may thus be stored in the form of cheese.
It is eaten with porridge and is also used for anointing the
body.
Sheep and goats are also milked in the mornings, but little
importance is attached to their yield, which is drunk by small
children and not used for dairy work. The woman milks and the
kids and lambs finish what is left in the udders. As they run with
their dams at pasture an evening milking is not taken ; but during
the day hungry herdboys often squeeze the udders and lick the
milk off their hands.
Some points that arise from an account of milking and
dairy-work deserve emphasis, (i) The present number and
distribution of cattle do not permit the Nuer to lead an
entirely pastoral life as they would like to do, and possibly did
at one time. On a generous estimate the average daily yield
to the byre is probably no more than twelve pints, or one and a
half pints per person. A mixed economy is, therefore, necessary.
(2) Furthermore the fluctuation in household resources, due to
epidemics and transmission of bride-wealth, is further accen-
tuated by the organic character of the staple diet, for cows only
produce milk for a certain period after calving and the yield is
not constant. It follows that a single family is not a self-
sufficient unit, as far as milk is concerned, for it cannot always
ensure an adequate supply. Therefore, since milk is considered
essential, the economic unit must be larger than the simple
family group. (3) Environmental conditions, as well as need
for cereal food to supplement their milk diet, prevent Nuer
from being entirely nomadic, but milk food enables them to
lead a roving life for part of the year and gives them mobility
and elusiveness, as their history shows and as has been recently
demonstrated in the Government campaign against them. Milk
requires neither storage nor transport, being daily renewed, but,
on the other hand, involves a straight dependence on water and
vegetation which not only permits, but compels, a wandering
4507 E
26 INTEREST IN CATTLE
life. Such a life nurtures the qualities of the shepherd — courage,
love of fighting, and contempt of hunger and hardship — rather
than shapes the industrious character of the peasant.
IV
Nuer are also interested in their cattle for meat, boiled and
roasted. They do not raise herds for slaughter, but sheep and
oxen are frequently sacrificed at ceremonies. There are always
ghosts and spirits in whose honour a sacrifice would at any
time be appropriate, and such sacrifices are generally long
overdue, so that there does not lack a proper excuse for a feast
when people desire one. Fertile cows are sacrificed in mortuary
rites, but, otherwise, only barren females are killed. At sacrifices
most people are interested more in the festal than the religious
character of the rites. Sometimes, as at marriage ceremonies,
the people who perform the ritual are different from those who
eat the meat, while at other ceremonies there is a general
scramble for the carcass. Desire for meat is shown without
shame on these occasions, and Nuer recognize that some men
sacrifice without due cause. In some years it is a custom in the
rains for young men to join together at a homestead with the
purpose of slaughtering oxen and gorging themselves with
meat. Except on such occasions, however, people ought not to
kill an ox solely for food — it being even thought that the ox may
curse them — and they only do so in severe famine. The Lou, who
are rich in cattle, have a reputation, of which they are rather
ashamed, for killing oxen for meat. Nevertheless, nowhere in
Nuerland are cattle ordinarily slaughtered for food, and a man
would never kill even a sheep or goat merely on the grounds that
he desired meat. On occasions of minor importance sheep or
goats are sacrificed rather than oxen, as they are less valuable.
Any animal which dies a natural death is eaten. Even when
a youth's favourite ox dies he must be persuaded to partake
of its flesh, and it is said that were he to refuse his spear might
avenge the insult by cutting his foot or hand on some future
occasion. Nuer are very fond of meat, and declare that on the
death of a cow, 'The eyes and the heart are sad, but the teeth
and the stomach are glad.' 'A man's stomach prays to God,
independently of his mind, for such gifts.'
INTEREST IN CATTLE 27
Though oxen are sacrificed and eaten they are not valued
only for these purposes, but also for display and for the prestige
their possession confers. Colour
and shape of horns are signifi-
cant, but the essential quali-
ties are bigness and fatness,
it being considered especially
important that the haunch
bones should not be apparent.
Nuer admire a large hump
which wobbles when the ani-
mal walks, and to exaggerate
this character they often mani-
pulate the hump shortly after
birth.
Like other pastoral peoples
in East Africa the Nuer ex-
tract blood from the necks of
their cattle, and this is a sup-
plementary article of diet in
dry season camps, where one
may generally see at least one
cow bled each evening. Cows
are bled for culinary purposes
more frequently than oxen.
The operation, called bar, con-
sists of tying a cord tightly
round a cow's neck so that
the veins stand out and one of
them can be stabbed, on the
head side of the cord, with a
small knife bound with cord or
grass to prevent it entering too
deeply. The blood spurts out,
and when a large gourd has
been filled they loosen the
cord and it ceases to flow.
Some dung is smeared over
the wound. If one examines
the neck of
Fig. 2. Gourd for storing cheese.
a cow one sees
a row of small cicatrices. Cows appear to be a little giddy after
the operation and may totter slightly, but otherwise seem none
28 INTEREST IN CATTLE
the worse for their experience. Indeed, it may well be, as Nuer
assert, that they are the better for it, for they lead a sluggish life.
The blood is boiled by women till it is fairly consistent and can be
used as a meat flavouring with porridge ; or the men let it stand
till it coagulates into a solid block, and, after roasting it in the
embers of a fire, cut it up and eat it.
Nuer do not regard the blood of cows as a staple article of diet
and it does not play an important part in their cuisine. Indeed,
they say that they do not perform the operation to acquire food,
though they confess that roasted blood is delicious, but for the
benefit of the cows. Bleeding is designed to cure a cow of any
unfitness by letting out the bad blood of the sickness. Also, Nuer
say, it makes the cow fat, for next day it will be more lively and
graze avidly. Bleeding, moreover, in their opinion, decreases the
desire of a cow to be served. Nuer say that if a cow is served too
frequently it may eventually become barren, whereas, if it is
bled now and again, it will only require to be served once and will
be in calf. Cattle are sometimes bled for medical reasons in the
rainy season, when people may be so replete that the blood is given
to the boys of the kraal and to the dogs. Sometimes they make
incisions in the noses of calves and let the blood flow to the ground
in order to make them fat. I have seen Nuer scarify their own legs
and the small of their backs to induce fleetness and strength.
The following two points seem to us to be significant, (i)
Whilst Nuer normally do not kill their stock for food, the end
of every beast is, in fact, the pot, so that they obtain sufficient
meat to satisfy their craving and have no pressing need to hunt
wild animals, an activity in which they engage little. (2)
Except when epidemics are rife the usual occasions of eating
meat are ritual and it is the festal character of rites which gives
them much of their significance in the hfe of the people.
V
Apart from milk, meat, and blood, cattle furnish Nuer with
numerous household necessities, and when we consider how few
are their possessions we can appreciate the importance of cattle
as raw material. The bodies and bodily products of cattle have
the following uses:
Their skins are used for beds, trays, for carr3dng fuel (Plate
XVII), cord for tethering and other purposes, flails (Fig. 15),
leather collars for oxen (Fig. 4), and for the tympana of drums.
They are employed in the manufacture of pipes, spears, shields,
t%J
Fig. 3. Bags made from the scrota of a bull and a giraffe.
30 INTEREST IN CATTLE
snuff-containers, &c. The scrota of bulls are made into bags to
contain tobacco, spoons, and other small objects (Fig. 3). Tail-
hairs are made into tassels used as dance ornaments by girls and
to decorate the horns of favourite oxen (Plate IV). Their bones
are used for the manufacture of armlets, and as beaters, pounders,
and scrapers. Their horns are cut into spoons (Fig. 14) and are
used in the construction of harpoons.
Their dung is used for fuel and for plastering walls, floors, and
the outsides of straw huts in cattle camps. It is also employed as
a plaster in minor technological processes and to protect wounds.
The ashes of burnt dung are rubbed over mens' bodies, and are
used to dye and straighten the hair, as a mouth wash and tooth
powder, in the preparation of sleeping-skins and leather bags,
and for various ritual purposes. Their urine is used in churning
and cheese-making, in the preparation of gourd-utensils, for tan-
ing leather, and for bathing face and hands.
The skins of sheep and goats are worn as loin garments by
married women (Plate XXIII {a)), used as rugs to sit on, made
into bags for storing tobacco and millet, and are cut into strips to
be tied round their ankles by youths when dancing. Their dung
and urine are not utilized.
The Bedouin Arab has been called the parasite of the camel.
With some justice the Nuer might be called the parasite of the
cow. It may, however, seem that the list we have compiled
does not cover a very wide range of uses, but so simple is Nuer
material culture that it accounts for a very considerable part of
their technology and contains items on which they are highly
dependent, e.g. the use of dung as fuel in a country where it is
difficult to obtain sufficient vegetable fuel for cooking, let alone
for the large fires that burn day and night in every byre and
windscreen.
We have seen that apart from their many social uses Nuer are
directly concerned with cattle as producers of two essential
articles of diet, milk and meat. We now perceive that the
economic value of cattle is more extensive. Taking into con-
sideration also the more general social value of cattle, briefly
indicated in Section I, we may already note that there is
over-emphasis on a single object, which dominates all other
interests and is consistent with those qualities of simplicity,
single-mindedness, and conservatism, so characteristic of
pastoral peoples.
PLATE IV
Ox with tassels hanging from its horns (Lou)
INTEREST IN CATTLE 31
VI
In later chapters we shall describe how the needs of cattle,
water, pasturage, protection from carnivorous beasts and
biting insects, and so forth, are attended to, and show in
what manner they determine human routine and affect social
relations. Leaving these broader issues on one side, we ask
here whether the Nuer, who are so reliant on their cattle and
who value them so highly, are competent herdsmen. It is
unnecessary to state that they give their beasts every attention
that their knowledge allows, but it is pertinent to inquire
whether their knowledge suffices. It was especially noted where
Nuer practice is not in accord with the conventions of farming,
and the reasons for the divergence were investigated. A few of
the more evident difficulties and some general observations on
Nuer husbandry are recorded below.
I. Since the cows are not brought back to the kraals at
midday the smaller calves must go without nourishment for
many hours each day. However, Capt. H. B. Williams, Director
of the Sudan Veterinary Department, tells me that Nuer oxen
have the reputation of being as good as any in the Sudan, so
that their development as young calves cannot be seriously
arrested. In the rains the cows are seldom milked before 9 to
10 a.m. and again at about 5 p.m., but in the dry season they
may be taken to pasture as early as 8 a.m. and not return till
about 5.30 p.m., so that they cannot suckle their calves for
about ten hours. However, this long interval is not easily
avoided, for in the dry season the grazing grounds are often
distant and owing to lack of good pasturage the cattle require
longer to feed than in the rains. In the rains it would be a
simple matter to pasture the herd at daybreak and bring it
home at midday, as many East African peoples do, for the
cows to suckle their calves and chew the cud. But Nuer say
that when the cattle come out of their hot smoky byres they
like to rest a while in the kraal before going to pasture, and
their lethargy, which contrasts with their eagerness to graze
after a night in the open in dry season camps, seems to justify
this statement. Nuer reahze that the heat and smoke of byres
are bad for the cattle, but they consider mosquitoes worse.
Fig. 4. Ox-bell and collar.
INTEREST IN CATTLE 33
Also by waiting till the dew has evaporated they consider they
lessen the risk of digestive troubles, for in the rains the ground
is cold and damp till a late hour, A further reason for keeping
the cattle late in the kraals is that if they are loosened early
they soon graze to repletion and begin to wander in all direc-
tions, since they are not usually herded in the rains.
2. It at once strikes a European that the condition of drinking
water at periods of the dry season leaves much to be desired,
especially if he has to drink it himself. Sometimes the pools
have almost dried up and contain foul, even slimy, water which
men and cattle drink. I have wondered why they do not move
sooner from these small pools, such as that shown on Plate
XXI (6), around which they camp in the early drought, to the
rivers and lakes where they make their final concentrations,
but I do not distrust their judgement, for they are fully aware
that dirty water is neither palatable to, nor good for, the cattle,
and when circumstances permit they are at pains to ensure that
they are supplied with clean water as often as they require it.
In moving camp they have to take into account a number of
desiderata : pasturage, fishing, the harvest of Balanites aegyptiaca,
the second millet harvest, &c., besides conditions of water.
3. Unlike some East African peoples Nuer do not keep too
many entire animals. If they err at all it is in keeping too few.
On the limited observations made it was estimated that there
is one adult bull to about thirty or forty adult cows. Nuer
try to select as stud bulls the calves of their best milch cows
so that they may breed good milch cows from them. They say
that if they did not castrate most of the bull calves the cows
would get no peace and there would be constant fighting in the
kraals and commotion in the byres. A calf is not castrated till
it is about eighteen months to two years old: 'When its dam
has had another calf and a third is in its womb.' It is thrown,
the scrotum is cut with a spear, and the testicles drawn out and
severed. There is little loss of blood and the animal soon
recovers. A calf may be castrated for sacrificial purposes at
any time, but otherwise Nuer prefer to perform the operation
in the dry season for there is less chance of inflammation than
in the rains. Bulls are not discouraged from fighting unless
they belong to the same herd, and fights are often cited in
34 INTEREST IN CATTLE
tradition as the cause of fission and migration of lineages. A
very large number of steers and oxen are slaughtered in
sacrifices.
4. Heifers are not served till their third year. Nuer know
when a cow is on heat by its be-
haviour in the kraal: it is restless, lows,
swishes its tail, sniffs at the vulvas of
other cows, and tries to mount them. If
a cow has mated in the grazing grounds
— for bulls run with the herd — the first
signs of pregnancy are said to be vulvar
changes. If you ask Nuer when a cow
which has been served at a certain time
will calve they can at once, and accur-
ately, tell you. They say that if a cow
has had no serious illness it will bear
about eight calves.
In my experience there is very slight
mortality among calves. Nuer give them
every attention. When a cow is seen to
be about to calve for the first time its
owner sits up with it all night, or accom-
panies it to pasture, to assist delivery.
An experienced cow is left to drop its calf
itself, but a man is usually present to
assist if it is in trouble. He must be
present if it calves in the bush, because
the calf is too weak to follow its dam,
which will stay with it, and they may be-
come separated from the herd and fall a
prey to wild beasts. If a calf dies in the
womb Nuer try to remove it, and when
it is necessary to turn it in the womb
they perform this operation. If the after-
birth does not fall, or if the cow does not
lick its calf, they administer medicines. When a calf dies they
resort to various devices to persuade its dam to give milk. They
stuff the head with straw (see Fig. 5), and rub some of the dam's
urine on it ; or, especially when a cow aborts, they stuff the whole
calf, insert stumps of wood to act as legs, and place it in front of
its dam and push its head against her teats, while they gently
squeeze and pull them and a boy blows up the vagina.
Fig. 5. Stuffed calf's head.
Fig. ^. Calf's weaning ring.
INTEREST IN CATTLE 35
Nuer say that if a calf is only a day or two old and its dam dies
it will also die, but once it knows the cak tin hor, 'the white milk'
which follows the colostrum, it can be saved. It is fed by hand
from a small gourd with a
funnel mouth and efforts
are made to get another
cow in lactation to suckle
it. Since Nuer believe,
erroneously it seems, that
it is dangerous for a calf
to drink the discoloured
milk at the top of the
colostrum, they milk this
off before allowing the
calf to suck, and if by
inadvertence it sucks first
they administer a purga-
tive. They regard it as
more serious if there is
any blood in the milk.
For the first three or four days a calf sucks all its dam's milk
except the part drawn off. Then close kinsmen, who live near by,
are summoned to eat porridge over which is poured the first milk
taken for human consumption. At this ceremony the end hairs of
the calf's tail are cut off and its owner spits on them and waves
them over the back of the dam, for otherwise the calf will sicken
because it resents people stealing its milk. Afterwards, however,
they can still say, ' We do not yet share with its calf ', for they take
very little milk for the first fortnight in order to give it a chance
to get strong and for its teeth to harden. When the calf is stronger
they take more milk and they then say that the calf shares {btdh)
the milk with men. It continues to suck till its dam is again in
calf and refuses it. Weaning devices are not as a rule employed,
but if the dam suckles when it is pregnant and it is found impracti-
cable to keep it apart from its calf in the pastures they place a
ring of thorns round the calf's muzzle (Fig. 6), which allows it
to graze but prevents it from sucking, for the thorns prick the
dam's udder and she kicks it aside. It will be seen from this
account how Nuer solve the herdsman's problem of making cows
provide for their masters without depriving their calves of
essential nourishment.
Small calves, after the adult herd has gone to the grazing
grounds, are housed till the late afternoon in byres in wet-season
villages, and tethered in the shade of a tree in dry-season camps.
36 INTEREST IN CATTLE
They are watered during the afternoon, and boys bring them
grasses, especially poon {Oryza Barthii), which is very fattening.
They begin to go to pasture with the older calves, under the
care of herdboys, in about their third month and are kept apart
from their dams by being driven in the opposite direction to that
taken earlier in the day by the adult herd. They run with the
herd when they are about a year old, by which time their dams
are again in calf.
We shall have opportunities for noting further the attention
Nuer give to their cattle and the wisdom of their methods. I
have merely given in this section a few examples to illustrate
a general conclusion reached in the course of my study: that
!- Nuer cattle husbandry could not in any important particular
be improved in their present oecological relations; that, con-
sequently, more knowledge than they possess would in no way
\ assist them ; and that, as will be shown, were it not for their
unceasing vigilance and care the cattle would not survive the
harsh conditions of their environment.
VII
It has been remarked that the Nuer might be called parasites
r of the cow, but it might be said with equal force that the cow
\^ is a parasite of the Nuer, whose lives are spent in ensuring its
welfare. They build byres, kindle fires, and clean kraals for its
comfort ; move from villages to camps, from camp to camp, and
from camps back to villages, for its health ; defy wild beasts for
its protection; and fashion ornaments for its adornment. It
lives its gentle, indolent, sluggish life thanks to the Nuer's
r devotion. In truth the relationship is symbiotic : cattle and men
sustain life by their reciprocal services to one another. In this
1 intimate symbiotic relationship men and beasts form a single
[ community of the closest kind. In a few paragraphs I direct
attention to this intimacy.
The men wake about dawn at camp in the midst of their
cattle and sit contentedly watching them tiU milking is finished.
They then either take them to pasture and spend the day watch-
ing them graze, driving them to water, composing songs about
them, and bringing them back to camp, or they remain in the
kraal to drink their milk, make tethering-cords and ornaments
PLATE V
Girl milking (Luu)
INTEREST IN CATTLE 37
for them, water and in other ways care for their calves, clean
their kraal, and dry their dung for fuel. Nuer wash their hands
and faces in the urine of the cattle, especially when cows urinate
during milking, drink their milk and blood, and sleep on their
hides by the side of their smouldering dung. They cover their
bodies, dress their hair, and clean their teeth with the ashes of
cattle dung, and eat their food with spoons made from their
horns. When the cattle return in the evening they tether each
beast to its peg with cords made from the skins of their dead
companions and sit in the windscreens to contemplate them and
to watch them being milked. A man knows each animal of his
herd and of the herds of his neighbours and kinsmen : its colour,
the shape of its horns, its peculiarities, the number of its teats,
the amount of milk it gives, its history, its ancestry and its
progeny. Miss Soule tells me that most Nuer know the points
of the dam and grand-dam of a beast and that some know the
points of its forebears up to five generations of ascent. A Nuer
knows the habits of all his oxen, how one bellows in the evenings,
how another hkes to lead the herd on its return to camp, and
how another tosses its head more than the rest are wont to do.
He knows which cows are restless during milking, which are
troublesome with their calves, which like to drink on the way
to pasture, and so forth.
If he is a young man he gets a boy to lead his favourite ox,
after which he takes his name, round the camp in the morning
and leaps and sings behind it ; and often at night he walks
among the cattle ringing an ox-bell and singing the praises of
his kinsmen, his sweethearts, and his oxen. When his ox comes
home in the evening he pets it, rubs ashes on its back, removes
ticks from its belly and scrotum, and picks adherent dung
from its anus. He tethers it in front of his windscreen so that
he can see it if he wakes, for no sight so fills a Nuer with con-
tentment and pride as his oxen. The more he can display the
happier he is, and to make them more attractive he decorates
their horns with long tassels, which he can admire as they toss
their heads and shake them on their return to camp, and their
necks with bells, which tinkle in the pastures. Even the bull
calves are adorned by their boy-owners with wooden beads and
bells (Fig. 13). The horns of young buUs, destined to be
38 INTEREST IN CATTLE
castrated later, are generally cut so that they will grow in a
shape that pleases their masters. The operation, called 7igat,
is probably performed towards the end of their first year and
usually takes place in the dry season, as it is said that a steer
may die if its horns are cut in the rains. The animal is thrown
and held down while its horns are cut through obliquely with
a spear. They grow against the cut. The beasts appear to
suffer much pain during the operation and I have sometimes
heard Nuer compare their ordeal to the initiation of youths
into manhood.
When a Nuer mentions an ox his habitual moroseness leaves
him and he speaks with enthusiasm, throwing up his arms to
show you how its horns are trained. 'I have a fine ox', he says,
'a brindled ox with a large white splash on its back and with
one horn trained over its muzzle' — and up go his hands, one
above his head and the other bent at the elbow across his face.
In singing and dancing they call out the names of their oxen
and hold their arms in imitation of their horns.
The attitude towards cattle varies with varying situations in
social life and with changes in social development. As soon as
children can crawl they are brought into close intimacy with the
flocks and herds. The kraal is their playground and they are
generally smeared with dung in which they roll and tumble.
The calves and sheep and goats are their companions in play
and they pull them about and sprawl in the midst of them.
Their feelings about the animals are probably dominated by
desire for food, for the cows, ewes, and she-goats directly satisfy
their hunger, often suckling them. As soon as a baby can drink
animal's milk its mother carries it to the sheep and goats and
gives it warm milk to drink straight from the udders.
The games of rather older children of both sexes centre round
cattle. They build byres of sand in camps and of moistened
ashes or mud in villages, and fill the toy kraals with fine mud
cows and oxen (Fig. 7), with which they play at herding and
marriage. The first tasks of childhood concern cattle. Very
small children hold the sheep and goats while their mothers
milk them, and when their mothers milk the cows they carry the
gourds and pull the calves away from the udders and tether
them in front of their dams. They collect urine in gourds and
INTEREST IN CATTLE 39
wash themselves in it. When they are a httle older and stronger
they have to clean the byres and kraals, assist in the milking,
and herd the small calves and the sheep and goats at pasture.
Fig. 7. Mud figures of oxen decorated with
tassels.
Food and play contacts with the cattle have changed to labour
contacts. At this age the interests of the sexes in cattle begin
to diverge and the divergence becomes more apparent as they
grow up. The labour of girls and women is restricted to the byres
and the kraals and is concerned mostly with the cows, while
boys herd the calves at pasture, as well as assisting in the kraal.
40 INTEREST IN CATTLE
and after initiation they herd the adult cattle and in the kraal
give their attention mainly to the oxen. The women are
dairy-maids ; the men herdsmen. Moreover, to a girl the cows
are essentially providers of milk and cheese and they remain
such when she grows up and is married and milks and churns
for her husband's people, whereas to a boy they are part of the
family herd in which he has property rights. They have
entered the herd on the marriage of his kinswomen and one
day he will marry with them. A girl is separated from the herd
on marriage ; a boy remains as its owner. When a boy becomes
a youth and is initiated into manhood the cattle become some-
thing more than food and the cause of labour. They are also a
means of display and marriage. It is only when a man marries
and has children and an independent household and herd,
when he has become an elder and man of position, that he often
uses cattle as sacrifices^ invests them with a sacred significance,
and employs them in ritual.
The Nuer and his herd form a corporate community with
solidarity of interests, to serve which the lives of both are
adjusted, and their symbiotic relationship is one of close
physical contact. The cattle are docile and readily respond to
human care and guidance. No high barriers of culture divide
men from beasts in their common home, but the stark naked-
ness of Nuer amid their cattle and the intimacy of their contact
with them present a classic picture of savagery. I ask the
reader to look at some of the illustrations, for example the
Frontispiece and Plates III, V, and XVII, which will convey
to him better than I can do in words the crudity of kraal life.
Cattle are not only an object of absorbing interest to Nuer,
having great economic utility and social value, but they live in
the closest possible association with them. Moreover, irres-
pective of use, they are in themselves a cultural end, and the
mere possession of, and proximity to, them gives a man his
heart's desire. On them are concentrated his immediate in-
terests and his farthest ambitions. More than anything else
they determine his daily actions and dominate his attention.
We have remarked on the over-emphasis on cattle produced by
their wide range of social and economic uses. So many physical,
psychological, and social requirements can be satisfied from
INTEREST IN CATTLE 41
this one source that Nuer attention, instead of being diffused in
a variety of directions, tends, with undue exclusiveness, to be
focused on this single object and to be introvertive, since the
object has a certain identity with themselves. We will now
examine briefly some linguistic material wherein we shall
perceive further evidence of this hypertrophy of a single
interest and of the identification of men with cattle to which I
have alluded.
VIII
Linguistic profusion in particular departments of life is one
of the signs by which one quickly judges the direction and
strength of a people's interests. It is for that reason, rather than
for its intrinsic importance, that we draw the reader's attention
to the volume and variety of the Nuer cattle vocabulary. Like
all the pastoral Nilotes they use an enormous number of words
and phrases about cattle and the tasks of herding and dairy-
work, and from this vast assortment we select for comment a
single class: the terms by which they describe cattle, chiefly
by reference to their colours. ' These terms are more than a
linguistic technique which enables Nuer to speak of cattle with
precision in situations of practical husbandry and in the many
social contexts in which they figure, for they establish associa- -^
tions on the one hand between wild creatures and cattle and on 1
the other hand between cattle and their masters ; they furnish
certain ritual categories ; and they greatly enrich the language
of poetry.
In naming a Nuer cow one has to notice its colours and the way
in which they are distributed on its body. When it is not of one
colour the distribution of colours is the significant character by
which one names it. There are ten principal colour terms: white
(bor), black {car), brown [lual), chestnut (dol), tawny (yan), mouse-
grey (lou), bay {thiang), sandy-grey {lith), blue and strawberry
roan (yil), and chocolate {gwir). When a cow is of a single colour
it is described by one of these terms. An animal may combine two
or more colours, but a combination of more than two, known as
cuany, is very rare. Normally there is a combination of white with
one other colour and twelve common distributions of this com-
■ I have recorded some information on this neglected subject among a
neighbouring people in 'Imagery in Ngok Dinka Cattle-Names', Bulletin of
the School of Oriental Studies, 1934.
4507 Q
•a o
Us
^85
44 INTEREST IN CATTLE
bination arc shown in Figs. 8 and 9. There are, however, many
more combinations, at least twenty-seven, one of the commonest
being varieties of a striped or brindled coat (nyang).
In describing a beast one often denotes both the form of dis-
tribution and the colour that is combined with white. Thus an ox
may be entirely mouse-grey (lou) ; have a mainly mouse-grey
colour with a white face {kwe looka), white back [kar looka), white
splash on barrel {bil looka), white shoulder [rol looka), or white
belly {reng looka) : be brindled mouse-grey {nyang looka) : be white
with large mouse-grey patches {rial looka), medium mouse-grey
patches {kwac looka), or a mouse-grey rump {jok looka), &c. There
are at least a dozen terms describing different combinations of
white and mouse-grey and there are a similar number of terms for
a combination of white with each of the other colours. A further
example is given to illustrate the wide range of variations : a white
shoulder and foreleg {rol) may be found on a cow of any colour, e.g.
rol cara, rol yan, rol thiang, rol yili, &c. There may also be a com-
bination of one form of distribution with another and, in this
case, the two combinations constitute the terms of reference and
there is no need to denote the colouring that occurs in them, e.g.
a white shoulder and foreleg {rol) may be combined with a white
face {kwe roal), black spots {rol kwac), speckling {rol cuor), brown
patches {rol paara), white back {kar roal), white face and black
ears {kur roal), &c. There are at least twenty-five terms which
include the rol distribution, and the other distributions likewise
have wide ranges of combinations with colours and with one
another.
As I shall elsewhere, and at length, analyse the principles of
colour terminology and abstract the rules of nomenclature,^ I
need no more than remark here that it is evident from the few
examples cited that there are several hundred colour permu-
tations.
Some colours and combinations of colours are associated with
animals, birds, reptiles, and fish, and this association is often
indicated by secondary terms of reference and by ritual usages, e.g.
lou (mouse-grey) is the bustard, nyang (striped) is the crocodile,
lith (sandy-grey) is associated with manlieth, the grey kestrel,
thiang (bay) is the tiang, dwai (brown with white stripes) is the
female sitatunga, kwe (white-faced) is the fish eagle, kwac (spotted)
is the leopard, cuor (speckled) is the vulture, gwong (spotted) is
the guinea-fowl, nyal (brown-spotted) is the python, &c. These
linguistic identifications and other colour associations lead to many
' ' Nuer Cattle Terms ', to appear in Sudan Notes and Records^
INTEREST IN CATTLE 45
fanciful elaborations of nomenclature, e.g. a black ox may be
called rual mini, charcoal-burning or won car, dark clouds ; a brown
ox riem dol, red blood, or rir dot, red tree-cobra ; a blue roan ox
bany yiel after the blue heron ; a mouse-grey ox duk lou, the shady
gloom of forests, &c. These fancy names add greatly to the list of
Nuer cattle terms.
Besides the vast vocabulary which refers to colours, distri-
bution of colours, and colour associations, cattle can also be
described by the shape of their horns and, as the horns of oxen
are trained, there are at least six common designations in use
besides several fancy names. Words denoting shape of horns
add considerably to the number of permutations, for they can
be combined with many of the colour and distribution terms,
e.g. a sandy-grey cow with horns which almost meet in a curve
above the head is a duot lieth, a shorthorn with rial markings is
a cot rial, a brindled ox with one horn trained across its face is a
gut nyang, &c. The ears of cattle, sheep, and goats are often
cut in different shapes and it is permissible, and with sheep and
goats usual, to describe them by reference to these incisions.
Sheep and goats have very different mixtures of colours from
those one finds among cattle, but the same terms can be used to
cover all combinations, because they are never exact descrip-
tions of colour dispositions but represent ideal distributions,
to one or other of which any actual disposition approximates.
A further range of permutations is created by prefixes which
denote the sex or age of an animal, e.g. tut, bull, yang, cow,
thak, ox, nac, heifer, math, male calf, dou, female calf, kol, calf
which has not yet begun to graze, and so forth. Thus one may
speak of a tut ma kar looka, dou ma rial, thak ma cuany, &c.
Indeed, if we were to count every possible mode of referring to
animals of the flocks and herds they would be found to number
several thousand expressions — an imposing and complicated
system of ramifications which bears eloquent witness to the
social value of cattle.
Furthermore, as we have mentioned, every man takes one of
his names from the term by which one of his oxen is described,
and these ox-names are the preferred salutations among age-
mates. A youth generally takes his first ox-name from the
beast his father gives him at his initiation, but he may later
46 INTEREST IN CATTLE
take further names from any oxen of his herd which delight
him. Men salute one another with these names and shower
them, with many a fanciful elaboration, on their companions at
dances. They also chant them when they display themselves
with their oxen in camps, sing them in their poems, and shout
them when they spear men, animals, or fish.
A man may be called by the identical name of his ox, e.g.
Bi(l)rial, Kwac(c)uor, Werkwac, and so forth, but generally one
part of the term is dropped and the other part is prefixed by a new
term, usually descriptive of some ornament worn by the ox or
some characteristic of it, not employed in defining its own name,
e.g. luth, a large bell (Fig. 4), gier, a small bell, lue, a long tassel,
dhuor, a short tassel (Plate IV), wak, the tinkling of an ox-bell,
lang, a brass ring attached to an ox's collar or tethering-cord (one
can be seen on the animal in the foreground of Plate II), rot,
bellowing of oxen, cwai, fatness, boi, shining whiteness, &c. Thus
a man whose favourite ox has rial distribution of colours may be
called Luthrial, Gierrial, Luerial, Dhuorrial, Boirial, and so on.
When ox-names are used between age-mates at dances they are
generally preceded by dance-names which are selected to harmon-
ize with the ox-names, euphony being considered of great impor-
tance in all these word formations. Ox-names are voluminous and
abstruse, and in describing them, as in describing cattle-colours, I
have not only made a meagre selection from the wealth at my dis-
posal, but have also chosen for illustration the simplest examples
and neglected the more obscure.
Names of cattle, especially of oxen, and ox-names of men are
used profusely in songs. The Nuer, like most pastoral peoples,
are poetic and most men and women compose songs which are
sung at dances and concerts or are composed for the creator's
own pleasure and chanted by him in lonely pastures and amid
the cattle in camp kraals. Youths break into song, praising
their kinsmen, sweethearts, and cattle, when they feel happy,
wherever they may be. I give a free translation of the first verses
of two songs, the first sung by girls as they sit together in the
evening after the day's work is done, and the second sung by its
creator when he is happy.
I. The wind blows wirawira;^
Where does it blow to ?
' Literally 'My wind'. The singer runs against it and seems by so doing to
add to its strength. This is the north wind which blows at the time of rich
INTEREST IN CATTLE 47
It blows to the river.
The shorthorn carries its full udder to the pastures;'
Let her be mUked by Nyagaak ;
My belly will be filled with milk.
Thou pride of Nyawal,
Ever-quarrelling Rolnyang.^
This country is overrun by strangers ;
They throw our ornaments into the river ;
They draw their water from the bank.^
Blackhair my sister,
I am bewildered.
Blackhair my sister,
I am bewildered.
We are perplexed ;
We gaze at the stars of God.*
2. White ox good is my mother
And we the people of my sister.
The people of Nyariau Bui.
As my black-rumped white ox.
When I went to court the winsome lassie,
I am not a man whom girls refuse.
We court girls by stealth in the night,
I and Kwejok
Nyadeang.s
We brought the ox across the river,
I and Kir j oak
And the son of my mother's sister
Buth Gutjaak.
Friend, great ox of the spreading horns,
Which ever bellows amid the herd.
Ox of the son of Bui
Maloa.6
It is not necessary to add more examples of cattle-terms and
pasture when the cows give plenty of milk: hence the connexion between
the first three lines and those which follow them.
' The cow has refused to suckle its calf or to be milked before going to graze.
2 Nyagaak is the sister of the poet. Pride (gweth) is the dance-name of a
girl, Nyawal. Rolnyang is a youth's ox-name.
3 The strangers are Government forces. The reference to drawing water
from the bank is obscure.
^ Blackhair is a girl's name. The Nuer are perplexed by foreign invasion
and the last line is a prayer to God to help them in their adversity.
5 The ox referred to in the first and fourth lines is the poet's ox. Kwejok
is a friend, whose mother is Nyadeang.
6 Buth is the birth-name of a friend whose ox-name is Gutjaak. The poet,
who is a son of Bui Maloa, addresses his ox as his friend in the final lines.
48 INTEREST IN CATTLE
their uses to demonstrate that we are deahng with a galaxy of
words in the arrangement of which a thesaurus of some mag-
nitude might be compiled. I need only emphasize that this
intricate and voluminous vocabulary is not technical and
departmental but is employed by every one and in manifold
situations of ordinary social life. I have only treated a
fragment of a fragment of the linguistic field relating to cattle,
I could enter into further detail, but, at best, I have only
surveyed, and in an amateur way, that field, which invites
broader and more speciahst research. My purpose has been to
draw attention to it and to show how a study of the dominant
interest of Nuer might be approached from this angle. The
subject is necessarily vast, because, as we have seen, it is not
possible to discuss with Nuer their daily affairs, social con-
nexions, ritual acts, or, indeed, any subject, without reference
to cattle which are the core round which daily life is organized
and the medium through which social and mystical relations
are expressed. Nor is Nuer interest in cattle confined to their
practical uses and social functions, but is displayed in their
plastic and poetic arts, in which they are the chief theme.
The over-emphasis on cattle is thus strikingly shown in lan-
guage, which, moreover, by compelling reference to cattle,
whatever be the subject of speech, continually focuses attention
on them and makes them the superlative value of Nuer hfe.
IX
Another way in which Nuer engrossment in cattle can be
illustrated — our last exemplification thereof — is by noting how
r readily and frequently they fight about them, for people risk
I their lives for what they greatly value and in terms of those
values.
r At the present time cattle are the main cause of hostility
1 towards, and suspicion of, the Government, not so much on
account of present taxation as of earlier tax-gathering patrols
which were little more than cattle raids and of the avowedly
plundering expeditions of the Egyptian Government era that
preceded them. Nuer war with the Dinka has been almost
entirely offensive and directed towards appropriation of herds
and annexation of grazing grounds. Cattle have also been the
INTEREST IN CATTLE 49
chief occasion of strife among Nuer themselves. Indeed, after
a successful raid on Dinka stock there is often further fighting
over the booty. Moreover, Nuer tribes raid one another for
cattle. Thus the Leek raid the Jikany, Rengyan, and other
western tribes, and cattle raids are of common occurrence along
tribal boundaries elsewhere, for to 'steal' {kwal) cattle from
another tribe is regarded as laudable. Within a tribe, also,
fighting frequently results from disputes about cattle between
its sections and between individuals of the same section, even
of the same village or homestead. Nuer fight on shght provo-
cation and most willingly and frequently when a cow is at stake.
On such an issue close kinsmen fight and homes are broken up.
When ownership of cattle is in dispute Nuer throw over caution
and propriety, showing themselves careless of odds, con-
temptuous of danger, and full of guile. As my Nuer servant
once said to me: 'You can trust a Nuer with any amount of
money, pounds and pounds and pounds, and go away for years
and return and he will not have stolen it ; but a single cow —
that is a different matter.'
Nuer say that it is cattle that destroy people, for 'more
people have died for the sake of a cow than for any other cause'.
They have a story which tells how, when the beasts broke up
their community and each went its own way and lived its own
life, Man slew the mother of Cow and Buffalo. Buffalo said she
would avenge her mother by attacking men in the bush, but
Cow said that she would remain in the habitations of men and
avenge her mother by causing endless disputes about debts,
bride-wealth, and adultery, which would lead to fighting and
deaths among the people. So this feud between Cow and Man
has gone on from time immemorial, and day by day Cow
avenges the death of her mother by occasioning the death of
men. Hence Nuer say of their cattle, 'They will be finished
together with mankind', for men wiU all die on account of
cattle and they and cattle will cease together.
' It must not, however, be supposed that Nuer live in con-
tinuous turmoil: the very fact that they are prepared to resist
any infringement of their rights in cattle induces prudence in the
relations between persons who regard themselves as members
of the same group. It may be said, furthermore, that the great
50 INTEREST IN CATTLE
vulnerability of cattle coupled with the extensive living-space
required for them are compatible only with a far recognition of
conventions in the settlement of disputes, or, in other words,
the existence of a tribal organization embracing a large terri-
tory, and of some feeling of community over yet larger areas.
Fighting about ownership of cattle and seizing cattle for
what are claimed as debts and compensation for losses are of a
somewhat different order to raiding for cattle over which no
rights, other than the power of the strong, are asserted. War
against foreign peoples, as distinct from warfare within a
tribe, is almost entirely for plunder. Nuer war against the
Dinka, therefore, differs from most primitive warfare in that its
primary object is acquisition of wealth, for cattle are a form of
wealth that not only lasts a long time and reproduces itself, but
is, also, easily seized and transported. Furthermore, it enables
invaders to live on the country without commissariat. Crops
and dwellings can be destroyed, but cattle can be confiscated
and taken home. This quality, which has given pastoral
peoples a bias in favour of the arts of war rather than the arts
of peace, has meant that the Nuer are not entirely dependent
on their own cattle, but can augment their herds and restore
the ravages of rinderpest, and have, in fact, for a long time
increased their stock, and hence supplemented their food-
supply, by raiding ; a condition that has shaped their character,
economy, and political structure. Skill and courage in fighting
are reckoned the highest virtues, raiding the most noble, as
well as the most profitable, occupation, and some measure of
political agreement and unity a necessity.
We hasten to add that an explanation of warfare between
Nuer and Dinka by reference to cattle and pastures alone is too
simple a reduction. Hostility is expressed in terms of cattle, and
desire for cattle accounts for some peculiarities of the struggle
and some characteristics of the political organizations involved
in it, but the struggle itself can only be fully understood as a
structural process and we present it as such later.
We now pass to a brief examination of the oecological system
of which Nuer and their cattle form part to discover the con-
ditions in which cattle-husbandry is practised and how far its
practice in a certain environment influences pohtical structure.
CHAPTER II
OECOLOGY
I
From a European's point of view Nuerland has no favourable
qualities, unless its severity be counted as such, for its endless
marshes and wide savannah plains have an austere, monotonous
charm. It is throughout hard on man and beast, being for most
of the year either parched or a swamp. But Nuer think that
they live in the finest country on earth and, it must be admitted,
for herdsmen their country has many admirable features. I
soon gave up trying to convince Nuer that there is any country
more suited for cattle husbandry than their own, an attempt
rendered more useless since a few of them have been taken to
Khartoum, which they consider to be the home of all white
men, and, having seen the desert scrub of that latitude, have
been confirmed in their opinion that their land is superior to ours.
The grasses necessary for the welfare of the herds depend for
existence on suitable conditions of soil and water. The soils of
Nuerland are heavy clays, broken by the sun into deep cracks in
the drought and sodden in the rains. They hold up water and
consequently enable certain species of grasses to survive the dry
months and provide pasture for the cattle. Nuer and their
cattle would not, however, be able to live if it were not that
there are more sandy elevated spots on which they can take
refuge in flood-time and where they can practise horticulture.
Surface water comes partly from rainfall and partly from the
flooding of the rivers which traverse Nuerland and is more than
adequate to make grass. In an average year the rains commence
in April, when a few showers fall and the sky is overclouded, but
it is not till the end of May that they set in with a wiU. At their
maximum, in July and August, the weather is cool, even cold in
the mornings and evenings, the sun is overcast during most of
the day, and a south-westerly wind prevails. Showers become
lighter and less frequent in October and usually have ceased
altogether by the middle of November when the north wind
begins to blow. It blows consistently down the valley of the Nile
52 OECOLOGY
till March. In March and April the heat is intense. The rainfall,
which is fairly even over all parts of Nuerland, is not so heavy
as farther east, on the Ethiopian Plateau, or south, in the basin
of the Victoria Nile and along the Nile-Congo divide, though
the effects are more felt, because clay beds hold up the water,
F M
N D
15
mm.
200
150
:
Rai
-ifall
i
Vlont
hly totals
Millimetres
100
-
50
m n n
n
Temperature and Rainfall estimates for Nuerland.
{Physical Department of the Egyptian Government.)
the flatness of the country prevents surface drainage, and the
annual flooding of the rivers occurs simultaneously.
The main rivers which so greatly influence Nuer life are
shown on the map on p. 8. They are the Nile itself, known in
these regions as the Bahr el Jebel, which derives its water from
the Plateau of the Great Lakes ; its western tributaries the Bahr
el Ghazal and the Bahr el Arab, which are fed by streams flowing
from the Nfle-Congo divide; the Baro, the lower reaches of
which are known as the Sobat, coming from the Ethiopian
OECOLOGY 53
Highlands ; and the Pibor which flows from the same direction
and also drains to a lesser extent the northern slopes of the
Plateau of the Great Lakes and the Sudan Plains. The Bahr el
Zeraf is another channel of the Bahr el Jebel.' All these rivers
flood at the time of the rains and, owing to its flatness, the
country is turned into a vast morass.
cm
2oor Rise and Fall oFthe Sobat in each month
(Centimetres)
Variation oF qauge reading
m. on tne Sobat
(Metres)
AMJ JASON D,
Rise and fall of the Sobat river.
(Physical Department of the Egyptian Government.)
The whole of Nuerland is an almost dead level plain, covered
during the rains with high grasses reaching to the waists, and
near streams, where they are higher and coarser, to the shoulders
of the tall Nuer. It approximately coincides with the extension
of true savannah in the Sudan. Here and there are patches of
thornwood forest, but often no tree is visible in any direction
^ For an excellent description of these rivers see H. E. Hurst and P. Phillips,
The Nile Basin, vol. i, 193 1.
54 OECOLOGY
and a desolate waste stretches everywhere to the horizon
(Plates VI and XI {a) and {b)). A belt of forest sometimes lines
a river where its banks are high, but never extends far inland.
North of the Sobat, on its lower reaches, true savannah gives
way to thorny savannah and beyond the southern extremities of
Eastern Nuerland one enters park-like savannah forest which
becomes thicker the farther south one proceeds, though it
usually turns to marshland as the Bahr el Jebel is approached.
The southern borders of Western Nuerland fringe ironstone
country, likewise covered with savannah forest. As a rule when
the rivers are in flood they have no banks and the country
lying on either side of them is swamp threaded with wide
lagoons, often running parallel to the main channel. This is
especially the case with the Bahr el Jebel and the greater part of
the Bahr el Ghazal and the Bahr el Arab, the Jebel and the
Ghazal being practically united by surface water in the rainy
season. The Bahr el Zeraf is bounded by swamp to a lesser
degree and the lower reaches of the Sobat not at all.
This vast plain is threaded with depressions, like that shown
on Plate XIX {b), which run in all directions, often crossing one
another, and linking up with the main rivers. Where continuous,
these depressions have the appearance of small rivers, though
water seldom flows in them. While rain is falling on the country
the main rivers flood into these depressions, making a network
of waterways which prevents drainage from the saturated earth,
so that the rainwater lies everywhere in deep puddles which
slowly extend till by the middle of June the whole country,
except for occasional higher land, is inundated. The water re-
mains several inches deep till September, and Nuerland has the
appearance of a great grass-covered swamp ; there are streams,
lagoons, and pools wherever there are shght depressions, and
islands, on which are perched villages, wherever there are
ridges and knolls. The rivers begin to fall about the same time
as the rains decline, the fall being most rapid on the Sobat
(diagram on p. 53). The blazing sun then quickly evaporates
the surface water, while the streams, instead of being channels
of overflow from the rivers, now feed them, and by the middle
of November the grasses are dry enough to be fired. By the
end of December a great part of the country has been burnt
PLATE VI
^
;^1^-
OECOLOGY 55
and is cracked into deep fissures. The wet and dry seasons
are therefore very pronounced and the transition from one to
the other sudden.
Scarcity of rain is probably more serious than low river water,
but both may inconvenience Nuer to the point of famine, because
sufficient water may not be held up in the clay beds to enable
the grasses to recover from firing; inland watercourses may
quickly dry up and compel movement to lakes and rivers earlier
than is desirable; and there may be a shortage of the marsh
pasturage which is usually the mainstay of the cattle at the end
of the dry season. Insufficient rain may also destroy the millet.
Moreover, it is probable that low rainfall over the whole of
North-East Africa causes insufficiency of rain and low rivers to
occur in Nuerland simultaneously. Western Nuerland is less
subject to drought than Eastern Nuerland, and to the west of
the Nile there is always water within easy reach of villages.
This seems largely due to the fact that the Bahr el Jebel and the
Bahr el Ghazal do not sink in their beds to any extent, since
they are fed by perennial streams and have enormous marsh
and lake reservoirs. High rainfall and high river water probably
also go together and in such years of flood it is difficult for the
cattle to find enough grazing to maintain life.
The main characteristics of Nuerland are: (i) It is dead fiat.
(2) It has clay soils. (3) It is very thinly and sporadically
wooded. (4) It is covered with high grasses in the rains. (5) It is
subject to heavy rainfall. (6) It is traversed by large rivers
which flood annually. (7) When the rains cease and the rivers
fall it is subject to severe drought.
These characteristics interact with one another and compose
an environmental system which directly conditions Nuer life and
influences their social structure. The determination is of so
varied and complex a nature that we do not attempt to sum-
marize its full significance at this stage of our description, but
shall ask ourselves a simpler question: to what extent are the
Nuer controlled by their environment as herdsmen, fishermen,
and gardeners ? We have demonstrated that their chief interest
is in their herds and shall first discuss how this interest, com-
bined with physical conditions, necessitates a certain mode of
life.
O P
OECOLOGY 57
Here we make only two observations of a more general order.
(i) Although the Nuer have a mixed pastoral-horticultural
economy their country is more suitable for cattle husbandry
than for horticulture, so that the environmental bias coincides
with the bias of their interest and does not encourage a change in
the balance in favour of horticulture. If it were not for rinder-
pest — a recent introduction — it might be possible to live a
purely pastoral life, but, as we shall see later, a purely horti-
cultural life would be precarious. (2) Nuer cannot, except in a
few favoured spots, live in one place throughout the year. The
floods drive them and their herds to seek the protection of
higher ground. Absence of water and pasture on this higher
ground compels them to move during the drought. Hence their
life is of necessity migratory, or, more strictly, transhumant. A
further reason that urges them to change their abode according
to the seasons is their inability at the present time to subsist
solely on the products of their cattle. A milk and meat diet has
to be supplemented by grain and fish; and whereas the most
suitable place for cultivation of millet is inland, on the edge of
slightly elevated ground, fish are found in rivers which are
generally distant from these elevated stretches.
II
Excess or insufficiency of water is the first problem that faces
Nuer. It is essential ^hat cattle should be protected from the
water that covers the country in the rains, for they quickly get
diseases of the hoof if they have to stand in water for long
periods. Village sites are selected on the only spots affording such
protection to man and beast : patches of slightly higher ground.
When the rains have ceased water supplies near villages are
soon exhausted, because, naturally, the highest and driest sites
have been selected for building, and it becomes necessary to
move to pools, lakes, lagoons, marshes, and rivers. Owing to the
vast rivers that traverse Nuerland and the thorough irrigation
they give to the country through a network of channels there is
seldom difficulty about finding surface water, though people
may have to go far to obtain it. Only in parts of Lou, Gaawar,
and Eastern Jikany, as far as is known, are they regularly
forced to dig wells in the beds of streams at the height of the
OECOLOGY 59
drought. Twenty years ago this was probably more customary
among the Lou than to-day, for they had not then the undis-
puted access to open water that most sections now enjoy. The
wells, which have to be redug each year, are two to three feet in
diameter and twenty to thirty feet in depth, and their excava-
tion takes two or three days of hard labour. The water, about a
foot deep, is clean and fresh and the wells are frequently cleaned
out, steps being cut in the walls for this purpose (Plate XV (6)).
Each household has its own well which is surrounded by shal-
low mud troughs where the cattle are watered three times daily.
Considerable labour is required to draw water for them and
much attention is devoted to prevent it from being fouled by
sheep and goats, which have their special troughs. Fights some-
times break out over these troughs.
The problerrLPi water is closely related to that of vegetatioiki!^
In their seasonal movements Nuer seek pasturage as well as
drinking-water and they take the cattle to where they know that
both can be obtained. When herdsmen drive the cattle from
camp to grazing grounds they do not guide them haphazardly
across the plain, but with purpose towards stretches of succu-
lent grasses. It is probable, too, that not only are daily and
seasonal movements influenced by distribution of grasses, but
also that the direction of Nuer expansion has been controlled by^
their habitat. Nuer claim that they have not overrun the
country of the Ngok Dinka because it is poor pasture land and
that they are little interested in the Shilluk kingdom for the same
reason.
The early rains are the season of fatness, for then the grasses
germinate, or renew their growth after the long drought, and the
cattle can graze on the young shoots to their content. As the
rains advance, grazing becomes more difficult, the ground being
flooded and the vegetation rank, and in years of high water
may be a serious problem. The cattle have to rely on the short
grasses that prevail on village ridges: a further reason that
compels Nuer to occupy these sites in the rains. When rain
ceases, the overcropped grasses on these ridges soon wither,
while the rank grasses of the plains impede the movements of
the herds and no longer provide good grazing. Therefore Nuer
hasten to burn them as soon as they are dry, since some species
Miks
10 20 30
Sketch-map showing the direction (indicated by arrows) of movement in the
dry season of the Zeraf tribes (after Mr. B. A. Lewis).
OECOLOGY 6i
send up new shoots a few days after being fired — probably those
that have roots long enough to tap water held up in the clay
beds and a clump formation that protects the core of the plants
from the flames. If it were not for this habit cattle could not
survive, at any rate inland, during a dry year. When the
grasses have been fired cattle can wander as they please, being
no longer frustrated by surface water and rank growth, and
satisfy their appetites on the new shoots. As water becomes
scarcer and grazing poorer Nuer fall back on permanent water
where they make large camps and the cattle can graze on marsh
plants that abound in numberless depressions and make good
milk. In May, when the new rains set in, they are able to return
to their villages. The few cattle Nuer possess, the vast spaces
they can exploit, and their nomadic existence ensure that there
is nowhere serious overgrazing.
Variation of water-supplies and vegetation thus forces Nuer
to move and determines the direction of their movements. In
a later section we shall see that fishing is another important
consideration in these movements. In late November or early
December the youths and some of the girls take the cattle from
the villages to camps, generally some miles away, leaving the
older people to harvest the second millet crop and repair huts
and byres. Usually a few lactating cows are left behind to pro-
vide milk for the small children. These early camps {wee jiom)
are made near pools in a place where the grasses have previously
been fired. In Lou country they are often made in thornwood
forest, where Balanites aegyptiaca is abundant, but in many
parts of Nuerland, especially to the west of the Nile, they are
formed on stream banks for fishing. When the second harvest
has been reaped the cattle are brought back to the villages to
eat the millet stalks, if they are near enough. As pools dry up,
pasturage is exhausted, or fishing becomes poor, the youths
make new camps, where they are joined by the married people,
and they may move camp several times before settling in their
final camping grounds [wee mai) at the side of lakes or rivers
in January and February. The early camps are small, being
manned by a few kinsmen, but they tend to get larger as the
season advances and water becomes scarcer, and the final camps
may contain several hundred persons.
62 OECOLOGY
The dry season movements of the Lou, the Eastern Jikany, and
the Zeraf River tribes are shown on the sketch-maps accompany-
ing this section. Of the Sobat tribes, the Lou stay inland as long
as they can and in a wet year may remain inland throughout the
dry season, falling back on the deeper pools, some large enough
to be called lakes, e.g. Muot tot, Muot dit, Fading, Fadoi, Gwong-
gwong, Yollei, Tepjor, and Nyerol. If forced to leave them the
Gun primary section moves north to the Sobat and south-west to
the flooded plains of the Bahr el Zeraf in the country of the Twic
Dinka, and the Mor primary section moves north-east to the
Nyanding river and east to the Geni and Pibor. In the old days
fighting frequently occurred if the Lou moved to these camping
sites, because the banks of the Sobat were in the hands of the
Balak Dinka while their occupation of the Geni and Pibor was
disputed by the Anuak and Beir and their move to the south-west
was trespass on Dinka grazing grounds. Even the lower reaches
of the Nyanding do not belong to the Lou, but to the Gaajok.
Therefore they probably only moved to these sites when abso-
lutely necessary.
The Eastern Jikany have not so much need to move as have the
Lou. Nevertheless they all fall back on rivers or marshes before
the drought is far advanced. The three primary sections of the
Gaajok move as follows: the Laang concentrate on the Sobat, the
Wangkac move south-east to the banks of the Pibor and Gila, and
the Vol either camp near the junction of the Wokau and Sobat or
northwards along the edges of the Machar swamp. The Gaagwang
move to the western end of Khor Makwai. The Gaajak primary
sections move as follows : the northern Thiang sections (Kang and
Lony) go to the Daga river while the southern Thiang section
(Tar) and the Reng, Nyayan, Wau, and Cany sections camp on
the banks of the Baro, Jokau, Adura, and Makwei, mainly in
Ethiopia.
The three Zeraf river tribes fall back on the Bahr el Jebel and
the Bahr el Zeraf and streams flowing into these two rivers, as
shown in Mr. Lewis's sketch. The Gaawar who Hve on the left
bank of the Zeraf can build their villages on high ground near that
river and do not have to move far in the drought.
The Western Nuer mainly camp on inland streams which are
no great distance from their villages. The Karlual primary section
of the Leek tribe mostly concentrate first on the Loogh stream
and afterwards on the Cal and Wangtac, all of which link up with
the Bahr el Ghazal. The Dok camp near pools in the beds of
inland streams. The Western Jikany move to the Jikany
marshes at the edge of the Bahr el Ghazal. Without giving
PLATE VII
a. Homesteads on mound (Lou)
^T uAi ■''*?PI~^Hr'
&. Homesteads on mound (Lou)
OECOLOGY 63
further details, which, indeed, I could not furnish, about the
disposition of the Western Nuer in the dry season, it may be said j:^
that they travel very much less than most of the Eastern Nuer
tribes, especially the Lou.
Different villages and sections tend to move about the same
time and to visit the same pools each year, though time and
place and, to some extent, degree of concentration vary accord-
ing to climatic conditions. Usually, however, the main dry
season camps are formed yearly at the same spots. When the
rains commence in May the older people return to the villages
to prepare the ground for sowing and they are joined there in
June by the youths and girls, who bring the cattle with them.
When the younger people return the whole camp is broken up
on the same day, and the cattle are driven to their village with
as few halts as possible. The move from villages to camps, on
the contrary, is less concerted and abrupt. The younger mem-
bers of two or three families form a small camp, after the firing
of the grasses in November, when and where they please.
Several days later they may be joined by other people from the
same village, or these may form a separate camp. There is still
water and some grazing on the village sites and there is still
work to be accomplished in gardens and homesteads. Thus,
while camps change into villages overnight, villages change
into camps more slowly. By the height of the drought every one
is in camp and the villages are silent and deserted.
The year thus consists of a period in villages and a period
in camps, and the camp period is divided into the early part of
the dry season when the younger people move from small camp
to small camp and the latter part of the dry season when every
one is concentrated in large camps round permanent water
which they do not leave till they return to their villages.
Ill
Nuer are forced into villages for protection against floods and
mosquitoes and to engage in horticulture, and are forced out
of villages into camps by drought and barrenness of vegetation
and to engage in fishing. I describe these villages and camps
very shortly.
Some tribes are better off for sites than others. The Lou
64 OECOLOGY
and the Eastern Jikany are especially fortunate in this respect,
but in Western Nuerland, though there are many suitable
ridges, the country is more inundated and there are few
elevated stretches of any size, except, as far as my observa-
tions go, among sections of the Leek and Dok tribes, which
are better off than those tribes which lie in between.
What is needed for a village site is not only room for building
but also for grazing and cultivations. Many villages are perched
on mounds (Plates VII and XVI), the surfaces of which are
accumulations of debris, at the edge of those rivers that have
banks in flood, and the cattle graze on the slopes or in near-by
thornwood forest, where also gardens are cultivated. More
usually homesteads are strung out along sandy ridges (Plate
VIII), a mile or two in length and a few hundred yards in
breadth, sites which allow greater spacial division between
them, and the gardens run at their backs and the grazing-grounds
to their fronts. In some parts of Nuerland, especially in the
favoured tribes I have mentioned, stretches of higher ground,
sometimes several miles in extent, enable the people to build
anywhere, except near the narrow depressions that thread them,
and httle groups of homesteads are dotted here and there, sur-
rounded, and separated, by their gardens and grazing grounds.
Nuer prefer to dwell in this greater privacy and show no in-
clination for true village hfe.
By making their villages on elevated ground Nuer keep them-
selves and their cattle above the flood that lies everywhere on
the vast plain beneath and gain some protection from the
swarms of mosquitoes which breed in the standing water. In
Western Nuerland I have seen low mud dykes at the foot of
occupied ridges to keep back the water in years of heavy
flood. As they bufld always on the highest spots the torrential
rains that fall daily after June run off the slope so that the hard
floors of kraals soon dry. How terrific are these rains may be
judged from Plate XIV, taken from under the awning of my
tent during a moderate shower in August, and how considerably
they flood all but village sites may be seen in Plate XII, which
shows a millet garden, on a higher level than the plain, in
October. On a tour of Western Nuerland in October of 1936,
a fairly dry year, we walked almost continuously in several
OECOLOGY 65
inches of water for seventeen days, apart from having to cross
numerous deep depressions. In some parts of Lou and Eastern
Jikany, particularly in Gaajak, people build in open thorn wood
forest, but generally Nuer prefer to make their villages in open
country, even where there is forest near-by, because their cattle
are better protected from wild beasts, insect pests, and damp,
and also, apparently, because millet does not do so well in
wooded country. Building in the open seems also to give free-
dom from termites.
A Nuer homestead consists of a cattle byre and huts. Their
byres are of size and workmanship that have evoked the admira-
tion of all travellers. Their form, and the appearance of huts,
may be seen in several plates, and their mode of construction is
excellently portrayed in Mr. Corfield's photograph (Plate XVIII).
It is only necessary to explain that the roofs are supported
by trunks of trees erected inside the barns. Both byres and huts
are of wattle and daub, though in Western Nuerland, where
there is less forest, bundles of millet stalks are employed as
rafters. Building and repairs generally take place early in the
dry season when there is plenty of straw for thatching and
enough millet to provide beer for those who assist in the work.
During the rains fences are erected from the byres along two
sides of the kraal and around huts to control the movements of
cattle and to prevent them messing the homesteads and damag-
ing the crops (Plate XVII). Grazing and the use of grass,
trees, &c., is the common right of all members of a village
community.
Families often change their place of residence from one part
of a village to another and from village to village, and, in the
case of small villages, if there have been many deaths, the cattle
are poorly, there have been fights within the village, or pastures
and cultivations are exhausted, the whole village community
may move to a new site. After about ten years both pastures
and gardens show evident signs of exhaustion on the smaller
ridges, and huts and byres need rebuilding after some five years.
In dry-season camps men sleep in windscreens and women
in beehive-huts, or both sexes in beehive-huts. These flimsy
shelters are erected a few yards from water, generally in a semi-
circle or in lines with their backs to the prevailing wind, and
66 OECOLOGY
are simply constructed, the roots of grasses, or occasionally
stems of millet, being tightly packed in a narrow trench to
make windscreens, and the tops of the grasses being bound
together and plastered with dung on the outside to make huts
(Plates XV {a), XIX («), and XXI {h)). The whole space within
a windscreen is occupied by a hearth of ashes on which the
men sleep around a fire, and the openings face the kraal. If
people do not intend to spend more than a few days at a site
they often sleep in the open and do not trouble to erect wind-
screens and huts. These light dwellings can be erected in a few
hours.
IV
Another circumstance that determines Nuer movements is
the abundance of insect life, which is an ever-present menace,
for the cattle get little rest from stinging flies and ticks from
morn till nightfall and would be worried to death if their
masters did not give them some protection.
Mosquitoes swarm in the rains, their ravages being terrible
from July to September, when, as soon as the sun sets, men and
beasts have to take refuge in huts and byres. The doors of the
huts are tightly closed, the air-holes blocked, and fires lit. In
the centre of the byres, which house the cattle, burn large dung
fires which fill them with smoke so dense that one cannot even
see the cattle. Young men sleep on platforms above the fires,
and if the smoke clears — the doors are shut, but it escapes
through the thatched roof — they descend to pile on more fuel.
By this means the beasts are enabled to obtain some rest at
night. At the end of the rains while the cattle are still in the
villages they are left in the kraals till their masters go to bed,
when they are shut in the byres to protect them from lions.
Fires are not made up so high at this season, for mosquitoes are
less troublesome, since there is no surface water, the millet has
been cut, and the grasses are close-cropped. Later those men
who remain in the villages, while the others go into camps, often
erect windscreens in the kraals and spend the night outside
with any cows which have been left behind. In the dry season
mosquitoes are absent except near pools and marshes, and even
in the vicinity of water they are not troublesome from January
to May, so that the herds can sleep in the open. Nevertheless,
OECOLOGY 67
they are surrounded by windscreens in which their masters
sleep at the side of dung fires from which smoke rises on every
side to envelop the camp.
Another unpleasant pest is the seroot fiy. It appears to be
seasonal, flourishing on cloudy days from May to July, though it
sometimes appears at other times of the year. The seroot attack
the cattle in the mornings and accompany them to pastures, where
they bite them so effectively that they are often driven, stained
with blood, back to camp, where dung fires are lit to give them
protection. On such days the cattle are unable to graze, fitfully,
for more than two or three hours. Another biting fly, the stomoxys,
is prevalent throughout the year, being especially conspicuous in
the dry season and early rains. It is probably this fly which is
responsible for the presence of trypanosomiasis in parts of Nuer-
land, especially among the Eastern Jikany, by transmitting the
trj^anosomes directly on its proboscis from beast to beast, for the
tsetse does not occur in Nuerland, except, perhaps, on its eastern
extremities, though its lethal character is well known to Nuer.
There are other insects which annoy the cattle, but it is not known
whether any of them carry disease in this latitude. Of these may
be mentioned the eight-legged cattle tick, which Nuer remove
from the bodies of their animals when they return to camp in the
evenings — though not frequently or systematically enough; an
insect called tharkwac which is said to live on the bodies of cattle,
though its bites do not cause bleeding ; a fly called miek ; and the
common black fly, like our house fly in appearance, which worries
the herds a good deal in hot weather — damp and cold seem to kill
it off. Red soldier ants occasionally infest the byres so that the
cattle have to be removed while the floors are sprinkled with ashes,
but they are rare in Nuerland. Nuer are helpless against most of
these insects, though, no doubt, smoke keeps them away to some
extent.
Speaking from personal experience I may say that one is con-
tinuously tormented by insects in Nuerland, especially by the
common black fly and the mosquito. It is evident that the
cattle suffer considerably from their attentions and there can be
little doubt that this constant irritation lowers their vitality and
affects their milk yield, for they seldom have any real rest. In
the circumstances their hardiness and endurance are remarkable.
Since several species of tsetse in the Southern Sudan carry
trypanosomes pathogenic to cattle it is fortunate for the
Nuer that they do not occur in his country. This immunity is
68 OECOLOGY
undoubtedly due to the absence of shady forest which, in its
turn, is probably due in the main to flooding and partly to firing.
The prevalence of tsetse in the forest belt that stretches along
the foot-hills of the Ethiopian escarpment has prevented the Nuer
from expanding as far eastwards as they might have done, for
it is clear that one of the reasons for their evacuation of Anuak-
land was the loss of stock. In relation to tsetse the Nuer, in their
present territory, occupy a more favourable position than most
Southern Sudan peoples.
A further consideration of great importance is the presence
of many microscopic organisms that cause disease in men and
cattle. This is not a subject about which we can say much.
With regard to stock it may, however, be said that it suffers from
many different diseases and that Nuer usually have some treat-
ment for them, though it may be doubted whether it has much,
if any, therapeutic value. The two most serious contagious
diseases are bovine pleuro-pneumonia, which in some years
causes heavy mortality among the herds, and rinderpest. Rin-
derpest entered the Sudan not more than fifty years ago and
Nuer refer to the period before it came, along with Arab in-
vaders, at the time the Boiloc age-set was being initiated, as
'the life of the cattle'. They have no way of combating the
scourge once it has attacked a herd, but they are aware that
an infected herd ought to be isolated. They are so used to it
now that they generally take the precaution of splitting up their
herd in the dry season, when it is prevalent, and placing beasts
in widely separated camps, so that if it breaks out in any part
of the country the cattle placed in another part may escape.
An animal which recovers from the disease is known to be im-
mune from further attacks and its value is thereby enhanced,
though Nuer are aware that its calves will not enjoy the same
immunity. Miss Soule tells me that Nuer claim to be able to
tell if an animal has had rinderpest by scraping the tip of the
horns and noting the colour beneath the surface. If it is
white the animal is immune. Though threatened with starva-
tion in the years of its visitation Nuer face the scourge with
resignation and detachment.
Rinderpest has caused and continues to cause terrible havoc
in the herds. It cannot exactly be estimated what, and how
PLATE VIII
Sandy rid.^e witli caltlt* byrrs on llu- hnii/on (l)i)k)
OECOLOGY 69
extensive, have been the social changes which have resulted
from this disturbance of the oecological equilibrium. Since
bride-wealth consists of cattle there must have been for a time
considerable dislocation of marriage arrangements, but stability
has been reached to-day by lowering the number of cattle
which have to be paid. A like assessment on a new footing does
not seem to have been achieved in homicide negotiations, in
which there is not the same goodwill between the parties as in
marriage, and blood-wealth appears to have been claimed up
to the present time at the old rate of payment, though Nuer
recognize that bride-wealth and blood-wealth should rise and
fall together. One cannot make any precise statement on the
matter, but it is likely that feuds were less easily concluded
than before and that consequently tribal relations were affected.
It may also be supposed that decrease in stock has led to a
general deterioration of the standard of living, for climatic
conditions do not allow adequate compensation by expending
further labour on horticulture. No doubt Nuer grew more millet
than before, but they must have suffered a decrease in their
total food-supply and, above all, in security. It will be seen in
a later section that Nuer can no more exist on a purely horti-
cultural economy than they can, at any rate since the introduc-
tion of rinderpest, exist on a purely pastoral economy. They
must have a mixed economy and can compensate on one side
or the other only to a very limited degree. Nuer sought to repair
their losses rather by extensive raiding of their Dinka neigh-
bours, passing on to them their own losses in stock. We know
that Nuer raided Dinka before rinderpest entered the country,
but it is likely that their foreign relations have been affected by
the greater stimulus to aggression. Other probable effects could
be noted, especially in kinship relations, but one can only guess
their importance, so we confine our speculations to a few of those
that may be supposed to have occurred in Nuer economy and
pohtical life.
V
It was noted in Chapter I that to exist Nuer have to make
use of a mixed economy, since their herds do not supply them
with adequate nourishment. It will be seen in a later section
that their millet harvest is often meagre and uncertain. Fish,
70 OECOLOGY
therefore, are an indispensable article of food, and the pursuit of
them influences seasonal movements.
The rivers teem with fish of many edible species which greatly
supplement the diet of the Nuer in the dry season and enable
them to survive years in which the crops fail or there are epi-
demics among the herds. In choosing camp sites opportunities
for fishing are considered no less than water and pasturage.
Nevertheless, Nuer do not regard themselves as a water-people,
and despise people like the Shilluk who, they say, live mainly by
fishing and hunting hippopotami. In spite of this suggestion of
superiority Nuer enjoy fishing and the feehng of well-being a
full fish diet gives them. One may judge how great their catch
must be at the height of the dry season from the fact that one
can see along the Baro and Sobat cattleless fishing camps [kal)
in which, except for a little grain, goat's milk, and an occasional
wild beast, people live on fish alone for several weeks. These are
poor people who either have no cattle or have one or two
cows which they have placed in the care of richer relatives, for
no Nuer would live without his cattle if he could help it, and
they are despised as persons of Anuak or Balak Dinka descent.
Some Nuer tribes fish more than others, according to their
opportunities. Thus Lou country has poor fishing compared
with Eastern Jikany which has a network of waterways. Tribes
and tribal sections jealously guard their fishing rights, and
people who want to fish extensively in a pool must first obtain
permission from its owners if they do not want to provoke
fighting.
It is the seasonal flooding, due to the rise and fall of the rivers
and the flatness of the country, that allows Nuer to kill fish in
such large numbers, for they are carried out of the rivers, where
they are little accessible to the simple methods of Nuer fishing,
into streams and lagoons where they are more vulnerable. The
best months are November and December, when the rivers
begin to fall and to drain streams and lagoons, which can be
dammed at suitable points and the fish speared in their efforts
to break downstream. Fishing from dams is conducted mainly
at night, fires being lighted behind the fishers, who fix their
attention on a line of withies placed upstream from the dam
and hurl their spears at whatever point of the line fish reveal
OECOLOGY 71
their presence by striking the withies. My friend the late
Mr. L. F. Hamer, who took the photograph reproduced in
Plate XXII (a), about sunrise, reckoned that quite a hundred fish
are caught from a dam in a single night. The dams are lowered
as the water falls.
Fig. 10. Instruments for attracting fish.
As the dry season advances a great number of fish are im-
prisoned in lakes and lagoons from which there is no outlet,
and as these dry up they are confined to a smaller and smaller
expanse of water and are killed by individual fishers using
barbed spears and long harpoons (Plates X and XXII (b)), and,
at the end of the season, by battues in which gaffs and basket
traps may also be employed. Fishing is fairly consistently pro-
ductive throughout the drought, its yield increasing somewhat
in the battue period and rising to a second peak at the
72 OECOLOGY
commencement of the rains, in April and May, when the rivers
begin to rise again and bring the fish into the shallows where
they are easily speared among the reeds and weeds. There is
very little fishing — by spear in occasional lagoons — after the
Nuer have returned to their villages, for these are mainly in-
land, away from open water, and the streams are too deep and
too infested with crocodiles to encourage fishing. Also, the fish
are widely distributed by flood water and protected by rank
marsh vegetation. Those Nuer whose villages are on the banks
of large rivers sometimes fish from dug-out canoes (Plate IX)
with long harpoons, but the Nuer possess very few good canoes
— and these they have exchanged or stolen from the Anuak —
for they have neither the wood nor the skill to make more than
crudely burnt-out vessels of palm and sycamore.
Nuer are fine spearsmen but, otherwise, not very ingenious
fishermen. Except for occasional fish spitted with amazing rapid-
ity from the bows of canoes by harpoons and spears as they jump
or rise to the surface, Nuer never see their prey, but either hurl
their spears at random into likely spots amid weeds and reeds
or set up grasses and withies to indicate the presence of fish.
A myth recounts that once upon a time all fish were visible to
the human eye but that God later made them invisible in the
water. The chief Nuer weapon is the barbed spear, though the
harpoon is considerably used. When fishing in shallows with
spears they sometimes strike the water with a ball of twine or
the shell of a giant snail attached to a stick (Fig. lo) to attract
the fish. Fish are eaten roasted or boiled.
VI
Nuerland is also very rich in game, though Nuer do not exploit
this source of food extensively. There are vast herds of tiang
and cob; other antelopes are plentiful; and buffalo, elephant,
and hippopotami abound. Nuer eat all animals except carni-
vora, monkeys, some of the smaller rodents, and zebra, and they
kill the last for its skin and tail in the south of Lou country,
where it roams in the dry season. Lions abound, especially to
the west of the Nile, and are a serious menace to the herds, but
Nuer only kill them and leopards, the skins of which are worn
by chiefs, if they attack cattle in kraals or grazing grounds, as
PLATE IX
Harpoon-tishiiiL; irum canoe (Sobat ri\'cr)
OECOLOGY 73
frequently happens, especially in the early dry season. They
seldom go out to hunt game animals, except gazelle and giraffe,
and only pursue those that approach their camps. Indeed, they
cannot be considered keen hunters and may even be said to
treat the sport with a measure of condescension, holding that
only absence of cattle makes a man engage in it other than
casually. Their flocks and herds provide them with meat, and
this fact probably accounts in part for their lack of interest in
hunting, which is, however, also to be related to the nature of
their country, for its open plains offer little opportunity for
successful hunting with the spear.
Nuer refuse all reptiles, except the crocodile and the turtle.
Ostriches, bustards, francolin and guinea-fowl, geese, duck,
teal, and other waterfowl, abound, but Nuer consider it shame-
ful for adults to eat them and, except in very severe famines
it is probable that only children, poor cattleless men, and an
occasional elderly person, eat them, and then rarely and secretly
in the bush. They do not keep domesticated fowls and show
particular repugnance to the idea of eating them. Likewise eggs
are not eaten. No insects are eaten, but the honey of wild bees
is collected after the firing of the grasses in December and
January and is either eaten in the bush or with porridge at
home.
The only tribe who possess more than an odd gun or two are
the Eastern Gaajak, who obtain them from Ethiopia. Nuer hunt
with dogs and the spear, and rely on their fleetness, endurance,
and courage. Hunting in the rains is therefore impossible,
because not only does rank vegetation impede pursuit but also
animals can choose their ground, there being everywhere suffi-
cient water and grazing. At the height of the dry season they
are forced to drink at the same pools, lagoons, and river inlets
as men, and the dry bare country allows open chase, for except
in tracking giraffe, Nuer hunt by sight and not by spoor. It is
only gazelles, which are worried by dogs, and buffaloes, which
prefer attack to flight, that can be tackled by one or two men.
Other animals, such as tiang, cob, and buck, are only killed
when they approach a camp and can be cut off by a large
number of young men. I saw several animals killed on the
Sobat by being surrounded on the landward side and driven into
74 OECOLOGY
the reeds, where their only escape was by swimming the river.
In this district, and perhaps elsewhere, Nuer, spurred by hunger,
leave camp after the first heavy showers to look for giraffe
tracks and pursue these animals relentlessly till they overtake
them. This is only possible at the time of the first rains when
the animals still have to approach camps to drink while their
large hooves stick in the moist earth and slow down their move-
ments. The Nuer, particularly those of the Zeraf river, have a
reputation for their courage and skill in hunting elephants,
which are surrounded and speared by large parties.
r Nuer hunting has thus the same character of simplicity as
1 their fishing. They use little cunning and, except for the spiked-
wheel trap, no mechanical devices. This trap is used by immi-
grant Dinka in most parts of Nuerland and by some of the Lou
Nuer, who, nevertheless, regard it as a Dinka contrivance and
accordingly tend to despise it, regarding its use as unworthy of
men possessing cattle but allowable to poor persons who can
thereby obtain meat and even cattle, for giraffe-hair is highly
valued for necklaces, such as those which can be seen on Nuer in
many plates (especially Plates I and XXVIII (&)). Hence one
finds a few members of a camp setting traps while the rest
abstain. They are set around drinking-pools at the end of the
dry season or in the early rains when surface water is still
restricted and the ground is not yet moist enough to rot the
leather nooses. In 1930 on the Sobat large numbers of giraffe
were captured by means of spiked-wheel traps, but it appears
to have been an exceptional year. If game-pits are dug at all
by Nuer they are very rare and found only on the borders of
Dinka tribes, and the burning of stretches of grass that have
survived the annual firing in order to spear or club the animals
that flee the flames is a very casual practice. In the dry season
in Western Nuerland, and probably in other parts, though not
on the Sobat, men harpoon hippopotami along the tracks to
their nightly grazing grounds. Hippopotamus hunting is not,
however, considered a Nuer practice, but a habit of Shilluk and
some of the Dinka tribes ; and even in those parts where Nuer
hunt them it is said to be only men with a few cattle who do so.
We may conclude that hunting does not give Nuer much meat
and that they do not greatly esteem it as a sport.
OECOLOGY 75
VII
In most years wild fruits, seeds, and roots are not an important
item in Nuer diet. Their country is mainly treeless and fruits
therefore few; only 'wild dates' [Balanites aegyptiaca), found in
occasional patches to the west of the Nile and in more extensive
stretches to the east, provide much sustenance. The fruits riperj
from about January to March and both the kernels and their
sweet fleshy covering are eaten. A number of other fruits are
eaten, mainly by children. Initiated youths refuse most of them.
The seeds and roots of water-lilies [Nymphaea lotus), which are
found in pools and lagoons in the early dry season are much
relished. The seeds of 'wild rice ' [Oryza Barthii) are collected
and a number of wild plants that grow on village sites are used
as seasoning for porridge. In famine years much greater atten-
tion is paid to the wild harvest. 'Wild dates' are then a great
stand-by and people eat a wide range of fruits, ripening mainly
in the early part of the drought, which they neglect when hunger
is not severe, and make use of bush-yams and the seeds of wild
sorghum and other grasses.
VIII
It will have been noted that fishing, hunting, and collecting
are dry-season occupations and produce during that season the
necessary supplement to an insufficient milk diet. In the wet
season, when these activities are no longer profitable and the
milk yield tends to fall off, the heavy rains, which cause the
changes responsible for these losses in supply, produce conditions
suitable for horticulture, impossible in the drought, that will
replace them. The variation in food-supply throughout the year
and its sufficiency for life at all seasons is thus determined by
the annual cycle of oecological changes. Without grain in the
wet season Nuer would be in a parlous state ; and since it can be
stored they can, also, to a small extent, make provision against
famine in the drought.
Climatic conditions together with flooding and the flatness
of the country make it impossible to cultivate most Central
African food plants in Nuerland, and the Nuer are especially
unfortunate in having no root crop as a reserve in famine years.
76 OECOLOGY
It may be doubted whether they could cultivate any crops other
than those they now sow, without extensive irrigation in the
dry season, and this is incompatible with their nomadism. These
crops do not make an imposing list. The staple crop is millet
{Sorghum vulgare) , from which they reap two harvests, and they
sow some maize near their huts and a few beans [Vigna), their
only garden vegetable, among the millet. Besides these three
food plants they cultivate a little tobacco under the eaves of
their huts and plant some gourds to grow up kraal fences. The
millet is consumed as porridge and beer; the maize is mainly
roasted, though some is eaten as porridge ; the beans are eaten
stewed or cooked with porridge ; the tobacco is used for smok-
ing in pipes, as snuff, and for chewing ; and the gourds are,
according to their species, either eaten or fashioned into dairy
vessels.
Millet, the main crop, is the only one of these plants to the
cultivation of which we need make more than a passing refer-
ence. Once it is established it stands up well to conditions which
would be fatal to most plants, and it is worthy of notice that
wild species of sorghum flourish in this latitude. Maize is also
hardy but, although it is important to Nuer because it is the
first grain to be harvested when their supplies are running short
or are exhausted, its quantity is negligible except on the banks
of the Baro. Nuer distinguish between many types of millet,
mainly by the colour of their seeds, and know which are early,
and which late, varieties, in what order they ripen, which give
good flour for porridge, and which have sweet stalks for chew-
ing. The plant thrives in black clays which hold up water, but
shows much adaptability and ripens also on the more sandy
stretches on which the Nuer build their villages, though in such
spots it has less resistance to drought and a second sowing is
very uncertain. Eastern Jikany is probably the best millet
country in Nuerland, and, although in many parts it is too
flooded for a second crop, it is the only section I have seen
which can generally be relied on to yield sufficient grain to
support the population. Lak and Thiang tribal areas in the
Zeraf Island, which I have not visited, are said to be good grain
country.
Even millet will not flourish in standing water, so that the
OECOLOGY 77
gardens have to be made on higher land. Where circumstances
permit homesteads to be built over a wide stretch of country
gardens can be made almost anywhere between them, but where
they are strung out along a ridge there is less choice, for the back
of the ridge is too hard, does not retain enough water, and is
required for grazing. Cultivations are therefore made behind
the huts and byres between the summit of the ridge and the
sunken plain. Among the Western Nuer, in the higher parts
where there is a slight slope, small dams are often constructed
to prevent the water running from the gardens. If, on the con-
trary, the millet is likely to be flooded where the ridge slopes
into the plain, drainage canals, sometimes several feet deep and
over fifty yards in length, are dug to take excess of water into
the bush. These gardens form a continuous line behind the
homesteads and if there is sufficient land there for cultivation
people do not cultivate elsewhere. If there is not enough land
they cultivate along the ridge beyond the confines of the village
where it is too damp for building but not too flooded for millet
gardens, or in near-by forest.
There is enough land for everybody on the Nuer scale of
cultivation and consequently questions of tenure do not arise.
It is taken for granted that a man has a right to cultivate the
ground behind his homestead unless some one is already using
it, and a man can choose any spot outside the village which is
not occupied by the gardens of others. New-comers are always
in some way related to some of the villagers, and kinsmen do not
dispute about gardens. Moreover, owing to the conformation
of Nuer villages, there is a rough correlation between the size
of population and the area of cultivable land, for where suitable
land is limited so is space for building.
It has been remarked that when once established millet shows
great resistance to climatic variations. The difficulty is for it
to become established. It frequently happens shortly after
sowing that there is a short period of drought in which the
young shoots wither and die. Sometimes this is due to the
hungry people sowing too soon, but it is generally unavoidable,
because if they wait too long before planting, the millet becomes
subject to smut and neither bears nor ripens properly. It also
frequently happens that heavy and violent rain destroys the
78 OECOLOGY
seedlings by beating them into the sticky clay or by washing the
soil from their roots. I have not known a season in Nuerland
in which drought or excessive rain did not in some measure
destroy the crops after sowing. Elephants do much damage
when the millet has passed the vulnerable seedUng period and
I have known the gardens of a village to be partially destroyed
three years in succession by these beasts. Weaver-birds take an
annual toll when the millet is ripening, but Nuer do nothing to
scare them unless they are in such numbers that they threaten
to consume the whole harvest, when they erect bird-watching
platforms in the gardens. In some years — I do not know how
often — the country is visited by locusts which cause immediate
and wholesale destruction. At various stages of growth guinea-
fowl, crows, ostriches, and some of the smaller antelopes do
much damage, and waterbuck show a great liking for the second
crop. Sometimes, though I think rarely, Nuer build huts in
their gardens, if they are some way from the village, to guard
the millet from these depredations, but generally they show
little concern, hoping that the proximity of dwellings will keep
animals away and, if it does not, accepting the consequences
with a detachment that appears sometimes to be almost
indifference.
HORTICULTURAL CALENDAR
Apr. May June July Aug.
Sept. Oct. Nov. Dec. Jan.
Sow maize Harvest maize
Sow first
Harvest first
millet
millet
Sow
Harvest
beans
beans
Sow jaak
Harvest jaak
millet
millet
Sow tobacco
Harvest tobacco
Sow second Harvest second
millet millet
The calendar shown above is an approximation, since the time
of sowing and harvesting depends on the commencement of the
first heavy rains, which is subject to annual variations. The horti-
cultural season generally begins at the end of March for those who
are able to spend the drought in or near their homes on river banks,
but most people have to wait till they can return to their villages
in April or May before they can start clearing their gardens.
Married people usually return to their villages in the first half of
PLATE X
OECOLOGY 79
May and start to prepare the ground around their homesteads for
maize. At the end of the month, or in June, the younger people
return with the cattle and assist in clearing last year's millet
gardens and in the hard work of hoeing new ground. Prolific
grasses spring up with the millet which has to be weeded at least
three, and maybe four or five, times. The more weeding done for
the first crop the less is needed for the second crop.
While the first crop is ripening adjacent strips are cleared for
the second crop (Plate XI (b)), which is sown shortly before the first
is harvested, at the end of August or in September, and it grows
together with the new shoots that spring from the roots of the
first crop after its stems have been chopped down. In Eastern
Nuerland the second sowing is made among the standing crop or
on an extension of the garden, but in Western Nuerland and in
parts of the Zeraf Island the ground is so flooded that it is made
on prepared mounds of earth. I would again draw attention to
Plate XII, where these mounds are illustrated, for it admirably
shows what Nuerland is like in the rains and conveys a clear impres-
sion of the obstacles confronting Nuer horticulture. The second
crop has also to be weeded several times and, as the same garden
will be used for the first sowing in the following rains, the more
thoroughly it is weeded now the less work there will be to do then.
In many parts of Nuerland the second harvest may be almost as
big as the first, and it is probable that where local conditions are
adverse to one crop they favour the other. It appears that while
the sown millet of the second crop requires much moisture and
will not flourish if the rains fall off early the millet which sprouts
from the old stems does best in drier conditions and may be spoilt
by late heavy showers, so that one or the other is Ukely to ripen,
and in moderate conditions both may thrive.
In some districts of the Lou and Eastern Jikany tribes the cul-
ture of a slow-maturing millet called jaak {Sorghum durra) has
been copied from the Dinka. The spot chosen for cultivation is
not burnt at the annual firing of the bush, but early in the follow-
ing rains when the new grasses have sprung up among the old.
Although this kind of millet has certain disadvantages it has
caught on in a few places because it requires less clearing and
weeding and is very hardy.
Nuer know nothing of crop rotation, having indeed no crops
to rotate, nor of manures, though the droppings of cattle and
the ashes of burnt weeds and brushwood no doubt act as
fertilizers. They never allow their homestead gardens to lie
fallow for a year to enable the soil to recuperate, but plant year
8o OECOLOGY
after year till it is utterly exhausted, when they change their
homestead site, often moving for some years to a new village.
Nevertheless, they know that each year of tillage means further
deterioration and judge the impoverishment of the soil by the
size of the millet plants and of their yield and by the presence
of certain weeds which only flourish on exhausted ground. A
garden is cultivated annually for from five to ten years and
after it has been left fallow for a few seasons it is tested to see
whether the soil is still hard and caked, or loose, soft, and
ready to bear a new crop. In gardens cultivated in bush sites
away from the villages, on the other hand, people are not
confined by the gardens of their neighbours, and it is possible
to hoe a virgin piece of ground each year at one end and to
abandon an exhausted piece at the other end.
Nuer gardens are very small. Measurements do not give so
clear an idea of their size as noting that in an average year the
people in most parts of Nuerland have just enough grain to last
them till the following harvest, if they are very economical
during the dry season and subsist mainly on milk and fish,
while in a bad year they may have to go without porridge for
several weeks. Also it may be noted that Nuer do not build
granaries, but find small grass and earthenware bins, kept in
their huts, sufficient for storage. However, we may judge how
important millet is to them less by its quantity than by its
place in their total food-supply, for it is not merely a subsidiary
item of diet of high nutritive value but is an essential food,
since they would be hard pressed to maintain life without it.
Nuer fully acknowledge that this is so and in no way despise
horticulture, but are, on the whole, industrious gardeners.
Nevertheless, they consider that horticulture is an un-
fortunate necessity involving hard and unpleasant labour and
not an ideal occupation, and they tend to act on the conviction
that the larger the herd the smaller need be the garden. They
are herdsmen and not peasants. When I have drawn atten-
tion to badly kept gardens or remarked that the crops received
no protection from beasts and birds they have been unper-
turbed, for, while it would be disgraceful to neglect cattle,
there are no strong feelings about inattention to gardens. When
I have asked them why they do not sow more millet I have
OECOLOGY 8i
often received some such reply as: 'Oh well, such is our
custom. We have cattle.'
I wish to emphasize, in conclusion, the following points:
(i) that Nuer cultivate only enough grain for it to be one element
in their food-supply and not enough to live on it alone ; (2) that
with their present climate and technology considerable increase
in horticulture would be unprofitable ; and (3) that the domin-
ance of the pastoral value over horticultural interests is in
accord with oecological relations which favour cattle husbandry
at the expense of horticulture. Nuer values and oecological "^
relations, therefore, combine to maintain the bias towards
cattle husbandry, in spite of rinderpest having rendered it a
more precarious occupation than hitherto.
IX
It has been mentioned that Nuer must have a mixed econ-
omy in the given oecological relations because no one source -\
of food is sufficient to keep them alive, and that the dominant
food-producing activity at each season is determined by the
oecological cycle. The different elements of diet, therefore,
,have an oecologically determined relation to one another and I
these relations can be roughly plotted.
Milk foods, millet in the form of porridge and beer, a little
maize, fish, and meat, are the main items of Nuer diet. Milk
is a staple food all the year round, though cows probably tend
to give a smaller yield towards the end of the rains owing to
insufficiency of pasturage, and it is said that there is a tendency
for them to calve in numbers after the first harvest and there-
fore to cease lactating a few weeks earlier — if this is correct it is
possibly due to the hot weather in February and March
bringing the cows on heat. This seasonal tendency and the
respective contributions of cattle and millet to Nuer food-
supply are brought out in a story in which Cow and Millet
have a dispute. Cow says that Millet is a person of no impor-
tance and that it is her milk that keeps people alive throughout
the year, while in time of famine they can eat her flesh and live.
Millet replies that Cow's claims are doubtless just, but that
when she is ripe the children are glad, for they chew the sweet
stems and rub out the grain between their hands and eat it,
4507 M
82 OECOLOGY
and there is plenty of porridge and beer. Cow argues that
anyhow porridge without milk is unpalatable and that her milk
will be finished by the time Millet is ripe. It is difficult to
confirm these variations in the yield of milk or to estimate their
importance, but the tendency for it to rise slightly in the dry
season is indicated in the chart opposite.
Millet is consumed as porridge and beer in large quantities
in the months between the first harvest and departure for dry-
season camps. If the harvest has been good people like to eat
their daily porridge in camps, and when the camp grain-supply
runs short women journey to the villages to replenish it. When
camps break up and people return to their villages millet con-
sumption increases, beer being again brewed, and in a good
year there is sufficient to satisfy requirements tiU the new
harvest is ripe. In a normal year Nuer can just tide over these
months if they are economical and have been careful not to use
much grain in camp. Only in the most favoured parts of the
country are they assured of an adequate supply throughout the
year. In most parts there is always a very narrow margin
between sufficiency and want, and in a bad year starvation is
not infrequent. If the crops fail people survive on milk, fish,
and wild fruits, and in extremity may kill some of their beasts.
Rinderpest is considered a worse calamity. When rinderpest and
failure of crops occur in the same year people expect the older
age-sets to be wiped out. Much suffering may be caused by ex-
cessive drought or flood, which injure both crops and grazing.
There are good and bad fishing years. Generally, over the
greater part of the country, consumption of fish is very low,
or entirely absent, at the height of the rains. It rises rapidly to
a peak at the commencement of the dry weather, and after
descending from this point of great abundance remains fairly
constant through the drought, rising again in the early rains.
Meat of domesticated stock is mainly eaten after harvest when
sacrifices and feasts take place. Cattle are seldom slaughtered
in the dry season and, in my experience, Nuer do not kill many
wild animals, so that consumption of meat is very low at this
time of year, though the deficiency is to some extent made up
by bleeding the cows and, in rinderpest years, by eating the
carcasses of its victims. On the whole the curve of meat con-
OECOLOGY 83
sumption follows that of grain consumption. We need not
consider consumption of bush products in a normal year,
though it may be kept in mind that they are very useful in famine
years, chiefly from January to April.
It will be seen from this chart of relative consumption, which
Nov.
Dec.
Jan.
Dry
Feb.
season
Mar. Apr.
May
Web
June
season
July Aug.
Sept. Oct.
^___
N
\
...v....
s
Fish
Milk
>
-^
^^^^^^
^d" meat"
v.^
^^^
X;;--.
ZZZ::^
is only presented as a virtual approximation, that fish largely
takes the place of grain and meat as the principal food from
January to June and that the time when there is most likely
to be a shortage of milk and when few, or no, fish are caught is
that when there is plenty of porridge and beer. 'The hungry
months', as the Nuer call them, are from May to August when
the supply of fish quickly diminishes and maize and millet are
still ripening. The months of plenty are from September to
the middle of December, when there is abundant millet, and
generally much meat, while the end of this period is the best
time for fishing. Nuer say that they put on flesh in the rains and
lose it in the drought. We may conclude, I think, that even in
normal years Nuer do not receive as much nourishment as they
require. Their diet is well-balanced, but there is not enough of
it at some seasons and there is no surplus to fall back on in the
all-too-frequent years of scarcity. Much of Nuer custom and
thought is to be attributed to this insufficiency of food. It is
wistfully related in one of their stories how once upon a time
Man's stomach led an independent life in the bush and lived
on small insects roasted by the firing of the grasses, for 'Man
was not created with a stomach. It was created apart from
him. ' One day Man was walking in the bush and came across
Stomach there and put it in its present place that it might feed
there. Although when it lived by itself it was satisfied with
tiny morsels of food, it is now always hungry. No matter how
much it eats it is soon craving for more.
84 OECOLOGY
Seasonal variation in quantity and kind of food is socio-
logically significant for several reasons, though not all are
relevant to the present study. It is, however, important to
note that abundance of millet is the main reason for holding
ceremonies in the rains, for ritual is seldom complete without
porridge and beer and, since it consists of sacrifice, of meat also.
Weddings, initiation rites, and religious ceremonies of various
kinds take place in the rains and early drought, generally
after the first millet harvest. This is also the main season for
raiding the Dinka. Nuer say that hunger and war are bad com-
panions and that they are too hungry to fight in the full dry
season ; and it is evident that they are then not so eager to come
to blows over personal and community quarrels as they are in
the rainy months, when they are replete with grain and meat
and, especially at wedding dances, sometimes slightly intoxi-
cated. Nor do the young men find dancing so attractive in the
drought, whereas in the rains they dance as much as possible
and think nothing of travelling many miles to attend weddings,
at which they dance from eventide till well into the morning.
The tempo of village life is different from that of camp life.
Owing to flooding at the height of the rains these joint activities
take place mainly at the beginning and end — chiefly at the end
— of the wet season.
Scarcity of food at times and the narrow margin that for
most of the year divides sufficiency from famine cause a high
degree of interdependence among members of the smaller local
groups, which may be said to have a common stock of food.
Jy Although each household owns its own food, does its own
cooking, and provides independently for the needs of its mem-
bers, men, and much less, women and children, eat in one
another's homes to such an extent that, looked at from outside,
the whole community is seen to be partaking of a joint supply.
Rules of hospitality and conventions about the division of meat
and fish lead to a far wider sharing of food than a bare state-
ment of the principles of ownership would suggest. Young men
eat at all byres in the vicinity; every household gives beer
parties which their neighbours and kinsmen attend; the same
people are given food and beer at the co-operative work parties
that assist in any difficult and laborious task; in camps it is
PLATE XI
■A
II. ( >pvn saxaniKih iii the dry season (J.ouj
b. Clearin,<; iiiillct garden for late sowin,^ (Lou)
OECOLOGY 85
considered correct for men to visit the windscreens of their
friends to drink milk, and a special gourd of sour milk is kept
for guests ; when an ox is sacrificed or a wild animal is killed
the meat is always, in one way or another, widely distributed ;
people are expected to give part of their catch of fish to those
who ask them for it ; people assist one another when there is a
shortage of milk or grain ; and so forth. This mutual assistance
and common consumption of food, which is especially evident
in compact dry-season camps, belongs rather to the subject of
domestic and kinship relations than to the present account.
I wish here only to stress the following points./ (i) This habit
of share and share alike is easily understandable in a com-
munity where every one is likely to find himself in difficulties
from time to time, for it is scarcity and not sufficiency that
makes people generous, since everybody is thereby insured
against hunger. He who is in need to-day receives help from
him who may be in like need to-morrow. (2) While the greatest
sharing is in the smaller domestic and kinship groups, there is
also so much mutual assistance and hospitality among members
of villages and camps that one may speak of a common economy
of these communities, which are treated in this book as the
smallest political groups in Nuerland and within which are taken
for granted ties of kinship, affinity, age-sets, and so forth.
X
I have examined Nuer food-supply in relation to their
oecology and I now give a brief account of their material culture
in the same terms. When a Nuer is born he enters not only
into a natural environment but also into a domesticated
environment which is the product of human labour; and this
inner world is constructed from the outer world, its form
and content being strictly limited by natural resources. I
am neither desirous nor capable of describing technological
procedures, and to excuse myself in some degree for this
omission I am including in this and in a second volume an
unusually large number of plates and figures illustrating many
examples of Nuer handiwork. However, some general remarks
on the limiting conditions of production seem pertinent.
86 OECOLOGY
Nuerland lacks the two raw materials that have played so
important a part in the manufacture of primitive tools: iron and
stone. Nuer have always been poor in iron objects. Till recently
they possessed very few iron spears, cherished as heirlooms, but
used instead the straightened horns of antelope and buck, ebony
wood, and the rib-bones of giraffe, all of which are still used to-day,
though almost entirely in dances (Fig. ii). Wooden hoes were
used for gardening, and are still sometimes so used to-day. Iron
bells (Fig. 4) are rare and highly prized even at the present time,
and in the old days iron rings and bracelets were important pieces
of property. Wooden bells and ivory and leather rings and brace-
lets took their place in common use. Nuer have no knowledge of
smelting and little of the blacksmith's art. I have never seen a
forge and, though there are certainly some blacksmiths, their art
is crude and may be regarded as a recent innovation, at any
rate in most parts of Nuerland. Spears bought from Arab mer-
chants are beaten out in the cold.
Nuerland also lacks any kind of hard stone. Indeed, outside
villages I have never seen a stone. They are sometimes brought
from neighbouring areas and employed as hammers, for smoothing
metal ornaments, for rubbing skins, and so on. Grinding grain seems
to have been a recent introduction. The grinder is made from
thomwood and the grindstone from baked marsh mud mixed
with finely pounded potsherds (Fig. 12). With clay, mud, and sand,
Nuer also make pots, grain-bins, pipes, toys, hearthstones, and
firescreens ; construct the walls of byres and huts ; and coat floors
and those parts of homesteads they wish to keep smooth and clean.
Nature, which denies them iron and stone, is niggardly in her
gift of wood. Large trees are rare. Thornwood and low brush-
wood provide timber for building, spear-shafts, harpoon-shafts,
clubs (Frontispiece and Plate VIII), pestles, neck-rests, baskets,
and winnowing trays. Probably no trees in Nuerland are suit-
able for carving and Nuer possess no wooden utensils. Even the
ebony they use for making spears is not found in their country.
Ambatch-trees grow in some parts of the swamp areas and from it
is made an object which serves as a parrying stick, tobacco pouch,
pillow, and seat (Plate XXIX). Byre and kraal fires are fed with
dried dung, and wood fuel is only employed in the kitchen —
grasses and millet stalks sometimes being used as a substitute.
Grasses, millet stalks, and some plants are also used for various
purposes: rafters, windscreens, thatching, cord, basketwork, &c.
Gourds are cultivated for dairy utensils.
Lacking iron and stone Nuer thus make use of vegetable and
earthy materials instead. Animal products are also a valuable
OECOLOGY B>7
source of material, as may be seen in the list of the uses of the
bodies and bodily products of cattle on pp. 28 and 30. The bodies
of wild animals replace those of cattle in some, but not many, of
these uses, e.g. the skins of tiang and cob are used as sleeping-
rugs, waterbuck-skin for the tympana of drums, giraffe-skin for
cord, the scrota of giraffe for bags (Fig. 3), the horns of buffalo for
spoons (Fig. 14) , and the bones and hides of various animals and the
tusks of elephants for armlets, leglets, wristlets, finger- rings, &c.,
and so forth. They can further be employed, though here also to
a limited extent, for purposes for which bovine products are un-
suitable, e.g. hippopotamus- and buffalo-hide for shields and san-
dals, elephant-hide also being used for the second purpose ; horns
and ribs, as mentioned above, for spear-points ; leopard- and genet-
skins for ritual and ceremonial apparel and so forth. Ostrich eggs
and shells of the giant land snail are manufactured into waist-bands
and the second are used for cutting heads of millet in harvesting.
I have indicated a few of the uses to which animal products
are put to give the reader a ^neral idea_of-tlie limitatioRs
imposed on Nuer economy by their environment and of the way
they manage to overcome the natural poverty of their country.
Taken with the earlier list of uses of cattle we can say that the
Nuer do not live in an iron age or even in a stone age, but in an
age, whatever it may be called, in which plants and beasts
furnish technological necessities.
Deficiency of food and other raw materials can be corrected by
trade. However, Nuer seem to have engaged very little in it^^
Many of their iron weapons and ornaments probably came through
Dinka hands from the so-called Jur peoples (Bongo-Mittu group)
and the iron-working sections of Dinka tribes to the west of the
Nile. Much of it was doubtless booty, though some was certainly
traded.' The Eastern Gaajak traded iron for ivory with the Galla of
Ethiopia, but I doubt whether much iron came from this source
before the Abyssinian conquest of the Western Galla towards
the end of the nineteenth century. At the beginning of the
present century there was some trading of ivory from Nuerland
to the Ethiopian markets at Gore and Sayo, and this
went on till very recently. A considerable distance separated
these markets from the main supply in the Zeraf Island,
and people from Eastern Gaajak country in the dry season, at any
rate in some seasons, took cattle, tobacco, and spearheads to the
Zeraf and returned with ivory. A good tusk fetched up to twenty
' Poncet, op. cit., p. 44.
88 OECOLOGY
head of cattle.' This trade was forbidden some years ago and has
probably not a long history, because early writers give no informa-
tion that might lead us to suppose that it went on previous to
the reconquest of the Sudan and the Abyssinian conquest
of Western Ethiopia. There was some ivory-trading on the Zeraf
between Nuer and Arabs from the middle of the nineteenth
century. Metal ornaments and Venetian glass beads seem to have
been offered in exchange. Nuer may have traded some tobacco,
and possibly an occasional canoe, from the Anuak of the Baro
and Gila rivers, but it is more likely that they usually acquired
these objects by raiding. I do not discuss here the very small
trade carried on to-day by Arab shopkeepers who live a hard,
and generally unprofitable, life here and there on the main water-
ways. Nuer obtain from them spears, hoes, fish-hooks, ornaments,
and occasionally an anvil, grindstone, &c., and they purchase
skins of oxen, and sometimes oxen, from the Nuer. Nuer economy
^- has not been much affected thereby. Nuer do not sell their laboufj
We may conclude, therefore, that trade is a very unimportant
social process among the Nuer. Many reasons can be adduced to
explain this fact. I mention only a few. Nuer have nothing to
trade except their cattle and have no inclination to dispose of
these ; all they greatly desire are more cattle, and, apart from the
difficulty that they have nothing to offer in exchange for them,
herds are more easily and pleasantly increased by raiding the
Dinka; they are seldom on such friendly terms with their
neighbours that trade can flourish between them, and, although
there may have been sporadic exchange with Dinka and Anuak,
it is probable that most objects obtained from these peoples were
spoil ; they have no money, no markets, and no transport other than
human porterage; &c. A further reason, and one that requires
emphasis, is the dominant interest of the Nuer in their herds.
This narrow focus of interest causes them to be inattentive to the
products of other people, for which, indeed, they feel no need and
often enough show contempt.
There is very little exchange within Nuerland itself, there being
no high specialization and no diversity in the distribution of raw
material. The only trading I have seen, apart from the exchange
of small commodities and minor services mentioned in the next
section, is the exchange of some cattle, mainly oxen, by the Lou for
Eastern Gaaj ok grain in a famine year. I think it most unlikely thaJL
such exchanges went on before British occuj)ation of .the country-,
though there was, according to Nuer, sometimes an exchange of an
animal for grain in bad years between persons of the same district.
' Bimbashi H. Gordon, Sudan Intelligence Report, no. 107, 1903.
OECOLOGY 89
Nuer have, it will be acknowledged, a low technology which,
taken with their meagre food-supply and scanty trade, may be
supposed to have some effect on their social relations and their
character. Social ties are narrowed, as it were, and the people
of village and camp are drawn closer together, in a moral sense,
for they are in consequence highly interdependent and their
activities tend to be joint undertakings. This is seen best in the
dry season, when the cattle of many families are tethered in a
common kraal and driven as a single herd to the grazing grounds
and daily activities are co-ordinated into a common rhythm of
life.
I risk being accused of speaking idly when I suggest that
a very simple material culture narrows social ties in another
way. Technology from one point of view is an oecological
process: an adaptation of human behaviour to natural cir-
cumstances. From another point of view material culture may
be regarded as part of social relations, for material objects are
chains along which social relationships run, and the more
simple is a material culture the more numerous are the relation-
ships expressed through it. I give, without further explana-
tion, a few examples. The simple family is attached to the hut,
the household to the byre, the joint family to the hamlet, the
village community to its ridge, and village communities are
linked together by paths. Herds of cattle are nuclei around
which kinship groups are clustered and the relationships
between their members operate through cattle and are ex-
pressed in terms of cattle. A single small artifact may be a
nexus between persons, e.g. a spear which passes from father
to son by gift or inheritance is a symbol of their relationship
and one of the bonds by which it is maintained. Thus people not
only create their material culture and attach themselves to it,
but also build up their relationships through it and see them in
terms of it. As Nuer have very few kinds of material objects
and very few specimens of each kind, their social value is in-
creased by their having to serve as the media of many relation-
ships and they are, in consequence, often invested with ritual
functions. Moreover social relationships instead of being dif-
fused along many chains of material links are narrowed by
the meagreness of culture to a few simple foci of interest. This
'<•". f\^
90 OECOLOGY
may be supposed to lead to a small range of relationship-forms
with a high degree of solidarity in the smaller local and kinship
groups and we may expect to find a simple social structure.
Some outstanding traits in Nuer character may be said to be
consistent with their low technology and scanty food-supply.
I again emphasize the crudity and discomfort of their lives.
All who have lived with Nuer would, I believe, agree that
though they are very poor in goods they are very proud in
spirit. Schooled in hardship and hunger — for both they express
contempt — they accept the direst calamities with resignation
and endure them with courage. Content with few goods they
despise all that lies outside them ; their derisive pride amaz£S_a
stranger. Reliant on one another they are loyal and generous to
.their kinsmen. One might even to some extent attribute their
pronounced individualism to resistance to the persistent claims
of kinsmen and neighbours against which they have no protec-
tion but stubbornness. The qualities which have been mentioned,
courage, generosity, patience, pride, loyalty, stubbornness, and
independence, are the virtues the Nuer themselves extol, and
these values can be shown to be very appropriate to their simple
mode of life and to the simple set of social relations it en-
genders, y.
XI
It is unnecessary to write more on what are generally called
economics. They can be further considered in an account of kin-
ship and family life. I will only ask the reader to bear in mind
the following matters, (i) One cannot treat Nuer economic rela-
tions by themselves, for they always form part of direct social
relationships of a general kind. Thus, di:a[si on of labour is part o f
general relationships between persons of different sexes and of
different ages, between spouses, T)etween parents and^xhildren,.
between kinsmen of one order or another, and so forth.. -(2) There
is some specialization, but it is occasional, and there are no occupa-
tions which can be called professions. Some women make better
pots, grindstones, and baskets tTian others ; only smiths can make
certain objects; there are only a few men who understand how
to make and put on the arm the tight-fitting bracelets which
young men wear to show their endurance, and so on ; and people
who want these things either ask for them in the name of kinship or
give the maker of them some millet for his services, or make him
a gift on some future occasion. The man who wants the object
TIME AND SPACE 95
and lunar changes repeat themselves year after year, so that a
Nuer standing at any point of time has conceptual knowledge
of what lies before him and can predict and organize his life
accordingly. A man's structural future is likewise already fixed
and ordered into different periods, so that the total changes in
status a boy will undergo in his ordained passage through the
social system, if he lives long enough, can be foreseen. Struc-
tural time appears to an individual passing through the social
system to be entirely progressive, but, as we shall see, in a sense
this is an illusion. Oecological time appears to be, and is,
cyclical.
The oecological cycle is a year. Its distinctive rhythm is the
backwards and forwards movement from villages to camps,
which is the Nuer's response to the climatic dichotomy of rains
and drought. The year [vMon) has two main seasons, tot and
mai. Tot, from about the middle of March to the middle of
September, roughly corresponds to the rise in the curve of rain-
fall, though it does not cover the whole period of the rains. Rain
may fall heavily at the end of September and in early October,
and the country is still flooded in these months which belong,
nevertheless, to the mai half of the year, for it commences at
the decline of the rains — not at their cessation — and roughly
covers the trough of the curve, from about the middle of Septem-
ber to the middle of March. The two seasons therefore only
approximate to our division into rains and drought, and the
Nuer classification aptly summarizes their way of looking at the
movement of time, the direction of attention in marginal months
being as significant as the actual climatic conditions. In the
middle of September Nuer turn, as it were, towards the life of
fishing and cattle camps and feel that village residence and
horticulture lie behind them. They begin to speak of camps as
though they were already in being, and long to be on the move.
This restlessness is even more marked towards the end of the
drought when, noting cloudy skies, people turn towards the life
of villages and make preparations for striking camp. Marginal
months may therefore be classed as tot or mai, since they belong
to one set of activities but presage the other set, for the concept
of seasons is derived from social activities rather than from the
climatic changes which determine them, and a year is to Nuer
96 TIME AND SPACE
a period of village residence [cieng] and a period of camp
residence {wee).
I have already noted the significant physical changes asso-
ciated with rains and drought, and some of these have been
presented in charts on pp. 52 and 53. I have also described, in
the last chapter, the oecological movement that follows these
physical changes where it affects human life to any degree.
Seasonal variations in social activities, on which Nuer concepts
of time are primarily based, have also been indicated and, on
the economic side, recorded at some length. The main features
of these three planes of rhythm, physical, oecological, and
social, are charted on the opposite page.
The movements of the heavenly bodies other than the sun
and the moon, the direction and variation of winds, and the
migration of some species of birds are observed by the Nuer,
but they do not regulate their activities in relation to them nor
use them as points of reference in seasonal time-reckoning. The
characters by which seasons are most clearly defined are those
which control the movements of the people: water, vegetation,
movements of fish, &c. ; it being the needs of the cattle and
variations in food-supply which chiefly translate oecological
rhythm into the social rhythm of the year, and the contrast
between modes of life at the height of the rains and at the
height of the drought which provides the conceptual poles in
time-reckoning.
""^Besides these two main seasons of tot and mai Nuer recognize
two subsidiary seasons included in them, being transitional
periods between them. The four seasons are not sharp divisions
fbut overlap. Just as we reckon summer and winter as the halves
of our year and speak also of spring and autumn, so Nuer reckon
tot and mai as halves of their year and speak also of the seaons
of rwil and jiom. Rwil is the time of moving from camp to
village and of clearing and planting, from about the middle of
March to the middle of June, before the rains have reached their
peak. It counts as part of the tot half of the year, though it
is contrasted with tot proper, the period of full village life and
horticulture, from about the middle of June to the middle of
September. Jiom, meaning 'wind', is the period in which the
persistent north wind begins to blow and people harvest, fish
I
U
D
O
u
I —
to
D
CO
Lju LU O
CO
z
08 3:
ZZ
<_
I-
u
z
LU
I--"
z-J
DO
CO
u. ^
Q -
z ^
CD ULca
u
CO
U
E
o E
c " >
*-> i_ I/)
So
n ^~
0) c
I- <u
D-T3
M Q.
0) 4>
> =
« N
•H E
>-
I-
z
LU
U
"C O
- 3
_ M Irt "•
a. CL V c
S I 5 «
_ l_ I-
00 (Q V
C 4) 'J
3
O i-
a
o
o
LU
J5 J> S
Do:3 'rt
E E
0« J:
0) c J
I- 4) o
a. -o lo
>-
\-
u
<
U
CO
"■o-c So
o S2 =
o
I- « *-• s
o s « =
98 TIME AND SPACE
from dams, fire the bush, and form early camps, from about the
middle of September to the middle of December. It counts as
part of the mai half of the year, though it is contrasted with
mai proper, from about the middle of December to the middle of
March, when the main camps are formed. Roughly speaking,
therefore, there are two major seasons of six months and four
%
>
T
minor seasons of three months, but these divisions must not be
regarded too rigidly since they are not so much exact units of
time as rather vague conceptualizations of changes in oeco-
logical relations and social activities which pass imperceptibly
from one state to another.
In the diagram above a line drawn from mid March to mid
September is the axis of the year, being an approximation to a
cleavage between two opposed sets of oecological relations and
social activities, though not entirely corresponding to it, as may
be seen in the diagram below, where village life and camp life
are shown in relation to the seasons of which they are the focal
points. Nuer, especially the younger people, are still in camp
for part of tot (the greater part of rwil) and are still in villages,
especially the older people, for part of mai (the greater part of
TIME AND SPACE 99
jiom) , but every one is in villages during tot proper and in camps
during mat proper. Since the words tot and mat are not pure
units of time-reckoning but stand for the cluster of social
activities characteristic of the height of the drought and of the
height of the rains, one may hear a Nuer saying that he is going
to 'tot ' or 'mat ' in a certain place.
JANUARY
1/ / I Part of the year spent in villages
^
Part of the year spent in camps
The year has twelve months, six to each of the major seasons,
and most adult Nuer can state them in order. In the list of
months given below it has not been possible to equate each Nuer
name with an English name, because our Roman months have
nothing to do with the moon. It will be found, however, that a
Nuer month is usually covered by the two English months
equated to it in the list and generally tends to coincide with the
first rather than the second.
teer Sept .-Oct. duong Mar. -Apr.
lath {boor) Oct. -No v. gwaak Apr.-May
kur Nov.-Dec. dwat May- June
tiop {in) dit Dec-Jan. kornyuot June-July
tiop {in) tot Jan.-Feb. paiyatni {paiyene) July-Aug.
pet Feb.-Mar. thoor Aug.-Sept.
100 TIME AND SPACE
Nuer would soon be in difficulties over their lunar calendar if
they consistently counted the succession of moons/ but there
are certain activities associated with each month, the associa-
tion sometimes being indicated by the name of the month.
The calendar is a relation between a cycle of activities and a
conceptual cycle and the two cannot fall apart, since the con-
ceptual cycle is dependent on the cycle of activities from which
it derives its meaning and function. Thus a twelve-month
j system does not incommode Nuer, for the calendar is anchored
lito the cycle of oecological changes. In the month of kur one
makes the first fishing dams and forms the first cattle camps, and
since one is doing these things it must be kur or thereabouts.
Likewise in dwat one breaks camp and returns to the villages,
and since people are on the move it must be dwat or thereabouts.
Consequently the calendar remains fairly stable and in any sec-
tion of Nuerland there is general agreement about the name of
the current month.
In my experience Nuer do not to any great extent use the
names of the months to indicate the time of an event, but
generally refer instead to some outstanding activity in process
at the time of its occurrence, e.g. at the time of early camps, at
the time of weeding, at the time of harvesting, &c., and it is
easily understandable that they do so, since time is to them a
relation between activities. During the rains the stages in the
growth of millet and the steps taken in its culture are often used
as points of reference. Pastoral activities, being largely un-
differentiated throughout the months and seasons, do not pro-
vide suitable points,
r^ There are no units of time between the month and day and
A night. People indicate the occurrence of an event more than a
day or two ago by reference to some other event which took place
at the same time or by counting the number of intervening
'sleeps' or, less commonly, 'suns'. There are terms for to-day,
to-morrow, yesterday, &c., but there is no precision about them.
When Nuer wish to define the occurrence of an event several
days in advance, such as a dance or wedding, they do so by
I There is some evidence of an intercalary month among the Eastern
Jikany, but I cannot be definite on this point, and I have not heard it men-
tioned in other parts of Nuerland.
TIME AND SPACE loi
reference to the phases of the moon: new moon, its waxing,
full moon, its waning, and the brightness of its second quarter.
When they wish to be precise they state on which night of the
waxing or waning an event will take place, reckoning fifteen
nights to each and thirty to the month. They say that only
cattle and the Anuak can see the moon in its invisible period.
The only terms applied to the nightly succession of lunar phases
are those that describe its appearance just before, and in, full-
ness.
The course of the sun determines many points of reference,
and a common way of indicating the time of events is by point-
ing to that part of the heavens the sun will then have reached
in its course. There are also a number of expressions, varying
in the degree of their precision, which describe positions of the
sun in the heavens, though, in my experience, the only ones
commonly employed are those that refer to its more conspicu-
ously differentiated movements: the first stroke of dawn,
sunrise, noon, and sunset. It is, perhaps, significant that there
are almost as many points of reference between 4 and 6 a.m.
as there are for the rest of the day. This may be chiefly due to
striking contrasts caused by changes in relations of earth to sun
during these two hours, but it may be noted, also, that the points
of reference between them are more used in directing activities,
such as starting on journeys, rising from sleep, tethering cattle in
kraals, gazelle hunting, &c., than points of reference during
most of the rest of the day, especially in the slack time between
I and 3 p.m. There are also a number of terms to describe the
time of night. They are to a very limited extent determined by
the course of the stars. Here again, there is a richer terminology
for the transition period between day and night than during the
rest of the night and the same reasons may be suggested to
explain this fact. There are also expressions for distinguishing
night from day, forenoon from afternoon, and that part of the
day which is spent from that part which lies ahead.
Except for the commonest of the terms for divisions of the
day they are little used in comparison with expressions which
describe routine diurnal activities. The daily timepiece is the
cattle clock, the round of pastoral tasks, and the time of day
and the passage of time through a day are to a Nuer primarily
102 TIME AND SPACE
the succession of these tasks and their relations to one another.
The better demarcated points are taking of the cattle from byre
to kraal, milking, driving of the adult herd to pasture, milking
of the goats and sheep, driving of the flocks and calves to pas-
ture, cleaning of byre and kraal, bringing home of the flocks and
calves, the return of the adult herd, the evening milking, and
the enclosure of the beasts in byres. Nuer generally use such
points of activity, rather than concrete points in the movement
of the sun across the heavens, to co-ordinate events. Thus a
man says, 'I shall return at mflking', T shall start off when the
calves come home', and so forth.
Oecological time-reckoning is ultimately, of course, entirely
determined by the movement of the heavenly bodies, but only
some of its units and notations are directly based on these move-
ments, e.g. month, day, night, and some parts of the day and
night, and such points of reference are paid attention to and
selected as points only because they are significant for social
activities. It is the activities themselves, chiefly of an economic
kind, which are basic to the system and furnish most of its
units and notations, and the passage of time is perceived in the
relation of activities to one another. Since activities are depen-
dent on the movement of the heavenly bodies and since the
movement of the heavenly bodies is significant only in relation
to the activities one may often refer to either in indication of
the time of an event. Thus one may say, 'In the 7V0W season '
or 'At early camps', 'The month of Dwat' or 'The return to
villages', 'When the sun is warming up' or 'At milking'. The
movements of the heavenly bodies permit Nuer to select natural
points that are significant in relation to activities. Hence in
linguistic usage nights, or rather 'sleeps', are more clearly
defined units of time than days, or 'suns', because they are
undifferentiated units of social activity, and months, or rather
'moons', though they are clearly differentiated units of natural
time, are little employed as points of reference because they are
not clearly differentiated units of activity, whereas the day,
the year, and its main seasons are complete occupational units.
Certain conclusions may be drawn from this quality of time
among the Nuer. Time has not the same value throughout the
year. Thus in dry season camps, although daily pastoral tasks
c
TIME AND SPACE 103
follow one another in the same order as in the rains, they do not
take place at the same time, are more a precise routine owing
to the severity of seasonal conditions, especially with regard
to water and pasturage, and require greater co-ordination and
co-operative action. On the other hand, life in the dry season
is generally uneventful, outside routine tasks, and oecological
and social relations are more monotonous from month to month
than in the rains when there are frequent feasts, dances, and
ceremonies. When time is considered as relations between acti-
vities it will be understood that it has a different connotation
in rains and drought. In the drought the daily time-reckoning
is more uniform and precise while lunar reckoning receives less
attention, as appears from the lesser use of names of months,
less confidence in stating their order, and the common East
African trait of two dry-season months with the same name
{Hop in dit and Hop in tot), the order of which is often inter-
changed. The pace of time may vary accordingly, since per-
ception of time is a function of systems of time-reckoning, but
we can make no definite statement on this question.
Though I have spoken of time and units of time the Nuer
have no expression equivalent to 'time' in our language, and
they cannot, therefore, as we can, speak of time as though it
were something actual, which passes, can be wasted, can be
saved, and so forth. I do not think that they ever experience
the same feeling of fighting against time or of having to co-
ordinate activities with an abstract passage of time, because
their points of reference are mainly the activities themselves,
which are generally of a leisurely character. Events follow a
logical order, but they are not controlled by an abstract system,
there being no autonomous points of reference to which activi-
ties have to conform with precision. Nuer are fortunate.
Also they have very limited means of reckoning the relative
duration of periods of time intervening between events, since
they have few, and not well-defined or systematized, units of
time. Having no hours or other small units of time they cannot
measure the periods which intervene between positions of the
sun or daily activities. It is true that the year is divided into
twelve lunar units, but Nuer do not reckon in them as fractions
of a unit. They may be able to state in what month an event
104 TIME AND SPACE
occurred, but it is with great difficulty that they reckon the
relation between events in abstract numerical symbols. They
think much more easily in terms of activities and of successions
of activities and in terms of social structure and of structural
differences than in pure units of time.
We may conclude that the Nuer system of time-reckoning
within the annual cycle and parts of the cycle is a series of
conceptualizations of natural changes, and that the selection of
points of reference is determined by the significance which these_
natural changes have for human activities.
II
In a sense all time is structural since it is a conceptualization
of collateral, co-ordinated, or co-operative activities : the move-
ments of a group. Otherwise time concepts of this kind could
not exist, for they must have a like meaning for every one
within a group. Milking-time and meal-times are approximately
the same for all people who normally come into contact with
one another, and the movement from villages to camps has
approximately the same connotation everywhere in Nuerland,
though it may have a special connotation for a particular group
of persons. There is, however, a point at which we can say that
time concepts cease to be determined by oecological factors and
become more determined by structural interrelations, being no
longer a reflection of man's dependence on nature, but a reflection
of the interaction of social groups.
The year is the largest unit of oecological time. Nuer have
words for the year before last, last year, this year, next year, and
the year after next. Events which took place in the last few
years are then the points of reference in time-reckoning, and these
are different according to the group of persons who make use of
them: joint family, village, tribal section, tribe, &c. One of the
commonest ways of stating the year of an event is to mention
where the people of the village made their dry season camps,
or to refer to some evil that befell their cattle, A joint family
may reckon time in the birth of calves of their herds. Weddings
and other ceremonies, fights, and raids, may likewise give points
of time, though in the absence of numerical dating no one can
say without lengthy calculations how many years ago an event
PLATE XIII
Girl in millet garden (Dok)
TIME AND SPACE 105
took place. Moreover, since time is to Nuer an order of events
of outstanding significance to a group, each group has its own
points of reference and time is consequently relative to struc-
tural space, locally considered. This is obvious when we examine
the names given to years by different tribes, or sometimes
by adjacent tribes, for these are floods, pestilences, famines,
wars, &c., experienced by the tribe. In course of time the names
of years are forgotten and all events beyond the limits of this
crude historical reckoning fade into the dim vista of long long
ago. Historical time, in this sense of a sequence of outstanding
events of significance to a tribe, goes back much farther than
the historical time of smaller groups, but fifty years is probably
its limit, and the farther back from the present day the sparser
and vaguer become its points of reference.
However, Nuer have another way of stating roughly when
events took place ; not in numbers of years, but by reference
to the age-set system. Distance between events ceases to be
reckoned in time concepts as we understand them and is
reckoned in terms of structural distance, being the relation
between groups of persons. It is therefore entirely relative to
the social structure. Thus a Nuer may say that an event took
place after the Thut age-set was born or in the initiation period
of the Boiloc age-set, but no one can say how many years ago
it happened. Time is here reckoned in sets. If a man of the
Dangunga set tells one that an event occurred in the initiation
period of the Thut set he is saying that it happened three sets
before his set, or six sets ago. The age-set system is discussed
in Chapter VI. Here it need only be said that we cannot
accurately translate a reckoning in sets into a reckoning in years,
but that we can roughly estimate a ten-year interval between
the commencement of successive sets. There are six sets in
existence, the names of the sets are not cyclic, and the order of
extinct sets, all but the last, are soon forgotten, so that an age-
set reckoning has seven units covering a period of rather under
a century.
The structural system of time-reckoning is partly the selection
of points of reference of significance to local groups which give
these groups a common and distinctive history ; partly the dis-
tance between specific sets in the age-set system; and partly
lo6 TIME AND SPACE
distances of a kinship and lineage order. Four generation-
Tsteps {kath) in the kinship system are Hnguistically differentiated
'relations, grandfather, father, son, and grandson, and within a
jsmall kinship group these relationships give a time-depth to
members of the group and points of reference in a line of ascent
by which their relationships are determined and explained.
Any kinship relationship must have a point of reference on a
line of ascent, namely a common ancestor, so that such a rela-
tionship always has a time connotation couched in structural
A
/
/ / / \ \. \
/ / / A \ \ \
./ / / A \ \
/ ^ / / \ \ \ \
/ / / / \ \ \ \.
A B
terms. Beyond the range of the kinship system in this narrow
sense the connotation is expressed in terms of the lineage
system. As this subject is treated in Chapter V, we limit
further discussion of it to an explanatory comment on the
diagram above. The base line of the triangle represents a given
group of agnates and the dotted lines represent their ghostly
agnatic forebears, running from this base to a point in lineage
structure, the common ancestor of every member of the group.
The farther we extend the range of the group (the longer
becomes the base line) the farther back in lineage structure is
the common ancestor (the farther from the base line is the apex
of the triangle) . The four triangles are thus the time depths of
four extensions of agnatic relationship on an existential plane
and represent minimal, minor, major, and maximal lineages of
a clan. Lineage time is thus the structural distance between
groups of persons on the line AB. Structural time therefore
TIME AND SPACE 107
cannot be understood until structural distance is known, since
it is a reflection of it, and we must, therefore, ask the reader to
forgive a certain obscurity at this point and to reserve criticism
till we have had an opportunity of explaining more clearly what
is meant by structural distance.
We have restricted our discussion to Nuer systems of time-
reckoning and have not considered the way in which an individual
perceives time. The subject bristles with difficulties. Thus an
individual may reckon the passage of time by reference to the
physical appearance and status of other individuals and to
changes in his own life-history, but such a method of reckoning
time has no wide collective validity. We confess, however, that
our observations on the matter have been slight and that a
fuller analysis is beyond our powers. We have merely indicated
those aspects of the problem which are directly related to the
description of modes of livelihood which has gone before and to
the description of political institutions which follows.
We have remarked that the movement of structural time is,
in a sense, an illusion, for the structure remains fairly constant
and the perception of time is no more than the movement of
persons, often as groups, through the structure. Thus age-sets
succeed one another for ever, but there are never more than six
in existence and the relative positions occupied by these six sets
at any time are fixed structural points through which actual
sets of persons pass in endless succession. Similarly, for reasons
which we explain later, the Nuer system of lineages may be con-
sidered a fixed system, there being a constant number of steps
between living persons and the founder of their clan and the
lineages having a constant position relative to one another. How-
ever many generations succeed one another the depth and range of
lineages does not increase unless there has been structural change.
These statements are discussed more fully on pp. 198-200.
Beyond the limits of historical time we enter a plane of
tradition in which a certain element of historical fact may be
supposed to be incorporated in a complex of myth. Here the
points of reference are the structural ones we have indicated. At
one end this plane merges into history; at the other end into
myth. Time perspective is here not a true impression of actual
distances like that created by our dating technique, but a
io8 TIME AND SPACE
reflection of relations between lineages, so that the traditional
[>Jevents recorded have to be placed at the points where the
lineages concerned in them converge in their lines of ascent.
PThe events have therefore a position in structure, but no exact
j position in historical time as we understand it. Beyond tradi-
tion lies the horizon of pure myth which is always seen in the
I same time perspective. One mythological event did not pre-
cede another, for myths explain customs of general social signifi-
cance rather than the interrelations of particular segments and
are, therefore, not structurally stratified. Explanations of any
qualities of nature or of culture are drawn from this intellectual
ambient which imposes limits on the Nuer world and makes it
self-contained and entirely intelligible to Nuer in the relation of
its parts. The world, peoples, and cultures all existed together
from the same remote past.
It will have been noted that the Nuer time dimension is
narrow. Valid history ends a century ago, and tradition,
generously measured, takes us back only ten to twelve genera-
tions in lineage structure, and if we are right in supposing that
lineage structure never grows, it follows that the distance be-
tween the beginning of the world and the present day remains
unalterable. Time is thus not a continuum, but is a constant
structural relationship between two points, the first and last
persons in a line of agnatic descent. How shallow is Nuer time
may be judged from the fact that the tree under which mankind
came into being was still standing in Western Nuerland a few
years ago !
I ^ Beyond the annual cycle, time-reckoning is a conceptualiza-
tion of the social structure, and the points of reference are a
' projection into the past of actual relations between groups of
persons. It is less a means of co-ordinating events than of
co-ordinating relationships, and is therefore mainly a looking-
backwards, since relationships must be explained in terms of
the past.
Ill
We have concluded that structural time is a reflection of
structural distance. In the following sections we define further
what we mean by structural distance, and make a formal.
TIME AND SPACE 109
preliminary, classification of Nuer territorial groups of a poli-j|
tical kind. We have classified Nuer socio-temporal categories. |
We now classify their socio-spatial categories.
Were a man to fly over Nuerland he would see, as on Plate
XVI, taken by the Royal Air Force in the dry season, white
patches with what look like tiny fungoid growths on them.
These are village sites with huts and byres. He would see that
between such patches are stretches of brown and black, the
brown being open grassland and the black being depressions
which are swampy in the rains ; and that the white patches are
wider and more frequent in some parts than in others. We find
Nuer give to these distributions certain values which compose
their political structure.
It would be possible to measure the exact distance between
hut and hut, village and village, tribal area and tribal area, and
so forth, and the space covered by each. This would give us a
statement of spatial measurements in bare physical terms. By
itself it would have very limited significance. Oecological space
is more than mere physical distance, though it is affected by it,
for it is reckoned also by the character of the country inter-
vening between local groups and its relation to the biological
requirements of their members. A broad river divides two
Nuer tribes more sharply than many miles of unoccupied bush.
A distance which appears small in the dry season has a different
appearance when the area it covers is flooded in the rains. A
village community which has permanent water near at hand is
in a very different position to one which has to travel in the dry
season to obtain water, pasturage, and fishing. A tsetse belt
creates an impassable barrier, giving wide oecological distance
between the peoples it separates (p. 133), and presence or
absence of cattle among neighbours of the Nuer likewise
determines the oecological distance between them and the
Nuer (pp. 132-3). Oecological distance, in this sense, is a re-
lation between communities defined in terms of density and
distribution, and with reference to water, vegetation, animal and
insect life, and so on.
Structural distance is of a very different order, though it
is always influenced and, in its political dimension, to a large
extent determined by oecological conditions. By structural
no TIME AND SPACE
j distance is meant, as we have already indicated in the pre-
ceding section, the distance between groups of persons in a
'social system, expressed in terms of values. The nature of the
country determines the distribution of villages and, therefore,
the distance between them, but values limit and define the
distribution in structural terms and give a different set of
distances. A Nuer village may be equidistant from two other
villages, but if one of these belongs to a different tribe and the
other to the same tribe it may be said to be structurally more
distant from the first than from the second. A Nuer tribe which
is separated by forty miles from another Nuer tribe is structurally
nearer to it than to a Dinka tribe from which it is separated by
only twenty miles. When we leave territorial values and speak
of lineages and age-sets, structural space is less determined by
environmental conditions. One lineage is closer to another than
to a third. One age-set is closer to another than to a third.
The values attached to residence, kinship, lineage, sex, and
age, differentiate groups of persons by segmentation, and the
relative positions of the segments to one another gives a per-
spective that enables us to speak of the divisions between them
as divisions of structural space. Having defined what is meant
by structural space we may now proceed with a description of
its political divisions.
IV
We cannot, owing to lack of adequate population statistics
(see p. 117) and survey records, present a map showing the
density of the different tribes, but a rough estimate is possible
for the whole of Nuerland. Jackson says that the area to the
east of the Nile amounts to some 26,000 square miles, ^ and
recent censuses put its population at about 144,000, making
about 5-5 to the square mile. The area to the west of the Nile
is no less sparsely inhabited and possibly has a lower density.
The total area of Nuerland is probably about 30,000 square
miles and the total population round about 200,000. We may
estimate that tribal density probably varies from about 4 to
10 to the square mile and that the average distribution for the
whole of Nuerland is from about 5 to 6 to the square mile. Given
the hydrological conditions of Nuerland and the present economy
' Jackson, op. cit., p. 62.
TIME AND SPACE iii
of the people it may be doubted whether it could support a much
larger population than it does. This is particularly the case to
the west of the Nile and it is likely, as Nuer themselves suggest,
that their expansion eastwards was due to over-population. It
is possible for there to be great local concentration in spite of
low density for tribal areas, for the estimates of square mileage
include vast stretches of land devoid of villages and camps,
which is grazed over in the drought or merely traversed in
seasonal movements. The degree of actual concentration, in
this sense, varies from tribe to tribe and from tribal section to
tribal section and from season to season.
I cannot figure such distributions more accurately than the
sketch-maps on pp. 57, 59, and 61, and I can only indicate them
verbally in the most general terms. As we have seen, the size of
a village depends on the space available for building, grazing, and
horticulture, and its homesteads are crowded or strung out ac-
cordingly, forming in most villages small clusters of huts and
byres which we call hamlets, each being separated from its
neighbours by gardens and unoccupied land on which calves,
sheep, and goats graze. The population of a village — we can
make no precise statement — may be anything from fifty to
several hundred souls and may cover from a few hundred yards
to several miles. A village is usually well demarcated by the
contiguity of its residences and the stretches of bush, forest, and
swamp which separate it from its neighbours. In my small
experience one may, in most parts of Nuerland, traverse five to
twenty miles from village to village. This is certainly the case
in Western Nuerland. On the other hand, where the nature of
the ground allows, villages may be much closer together and
foUow one another at short intervals over wide areas. Thus the
greater part of the Lou are concentrated within thirty miles of
Muot tot, the greater part of the Dok within ten miles of Ler,
while the Lak, Thiang, and part of the Gaawar run fairly con-
tinuously along a wide ridge between the Nile and the Zeraf.
Villages are always joined to their neighbours by paths created
and maintained by their social interrelations. In every part of
Nuerland there are also large areas, inundated in the rains, with
no, or very few, villages. Those parts of the tribal areas left
blank or shaded to show dry-season occupation in the sketch-
112 TIME AND SPACE
maps are mainly, and often entirely, without village sites,
and in Western Nuerland the whole area between the Nile and
the Bahr el Ghazal is very sparsely dotted with small villages ;
and the same is probably true of the country to the north of the
Bahr el Ghazal.
I am compelled to describe the distribution of dry season
camps as vaguely as the distribution of villages. Early camps
may be found almost anywhere and often comprise only a few
households; but the location of larger camps, formed later in
the season, can be known in advance because there are only a
few places which provide adequate water. Their size depends
largely on the amount of water and grazing and their population
varies from about a hundred to over a thousand souls. These
concentrations are never tribal but comprise larger or smaller
tribal sections. Round a lake a camp may be distributed in
several sections with a few hundred yards between each ; or one
may then speak of contiguous camps. In any camp a few wind-
screens are generally adjacent to, almost touching, one another,
and such a group may often at a glance be seen to be a distinct
unit with its own section of the common kraal. Along the left
bank of the Sobat and the right bank of the Baro one may
observe from a steamer camps almost anywhere, separated from
one another by only a few miles ; but up streams, like the Nyan-
ding and Filus, in which only isolated pools of water remain,
several miles separate camps. Some large camps in the interior of
Lou country are separated by more than twenty miles of bush.
Great rivers flow through Nuerland and it is often these
natural boundaries which indicate the lines of political cleavage.
The Sobat separates the Gaajok tribe from the Lou tribe; the
Pibor separates the Lou tribe from the Anuak people ; the Zeraf
separates the Thiang and Lak from the Dinka; the Ghazal
separates the Karlual primary section of the Leek tribe from
its other two primary sections ; and so forth. Marshes and areas
inundated in the rains likewise separate political groups. The
Macar swamps divide the Eastern Gaajak from the Gaajok and
Gaagwang; water-covered stretches divide, in the rains, the
Rengyan from the Wot, Bor, &c. ; and so on.
This account of distribution is unavoidably vague, but the
main conditions and their significance can easily be summed up.
TIME AND SPACE
village is not an unsegmented unit but is a re-
lation between a number of smaller units.
The village is a very distinct unit. It is some-
times referred to as thur, a ridge of high ground,
but generally as cieng, a word which may be
translated 'home', but which has such a variety
of meanings that we shaU devote a special section
to it. A village comprises a community, linked
by common residence and by a network of kin-
ship and affinal ties, the members of which, as
we have seen, form a common camp, co-operate
in many activities, and eat in one another's
byres and windscreens. A village is the smallest
Nuer group which is not specifically of a kinship
order and is the political unit of Nuerland.
The people of a village have a feeling of strong
soUdarity against other villages and great affec-
tion for their site, and in spite of the wandering
habits of Nuer, persons born and bred in a village
have a nostalgia for it and are likely to return
to it and make their home there, even if they
have resided elsewhere for many years. Members
of a village fight side by side and support each
other in feuds. When the youths of a village go
to dances they enter the dance in a war line {dep)
singing their special war chant.
A cattle camp, which people of a village form
in the drought and in which members of neigh-
bouring villages participate, is known as wee.
While this word has the meaning of 'camp' in
contrast with eieng, 'village', both words are
used in the same general sense of local commun-
ity. Thus when it is said of a certain clan that
they have no wee we are to understand that
they nowhere in a tribal section or village form
a dominant nucleus of the community and that,
therefore, no local community takes its name
from them. A large camp is called after the
dominant lineage in it or after the village community who
Fig. II. Horn
and ebony
spears.
Ii6 TIME AND SPACE
occupy it, and small camps are sometimes named after an
old person of importance who has erected his windscreen there.
We have seen that the social composition of a camp varies at
different times of the drought from the people of a hamlet to the
people of a village, or of neighbouring villages, and that men
sometimes camp with kinsmen living, in camps other than those
of their own villages. Consequently, while local communities of
the rains tend to be also local communities in the drought their
composition may be somewhat different. We again emphasize
that not only are the people of a camp living in a more compact
group than the people of a village, but also that in camp life
there is more frequent contact between its members and greater
co-ordination of their activities. The cattle are herded together,
milked at the same time, and so on. In a village each house-
hold herds its own cattle, if they are herded at aU, and performs
its domestic and kraal tasks independently and at different
times. In the drought there is increasing concentration and
greater uniformity in response to the greater severity of the
season.
We sometimes speak of a district to describe an aggregate of
villages or camps which have easy and frequent intercommunica-
tion. The people of these villages take part in the same dances,
intermarry, conduct feuds, go on joint raiding parties, share dry
season camps or make camps in the same locality, and so on.
This indefinite aggregate of contacts does not constitute a Nuer
category or a political group, because the people do not see
themselves, nor are seen by others, as a unique community, but
'district' is a term we employ to denote the sphere of a man's
social contacts or of the social contacts of the people of a village
and is, therefore, relative to the person or community spoken
about. A district in this sense tends to correspond to a tertiary
or a secondary tribal segment, according to the size of the tribe.
\ in the smallest tribes a whole tribe is a man's district, and a dis-
I trict may even cut across tribal boundaries in that in a large tribe
a border village may have more contacts with neighbouring vil-
lages of another tribe than with distant villages of its own tribe.
The sphere of a man's social contacts may thus not entirely
coincide with any structural division.
A number of adjacent villages, varying in number and total
TIME AND SPACE 117
extension according to the size of the tribe, are grouped into
small tribal sections and these into larger ones. In the larger
tribes it is convenient to distinguish between primary, secon-
dary, and tertiary tribal sections. These sections, of whatever
size, are, like a village, spoken of as 'cieng'. Since the next
chapter is devoted to these tribal segments no more is said
about them here.
VI
The main Nuer tribes are shown on p. 8. The name 'Jagei'
inserted to the west of the Nile includes a number of small
tribes — Lang, Bor, Rengyan, and Wot. There are also some
small tribes — if they are rightly regarded as tribes, for little
research was done in the area — in the vicinity of the Dok Nuer:
Beegh, Jaalogh, (Gaan)Kwac, and Rol. A crude census, com-
piled from various Government sources, gives, in round figures,
the more recent estimates for the larger tribes as follows :
Sobat Nuer: Gaajak, 42,000 ; Gaagwang, 7,000 ; Gaajok, 42,000;
Lou, 33,000. Zeraf Nuer: Lak, 24,000; Thiang, 9,000; Gaawar,
20,000. Western Nuer: Bui, 17,000; Leek, 11,000; the three
Western Jikany tribes, 11,000; the various Jagei tribes, 10,000;
Dok, 12,000; Nuong, 9,000. These figures are probably more
accurate for the Eastern than for the Western Nuer. Estimates
show wide discrepancies and there has been much guesswork. On
the basis of those recorded, the Sobat Nuer are 91,000, the Zeraf
Nuer 53,000, and the Western Nuer 70,000, making a total of
214,000 for the whole of Nuerland. The population of only a few
tribal segments is known. Among the Lou the Gun primary sec-
tion number about 22,000 and the Mor primary section about
12,000; among the Gaawar the Radh primary section number
about 10,000 and the Bar primary section about 10,000 ; and
among the Lak tribe the Kwacbur primary section number about
12,000 and the Jenyang primary section number about 12,000.
It will be remarked that the tribes of Western Nuerland are
generally smaller than those on the Zeraf, and those on the Zeraf
smaller than those on the Sobat. Tribes tend to be larger the far-
ther eastwards one proceeds. Their territories also tend to be more
extensive. It may be suggested that the larger population of the
Eastern Nuer tribes is due to the integrating effects of conquest and
settlement and to the absorption of large numbers of Dinka which
resulted therefrom, but we do not think that such explanations
Ii8 TIME AND SPACE
account for their maintaining a semblance of tribal unity over
such large areas in the absence of any central government. It is
evident that the size of tribal populations is directly related to the
amount of high land available for wet season occupation, and also
to its disposition, for tribes like the Lou and Eastern Gaajok and
Gaa j ak can have a concentration of homesteads and villages on wide
stretches of elevated land, which is not possible for the smaller tribes
of Western Nuerland, whose only building sites are small and sparsely
distributed ridges. But we hold that this fact in itself would not
determine the lines of political cleavage, which can only be under-
stood by taking into consideration also the relation between village
sites and dry season water-supplies, pasturage, and fishing.'' We
have noted earlier hoW tribal sections move from their villages to
dry season pastures, each having spatial distinction in the rains
which is maintained in the drought, but whereas in Western Nuer-
land there is always plenty of water, grazing, and fishing, generally
not far from the villages, and it is possible for wet season village
communities, isolated by inundated tracts in the rains, to main-
tain their isolation and independence in dry season camps, in the
larger tribes of Eastern Nuerland, such as the Lou, drier
conditions compel greater concentration and wider seasonal
movements, with the result that village communities not only
have a greater spatial, and we may say also moral, density in
the drought than in the rains, but have to mix with one another
and share water and pastures and fishing. Distinct villages are
found side by side around a pool. Moreover, people of one section
have to cross the territories of other sections to reach their camps,
which may be situated near the villages of yet another section.
Families and joint families often camp with kinsmen and affines
who belong to other villages than their own, and it is a common
practice to keep cattle in two or more parts of the country to avoid
total loss from rinderpest, which is a dry season pestilence. It is
understandable, therefore, that local communities which, though
they are isolated in the rains, are forced in the drought into rela-
tions that necessitate some sense of community and the admission
of certain common interests and obligations, should be contained
in a common tribal structure. The severer the dry season condi-
tions the greater the need for some measure of contact and there-
fore of forbearance and recognition of interdependence./The Zeraf
tribes move more than the Sobat tribes and less than the Western
Nuer tribes, the Gaawar moving more than the Thiang and Lak.
We may again point out that on the whole where there is plenty
of elevated country which permits concentration in the rains there
is likewise the greater need for large concentrations in the dry
TIME AND SPACE 119
season, since water, fishing, and pasture are found away from
these elevated areas.
These facts seem to explain to some extent the political prepon-
derance of pastoral peoples in East Africa. There may be wide
dispersal of communities and low density of population, but there
is seasonal contraction and wide interdependence. The variation
in their circuits of transhumance also helps us to understand
the variation in the sizeof Nuer tribes. It may be noted that, though
the size and cohesion of tribes vary in different parts of Nuer-
land, nowhere do environmental conditions permit complete auto-
nomy and exclusiveness of small village groups, such as we find
among the Anuak, or such high density of population and such
developed political institutions such as we find among the Shilluk.
Thus, on the one hand, environmental conditions and pastoral
pursuits cause modes of distribution and concentration that pro-
vide the lines of political cleavage and are antagonistic to political
cohesion and development; but, on the other hand, they necessi-
tate extensive tribal areas within which there is a sense of com-
munity and a preparedness to co-operate.
Each tribe has a name which refers alike to its members and
to their country {rol), e.g. Leek, Gaawar, Lou, Lak, &c. (see
map on p. 8). Each has its particular territory and owns and
defends its own building sites, grazing, water-supplies, and
fishing-pools. Not only do large rivers or wide stretches of
uninhabited country generally divide adjacent sections of con-
tiguous tribes, but these sections tend to move in different
directions in the drought. Conditions are doubtless changing in
this respect, but we may cite as examples how the Gaawar tribe
tend to move eastwards towards the Zeraf and not to come into
contact with the Gun primary section of the Lou tribe, who
cluster around their inland lakes or move to the Sobat and Pibor ;
how the Mor tribal section of the Lou move to the Nyanding and
Upper Pibor in the direction of the Gaajok, who do not coalesce
with them but move to the upper reaches of the Sobat and the
lower reaches of the Pibor ; and how the Western Jikany move
towards the marshes of the Nile, while the Leek move north-
west to the junction of the Bahr el Ghazal with its streams and
lagoons.
Tribesmen have a common sentiment towards their country"^
and hence towards their fellow tribesmen. This sentiment is-J
evident in the pride with which they speak of their tribe as the
120 TIME AND SPACE
object of their allegiance, their joking disparagement of other
tribes, and their indication of cultural variations in their own
tribe as symbols of its singularity. A man of one tribe sees the
people of another tribe as an undifferentiated group to whom
he has an undifferentiated pattern of behaviour, while he sees
himself as a member of a segment of his own tribe. Thus when
a Leek man says that So-and-so is a Nac (Rengyan) he at once
defines his relationship to him. Tribal sentiment rests as much
on opposition to other tribes as on common name, common
territory, corporate action in warfare, and the common lineage
structure of a dominant clan.
How strong is tribal sentiment may be gathered from the fact
that sometimes men who intend to leave the tribe of their birth
to settle permanently in another tribe take with them some
earth of their old country and drink it in a solution of water,
slowly adding to each dose a greater amount of soil from their
new country, thus gently breaking mystical ties with the old and
building up mystical ties with the new. I was told that were a
man to fail to do this he might die of nueer, the sanction that
punishes breach of certain ritual obligations.
A tribe is the largest group the members of which consider it
their duty to combine for raiding and for defensive action. The
younger men of the tribe went, till recently, on joint expedi-
tions against the Dinka and waged war against other Nuer
tribes. Wars between tribes were less frequent than attacks on
the Dinka, but there are many examples in recent Nuer history
of border disputes among tribes and even of one tribe raiding
another for cattle, and such fighting is traditional among the
Nuer. The Leek tribe raided the Jikany and Jagei tribes, and a
Leek tribesman told me, 'The cattle with which my father mar-
ried were Gee (Jagei) cattle.' Poncet remarks: 'Les Elliab (Dok)
se battent avec les Egnan (Nuong) du sud et les Reian (Reng-
yan) du nord ; les Ror de I'interieur avec ces derniers et les Bior
(Bor) de Gazal (Ghazal river). Toutes leurs querelles viennent
des paturages qu'ils se disputent, ce qui n'empeche pas qu'ils
voyagent les uns chez les autres sans aucun danger, a moins
cependant qu'on n'ait un parent a venger.'i In theory a tribe was
regarded as a military unit, and if two sections of different tribes
' Op. cit., p. 39.
PLATE XIV
m0%Mf-.
August shower (Lou)
TIME AND SPACE 121
were engaged in hostilities each could rely on the support of the
other sections of the same tribe ; but, in practice, they would
often only join in if the other side was receiving assistance from
neighbouring sections. When a tribe united for warfare there
was a truce to disputes within its borders.
Tribes, especially the smaller ones, often united to raid
foreigners. The Leek united with the Jagei and Western Jikany,
and the Lou with the Gaawar to attack the Dinka ; the Lou with
the Eastern Jikany tribes to attack the Anuak; and so on.
These military alliances between tribes, often under the aegis of
a Sky-god, speaking through his prophet (p. 188), were of short
duration, there was no moral obligation to form them, and,
though action was concerted, each tribe fought separately under
its own leaders and lived in separate camps in enemy country.
Fighting between Nuer of different tribes was of a different
character from fighting between Nuer and Dinka. Inter-tribal
fighting was considered fiercer and more perilous, but it was
subject to certain conventions: women and children were not
molested, huts and byres were not destroyed, and captives were
not taken. Also, other Nuer were not considered natural prey, as
were the Dinka.
Another defining characteristic of a tribe is that within it
there is cut, blood-wealth paid in compensation for homicide,
and Nuer explain the tribal value in terms of it. Thus Lou
tribesmen say that among themselves there is blood-wealth, but
not between themselves and the Gaajok or the Gaawar; and
this is the invariable definition of tribal allegiance in every part
of Nuerland. Between tribesmen there is also nwk, compensa-
tion for torts other than homicide, though the obligation to pay
it is less generally stressed or carried out, while between one
tribe and another no such obligation is acknowledged. We may
therefore say that there is law, in the limited and relative sense
defined in Chapter IV, between tribesmen, but no law between
tribes. If a man commits an offence against a fellow tribesman
he places himself and his kin in a legal position towards this
man and his kin, and the hostile relations that ensue can be
broken down by payment of cattle. If a man commits the same
act against a man of another tribe no breach of law is recognized,
no obligation is felt to settle the dispute, and there is no
122 TIME AND SPACE
machinery to conclude it. Local communities have been classed
as tribes or tribal segments by whether they acknowledge the
obligation to pay blood-wealth or not. Thus the Gun and the
Mor are classed as primary segments of the Lou tribe, while
the Eastern Gaajok, Gaajak, and Gaagwang have been classed
as three tribes and not as three primary segments of a single
Jikany tribe.
I ^It may have happened that border cases between different
tribes were sometimes settled by payment of compensation, but
I have no record of such settlements other than the doubtful
statement given on p. 189, and were they to have taken place it
would in no way invalidate our definition of tribal structure.
However, it must be understood that we are defining a tribe in
the most formal way and that, as we shall show later, the
acknowledgement of legal responsibility within a tribe does not
mean that, in fact, it is easy to obtain compensation for a tort.
There is little solidarity within a tribe and feuds are frequent
and of long duration. Indeed, the feud is a characteristic institu-
tion of tribal organization.
^^h tribe has been defined by (i) a common and distinct name ;
(2) a common sentiment ; (3) a common and distinct territory ;
(4) a moral obligation to unite in war ; and (5) a moral obligation
to settle feuds and other disputes by arbitration. To these five
points can be added three further characteristics, which are
discussed later: (6) a tribe is a segmented structure and there
is opposition between its segments ; (7) within each tribe there
is a dominant clan and the relation between the lineage struc-
ture of this clan and the territorial system of the tribe is of great
structural importance ; (8) a tribe is a unit in a system of tribes ;
and (9) age-sets are organized tribally.
VII
Adjacent tribes are opposed to one another and fight one an-
other. They sometimes combine against Dinka, but such com-
binations are loose and temporary federations for a specific end
and do not correspond to any clear political value. Occasionally
a tribe will allow a section of another to camp in its territory
and there may be more contacts between persons of border
villages or camps of different tribes than between widely sepa-
TIME AND SPACE 123
rated communities of the same tribe. The first may have more
social contacts ; the second be structurally nearer. But between
Nuer tribes there is no common organization or central ad-
ministration and hence no political unity that we can refer to
as national. Nevertheless, adjacent tribes, and the Dinka who
face them, form political systems, since the internal organiza-
tion of each tribe can only be fully understood in terms of their
mutual opposition, and their common opposition to the Dinka
who border them.
Beyond these systems of direct political relations the whole
Nuer people see themselves as a unique community and their
culture as a unique culture. Opposition to their neighbours
gives them a consciousness of kind and a strong sentiment of
exclusiveness. A Nuer is known as such by his culture, which
is very homogeneous, especially by his language, by the absence
of his lower incisors, and, if he is a man, by six cuts on his brow.
All Nuer live in a continuous stretch of country. There are no
isolated sections. However, their feeling of community goes
deeper than recognition of cultural identity. Between Nuer,
wherever they hail from, and though they be strangers to one
another, friendly relations are at once estabhshed when they
meet outside their country, for a Nuer is never a foreigner to
another as he is to a Dinka or a Shilluk. Their feeling of
superiority and the contempt they show for all foreigners and
their readiness to fight them are a common bond of communion,
and their common tongue and values permit ready intercom-
munication,
Nuer are well aware of the different divisions of their country,
even if they have never visited them, and they all look upon the
area to the west of the Nile as their common homeland, to the
occupants of which they still have distant kinship ties. People
also journey to visit kinsmen in other tribes and often settle for
long periods far from their homes, sometimes in different tribes,
in which, if they stay there long enough, they become perman-
ently incorporated. Constant social intercourse flows across the
borders of adjacent tribes and unites their members, especially
members of border communities, by many strands of kinship
and affinity. If a man changes his tribe he can at once fit himself
into the age-set system of the tribe of his adoption, and there is
124 TIME AND SPACE
often co-ordination of sets between adjacent tribes. A single
clan is sometimes dominant in more than one tribal area, domi-
nant clans are linked in a general clan system, and the principal
clans are found in every part of Nuerland. We have noted
how in the days of ivory trading Gaajak tribesmen journeyed
through the territories of other tribes as far as the Zeraf .
The limits of the tribe are therefore not the limits of social
intercourse, and there are many ties between persons of one
tribe and persons of another tribe. Through association with
the clan system and by proximity the people of one tribe may
consider themselves nearer to a second than to a third. Thus the
three Eastern Jikany tribes feel a vague unity in relation to the
Lou, and so do the Bor and Rengyan in relation to the Leek.
But, also, individuals, and through individuals kinship groups
and even a village, have a circuit of social relationships that
cut across tribal divisions, so that a traveller who crosses the
border of his tribe can always establish some links with in-
dividuals of the tribe he visits in virtue of which he will receive
hospitality and protection. If he is wronged, his host, and not he,
is involved in legal action. However, there is a kind of inter-
national law, in the recognition of conventions in certain matters,
beyond political boundaries and the limits of formal law. Thus,
though it is considered more risky to marry outside the tribe
than within it, since divorce may prove more detrimental in
that the return of bride-wealth is less certain, the rules of
marriage are acknowledged on both sides and it would be con-
sidered improper to take advantage of political cleavage to break
them. Tribes are thus politically exclusive groups, but they are
not coterminous with an individual's sphere of social relations,
though this sphere tends to follow the lines of political cleavage,
in the same way as a man's district tends to be equated to his
tribal segment. The relation between political structure and
general social relations is discussed in the chapters that follow.
Here we may note that it is desirable to distinguish between
(i) political distance in the sense of structural distance between
segments of a tribe, the largest political unit, and between tribes
in a system of political relations ; (2) general structural distance
in the sense of non-political distance between various social
groups in the Nuer-speaking community — non-political struc-
TIME AND SPACE 125
tural relations are strongest between adjacent tribes, but a
common social structure embraces the whole of Nuerland ; and
(3) the social sphere of an individual, being his circuit of social
contacts of one kind or another with other Nuer.
VIII
The political structure of the Nuer c an only be understood
in relat ion to their neighbours, with whom they form a single
political system. Contiguous Dinka and Nuer tribes are seg-
ments within a common structure as much as are segments of
the same Nuer tribe. Their social relationship is one of hostility
and its expression is in warfare.
The Dinka people are the immemorial enemies of the Nuer.
They are alike in their oecologies, cultures, and social systems,
so that individuals belonging to the one people are easily assi-
milated to the other ; and when the balanced opposition between
a Nuer political segment and a Dinka political segment changes
into a relationship in which the Nuer segment becomes entirely
dominant, fusion and not a class structure results.
As far as history and tradition go back, and in the vistas of
myth beyond their farthest reach, there has been enmity be-
tween the two peoples. Almost always the Nuer have been the
aggressors, and raiding of the Dinka is conceived by them to be
a normal state of affairs and a duty, for they have a myth, like
that of Esau and Jacob, which explains it and justifies it. Nuer
and Dinka are represented in this myth as two sons of God
who promised his old cow to Dinka and its young calf to Nuer.
Dinka came by night to God's byre and, imitating the voice of
Nuer, obtained the calf. When God found that he had been
tricked he was angry and charged Nuer to avenge the injury
by raiding Dinka's cattle to the end of time. This story,
familiar to every Nuer, is not only a reflection of the political
relations between the two peoples but is also a commentary on
their characters. Nuer raid for cattle and seize them openly and
by force of arms. Dinka steal them or take them by treachery.
All Nuer regard them — and rightly so — as thieves, and even the
Dinka seem to admit the reproach, if we attribute correct signi-
ficance to the statement made to Mr. K. C. P. Struve in 1907
by the Dinka keeper of the shrine of Deng dit at Luang Deng.
126 TIME AND SPACE
After recounting the myth of the cow and calf, he added, 'And
to this day the Dinka has always lived by robbery, and the
Nuer by war.'^
Fighting, like eatt4eJiusbandry, is one of the chief activities
and dominant interests of all Nuer men, and raiding Dinka for
cattle is one of their principal pastimes. Indeed jaang, Dinka,
is sometimes used to refer to any tribe whom the Nuer habitually
raid and from whom they take captives. Boys look forward to
the day when they will be able to accompany their elders on
these raids against the Dinka, and as soon as youths have been
initiated into manhood they begin to plan an attack to enrich
themselves and to establish their reputation as warriors. Every
Nuer tribe raided Dinka at least every two or three years, and
some part of Dinkaland must have been raided annually. Nuer
have a proper contempt for Dinka and are derisive of their
fighting qualities, saying that they show as little skill as courage.
Kur jaang, fighting with Dinka, is considered so trifling a test
of valour that it is not thought necessary to bear shields on
a raid or to pay any regard to adverse odds, and is contrasted
with the dangers of kur Nath, fighting between Nuer themselves.
These boasts are justified both in the unflinching bravery of the
Nuer and by their military success.
The earliest travellers record that Nuer held both banks of the
Nile, but it is probable that the entire Zeraf Island was at one time
occupied by Dinka and it is certain that the whole of the country
from the Zeraf to the Pibor and, to the north of the Sobat, from
the confines of Shillukland to the Ethiopian scarp was, with the
exception of riverain settlements of Anuak, still in their hands as
late as the middle of last century, when it was seized by the Nuer
in two lines of expansion, to the north of the Sobat and to the
south of it. This is known from the statements of both Nuer and
Dinka, the evidence of genealogies and age-sets, and the records of
travellers, who frequently refer to the struggle between the two
peoples, the dominant position of the Nuer among their neigh-
bours, the awe they inspired in them, and their bravery and
chivalry.2 The conquest, which seems to have resulted in absorp-
' Sudan Intelligence Report, no. 152, 1907.
^ Werne, op. cit., p. 163; Abd-el-Hamid, op. cit., pp. 82-3; Philippe Terra-
nuova D'Antonio, 'Relation d'un voyage au Fleuve Blanc', Nouvelles annates
des voyages, Paris, 1859. Lejean, op. cit., p. 232; Poncet, op. cit., pp. 18, 26,
39, 41-2, and 44; Petherick, op. cit., vol. ii, p. 6; Heuglin, Reise in das Gebiet
TIME AND SPACE 127
tion and miscegenation rather than extermination, was so rapid
and successful that the whole of this vast area is to-day occupied
by Nuer, except for a few pockets of Dinka on the Sobat, Filus, and
Atar. Apart from these independent units there are many local
communities in Eastern Nuerland of Nuer who acknowledge that
they are of Dinka descent, and small lineages of Dinka origin are
found in every village and camp. Some Dinka tribes took refuge
with compatriots to the south, where the Gaawar and Lou con-
tinued to raid them. The Western Nuer likewise persistently
raided all the Dinka tribes that border them, particularly those to
the south and west, obtained a moral ascendency over them, and
compelled them to withdraw farther and farther from their
boundaries. To the west of the Nile, as to the east, Dinka captives
were assimilated, and there are many small lineages of Dinka
descent in every tribe and these are often preponderant in local
communities. Of all the Dinka only the Ngok, to the south of the
Sobat, were left in peace, probably on account of their poverty of
stock and grazing, though their immunity has a mythological
sanction. It seems also that the Atwot were not considered such
legitimate prey as the Dinka on account of their Nuer origin, and
it is probable that they were seldom molested as they are remotely
situated.
The favourite season for raiding Dinka was at the end of the
rains, though they were also invaded at their commencement.
Leek tribesmen told me that when they raided Dinka to the south-
west they used to sleep the first night near the villages of the Wot
tribe and the second night in the bush. They took no food with
them and ate only what fish they might hastily spear on the way,
travelling with all speed throughout the day and part of the night.
On the third day they attacked the Dinka villages or camps at
dawn. The Dinka seldom put up any resistance, but loosened
their cattle and tried to drive them away. No one seized cattle
till the enemy had been dispersed. Then each took what prizes he
could, often not troubling to tether his captures but slashing their
rumps in sign of ownership. Afterwards the beasts were tied up
des Weissen Nil und seiner westlichen Zufliisse in den Jahren 1862-1864, 1869,
p. 104. Georg Schweinfurth, The Heart of Africa (English translation), 1873,
vol. i, pp. 1 18-19 ; Gaetano Casati, Ten Years in Equatoria and the Return with
Emin Pasha (English translation), 1891, vol. i, p. 39; Romolo Gessi Pasha,
Seven Years in the Soudan (English translation), 1892, p. 57. The maps in-
tended by travellers to show the positions of the peoples of this area are
rather vague and are not in entire agreement. The reader may consult those
in Marno, op. cit., Poncet, op. cit., Heuglin, op. cit., the map compiled by
Lejean from information supplied by the Poncet brothers in Bulletin de la
Socidti de Geographie (Paris), i860; the maps prepared by V. A. Malte-Brun
in Nouvelles annales des voyages, 1855 and 1863, and others of this period.
128 TIME AND SPACE
in the enemy's kraal, the oxen being mainly slaughtered for food.
If the Dinka gathered reinforcements and returned to fight they
were met in full battle formation. Nucr fight in three divisions
with two or three hundred yards between each, and if one division
is engaged the others advance or retreat parallel to it according
to the fortunes of war. A party of scouts are in advance of the
central division and they charge up to the enemy, hurl their spears
at them, and fall back on the main body.
The raiders spent several weeks in Dinkaland and sometimes re-
mained there throughout the dry season, living on the milk and
flesh of captured cattle, on pillaged grain, and on fish. Using a
captured kraal as a base they extended their raid against distant
camps. Nuer migrations seem to have been conducted on these
lines, the raiders settling permanently in Dinka country and by
systematic raiding compelling the inhabitants to withdraw farther
and farther from the points of occupation. In the following season
a new series of raids was initiated and the process was repeated till
the Dinka were compelled to seek refuge with their kinsfolk of
another tribe and leave their country to its invaders. If settle-
ment was not contemplated, however, the raiders returned home
when they considered that they had sufficient booty.
Before camp was broken up there took place a custom highly
indicative of Nuer sense of equality and justice. It was recognized
that the whole force was jointly responsible for the success of the
raid and there was therefore a redistribution of the booty. The
prophet whose revelations sanctioned the raid first made a round
of the camp and selected from each household a cow for the divine
spirit of whom he was the mouthpiece. By this time a household
possessed some fifty head, so that it was no hardship to be asked
to give one to the spirit. There then took place a general scramble
and everybody rushed amid the herd to earmark beasts for him-
self. A man who could first seize an animal, tether it, and cut its
ear had absolute claim to it. The man who originally captured a
cow had the advantage that it was tethered near his windscreen,
but if he and members of his household had an undue share of the
booty they could not earmark all the beasts before they were
seized by others. As might be supposed, men frequently sustained
injury in these scrambles, for if two men seized the same cow they
fought with clubs for possession of it. One must not use the spear
on these occasions. Men of neighbouring camps took part in one
another's redistributions and there must have been great confusion.
Captives, women of marriageable age, boys, and girls, were not
redistributed but belonged to their original captor. Older women
and babies were clubbed and, when the raid was on a village, their
TIME AND SPACE 129
bodies were thrown on the flaming byres and huts. Captives
were placed in the centre of the camp, the women being sometimes
bound at night to render them more secure. Sexual intercourse
is taboo on. a^raijd. N.or may Nuer eat ^th a, captive. A boy
captive may not even draw water for them to drink. Only when
Abjdlang
Tribal Distribution about i860 (after V. A. Malte-Brun, Nouvelles annales
des voyages, 1863).
an ox has been sacrificed in honour of the ghosts, after the return
home, and they have been informed that strangers have entered
their homesteads, may Nuer have sexual relations with captives or
eat with them.
In the following section we describe other foreign contacts,,
but till European conquest the only foreign relations which
may be said to have been expressed in constant warfare were
those with the various Dinka tribes which border Nuerland.
130 TIME AND SPACE
We have not enumerated them, for their names are irrelevant.
The fighting between the two peoples has been incessant from
time immemorial and seems to have reached a state of equili-
brium before European conquest upset it, (Malte-Brun's map
compared with modern maps suggests that tribal positions have
not greatly altered since i860.) In the earlier part of the historic
period, from about 1840 to near the end of the century, Nuer
appear to have been expanding in search of new grazing, but
continued raiding for cattle, an aggressive action which we
attribute to the structural relations between the two peoples,
but which was, no doubt, intensified by rinderpest.
Though Dinka relations with Nuer are extremely hostile and
war between them may be caUed an established institution,
they have, nevertheless, united occasionally to make war
against the Egyptian Government and there have sometimes
been joint social gatherings. In times of famine Dinka have
often come to reside in Nuerland and have been readily accepted
and incorporated into Nuer tribes. In times of peace, also,
Dinka visited their relations who had been captured or who had
settled in Nuerland, and, as remarked earlier, there seems in
parts to have been some trade between the two peoples. The
strands of social relationships of a general kind, which are often
numerous across the boundaries of adjacent Nuer tribes and
which stretch across Nuerland, are thus prolonged weakly be-
yond the limits of Nuerland in occasional and hazardous con-
tacts with foreigners.
AU Dinka come into the category of Jaang, and Nuer feel
that category to be nearer to themselves than other categories
of foreigners. These foreign peoples, with all of whom the Nuer
have reached a state of balanced hostility, an equilibrium of
opposition, expressed occasionally in fighting, are, with the
exception of the Beir, generally classed as Bar, cattleless people
or people possessing very few cattle. A further category are the
Jur, cattleless people whom the Nuer regard as lying on the
periphery of their world, such as the Bongo-Mittu group of
peoples, the Azande, the Arabs, and ourselves. However, they
have separate names for most of these peoples.
We have remarked that Nuer feel Dinka to be nearer to them-
selves than other foreigners, and in this connexion we draw
TIME AND SPACE 131
attention to the fact that Nuer show greater hostihty towards,
and more persistently attack, the Dinka, who are in every
respect most akin to themselves, than any other foreign people.
This is undoubtedly due, in some degree, to the ease with which
they can pillage the vast Dinka herds. It may also, in part, be
attributed to the fact that of all neighbouring areas Dinkaland
alone opposes no serious oecological handicaps to a pastoral
people. But it may be suggested further that the kind of war-
fare that exists between Nuer and Dinka, taking into considera-
tion also the assimilation of captives and the intermittent
social relations between the two peoples between raids, would
seem to require recognition of cultural affinity and of like values.
War between Dinka and Nuer is not merely a clash of interests,
but is also a structural relationship between the two peoples, and,
such a relationship requires a certain acknowledgement on both
sides that each to some extent partakes of, the feelings and.
habits of the other. We are led by this reflection to note that
political relations are profoundly influenced by the degree of
cultural differentiation that exists between the Nuer and their
neighbours. The nearer people are to the Nuer in mode of liveli-
hood, language, and customs, the more intimately the Nuer re-
gard them, the more easily they enter into relations of hostility
with them, and the more easily they fuse with them. Cultural
differentiation is strongly influenced by oecological divergences,
particularly by the degree to which neighbouring peoples are
pastoral, which depends on their soils, water-supplies, insect
life, and so forth. But it is also to a considerable extent indepen-
dent of oecological circumstances, being autonomous and his-
torical. The cultural similarity of Dinka and Nuer may be
held largely to determine their structural relations; as, also,
the relations between the Nuer and other peoples are largely
determined by their increasing cultural dissimilarity. The
cultural cleavage is least between Nuer and Dinka; it widens
between Nuer and the Shilluk-speaking peoples ; and is broadest
between the Nuer and such folk as the Koma, Burun, and
Bongo-Mittu peoples,
Nuer make war against a people who have a culture like their i
own rather than among themselves or against peoples with cul-li
tures very different from their own. The relations between
132 TIME AND SPACE
social structure and culture are obscure, but it may well be that
had the Nuer not been able to expand at the expense of the
Dinka, and to raid them, they would have been more antagonistic
to people of their own breed and the structural changes which
would have resulted would have led to greater cultural hetero-
geneity in Nuerland than at present exists. This may be an
idle speculation, but we can at least say that the vicinity of a
people like themselves who possess rich herds that can be
plundered may be supposed to have had the effect of directing
the aggressive impulses of Nuer away from their fellow-country-
men. The predatory tendencies, which Nuer share with other
nomads, find an easy outlet against the Dinka, and this may
account not only for the few wars between Nuer tribes but also,
in consequence, be one of the explanations of the remarkable
size of many Nuer tribes, for they could not maintain what
unity they have were their sections raiding one another with the
persistence with which they attack the Dinka.
IX
Nuer had little contact with the Shilluk, a bufifejLQf Dinka_
dividing them in most places, and where they have a common .
border, warfare seems to have been restricted to incidents involving
only frontier camps. The powerful Shilluk kingdom, well organized
and comprising over a hundred thousand souls, could not have
been raided with the same impunity as the Dinka tribes, but the
characteristic reason Nuer give for not attacking them is other-
wise : 'They have no cattle. The Nuer only raid people who possess
cattle. If they had cattle we would raid them and take their cattle,
for they do not know how to fight as we fight.' There is no actual
or mythological enmity between the two peoples.
The Anuak, who also belong to the Shilluk-Luo group, border
the Nuer to the south-east. Though they are almost entirely
horticultural to-day, they possessed herds in the past and in
Nuer opinion their country has better grazing than Shillukland.
It was overrun by the Nuer over half a century ago as far as the
foothills of the Ethiopian scarp, but was quickly abandoned, prob-
ably because of tsetse, for the Anuak put up little resistance. Nuer
continued to raid them up to thirty years ago, when they obtained
rifles from Abyssinia and were better able to resist and even to
take the offensive. In spite of two reverses they finally succeeded
in penetrating Lou country, where they inflicted heavy casualties
TIME AND SPACE 133
and captured many children and cattle, a feat which brought the
Government down the Pibor, thereby closing hostilities. Many
evidences show that at one time the Anuak extended far westward
of their present distribution and were displaced from these sites,
or assimilated, by the Nuer.
The other peoples with whom Nuer come into contact may
be mentioned very briefly as their interrelations have little
political importance. Another south-eastern neighbour is the
Beir (Murle) people. As far as I am aware the Nuer did not raid
them often and those few who know something of them respect
them as devoted herdsmen. To the north-east of Nuerland the
Gaajak have for several decades had relations with the Galla of
Ethiopia. These appear to have been peaceful and there was a
certain amount of trade between the two peoples. Absence of
friction may be attributed largely to the corridor of death that
divides them, for when the Galla descend from their plateau they
quickly succomb to malaria while any attempt on the part of the
Nuer to move eastwards is defeated by the tsetse belt that runs
along the foot-hills. The Gaajak raided the.Burun and Koma (both
often referred to vaguely as 'Burun') for captives, and they were
too small in numbers and too unorganized to resist or retaliate. To
the north-west the Jikany, Leek, and Bui tribes occasionally raided
the Arabs and the communities of the Nuba Mountains ; and, to
judge from a statement by Jules Poncet, the trouble over water
and grazing in the dry season that occurs to-day between Nuer and
Arabs has long occurred.'
The Arab slavers andivoiv traders, who caused so much misery _
and destruction among thej3eoples^gf_the Southern Sudan a fter_
the conquest of the Northern Sudan by Muhammad Ali jn i82Xt_
very little inconvenienced the Nuer. They ^sometimes pillaged
riverside villages, but I know of no record of their having penetrated
far inland, and it was only the more accessible sections of the Zeraf
River tribes that appear to have suffered to any extent from their
depredations. I do not believe that anywhere were the Nuer
deeply affected by Arab contact.^ The Egyptian Government and,
later, the Mahdist Government, which were supposed to be in
control of the Sudan from 1821 to the end of the century, in no
way administered the Nuer or exercised control over them from
the riverside posts they established on the fringes of their country.
' Poncet, op. cit., p. 25.
^ I find myself unable to accept Casati's statement of the position (op. cit.,
vol. i, p. 38), but consider that greater reliance should be placed on the opinion
of other authorities. See Romolo Gessi's letter to the Editor of Esploratore in
1880 (op. cit., p. xx), and Lejean's report from Khartoum in i860 (op. cit.,
P- 215)-
134 TIME AND SPACE
The Nucr sometimes raided these posts and were sometimes raided
from them,' but on the whole it may be said that they pursued
their Hves in disregard of them.
This disregard continued after the reconquest of the Sudan by
Anglo-Egyptian forces and the establishment of the new administra-
tion. The Nuer were the last important people to be brought under
control and the administration of their country cannot be said to
have been very effective till 1928, before which year government
consisted of occasional patrols which only succeeded in alienating
them further. The nature of the country rendered communica-
tions difficult and prevented the establishment of posts in Nuerland
itself, and the Nuer showed no desire to make contact with those
on its periphery. Little control was exercised and it was impossible
to enforce decisions.^ A further difficulty was the absence of Nuer
who had travelled in foreign parts and spoke Arabic, for their
place was usually taken, as interpreters and in other capacities,
by Dinka and Anuak, who were distrusted, and rightly so, by the
Nuer against whom they lodged every kind of complaint.
The truculence and aloofness displayed by the Nuer is conform-
able to their culture, their social organization, and their character.
The self-sufficiency and simplicity of their culture and the fixation
of their interests on their herds explain why they neither wanted
nor were willing to accept European innovations and why they
rejected peace from which they had everything to lose. Their
i political structure depended for its form and persistence on balanced
antagonisms that could only be expressed in warfare against their
neighbours if the structure were to be maintained. Recognition
of fighting as a cardinal value, pride in past achievements, and a
deep sense of their common equality and their superiority to other
peoples, made it impossible for them to accept willingly domina-
tion, which they had hitherto never experienced. Had more been
known about them a different policy might have been instituted
earlier and with less prejudice.^
' See, for example, Casati, op. cit., p. 221.
^ For an account of conditions at this time see the Sudan Intelligence
Reports, especially those by Kaimakam F. J. Maxse (no. 6i, 1899), Capt. H. H.
Wilson (no. 128, 1905), and O'Sullivan Bey (no. 187, 1910).
^ For abusive references to the Nuer see Sir Samuel Baker, The Albert
N'Yanza, 1913 (first published in 1866), pp. 39-42 ; Capt. H. H. Austin, Among
Swamps and Giants in Equatorial Africa, 1902, p. 15 ; Count Gleichen, op. cit.,
1905, vol. i, p. 133; C. W. L. Bulpett, A Picnic Party in Wildest Africa, 1907,
pp. 22-3 and 35 ; Bimbashi Coningham, Sudan Intelligence Report, no. 192,
1910; H. Lincoln Tangye, In the Torrid Sudan, 1910, p. 222; E. S. Stevens,
My Sudan Year, 1912, pp. 215 and 256-7; H. C. Jackson, op. cit., p. 60; The
Story of Fergie Bey. Told by himself and some of his Friends, 1930, p. 113;
and J. G. Millais, Far away up the Nile, 1924, pp. 174-5.
PLATE XV
v*..-;-;
a. Windscreen (Lou)
b. A WL'll in thr l)c'(l of the X\-anding (Lou)
TIME AND SPACE 135
In 1920 large-scale military operations, including bombing and
machine-gunning of camps, were conducted against the Eastern
Jikany and caused much loss of life and destruction of property.
There were further patrols from time to time, but the Nuer re-
mained unsubdued. In 1927 the Nuong tribe killed their District
Commissioner, while at the same time the Lou openly defied the
Government and the Gaawar attacked Duk Faiyuil Police Post.
From 1928 to 1930 prolonged operations were conducted against
the whole of the disturbed area and marked the end of serious
fighting between the Nuer and the Government. Conquest was a
severe blow to the Nuer, who had for so long raided their neighbours
with impunity and whose country had generally remained intact.
X
In our account of Nuer time-reckoning we noted that in one
department of time their system of reckoning is, in a broad
sense, a conceptualization, in terms of activities, or of physical
changes that provide convenient points of reference for activi-
ties, of those phases of the oecological rhythm which have pecu-
liar significance for them. We further noted that in another
department of time it is a conceptualization of structural rela-
tions, time units being co-ordinate with units of structural
space. We have given a brief description of these units of struc-
tural space in its political, or territorial, dimension and have
drawn attention to the influence of oecology on distribution
and hence on the values given to the distribution, the interrela-
tion between which is the political system. This system is not,
however, as simple as we have presented it, for values are not
simple, and we now attempt to face some of the difficulties we
have so far neglected. We start this attempt by asking what it is
the Nuer mean when they speak of their cieng.
Values are embodied in words through which they influence
behaviour. When a Nuer speaks of his cieng, his dhor, his gol,
&c., he is conceptualizing his feelings of structural distance,
identifying himself with a local community, and, Jby^^o_ doing,
cutting himself off from other communities of the same kind._
An examination of the word cieng will teach us one of the most
fundamental characteristics of Nuer local groups and, indeed,
of all social groups : their structural relativity.
136 TIME AND SPACE
What does a Nuer mean when he says, T am a man of such-and-
such a cieng'? Cieng means 'home', but its precise significance
varies with the situation in which it is spoken. If one meets an
Enghshman in Germany and asks him where his home is, he
may reply that it is England. If one meets the same man in
London and asks him the same question he will tell one that his
home is in Oxfordshire, whereas if one meets him in that county
he will tell one the name of the town or village in which he lives.
If questioned in his town or village he will mention his particular
street, and if questioned in his street he will indicate his house. So
it is with the Nuer. A Nuer met outside Nuerland says that his
home is cieng Nath, Nuerland. He may also refer to his tribal
country as his cieng, though the more usual expression for this is
rol. If one asks him in his tribe what is his cieng, he will name his
village or tribal section according to the context. Generally he will
name either his tertiary tribal section or his village, but he may
give his primary or secondary section. If asked in his village he will
mention the name of his hamlet or indicate his homestead or the
end of the village in which his homestead is situated. Hence if a
man says 'Wa ciengda , 'I am going home', outside his village he
means that he is returning to it ; if in his village he means that
he is going to his hamlet ; if in his hamlet he means that he is going
to his homestead. Cieng thus means homestead, hamlet, village,
and tribal sections of various dimensions.
The variations in the meaning of the word cieng are not due
to the inconsistencies of language, but to the relativity of the
group-values to which it refers. I emphasize this character of
structural distance at an early stage because an understanding
of it is necessary to follow the account of various social groups
which we are about to describe. Once it is understood, the
apparent contradictions in our account will be seen to be contra-
dictions in the structure itself, being, in fact, a quality of it.
The argument is here introduced in its application to local com-
munities, which are treated more fully in the next chapter, and
its application to lineages and age-sets is postponed to Chapters
V and VI.
A man is a member of a political group of any kind in virtue
of his non-membership of other groups of the same kind. He
sees them as groups and their members see him as a member of
a group, and his relations with them are controlled by the
structural distance between the groups concerned. But a man
TIME AND SPACE 137
does not see himself as a member of that same group in so far as
he is a member of a segment of it which stands outside of and is
opposed to other segments of it. Hence a man can be a member
of a group and yet not a member of it. This is a fundamental y
principle of Nuer poHtical structure. Thus^a man is a member of /;
his tribe in its relation to other tribes, but he is not a member
of his tribe in the relation of his segment of it to other segments
of the same kind. Likewise a man is a member of his tribal
segment in its relation to other segments, but he is not a member
of it in the relation of his village to other villages of the same
segment. A characteristic of any political group is hence its
invariable tendency towards fission and the opposition of its seg-
ments, and another characteristic is its tendency towards fusion
with other groups of its own order in opposition to political
segments larger than itself. Political values are thus always,
structurally speaking, in conflict. One value attaches a man to
his group and another to a segment of it in opposition to other
segments of it, and the value which controls his action is a func-
tion of the social situation in which he finds himself. For a man
sees himself as a member of a group only in opposition to other
groups and he sees a member of another group as a member of
a social unity however much it may be split into opposed seg-
ments.
Therefore the diagram presented on p. 114 illustrates political
structure in a very crude and formal way. It cannot very
easily be pictured diagrammatically, for political relations
are relative and dynamic. They are best stated as tendencies
to conform to certain values in certain situations, and the value
is determined by the structural relationships of the persons who
compose the situation. Thus whether and on which side a
man fights in a dispute depends on the structural relationship
of the persons engaged in it and of his own relationship to
each party.
We need to refer to another important principle of Nuer
political structure : the smaller the local group the stronger the
sentiment uniting its members. Tribal sentiment is weaker than
the sentiment of one of its segments and the sentiment of a seg-
ment is weaker than the sentiment of a village which is part of
it. Logically this might be supposed to be the case, for if unity
4507 -J
138 TIME AND SPACE
within a group is a function of its opposition to groups of the
same kind it might be surmised that the sentiment of unity
within a group must be stronger than the sentiment of unity
within a larger group that contains it. But it is also evident
that the smaller the group the more the contacts between its
members, the more varied are these contacts, and the more
they are co-operative. In a big group like the tribe contacts
between its members are infrequent and corporate action is
limited to occasional military excursions. In a small group like
the village not only are there daily residential contacts, often
of a co-operative nature, but the members are united by close
agnatic, cognatic, and affinal ties which can be expressed in
reciprocal action. These become fewer and more distant the
wider the group, and the cohesion of a political group is un-
doubtedly dependent on the number and strength of ties of a
non-political kind.
It must also be stated that political actualities are confused
and conflicting. They are confused because they are not always,
even in a political context, in accord with political values,
though they tend to conform to them, and because social ties of
a different kind operate in the same field, sometimes strengthen-
ing them and sometimes running counter to them. They are
conflicting because the values that determine them are, owing
to the relativity of political structure, themselves in conflict.
Consistency of political actualities can only be seen when the
dynamism and relativity of political structure is understood
and the relation of political structure to other social systems
is taken into consideration.
CHAPTER IV
THE POLITICAL SYSTEM
I
NuER tribes are split into segments. The largest segments we
caU_jgrimary tribal sections and these are further segmented
iiltQ_secondary tribal sections which are further segmented into
t ertiar y tribal sections. Experience shows that primary, secon-
dary, and tertiary are sufficient terms of definition, and in the
smallest tribes probably fewer terms are required. A tertiary
tribal section comprises a number of village communities which
are composed of kinship and domestic groups.
Thus, the Lou tribe, as shown in the diagram below, is seg-
mented into the Gun and Mor primary sections. The Gun
primary section is segmented into the Rum j ok and Gaatbal
secondary sections. The Gaatbal secondary section is further
segmented into the Leng and Nyarkwac tertiary sections. Only
a few segments are shown diagrammatically : Gaaliek is split
into Nyaak and Buth, Rumjok into Falker, Nyajikany, Kwac-
gien, &c., and so on.
LOU TRIBE
Mor prim. sect.
Gun prim. sect.
Gaaliek sec. sect.
Rumjok sec. sect.
Jimac sec. sect.
Leng tert. sect.
Jaajoah sec. sect.
Nyarkwac tert. sect.
Gaatbal
sec. sect.
The diagram on p. 140 shows the primary sections of the
Eastern Gaagwang tribe and the primary and secondary sec-
tions of the Eastern Gaajak and Gaajok tribes. They are
140 THE POLITICAL SYSTEM
presented as accurately as my knowledge permits, but anyone
acquainted with the difficulties of unravelling the complex
system of Nuer tribal divisions will not be surprised if he dis-
covers sections which he knows under different names or others
which he may think should not have been omitted. I am not
certain of the secondary sections of the Gaagwang tribe, which
I have not visited.
EASTERN JIKANY TRIBES
Gaagwang
Gaajok tribe tribe Gaajak tribe
Thiur sec. sect.
Gaatcika
prim. sect.
Nyayan sec. sect.'
Laang
prim. sect.
Dwong sec. sect.
Kwith sec. sect.
Cany sec. sect.
Wau sec. sect.
Wangkac
prim. sect.
Minyaal sec. sect.
Nyingee
prim. sect.
Kong sec. sect.^
Wang sec. sect.
Col sec. sect.^
Nyathol sec. sect.
Dhilleak sec. sect.*
Pwot sec. sect.
Nyaang
prim. sect.
Tar sec. sect.
Kwal sec. sect.
Yol
prim. sect.
Kang sec. sect.
Yiic sec. sect.
Cam sec. sect.
Lony sec. sect.
Kwul sec. sect.
Gaagwong
prim. sect.
Reng
prim. sect.
Thiang
prim. sect.
Among the Western Jikany the Gaagwang appear to be
classed as part of the Gaajok tribe whose country stretches on
both sides of the Bahr-el-Ghazal, the Gaajak tribe living to
the south of that river. The primary sections of these two
tribes, Gaagwong, Reng, Thiang, Laang, Wangkac, and Yol,
are the same as in the east, but some secondary sections which
are important to the north of the Sobat are not to be found,
except as very small clusters, on the Bahr-el-Ghazal, and vice
versa. The reason for this is that certain lineages migrated east-
wards while others remained in the homeland.
After having satisfied myself that the segmentation of other
With which goes Nyajaani,
Also called Nyaruny.
^ Also called Tick and Yaar.
* Also called Gying.
THE POLITICAL SYSTEM 141
Nuer tribes is on the same pattern as that of the Lou and Jikany
tribes I did not make detailed lists of their divisions, being
interested in new and different inquiries. I include, however,
some diagrammatic representations of tribal segmentation in
Gaawar, Lak, and Thiang countries, for which I am indebted
to Mr. B, A. Lewis, at one time Commissioner for Zeraf River
District.
GAAWAR TRIBE
Radh prim. sect.
Bar prim. sect.
Kerf ail sec. sect.
Bang tert. sect.
Lidh
Nyadakwon sec. sect.
Jamogh tert. sect.
sec. sect
Per sec. sect.
Caam tert. sect.
Nyaigua sec. sect.
Gatkwa
Gatkwa tert. sect.
sec. sect
Jitheib sec. sect.
LAK TRIBE
Jenyang prim, sect, Kwacbur prim. sect.
Kudwop sec. sect.
Nyawar tert. sect.
Dongrial tert. sect.
Nyapir sec. sect.
Thiang tert. sect.
Kar. tert. sect.
Cuak tert. sect.
Tobut
sec. sect.
Lak
sec. sect.
THIANG TRIBE
Riah prim. sect. Bang prim. sect.
Juak sec. sect.
Gul tert. sect.
Bedid tert. sect.
Manyal sec. sect.
Dwong tert. sect.
Kwoth sec. sect.
Giin sec. sect.
Cuol sec. sect.
Nyangur
sec. sect.
142 THE POLITICAL SYSTEM
It will have been observed that I have not attempted to list
aU the sections of each tribe, but have merely tried to indicate
I the mode of segmentation so that the relation between tribal_
i divisions and lineages may be more clearly understood in the
following chapter, .
II
Segments of a tribe have many of the characteristics of the
tribe itself. Each has its distinctive name, its common senti-
ment, and its unique territory. Usually one section is clearly
divided from another by a wide stretch of bush or by a river.
Segments of the same tribe also tend to turn in different direc-
tions for their dry season pastures, as is shown on the sketch-maps
on pp. 57, 59, and 6i, so that the spatial divisiojisj:if the raina.-
are maintained, and may be accentuated during the drought,
though, as we have pointed out, in the larger tribes to the east
of the Nile severity of natural conditions may also produce
closer interrelations than in the smaller tribes to the west.
The smaller the tribal segment the more compact its terri-
' tory, the more contiguous its members, the more varied and
^^ \ more intimate their general social ties, and the stronger there-
fore its sentiment of unity. As we shall see, a tribal segment
is crystallized around a lineage of the dominant clan of the
tribe and the smaller the segment the closer the genealogical
relationship between members of this clan fragment. Also the
smaller the segment the more the age-set system determines
behaviour and produces corporate action within it. Political
cohesion thus not only varies with variations of political dis-
tance but is also a function of structural distance of other
kinds.
Each segment is itself segmented and there is opposition be-
tween its parts. The members of any segment unite for war
against adjacent segments of the same order and unite with
these adjacent segments against larger sections. Nuer them-
selves state this structural principle clearly in the expression of
their political values. Thus they say that if the Leng tertiary
section of the Lou tribe fights the Nyarkwac tertiary section —
and, in fact, there has been a long feud between them — the
villages which compose each section will combine to fight ; but
THE POLITICAL SYSTEM 143
if there is a quarrel between the Nyarkwac tertiary section and
the Rum j ok secondary section, as has occurred recently over
water rights at Fading, Leng and Nyarkwac will unite against
their common enemy Rumjok which, in its turn, forms a coali-
tion of the various segments into which it is divided. If there
is a fight between the Mor and the Gun primary sections, Rum-
jok and Gaatbal will unite against the combined Mor sections:
Gaaliek, Jimac, and Jaajoah. If there is fighting against the
Gaajok or the Gaawar the primary sections, Gun and Mor,
will, at any rate in theory, combine and a united Lou tribe
will take the field, since both sections belong__tothe sanieu-
poHtical group and since theiFHommant lineages belong to thei
same clan. Certainly they used to unite in raids oiTTRe Diiika:
Among the Eastern Gaajok, Minyal, Wang, and Nyathol
sections combine against Yol. Also Thiur, Dwong, and Kwith
sections unite for war. These fights between tribal sections and
the feuds that result from them, though based on a territorial
principle, are often represented in terms of lineages, since there
is a close relation between territorial segments and lineage seg-
ments, and Nuer habitually express social obligations in a
kinship idiom. Thus in telling me that Wangkac and Yol would
unite for war against any other section Nuer stated the proposi-
tion by saying that the wangkac and yol lineages, which are
the dominant lineages in these sections, would unite because
their ancestors were sons of the same mother. We shall see in
Chapter V that Nuer generally speak in such terms.
This principle of segmentation and the opposition between
segments is the same in every section of a tribe and extends
beyond the tribe to relations between tribes, especially among
the smaller Western Nuer tribes, which coalesce more easily and
frequently in raiding the Dinka and in fighting one another than
the larger tribes to the east of the Nile. Thus a man of the
Fadang section of the Bor tribe exemplified it when he told me,
'We fight against the Rengyan, but when either of us is fighting
a third party we combine with them'. It can be stated in
hypothetical terms by the Nuer themselves and can best be
represented in this way. In the diagram on p. 144, when Z' fights
Z^ no other section is involved. When Z' fights Y', Z^ and Z^
unite as Y^. When Y' fights X', Y' and Y^ unite, and so do X'
144 THE POLITICAL SYSTEM
and X^. When X^ fights A, X', X^, Y', and Y^ all unite as B.
When A raids the Dinka A and B may unite.
A B
X Y
X'
Y«
X2
z'
z^
Y*
The larger tribal sections were almost autonomous groups and
acted as such in their enmities and alliances. At one time they
would be fighting among themselves and at another would be
combining against a third party. These combinations were not
always as regular and simple as they were explained to me and as
I have stated them. I give a few examples of fighting between
tribal sections. One of the worst wars in Nuer history occurred
in the last generation between the Gun and the Mor moieties of
the Lou tribe. It was known as kur luny yak, the war of letting
loose the hyena, because so many people were killed that the
dead were left for the hyenas to eat. It is said that in this struggle
men displayed unusual ferocity, even cutting off arms to seize
quickly ivory armlets. There was a lengthy, and more recent, feud
between the Leng and the Nyarkwac tertiary sections of the Lou
tribe which has continued up to the present. It sprang out of an
earlier fight between Thiang and Yol which once formed sub-sec-
tions of the Nyarkwac. The ancestors of the dominant lineages in the
Leng and Yol divisions were brothers, whereas the ancestor of the
dominant lineage of the Thiang division stood to these brothers in
the relationship of sister's son. For a long while Yol and Thiang
lived peaceably together, but some thirty years ago a fight broke out
between them, and Thiang, defeated, fled to seek protection among
the Leng section. The Yol sent messages to the Leng telling them
that they were not to receive their enemies or give them asylum.
The Leng replied that the ancestor of the leng lineage was the
maternal uncle of the ancestor of the thiang lineage and that they
could not refuse asylum to their sisters' sons. This attitude involved
Yol (Nyarkwac) in a second war, this time against a combination
of Leng and Thiang. Other recent Lou feuds were between the
THE POLITICAL SYSTEM 145
Falkir and Nyajikany divisions of the Rum j ok secondary section and
between various local communities of the Mor primary section, par-
ticularly between two divisions of the Jimac secondary section.
In Eastern Gaajok country the Yol primary section joined the
Gaagwang tribe, who seem to have identified themselves so much
with the Gaajok tribe that we may almost speak of them, as we
can to the west of the Nile, as a single tribe separated from the
Gaajak by the wide Macar swamps, against several, if not all,
sections of the Gaajak tribe. Yol fought Nyayan while Gaagwang
fought Reng and Kang. About half a century ago the Laang and
Wangkac primary sections of the Gaajok tribe were involved in a
long feud and there was also warfare between the Yol and Wangkac
sections in which Yol, assisted by their allies the Gaagwang tribe,
were victorious ; the Wangkac being so heavily defeated that they
moved southwards to the banks of the Pibor river. Here, they say,
they were attacked by Turuk (Arabs of some kind) and moved
north again to the site of their old homes. They were too exhausted
to resume their feud with the Yol section. In spite of these internal
feuds, if any section of the Gaajok tribe is engaged in warfare with
the Lou tribe all its sections will come to assist the threatened
section if it is not strong enough to resist whatever section of the
Lou is opposed to it. There have also been feuds between the
Eastern Gaajak sections, e.g. between Thiang and Reng. When
two tribes fight, other tribes remain neutral, and if two sections
of a tribe are at war with one another the other sections of
each may leave them to fight it out if they are well matched
and do not appeal for assistance. Some of Miss Soule's infor-
mants pointed out that when there was trouble a few years
ago between the Yol section of the Gaajok tribe and the Lony
section of the Gaajak tribe they were strong enough to fight their
own battle, but had Lony not been strong enough to fight single-
handed then the Kaang and Tar, and possibly other, sections of
the Gaajak would have come to its assistance, in which case the
Gaajok sections would have joined Yol. They also pointed out
that at present there is trouble between the Luluaa section and
the Wang section. There is also trouble between various sections
of Wangkac. If Luluaa and Wang begin to fight then the Wangkac
sections will compose their differences and join Luluaa.
In accordance with the general tendency to the west of the Nile
the Western Gaajok and Gaajak tribes are not only smaller but less
united than the Eastern Gaajok and Gaajak. Both, on the Bahr-el-
Ghazal, had frequent and bitter internal feuds. There was a fierce
battle between the Gai division of the Gaagwong primary section
and two other divisions of the same section, the Kwoth and the
146 THE POLITICAL SYSTEM
Bor, whose dominant lineages spring from one mother. The Kwoth
and Bor divisions were defeated and migrated southwards to settle
at Kwac in Rengyan country. The same Gai division also had a feud
with the Reng primary division, after which it moved into Karlual
country. There were numerous other feuds in the Gaajak tribe.
The Gaajok tribe once lived entirely on the left bank of the Bahr-
el-Ghazal and their present extension on the right bank is a
consequence of migration following feuds.
The Leek tribe once all lived on the right bank of the Bahr-el-
Ghazal. Here two of its primary sections, Cuaagh and Deng who
lived to the west of the Gany river, fought the third primary section,
the Keunyang (Karlual), who lived to the east of that river, and
being defeated crossed the Bahr-el-Ghazal and settled on its left
bank. The story is that some aristocrats of the Nyapir section and
some aristocrats of the Nyawah section used offensive expres-
sions in songs about one another. These songs led to fighting
between the young men, one of whom on each side was killed.
There was further fighting in consequence, and finally Deng
and Cuaagh crossed the river. In the following year they
recrossed the Bahr-el-Ghazal to camp in the dry season on the
right bank and on their return to their villages they drove before
them herds belonging to Keunyang. Six of their girls came back
to collect their dairy vessels which they had left in a camp hut and
were waylaid and killed by some Keunyang men. This act was
considered a serious breach of the rules of war, for the Nuer do
not kill their own womenfolk. Because of it the Deng uttered a
curse, according to which it is forbidden to an aristocrat of Keun-
yang who crosses the Bahr-el-Ghazal and settles among the Deng or
Cuaagh, and also to an aristocrat of Deng or Cuaagh who moves
southwards and settles among the Keunyang, to build a cattle-
byre in the usual manner. The curse also causes an aristocrat
who thus changes his residence to beget only boys for his first
few children because of the slain girls. When the Government
raided Karlual (Keunyang) country many Keunyang aristocrats
crossed the river to live in Deng and Cuaagh country. At the
present day many Deng and Cuaagh spend the dry season in
Keunyang country because their own country is not rich in good
pasture grasses but consists mainly of marsh grasses which are not
so nourishing.
Within each of these primary sections there were constant feuds.
Thus in Karlual country, Riaagh, Gom, Jiom, Nyaagh, Jikul, and
Ngwol sections have been frequently at feud with one another.
It would be tedious to recount the occasions and outcome of these
petty feuds. I want only to make it clear that the villages occupied
PLATE X\'I
—
O
THE POLITICAL SYSTEM 147
by these minor sections, Tutgar (Ngwol), Nyang (Riaagh), Nyueny
(Juak), Kol (Jikul), &c., are only a few miles from their nearest
neighbours, all of them being contained in a radius of five miles. It
is between villages and tertiary tribal sections that fights most
frequently take place and feuds develop.
I could give many more examples of feuds, but to no purpose,
for those I have cited amply illustrate the lack of political con-
trol in Nuer tribes. We may conclude that a man's tribe only
claims his allegiance in intertribal fighting and in wars against
the Dinka. In normal times a man thinks and acts as a member
of very much smaller local groups with the members of which
he has manifold contacts.
Ill
We may use the diagram on p. 144 to emphasize the principle
of contradiction in political structure. A member of TJ- tertiary
section of tribe B sees himself as a member of Z^ in relation to
Z', and all other members of Z^ see themselves as members of
that group in relation to Z^ and are so regarded by members
of Z'. But he regards himself as a member of Y^ and not of Z^
in relation to Y' and is so regarded by members of Y'. Likewise
he regards himself as a member of Y and not of Y^ in relation
to X, and as a member of the tribe B, and not of its primary
section Y, in relation to tribe A. Any segment sees itself as an
independent unit in relation to another segment of the same
section, but sees both segments as a unity in relation to another
section ; and a section which from the point of view of its mem-
bers comprises opposed segments is seen by members of other
sections as an unsegmented unit. Thus there is, as we have
pointed out earlier, always contradiction in the definition of a
political group, for it is a group only in relation to other groups.
A tribal segment is a political group in relation to other seg-
ments of the same kind and they jointly form a tribe only
in relation to other Nuer tribes and adjacent foreign tribes
which form part of the same political system, and without
these relations very little meaning can be attached to the con-
cepts of tribal segment and tribe. We make_here the^^ame__
point ^sjffe m r de_..iii- discussing: llie word cieng: t hat poli tical-
values are relative and-thatiJia-political sys tem is an equilibrium
148 THE POLITICAL SYSTEM
between opposed tendencies -tojwards iission and f iision,-h£tween
the-lendenejrof all groups to segment, and the tendency of all
groups to combine with segments ol the same order. The ten-
dency towards fusion is inherent in the segmentary character
of Nuer pohtical structure, for although any group tends to
split into opposed parts these parts must tend to fuse in relation
to other groups, since they form part of a segmentary system.
Hence fission and fusion in political groups are two aspects
of the same segmentary principle, and the Nuer tribe and its
divisions are to be understood as an equilibrium between these
two contradictory, yet complementary, tendencies. Physical
environment, mode of livelihood, poor communications, a simple
technology, and sparse food-supply — all, in fact, that we call
their oecology — to some extent explain the demographic fea-
tures of Nuer political segmentation, but the tendency towards
segmentation must be defined as a fundamental principle of
their social structure.
There must always, therefore, be something arbitrary about
our formal definition of a tribe by the characters we have earlier
listed. The political system is an expanding series of opposed
segments from the relations within the smallest tribal section to
intertribal and foreign relations, for opposition between seg-
ments of the smallest section seems to us to be of the same
structural character as the opposition between a tribe and its
Dinka neighbours, though the form of its expression differs.
Often it is by no means easy to decide whether a group should be
regarded as a tribe or as the segment of a tribe, for political
structure has a dynamic quality. Using payment of blood-
wealth as the principal criterion we class the Eastern Gaajok and
Gaajak as distinct tribes because there is no compensation for
homicide between them, but they consider themselves to be a
single community in relation to the Lou. The tribal value is still
recognized throughout Lou country, but, in fact, the Gun and
the Mor sections are largely autonomous and it may be doubted
whether compensation for homicide would actually be paid
between them, though people say that it ought to be paid. It
appears that so many people were killed in feuds between the
Yol and Wangkac primary sections of the Gaajok tribe that all
payments for homicide lapsed. On the other hand, I was told
THE POLITICAL SYSTEM 149
that at the time when the influence of the Lou prophets Ngun-
deng and Gwek was at its height compensation was for a time
paid between the Lou and the Gaajok. In the larger tribes the
segments recognize a formal unity, but there may be little
actual cohesion. The tribal value is still affirmed, but actual
relations may be in conflict with it since they are based on local
allegiances within the tribe, and, in our opinion, it is this conflict
between rival values within a territorial system which is the
essence of the political structure.
Nuer tribes are an evaluation of territorial distribution, and
tribal and intertribal and foreign relations are standardized
modes of behaviour through which the values are expressed.
The tribal value is, therefore, relative and at any time is attached
to a certain extension of an expanding series of structural rela-
tions without being inevitably fixed to that extension. More-
over, it is not only relative because what we designate a tribe
to-day may be two tribes to-morrow, but it can only be said to
determine behaviour when a certain set of structural relations
are in operation, mainly acts of hostility between tribal seg-
ments and between a tribe and other groups of the same struc-
tural order as itself, or acts likely to provoke aggression. A
tribe very rarely engages in corporate activities, and, further-
more, the tribal value determines behaviour in a definite and
restricted field of social relations and is only one of a series of
political values, some of which are in conflict with it. The same
is true of its segments. We would, therefore, suggest that Nuer
political groups be defined, in terms of values, by the relations
between their segments and their interrelations as segments of
a larger system in an organization of society in certain social
situations, and not as parts of a kind of fixed framework within
which people live.
We do not doubt that there is an interdependence between
the various sectional interrelations and the entire political
system of which they form part, but this cannot easily be demon-
strated. It has been noted that the smaller the local group the
more cohesive it is and the more contacts of various kinds its
members have with one another. There is less solidarity the
wider we extend the circle from a village to adjacent tribes. It
might be assumed, therefore, that there is always greater
150 THE POLITICAL SYSTEM
opposition between two groups than between the segments of
either and that the segments are held together, as it were, by
this external pressure, but we cannot admit that this view
accords with the facts, because greater hostility appears to be
felt between villages, groups of villages, and tertiary tribal
sections than between larger tribal sections and between tribes.
Probably the raids conducted tribally and in tribal federation
against the Dinka had an integrating action, but the Dinka
were not aggressive against the Nuer and it seems that the
maintenance of tribal structure must rather be attributed to
opposition between its minor segments than to any outside
pressure. If this be so, and a consideration of the institution of
the feud suggests that it is so, we arrive at the conclusion that
the more multiple and frequent the contacts between members
of a segment the more intense the opposition between its parts.
However paradoxical this conclusion may at first seem we are
led to it both by observation and by reflection on what con-
stitutes a segmentary system.
IV
~^We have used the term 'feud' in the last section in the sense
of lengthy mutual hostility between local communities within
a tribe. This broad and slightly vague usage seems justified by
convention and also, as we shall show, because, although
responsibility for homicide and the duty of exacting vengeance
directly fall only on the close agnatic kin of slayer and slain, the
communities to which the two parties belong are, in one way or
another, involved in the hostility that ensues and, often enough,
in any fighting that may result from the dispute. Strictly,
however, the word might be considered more appropriately used_
to describe the relations between the kin on both sides in -a
sitaation^ homiciderTof It-then refers to a specific institution.
We sometimes7 therefore, speak of the 'blood-feud ' to emphasize
this restricted and more clearly defined meaning.
"•^Blood-feuds are a tribal instituttonTior they can only occur
whefe"ar^eacli of law is recognized since they are the way in
which reparation is obtained. Fear of incurring a blood-feud is,
in fact, the most important legal sanction within a tribe and the
main guarantee of an individual's life and property. If a com-
THE POLITICAL SYSTEM 151
munity of one tribe attempts to avenge a homicide on a com-
munity of another tribe a state of intertribal war, rather than
a state of feud, ensues, and there is no way of settling the dispute
by arbitration.
As Nuer are very prone to fighting, people are frequently
killed. Indeed it is rare that one sees a senior man who does not
show marks of club or spear. A Nuer gave me the following
causes of fighting: a dispute about a cow; a cow or goat eats
a man's miUet and he strikes it ; a man strikes another's little
son ; adultery ; watering rights in the dry season ; pasturage
rights ; a man borrows an object, particularly a dance ornament,
without asking its owner's permission. A Nuer will at once fight
if he considers that he has been insulted, and they are very
sensitive and easily take offence. When a man feels that he has
suffered an injury there is no authority to whom he can make
a complaint and from whom he can obtain redress, so he at once
challenges the man who has wronged him to a duel and the
challenge must be accepted. There is no other way of settling a
dispute and a man's courage is his only immediate protection
against aggression. Only when kinship or age-set status in-
hibits an appeal to arms does a Nuer hesitate to utter a chal-
lenge, for it does not occur to him to ask advice first, and no one
would listen to unsolicited advice. From their earliest years
children are encouraged by their elders to settle all disputes by
fighting, and they grow up to regard skill in fighting the most
necessary accomplishment and courage the highest virtue.
Boys fight with spiked bracelets. Men of the same village or
camp fight with clubs, for it is a convention that spears must not
be used between close neighbours lest one of them be killed and
the community be split by a blood-feud. It is also a convention
that no third person may take part in the fight, even though he
be a close kinsman of one of the combatants. Once a fight has
begun neither party can give way and they have to continue
till one or the other is badly injured unless, as generally hap-
pens, people pull them away from each other, loudly protesting,
and then stand between them.
When a fight starts between persons of different villages it is
with the spear; every adult male of both communities takes
part in it ; and it cannot be stopped before considerable loss of
152 THE POLITICAL SYSTEM
life has ensued. Nuer know this and, unless they are very angry,
are reluctant to start a fight with a neighbouring village and are
often willing to allow a leopard-skin chief or the elders to inter-
vene. I have seen a fight of this kind prevented by the media-
tion of elders on both sides, but it was clear that their mediation
would have availed little if the young men had been eager to
come to blows. To-day such fights are less common because
fear of Government intervention acts as a deterrent, but I have
seen camps and even tribal sections massed for war and on
the verge of fighting, and at one time fights must have been
very frequent.
Tribes sometimes raided each other for cattle, but fighting
between them was rare. Inter-community fighting and the feuds
that result from it are part of the political relations that exist
between segments of a common tribal organization. Thus a
Leek man said to me: 'We have our fighting among ourselves
and the Gaajok have their fighting among themselves. We do
not fight the Gaajok. We fight only among ourselves. They
have their own fights.' People are killed in these fights and
blood-feuds thereby commence. Within a tribe there is a
method by which they can be concluded by arbitration.
V
We give a concise account of the procedure for settling a
blood-feud, without describing the detail of ritual. As soon as a
man slays another he hastens to the home of a leopard-skin chief
to cleanse himself from the blood he has spilt and to seek
sanctuary from the retaliation he has incurred. He may neither
eat nor drink till the blood of the dead man has been let out of
his body, for it is thought to pass into it in some way, and to this
end the chief makes one or two vertical incisions on his arm by a
downward stroke from the shoulder with a fishing spear. The
slayer presents the chief with a steer, ram, or he-goat, which the
chief sacrifices. This rite and the mark of Cain on the arm are
known as hir. As soon as the kinsmen of the dead man know that
he has been killed they seek to avenge his death on the slayer,
for vengeance is the most binding obligation of paternal kinship
and an epitome of all its obligations. It would be great shame
to the kinsmen were they to make no effort to avenge the homi-
THE POLITICAL SYSTEM 153
cide. By living with the chief as his guest from the moment his
arm has been cut till the final settlement, the slayer has asylum,
for the chief is sacred and blood must not be shed in his home-
stead. It is possible that men only take refuge with a chief when
the danger of vengeance is very great, but it seems to be the
general practice.
While the slayer is at the chief's home the avengers keep
watch on him {bim) from time to time to see if he leaves
his sanctuary and gives them a chance to spear him. They
take any opportunity that offers to kill him, but they are not
very persistent in seeking an opportunity. This state of affairs
may go on for some weeks before the chief opens negotiations
for settlement with the dead man's people, for his overtures are
not likely to meet with a favourable response till the mortuary
ceremony has been held and tempers have cooled a little. The
negotiations are leisurely conducted. The chief first finds out
what cattle the slayer's people (jithunga) possess and that they
are prepared to pay compensation. I do not think that it would
often happen that they would refuse to pay blood-cattle, unless
they lived very far from the avengers or there were a number
of unsettled feuds between the sections concerned, though they
might not have the intention of handing them all over. He then
visits the dead man's people [jiran) and asks them to accept
cattle for the life. They usually refuse, for it is a point of
honour to be obstinate, but their refusal does not mean that they
are unwilling to accept compensation. The chief knows this and
insists on their acceptance, even threatening to curse them if
they do not give way, and his exhortations are supported by the
advice of distant paternal kinsmen and cognatic relatives who
wiU not receive any of the cattle and need not, therefore, show
such pride and stubbornness, but who have a right to make their
opinion known in virtue of their relationship to the dead man.
The voice of compromise is also supported by the bias of custom.
Nevertheless, the_close-ktnsm€n musi refuse to Usten to i± till
the chief has reached the Jimit of his^argumetttSj and when they
give way they declare that they are accepting the cattle only in
order to honour him and not because they are ready to take
cattle for the life of their dead kinsman.
In theory forty to fifty head of cattle are paid, but it is
154 THE POLITICAL SYSTEM
unlikely that they will all be paid at once and the debt may
continue for years. The ceremonies of atonement are performed
when some twenty have been handed over, and then the slayer's
kin may go about without fear of being waylaid, for the time
being at any rate, for they are not safe from vengeance till all the
cattle have been paid, and possibly not even then. The chief
takes the cattle to the dead man's home. The slayer's people
would not venture to accompany him. They are partly dis-
tributed among the kinsmen of the dead man and partly used to
marry a wife to his name to give him heirs. Even if a man on
either side has been killed cattle must be paid by both parties,
though perhaps only twenty head by each, for the ghosts must
be appeased and the honour of the living must be upheld. Also,
sacrifices must be performed to rid the villages of death, which
is loose in them and must be sent into the bush, and kinsmen
on both sides must be purified from their uncleanness. For his
part in the proceedings the chief receives, apart from meat of
sacrifices, two beasts, but he has to give one of these to an
agnatic kinsman who assists him. Often he gains nothing, as he
is expected to give the slayer a cow to help him pay compensa-
tion and, moreover, he has had the expense of providing him
with lengthy hospitality.
A homicide does not concern only the man who has committed
it, but his close agnatic kinsmen also. There is mutual hostiHty
between the kinsmen on both sides and they may not, on
penalty of death, that will inevitably fall on those who break
the injunction, eat or drink with one another or from the same
dishes and vessels the other has eaten or drunk from, even if in
the home of a man who is not kin to either party. This prohibi-
tion ceases after the cattle have been paid and sacrifices made,
but close kinsmen on either side will not eat with one another for
years, even for a generation or two, for reasons of sentiment.
'A bone (the dead man) lies between them.' Indeed, all Nuer
recognize that in spite of payments and sacrifices a feud goes
on for ever, for the dead man's kin never cease 'to have war in
their hearts '. For years after cattle have been paid close agnates
of the slayer avoid close agnates of the dead man, especially
at dances, for in the excitement they engender the mere knock-
ing against a man whose kinsman has been slain may cause a
THE POLITICAL SYSTEM 155
fight to break out, because the offence is never forgiven and the
score must finally be paid with a hfe. When the dead man is
married a wife the bride is rubbed with ashes by her dead hus-
band's kinsmen and God is invoked by them that she may bear
a male child who will avenge his father. Such a child is a gat ter,
a child of feud. At the sacrifices the ghost is told that his kins-
men have accepted cattle and will marry him a wife with them,
but he is assured that one day they will avenge him properly
with the spear, 'A Nuer is proud and wants a man's body in
vengeance and not his cattle. When he has killed a man he has
paid the debt and then his heart is glad.' Hence, though the
chief admonishes the relatives of the dead man at the cere-
monies of settlement that the feud is ended and must not be
reopened, Nuer know that 'a feud never ends'. There may be
peace for a time for the reasons that persuaded the kinsmen to
accept compensation and on account of the cattle they have
received, but enmity continues and the people on both sides
remain jiter, people who are at feud, even if there is no overt
hostility. There is no frequent fighting or continuous unabated
hostility, but the sore rankles and the feud, though formally
concluded, may at any time break out again.
VI
We have said that feuds create a state of hostility between
lineages and thereby, as we shall explain further, between whole
tribal sections ; and that there is no very great difference between
the occasional efforts to exact vengeance when feuds are still
unsettled and the latent hostility that persists when they have
been settled. This is only true, however, when the homicides
are between primary, secondary, and tertiary tribal sections.
In smaller groups it is not so, for, in spite of the strength of the
feelings aroused and their persistence after compensation has
been paid, feuds have there to be settled more quickly and are
not so likely to break out again after settlement.
What happens when a man kills another depends on the
relationship between the persons concerned and on their
structural positions. There are different payments for a true
Nuer, for a Dinka living in Nuerland, and, among the Eastern
Jikany, for a member of the aristocratic clan (see p. 217). The
156 THE POLITICAL SYSTEM
ability to prosecute a feud and thereby to obtain redress by a
life or by payment of cattle depends to some extent on the
strength of a man's lineage and on his kin relationships. But
the intensity of a feud and the difficulty of settling it depend
chiefly on the size of the group concerned in it. If a man kills
another who is closely related to him — his paternal cousin, for
example — cattle are still paid, though fewer, probably about
twenty head. One of the sources of contribution, the father's
brothers, or their sons, would be the recipients of compensation
and therefore cannot pay it. None the less some cattle have
to be paid as it is necessary to compensate the family of the dead
man, to provide the ghost with a wife, and to perform the due
sacrifices. I was told that in such cases the matter is quickly
settled. Probably a blood-feud can be settled more easily if it
is within a clan, for Nuer consider it wrong for clansmen to
engage in a feud. After payment has been made they say : 'The
feud has been cut behind, we have returned to kinship.' It is
also said that if there has been much intermarriage between
two groups a feud is unlikely to develop.
When a man kills a fellow villager or a man of a nearby village
with which his village has close social relations a feud is soon
settled, because the people on both sides have got to mix and
because there are sure to be between them many ties of kinship
and affinity. It is pointed out to the ghost that cattle have been
paid and that it is impossible to avenge him by taking life
because no one would be left alive if the feud were continued
between kinsmen and neighbours. Corporate life is incompat-
ible with a state of feud. When a man spears another of a nearby
village it is customary for the people of the wounder's village to
send the spear which caused the injury to the wounded man's
people so that they can treat it magically to prevent the wound
proving fatal. They also send a sheep for sacrifice. By so doing
they intimate their hope that the wound will soon heal and that,
in any case, they do not want to incur a feud on account of a
personal quarrel. After this courtesy, even if the man dies his
kinsmen are likely to accept compensation without great reluc-
tance. If a man dies many years after having received a wound
the death is attributed to this wound, but compensation is
likely to be accepted without demur, and on a reduced scale.
>2;
PLATE XVII
'tsi'Xf f- 'jjft, :>
'r^^or
_- icafife^
"^«g
/
^hJv''
.:9H^BtoBB^H^K«i^BQ>IKI
Boy collectiii- (liniL^-fiiel (Lou)
THE POLITICAL SYSTEM 157
When a man has killed a neighbour a cow is often immediately
paid over in earnest so that the community may remain at peace.
It must not be supposed, however, that the ease with which
feuds are settled is an indication of lack of strong indignation
or that the difficulty with which they are settled is an indication
of greater indignation.
Feuds are settled with comparative ease in a restricted social
milieu where the structural distance between the participants
is narrow, but they are more difficult to settle as the milieu
expands, until one reaches intertribal relations where no com-
pensation is offered or expected. The degree of social control
over feuds varies with the size of the tribal segment, and Nuer
themselves have often explained this to me. Long and intense
feuds may take place between tertiary tribal sections, but an
effort is generally made to close them, for a segment of this size
has a strong sense of community, close lineage ties, and some
economic interdependence. However, it is far less easy to stop
a feud between persons of different tertiary sections than to stop
a feud in a village or between neighbouring villages, where a
quick and permanent settlement is assured, and unsettled feuds
between sections of this size tend to accumulate. This is
especially the case where there has not been a single death
resulting from a personal quarrel but several deaths in a fight
between the two sections. When a fight has taken place between
secondary tribal sections there is little chance of exacting ven-
geance except by a general fight, and people feel less the need
to submit to mediation since they have few social contacts and
these of a temporary kind, for the relative ease with which
feuds are settled is an indication of the cohesion of the com-
munity. The larger the segment involved the greater the
anarchy that prevails. People say that there is payment of
blood-cattle between primary tribal sections, but they do not
greatly feel the need of paying it. The tribe is the last stage in
this increasing anarchy. It still has nominal political unity,
and it is held that feuds between its most distant members can
be settled by compensation, but often they are not settled, and
if many men are killed in a big fight between large sections
nothing is done to avenge them or to pay compensation for
their deaths. Their kinsmen abide their time till there is another
158 THE POLITICAL SYSTEM
fight. The poHtical integument may in consequence be stretched
eventually to the breaking-point and the tribe split into two.
The rent between sections enlarges until they have little to do
with each other beyond occasional uniting for raids ; and feuds
between their members are settled, if at all, with greater difficulty
and casualness.
VII
The likelihood of a homicide developing into a blood-feud, its
force, and its chances of settlement are thus dependent on the
structural interrelations of the persons concerned. Moreover, the
blood-feud may be viewed as a structural movement between
political segments by which the form of the Nuer political
system, as we know it, is maintained. It is true that only close
agnatic kin of both sides are immediately and directly involved,
but feuds between persons belonging to different tribal sections
sooner or later influence the interrelations of the whole com-
munities to which they belong.
The kinsmen of a slain man try to kill the gwan thunga, the
slayer, but they have a right to kill any of his close agnates
{gaat gwanlen). They must not kill sons of the mother's brother,
father's sister, or mother's sister, because these people do not
belong to the slayer's lineage. Also, only the minimal lineages
on both sides are directly involved in the feud. However, the
significance of the feud may be held to lie less in the ease of
settlement within smaller groups than in the difficulties of
settlement within larger groups, which participate indirectly in
the conflict. We have noted that people at feud may not eat in
the same homesteads, and, as a man eats in all the homesteads of
his viUage, the villagers at once come within the scope of the
prohibition and enter into a state of ritual opposition to each
other. All the people of a village are generally in some way
related to one another and have also a strong sense of com-
munity, so that if any fighting occurs between their village and
another on account of a feud in which some of their members are
involved the whole village is likely to be drawn into it. Thus at
dances men of each village who attend them arrive in war
formation and maintain an unbroken line throughout the dance,
so that if one of them is attacked the others are at his side and
THE POLITICAL SYSTEM 159
are bound to assist him. People not directly affected by the
feud may thus be compelled to assist the principals.
We have noted, moreover, that the intensity of a feud and the
manner of its prosecution depend on the structural relationship
of the persons concerned in the political system. A feud cannot
be tolerated within a village and it is impossible to maintain
one for a long period between nearby villages. Consequently,
although fights most frequently occur within a village or be-
tween neighbouring villages and camps, a feud, in the sense of a
relation between parties between whom there is an unsettled
debt of homicide which can be settled either by vengeance or by
payment of compensation — a temporary state of active hostility
which does not compel immediate settlement but which requires
eventual conclusion— can only persist between tribal sections
which are near enough to each other for the maintenance of
actively hostile relations and far enough from each other for
these relations not to inhibit essential social contacts of a more
peaceable kind. A feud has little significance unless there are
social relations of some kind which can be broken off and
resumed, and, at the same time, these relations necessitate
eventual settlement if there is not to be complete cleavage.
The function of the feud, viewed in this way, is, therefore, to
maintain the structural equilibrium between opposed tribal
segments which are, nevertheless, politically fused in relation to
larger units.
Through the feud whole sections are left in a state of hostility
towards one another, without the hostility leading to frequent
warfare, for the scope of direct vengeance is limited to small kin-
ship groups and their efforts to exact it are not incessant. There
is a fight between two sections and some people are killed on
either side. Only those lineages which have lost a member
are in a state of direct feud with the lineages which they have
deprived of a member, but through common residence, local
patriotism, and a network of kinship ties, the whole sections
participate in the enmity that results, and the prosecution of
the feuds may lead to further fighting between the communities
concerned and to a multiplication of feuds between them. Thus
when the Nyarkwac section of the Lou tribe fought the Leng
section, the lam lineage and the people who live with them were
i6o THE POLITICAL SYSTEM
ranged against the mar, kwoth, and malual lineages and the
people who live with them ; the manthiepni lineage was ranged
against the dumien lineage; and so forth. Only these minimal
lineages were involved with one another in the feuds that
resulted, and not collateral lineages, even though they took part
in the fight in other sectors of it, but the hostility between the
sections was common to all their members. A good example of
how Nuer feel in such matters is furnished by their reactions
at Muot Dit cattle camp when the Government seized hostages
to compel them to hand over two prophets. Their main com-
plaint to me was that the hostages did not belong to the same
lineages as the prophets and were therefore not directly con-
cerned in the issue. The Government was looking at the affair
in territorial terms, they in kinship terms on analogy with the
conventions of a feud.
Apart from ritual observances, kinship obligations, community
sentiment, and so forth, there is a further reason why blood-
feuds between small lineages, especially when there are many of
them, develop into states of feud, and tend to maintain hostile
feelings between communities. As is explained in Chapter V
every community is associated with a lineage in such a way that
all persons in the community who are not members of the lineage
are assimilated to it in political relations which are, therefore,
often expressed in lineage values. Hence a blood-feud between
small agnatic groups is translated into a feud, in the more
general sense, between the lineages with which these groups
are associated through the expression of the disturbed rela-
tions in terms of their structure, and the communities associated
with the lineages are involved in mutual hostility.
j Hostility between smaller segments of a tribe may involve
/ the larger segments of which they form part. A quarrel between
i two villages may thus, as we have noted, bring about a fight
between secondary, and even primary, tribal sections. The
interrelations between larger sections are operated, as it were, by
the interrelations between smaller sections. When a section in
which there are unsettled feuds fights another section all quarrels
are neglected for the time and the whole section combines for
action.
The feud is a political institution, being an approved and
THE POLITICAL SYSTEM i6i
regulated mode of behaviour between communities within a
tribe. The balanced opposition between tribal segments and
their complementary tendencies towards fission and fusion,
which we have seen to be a structural principle, is evident in the
institution of the feud which, on the one hand, gives expression
to the hostility by occasional and violent action that serves to
keep the sections apart, and, on the other hand, by the means
provided for settlement, prevents opposition developing into
complete fission. The tribal constitution requires both elements
of a feud, the need for vengeance and the means of settlement.
The means of settlement is the leopard-skin chief, whose role
we shall examine later. We therefore regard the feud as essential
to the poHtical system as it exists at present. Between tribes
there can only be war, and through war, the memory of war,
and the potentiality of war the relations between tribes are
defined and expressed. Within a tribe fighting always produces
feuds, and a relation of feud is characteristic of tribal segments
and gives to the tribal structure a movement of expansion and
contraction.
There is, of course, no clear-cut distinction between fighting
against another tribe and fighting against a segment of one's
own tribe. Nuer, however, stress that the possibility of arbitra-
tion and payment of blood-wealth for deaths resulting from
a fight within a tribe makes it a ter, feud, and that this is different
from a fight between tribes, Mr, in which no claims for com-
pensation would be recognized. Both differ from raiding of
Dinka, pec, and from individual duelling, dwac, though all
fighting is kur in a general sense. But it is obvious that a fight
in a village which at once leads to payment of compensation for
deaths and a fight between tribes in which there is no compensa-
tion for deaths are two poles, and that the further we go from
a village community the fights between tribal sections become
more like fights between tribes in that blood-wealth is paid with
increasing difficulty and infrequency, so that between primary
sections the tribal value, the feeling that blood- wealth can, and
even ought to, be paid, alone distinguishes their fights from
fights between tribes. Here again we would emphasize the
conclusion that the tribal value is relative to the structural
situation,
4507 Y
i62 THE POLITICAL SYSTEM
We would emphasize further that blood-feuds only directly
involve a few persons and that though they sometimes cause
violence between whole local communities — a feud in a wider
sense — ordinary social contacts continue in spite of them. The
strands of kinship and affinity, of age-set affiliations, and of
military and even economic interests remain unbroken ; and
these strands act like elastic between the sections, being capable
of considerable expansion by disturbed political relations, but
always pulling the communities together and keeping them as
a single group in relation to other groups of the same kind. As we
have explained, these strands lessen in number and strength the
larger the community, but they stretch even beyond tribal
frontiers. Increasing anarchy, increasing difficulty in settling
feuds, goes together with lessening frequency of social contacts
of all kinds. Social cohesion increases as the size of the com-
munity narrows.
VIII
There are, of course, disputes between Nuer other than
those about homicides, but they can be treated briefly and in
direct relation to homicide and the feud. In a strict sense
"Nuer have no law. There are conventional compensations for
dajnage,_a4ulterY2 _loss of limb, ^an d so forth, but there is no
) authority with power to_adjudic ate on such matters or to_en-
I force a verdict. In Nuerland legislative, judicial, and executive
functions are not invested in any persons or councils. Between
members of different tribes there is no question of redress ; and
even within a tribe, in my experience, wrongs are not brought
forward in what we would call a legal form, though compensa-
tion for damage [ruok) is sometimes paid. A man who considers
that he has suffered loss at the hands of another cannot sue
the man who has caused it, because there is no court in which to
cite him, even were he willing to attend it. I lived in intimacy
for a year with Nuer and never heard a case brought before an
individual or a tribunal of any kind, and, furthermore, I reached
the conclusion that it is very rare for a man to obtain redress
except by force or threat of force. The recent introduction of
Government courts, before which disputes are now sometimes
THE POLITICAL SYSTEM 163
settled, in no way weakens this impression, because one well
knows how among other African peoples cases are brought before
courts under Government supervision which would not pre-
viously have been settled in a court, or even settled at all, and
how for a long time after the institution of such Government
tribunals they operate side by side with the old methods of
justice.
Before discussing the chief characteristics of Nuer legal pro-
cedure I record that, according to verbal information, for I have
never observed the procedure, one way of settling disputes is to
use a leopard-skin chief as mediator. Thus, I was told that a man
who has a cow stolen may ask a leopard-skin chief to go with him
to request the return of the cow. The chief goes first, with several
of the elders of his village, to the plaintiff's homestead, where he
is given beer to drink. Later they go, with a deputation from the
plaintiff's village, to the defendant's village, and here also the chief
may be presented with some beer or a goat. The chief is considered
to be neutral and a certain sanctity attaches to his person so that
there is little likelihood of the deputation being injured. The
visiting elders sit with elders of the defendant's village and the
chief in one of the byres and talk about the matter in dispute. The
owner of the animal gives his view and the man who has stolen
it attempts to justify his action. Then the chief, and anybody
else who wishes to do so, expresses an opinion on the question.
When every one has had his say the chief and elders withdraw to
discuss the matter among themselves and to agree upon the
decision. The disputants accept the verdict of the chief and the
elders and, later, the owner of the animal gives the chief a young
steer or a ram unless he is a very poor man, when he gives him
nothing.
If a man has a dispute with another of the same neighbourhood
they may both go to the homestead of a local leopard-skin chief
and lay their spears on the ground in his byre. A man would not
plant his spear upright in a chief's byre, and I was told that were
he to do so a bystander might appropriate it, as disrespect had
thereby been shown to the chief. When both men have stated
their views the chief and the elders discuss the matter outside the
byre and re-enter it to acquaint the disputants of their decision.
The person in whose favour the decision has been given hands his
spear to the chief, who either gives it to a friend or spits on it and
returns it to its owner. It was clear from the way in which my
informants described the whole procedure that the chief gave his
final decision as an opinion couched in persuasive language and
i64 THE POLITICAL SYSTEM
not as a judgement delivered with authority. Moreover, whilst
the sacredness of the chief and the influence of the elders carry
weight, the verdict is only accepted because both parties agree to
it. No discussion can be held unless both parties want the dispute
settled and are prepared to compromise and submit to arbitration,
the chief's role being that of mediator between persons who wish
other people to get them out of a difficulty which may lead to
violence. The man against whom the decision is pronounced may
give way to honour the elders and the chief where he would not
give way directly and without their intervention, for he does not
lose prestige by accepting their verdict. If there is any doubt about
the facts, certain oaths, which are in the nature of ordeals, may
be employed, such as swearing statements on the chief's leopard
skin.
For a dispute to be settled in this way not only is it necessary
that both parties should want the matter amicably settled, but
it is also necessary that they should themselves reach agreement
during the discussion. No one can compel either party to accept
a decision, and, indeed, a decision cannot be reached unless there
is unanimity, since the elders are of both parties to the dispute.
They go on talking, therefore, till every one has had his say and
a consensus has been reached.
The five important elements in a settlement of this kind by
direct negotiation through a chief seem to be (i) the desire of the
disputants to settle their dispute, (2) the sanctity of the chief's
person and his traditional role of mediator, (3) full and free dis-
cussion leading to a high measure of agreement between all
present, (4) the feeling that a man can give way to the chief and
elders without loss of dignity where he would not have given
way to his opponent, and (5) recognition by the losing party of
the justice of the other side's case.
I repeat that I have not seen this method employed and add
that I am of the opinion that it is very rarely used and only
when the parties are fairly close neighbours and belong to com-
munities closely linked by many social ties. In theory any
member of a tribe can obtain redress from any other, but we
know of no evidence that leads us to suppose that it was often
obtained. Before summing up what we judge to be the nature
and scope of legal relations in Nuerland we record a few brief
examples of typical acts likely to lead to violence if some
reparation is not made.
THE POLITICAL SYSTEM 165
When a Nuer speaks of a person having stolen (kwal) an animal,
he means that he has taken it without permission and by stealth
and by no means implies that he ought not to have taken it.
Within a tribe an abductor of cattle always considers that he is
taking what is owing to him. It is a debt (ngwal) which he is
settling in this way, because the man who owes him cattle has not
repaid them of his own accord. The legal issue, therefore, is
whether he is right in assuming a debt and whether he should
have taken the particular animals he took. So well is this practice
of helping yourself to what is owing to you an established habit,
that it may be said to be a customary way of settling debts. Thus
the final cattle of a homicide-payment are often seized in the
pastures, and it often happens that when a bridegroom and his
kin do not hand over all the cattle they have promised, the wife's
brothers try to seize the animals still owing to them. In other
circumstances a man will steal a cow which he is owed, sometimes
employing the services of a magician to charm the owner so that
he may not guard his herd on the day the theft is planned; for
example a man who lent another an ox for sacrifice in sickness, at
his daughter's wedding, in time of famine, and so on, and has not
received a heifer in return, although the debtor possesses one.
Having seized a cow from the debtor's herd he is quite willing to
return it if he receives instead the heifer owing to him. The debtor
will then either try to steal his cow back again or he will open
a discussion which will result in his paying a heifer and receiving
back his cow.
The only quarrels within a village or camp about ownership of
cattle that I have witnessed have concerned obligations of kinship
or affinity and have eventually been settled by one party giving
way on account of his relationship to the other. If a man seizes
cattle from a kinsman or neighbour he walks into his kraal and
takes them. If the owner has a strong case he may resist : otherwise
he lets the cattle go, for he knows that the man will be supported
by public opinion in the community. If a man seizes cattle from
a man of a different village he adopts different tactics. With one
or two friends he watches the cattle in their grazing-ground till
a suitable opportunity occurs. I have never heard of a Nuer
stealing a cow from a fellow tribesman merely because he wanted
one. He has, on the other hand, no hesitation in stealing cows
from persons belonging to neighbouring tribes and will even go
with friends to another tribe in order to steal from them. This
theft {kwal) is not considered in any way wrong.
If a man commits adultery he pays an indemnity of five cows
and an ox, unless the man is impotent, when the adulterer can
i66 THE POLITICAL SYSTEM
claim a cow on the marriage of a female child of his adultery.
Even when the husband is not impotent, if the adulterer can show
that there is fruit of his adultery he can claim back the cattle he
has paid as an indemnity, except for a cow known as yang kule,
the cow of the sleeping-skin, which has ritual significance. But
adultery within a small local community is probably rare because
the people are all related to one another, and therefore a man con-
siders that not only would it be wrong to commit adultery with
their wives but also that it would be, in a greater or lesser degree,
incestuous. If the two men are close kinsmen the adulterer pro-
vides an ox for sacrifice but is unlikely to pay an indemnity. If
they are not close kinsmen the husband may try to seize the
adulterer's cattle, but he only takes this step if he actually catches
the offender committing adultery. The adulterer, to avoid a fight,
runs away, and if he fears that his cattle may be seized he places
them in the kraals of friends and kinsmen. This makes it difficult
for the husband to seize them, for even if he knows where they are
he does not want to involve himself in quarrels with a number of
his neighbours by raiding their kraals. Nuer do not consider it
immoral to commit adultery with the wives of persons living in
other villages. If the husband discovers the offence he may try to
seize the offender's cattle, but in doing so he runs the risk of resist-
ance and some one may get killed and a feud result. A herd is the
joint property of brothers, even though it is divided among them,
and they do not readily acquiesce in the loss of cattle on account
of adultery. In my experience it is very seldom indeed that a man
obtains compensation for adultery. Adultery with the wife of a
man of another tribe is a matter of no importance. What can he
do anyhow ?
Likewise fornication with an unmarried girl is compensated by
payment of a heifer and a steer. But it is most unlikely that the
payment will be made. If the girl's kinsmen know that she is
having relations with a man who possesses cattle and is likely to
marry her they turn a blind eye to the affair. If he has no cattle,
or if the girl is already engaged, and one of her brothers catches
him in congress with her he fights him, unless, as usually happens,
the man runs away, for it is not considered cowardly to run away
in these circumstances. The girl's kinsmen may then come to his
kraal and, if he possesses them, take a male calf and a female calf,
and if they are strong enough they may not be resisted. This is
what Nuer tell one, but I have never known any one pay a male
and a female calf, although after every dance I have seen the youths
and maidens have paired off and there has been much promiscuous
intercourse and very little effort to conceal it. It frequently
PLATE XVIII
m
Building a cattle byre (Eastern Jikany)
THE POLITICAL SYSTEM 167
happens that a man impregnates an unmarried girl and he is then
expected to marry her. The girl's kinsmen may raid his kraal and
seize some of his cattle, but he will try to circumvent this by
hiding his beasts in the kraals of relatives and neighbours. If
later he marries the girl the cattle her kinsmen have seized
count as part of the bride-wealth, and if he refuses to marry her
they count as payment for possession of the child, so that in cither
case he pays no indemnity but only a fee to establish his rights.
Here again, in point of fact, it is very difficult for the girl's
brothers to seize the man's cattle, unless he is willing to let them
do so, and there is always the risk of a fight which may become
general. One does not fornicate with the girls of one's own village,
for they are generally related to one, so that when an issue of the
kind arises it is generally between persons of different villages of
the same district. If the young man can avoid being hit over the
head with a club at the time of the incident and keeps clear of the
girl's village for a few months he is not likely to pay an indemnity
or to suffer any other consequences. If he has put the girl in child
he will, in normal circumstances, send a kinsman to say that he
intends to marry her. The girl is then regarded as engaged and the
young man becomes son-in-law to her parents and one does not
injure one's son-in-law. Even if he refuses to marry their sister,
the brothers hesitate to attack the father of her child.
It is possible to obtain from Nuer a list of compensations for
injuries to the person: e.g. ten cattle for a broken leg or skull, ten
cattle for the loss of an eye, two cattle for a girl's broken teeth, &c.
For a flesh wound, however serious, there is no compensation unless
the man dies. In different parts of Nuerland the number of cattle
to be paid in compensation varies. I have not recorded any
instance of a man receiving such compensation, except from a
Government court, but Nuer say that he would receive it if his
kinsmen were strong enough to retaliate.
It is alleged that, in the old days, if a man died of magic his
kinsmen would try to slay the magician {gwan wal), though I have
not recorded a case of a magician having been killed. Nuer point
out that a magician does not use his magic against persons in his
own community but only against persons of other villages, so that
it is not easy to revenge yourself on him since he will be supported
by his village who regard powerful magic as of value to their
community. It is also said that in the old days a witch {peth) was
sometimes slain, though I cannot say how often, if at all, this
occurred.
Many disputes arise^about bride-wealth : the husband's people,
do not pay what they have promised or there is a divorce and the
i68 THE POLITICAL SYSTEM
wife's people do not return all the cattle that have been paid. In
such circumstances the debtor does not deny the debt but puts
forward a counterbalancing claim on some account or says that
he has no cattle to settle it. Often enough he says this even when
he has got cattle. The claimant can only be sure of getting his due
if he takes it by force from the debtor's kraal or from the herd in the
grazing-grounds. If he is strong and has the backing of a powerful
lineage he is not resisted, since he has right on his side. Such matters
are easily settled within a village and among people who share a
common dry season camp because every one realizes that some
agreement must be reached by discussion and that it must accord
with justice. But when the parties belong to different, and maybe
hostile, villages, settlement is not so easy. The leopard-skin chief
may be employed, in the manner described, to bring the parties
together for discussion and some agreement may then be reached,
but many such debts are never settled. They are remembered for
years. Perhaps some day, maybe in the next generation, an
opportunity will occur to steal the cattle.
If a wife dies in her first pregnancy or childbirth the husband
is held responsible. There is no question of a feud arising, but the
husband loses the bride-cattle he has paid, since these now become
blood-cattle for the loss of his wife. The husband is only responsible
if the death occurs during childbirth before expulsion of the
afterbirth If there is any dispute about the mode of death or the
number of cattle that are still owing it is settled by a mediator called
kuaa yiika or kuaa yiini, 'the chief of the mats', an office which
pertains to certain lineages. This man has no other office and is not
an important person in virtue of his role of arbitrator in disputes
of this kind. It is easy to obtain compensation, as the father-in-
law has the bride-wealth in his possession. There is, moreover,
a tie of affinity and neither party is likely to resort to violence.
Using these brief notes on some exemplifications of the ten-
dencies in Nuer law we may now state what these tendencies
are. We speak of 'law' here in the sense which seems most
appropriate when writing of the Nuer, a moral obligation to
settle disputes by conventional methods, and not in the sense
of legal procedure or of legal institutions. We speak only of civil
law, for there do not seem to be any actions considered injurious
to the whole community and punished by it. The informants
who said that witches and magicians were sometimes killed
stated that it was always individuals or groups of kin who way-
laid them and killed them in vengeance.
THE POLITICAL SYSTEM 169
The first point to note about Nuer law is that it has not
everywhere the same force within a tribe, but is relative to the
( position of persons in social structure, to the distance between
them in the kinship, lineage, age-set, and, above all, in the
political, systems. In theory one can obtain redress from any
member of one's tribe, but, in fact, there is little chance of doing
so when he is not a member of one's district and a kinsman. The
wider the area which contains the parties to a dispute the
weaker the feeling of obligation to settle it and the more difficult
the task of enforcing settlement, and, consequently, the less the
likelihood of it being settled. Within a village differences be-
tween persons are discussed by the elders of the village and
agreement is generally and easily reached and compensation
paid, or promised, for all are related by kinship and common
interests. Disputes between members of nearby villages, be-
tween which there are many social contacts and ties, can also
be settled by agreement, but less easily and with more likelihood
of resort to force. The nearer we get to the tribe the less the
chances of settlement. Law operates weakly outside a very
limited radius and nowhere very effectively. The lack of social
control to which we have often referred is thus shown in the
weakness of law, and the structural interrelations of tribal
segments are seen in the relativity of law, for Nuer law is rela-
tive like the structure itself.
A cogent reason why there is little chance of redress between
members of different secondary and primary tribal sections is
that the basis of law is force. We must not be mislead by an
enumeration of traditional payments for damage into supposing
that it is easy to exact them unless a man is prepared to use
force. The club and the spear are the sanctions of rights. What
chiefly makes people pay compensation is fear that the injured
man and his kin may take to violence. It follows that a member
of a strong lineage is in a different position from that of a member
of a weak lineage. Also the chances of a man obtaining redress
for an injury are less the further removed he is from the man
who has injured him, since the opportunity for violence and the
effectiveness of kinship backing lessen the wider the distance
between the principals. Since self-help, with some backing of
public opinion, is the main sanction it is only operative when
4S07 z
170 THE POLITICAL SYSTEM
people are within easy striking distance of one another. This is
one of the main reasons why it is difficult to settle feuds when
the parties belong to different primary or secondary tribal
sections.
Most disputes occur in villages or camps and between persons
of nearby villages, since people who live close to one another
have more opportunities for dispute than those who live far from
Fig. 12. Baked-mud grindstone with wooden grinder.
one another. Such quarrels are usually complicated by notions
of kinship, affinity, age, and so forth and are very often the
result of breaches of specific patterns of social behaviour rather
than simple breaches of general social regulations. They are,
therefore, generally settled in harmony with these traditional
patterns. But, if they are not settled by the mediation of kins-
men, they are likely to lead to violence, because, as we have
remarked, Nuer are at once prepared to fight if they are wronged
or insulted, unless kinship, or great disparity in age, restrains
them. If a man refuses to pay compensation for a wrong, there-
THE POLITICAL SYSTEM 171
fore, he runs a great risk of his skuU being cracked with a club,
and even of being speared if excitement runs high. And this is
what often happens.
For this reason we have said that Nuer law, in so far as it
applies to a study of political relations, must be treated in
connexion with the blood-feud. Disputes can often be settled
on account of close kinship and other social ties, but between
tribesmen as such they are either settled by the aggrieved
party using force, and this may result in homicide and blood-
feud, or by the debtor giving way in the knowledge that force
may be used and a feud result. It is the knowledge that a Nuer
is brave and will stand up against aggression and enforce his
rights by club and spear that ensures respect for person and
property.
The Nuer has a keen sense of personal dignity and rights.
The notion of right, cuong, is strong. It is recognized that a man
ought to obtain redress for certain wrongs. This is not a con-
tradiction of the statement that threat of violence is the main
sanction for payment of compensation, but is in accord with it,
for a man's kinsmen wiU only support him if he is right. It is
doubtless true that if a man is weak it is unlikely that his being
in the right will enable him to obtain satisfaction, but if he is in
the right he will have the support of his kin and his opponent
will not, and to resort to violence or to meet it the support of
one's kin and the approval of one's community are necessary.
One may say that if a man has right on his side and, in virtue
of that, the support of his kinsmen and they are prepared to use
force, he stands a good chance of obtaining what is due to him,
so long as the parties to the dispute live near one another.
When we speak of a man being in the right we do not suggest
that disputes are mostly a clear issue between right and wrong.
Indeed, it would be correct to say that, usually, both parties are
to some extent right and that the only question which arises
is. Who has the greater right ? To state the matter in a different
way : a Nuer dispute is usually a balance of wrongs, for a man
does not, except in sexual matters, wantonly commit an act of
aggression. He does not steal a man's cow, club him, or with-
hold his bride-cattle in divorce, unless he has some score to
settle. Consequently it is very rare for a man to deny the
172 THE POLITICAL SYSTEM
damage he has caused. He seeks to justify it, so that a settle-
ment is an adjustment between rival claims. I have been told
by an officer with wide experience of Africans that Nuer defen-
dants are remarkable in that they very seldom lie in cases
brought before Government tribunals. They have no need to,
since they are only anxious to justify the damage they have
caused by showing that it is retaliation for damage the
plaintiff has inflicted earlier.
IX
Feuds are settled through the leopard-skin chief and he plays
a minor role in the settlement of disputes other than homicide.
It might be supposed that this functionary has a position of
great authority, but this is not so. Indeed, on the same grounds
as we have said that the Nuer have no law we might say that
they also have no government. We devote a few lines to stating
what are the ritual qualifications of the leopard-skin chief and
then assess the part he plays in feuds and quarrels.
The few references in the writings of early travellers about
Nuer leaders do not suggest that they were persons of very great
authority. J The absence of any persons with sufficient authority
or, except for a few prophets, with sufficient influence through
whom an administrative system might be built up is stated in
very direct terms by the earliest British officers to enter Nuer-
land.2 The 'sheikhs ' described in these early reports as lacking
authority are probably the persons who were later known
among Europeans as leopard-skin chiefs. A leopard-skin chief,
kuaar muon, has a sacred association with the earth [mun) which
gives him certain ritual powers in relation to it, including the
power to bless or curse. However, lest it be supposed that the
power to utter curses enables a chief to wield great authority I
at once record that I have never observed a chief exercise this
power. There are stories which recount the dire effects of a curse,
but I think that as a rule a chief only ventures to threaten to
utter one when he is officiating in a ritual capacity in the settle-
ment of feuds, on which occasions he is expected to do so, the
' Werne, op. cit., p. 207 ; Poncet, op. cit., p. 40; Brun-Rollet, op. cit., p. 222.
Brun-Rollet's account is unacceptable.
2 Kaimakam G. Hawkes, S.I.R., no. 98, 1902 ; Bimbashi H. Gordon, S.I.R.,
no. 107, 1903.
THE POLITICAL SYSTEM 173
threat forming part of the proceedings. Certainly to-day chiefs
have no authority in virtue of their power of cursing. He is also
known as kuaar twac because he alone wears a leopard-skin
{twac) across his shoulders. A chief may be seen wearing one
in Mr. Corfield's photograph on Plate XXIV. The word kuaar
has ritual associations in all the Nilotic languages, but, without
discussing further what word might best define its scope of
reference in Nuer, we intend to refer to the person in this
account, as we have done heretofore, as a chief, with the caution
that we do not thereby imply that he has any secular authority,
for we hold that his public acts are mainly ritual.
Nevertheless, his function is political, for relations between
political groups are regulated through him, though he is not
a political authority controlling them. His activities are
chiefly concerned with settlement of blood-feuds, for a feud
cannot be settled without his intervention, and his political
significance lies in this fact. Chiefs sometimes prevent fights
between communities by running between the two lines of com-
batants and hoeing up the earth here and there. The older men
then try to restrain the youths and obtain a settlement of the
dispute by discussion. However, we think that a fight can only
be prevented in this way when the disputants are close neigh-
bours and anxious not to kill one another.
Besides the part they play in feuds, chiefs perform ritual to
cleanse parties to incestuous congress, and they possess slight
rain-making powers, though Nuer do not attach much im-
portance to the art. On the whole we may say that Nuer chiefs
are sacred persons, but that their sacredness gives them no
general authority outside specific social situations. I have never
seen Nuer treat a chief with more respect than they treat other
people or speak of them as persons of much importance. They
regard them as agents through whom disputes of a certain kind
can be settled and defilement of a certain kind can be effaced,
and I have often heard remarks such as this: 'We took hold of
them and gave them leopard skins and made them our chiefs to
do the talking at sacrifices for homicide.' Their ritual sphere
seldom extends beyond a tribal section.
Only certain lineages are chiefs and only some men of these
lineages practise as such. It is, perhaps, significant that in many
174 THE POLITICAL SYSTEM
parts of Nuerland, including most of the area with which I am
acquainted, chiefs do not belong to the dominant clans in the
tribes in which they function, though some of them are said to
be aristocrats in parts of Eastern Gaajak, Gaawar, and Leek.
Most of those whose clans I know belong to the gaatleak and
JIMEM clans which nowhere have aristocratic status. Since
quarrels between tribal sections are expressed in terms of line-
ages of the dominant clan associated with the sections, as will
be explained in the following chapter, the chief, having no posi-
tion in the system of dominant lineages, is thereby more fitted
to mediate between them. He is not a member of the hereditary
owners of the tribal land, but is a stranger living there. A chief
may act as such in whatever tribe he resides in. If a chief is
killed the ceremonies connected with payment of compensation
are performed by an aristocrat of the tribe. This is probably
because, even when the chiefs of an area are not all members
of one clan, they are believed to have a kind of kinship through
their common badge of the leopard skin and cannot marry into
one another's families. We regard chiefs as a category of ritual
experts and do not consider that they comprise in any way a
class or rank. We believe their social function to be a mechanism
by which the equilibrium of the political system is maintained
through the institution of the feud. The slight authority of
chiefs and, in many parts, their position outside the dominant
clan, accord with this view.
In taking the view that to regard the leopard-skin chief as a
political agent or a judicial authority is to misunderstand the
constitution of Nuer society and to be blind to its fundamental
principles, we have to account for the part he plays in the settle-
ment of feuds. We have stated that he has no judicial or execu-
tive authority. It is not his duty to decide on the merits of a case
of homicide. It would never occur to Nuer that a judgement
of any kind was required. Likewise he has no means of com-
pelling people to pay or to accept blood-cattle. He has no
powerful kinsmen or the backing of a populous community to
support him. He is simply a mediator in a specific social situa-
tion and his mediation is only successful because community
ties are acknowledged by both parties and because they wish
to avoid, for the time being at any rate, further hostilities.
THE POLITICAL SYSTEM 175
Only if both parties want the affair settled can the chief inter-
vene successfully. He is the machinery which enables groups
to bring about a normal state of affairs when they desire to
achieve this end.
It is true that a leopard-skin chief has always in these circum-
stances to persuade, by exhortations and threats, the kinsmen
of the dead man to accept compensation, but this pressure must
not be regarded as a command. It is quite clear from many
Nuer statements on the matter that the chief's threats are
encouraged to their furthest point in order that by yielding to
his persuasion the kinsmen of the dead man may not dishonour
themselves by neglecting to exact a life for the life of their
kinsman.
A chief's threats might go no further than saying that if the
kinsmen would not listen to him neither would he, when they
were in a similar difficulty, listen to them. But I was told that were
the people to refuse mediation with undue obstinacy the chief
might threaten to leave their homestead and to curse them. He
would take an ox and rub its back with ashes and begin to address
it, saying that if the injured party were to insist on revenge many
of them would be killed in the endeavour and they would throw
spears vainly at their enemies. I was told that he would then
raise his spear to slaughter the animal, but this would be as far
as people would care to let him go. Having asserted their
pride of kin, one of the dead man's family would seize his upraised
arm to prevent him from stabbing the ox, saying: 'No! Do not
kill your ox. It is finished. We will accept compensation.' My
informant, whose statements were supported by others, further
added that if people were to insist on refusing the mediation of a
leopard-skin chief he would take a short-homed ox and, after
invoking God, slay it and rub the hairs off its head so that the
members of the lineage who rejected his mediation might perish
in prosecuting their feud.
We conclude, therefore, that a chief's curse is not in itself the
real sanction of settlement, but is a conventional, ritual, opera-
tion in the settlement of feuds, which is known to every one in
advance and is allowed for in their calculations. The threat of
it is compelled by those on whom it would fall were it uttered.
These affairs are like a game in which everybody knows the
rules and the stages of development: when one is expected to
give way, when to be firm, when to yield at the last moment.
176 THE POLITICAL SYSTEM
and so forth. This conclusion is based on many statements (I
have only once been present at discussions between a chief and
the kinsmen of a slain man, and then the circumstances were
unusual). It may, however, be said with certainty that no
amount of pressure from a leopard-skin chief, if it is ever
exerted, can settle feuds expeditiously, if at all, between the
larger tribal sections. In other disputes the chief acts rarely, and
only when both sides strongly desire settlement. He has no
jurisdiction to hear cases in a locality. Here, again, I was told
that if one party refused to accept his decision as arbiter he
might pass the man his leopard-skin, an action tantamount to
a curse. The man must then make the chief a gift before he will
consent to take back the skin. However, this probably happens
only when a man refuses to accept a decision to which every one
else, including his own elders, have agreed. I was also told that
in disputing a chief's words a man will do so respectfully, first
spitting into the chief's hands as a sign of goodwill. No doubt
a chief is shown respect on such occasions, but the chiefs I have
seen were treated in everyday life like other men and there is no
means of telling that a man is a chief by observing people's
behaviour to him. His role in disputes may be regarded as a
means by which neighbours, who wish a difficulty settled with-
out resort to force and who acknowledge that the other side
have a good case, can negotiate.
X
We have considered the position of the leopard-skin chief at
some length because it is structurally important. He in no way
represents or symbolizes the unity and exclusiveness of political
groups, but is a mechanism by which, through the institution
of the feud, these groups interact and maintain their structural
distance. There are other people in Nuerland with ritual powers,
of one kind or another, which sometimes make a man well known
and, occasionally, very influential, but none of them are politic-
ally important, except the prophets whose activities we discuss
later. They neither govern nor judge and their sacred functions
are not, like those of the leopard-skin chief, specifically related
to the interaction of local groups. However, we do not entirely
PLATE XIX
a. Cattle camp (Leek)
b. Typical swampy depression in No\-embcr (Western Jikany)
THE POLITICAL SYSTEM 177
pass them by because sacred powers often give a man prestige
from which he may attain local eminence as an important elder,
if they are combined with wealth, ability, and wide kinship con-
nexions.
Next to prophets and leopard-skin chiefs the ritual status that
carries with it most prestige is that of the wut ghok, the Man of the
Cattle. Certain lineages have hereditary ritual powers in rela-
tion to cattle and are asked to cure sick beasts and to make barren
cows fruitful, though only some members of these lineages use
their powers. Like leopard-skin chiefs, Men of the Cattle are often
members of stranger lineages and not of the aristocratic clan of
their tribe. I was told that their curse is feared, as it can be
directed to the cattle, and that Nuer do not care to offend them,
but, outside tradition, I have not recorded any occasion of its
utterance. Apart from the few Men of the Cattle who play a part
in the regulation of Age-sets (see Chap. VI), and those who are
sometimes consulted about migration to new pastures, they have
no public functions. One wut ghok of the Eastern Gaajok became
very rich and powerful a generation ago, but his prestige was
largely due to his possession of magic.
Besides the kuaa muon, who has a ritual relationship to the earth,
and the wut ghok, who has a ritual relationship to the cattle, there
are a number of totemic specialists whose ritual connexion with
lions, crocodiles, weaver birds, &c., enables them to influence the
behaviour of these creatures. A totemic specialist is a possessor
(gwan) of the spirit {kwoth) of his totem. Totemic specialists have
no political significance and have no social influence in virtue of
their powers alone. There is a war specialist whose duty it is to
shake a spear in the face of the enemy and to deliver an invocation
against them. He is known as gwan muot, possessor of the spear,
or ngul, and is often, perhaps always, a member of a senior
lineage of the dominant clan of the tribe, for he invokes the spear
by the spear-name of the clan. There are also magicians of various
kinds : leeches, diviners, owners of medicines, and owners of fetishes.
Of these specialists only owners of fetishes become prominent mem-
bers of their communities on account of their ritual powers. Nuer
are very much afraid of fetish spirits and believe therh to be so
powerful that they will even purchase them with cattle. A fetish-
owner may become the most influential man in the village, and I have
been surprised at the respect and fear with which his neighbours
sometimes treat him. Nevertheless, he has no defined authority in
controlling the relations of the villagers to one another nor does he
represent the village in its relations with neighbouring communities.
«" A a
178 THE POLITICAL SYSTEM
XI
Ritual status gives a man vague influence in his locality ; but
authority only in specific ritual situations. Sex and age are
two, more general, conditioning attributes of local influence.
Women and children have always a position inferior to that of
men. Women occasionally gain a reputation as prophets and
magicians but, as a rule, they play no leading part in public
affairs. Among the Nuer, relations between the sexes, and
between man and wife, are more equitable and give females
more privilege than in any other tribe I have visited in
the Southern Sudan. Nevertheless, they are subject to men:
daughters to their fathers and wives to their husbands. Boys
are under the command of their parents and elder brothers and
only become full tribesmen, with the privileges and responsi-
bilities of such, at initiation. The relations between the sexes
and between children and adults belong rather to an account of
domestic relations than to a study of political institutions.
When a lad has passed through initiation he becomes 'a man '
and when he has married and begotten several children he
becomes 'a true man', what we have called an elder. We have
from time to time spoken of the part played by elders in
homicides and other disputes. When a local community acts
corporately and leadership and advice are required, these
functions rest with the elders. They decide when seasonal moves
are to be made and where camps are to be formed, negotiate
marriages, advise on questions of exogamy, perform sacrifices,
and so forth. Their opinions on such matters are readily ac-
cepted by the younger men, who take little part in the talk
unless they are directly concerned in the issue. When the elders
disagree there is much shouting and argument, for every one
who wishes to speak does so and as often and loud as he pleases.
The words of some elders count for more than the words of
others and one easily observes that their opinions are usually
agreed to.
These elders are members of central age-sets, at the present
time the Maker and Dangunga, for members of the most senior
sets, the Thut and Boiloc, take little part in public life. We
discuss in Chapter VI the relations between age-sets. Here we
THE POLITICAL SYSTEM 179
remark only that there is no constituted authority within each
set, all members being of equal status, and that, whilst members
of junior sets respect those of senior sets, the authority of older
men is personal, very indefinite, and based on analogy with
domestic relations within the family. Behaviour between indi-
viduals is influenced by the distance between them in the age-
set system, but age-sets are not a political institution in the
sense of the system having an administrative, military, or
judicial organization.
Age by itself does not give a man social position. He must
have other qualifications as well. Those elders with most
influence are the gaat twot, the children of bulls. Such a man is
called a tut, bull, and in strict usage this is equivalent to dil,
tribal aristocrat. As will be fully explained in Chapter V a dil
is a member of the dominant clan in each tribe and, in virtue
of his membership, has within that tribe a slightly superior
social position. This clan is not a ruling class and the enhanced
prestige of its members is very indefinite. The clan system has
no hereditary leadership ; a senior lineage does not rank higher
than others; there is no 'father of the clan'; and there is no
'council of clan elders '. Tut is also used in a wider sense to refer
to men of social position who do not belong to the dominant
clan but to other lineages which have long been domiciled in the
tribe. A tut in this rather wide sense of 'man of good standing',
or 'social leader', is usually a scion of an important lineage, the
head of his own family, and master of his homestead and herd.
He is generally also the eldest surviving son of his father's
family and, therefore, head also of the joint family, the master
of the hamlet. If he is to gain a social reputation he must also
possess sufficient cows to be able to entertain guests and to
attract young kinsmen to reside in his byre. Round such a man's
homestead are clustered the homesteads of his brothers and
married sons and, often enough, the homesteads of his sisters'
husbands and daughters' husbands. To be a social leader,
whose opinion is readily agreed to, he must also be a man of
character and ability.
The authority of a gat twot, or tut wee, 'bull of the camp', as
he is often called, is never formalized. He has no defined status,
powers, or sphere of leadership. Lineage, age, seniority in the
i8o THE POLITICAL SYSTEM
family, many children, marriage alliances, wealth in cattle,
prowess as a warrior, oratorical skill, character, and often ritual
powers of some kind, all combine in producing an outstanding
social personality who is regarded as head of a joint family and
of a cluster of cognatic kinsmen and affines, as a leader in village
and camp, and a person of importance in the rather vague social
sphere we call a district. It is easy to see in village or camp who
are its social leaders, and it is these people who have furnished the
administration with most government chiefs, for the influence
of the leopard-skin chief is mainly restricted to the circuit of his
ritual functions and only if he is also a gat twot does he exercise
influence beyond these limits.
When we ask, however, in what way a tut acts as a leader in
his community it is difficult to answer. As the chief man of his
family and joint family he takes the most prominent part in
settling the affairs of these groups, but he cannot on that ac-
count be said to have political authority, for these domestic
groups act independently of others in the village, though some
co-ordination between them is imposed by their common require-
ments. A joint family decide on the advice of their tut to change
camp and the tut is supposed to drive in the first tethering peg
in the new camp, if he is present, but other joint families of the
same camp may decide not to move till another day. Leadership
in a local community consists of an influential man deciding to
do something and the people of other hamlets following suit
at their convenience. When villagers co-operate, there is no
appointed leader to organize their activity. If some members
of a village are attacked the others rush to their assistance,
headed by the swiftest and bravest, but there is no one who
calls on them to do so or who organizes their resistance. A
village is a political unit in a structural sense, but it has no
political organization. There is no headman or other appointed
leader invested with authority who symbolizes its unity and no
village council. Beyond his domestic groups a tut has authority
in his village only in the sense that he takes a prominent part
in questions of procedure and other discussions. Outside his
village he is a well-known person who is generally respected in
his district, but he has no political status.
In larger groups than a village and camp there is far less
THE POLITICAL SYSTEM i8i
co-ordination of activities and less scope for leadership. Only
in war is there any lengthy direct co-operation. Men noted for
their prowess and ability stir up among the youths enthusiasm
for a raid against the Dinka or a fight against another tribal
section, and direct what simple tactics are employed, but these
men have no political status or permanent leadership. The
warriors mobilize in local divisions of their own accord, for
there are no regiments and companies under officers, and in
fighting they follow the most forward and courageous among
them. Some of these warriors become renowned and their
reputation quickly attracts recruits for raids. Two of the most
famous war leaders were Latjor who led the Jikany tribes, and
Bidiit who led the Lou, eastwards. Neither had any ritual
qualifications, but both were men of outstanding ability who
were members of the dominant clans of their tribes. It is not
said by Nuer that either established any political control, or
even had great authority, in their tribes. The role of prophets
in war is examined later. Between tribal segments there are no
other joint activities that require organization and direction,
XII
The lack of governmental organs among the Nuer, the absence
of legal institutions, of developed leadership, and, generally, of
organized political life is remarkable. Their state is an acepha-
lous kinship state and it is only by a study of the kinship system
that it can be well understood how order is maintained and
social relations over wide areas are established and kept up.
The ordered anarchy in which they live accords well with their
character, for it is impossible to live among Nuer and conceive
of rulers ruling over them.
The Nuer is a product of hard and egalitarian upbringing, is
deeply democratic, and is easily roused to violence. His tur-
bulent spirit finds any restraint irksome and no man recognizes
a superior. Wealth makes no difference. A man with many
cattle is envied, but not treated differently from a man with few
cattle. Birth makes no difference. A man may not be a member
of the dominant clan of his tribe, he may even be of Dinka
descent, but were another to aUude to the fact he would run a
grave risk of being clubbed.
i82 THE POLITICAL SYSTEM
That every Nuer considers himself as good as his neighbour is
evident in their every movement. They strut about hke lords of
the earth, which, indeed, they consider themselves to be. There
is no master and no servant in their society, but only equals who
regard themselves as God's noblest creation. Their respect for
one another contrasts with their contempt for all other peoples.
Among themselves even the suspicion of an order riles a man and
he either does not carry it out or he carries it out in a casual and
dilatory manner that is more insulting than a refusal. When a
Nuer wants his fellows to do something he asks it as a favour to
a kinsman, saying, ' Son of my mother, do so-and-so ', or he includes
himself in the command and says : ' Let us depart ', ' Let the people
return home', and so forth. In his daily relations with his fellows
a man shows respect to his elders, to his 'fathers', and to certain
persons of ritual status, within the circuit of its reference, so long
as they do not infringe on his independence, but he will not submit
to any authority which clashes with his own interests and he does
not consider himself bound to obey any one. I was once discussing
the Shilluk with a Nuer who had visited their country, and he
remarked, 'They have one big chief, but we have not. This chief
can send for a man and demand a cow or he can cut a man's throat.
Whoever saw a Nuer do such a thing? What Nuer ever came
when some one sent for him or paid any one a cow ? '
I found Nuer pride an increasing source of amazement. It is
as remarkable as their constant aloofness and reticence. I have
already described how Nuer would interrupt my inquiries. I
mention here three incidents typical of the cavalier way in which
they treated me. On one occasion I asked the way to a certain
place and was dehberately deceived. I returned in chagrin to
camp and asked the people why they had told me the wrong way.
One of them replied, 'You are a foreigner, why should we tell you
the right way ? Even if a Nuer who was a stranger asked us the
way we should say to him, "You continue straight along that path",
but we would not tell him that the path forked. Why should we
tell him ? But you are now a member of our camp and you are
kind to our children, so we will tell you the right way in future.'
In this same camp, at the end of my stay, when I was sick and
being removed by steamer, I asked the people to carry my tent
and belongings to the river's edge. They refused, and my servant,
a Nuer youth, and I had to do it ourselves. When I asked him
why the people were churlish, he replied, 'You told them to carry
your belongings to the river. That is why they refused. If you had
asked them, saying, "My mother's sons, assist me", they would
not have refused.'
THE POLITICAL SYSTEM 183
On one occasion some men gave me information about their
lineages. Next day these same men paid me a visit and one of
them asked me, 'What we told you yesterday, did you believe it ?'
When I replied that I had believed it they roared with laughter
and called to others to come and share the joke. Then one of them
said, 'Listen, what we told you yesterday was all nonsense. Now
we will tell you correctly.' I could relate many such stories.
The Nuer have been rightly described as dour, and they are
often gruff and curt to one another and especially to strangers.
But if they are approached without a suggestion of superiority they
do not decline friendship, and in misfortune and sickness they
show themselves kind and gentle. At such moments they permit
themselves to show sympathy which their pride stifles at other
times, for even when Nuer approve of one they cannot bear that
one shall see it and are the more truculent to hide their friendli-
ness. Never are they truckling or sycophantic. When a Nuer
wants a gift he asks for it straight out, and if you refuse it he
remains in good humour. Their only test of character is whether
one can stand up for oneself. One rises in Nuer estimation the
more one lives their kind of life and accepts their values.
It you wish to live among the Nuer you must do so on their
terms, which means that you must treat them as a kind of kins-
men and they will then treat you as a kind of kinsman. Rights,
privileges, and obligations are determined by kinship. Either a
man is a kinsman, actually or by fiction, or he is a person to whom
you have no reciprocal obligations and whom you treat as a poten-
tial enemy. Every one in a man's village and district counts in
one way or another as a kinsman, if only by linguistic assimilation,
so that, except for an occasional homeless and despised wanderer,
a Nuer only associates with people whose behaviour to him is on
a kinship pattern.
Kinsmen must assist one another, and if one has a surplus of a
good thing he must share it with his neighbours. Consequently
no Nuer ever has a surplus. But the European has a surplus and
if his possessions are of any use to the Nuer he ought, in their
opinion, to share them with the people among whom he is living.
Travellers have often remarked that the Nuer have plagued them
for gifts. They beg from one another with equal persistence. No
Nuer is expected to part with his cattle or household property and,
except in special circumstances, these would not be asked for. But
were a man to possess several spears or hoes or other such objects
he would inevitably lose them. Deng, a Government chief and a
man of standing, told me, as I was leaving his village on the Pibor
River, that he was grateful for the fishing spears I had distributed
i84 THE POLITICAL SYSTEM
among his kinsmen, but added that they would not be able to
keep them when his relatives at Fadoi came to spend next dry
season on the Pibor.
The only way of keeping tobacco among the Nuer is to deny
that one possesses it and to keep it well hidden. When I gave
Deng a big lump of Anuak tobacco he managed to place a small
piece of it in his pipe, but he had at once to distribute the rest of it.
When I used to give tobacco to youths at Yakwac they generally
took a small piece for immediate use as snuff and asked me to
hide the rest, so that they could come and get a bit when they
wanted it without any one knowing that they possessed it. I had
hiding-places all over my tent. No Nuer can resist the pleadings
of his kinsman for tobacco. Age-fellows do not even ask for snuff
or tobacco, but, if they find it in a man's byre, they just take it.
My own system was to give away at the first opportunity anything
I possessed which Nuer might covet and to rest in poverty and
peace. Arab merchants are driven almost crazy by Nuer demands
for gifts, but they generally speak Nuer well and have considerable
knowledge of Nuer habits and so are able to hold their own. Never-
theless I have often observed that they make gifts where no return
is to be expected.
Nuer are most tenacious of their rights and possessions. They—
take easily but give with difficulty. This selfishness arises from.-.
their education and from the nature of kinship obligations. A
child soon learns that to maintain his equality with his peers he
must stand up for himself against any encroachment on his person
and property. This means that he must always be prepared to
fight, and his willingness and ability to do so are the only protec-
tion of his integrity as a free and independent person against the
avarice and bullying of his kinsmen. They protect him against
outsiders, but he must resist their demands on himself. The
demands made on a man in the name of kinship are incessant and
imperious and he resists them to the utmost.
XIII
Some personal recoUections and general impressions were
recorded in the last section in order that it might be understood
what are Nuer feelings about authority. It is the more remark-
able that they so easily submit to persons who claim certain
(^supernatural powers. Fetishes are a recent introductJQELJx)
Nuerland and they inspire much apprehension among the
people, so that, in recent years, their owners have often won
prestige in their villages and made themselves feared in their
THE POLITICAL SYSTEM 185
districts, and occasionally even in large tribal sections. These
fetish owners are not, however, in any sense tribal leaders and
cannot compare in social importance with the prophets.
Owing to the fact that Nuer prophets had been the foci of
opposition to the Government they were in disgrace and the
more influential of them under restraint or in hiding during my
visits to Nuerland, so that I was not able to make detailed
observations on their behaviour.' Without discussing Nuer
religious categories it may be said that a prophet is a man who
is possessed by one of the sky-spirits, or Gods, whom Nuer
regard as sons of the Sky-god. Nuer have great respect for
these spirits and fear, and readily follow, those whom they
possess. Prophets, consequently, achieved greater sanctity and
wider influence than any other persons in Nuer society. A
prophet is known as guk and is sometimes referred to as cok
kwoth, an ant of God. He also comes into the general category
of gwan kwoth, possessor of a spirit.
The first prophet to have gained great influence appears to have
been Ngundeng who died in 1906. He was a Lou tribesman of the
GAATLEAK clan and an immigrant from the Eastern Jikany. He
had practised as a leopard-skin chief before he had acquired a
reputation as a prophet by prolonged fasts and other erratic
behaviour, by his skill in curing barrenness and sickness, and by
his prophecies. Women came to him to be made fruitful from all
over Lou, from the Eastern Jikany tribes, and even from the west
of the Zeraf and Nile. Many brought oxen which Ngundeng
sacrificed to Deng, the Sky-god which possessed him. He then
anointed them with his spittle. When small-pox threatened the
Lou he went out to meet it and to stay its advance by sacrifice of
oxen. He foretold cattle epidemics and other events and led
expeditions against the Dinka.
When Ngundeng died, the spirit of Deng eventually entered
his son Gwek, who began to prophesy and to cure barrenness and
sickness, as his father had done. However, he never displayed the
pathological qualities of his father, who appears to have been a
genuine psychotic. The spirit of Deng had passed over his elder
brothers, or did not long reside in them. Nuer say that a spirit
eventually returns to the lineage of the man whom it first
' The Government has always looked askance at prophets and opposed their
influence. See, for some pejorative references to them, Jackson, op. cit.,
pp. 90-1 ; Fergusson, Appendi.x to Jackson, p. 107; C. A. Willis, 'The Cult of
Deng', S.N. and R., vol. -xi, 1928, p. 200.
4507 B b
^(^
o^oM*''^'^
i86 THE POLITICAL SYSTEM
possessed, even if a generation or two later, and the chosen vessel
generally becomes aware of its entry by experiencing severe sick-
ness with delirium. Ordinary Nuer, particularly if they are young,
do not want to be possessed, and it seems that it is often an
abnormal person who is possessed in the first place, while it is the
most ambitious of his sons on whom his mantle falls, for he would
seem to welcome possession, although not undergoing fasts to
procure it. Gwek was killed by Government forces in 1928.
Another famous prophet, Diu or Dengleaka, of the Gaawar tribe,
was a captured Dinka who set about to acquire a spirit by fasting
and solitude. He later achieved fame and power by successful
campaigns against the Dinka, whose country his followers occu-
pied, and against Arab slavers. Like Ngundeng he had a reputa-
tion for performing miracles. Dengleaka died in 1908 and his son
Dwal became possessed with the spirit of the Sky-god, Diu. He
is at present a political prisoner. The only prophet I have met,
Buom, of Dok country, was possessed by the spirit of a Sky-god,
Teeny. He was regarded as selfish and greedy by his neighbours,
having been cunning enough to get himself accepted as a Govern-
ment chief. He showed too great ambition, however, and is now
in exile. Well-known prophets, in other parts of Nuerland, were Mut
of the Eastern Gaajak, Kulangin Western Nuerland, and others.
Brief reference must be made to the very remarkable pyramid
erected by Ngundeng, and added to by Gwek, in the Rumjok
section of the Lou tribe. It was fifty to sixty feet high with large
elephant tusks planted round the base and at the summit. Dr.
Crispin's photograph on Plate XXV, taken in 1901, shows the
pallisade of ivory tusks, the type of material of which it was
constructed, and weathering, due to rains, round the base. It was
blown up by Government forces in 1928. The material used in
construction consisted of ashes, earth, and debris excavated from
sites of cattle camps. People came from all over Lou and the
Eastern Jikany tribes, bringing oxen with them for sacrifice, to
assist in the building. Nuer say that it was built in honour of the
Sky-god Deng and for the glory of his prophet Ngundeng. There
is no doubt that the cult of Deng was of Dinka origin and probably
the idea of a mound came from the same source. Besides the
famous Lou pyramid there is said to be a smaller one at Thoc in
Eastern Jikany country, built by a prophet called Deng, son of Dul.
Nuer are unanimous in stating that these prophets are a
recent development. They say that Deng descended from the
sky in recent times — within living memory, in fact — and that
he was the first, or almost the first, of the Sky-gods to come to
PLATE XX
Ciittle tra\-elliiif; (l^ou)
THE POLITICAL SYSTEM 187
earth. They state that in the old days there were no prophets ;
only the ritual officials mentioned earlier. The records of
European travellers are not explicit enough to confirm or reject
this statement. Poncet says that among the Nuer in his day
were rich and important persons, honoured after their death,
whom he calls devins ou sorciers diud jongleurs,^ and Brun-RoUet
says that the Nuer had a kind of pope for whom they had
veneration akin to worship, but his account is too fanciful to
carry weight. 2 Whilst it is difficult to believe that there were
Fig. 13. Calf's bell-necklace of palm nuts.
no cases of possession sixty years ago we must, in the absence
of conflicting evidence, accept so unanimous a declaration by
Nuer that there was no possession by Sky-gods, and it seems
fairly certain that if there were any prophets at that time their
influence was restricted to small localities and had not the tribal
significance of more recent times. There is some evidence that
the development of Nuer prophets was related to the spread
of Mahdism from the Northern Sudan. However that may
be, there is no doubt that powerful prophets arose about the
time when Arab intrusion into Nuerland was at its height
and that after the reconquest of the Sudan they were more
respected and had more influence than any other persons in
Nuerland.
Nevertheless, we consider that the power of these prophets,
even of the most successful ones, has been exaggerated and that
' Poncet, op. cit., p. 40. ^ Brun-Rollet, op. cit., p. 222.
i88 THE POLITICAL SYSTEM
their tribal position has been misunderstood. The earHest
Government officials to enter Lou country have recorded that
Ngundeng was greatly feared and respected and were of the
opinion that if the Lou were to be administered he would have
either to be reconciled or removed. However, his son, Gwek, was
not supported against the Government by some sections of the
tribe. Mr. Struve, then Governor of the Upper Nile Province,
reported that Dwal, son of Diu, had 'rather shaky authority'
among the Gaawar. I received the impression in 1932 that
Buom had far more power in Dok country as a government
chief than he ever had in pre-government days as a prophet.
A prophet's curse is feared, but armed intervention of Govern-
ment forces is a weightier sanction. Buom was attempting to
exercise unprecedented judicial functions and his banishment
met with no popular hostility and with little regret. There is no
good evidence that the earlier prophets were more than spiritual
persons, whose ritual powers were used particularly in warfare,
though some of the later ones appear to have begun to settle
disputes, at any rate in their own villages and districts. Of them
all, perhaps Gwek came nearest to exercising political functions
and to imposing his authority outside his own district, but the
hostility between tribes and between tribal segments rendered
effective personal control impossible.
The only activities of prophets which can truly be called
tribal were their initiation of raids against the Dinka and their
rallying of opposition to Arab and European aggression, and it
is in these actions that we see their structural significance and
account for their emergence and the growth of their influence.
All the important prophets about whom we have information
gained their prestige by directing successful raids on the Dinka,
for these raids were carried out in the names of the spirits which
promised rich booty by their lips. No extensive raids were
undertaken without the permission and guidance of prophets,
who received instructions from the Sky-gods in dreams and
trances about the time and objective of attack, and often they
accompanied them in person and performed sacrifices before
battle. They took for themselves part of the spoil and to some
extent supervised the division of the rest of it. The warriors
sang war hymns to the Sky-gods before starting on raids and the
THE POLITICAL SYSTEM 189
sacrifices to them by their prophets were beheved to ensure
booty and safety.
For the first time a single person symbohzed, if only to a
moderate degree and in a mainly spiritual and uninstitution-
alized form, the unity of a tribe, for prophets are tribal figures.
But they have a further significance, for their influence extended
over tribal boundaries. Gwek had great influence in Gaajok, and
it is said that because of it the Lou and the Gaajok for a time
paid compensation for homicide between them. His influence
reached as far as the Eastern Gaagwang and Gaajak. Dengleaka
had a similar influence in the valley of the Zeraf, especially
among the Thiang. Some of the Western Nuer prophets had a
reputation among a number of neighbouring tribes which united
for raids at the direction of their spirits. They were not a
mechanism of tribal structure like leopard-skin chiefs, but were
pivots of federation between adjacent tribes and personified the
structural principle of opposition in its widest expression, the
unity and homogeneity of Nuer against foreigners. Probably,
the coalition of tribes and the organization of joint raids is very
largely their achievement — though one cannot be sure of this
in the absence of historical records — ^and has made them im-
portant and powerful figures in Nuerland. This interpretation
explains how it is that prophets began half a century ago, or at
any rate came to the fore then. Certain structural changes were
taking place in response to changed conditions : the development
of functions that were more purely political than any exercised
by individuals before and of a greater degree of unity among
neighbouring tribes than there had been hitherto. As Sky-gods
passed at the death of prophets into their sons we are further
justified in suggesting growth of hereditary political leadership
which, with the strong tendency towards federation between
adjacent tribes we attribute to the new Arab-European menace.
Opposition between Nuer and their neighbours had always been
sectional. They were now confronted by a more formidable and
a common enemy. When the Government crushed the prophets
this tendency was checked. As we understand the situation, the
prophets were inevitably opposed to the Government because it
was this opposition among the people which led to their emerg-
ence and was personified in them.
igo THE POLITICAL SYSTEM
XIV
We have tried to show how distribution depends on oecology
and how the hnes of poHtical cleavage tend to follow distribu-
tion in relation to modes of livelihood. But consideration of
oecology only helps us to understand some demographic features
of Nuer tribes and tribal segments and not the nature of their
structural relations. These can only be understood in terms of
certain structural principles and we have tried to isolate these
principles, though not, we admit, at a very deep level of
analysis. The chief points we have made are here briefly sum-
marized.
(i) Nuer attach values to their geographical distribution and
these evaluations give us socio-spacial units and relate these
units into a system. (2) In all such units there is evident a
tendency to segment into opposed segments, and also for these
segments to fuse in relation to other units. (3) The smaller
is the segment the greater is its cohesion, and it is for this
reason that a segmentary system exists. (4) The political
system of the Nuer can only be understood in relation to a
whole structure of which other peoples form part and, likewise,
the character of all Nuer communities must be defined by
their relations with other communities of the same order within
the whole political system. (5) The social system is much wider
than spheres of actual political relations and cuts across them.
(6) Political values depend on more than residential relations.
Political relations can be isolated and studied independently of
other social systems, but they are a specific function of the whole
set of social relations. These are mainly of a kinship type and
the organization of kinship relations into political relations in
certain situations is one of our major problems. (7) The
structural relations between Nuer tribes and other peoples and
between tribe and tribe are maintained by the institution of
warfare and the structural relations between segments of the
same tribe are maintained by the institution of the feud. (8)
There is no central administration, the leopard-skin chief being
a ritual agent whose functions are to be interpreted in terms of
THE POLITICAL SYSTEM igi
the structural mechanism of the feud. (9) Law is relative to the
structural distance between persons and has not the same force
in different sets of relations. (10) The new conditions of Arab-
European intrusion have probably caused the development of
prophets, with embryonic juridical functions, and greater
intertribal solidarity.
CHAPTER V
THE LINEAGE SYSTEM
I
Many characteristics of Nuer lineages and clans belong to the
study of kinship which we hope to make in a subsequent
volume. Here only^those characters which are strictly relev ant .
to the territorial system are discussed. We begin, however, by
a formal definition of clan, lineage, and kin. A Nuer clan is the
largest group of agnates who trace their descent from a common
ancestor _and between whom marriage is forbidden and sex ual,
relations considered incestuous. It is not merely an undifferen-
tiated group of persons who recognize their common agnatic
kinship, as are some African clans, but is a highly segmented
genealogical structure. We refer to these genealogical segments
of a clan as its lineages. The relationship of any member of
a lineage to any other member of it can be exactly stated
in genealogical terms and, therefore, also his relationship to
members of other lineages of the same clan can be traced,
since the relationship of one lineage to another is genealogically
known. A clan is a system of lineages and a lineage is a genea-
logical segment of a clan. QnejrnjglrLs peak of the whole clan as
a. lineage, b ut we prefer to speak of lineages as segments of it.
and to define them as such. Alternatively one may speak of a
lineage as an agnatic group the members of which are genea-
logically linked, and of a clan as a system of such groups, the
system being, among the Nuer, a genealogical system. In the
diagram on p. 193 clan A is segmented into maximal lineages B
and C and these bifurcate into major lineages D, E, F, and G.
Minor lineages H, I, J, and K are segments of major lineages
D and G, and L, M, N, and O are minimal lineages which are
segments of H and K. It was found that words to describe four
stages of lineage segmentation are sufficient when speaking of
even the largest clans. The commonest Nuer word for a lineage
is thok dwiel and the smallest genealogical unit they describe as
a thok dwiel has a time depth of from three to five generations
from living persons. A whole clan is thus a genealogical
THE LINEAGE SYSTEM 193
structure, and the letters in the diagram represent persons from
whom the clan and its segments trace their descent and from
whom they often take their names. There must be at least
twenty such clans in Nuerland, without taking into account
numerous small lineages of Dinka origin.
A
B I C
D
H
K
M
N
O
A lineage in the sense in which we generally employ this
word is a group of living agnates, descended from the founder
of that particular line. Logically it also includes dead persons
descended from the founder — and we sometimes use the word to
include them also — but these dead persons are only significant
in that their genealogical position explains the relationships
between the living. It is clear from the context whether the
word is used in a more or less inclusive sense.
Clans and lineages have names, possess various ritual symbols,
and observe certain reciprocal ceremonial relations. They have
spear -name s_w hich a re shouted out at ceremonies, honorific
titles by which people are sometimes addressed, totemic and
other mystical affiliations, and ceremonial status towards one_
another. .
Agnatic kinship between lineages is caUed huth. Buth is
always an agnatic relationship between groups of persons, and
only between persons in virtue of their membership of groups.
Buth agnation is to be distinguished from kinship in the sense
of relationship between persons, e.g. between a man and his
father's brother and mother's brother. Cognation in this sense
the Nuer call mar. Any person to whom a man can trace any
genealogical link, whether through males or females, is mar to
him. A man's mar are consequently all his father's kin and all
his mother's kin, and we call this cognatic category his kindred.
4507 c c
194 THE LINEAGE SYSTEM
In normal usage the word refers to close relatives only. There-
fore, as mar includes close agnates, the word buth is used only in
reference to distant agnates, between whom and the speaker
there is lineage cleavage. There is buth relationship between
lineages of the same clan and also between related clans which
have the same ancestor but do not compose an exogamous unit.
A Dinka can be given buth relationship to a Nuer lineage by
being adopted into a collateral lineage. We thus formally distin-
guish between the lineage system, which is a system of agnatic
groups, and the kinship system, which is a system of categories
of relationship to any individual ; and we speak of these relation-
ships as a man's paternal kin and his maternal kin, and both
together as his kindred.
Political and lineage groups are not identical, but tjiey^jiaye
a certam correspon dence and o ffehn3eaj"the3ame jame., for a
tribal a.r£a-andits-^ivisions^are often, called after- the -claas-and-
lineages which are supposed to have first occupied -them. This
makes their inter-relation a very difficult problem to investigate
and has led to some confusion in writings on the Nuer. Thus
Gaawar is the name of a tribal area, of the tribesmen who
inhabit it, and of the members of a clan which have in that
area a socially dominant status. Likewise Gaajok and Gaajak
are names of territories, tribes, and lineages. To clarify des-
cription clan and lineage names are therefore printed in small
capitals. Hence, when we say that a man is a Gaawar we mean
that he is a Gaawar tribesman, a man who resides in Gaawar
country, whereas when we say that he is a gaawar we mean
that he is a member of the gaawar clan, a descendant in the
male line from War, the founder of the dominant clan in
Gaawar country.
II
Three trees of clan descent are presented in the following
section in a form conventional to us, and which would also
commend itself to Nuer, who sometimes speak of a lineage as
kar, a branch, as illustrations of the way in which lineages split
up, each being a branch of a larger one. The jinaca are the
dominant clan in the Lou and Rengyan tribes, the gatgankiir
are the dominant clan in the Jikany tribes, and the (gaa)thiang
THE LINEAGE SYSTEM 195
are the dominant clan in the Thiang tribe. Only one line of
descent is shown from the roots to the twigs, from the clan
ancestor to the minimal lineages, but it is indicated where the
other main branches and boughs, the other maximal, major,
and minor lineages leave the stem,
III
We have defined lineage and clan and give some examples
of them in diagrams. In this section we note some character-
istics of both which are germane to our inquiry. It may at
once be said that it is not easy to discover a Nuer's clan, for
a clan is not to Nuer an abstraction and there is no word in i^
their language that can be translated 'clan ' in ours. One may
obtain the name of a man's clan by asking him who was his
* ancestor of yore ' or his 'first ancestor ' (gwandong) or what are
his 'seeds ' {kwai) , but it is only when one already knows the clans
and their lineages and their various ritual symbols, as the
Nuer does, that one can easily place a man's clan through his
lineage or by his spear-name and honorific salutation, for Nuer
speak fluently in terms of lineages. A lineage is thok mac, the
hearth, or thok dwiel, the entrance to the hut, or one may
speak of kar, a branch. Thok dwiel is the commonest expression
to denote a line of agnatic descent in those situations when
genealogical exactness and precision are relevant, but in normal
everyday usage Nuer employ the word cieng, as will be ex-
plained later. The clan has itself a position analogous to that
of a lineage in the clan system, and it is significant to the Nuer
not so much as a unique group as a segment in a system of
groups, for it acquires its singularity only as part of a system.
A lineage is a relative term, since its range of reference depends
on the particular person who is selected as the point of departure
in tracing descent. Thus, if we were to begin with a father, the
thok dwiel would include only sons and daughters, but if we
were to take a grandfather as our point of departure, it would
include all his sons and daughters and the children of his sons.
A larger and larger number of agnates would be included the
higher up in the line of ascent we were to take the point of
departure for counting descendants. It might be maintained
that the smallest possible lineage are the sons and daughters of
196 THE LINEAGE SYSTEM
one man, but Nuer do not refer to them as a thok dwiel. They
are, with their father and their mother, a family and household.
One cannot say definitely how far up a line of ascent Nuer will
MINIMAL LINEAGES
MINOR LINEAGES
MAJOR LINEAGES
MAXIMAL LINEAGES
CLAN
go in selecting the apex of a minimal lineage. They may go back
only two steps, to the grandfather, making in aU three genera-
tions of agnates, but a minimal lineage of four or five
generations is more usual. Apart from Nuer usage, we con-
sider it important to define lineages as groups wit h a depth of
at j£aslihr£e g€ne£a4iQn&,_sinc£jthey are then distinct structural 1
THE LINEAGE SYSTEM
segments in a system ofsuch_seg ments and jiol
wTth do mestic ^oups.
197
:-Confused_
The Nuer clan, being thus highly segmented, has many of
the characteristics which we have found in tribal structure. Its
MINIMAL LINEAGES
MINOR UNEAGES
MAJOR LINEAGES
MAXIMAL LINEAGES
CLAr4
lineages are distinct groups only in relation to each other.
Thus, in the diagram on p. 193, M is a group only in opposition
to L, H is a group only in opposition to I, D is a group only in
opposition to E, and so on. There is always fusion of collateral
lineages of the same branch in relation to a collateral branch,
e.g. in the diagram, L and M are a single minor lineage, H, in
opposition to I, and not separate lineages ; and D and E are a
igS THE LINEAGE SYSTEM
single maximal lineage, B, in opposition to C, and not separate
lineages. Hence two lineages which are equal and opposite are
composite in relation to a third, so that a man is a member of a
lineage in relation to a certain group and not a member of it in
relation to a different group. Lineage values are thus essentially
niNOR LINEAGES
MAJOR LINEAGES
MAXIMAL LINEAGES
CLAN
relative like tribal values, and we suggest later that the processes
of lineage segmentation and political segmentation are to some
extent co-ordinate.
Since humans propagate themselves, the clans might be sup-
posed to be further and further removed from their founder,
and the living representatives of a clan to become wider and
wider separated in lineage structure. However, we do not
believe that this is the case. In theory every man is a potential
founder of a lineage, but, in fact, lineages spring from very few
THE LINEAGE SYSTEM 199
names. The others, for one reason or another, drop out, so that
only certain Hnes of descent are remembered. Also, in those
lines that persist names drop out of the steps in ascent to the
founder of the clan, so that the distance in generations from
the founder of a clan to the present day remain fairly constant.
We have suggested in our discussion on Nuer time-reckoning
that there is always this difference between a true genealogy
and a genealogy which Nuer deem to be true. The evidence for
the assertion depends partly on a comparative study of East
African genealogies and partly on a study of Nuer genealogies.
We mention some of the reasons, arising from reflections on
Nuer genealogies, that have led us to this conclusion, since the
validity of some arguments, developed later in the chapter,
depends on their acceptance.
(i) All the main clans have about ten to twelve generations
from the present day to the ancestors who gave rise to them.
There is no reason to suppose that the Nuer came into existence
ten to twelve generations ago. (2) When a Nuer is asked his
lineage he gives it by reference to an ancestor, the founder of his
minimal lineage, who is from three to six, generally four to live,
steps in ascent from the present day. These steps are certain and
agreed upon. This is understandable, since five steps represent a
man, his father, his grandfather, and his grandfather's father and
grandfather, and since a man instructs his children in the names of
his immediate forbears. It is evident that after five or six genera-
tions the names of ancestors become lost. Young men often do not
know them, and there is frequent confusion and disagreement among
older persons. The founder of the minor lineage must be placed some-
where between the founder of the minimal lineage and the founder of
the major lineage ; the founder of the major lineage must be placed
somewhere between the founder of the minor lineage and the
founder of the maximal lineage ; and the founder of the maximal
lineage must be placed somewhere between the founder of the
major lineage and the founder of the clan. The names of these
founders of lineage-branches must go into the line of ascent some-
where, and in a definite order, because they are significant points
of reference. It is immaterial whether other names go in or not,
and their order is without significance. Consequently some in-
formants put them in and some leave them out, and some put
them in one order and others in a different order. It is evident,
moreover, that since the minimal lineage consists of four or five
actual steps in ascent, there has been telescoping of the agnatic
200 THE LINEAGE SYSTEM
line from the founder of the minimal lineage further up the line
of ascent to the founder of the clan, for the founder of the minimal
lineage was himself the extremity of another minimal lineage
which has, by increase in generations, become the minor lineage,
and so on. Consequently, even were the supposed founder of the
clan the real founder of it, there ought to be at least sixteen steps
from him to the present day, assuming that minimal lineages have
always been of the same character as at the present time. The
length of each fork in the tree of descent ought logically to be of
equal length, whereas the twig, so to speak, is longer than the
branch or stem from which it springs. (3) There is another way
in which only significant ancestors, i.e. ancestors who form the
apex of a triangle of descent, are denoted in genealogical trees, and
irrelevant ancestors, i.e. ancestors who do not give their name to
a group of descendants, are obscured and finally forgotten. Not
only do links drop out of the direct line of descent, but also collateral
lines merge. It is clear from a study of Nuer genealogies that the
descendants of one or two brothers become numerous and domin-
ant, that the descendants of others die out, and that the descendants
of yet others are relatively few and weak and attach themselves,
as is explained in later sections, by participation in local and
corporate life, to a stronger and dominant collateral line. They
become assimilated to this line in ordinary lineage reference and
eventually are grafted on to it by misplacement of their founder,
who becomes a son instead of a brother of its founder. The
merging of collateral lines higher up a lineage seems to be com-
mon, and to be more frequent and necessary the higher up one pro-
ceeds. It is necessary because, as will be seen later, the lineage
system provides one of the principles of political organization.
The structural form of clans remains constant, while actual
lineages at any point in time are highly dynamic, creating new
bifurcations and merging old ones. They may therefore be
presented as trees. But a presentation more useful for socio-
logical analysis is in terms of structural distance, for lineages
are groups of living agnates, and the distance between them
varies with their relative positions in clan structure. Thus, in
the diagram below, the line A-B represents the jinaca clan.
The agnatic distance between the mar minimal lineage and
other lineages of the same clan is represented both on the line
A-B and also in its time depth by their point of convergence
in ascent on the line B-C. The wider the range of agnation
specified the further back their point of convergence, so that
PLATE XXI
a. Cattle yraziny on rielye (Leek
)^'"-'-^"f->
b. Early dry season cattle camp at forest pool (Lou)
THE LINEAGE SYSTEM 201
the depth of a Hneage (the vertical Hne of ascent) is always in
proportion to its width (the base line representing living lineage
groups in the clan system).
A Nuer clan, therefore, is a system of lineages, the relation-
ship of each lineage to every other lineage being marked in its
structure by a point of reference in ascent. The distance to this
J I N ACA
of Lou and
Rengyan tribes
C
'J IN ACA
of Lou tribe
DIEL MALUAL KWOTH MAR
jToint is what we rqlj the ti me depth of a lineag e. In theory the
genealogical relationship between any two clansmen can be
traced through this point, and Nuer can actually trace it if they
take the trouble. However, they do not consider it necessary
to know the exact genealogical relationship between persons
who are known to be distantly related by membership of their
respective lineages. Thus it is sufficient for a man of the
GAATBAL lineage to know that another man is of the gaaliek
lineage without his having to know the man's exact descent, for
these two lineages stand to one another in a certain structural
relationship, and therefore the two men stand to one another at
that distance. Nuer are conversant up to a point — generally
up to the founders of their minimal and minor lineages — with
the full range of their genealogical relationships. Beyond this
point they reckon kinship in terms of lineages. It is necessary
4507 J) d
202 THE LINEAGE SYSTEM
for a Nuer to know not only that a man is a fellow clansman,
but also to know what lineage he belongs to, in deciding ques-
tions of exogamy and ceremonial. The relationship between a
lineage and others of the same clan is not an equal relationship,
for lineages are structurally differentiated units which stand to
one another at different and exact structural distances.
It is interesting to note how the Nuer themselves figure a
lineage system. When illustrating on the ground a number of
related lineages they do not present them the way we figure
them in this chapter as a series of bifurcations of descent, as a
tree of descent, or as a series of triangles of ascent, but as a
number of lines running at angles from a common point. Thus
in Western Nuerland a man illustrated some of the gaatgan-
KIIR lineages, using the names of their founders, by drawing
the figure above on the ground. This representation and Nuer
comments on it show several significant facts about the way in
which Nuer see the system. They see it primarily as actual
relations between groups of kinsmen within local communities
rather than as a tree of descent, for the persons after whom the
lineages are called do not all proceed from a single individual.
Jok, Thiang, and Kun are three sons of Kir and founders of
the maximal lineages gaajok, gaajak, and gaagwong of the
GAATGANKiiR clau. Thiang and Kun are shown next to each
other because jointly they form the lineage framework of the
Gaajak tribe. The Gying lineage does not belong to the gaat-
GANKiiR clan, but it is shown next to Kun because of the
proximity of the Reng section, of which it forms part, to the
THE LINEAGE SYSTEM 203
Gaagwong section. Nyang is shown as a short hne at the side
of Jok because, although the hneage which springs from him
belongs to the group of lineages founded by Thiang, they live
in the Gaagwang tribe together with a lineage descended from
Jok, and the Gaagwang tribe is closely associated with the
Gaajok tribe. The Nuer, outside certain ritual situations,
evaluate clans and lineages in terms of their local relations.
Herein lies the importance of these groups for this study.
IV
Nuer lineages are not corporate, localized, communities,
though they are frequently associated with territorial units, and
those members of a lineage who liv-e in an. area .associated with
J.t see themselves as a residential group, and the value, or
concept, of lineage thus functions through the political system.
Every Nuer village is associated with a lineage, and, though the
members of it often constitute only a small proportion of the
village population, the village community is identified with
them in such a way that we may speak of it as an aggregate of
persons clustered around an agnatic nucleus. The aggregate
is linguistically identified with the nucleus by the common
designation of the village community by the name of the
lineage. It is only in reference to rules of exogamy, certain
ritual activities, and to a very limited extent to responsibility
for homicide, that one needs to regard lineages as completely
autonomous groups. In social life generally they function
within local communities, of all sizes from the village to the
tribe, and as part of them.
A Nuer rarely talks about his lineage as distinct from his
community, and in contrast to other lineages which form part
of it, outside a ceremonial context. I have watched a Nuer
who knew precisely what I wanted, trying on my behalf to
discover from a stranger the name of his lineage. He often
found great initial difficulty in making the man understand the
information required of him, for Nuer think generally in terms
of local divisions and of the relations between them, and an
attempt to discover lineage affiliations apart from their com-
munity relations, and outside a ceremonial context, generally
led to misunderstanding in the opening stages of an inquiry.
204 THE LINEAGE SYSTEM
I must here again refer to the term cieng, which has been a source
of confusion in Nuer studies. A Nuer does not normally say
that he is a man of such-and-such a thok dwiel (lineage) when he
denotes his social position, but says that he is a man of a certain
local community, cieng. Thus he says he is a man of cieng Mar,
cieng Pual, cieng Leng, cieng Gaatbal, and so on (see diagram
on p. 196). What he is telling you is that he is a member of a
group of people who live together in a village or district or tribal
section. In ordinary situations of social life it is irrelevant whether
he is, or is not, a member of the lineages from which these local
communities derive their names. Moreover, since in ordinary
speech a lineage-name has a local rather than a strict kinship
connotation, those who share community life with the members of
the lineage speak of themselves as though they also were members
of it, because politically they are identified with it. Thus the word
cieng is often used in an ambiguous sense which leaves lineage
affiliation uncertain, since it is irrelevant, and it may, in conse-
quence, be difficult, without much probing, to discover to what
lineage a man belongs.
A cieng, in the sense of 'homestead', is called after the man who
owns it, e.g. the homestead of Rainen is called 'cieng Rainen'.
When Rainen is dead and his sons and younger brothers and
nephews live in his home they may call the hamlet after him, and
it will be said that they are all members of cieng Rainen. If Rainen
was an important man, and proves to have been the begetter of a
strong line of descent, the whole village wherein live his agnatic
heirs, and strangers who have intermarried with them or in other
ways become attached to them, may thus become known as ' cieng
Rainen'. In course of time his descendants multiply and consti-
tute the nucleus of a tribal section which is called 'cieng Rainen'.
Hence it has come about that many tribal sections are called after
persons, e.g. cieng Minyaal, cieng Dumien, cieng Wangkac, &c.
A lineage thus becomes identified in speech with the territory it
occupies; the district occupied by the major lineage of wangkac,
for example, being known as cieng Wangkac. A Nuer then talks
about the local community and the lineage which is its political
nucleus as interchangeable terms. He even speaks of cieng wang-
kac when he means the wangkac lineage. This habit confuses
a European, since the wangkac lineage and the people who live
in the Wangkac section are by no means the same.
If you ask a Lou man what is his cieng you are asking him where
he lives, what is his village, or district. Suppose that he replies
that his cieng is cieng Pual. You can then ask him what his cieng
forms part of, and he will tell you that it is a part of cieng Leng,
THE LINEAGE SYSTEM 205
giving the name of a tertiary tribal section. If you continue to
interrogate him he will inform you that cieng Leng is part of cieng
Gaatbal, a secondary tribal section, and that cieng Gaatbal is a
division of the Gun primary section of the Lou tribe. But he has
told you nothing about his clan affiliation. He may or may not
be a member of the pual lineage, which forms part of the leng
lineage, which forms part of the gaatbal hneage, which is part
of the clan of jinaca or gaatgannaca, the people of Nac or
the children of the children of Nac. In the same way, if a
Jikany man tells you that he belongs to cieng Kwith of the Gaajok
it does not follow that he is descended from Kwe (Kwith) who
is descended from Jok the founder of the gaajok lineage of
the gaatgankiir clan, the children of the children of Kir. He
may merely mean that he lives in a tribal section occupied
by the kwith lineage of this clan. But these men will not tell
you that they are not members of the dominant lineages in these
sections, and will allow you to assume that they are, for in
community relations there is a degree of Hnguistic assimilation of
all residents other than members of the dominant lineage to
that lineage, and people do not wish the fact that they are
strangers in the tribal area to be publicly stressed, especially if
they are of Dinka origin.
The assimilation of community ties to lineage structure, the
expression of territorial affiliation in a lineage idiom, and the
expression of lineage affiliation in terms of territorial attach-
ments, is what makes the lineage system so significant for a
study of political organization.
V
In emphasizing the relations between lineages and local com-
munities we speak mainly of those lineages which are segments
of the dominant clans in the different tribes. It is these which
have the greater political importance. We speak of them more
fully in later sections. Here we give a preliminary account of
how they are associated with tribal segments and how they act
in those segments as a framework of the political structure.
The two diagrams which follow show the main lineages of the
dominant clans among the Lou and Eastern Jikany tribes and
the larger tribal sections in which they are dominant. The lines
of descent are traced only as far as is necessary to illustrate the
argument which follows. They may be compared with the trees
2o6 THE LINEAGE SYSTEM
of descent of the two clans and with the maps on pp. 56 and 58,
which show the distribution of the sections of the two tribes.
Mor primary section
Gaaliek secondary
section
Nyang (founder of
GAALIEK maximal
lineage)
LOU TRIBE
Gun primary section
Gaatbal secondary section
Denac (founder of
I jiNACA clan)
Bal (founder of
GAATBAL maxi-
mal lineage)
Campi
Gilgil
Rumjok secondary
section
Dak (founder of
RUMJOK maximal
lineage)
Bang (founder of Ling (founder
NYARKWAC major of LENG major
lineage) lineage)
In the Lou tribe it will be seen that the descendants of Nyang,
son of Denac, form the nucleus of the Gaaliek secondary section ;
that the descendants of Bal, another of his sons, form the nucleus
of the Gaatbal secondary section, and that the descendants of
Dak, his third son, form the nucleus of the Rumjok secondary
section. The sections shown in the map, which are not accounted
for in the genealogical tree, are the Jimac and the Jaajoah sections,
which have clan nuclei of foreign origin. It must be made clear
that when a tribal section is called after a lineage it does not
mean that all members of the lineage live in it, though prob-
ably most of them do, and it certainly does not mean that
they alone live in it, for investigations show that they form only
a small minority in the total population of the section. Large
stranger lineages are included in the area designated by a title
taken from any of the sons of Denac, e.g. the thiang lineage in the
Gaatbal primary section. Also there are innumerable small
lineages of Nuer strangers and of Dinka clustered around hneages
of GAATNACA dcsceut. Thus, if one visits the villages and cattle
camps called after lineages which spring from the gaaliek maximal
lineage, e.g. the lineages of jaanyen and kuok, one will find them
occupied by a relatively small number of persons of these lineages,
THE LINEAGE SYSTEM 207
while most of the residents will be found to have sprung from other
Nuer and Dinka clans. Thus, at the cattle camp at Muot Dit in
1930 there were not only various minimal lineages of the rue
branch of the rum j ok maximal lineage of the gaatnaca clan,
the 'owners' of the site, but also a kaang lineage from Lang
country, a kan lineage from Bui country, a kong lineage from
Lang country, all segments of clans dominant to the west of the
Nile, and many Dinka lineages. Likewise at the village of Par-
kur in Leek country in 1930 hamlets were occupied by a nyapir
Hneage of the dominant gaatbol clan, a cuor Hneage of Dinka
which has close association with the jikul clan, a geng lineage
from Beegh country, a rual lineage from Bui country, a
KWACUKUNA lineage of the jimem clan, and other lineages.
The descendants of Jok'-Kir form the aristocratic nucleus
of the Gaajok tribe ; the descendants of Thiang^-Kir and Kun-Kir
form the aristocratic nucleus of the Gaajak tribe ; and the descend-
ants of Gwang-Jok and Nyang-Thiang together form the aristo-
cratic nucleus of the Gaagwang tribe. On the map on p. 58 the
names of these lineages figure as tribal sections, for they have given
their names to the areas in which they are dominant. Thus the
names of three major Hneages descended from Jok, laang, yol,
and WANGKAC appear on the map as Laang, Yol, and Wangkac.
None of the foreign lineages is of sufficient size to give its name to
a large tribal division, though they give their names to small
sections and villages. 3 Here again, it must not be supposed that
the descendants of Jok form more than a fragment of the Gaajok
tribe. In the same way the Gaagwang tribe comprises many
foreign elements which probably far outnumber the descendants
of Nyang-Thiang and Gwang-Jok. The same applies to the Gaajak
tribe. The gaagwong lineage is so closely associated with the
GAAJAK lineage in tribal life that they form together a twin
nucleus of the Gaajak tribe. The secondary sections of Lony, Kang,
and Tar, shown on the map, are aU called after lineages which spring
from Thiang-Kir. Likewise the secondary sections of Nyayan,
Cany, and Wau are all called after lineages which spring from
Kun-Kir. Hence the only section in the map not accounted for
I Also called Majok. 2 Also called Mathiang.
3 Village-names are generally place-names and not names of lineages, but
one may refer to the communities by the names of their chief lineages. It
often happens that these lineages are of stranger or Dinka descent, and though
the communities may be referred to by their names, it is recognized that the
sites belong to lineages of the dominant clan of the tribe. Hence a village may
be associated with two lineages. Thus the juak, ngwol, &c., of the Leek tribe
are stranger or Dinka lineages after whom village communities are called, but
the sites are the land of the keunyang lineage, and they alone are diel in them.
■?^
6fl
K
n
i
a
a
H
«
Ui
O
<
>-.
<
^
<
<o
O
to
Si
Z
O
^^^
^
hn
o
nJ
■<
o
•<
d
a
o
3
o
<u
y
£<
«<!
<
<
a>
c ^
bf)
a
C/5
w
PQ
»— (
H
o 5
«
3 "S
m
«2 t^
o
-^ <
u <
jao
bO
5 o 2 So
-0 fe S i
o
< <u
<; 60
bo "
a IS
o
o 'aT
►-> bo
^ ni
o a
o S
h)
o
<l>
>•
bo
n1
_^
o
n
>^
<D
Ui
2 t^
j2 B
<
<
01
a
bD
^
a
o
u
rt
n
^
'(?
'V
a
THE LINEAGE SYSTEM 209
in the diagram is the primary section of Reng, the Kong and
Dhilleak divisions of which have clan nuclei of different origin to
the other sections.
In all these tribes and tribal sections there is much admixture
of lineages in communities. The same conditions are found in
the Zeraf valley and also, and perhaps to a larger degree, in
Western Nuerland. Nuer say that feuds and quarrels between
lineages chiefly led to their dispersal, and they can cite many
examples. Thus the descendants of Nyang-Thiang left the
other sons of Thiang and joined the people with whom they
form the Gaagwang tribe, and the nyaruny lineage left their
kinsmen of the Thiang primary section and joined the Reng
primary section of the Gaajak tribe. After fights, a whole
community, led by its dominant lineage, may in this way move
to a different section or tribe and take up permanent residence
with them. Migration led to further dispersal, since some
lineages remained in the homeland to the west of the Nile and
others crossed the Nile and Zeraf and settled to the east of these
rivers. Nuer say that in the earliest stages of the migration the
warriors used to raid Dinka and return home to their kinsmen
after each raid. Then they settled to the east, but kept in close
contact with the people of their lineage to the west. But as they
shifted farther away the contact lessened and finally, in most
cases, ceased altogether. When lineages migrated they must
already have been, nuclei of heterogeneous accretions and not
exclusive groups of agnates: but admixture was no doubt
hastened and made more complete by movement.
Two factors other than migration, quarrels, intermarriage,
&c., have probably contributed to the dispersal of lineages.
The Nuer are mainly a pastoral people with dominant pastoral
interests and do not feel themselves bound by economic neces-
sity or by ritual ties to any particular spot. Where his cattle are is
the Nuer's home ; his hearth is the droppings of his cattle and his
altar a light pole [riek) which he plants wherever he may wander,
The spirits which give him protection and the ancestral ghosts
that watch over his welfare are no more bound to the soil than
he is and they are present wherever are the herds. Beasts
dedicated to ghosts and spirits are his wandering shrines. Also
the Nuer have no organized cult of ancestral ghosts. The dead
4507 E e
210 THE LINEAGE SYSTEM
are buried quickly and crudely in unvisited and unremembered
tombs ; only in very rare cases are sacrifices offered to them ;
and there are no sacred places associated with them.
Thus Nuer have always felt themselves free to wander as they
pleased, and if a man is unhappy, his family sick, his herds de-
clining, his garden exhausted, his relations with some of his
neighbours uncongenial, or merely if he is restless, he moves to
a different part of the country and resides with some kinsmen.
It is seldom that a man goes alone, for brothers are a corporate
group and, especially if they are sons of one mother, stick to-
gether. So, frequently as a result of quarrels, a group of brothers
will often leave a village and settle elsewhere. Nuer say that
they usually make for the home of a married sister, where they
are certain to be well received. Here they are respected as
jiciengthu, in-laws, and their children accepted as gaat ndr,
children of the mother's brother, while the people they join are
to them cieng conyimar, the people of my sister's husband, and
gaat nyal, children of a female agnate, and gaat waca, children
of the father's sister. A man who changes his residence thus
becomes a member of a different community and enters into
intimate relations with the dominant lineage of that community.
When a gaatiek man said to me, 'Now that I have come to
settle in Cieng Kwoth I am a man of Cieng Kwoth ', he meant
that outside ceremonial situations he identified himself with the
KWOTH lineage rather than with his own.
Even one man is a potential lineage and several brothers even
more so. A minimal, and then a minor, lineage comes into being
which has only ritual status towards the other lineages of its
clan, whereas with the people in whose village and district its
members have grown up it has a mutuality of interests and a
community of experience. The group thus develops into a dis-
tinct lineage. It intermarries with the other people of its home
and very often it intermarries so frequently with the dominant
lineage of the district that further marriage between them
becomes impossible without breaking incest regulations. In
this way lineages twine around one another and a texture of
cognatic relationships unites aU members of the community.
Only a few such lineages establish themselves and survive as
lineages. Many either die out or lose much of their individuality
PLATE XXII
«<t ^a j r » . ' »
rt. Spearing hsh from dam (Eastern Gaajok)
JK.
tf'rt i?rii
H
niaMMaBliI**
if. Harpoon-fishing in Lake Eadoi (Lou)
THE LINEAGE SYSTEM 211
and become attached to larger and stronger lineages by processes
which we will explain later.
Nevertheless, a Nuer lineage never entirely merges into
another clan. There are always certain ritual observances
which cannot be shared. If intermarriage creates community
ties between two lineages, it likewise keeps them distinct, since
one may only marry those who are not of one's agnatic group,
and even if two lineages intermarry to such an extent that cog-
natic relationship is a bar to further marriage between them,
then each may marry into collateral lineages of the other's clan
and children of their daughters. If a mythological link forges
a closer union between them it is also a record of their different
lines of descent, explaining, indeed, how people of divergent
descent are living together amicably. Thus a lineage, however
far removed from its homeland and however widely separated
from its kinsfolk, never becomes entirely absorbed or loses its
ritual heritage. A lineage can only merge with a collateral
lineage of the same clan.
But though lineages maintain their autonomy, the lineage
value only operates in the restricted field of ceremonial and is,
therefore, only occasionally a determinant of behaviour. Com-
munity values are those which constantly direct behaviour, and
these operate in a different set of social situations to lineage
values. While lineage values control ceremonial relations be-
tween groups of agnates, community values control political
relations between groups of people living in separate villages,
tribal sections, and tribes. The two kinds of value control dis-
tinct planes of social life.
It is, as we explain in the following sections, only the close
association between a tribe and its dominant clan, and lineages
in one way or another related to this dominant clan, that makes
the agnatic principle in lineage structure politically important,
for these lineages function as values in the political system
which gives them corporate substance.
VI
In spite of so much dispersion and admixture of clans there is
in every tribe a definite relation between its political structure
and the clan system, for in each tribe a clan, or a maximal lineage
212 THE LINEAGE SYSTEM
of a clan, is associated with the poHtical group in which it
occupies a dominant position among the other agnatic groups
that Hve in it. Moreover, each of its segments tends to be asso-
ciated with a segment of the tribe in such a way that there is a
correspondence, and often linguistic identification, between the
parts of a clan and the parts of a tribe. Thus, if we compare
the diagram on p. 144 with that on p. 193 and suppose clan A to
be the dominant clan in tribe B, then maximal lineages B and C
correspond to primary sections X and Y, major lineages D and
E correspond to secondary sections X' and X^, major lineages
F and G correspond to secondary sections Y' and Y^, and minor
lineages J and K correspond to tertiary sections Z' and Z^,
Minimal lineages are associated with villages composing the
tertiary sections. It is for this reason that we have spoken of the
dominant clan as forming a framework on which the political
system of the tribe is built up through a complex series of kin-
ship links. The system of lineages of the dominant clan is a con-
ceptual skeleton on which the local communities are built up
into an organization of related parts, or, as we would prefer to
state it, a system of values linking tribal segments and provid-
ing the idiom in which their relations can be expressed and
directed.
The J IN AC A of Lou country lived originally with their kinsmen
the jiNACA of Rengyan country to the west of the Nile, but they
spHt off from them and, having crossed the NUe, conquered present-
day Lou country. Here they were the first occupants, or at any
rate the strongest element among the first occupants. Probably
JINACA men frequently crossed the Nile to join their kinsmen to
the east and amalgamated with them. Such persons would at once
be members of the dominant clan and be spoken of as did, members
of the aristocratic group of lineages. But members of other clans
which settled in Lou country during or after the occupation were
classed as strangers {rul). In the same manner those lineages of
the GAATGANKiiR clau who crossed the Nile and settled to the
north of the Sobat held there a privileged position among other
Nuer who joined them.
Every Nuer tribe has in this way its diel, its superior clan,
though in the case of some tribes we are not certain of the correct
designation of these clans. Among the Gaawar they are the gaa-
WAR, among the Thiang the (gaa)thiang, among the Leek the
GAATBOL, among the Wot, and probably also among the Ror, the
THE LINEAGE SYSTEM 213
jiDiET, and among the Begh the JIKOI, &c. Where the correct clan
name is unknown one may refer to these dominant elements as the
aristocrats {diet) or the 'bulls' {tut) of such-and-such a tribe, as the
Nuer themselves frequently do, e.g. a dil or tut Bura, an aristocrat
of the Bor tribe, dil or tut Wotni, an aristocrat of the Wot tribe,
dil or tut Beeka, an aristocrat of the Beegh tribe, dil or tut Laka, an
aristocrat of the Lak tribe, &c. One may always speak of an
aristocratic clan or lineage by its proper name or by reference to
the tribe in which it has a privileged position, e.g. gaatn aca, children
of Nac, or diel Looka, aristocrats of the Lou tribe, gaatbol,
children of Bui, or diel Leegni, aristocrats of the Leek tribe.
There are four essential points to be remembered about these
aristocratic clans, (i) Not every clan has superior status in a
tribe. Some clans, e.g. the jimem and the jakar, have no wee,
no local community, as the Nuer say. Others have village-sites
where they have resided for a long time and which are called
after them, but they are not diel in the tribe where these sites
are found. Many are like cuttings which have lost all contact
with the parent stem, but these may at once be designated
Dinka lineages that have sprung from immigrants, whose des-
cendants sometimes know the Dinka country of their origin but
do not know their position in its system of lineages. Conse-
quently they cannot trace ascent for as many generations as true
Nuer, and their lineages are narrower in range and territorially
more restricted. One finds them in little local pockets in a
single tribe, whereas the Nuer clans are distributed throughout
many tribes.
(2) Not every member of a Nuer clan lives in the tribe where
it has superior status, for most clans are found in all parts of
Nuerland. Most jinaca live in Lou and Rengyan tribal areas,
where they are diel, but many are also found among the Eastern
Jikany tribes and elsewhere. Likewise, most gaawar live in
Gaawar country, but they are also found in most, perhaps all,
Nuer tribes. Tribes are territorial groups with an unbroken
social extension, whereas clans are kinship groups dispersed
far and wide. Consequently a tribe is a community and can
have corporate functions, but a clan is never a community and
can never act corporately. The Lou tribe unite for war. The
jinaca never unite. Also a man can change his tribe by chang-
ing his place of residence, but he can never change his clan. A
214 THE LINEAGE SYSTEM
Lou man who goes to live in Gaawar country becomes a Gaawar
man. A jinaca man remains a nac wherever he hves.
(3) A clan is, as we have earlier emphasized, not numeri-
cally preponderant in the tribe where it is dominant, e.g. the
JINACA are only a small minority in the Lou tribe and the
GAAJOK lineage are only a small minority in the Gaajok tribe.
(4) A man is a ^z/^. aristocrat, only in the one tribe where his
clan has superior status. Thus a dil Leegni, an aristocrat of the
Leek tribe, is an aristocrat there and nowhere else. If he goes
to live in Bui country or in one of the Jikany countries he is no
longer a dil but a rul, stranger. In the same way a member of
the JINACA is a dil in Lou country, but if he goes, as many of his
clan do, to reside in Gaajok country, he is not a dil there but a
stranger, rul. The jinaca are diel Looka, but the gaajok are
diel Gaajok. The status of diel is dependent on residence on land
owned by their clan. The only exception to this statement is to
be found in such cases where a clan is dominant in two or more
tribes, e.g. the jinaca in Lou and Rengyan and the gaatgan-
KiiR in the Jikany countries to the west and to the east of the
Bahr el Jebel. If a man of the jinaca moves from Rengyan to
Lou he still remains a dil, because both countries belong to his
clan. Likewise a man of the gaajok lineage of the gaatgan-
KiiR can move from Gaajok to Gaagwang, or Gaajak, country,
on either side of the Nile, and still remain a dil, for various
lineages of his clan are dominant in all these tribes.
Like most words which denote sociological status, dil is used by
Nuer in various contexts with various meanings. In this book it
is used with the precise meaning given to it in the preceding
paragraphs. It is, however, permissible to use the word to denote
a true member of any lineage, whether dominant in the tribe or
not. For example, the juak lineage of the Leek tribe are of Dinka
origin and are not diel Leegni, but a man who dwells among them
can obviously be either a dil Juaka, a true member of the juak
lineage, or a person who has attached himself to them for some
reason or other. In the same manner one may speak of a dil
jiMEM, although there is no tribe in which the jimem have the
status of diel, for a man may be a true member of the clan or a
Dinka who has attached himself to it. Also a member of the jinaca
who leaves Lou and settles in Gaajok still remains a gat dila Looka,
a child of Lou aristocracy, and will call himself such, meaning that
when he is in Lou country he is an aristocrat.
THE LINEAGE SYSTEM 215
The words 'hit' or 'gat twot' , son of a bull, are used in the same
manner as ' dW and 'gat dila' . Here again a man may be a tut of
his lineage, in contrast to accretions of strangers and Dinka who
live in the district associated with it, without being a tut of the
tribe of which that district forms part, e.g. a man may be a tut
of the Jaajoah secondary section of the Lou tribe without being
a tut Looka, because the aristocratic clan of the whole of Lou
country are the jinaca, and the jaajoah are not members of this
clan. In other words a man may call himself a tut of the Jaajoah
to emphasize that he is a jaajoah and not merely a Jaajoah, but
he does not thereby imply that he is a tut Looka. In a still more
general sense tut may, as we have noted, merely mean a pater-
familias or even a male person. One must judge from the context
how the expression should be translated. In this book we use
the words tut or gat twot in the sense, defined on pp. 179-80, of an
elder, and we use the word dil to refer to a tribal aristocrat.
It is difficult to find an English word that adequately describes
the social position of diel in a tribe. We have called them aristo-
crats, but do not wish to imply that Nuer regard them as of superior
rank, for, as we have emphatically declared, the idea of a man
lording it over others is repugnant to them. On the whole — we
will qualify the statement later — the diel have prestige rather
than rank and influence rather than power. If you are a dil
of the tribe in which you live you are more than a simple
tribesman. You are one of the owners of the country, its village
sites, its pastures, its fishing pools and wells. Other people live
there in virtue of marriage into your clan, adoption into your
lineage, or of some other social tie. You are a leader of the tribe
and the spear-name of your clan is invoked when the tribe
goes to war. Wherever there is a dil in a village, the village
clusters around him as a herd of cattle clusters around its bull.
I have described the position of diel as I judged it to be among
the Lou tribe. I had the impression that to the west of the Nile
their status was less pronounced, whereas among the Eastern
Jikany tribes, on the periphery of Nuer expansion, it was more
emphasized. In the Karlual area of the Leek tribe, the only part
of western Nuerland I know more than superficially, the aristo-
cratic prestige of a dil is recognized, but there are clans of
strangers so well and long entrenched in the districts and villages
in which they are found to-day that a dil has no legal privilege.
2i6 THE LINEAGE SYSTEM
At a cursory glance, I gathered that conditions were elsewhere
the same in western Nuerland, except, perhaps, among the
Jikany where the status of diel may be more enhanced. Among
the Eastern Jikany one finds the greatest insistence on social
differentiation and legal privilege. The category of diel tends to
be more stressed in the larger tribes than in the smaller, and
when its structural function has been examined it will be under-
stood why this is so.
In a Nuer village or cattle camp there are seldom more than
a few families of diel. Most people are either rul, Nuer of other
clans, or jaang, persons of Dinka descent who have not been
adopted into Nuer lineages. A rul is a Nuer who in a certain
tribe is not a dil, though he may be a dil in another tribe. I have
already described how lineages split off from their localized
kinship groups, wander, join persons of other clans, and become
members of a new community. This process was well described
to me by a man of the Dok tribe. Members of a lineage beget
children and they become numerous and spread over the country-
side, wandering here, there, and everywhere. Then their close
relationship ends and they go to live in the midst of other clans
who are distantly related to them. Here they dwell as friends
and slowly forge new cognatic relationships by intermarriage.
Hence lineages are much mixed in all local communities.
Also Nuer say that no dil dweUs in a social milieu composed
entirely of fellow aristocrats, for lineages of diel split up and
segments seek autonomy by becoming the nuclei of new social
agglomerations in which they are the aristocratic element. Thus
lineages of diel split up not only on account of internal dissension,
but because a man of personality likes to found his independent
settlement where he will be an important person rather than
remain a younger brother in a group of influential elder relatives.
I was told that this process by which any man, especially a dil,
could become a local leader is felt to be ingrained in their social
system and is a reason why they object to the creation by the
Government of a few local 'chiefs' whose position tends to
become formalized, permanent, and hereditary. To them this
is a rigid interpretation of status, based on territorial rather
than personal qualifications, which stabihzes the superiority
of a single man or lineage. Every man of standing feels that
PLATE XXIII
a. Section of camp kraal (Lou)
b. Sobat ri\-cr in the dry season (Lou)
THE LINEAGE SYSTEM 217
he should be a 'chief. It does not follow that a man must be
an aristocrat to gain influence among his fellow villagers. He
may be a tut of some other lineage than the dominant one of his
tribe who, by outstanding character, has been able to establish
himself and his kin as social leaders in his locality.
There is thus in every tribe some differentiation of status,
but the people so differentiated do not constitute classes, and
'strangers ' and ' Dinka ' are properly to be regarded as categories
rather than as groups. Their relation to the aristocrats in the
tribal system and the ways in which the different elements
are integrated into communities will engage our attention in
later sections.
VII
It is only in assessment of blood-cattle that social differentiation
between aristocrats and Nuer strangers is of great significance, and
then only among the Jikany tribes, principally among the Eastern
Jikany . Among the Eastern Jikany the kinsmen of a dil, aristocrat,
who was killed had to be compensated by payment of more cattle
than the kinsmen of a slain rul, stranger, ox jaang, Dinka. It is not
easy to know to what extent this privilege could be enforced, or
even to discover the relative assessments of blood-cattle. There
was undoubtedly considerable elasticity in reckoning who counted
as equivalent to a dil in situations of homicide and variation in the
number of cattle paid for homicide. It was asserted by several in-
formants that in pre-government days a true Nuer, called in the
Jikany tribes a gat Geeka, counted the same as a dil Jikany for
purposes of assessing blood-cattle, and I was given the following
blood-cattle assessments : a Jikany aristocrat, 40, a Nuer stranger,
40, a Dinka adopted by a Jikany aristocrat, 20, a Dinka who had
been adopted by a Nuer stranger, 20, and an unadopted Dinka, 6.
In more recent practice the following payments seem to have
been regarded as usual: a Jikany aristocrat, 20, a Nuer stranger,
17, a Dinka who has settled permanently in the country, 16, a
Dinka who has not been long settled, 10.
I was not long enough among the Eastern Jikany to investigate
this matter thoroughly, but I received the impression that the
second list was coloured in its variation of assessment, as it was
certainly influenced in amounts of assessment, by recent Govern-
ment decisions ; though the informants who gave it insisted that
it was in proportion to the ancient assessments. Aristocrats
assured me that in the past more cattle were always paid for the
homicide of an aristocrat than for that of a Nuer stranger. Members
4507 p f
2i8 THE LINEAGE SYSTEM
of Nuer clans, other than the gaatgankiir, who have settled in
Jikany tribes equally assured me that the payments were the
same for both. No doubt they were both hoping to influence
Government practice by their statements. On the whole, I consider
that among the Eastern Jikany there was probably a difference
between payments for homicide of an aristocrat and of a stranger,
but that there was much flexibility in the assessments, which
depended generally on the special circumstances of each case : the
length of the dead man's residence in the country, marriage
alliances between his family and aristocratic lineages, the strength
of his lineage and local community, whether he had been killed by
a member of his village or by a man of another village, and so forth.
Probably the same is true of the Western Jikany tribes, where I
was told that the customary compensation for an aristocrat was
from 40 to 50 head of cattle, for a stranger or Dinka who had settled
in the country, 30, and for a Dinka who was staying there but had
not yet built a homestead, 20. The practice of the Jikany is not
typical of Nuerland as a whole.
Nevertheless, all over Nuerland Nuer and Dinka were differen-
tiated by assessment of their value in blood-cattle, though the
definition of a Dinka, in this respect, varied in different tribes.
Among the Lou the practice was to reckon aristocrats and
strangers together at 40 head of cattle. A Dinka who was born
in Lou country was said to have become a Nuer [caa nath) and
a member of the community in which he was living {caa ran wee),
so that his life also was assessed at 40 head of cattle. On the
other hand, a Dinka who had been captured in war and brought
to Lou country was reckoned at 16 head of cattle, while a Dinka
visiting relatives or affines in Lou country was reckoned at
only 6 head of cattle. I was told that a Dinka who had been
adopted held an inferior position in this respect to that held by
his children, who counted as true Nuer. In the Jagei tribes
Nuer strangers and Dinka who were permanent members of a
community appear to have been reckoned at 40 head of cattle,
like an aristocrat, while a stranger or a Dinka who had not
built a byre was reckoned at only 10.
The building of a byre was stressed because a man who
built a homestead in a village had clearly the intention of
remaining there and the community gained by an addition
to its herd. Such a man seems to have been reckoned equal
to an aristocrat in all parts of Nuerland, except among the
THE LINEAGE SYSTEM 219
Jikany where, I was told, a Dinka could never lose his inferior
status, which was transmitted to his descendants. The accep-
tance of a permanent member of a community as equal to an
aristocrat conforms to the general tendency among the Nuer
for descent to be subordinated to community, a tendency which
we shall constantly stress.
We again emphasize the point to which we drew attention in
the section on law, that the degree of responsibility acknowledged
for a tort, the chances of compensation being offered for it, and
the amount of compensation paid, depend on the relations of
the persons concerned in social structure. Thus, if a man kills
an unadopted Dinka of his own household, who was not born
in Nuerland, there is no redress, but his household will protect
their Dinka against outsiders and will avenge his death at their
hands. Dol, who is himself of Dinka descent, told me: Tf you
curse a Dinka of your household, well, you curse him, that is
all. If he is angry you tell him that you will kill him and that
nothing will happen. You will merely clean your spear on the
ground and hang it up. But if another man curses a Dinka of
your household you fight him, for the Dinka is your brother.
You ask the man whether he is your Dinka or his Dinka.'
The position of a Dinka in his own domestic circle is thus
different from his position in relation to members of a wider
group. He is only a. jaang to the joint family which considers him
'their man.' To people standing outside this joint family he is
a member of that gol, joint family, and it is not their business
to differentiate his status within it. I was told that if an out-
sider called such a Dinka 'jaang', the sons of the man who had
captured him would resent the insult and might start a fight to
wipe it out, for to them he is ' deniar' , 'my brother', in relation
to outsiders. They ask, ' Who is a jaang ? Did your father seize
him or did ours ? ' The acceptance of Dinka born in Nuerland
as full members of Nuer households, hamlets, and villages is
even more pronounced.
The status of Dinka is thus a relative one, and a man may be
considered to belong to this category in one situation and not in
another. This is obviously the case in social life generally,
because ordinarily no one differentiates a man of Dinka descent
from a man of Nuer descent, but we believe that it was also the
220 THE LINEAGE SYSTEM
case in questions of homicide, since the social situation was
composed of the structural relations of the slayer and his kin to
the dead man and to other people concerned in the dispute.
In our opinion the uncertainty and contradiction which was
often evident in the statements of Nuer about assessments of
blood-cattle is to be accounted for by relativity of status, it being
relative always to the structural distance between persons, and
therefore not rigidly definable.
Likewise ml is a very plastic concept. If a Leek man goes to
Gaajak country to steal cattle and is killed no compensation
would be paid for the homicide. A Leek man travelling through
Gaajak country without intent to cause loss to its owners would
not wantonly be slain. If the man were visiting kinsmen or
affines and was killed in a quarrel, his hosts would consider that
they were under an obligation to avenge him, though, perhaps,
not under a very strong obligation. But a Leek man who has
built his homestead in Gaajak country and has married into
the viUage where he resides is a member of that community.
If another member of his village kills him, it may be held that
he is a rul and therefore his death may be paid for with fewer
cattle than that of an aristocrat. But if a member of another
village kills him, his community are not likely to accept this
definition of status, because one does not differentiate between
members of one's community on grounds of descent in its rela-
tion with other political segments. In political relations com-
munity ties are always dominant and determine behaviour.
VIII
CWe have observed that within a tribe there are three cate-
)ries of persons : diel, rul, and jaang. The diel are an aristo-
cratic clan, numerically swamped in the tribe by strangers and
Dinka, but providing a lineage structure on which the tribal
organization is built up. The problem is how strangers and
Dinka are attached to the dominant clan in such a way that the
clan becomes, through the relations of other members of the
tribe to it, the framework of the political system. As Nuer
reckon all social ties in a kinship idiom it is clear that only the
recognition of mutual bonds of kinship could lead to this result.
Such recognition is accorded in several ways. We will start by
THE LINEAGE SYSTEM 221
an examination of adoption. A Nuer cannot be adopted into a
lineage other than that into which he is born, so that the custom
concerns only Dinka.
We have already described how Nuer scorn Dinka and per-
sistently raid them, but they do not treat those Dinka who are
permanent members of their community differently from its
Nuer members, and we have seen that persons of Dinka descent
form probably at least half the population of most tribes. These
Dinka are either children of captives and immigrants who have
been brought up as Nuer, or are themselves captives and immi-
grants who are residing permanently among Nuer. They are
'Jaang-Nath', 'Dinka-Nuer', and, it is said, 'caa Nath', 'they
have become Nuer'. As we have explained, once their member-
ship of a community is recognized, in most of Nuerland, their
legal status is the same as that of a free-born Nuer, and it is only
in relation to ritual and rules of exogamy that attention is drawn
to their origin. In structural relations of a political kind they
are undifferentiated members of a segment. Although in his
domestic and kinship relations a Dinka has not so strong a
position as a Nuer, because he has not the same range of kinship
links, I have never observed that they suffer any serious disa-
bilities, far less degradation. In answer to my question whether
a captured Dinka would not work harder in the kraal than a son
of the family, I was told that he was a son and would enjoy the
same privileges as the other sons, being given an ox by his father
at initiation and, later, bride-cattle for a wife. The only
foreigners who suffer serious social inequality are certain small
pockets of Dinka and Anuak who have been conquered but not
absorbed into Nuer society and culture. Such pockets, like the
Balak Dinka and the Anuak on the Sobat river, enjoy neither
the privileges of Nuer citizenship nor the freedom of foreigners.
Such pockets do not truly constitute part of a Nuer tribe.
Captured Dinka boys are almost invariably incorporated into
the lineage of their Nuer captors by the rite of adoption, and
they then rank as sons in lineage structure as well as in family
relations, and when the daughters of that lineage are married
they receive bride-cattle. A Dinka boy is brought up as a child of
his captor's household. He is already incorporated into the
family and joint family by his acceptance as a member of these
222 THE LINEAGE SYSTEM
groups by their other members and by outsiders. People say
' caa dil e cieng' or ' caa ran wee' , 'he has become a member of
the community ', and they say of the man who captured him that
'he has become his father', and of his sons that 'they have
become his brothers'. He is already a member of the gol, the
household and joint family. Adoption gives him a position in
lineage structure, and thereby ceremonial status, for by adoption
he becomes a member of his captor's thok dwiel, lineage.
I was told that the captor will seldom himself give the lad huth
agnatic affiliation to his lineage, and that the rite is usually per-
formed by a kinsman at the request of his sons and with the consent
of their minimal lineage. A representative of the lineage invites
the Dinka, now grown-up and initiated, to attend the sacrifice of
an ox or sheep in his kraal. The head of the joint family provides
the sacrificial beast and the representative of the lineage drives the
tethering-peg into the earth at the entrance to the byre and walks
up and down the kraal invoking the spear-name of the clan and
calling on the spirits and ancestral ghosts of the lineage to take
note that the Dinka is now a member of it and under their protec-
tion. He then spears the beast and the Dinka is smeared with the
undigested contents of its stomach while the ghosts and spirits are
asked to accept him. He is especially smeared on the soles of his
feet, for this binds him to his new home. If he leaves it he will die.
The beast is then cut up, and a son of the house, or the representa-
tive of the lineage, and his new brother divide the skin and the
scrotum, which the Dinka cuts. The Dinka also takes the neck as
his portion. On all future occasions when animals are sacrificed by
members of the lineage the Dinka will receive his share of the meat,
for he is now a member of the lineage. The cutting of the scrotum
is the symbolic act which makes the man a member of the lineage,
because only an agnatic relative may cut the scrotum of a sacri-
ficial beast. 'A man who has cut the scrotum of your beast, if he
has sexual relations with your daughter, he will die.'
A girl captive is not adopted into the lineage, but people say
'caa lath cungni', 'she is given a right to receive bride-wealth '. ' Her
children are become people who partake in the bride-cattle'. This
means that when she is married, or her daughters are married, the
sons of the family in which she has been brought up will receive the
cattle due to brothers and maternal uncles, and that in return, when
the daughters of the sons and daughters of this family are married,
she, or her sons, can claim the cow due to the paternal aunt and the
cow due to the maternal aunt . She has become a daughter to her cap-
tor and a sister to his sons, but she is not a member of their lineage.
PLATE XXIV
n
A leopard-skin chief
THE LINEAGE SYSTEM 223
By adoption Dinka men are grafted on to the lineage of their
captors. They trace their ascent up the Hneage to its ancestor
and they become a new point of its growth. The fusion is
complete and final. The spirits of the lineage become their
spirits, its ghosts become their ghosts, and its spear-name and
honorific name become their symbols. Indeed it is almost
impossible without a prolonged stay in a Nuer village or camp
to discover who are and who are not of pure Nuer origin. I have
for weeks considered men to be true Nuer, whereas they were
descended from captured Dinka, for a man whose Dinka grand-
father had been adopted into a Nuer lineage regards himself
as being just as much a member of the lineage as the man whose
grandfather adopted his grandsire, and he is so regarded by
other members of the lineage and by persons not of it. Thus,
when a man gives his descent from E through D and C, and
another man gives his descent from E through J and K, one
naturally assumed that D and J were sons of E. There is no
means of knowing that, in fact, J was a captured Dinka who
was adopted into the lineage, unless someone volunteers the
information — a most unlikely happening in Nuerland. Moreover,
it is impohte to ask strangers whether their grandfathers were
captured Dinka, and, even if they were of Dinka origin, they
would not readily say so. One can, of course, always ask other
people ; but only those who are members of the same lineage are
likely to be fully acquainted with the man's ancestry, and they,
in all probability, will not tell one if he is of Dinka origin, for he
is their agnatic kinsman as far as outsiders are concerned.
A very large number of Dinka in all tribes have been incor-
porated by adoption into Nuer lineages. Since, as mentioned
later, adopted Dinka and their descendants can marry into
collateral lineages, it would not be accurate to say that they are
adopted into clans. Probably most captured Dinka were adopted
into Nuer lineages, but there are also many Dinka lineages
descended from men who came of their own free will to settle in
Nuerland, either to escape famine, largely caused by Nuer raid-
ing, in their own country, to visit captured sisters, or to reoccupy
the sites from which Nuer raids had ousted them. Such immi-
grants were unmolested and were permitted to settle or return to
Dinkaland as they might choose. A Dinka who decided to settle
224 THE LINEAGE SYSTEM
became some Nuer's jaang, his Dinka, and rande, his man, and
the Nuer would give him an ox and maybe a cow or two when
he had given proof of his fideHty and attachment to his new home.
I was told that he might even be given a daughter of the house in
marriage without payment of bride-wealth if she were blind or
lame and no Nuer would contemplate taking her as a bride. Often
a widow lives in concubinage with such a Dinka, who thus obtains
a 'wife ' in the sense of cook, housekeeper, and mate ; and even
if the children she may bear him do not count as his descen-
dants he can gain their affection. If a Dinka settles at the home
of the husband of his sister, the husband may give him a cow or
two in acknowledgement of affinity.
There must also have been pockets of the original Dinka
occupants of country overrun by the Nuer who submitted and
gave up their language and habits in favour of those of the
Nuer. At any rate, there are to-day in all tribes many small
Dinka lineages, and villages are often called after them. Such
lineages are numerically preponderant in the communities where
I spent most of my time, Yakwac camp and Nyueny village.
The way in which these lineages are woven into the lineage
texture of the dominant clan of the tribe is discussed in the next
two sections.
Here we summarize points which have already emerged from
our description of thepositionof Dinka in relation to Nuer.^
(i) Jaang, Dinka, has many meanings : any foreigners whom the
Nuer habitually raid ; Dinka living in Dinkaland and raided by
Nuer ; Dinka of unabsorbed pockets in Nuerland or on its con-
fines ; recent Dinka immigrants ; certain clans which are said to
be Dinka in origin, e.g. the gaatgankiir; members of small
Dinka lineages which are Nuer in every character except in
origin ; descendants of adopted Dinka ; adopted Dinka. One can
only judge the meaning Nuer attach to the word by its context
and the tone in which it is spoken. (2) It is only those Dinka .
who are regarded as members of a Nuer tribe who concern .
our present discussion. Their status is relative to the social
situation in which the question of status arises, and cannot be
rigidly defined. (3) Nuer conquest has not led to a class or sym-
biotic system, but, by the^ustom ol_adpption.,..has absorbed ^
jthe.i:;Qiiq.Ufiieji_Dinka into its kinship system, and through
THE LINEAGE SYSTEM 225
the kinship system has admitted them into its poHtical structure
on a basis_ot equalit}^" ' ' "" """"* ' " *
IX
A large number of Dinka who were not captured as children
are not adopted into Nuer lineages, and Nuer strangers cannot
be adopted into lineages of the dominant clan or into any other
Nuer lineages. Nevertheless, members of all local communities,
whilst they see themselves as distinct segments in relation to
other local segments, express their relations to one another in
terms of kinship. This is brought about by intermarriage.
We m&nti'on the ru1gsjTfjeyoggjIiy_g^JlHgfly at; po'^t^ible and
on ly in so far as t hey have a direct bea r ing on the political
^systguL Nuer generally marry within their tribe, though they
sometimes marry women of other tribes, especially if they live
near the border. Sometimes, also, a man marries in one tribe
and then, taking his wife and family with him, goes to live in
another tribe. In recent times there have been occasional
marriages with the Ngok, and possibly with other Dinka tribes.
There are no exogamous rules based on locality. They are
determined by lineage and kinship values. A man may not marry
into his clan and, a fortiori, into his lineage. In most clans a
man may marry into his mother's clan, but not into her maximal
lineage, though this rule is less exact. A man may not marry
any woman to whom he is in any way closely related. A Dinka
adopted into a lineage may not marry into that lineage, but
may marry into collateral lineages of the same clan.
The rules of exogamy have been cursorily described. Never-
theless we consider them important, for the values which chiefly
regulate behaviour between one person and another in Nuer
society are kinship values. Nuer rules of exogamy break down
the exclusiveness of agnatic groups by compelling their members
to marry outside them, and thereby to create new kinship ties.
As the rules also forbid marriage between near cognates, a small
local community like a village rapidly becomes a network of
kinship ties and its members are compelled to find mates out-
side it. Any stranger who enters the village, if he is not already
related to most of its members, rapidly enters into affinal
relations with them and his children become their kinsmen.
Consequently the population of a Nuer village or cattle camp
4S07 (^ g
226 THE LINEAGE SYSTEM
can be placed on a single genealogical chart showing lines of
descent and affinity, and, since affinity is fundamentally a
relationship through kinship, we may say that all members of
a village or camp are united by kinship ties and, therefore, are
generally unable to marry into it. Consequently they are
forced to take spouses from neighbouring villages of their
district. Normally a man marries a girl who lives within visiting
distance of his village. Hence a network of kinship ties stretches
over a district and links in diverse ways the members of distinct
political groups.
Looked at from the angle of a single village the circle of close
kinship relations is limited to a small radius, and they tend to
become fewer and more distant the nearer its periphery is
approached. But the circumference of one such circle is
intersected by other circles, so that there is no limit to the
extension of a continuous series of kinship links. Exogamoiis.
rules, therefore, prevent the formation of autonomous agnatic
groups and create extensive kinship ties within, and beyond,
the tribal structure. Thus the kinship system bridges the gaps
in political structure by a chain of links which unite members
of opposed segments. They are like elastic bands which enable
the political segments to fall apart and be in opposition and yet
keep them together. This relation between kinship and political
structure poses a set of complex problems. Here we wish to
demonstrate only one point: the way in which dominant
lineages serve as a political framework by the accretion of other
lineages to them within local communities.
We have seen how every local community is associated with a
lineage and that the members of this lineage who live in the
community are numerically swamped by members of other
lineages. We have also seen how all members of the commun-
ity are in some way or other related by kinship. What gives
a pattern to this complicated criss-crossing of cognatic threads
is their relation to the dominant lineage of the community.
The Nuer have a category of gaat nyiet, children of girls,
which includes all persons who are in the relationship of
sister's son and daughter's son to a lineage. As a whole lineage
can be spoken of as gaat nyiet to another if there is one such
female link between them anywhere in the Unes of their descent,
THE LINEAGE SYSTEM 227
and as there must be such a hnk if they live in the same com-
munity, owing to the rules of exogamy, it follows that people
who live together are all gaat nyiet to one another. However, it
is in relation to the dominant lineage of a community that the
concept is mainly employed and is politically important. When
people are not members of this lineage it is stressed that they
are gaat nyiet to it. Nuer of other clans can never more closely
identify themselves with the dominant lineage, because, for
ritual reasons, they must remain autonomous units, but
politically they accrete themselves to it through this kinship
category. Moreover, outside ritual situations, being gaat nyiet
to a dominant lineage gives people complete equality with it,
and their accretion to it is often expressed in terms of lineage
structure, so that a man will often give his ascent to the woman
of the dominant lineage who bore one of his ancestors, and thus
graft himself, through her, on to their tree of descent ; though
this is more usually done by Dinka than by Nuer. It is, how-
ever, the common practice for children of strangers who have
been brought up at the home of their maternal kinsmen, who
are aristocrats, to regard themselves as members of their
mother's lineage, except in ceremonial situations, and to
consider its members, rather than their father's lineage, as their
true kinsmen.
Dinka who have not been adopted commonly trace their
ascent to a female Nuer forebear and through her they graft
themselves into a Nuer lineage and are accepted as members of
it in ordinary social relations. Thus a Dinka often gives his
ascent to the dominant lineage of his community through a
woman, and sometimes through two or three female links, and
though this is generally evident from female prefixes it cannot
always be known. These Dinka individuals incorporate them-
selves into the structure of a Nuer lineage through their mothers,
since they have no lineage structure of their own. This is
different from the stressing of a female link {gaat nyiet) which
unites a group of Nuer strangers or of Dinka to the dominant
lineage of their tribal section and also from matrilineal modes of
reckoning descent due to matrilocal conditions of residence, which
may be temporary.
Owing to exogamous rules lineages are thus linked by
228 THE LINEAGE SYSTEM
innumerable cognatic ties, so that however many Hneages there
may be in a local community their members are all related to
one another by some kind of cognation and affinity. A lineage
remains an exclusive agnatic group only in ritual situations. In
other situations it is merged in the community, and cognation
[mar] takes the place of lineage agnation {buth) as the value
through which people living together express their relations to
one another. The agnatic structure of the dominant lineage is
not stressed in ordinary social relations, but only on a political
plane where relations between territorial segments are con-
cerned, for the assimilation of territorial segments to segments
of the dominant lineage means that the interrelations of the
one are expressed in terms of the other.
In every small tribal segment there is a lineage of the dominant
clan of the tribe associated with it, and the members of the
segment are joined to this lineage by adoption, cognatic kinship,
or kinship fictions, in such a way that one may speak of them
as an accretion around a lineage nucleus. As these different
nuclei are lineages of the same clan, or, as we shall see in the
next section, assimilated to it, the structure of the dominant
clan is to the political system like the anatomical structure to
the system of an organism.
X
We have seen how Dinka and strangers are linked to the
framework of the dominant clan by adoption and cognation
and how these links form an embracing kinship system which
provides the non-political texture of the political system. Kin-
ship values are the strongest sentiments and norms in Nuer
society and all social interrelations tend to be expressed in a
kinship idiom. Adoption and the assimilation of cognatic to
agnatic ties are two ways in which community relations are
translated into kinship relations : in which living together forces
residential relations into a kinship pattern. A third way is by
mythological creation of kinship fictions, and this way is appro-
priate to relations between dominant lineages and stranger and
Dinka groups, living with them in the same tribal segments,
which are too large and occupy too distinct a territory for
incorporation by either of the other two methods. It is the way
PLATE XXV
a. Ngundeng's pyramid (Lou)
jawwPPWlJll'! jijli
6. >\.L;LiiidLii,L;'s jiyraiiiid (Luu)
THE LINEAGE SYSTEM 229
in which large pockets of strangers and Dinka are incorporated
into the conceptual scheme of a tribe.
It has frequently been emphasized that political relations
are often expressed in speech as lineage relations, in the sense
that one talks of a local community as though it were a lineage,
thereby assimilating to a dominant lineage those who share the
same community life with it; and that lineage relations are
often expressed as political relations, in the sense that one talks
of a lineage as though it were identical with the local community
in which it is only a nucleus, thereby depriving the lineage of
its unique agnatic status and giving it a general residential
value. In conformity with this way of describing community
interrelations, they are personified in myths and derived from
personal relationships of a kinship kind.
We do not propose to give a collection of Nuer myths. We
have so far only mentioned one myth explaining group inter-
relations : that which tells why the Nuer raid the Dinka. There
are very few myths of this general kind. Most relate to clans
and lineages in their corporate territorialized form and explain
their association with one another as tribes and tribal segments,
particularly the relations between dominant lineages and large
stranger lineages living with them. We are not always able
to explain mythological relations by the present-day political
system, but this can often be done, and where we fail to do it
we attribute our inability to ignorance, especially to ignorance
of tribal history.
The two large Lou tribal sections, the Jimac and the Jaajoah
which appear on the map on p. 56, but not on the clan tree of the
jiNACA, the dominant clan of the tribe, on p. 196, are divisions
called after the jimac and jaajoah lineages. They are said to be
gaat nyiet, children of daughters of the founder of the jinaca clan,
and there is a myth accounting for this maternal link. Denac was
said to have had, according to the Lou story, four sons, named
Yin, Dak, Bal, and Bany, by one wife, and Nyang and two nameless
brothers by a second wife. These wives are sometimes said to have
been called Nyagun and Nyamor and the two primary sections of
the tribe. Gun and Mor, to be named after them. Nyang's two
brothers were eaten by an ogre. When, afterwards, the sons of
Denac went fishing, the four sons of one mother went by themselves
and Nyang by himself, for he would not accompany his half-
230 THE LINEAGE SYSTEM
brothers, but pined for the sons of his mother. When he caught
a fish someone would come and steal it from him, for he was all by
himself and only a boy. When he came home he would not sit with
the other boys facing his father, but sat apart with his back to him,
and when his father asked him why he was troubled he replied that
he was thinking of his brothers whom the ogre had eaten. His
father said to him, ' Never mind, take your two sisters and let them
be your brothers.' So when Nyang went fishing he was accompanied
by his sisters Nyabil and Fadwai. Nyang is the founder of the
GAALiEK lineage, Nyabil of the jimac lineage, and Fadwai of the
JAAJOAH lineage. These lineages together form the kinship frame-
work of the Mor primary section of the Lou tribe and the myth
explains their association. This maternal link has not prevented
intermarriage between the gaaliek and the jimac. Apart from
questions of ritual and exogamy the descendants of Nyabil and
Fadwai are treated as though these daughters had been sons, and
they possess a mythological patent which gives them equal status
in the tribe with the diel. In tracing their agnatic ascent members
of these lineages do not go further back than their ancestress.
From her they continue to her father, Denac.
In the Gaawar tribe there is an important jakar lineage which
is mythologically attached to the gaawar, the aristocrats of
the tribe, in the following manner. A man called Kar, or Jakar,
descended from heaven by a rope that connected the sky with a
tamarind tree, probably the tree in Lang country beneath which
mankind is said to have been created. He was later followed by
War, the founder of the gaawar clan, who was found sitting in the
tree by Kar's sister who was gathering firewood accompanied by
her dog. She returned to tell her brother that she had found a man
whose head was covered in blood. Kar tried to persuade him to
come to the village but he refused to do so. They then sacrificed
an ox and roasted its flesh and the smell so attracted War, who
was very greedy, that he climbed down the tree and entered the
village. When he had eaten he wanted to return to heaven, but Kar
cut the rope. Mr. B. A. Lewis has kindly furnished me with what
he says is a less common version, found in the Gaawar tribe. War
fell from heaven in a rainstorm and was discovered by a dog which
belonged to Logh, but was with Kwec's wife when she was looking
for wood in the forest and War was found. Kwec's wife took him
home with her and a dispute arose between Kwec and Logh about
the ownership of the foundhng. Logh claimed War on the ground
that it was his dog which had discovered him, and Kwec claimed
him because his wife had found him. Then Kar joined in the
discussion, saying that War was his brother.
THE LINEAGE SYSTEM 231
This myth brings War and Kar and Logh into some kind of
relationship to one another and is to be explained by the fact that
the two main clans in the Gaawar tribe, next to the aristocratic
clan of the gaawar, are the jakar and the jalogh. The jalogh
are presumably the same clan as live to the south of Dok country
where a small territory is named after them. Kwec was, doubtless,
the founder of the kwec lineage after whom a small territory,
next to the Jalogh country, is named. We may surmise that, since
both lineages are found in Gaawar country at the present day and
in its present site to the east of the Nile, they also had close
relations with the gaawar when all three clans lived in their
homeland to the west of the Nile.
The richest clan mythology is that of the gaatgankiir, and
it clearly illustrates the mythological integration of lineages of
different origins to the dominant lineage system in a political
structure, and shows how territorial relations are given a kin-
ship value.
There are several versions of many of the incidents relating to
Kir, the founder of the gaatgankiir clan, and we give an abridge-
ment of one of these. A Dinka of the Ngok tribe, called Yul, saw
a stalk of a gourd on a river bank and, having followed it a long
way, arrived at a huge gourd. He cut this gourd open and out of
it came Kir with various ritual objects. Yul's wife suckled the
child as well as her own baby, Gying. When Kir grew up he turned
out to be a witch and magician and the sons of Yul tried to kill
him because his evil powers were destroying the cattle. Only
Gying remained Kir's friend, and said to him, as he fled from Yul's
home, that he would one day follow after him and join him.
In his flight Kir came to the Nile where he saw a man, called Tik,
in the river and asked him for help. Tik struck the waters of the
Nile and cut them in two and Kir crossed over to the west bank.
Kir told Tik that when he had found a place to settle in Tik was to
come after him. Tik accompanied Kir till they met a man of the
Wot tribe who took them to his home where the jidiet, the domi-
nant clan of the Wot, sacrificed a black ox so that the lethal power
of witchcraft might leave Kir's eyes and enable him to look at people
and cattle without killing them. Kir then dug a hole for himself in a
termite-mound near a cattle camp of the gaawar, where he per-
formed many strange feats. Eventually the gaawar offered sacrifices
and persuaded him to leave the mound and took him to their camp.
Kir was then given a wife, Nyakwini, who bore him Thiang
before he killed her with his witchcraft. He then married Nyabor
who bore him Kun. He likewise killed her. The people then gave him
232 THE LINEAGE SYSTEM
a lame woman, Duany, who bore him Jok. In Lou and Eastern
Jikany versions the three wives were all daughters of Gee, the
founder of the gaatgangeeka family of clans, and in versions to
the west of the Nile the first two were gaawar and Duany a nyapir
of the Bui tribe, but all accounts make Nyakwini and Nyabor more
closely related to one another than either to Duany. After Duany
had borne Jok she killed Kir with witchcraft, for she, also, was a
witch. Later Thiang, her dead husband's eldest son, cohabited
with her and begat Nyang.
In all the variants of the Kir myth the parts played by Gying
and Tik are stressed. Gying was suckled with him and after-
wards joined him and lived with him as a brother. When Kir died
his eldest sons Thiang and Kun possessed cattle, but the youngest
son, Jok, and Gying had no cattle. Thiang wanted to prevent
Gying from acquiring cattle, but Kun gave him some, so Thiang
said that Kun and Gying were to live together. Tik had saved
Kir's life and had gone to live with him. There is a further story
of how Gying and Tik were threatened by an ogre and shared a hut
together and became like brothers, so that the lineages descended
from these two men do not intermarry.
Without recording further details we may note how actual
political relations are mythologically represented in the characters
of these stories. The two largest segments of the Gaajak tribe
which are named after nuclei of strangers are the Kong section,
the stranger nucleus of which is a lineage descended from Tik, and
the Dhilleak section, the stranger nucleus of which is a lineage
descended from Gying, and these two sections live together as
parts of the Reng primary section (see diagram on p. 140, and
sketch-map on p. 58). The myth tells also how Jok and Nyang
are sons of the same mother, Duany, Jok being begotten by Kir
and Nyang by Thiang. This is a mythological representation of
the structure of the Gaagwang tribe which has dominant lineage
nuclei descended from both Nyang and Jok, and also of the
political relations between the Gaagwang tribe and the Gaajok
tribe, for these relations, especially to the west of the Nile, are of
close alliance compared to the more distant relations between
the Gaagwang tribe and the primary sections of the Gaajak tribe
that also border them, the Thiang and the Reng. Thiang and Kun
were begotten by Kir and borne by women who are generally
represented as sisters, and the sections in which their descendants
are dominant are the Thiang and Gaagwong primary sections of
the Gaajak tribe, the third primary section of which, the Reng, has
nuclei descended from Thiang, Gying, and Tik, whose relationships
in the myth have already been noted.
i
Hh
234 THE LINEAGE SYSTEM
In every Nuer tribe there are similar stories which explain
the relations between the aristocratic clan and large stranger
lineages living with it. Other myths explain the relations
between these stranger lineages. Thus the lineages living
at Nyueny and in neighbouring villages in Leek country,
JUAK, NGWOL, JiKUL, &c., are all mythologically related to one
another and to the dominant clan of the Leek tribe. These
myths also explain the ritual symbols and observances of the
lineages mentioned in them.
Actual interrelations of a political kind are thus explained
and justified in mythological interrelations, and wherever, as
far as we know, large lineages of different clans are politically
associated there is a myth bringing their ancestors into some
social relationship. This is especially the case between dominant
lineages and stranger or Dinka lineages, and the mythological
link gives them equality and fraternity in community life, while
permitting ritual exclusiveness and intermarriage between them.
Complete assimilation is impossible, for there must always be
ritual distinction or the clan and lineage systems would collapse.
Strangers have to be incorporated into the community of the
dominant lineage and excluded from its agnatic structure.
XI
By adoption, the recognition of the equivalence of cognatic
and agnatic ties in community life, and by mythological rela-
tionships, all persons in a tribal segment have kinship relation-
ships of some kind to one another and the segments themselves
are given a kinship relationship to each other within the political
system. Although the categories of diel, ml, and jaang create
social differentiation, it is on a ritual and domestic, rather than
on a political, plane and is only indicated in certain situations
of social life.
This is evident in the Nuer use of the three words denoting
the three statuses. It is a common Nuer practice when addressing
people and speaking publicly about them to use words which
denote a closer relationship between them and the speaker than
their actual relationship. This is commonly done with kinship
terms and, also, in defining the status of a person in his tribe.
Nuer do not emphasize that a man is a stranger or Dinka by
THE LINEAGE SYSTEM 235
alluding to him as such in ordinary social life, for it is in rare
situations that his being other than an aristocrat is relevant : to
some extent in payment of blood-cattle, in questions of exogamy,
and at sacrifices and feasts. A stranger who has made his home
with aristocrats is treated as a social equal and regards himself
as such. People do not call him a rul, for he is a member of their
community. They may even refer to him as a dil out of politeness.
In the same way people do not refer to an adopted Dinka as
'Jaang ' ; for he is by adoption a brother of aristocrats or of Nuer
of other lineages. One would not ordinarily speak of unadopted
Dinka residents as 'Jaang' , but as 'rul'. Just as strangers tend
to be linguistically assimilated to aristocrats, so Dinka tend to
be assimilated to strangers, and people speak only of unconquered
Dinka of Dinkaland by the contemptuous expression 'Jaang'.
Nuer do not make distinctions of status between people who
live with them, share their fights, partake of their hospitality,
and are members of their community against other communities.
Community of living overrides differentiation of descent.
We again emphasize that the designations 'aristocrat',
'stranger', and 'Dinka' in a Nuer tribe are relative terms, being
defined by the relations of persons in the social structure in
specific situations of social life. A man is a stranger, or Dinka,
in reference to a few, mainly ritual, situations, but is not indi-
cated as such on other occasions; and a man is a stranger or
Dinka in relation to members of a social group, but is not con-
sidered by them to have a differentiated status in relation to
another group. A stranger is a stranger to you, your stranger,
but is one of you vis-a-vis other people. A Dinka is a Dinka to
you, your Dinka, but he is your brother vis-a-vis other people.
In political structure all members of a segment are essentially
undifferentiated in its relation to other segments.
How can it be explained that among a people so democratic
in sentiment and so ready to express it in violence a clan is
given superior status in each tribe ? We believe that the facts
we have recorded provide an answer in terms of tribal structure.
Many Nuer tribes are large in area and population — some of them
very large — and they are more than territorial expressions, for
we have shown that they have a complex segmentary structure
which the Nuer themselves see as a system. As there are no
236 THE LINEAGE SYSTEM
tribal chiefs and councils, or any other form of tribal govern-
ment, we have to seek elsewhere for the organizing principle
within the structure which gives it conceptual consistency and
a certain measure of actual cohesion, and we find it in aristo-
cratic status. In the absence of political institutions providing
central administration in a tribe and co-ordinating its segments,
it is the system of lineages of its dominant clan which gives it
structural distinctness and unity by the association of lineage
values, within a common agnatic structure, with the segments
of a territorial system. In the absence of a chief or king, who
might symbolize a tribe, its unity is expressed in the idiom of
lineage and clan affiliation.
XII
In the Kir myth not only are the ancestors of important
lineages, and, through them, the lineages and the territorial
segments in which they are incorporated brought into relation-
ship, but the ancestors of clans, and, through them, the clans
and the tribes in which these clans are dominant, are linked
together. Thus Kir, in various versions of the myth, is adopted
by Gee, the founder of the gaatgangeeka family of clans;
meets Wot, in whom the Wot tribe is personified ; has relations
with the GAAWAR ; and so forth. The myth, thus, also mirrors
intertribal relations and brings the whole of Nuerland into a
single kinship structure, which we call the clan system as dis-
tinct from the lineage system of a clan.
The clan is the farthest range to which agnatic kinship is
traced when the marriage of two persons is in question, but
some clans have, nevertheless, an agnatic relationship to other
clans, though Nuer do not regard it quite in the same way as
relationship between lineages of a clan. They give the impres-
sion, when speaking of the ancestor of a clan, that they regard
him as an historical figure, clearly delineated against a back-
ground of tradition, while, when speaking of the ancestor of a
family of clans, they seem to regard him as a vaguer figure
obscured in the dimness of myth.
We here note again that the dominant lineages in more than
one tribe sometimes form part of the same clan structure. Thus,
the lineages dominant in the Gaajok, Gaajak, and Gaagwang
PLATE XXVI
W
THE LINEAGE SYSTEM 237
tribes, to the east and to the west of the Nile, are all segments
of the GAATGANKiiR clan. Also the dominant lineages of the
Rengyan tribe, to the west of the Nile, and the Lou tribe, to
the east of the Zeraf, are part of the jinaca clan. This distribu-
tion is easily accounted for, because we know that till recently
Fig. 14. Bufifalo-horn spoons.
lineages
the Eastern gaatgankiir and the Eastern jinaca
lived with the other lineages of these clans in the Jikany and
Rengyan areas to the west of the Nile.
There are also more general and mythological relationships
between clans. In recounting these relationships Nuer personify
the tribes and give them a kinship value by assimilating them
to their dominant clans. Thus, they speak of Bor, Lang, Lou,
Thiang, Lak, &c., as though they were persons and could have
kinship relations between them like those between persons, and
as though all the members of these tribes were of the same
238 THE LINEAGE SYSTEM
descent. By so doing they stress community relations and
obscure clan differentiation in a political context. This habit
often makes their statements appear confused, and even contra-
dictory, but it accords with a strong tendency in social life, as
we have seen in discussing the various meanings of the word
'cieng' , to identify the lineage system and the political system
in a specific set of relations.
Many Nuer regard their ancestors Gee and Ghaak as the
progenitors of all true Nuer, though one is given different
classifications in different parts of Nuerland. Among the Lou
it is said that all the tribes are descended from Gee, except the
Jikany and Gaawar tribes. These two alone are differentiated
because their proximity makes them significant for the Lou,
whereas all the other tribes, which have no direct relations with
the Lou, are vaguely classified as children of Gee. Among the
Eastern Jikany all true Nuer tend to be classed as 'Gee' in
contrast to 'Kir', the Jikany themselves. In the valley of the
Zeraf and in Western Nuerland, where the tribes have a very
much wider range of intertribal contacts, there is a wider range
of differentiation. The Nuer tribes there fall into three classes:
the Gee group, consisting of Bor, Lang, Rengyan, Bui, Wot,
Ror, Thiang, and Lou, stretches in an unbroken line, from
north-west to south-east, across the centre of Nuerland; the
Ghaak group, consisting of Nuong, Dok, Jaloogh, Beegh,
Gaankwac, and Rol, occupies the south-western part of Nuer-
land; and the Ril group, consisting of the Leek and Lak,
occupies the lower reaches of the Zeraf and the Ghazal near their
junction with the Nile. I have sometimes heard the Bui included
in the Ril group. However, in Dok country and in adjacent
tribal areas people further distinguish between the tribes else-
where classed as sons of Ghaak and divide them into a Ghaak
group, comprising Beegh, and Jaalogh, and a Gwea group,
comprising Dok, Nuong, Gaankwac, and Rol.
In these classifications we note a further exemplification of
what we have often observed elsewhere about Nuer classifica-
tions: their segmentary tendency and their relativity. Whilst,
for example, other Nuer see the Dok and the Beegh as Ghaak
they only see themselves as an undivided Ghaak group in
opposition to the Gee fraternity and otherwise see themselves
THE LINEAGE SYSTEM 239
as parts of opposed segments, Gwea and Ghaak. It will be noted
that these groups of tribes, often represented as families of
clans, occupy distinct sections of Nuerland. Before the
period of eastwards migration they ran from north to south, in
three or four groups, to the west of the Nile. Territorial contiguity
and a common clan structure, such as we find among the Jikany
tribes, or close relationship within a clan system, such as we find
among the Ghaak group of tribes, go together, and the values of
the two systems may be assumed to interact. The segmentation
of lineages within a tribe in relation to its political segmentation
is thus repeated in the whole clan system of the Nuer, the
segments of which are co-ordinated with the political segmenta-
tion of Nuerland. Those tribes which are adjacent to one
another have a common opposition to other groups of tribes and
this relation is reflected in the tendency for them to be repre-
sented, through their dominant lineages and clans, as closely
related on a mythological and ritual plane.
Gee and Ghaak and Gwea are represented as brothers, sons of
amythological ancestor, sometimes called Ghau, the World, and
sometimes Ran, Man, whose father is said to be Kwoth, God. Ril
is often described as one of their brothers as well, though he is
sometimes represented to be a son of a daughter of Gee, called Kar.
All sons of Gee have agnatic kinship {huth) which allows them
to partake of one another's sacrifices. In these ritual situations
only the true sons of Gee, the jinaca, the gaathiang, and jidiet,
and other clans descended from Gee, have huth interrelationship,
but in other situations the tribes in which these clans have a
dominant status are represented as brothers or first cousins.
Thus Thiang is said to have been the eldest son of Gee, Nac
(Rengyan and Lou) the second son, Ror and other tribes younger
sons ; and Rengyan (Nac) and Wot (Dit) are said to have been
twins, as also Bor and Lang, sons of Meat.
Some tribes stand outside this big family. The Jikany tribes
have dominant lineages of Dinka origin, descended from Kir,
who was found in a gourd by a man of the Ngok Dinka, but, as
was earlier explained, they are mythologically related to the
Gee group, because Gee is variously represented as the protector
or father-in-law of Kir. The gaatgankiir have a huth relation-
ship with some lineage system of the Ngok Dinka, and there-
240 THE LINEAGE SYSTEM
fore, in an unprecise political sense, the Jikany and Ngok Dinka
tribes have, by analogy, a fraternal relationship. It may safely
be assumed that at one time they had close intertribal relations.
The GAAWAR clan have also an independent origin, their ancestor
having descended from heaven. However, a number of mytho-
logical links unite him to the founders of various clans which
are dominant in tribes of the Ghaak group (see pp. 230-1), and
the Gaawar tribe therefore belongs to this group. Although the
Nile now separates it from the other members of the group it
was at one time their most northern extension on the west bank.
Owing to huth relationship between gaawar and other clans of
the family of clans descended from Kwook, Gaawar are said to
go with the people of the Fadang section of the Bor tribe and
the Atwot people, who are said to have lived at one time between
the present Rengyan and Dok tribal areas.
Through the recognition of agnatic relationship between
exogamous clans and of cognatic and mythological ties between
clans not considered to be agnates, all the Nuer tribes are by^
assimilation of political to kinship values conceptualized as a
single social system. A number of clans are not associated with
tribes, but their lineages are included in this system by the
affiliation of the clans to one or other of the large families of
clans. Thus, the jimem, jikul, gaatleak, and jither are
descended from Gee and belong to the Gee group ; the jikul
are mythologically attached to the Ril group and the jakar to
the Ghaak group ; and so on. The whole of the Nuer are brought
into a single kinship or pseudo-kinship system and all the
territorial segments of Nuerland are interconnected by that
system.
XIII
In our view the unusual degree of genealogical segmentation
in the Nuer lineage system is to be understood in terms of tribal
structure, which is, as we have seen, characterized by its ten-
dency towards segmentation. The association of the lineage
system with the tribal system means that as the tribe splits into
segments so will the clan split into segments, and that the lines of
cleavage will tend to coincide, for.liaeag es are no t corporate,
groups_Jbiit-are-£gibodied.in l ocal commu nities throughjvhich
they function s^ucturaUY. Just as a man is a member of a tribal
THE LINEAGE SYSTEM 241
segment opposed to other segments of the same order and yet also
a member of the tribe which embraces all these segments, so also
he is a member of a lineage opposed to other lineages of the same
order and yet also a member of the clan which embraces aU
these lineages, and there is a definite correspondence between
these two sets of affiliations, since the lineage is embodied in
the segment and the clan in the tribe. Therefore, the distance
in clan structure between two lineages of a dominant clan tends
to correspond to the structural distance between the tribal
segments with which they are associated. Hence the tribal
system draws out and segments the dominant clans and gives
them their characteristic lineage form. Evidence in support of
this contention could be cited from any Nuer tribe ; we propose
to examine only a few typical examples.
We have observed that in the Lou tribe the Gaatbal and Rum j ok
secondary sections form the primary section of Gun in opposition
to the Mor primary section, and how the dominant lineages of the
Gaatbal and the Rum j ok sections are descended from one wife of
Denac and the dominant lineage of the Mor section is descended
from a different wife, so that Gaatbal and Rumjok are to one
another in a relation analogous to that of full brothers and the
Gun stand to the Mor in a relation analogous to that of half-
brothers. We have noted, likewise, how the dominant lineages of
the Gaajak are descended from two closely related wives of Kir
while the dominant Uneages of the Gaajok and Gaagwang, which
have a very close alliance, are descended from a third wife.
The GAATGANKiiR lineages in their relation to the segmentary
structure of the Jikany tribes provide an excellent test of the
hypothesis that lineage structure is twisted into the form of the
political structure, for the same lineages are found at extremities
of Nuerland where pohtical conditions are not identical. Had I had
more time in the Jikany countries, or had I formulated the prob-
lem more clearly in the little time I had to spend there, I might
have been in a position to state my conclusions more dogmatically.
We will briefly analyse the lineage system of the gaatgankiir
in its relation to two of the Gaajak primary sections.
Thiang was the eldest son of Kir. He had two wives, Nyagaani
and Baal. From these two wives spring the three main lineages of
cieng Thiang, the Thiang primary tribal section, tar, lony (or
gek), and rang. What is said to have happened is shown in the
diagram on p. 242 . Tar , being the only son of his mother, has founded
an independent lineage and tribal section, that which lives in the
4507 J i
242 THE LINEAGE SYSTEM
extreme south of Eastern Gaajak country. The other four lineages
are all sprung from Nyagaani and are collectively known as cieng
Nyagaani. At first her four sons stuck together, but later Lony's
family increased and became more powerful than those of his
brothers and tried to lord it over them, especially over Lem, the
eldest. Kang, son of Lual, took the lead against Lony, and com-
pelled him to migrate. Owing to the prominent part played by
Kang, the lineages descended from Lem, Leng, and Lual are
collectively known as cieng kang in contrast to cieng lony. These
two lineages live to the extreme north of Eastern Gaajak country.'
When the brothers are spoken about, as quarrelling, migrating,
and so forth, it must be understood that the lineages and the local
communities of which they form part are being personified and
dramatized.
Kird'
Nyagaani ^= Thiang = Baal
Tarc3'
Lony
-<
S
5>'
TO
s
CO
Kang
secondary sect.
Lony
secondary sect.
Tar
secondary sect.
V. y
Thiang pnmarij section
We see in this diagram how the splitting and merging of lineages,
determined by the logic of lineage structure, follows the lines of
tribal fission and fusion. Thus the descendants of Lem, Leng, and
Lual, who live together, are merged in opposition to the lony, and
the LONY, who live adjacent to them, are merged with them in
opposition to the tar. The diagram does not show us the lines of
descent which have become completely merged in those recorded,
* The distribution of the three divisions of the Thiang primary section of
the Gaajak tribe contrasts with the unbroken territory of the segments of
primary sections elsewhere in Nuerland. I have not visited the area, and
cannot explain this unusual distribution by historical events nor state its
structural consequences.
THE LINEAGE SYSTEM 243
because such lines, having no localities specifically associated with
them and therefore no community value, are not differentiated.
That the diagram does not truly record historic lineage growth,
but is a distortion of it, is further suggested by the fact that there
are, on an average, five generations from the present day to Lem
and Leng, six to Lual, and seven to Lony, who was the youngest of
the four brothers.
The GAAGWONG lineage, which is the nucleus of the Gaagwong
section, is called after Gung, son of Kun, son of Kir. The gaag-
wong maximal lineage sphts into several major lineages. To make
illustration easier only those important among the Eastern Jikany
are figured in the diagram : cany, wau, and taiyang (nyayan and
NYAjAANi), the descendants of Buok, Wau, and Gee.
Gunq
Ten I
Wau
Gee
T
ai u
Z3
a n^
S' >
CO >
C\
\Na u la I y a n q
secondary secondary
section section
( Nyayan and
N y a J a a n i )
Buok
Cany
>
Z
-<
at
CD
Cany
secondary
section
Gaagwong primary section
The CANY major hneage split up, as shown in the diagram on p. 244,
into a number of smaller lineages of three to four generations in
depth. In this diagram the traditional representation of cleavage
between sons of the same father but of different mothers is shown
244 THE LINEAGE SYSTEM
in the lines which spring from Diu through his three wives:
Mankwoth, Thul, and Mankang. The descendants of Dup are
normally referred to as cieng mankwoth because they live with
them. Common residence has reacted on the lineage structure, so
that the dup lineage has become very largely merged in the mank-
woth lineage. For this diagram I am indebted to the Government
Canyd*
Wurum KolmuCrf*
(Kolthiang) d"
cr
^3
—I
cr.
v2
Dup Mankwoth =Diu= Thul
?
C3
c5
Mankang
o "^
Jokrual
3 ^
a-
Duwar
Wiu<^
V
Cany secondary section
files of Nasser District. I have no record of the divisions of the
WAU lineage.
In Western Nuerland wau and cany are politically unimportant
lineages and are merged with cieng Taiyang, which is contrasted
with cieng Jueny, called after Jueny, son of Teng, son of Gung,
who founded a lineage politically unimportant among the Eastern
Jikany. The taiyang lineage has two branches called after his
wives Nyayan and Nyajaani. In Western Nuerland there is a
third lineage springing from a third wife, Nyakoi.
It is interesting to observe that Jueny, whose line is politically
unimportant among the Eastern Jikany, is there given as a son of
Gee and merged in the taiyang line, while Duob is given as a son
of Nyajaani and merged in her lineage. Here again we see how
Hneage structure is influenced by political relations.
THE LINEAGE SYSTEM
Gungc^
245
Teng<^
Dit Nyaluac = Tai
(Nyajaani) ?
Gee^
y/ng
Juenyc^'
NYAJAANI
Lin cage
JUENY
Lineage
Nyayan = Nyakoi
^ Duobc^' ^
NYAYAN
Lineage
DUOB
Lineage
A wider and deeper analysis is required to prove the thesis we
have put forward and of which we have given a few out of many
examples. It is, however, supported by evidence of a different
kind. We found that it was always easier to obtain a more
complete record, and a longer line, of descent from members of
dominant lineages in the larger tribes than in the smaller tribes,
showing that greater attention is paid to the lineage system in
the larger tribes and that the clan structure is broadened and
deepened to serve its structural function. Also, we found it easy
to obtain from any adult member of an aristocratic lineage an
account of the other maximal and major lineages of his clan and
a long list of ancestors, some nine or ten at least, giving a con-
sistent length from the founder of the clan ; whereas we found
that we could not elicit the same information from members
of clans which have no tribal associations. They were often able
to trace their descent back for only some four to six generations,
the time depth they gave was seldom consistent, and they were
usually unable to give a coherent account of the other lineages
of their clan. We attribute this fact to the absence of systemiza-
tion through association with tribal structures. A lineage does
not stand in territorial opposition to other lineages of its clan,
but has with them only a vague ceremonial relationship, and
this relationship may never be expressed in corporate action.
Consequently there is generally a complete absence of any
elaborate system of lineages like those of the dominant clans.
There are many jimem clansmen and doubtless one could by
r;
t
246 THE LINEAGE SYSTEM
putting together their genealogies construct some sort of tree
of descent from Mem, in which the agnatic relationship between
various lineages could be indicated ; but it would be
very unlike the spontaneous statements that at once
delineate the lineage system of large clans like the
jiNACA, associated with tribal territories.
It is also very noticeable that Nuer knowledge of
the lineage system of a dominant clan tends to be
restricted to those parts of the system that corre-
spond to segments of their tribe. Thus the jinaca
lineages which are associated with segments of the
Lou tribe are well known to Lou tribesmen, but they
have no, or very little, knowledge of the jinaca line-
ages of the Rengyan tribe. Likewise I experienced
much difficulty in obtaining from Gaajok and
Gaajak tribesmen a clear account of the lineages
of the gaatgankiir clan which form the dominant
nucleus of the Gaagwang tribe, though they were
well informed about the lineages of the same clan
which are associated with segments of their own
tribes.
It follows from our account that, as we have sug-
gested before, the lineage system of a clan can only
to a very limited extent be considered a true Vecord
of descent. Not only does its time depth appear to
be limited and fixed, but also the distance between
collateral lineages appears to be determined by the
political distance between the sections with which
these lineages are associated, and it may be supposed
that a lineage only persists as a distinct line of
descent when it is significant politically. Ancestors
above the founder of a minimal lineage are relevant
only as points of departure for denoting lines of
descent when these lines are rendered significant by
the political role of the lineage system. We have
suggested that the depth of lineages is a function of the range
of counting agnation on an existential plane, and we now
further suggest that the range of counting agnation is largely
determined by its organizing role in political structure.
Fig. 15.
Leather
flail.
THE LINEAGE SYSTEM 247
Nuer consider that lineage cleavage arises from a fundamental
cleavage in the family between gaatgwan, children of the father,
and gaatman, children of the mother. Where there are two
wives and each has sons, the lineage bifurcates from this point.
A lineage bifurcation is a polygamous family writ large. Thok
dwiel, lineage, means this; it is 'the entrance to the hut', the
mother's hut. The tiny twigs we see in the gol, household, grow
into the great branches of the lineages. It is for this reason that
lineages are so often named after women, the mothers from
whose wombs sprang the different lines of descent. As we under-
stand the process, what happens is that certain lineage groups
gain political importance and exclusiveness, becoming nuclei
of tribal sections, and that only by so doing is their structural
position stabilized and are the points of their bifurcation rendered
fixed and permanent points of convergence in lineage structure.
This explains how it is that in only a few out of a vast number
of polygamous families is maternal descent structurally signifi-
cant, and why the bifurcation occurs in the lineage where it
occurs in the tribe.
This tendency towards co-ordination of territorial segmenta-
tion and Uneage segmentation can be seen in the various stages
of territorial expansion between the household and the tribe.
When brothers of an influential family live in different parts of a
village and gather around them a cluster of relations and depen-
dants, these hamlet-groups are named after them and they
become the point at which the lineage is likely to bifurcate.
Thus, if the brothers are called Bui and Nyang, people speak
of the gol of Bui and the gol of Nyang, and if later the grand-
children of one of them move to a different village site the
lineage will split into two branches. Such minimal lineages as
those pictured in the diagrams on pages 196-7 occupy adjacent
villages, or widely separated divisions of the same large straggling
village, and make separate camps along the same stretch of river
or adjacent camps around a small lake. The points of divergence
of lineages from clan trees are thus related to the size and
distribution of inhabited sites in a tribal area.
The association of the tribal sy stem with a clan may thus
be supposed to influence the form of \he. lineage structure . We
may further emphasize the morphological consistency between
248 THE LINEAGE SYSTEM
the two structures. There are always more villages than tertiary
segments in a tribe and more tertiary segments than secondary
segments, and so on, so that, since each territorial unit is asso-
ciated with a lineage, the narrowing of such units from the
multitude of villages to the single unit of the tribe must be
reflected in the conceptual structure of the lineage system,
there being a multitude of minimal lineages, fewer minor
lineages, and so forth, till the single unit of the clan is reached.
If this suggestion is accepted it is evident that the lineages are
in number and structural position strictly limited and con-
trolled by the system of territorial segmentation. The two
systems may thus be represented diagrammatically by the
same figure, though the correspondence is not exact.
Tribe Clan
Primary tribal sections/ \ Maximal lineages
Secondarij tribal sections/ \Major lineages
Tertiary tribal sections/ \Mincr lineages
Village communiticsZ AMinimal lineages
CHAPTER VI
THE AGE-SET SYSTEM
I
All male Nuer are initiated from boyhood to manhood by a
very severe operation {gar) . Their brows are cut to the bone with
a small knife, in six long cuts from ear to ear. The scars remain
for life, and it is said that marks can be detected on the skulls
of dead men. They are particularly clear in Plates XXVI (6),
XXVII, and XXVIII. The ceremonial of initiation is more
complex, and the age-set system has greater social importance
among the Nuer than among other Sudan Nilotes.
We have described, and others have described, elsewhere the
ritual of initiation. Though we have since collected further
information on the details of the rites, we consider that to trans-
scribe it here would be out of accord with the plan of this book.
The barest facts are related. Boys are now usually initiated
between the ages of 14 and 16 ; in the old days at a somewhat
later age, perhaps from 16 to 18. Whether a boy is initiated in
one year or in another year may depend on the milk and millet
supply. A boy must obtain his father's consent to the operation,
but it cannot weU be refused, for the boy would then run away
to the home of a kinsman and the father would be humiliated.
He afterwards goes to a member of his father's age-set who
performs a rite to give him the blessing of the set. A member
of the clan likewise gives him its blessing, and his father and
maternal uncle bless him. The boys make their own arrange-
ments with the operator and each presents him with a fishing
spear. The operator is any man who has learnt the art.
Several boys are initiated at the same time, for it is thought
that were a boy to be initiated by himself he would be lonely
and might die. Also, it is easier to cater for the boys and to give
them the care and attention they require during convalescence
if they are initiated in batches. Usually some four to a dozen
boys pass through the rites together. Initiation can take place
at any season, but it almost invariably takes place at the end
of the rains, when there is plenty of food and the north wind
blows and cicatrizes the wounds. Each village acts independently
4507 K k
250 THE AGE-SET SYSTEM
in arranging for its boys to be initiated. After the operation
the boys Uve in partial seclusion and are subject to various
taboos. This is a time of mild licence and they pass out of it by
a special rite. On the day of the cutting, and on the day of the
passing out of seclusion, sacrifices are made and there is festivity,
which includes licentious horseplay and the singing of lewd
songs. Only age-mates of the father of the initiate in whose
homestead the feasting takes place attend it: others keep at a
distance lest they see the nakedness of their kinswomen and
mothers-in-law,
II
All boys initiated during a number of successive years belong
to a single age-set (nc). There has, till recently, been a four-year
interval between the end of one such set and the commencement
of the next. These four years are known as the time 'when the
knife is hung up', and at the end of that period it is said 'the
knife is brought out' and boys may then be initiated again.
A certain wid ghok, 'Man of the Cattle', is, in each tribe, respon-
sible for opening and closing the initiation periods and thereby
dividing the sets. He performs the appropriate rite in his
district, and when the news goes round other districts begin or
cease initiation. He derives prestige from his functions, but
they are ritual and give him no political authority. Sometimes,
in Western Nuerland, they are performed by a prophet, but it
is possible that in these cases the prophet is also a Man of the
Cattle. The age-sets are organized independently in each tribe,
at any rate in the larger tribes, but it often happens that when
a new set has been started in one tribe an adjacent tribe will
follow its lead, so that the names and periods of sets in neigh-
bouring tribes are frequently the same. Also, though in different
parts of Nuerland the names of the sets are different and the
open and closed periods do not coincide, it is easy for a man who
moves from one part to another to perceive in which set he
would have been initiated had he been brought up in that part
of the country.
To-day there are no closed periods and boys are initiated every
year. The Man of the Cattle announces every few years that he
is about to cut the sets and performs a ceremony by which all
PLATE XXVII
Initiation of boys (near Nasser, h^astcrn (^.aajnk)
THE AGE-SET SYSTEM 251
youths initiated up to that year fall into one set and all youths
initiated after that year fall into a junior set. The number of
years during which a set runs before being cut is variable, and
this has probably always been the case. We conclude on the
evidence that ten years may be regarded as an average period
between the commencement of one set and the commencement
of the next. It was found that generally two sets, but occasion-
ally one, intervened between the set of a man and that of his
eldest son. In the case of younger sons two or three sets usually
intervene. It may be accepted that on an average the genera-
tions of grandfather — father — son cover six sets.
At the time of my inquiry into the age-set system there were
members of six sets alive, but there were only a few survivors
of the senior set and the members of the next senior were very
infirm. Sets with no living members are not recorded in the lists
below. Their names are irrelevant to an understanding of the sys-
tem and their order is so ill-remembered that the statements of
two informants on the matter are seldom in entire agreement. It
should be noted, however, that as far back as Nuer recitation
carries us the names are not uniform for the whole of Nuerland
and that they are not repeated. There is no cycle of names such
as we find in many parts of East Africa. The Lou and the
Eastern Jikany tend to have common names ; also the tribes of
western Nuerland ; while the tribes of the Zeraf have some names
in common with the Sobat river tribes and some with the tribes
to the west of the Nile.
Western Jikany and
Lou tribe
Eastern Jikany
Lak tribe
Leek tribes
Thut
Thut
Thut
Lilnyang
Boiloc
Boiloc
Boiloc
Ruob
Maker
Maker
Ruob
Wangdel
Dangunga
Dangunga
Wangdel
Tangkwer
Luac
Carboc
Wooni
Rol
Lithgac
Lithgac
Kec
Juong
Rialmac
Rialmac
Pilual
Bildeang
Each age-
set has two or
three subdivisions
^Each year ii
initiation period may receive a separate name and constitute a
division though probably two years often have the same name and
the divisions are usually two-year periods. But though a set is
thus stratified internally and the divisions are called by different
252 THE AGE-SET SYSTEM
names, all members of the set are known by the name of the first
division, and this common name persists while the others eventu-
ally fall into disuse. Thus to-day one seldom hears of Maker indit
and Ngwak, but only of Maker, under which title both divisions are
included. Likewise one hears small mention of Gwong indit, Carboc,
and Nyamnyam, and reference is normally to Dangunga (Gwong),
which term covers all three divisions. The senior division is called
indit, the greater, and when the segmentary names are dropped
the indit is dropped also, since its purpose is to distinguish the
primary division from the later ones. Hence we have Thut indit,
Maker indit, and Boiloc indit, the elder Thut, the elder Maker, and
the elder Boiloc, but the indit ending in these names is eventually
dropped and the complete sets become known as Thut, Maker, and
Boiloc.
In recent years the matter has become somewhat complicated
by the absence of well-defined closed and open initiation periods.
Thus in my earlier visits to Lou and Eastern Gaajok I heard people
speak of Lieth indit, Lieth incar [Lieth intot), Caiyat {Pilual), and
Rialmac [Rialdang) as four divisions of the Lithgac age-set, but
this was because there had been no declaration by the Man of the
Cattle separating them into different age-sets. On a later expedi-
tion I found that the Lieth indit, the Lieth incar, and Caiyat had
been declared to be a single age-set, and the Rialmac to start a new
age-set, a second division of which, the Kwekoryoamni, has since
been initiated. Similarly among the Western Jikany and Leek
the Bildeang age-set has recently been declared separate from the
Juong. At one time in Eastern Gaajok the Lithgac were cut off
from the Rialmac, while in the adjacent country of the Eastern
Gaajok they had not yet been separated and remained for the
time being a single set. It may thus happen that, in modern times,
a sub-division may for a time be regarded as the junior segment of
one set and later become the senior segment of the next set. Below
are given the sub-divisions of the age-sets in the Lou and Western
Jikany tribes :
Thut
Boiloc
Maker
Lou
iThut indit
Muothjaang
Lilcoa
(Boiloc indit
Golyangkakeat
Laibwau
{ Maker indit
[Ngwak
Western Jikany
ILilnyang
Lilcoa
Lilcuath
Ruob
Wangdel
iRuob
( Nomalith
j Wangdel
1 Wathcar
Gwong
Luac
Lithgac
Rialmac
THE AGE-SET SYSTEM
Western Jikany
253
Lou
I Gwong indit
Carboc
Nyamnyam
I Luac indit
Karam
Camthoari
ILieth indit (inbor)
Lieth intot {incur)
Caiyat {Pilual)
I Rialmac (Rialdang) indit
\ Kwekoryoam
Tangkwer
Rol
Juong
I Tangkwer
\ Karam
[Rol
\ Pilual
{ Juong
\ Majaani
Bildeang Bildeang
III
In seeking to understand how membership of an age-set deter-
mines a man's behaviour we have first to reahze that there is no
purposive education or moral training in the procedure of
initiation. Also, many of the characteristic features of the age-
set system in Kenya, where it is most highly developed, are
absent in the Nuer variation. There are not three distinct
age-grades of boys, warriors, and elders through which the sets
pass, for a boy who is initiated into manhood remains in this
grade for the rest of his life. Warriors are not prohibited from
marrying and they neither enjoy privileges nor suffer restric-
tions different from those of other male adults. The sets have
no administrative, juridical, or other specific political functions
and the country is not handed over to their care. The sets have
no definite military functions. Indeed^ we are o fthe opinion
that the Nuer age-set system ought not to be des crib ed as
a military organization, though some writers give it this
character. Youths who have recently been initiated are^anxious
for their first raid, and consider that they ought to earn for their
set a reputation for valour, and it is likely that raids were
generally conducted in the main by men of the most junior set.
However, there is no grade of warriorhood through which the
age-sets pass nor a grade of elderhood into which they enter.
Were boys and old men to take part in warfare against other
Nuer they would probably be killed, and it is understandable
that raids are the occupation of the strongest and fleetest,
254 THE AGE-SET SYSTEM
though many middle-aged men accompany the expeditions
and always have their share in inter-tribal fighting and local
disputes.
The age-set system of a tribe is in no way its military organiza-
tion. Men fight by villages and by tribal sections and not by
sets. The war companies are local units and not age-set units,
and within a company men of different sets fight side by side,
though, especially in raids, most of the warriors would be mem-
bers of the two most junior sets. Kinship and local ties deter-
mine a man's place in the ranks. Hence the age-sets are not
regiments, though wars and raids are often spoken of as the
actions of a certain set because they took place during the
initiation period of this set and its members took the most
prominent part in them, since skill in arms, love of adventure,
and desire for booty are the privileges of youth.
It is in more general social relations, chiefly of a domestic
and kinship order, and not in political relations that behaviour
is specifically determined by the positions of persons in age-set
structure. When a boy passes into the grade of manhood his
domestic duties and privileges are radically altered. His change
of status is epitomized in the taboo on milking which comes into
force on the day of his initiation and continues for the rest of
his life, but it is expressed also in other domestic tasks, in habits
of eating, and so forth. At initiation a youth receives from his
father or uncle a spear and becomes a warrior. He is also given
an ox, from which he takes an ox-name, and becomes a herds-
man. From now on, till he is a husband and father, his chief
interests are dancing and love-making. Then he becomes 'a
true man': 'he has fought in war and not run away; he has
duelled with his age-mates ; he has cultivated his gardens ; he
has married a wife'.
There is a sudden and great change in status from boyhood
to manhood, but the modes of behaviour which differentiate
these two grades do not distinguish one set from another, for
the privileges of manhood are enjoyed by members of all the
sets equally. Nevertheless, the sets are stratified by seniority
and there are well-defined relationships between them. Before
summarizing these patterns of inter-set behaviour we touch on
some general characteristics of the whole system.
THE AGE-SET SYSTEM 255
The age-set system is a further exempHfication of the seg-
mentary principle which we have seen to be so evident a quahty
of social structure. Tribes segment into sections and their sec-
tions further segment, so that any local group is a balanced
relation between opposed segments. Clans segment into lineages
and their lineages further segment, so that any lineage group is
a balanced relation between opposed segments. Likewise the
institution based on age is highly segmentary, being stratified
into sets which are opposed groups, and these sets are further
stratified into successive sections. We may therefore speak of
structural distance in this new dimension. Just as the distance
between political segments varies according to their positions
in the political structure, and the distance between lineage seg-
ments varies according to their positions in the lineage struc-
ture, so the distance between age-set segments varies according
to their positions in the age-set structure. The structural
distance between any two sets is the social relation between
those sets and the determinant of behaviour between their
members.
The relativity of values which we noted in discussing the
political and lineage systems may also be seen in the age-set
system. We have noted that a set which is seen as an unseg-
mented whole by members of other sets is internally segmented,
and that members of each of its segments see themselves as
exclusive units in relation to the others, though these divisions
close as the set becomes more senior and has a new position
in relation to sets, since created, below it. Also, there is a
tendency for members of two successive sets, adjacent segments
of the structure, to fuse in relation to a third in feelings and in
ceremonial actions. A Rialmac youth said : 'We and the Lithgac
are about the same age and we can be free in speaking to them,
but we must show respect to an older man, even if he is not of
our fathers' age-set.' Although there are six sets with living
members there are very few survivors of the two senior sets, and
from the point of view of a young man they are merged with
the one that follows them. Only four sets count, and, seen by
individuals, they merge into two generation groups of equals
and brothers, and seniors and fathers or juniors and sons. To a
Lithgac son of a Maker father all members of the Maker age-set
256 THE AGE-SET SYSTEM
are his fathers, and the Lithgac and the Luac tend to see them-
selves as a single group in relation to the Maker, having a like
attitude of respect towards them. But in relation to the Dan-
gunga and to the Lithgac, the Luac identify themselves with the
one or the other according to the direction of attention, and this
is determined by the social situation. Any set tends to see the
set senior to it as equals in relation to junior sets and the set
junior to it as equals in relation to senior sets. It is possibly
this contradiction which creates segmentation in any set. Thus,
at sacrificial feasts men eat according to their position in the
age-set structure, but which sets sit and eat together depends
on the set of the owner of the feast and on the number of sets
present at it. If a Dangunga kills an ox and there are Maker
present, but not Boiloc, then Dangunga eat with Maker and
Luac with Lithgac and the Rialmac by themselves ; but if there
are Boiloc present then the Maker eat with them and the Dan-
gunga with the Luac, and the Lithgac with the Rialmac. The
Dangunga would not eat with the Boiloc because they are the
set of their fathers or fathers-in-law; and for the same reason
the Luac must eat with the Lithgac if the Dangunga go with the
Maker.
The age:-set ^system, differs^om the_ territorial ,and._Jineage_
systems in one important respect. Whereas the people of a terri-
torial segment remain, or most of them remain, in the same
structural relation to other territorial segments for their whole
lives, and whereas the members of a lineage have a fixed
relationship to other lineages, an age-set group changes its
position in relation to the whole system, passing through points
of relative juniority and seniority. This mobility of age-set
groups is peculiar to the system and is a necessary characteristic
of it, for it is an institution based on the succession of genera-
tions. Probably, for oecological reasons, the actual political con-
figuration remains very much the same from generation to
generation. People pass through the political system without
their structural position in it changing to any extent during
their passage. It is the same with the lineage system. Never-
theless the mobility of groups through the age-set structure
and their changing position in it should not be allowed to
obscure the constancy of its structural form. There have pro-
PLATE XX\in
Youth (Zeraf river)
b. Youth (Louj
Al.m V.cr.\! i'i\xTj
THE AGE-SET SYSTEM 257
bably always been the same number of sets in existence at any
time and these sets always have the same relative positions to
each other in the system, regardless of actual groups of men
composing them.
It is significant that among the Nuer, as among other East
African peoples, the age-set system is the first institution to
undergo rapid and great modification under European rule and
that the other social systems do not appear to be affected by
the changes in its constitution. This tends to confirm the opinion
we have earlier expressed, that whilst the age-set system is
combined with the territorial and lineage systems in the same
social cadre and is consistent with them, the consistency is not
an interdependence.
IV
Within the age-set system the position of every male Nuer
is structurally defined in relation to every other male Nuer and
his status to them is one of seniority, equality, or juniority. It
is difficult to describe these statuses in terms of behaviour,
because the attitudes they impose are often of a very general
nature. The following points may, however, be noted, (i) There
are certain ritual observances and avoidances, chiefly between
members of the same set, but also between sets. The most
important of these are the segregation of the sets at sacrificial
feasts, to which we have referred, and the stringent prohibition
on members of a set burying an age-mate or partaking of the
meat of beasts sacrificed at his mortuary ceremony ; but there
are a number of other ritual injunctions. (2) A man may not
marry, or have sexual relations with, the daughter of an age-
mate, for she is his 'daughter' and he is her 'father'. Also,
while a man may always have sexual relations with the daughter
of one of his father's age-mates he ought not to marry her
unless either his father, or her father, is dead, and then only
after the parties to the marriage have exchanged beasts in
atonement to the age-set of the fathers. (3) Members of the
same age-set are on terms of entire equality. A man does not
stand on ceremony with his age-mates, but jokes, plays, and eats
with them at his ease. Age-mates associate in work, war, and
in all the pursuits of leisure. They are expected to offer one
4507 L 1
258 THE AGE-SET SYSTEM
another hospitality and to share their possessions. Fighting is
considered an appropriate mode of behaviour between age-
mates, but a man ought not to fight a man of a senior set. The
comradeship between age-mates springs from a recognition of a
mystical union between them, linking their fortunes, which
derives from an almost physical bond, analogous to that of true
kinship, for they have shed their blood together. (4) Members
of a set are expected to show respect to members of senior sets,
and their deference to them can be seen in discussions, in
etiquette, in division of food, and so forth. Whenever there is
a question about the propriety of speech or action it is judged
by reference to the relative positions of the persons concerned
in the age-set structure, if kinship status is not also involved.
Since every man has a known age-relationship to every other
man in Nuerland with whom he is likely to come into contact,
their social attitude to him, and his social attitude to them, is
determined in advance by the distances between them in age-
set structure, unless kinship takes precedence. Although it is
possible for a man to avenge a breach of these patterns of
behaviour by a curse, if it is a very serious breach, the ordinary
sanctions of conduct are a man's conscience and desire for
approval.
It will have been noted that the relations between the sets
are defined in the idiom of family relationships. The members
of a man's father's age-set are his 'fathers ' and the members
of his father's brothers' 'age-sets' are, in a less precise sense,
also his 'fathers'. The sons of a man's set are his 'sons', and
they may fall into several sets. The wives of members of
a man's father's set are his 'mothers', and the wives of
members of his sons' sets are his 'daughters'. All members of a
man's own set are likewise 'brothers', though here the analogy
is seldom expressed because the comradeship between age-mates
is strongly affirmed in the idiom of the system, for they are all
nc, age-mates, to one another. As, in any case, a man commonly
addresses all persons much senior to himself as 'father' and
'mother', all persons much junior to himself as 'son' and
' daughter', and all persons of about the same age as himself as
' brother ' and 'sister ', the terminology of address between differ-
ent sets is not a differentiating one and it cannot be said how
THE AGE-SET SYSTEM 259
far it is determined by specific age-set relationships. When
speaking about sets senior to his own, but not that of his father
or the set immediately senior to his own, a Nuer sometimes
speaks about them collectively as though all their members
were his fathers-in-law and their wives his mothers-in-law, for
he is courting their daughters and is likely to marry one of
them and so is circumspect in his dealings with their parents.
Thus a Lithgac son of a Maker father regards members of the
Dangunga set and their wives as potential fathers-in-law and
mothers-in-law.
The age-set system thus influences persons through a kinship
idiom and on the pattern of kinship. The sets never act cor-
porately, but they function locally between individuals and, in
ceremonial situations, between small aggregates of persons who
live near to one another, for a man only has frequent contacts
with those members of his set and of other sets who live in his
district. No doubt relative positions in age-set structure to
some extent determine behaviour between neighbours, and it
can sometimes be observed that they determine it, but it is
difficult to say to what extent, for men who live near one another
are not only members of the same age-set or of different age-
sets but are also kinsmen or affines. The age-set patterns of
behaviour are, except in specific rites, of so general a nature
that they cannot be isolated in a community where everyone is
related in a number of different ways to everybody else. We
have noted how persons who live together are always able to
express their relations to one another in the language of kinship
and how, when they are not actual kinsmen, they are recognized
as equivalent to such by adoption or through some traditional,
or mythological, connexion. The age-set stratification of all
men, and by analogy all women, into groups whose inter-
relations are on the pattern of family relationships is one of the
ways by which community relations are expressed in kinship
patterns and is comparable to the classificatory system of kin-
ship nomenclature in its assimilation of social relations to a few
elementary types. Age-relations are part of the general social
ties of a kinship type which unite all persons living in a com-
munity. The members of a local group have group relations
only with other groups of the same kind and it is these relations
26o THE AGE-SET SYSTEM
which we call political. They also have manifold contacts with
one another — economic, ceremonial, food, play, and so forth —
and political relations may be viewed as a specific organization
of the texture of social ties, which control these contacts, in
certain situations. It is this action of the age-set system, in
establishing ties between members of local communities and in
giving them a kinship value, that we chiefly stress in a political
context rather than its indication of leadership, for outside
small kinship and domestic groups the authority derived from
seniority is negligible, and the sets lack leadership and admini-
strative and judicial functions.
The age-set system has been briefly treated because of this
action and also because, in the larger tribes at any rate, it is a
tribal institution. It segments the male population of a tribe
into stratified groups which stand in a definite relationship to
one another and it cuts across territorial divisions, giving
identity of status where there is political disparity and differ-
entiating status where there is political identity. However, the
political system and the age-set system do not seem to be inter-
dependent. Both are consistent in themselves and to some
extent overlap and influence one another, but it is easy to
conceive of the political system existing without an age-set
organization. There is evidence in East Africa that political
development produces atrophy of the age-set organization. In
conclusion we would again emphasize that adjacent tribes
co-ordinate their sets and that the sets of any tribe are easily
translated into the sets of another tribe. Initiation rites, more
than anything save language, distinguish Nuer culture and give
Nuer that sense of superiority which is so conspicuous a trait of
their character. Only in the sense that age-sets are organized
tribally and are common to all tribes can there be said to be a
correspondence between the age-set system and the political
system. There is no positive structural correspondence of the
kind we have noted between the lineage system of dominant
clans and tribal segmentation. It may be said, therefore, that
whereas the political system and the lineage system of dominant
clans are interdependent, the political system and the age-set
systems are only a combination, in Nuer society. We may add
that the common assumption that an age-set system merely by
THE AGE-SET SYSTEM 261
stratification integrates the members of a tribe has httle to
commend it.
In the way we have written this book we have in some
measure broken away from the tradition of lengthy monographs
on primitive peoples. These weighty volumes generally record
observations in too haphazard a fashion to be either pleasant
or profitable reading. This^ deficie ncy is due to a bsenceof a
tiody-of-scientifiG- theory -in_^QciaI Anthropology, for facts can
Qnly—be-^ected -and -arranged in the light of theory. It is
aggravated by the error of confusing documentation with
illustration. We have also tried to describe Nuer social organiza-
tion on a more abstract plane of analysis than is usual, for usually
abstract terms are mistaken for abstractions. Whether we
have succeeded in doing so is for the reader to judge, but in case
it be said that we have only described the facts in relation to a
theory of them and as exemplifications of it and have sub-
ordinated description to analysis, we reply that this was our
intention.
It is difficult to know how far one is justified in pressing an
abstraction. Once one has a theoretical point of view it is fairly
simple to decide, what facts are significant, since they are
significant or otherwise to the theory, but it may be doubted
whether it is wise, in discussing the political institutions of
a primitive people, to give only the barest reference to their
domestic and kinship life. Can this be done successfully?
That is precisely the question we have asked ourselves, and we
have concluded that an answer can only be given to it by making
the attempt.
I. We first gave an account of Nuer absorption in their cattle
and showed how this value in their system of oecological rela-
tions necessitates a certain mode of distribution and trans-
humance. We then described the concepts of time and space
that arise very largely from ways of livelihood and disposition
of settlements. We then examined the territorial sections which,
through the values attached to them, form a political system.
We further noted that structural distance in the lineage systems
262 THE AGE-SET SYSTEM
of the dominant clans is a function of structural distance in the
tribal systems and that there is no comparable interdependence
between age-set structure and political structure.
2. By social structure we mean relations between groups
which have a high degree of consistency and constancy. The
groups remain the same irrespective of their specific content of
individuals at any particular moment, so that generation after
generation of people pass through them. Men are born into
them, or enter into them later in life, and move out of them at
death ; the structure endures. In this definition of structure the
family is not considered a structural group, because families
have no consistent and constant interrelations as groups and
they disappear at the death of their members. New families
come into being, but the old for ever vanish. We do not suggest
that the family is for this reason of less importance than struc-
tural groups ; it is essential for the preservation of the structure,
for it is the means by which new persons are born into its seg-
ments and the system is maintained. Nor do we suggest that
the relations we consider structural are between groups that do
not in any way vary. Territorial, lineage, and age-set systems
change, but more slowly, and there is always the same kind of
interrelationship between their segments. We do not, however,
insist on this limiting definition of structure and our description
and analysis do not depend on it.
3. Structural relations are relations between groups which
form a system. By structure we therefore further mean an
organized combination of groups. The territorial distribution
of a Nuer tribe is not a haphazard aggregate of residential units,
but every local group is segmented and the segments are fused
in relation to other groups, so that each unit can only be defined
in terms of the whole system. Similarly a lineage or age-set can
only be defined in terms of the systems of which they form part.
We have tried to show this in our account.
4. By structure we mean relations between groups of persons
within a system of groups. We stress that it is a relation
between groups, for relations between individuals may also be
arranged on" a regular plan, e.g., kinship relationships may be
spoken of as a kinship system. By 'group' we mean persons
who regard themselves as a distinct unit in relation to other
THE AGE-SET SYSTEM 263
units, are so regarded by members of these other units, and who
all have reciprocal obligations in virtue of their membership of
it. In this sense a tribal segment, a lineage, and an age-set are
groups, but a man's kindred are not a group. A kinship relation-
ship is a category and the kinship system a co-ordination of
categories in relation to an individual. In our opinion strangers
and Dinka ought to be described as persons of certain categories
rather than as members of social groups and that the relations
between them and aristocrats are not, strictly speaking, to be
described as structural relations.
5. The social structure of a people is a system of separate but
interrelated structures. This book deals mainly with the political
structure. Faced with the initial difficulty of defining what is
political we decided to regard the relations between territorial
groups as such, taking the village as our smallest unit, for
though a village is a network of kinship ties it is not a kinship
group, but a group definable only by common residence and
sentiment. We found that the complementary tendencies to-
wards fission and fusion, which we have called the segmentary
principle, is a very evident characteristic of Nuer political struc-
ture. The lines of political cleavage are determined chiefly
by oecology and culture. Harsh environment together with
dominant pastoral interests cause low density and wide gaps in
the distribution of local communities. Cultural differences
between the Nuer and their neighbours also cause varying
degrees of political distance. Oecological and cultural relations
often combine to produce fission. In Nuerland itself culture is
homogeneous, and it is oecological relations that chiefly deter-
mine the size and distribution of segments.
6. These tendencies in, or principles of, political structure
control actual behaviour between persons through values. These
values appear contradictory. They are only seen to be con-
sistent when we view structure as sets of relations defined by
reference to specific social situations. By political values we
mean the common feeling and acknowledgement of members of
local communities that they are an exclusive group distinct from,
and opposed to, other communities of the same order, and that
they ought to act together in certain circumstances and to observe
certain conventions among themselves. It does not follow that
264 THE AGE-SET SYSTEM
behaviour always accords with values and it may often be found
to be in conflict with them, but it always tends to conform to
them.
7. Not only can we speak of the relations between territorial
groups as a political system, the relations between lineages as a
lineage system, the relations between age-sets as an age-set
system, and so forth, but also in a society there is always some
relationship between these systems in the whole social structure,
though it is not easy to determine what this relationship is.
We have shown that there is interdependence of a kind between
the Nuer lineage system and their political system. This does
not mean a functional relationship between clan groups and
territorial groups, although they have a certain association, for
clans, and even their lineages have no corporate life. Nor does
it mean that when a man behaves in a certain way to a fellow
clansman and in a different way to a fellow tribesman that
there is a functional relationship between these two modes of
behaviour. Nor, again, does it mean that there is a functional
relationship between those members of a dominant clan who
live in a tribe and the tribe of which they form part. But it
means that there is structural consistency between the two
systems — a consistency between abstractions. We are unable
to show a similar interdependence between the age-set system
and the political system.
8. Can we speak of political behaviour as a distinct type of
social behaviour? We have assumed that certain activities,
such as war and feuds, may be called political, but we do not
consider that much is gained by so designating them. It is only
on the more abstract plane of structural relations that a specific
sphere of political relations can be demarcated. The behaviour
of persons to one another is determined by a series of attach-
ments, to family, joint family, lineage, clan, age-set, &c., and
by kinship relationships, ritual ties, and so forth. These strands
of relationships give to every man his sphere of social contacts.
His field of actual contacts is hmited ; his field of potential
contacts is unlimited. We distinguish a man's social sphere in
this sense from structural space, the distance between social
segments, which are groups of people who compose units in a
system. We do not therefore say that a man is acting politically
THE AGE-SET SYSTEM 265
or otherwise, but that between local groups there are relations
of a structural order that can be called political.
9. We do not describe the different social ties which exist
between persons living in the same district, but we may say
that, in our view, the relations between this network of in-
dividual relationships, that together make up a community, and
political structure, the relations that exist between territorial
segments, present a problem of considerable importance, and
we make some comments on it. {a) Social relationships are
ordered by the political structure, so that a man's social sphere,
and the joint social sphere of a number of persons living in the
same village, tend always to be limited by the extension of
their political groups, {b) Local communities, relations between
which constitute the political structure, are only groups because
of these many and varied relationships between the individuals
who compose them. But it is the organization of these relation-
ships into groups standing in a certain relation to one another
within a system that interests us in our present discussion and
we only study them in this organized form ; just as one can, for
certain purposes, study the relation between organs of the body
without studying the interrelation of the cells that compose
the organs, (c) In our view the territorial system of the Nuer
is always the dominant variable in its relation to the other
social systems. Among the Nuer, relationships are generally
expressed in kinship terms, and these terms have great emotional
content, but living together counts more than kinship and, as
we have seen, community ties are always in one way or another
turned into, or assimilated to, kinship ties, and the lineage
system is twisted into the form of the territorial system within
which it functions.
10. We have defined structure by what amounts to the pre-
sence of group segmentation and have discussed some Nuer
systems from this point of view. We again emphasize that we
do not insist on our definition and that we recognize that struc-
ture can be otherwise defined. But having so defined it, frequent
allusion to a principle of contradiction in it was forced on us.
To avoid misunderstanding, however, we would remark that
the contradiction we have alluded to is on the abstract plane
of structural relations and emerges from a systematization of
"♦so? M m
266 THE AGE-SET SYSTEM
values by sociological analysis. It is not to be supposed that we
mean that behaviour is contradictory or that groups stand in
contradiction to one another. It is the relations of groups
within a system that constitutes and exemplifies the principle.
There may sometimes be conflict of values in the consciousness
of an individual, but it is structural tension to which we refer.
Likewise when we refer to the relativity of the structure we do
not mean that a group is anything than an actual mass of
people who can be seen and counted and plotted in space and
time. We mean that on the plane of structural relations its
position in a system is relative to the functioning of the system
in changing situations.
II. Besides making a contribution to the ethnology of the
Nilotes we have attempted in this book a short excursion into
sociological theory, lmt^wexan.X)nly make a theoreticaLanalysis
up to a certain p oint, beyond which we perceive vaguely how
furJtiieiLanaJ^isis, might be made. Our experience in research and
in writing this essay has intimated the lines of more extensive
treatment. Social anthropology deals at present in crude con-
cepts, tribe, clan, age-set, &c., representing social masses and a
supposed relation between these masses. The science will make
little progress on this low level of abstraction, if it be considered
abstraction at all, and it is necessary for further advance to use
the concepts to denote relations, defined in terms of social
situations, and relations between these relations. The task of
exploring new country is particularly difficult in the discipline
of politics where so little work has been done and so little is
known. We feel like an explorer in the desert whose supplies
have run short. He sees vast stretches of country before him
and perceives how he would try to traverse them ; but he must
return and console himself with the hope that perhaps the little
knowledge he has gained will enable another to make a more
successful journey.
INDEX
Adoption, of Dinka, 221-4, 228, 234.
Age-set(s), {ric), Chap. VI, 4, 16, 107,
123, 151, 169, 178-9, 250.
— and initiation, 250 ; and time-
reckoning, 105 ; and tribal warfare,
162.
— distance, 113.
— have no pohtical functions, 253.
— lapse of time between, 251.
— membership determines behaviour,
253-
— naming of, 252.
— not a military organization, 253,
254-
— ritual observances of, 257.
— respect shown to senior sets, 258.
— • stratified by seniority, 253.
— sub-divisions of, 251-3.
— typical examples of, 251.
Age-set system, the, 6, 249-66;
general characteristics of, 254-7 ;
has great social importance, 249 ;
is a tribal institution, 260.
Aliab Dinka, i.
Ancestral ghosts, 18, 209, 222.
Antonio, Philippe Terranuova D',
126.
Anuak, the, 3, 5, 16; overrun by
Nuer, 132-3.
Arabs, 133, 188.
Arnaud, i.
Atwot, 3, 127.
Austin, Capt. H. H., 134.
Azande, 15.
Bahr el Arab, river, 52, 54.
Bahr el Ghazal, river, i, 3, 52, 54.
Bahr el Jebel (Nile), 52, 54.
Bahr el Zeraf, river, i, 53.
Baker, Sir Samuel, 134.
Balak Dinka, 62.
Balanites aegyptiaca, 33, 60, 75.
Bar (bleeding of cattle), 27-8.
Baro, river, 52.
Beir (Murle) people, 133.
Bibliography, i, 2, 126-7, I34-
Blood-feuds, see under Feuds.
Blood-wealth, 121-2, 217-18; see
also under Cattle Payments.
Bovine pleuro-pneumonia, 68.
Bride-wealth, 17, 25, 69; disputes
about, 167-8, 222, 224; see also
under Cattle Payments.
Brun-Rollet, i, 2, 173, 187.
Burial, 209-10.
Burun peoples, 133.
Buth, see under Kinship.
Camps, dry-season, 65-6, 112, 1 15-16;
see also under wee.
Cany lineage, 243-4.
Casati, G., 127, 134.
Categories, social, 7, 109, 1 14-17,
130, 226-7, 234-5-
Cattle: a Nuer's dearest possession,
16.
— and Dinka, 218.
— and time-reckoning, 101-2.
— bleeding of, 27-8.
— breeding of, 34-6.
— byres, 65, 218.
— camps, 60, 62, 65, 112, 115-16.
— castration of, 33.
— cause hostilities, 48-9.
— census of, 20.
— chiefly used for milk, 21.
— colour-terms, 41-8.
— debts, 165.
— disposal of, in the family, 17-18.
— furnish household necessities, 28-9.
— influence behaviour, 16, 18, ig,
36-8, 40-1.
— interest in. Chap. I.
— intimacy with men, 36-7.
— it is cattle that destroy people, 49.
— milking of, 22-3, 31.
— names, 46; used in songs, 46-7.
— owned by families, 17.
— pasturing of, 17, 31-2.
— payments, 20, 22, 161, 217-18,
235-
— personal names derived from, 18.
— play a foremost part in ritual, i8.
— • sacrifices, 26-7 ; see also under
Sacrifices.
— uses of, 21, 26, 28-30.
— vocabulary, 41-8.
Churning, 24.
Cieng (residence), 96, 115, 117, 147,
195. 238, 242; meaning of, 135-6,
204-5 ; named after its owner, 204.
Clan, the, 4, 6, 179, 194, 236, 247; a
man can never change his clan,
213; characteristics of, 197; defini-
tion of, 192 ; dominant clan, 7, 174,
212-13, 214, 220, 225; is a system
of lineages, 6, 201 ; mythological
relationship, 237, 239; structural
form of, 200, 241.
Climate, 51-2, 75.
Corfield, F. D., 65, 173.
Coriat, P., 2.
268
INDEX
Couret, C. L. du (Hadji-Abd-el-
Hamid Bey), i, i6, 126.
Crazzolara, Father J. P., 2.
Crispin, Dr. 186.
Crop-rotation, absence of, 79-80.
Culture, material, 85-90.
Cuong (right), 171.
Cut (blood-wealth), 121; see also
under Blood-wealth.
Death, during childbirth, 168.
Demar ('my brother'), 219.
Deng, 185.
Dep (war line), 115.
DJiov (section of a village), 114, 135.
DU (aristocrat), 179, 212, 214, 220,
230. 234.
" — and blood-cattle payments, 217-
18.
— is a relative term, 235.
— status of, 216-17.
— various meanings of, 214-16.
Dinka, 3. 5, 7, 16, 20, 50, 69, 87, 93,
131, 150, 181, 188, 193, 194, 235,
263.
— adoption by Nuer, 221-4.
■ — and mythology, 125-6.
— captured Dinka, 221-2.
— enemies of the Nuer, 125.
— have same culture and social
systems as Nuer, 125.
— incorporation into Nuer lineages,
227.
— intermarriage with Nuer, 225.
— ritual of adoption, 222.
— status in Nuer society, 218-20.
— value in blood-cattle, 218.
Disputes, see under Feuds.
District, 116, 226.
Divorce, 167.
Domestic groups, 4-5.
Dwac (duelling), 161.
Dwil (or ut), living-hut, 114.
Eastern Jikany, 217-18, 233, 251.
Economics, 90-2.
Elders [gaat twot), 163-4, ^^g, 178;
authority of, 179-80.
Emin Pasha, 127.
Exogamy, rules of, 6, 17, 192, 202,
225-8, 235, 257.
Family relationships, and Age-sets,
258.
Fergusson, Capt. V. N., 1, 2, 185.
Fetishes, 177, 184.
Feuds, 5, 69, 145, 146, 147, 150-62,
166, 209.
— a political institution, 160.
— and sacrifices, 156.
Feuds, between lineages, 209.
— blood-feuds, 161; a tribal institu-
tion, 150.
— causes of, 151.
— gat ter (child of feud), 155.
— go on for ever, 154.
— people may not eat in the same
homestead, 158.
— settlement of, 152-8, 172.
Fieldwork, conditions of, 7-15.
Fishing {also Fish), 60, 91 ; a dry-
season occupation, 75 ; harpoon
fishing, 72 ; seasonal consumption
of fish, 82-3 ; methods of, 70-2.
Food, 56, 75, 81-5.
Gaagwang tribe, 139, 207, 236, 243.
Gaagwong lineage, 243, primary .sec-
tion, 243.
Gaajak tribe, 139, 194, 207, 232, 236.
Gaajok tribe, 139, 194, 236.
Gaaliek secondary section, 139.
Gaatbal secondary section, 139.
Gaatgankiir clan, 194, 197, 202-3,
207-8, 214, 224, 231-2, 237, 239-41.
Gaatgwan ('children of the father'),
247.
Gaatleak clan, 174.
Gaatman ('children of the mother'),
247.
Gaat nyiet ('children of girls'), 226-7.
Gaawar tribe, 62, 119, 121, 141, 194,
212, 230-1.
Galla people, 13, 87, 133.
Gar (initiation rite), 249.
Gessi, R., 127.
Gier (small bell), 46.
Gleichen, Count, i, 134.
Gol (hearth, joint family), 65, 135,
219, 222, 247.
Government courts, 162-3, 172.
Grasses, 58-60.
Guk (prophet), 185.
Gun primary section, 62, 139.
Gwandong ('first ancestor'), 195.
Gwan kivoth (possessor of a spirit),
185.
Gwan thunga (slayer), 158.
Gwan wal (magician), 167.
Hadji-Abd-el-Hamid Bey [see under
Couret) .
Hamer, L. F., 71.
Heuglin, 126.
Homestead, the, 114, 194, 197, 202,
224, 231, 237, 239, 241.
Homicide, 152-5, 162, 165, 217, 220;
— - and cattle payments, 155, 156; see
also under Feuds.
Horticulture, 16, 56, 63, 75-81.
Horticultural calendar, 78.
INDEX
269
Huffmann, Miss Ray, 2.
Hunting, 72-4, 91 ; a dry-season
occupation, 75.
Hurst, H. E. (and Phillips, P.), 53-
Initiation {see also under Age-set), 6,
40, 178 ; and taboos, 254 ; a time of
mild licence, 250 ; boys initiated in
groups, 249; operation of {gar),
249 ; period of, 249 ; radically
changes duties and privileges, 254.
— rites: distinguish Nuer culture, 260.
■ — sometimes organized by prophets,
250.
Insects, 66-8.
Jaang {see also under Dinka), 126,
216, 217, 219, 220, 224, 234-5.
Jackson, H. C, 2, no, 134, 185.
Jiciengthu (in-laws), 210.
J ikany tribes, 49, 62, 140, 181, 207-8,
217, 218, 231-3, 239; said to be of
Dinka origin, 3-4.
JiMEM clan, 174, 245.
JiNACA clan, 194, 196, 200-1, 206,
214, 237, 246.
J torn (season), 96-9.
Jur (cattleless peoples), 130.
Kal (fishing-camps), 70.
Kinship: agnatic {buth) relationship,
191, 193, 194, 222, 225-6, 228-9,
239-40; and tribal warfare, 162;
and age-set system, 259-60; (see
also under Mar).
— group, 6.
— system, 106, 165, 181, 183, 226,
228.
— ties, 17.
Kir myth, 231-2, 236.
Kuaa yiika (mediator), 168,
Kuaar muon, see under Leopard-skin
chief.
Kur (fight between tribes), 161.
Kwai ('seeds', ancestors), 195.
Kwal (theft), 165.
Lak tribe, 141, 251.
Lang (brass ring), 46.
Lango, the, 3.
Leek tribe, 49, 121, 146, 212, 213, 214,
234. 251-
Legal forms, 6, 17, 162-8, 219; adul-
tery, indemnity for, 162, 165-6;
and bridewealth, 167-8 ; and feuds,
171; and fornication, 168-9;
basis of law is force, 169; disputes
about ownership of cattle, 165 ;
Nuer have no law, 162 ; sanctions,
150; status of Dinka, 221; theft,
165.
Lejean, G., i, 126.
Leng tertiary section, 139.
Leopard-skin chief {Kuaar Muon), 5,
6, 16, 161, 168, 189; as mediator,
163-4, 172. 174-5; exhortations
and threats of, 175-6; has a poli-
tical function, 173; has no great
authority, 172, 174; has sacred
associations with the earth {mun),
172; is a sacred person, 5; is a
ritual agent, 173, 190; lineage
aftiliations of, 173-4; role in homi-
cide, 152-5.
Lewis, B. A., 62, 230-1, 141.
Lineages, 4, 107, 183; admixture of,
209 ; and mythology {see under
Myths); are agnatic, 6, 193, 200;
as Nuer know them, 199, 202-3;
buth relationship, 194 {see also
under Kinship) ; characteristics of,
195 ; cleavage of, 247 ; definition of,
191, 196; dispersal of, 209 ; distance,
113; founders of, 199; have spear-
names, 193; identified with terri-
torial sections, 204 ; include dead
persons, 193 ; major lineages, 6,
199, 245 ; maximal lineages, 6, 192,
199, 245, 247 ; minimal lineages, 6,
199, 210; minor lineages, 6, 199,
210.
— of Jikany tribes, 208.
— of Lou tribe, 206, 241.
— of the dominant clan, 211-13 ; one
man a potential lineage, 210; rela-
tive term, 195; ritual relations be-
tween, 210 ; spring from few names,
199 ; structure, 197-8 ; time-depth
of, 201.
Lineage-system, the. Chap. V, 6 ; as
a record of descent, 246 ; of the
Gaatgankiir clan, 241-3; and
time-reckoning, 106-7.
Lou tribe, 26, 58, 62, iii, 139, 142,
148, 181, 205, 206, 213, 229-30,
237, 238, 241, 251-3.
Lue (tassel), 46.
Luo, 3, 14, 132.
Luth (large bell), 46.
Magic, 167.
Magicians, 177, 178.
Mahdi, revolt of the, i.
Mahdism, 187.
Alai (season), 95-9.
Malte-Brun, V. A., 127, 129, 130.
Man of the Cattle, see under IVut
Ghok.
Mar (cognation), 193, 194, 228.
Marno, Ernst, i, 2, 16, 127.
Marriage, 17, 20, 40, 104, 124, 166,
167, 178, 192, 211, 222, 225.
270
INDEX
Meat, 26 ; seasonal consumption of,
82-3.
Migration, causes of, 127; seasonal
movements of Nuer, 62-3.
Military organization, 254.
Milk, churning, 24-5 ; consumed in
various ways, 23-4 ; importance of,
21 ; place in food-supply, 81-2.
Milking, Nuer process of, 22-3.
Months, 99.
Mor primary section, 62, 139, 143.
Myths, 49, 81-2, 83, 108, 229-34,
237, 239; of Gaatgankiir clan,
231-2, 236; of Gaawar tribe,
230-1 ; of Lou tribe, 229-30; poli-
tical relations are mythologically
represented, 232.
Nath (Nuer), 3.
Nation, definition of, 5.
Ngat (cutting of bull's horns), 38.
Ngok Dinka, 127, 225.
Ngundeng, 185, 188.
Ngundeng pyramid, the, 186.
Ngwal (debt), 165.
Nilotes, Cultural classification of, 3.
Nuer, the, a unique community, 123 ;
aloofness of, 134; census of, 3;
character of, 183-4; common
origin with Dinka, 3,4; competence
as herdsmen, 31-40; diet, 81-2;
difficult to question, 12-13 ; dis-
tribution of, 5 ; earlier papers on,
1-2; earlier travellers to, 126;
Eastern and Western, 5 ; frequently
fight over cattle, 48-50; handi-
work of, 85-7; are deeply demo-
cratic, 1 8 1-2; months, 99; migra-
tion of, 62, 64 ; physique of, 3 ; pre-
eminently pastoral, 16, 209; pride
of, 90, 182 ; raid Dinka for cattle,
50, 84, 126-9; sense of personal
dignity, 171 ; similar to Dinka, 125 ;
sub-divisions of, 3; tribes, 117-25,
139, 238; unaffected by Arab con-
tact, 133; unite with Dinka, 130;
why Nuer raid the Dinka, 131-2,229.
Nuerland, main physical character-
istics of, 55.
Nyang myth, 229-30.
Nyarkwac tertiary section, 139.
Nymphaea lotus, 75.
Oecology, Chap. II ; and time, 94-
104; and space, 109.
Ox-bell, 32.
Ox -names, 18, 37, 44, 46.
Pec (raiding Dinka), 161.
People, definition of, 5.
Peth (witch), 167.
Petherick, Mr. and Mrs., i, 126.
Pibor, river, 53, 112.
Political distance, 114, 124.
— structure, 4, 94, 125, 147 ; and clan
system, 211.
Political .sy.stem, the, 5, 20, 139-91,
228; and blood-wealth, 148-9;
fighting between segments, 142,
147 ; influences lineage .structure,
247 ; of Eastern Jikany tribes, 140 ;
of Gaawar tribe, 141 ; of Lak tribe,
141 ; of Lou tribe, 139-40; of Thiang
tribe, 141 ; tribal segments, 139.
Poncet, Jules, i, 3, 120, 126, 127, 133,
172, 187.
Poon {Oryza Barthii) grass, 36, 75.
Population, density of, no; distribu-
tion of, 111-13; of tribes, 117.
Prophets, 6, 16, 149, 172, 176, 178,
185-9 ; are possessed by spirits, 186 ;
and earlier travellers, 187; exag-
gerated power of, 187; initiate
raids against Dinka, 188.
Rainfall, 51-2, 55, 95-
Pic, see under Age-sets.
Riek (sacred pole), 209.
Rinderpest, 19, 82, 91 ; a cause of
migration, 56 ; a cause of social
changes, 69 ; decimates the herds,
68 ; prevents dependence on milk
foods, 92-3.
Ritual, 28, 152, 166, 172, 176, 177,
211.
Rivers, 52 ; flooding of, 53-4.
Rul (stranger), 212, 216-17, 220,
234-5 ; is a relative term, 235.
Rumjok secondary section, 139.
Ruok (compensation), 121, 162.
Ruon (year), 95.
Rwil (season), 96-9.
Sacrifices, 26, 27, 85, 128, 152, 155,
156, 165, 189, 222, 235.
Schweinfurth, G., 127.
Seasons, 95.
Secondary sections, 5, 139, 169.
Seligman, C. G. and B. Z., 2.
Seroot fly, 67.
Shilluk people, 3, 5, 182; have little
contact with Nuer, 132.
Sky-gods, 6, 121, 185, 186, 188, 189.
Sobat, river, i, 3, 5, 52, 112.
Soil, 52.
Songs, 46-7.
Sorghum vulgare (millet), 21, 55,
76-8, 81-2, 84.
Soule, Miss, 68, 145.
Space, social categories of. Chap. Ill ;
measurements of, 109-38; and
population, iio-ii.
INDEX
271
Stigand, Major C. H., i, 2.
Structure, social, 4, 50, 131, 138-9,
147-50, 158-9, 169, 176, 189, 262-6;
structural time, 104-8 ; structural
distance, 108-13, 135-6, 157, 176,
200-1, 241 ; see also under Systems.
Struve, K. C. P., 125, 188.
Stumoxis fly, 67.
Sudan Intelligence Reports, i, 88,
134. 172-
Sudan Notes and Records, i, 13, 42,
126.
Systems, social, 4, 5, 6, 125, 135,
190-1, 212, 243, 255-7; ^^^ ^'^o
under Structure.
Taboos, on milking, 254 ; on sexual
intercourse, 129.
Temperature, 51, 52.
Ter (feud), 161.
Terms, definition of, 5-7.
Tertiary sections, 5, 139.
Thiang clan, 194-5, 198.
Thiang tribe, 141, 212; primary sec-
tion, 242.
Thibaut, i.
Thok dwiel (lineage), 192, 204, 222,
247.
Thok mac (lineage), 195.
Time-reckoning, 94-108, 199; daily
time-piece is cattle-clock, loi ; his-
torical time, 105, 107 ; and age-set
system, 105 ; and mythology, 108 ;
and the individual, 107.
— Oecological time, 94 ; the year the
largest unit of, 104.
— Structural time, 94, 104, 107; a
reflection of structural distance,
108.
Tools, manufacture of, 86.
Tot (season), 95-9.
Totemic specialists, 177.
Trade, 87-8.
Travellers, early, 126, 172.
Tribal limits, not limits of social
intercourse, 124.
Tribal structure, 235-6, 240, 245.
Tribe, the, 4, 5 ; and blood-wealth
{cut), 121, 161; and warfare.
120-1, 161; census of, 117; and
territorial distribution, 149; defi-
nition of, 122, 148 ; largest group to
combine for defence, 5, 120; largest
political segment, 5, 147 ; main
Nuer tribes, 117; mythological re-
lationship between tribes, 238-40 ;
names of, 119 ; other tribes on same
pattern as Lou tribe, 140; seg-
ments of, 5, 137, 140-7, 228, 234;
tribal sentiment, 1 20 ; tribal system.
Chap. IV; tribal value, 149, 161;
intra-tribal relations, 118-19.
Tilt ('bull', elder), 44, 179-81, 213,
215-
Vigna (beans), 76.
Village, the, a political unit,
names of villages, 207.
— sites, 63-5 ; communities, 92.
115;
War specialists (gwan miiot), 177.
Water-supply, 56, 60.
Weaning ring, 35.
Wee (dry-season camp), 96, 115,
213.
Wee jiom (early cattle-camps), 60.
Wee mat (late cattle-camps), 60.
Wells, 56, 58.
Werne, Ferdinand, i, 16, 126, 172.
Westermann, Prof. D., 2.
Western Jikany, 251, 252-3.
Wild fruits, seeds, and roots, 75.
Williams, Capt. H. B., 31.
Willis, C. A., 185.
Windscreens, 65, 66.
Wutghok (Man of the Cattle), 16, 177;
has ritual relationship to cattle,
177; determines initiation periods,
250.
Zeraf, river, 5, 112.
PRINTED IN
GREAT BRITAIN
AT THE
UNIVERSITY PRESS
OXFORD
BY
JOHN JOHNSON
PRINTER
TO THE
UNIVERSITY
iwU'
University of California Library
This book is DUE on the 1^ J^^^S^J^ P^^ below.
310/8
Phone Renewals
125-9188
?
QU.
m
4K26^.:
\j\
SMN I ?005
^AN 1 7 200G
ore ^ 5 ?005
L 005 866 331 1
