TUC R I T I I A I

1 1 1 LB. 1 1 1 1 %J i \ L
P
i""^ /^%
H & ^^^^ ^^^r ^^^y v^^g
Structure and
Anti-Structure
VICTOR TURNER
The Lew/s Hen/y Morgan Lectures | 7966
presented at The University of Rochester,
Rochester, New York
Cornell Paperbacks
Cornell University Press
I THACA, NEW Y ORK
lite.*..-
SYMBOL, MYTH, AND RITUAL SERI ES
General Editor: Victor Turner
Raymond Firth, Symbols: Public and Private*
Eva Hunt , The Transformation of the Hummingbird: Cultural Roots of a
Zinacanlecan Mythical Poem
Bennetta Jules-Rosette, African Apostles: Ritual and Conversion in the
Church of John Maranke*
Sally Falk Moore and Barbara G. Myerhoff, eds., Symbol and Politics in
Communal Ideology: Cases and Questions^
Barbara G. Myerhoff, Peyote Hunt: The Sacred Journey of the Huicho'l
Indiansj
Victor Tur ner , Dramas, Fields, and Metaphors: Symbolic Action in Human
Socielyj
Victor Tur ner , Revelation and Divination in Ndembu Ritual^
Victor Tur ner , The Ritual Process: Structure and Anti-Structurej
Roy Wagner, Lethal Speech: Daribi Myth as Symbolic Obviation
*Also available in a Cornell Paperbacks edition.
tAvailable from Cornell University Press only in a Cornell Paperbacks edition.
To the memory of Allan Holmberg
this book is respectfully dedicated.
Copyright i g6g by Victor W. Turner
Foreword to Cornell Paperbacks edition copyright 1977 by Cornell University
All rights reserved. Except for brief quotations in a review, this book, or parts
thereof, must not be reproduced in any form without permission in writing
from the publisher. For information address Cornell University Press, 124
Roberts Place, Ithaca, New York 1 4850.
First published i g6g by Aldine Publishing Company.
First published, Cornell Paperbacks, 1 9 7 7 .
Seventh printing 1 9 9 1 .
Library of Congress Cataloging in Publication Data
(For library cataloging purposes only)
Turner, Victor Witter.
The ritual process.
(Symbol, myth, and ritual series) (Cornell paperbacks ; CP- 1 63 )
Reprint of the ed. published by Aldine Pub. Co., Chicago, in series: The
Lewis Henry Morgan lectures, i g66.
Bibliography: p.
Includes index.
1 . Rites and ceremonies. I. Ti de. II. Series: The Lewis Henry Morgan
lectures ; 1966.
[ GN47 3 . T8 2 1 9 7 7 ] 3 0 1 . 2' 1 7 6- 56627
ISBN 0- 801 4- 9 1 63- 0
Printed in the United States of Ameri ca
The paper in this book meets the minimum requirements
of the American National Standard for Information Sciences
Permanence of Paper for Printed Library Materials, ANSI Z39.48- 1984.
Foreword to the
Cornell Paperbacks Edition
Recent l y bot h t he r es ear ch and t heoret i cal concer ns of many
ant hr opol ogi st s have once agai n be e n di r ect ed t owar d t he rol e
of symbol srel i gi ous, myt hi c, aest het i c, political, a n d even
economi ci n social and cul t ural processes. Wh e t h e r this r e-
vival is a bel at ed r es pons e to devel opment s in ot he r disciplines
(psychology, et hol ogy, phi l osophy, linguistics, to n a me onl y a
few), or whe t he r it reflects a r e t ur n t o a cent ral concer n aft er a
per i od of negl ect , is difficult to say. I n r ecent field st udi es, an-
t hropol ogi st s have been collecting myt hs and ri t ual s in t he con-
text of social act i on, and i mpr ovement s in ant hr opol ogi cal field
t echni que have pr oduc e d dat a t hat ar e r i cher and mor e refi ned
t han her et of or e; t hese new dat a have pr obabl y chal l enged
t heoret i ci ans t o pr ovi de mor e ade quat e expl anat or y frames.
What ever may have been t he causes, t he r e is no de nyi ng a r e-
newed curi osi t y about t he nat ur e of t he connect i ons bet ween
cul t ur e, cogni t i on, and per cept i on, as t hese connect i ons ar e
reveal ed in symbolic forms.
Al t hough excel l ent i ndi vi dual monogr a phs and articles in
symbolic ant hr opol ogy or compar at i ve symbol ogy have recent l y
appe ar e d, a c ommon focus or for um t hat can be pr ovi ded by a
topically or gani zed series of books has not been available. T h e
v
vi
FOREWORD
pr e s e nt series is i nt e nde d t o fill this l acuna. It is des i gned to
i ncl ude not onl y field mon ogr a ph s and t heoret i cal and compar -
ative st udi es by ant hr opol ogi s t s , but also work by schol ars in
ot he r di sci pl i nes, bot h scientific and humani st i c. T h e appear -
ance of st udi es in such a f or um encour ages emul at i on, and em-
ul at i on can pr oduc e frui t ful new t heori es. It is t her efor e our
h ope t hat t he series will serve as a hous e of many mansi ons,
pr ovi di ng hospi t al i t y for t he pr act i t i oner s of any discipline t hat
has a seri ous a n d creat i ve concer n with compar at i ve symbology.
To o oft en, disciplines a r e seal ed off, in sterile pedant r y, fr om
significant i nt el l ect ual i nfl uences. Nevert hel ess, our pr i mar y
ai m is t o br i ng to publ i c at t ent i on wor ks on ri t ual and myt h
wr i t t en by ant hr opol ogi s t s , and our r e ade r s will find a variety
of strictly ant hr opol ogi cal appr oache s r angi ng fr om formal
anal yses of syst ems of symbol s to e mpat he t i c account s of divin-
at or y a n d i ni t i at ory ri t ual s.
Thi s book is based on t he Lewis He n r y Mor gan Lect ur es at
t he Uni versi t y of Roches t er whi ch I del i ver ed in 1966. It was in
t he cour s e of t hese l ect ur es t hat I crossed t he t hr es hol d be-
t ween t he st udy of r i t ual in an Afri can tribal cont ext a n d t he
analysis of pr ocessual symbol s in cross-cul t ural and t r ans t em-
por al t er ms . The Ritual Process and s ubs equent books of mi ne
have pr oduc e d t hei r s har e of cont r over s y over t he years. Mor e
t han once I have be e n accused of over gener al i zi ng a n d of mis-
appl yi ng concept s like "l i mi nal i t y" a n d "communi t as . " Th e s e
t er ms , it is ar gue d, may adequat el y descri be or account for
social and cul t ur al processes and p h e n o me n a f ound in prelit-
er at e societies, but have l i mi t ed use in expl ai ni ng soci ocul t ural
syst ems of muc h gr e at e r scale and compl exi t y.
T o at t e mpt t o ans wer such criticisms is probabl y a futile exer -
cise. I am unabl e, however , t o resist quot i ng t he adage " Th e
pr oof of t he puddi n g is i n t he eat i ng. " Thi s book has been cited
r epeat edl y by schol ars i n such di ver s e fields as hi st ory, t he his-
t ory of rel i gi ons, Engl i sh l i t er at ur e, political science, t heol ogy,
and dr a ma , as well as i n ant hr opol ogi cal and sociological books
a n d articles conce r ne d wi t h ri t ual a n d semiotics, part i cul arl y in
Afri can cont ext s; its r ecept i on e nc our a ge d me t o ext end t he
FOREWORD Vll
compar at i ve ent er pr i s e. In Dramas, Fields, and Metaphors, an-
ot he r work in t he Symbol , Myt h, and Ritual series, several case
st udi es ar e based on t he as s umpt i on, first devel oped her e, t hat
society is a process r at he r t han an abst ract system, whet her of
social st r uct ur al rel at i ons or of symbols and meani ngs .
Society, mor eover , is a process in which any living, relatively
wel l -bonded h u ma n gr oup al t ernat es bet ween fixed andt o
bor r ow a t e r m fr om our J apane s e friends"floating worl ds. "
JBy^er baLand_j i pnvr ^al _means of das s m^cadon^^e^i mpos e
xupj pn. pur sel yes i nnume r abl e const rai nt s and boundar i es to
keep chaos at bay, but oft en at t he cost of failing t o make j i i s -
,_cpveries a n d i nvent i ons : t h a t is -to say . . noLal l i nst ances. of s ub- ,
versi on of t he nor mat i ve ar e devi ant or cr i mi nous. Yet -i n-order
jto live, t o br eat he, andj p. gener at e. novel t y, h uma n bei ngs have
had t o creat eby st r uct ur al meansspaces a n d t i mes in t he
cal endar or , in t he cul t ural cycles of t hei r mo s t cher i shed -
gr oups whi ch cannot be capt ur ed in t he classificatory net s of
. their quot i di an, r out i ni zed s phe r e s . oLact i on. . These_Jmi i nal
areas. pf t i me a n d spaceri t ual s, carnivals, dr amas , and latterly
filmsare ope n to t he play of t hought , feel i ngs and_wj l l ; Jn
t hem ar e ge ne r at e d new model s, often fantastic, s ome of which
may have sufficient power and plausibility to repl ace event ual l y
t he force-backed political a n d j ur al model s t hat cont rol t he cen-
t ers of a society' s ongoi ng life.
T h e ant i st r uct ur al liminality pr ovi ded in t he cores of ri t ual
and aes t hedc for ms r epr es ent s t he reflexivity of t he social pr o-
cess, wher ei n society becomes at once subject and di r ect object;
it r epr es ent s also its subj unct i ve mood, wher e supposi t i ons, de-
sires, hypot hes es , possibilities, and so fort h, all become legiti-
mat e. We have been t oo pr on e to t hi nk, in static t er ms , t hat cul-
t ural s uper s t r uct ur es ar e passive mi r r or s , me r e reflections of
s ubs t r uct ur al pr oduct i ve modes and rel at i ons or of t he political
processes t hat enfor ce t he domi nance of t he product i vel y privi-
l eged. If we wer e as dialectical as we claim to be, we woul d see
t hat it is mor e a mat t er of an existential be ndi ng back upon
oursel ves: t he s ame pl ur al subject is t he active s upe r s t r uct ur e
t hat assesses t he s ubs t r uct ur al and st r uct ur al modal i t i es t hat we
also ar e. Ou r concr et eness, our substantiality is with us in our
vm
FOREWORD
reflexivity, even in t he l udi c play domai n of cert ai n of our
l i mi nal mome n t s : play is mor e seri ous t han we, t he i nher i t or s
of We s t e r n Pur i t ani s m, have t hought .
The Ritual Process r epr es ent s an at t empt to free my own
t hought , a n d I h ope t hat of ot her s in my field as well, fr om
gr oove d de pe n de n c e on "s t r uct ur e" as t he sole sociological di -
me ns i on. A good deal of t he philistinism pe r haps ri ght l y
ascr i bed t o o u r di sci pl i ne has been due t o this "single vision," as
Wi l l i am Bl ake woul d have called itthis obdur at e evasion of
t he ri ch compl exi t i es of cul t ural creat i on. Ou r goal s houl d be t o
s t udy ma n alive a n d woman alive, in t he many levels of t hei r
mut ua l deal i ngs. Thi s book is, if not hi ng mor e , a modes t st ep
t owar d real i zi ng t hat goal.
VICTOR TURNER
University of Chicago
Contents
Planes of Classification in a Ritual of Life and Death
Paradoxes of Twinship in Ndembu Ritual 44
Liminality and Communitas 94
Communitas: Model and Process 1 3 1
Humility and Hierarchy: The Liminality of
Status Elevation and Reversal 166
Bibliography 204
Index 209
1
Planes
of Classification
in a Ritual
of Life and Death
MORGAN AND RELIGION
It must first be said t hat for me, as for many others, Lewis Henry
Morgan was one of the lodestars of my student days. Everything he
wrote bore the stamp of a fervent yet pellucid spirit. But, in under-
taking to deliver the Morgan Lectures for 1966, I was immediately
conscious of one profound, and it might seem crippling, disadvan-
tage. Morgan, t hough he faithfully recorded many religious cere-
monies, had a marked^ disinclination to give the study of religion,
the same piercing attention he devoted to kinship and politics. Yet
religious beliefs~and pracfices~weie t h e ma i n subject mat t er of my
talks. Two quotations especially emphasize Morgan' s attitude. The
first is taken from his seminal classic Ancient Society ( 1 877) : " T h e
growth of religious ideas is environed with such intrinsic difficulties
that it may never receive a perfectly satisfactory exposition. Religion
deals so largely with t he imaginative and emotional nat ure, and
consequently with such uncertain elements of knowledge, t hat all
primitive religions are grotesque and to some extent unintelligible"
(p. 5). The second consists of a passage from Merle H. Deardorff' s
(1 95 1 ) scholarly study of the religion of Handsome Lake. Morgan' s
account of Handsome Lake' s syncretic gospel in his book League of the
I
2
The Ritual Process
Ho-de-no-sau-nee or Iroquois was based on a set of notes made by young
Ely S. Parker (a Seneca Indian, who was later to become General
Ulysses S. Grant' s military secretary), consisting of the texts and
translations of Handsome Lake' s grandson' s Good Message recitals
at Tonawanda. According to Deardorff, " Mor gan followed Ely's
notes faithfully in reporting what Ji mmy Johnson, the prophet' s
grandson, said, but he departed widely from Ely's glosses on it and
i t sceremoni al accompani ment " (p. 98; see also Wi l l i amFent on, 1 941 ,
PP- I 5 I - I 5 7 ) -
The correspondence between Morgan and Parker shows that if
Morgan had listened more carefully to Ely, he might have avoided
the general criticism of his " Le ague " made by Seneca who read i t :
"Ther e' s nothing actually wrong in what he says, but it isn' t right
either. He doesn' t really understand what he is talking about . " Now,
what did these Seneca "r eal l y" mean by these extraordinary re-
marks, which seem to be addressed to Morgan' s work on the religious,
rat her t han the political, aspects of Iroquois culture. To my mind,
the Seneca comments are related to Morgan^s_djsiciisLoLthe_llimag--.
inative and. emot i onal . " his reluctance to concede t hat religion has
an i mport ant rational aspect, and his belief t hat what appears
"gr ot es que" to the highly "evol ved" consciousness of a nineteenth-
century savant must be, ipso facto, largely "uni nt el l i gi bl e. " They
also betray in him a related unwillingness, if not incapacity, to make
t hat empat het i c exploration of Iroquois religious life, t hat at t empt
to grasp and exhibit what Charles Hockett has called " t h e inside
vi ew" of an alien culture, which might well have made compre-
hensible many of its seemingly bizarre components and interrela-
tions. Indeed, Morgan might have pondered with salutary effect
Bachofen's (1960) words to him in a letter: " Ge r ma n scholars pro-
pose to make antiquity intelligible by measuring it according to
popul ar ideas of the present day. They only see themselves in the
creation of the past. To penetrate to the structure of a mi nd different
from our own, is hardy wor k" (p. 1 3 6). Upon this remark, Professor
Evans-Pritchard (1965b) has recently commented t hat " i t is indeed
hardy work, especially when we are dealing wi t h such difficult
Planes of Classification
3
subjects as primitive magic and religion, in which it is all too easy,
when translating the conceptions of the simpler peoples into our
own, to t ranspl ant our t hought into t hei r s " (p. 1 09). I would like
to add as a proviso here t hat in matters of religion, as of art, t here
are no " s i mpl e r " peoples, only some peoples with simpler tech-
nologies t han our own. Man' s "i magi nat i ve" and " e mot i onal " life
is always and everywhere rich and complex. Just how rich and com-
plex t he symbolism of tribal ritual can be, it will be par t of my task
to show. Nor is it entirely accurate to speak of t he " structure of a
mi nd different from our own. " It is not a mat t er of different cognitive
structures, but of an identical cognitive structure articulating wide
diversities of cultural experience.
Wi t h t he development of clinical depth-psychology, on the one
hand, and of professional anthropological field work, on the other,
many products of what Morgan called " t h e imaginative and emo-
tional n a t ur e " have come to be regarded with respect and attention
and investigated with scientific rigor. Freud has found in the fanta-
sies of neurotics, in the ambiguities of dream imagery, in wit and
punning, and in the enigmatic utterances of psychotics clues to t he
structure of the normal psyche. Levi-Strauss, in his studies of the
myths and rituals of preliterate societies, has detected, so he assever-
ates, in their underlying intellectual structure similar properties to
those found in the systems of certain modern philosophers. Many
other scholars and scientists of the most impeccable rationalist
pedigree have thought it well worth their while, since Morgan' s day,
to devote whole decades of their professional lives to t he study of
religion. I need only instance Tylor, Robertson-Smith, Frazer, and
Herbert Spencer; Durkhei m, Mauss, Levy-Bruhl, Hubert , and
Her z; van Gennep, Wundt , and Max Weber to make this poi nt .
Anthropological field workers, including Boas and Lowie, Mal i n-
owski and Radcliffe-Brown, Griaule and Dieterlen, and a host of
their coevals and successors, have labored mightily in t he vineyard
of preliterate ritual, maki ng meticulous and exacting observations of
hundreds of performances and recording vernacular texts of myt hs
and prayers from religious specialists wi t h loving care.
Silken- Univer
Library
4 The Ritual Process
Most of these thinkers have taken up the implicitly theological
position of trying to explain, or_explain away, religious phenomena
as the product ofjr^sycho]ogical or sociological causes of the most
diverse and even_com^icjMg^tyjp^ *h
m a n
Y pret erhuman
SQgilLL but none of t hem has_denied the extreme i mport ance of
religious beliefs and practices, for bot h the mai nt enance and radi-
cal t r a ns f or ma t i on^ human social and psychical structures. The
reader will perhaps be relieved to hear that I have no intention of
entering the theological lists but will endeavor, as far as possible, to
confine myself to an empirical investigation of aspects of religion
and, in particular, to elicit some of the properties oi ^Jncan~ri t uaI7~
Rat her will I try, in fear and trembling, owing to my high regard .
/ ' for his great scholarship and standing in our discipline, to witlv"
stand Mor gan^ casual challenge to posterity, and demonstrate that
modern anthropologists, working with _the best of the conceptual
tools bequeat hed to t hem, can...now make intelligible many of the
|^ cryptic phenomena of religion in preliterate societies. <*
RITUAL STUDIES IN CENTRAL AFRICA
Let us begin with a close look at some ritual performed by the
people among whom I did two and a half years' field work, the
Ndembu of northwestern Zambi a. Like Morgan' s Iroquois, the
Ndembu a r e j n a t r i h n g a l j mc y ^ withJb-unting,^
to whi ch they a t t a c h a high ritual value. The Ndembu belong to a
great congeries of West and Central African cultures, which conjoin
with considerable skill in wood-carving and the plastic arts an elab-
orat e development of ritual symbolism. Many of these peoples have
complex initiation rites with long periods of seclusion in the bush for
t he training of novices in esoteric lore, often associated with the
presence of masked dancers, who portray ancestral spirits or deities.
The Ndembu, together with their nort hern and western neighbors,
t he Lunda of the Kat anga, the Luvale, the Chokwe, and the Luchazi,
at t ach great i mport ance to ri t ual ; their eastern neighbors, the
Planes of Classification
5
Kaonde, the Lamba, and the Ila, although they practice much
ritual, appear to have had fewer distinct kinds of rites, a less exuberant
symbolism, and no boys' circumcision ceremonies; and their diverse
religious practices are less closely articulated with one anot her.
When I began field work among the Ndembu, I worked in the
tradition established by my predecessors in the employment of the
Rhodes-Livingstone Institute for Sociological Research, located at
Lusaka, the administrative capital of Nort hern Rhodesia (now
Zambi a). Thi s was the earliest established research institute in
British Africa, founded in 1938, and was intended to be a center
where the probl em of establishing permanent and satisfactory rela-
tions between natives and non-natives mi ght form the subject of
special study. Under the directorship of Godfrey Wilson and Max
Gluckman, and later of Elizabeth Colson and Clyde Mitchell,
research officers of the institute had made field studies of tribal
political and j ur al systems, of marriage and family relationships, of
aspects of urbanization and labor migration, of comparative village
structure, and of tribal ecological and economic systems. They had
also done a good deal of mappi ng work and had classified all the
tribes of what was then Nort hern Rhodesia into six groups in terms
of their descent systems. As Lucy Mai r (1960) has pointed out, the
contribution of the Rhodes-Livingstone Institute to t he shaping of
policy, like t hat of t he other research institutes in British Africa, lay
not in " prescribing the action appropri at e to specific situations " but
" rather in the analysis of situations in such a way t hat t he policy-
makers (could) see more clearly the forces with which they were
dealing." (pp. g8- i o6).
Among these "for ces/ ' ritual had_a_very low priority at the
time I began field work. Indeed, interest in ritual has never been
strong among Rhodes-Livingstone researchers: Professor Raymond
Apthorpe (1961) pointed out t hat of t he 99 publications of the
institute until that time dealing with various aspects of African life
during the last thirty or so years, only three had taken ritual for
their subject (p. ix). Even now, five years later, of t he 3 1 Rhodes-
Livingstone Papersshort monographs on aspects of Central
6 The Ritual Process
African tribal lifeonly four have made ritual their mai n t opi c, t wo
of t hem by the present aut hor. Evidently, Morgan' s at t i t ude to
"pr i mi t i ve rel i gi ons" still persists in many quart ers. Yet t he insti-
tute' s first director, Godfrey Wilson, took a lively interest i n t he
study of African ri t ual . His wife, Moni ca Wilson (1 95 4), wi t h wh om
he did intensive field research into t he religion of t he Nyakyusa
people of Tanzani a, and who has, herself, publ i shed out s t andi ng
studies of_citual, has pertinently wr i t t e n : " Ri t ual s reveal val ues at
their(jdeepest J gve l . . . me n e x p j ^ s s ^ r i t ual j vhat moves t hem most ,
and since t he form of expressiojiJs_conventionalized and obl i gat ory,
I t is~tEe~values of the_group t hat are revealed. I see in t he st udy of
rituals t he key to an underst andi ng of t he essej ui al const i t ut i on of
human societies" (p. 241 ) .
IfWIIsor?s view is correct, as I believe it is, t he study of t r i bal
ritual would certainly have been mj di e j [ pj nt ^f j ^^
aspiration " to study . . . t he probl em of_estabJisJy^gj3^^
satisfactory relations bet ween natives and non-native.s," for "s at i s -
fa^t5ryre1a"tiSn"s""llep^nd on a deep mut ual underst andi ng. I n con-
t r a s V^ e l t u d ^ o f religion has been pr omi nent in t he work of r esear ch
institutes in East and West Africa, especially in t he peri od j us t
before and after t he at t ai nment of political i ndependence. I n t he
social sciences generally, it is, I think, becomi ng widely recogni zed
t hat religious beliefs and practices_are_spmething more t han " gr ot -
esque " reflections or expressions of economic, poHt i cal , and social
relajdonshigsj_rather are they comi ng to be seen as deasi ye Jcevs^to
t he underst andi ng of how_p^ojjIe7thmk^andfeel about those rel a-
tionshirjs, and about t he nat ur al j md social e nvi r onme ^s ^nj whi ch
t hey operat e.
PRELIMINARY FIELD WORK ON NDEMBU RITUAL
I have dwelt on this "religious unmusi cal i t y" (to use t he t e r m Ma x
Weber quite unjustifiably applied to himself) of social scientists of
my generation wi t h regard to religious studies mai nl y to under l i ne
Planes of Classification
7
t he rel uct ance I felt at first to collect ritual dat a. For t he first nine
mont hs of field work, I amassed considerable guannt i e^ofj dat aj pn
kinship, vi l l age^t r ur t ur e, marri age and divorce, family and indi-
vj dual budget s, tribal and village politics, and t he agricultura]j:ycle.
I filled my notebooks wi t h genealogies; I made village hut-plans
and collected census mat eri al ; I prowled around to catch the rare
and unwary kinship t erm. Yet I felt uneasily t hat I was always on
t he outside looking in, even when I became comfortable in my use
of the vernacular. F o r i was onst andy_aware of t he t huddi ng of
nt uaLj i r ums- UL- t he vicinity of my camp, and t he people I knew
would oft enj ake their l eaygofme to spend days at a time at t endi ng
such exotically named r i t e s a s Nkula, Wubwane'u, and Wubinda.
Eventually, I was forced to recognize t hat if I want ed to know
what even a segment of Ndembu culture was really about , I would
have to overcome my preiudice_agajnst ritual and start to investi-
gat ej l ^ -
It is t rue t hat almost from t he beginning of my stay among the
Ndembu I had, on invitation, at t ended t he frequent performances of
the girls' pubert y rites (Nkang'a) and had tried to describe what I
had s e e n a s accurately j as j os s i bl e. But it is one thing to observe
people_performing t he stylized gestures an?s i ngi ng t he cn^i Fs ongT
of ritualjperformances and quite anoAeijQxeaclLan adequat e under-
standing of what the movements and words mean to them. To obtain
enught enment , I had recourse at first to t he District Notebook, a com-
pilation of r andom jottings by officers of the Colonial Administration
on events and customs t hat struck t hem as interesting. Here I found
short accounts of Ndembu beliefs in a Hi gh God, in ancestral spirits,
and of different kinds of rites. Some were accounts of observed cere-
monies, but most of t hem were based on the reports of Ndembu local
government employees, such as messengers and clerks. At all events,
they hardl y rjrp3dd_ejd_satisfactory explanations of t he long, compli-
cated pubert y rites I had seen, t hough they gave me some prelimi-
nary information about the kinds of rites I had not seen.
My next move was to set up a series of interviews with an excep-
tionally capabl e chief, entitled I kdenge^who had a soundknowl edge
8 The Ritual Process
of English. Chief Ikelenge at once grasped what I want ed and gave
me an inventory of t he names of t he principal Nde mbu rituals, wi t h
brief accounts of t he mai n features of each. I soon discovered t hat
t he Ndembu were not at all resentful of a stranger' s interest i n their
ritual system and were perfectly prepared to admi t to its perform-
ances anyone who treated their beliefs wi t h respect. I t was not long
before Chief Ikelenge invited me to at t end a performance of a ri t ual
belonging to t he gun-hunt ers' cult, Wuyang'a. It was at this perform-
ance t hat I became aware t hat at least one set of economic activities,
namely hunt i ng, could hardl y be understood wi t hout a grasp of the
ritual idiom pert ai ni ng to t he chase. Th e accumul at i on of symbols in-
dicative at once of hunt i ng power and virility gave me an i nsi ght as
well into several features of Ndembu social organization, not abl y the
stress on the i mport ance of cont emporaneous links bet ween mal e kin
in a matrilineal society whose structural continuity was t hr ough
women. I do not want to dwell upon this probl em of t he ri t ual i zat i on
of sex roles at t he moment , but merely wish to stress how cert ai n
regularities t hat emerged from the analysis of numeri cal dat a, such
as village genealogies and censuses and records of succession to
office and i nheri t ance of propert y, became fully intelligible onl y in
t he light of values embodi ed and expressed in symbols at ri t ual
performances.
Ther e were limits, however, to t he assistance Chief Ikel enge was
able to offer me. I n t he first place, his position and its mani fol d roles
prevented hi m from leaving his capital village for long, a n d his
relations wi t h t he local mission, whi ch were of political i mpor t ance
to hi m, were too delicate, in a situation where gossip carries news fast,
to permi t hi m t he l uxury of at t endi ng many pagan ceremoni es.
Moreover, my own research was rapidly becoming a microsociologi-
cal investigation of the ongoing process of village life. I move d my
camp from t he chief' s capital to a cluster of commoner villagers.
Ther e, in t i me, my family came to be accepted as more or less a par t
of t he local communi t y, and, wi t h eyes j ust opened to t he i mpor t ance
of ritual in t he lives of t he Ndembu, my wife and I began to percei ve
many aspects of Ndembu culture t hat had previously been invisible
Planes of Classification
9
to us because of our theoretical blinkers. As Nadel has said, facts
change wi t h theories and new facts make new theories.
It was about this time t hat I read some remarks in the second
Rhodes-Livingstone Paper to be published, The Study of African
Society, by Godfrey and Moni ca Wilson (1939), to the effect that in
many African societies where ritual is still a going concern, there
are a number of religious specialists who are prepared to offer inter-
pretations of it. Later, Moni ca Wilson (1957) was to write that " a n y
analysis not based on some translation of the symbols used by people
of that culture is open to suspicion" (p. 6). I then began to seek out
Ndembu ritual specialists to record interpretative texts from t hem
about rites I had observed. Our entree to performances, and access
to exegesis, was no doubt helped by the fact that, like most ant hro-
pological field workers, we distributed medicines, bandaged wounds,
and, in the case of my wife (who is a doctor' s daught er and bolder
in these matters t han I ) , injected with serum persons bitten by
snakes. Since many of the Ndembu cult rituals are performed for
the sick, and since European medicines are regarded as having
mystical efficacy of the same kind as their own t hough greater in
potency, the curative specialists came to regard us as colleagues and
to welcome our at t endance at their performances.
I remembered having read in Dr. Livingstone' s Missionary Travels
how he had made a strict point of consulting the local medicine
men about t he condition of patients, and how this had made for
good rapport with an influential section of the Central African popu-
lation. We copied his example, and this may have been one reason
why we were allowed to at t end the esoteric phases of several rites and
obtain what cross-checking suggested were reasonably reliable inter-
pretations of many of the symbols employed in t hem. By " r e l i abl e "
I mean, of course, that the interpretations were, on the whole,
mutually consistent. They might, in fact, be said to constitute t he
standardized hermeneutics of Ndembu culture, rat her t han the free
associations or eccentric views of individuals. We also collected
interpretations from Ndembu who were not ritual specialists, or at
least not specialists in t he ritual immediately under consideration.
10
The Ritual Process
Most Ndembu, bot h men and women, were members of at least one
cult association, and it was hard to find an elderly person who was
not an " e xpe r t " in the secret knowledge of mor e t han one cult. I n
this way we gradually built up a body of observational dat a and in-
terpretative comments, which, when submitted to analysis, began to
exhibit certain regularities from which it was possible to elicit a
structure, expressed in a set of pat t erns. Lat er we shall consider some
of the characteristics of these pat t erns.
In all this time, we never asked for a ritual to be performed solely
for our anthropological benefit; we held no brief for such artificial
^play-acting. Ther e was, in fact, no dear t h of spontaneous perform-
ances. One of our major difficulties was frequently in deciding on
a given day whi ch of two or more performances to at t end. As we
became increasingly a par t of the village scene, we discovered t hat
very often decisions to perform ritual were connected wi t h crises in
the social life of villages. I have written elsewhere at some length
on the social dynamics of ritual performances and do not i nt end to
give t hem more t han passing ment i on in these lectures. Here I merely
indicate t hat among t he Ndembu there is a close connection between
social conflict and ritual at t he levels of village and " vicinage (a
t erm I use for discrete clusters of villages), and t hat a multiplicity
of conflict situations is correlated with a hi gh' frequency of ritual
performance.
ISOMA
My mai n ai m in this chapt er is to explore t he semantics of ritual
symbols in Isoma, a ritual of the Ndembu, and to construct from the
observational and exegetical dat a a model of t he semantic structure
of this symbolism. Th e first step in such a task is to pay close attention
to t he way t he Ndembu explain their own symbols. My procedure
will be to begin with particulars and move to generalization, letting
the reader into my confidence at every step al ong this road. I am
now going to look closely at a kind of ritual whi ch I observed on
Planes of Classification
II
three occasions and for which I have a considerable quant i t y of
exegetical material. I must crave the reader' s indulgence for the fact
t hat I shall have to mention a number of Ndembu vernacular terms,
for an i mport ant par t of the Ndembu explanation of symbols rests
upon folk etymologizing. The meani ng of a given symbol is often,
though by no means invariably, derived by Ndembu from the name
assigned to it, t he sense of which is traced from some pri mary word,
or etymon, often a verb. Scholars have shown t hat in other Bantu
societies this is often a process of fictitious etymologizing, dependent
on similarity of sound rat her than upon derivation from a common
source. Nevertheless, for the people themselves it constitutes part
of the "expl anat i on" of a ritual symbol; and we are here trying to
discover " the Ndembu inside view, " how the Ndembu themselves
felt and t hought about their own ritual.
Reasons for Performing Isoma
The Isoma (or Tubwiza) ritual belongs to a class (muckidi) of rituals,
recognized as such by Ndembu, known as "women' s r i t ual s " or
"ri t ual s of procreat i on, " which itself is a subclass of "ri t ual s of t he
ancestral spirits o r ' shades' "a t erm I borrow from Moni ca Wilson.
The Ndembu word for " r i t ua l " is chidika, which also means " a
special engagement " or an "obl i gat i on. " This is connected with the
idea t hat one is under an obligation to venerate t he ancestral shades,
for, as Ndembu say, " a r e they not the ones who have begotten
or borne y o u ? " The rituals I am speaking of are in fact performed
because persons or corporate groups have failed to meet this obliga-
tion. Either for his own default or as representative of a group of kin,
a person is believed to have been "caught , " as Ndembu say, by a
shade and afflicted with a misfortune thought to be appropri at e to
his sex or social role. The misfortune appropri at e to women consists
in some kind of interference with the victim' s reproductive capacity.
Ideally, a woman who is living at peace with her fellows and is
mindful of her deceased kin should be marri ed and a mot her of " live
12
The Ritual Process
and lovely chi l dr en" (to t ransl at e an Nde mbu expression). But a
woman who is either quarrel some hersel f or a member of a group
riven wi t h quarrels, and who has simultaneously "forgotten her
[deceased mot her or mot her' s mot her or some other senior deceased
matrilineal kinswoman' s] shade in her liver [or, as we would say,
' h e a r t ' ] , " is in peril of havi ng her procreat i ve power (lusemu)
"t i ed u p " (ku-kasila) by t he offended s hade.
The Ndembu, who practice mat ri l i neal descent combined with viri-
local marri age, live in small, mobi l e villages. The effect of this ar-
rangement i s t hat women, t hrough wh om chi l dren derive their pri mary
lineage and residential affiliation, spend much of their reproductive
cycle in t he villages of their husbands a n d not of their matrilineal kin.
Ther e is no rule, as t here is, for exampl e, among the matrilineal
Tr obr i and Islanders, t hat t he sons of wome n living in this form of
marri age should go to reside in t he villages of their mothers' brothers
and ot her mat ri ki n on reachi ng adol escence. One consequence of
this is t hat every fruitful marri age becomes an arena of covert strug-
gle between a woman' s husband and her brothers and mother' s
brothers over the residential affiliation of her children. Since there is
also a close bond between a woman a n d her children, this usually
means t hat after a short or long per i od a woman will follow her
children to her village of mat ri l i neal affiliation. My figures on
Ndembu divorce indicate t hat t he t r i bal ratios are the highest among
all the matrilineal societies in Cent r al Africa for which reliable
quant i t at i ve dat a existand all have high divorce rates. Since
women ret urn to their mat ri ki n on di vorceand a fortiori to their
children resident among those ki ni n a very real sense village con-
tinuity, t hrough women, depends upon mari t al discontinuity. But,
while a woman is residing wi t h her hus band wi t h her young children,
and thus fulfilling the valid nor m t hat a woman should please him,
she is not fulfilling an equally valid n o r m t hat she should contribute
children to t he cont emporaneous member s hi p of her matrilineal
village.
Interestingly, it is the shades of di r ect matrilineal kinswomen
own mothers or own mot hers' mot her s t hat are held to afflict
women wi t h reproductive disorders, resul t i ng in t emporary barren-
Planes of Classification
13
ness. Most of these victims are residing wi t h their husbands when
divination decrees t hat they have been caught with infertility by
their matrilineal shades. They have been caught, so Ndembu reg-
ularly say, because they have "for got t en" those shades who are not
only their direct ascendants but also t he i mmedi at e progenetrices of
their mat ri ki nwho form the core membership of villages different
from those of their husbands. The curative rites, including Isoma,
have as one social function t hat of "causi ng t hem to r e me mbe r "
these shades, who are structural nodes of a locally residing matriline-
age. The condition of barrenness these shades bri ng about is consid-
ered to be a t emporary one, to be removed by performance of
the appropri at e rites. Once a woman remembers the afflicting
shade, and thus her pri mary allegiance to matrikin, t he interdiction
on her fertility will cease; she can go on living with her husband
but wi t h a sharpened awareness of where her and her children' s
ultimate loyalties lie. Th e crisis brought on by this contradiction
between norms is resolved by rituals rich in symbolism and preg-
nant wi t h meani ng.
Processual Form
Isoma shares wi t h t he other women' s cults a common diachronic
profile or processual form. I n each a woman suffers from gynecolo-
gical disorders; then either her husband or a mat ri ki nsman seeks
out a diviner, who denominates t he precise mode of affliction in
which t he shade, as Ndembu say, has "come out of t he grave to
catch her . " Dependent upon t hat mode, the husband or kinsman
employs a doctor (chimbuki) who "knows the medi ci nes" and t he
correct ritual procedures for appeasing t he afflicting shade to act as
master of ceremonies for t he coming performance. Thi s doctor t hen
summons other doctors to help hi m. These are either women who
have undergone exposure to t he same kind of ritual and have thus
gained entry into t he curative cult, or men closely linked by matrilin-
eal kinship or affinity to a previous patient. The patients (ayeji) may
be regarded as " candidates " for membership of t he cult, t he doctors
The Ritual Process
as its " a de pt s . " Th e afflicting shades (akishi) ar e believed to have
been former adept s. Cul t member s hi p t hus transects village and
lineage member shi p and bri ngs i nt o t empor ar y operation what may
be termed " a communi t y of sufferi ng"or, rat her, of "former
sufferers " from t he same t ype of affliction as n ow besets the candi dat e
pat i ent . Member s hi p of a cult such as Isoma cut s across even tribal
boundari es, for member s of t he cul t ural l y a n d linguistically related
Luvale, Chokwe, and Luchazi tribes ar e ent i t l ed to at t end Ndembu
Isoma rites as adept s, and as such t o perform r i t ual tasks. Th e " senior "
(mukulumpi) or " g r e a t " (weneni) adept is usual l y a man, even for such
women' s cults as Isoma; as in most mat ri l i neal societies, while social
pl acement is t hr ough women, aut hor i t y is in t he hands of men.
Women' s cults have t he t ri part i t e di achr oni c structure made
familiar to us by t he work ofvan Gennep. Th e first phase, called Ilembi,
separates t he candi dat e from t he profane wor l d; t he second, called
Kunkunka (literally, " i n t he grass h u t " ) , part i al l y secludes her from
secular life; while t he t hi rd, called Ku-tumbuka, is a festive dance,
cel ebrat i ng t he removal of t he shade' s i nt erdi ct i on and t he candidate' s
r et ur n to nor mal life. I n Isoma this is signalized by t he candidate' s
beari ng a child and raising it to t he t oddl i ng st age.
Indigenous Exegesis of Symbols
So much for t he br oad social and cul t ural settings of Isoma. If we now
desire to penet rat e t he i nner st ruct ure of i deas contained in this
ri t ual , we have to underst and how t he Nde mbu themselves interpret
its symbols. My met hod is perforce t he reverse of t hat of those num-
erous scholars who begin by eliciting t he cosmology, whi ch is often
expressed in t erms of mythological cycles, and then explain specific
rituals as exemplifying or expressing t he "s t r uct ur al model s " they
find in t he myt hs. But t he Nde mbu have a pauci t y of myths and
cosmological or cosmogonic nar r at i ves. l t is therefore necessary to
begin at t he ot her end, wi t h t he basic building-blocks, t he "mol e-
cules, " of ri t ual . These I shall call "s ymbol s , " and for t he moment I
shall eschew involvement in t he l ong debat e on t he difference be-
Planes of Classification
15
tween such concepts as symbol, sign, and signal. Since the preliminary
approach is from the " i ns i de " perspective, let us rat her first inquire
into the Ndembu usage.
I n an Ndembu ritual context, almost every article used, every ges-
ture employed, every song or prayer, every unit of space and time,
by convention stands for something other t han itself. It is more t han
it seems, and often a good deal more. The Ndembu are aware of the
expressive or symbolic function of ritual elements. A ritual element
or unit is called chijikijilu. Literally, this word signifies a " l andmar k"
or "bl aze. " Its etymon is ku-jikijila, " to blaze a t r ai l "by slashing
a mark on a tree with an ax or breaking one of its branches. This t erm
is drawn originally from t he technical vocabulary of hunt i ng, a voca-
tion heavily invested with ritual beliefs and practices. Chijikijilu also
means a "beacon, " a conspicuous feature of the landscape, such as
an ant hill, which distinguishes one man' s gardens or one chief' s
realm from another' s. Thus, it has two mai n significations: (1) as a
hunter's blaze it represents an element of connection between known
and unknown territory, for it is by a chain of such elements t hat a
hunt er finds his way back from t he unfamiliar bush to the familiar
village; (2) as bot h blaze and beacon it conveys the notion of t he struc-
tured and ordered as against the unstructured and chaotic. Its ritual
use is already met aphori cal : it connects t he known world of sensorily
perceptible phenomena wi t h the unknown and invisible realm of the
shades. It makes intelligible what is mysterious, and also dangerous.
A chijikijilu has, further, bot h a known and an unknown component.
Up to a point it can be explained, and there are principles of explana-
tion available to Ndembu. It has a name (ijina) and it has an appear-
ance (chimwekeshu), and bot h of these are utilized as the starting points
of exegesis (chakulumbwishu),
The Name "Isoma"
At t he very outset, t he name Isoma itself has symbolic value. My
informants derive it from ku-somoka, " to slip out of place or fastening. "
This designation has multiple reference. In the first place, it refers to
1 6 The Ritual Process
The Mask "Mvweng'i"
Th e shade t hat has emerged in Isoma manifests itself in other ways,
too. I t is t hought to appear in t he pat i ent ' s dreams dressed like one
of t he masked beings i n t he boys' circumcision rites (Mukanda). These
masked beings, known as makishi (singular ikishi), are believed by
women t o be shades of anci ent ancestors. Th e one known as Mvweng'i
wears a bark kilt (nkambi), like t he novices dur i ng their seclusion after
circumcision, and a costume consisting of many strings made. from
bark cloth. He carries a hunt i ng bell (mpwambu) used by hunters to
keep in t ouch wi t h one anot her i n t he deep bush or to summon their
dogs. He is known as " gr a ndf a t he r " (nkaka), appears after the boys'
circumcision wounds are healed, and is greatly feared by women. If
a woman touches Mvweng'i, it is t hought t hat she will have miscar-
riages. A song traditionally sung when this ikishi first appears near
t he specific condi t i on t he rites ar e intended to dispel. A woman who
is " caught in Isoma " is very frequently a woman who has had a series
of miscarriages or abort i ons. Th e unborn child is thought to "sl i p
o u t " before its t i me has come t o be born. In the second place, ku-
somoka means " to leave one' s gr oup, " perhaps also with the implica-
tion of pr emat ur i t y. Thi s t heme seems to be related to the notion of
"for get t i ng" one' s mat ri l i neal at t achment s. In discussing the mean-
ing of t he word Isoma, several informants mentioned the t erm lufwisha
as indicative of t he pat i ent ' s condi t i on. Lufwisha is an abstract noun
derived from ku-fwisha, itself deri ved from ku-fwa, " to di e. " Ku-fwisha
has bot h a generic sense and a specific one. Generically, it means " to
lose relatives by de at h, " specifically " to lose chi l dren. " The noun
lufwisha means bot h " t o give bi r t h to a dead chi l d" and the "con-
st ant dyi ng of chi l dr en. " On e i nformant told me : " If seven children
die one after t he ot her, it is lufwisha." Isoma is thus a manifestation of
a shade t hat causes a woman to bear a dead child or brings deat h on a
series of infants.
Planes of Classification
17
the lodge where the novices are secluded in the bush runs as follows:
Kako nkaka eyo nkaka eyo eyo nkakayetu nenzi, eyo eyo, nkakayelu, mwanta;
" Grandfather, O Grandfather, our grandfather has come, our grandfather,
the chief;"
mbwemboye mbwemboyeyawume-e
"the glans penis, the glans is dry,
mwang'u walulemba mbwemboyeyawumi.
a scattering oitulemba spirits, the glans is dry."
The song represents for Ndembu a concentration of masculine
power, for nkaka also signifies " a n owner of slaves," and a "chi ef"
owns many slaves. The dryness of the glans is a symbol of t he attain-
ment of an auspicious masculine adul t status, one of t he aims of the
Mukanda circumcision rites, for the glans of an uncircumcised boy is
regarded as wet and filthy, hence inauspicious, beneat h the prepuce.
Tulemba spirits, propitiated and exorcised in anot her type of ritual,
cause infants to sicken and pine. Mvweng'i drives t hem from the boys.
The strings of his costume are believed to " tie up " (ku-kasila) female
fertility. I n brief, he is a symbol of mat ur e masculinity in its pur e
expressionand his hunt i ng attributes further support thisand as
such is dangerous to women in their most feminine role, t hat of mot her.
Now, it is in t he guise of Mvweng'i t hat t he shade appears to the vic-
tim. But here there is some ambiguity of exegesis. Some informants
say t hat the shade is identified wi t h Mvweng'i, others t hat shade
(mukishi) and masker (ikishi) operate in conjunction. The latter say
t hat t he shade rouses Mvweng'i and enlists his aid in afflicting the
victim.
It is interesting to note t hat the shade is always the spirit of a de-
ceased female relative, while Mvweng'i is almost maleness personified.
This motif of linking reproductive disorder to the identification of a
female wi t h a type of masculinity is found elsewhere in Ndembu
ritual. I have mentioned it in connection wi t h rites to cure menstrual
difficulties in The Forest of Symbols (1 967): " Wh y then is t he woman
patient identified with mal e bloodspillers ? The [social] field context
of these symbolic objects and items of behavior suggests t hat t he
Ndembu feel t hat t he woman, in wasting her menst rual blood and in
1 8 The Ritual Process
failing to be ar children, is actively renouncing her expected role as a
mat ur e mar r i ed female. She is behaving like a male killer [i.e. a
hunt er or homi ci de], not like a female nouri sher" (p. 42. For a
fuller analysis of the Nkula curative rites, see Turner, 1968, pp. 5 4-87).
Th e si t uat i on in Isoma is not dissimilar. It should be noticed t hat in
these cults, t he victim is in various episodes and symbolisms often
identified wi t h the shade that afflicts her : she is being persecuted,
one mi ght say wi t h fair legitimacy, by a part or aspect of herself, pro-
j ect ed ont o t he shade. Thus a cured victim in Isoma will" become, in
Nde mbu t hought , herself an afflicting shade after death, and as such
will be identified with or closely conjoined to the masculine power
Mvweng'i.
But it woul d, I think, be erroneous to see in the Isoma beliefs merely
an expression of the "mascul i ne protest. " This unconscious attitude
may well be more prominent in the Nkula rites t han in Isoma. The
st ruct ural tension between matrilineal descent and virilocal marriage
seems to domi nat e the ritual idiom of Isoma. It is because the woman
has come too closely in touch with the " man' s s i de" in her marriage
t hat her dead mat ri ki n have impaired her fertility. The right relation
t hat should exist between descent and affinity has been upset; the
mar r i age has come to outweigh the matrilineage. The woman has
been scorched by the dangerous fires of male sacredness. I use this
met aphor because Ndembu themselves do: if women see the flames
of t he boys' seclusion lodge when it is burned down after the circum-
cision ri t ual , it is believed that they will be striped as with flames, or,
like t he zebra (ng'ala), with leprosy, or, alternatively, will run mad
or become simpletons.
Aims of Isoma
Thus the implicit aims of Isoma include: restoration of the right rela-
tion bet ween matriliny and marri age; reconstruction of the conjugal
relations bet ween wife and husband; and maki ng the woman, and
hence t he mar r i age and lineage, fruitful. The explicit aim of the rites,
Planes of Classification
19
as Ndembu explain it, is to remove the effects of what they call a
chisaku. Broadly, chisaku denotes "misfortune or illness due to t he
displeasure of ancestral shades or a breach of t aboo. " Mor e specific-
ally, it also denotes a curse spoken by a living person to arouse a shade
and may include medicines concocted to har m an enemy. In t he
case of Isoma, the chisaku is of a particular kind. It is believed t hat a
matrilineal relative of t he victim has gone to the source (kasulu) of a
stream in the vicinity of the village of her mat ri ki n and there spoken
a curse (kumushing'ana) against her. The effect of this curse has been
" t o awake n" (ku-tonisha) a shade who was once a member of the
Isoma cult. As one informant said (and I translate literally): "At Isoma
they behead a red cock. Thi s stands for t he chisaku or misfortune
through which people die, it must go away (chisaku chafwang'a antu,
chifumi). The chisaku is deat h, which must not happen to t he woman
patient; it is sickness (musong'u), which must not come to her; it is
suffering (ku-kabakana), and this suffering is from the grudge (chitela)
of a witch (muloji). A person who curses anot her wi t h deat h has a
chisaku. The chisaku is spoken at the source of a river. If someone
passes there and steps on it (ku-dyata) or crosses over it (ku-badyika),
bad luck (malwa) or lack of success (ku-halwa) will go with her wherever
she goes. She has gotten it at t hat place, the stream source, and she
must be treated (ku-uka) there. The shade of Isoma has come out as
the result of t hat curse, and comes like Mvweng'i."
As the reader can see, t here is in all this a strong overtone of witch-
craft. Unl i ke other women' s rites, Isoma is not performed merely to
propitiate a single shade; it is also aimed at exorcising malign mystical
influences emanat i ng from the living as well as t he dead. Ther e is
here a grisly alliance of witch, shade, and t he Ikishi Mvweng'i to be
dealt with. The rites involve symbolic reference to all these agencies.
It is significant that a matrilineal relative should be regarded as the
precipitating cause of the affliction, t he arouser of these two grades
of ancestral beings, remote and near, Mvweng'i and t he female shade.
It is also significant t hat t he rites are performed, whenever possible,
near a village inhabited by t he victim' s matrilineal kin. Furt hermore,
she is partially secluded at t hat village for a considerable time afterward,
20
The Ritual Process
and her hus band must reside wi t h her uxorilocally duri ng that
peri od. Ther e seems to be some ambi gui t y in my informants' accounts
about t he i nt erpret at i on of t he precipitating curse. I t is felt to smack
of witchcraft and hence to be " b a d , " but, at the same time, to be
partially justified by t he victim' s neglect of her matrilineal ties past
and present. Th e rites ar e part i al l y to effect a reconciliation between
t he visible and invisible part i es concerned, t hough they contain
episodes of exorcism as well.
PREPARATI ON OF SACRED SITE
So much for t he social settings and t he major beliefs underlying Isoma.
Now let us t ur n to t he rites themselves, and consider the interpreta-
tions of symbols in or der of their occurrence. These will expand our
pi ct ure of t he belief st ruct ure, for Ndembu, who, as I said, have re-
mar kabl y few myt hs, compensat e for this by a wealth of item-by-item
exegesis. Ther e ar e no short cuts, t hrough myt h and cosmology, to
t he structure:in Levi-Strauss' s senseof Ndembu religion. One has
t o proceed atomistically and piecemeal from " bl a z e " to " bl aze , "
" b e a c o n " to " be a c on , " if one is properly to follow the indigenous
mode of t hi nki ng. I t is only when t he symbolic pat h from t he unknown
to t he known is compl et ed t hat we can look back and comprehend its
final form.
As wi t h all Nde mbu rites, t he pat t er n of procedure in each specific
case is set by t he di vi ner originally consulted about the patient' s
affliction. He is t he one who establishes t hat the woman has lost a
succession of chi l dren by mi scarri age or deat h in infancymisfortunes
summari zed i n t he t er m lufwisha. I t is he who decrees t hat t he rites
must begin at t he hol e or bur r ow, either of a giant rat (chituba) or of
an ant -bear (mfuji). Wh y does he make this rat her odd prescription?
Ndembu explain it as follows: Bot h these animals stop up their
burrows after excavat i ng t hem. Each is a symbol (chijikijilu) for t he
Isoma shade-manifestation whi ch has hi dden away the woman' s
fertility (lusemu). Th e doct or adept s must open t he blocked ent rance
Planes of Classification
2 1
of the burrow, and thus symbolically give her back her fertility, and
also enable her to raise her children well. The diviner decides whi ch
of these species has hidden the fertility in the part i cul ar case. The
burrow must be near the source of the stream where the curse was
uttered. The ut t erance of a curse is usually accompanied by the burial
of "medi ci nes, " often pressed (ku-panda) into a small antelope' s horn.
From my knowledge of ot her Ndembu rites, I strongly suspect t hat
these are hi dden near the river source. The animal' s burrow provides
the reference point of orientation for the spatial structure of t he sacred
site. The rites I am discussing here are " t h e rites of separation, "
known as ku-lembeka or ilembi, a t erm Ndembu connect materially wi t h
ways of using medicines or medicine containers promi nent in some
kinds of women' s cults, and etymologically with ku-lemba, " to suppli-
cate, beg forgiveness, or be peni t ent . " The notion of propitiation is
prominent in them, for the doctors ar e partly pleading on the patient' s
behalf with t he shades and other pret erhuman entities to give her
back her mot herhood.
In all ilembi rites one of the first steps is for the doctor adepts, led
by the senior adept or " master of ceremonies, " to go into the bush to
collect t he medicines they will treat t he patient wi t h later. Thi s
episode is known as ku-lang'ula or ku-hukulayitumbu. In Isoma, before
this step is taken t he patient' s husband, if she has one currently,
constructs for her use duri ng the subsequent seclusion period a small
round grass hut , j ust outside t he ring of a dozen or so huts t hat con-
stitutes an Ndembu village. Such a hut (nkunkd) is also made for girls
undergoing seclusion after their pubert y rites, and t he Isoma hut is
explicitly compared wi t h this. The patient is like a novice. Just as a
puberty novice is " g r o wn " into a woman, according to Ndembu
thinking, so the Isoma candi dat e is to be regrown into a fertile woman.
What has been undone by the curse has to be done all over again,
although not in precisely t he same way, for life crises are irreversible.
There is analogy but not replication.
A red cock, supplied by the husband, and a white pullet, supplied
by the patient' s matrikin, are then collected by the adepts, who pro-
ceed to t he particular st ream source where divination previously
22
The Ritual Process
i ndi cat ed t hat t he curse was laid. They then examine the ground care-
fully for signs of a gi ant rat ' s or ant -bear' s burrow. When they find it,
t he senior adept addresses t he ani mal as follows: " Gi ant rat (ant-
bear ) , if you ar e t he one who kills children, now give the woman back
her fertility, ma y she raise children wel l . " Here the animal seems to
represent t he whol e " t r oi ka " of afflicting agencieswitch, shade,
and ikishi. Th e next task is to tie hanks of grass into two knots, one
above t he filled-in ent rance to t he burrow, the other about four feet
away above t he t unnel made by the ani mal . Th e clods beneat h these
ar e removed by hoe, and t he senior adept and his major mal e assistant
begin to dig de e p holes t here, known as makela (singular, ikela), a
t er m reserved for holes serving a magico-religious purpose. Next,
t wo fires are ki ndl ed at a distance of about ten feet from t he holes and
nearer t he second t han the first. One fire is said to be " on t he right-
hand si de, " (i. e. , looking from t he ani mal ' s burrow to the new hole)
and is reserved for t he use of t he mal e adept s; t he other, " on t he left-
hand si de, " is for t he women. Th e senior adept then puts down a piece
of broken cal abash near t he first burrow-ent rance hole, and female
adept s, led by t he pat i ent ' s mot her if she is an adept , put in it some
portions of edi bl e roots from their gardens, including cassava rhiz-
omes and sweet pot at o t ubers. I n ritual idiom these represent " t h e
b o d y " (mujimba) of t he pat i ent . I t is significant t hat they are supplied
by women, not abl y by women of t he pat i ent ' s matrilineage.
After t he senior adept and his principal male assistant have inaug-
urat ed t he di ggi ng, t hey hand over their hoes to other mal e adepts,
who cont i nue t o excavat e t he holes unt i l they are about four to six
feet deep. Th e bur r ow ent rance is known as " t he hole of t he giant
r a t " (or " a n t - b e a r " ) , t he ot her as " t h e new hol e. " The ani mal is
known as t he " wi t ch " (muloji), and t he burrow ent rance is said to be
" h o t " (-tata). Th e ot her hole is called ku-fomwisha or ku-fomona,
verbal nouns t hat signify respectively " cooling down " and " domes-
t i cat i ng. " Wh e n t hey have reached t he appropri at e dept h, t he
adept s commence t o dig t oward one anot her until they meet about
halfway, havi ng compl et ed a t unnel (ikela dakuhanuka). This has to
be wi de enough for one person to pass t hrough. Ot her adepts break
Planes of Classification
23
or bend the branches of trees in a wide ring around the whole scene
of ritual activity, to create a sacred space t hat rapidly achieves
structure. To ring something around is a persistent t heme of Ndembu
ri t ual ; it is usually accompanied by the process of maki ng a clearing
(mukombela) by hoe. In this way a small realm of order is created in
the formless milieu of t he bush. The ring is known as chipang'u, a
t erm t hat is also used for t he fence around a chief' s dwelling and his
medicine hut .
COLLECTION OF MEDICINES
While t he j uni or adepts prepare the sacred site, the senior adept and
his principal assistant go to the adjacent bush to find medicines.
These are collected from different species of trees, each of which has
a symbolic value derived from the attributes and purposes of Isoma.
In most Ndembu rituals there is considerable consistency in the sets
of medicines used in different performances of the same kind of
ritual, but in t he Isoma rites I at t ended there was wide variation
from performance to performance. The first tree from whi ch portions
are taken for medicine (yitumbu) is always known as t he ishikenu, and
it is here t hat invocation is made, either to the afflicting shade or to
the species itself, whose power (ng'ovu) is said to be " awake ne d"
(ku-tona) by the words addressed to it. At one performance I at-
tended, the senior adept went to a kapwipu tree (Swarlzia madagas-
cariensis), which is used because its wood is hard. Hardness repre-
sents the health and strength (wukolu) desired for the patient. The
senior adept cleared the base of the tree of weeds with his ritual hoe,
then put the pieces of edible tubers representing the patient' s body
on the cleared space (mukombela) and spoke as follows: " Wh e n this
woman was pregnant before, her lips, eyes, palms and t he soles of
her feet t urned yellow [a sign of anemi a] . Now she is pregnant again.
This time make her strong, so t hat she may bear a living child, and
may it grow st rong. " The doctor then cut bark chips from anot her
tree of the same species wi t h his medicine ax, and put t hem in his
24
The Ritual Process
piece of br oken cal abash. After t hat he proceeded to cut bark chips
from sixteen furt her species of t rees.
1
I t woul d t ake too long to discuss t he meani ng of each of these here,
suffice it to say t hat many Ndembu can at t ach not merely a single
significance but in some cases (such as musoli, museng'u, and mukom-
bukombu) ma n y connot at i ons to a single species. Some of these species
are used i n ma n y different kinds of rituals and in herbalist practice,
(where, however, different types of associational linkages are utilized
from those empl oyed in ritual, dependi ng more on taste and smell
t han on nat ur al propert i es and etymology). Some (e.g., kapwipu,
mubang'a) ar e used because they have tough (hence " st rengt heni ng")
wood, ot hers (e. g. , mucha, musafwa, mufung'u, museng'u, musoli, and
mubulu) because t hey ar e fruit-bearing trees, representing the ritual
i nt ent i on of maki ng t he pat i ent fruitful once mor e; but all share t he
ritually i mpor t ant propert y t hat bark string cannot be taken from
t hem, for this woul d " t i e u p " t he fertility of the pat i ent . In this
sense, t hey ma y all be said to be coxmteT-Mvweng'i medicines, for,
as t he r eader will recall, his costume is largely made up of bark
strings, deadl y to women' s procreation.
I cannot refrain, however, from mentioning in more detail a
smaller set of Isoma medicines, from anot her performance, for t he
nat i ve i nt er pr et at i on of these throws light on many of t he ritual' s
underl yi ng i deas. He r e t he doctors went first to a chikang'anjamba or
chikoli tree {Strychnos spinosa). Thi s they described as t he mukulumpi,
" s e n i or " or " e l de r , " of t he medicine. After invoking its powers,
they took a por t i on of one of its roots and some leaves. Chikang'-
anjamba means " t he el ephant fails" (to uproot it), on account of its
tenacity and t oughness. Its alternative name, chikoli, they derived
from ku-kola, " t o be strong, heal t hy, or firm," a designation t hat
accords wi t h its ext r eme toughness and durability. Thi s same tree
1
Mubang'a (Afrormosia angolensis), mulumbulumbu, mucha {Parinari mobola), musesi
wehata [Erythrophloeum qfricanum), musesi wezenzela (Burkea qfricana),musafwa,mufung'u
(Anissophyllea fruticulosa or boehmii), katawubwang'u, musoli (Vangueriopsis lanciflora),
kayiza {Strychnos stuhlmannii), wunjimbi, museng'u (Ochna pulchra), wupembi, muleng'u
{Uapaca species), mukombukombu (Tricalysia angolensis), and mubulu.
Planes of Classification
25
provides medicine for the circumcision rites, where it is thought to
confer on t he novices exceptional virility. In Isoma, its use stresses
the connection between these rites and Mukanda, the circumcision
rites, while it is also a specific against the infirmityand in many
cases the anemiaof the patient. A comparison of the domi nant
medicines of these two performances shows that the same principle
or idea can be expressed in different symbols. The domi nant medicine
of the first performance, kapwipu, is also a strong tree, and one from
which is often taken the forked branch that forms the central ele-
ment of shrines set up to the shades of hunters, considered to be
" tough and virile men. " Such shrine trees, peeled of bark, are ex-
ceptionally resistant to t he action of termites and other insects.
Decoctions of kapwipu leaves and bark are also used as aphrodisiacs.
The second medicine collected in this performance represents
anot her t heme of Ndembu ri t ual t hat of representing the patient' s
inauspicious condition. Thi s is the mulendi tree. It has a very slippery
surface, from which climbers are prone to slip (ku-selumuka) and
come to grief. In the same way the patient' s children have tended
to "sl i p o u t " premat urel y. But the "glossiness" (ku-senena) of this
tree also has therapeutic value, and this side of its meani ng is promi-
nent in other rites and treatments, for its use makes t he "di s eas e"
(musong'u) slip away from t he pat i ent .
2
It is, indeed, not uncommon
for Ndembu symbols, at all levels of symbolism, to express simulta-
neously an auspicious and an inauspicious condition. For example,
the name Isoma itself, meani ng " t o slip out , " represents both the
patients' undesirable state and the ritual to cure it.
Here we come across anot her ritual principle, expressed in t he
Ndembu term ku-solola, " to make appear, or reveal. " What is. made
sensorily perceptible, in the form of a symbol (chijikilu), is thereby
made accessible to the purposive action of society, operating through
its religious specialists. I t i s t he " hidden " (chamusweka) t hat i s " danger-
ous " or "noxi ous " (chafwana). Thus, to name an inauspicious con-
dition is halfway to removing that condition; to embody the invisible
2
See also Turner, 1 9 67 , pp. 3 25 - 3 26.
26
The Ritual Process
act i on of wi t ches or shades in a visible or tangible symbol is a big
step t owar d remedyi ng it. This is not so very far removed from t he
pract i ce of t he modern psychoanalyst. When something is grasped
by t he mi nd, made capable of being t hought about , it can be dealt
wi t h, mas t er ed. Interestingly enough, the principle of revelation
itself is embodi ed in an Ndembu medicine-symbol used in Isoma.
Thi s is t he musoli tree (whose name is derived by informants from
ku-solola), from whi ch leaves and bark chips are also taken. It is
wi del y used in Ndembu ritual, and its name is linked wi t h its nat ural
pr oper t i es. I t produces many small fruits, which fall to the ground
and l ur e out of hiding various species of edible animals, which can
be killed by t he hunt er. It literally " makes t hem appear . " In hunt -
i ng cults, its empl oyment as medicine is intended to produce animals
t o t he vi ew {ku-solola anyama) of the hi t hert o unlucky hunt er ; in
women' s cults, it is used " t o make children appe ar " {ku-solola
anyana) t o an unfruitful woman. As in so many cases, there is in t he
semant i cs of this symbol a union of ecology and intellect t hat results
i n t he mat eri al i zat i on of an idea.
T o r et ur n to the medicine-collecting: the doctors next
collect root s and leaves from a chikwata tree {Zizyphus mucronata),
a species in whose therapeutic meani ng etymology once more
combi nes wi t h its nat ural characteristics. Chikwata has "st r ong
t hor ns , " whi ch " cat ch" {ku-kwata) or arrest t he passer-by. It is thus
said bot h to represent " s t r e ngt h" and, by its thorns, to "pi erce
di sease. " I coul d, if time permitted, expatiate upon the ritual theme
of " c a t c h i n g " or "s nat chi ng, " whi ch is expressed in many symbols.
I t per vades t he i di om of hunt i ng symbolism, as mi ght be expected,
but is also exemplified in t he phrase " t o catch a chi l d" {ku-kwata
mwana), whi ch means " to give bi r t h. " But I will pass on to the next
medi ci ne species from which portions are taken, musong'asong'a
{Ximenia caffra), again a hardwood tree, maki ng thus for health and
st r engt h, but also derived by folk etymology from ku-song'a, " t o
come t o fruit or develop fruit," a t erm t hat is metaphorically applied
t o gi vi ng bi r t h t o children, as in ku-song'a anyana. The muhotuhotu tree
(Canthium venosum) is used for medicine "because of its name. "
Planes of Classification
27
Ndembu derive this from ku-hotomoka, " t o fall suddenly, " like a
branch or fruit. The inauspicious condition, it is hoped, will suddenly
cease by its application. Next, medicine is taken from the mutunda
tree, whose derivation is from ku-lunda, which means " to be higher
t han those around i t . " In Isoma it stands for the good growth of an
embryo in the womb and t he child' s continued exuberant growth
thereafter. Mupapala (Anthocleista species) is the name of the next
medicine species, and once more we have a representation of t he
patient' s inauspicious condition. Ndembu derive its name from
kupapang'ila, which means " t o wander about in confusion" without
knowing where one is. One informant put it in this way: " A woman
goes this way and t hat wi t hout children. She must not do this any
more. That is why we cut mupapala medi ci ne. " Behind this idea,
and behind the idea of "sl i ppi ng out , " is t he notion t hat it is good
and appropri at e when things adhere to their proper place and when
people do what is appropri at e for t hem to do in their stage of life
and status in society.
In anot her performance of Isoma, the principal medicine or
"domi nant s ymbol " was not a particular species of tree but any
kind of tree whose roots were thoroughly exposed to view. Such a
tree is called wuvumbu, derived from the verb ku-vumbuka, meani ng
" t o be unear t hed" and " to emerge from hi di ng, " for example, like
a hunted ani mal . Thus, one informant adumbrat ed its meani ng as
follows: " We use wuvumbu tree to bring everything to the surface.
In just the same way everything in Isoma must be cl ear " (-lumbuluka).
Another variant upon the theme of "revel at i on. "
Hot and Cool Medicines: Apertures of Death and Life
Sometimes a portion of wood is taken from a decayed, fallen tree.
This, once more, represents the patient' s musong'u, or diseased,
afflicted condition. Equipped with this array of strengthening, fecun-
datory,revelatory, clarifying, health-giving, affixing medicines, some
28
The Ritual Process
of whi ch in addi t i on represent t he manne r of t he pat i ent ' s affliction,
t he adepts r et ur n t o t he sacred site wher e t r e a t me nt will be given.
They now compl et e t he arrangement s t hat gi ve t hat consecrated
space its visible st r uct ur e. Th e medi ci ne l eaves a n d bark fragments
are pounded by a female adept in a consecr at ed meal -mort ar. Then
they are soaked i n wat er and t he liquid medi ci ne is divided into two
portions. One is put i nt o a large, thick pi ece of ba r k (ifunvu) or into
a potsherd (chizanda), and is t hen heat ed on a fire t hat is kindled
j ust outside t he hol e dug t hr ough t he e nt r ance t o t he giant rat ' s or
ant -bear' s bur r ow. Th e ot her is pour ed col d i nt o an izawu, a t erm
t hat refers to ei t her a clay pot or a medi ci ne t r ough, or into a broken
calabash, and this is placed by t he " n e w h ol e . " (See Figure i ) .
Accordi ng t o one i nformant , t he holes s t and for "gr aves (tulung'a)
and for procreat i ve power (lusemu) " i n ot her wor ds , for t omb and
womb. Th e same i nformant cont i nued: " T h e ikela (hole) of heat
is t he ikela of de at h. Th e cool ikela is life. T h e ikela of t he giant rat
is t he ikela of t he mi sfort une or gr udge (chisaku). T h e new ikela is t he
ikela of maki ng wel l (kuhandisha) or cur i ng. An ikela is located at or
near t he source of a s t r eam; this represent s lusemu, t he ability to
pr oduce offspring. Th e new ikela shoul d flow a wa y from t he pat i ent
(muyeji); in this way t he bad things mus t l eave her . Th e circle of
broken trees is a chipang'u. [This is a mul t i vocal t e r m t hat stands for
( i ) an encl osure; (2) a ri t ual encl osure; (3) a fenced court yard
ar ound a chief' s dwel l i ng and medi ci ne h u t ; (4) a ri ng ar ound t he
moon. ] Th e woma n wi t h lufwisha [i. e. , wh o has lost t hree or four
chi l dren by st i l l bi rt h or infant mor t al i t y] mus t go i nt o t he hole of
life and pass t hr ough t he t unnel t o t he hol e of de at h. Th e bi g doctor
sprinkles her wi t h cold medi ci ne, whi l e his assi st ant sprinkles her
wi t h hot me di ci ne . "
FIGURE i.Isoma: the ritual scene. The couple to be treated sit in the
" hot " hole of a tunnel representing the passage from death to life. A
medicine fire is tended behind them by a doctor. A calabash of cold
medicines stands in front of the " cool " hole, where the entrance to
the tunnel can be seen. Doctors wait here for the patients to emerge.
3
The Ritual Process
/
/
/
Kuhandisha
X
Female
adepts
Women' s fire
Chisaku
\
\
\
River source
O Cool medicine
\
\
\
c
H
IKELA OF LIFE OR
HEALTH
Male
adepts
X
Men's fire
IKELA OF DEATH OR
WITCHCRAFT
^<^Red cock
Hot medicine over fire
/
/
Animal' s
burrow
(blocked)
Fence _ **
(Chipang'uj
Schematic Representation of the Spatial Symbolism of thehoma Ritual
Planes of Classification
3i
We are now beginning to see the development of a whole series of
classifications, symbolized in spatial orientations and in different
kinds of objects. They are for the most par t arranged in a set of
what Levi-Strauss mi ght well call "bi nar y discriminations. " But,
before we analyze the pat t ern, a few more variables have to be fed
into the system. At performances I observed, the patient' s husband
entered t he " c ool " ikela with her, standing on the "r i ght - hand s i de "
nearer t he men' s fire, while she stood on the left. Then, after havi ng
been splashed with cool and hot medicine, she entered t he connect-
ing tunnel first, while he followed her. As a variant t he senior adept
(or " bi g doct or ") swept bot h wife and husband wi t h cool and hot
medicine. Then his assistant took over for a while and did likewise.
White and Red Fowls
When t he patient first enters the cool ikela, she is given the young
white pullet to hold; duri ng the rites she clasps it against her left
breast, where a child is held (see figure 2). Both husband and wife,
incidentally, are naked except for narrow waist-cloths. This is said
to represent t he fact t hat they are at once like infants and corpses.
The adepts, in contrast, are clothed. The mat ure red cock is laid,
trussed up by t he feet, on t he right of the hot ikela, in fact on the
men' s side, ready to be sacrificed by beheadi ng at t he end of t he
rites. Its blood and feathers are poured into the hot ikela as the final
act of the rites, as t he antithesis of the reception of t he white pullet
by the woman patient, whi ch begins the rites. The white chicken is
said to stand for ku-koleka, "good luck or st rengt h, " and ku-tooka,
"whiteness, purity, or auspiciousness." But the red cock, as we have
seen, represents the chisaku, or mystical misfortune, t he "sufferi ng"
of the woman. The white pullet, according to one informant, also
stands for lusemu, procreative capacity. " Th a t is why it is given to
the woman, " he said, "for she is t he one who becomes pregnant
and gives bi rt h to children. A man is j ust a man and he can' t be
pregnant. But a man gives power to women to have children who
32
The Ritual Process
FIGURE 2. Isoma: the woman pati ent holds the whi te pullet against her
left breast, representing the side of nurturance.
can be seen, who ar e visible. Th e r ed cock stands for t he man,
per haps t he gr udge is t h e r e " [i. e. , agai nst hi m] . " I f t he woman
still has no chi l dren after t he rites, t he gr udge would be with the
wo ma n " [i.e., woul d not be connect ed wi t h her mari t al situation,
but woul d have arisen i n ot her sets of r el at i ons ] . Finally, it is prob-
abl y of significance, al t hough unst at ed, t h a t t he red cock remains
Planes of Classification
33
trussed up and unmoving through the rites, while the white hen
accompanies the woman as she moves through the tunnel from " l i f e "
to " d e a t h " and back to " l i f e " again. In other Ndembu ritual con-
texts, movement represents life and stillness deat h: the cock is con-
secrated for slaughter.
TH E CURATIVE PROCESS
The rites in the makela follow a processual pat t ern. The first phase
consists of a passage from the cool to the hot ikela, the woman leading
and the man following. At the hot ikela the doctors mingle their
splashings of medicine with exhortations to any witches or curse-
layers to remove their inimical influences. Next the marital pair,
FIGURE 3. Isoma: the doctor beside the calabash splashes the patients
with medicine, while the men stand on the right of the tunnel's longi-
tudinal axis singing the kupunjila "swayi ng" song.
Bilkent Uni vers
Library
34
The Ritual Process
in t he s ame or de r , ret urn to the cool ikela, where they are again
splashed wi t h me di ci ne (See figure 3 ) . Then they cross once more
to t he hot ikela. Th e r e follows a t emporary lull, duri ng which the
hus band is e s cor t e d out of the ikela to fetch a small cloth to wipe the
medi ci ne from t h e faces of the couple and the body of the pullet.
He r et ur ns t o t h e cool ikela, and after further medication, there is a
pr ol onged i nt e r va l , dur i ng which beer is brought and drunk by the
at t ender s a n d t h e husband. The patient, herself is forbidden to
dri nk any. Aft er beer , beginning again in the cool ikela, the splashing
FIGURE 4. Isoma: the husband prepares to follow his wife through the
tunnel.
is r es umed. Th i s t i me ar ound, the husband leads the way to the hot
ikela (See fi gure 4). Th e y ret urn to the cool ikela in the same order.
After spl ashi ng, t h e r e is anot her interval for beer. Then the sequence
Planes of Classification
35
FIGURE 5.Isoma: the cock is beheaded over the fire and its blood is
scattered in the " hot " hole.
cool-hot-cool follows, t he wife leading. Finally, t here is a like se-
quence at the end of which the red cock is beheaded and its blood
poured into t he hot ikela (See figure 5). Then t he couple are swept
once more with bot h types of medicine and cold wat er is poured
over t hem (See figure 6). In all, t he couple are splashed t went y
times, thirteen of t hem in the cool ikela, seven in t he hot, a ratio of
nearly two to one.
While splashing goes on, the mal e adepts on t he right and t he
female adults on the left sing songs from the great life-crisis and
initiation rites of the Ndembu: from Mukanda, boys' circumcision;
Alung'ong'i, the rites of a funerary initiation; Kayong'u, initiation into
divining; Nkula, a traditional women' s cult; and Wuyang'a, initia-
tion into hunt ers' cults. Periodically, they sing t he Isoma song
FIGURE 6.Isoma: cold water is poured over the couple.
Planes of Classification
37
"mwanamiyaya punjila," accompanied by a swaying dance called
kupunjila, which represents the danci ng style of the Mvweng'i ikishi
and, further, mimes the contractions of an abortive labor.
FIGURE J. Isoma: wife and husband squat in the newly-made seclusion
hut, where the white pullet will also be kept until it lays its first egg.
The hut is built just outside the village. The doctor holds in his right
hand the knife with which he beheaded the cock.
CLASSIFICATORY STRUCTURE: TRIADS
There is enough dat a to at t empt to analyze the structure of the rites
so far. First, t here are three sets of triads. Ther e is t he invisible
triadwitch, shade, and Mvweng'ito which is opposed the visible
triaddoctor, patient, and patient' s husband. In t he first triad, t he
witch is the medi at or between the dead and the living in a hostile
and lethal connection; in the second, the doctor is the medi at or
38
The Ritual Process
bet ween t he l i vi ng a n d t he dead in a conciliatory and life-giving
connect i on. I n t he first, t he shade is female and t he ikishi male,
whi l e t he wi t ch ma y be of either sex; in t he second, the patient is
female and her h us ba n d mal e. Th e doctor medi at es between the
sexes, in t hat he t reat s bot h . The Ndembu doctor, in fact, has many
at t r i but es t hat ar e r e gar de d as feminine in Nde mbu cul t ure; he can
pound medi ci ne i n a me a l mor t ar , a task normal l y undert aken by
wome n ; a n d he handl es women and talks to t hem about private
mat t er s i n a way t hat woul d be impermissible to men in secular roles.
On e t e r m for " doct or , " chimbanda, is said by Nde mbu themselves to
be connect ed wi t h t he . t e r m mumbanda, st andi ng for " woma n . "
I n bot h t ri ads t her e ar e close bonds of relationship between two
of t he par t ner s . I n t he first, t he shade and t he wi t ch are believed to
be mat r i l i neal ki n; i n t he second, t he husband and t he wife are
l i nked by affinity. Th e first pai r afflicts t he second pai r with misfor-
t une. Th e t hi r d par t ne r , Mvweng'i, represents t he mode of t hat mis-
fort une, and t he ot her t hi r d par t ner , t he doct or, t he mode of its
r emoval .
T h e t hi r d t r i ad is represent ed by t he 2 : i rat i o bet ween the cold
and hot "abl ut i ons, whi ch further may be hel d to symbolize the
ul t i mat e vi ct ory of life over deat h. Her ei n is cont ai ned a dialectic
t hat passes from life t hr ough deat h to r enewed life. Perhaps, at the
level of " deep s t r uct ur e, " one mi ght even connect the movement
of t he pat i ent in t he t unnel wi t h her act ual movement through
mar r i age from village t o village, mat r i ki n to spouse' s kin, and back
agai n on t he de at h or di vorce of t hat spouse.
CLASSIFICATORY STRUCTURE : DYADS
Th e ot her st ruct ural features of t he rites may be arrayed in terms
of criss-crossing bi nary oppositions. I n t he first pl ace, there is the
maj or opposition bet ween t he ri t ual site and t he wild bush, which
is roughl y similar t o t hat ma de by El i ade bet ween "cos mos " and
Planes of Classification
39
"chaos . " The other oppositions are best arranged in three sets in
columnar form, as follows:
Longitudinal
Burrow/new hole
Latitudinal
Left-hand fire/
Altitudinal
Below surface/above
Grave/fertility
Death/life
right-hand fire
Women/ men
Patient/patient's
surface
Candidates/adepts
Animals/humans
Mystical misfortune/
husband
Cultivated roots/bush Naked/clothed
Hot medicine/cool
curing
medicines
Whi te pullet/red cock Medi ci ne roots/
Fire/absence of fire
Blood/water
medicine
medicine leaves
Shades/living
Whi te pullet/red cock
Red cock/white pullet
These sets of pairs of opposed values lie along different planes in
ritual space. The first set is longitudinal and is spatially polarized by
the "ikela of l i fe" and the "ikela of deat h. " The second set is latitud-
inal and is spatially bounded by the mal e fire on the right and t he
female fire on the left. The third set is altitudinal and is spatially
bounded by the surface of t he ground and the floor of t he combined
makela and connecting t unnel . These oppositions ar e made by t he
Ndembu themselves in exegesis, in practice, or in bot h. I n terms of
spatial orientation the mai n oppositions ar e: ani mal -made hole/
man-made hole; left/right; below/above. These correspond respec-
tively to the paired values: death/life; female/male; candidates/
adepts. But, since these sets of values transect one another, they
should not be regarded as equivalent.
In Isoma, t he Ndembu are not saying, in t he nonverbal language
of ritual symbols, t hat deat h and feminity, and life and masculinity,
are equivalent; nor are they saying t hat candidates are in a feminine
role in relation to adepts (though they are certainly in a passive
role). Equivalences may be sought within each set (or col umn), not
between t hem. Thus, t he animal' s blocked lair-entrance is regarded
as similar to t he filled-in graves of people, to deat h, whi ch blocks up
40
The Ritual Process
life; t o t he myst i cal mi s f or t une t hat results in t he deaths of infants;
to " h e a t , " whi ch is a e u p h e mi s m for witchcraft and for grudges t hat
" b u r n " ; t he r ed cock, whos e color stands for " t he blood of witch-
cr aft " (mashi awuloji) i n Isoma ( Nde mbu witchcraft is necrophagous,
and i n ant i -wi t chcraft r i t es, r e d stands for t he blood exposed in
such feasts [see Tu r n e r , 1 967, p . 70]); and to " b l o o d " as a general
symbol for aggressi on, da n ge r , a nd, in some contexts, ritual impurity.
Th e new hol e, ma d e i n t h e di r ect i on of t he river source, symbolizing
t he spri ng of fertility, is r e gar de d, on t he ot her hand, as having
affinities wi t h fertility, life, cur at i ve procedures, coolness or cold-
nessa synonym for f r eedom from t he attacks of witches or shades
and hence for " h e a l t h " (wukolu); wi t h t he absence of "fi r e"i n
this cont ext a s ymbol for t h e was t i ng and dangerous power of witch-
craft; wi t h t he wh i t e pul l e t whi ch in this ri t ual represents and
even embodi es t he pat i e nt ' s fertility and by its color symbolizes (as
I have shown el s ewher ee. g. , 1967, pp. 69-70) such desirable
qualities as " goodne s s , he al t h, strength, puri t y, good fortune,
fertility, food, e t c . " ; a n d finally wi t h wat er, whi ch has much the
same r ange of senses as " whi t e ne s s , " t hough in t erms of process
r at her t han st at e.
Thes e positive a n d ne gat i ve qualities are suprasexual in their
at t r i but i on, a n d I bel i eve t h a t it woul d be a mi st ake to equat e them
too nar r owl y wi t h s exual differences. Th e l at t er ar e more closely
linked wi t h t he l eft - hand/ r i ght - hand opposition. I n this set, it can
har dl y be said t h a t t h e pat i e nt , her whi t e pullet, and t he cultivated
roots suppl i ed by t he wo me n have t he inauspicious connotations
allocated t o t he gr a ve / de a t h / h e a t symbolism of t he first set. I men-
tion this because ot he r wr i t er s , such as Herz, Needham, Rigby, and
Bei del man, admi t t e dl y i n r e gar d to ot her cultures, have tended to
list as member s of t he s a me set such pairs as left/right, female/male,
i nauspi ci ous/ auspi ci ous, i mpur e / pur e , etc. , thus regardi ng the link-
age bet ween femi ni ni t y a n d inauspiciousness as a frequentindeed,
almost a uni ve r s al - humani t e m of classification. Nor should the
bel ow/ above di chot omy b e correl at ed, in Nde mbu cul t ure, with the
sex division. T h e set of t e r ms ar r ayed under these heads is once more
Planes of Classification
41
sex-free, since, for example, t he patients below and t he doctors above
contain members of bot h sexes.
SITUATION AND CLASSIFICATION
In other types of ritual contexts other classifications apply. Thus,
in male circumcision rites, females and female attributes may be
regarded as inauspicious and polluting, but the situation is reversed
in girls' pubert y rites. What is really needed, for the Ndembu and,
indeed, for any other culture, is a typology of culturally recognized
and stereotyped situations, in which the symbols utilized are classi-
fied according to the goal structure of t he specific situation. Ther e
is no single hierarchy of classifications t hat may be regarded as
pervading all types of situations. Rat her, there are different planes
of classification which transect one another, and of which the con-
stituent binary pairs (or triadic rubrics) are only temporarily con-
nected: e.g., in one situation the distinction red/ white may be homo-
logous with male/female, in anot her with female/male, and in yet
another with meat/flour without sexual connotation.
Planes of Classification
Indeed, single symbols may represent t he points of interconnection
between separate planes of classification. It will have been noted
that the opposition red cock/white pullet in Isoma appears in all
three columns. In the life/death pl ane, t he white pullet equals life
and fertility as against t he red cock, which equals deat h and witch-
craft; in the right/left plane, the cock is masculine and t he pullet
feminine; and in the above/below plane, t he cock is above, since it
is to be used as " me di ci ne " (yitutnbu), poured down from above,
while the pullet is below, since it is closely linked, as child to mot her,
with the patient who is being medicated. This leads me to the pr ob-
lem of the "pol ys emy" or multivocality of many symbols, t he fact
42
The Ritual Process
t hat they possess ma n y si gni fi cat i ons simultaneously. One reason
for this may be found i n t he i r " n o d a l " function with reference to
intersecting sets of cl assi fi cat i ons. Th e binary-opposition red cock/
whi t e hen is significant i n a t l east t hr ee sets of classifications in Isoma.
If one is looking at omi s t i cal l y at each of these symbols, in isolation
from one anot her a n d fr om t h e ot her symbols in the symbolic field
(in t erms of i ndi genous exegesi s or symbol context), its multivocality
is its most st ri ki ng feat ur e. If, on t he ot her hand, one is looking at
t hem holistically in t e r ms of t he classifications t hat structure the
semantics of t he whol e r i t e i n whi ch t hey occur, then each of the
senses allocated to t he m a ppe a r s as t he exemplification of a single
pri nci pl e. I n bi nar y oppos i t i on on each pl ane each symbol becomes
uni vocal .
COGNITION AND EX I STENCE IN RITUAL SYMBOLISM
I conclude this chapt er by r el at i ng its findings to the standpoint of
Levi-Strauss in The Savage Mind. Levi-Strauss is qui t e correct in
stressing t hat la pensee sauvage cont ai ns propert i es such as homo-
logies, oppositions, cor r el at i ons , and transformations which are also
characteristic of sophi st i cat ed t hi nki ng. In t he case of the Ndembu,
however, t he symbol s t he y use i ndi cat e t hat such properties are
wr apped up in a mat e r i al i nt e gume nt shaped by their life experi-
ence. Opposi t i on does not a p p e a r as such but as the confrontation
of sensorily per cept i bl e obj ect s , such as a hen and a cock of different
ages and colors, in va r yi ng spat i al relationships and as undergoing
different fates. Al t hough Levi - St r auss devotes some attention to
t he role of r i t ual and my t h i c a l symbol s as instigators of feeling and
desire, he does not devel op t hi s line of t hought as fully as he does his
work on symbols as fact ors i n cogni t i on. (I have considered this
elsewhere at some length-for i nst ance, 1 967, pp. 28-3 0, 54-55O
Th e symbols and t hei r r el at i ons as found in Isoma are not only a set
of cognitive classifications for or der i ng t he Nde mbu universe. They
ar e also, and per haps as i mpor t ant l y, a set of evocative devices for
Planes of Classification
43
rousing, channeling, and domesticating powerful emotions, such as
hate, fear, affection, and grief. They are also informed with purpos-
iveness and have a " conat i ve " aspect. In brief, the whole person,
not just the Ndembu " mi n d, " is existentially involved in the life or
death issues with whi ch Isoma is concerned.
Finally, Isoma is not "gr ot es que" in the sense t hat its symbolism
is ludicrous or incongruous. Every symbolic item is related to some
empirical item of experience, as the indigenous interpretations of
the vegetable medicines clearly reveal. From the standpoint of
twentieth-century science, we may find it strange t hat Ndembu feel
that by bringing certain objects into a ring of consecrated space
they bring with these the powers and virtues they seem empirically
to possess, and t hat by mani pul at i ng t hem in prescribed ways they
can arrange and concentrate these powers, rat her like laser beams,
to destroy mal i gnant forces. But, given t he limited knowledge of
natural causation transmitted in Ndembu culture, who can doubt
that under favorable circumstances the use of these medicines may
produce considerable psychological benefit? The symbolic expres-
sion of group concern for an unfortunate individual' s welfare,
coupled with the mobilization of a bat t ery of " g o o d " things for her
benefit, and the conjunction of the individual' s fate wi t h symbols of
cosmic processes of life and deat hdo these really add up for us to
something merely "uni nt el l i gi bl e"?
Paradoxes
of Twinship
in Ndembu Ritual
TWINSH IP IN KI NSH I P AND LIFE :
SOME AFRI CAN EXAMPLES
I n t he first chapt er I anal yzed one ki nd of Ndembu ritual per-
formed to remedy a deficiency: e. g. , a woman' s t emporary incapacity
to produce or raise living chi l dr en. I now wish to consider an
Ndembu ritual whose raison d'etre is a n i mmoder acy of a different
sort. Thi s is the Wubwang'u r i t ual , whi ch is performed to strengthen
a woman who is expect ed t o be a r or wh o has already borne a set
of twins (ampamba). Her e t he difficulty is one of excess rat her than
defect, of overperformance r a t he r t h a n underperformance. The
beari ng of twins constitutes for t he Nd e mb u what we would call a
par adoxt hat is, a t hi ng t h a t conflicts wi t h preconceived notions
of what is reasonabl e or possi bl e. Th e r e ar e several absurdities in
t he physiological fact of t wi ns hi p for t he Ndembu. In the first
place, as we have seen, a hi gh cul t ur al pr e mi um is placed on
fertility (lusemu); yet her e we have a n exuber ance of fertility that
results in physiological a n d economi c distress. I n a society without
cattle or t he not i on t hat sheep a n d goat s can be milked for human
consumption, it is difficult for a mot h e r t o suppl y twins with ade-
quat e nouri shment by l act at i on. Oft en t hei r survival may depend
upon t he chance t hat anot he r wo ma n has recent l y lost a child, has
milk available, and is wi l l i ng t o nur s e one of t he twins. And even if
44
Paradoxes of Twinship in Ndembu Ritual
45
the twins survive until they are weaned, it may be difficult for their
parents alone to provide t hem with their subsistence. For this reason
they are symbolically represented in the rites as a charge upon the
community.
One way in which this is expressed is in a ceremonial dance
where the mot her of twins, clad only in a strip of bark cloth with a
frontal flap of leather or cloth, and carrying a flat, round winnowing
basket (Iwalu), makes t he round of all t he villages in a vicinage. As
she dances she raises t he flap to expose to all t he source of her ex-
cessive fecundity, and solicits offerings of food, clothing, and money
by circling her basket before t he onlookers. This dance exhibits
several motifs characteristic of Wubwang'u. One is t he suspension
of the rules of modesty, which are normally rigorously i ncumbent on
Ndembu women; anot her is the ritual power of vulnerability or
weaknessa motif pursued much further in Chapt er 3. Here I will
point out only t hat twinship is regarded simultaneously as a blessing
and a misfortune, bot h of which involve the wider communi t y in
the welfare of t he ritual subject.
But Wubwang'u exhibits anot her paradox in t he social order.
Professor Schapera (and other scholars) have drawn attention to
the fact t hat wherever kinship is structurally significant, and provides
a frame for corporate relationships and social status, the bi rt h of
twins is a source of classificatory embarrassment. For it is widely
held, in Africa and elsewhere, t hat children born duri ng a single
parturition are mystically identical. Yet, under t he ascriptive rules
associated with kinship systems, there is "only one position i n t he
structure of the family or corporate kin-group for t hem to occupy.
There is a classificatory assumption t hat human beings bear only
one child at a time and t hat there is only one slot for t hem to occupy
in the various groups articulated by kinship which t hat one child
enters by bi rt h. Sibling order is anot her i mport ant factor; older
siblings exert certain rights over j uni or siblings and may in some
cases succeed to political office before t hem. Yet twinship presents
the paradoxes t hat what is physically double is structurally single
and what is mystically one is empirically two.
46
The Ritual Process
Afr i can societies resolve this dilemma in various ways. One
r e me d y for t he structural contradiction produced by twinship is to
p u t t he t wi ns to deat h. This practice is followed by the Bushmen of
t h e Ka l a h a r i , of whom Baumann writes: " L'infanticide est friquentpar
suite des conditions konomiques difficiles, mais le meurtre des jumeaux ou
de Vun d'entre eux est du a la croyance qu'ils portent malheur " (Baumann
a n d We s t e r ma nn, 1962, pp. 1 00- 1 01 ) . The paradox is here resolved
by t h e dest r uct i on of one or bot h of t he twins, who are believed to
br i n g (myst i cal ) misfortune. Ot her societies do not destroy twins
b u t r e move t hem from t he kinship system to which they belong by
bi r t h a n d confer on t hem a special status, often with sacred attributes.
Th u s , a mo n g t he Ashanti, according to Rat t ray (1 923 ), "t wi ns, if
b o t h of t h e same sex, belong, as of right, to the chief, and become,
if gi rl s, hi s pot ent i al wives, if boys, elephant-tail switchers at the
cour t . T h e y must be shown to him as soon as possible after birth,
be i ng car r i ed to t he ' pal ace' in a brass basin. Twins, on state
occasi ons, ar e dressed in white, each al i ke" (p. 99).
Wh i t e , among the Ashanti, is a symbol, inter alia, for divinity and
t h e " s p i r i t u a l " and fertilizing fluidswater, semen, and saliva.
T h e e l e phant is also connected with exuberant fertility, as is evi-
de n c e d i n t he girl' s pubert y ritual, during which the novice "touches
t h r e e r oas t ed pieces of elephant' s ear, while t he following words are
addr e s s e d t o her : ' Ma y the elephant give you her womb that you
ma y b e a r t en chi l dr e n' " (1 923 , p. 73 ). Ashanti chiefs have many
of t h e at t r i but es of "di vi ne ki ngs " and are believed to transcend
t he cl eavages between sectional groups in their realms, with whose
wel far e a n d fertility their own are mystically identified. Thus, twins
ar e lifted out of the secular structure and participate in and sym-
bol i ze t h e sacredness and fertility of the chief. But twins born in the
r oyal fami l y itself are killed, for such an event is said to be "hat eful "
t o t he Gol de n Stool, supreme insignium and expression of Ashanti
r oyal t y ( 1 923 , p. 66). Thi s is presumabl y because twins would intro-
duc e cont r adi ct i on into t he st ruct ure of the royal matrilineage, giving
ri se t o pr obl ems of succession, inheritance, and precedence.
Ac c or di ng to Evans-Pri t chard (1956), the Nuer of t he Nilotic
Paradoxes of Twinship in Ndembu Ritual
47
Sudan assert t hat twins are one person and t hat they are bi rds:
"Thei r single social personality: is something over and above their
physical duality, a duality which is evident to the senses and is
indicated by the plural form used when speaking of twins and by
their treatment in all respects in ordinary social life as two quite
distinct individuals. It is only in certain ritual situations, and
symbolically, t hat t he unity of twins is expressed, particularly in
ceremonies connected with marri age and deat h, in which t he
personality undergoes a change " (pp. 1 28 - 1 29) . In this society,
twins are not removed from t he social structure, but they neverthe-
less acquire a ritual and symbolic value. They are symbolically
identified with birds, not only on account of the resemblance between
"t he multiple hat chi ng of eggs and the dual birth of bi r ds " (p. 1 3 0) ,
but also because twins, like birds, are classified by t he Nuer as
"people of t he above" and "chi l dren of God. " "Bi rds are children
of God on account of their being in t he air, and twins belong to the
air on account of their being children of God by the manner of
their conception and bi r t h" (p. 1 3 1 ) . The Nuer thus resolve the
paradox of twinship by relating the single personality of twins to the
sacred order, and their physical duality to the secular order. Each
aspect operates on a distinct cultural level, and the concept of
twinship mediates between t he levels.
In many societies, twins have this mediating function between
animality and dei t y: They are at once more t han human and less
than human. Almost everywhere in tribal society they are hard to
fit into the ideal model of t he social structure, but one of the para-
doxes of twinship is t hat it sometimes becomes associated with
rituals t hat exhibit t he fundamental principles of t hat st ruct ure;
twinship takes on a contrastive character analogous to the relation-
ship of ground to figure in Gestalt psychology. Indeed, one often
finds in human cultures t hat structural contradictions, asymmetries,
and anomalies are overlaid by layers of myt h, ritual, and symbol,
which stress t he axiomatic value of key structural principles with
regard to the very situations where these appear to be most in-
operative.
48 The Ritual Process
Among ma n y Bant u- speaki ng peoples, including the Ndembu,
twins ar e n e i t h e r p u t t o deat h nor permanent l y assigned a special
status as a mo n g t he Ashant i . But, at the life crises of their birth,
mar r i age, a n d de a t h , special rituals are performed and they have
almost al ways a l at ent l y sacred character, which becomes visible
at all rites c onc e r ni ng t wi n births. Moreover, the parent s of twins
and cer t ai n of t he i r siblings, especially the one immediately follow-
ing t he m i n b i r t h or der , fall within t he penumbr a of this sacredness.
For e xampl e , Mo n i c a Wi l son (1957) writes:
Twin birth is a fearful event to the Nyakyusa. The parents of twins and
twins themselves are abipasya, the fearful ones, felt to be very dangerous to
their relatives and immediate neighbours, and to cattle, causing them to
suffer from di arrhoea or purging, and swollen legs, if any contact takes
place. Therefore, the parents are segregated and an elaborate ritual is per-
formed, in which a wide circle of kinsmen and neighbours and the family
cattle participate. The infants are naturally segregated with their mother, but
it is the danger from the parents rather than from the twins themselves that
is emphasized. Ilipasa is commonly used to mean " twins," " twin birth," but
it is more accurately translated as " abnormal birth," for it is used of a child
born feet foremost (unsolola) as well as for any multiple birth, and the same
ritual is performed whatever the type of ilipasa (p. 1 5 2).
T h e a i m of t h e Nyakyus a rites is to rid twins and their parents of
t he dange r ous cont agi ousness of their condition. The parents must
be t r eat ed wi t h medi ci nes and ri t ual so t hat they may produce one
child at e a c h bi r t h hencefort h and so t hat they may not affect their
nei ghbor s wi t h myst i cal illness. Among t he Nyakyusa and other
Bant u societies, s uch as t he Suku of t he Congo, of whose twinship
rites va n Ge n n e p (1909) has wri t t en, and the Soga of Uganda
(Roscoe, 1 924, p . 1 2 3 ) , t wi nshi p rites involve the whole local com-
muni t y. Va n Ge n n e p dr aws at t ent i on to t he fact t hat at the Suku
rites of r e i nt e gr at i on, following a long " l i mi nal " period during
whi ch t he t wi ns ar e secluded from cont act with the public life for
six years, t he r e is a "r i t ual i st i c traversing of the territory belonging
to t he soci et y as a whol e and a (general) sharing of food" by the
villagers ( p. 47) . I have al ready ment i oned how the Ndembu
Paradoxes of Twinship in Ndembu Ritual
49
regard twins as a charge upon the whole community. This may be
regarded as anot her instance of a widely prevalent social tendency
either to make what falls outside the norm a mat t er of concern for
the widest recognized group or to destroy the exceptional pheno-
menon. In the former case, the anomalous may be sacralized, re-
garded as holy. Thus, in eastern Europe, idiots used to be regarded
as living shrines, repositories of a sacredness that had wrecked their
natural wits. They were entitled to food and clothing from everyone.
Here the anomaly, the "st one that the builders rejected," is removed
from the structured order of society and made to represent the simple
unity of society itself, conceptualized as homogeneous, rat her t han
as a system of heterogeneous social positions. Among Ndembu, too,
the whole biology of twinning is sacralized and made into a mat t er
for everyone, not just for the mother' s close kin. The mother' s
affliction with too much of a good thing becomes the community' s
responsibility. It also becomes an occasion on which t he communi t y
can celebrate and extol some of its crucial values and principles of
organization. The paradox t hat what is good (in theory) is bad
(in practice) becomes the mobilizing point of a ritual t hat stresses
the overall unity of the group, surmount i ng its contradictions.
To repeat : there are two things t hat can be done about twinship in
a kinship society. Either you can say, like t he little boy on first seeing
a giraffe, " I don' t believe i t , " and deny the social existence of the bio-
logical fact; or else, having accepted the fact, you can try to cope
with it. If you try to cope, you must make it, if you can, appear to be
consistent with the rest of your culture. You may, for example, in some
situations focus attention upon the duality of twins, and in others
upon their unity. Or you can reflect upon natural and social pro-
cesses whereby what were originally two separate and even opposed
elements fuse to form something new and uni que. You can examine
the process whereby two become one. Or you can examine the converse
of this, the process whereby one becomes two, the process of bifurca-
tion. Still further, you can regard the number Two as being itself
representative of all forms of plurality as opposed to unity. Two
represents the Many as opposed to t he One, as derived from it, or
as fused with it again.
5
The Ritual Process
Fur t h e r mor e , if you pay at t ent i on to the Two, disregarding the
On e for t h e mome n t , you may regard it as comprising either a pair
of similars, a di oscur al pai r like Castor and Pollux, or a pair of
opposi t es, like mal e and female, or life and deat h, as in the Isoma
r i t ual . Nd e mb u , in t he symbolic idiom of the twinship ritual, have
elected t o e mphas i ze t he aspect of opposition and complementarity.
Al t hough t wi ns, in nat ur e, are frequently of the same sex, and,
i ndeed, i dent i cal t wi ns are always of t he same sex, Ndembu stress
in Wubwang'u t he equal but opposite aspect of duality. Pursuing this
view fur t her , wh e n t hey exhibit t he process of uni t i ng the compo-
nent s of t he dyad, t hey represent this process as a coincidence of
opposi t es, a n d not as a doubl i ng of similars. Sexual symbolism is
used t o por t r a y t hi s process, but I hope to show t hat very much more
t h a n sexual i nt er cour se is i nt ended by it. Th e idiom of sexuality is used
to r epr es ent t he processes by whi ch social forces approximately
equal i n s t r e ngt h a n d opposite in quality are exhibited as working
in h a r mon y . I n this chapt er I shall be mai nl y concerned with the
social referent s of symbols t hat also represent aspects of sexuality.
Th e fusion of a pl ural i t y of sociocultural referents wi t h a plurality
of or gani c referent s (including those wi t h a sexual character) in a
single vi si bl e represent at i on, invested by believers wi t h an extra-
or di nar y powe r , and possessing a new quality of human communi-
cat i on, is a n i mpor t ant characteristic of religious symbols. To say-
t hat ei t her set of referents, cul t ural or organi c, is " bas i c" or
" p r i ma r y , " a n d t hat t he ot her is reducible to it, is to overlook the
qual i t at i ve difference from either set presented by t he pat t ern of
t hei r i nt e r de pe nde nce .
TH E PLOT OF TH E NDEMBU TWIN RITUAL
T h e uni fyi ng of a pai r of opposites, domi nant l y expressed in symbols
for mal e- femal e difference, opposition and uni on, constitutes what
ma y be cal l ed t he r i t ual " p l o t " of Wubwang'u. I propose to select
t wo i mpor t a n t episodes in t he ritual and to exami ne each in turn
Paradoxes of Twinship in Ndembu Ritual
51
with reference to its symbolism. Like most Ndembu cults of afflic-
tion, the cult association of Wubwang'u is made up of persons who
have themselves undergone as patients t he ritual t reat ment char-
acteristic of Wubwang'u. The afflicting spirit is believed to be t hat
of a deceased member of t he cult. The adepts or doctors collect
vegetable medicines for the patient, adorn themselves in a special
W
a y , and then wash the patient with pounded leaf-medicines and
give her medicine to drink mixed with water. A shrine is made for
the patient near the door of her hut, and cult members perform a
number of rites in connection with it. Both men and women may
act as doctors, for men who were themselves members of a pai r of
twins, who were sons or fathers of twins, or whose wives, mothers,
or sisters have been successfully treated by t he Wubwang'u procedure,
have the right to learn t he medicines and techniques of Wubwang'u.
According to my records, the afflicting spirit is always t hat of a
woman, and in the majority of cases is believed to be t he patient' s
own mother' s mother.
Wubwang'u may be performed for a woman who has j ust borne
twins or for a woman who is expected to bear twins. It is expected,
for example, t hat a woman whose own mother, mother' s mother,
or both have borne twins, or who was one of a twin pai r herself, will
have twins. If such a woman experiences any form of reproductive
disorder duri ng pregnancy, Wubwang'u may be performed for her
often without consulting a diviner. Ot her women, unconnected in
any way with twinning, may become patients in Wubwang'u, if they
have suffered from reproductive troubles. This is often because
relatives of the ailing woman have consulted a diviner, who has
consulted his symbolic objects and decided t hat a spirit " i n
Wubwang'u for m" has caught her. All Ndembu rituals concerned
with female reproduction have bot h a specific and a general aspect,
dealing explicitly with a part i cul ar culturally defined disorder but
being held capable of curing other kinds. Thus, Nkula is properly
for menstrual troubles but is also performed for miscarriage, frigidity,
and barrenness, while Isoma is for miscarriage and stillbirth but also
deals with menstrual disorders. Similarly, Wubwang'u as a generic
52
The Ritual Process
curat i ve r i t ual is bel i eved to benefit women suffering from a variety
of r e pr oduct i ve di s or der s . But its mai n symbolic emphases are on
t wi nbi r t h, j us t as t hos e of Nkula are on menorrhagi a, and Isoma on
mi scarri ages.
Th e t wo epi s odes (of which the second is subdivided into two
phases) t o wh i c h I woul d like to draw attention ar e: (i ) the Rites
of t he Ri ve r So u r c e ; a n d (2) the Maki ng of the Twi n Shrine, with
t he Frui t ful Con t e s t of t he Sexes. In t he first, the unity of the sexes
i n mar r i age is r e pr e s e nt e d as a mystery; in the second, the sexes
ar e r epr es ent ed i n t he i r division and opposition.
Properties of Ritual Symbols
Each of t hese epi s odes is charged with symbolism.
1
Such symbols
exhi bi t t he pr ope r t i e s of condensation, unification of disparate referents,
and polarization of meaning. A single symbol, in fact, represents many
t hi ngs at t h e s a me t i me : it is multivocal, not univocal. Its referents
ar e not all of t h e s a me logical order but are drawn from many
domai ns of s oci al experi ence and ethical evaluation. Finally, its
referents t e nd t o cl ust er around opposite semantic poles. At one
pol e t he r efer ent s a r e t o social and moral facts, at the other, to
physi ol ogi cal fact s. Th u s , t he mudyi (Diplorrhyncus condylocarpon) tree,
cent ral s ymbol of t h e girls' pubert y ri t ual , means simultaneously
breast mi l k a n d mat r i l i ny, while t he mukula (Pterocarpus angolensis)
t ree st ands for t h e bl ood of circumcision and the moral community
of ma t ur e t r i be s me n. Such symbols, t hen, uni t e the organic with
t he s oci omor al or de r , procl ai mi ng t hei r ul t i mat e religious unity,
over and a b o v e conflicts between and wi t hi n these orders. Powerful
drives a n d e mot i ons associated with h uma n physiology, especially
wi t h t he phys i ol ogy of reproduct i on, are divested in the ritual process
1
See Turner, 1 9 67 , for a discussion of what I consider to be the kinds of data
from which the mai n semantic components and properties of religious symbols
may be inferred, and I will not repeat the whole argument here.
Paradoxes of Twinship in Ndembu Ritual
53
of their antisocial quality and attached to components of the nor-
mative order, energizing t he latter wi t h a borrowed vitality, and
thus maki ng t he Durkhei mi an "obl i gat or y" desirable. Symbols
are both t he resultants and the instigators of this process, and
encapsulate its properties.
RITES OF TH E RIVER SOURCE:
COLLECTION OF MEDICINES
The Rites of the River Source at Wubwang'u exemplify most of these
properties. They form par t of a sequence of ritual activities t hat
makes up t he first phase of this ritual of twinship. As in Isoma, and
indeed other Ndembu rituals of affliction, t he collection of medicines
(ku-hukula yitumbuliterally, " t o snatch or steal medi ci nes"or
ku-lang'ulayitumbu) is the first activity in t he sequence. Th e Wubwang'u
doctor adepts who perform this carry wi t h t hem into t he bush a
number of foods in t he senior practitioner' s winnowing basket (Iwalu).
These may include a cassava root, beans, groundnuts, a l ump of salt,
maize grains, portions of t he meat of domestic animals and wild pig,
and other comestibles. They bri ng t he white beer made from maize
or bulrush millet, the color of which makes it an appropri at e libation
for the shades, who are symbolically " wh i t e " {a-tooka) beings. They
also carry white clay in a phallus-shaped calabash (see Fi gure 8) and
powdered red clay in t he shell of a wat er mollusc (nkalakala) (see
Figure 1 7, p. 74). According to informants, " t he foods are brought to
strengthen t he bodies of t he mot her and chi l dren, " while t he white
clay is " t o make t he children strong, pure, and fort unat e. " Several
informants held t hat the red clay means " ba d luck (ku-yindama), lack
of strength (kubula kukoleka), and lack of success (ku-halwa)." But, as
we shall see below, p. 69, this same red clay at the Rites of t he River
Source represents " t h e blood of t he mot her. " This is yet anot her
example of the way in which the same symbols have varying sig-
nificance in different contexts. The binary-opposition white/red at
different episodes of Wubwang'u represents strength/weakness, good
54
The Ritual Process
FIGURE 8. Twi n ceremony: an adept carries the ritual winnowing
basket, containing a calabash of white beer and a phallus-shaped
calabash filled with white clay. She is receiving a medicine branch.
l uck/ bad l uck, heal t h/ di s eas e, pur i t y of heart / a grudge causative of
wi t chcraft , s e me n/ ma t e r na l blood, masculinity/femininity.
T h e ba n d of ade pt s is he ade d by a mal e and a female senior practi-
t i oner. Thes e ade pt s ar e accompani ed by their chi l dren; indeed,
Wubwang'u is t he onl y ki nd of Nde mbu ritual in which children are
enjoined t o par t i ci pat e i n collecting " me di ci ne s " (yitumbu), to use a
t r adi t i onal but not whol l y appr opr i at e t er m for t he vegetable sub-
Paradoxes of Twinship in Ndembu Ritual
55
stances. Each child carries a leafy branch taken from every "me di -
cine " tree or bush visited. Bawdy songs are sung duri ng t he medicine
collection " t o make the pat i e nt " strong, and a double hunting-bell
(mpwambu) is rung by the principal doctor. Its purpose is " t o open
the ears of the unborn children so t hat they may know they are t wi ns. "
The singing and bell-ringing are also " t o arouse the shades" (ku-
tnnisha akishi), for each doctor adept has a guardi an shade who was
formerly a Wubwang'u cult-member. Furt hermore, they are held to
"r ous e" t he medicine trees, t he species from which Wubwang'u
medicine potions and lotions will be prepared. Wi t hout these stim-
ulating sounds, it is believed the trees would remai n merely as trees;
with t hem arid with their accompanyi ng rites of sacralization, they
become magically efficacious powers, akin to the " vi r t ue s " possessed
by herbs in Western folk-therapy..
In a text on medicine collection which I cite in full on pp. 86-88,
there is a passage t hat r uns: " The r e must be a renewal (or causing to
rise up) and scattering of those former (or traditional) words and a
cutting (of medicines). " These " wor ds " are the songs and prayers of
Wubwang'u, and they mystically affect the cutting of medicine plants.
An example of prayer is to be found when t he domi nant symbolic
medicine of t he rites is consecrated, t he kata wubwang'u tree. First t he
senior practitioner dances around it in a circle because " h e wants to
please the shade, " for it is t he big tree of the Wubwang'u shade
"big " t hat is, in ritual status, for all t he trees I have seen treated in
this way were slender young specimens. Then he digs a hole over its
tap root and places t he items of food in it, while he utters the following
prayer:
Eyi mufu wami kanang'a wading'i naWubwang'u,
" You, O my dead [kinswoman] who had Wubwang'u,
neyi muntu wunatnwidyikili dehi muWubwang'u,
if you have come out to someone today in Wubwang'u,
ifuku dalelu mukwashi chachiwahi
this very day you must help her well,
ashakami chachiwahi nawanyana.
that she may sit well with children."
56
The Ritual Process
A libation of beer is t h e n p o u r e d i nt o the hole on t he food so that
" t h e shades may come t o e a t a n d dr i nk t here. " Then the doctor fills
his mout h wi t h wat e r or b e e r a n d powdered white clay (mpemba or
mpeza) and blows i t ove r t h e l a ugh i n g scattering onlookers in sign
of blessing. Next t he p a t i e n t is ma d e to stand touching the tree and
facing east whi l e st ri ps of b a r k a r e cut from it into the winnowing
basket (see Fi gures 9 a n d 1 o) a n d a fronded branch is cut and given
FIGURE 9 .Twi n Cer emony: t he patient stands touching the medicine
tree while facing east, t he di rect i on of rebirth. The doctor cuts portions
of bark into t he wi nnowi ng basket with his ritual ax.
Paradoxes of Twinship in Ndembu Ritual
57
FIGURE i o.Twi n ceremony: this figure illustrates the ritual identifi-
cation of twinsin this case of opposite sex. The man in white is twin
to the female patient whose back is to the molu waWubwang'u vine,
from which medicine fronds are being cut. H e must stand near her at
every medicine cutting.
to a woman adept to carry. According to one adept, "s he faces east be-
cause everything comes from the east (kabeta kamusela) where the sun
rises; when someone dies, his face is turned toward the east, meani ng
58
The Ritual Process
t hat he will be bor n agai n, but a s t er i l e pe r s on (nsama) or a witch
(muloji) is buri ed facing west so t h a t h e wi l l di e forever. " In brief,
t he east is t he auspicious and l i fe-gi vi ng di r e ct i on.
Th e kata wubwang'u t ree, as i n Isoma, is k n o wn as " t h e el der " or
" t h e place of gr eet i ng, " and is a mul t i voc a l s ymbol (i.e., one having
many designations). Such a s ymbol is r e g a r d e d as t he critical site of
transition from secular t o sacred ways of b e h a v i n g . I n Wubwang'u, a
clear distinction is made bet ween me di c i ne s col l ect ed in t he dry bush
(yitumbu ya mwisang'a) and t hose col l ect ed i n t h e streamside forest
(yitumbu yetu). Th e bush is r egul ar l y as s oci at ed wi t h bot h hunting
and virility, whi l e t he st r eamsi de forest is l i nked wi t h femininity.
Wome n make gardens in t he r i ch bl ack al l uvi al soil beside streams,
and soak t hei r cassava root s in pool s n e a r b y . I n Wubwang'u, t here is a
separat e " e l d e r " t ree for t he bus h a n d on e for t he st ream. Kata
wubwang'u is t he " e l d e r " for t he bus h. T h e frui t of t hi s tree is divided
i nt o t wo symmet ri cal por t i ons, wh i c h Nd e mb u compar e explicitly
wi t h twins (ampamba or ampasa). A n u mb e r of ot h e r trees of t he dry
bush ar e next visited for bar k s cr api ngs a n d leafy br anches. Below is
a list cont ai ni ng t he names of each speci es, fol l owed by an abbreviated
nat i ve expl anat i on of why it is us ed.
SPECIES
Ndembu Term Botanical Name
I. Kata Wubwang'u
p
a. Museng'u Ochna pulchra
3 . Mung'indu Swartzia
madagascariensis
4. Mucha Parinari mobola
5 . Mufung'u ? Arisophyllea
boehmii
6. Kapepi Hymenocardia
acida
7. Musoli Vangueriopsis
lanciflora
NDEMBU EXPLANATION
"Doubl e- frui t twi ns"
" One flower makes many small
frui tstwi ns are like one person"
" Bears fruits, thus will give
mother many chi ldren"
Same as 3
Same as 3
Same as 3 " has thin fruits, like
leaves, these are sour (batuka),
used as reli sh "
" From ku-solola, 'to make visible'
to make a woman with no
chi ldren to produce young ones"
Paradoxes of Twinship in Ndembu Ritual
59
SPECIES NDEMBU EXPLANATION
Ndembu Term
8. Mukula
Botanical Name
Pterocarpus
angolensis
" Its red gum is called 'blood'
to give a woman enough blood at
time of birth "
9. Mudumbila
"H as fruits, gives a woman
fertility"
10. Muhotuholu
Canthium venosum "From ku-hotomoka, 'to fall sud-
denly,' so must the woman's
trouble slip from her body "
1 1 . Mudeng'ula
12. Mwang'alala
Paropsia brazzeana
"H as fruits" (see 3)
"From ku-mwang'a, 'to scatter,'
means to scatter the disease "
To this set of vegetable medicines is added a portion of hornets' nest.
"Perhaps this is because of its many young, " one informant guessed.
That completes t he list of bush medicines. Next, a number of
medicines are obtained from the streamside (gallery) forest. The
"el der " tree for the streamside is a creeper called molu waWubwang'u,
"t he vine of IVubwang'u." Ndembu say: "Molu waWubwang'u grows
into many different branches and spreads to form a large place of
its own. In just the same way a woman should have as many children
as the creeper has branches. " Its later use in Wubwang'u is twofold:
first, it is intertwined among the children' s medicine branches, which
have been set upright near the patient' s hut to form a tiny double
enclosure like the letter m, which serves as a shrine for t he afflicting
shade; second, it is draped over the patient' s shoulders and around
her breasts. This use recalls its role as a medicine for maki ng a
woman' s breast milk white, if it becomes yellow or reddish. This dis-
colored milk is called nshidi (" sin " ) . If the milk is red or yellow, witch-
craft is felt to be somehow involved in t he anomal y; the mot her herself
may be a witch, or someone else is bewitching her. Molu medicine
restores the milk to its proper color (see also Turner, 1967, p. 3 47).
White things are believed by Ndembu to stand for such virtues and
values as goodness, purity, good health, good luck, fertility, openness,
social communion, and a number of other auspicious qualities. Thus,
molu, the domi nant symbol of the streamside, stands for motherhood,
6o
The Ritual Process
l act at i on, t h e br east s, a n d fertility. Like mudyi, molu represents the
n u r t u r a n t as pect s of mot her hood.
T h e ot h e r s t r eams i de medi ci nes are t hen collected. These are, in
or de r of col l e ct i on:
SPECIES
Ndembu Term
, Molu waWubwang'u
2. Musojisoji
3 . Muhotuhotu
4. Mudyi
5 . Katuna
6. Mutung'ulu
Botanical Name
Possibly a species
of Convolvulaciae
Canthium venosum
Diplorrhyncus
condylocarpon
(Uvariastrom
hexalobodies)
Harungana
madagascariensis
NDEMBU EXPLANATION
" It grows into many different
branches and forms a large place
of its own; it spreadsthus a
woman should have as many
children as the creeper has
branches "
" It has many fruits, will make
woman fertile "
See bush medicines above, p. 59
(10)
"Because it is used in Nkang'a,
the girls' puberty rites, to make
a woman mature and fruitful"
" Katuna has red sap. As a child
is born accompanied by blood,
so should a mother have much
blood"
" It has many spreading roots
a woman should have many
children. Ku-tung'ula means 'to
speak of a person behind his
back' perhaps the grudge
(chitela) comes from this "
Commentary
T h e gr e at maj or i t y of these species represent t he woman' s desired
fruitfulness. Some ar e connect ed wi t h t he i dea of mat ernal blood.
On e a de pt vouchsafed t he i nformat i on t hat an unborn child "eats
food t h r ough t he bl ood of t he mot her , " t hereby indicating some
knowl e dge of t he physi ol ogy of r epr oduct i on. Wh a t is of great interest
is t he connect i on of such medi ci nes as muhotuhotu and mutung'ulu with
t r oubl e, backbi t i ng, and grudges. These r un like a red thread through
t he i deol ogi cal s t r uct ur e of Wubwang'u, and are in fact connected
Paradoxes of Twinship in Ndembu Ritual
61
with its red symbolism. Thus, from the powdered red clay brought
by the senior practitioner, the children who accompany their doctor
parents into the bush decorate their faces (see Figure 1 1 ) . Those of
them who are twins draw a red circle around their left eye, and, with
powdered white clay, a white circle around the right eye. These are
"for the shades of twins or mothers of twins, " informants told me.
According to one of them, t he red circle "represents blood, " while
the white one stands for " s t r e ngt h" or "good luck'.' But another said
explicitly t hat the red circle stands for " t h e gr udge " (chitela), and
since it was around the left, or "femi ni ne, " eye, "per haps the grudge
comes from t hat side. " Asked what he meant by this, he went on to
say that perhaps there was ill-feeling between the pat i ent and her
grandmother when t he latter, now an afflicting Wubwang'u shade,
was alive. On t he other hand, he continued, t he shade mi ght have
been angered by quarrels in the matrilineal kin-group (akwamama,
" those on t he mother' s side ") and have punished one of its members.
In any case, he said, grudges are found more often i n t he matrilineage
(ivumu, or " womb" ) t han among pat ernal kin, who have goodwill
toward one another. This was a conscious at t empt to interrelate the
binary oppositions male/female, patrilaterality/ matriliny, good will/
grudge, white/red in a completely consistent manner.
Implicit in this interpretation, too, is the paradox of twinship itself.
Twins are bot h good luck and reasonable fertilityand in this
respect have an affinity wi t h t he ideal relationship t hat should inter-
link patrilateral ki nand bad luck and excessive fertility. Th e
Ndembu, incidentally, regard twins of opposite sex as being more
auspicious t han twins of t he same sexa view widely held in African
societiespossibly for t he reason t hat twins of t he same sex occupy
the same sibling position in t he kinship and political structure.
Apart from t he twin-fruit symbolism of kata wubwang'u and t he
many-in-one symbolism of museng'u, t he medicines themselves do not
make explicit reference to twinship. Rat her do they cumulatively
represent exuberant fertility. But t he sharp distinction made in the
rites between medicines of t he bush and those of the gallery forest,
a distinction connected by informants with t hat between masculinity
and femininity, is related to t he mai n dualistic t heme of Wubwang'u.
62
The Ritual Process
FIGURE 1 1 . Twi n ceremony: children are marked with white and red
circles ar ound their eyes, distinguishing them in categories of twin and
non-twin.
Paradoxes of Twinship in Ndembu Ritual
63
TH E RITES OF TH E RIVER SOURCE :
TH E STREAM AND TH E ARCH
The mudyi tree (the "mi l k- t r ee") , focal symbol of t he girls' pubert y
rites, also appears in t he twin ritual. Characteristically, it appears
in an episode t hat portrays t he mystical unity of opposites. After t he
medicines for the basket have been collected, the senior mal e pract i -
tioner cuts a pliant wand of mudyi, and anot her of muhotuhotu. These
are taken near the source of a stream (see Figure 1 2) . Th e wands are
FIGURE 1 2. Twin Ceremony: the ritual party arrives at the river source,
"where procreative capacity begins," bearing branches of medicine
trees.
planted on either bank of the stream, opposite one another, their tips
are bent over to form an arch, and they are bound together. The
muhotuhotu wand lies on top of the mudyi wand. The complete arch
is called mpanza or kuhimpa, a verbal noun meani ng "exchangi ng. "
64
The Ritual Process
T h e muhotuhotu t r e e is us e d i n various ritual contexts. Its meaning
t ends t o be as s oci at e d b y Nd e mb u with certain of its nat ural proper-
ties, a n d al so wi t h t wo ve r bs from whi ch certain ritual experts derive
some of its r ef er ent s . T h i s ha bi t of etymologizing, as I mentioned in
Ch a pt e r I , is h i gh l y char act er i s t i c of Central African exegetics.
Wh e t h e r t h e e t y mol ogi c a l expl anat i on of t he names of ritual objects
a n d act i ons is t r u e or fal se is uni mpor t ant . Ndembu are merely utiliz-
i ng one of t he pr oces s es t h a t give richness to t he semantic content of all
l anguages homonymywhi ch may be described as a kind of serious
pun n i n g. I f t wo s i mi l a r l y soundi ng words of different derivation
can l end on e a n o t h e r c e r t a i n of their senses, semantic enrichment
is effected. H o mo n y my is exceptionally useful in ritual where, as I
have sai d, r el at i vel y few symbol s must represent a multiplicity of
ph e n ome n a .
Muhotuhotu is s o me t i me s deri ved from t he verb ku-kotumuna, which
me ans " t o fall s u d d e n l y . " I t is said t hat toward the end of t he dry
season t he l eaves of t hi s t r e e tend to fall off simultaneously, leaving
t h e boughs s udde n l y b a r e . I n t he same way, when muhotuhotu is used
as me di ci ne , di seases, mi sfort unes, and t he effects of witchcraft/
sorcery wi l l " fall of f " t h e pat i e nt t reat ed wi t h it. Whenever Ndembu
use a me di ci ne b r o o m for sweepi ng t he body wi t h pounded leaf-
medi ci ne, muhotuhotu f or ms one of its t hree components. This broom
is used mos t t ypi cal l y i n ant i -wi t chcraft ri t ual .
But t he r a di c a l -hotu- h a s anot her derivative, whi ch also influences
t he me a n i n g of muhotuhotu. Thi s is t he verb ku-hotomoka, t he sense of
whi ch was gi ven t o me i n t hi s peri phrast i c formul at i on: " a tree which
lodges on a n ot h e r t r e e falls down suddenl y when t he wind blows; its
falling is cal l ed ku-hotomoka. Somet i mes it means a tree which grows
over t he body of a n o t h e r t r ee. A disease lies on a person' s body, and
t he doct or desi res t h a t i t s houl d slip off."
I n t he specific s i t uat i on found i n Wubwang'u, however, muhotuhotu
is said t o s t and for " t h e ma n " (iyala), while t he mudyi wand stands for
" t h e w o ma n " (mumbanda). All t he adept s I have questioned agree
t hat this is t he case, poi n t i n g out t hat muhotuhotu was placed above
mudyi. Fur t h e r mor e , t he y say t hat t he tying together of t he wands
Paradoxes of Twinship in Ndembu Ritual
65
stands for their sexual union (kudisunda). Sometimes a wand of
kabalabala (Pseudolachnostylis species) wood is used instead ofmuhotuhotu.
A forked bough of this wood is frequently used as a shrine in the
hunters' cult. It is a tough, termite-resistant wood, and is compared
in the boys' circumcision ri t ual with an erect phallus. I t is used t here
as a medicine to induce male potency. Here the connection wi t h
virility is quite clear.
Another cluster of referents is associated with the form of the arch
over the stream. Its title mpanza means " t h e cr ot ch" or bifurcation
the human body. According to one informant: "Mpanza is the place
where the legs j oi n. It is t he place of the organs of reproduction in
men and women. " The same symbol appears in the girls' pubert y
ritual, where a tiny bow (kawuta) of mudyi wood is placed at the apex
of the girl novice' s seclusion hutjust where a pole of mudyi wood is
tied to a pole of red mukula wood. The bow, draped with white beads
representing children, stands for the novice' s desired fertility. The
point of j unct i on between the poles is also called mpanza. This bi-
furcation, basic to biological and social continuity, reappears in the
dualistic symbolism of twinship.
The t erm mpanza is used at boys' circumcision for a tunnel of legs
belonging to senior officiants and circumcisers, beneath which t he
junior guardians who tend the novices duri ng seclusion are obliged
to pass. This tunnel is bot h an entrance to t he situation of circumcision
and also a magical mode of strengthening the genitalia of t he j uni or
guardians. The tunnel symbolism in this ritual recalls t hat found in
Isoma.
The mpanza motif recurs in the Wubwang'u ritual itself. Duri ng the
rites performed later at the village shrine, mal e doctors plunge under
each other' s outstretched legs (see Figure 1 9, p. 77). Also the patient
herself is passed through the doctors' legs. This is called kuhanwisha
muyeji mwipanza. The Isoma tunnel, the reader may recall, was called
ikela dakuhanuka, where kuhanuka has t he same root as kuhanwisha.
So far, then, the arch stands for the fertility resulting from com-
bined masculinity and femininity. Th e siting of t he mpanza near the
source of a stream is also significant. Such a source (ntu or nsulu) is
66
The Ritual Process
said by Nd e mb u t o be " w h e r e pr ocr e at i ve capaci t y (lusemu) begins. "
Wat er is classified by r i t ua l speci al i st s in t he category of " whi t e "
symbols. As such it h a s t h e ge ne r i c senses of "goodness, " "pur i t y, "
"good l uck, " and " s t r e n g t h , " wh i c h it shares wi t h other symbols of
this class. (A funct i on of t he s e r i t es, i nformant s told me, is " t o wash
away di seases" (nyisong'u). T h e doct or s ' feet are washed "s o as to
purify t h e m" (nakuyitookesha), for t h e r e is an element of i mpuri t y in
Wubwang'u, i n its r i ba l dr y a n d aggressiveness. ) But wat er has addi-
tional senses c or r e s pondi ng t o its pecul i ar propert i es. In t hat water
is " c o o l " (atuta) or " f r e s h " (atontola), it st ands for "bei ng al i ve"
(ku-handa), as oppos ed t o t h e b u r n i n g heat of fire, which, like fever,
means " d y i n g " (ku-fwila), especi al l y dyi ng as t he result of witch-
craft. Agai n, wat er , i n t h e for m of r ai n and rivers, stands for "i n-
cr eas e" or " mu l t i p l i c a t i o n " (ku-senguka), for fertility in general.
Th e symbol i sm of mpanza i n t h e t wi ns hi p rites suggests t hat human
fertility is to be connect ed wi t h t he fertility of nat ur e.
Th e mot i f of " c o o l n e s s " is al so exemplified when t he senior
female pr act i t i oner r e move s a pi e ce of black alluvial soil (malowa)
from t he s t r eam i mme di a t e l y be l ow t he ar ch. Thi s piece is placed in
t he medi ci ne basket a n d l at e r for ms one of t he component s of the
village shri ne for t he Wubwang'u s pi r i t . Informant s say t hat t he use of
malowa her e paral l el s its us e i n t h e gi rl s' puber t y rites. Ther e malowa
st ands for mar i t al happi nes s (wuluwi), a t er m related to luwi, meani ng
" me r c y " or " ki n dn e s s . " I n ma n y ot he r contexts it is said to be used
because it is " c o o l " fr om its c on t a c t wi t h wat er. Being "cool , " it
weakens diseases, whi ch, as i n Isoma r i t ual , are t hought to be " hot . "
But it is also l i nked by i nf or mant s wi t h fertility, since crops grow
exuber ant l y i n i t .
After t he br i dal ni ght t h a t follows t he girls' puber t y ri t ual , the
novice' s i nst ruct ress (nkong'u) put s s ome malowa soil in contact with
t he br i de and t he gr oom, t h e n s cat t er s fragments of it on t he threshold
of every hut i n t he vi l l age i n h a bi t e d by a mar r i ed couple. Ndembu
say t hat this means t hat " t h e coupl e now love one anot her properly
and t he instructress wi shes t o j oi n al l t he mar r i ed peopl e in t he village
wi t h t hat s ame l ove. " Th i s n o t i o n t h a t mar r i age should ideally be
Paradoxes of Twinship in Ndembu Ritual 67
fruitfully peaceful is stated qui t e explicitly by Ndembu women.
They say t hat t he sort of husband they prefer is a good-tempered,
hard-working, and quiet-speaking man. A man like this, they say, will
"father ten chi l dren. " This ideal type, as seen by women, is the exact
opposite of t he mal e personality-type extolled in t he hunt ers' cults,
the sort of man who, it is said in a hunt ers' ritual song, "sleeps wi t h
ten women a day, and is a great thief." Indeed, women are recom-
mended in such contexts to give their hearts to these tough, quarrel-
some, and lustful men of the bush. The two antithetical ideals coexist
in Ndembu society as in our own, as any reader of the novel Gone with
the Wind will recognize. This novel, incidentally, is also based on a
dualistic t hemet hat of t he Nort h versus t he South, and of capitalism
versus landowning. Moreover not only t he fruitful union but also t he
combat of t he sexes is shown in various episodes of t he twin ritual.
Thus, the mpanza arch represents fertile, legitimate love between
man and woman. The mal e and female principles " e xchange " their
qualities, t he opposite banks of the stream are j oi ned by the arch.
The water of life flows t hrough it, and coolness and health are t he
prevailing modes.
After t he mpanza is made, t he pat i ent stands on a log placed in t he
middle of the water (see Figure 1 3 ) . Th e female adepts and their
daughters line up on t he log behind her in order of seniority. The
senior male practitioner brings the small calabash (ichimpa), openly
compared by informants with a phallus (ilomu) and of the type used
to give novices their training in sexual technique at t he girls' pubert y
ritual, and takes powdered white clay (mpemba) out of it. The mal e
doctors have previously added certain ingredients to t he white d a y -
sman portions of mpelu, or pieces of ani mal or organic mat t er, used as
ingredients of contagious magi c. In Wubwang'u these are classified
as " wh i t e " symbols and i ncl ude: powdered white portions of t he
goliath beetleused also as a charm in hunt i ng cults; some hairs
from an albino (mwabi), regarded as an auspicious bei ng; whi t e
feathers from the gray parrot (kalong'u); and white pigeon feathers
(kapompa). These are all connected with hunt i ng and masculinity as
well as with whiteness. Th e white clay itself refers qui t e explicitly
68
Paradoxes of Twinship in Ndembu Ritual
69
to semen (matekela), which, in its turn, is said to be "bl ood purified
by wat er. " Th e senior practitioner faces t he patient, puts t he white
powder in his mout h, and blows it over t he patient' s face and chest.
Next the senior female practitioner, standing just behind the patient,
takes some powdered red clay (mukundu) from t he shell of a large
water snail called nkalakala, puts it in her mout h, and blows it over t he
patient' s face and chest.
The act of blowing (ku-pumina or ku-pumbila) stands bot h for
orgasm and for blessing wi t h the good things of life (ku-kiswila nkisu).
It affords yet anot her example of t he semantic bipolarity of ritual
symbols. Th e blowing on of white, t hen red, clay dramatizes t he
Ndembu theory of procreation. My best informant, Muchona, inter-
preted t he rite as follows: " The white clay stands for semen and t he
red clay for mat ernal blood. The father first gives blood to t he mother,
who keeps it in her body and makes it grow. Semen is this blood mixed
and whitened wi t h water. It comes from t he power of the father. I t
remains in t he mot her as a seed of life" (kabubu kawumi). Muchona,
and some others, took t he view t hat both white and red clay should be
contained in t he snail's shell to represent the union of mal e and female
partners in t he conception of a child. But in each of t he performances
of Wubwang'u I witnessed, white and red clay were kept in separate
containers. What is interesting about Muchona' s view is t hat in it
he stresses t he unitive aspect of the rite.
TH E MAKING OF TH E
TWIN SH RINE IN TH E VILLAGE
Dualism prevails in t he public rite t hat follows in the patient' s village.
This is emphatically represented bot h in t he bi nary structure of the
twin shrine and in t he explicit opposition of the sexes in mi me, dance,
and song. Th e doctors ret urn from t he river bearing leafy fronds, like
a Palm Sunday processionone made up largely of women and child-
ren, however (see Figure 1 4) . Levi-Strauss would perhaps regard the
presence of t he children in t he medicine-gatheringhighly anomal -
ous in Ndembu ritualas a sign t hat children were "me di at or s "
7o
The Ritual Process
FIGURE 1 4. Twi n ceremony: the ritual party returns from the river
beari ng fronds, like a Palm Sunday procession.
bet ween t he me n a n d t he women, but Ndembu themselves look on
t hem as s ymbol s (yinjikijilu) of twinship (Wubwang'u) and fertility
(lusemu). Th e y wa n t t he m also to " b e st rengt hened, " for all who fall
wi t hi n t he a mb i t of Wubwang'u, by bi rt h or bearing, are believed to
be weakened a n d i n need of mystical invigoration.
Th e t wi ns hi p s hr i ne in t he village is constructed about five yards
i n front of t h e pat i e nt ' s hut . It is made from t he leafy branches col-
lected i n t h e bus h, one from each medi ci ne species, in t he form of a
Paradoxes of Twinship in Ndembu Ritual
7i
semicircle about a foot and a half in diameter. A partition of branches
is made across the center, dividing it into two compartments. Each
of these compart ment s is eventually filled with sets of ritual objects.
But, at different performances I witnessed, the senior officiants had
different views as to how the compart ment s should be regarded, and
this influenced the choice of objects. One school of t hought held t hat
what is called the "l eft - hand" compart ment should cont ai n: (1 ) a
foundation of black river mud (malowa) taken from under the patient' s
feet at the Rites of the River Source, " t o weaken t he shades causing
the Wubwang'u condi t i on"; (2) a black clay pot (izawu) dotted wi t h
white and red clay taken from the phallus-shaped calabash and the
shell of t he water mollusc (see Figure 1 5 ) ; and (3) in the pot, cold
water mixed with bark chips from the medicine trees (see Figures 16
and 1 7) . I n contrast, the right-hand compart ment should contain a
small calabash of sacralized honey beer (kasolu), normally a man' s and
a hunter' s drink, used as a sacred beverage in hunt ers' cults. It is
far more intoxicating t han other Ndembu beers, and its " h e a dy "
quality is considered appropri at e to the sexual jesting t hat character-
izes the rites. Honey, too, is a symbol for the pleasure of sexual
intercourse (see, for example, the song on p. 80). I n this variant, t he
left-hand compart ment is regarded as female, and t he right-hand
one as male. Each compart ment is called chipang'u, meani ng "encl o-
s ur e" or "fence, " usually surrounding a sacralized space, such as
the dwelling place and medicine hut of a chief. The pat i ent is splashed
with medicine from t he pot, while the adepts, male and female, drink
the beer together. In this form of the ritual, the mai n dualism is t hat
of sex.
But, in anot her vari ant t he one described on p. 87the left-hand
compart ment is made smaller t han the other one. Here t he opposi-
tion is between fertility and sterility. The right-hand compart ment of
chipang'u represents fertility and the beneficent and fertile shades; the
left-hand compart ment is said to be that of sterile persons (nsama)
and the shades of sterile or malevolent persons (qyikodjikodji). A large
clay pot, decorated with red and white clay as in t he first form of the
rites, is placed in the large compart ment . This is actually known as t he
"gr andmot he r " (nkaka yamumbanda), and represents t he afflicting
72
The Ritual Process
FIGURE 1 5 . Twi n ceremony: the construction of the twin shrine. The
medicine pot is decorated wi th whi te and red dabs. In the basket is a
plump cassava root, whi ch is the "food" mentioned on p. 5 3 .
shade who was once a mo t h e r of twins. The ot her compart ment is
t he interesting one for ant hr opol ogi cal i nqui ry. Ther e is an enigmatic
phrase in t he nar r at i ve of t he act ual rites (see below, p. 87): nyisoka
yachifwifwu chansama, whi ch literally means "shoot s of a bundl e of
leaves of a sterile pe r s on. " T h e t er m nsama represents a homonym,
really a sinister pun. On e sense of t he word is " a bundl e of leaves or of
Paradoxes of Twinship in Ndembu Ritual
73
FIGURE 1 6. Twi n ceremony: all the adepts' hands collectively pour
water into the medicine pot, each one adding his " strength."
grass." When a hunt er wishes to obtain honey, he climbs up a tree to
a hive (mwoma) and draws up after him on a rope a bundl e of grass or
leaves. He throws the rope over a bough, sets fire to this nsama bundl e,
and hauls it up under t he hive. It smokes furiously and the smoke
drives out the bees. The blackened remains of t he bundl e are also
called nsama. Nsama also means " a sterile or barren person, " perhaps
74
The Ritual Process
FIGURE 1 7 . Twin ceremony: the twin shrine is ready. It is obviously a
binary shrine with two compartments, wound around with the molu
waWubwang'u vine. In the left compartment is the black medicine pot,
under which can be seen the black mud. In the right is the calabash
containing sacralized honey beer, daubed with red and white clay.
in t he sense i n whi ch we speak of " a bur nt - out case. " Black is often,
but not always, t he color of sterility in Nde mbu ritual.
I n Wubwang'u, wh e n t he adept s r et ur n from the bush with their
leafy branches, t he seni or pr act i t i oner snatches leaves from t hem and
ties t hem i nt o a bundl e kn own as nsama yawayikodjikodji abulanga
kusema anyana, " t he bundl e of t he mischievous shades who fail to bear
chi l dren, " or nsama for s hor t . Th e n this chimbuki (doctor) takes a
calabash cup {chikashi or lupanda) of mai ze or kaffir corn beer and
pours it on t he nsama as a l i bat i on, saying, "All you shades without
Paradoxes of Twinship in Ndembu Ritual
75
children, here is your beer. You cannot drink the beer t hat is already
poured into this big pot " (in t he right-hand compart ment ). " Th a t
is t he beer for t he shades who bore chi l dren. " He then put s the piece
of black river mud in the chipang'u and lays the nsama bundl e on top
of it. The malowa black clay is said " t o weaken the shades causing
disease."
Another difference between t he two forms of the chipang'u " e n -
closure" is t hat in t he one stressing sexual dualism, an arrow is
inserted behi nd t he pot in t he left-hand compart ment , point down-
ward (see Fi gure 1 8) . Thi s arrow stands for the patient' s husband.
Arrows wi t h this meani ng appear in several Ndembu rituals, and
the name for bridewealth paid by t he husband is nsewu, " ar r ow. "
In the rites stressing t he dichotomy between fertility and sterility
the arrow is not employed. I n t he latter there seems to be an equation
made between sterility and twinship, for twins frequently di e; too
much is t he same as too little. I n bot h types, however, the molu
waWubwang'u river creeper is woven laterally t hrough t he vertical
leafy branches of t he shrine.
The pat i ent is made to sit on a mat before this shrine, and over her
shoulders are draped vines ormolu waWubwang'u, to give her fecundity
and especially a good supply of milk (see Figure 1 9) . She is then
steadily splashed with medicine while what I will call t he "ri t es of
the fruitful contest of t he sexes" rage hilariously in the danci ng place
between t he shrine and t he patient' s hut . It is considered appropri at e
if pieces of medicine leaves are seen to adhere to her skin. These are
yijikijilu, or "symbol s, " of the Wubwang'u manifestation of t he shades.
They make the shade in this twinship form "vi si bl e" to all, though
transubstantiated into leaves.
TH E FRUITFUL CONTEST OF TH E SEXES
The next aspect of Wubwang'u to which I would like to call attention
is the cross-sexual joking t hat marks two of its phases. Here we have
an expression of t he " t wi n " paradox as a joke or, as Ndembu say,
Paradoxes of Twinship in Ndembu Ritual
77
FIGURE 19.Twin ceremony: the patient's shoulders are draped with
molu waWubwang'u vines, to give her fecundity and a good supply of
milk. Here a male doctor can be seen plunging under another doctor's
legs, to give sexual strength (see pp. 65 and gi ).
" a joking rel at i onshi p" (wusensi). The specific reference of the rites
is to t he division of humani t y into men and women, and to t he arousal
of sexual desire by stressing t he difference between t hem in t he form of
antagonistic behavior. The shades of t he dead, in t hat they are
believed to give their names and personal characteristics to infants of
both sexes, and in a certain sense to be reborn in them, in a way have
78
The Ritual Process
no sex. It is t hei r gener i c h uma n i t y t hat is stressed, or perhaps their
bisexuality. But t he l i vi ng a r e differentiated by sex, and sex differences
are, as Gl uckman ( 1 95 5 ) wr i t es , "exagger at ed by cus t om" (p. 61 ) .
I n Wubwang'u, Nd e mb u a r e obsessed by t he hilarious contradiction
t hat t he mor e t he sexes stress t hei r differences and mut ual aggression,
t he mor e do t hey desi r e sexual congress. They sing ribald and
Rabel ai si an songs dur i n g t h e collection of " medi ci nes" in t he bush
and t oward t he end of t h e publ i c dance, while t he pat i ent is being
sprinkled wi t h t hose medi ci nes , some of whi ch emphasize sexual
conflict and some of whi ch a r e di t hyr ambs i n praise of sexual union,
frequently specified as adul t er ous . These songs are believed to
" s t r e ngt he n" (ku-kolesha) b o t h t he medicines and t he pat i ent . They
are also believed t o ma k e t h e at t enders strong, bot h sexually and
bodily.
First, before si ngi ng t h e r i bal d songs, Nde mbu chant a special
formula, "kaikaya wo, kakwawu weleli" ( "her e anot her t hi ng is
done " ) , whi ch has t he effect of legitimizing t he ment i on of matters
t hat otherwise woul d be wh a t t hey call " a secret t hi ng of shame or
modesty " (chuma chakujinda chansonyi). Th e same formula is repeat ed
i n legal cases concer ni ng s uch mat t ers as adul t ery and breaches of
exogamy, wher e sisters a n d daught er s or in-laws (aku) of t he plaintiffs
and defendants ar e pr es ent . Nde mbu have a cust omary phrase
explaining Wubwang'u songs. " Th i s singing is wi t hout shame because
shamelessness is [a char act er i st i c] of t he curat i ve t reat ment of
Wubwang'u" (kamina kakadi nsonyi mulong'a kaWubwang'u kakuuka
nachu nsonyi kwosi). I n brief, Wubwang'u is an occasion of licensed
disrespect and prescri bed i mmodest y. But no sexual promiscuity is
displayed i n act ual be havi or ; i ndecency is expressed by word and
gesture only.
Th e songs, at bot h phases, ar e i n serial order. First, members of
each sex belittle t he sexual organs and prowess of members of the
opposite sex, and extol t hei r own. Th e wome n j eeri ngl y assert to their
husbands t hat t hey have secret lovers, and t he me n ret ort t hat all they
get from t he women ar e vener eal diseases, a consequence of adultery.
Afterward bot h sexes pr ai se i n lyrical t er ms t he pleasures of inter-
course as such. Th e whol e at mos pher e is buoyant and aggressively
Paradoxes of Twinship in Ndembu Ritual
79
jovial, as men and women strive to shout one anot her down (see
Figure 20 ). The singing is thought to please the strong and merry
Wubwang'u shade-manifestation.
Nqfuma mwifundi kumwemweta,
" I am going away to teach her how to smile,
lyayi lelu iyayi kumwemweta.
Your mother, today, your mother how to smile.
Kakweji nafu namweki,
The moon which has gone appears,
FIGURE ao. Twi n ceremony: the men and women cheerfully revile one
another, to vocally symbolize the fruitful contest of the sexes.
The Ritual Process
Namoni iyala hakumwemweta.
I have seen the man on whom to smile.
Eye iyayi eye!
Mother!
Twaya sunda kushiya nyisong'a,
Come and copulate to leave diseases,
Lelu tola mwitaku mwazowa.
Today look at a wet vulva
Nyelomu eyeye, nyelomu!
Mother of penis! Mother of penis!
Yeyuwamuzang'isha.
That will give you much pleasure.
Nashinkaku. Nashinki dehi.
I do not close. I have closed already.
Wasemang'ayami wayisema,
You are giving birth, I am the one who gives birth
Nimbuyi yami.
I am the elder of the twins.
Mwitaku mweneni dalomu kanyanya,
A large vulva, a small penis,
Tola mwitaku neyi mwihama dachimbu,
Look, a vulva as on a lion's brow,
Nafumahu ami ng'ang'ayanyisunda.
I am going away, I, a veritable witchdoctor of copulation.
Kamushindi ilomu,
I will rub your penis,
Yowu iyayi,yowu iyayi\
Mother, O mother!
Mpang'ayeyiyobolang'a chalala.
Your swollen scrotum stimulates the vulva indeed.
Mwitaku wakola nilomu dakola,
A strong vulva and a strong penis,
Komanayowana neyi matahu, wuchi wawutowala sunjiyakila.
H ow it tickles like grass! Copulati on is like sweet honey.
Ilomu yatwahandang 'a,
The penis is making me strong,
Eyi welili neyi wqyobolang'a,iwu mutong'a winzeshimu.
You did something when you played with my vulva, here
the basket, fill it."
Paradoxes of Twinship in Ndembu Ritual 81
CROSS- SEXUAL AND CROSS- COUSIN JOKING
What is conspicuous is the perfect equality between the sexes in this
jesting and mut ual "fl yt i ng"t o borrow a term from the Scottish
Chaucerian poets for competitive lampooning verses. There is no
hint that this is a "r i t ual of rebel l i on" in Gluckman' s (1954) sense.
What is represented in Wubwang'u seems rat her to be associated with
the conflict between virilocality, which interlinks male kin together
and expels female kin from their natal villages, and matriliny, which
asserts the ul t i mat e structural paramount cy of descent through
women. These principles are fairly evenly balanced in secular life,
as I have suggested inSchism and Continuity in an African Society (1 95 7).
Ndembu explicitly connect Wubwang'u joking with t he customary
joking between cross-cousins. Both kinds are called wusensi, and bot h
involve an element of sexual repartee.
The i mport ance of cross-cousinship (wusonyi) in Ndembu society
derives in great par t from the opposition between virilocality and
matriliny. For villages tend to contain almost half as many children
as sisters' children of men of the senior generation of matrilineal kin
(Turner, 1957, Tabl e 1 0, p. 71 ) . These are grouped together as
members of a single genealogical generation in opposition to t he
senior adjacent generation. But cross-cousins are also divided from
one anot her: children of mal e villagers compete with their cross-
cousins for their fathers' favors and attentions. Virilocality in a society
with matrilineal descent also gives an individual two villages in which
he has strong legitimate claims to reside, those respectively of his
father's and his mother' s kin. In practice, many men are torn between
competing loyalties to one or the other, to the father' s or to the
mother' s side. Yet, as t he child of his father and mother, each man
represents t he union of bot h.
I consider t hat the approxi mat e equality of ties t hrough t he mal e
and female sides in Ndembu society, with neither set regarded as
axiomatically domi nant , is symbolized in Wubwang'u by t he ritual
opposition between men and women. Cross-cousinship is the kinship
bond t hat most fully expresses the fruitful tension between these
principles, for it expresses the residential unity of matrilineally
82
The Ritual Process
and pat ri l at eral l y l i nked ki n. Cross-cousins of opposite sex are en-
couraged to mar r y, a nd, before marri age, may indulge in love play
and ribald j oki ng wi t h one anot her. For marri age produces a
t emporary unity of t he sexes, whose differences, stereotyped and
exaggerated by cust om, have become associated with equal and
opposed principles of social organi zat i on. Hence, it is not inconsistent
with t he Nde mbu way of l ooki ng at things t hat they compare Wub-
wang'u cross-sexual j oki ng wi t h cross-cousin j oki ng. Wubwang'u too,
for all its ri bal dry, cel ebr at es t he institution of marri agei n the
symbolism of t he mpanza ar ch, and of an arrow representing the
husband, inserted i n t he chipang'u shrine. Thi s arrow stands for the
patient' s hus band. I n t he gi rl s' puber t y ritual, an arrow placed in
t he mudyi tree symbolizes t he bri degroom, and indeed the term for
t he mai n mar r i age p a y me n t is nsewu, which means " ar r ow. " The
procreative urge is domes t i cat ed into t he service of society t hrough
t he institution of ma r r i a ge ; t hat is what the symbolism suggests. And
marri age bet ween cross-cousins, bot h mat ri l at eral and patrilateral,
is t he preferred form.
COMPETITION FOR RESIDENTIAL FILIATION
BETWEEN MATRI LI NY AND VIRILOCALITY
Ndembu society, t o r epeat , is regul at ed by two residential principles
of almost equal s t r e ngt h: mat ri l i neal descent and virilocality-
patrilocality. These pr i nci pl es t end to become competitive rat her
t han coadapt i ve, as I have ar gued in Schism and Continuity (1 95 7),
and this is par t l y owi ng t o ecological reasons. Ndembu grow a staple
crop, cassava, whi ch flourishes on many kinds of soil, and hunt forest
animals widely di st r i but ed t hr ough t hei r territory. They do not keep
cattle, and men at t ach hi gh val ue to hunt i ng, whi ch can be carried
on all over Nde mbu count r y. Wat e r is available everywhere. There
is not hi ng to pi n down popul at i ons t o limited tracts of territory.
Given t he existence of t wo maj or modes of filiation, t here is no
ecological wei ght i ng i n favor of either principle. It is where an
Paradoxes of Twinskip in Ndembu Ritual
83
African communi t y is anchored to limited tracts of fertile land
or can exploit only a single category of movable resources (such as
large livestock), t hat one tends to find the regular paramount cy
in many fields of activity of a single kinship principle of organiza-
tion: patriliny or matriliny. Under Ndembu ecological conditions,
residential filiation t hrough male links (husband and father) com-
petes freely with matriliny. At one moment a given village may
exhibit in its residential composition the dominance of one mode,
and, at anot her moment , of the other.
I believe t hat this structural competition between major principles
of residential filiation is a crucial factor in accounting for (1 ) the
way Ndembu treat twins, and (2) their conceptualization of duality,
in terms not of a pai r of similars but of a pai r of opposites. The unity
of such a pai r is t hat of a tensed unity or Gestalt, whose tension is
constituted by ineradicable forces or realities, implacably opposed,
and whose nat ure as a uni t is constituted and bounded by t he very
forces t hat contend within it. If these mutually involved irrepress-
ibles belong together in a human being or a social group, they can
also constitute strong unities, the more so if bot h principles or
protagonists in t he conflict are consciously recognized and accepted.
These are self-generated nat ural unities, to be distinguished from
the arbitrary flat unities t hat can be externally reduplicated. But,
they are also not quite like the dialectical pairs of opposites of Hegel
or Marx, of which one part y, after mastering the other, gives rise
to new contradictions within itself. Given t he persistence of Ndembu
ecology, the parties to this tensed unity belong together and, in their
very opposition, frame it, constitute it. They do not break each
other down; in a way they provoke each other, as in symbolic form
the mutually t aunt i ng sexes do in Wubwang'u. Only socioeconomic
change can break this kind of social Gestalt.
In Schism and Continuity I tried to analyze various aspects of this
kind of uni t y: matriliny versus virilocality; the ambitious individual
versus the wider interlinking of matrilineal ki n; the elementary
family versus the uterine sibling group, an opposition t hat may also
be seen in terms of tension between patrifilial and matrilineal
84
The Ritual Process
principles; t he forwardness of yout h versus the domineering elders;
status-seeking versus responsibility; sorceri smi . e. , hostile feelings,
grudges, and intriguesversus fri endl y respect for others, etc. All
these forces and principles can be cont ai ned wi t hi n Ndembu uni t y;
they belong to it, they color it, t hey a r e it. What cannot be con-
tained are modern pressures and t he ma ki ng of money.
What happens, t hen, in t he course of t he Wubwang'u ritual ? The
opposing principles are not pe r ma ne nt l y reconciled or blended.
How can they be while Nde mbu r e ma i n at t he level of technology
and wi t h t he specific ecology I have descri bed? But, instead of
coming against one anot her i n t he bl i nd ant agoni sm of material
interest, "seei ng not hi ng but t hems el ves , " as it were, they are re-
instituted against one anot her in t he t r anscendant , conscious, re-
cognizant uni t y of Ndembu society whos e principles they are. And
so, in a sense, for a time, they act ual l y become a play of forces
1
instead
of a bitter bat t l e. The effects of such a " p l a y " soon wear off, but
t he sting is temporarily removed from cer t ai n troubled relationships.
TWINSH IP AS MYSTERY AND ABSURDITY
Th e ritual episodes I have discussed, t hough only superficially
t he Rites of t he River Source, and t he Doubl e Shri ne wi t h t he Fruit-
ful Contest of t he Sexesrelate t o t wo aspects of t he par adox of
twinship. Th e first is to be found in t he fact t hat t he not i on 2 = 1
may be regarded as a mystery. I ndeed, t he Nde mbu characterize
t he first episode by a t er m t hat l argel y conveys this sense. Thi s is
mpang'u, whi ch is applied to t he cent r al and most esoteric episode
of a ritual. The same word also means " a secret saying or pass-
wor d, " such as is used by novices and t hei r guar di ans i n t he cir-
cumcision lodge. Th e rites by t he s t r eam source ar e as much a
religious mystery as those of t he anci ent Greeks and Romans or of
1
My sister-in-law, Mrs. Helen Barnard, of Wellington University, New Zealand,
has pointed out to me how similar this viewpoint is to the Hindu notion of a Ma.
Paradoxes of Twinship in Ndemba Rituals
85
modern Christians, since they relate to hidden or inexplicable
matters beyond human reason. The second aspect is the Ndembu
feeling t hat 2 = 1 is an absurdity, a huge and even brutal joke. So
much of their ritual is devoted to the procurement of fertility of
various sorts, yet the mot her of twins has been endowed with too
much of it at one time.
What is interesting about both the mystery and the absurdity
of twinship here is that Ndembu, in the ritual of Wubwang'u, have
elected to exhibit the major sets of complementary and antithetical
dyads recognized in their culture. Yet, in its aspect of mystery,
there is also the clear emergence of the sacred color-triangle white-
red-black (see Turner, 1967, pp. 69-81 ) . These colors constitute,
for Ndembu, classificatory rubrics under which a hierarchy of ritual
objects, persons, activities, episodes, gestures, events, ideas, and
values are assembled and arrayed. At the river source the white clay
and the red clay are brought into conjunction with cool black river
mud, the ensemble being interpreted to mean the union of the sexes
in peaceful, fruitful marri age. But, clearly, the triangle, from its
appearance in other, more complex and basic rituals, notably
those of life crisis, has a deeper significance than this situational
specification within its total semantic wealth. It represents t he whole
cosmic and social order recognized by Ndembu in its harmony and
balance, wherein all empirical contradictions are mystically resolved.
The disturbance brought about by the Wubwang'u manifestation
of the shades is here ritually countered by a portrayal of quintessen-
tial order, a port rayal t hat is believed to have efficacy and is not a
mere assemblage of cognitive signs.
Wubwang'u is a ritual t hat moves regularly from the expression of
jocose disturbance to that of cosmic order and back to di st urbance
to be finally resolved by t he transfer of the patient to partial seclusion
from secular life until the dangerous condition has been removed
from her. This oscillation is to some extent homologous with the
processual structure of Isoma. But t he major difference between these
rites is the constant emphasis in Wubwang'u on opposition between
the sexes and the social principles of filiation derived from the parents
86
The Ritual Process
of opposite sex. I n Isoma, t he sex dyad was s ubor di nat ed t o t he life/
deat h antithesis. I n Wubwang'u, sexual opposi t i on is t he ma i n t heme.
AN NDEMBU VIEW OF WUBWANG' U
I fear t hat I have not as yet allowed t he Nde mbu t o speak sufficiently
for themselves about t he meani ng of Wubwang'u. T o gi ve their
"i nsi de vi ew, " and to enabl e t he reader to compar e t hei r inter-
pret at i on with mine, I will translate comment s I r ecor ded from
Wubwang'u adepts either duri ng act ual performances of t he rites
or shortly afterward in informal discussions.
I will begin with a succinct account of t he whol e pr ocedur e as
related by an experienced mal e doct or:
Neyi nkakayindi wavwalili ampamba,
" If her [patient's] grandmother gave birth to twins,
neyi nkakayindi nqfwi dehi
and if her grandmother is already dead,
chakuyawu nakuhong'a kutiya mukwakuhong'a
when they go to divine the diviner answers
nindi nkakayeyi diyi wudi naWubwang'u
and says: ' Your grandmother is the one who has Wubwang'u,
diyi wunakukwati nakutwali.
she is the one who has caught you,
kulusemu IwaWubwang'u
who has brought you to the reproductive state of Wubwang'u,
dichu chochina hikukeng'a walwa
and so, therefore, she desires beer
nakumwimbila ng'omaya Wubwang'u
for the playi ng of the drums [or dance] of Wubwang'u.
Neyi wudinevumu akumujilika hakuvwala chachiwahi.
If you have a womb [e.g., are pregnant], she forbids you to
give birth well.
Neyi eyi navwali dehi chachiwahi
If you have already gi ven birth well,
kunyamuna mazu amakulu
there [must be] a renewal and scattering of those former words
Paradoxes of Twinship in Ndembu Ritual
hikuyimwang'a hikuteta acheng'i
and a cutting [of medicines] [i.e., the rites must be
performed again]
nakuwelishamu mwana mukeki.
in order that the baby may be washed in [them],
Neyi nawa aha mumbanda navwali ampamba
Sometimes when a woman has borne twins
akuya ninyana mwisang'a
they will go with children into the bush
nakumukunjika kunyitonduyakumutwala kumeji
and stand her beside the trees and take her through to the water
nakusenda nyolu
and carry vines [of the molu waWubwang'u creeper]
yakupakata nakukosa mamayawu
for draping [over and under her arms] and wash their mother
ninyana hamu hikutwala anyana ku mukala.
and the children in just the same wayand convey the
children to the village.
Kushika kuna ku mukala
When they arrive there at the village,
hikutung'a chipang'u kunonayilumbu
they construct a [small] enclosure [for a shrine] and pick up
medicines
hikusha mu mazawu izawu dimu danyanya dakusha
and put them in medicine troughs [or clay pots]one small
trough [or pot]
nyisoka yachifwifu chansarna
for green shoots from a bundle of leaves for a sterile person
hikwinka muchipang'u china chanyanya
they put in that small enclosure,
hikunona izawu hikwinka mu chipang'u cheneni.
they take [another] medicine trough and put it in the large
enclosure.
Akwawu anading'i nakuhang'ana nanyoli
Others were dancing with creepers,
asubolang'a nyoli nakutenteka mu chipang'u.
they strip off the creepers and put them away in the enclosure.
88
The Ritual Process
Kushalayemweni imbe-e hakuwelisha anyana hamu
They remain there themselves singing and wash the children
[with medicine]
nakuhilisha munyendu;
and pass them under [their] legs;
chikukwila namelele hikuyihang'a;
this is done in the late afternoon, when they chase them;
mwakukama nawufuku kunamani.
when they sleep at night it is all over.
Mqfuku ejima anyana ching'a kuyiwelisha mu mazawu,
Every day they must wash the children [with medicine] in
the troughs,
hefuku hefuku diku kukula kwawanyana ampamba
day after day until the twins grow up."
Commentary
Thi s account is Wubwang'u in a nutshell, But, nat ural l y, it leaves out
many of those fascinating details t hat for anthropologists const i t ut e
t he major clues to a culture' s pri vat e universe. It makes clear t hat
t he afflicting shade in Wubwang'u is typically a deceased mot her
of twins (nyampasa). She was herself a member of t he cult, for i n
Ndembu ritual t hought , as I have not ed, only a deceased cult-
member can afflict t he living in t he mode of manifestation t reat ed
by t hat cult. Again, t he text makes pl ai n t hat affliction is in t he
matrilineal descent line. However, glosses by other informants
insist t hat a mal e shade can " c ome t hrough in Wubwang'u" if he
was a father of twins (sampasa) or a twin himself. I have never re-
corded a single instance of this, however. Wubwang'u is not t hought
of as an i ndependent spirit but as t heway in which an ancest ral shade
makes its displeasure wi t h t he living known.
According to ot her informants, it is " t h e women who explain
to t he men t he medicines and curative techniques of Wubwang'u."
One doctor' s sister t aught hi m; she was a nyampasa, a mot her of
twins. He went on to say t hat bot h twins had di edand, i ndeed,
Paradoxes of Twinship in Ndembu Ritual 89
it is very common for one or both to die, for Ndembu say that a
mother will either favor one with milk and food supplies and neglect
the other, or try to feed both equally on a supply that is sufficient for
only one. Twi ns are known by special t erms: the elder is mbuya, the
younger kapa. The child following them in birth order is called
chikomba, and it is his dut y to play the ritual drums at a performance
of Wubwang'u. Often the rites are performed for chikomba and his
mother when he is a toddler, to " make him st rong. " A chikomba
can also become a Wubwang'u doctor. Although men learn the
medicines from female adepts in the cult, they become the principal
doctors and masters of ceremonies. One sign of their status is the
double hunting-bell (mpwambu), which once more represents the
duality of twinship.
TH E H OPPING WITH TH E ARROW
The conclusion of the rites further emphasizes the sexual division.
At sundown, the senior practitioner takes the winnowing basket,
which has been laid on the pot in the "femal e" compart ment , puts
it on the patient' s head, then raises and lowers it several times. Then
he puts the remaining ritual equipment on the basket and holds the
ensemble aloft. He next takes the arrow and places it between his
big toe and second toe and invites the patient to hold his waist.
The pai r then hop on their right legs straight to the patient' s hut .
Two hours later she is taken out and washed with what remains
of the medicine in the clay pot, or medicine trough.
I conclude this description of the rites of the Double Shrine with
a text that describes in full the episode of hopping with the arrow:
Imu mumuchidika.
" Thi s is what is in the ritual.
Neyi chidika chaWubwang'u chinamani dehi namelele
When the ritual of Wubwang'u is already finished in the late afternoon
chimbuki wukunona nsewu
the doctor takes the arrow
9
The Ritual Process
wukwinka mumpasakanyijanyinuyakumwendu wachimunswa.
and puts it in the cleavage of the toes of the left foot.
Muyeji wukwinza wukumukwata nakumukwata mumaya.
The patient comes and catches him around the wai st.
Chimbuki neyi wukweti mfumwindi
If the doctor should catch her husband
mumbanda wukumukwata mjummindi mumaya
the woman will clasp her husband around the wai st
hiyakuya kanzonkwela mwitala
and they will go hopping into the hut
nakuhanuka munyenduyawakwawu adi muchisu.
and they will pass through the legs of other people who are in the
doorway.
lyala ning'odindi akusenda wuta ninsewu mwitala dawu.
The man and his wife will carry a bow and arrow i nto their hut.
Chimbanda wayihoshang'a
The doctor says to them:
nindi mulimbamulimba
' Get into the kraal [as a man says to his sheep or goats],
ing'ilenu mwitala denu ing'ilenu mwitala
enter your hut, enter your hut.'
Chakwing'ilawu anlu ejima hiyakudiyila kwawu kunyikala yawu.
When they went in, all the people went away to their own villages.
Tunamanishi.
We have finished."
Commentary
It is wort h noting t hat t he t erm for "bet ween t he t oes , " mumpasakanyi,
is etymologically connected with t he t er m mpasa, t he r i t ual wor d for
" t wi ns . " I n Ndembu ri t ual generally, t he ar r ow st ands for t he ma n
or husband and is held in t he ri ght hand, whi l e t he bow represent s
t he woman and is held in t he left hand. Bow a n d ar r ow t oget her
symbolize marri age. " T o h o p " (kuzonkwela) st ands for sexual i nt er-
course, and has this meani ng in t he boys' ci rcumci si on ri t es, when
t he novices are forced t o hop on one leg as par t of t hei r discipline
Paradoxes of Twinskip in Ndembu Ritual 91
during seclusionr I n Wubwang'u the doctor and t he pat i ent hop on
their right legs, for the right is t he side of strength. The phrase
"mulimbamulimba" is shouted at domestic animals when they are
herded into their kraals at night. It signifies the bestial aspect of
twinship, which, as a mode of multiple bi rt h, is considered more
appropri at e to animals t han men. The tunnel of legs made by t he
adepts under which t he father and mot her of twins must pass
resembles t hat at t he circumcision rites under which t he j uni or
guardians of t he novices must pass. Thi s tunnel, as we have seen,
is made by the senior men in Mufcanda, and signifies (1 ) sexual
strength for the j uni or guardians passing beneat h it, and (2) t he
rite de passage from j uni ori t y to seniority. In Wubwang'u it appears to
mean, by homology, the incorporation of t he parent s of twins into
the cult association of Wubwang'u into which they are born from
the bodies of t he adepts.
Conclusion
1 . Forms of Duality
The ritual of twinship among t he Ndembu throws into high
relief most types of duality recognized by t he Ndembu. The cleavage
between men and women, the opposition of mean and private
grudgery and social feeling, and between sterility and fruitfulness,
are shared by Wubwang'u and Isoma. But Wubwang'u has certain
special features of its own. It exhibits fully the animality and t he
humani t y of sex, in t he forms of excessive production of children,
as juxtaposed to the mystery of marri age, which unites dissimilars
and curbs excess. The couple are at once praised for their excep-
tional contribution to society and cursed for their excess in so doing.
At the same time, t he deep contradiction between matrilineal de-
scent and patrilaterality emerge in the boisterous joking relationship
between t he sexes, which is explicitly compared to t he joking re-
lationship between cross-cousins. Ther e is a strong strain, moreover,
92
The Ritual Process
of egalitarianism in t he ri t es; t he sexes are por t r aye d as equal
t hough opposed. This equality exposes s omet hi ng profound in the
nat ur e of all social systemsan idea I develop mo r e fully in Chapt er
Thr ee. An event, such as t wi nni ng, t hat falls out s i de t he orthodox
classifications of society is, paradoxically, ma de t h e ri t ual occasion
for an exhibition of values t hat rel at e t o t he c ommun i t y as a whole,
as a homogeneous, unst ruct ured uni t y t hat t r ans cends its differen-
tiations and contradictions. Thi s t heme, of t he dual i s m between
" s t r uct ur e " and " communi t as , " and t hei r ul t i ma t e resolution in
"soci et as, " seen as process r at her t hen timeless ent i t y, dominates
t he next t hree chapt ers of this book.
2 . Prescribed Obscenity
I t woul d be appr opr i at e here to ment i on a n i mpor t ant and
unjustly neglected paper by Professor Evans - Pr i t char d, " Some
Collective Expressions of Obsceni t y in Africa, " r ecent l y republished
in his collection of essays The Position of Women in Primitive Society
(1 965 a). Thi s paper makes t he following poi nt s :
(1) There are certain types of obscene behavi or [i n Afri can society] the
expression of whi ch is always collective. These are usually prohibited, but
are permitted or prescribed on certain occasions;
(2) These occasions are all of social i mportance, and fall roughly under
two headings, Religious Ceremonies and Joi nt Economi c Undertaki ngs
(p. 1 01 ) .
He explains t he obscenity as follows:
(1) The wi thdrawal by society of its normal prohibitions gives special
emphasis to the social value of the acti vi ty;
(2) It also canalizes human emotion into prescri bed channels of expres-
sion at periods of human crisis (p. 1 01 ) .
Wubwang'u falls neat l y i nt o this cat egory of rites of prescri bed
and stereotyped obscenity, al t hough it cont ai ns cruci al episodes
extolling marri age, whose net work of rel at i onshi ps is charact eri st i c-
ally inhibitory of expressions of obscenity. Wh a t we ar e confronted
wi t h i n t he twinship rites is i n fact a domest i cat i on of t hose wi l d
Paradoxes of Twinship in Ndembu Ritual
93
impulses, sexual and aggressive, which Ndembu believe are shared
by men and animals. The raw energies released in overt symbolisms
of sexuality and hostility between t he sexes are channeled toward
master symbols representative of structural order, and values and
virtues on which that order depends. Every opposition is overcome
or transcended in a recovered unity, a unity that, moreover, is rein-
forced by t he very potencies t hat endanger it. One aspect of ritual
is shown by these rites to be a means of put t i ng at the service of the
social order t he very forces of disorder t hat inhere in man' s mam-
malian constitution. Biology and structure are put in right relation
by the activation of an ordered succession of symbols, which have
t he twin functions of communication and efficacy.
Communitas
FORM AND ATTRIBUTES
OF RITES OF PASSAGE
I n this Chapt er I take up a t heme I have discussed briefly elsewhere
(Turner, 1967, pp. 9 3 - 1 1 1 ) , note some of its variations, and consider
some of its further implications for t he study of culture and society.
This t heme is in the first place represented by the nat ure and char-
acteristics of what Arnold van Gennep (1909) has called the "l i mi nal
phas e " of rites de passage. Van Gennep himself defined rites de passage
as "ri t es which accompany every change of place, state, social posi-
tion and age. " To point up the contrast between " s t a t e " and " t r an-
sition, " I employ " s t a t e " to include all his other terms. It is a more
inclusive concept t han " s t at us " or "office," and refers to any type
of stable or recurrent condition t hat is culturally recognized. Van
Gennep has shown t hat all rites of passage or "t r ans i t i on" are
marked by t hree phases: separation, margin (or limen, signifying
"t hr es hol d" in Lat i n) , and aggregation. The first phase (of separa-
tion) comprises symbolic behavior signifying the det achment of the
individual or group either from an earlier fixed point in the social
structure, from a set of cultural conditions (a " s t at e " ) , or from bot h.
Duri ng the intervening " l i mi nal " period, the characteristics of t he
ritual subject (the "passenger ") are ambi guous; he passes t hrough a
cultural realm t hat has few or none of t he attributes of the past or
coming state. I n t he third phase (reaggregation or reincorporation),
Liminality and Communitas
the passage is consummat ed. The ritual subject, individual or cor-
porat e, is in a relatively stable state once more and, by virtue of this,
has rights and obligations vis-a-vis others of a clearly defined and
"s t r uct ur al " t ype; he is expected to behave in accordance with
certain customary norms and ethical standards binding on incum-
bents of social position in a system of such positions.
Liminality
The attributes of liminality or of liminal personae (" threshold people ")
are necessarily ambiguous, since this condition and these persons
elude or slip through the network of classifications t hat normally
locate states and positions in cultural space. Liminal entities are
neither here nor t here; they are betwixt and between the positions
assigned and arrayed by law, custom, convention, and ceremonial.
As such, their ambiguous and i ndet ermi nat e attributes are expressed
by a rich variety of symbols in the many societies t hat ritualize social
and cultural transitions. Thus, liminality is frequently likened to
death, to being in the womb, to invisibility, to darkness, to bisexu-
ality, to t he wilderness, and to an eclipse of the sun or moon.
Liminal entities, such as neophytes in initiation or pubert y rites,
may be represented as possessing nothing. They may be disguised as
monsters, wear only a strip of clothing, or even go naked, to demon-
strate t hat as liminal beings they have no status, property, insignia,
secular clothing indicating rank or role, position in a kinship system
in short, nothing t hat may distinguish t hem from their fellow
neophytes or initiands. Thei r behavior is normally passive or humbl e;
they must obey their instructors implicitly, and accept arbitrary^
punishment without complaint. It is as though they are being re-
duced or ground down to a uniform condition to be fashioned anew
and endowed with additional powers to enable t hem to cope with
their new station in life. Among themselves, neophytes tend to develop
an intense comradeship and egalitarianism. Secular distinctions of
rank and status disappear or are homogenized. The condition
96
The Ritual Process
of the patient and her husband in Isoma had some of these at t r i -
butespassivity, humility, near-nakednessin a symbolic mi l i eu t hat
represented both a grave and a womb. I n initiations wi t h a l ong
period of seclusion, such as the circumcision rites of many t r i bal
societies or induction into secret societies, there is often a ri ch pr o-
liferation of liminal symbols.
Communitas
What is interesting about liminal phenomena for our present pur -
poses is the blend they offer of lowliness and sacredness, of homo-
geneity and comradeship. We are presented, in such rites, wi t h a
"mome nt in and out of t i me, " and in and out of secular social
structure, which reveals, however fleetingly, some recogni t i on (in
symbol if not always in language) of a generalized social bond t hat
has ceased to be and has simultaneously yet to be fragment ed i nt o a
multiplicity of structural ties. These are the ties organi zed in t er ms
either of caste, class, or rank hierarchies or of segment ary oppositions
in t he stateless societies beloved of political anthropologists. It is as
though there are here two major " mode l s " for human i nt errel at ed-
ness, j uxt aposed and alternating. The first is of society as a st r uc-
tured, differentiated, and often hierarchical system of politico-legal-
economic positions wi t h many types of evaluation, separat i ng me n
in terms of " mo r e " or " l e s s . " The second, which emerges recogniz-
ably in t he liminal period, is of society as an unst ruct ured or r udi -
mentarily structured and relatively undifferentiated comitatus, com-
muni t y, or even communi on of equal individuals who submi t t o-
gether to the general aut hori t y of the ritual elders.
I prefer the Lat i n term " communi t as " to " communi t y, " to dis-
tinguish this modal i t y of social relationship from an " ar ea of common
living. " Th e distinction between structure and communi t as is not
simply t he familiar one between "s ecul ar " and "s acr e d, " or t hat ,
for example, bet ween politics and religion. Cert ai n fixed offices in
tribal societies have many sacred at t ri but es; i ndeed, every social
Liminality and Communitas
97
position has some sacred characteristics. But this "sacred " component
is acquired by the incumbents of positions during the rites de passage,
through which they changed positions. Something of the sacredness
of that transient humility and modelessness goes over, and tempers
the pride of the i ncumbent of a higher position or office. This is not
simply, as Fortes (1962, p. 86) has cogently argued, a mat t er of
giving a general st amp of legitimacy to a society's structural positions.
It is rat her a mat t er of giving recognition to an essential and generic
human bond, wi t hout whi ch there could be no society. Liminality
implies t hat the high could not be high unless the low existed, and
he who is high must experience what it is like to be low. No doubt
something of this thinking, a few years ago, lay behind Prince
Philip' s decision to send his son, the heir apparent to the British
throne, to a bush school in Australia for a time, where he could
learn how " t o rough i t . "
Dialectic of the Developmental Cycle
From all this I infer t hat , for individuals and groups, social life is a
type of dialectical process t hat involves successive experience of high
and low, communi t as and structure, homogeneity and differentia-
tion, equality and inequality. The passage from lower to higher
status is t hrough a ^imbo.. of-statuslessness_ In such a process, the
opposites, as it were, constitute one anot her and are mutually indis-
pensable. Furt hermore, since any concrete tribal society is made up
of multiple personae, groups, and categories, each of which has its
own developmental cycle, at a given moment many incumbencies of
fixed positions coexist with many passages between positions. In
other words, each individual' s life experience contains alternating ex-
posure to structure and communitas, and to states and transitions.
TH E LIMINALITY OF AN INSTALLATION RITE
One brief example from the Ndembu of Zambi a of a rite de passage
t hat concerns the highest status in t hat tribe, that of the senior chief
98
The Ritual Process
Kanongesha, will be useful here. It will al so expand our knowledge
of the way the Ndembu utilize and expl ai n t hei r ritual symbols. The
position of senior or paramount chief a mon g the Ndembu, as in
many other African societies, is a par adoxi cal one, for he represents
both the apex of the structured politico-legal hierarchy and the total
communi t y as an unstructured unit. He is, symbolically, also the
tribal territory itself and all its resources. I t s fertility and freedom
from drought, famine, disease, and insect pl agues are bound up with
his office, and with both his physical and mor al condition. Among
t he Ndembu, the ritual powers of the seni or chief were limited by
and combined with those held by a seni or headman of t he au-
tochthonous Mbwel a people, who made submission only after long
struggle to their Lunda conquerors led by t he first Kanongesha. An
i mport ant right was vested in the he a dma n named Kafwana, of t he
Humbu, a branch of the Mbwel a. Thi s was the right to confer and
periodically to medicate the supreme symbol of chiefly status among
tribes of Lunda origin, t he lukanu bracel et , made from human gen-
italia and sinews and soaked in t he sacrificial blood of mal e and
female slaves at each installation. Kafwana' s ritual title was Chi v-
wikankanu, " t h e one who dresses wi t h or put s on t he lukanu." He
also had the title Mama yaKanongesha, " mot h e r of Kanonges ha, "
because he gave symbolic birth to each new i ncumbent of t hat office.
Kafwana was also said to teach each new Kanonges ha t he medicines
of witchcraft, which made hi m feared by his rivals and subordi nat es
perhaps one indication of weak political cent ral i zat i on.
Th e lukanu, originally conferred by t he head of all t he Lunda, t he
Mwant i yanvwa, who ruled in the Kat anga many miles to t he nor t h,
was ritually treated by Kafwana and hi dden by hi m dur i ng inter-
regna. The mystical power of the lukanu, and hence of t he Kanon-
gesha-ship, came jointly from Mwant i yanvwa, t he political fountain-
head and, Kafwana, the ritual source: its empl oyment for t he benefit
of t he land and the people was in t he hands of a succession of
individual incumbents of the chieftainship. Its origin in Mwant i yan-
vwa symbolized the historical unity of t he Nde mbu people, and t hei r
political differentiation into subchiefdoms under Kanonge s ha; its
Liminality and Communitas
99
periodic medication by Kafwana symbolized the landof which Kaf-
wana was the original " owne r " and the total communi t y living on
it. The daily invocations made to it by Kanongesha, at dawn and
sunset, were for t he fertility and continued health and strength of the
land, of its ani mal and vegetable resources, and of t he peoplein
short, for the commonweal and public good. But the lukanu had a
negative aspect; it could be used by Kanongesha to curse. If he
touched t he eart h with it and uttered a certain formula, it was
believed t hat the person or group cursed would become barren, their
land infertile and their game invisible. In t he lukanu, finally, Lunda
and Mbwel a were united in the j oi nt concept of Ndembu land and
folk.
In t he relationship between Lunda and Mbwela, and between
Kanongesha and Kafwana, we find a distinction familiar in Africa
between t he politically or militarily strong and the subdued autoch-
thonous people, who are nevertheless ritually potent. Iowan Lewis
(1963) has described such structural inferiors as having " t h e power
or powers of the we ak" (p. 1 1 1 ) . One well-known example from the
literature is to be found in Meyer Fortes' s account of the Tallensi of
northern Ghana, where t he incoming Namoos brought chieftainship
and a highly developed ancestral cult to the autochthonous Tal e,
who, for their part , are thought to have i mport ant ritual powers in
connection with the eart h and its caverns. In the great Golib
Festival, held annually, t he union of chiefly and priestly powers is
symbolized by the mystical marriage between chief of Tongo, leader
of the Namoos, and the great earth-priest, the Gol i bdaana, of the
Tale, portrayed respectively as " h us ba n d" and "wi fe. " Among
Ndembu, Kafwana is also considered, as we have seen, symbolically
feminine in relation to Kanongesha. I could multiply examples of
this type of dichotomy many times from African sources alone, and
its range is world-wide. The point I would like to stress here is t hat
there is a certain homology between the ^wjeakness.".and "jDassivity^"
ojLHrmnality in diachronic transitions between states and statuses,
and the "s t r uct ur al " or synchronic inferiority of certain personae,
groups, and social categories in political, legal, and economic
IOO
The Ritual Process
systems. Th e " l i mi n a l " and t he "i nfer i or " condi t i ons are often asso-
ciated with ri t ual powers and wi t h t he t ot al communi t y seen as
undifferentiated.
To r et ur n to t he installation rites of t he Kanonges ha of t he
Nde mbu: Th e liminal component of such rites begi ns with t he con-
struction of a small shelter of leaves about a mi l e away from t he
capital village. Thi s hut is known as kafu or kafwi, a t er m Nde mbu
derive from ku-fwa, " t o di e, " for it is here t h a t t h e chief-elect dies
from his commoner st at e. I mager y of deat h abounds in Nde mbu
liminality. For exampl e, t he secret and sacred si t e wher e novices are
circumcised is known as ifwilu or chifwilu, a t e r m also derived from
ku-fwa. Th e chief-elect, clad i n not hi ng but a r agged waist-cloth, and
a ri t ual wife, who is either his senior wife (mwadyi) or a special
slave woman, known as lukanu (after t he r oyal bracelet) for t he
occasion, similarly clad, ar e called by Kaf wana t o ent er t he kafu
shelter j ust after sundown. Th e chief himself, i nci dent al l y, is also
known as mwadyi or lukanu i n these rites. Th e coupl e ar e led t her e as
t hough they were infirm. Ther e they sit cr ouched in a post ure of
shame (nsonyi) or modesty, while they are was hed wi t h medi ci nes
mi xed wi t h wat er br ought from Kat ukang' onyi , t he river site wher e
t he ancestral chiefs of the sout hern Lunda di aspor a dwel t for a whi l e
on t hei r j our ney from Mwant i yanvwa' s capi t al before separat i ng t o
carve out realms for themselves. Th e wood for this fire must not be
cut by an ax but found lying on t he ground. Thi s means t hat it is t he
pr oduct of t he ear t h itself and not an artifact. On c e mor e we see t he
conjunction of ancestral Lundahood and t he cht honi c powers.
Next begins t he rite of Kumukindyila, whi ch means literally " t o
speak evil or insulting words against h i m" ; we mi ght call t hi s ri t e
" T h e Reviling of t he Chief-Elect. " I t begins whe n Kafwana makes
a cut on t he underside of the chief' s left ar mon whi ch t he lukanu
bracelet will be dr awn on t he morrowpresses medi ci ne i nt o t he
incision, and presses a mat on t he upper side of t he a r m. Th e chi ef
and his wife are t hen forced rat her roughly t o sit on t he mat . T h e
wife must not be pregnant , for t he rites t hat follow ar e hel d t o dest roy
fertility. Moreover, t he chiefly couple must have refrai ned from sex-
ual congress for several days before t he rites.
Liminality and Communitas
101
Kafwana now breaks into a homily, as follows:
Be silent! You are a mean and selfish fool, one who is bad-tempered! You
do not love your fellows, you are only angry with them! Meanness and
theft are all you have! Yet here we have called you and we say that you
must succeed to the chieftainship. Put away meanness, put aside anger, give
up adulterous intercourse, give them up immediately! We have granted
you chieftainship. You must eat with your fellow men, you must live well
with them. Do not prepare witchcraft medicines that you may devour your
fellows in their hutsthat is forbidden! We have desired you and you only
for our chief. Let your wife prepare food for the people who come here to
the capital village. Do not be selfish, do not keep the chieftainship to your-
self! You must laugh with the people, you must abstain from witchcraft, if
perchance you have been given it already! You must not be killing people!
You must not be ungenerous to people!
But you, Chief Kanongesha, Chifwanakenu ["son who resembles his
father"] of Mwantiyanvwa, you have danced for your chieftainship because
your predecessor is dead [i.e., because you killed him]. But today you are
born as a new chief. You must know the people, O Chifwanakenu. If you
were mean, and used to eat your cassava mush alone, or your meat alone,
today you are in the chieftainship. You must give up your selfish ways,
you must welcome everyone, you are the chief! You must stop being adult-
erous and quarrelsome. You must not bring partial judgments to bear on
any law case involving your people, especially where your own children
are involved. You must say: " If someone has slept with my wife, or wronged
me, today I must not judge his case unjustly. I must not keep resentment
in my heart."
After this harangue, any person who considers t hat he has been
wronged by t he chief-elect in the past is entitled to revile hi m and
most fully express his resentment, going into as much detail as he
desires. Th e chief-elect, duri ng all this, has to sit silently wi t h down-
cast head, " t h e pat t ern of all pat i ence" and humility. Kafwana
meanwhile splashes the chief with medicine, at intervals striking his
buttocks against hi m (kumubayisha) insultingly. Many informants
have told me t hat " a chief is j ust like a slave (ndung'u) on the ni ght
before he succeeds." He is prevented from sleeping, partly as an
ordeal, partly because it is said that if he dozes off he will have bad
dreams about the shades of dead chiefs, " who will say t hat he is
102
The Ritual Process
wr ong to succeed t h e m, for has he not killed t h e m? " Kafwana, his
assistants, a n d ot he r i mpor t ant men, such as village headmen, man-
handl e t he chief a n d his wifewho is similarly reviledand order
t hem to fetch fi rewood and perform ot her menial tasks. The chief
may not resent any of t hi s or hold it against t he perpetrators in times
t o come.
ATTRI BUTES OF LIMINAL ENTITIES
Th e phase of r eaggr egat i on in this case comprises t he public installa-
t i on of t he Ka n on ge s h a wi t h all pomp and ceremony. Whi l e this
woul d be of t he ut mos t interest in study of Ndembu chieftainship,
and to an i mpor t ant t r end i n current British social anthropology, it
does not concern us her e. Ou r present focus is upon liminality and
t he ri t ual powers of t he weak. Thes e are shown under two aspects.
First, Kafwana a n d t he ot her Nde mbu commoners are revealed as
privileged to exer t aut hor i t y over t he supreme aut hori t y figure of the
t ri be. I n l i mi nal i t y, t h e under l i ng comes uppermost . Second, t he
supreme political aut hor i t y is por t r ayed " a s a^slave," recalling t hat
aspect of t he cor onat i on of a pope i n west ern-Chri st endom when he
is called upon t o be t he "serous servorum Dei." Par t of t he rite has, of
course, what Mon i c a Wi l son (1 95 7, pp. 46-54) has called a " pr o-
phylactic funct i on. " T h e chief has to exert self-control in t he rites
t hat he may be abl e t o have self-mastery thereafter in face of the
t empt at i ons of powe r . But t he role of t he humbl ed chief is only an
ext reme exampl e of a r ecur r ent t heme of liminal situations. This
t heme is t he s t r i ppi ng off of prel i mi nal and postliminal attributes.
Let us look at t he ma i n i ngredi ent s of t he Kumukindyila rites. The
chief and his wife ar e dressed i dent i cal l y in a ragged waist-cloth and
share t he same namemwadyi . Thi s t er m is also applied to boys
undergoi ng i ni t i at i on a n d t o a man' s first wife in chronological order
of marri age. I t is a n i ndex of t he anonymous state of "i ni t i and. "
These at t ri but es of sexlessness a n d anonymi t y are highly character-
istic of liminality. I n ma n y ki nds of initiation where t he neophytes
are of bot h sexes, mal es a n d females ar e dressed alike and referred to
Liminality and Communitas
by the same t erm. This is true, for example, of many baptismal cere-
monies in Christian or syncretist sects in Africa: for example, those
of the Bwiti cult in the Gabon (James Fernandez; personal com-
muni cat i on). It is also true of initiation into the Ndembu funerary
association of Chiwila. Symbolically, all attributes that distinguish
categories and groups in the structured social order are here in abey-
ance; the neophytes are merely entities in transition, as yet wi t hout
place or position.
Ot her characteristics are submissiyeness and silence. Not only t he
chief in t he rites under discussion, but also neophytes in many riles de
passage have to submit to an aut hori t y that is not hi ng less t han t hat
of the total communi t y. Thi s communi t y is the repository of t he
whole gamut of the culture' s values, norms, attitudes, sentiments, and
relationships. Its representatives in t he specific ritesand these may
vary from ritual to ritualrepresent the generic authority of t radi -
tion. I n tribal societies, too, speech is not merely communication but
also power and wisdom. The wisdom (mana) t hat is i mpart ed in
sacred liminality is not j ust an aggregation of words and sentences;
it has ontological value, it refashions t he very being of the neophyt e.
That is why, in the Chisungu rites of the Bemba, so well described by
Audrey Ri chards (1956), the secluded girl is said to be "gr own into
a woman " by the female eldersand she is so grown by the verbal
and nonverbal instruction she receives in precept and symbol, especi-
ally by t he revelation to her of tribal sacra in the form of pottery
images.
The neophyte in liminality must be a tabula rasa, a blank slate,
on which is inscribed the knowledge and wisdom of the group, in
those respects that pert ai n to the new status. The ordeals and humi l i -
ations, often of a grossly physiological character, to which neophytes
are submitted represent partly a destruction of t he previous status
and part l y a tempering of their essence in order to prepare t hem to
cope wi t h their new responsibilities and restrain t hem in advance
from abusing their new privileges. They have to be shown t hat in
themselves they are clay or dust, mere mat t er, whose form is impres-
sed upon t hem by society.
104
The Ritual Process
Anot her liminal t heme exemplified in t h e Nde mbu installation
rites is sexual continence. Thi s is a per vasi ve t heme of Ndembu
ritual. Indeed, the resumpt i on of sexual r el at i ons is usually a cere-
moni al mar k of the ret urn to society as a s t r uct ur e of statuses. While
this is a feature of certain types of religious behavi or in almost all
societies, in preindustrial society, wi t h its s t r ong stress on kinship as
t he basis of many types of group affiliation, sexual continence has
additional religious force. For kinship, or rel at i ons shaped by t he
idiom of kinship, is one of t he mai n factors i n st ruct ural differentia-
tion. Th e undifferentiated charact er of l i mi nal i t y is reflected by t he
discontinuance of sexual relations and t he abs ence of marked sexual
polarity.
I t is instructive to analyze t he homiletic of Kafwana, in seeking to
grasp t he meani ng of liminality. Th e r eader will r emember t hat he
chided t he chief-elect for his selfishness, meannes s , theft, anger, wi t ch-
craft, and greed. All these vices represent t he desi re t o possess for one-
self what ought to be shared for t he common good. An i ncumbent of
high status is peculiarly t empt ed to use t he aut hor i t y vested in hi m by
society to satisfy these pri vat e and pri vat i ve wishes. But he should
regard his privileges as gifts of t he whol e communi t y, whi ch in t he
final issue has an overright over all his act i ons. St r uct ur e and t he
high offices provided by st ruct ure are t hus seen as i nst rument al i t i es
of t he commonweal , not as means of personal aggr andi zement . Th e
chief must not "keep his chieftainship to himself. " He " mus t l augh
wi t h t he peopl e, " and l aught er (ku-seha) is for t he Nde mbu a " whi t e "
quality, and enters into t he definition of " wh i t e n e s s " or " wh i t e
t hi ngs. " Whiteness represents t he seamless we b of connect i on t hat
ideally ought to include bot h t he living a n d t he dead. I t is r i ght
relation bet ween people, merel y as h uma n bei ngs, and its fruits ar e
heal t h, strength, and all good things. " Wh i t e " l aught er , for exampl e,
whi ch is visibly manifested in t he flashing of t eet h, represent s fellow-
ship and good company. I t is t he reverse of pr i de (winyi), a n d t he
secret envies, lusts, and grudges t hat result behavi oral l y in wi t chcraft
(wuloji), theft (wukombi), adul t ery (kushimbana), meanness (chifwa),
and homicide (wubanji). Even when a ma n has become a chief, he
Liminality and Communitas
105
must still be a member of the whole community of persons (antu),
and show this by "l aughi ng with t hem, " respecting their rights,
"welcoming everyone, " and sharing food with t hem. The chastening
function of liminality is not confined to this type of initiation but
forms a component of many other types in many cultures. A well-
known example is the medieval knight' s vigil, during t he night before
he receives the accolade, when he has to pledge himself to serve the
weak and the distressed and to meditate on his own unworthiness.
His subsequent power is t hought partially to spring from this pro-
found immersion in humility.
The pedagogics of liminality, therefore, represent a condemnation
of two kinds of separation from t he generic bond of communitas.
The first kind is to act only in terms of t he rights conferred on one
by the incumbency of office in t he social structure. Th e second is to
follow one' s psychobiological urges at the expense of one' s fellows. A
mystical character is assigned to t he sentiment of humankindness in
most types of liminality, and in most cultures this stage of transition
is brought closely in touch with beliefs in the protective and punitive
powers of divine or pret erhuman beings or powers. For example,
when the Ndembu chief-elect emerges from seclusion, one of his
subchiefswho plays a priestly j ol e at t he installation ritesmakes
a ritual fence around the new chief' s dwelling, and prays as follows
to the shades of former chiefs, before the people who have assembled
to witness t he installation:
Listen, all you people. Kanongesha has come to be born into the chieftain-
ship today. Thi s white clay [mpemba], with which the chief, the ancestral
shrines, and the officiants will be anointed, is for you, all the Kanongeshas
of old gathered together here. [Here the ancient chiefs are mentioned by
name.] And, therefore, all you who have died, look upon your friend who
has succeeded [to the chiefly stool], that he may be strong. H e must con-
tinue to pray well to you. He must look after the children, he must care for
all the people, both men and women, that they may be strong and that he
himself should be hale. Here is your white clay. I have enthroned you,
O chief. You O people must give forth sounds of praise. The chieftainship
has appeared.
The Ritual Process
The powers t hat shape t he neophytes in liminality for t he i n c um-
bency of new status are felt, in rites all over t he world, to be mo r e
t han human powers, t hough they are invoked and channel ed by
t he representatives of t he communi t y.
%
LIMINALITY CONTRASTED WITH STATUS SYSTEM
Let us now, rat her in t he fashion of Levi-Strauss, express t he differ-
ence bet ween t he properties of liminality and those of t he s t at us
system in terms of a series of bi nary oppositions or di scri mi nat i ons.
They can be ordered as follows:
Transition/state
Totality/partiality
Homogeneity/heterogeneity
Communitas/structure
Equality/inequality
Anonymity/systems of nomenclature
Absence of property/property
Absence of status/status
Nakedness or uniform clothing/distinctions of clothing
Sexual continence/sexuality
Minimization of sex distinctions/maximization of sex distinctions
Absence of rank/distinctions of rank
Humility/just pride of position
Disregard for personal appearance/care for personal appearance
No distinctions of wealth/distinctions of wealth
Unselfishness/selfishness
Total obedience/obedience only to superior rank
Sacredness/secularity
Sacred instruction/technical knowledge
Silence/speech
Suspension of kinship rights and obligations/kinship rights and
obligations
Continuous reference to mystical powers/intermittent reference to
mystical powers
Foolishness/sagacity
Liminality and Communitas
Simplicity/complexity
Acceptance of pain and suffering/avoidance of pain and suffering
Heteronomy/degrees of autonomy
This list could be considerably lengthened if we were to widen
the span of liminal situations considered. Moreover, t he symbols in
which these properties are manifested and embodied are manifold
and various, and often relate to the physiological processes of deat h
and birth, anabolism and katabolism. The reader will have noticed
immediately t hat many of these properties constitute what we think
of as characteristics of t he religious life in the Christian tradition.
Undoubtedly, Muslims, Buddhists, Hi ndus, and Jews would num-
ber many of t hem among their religious characteristics, too. What
appears to have happened is t hat wi t h t he increasing specialization
of society and culture, wi t h progressive complexity in t he social divi-
sion of labor, what was in tribal society principally a set of transi-
tional qualities " betwixt and between " defined states of culture and
society has become itself an institutionalized state. But traces of t he
passage quality of the religious life remai n in such formulations as :
" Th e Christian is a stranger to t he world, a pilgrim, a traveler, wi t h
no place to rest his head. " Transition has here become a permanent
condition. Nowhere has this institutionalization of liminality been
more clearly marked and defined t han in the monastic and mendi -
cant states in t he great world religions.
For example, the Western Christian Rul e of St. Benedict "provi des
for the life of men who wish to live in community and devote themselves
entirely to God' s service by self-discipline, prayer, and work. They are
to be essentially families, in the care and under the absolute control of a
father (the abbot ) ; individually they are bound to personal poverty,
abstention from marriage, and obedience to their superiors, and by t he vows
of stability and conversion of manners [originally a synonym for
"common life," "monast i ci t y" as distinguished from secular life]; a
moderate degree of austerity is imposed by" the night office, fasting,
abstinence from fleshmeat, and restraint in conversation" (Attwater,
1961, p. 5 1 my emphases). I have stressed features t hat bear a
remarkable similarity to the condition of the chief-elect duri ng his
io8
The Ritual Process
transition to t he publ i c i ns t al l at i on r i t es, when he enters his kingdom.
Th e Nde mbu circumcision r i t es (Mukanda) present further parallels
bet ween t he neophyt es a n d t h e mo n k s of St. Benedict. Erving Goff-
man (Asylums, 1962) discusses wh a t h e calls t he "characteristics of
total institutions. " Amon g t hes e h e i ncl udes monasteries, and de-
votes a good deal of at t ent i on t o " t h e stripping and leveling pro-
cesses whi ch . . . di rect l y cut acr oss t he various social distinctions
wi t h whi ch the recruits e n t e r . " H e t he n quotes from St. Benedict' s
advice to t he abbot : " Le t h i m ma k e n o distinction of persons in t he
monast ery. Let not one be l oved mor e t han anot her, unless he be
found to excel in good works or i n obedi ence. Let not one of noble
bi rt h be raised above h i m wh o was formerly a slave, unless some
other reasonable cause i n t e r v e n e " ( p. 1 1 9 ) .
Her e parallels wi t h Mukanda a r e st ri ki ng. The novices are "s t r i p-
p e d " of their secular cl ot hi ng wh e n t hey ar e passed beneat h a
symbolic gat eway; t hey ar e " l evel ed " i n t hat their former names are
discarded and all ar e assi gned t h e common designation mwadyi, or
"novi ce, " and t reat ed al i ke. On e of t he songs sung by circumcisers
to t he mot hers of t he novi ces on t he ni ght before circumcision con-
tains t he following l i ne: " Eve n if your chi l d is a chief' s son, t omorrow
he will be like a sl ave"j ust as a chief-elect is t r eat ed like a slave
before his installation. Mor eover , t he senior i nst ruct or in t he seclu-
sion lodge is chosen par t l y becaus e he is fat her of several boys under -
going t he rites and becomes a fat her for t he whol e gr oup, a sort of
" a bbot , " t hough his title Mfumwa tubwiku, means literally " h us ba n d
of t he novices, " to emphasi ze t hei r passi ve rol e.
MYSTICAL DANGER
AND TH E POWERS OF TH E WEAK
One may well ask why i t is t h a t l i mi nal si t uat i ons and roles are
almost everywhere at t r i but ed wi t h magi co-rel i gi ous propert i es, or
why these should so often b e r e ga r de d as danger ous , i nauspi ci ous,
or polluting to persons, objects, event s, a n d rel at i onshi ps t hat have
Liminality and Communitas
109
not been ritually incorporated into t he liminal context. My view is
briefly t hat from the perspectival viewpoint of those concerned with
the mai nt enance of "st r uct ur e, " all sustained manifestations of com-
munitas must appear as dangerous and anarchical, and have to
be hedged around with prescriptions, prohibitions, and conditions.
And, as Mar y Douglas (1966) has recently argued, t hat which can-
not be clearly classified in terms of traditional criteria of classifica-
tion, or falls between classificatory boundaries, is almost everywhere
regarded as "pol l ut i ng" and " dange r ous " (passim).
To repeat what I said earlier, liminality is not the only cultural
manifestation of communi t as. In most societies, t here are other areas
of manifestation to be readily recognized by t he symbols t hat cluster
around t hem and the beliefs t hat at t ach to them, such as " t he powers
of the weak, " or, in other words, the permanent l y or transiently
sacred attributes of low status or position. Wi t hi n stable structural
systems, there are many dimensions of organization. We have already
noted t hat mystical and moral powers are wielded by subjugated
autochthones over the total welfare of societies whose political frame
is constituted by t he lineage or territorial organization of incoming
conquerors. I n other societiesthe Ndembu and Lamba of Zambi a,
for examplewe can point to the cult associations whose members
have gained entry t hrough common misfortune and debilitating
circumstances to t herapeut i c powers with regard to such common
goods of manki nd as heal t h, fertility, and climate. These associations
transect such i mport ant components of t he secular political system
as lineages, villages, subchiefdoms, and chiefdoms. We could also
mention t he role of structurally small and politically insignificant
nations within systems of nations as upholders of religious and moral
values, such as t he Hebrews in t he ancient Near East, t he Irish in
early medieval Christendom, and t he Swiss in modern Europe.
Many writers have dr awn attention to t he role of t he court jester.
Max Gl uckman (1965), for example, writes: " T h e court jester
operated as a privileged arbiter of morals, given license to gibe at
king and courtiers, or lord of t he manor . " Jesters were "usual l y
men of low classsometimes on the Continent of Europe they were
no
The Ritual Process
priestswho clearly moved out of their usual estate. . . . I n a system
where it was difficult for others to rebuke t he head of a political
unit, we mi ght have here an institutionalized joker, operat i ng at t he
highest point of the unit . . . a j oker able to express feelings of out-
raged moral i t y. " He further ment i ons how jesters at t ached to many
African monarchs were "frequent l y dwarfs and other oddi t i es. "
Similar in function to these were t he dr ummer s in the Barotse royal
barge in whi ch t he king and his court moved from a capital in t he
Zambezi Flood Plain to one of its margi ns duri ng the annual floods.
They were privileged to throw into t he wat er any of the great nobles
" wh o had offended t hem and their sense of justice duri ng t he past
ye ar " (pp. 1 02- 1 04) . These figures, representing the poor and t he
deformed, appear to symbolize t he moral values of communi t as as
against t he coercive power of supreme political rulers.
Folk literature abounds in symbolic figures, such as "hol y beg-
gars, " " t hi r d sons, " "l i t t l e tailors, " and "si mpl et ons, " who strip off
t he pretensions of holders of hi gh r ank and office and reduce t hem
to t he level of common humani t y and mort al i t y. Again, in t he
traditional "Wes t er n, " we have all r ead of t he homeless and
mysterious " s t r ange r " wi t hout weal t h or name who restores ethical
and legal equi l i bri um to a local set of political power relations by
eliminating t he unjust secular "bos s es " who are oppressing t he
smallholders. Member s of despised or out l awed ethnic and cul t ural
groups pl ay major roles in myt hs and popul ar tales as representatives
or expressions of universal huma n values. Famous among these ar e
t he good Samari t an, t he Jewi sh fiddler Rot hschi l d in Chekhov' s tale
"Rot hschi l d' s Fi ddl e, " Mar k Twai n' s fugitive Negro slave J i m in
Huckleberry Finn, and Dostoevsky' s Sonya, t he prostitute who redeems
t he woul d-be Nietzschean " s u p e r ma n " Raskolnikov, in Crime and
Punishment.
All these myt hi c types are st ruct ural l y inferior or " mar gi nal , " yet
represent what Henr i Bergson woul d have called " o p e n " as agai nst
"closed moral i t y, " t he l at t er bei ng essentially t he normat i ve system
of bounded, structured, particularistic groups. Bergson speaks of how
an i n-group preserves its i dent i t y agai nst member s of out -groups,
Liminality and Communitas
i n
protects itself against threats to its way of life, and renews the will to
maintain the norms on which t he routine behavior necessary for its
social life depends. I n closed or structured societies, it is t he margi nal
or "i nferi or" person or the "out s i der " who often comes to sym-
bolize what Davi d Hume has called " t h e sentiment for humani t y, "
which in its t urn relates to t he model we have termed " communi t as. "
MILLENARIAN MOVEMENTS
Among the more striking manifestations of communi t as are to be
found the so-called millenarian religious movements, whi ch arise
among what Nor man Cohn ( 1 9 61 ) has called "upr oot ed and des-
perate masses in town and countryside . . . living on the margi n of
society" (pp. 3 1 - 3 2) (i.e., structured society), or where formerly
tribal societies are brought under the alien overlordship of complex,
industrial societies. The attributes of such movements will be well
known to most of my readers. Here I would merely recall some of t he
properties of liminality in tribal rituals t hat I ment i oned earlier.
Many of these correspond pretty closely wi t h those of millenarian
movement s: homogeneity, equality, anonymity, absence of propert y
(many movements actually enjoin on their members t he destruction
of what property they possess to bring nearer t he coming of t he
perfect state of unison and communi on they desire, for propert y
rights are linked with structural distinctions bot h vertical and hori-
zontal), reduction of all to t he same status level, t he wearing of
uniform apparel (sometimes for bot h sexes), sexual continence (or
its antithesis, sexual communi t y, bot h continence and sexual com-
munity liquidate marri age and the family, which legitimate struc-
tural status), minimization of sex distinctions (all are "equal i n t he
sight of God" or t he ancestors), abolition of rank, humility, dis-
regard for personal appearance, unselfishness, total obedience to t he
prophet or leader, sacred instruction, t he maximization of religious,
as opposed to secular, attitudes and behavior, suspension of kinship
rights and obligations (all are siblings or comrades of one anot her
regardless of previous secular ties), simplicity of speech and nanner s,
1 1 2
The Ritual Process
sacred folly, accept ance of p a i n a n d suffering (even to t he point of
undergoi ng mar t yr dom) , a n d so for t h.
It is not ewort hy t hat ma n y of t hese movements cut right across
tribal and national di vi si ons d u r i n g their initial moment um. Com-
muni t as, or the " ope n s oci et y, " differs in this from structure, or the
"closed society," in t hat i t is pot ent i al l y or ideally extensible to the
limits of humani t y. I n pr act i ce , of course, the impetus soon becomes
exhausted, and t he " mo v e me n t " becomes itself an institution among
other institutionsoften on e mo r e fanatical and militant than the
rest, for the reason t hat i t feels itself to be the uni que bearer of
universal human t r ut hs . Mos t l y , such movements occur during
phases of history t hat a r e i n ma n y respects "homol ogous" to the
liminal periods of i mpor t a nt r i t ua l s i n stable and repetitive societies,
when major groups or soci al cat egor i es in those societies are passing
from one cultural st at e t o a n o t h e r . Th e y are essentially phenomena
of transition. Thi s is pe r h a ps w h y i n so many of these movements
much of their myt hol ogy a n d s ymbol i s m is borrowed from those of
traditional rites de passage, e i t h e r i n t he cultures in which they
originate or in t he cul t ur es wi t h wh i c h t hey are in dramat i c contact.
H I PPI ES, COMMUNITAS,
AND TH E PO WERS OF TH E WEAK
I n moder n West ern society, t h e val ue s of communi t as are strikingly
present in t he l i t er at ur e a n d be h a vi or of what came to be known
as t he " be a t gener at i on, " wh o we r e succeeded by the "hi ppi es, "
who, in t urn, have a j uni or di vi s i on known as the " teeny-boppers. "
These are t he " c o o l " me mb e r s of t he adolescent and young-adult
categorieswhich do not h a ve t h e advant ages of national rites de
passagewho " o p t o u t " of t he s t a t us - bound social order and acquire
t he stigmata of t he l owl y, dr es s i ng l i ke " bums , " itinerant in their
habi t s, "fol k" in t hei r mus i cal t as t es , and menial in the casual
empl oyment they unde r t ake . T h e y st ress personal relationships rather
t han social obligations, a n d r e ga r d s exual i t y as a polymorphic instru-
Liminality and Communitas
" 3
merit of i mmedi at e communi t as rat her t han as the basis for an
enduring structured social tie. The poet Allen Ginsberg is particularly
eloquent about the function of sexual freedom. The "s acr e d"
properties often assigned to communitas are not lacking here, either:
this can be seen in their frequent use of religious terms, such as
"s ai nt " and "angel , " to describe their congeners and in their
interest in Zen Buddhism. The Zen formulation "al l is one, one is
none, none is al l " well expresses the global, unstructured character
earlier applied to communi t as. The hippie emphasis on spontaneity,
immediacy, and "exi st ence" throws into relief one of the senses in
which communitas contrasts wi t h structure. Communi t as is of the
now; structure is rooted in the past and extends into the future through
language, law, and custom. While our focus here is on traditional
preindustrial societies it becomes clear t hat the collective dimensions,
communitas and structure, are to be found at all stages and levels of
culture and society.
STRUCTURE AND COMMUNITAS
IN KINSH IP BASED SOCIETIES
I. Tallensi
There are some further manifestations of this distinction found in the
simpler societies. These I shall consider in terms, not of passages
between states, but rat her of binarily opposed states t hat in certain
respects express the distinction between society regarded as a struc-
ture of segmentarily or hierarchically opposed parts and as a homo-
geneous totality. In many societies, a terminological distinction is
made between relatives on the father' s and mother' s side, and these
are regarded as quite different kinds of people. This is especially the
case with regard to the father and the mother' s brother. Where there
is unilineal descent, propert y and status pass either from father to
son or from mother' s brot her to sister's son. In some societies, bot h
lines of descent are used for purposes of inheritance. But, even i n this
U 4
The Ritual Process
instance, t he types of p r o p e r t y a n d status t hat pass in each line are
very different.
Let us begin by cons i de r i ng a society i
n
whi ch there is unilineal
descent only in the pat e r nal l i n e . This example is drawn once more
from the Tallensi of Gh a n a , o n which we have rich information.
Our probl em is to discover wh e t h e r in a bi nary discrimination at one
structural level of t he t ype " s t r uc t ur a l superiority-structural inferi-
ori t y, " we can find a n y t h i n g appr oxi mat i ng t he ritual " power of the
weak, " which, in its t ur n, c a n be shown to relate to the model of
communi t as. Fort es (1949) wr i t e s :
The dominant line of descent confers the overtly significant attributes of
social personalityjural status, rights of inheritance and succession to
property and office, political allegiance, ritual privileges and obligations;
and the submerged line [constituted by matri-filiation; I would prefer
"submerged side," since the link is a personal one between ego and his
mother and through her both t o her patrilineal kin and to her cognates]
confers certain spiritual characteristics. Among the Tallensi it is easy to see
that this is a reflex of the fact t hat the bond of uterine descent is maintained
as a purely personal bond. It does not subserve common interests of a material,
jural, or ritual kind; it unites individuals only by ties of mutual interest and
concern not unlike those that prevail between close collateral kin in our
culture. While it constitutes one of the factors that counterpoise the exclusive-
ness of the agnatic line, it does not create corporate groups competing with the
agnatic lineage and clan. Carryi ng only a spiritual attribute, the uterine tie
cannot undermine the j ural and politico-ritual solidarity of the patrilineal
lineage (p. 3amy emphases).
Here we have t he oppos i t i on pat r i l i neal / mat r i l at er al , which has
t he functions domi nant / s ubme r ge d. T h e pat r i l i neal tie is associated
wi t h property, office, pol i t i cal al l egi ance, exclusiveness, and, it may be
added, particularistic a n d s e gme nt ar y i nt er es t s . It is the "s t r uct ur al "
link par excellence. Th e ut e r i ne t i e is associated with spiritual
characteristics, mut ual i nt erest s a n d conce r ns , and collaterality. It is
counterpoised to exclusiveness, wh i c h pr es umabl y means that it
makes for inclusiveness a n d does not s e r ve mat e r i al interests. In brief,
matrilaterality represents, in t he di me n s i on of kinship, the notion of
communitas.
Liminality and Communitas
" 5
An example, drawn from the Tallensi, of the "s pi r i t ual " and
"communi t ar i an" character of matrilaterality is to be found in t he
consecration rites of the so-called bakologo, or diviner' s shrine. This
shrine is by definition, says Fortes (1949), a "femal e" shri ne:
That is to say, the ancestors associated with it come, by definition, from a
matrilateral lineage of the diviner; and the dominant figure among them is
usually a woman, " a mother." The bakologo . . . is the very incarnation of
the vindictive and jealous aspect of the ancestors. It persecutes the man in
whose life it has intervened relentlessly, until he finally submits and " accepts
itthat is, until he undertakes to set up a shrine to the [matrilateral]
bakologo spirits in his own home so that he can sacrifice to them regularly.
Every man, and not only those who have suffered exceptional misfortunes, is directed
by the religious system of the Tallensi to project his deeper feelings of
guilt and insecurity largely on to the mother image embodied in the bakologo
complex. Usually, also, a man does not immediately yield to the demands of
the bakologo ancestors. H e temporizes, evades, and resists, perhaps for years,
until he is at last forced to submit and accept the bakologo. Ni ne out of ten
men over forty have bakologo shrines, but not every man has a talent for
divining, so most men simply have the shrine and do not use it for divining
(p. 3 25 my emphases).
I have given Fortes' s account at some length because I think it
brings out vividly, not only the opposition and tension between
patrilineal and matrilateral kinship bonds, but also t he tension pro-
duced in individual pysches as they mat ur e between structural and
communitarian ways of looking at Tallensi society. We must re-
member t hat the dogma of patriliny, what Homans and Schneider
would call the " h a r d " descent line t hrough which rights over status
and property are transmitted, is domi nant and colors t he values of
the Tallensi at many levels of society and culture. From t he stand-
point and perspective of persons occupying positions of authority in
the patrilineal structure, social links t hrough women, symbolizing
the widest Tal e communi t y where it transects the narrow corporate
bonds of descent and locality, must needs appear to have a disruptive
aspect. Tha t is why, in my opinion t he Tallensi have t he bakologo
"mot her i mage, " which "per secut es" and "i nt er venes " in a mat ur e
n6
The Ritual Process
man' s life until he " a c c e pt s " i t . For , as men mat ur e and interact
wi t h one anot her in ever wi der r ange s of social relations, they become
increasingly conscious t hat t hei r pat r i l i ne age s are merely part s of the
Tallensi whole. For t hem, in a pe r f e ct l y literal fashion, t he wider
communi t y intervenes, by br e a ki n g i nt o t he self-sufficiency and
relative aut onomy of segment ar y l i n e a ge and clan affairs. The global
sentiments annual l y stressed at s uch g r e a t integrative festivals as the
Golib, where, as I ment i oned ear l i er , t her e is a ki nd of mystical
marri age bet ween represent at i ves of t h e Namoos invaders and the
Tal e aut ocht hones, become mo r e a n d mor e meaningful to " me n
over forty," who par t i ci pat e as fami l y a n d sublineage heads, and no
longer as minors under pat e r nal a ut h or i t y . Nor ms and values "from
wi t hout " break into t he excl usi veness of lineage loyalties.
I t is clearly appr opr i at e t hat c o mmu n i t a s should her e be symbol-
ized by matrilateral ancestors, especi al l y by mot her images, since, in
this virilocal, patrilineal society, wo me n ent er t he lineage patri-
segments from wi t hout , and, as For t e s has shown, mat ri l at eral kin
for t he most par t reside out si de a ma n ' s "fi el d of cl anshi p. " It is
underst andabl e, too, t hat s uch spi r i t s shoul d be represented as
"vi ndi ct i ve" and " j e a l o u s " : i t is t h e " mo t h e r s " (who are the
founders of dugs, or mat r i - segment s) wh o i nt r oduce divisions into the
ideal uni t y of t he pat ri l i neage. T o put i t briefly, at cer t ai n life crises,
such as adolescence, t he a t t a i n me n t of el der hood, a n d deat h, varying
in significance from cul t ur e t o cul t ur e , t he passage from one struc-
t ural status to anot her ma y be a c c ompa n i e d by a st rong sentiment
of "humanki ndnes s , " a sense of t he gener i c social bond bet ween all
members of societyeven i n s ome cases t r ans cendi ng tribal or
nat i onal boundari esregardl ess of t hei r s ubgr oup affiliations or in-
cumbency of st ruct ural posi t i ons. I n e xt r e me cases, such as the
accept ance of t he s haman' s vocat i on a mo n g t he Saor a of Middle
I ndi a (Elwin, 1 9 5 5 ), this ma y r esul t i n t he t r ans for mat i on of what is
essentially a liminal or ext r as t r uct ur al phas e i nt o a pe r mane nt con-
dition of sacred " out s i der hood. " T h e s h a ma n or pr oph e t assumes a
statusless status, ext ernal t o t he secul ar social s t r uct ur e , whi ch gives
Liminality and Communitas
117
him the right to criticize all structure-bound personae in terms of a
moral order bi ndi ng on all, and also to medi at e between all segments
or components of the structured system.
In societies in which kinship is what Fortes calls an "i rreduci bl e
pri nci pl e" of social organization, and where patrilineality is the
basis of social structure, an individual' s link to other members of his
society t hrough the mot her, and hence by extension and abstraction
" wome n" and "femi ni t y, " tends to symbolize t hat wider communi t y
and its ethical system t hat encompasses and pervades the politico-
legal system. Fascinating correlations can be shown in many societies
to exist between this conversion to t he perspective of communi t as
and the assertion of individuality as against status incumbency. For
example, Fortes (1949) has shown us the individuating functions of
the tie between sister's son and mot her' s brot her among the Tallensi,
which, he says, "i s an i mport ant breach in the genealogical fence
enclosing t he agnatic lineage; it is one of the mai n gateways of an
individual' s social relations with members of other clans t han his
own" (p. 3 1 ) . By matrilaterality, the individual, in his integral
character, is emanci pat ed from t he segmental status incumbencies
determined by patriliny into the wider life of a communi t y which
extends beyond the Tallensi proper into tribal groups of similar
religious culture.
Now for a look at a concrete example of t he way in which the con-
secration of a bakologo shrine makes t he wider Tallensi communi t y
visible and explicit t hrough matrilateral ties. All rituals have this
exemplary, model-displaying charact er; in a sense, they mi ght be
said to " cr e at e " society, in much the same way as Oscar Wilde held
life to be " a n imitation of ar t . " In t he case I cite (Fortes, 1949),
a man named Naabdi ya " acce pt e d" as his bakologo ancestors his
mother' s father, his mot her' s father' s mot her, and his mot her' s
father's mother' s mot her. I t was t he clansmen of the last-mentioned
who came to set up t he shrine for their classificatory " gr ands on"
Naabdiya. But, to reach t hem, Naabdi ya had first to go to his
mother' s brother' s people; they then escorted hi m to his mother' s
n8
The Ritual Process
mot her' s brot her' s l i neage, t we l ve miles from his own settlement. At
each place, he had t o s acr i fi ce a fowl and a gui nea fowli.e., a
domesticated and an un dome s t i c a t e d bi r dt o t he lineage bogar, or
shrine of its founding a n c e s t or .
Th e lineage of t he d o mi n a n t ancest or, or, mor e often, the ancestress
of t he bakologo compl ex, n e a r l y al ways a mat ri l at eral ancestress, has
t he responsibility for s e t t i ng u p t he shri ne for t he afflicted person.
Th e lineage head sacrifices t h e t wo bi rds provi ded by t he patient at
his lineage shrine, e xpl a i n i n g t o t he ancestors t he nat ure of the
occasion t hat has br ought t h e i r sister' s son or mat ri l at eral grandson
to plead wi t h t hem. He asks t h e m to bless t he setting up of t he new
shrine, to assist t he c a n d i d a t e t o become a successful diviner, and to
gr ant hi m prosperity, c h i l dr e n , a n d heal t hi . e. , general good things.
The n he scoops some s e d i me n t from t he bot t om of the pot, which is
t he most i mpor t ant c o mp o n e n t of a bogar shri ne, into a tiny pot for
t he candi dat e to t ake h o me t o a dd to his new shri ne. " I n this way, "
says Fortes, " t h e di r ect c ont i nui t y of t he new bakologo shrine with
t he mat ri l at eral l i neage bogar is t angi bl y s ymbol i zed" (p. 326).
Thus , t wo shrines t wel ve mi l e s a pa r t a n d it must be remembered
t hat Tal el and itself is " s c a r c e l y t went y mi l es wi de " and several
intervening shrines ar e di r e ct l y and " t a n g i b l y " linked by t he rites.
Th e fact t hat cont i nuous phys i cal cont act bet ween the lineages
involved is hardl y possi bl e is not ideologically i mport ant here, for
t he bakologo shrines ar e s ymbol s a n d expressions of Tal e community.
" Ni n e out o f t e n " ma t u r e me n have a cl ust er of bakologo ancestors
each. All these me n ar e r i t ual l y connect ed t hr ough t hem to a plura-
lity of settlements. Conver s el y, each l i neage bogar has linked to it a
number of bakologo shri nes t h r ough sor or al or sisterly connections.
Such linkages ar e pat e nt l y i n t hei r aggr egat e and transection more
t han merely personal or s pi r i t ual t i es; t hey represent the ties of
communi t as count er i ng t h e cl eavages of s t r uct ur e. They are, more-
over, bonds creat ed fr om t h e " s u b me r g e d " side of kinship, the
j ural l y weaker or inferior s i de. On c e mor e we have manifested the inti-
mat e connection be t we e n c ommun i t a s a n d t he powers of the
weak.
Liminality and Communitas
" 9
2. Nuer
It is this perennial tensed opposition between communitas and struc-
ture that, to my mi nd, lies at the back of bot h the sacred and "affec-
t i onal " aspects of the mot her' s brother/sister' s son relationship in
many patrilineal societies. In these societies, as numerous scholars
have shown, t he mot her' s brother, who has weak j ur al authority
over his nephew, nevertheless may have a close personal tie of
friendship with him, may give him sanctuary from pat ernal harsh-
ness, and, very often, has mystical powers of blessing and cursing
over hi m. Here weak legal authority in a corporate group setting is
countered by strong personal and mystical influences.
Among the Nuer of the Sudan, t he role of "leopard-skin pri est "
interestingly links the symbolic value of t he mot her' s brot her in
patrilineal society with some of the other attributes of liminal,
marginal, and politically weak figures we have already considered.
According to Evans-Pritchard (1956), " i n some myths of the Ji kany
tribes [of t he Nuer] the leopard-skin [insignium of priestly office]
was given by the ancestors of the [territorially] domi nant [agnatic]
lineages to their maternal uncles t hat they mi ght serve as tribal priests.
The structurally opposed lineages of t he clan were then in t he
common relationship of sisters' sons to the line of priests, which thus
had a mediatory position between t h e m" (p. 293my emphasis). As
well as being categorical mot hers' brothers to the political segments,
leopard-skin priests are " i n t he category of ml, strangers, and not of
diel, members of the clans which own t he tribal territories. . . .
[They] have no tribal territories of their own but live, as families
and small lineages, in most or all territories owned by other clans.
They are like Levi, divided in Jacob and scattered in I sr ael "
(p. 292). (Something of this priestly charact er adheres to the scat-
tered lineages of circumcisers and rain makers among the Gisu of
Uganda. ) Nuer leopard-skin priests have " a mystical relationship . . .
with the eart h in virtue of which their curses are t hought to have
special potency, for . . . they can affect not only a man' s crops but
120
The Ritual Process
his welfare generally, si nce h u ma n activities all take place on the
e a r t h " (p. 2 9 1 ) . Th e pri est ' s ma j o r role is in connection with homi-
cide, where he gives t he sl ayer s anct uar y, negotiates a settlement,
performs sacrifice to enabl e n o r ma l social relations to be resumed,
and rehabilitates t he sl ayer. Th i s general i zed mot her' s brot her thus
has many of t he at t ri but es of c ommun i t a s wi t h whi ch we are becom-
ing familiar: he is a s t r anger , a me di at or , acts for the whole com-
muni t y, has a mystical r e l at i ons hi p wi t h t he whole eart h it dwells
upon, represents peace as a ga i n s t feud, and is unaligned with any
specific political segment .
3 . Ashanti
Lest it may be t hought t h a t s t r uc t ur e is universally associated with
pat ri l i ny and mascul i ni t y, a n d c ommun i t a s wi t h matrilaterality and
femininity, in societies a r t i c ul a t e d by t he principle of unilineal
descent, it is wor t h t aki ng a br i e f gl ance at a well-known matrilineal
society, t he Ashant i of Gh a n a . T h e Ashant i belong to a group of
West African societies wi t h h i gh l y devel oped political and religious
systems. Yet, uni l i neal ki ns hi p still has considerable structural
i mport ance. Th e l ocal i zed mat r i l i ne age t raci ng descent from a
known common ancestress for a pe r i od of t en to twelve generations
is t he basic uni t for pol i t i cal , r i t ua l , a n d legal purposes. Fortes (1950)
has described t he s egment ar y c h a r a c t e r of t he lineage: "each seg-
ment bei ng defined i n r el at i on t o ot her segment s of like order by
reference to common a n d di ffer ent i at i ng ancestresses" (p. 255).
Succession to office and i n h e r i t a n c e of pr oper t y are matrilineal, and
t he wards of sections of As h a n t i villages ar e each inhabited by a
nucl ear mat ri l i neage s ur r oun de d by a fringe of cognates and
affines.
Th e name for a mat r i l i ne age is abusua, whi ch, according to
Rat t r ay, (1 923 ) is " s y n on y mous wi t h mogya, bl ood" (p. 35)as in
t he proverb abusua bako mogya bako, " o n e cl an one bl ood. " It has
sometimes been debat ed wh e t h e r As hant i ki nshi p ought not to be
classified as a system of " d o u b l e des cent . " Thi s view derives from
Liminality and Communitas
121
Rattray' s reports (1 923 , p. 45-46) on a mode of social categorization
known by Ashanti as ntoro (literally, " s e me n" ) , which he regarded
as an exogamous division based on transmission by and through
males only. Fortes (1950, p. 266) has emphasized the minimal
significance of this patrilineal element for the kinship system, and
for the politico-legal order. He speaks of the ntoro as "named quasi-
ritual divisions," but these are neither exogamous nor organized
groups in any sense. Yet, from the point of view of the present paper,
the ntbro divisions are of the utmost importance. One of the reasons
for the neglect of the communitas dimension of society, with its pro-
found implications for the underst andi ng of many ritual, ethical,
esthetic, and, indeed, political and legal phenomena and processes,
has been a propensity to equate the "s oci al " with the "social
structural. " Let us then follow the clue of ntoro into many dark
corners of Ashanti culture.
In the first place, the father-son link, the basis of the ntoro division,
is the structurally inferior link. Yet, the symbols with which it is
associated build up into a picture of formidable communitas value.
According to Rat t ray (1 923 ), the Ashanti believe t hat it is the
"male-transmitted ntoro or semen, mingling with the blood [a
symbol of the matrilineage] in the female, which accounts for the
physiological mysteries of conception . . . ntoro . . . is . . . used at
times synonymously with sunsum, that spiritual element in a man or
a woman upon which depends . . . that force, personal magnetism,
character, personality, power, soul, call it what you will, upon which
depend health, wealth, worldly power, success in any venture, in fact
everything t hat makes life at all worth l i vi ng" (p. 46). Once again,
we are beginning to come across the peculiar linkages between
personality, universal values, and "s pi r i t " or " s oul " t hat appear to
be the stigmata of communitas.
Rattray (1923) was able to enumerat e nine ntoro divisions, though
he says there may have been more. These, of course, cut across
the membership of the segmentary abusua matrilineages. One ntoro
is traditionally considered to be " t h e first ntoro ever bestowed upon
man, the Bosommuru ntoro" (p. 48). The myt h told in connection
1 2 2 The Ritual Process
with its establishment, in Rat t r ay' s vi ew, i l l umi na t e s t he way in
which Ashanti think about ntoro in general :
Very long ago one man and one woman came down from the sky and one
woman came up from the earth.
From the Sky God (Onyame), also came a python (onini), and it made its
home in the river now called Bosommuru.
At first these men and women did not bear chi l dren, they had no desire,
and conception and birth were not known at that ti me.
One day the python asked them if they had no offspri ng, and on being
told that they had not, he said he would cause the wome n to conceive. He
bade the couples stand face to face, then he pl ung ed into the river, and
rising up, sprayed water upon their bellies wi th the words kus kus (used in
most ceremonies in connection with ntoro and Onyame ) , and then ordered
them to return home and lie together.
The women conceived and brought forth the first chi ldren in the world,
who took Bosommuru as their ntoro, each mal e passi ng on this ntoro to his
children.
If a Bosommuru ntoro man or woman sees a dead python (they would
never kill one) they sprinkle white clay upon it and bury it (pp. 48 - 49 ) .
Thi s myt h symbolically relates ntoro, as b o t h s e me n and a social
division, wi t h t he Sky God (who is also a r a i n a n d wat e r god) , with
water, wi t h a river, and wi t h t he fertilization of wo me n . Ot he r ntoro
divisions, such as Bosomtwe, whi ch is a l ar ge l a ke i n cent r al Ashant i ,
and Bosompra, a river rising in As hant i , a r e c on n e c t e d wi t h bodies
of wat er. Th e major Ashanti gods ar e male dei t i es, t he sons of
Onyame, t he mal e Hi gh God. Fur t he r mor e , t h e y ar e all connected
wi t h wat er, t he master symbol of fertility, a n d b y ext ensi on of those
good things t he Ashanti hold in common, r egar dl es s of t hei r sub-
group affiliations. Rat t r ay ( 1 9 23 ) quot es As h a n t i as s ayi ng: " On y a me
decided to send those children of his own d o wn t o t h e e a r t h in order
t hat t hey mi ght receive benefits from, a n d confer t h e m upon , man-
kind. All these sons bore t he names of what ar e n o w r i ver s or lakes . . .
and every ot her river or wat er of any i mpor t a n c e . T h e t r i but ar i es of
these agai n are their chi l dr en" ( pp. 1 4 5 - 1 4 6 ) . H e a d d s : " Wh a t has
been said is sufficient to show t hat wat er s i n As h a n t i . . . ar e all looked
Liminality and Communitas
123
upon as containing the power or spirit of the divine Creator, and
thus as being a great life-giving force. ' As a woman gives bi rt h to a
child, so may wat er to a god,' once said a priest to me " (p. 146).
Ot her bodily fluids are symbolically connected with " t h e ntoro
element in man, " says Rat t r ay (1923, p. 54), such as saliva; and wat er
is sprayed from the mout h of the Ashanti king, during rites associated
with the Bosommuru river, accompanied by t he words: "Life to
me, and may this nat i on prosper. " The white symbolism in t he
Bosommuru myt h recurs in many ritual contexts where t he water
gods are worshiped, while t he priests of the Hi gh God and other
deities regularly wear white vestments. I have discussed white
symbolism and its connotations of semen, saliva, health, strength,
and auspiciousness in many African and other societies in several
published works (Turner, 1961; 1962; 1967). Ashant i white sym-
bolism is not dissimilar in its semantics to Ndembu white symbolism.
To summarize our Ashanti findings to this poi nt : Ther e would
appear to be a nexus between the father-child bond, ntoro (as
semen, spirit, and social division with widely scattered member-
ship) ; masculinity (represented by t he father image, Onyame, his
sons, and the mythical pyt hon, a mal e symbol ); saliva; wat er;
blessing by blowing wat er; lakes; rivers; the sea; whi t e symbolism;
and priesthood. In addition, chiefs, and especially t he king, are
clearly associated, in t he Adae and other ceremonies, wi t h the Sky
God and wi t h rivers, especially t he Tano, as the messages of the
talking dr um played at the Adae rites suggest (Rat t ray, 1923, p. 101).
The feminine principle and the abusua are linked, as we have seen,
with blood and t hrough blood, with a rich variety of red symbols.
Nearly everywhere blood and redness have both auspicious and in-
auspicious connotations. In Ashanti, red is associated with war
(Rattray, 1927, p. 134), with witchcraft (pp. 29, 30, 32, 34), with
the revengeful ghosts of victims (p. 22), and with funerals (p. 150).
In some cases there is a direct opposition between t he white (male)
symbolism and the red (female) symbolism. For example, the river
god Tano or Ta Kora, according to Rat t ray (1923), "seems especi-
ally indifferent or even hostile to women. They are ungrateful
124
The Ritual Process
creatures (bonniaye), he de cl ar e s . No women are allowed to touch
his shrine, and he has n o f e mal e akomfo (priests) of his own. Men-
struating women ar e one of hi s t aboos " (p. 183). It will be recalled
t hat t he Ta no Ri ver pl ays a n i mpor t ant role in t he Adae rites of the
Asantehene, pa r a moun t c hi e f of t he nat i on. Witchcraft and the red
symbolism of funerary r i t u a l have a relationship with abusua
membershi p, since i t is ma t r i l i n e a l kin who accuse one another of
witchcraft, and most d e a t h s a r e at t ri but ed to witchcraft. There is
anot her sinister me a n i n g h e r e conceal ed in notion of the blood tie.
Red symbolism is also as s oci at e d wi t h t he cult of t he earth, Asase Ya,
regarded as " a female d e i t y " ( Rat t r ay, 1927). According to Rat t ray,
"s he did not t aboo me n s t r ua t i on {kyiri bara); she liked human
bl ood" (p. 342).
I could mul t i pl y ci t at i ons from Rat t r ay' s (1927) magnificently
detailed dat a on r ed s ymbol i s m to demonst rat e the relationship
Ashanti make bet ween femi ni ni t y, deat h, killing, witchcraft, in-
auspiciousness, me ns t r ual pol l ut i on, and t he sacrifice of men and
beasts. For exampl e, t h e As h a n t i have a " r e d " suman, or "fetish,"
whi ch is " i n t he na t ur e of a s capegoat or somet hi ng t hat takes upon
itself t he evils and sins of t h e wo r l d " (p. 1 3). Thi s is steeped in red
esono dye ( made from t he powde r e d bark of t he adwino tree, prob-
ably a species of Pterocarpus), whi ch is " a substitute for human
bl ood, " and is used i n t h e e a r t h cul t . Esono also stands for menstrual
blood. Thi s so-called kunkuma fetish is also "st ai ned and clotted
with t he blood of sheep a n d fowls t hat have been sacrificed upon it, "
and in it is " h i dde n a pi e ce of fiber (baha) used by a menstruating
woma n " (p. 13). He r e we see sacrificial bl ood and menstruation
brought into r el at i onshi p wi t h breaches of t he nat ural and social
orders"evi l s and s i ns . " O n e final exampl e, perhaps the most
interesting of all, mus t suffice. On c e a year t her e is a ritual violation
of the original ntoro s hr i ne, t h e Bosommur u ntoro ment i oned earlier.
This ntoro is often t hat of t h e As ant ehene himself. On the day of the
rites " t h e king is s me ar e d wi t h t he red esono d y e " (p. 136). In this
way t he whiteness of t he ntoro a n d t he ri ver Bosommur u is violated.
When t he shrine is l at er pur i fi ed, wat er from a number of sacred
Litninality and Communitas
125
rivers is mixed with white clay in a bowl, and the shrine is sprinkled
with it.
In many patrilineal societies, especially those with t he blood feud,
it is descent t hrough males that is associated with ambivalent blood
symbolism. But, in Ashanti, where matriliny is the domi nant
articulating principle, the male-to-male link of descent is regarded as
almost totally auspicious and connected with the Sky God and the
great river gods, who preside over fertility, health, strength, and all
the life values shared by everyone. Once more we meet with the
structurally inferior as the morally and ritually superior, and secular
weakness as sacred power.
LIMINALITY, LOW STATUS, AND COMMUNITAS
The time has now come to make a careful review of a hypothesis
that seeks to account for t he attributes of such seemingly diverse
phenomena as neophytes in the liminal phase of ritual, subjugated
autochthones, small nations, court jesters, holy mendicants, good
Samaritans, millenarian movements, " dh a r ma bums , " mat ri l at era-
lity in patrilineal systems, patrilaterality in matrilineal systems, and
monastic orders. Surely an ill-assorted bunch of social phenomena!
Yet all have this common characteristic: they are persons or prin-
ciples that ( 1 ) fall in the interstices of social structure, (2) are on its
margins, or (3) occupy its lowest rungs. This leads us back to the
problem of the definition of social structure. One authoritative
source of definitions is A Dictionary of the Social Sciences (Gould and
Kolb, 1 9 64) , in which A. W. Eister reviews some major formulations
of this conception. Spencer and many modern sociologists regard
social structure as " a more or less distinctive arrangement (of which
there may be more t han one type) of specialized and mutually
dependent institutions [Eister' s emphasis] and the institutional organ-
izations of positions and/ or of actors which they imply, all evolved in
the natural course of events, as groups of human beings, wi t h given
needs and capacities, have interacted wi t h each other (in various
126
The Ritual Process
types or modes of i nt er act i on) a n d sought to cope wi t h their environ-
me n t " (pp. 668-669). Ra y mo n d Fi r t h' s ( 1 95 1 ) mor e analytical con-
ception runs as follows: " I n t he t ypes of societies ordinarily studied
by anthropologists, t he social s t r uct ur e may i ncl ude critical or basic
relationships ari si ng si mi l arl y f r om a class system based on relations
wi t h t he soil. Ot he r aspect s of soci al st r uct ur e arise t hrough member-
ship in ot her ki nds of per si st ent gr oups , such as clans, castes, age-sets,
or secret societies. Ot h e r bas i c rel at i ons agai n ar e due to position in
a kinship s ys t em" ( p. 3 2) .
Most definitions cont ai n t h e not i on of an ar r angement of positions
or statuses. Most i nvol ve t he i nst i t ut i onal i zat i on and perdurance
of groups and r el at i onshi ps. Classical mechani cs, t he morphology
and physiology of ani mal s a n d pl ant s, and, mor e recently, with
Levi-Strauss, st r uct ur al l i ngui st i cs have been ransacked for con-
cepts, models, a n d homol ogous forms by social scientists. All share
in common t he not i on of a s uper or gani c ar r angement of part s or
positions t hat cont i nues, wi t h modi fi cat i ons mor e or less gradual,
t hrough t i me. Th e concept of " conf l i ct " has come to be connected
wi t h t he concept of "s oci al s t r uct ur e , " since t he differentiation of
part s becomes opposi t i on be t we e n part s, and scarce status becomes
t he object of struggles be t we e n persons and groups who lay claim
to it.
Th e other di mensi on of " s o c i e t y " wi t h whi ch I have been con-
cerned is less easy t o defi ne. G. A. Hi l l ery (1 95 5 ) reviewed g4
definitions of t he t e r m " c o mmu n i t y " and r eached t he conclusion
t hat "beyond t he concept t h a t peopl e ar e involved in communi t y,
t here is no compl et e a gr e e me n t as to t he nat ur e of communi t y"
(p. 1 1 9 ) . Th e field woul d, t her efor e, seem t o be still open for new
at t empt s! I have t ri ed t o es chew t he not i on t hat communi t as has a
specific territorial locus, often l i mi t ed in char act er , whi ch pervades
many definitions. For me , c ommuni t a s emerges wher e social struc-
t ure is not. Per haps t he bes t wa y of put t i ng this difficult concept
into words is Ma r t i n Bube r ' s t hough I feel t hat perhaps he should
be regarded as a gifted nat i ve i nfor mant r at her t han as a social
scientist! Buber ( 1 961 ) uses t he t e r m " c o mmu n i t y " for "com-
Liminality and Communitas
127
muni t as ": "Communi t y is the being no longer side by side (and,
one might add, above and below) but with one another of a mul t i t ude
of persons. And this mul t i t ude, though it moves towards one goal,
yet experiences everywhere a turning to, a dynamic facing of, the
others, a flowing from / to Thou. Communi t y is where community
happens" (p. 5 1 ) .
Buber lays his finger on the spontaneous, immediate, concrete
nature of communitas, as opposed to the norm-governed, institu-
tionalized, abstract nat ure of social structure. Yet, communitas is
made evident or accessible, so to speak, only through its j uxt a-
position to, or hybridization with, aspects of social structure. Just
as in Gestalt pyschology, figure and ground are mutually det ermi na-
tive, or, as some rare elements are never found in nat ure in their
purity but only as components of chemical compounds, so com-
munitas can be grasped only in some relation to structure. Just
because the communitas component is elusive, hard to pin down,
it is not uni mport ant . Here the story of Lao-tse's chariot wheel may
be apposite. The spokes of the wheel and the nave (i.e., the central
block of the wheel holding the axle and spokes) to which they are
attached would be useless, he said, but for the hole, the gap, the
emptiness at the center. Communitas, with its unstructured char-
acter, representing the " qui ck" of human interrelatedness, what
Buber has called das Zwischenmenschliche, might well be represented
by the "emptiness at the center, " which is nevertheless indispensable
to the functioning of the structure of the wheel.
It is neither by chance nor by lack of scientific precision t hat ,
along with others who have considered the conception of com-
munitas, I find myself forced to have recourse to met aphor and ana-
logy. For communitas has an existential quality; it involves the
whole man in his relation to other whole men. Structure, on the other
hand, has cognitive qual i t y; as Levi-Strauss has perceived, it is
essentially a set of classifications, a model for thinking about culture
and nature and ordering one' s public life. Communi t as has also an
aspect of potentiality; it is often in the subjunctive mood. Relations
between total beings are generative of symbols and metaphors and
128
The Ritual Process
compari sons; ar t and religion are t hei r p r o d u c t s r at her t han legal
and political structures. Bergson saw i n t h e wo r d s and writings of
prophet s and great artists t he creat i on of a n " o p e n mor al i t y, " which
was itself an expression of what he cal l ed t h e Han vital, or evolution-
ary "life-force. " Prophet s and artists t end t o b e l i mi nal and marginal
people, " e dge me n, " who strive wi t h a p a s s i o n a t e sincerity to rid
themselves of t he cliches associated wi t h s t a t u s i ncumbency and
role-playing and to ent er into vital r el at i ons w i t h ot her men in fact
or i magi nat i on. I n t hei r product i ons we ma y c a t c h glimpses of that
unused evolutionary pot ent i al in ma n ki n d w h i c h has not yet been
externalized and fixed i n st ruct ure.
Communi t as breaks in t hr ough t he i nt e r s t i ce s of structure, in
l i mi nal i t y; at t he edges of st ruct ure, i n ma r gi n a l i t y ; and from
beneat h st ruct ure, i n inferiority. I t is al mos t e ve r ywhe r e held to be
sacred or " hol y, " possibly because it t r ans gr e s s e s or dissolves the
norms t hat govern st ruct ured and i ns t i t ut i onal i ze d relationships and
is accompani ed by experiences of u n p r e c e d e n t e d pot ency. The
processes of "l evel i ng" and " s t r i ppi ng, " t o wh i c h Goffman has
dr awn our at t ent i on, often appear t o flood t h e i r subjects wi t h affect.
Inst i nct ual energies ar e surely l i berat ed by t h e s e processes, but I am
now inclined to t hi nk t hat communi t as is n o t solely t he product of
biologically i nheri t ed drives released f r om cul t ur al constraints.
Rat he r is i t t he pr oduct of peculiarly h u ma n facul t i es, whi ch include
rationality, volition, and memor y, a n d wh i c h devel op wi t h experi-
ence of life i n societyjust as among t h e Ta l l e n s i it is only mat ur e
men who under go t he experiences t h a t i n d u c e t he m to receive
bakologo shrines.
Th e not i on t hat t here is a generi c b o n d be t we e n men, and its
related sent i ment of " humanki ndne s s , " a r e n o t epi phenomena of
some ki nd of herd instinct but ar e pr oduct s of " me n in t hei r whole-
ness wholly at t endi ng. " Li mi nal i t y, ma r g i n a l i t y , and st ruct ural
inferiority are conditions in whi ch ar e f r e que n t l y generat ed myths,
symbols, rituals, philosophical systems, a n d wor ks of art . These
cul t ural forms provi de me n wi t h a set of t e mp l a t e s or model s which
are, at one level, periodical recl assi fi cat i ons of real i t y and man' s
Liminality and Communitas
129
relationship to society, nat ure, and culture. But they are more t han
classifications, since they incite men to action as well as to t hought .
Each of these productions has a multivocal character, having many
meanings, and each is capabl e of moving people at many psycho-
biological levels simultaneously.
There is a dialectic here, for the immediacy of communi t as gives
way to the mediacy of structure, while, in rites de passage, men are
released from structure into communi t as only to ret urn to structure
revitalized by their experience of communi t as. What is certain is
that no society can function adequately without this dialectic.
Exaggeration of structure may well lead to pathological manifesta-
tions of communitas outside or against " t h e l aw. " Exaggeration of
communitas, in certain religious or political movements of the
leveling type, may be speedily followed by despotism, overbureau-
cratization, or other modes of structural rigidification. For, like the
neophytes in the African circumcision lodge, or the Benedictine
monks, or the members of a millenarian movement, those living in
community seem to require, sooner or later, an absolute authority,
whether this be a religious commandment , a divinely inspired
leader, or a dictator. Communi t as cannot stand alone if the material
and organizational needs of human beings are to be adequately
met. Maximization of communi t as provokes maximization of struc-
ture, which in its t urn produces revolutionary strivings for renewed
communitas. The history of any great society provides evidence at
the political level for this oscillation. And t he next chapt er deals wi t h
two major examples.
I mentioned earlier t he close connection t hat exists between struc-
ture and property, whet her this be privately or corporately owned,
inherited, and managed. Thus, most millenarian movements try to
abolish property or to hold all things in common. Usually this is
possible only for a short timeuntil t he dat e set for t he coming of
the millennium or the ancestral cargoes. When prophecy fails,
property and structure ret urn and the movement becomes institu-
tionalized, or the movement disintegrates and its members merge
into the environing structured order. I suspect t hat Lewis Henry
130
The Ritual Process
Mor gan (1877) himself l onged for t h e comi ng of worl d-wi de com-
muni t as. For exampl e, in t he l ast s onor ous par agr aphs of Ancient
Society, he has this to say: " A me r e p r o p e r t y career is not t he final
destiny of manki nd, if progress is t o be t h e l aw of the future as it has
been of the past . . . t he di ssol ut i on of soci et y bids fair to become t he
t ermi nat i on of a career of whi ch p r o p e r t y is t he end and ai m;
because such a career cont ai ns t h e el ement s of self-destruction.
Democracy in government , b r o t h e r h o o d i n society, equal i t y in
rights and privileges, and uni ver s al e duc a t i on, foreshadow t he next
higher pl ane of society to whi ch e xpe r i e nce , intelligence and know-
ledge are steadily t e ndi ng" (p. 5 5 2).
What is this "hi gher p l a n e " ? I t is h e r e t hat Mor gan seemingly
succumbs to t he error made by s uch t hi nke r s as Rousseau and Ma r x:
t he confusion between communi t as , wh i c h is a dimension of all
societies, past and present, and a r c h a i c or pri mi t i ve society. " I t will
be a revi val , " he continues, " i n a h i gh e r for m, of t he liberty, equal i t y
and fraternity of t he anci ent ge n t e s . " Ye t , as most ant hropol ogi st s
woul d now confirm, cust omar y n o r ms a n d differences of status and
prestige in preliterate societies al l ow of l i t t l e scope for i ndi vi dual
liberty and choicethe i ndi vi dual i st is often regarded as a wi t ch;
for t rue equality bet ween, for e xa mpl e , me n and women, elders and
j uni ors, chiefs and commoner s ; wh i l e frat erni t y itself frequently
succumbs to t he sharp di st i nct i on of s t at us bet ween older and j uni or
sibling. Membershi p of ri val rous s e gme nt s i n such societies as the
Tallensi, Nuer, and Ti v does not al l ow even of t ri bal br ot her hood:
such membershi p commi t s t he i n di vi dua l t o st ruct ure and to t he
conflicts t hat are i nseparabl e from s t r uc t ur a l differentiation. How-
ever, even in t he simplest societies, t h e di st i nct i on bet ween st ruct ure
and communi t as exists and obt ai ns symbol i c expression i n the
cul t ural at t ri but es of l i mi nal i t y, ma r gi na l i t y, and inferiority. I n
different societies and at different pe r i ods i n each society, one or the
ot her of these "i mmor t al a n t a gon i s t s " (t o bor r ow t erms t hat Fr eud
used in a different sense) comes uppe r mos t . But t oget her t hey con-
stitute t he " h u ma n condi t i on, " as r e ga r ds man' s relations wi t h his
fellow man.
4
Communitas:
Model and Process
MODALITIES OF GOMMUNITAS
This chapter springs fairly naturally from a seminar I ran at Cornell
University with an interdisciplinary group of students and faculty,
on various aspects of what may be called t he meta-structural aspects
of social relations. I was reared in t he orthodox social-structuralist
tradition of British anthropology, which, to put a complex argument
with crude simplicity, regards a "soci et y" as a system of social
positions. Such a system may have a segmentary or a hierarchical
structure, or both. What I want to stress here is t hat t he units of
social structure are relationships between statuses, roles, and offices.
(Here, of course, I am not using "s t r uct ur e" in the sense favored by
Levi-Strauss.) The use of social-structural models has been ex-
tremely helpful in clarifying many dark areas of culture and society,
but, like other major insights, the structural viewpoint has become _
in the course of time a fetter and a fetish. Fd^kij^xpejdjneejjmd
gpnprpl rpaH i ng i n thr p r j g j mH ^ r mTa n i H p ^ me that t he
"social" i s_not i dent i cal wi t h t he "social-structural. " Ther e are
. ntli pr modal i t i es of social relationship. _^
Beyond the structural lies not only t he Hobbesian " wa r of all
against al l " but also communi t as, a mode of relationship already
recognized as such by our seminar. Essentially, comBHmitas__is__a_
relationship between concrete, historical, idiosyncratic individuals.
132
The Ritual Process
These individuals are not s ^ me n t a l i z e d i nt o roles^and^ statuses but
confront one anot her r at he r i n .the r n^
a n d Th o u . " Along wi t h this di r ect , i mme di a t e , and t ot al confronta-
tion of human identities, t he r e t ends t o go a j n o d e l of society as a
homogeneous, unst r uct ur ed c ommun i t a s , whose boundari es are"
ideally coterminous wi t h t hose of t he h u ma n species.) Communi t as
is in this respect strikingly di fferent from Dur kh e i mi a n "solida.ri.tyJ' -
t he force of whi ch depends upon an i n- gr oup/ out - gr oup contrast.
To some extent, communi t as is t o sol i dar i t y as He nr i Bergson's
" ope n mor al i t y" is to his "cl os ed mor a l i t y . " Bm_t h^^poj u^nei t y_
amLJmmedi ac_y_of_cai mmi ni ^ t h e j u r ^ p o l i t i c a l
character_j3f_sj:rurture^^c^ji__s_eidpjn_be j xi ai nt ai ne d for very long. j
Conmi uni t as itself soon devel ops a s t r uct ur e , in whi ch free relation-'
slugs bet ween individuals be come conve r t e d i nt o norm-governed
relaUonships_Jbetyy^eji_^ociai_p^r^on^. \ Th u s , it is necessary to
distinguish bet ween: ( i ) existential or spontaneous communi t as
approxi mat el y what t he hi ppi es t oday woul d call " a happeni ng, "
and William Blake mi ght have cal l ed " t h e wi nged mome nt as it
flies" or, later, " mut ua l forgiveness of e a c h v i c e " ; ( 2) normative
communi t as, where, unde r t he i nfl uence of t i me, t he need to
mobilize and .organize. r e s ^ur c e ^_a n dj h e necessi t y for social control
among_the_members of t he gr oup i n pur s ua n c e of these goals, the
existential. communitas is or gani zed i nt o a pe r dur i ng social system;
. and (3) ideological communi t as , whi ch is a l abel one can appl y to a
vari et yof . Ut opi an . model s , of - soci et i es. . based- on. existential com-
j nuni t as . '
Ideological communi t as is at once a n a t t e mpt t o describe the
external and visible effectsthe out wa r d for m, it mi ght be saidof
an i nward experience of exi st ent i al c ommun i t a s , and t o spell out
t he opt i mal social condi t i ons un de r wh i c h s uch experi ences might
be expected to flourish and mul t i pl y. Bot h nor ma t i ve and ideological
communi t as are al ready wi t hi n t he d o ma i n of s t r uct ur e, and it is
t he fate of all spont aneous c ommun i t a s i n hi st ory t o under go what
most people see as a " de cl i ne a n d fal l " i nt o s t r uct ur e a n d law. In
Communitas: Model and Process
*3 3
religious movements of t he communi t as type, it is not only t he
charisma of t he leaders t hat is "r out i ni zed" but also t he com-
munitas of their first disciples and followers. It is my intention to
trace the broad outlines of this widely distributed process with
reference to two well-known historical examples: t he early Fr an-
ciscans of medieval Europe, and the Sahajiyas of fifteenth- and
sixteenth-century Indi a.
Furthermore, structure tends to be pragmat i c and this-worldly;
while communitas is often speculative and generates imagery and
philosophical ideas. One example of this contrast, to which our
seminar gave a great deal of attention, is t hat kind of normat i ve
communitas t hat characterizes the liminal phase of tribal initiation
rites. Here there is usually a great simplification of social structure
in the British anthropological sense, accompani ed by a rich proli-
feration of ideological structure, in the form of myt hs and sacra,
in the Levi-Strauss sense. Rules t hat abolish mi nut i ae of structural
differentiation in, for example, t he domai ns of kinship, economics,
and political structure liberate t he human structural propensity
and give it free reign in the cultural real m of myt h, ritual, and
symbol. It is not tribal initiation, however, but t he genesis of
religious movements t hat concerns us heret hough bot h may
possibly be said to exhibit a " l i mi nal " character, in t hat they arise
in times of radical social transition, when society itself seems to be
moving from one fixed state to another, whet her t he terminus ad
quern is believed to be on eart h or in heaven.
In our seminar, also, we frequently came across instances, in
religion and literature, in whi ch normat i ve and ideological com-
munitas are symbolized by structurally inferior categories, groups,
types, or individuals, rangi ng from t he mot her' s brot her in patrilineal
societies, to conquered aut ocht honous peoples, Tolstoy' s peasants,
Gandhi' s harijans, and the " holy poor " or " God' s poor " of medieval
Europe. For example, today' s hippies, like yesterday' s Franciscans,
assume the attributes of t he structurally inferior in order to achieve
communitas.
Bilkent University
I- i hrarv
134
The Ritual Process
IDEOLOGICAL AND
SPONTANEOUS COMMUNITAS
The scattered clues and indications we have e n c o u n t e r e d in pre-
literate and prei ndust ri al societies of t he existence i n t h e i r cultures,
not abl y in liminality and st ruct ural inferiority, of t h e egalitarian
model we have called normat i ve communi t as, b e c o m e i n complex
and l i t erat e societies, bot h anci ent and moder n, a p o s i t i v e torrent
of explicitly formulated views on how men may bes t l i v e together in
comradely har mony. Such views may be called, a s w e have just
not ed, ideological communi t as. I n order to convey t h e wi de gener-
ality of these formulations of t he ideal structureless d o m a i n , I would
like to adduce, almost at r andom, evidence from s o u r c e s f ar removed
from one anot her i n space and t i me. I n these s our ces , b o t h religious
and secular, a fairly regul ar connection is ma i n t a i n e d between
liminality, st ruct ural inferiority, lowermost s t at us , a n d structural
outsiderhood on t he one hand, and, on t he ot he r , s u c h universal
huma n values as peace and har mony bet ween al l m e n , fertility,
heal t h of mi nd and body, universal j ust i ce, c o mr a d e s h i p and
brot herhood bet ween all men, t he equality before Go d , t he law or
t he life force of men and women, young and ol d, a n d pe r s on s of all
races and et hni c groups. And of especial i mp o r t a n c e i n all these
Utopian formulations is t he persisting adhesi on b e t w e e n equality
and absence of propert y. Take, .for exampl e, Go n z a l o ' s i deal com-
monweal t h i n Shakespeare' s Tempest (Act I I , S c e n e i , lines 1 41 -
163), i n whi ch Gonzal o addresses t he vi l l ai nous An t o n i o and
Sebastian t hus :
Gonzalo:
V the commonwealth I would by contraries
Execute all things; for no kind of traffic
Would I admit; no name of magistrate;
Letters should not be known; riches, poverty
And use of service, none; contract, succession
Bourn, bound of land, tilth, vineyard, none;
No use of metal, corn, or wine, or oil;
Communitas: Model and Process
*35
No occupation; all men idle, all;
And women too, but innocent and pure;
No sovereignty;
Sebastian:
Yet he would be king on't.
Antonio:
The latter end of his commonwealth forgets the beginning.
Gonzalo:
All things in common nature should produce
Without sweat or endeavour; treason, felony,
Sword, pike, knife, gun, or need of any engine,
Would I not have; but nature should bring forth,
Of its own kind, all foison, all abundance,
To feed my innocent people.
Sebastian:
No marrying 'mong his subjects?
Antonio:
None, man; all idle; whores and knaves.
Gonzalo :
I would with such perfection govern, sir,
To excel the golden age.
Gonzalo' s commonweal t h has many attributes of communi t as.
Society is seen as a seamless and structureless whole, rejecting alike
status and contractthese evolutionary poles of Sir Henry Mai ne' s
entire system of societal developmenteschewing private property,
with its bourns and bounds of land, tilth, and vineyard, and relying
on nature' s bount y to supply all needs. Here he is, of course, rat her
meretriciously accommodat ed by t he Cari bbean setting; in more
spartan circumstances, men would have had to work if only to keep
warm. Thus he circumvents the crucial difficulty of all Utopias
that they have to produce life's necessities t hrough workin
economists' j argon, to mobilize resources. To mobilize resources
also means to mobilize people. Thi s implies social organization,
with its " e n d s " and " me a n s " and necessary "deferment of grati-
fications," and all these entail the establishment, however transient,
of orderly structural relations between man and man. Since, under
136
The Ritual Process
these conditions, some must initiate and c o mma n d , a n d ot h e r s must
respond and follow, a system for the pr oduct i on a n d di s t r i but i on of
resources contains within it the seeds of s t r uct ur al s e g me n t a t i o n and
hierarchy. Gonzal o gets around this awkwar d fact b y as s umi ng an
incredible fecundity of nat ureand thereby i n d i c a t i n g t h e absurdi t y
of his whole noble dr eam. Shakespeare also, as h e of t e n does, puts
valid argument s into the mouths of less t han wo r t h y c h a r a c t e r s when
he makes Sebastian say, " Ye t he would be ki ng o n ' t . " H e r e we may
be able to detect the intuition that whenever a p e r f e c t equal i t y is
assumed in one social dimension, it provokes a p e r f e c t i nequal i t y
in anot her.
A final communi t as value stressed by Go n z a l o is t h a t of the
innocence and puri t y of those who live wi t hout s o v e r e i g n t y . We have
t he assumption here, later to be devel oped mo s t e l a bor a t e l y by
Rousseau, of the nat ural goodness of h u ma n b e i n g s l i vi ng in a
propertyless, structureless state of abs ol ut e e q u a l i t y . Indeed,
Gonzalo suggests t hat among his i nnocent pe opl e t h e r e woul d be
no treason, felony, sword, pike, knife, gun wi t h w h i c h h e appears
to equat e the need of any engine, as t hough wa r , c onf l i c t , or indeed
any "pol i t i cki ng" were necessarily connect ed wi t h t echnol ogy,
even of t he most rudi ment ary sort.
Gonzalo' s commonweal t h cleaves closer t h a n a l mo s t a n y other
type of ideological communi t as to what Bu b e r ( 1 9 5 8 , 1961) has
called "das Zwischenmenschliche," or s pont aneous c o mmu n i t a s . When
Buber uses t he t erm " c ommun i t y " he is not , i n t h e first place,
talking about persisting social groups wi t h i n s t i t ut i on a l i z e d struc-
tures. He does believe, of course, that such gr oups c a n b e founded
in communi t y, and t hat some types of gr oups , l i ke t h e kvuzoth and
kibbutzim of Israel, best preserve its spirit. Ye t , for Bu b e r , com-
muni t y is quintessentially a mode of r e l at i ons hi p b e t we e n total
and concrete persons, between "I" and "Thou." T h i s r el at i onshi p
is always a " happe ni ng, " something t hat arises i n i n s t a n t mut ual i t y,
when each person fully experiences the bei ng of t h e o t h e r . As Buber
(1961) put s i t : " On l y when I have to do wi t h a n o t h e r essentially,
t hat is, in such a way t hat he is no longer a p h e n o me n o n of my / ,
Communitas: Model and Process
J37
but instead is my Thou, do I experience reality of speech wi t h
anotherin the irrefragable genuineness of mut ual i t y" (p. 7 2) .
But Buber does not restrict communi t y to dyadic relationships. He
also speaks of an " essential We" by which he means " a communi t y
of several i ndependent persons, who have a self and self-respon-
sibility. . . The We includes the Thou. Onl y men who are capable
of truly saying Thou to one anot her can truly say We with one
another. . . . No part i cul ar kind of group-formation as such can be
adduced as an example of the essential We, but in many of t hem
the variety which is favourable to the arising of the We can be seen
clearly enough. . . . It is enough to prevent the We arising, or being
preserved, if a single man is accepted, who is greedy of power and
uses others as a means to his own end, or who craves of i mport ance
and makes a show of himself " (pp. 2 1 3 - 2 1 4 ) .
In this and other similar formulations, Buber makes it clear t hat
the "essential We" is a transient, if highly potent, mode of relation-
ship between integral persons. To my mi nd, the "essential We"
has a liminal character, since perdurance implies institutionalization
and repetition, while communi t y (which roughly equals spontaneous
communitas) is always completely uni que, and hence socially t ran-
sient. At times Buber appears to be misled about t he feasibility of
converting this experience of mut ual i t y into structural forms. Spon-
taneous communi t as can never be adequately expressed in a
structural form, but it may arise unpredi ct abl y at any time between
human beings who are institutionally reckoned or defined as mem-
bers of any or all kinds of social groupings, or of none. Just as i n
preliterate society the social and individual developmental cycles
are punct uat ed by more or less prolonged instants of ritually guarded
and stimulated liminality, each with its core of potential com-
munitas, so the phase structure of social life in complex societies is
also punct uat ed, but wi t hout institutionalized provocations and
safeguards, by i nnumerabl e instants of spontaneous communi t as.
In preindustrial and early industrial societies wi t h multiplex
social relations, spontaneous communitas appears to be very fre-
quently associated with mystical power and to be regarded as a
138
The Ritual Process
charism or grace sent by t he de i t i e s or ancest ors. Nevertheless, by
impetrative ritual means , a t t e mp t s a r e ma de , mostly in the phases
of liminal seclusion, to cause t h e de i t i e s or ancestors to bring this
charism of communi t as a mo n g me n . But t her e is no specific social
form t hat is held to express s p o n t a n e o u s communi t as. Rat her is it
expected best to arise i n t he i n t e r va l s bet ween incumbencies of
social positions and statuses, i n w h a t us e d t o be known as " t he inter-
stices of the social s t r uct ur e . " I n c o mp l e x industrialized societies,
we still find traces in t he l i t ur gi es of chur ches and other religious
organizations of i nst i t ut i onal i zed a t t e mp t s t o pr epar e for the coming
of spontaneous communi t as . T h i s moda l i t y of relationship, how-
ever, appears to flourish best i n s pont a ne ous l y liminal situations
phases betwixt and bet ween s t at es w h e r e social-structural role-playing
is domi nant , and especially b e t we e n s t at us equal s.
Some at t empt s have be e n ma d e fai rl y recently in America and
Western Europe to r e- cr eat e t h e r i t ua l conditions under which
spontaneous communi t as ma y b e , d a r e one say it, invoked. The
beats and t he hippies, by t he e cl e ct i c a n d syncretic use of symbols
and liturgical actions dr a wn f r om t h e r eper t oi r e of many religions,
and of " mi nd- e xpa ndi ng" dr ugs , " r o c k " musi c, and flashing lights,
try to establish a " t o t a l " c o mmu n i o n wi t h one anot her. This, they
hope and believe, will e nabl e t h e m t o r each one anot her through
t he "dereglement ordonne de tous les j e a r , " i n t ender, silent, cognizant
mut ual i t y and in all concr et enes s . T h e ki nd of communi t as desired
by tribesmen in t hei r ri t es a n d b y hi ppi e s in their "happeni ngs "
is not t he pl easurabl e a n d effort l ess comr ades hi p t hat can arise
between friends, coworkers, or pr ofes s i onal colleagues any day. What
they seek is a t ransformat i ve e x p e r i e n c e t hat goes to the root of
each person' s bei ng a n d finds i n t h a t r oot something profoundly
communal and shared.
The often made et ymol ogi cal h o mo l o g y bet ween t he nouns " exist-
e nce " and " e cs t as y" is pe r t i ne nt h e r e ; t o exist is to "s t and outside"
i.e., to stand outside t he t ot al i t y of s t r uc t ur a l positions one normally
occupies in a social syst em. T o exi s t is t o be in ecstasy. But, for the
hippiesas indeed for ma n y mi l l e n a r i a n a n d "ent husi ast i c" move-
Communitas: Model and Process
39
m e n t s
t h e ecstasy of spontaneous communitas is seen as the end of
human endeavor. In the religion of preindustrial societies, this state
is regarded rather as a means to the end of becoming more fully
involved in the rich manifold of structural role-playing. In this
there is perhaps a greater wisdom, for human beings are responsible
to one another in the supplying of humble needs, such as food, drink,
clothing, and the careful teaching of material and social techniques.
Such responsibilities imply the careful ordering of human relation-
ships and of man' s knowledge of nat ure. There is a mystery of
mutual distance, what the poet Rilke called " t he circumspection
of human gesture," which is just as humanly i mport ant as the
mystery of intimacy.
Once more we come back to the necessity of seeing man' s social
life as a process, or rat her as a multiplicity of processes, in which
the character of one type of phasewhere communitas is paramount
differs deeply, even abyssally, from that of all others. The great
human temptation, found most prominently among Utopians, is to
resist giving up the good and pleasurable qualities of that one phase
to make way for what may be the necessary hardships and dangers
of the next. Spontaneous communitas is richly charged with affects,
mainly pleasurable ones. Life in "s t r uct ur e" is filled with objective
difficulties: decisions have to be made, inclinations sacrificed to the
wishes and needs of the group, and physical and social obstacles
overcome at some personal cost. Spontaneous communitas has
something "magi cal " about it. Subjectively there is in it the feeling
of endless power. But this power untransformed cannot readily be
applied to the organizational details of social existence. It is no
substitute for lucid thought and sustained will. On the other hand,
structural action swiftly becomes arid and mechanical if those in-
volved in it are not periodically immersed in the regenerative abyss
of communitas. Wisdom is always to find the appropri at e relation-
ship between structure and communitas under the given circum-
stances of time and place, to accept each modality when it is par a-
mount without rejecting the other, and not to cling to one when its
present impetus is spent.
140
The Ritual Process
Gonzalo' s commonweal t h, as Shakespeare appears i r oni cal l y t o
indicate, is an Edeni c fantasy. Spontaneous communi t as is a p h a s e ,
a moment , not a per manent condition. Th e moment a di ggi ng s t i c k
is set in the eart h, a colt broken in, a pack of wolves defended a g a i n s t ,
or a human enemy set by his heels, we have t he germs of a s oc i a l
structure. Thi s is not merely t he set of chains in whi ch me n e v e r y -
where are, but the very cul t ural means t hat preserve t he di gni t y a n d
liberty, as well as t he bodily existence, of every man, wo ma n , a n d
child. Ther e may be manifold imperfections in t he s t r uct ur al me a n s
employed and t he ways in whi ch they are used, but , s i nce t h e
beginnings of prehistory, t he evidence suggests t hat such me a n s a r e
what makes man most evidently man. Thi s is not to say t h a t s p o n -
taneous communi t as is merel y " n a t ur e . " Spont aneous c o mmu n i t a s
is nat ure in dialogue wi t h st ruct ure, marri ed to it as a w o ma n is
marri ed to a man. Toget her t hey make up one st ream of life, t h e o n e
affluent supplying power, t he ot her alluvial fertility.
FRANCISCAN POVERTY AND COMMUNITAS
Between Gonzalo' s commonweal t h and models of closely i n t e g r a t e d
structural systems lies an abundance of ideal social forms. At t i t u d e s
to propert y distinguish t he communi t as set of models from t h e mo r e
empirically oriented models, whi ch combi ne in varyi ng p r o p o r t i o n s
components of t he communi t as type wi t h a clear r ecogni t i on of t h e
organizational advant ages of institutionalized st ruct ures. I t is e s s e n-
tial to distinguish bet ween t he ideal models of communi t as p r e s e n t e d
in literature or procl ai med by founders of movement s or a c t ua l c o m-
munities, and t he social process t hat results from ent hus i as t i c a t -
tempts by the founder and his followers to live in a c c or da nc e wi t h
these models. It is only by studying social fields, of what e ve r d o mi n -
ant character, over time, t hat one can become awar e of t h e i l l u-
mi nat i ng nuances of behavi or and decision t hat t hr ow i nt o l i gh t
the developmental st ruct ure of the relationship bet ween i de al a n d
praxis, existential communi t as and normat i ve communi t as .
Communitas: Model and Process
141
One of the great classic instances of such a development may be
found in the history of the Franciscan order of the Catholic Church.
M. D. Lambert , in his recent book Franciscan Poverty (1961) which
draws on the major pri mary and secondary sources of Franciscan
history and doctrine, has made an admi rabl y lucid reconstruction
of the course of events t hat flowed from t he at t empt of St. Francis t o
live, and to encourage others to live, in terms of a certain view of
poverty. He examines t he vicissitudes over time of t he group t hat
Francis founded, in their relationship to t he structured Church and,
implicitly, to the environing secular society. In so doing, he reveals
a processual paradi gm of the fate of spontaneous communi t as when
it enters social history. Subsequent movements, bot h religious and
secular, tend to follow, at varying tempi, the pat t ern of Francis-
canism in its dealings with t he world.
COMMUNITAS AND SYMBOLIC TH OUGH T
The gist of Lambert ' s cautious deductions about Francis' s own way
of thinking and his ideas on poverty goes something like this. I n t he
first placeand here Francis is like many other founders of com-
munitas-type groups"hi s thought was always i mmedi at e, personal
and concrete. Ideas appeared to hi m as images. A sequence of
thought for him . . . consists of leaping from one picture to t he
next . . . . When, for instance, he wishes to explain his way of living
to Pope Innocent I I I he turns his plea into a par abl e; on other
occasions, when he wishes t he brothers to underst and his intentions,
he chooses to do so by symbols. The luxury of t he brothers' table is
demonstrated by Francis disguised as a poor stranger. The wicked-
ness of touching money is conveyed by an acted parabl e imposed on
an offender by Francis as a penance" (p. 33). This concrete, per-
sonal, imagist mode of thinking is highly characteristic of those in
love with existential communitas, wi t h t he direct relation between
man and man, and man and nat ure. Abstractions appear as hostile to
live contact. William Blake, for example, a great literary exponent
142
The Ritual Process
of communi t as in his Prophetic Books, wr ot e t h a t " h e woul d do good
to others must do it in Mi nut e Par t i cul ar s ; Ge n e r a l Good is t he
plea of t he Hypocri t e and Scoundr el . "
Again, like ot her seers of c ommuni t a s a n c i e n t a n d moder n
Francis made several crucial decisions on t he bas i s of dr e a m sym-
bolism. For example, before he deci ded t o r esi gn t h e official l eader-
ship of t he Or der in 1220, he " d r e a mt of a l i t t l e b l a c k hen, whi ch,
try as she might, was too small t o cover al l h e r br ood wi t h her
wi ngs. " A little later, his deficiencies as a l egi s l at or wer e revealed
to hi m in anot her dr eam, in whi ch h e " t r i e d v a i n l y t o feed his
starving brothers wi t h cr umbs of br e a d t h a t s l i ppe d t hr ough his
fingers" (p. 34). I t was no doubt t he ver y concr e t e ne s s of his t hi nk-
ing and, if we were in full possession of t he facts a b o u t his social field
environment, t he multivocality of his s ymbol i s m t h a t ma d e Franci s
a poor legislator. Th e creation of a social s t r uc t ur e , especi al l y wi t hi n
t he prot obureaucrat i c frame of t he Ro ma n eccl es i a, woul d have
demanded a propensity for abst r act i on a n d ge ne r al i zat i on, a skill in
t he production of univocal concept s, a n d a ge n e r a l i z i n g foresighted-
ness; and these would have r un count e r t o t h e i mme di a c y , spon-
taneity, and, indeed, downr i ght ear t hi nes s of Fr anci s ' s vision of
communi t as. Besides, Francis, like ot her s befor e a n d aft er hi m, was
never able to overcome t he numer i cal l i mi t at i ons t h a t seem to be
set upon groups t hat maxi mi ze exi st ent i al c o mmu n i t a s . " Fr anci s
was a supreme spiritual mast er of small gr oups : b u t h e was unabl e
to provide t he impersonal or gani zat i on r e qui r e d t o mai nt ai n a
world-wide or de r " (p. 36).
Recently, Mar t i n Buber (1966) has conf r ont ed t hi s pr obl e m and
has argued t hat " a n organi c c o mmo n we a l t h a n d onl y such com-
monweal t hs can j oi n together to for m a s ha pe l y a n d ar t i cul at ed
race of menwill never build itself u p out of i ndi vi dual s , but only
out of small and even smaller communi t i e s : a n a t i o n is a communi t y
to t he degree t hat it is a communi t y of c o mmu n i t i e s " ( p. 136). He
thus proposes to ci rcumvent t he pr obl e m i mpos e d on Fr anci s of
drawi ng up in advance a detailed cons t i t ut i on b y a l l owi ng his com-
muni t y of communities to struggle i nt o gr a dua l c oh e r e n c e . Thi s is t o
Communitas: Model and Process
143
be achieved by "gr eat spiritual t act " informing a relationship
between centralism and decentralization and between idea and
reality"with the constant and tireless weighing and measuring
of the right proportion between t h e m" (p. 137).
Buber, in short, wishes to preserve t he concreteness of communitas
even in the larger social units, in a process he regards as analogous
to organic growth, or to what he has called " t he life of dialogue. "
Centralizationbut only so much as is indispensable in the given conditions
of time and place. And if the authorities responsible for the drawi ng and
re-drawing of lines of demarcation keep an alert conscience, the relations
between the base and the apex of the power-pyramid will be very different
from what they are now, even in states that call themselves Communist,
i.e., struggling for community. There will have to be a system of representa-
tion, too, in the sort of social pattern I have in mi nd; but it will not, as now,
be composed of the pseudorepresentatives of amorphous masses of electors
but of representatives well tested in the life and work of the communes. The
represented will not, as they are today, be bound to their representatives by
some windy abstraction, by the mere phraseology of a party-programme,
but concretely, through common action and common experience (p. 1 3 7 ) .
Buber's phraseology, which strikingly recalls that of many African
leaders of one-party states, belongs to the perennial speech of
communitas, not rejecting the possibility of structure, but conceiving
of it merely as an outgrowth of direct and i mmedi at e relations
between integral individuals.
Unlike Buber, St. Francis, as a member of the Catholic Church,
was under the obligation of maki ng a Rul e for his new fraternity.
And, as Sabatier (1905) said, " Never was man less capable of maki ng
a Rule t han Fr anci s " (p. 253). His Rul e was in no sense a set of
ethical or legal prescriptions and prohibitions; rather, it was a con-
crete model for what he considered should be the total "vitafratrum
minorum." Elsewhere (see Tur ner , 1967, pp. 98-99), I have stressed
the importance for liminarsas persons undergoing ritualized t ran-
sitions may be termedof doing wi t hout property, structural status,
privileges, material pleasures of various kinds, and often even cloth-
ing. Francis, who conceived of his friars as liminars in a life t hat was
144
The Ritual Process
merely a passage to t he u n c h a n g i n g st at e of heaven, laid great
emphasis on t he i mpl i cat i ons of b e i n g " wi t h o u t " or " not having. "
Thi s may be best expressed i n L a mb e r t ' s succi nct formulation of
Francis' s posi t i on"spi r i t ual d e n u d a t i o n . "
Francis himself t hought i n t e r ms of pover t y, celebrated by him
in t r oubadour fashion as " M y L a d y Pove r t y. " As Lamber t writes,
" We can accept it as a n a xi om t h a t t h e mor e r adi cal t he version of
poverty t hat is present ed t o us, t h e mo r e likely it is t o reflect the true
wishes of Franci s. " He goes on t o s ay " t h a t t he Rul e of 1221, taken
as a whole, gives t he i mpr es s i on t h a t Fr anci s wi shed his friars to cut
adrift entirely from t he c o mme r c i a l s ys t em of t he world. He is in-
sistent, for exampl e, t hat t h e n e e d for gi vi ng advi ce to postulants
about t he disposal of t hei r goods s h oul d not involve the brothers
in secular busi ness" (p. 38). I n Ch a p t e r 9 of t he Rul e he tells the
brot hers t hat they shoul d rej oi ce " w h e n t hey find themselves among
mean and despised persons, a mo n g s t t h e poor a n d weak and infirm
and t he lepers and t hose t h a t b e g i n t he s t r e e t " (Boehmer, 1904,
p. 10). Francis, in fact, hol ds cons i s t ent l y t hat t he poverty of
Franciscans should be t ake n t o t h e l i mi t s of necessity.
A detailed exampl e of t hi s p r i n c i p l e ma y be found in t he prohi-
bition of money to t he friars. " An d i f we shoul d find coins anywhere
let us pay no mor e at t ent i on t o t h e m t h a n t o t he dust t hat we tread
under our feet " (Boehmer, 1904, p . 9). Al t hough Francis here uses
t he t er m denarius, an act ual coi n, for " mo n e y , " elsewhere he equates
denarius wi t h pecunia, " a n y t h i n g t h a t pl ays t he rol e of money. " This
equivalence involves a r adi cal wi t h d r a wa l from t he world of buying
and selling. I t went far be yond t h e " p o v e r t y " r ecommended by the
older religious orders, for t he y st i l l r e t ai ne d t hei r communities to
some extent wi t hi n t he f r ame wor k of t h e secul ar economic system.
Francis by his Rul e i ns ur ed t h a t , as La mbe r t says, " t h e normal
sources for t he mai nt e nance of life we r e of a del i berat el y transient
and insecure nat ur e : t hey cons i s t ed i n r ewar ds i n ki nd for menial
l abour outside t he set t l ement s e ke d o u t wi t h t he product s of begging
expedi t i ons. " [Parallels wi t h t h e be h a vi or of t he hippies of the
Hai ght -Ashbury communi t y i n S a n Fr anci s co will no doubt leap
Communitas: Model and Process
H5
to the mind of the modern American reader!] " Th e Rul e of 1221
forbids the friars to take posts of authority. . . . Early followers, like
Brother Giles, always took on irregular tasks, such as grave-digging,
basket-weaving, and water-carrying, none of t hem providing any
security in times of deart h. The prescribed met hod of begging, pass-
ing indiscriminately from door to door . . . precluded mitigation of
the instability t hrough recourse to wealthy, regular pat r ons "
(pp. 4-I-4
2
)-
FRANCIS AND PERMANENT LIMINALITY
In all this, Francis appears quite deliberately to be compelling the
friars to inhabit the fringes and interstices of the social structure of
his time, and to keep t hem in a permanent l y liminal state, where, so
the argument in this book would suggest, the optimal conditions
inhere for t he realization of communi t as. But, in keeping with his
habit of thinking in "pr i mar y, visual images, " Francis nowhere
defined in unambi guous j ur al terms what he meant by poverty and
what this entailed with regard to property. For hi m t he ideal model
of poverty was Christ. For example, in t he Rul e of 1221, he said
of the friars:
And let them not be ashamed, but rather remember that our Lord Jesus
Christ, the Son of the living omnipotent God, set his face as the hardest
flint, and was not ashamed to be made a poor man and a stranger for us
and lived on alms, himself and the blessed Virgin and his disciples (Boehmer,
pp. 10, 1 1 , lines 6- 1 0 ) .
According to Lamber t :
The key figure in Francis's mind . . . is the image of the naked Christ. . . .
Nakedness was a symbol of great importance for Francis. He used it to
mark the beginning and the end of his converted life. When he wished to
repudiate his father's goods and enter religion, he did so by stripping him-
self naked in the bishop's palace at Assisi. At the end, dying in the Portiun-
cula, he forced his companions to strip him so that he could face death
unclothed on the floor of the hut. . . . When he slept it had to be on the
146
The Ritual Process
naked ground. . . . Twice he even chose t o a b a n d o n the table of the friars
to sit on the naked ground to eat his meal , i mpel l ed on each occasion by the
thought of Christ's poverty (p. 61 ) .
Nakedness represented povert y, a n d p o v e r t y t he literal absence of
property. Francis declared t hat j us t a s Ch r i s t and the apostles re-
nounced material goods in or der t o c a s t t hems el ves on providence
and live on alms, so should t he fri ars. As L a mb e r t points out, " the
one apostle who did not do this, a n d w h o k e p t a reserve in his bag,
was t he betrayer, J udas " (p. 66) .
Th e poverty of Christ cl earl y h a d " i mme n s e emotional signi-
fi cance" for Francis, who r e gar de d n a k e d n e s s as t he master symbol
of emanci pat i on from st r uct ur al a n d e c o n o mi c bondageas from
t he constraints set upon hi m by hi s e a r t h l y father, the wealthy
mer chant of Assisi. Religion for h i m wa s c ommun i t a s , between man
and God and man and man, ver t i cal l y a n d hori zont al l y, so to speak,
and poverty and nakedness wer e b o t h expr es s i ve symbols of com-
muni t as and i nst rument s for a t t a i n i n g i t . B u t hi s imaginative notion
of poverty as t he absolute pover t y of Ch r i s t was hard to sustain in
practice by a social gr oup forced b y t h e Ch u r c h to institutionalize
its arrangement s, routinize not me r e l y t h e c h a r i s ma of its founder but
also t he communi t as of its s pont ane ous be gi n n i n gs , and formulate in
precise legal terms its collective r e l a t i o n s h i p t o povert y. Propert y and
st ruct ure are undisseverably i n t e r r e l a t e d, a n d t he constitutions of
persisting social units i ncor por at e b o t h d i me n s i o n s as well as t he core
values t hat legitimatize t he exi st ence a n d f or ms of bot h.
As t he Franciscan Or de r e n d u r e d , i n t i me it developed in t he
direction of becomi ng a st r uct ur al s y s t e m, a n d as i t did so, t he heartfelt
simplicity of Francis' s formul at i ons o n p r o p e r t y in t he original Rul e
gave way to more legalistic def i ni t i ons . I n fact he had given two
laconic instructions only, i n his first R u l e of 1 2 2 1 and in his revised
Rul e of 1 2 2 3 . I n t he former, he s ays obl i que l y i n a chapt er con-
cerned primarily wi t h t he ma n u a l l a b o r of t he friars and wi t h
reference to t he ownershi p of t h e i r s e t t l e me nt s onl y: " Le t t he
brot hers be careful, wherever t hey m a y b e , i n hermi t ages or in ot her
settlements, not to appr opr i at e a s e t t l e me n t t o themselves or mai nt ai n
Communitas: Model and Process
147
it against anyone" (Boehmer, pp. 8, 1 1 , lines 5- 7). In 1 223 this was
extended: " Le t the brothers appropri at e nothing to themselves,
neither a house nor a settlement nor anyt hi ng. " One might think
this quite unequivocal, but any developing structure generates
problems of organization and values t hat provoke redefinition of
central concepts. This often seems like temporizing and hypocrisy,
or loss of faith, but it is really no more t han a reasoned response to
an alteration in the scale and complexity of social relations, and with
these, a change in t he location of t he group in t he social field it
occupies, with concomitant changes in its major goals and means
of attaining t hem.
SPIRITUALS VERSUS CONVENTUALS:
CONCEPTUALIZATION AND STRUCTURE
From the first t he Franciscan Or der burgeoned, and within a few
decades from t he deat h of its founder, we find t he bret hren in many
parts of Italy, Sicily, France, Spain and even undert aki ng missionary
journeys to Armeni a and Palestine. Fr om t he first, too, t he poverty
and vagabondageindeed, t he enthusiasmof the friars had caused
them to be viewed wi t h suspicion by t he secular clergy, organized in
local divisions, such as sees and parishes. Under these circumstances,
as Lambert points out, Francis' s idea of povertywhich, as we have
seen, is associated wi t h existential communitasis "so extreme as to
cause immense difficulties as soon as it has to be applied, not to a
band of wanderi ng friars, but to a developing order wi t h its prob-
lems of dwelling places, learning, sick friars, and the l i ke" (p. 68).
More difficult t han these were t he problems of structural continuity
concerning t he mani pul at i on of resources t hat threw into sharp relief
the question of the nat ure of property. This latter question became
almost an obsession with the order duri ng the century following
Francis's deat h, and resulted i n its division into two major branches
one might even call t hem camps or factions: the Conventuals,
who in practice relaxed the rigor of Francis' s ideal, and t he Spirituals,
148
The Ritual Process
who, wi t h their doct ri ne oCusus pauper, practiced, if a n y t h i n g ,
a
more
severe observance t han t hei r founder.
To ant i ci pat e a little, it is significant t hat ma n y of t h e l e ade r s of
the Spirituals had close contacts with J oachi mi s m, a mi l l e n a r i a n
movement based on t he genui ne and spurious wo r k s o f a twelfth-
century Cistercian abbot , J oachi m of Flora. It is c u r i o u s h o w often
in history notions of cat ast rophe and crisis are c o n n e c t e d w i t h what
one mi ght call "i ns t ant communi t as . " Perhaps i t i s n o t r e al l y so
curious, for clearly if one anticipates t he swift c o mi n g o f t h e worl d' s
end, t here is no point in legislating into exi s t ence a n el abor at e
system of social institutions designed to resist t he i n c u r s i o n s of time.
One is t empt ed to speculate about t he r e l at i ons hi p b e t w e e n the
hippies and t he Hydr ogen bomb.
But at first this cleavage in t he order had not b e c o m e pal pabl e ,
t hough every t hi ng favored devel opment away f r o m t h e pristine
poverty of Francis. As Lambe r t wri t es:
The influence of successive popes was naturally enough di r e c t e d towards
making the Franci scans, like the rival order of Domi ni cans, a fi tti ng instru-
ment of policy, both spiritual and political. For that pur pos e a n extreme
poverty tended generally to be an encumbrance. Benef actors from the
outside world, who were attracted by the austerity of Fr a nc i s c a n poverty,
played their part in weakeni ng it by donations often di ffi cult t o refuse. The
friars themselves, the only true guardi ans of their own obs e r v anc e , too often
were not sufficiently concerned to protect their poverty ag ai ns t those out-
side who from the highest motives wished to lighten thei r bur de n. Indeed it
was the members of the order above all, and not any f i g ures , however
exalted, from the outside world, who were responsible for t h e evoluti on of
the Franci scan ideal whi ch in the first twenty years carri ed t h e brothers so
far and so fast from the primitive life of Franci s and his c ompani ons (p. 70).
Interestingly, for several years before his de at h, F r a n c i s h a d relin-
quished t he government of his order and spent his t i me l a r g e l y in the
company of a small circle of compani ons in h e r mi t a g e s i n TJmbria
and Tuscany. A man of direct and i mmedi at e r e l a t i o n s h i p s , com-
muni t as, for hi m, had always to be concret e a n d s p o n t a n e o u s . He
Communitas: Model and Process
H9
may even have been dismayed at t he success of his own movement ,
which already in his lifetime had shown signs of t he structuring and
routinization it was to undergo under the influence of successive
generals and under t he external formative force of a series of papal
bulls. Francis' s very first successor, Elias, was what Lamber t calls
"t he essential organizing figure which in so many religious societies
has translated t he lofty ideals of their founders into terms acceptable
to later followers" (p. 74). It is significant t hat it was Elias who was
the driving force behi nd t he building of the large basilica at Assisi
to house St. Francis' s body, for which good office t he municipality
of Assisi in 1937 erected a memorial to hi m. As Lamber t says, " h e
made a more lasting contribution to the development of the city
then ever he did to t he evolution of t he Franciscan i deal " (p. 74).
With Elias, structure, bot h material and abstract, had begun to
replace communitas.
As the new order grew in numbers and spread across Europe, it
developed all the technical apparat us of vows and superiors along
with the quasi-political structure characteristic of religious orders
of the timesand, indeed, of later times. Thus, in their centralized
government, the friars had a minister general at the top and below
him a number of provincials, each of whom was the superior of a
provincei.e., the division of a religious order that comprised all its
houses and members in a given district; its territorial boundaries
were very often but not necessarily coterminous with those of a civil
state. The provincial was responsible to his superior general for the
administration of his province and for the mai nt enance of religion
in it, chiefly by means of visitations. He convened t he provincial
chapter and was a member of the general chapt er of t he order. Both
types of chapter had legislative, disciplinary, and elective functions.
Among the Franciscans some of t he provinces, for example, were
Provence, the Mar ch of Ancona, Genoa, Aragon, Tuscany, and
England. Anthropologists who have studied centralized political
systems bot h in preliterate and feudal societies will find little diffi-
culty in realizing the possibilities of structural opposition inherent
within such a hierarchy. Furt hermore the Franciscans were exempt
15
The Ritual Process
religious who were subject to t hei r own s u p e r i o r s onl y, not to the
local ordi nary (i.e., to clerics wi t h or di na r y j u r i s d i c t i o n i n the exter-
nal forum over a specified territory, s uch as b i s h o p s i n t hei r dioceses).
I n fact they were responsible directly a n d n o t me d i a t e l y to t he papacy.
St ruct ural conflict was therefore possible b e t w e e n t h e or der and the
secular clergy.
Rivalries also existed wi t h ot her or der s , a n d di s put e s bet ween the
Franciscans and Domi ni cans on poi nt s of t h e o l o g y a n d organization,
as well as for influence wi t h t he papacy, we r e p r o mi n e n t features of
medieval church history. And, of course, t h e effect i ve social field of
t he Franciscan Or der was not l i mi t ed t o t h e c h u r c h but contained
many profane, political influences. For e x a mp l e , on e is struck, in
readi ng Lamber t ' s narrat i ve, by t he i mp o r t a n c e of t he support
mustered by t he Spiritual faction a mon g t h e Fr a n c i s c a n s from such
monarchs as J ames I I of Ar agon and F r e d e r i c k I I of Sicily as well
as from such queens as Escl ar monde d e F o i x a n d Sancia, her
daught er, who became t he queen of Ro b e r t t h e Wi s e of Napl es. At a
time when t he Convent ual division of t h e o r d e r h a d most influence
wi t h t he papacy and were embol dened b y t hi s t o per s ecut e and im-
prison many Spirituals, such monar chs g a v e s a n c t u a r y and protec-
t i on to leaders of t he Spiritual gr oup.
DO MI NI UM A ND USTJ S
Some day anthropologists should t ur n t he i r full a t t e n t i on to t he often
splendidly document ed domai n of me di e val r e l i gi ous politics, where
they would be able to follow political pr oce s s e s over t i me in some
detail for centuries. Her e I wish merel y t o p o i n t o u t t h a t St. Francis' s
original band of free compani onsa g r o u p i n wh i c h normat i ve
communi t as had hardl y disengaged i t sel f f r om existential com-
muni t as^-coul d not have persisted if it h a d n o t or gani ze d to endure
in a complex political field. Yet t he me mo r y of t h e ori gi nal comm-
unitas, exemplified by t he life, visions, a n d wo r d s of St . Franci s, was
always kept alive i n t he order, especially b y t h e Spi r i t ual snot abl y
Communitas: Model and Process
such men as J ohn of Parma, Angelo da Clareno, Olivi, and Ubert i no.
But since, by successive papal bulls and by t he writings of St.
Bonaventura, the doctrine of absolute poverty had been juristically
a s
well as theologically defined, the Spirituals were forced into a
"st r uct ur al " at t i t ude toward poverty.
Under the formal definition, the notion of property had been
separated into two aspects: dominium (or proprietas) and usus. Domin-
ium means essentially rights over property, usus the actual mani -
pulation and consumption of property. Now, Pope Gregory I X
declared that the Franciscans should retain usus but renounce
dominium of every kind. At first the Franciscans asked their original
benefactors to keep dominium, but before long they found it more con-
venient to have a comprehensive arrangement , and put the dominium
over all their goods into the hands of the papacy. It was over the
practical consequences of usus that the ideological component of the
split between the Conventuals and the Spirituals first took shape
and became eventually a diacritical symbol of their opposition. For
the Conventuals, more structurally oriented, took full cognizance of
the needs of t he order in a complex political milieu. Thus, to do
their evangelical and charitable work effectively, they felt that they
had to construct solid buildings, both churches and dwellings. To
defend the specific religious position of St. Francis, they had to train
their more intellectual bret hren in philosophy and theology, for
they had to hold their own in the refined arenas of Paris and Florence
against the subtle Dominicans and in face of the growing threat of
the Inquisition. Thus they needed resources, including pecuniary
resources, even coins, to spend on bricks and books.
Among the Conventuals it was more and more left to the discretion
of the local superior to decide how far t he friars might go in the
exercise of usus. According to the Spiritualsand all this came out
during the famous papal investigation of the affairs of the order in
l
39> 83 years after the deat h of the founderConventual " u s e "
became "abus e . " Ubert i no, their spokesman, brought forward much
documentary evidence concerning the practice of cultivation for
profit, the use of granaries and cellars for wine, t he reception of
The Ritual Process
bequests of horses and ar ms . He even a c c u s e d t h e m of exerting
dominium:
Again in just the same way those who can, t ake wi t h t h e m bursarii, who are
their servants, who so spend money at the o r d e r of t he brothers, that in
every respect the brothers appear to have d o mi n i o n not only over the
money, but also over the servants spending it. An d sometimes the brothers
carry a box with the money inside; and on t he occas i ons when this is carried
by the boys, they often know nothing of the c ont e nt s , a n d it is the brothers
who carry the keys. And although the servants ma y sometimes be called
nuntii (a nuntius was an official who was t he a ge n t of t he alms-giver in the
original Papal definition) of those persons wh o ga ve t he money for the
brothers; yet neither the servants nor those wh o de pos i t with them, know
that the money is in the dominion of anyone o t h e r t h a n the brothers . . .
(Quoted by Lambert, 1 9 61 , p. 1 9 0).
But the Spirituals' at t i t ude t owar d usus w a s b e s t expressed in the
doct ri ne of usus pauper, whi ch hel d i n effect t h a t t he friars' use of
goods should be in effect rest ri ct ed t o t h e b a r e mi n i mu m sufficient
to sustain life; i ndeed, some Spi r i t ual s p e r i s h e d from their own
austerities. In this they cl ai med t hey we r e a d h e r i n g to the spirit of
t hei r great founder' s view of pover t y. O n e f e a t u r e of this apparently
admi r abl e at t i t ude made it ul t i mat el y i n t o l e r a b l e t o the structured
chur ch. Thi s was t he Spi ri t ual s' e mph a s i s o n t h e conscience of the
individual as t he supreme ar bi t er of w h a t cons t i t ut ed poverty,
al t hough this conscience oper at ed wi t h r e f e r e n c e to the rigorous
st andards of usus pauper. Some Spi r i t ual s w e n t a s f ar as to imply that
any relaxation of this rigor r an c oun t e r t o t h e i r profession of the
vow of poverty and was t hus a mor t a l s i n. I f t h i s posi t i on were valid,
many of the Convent ual s could be r e g a r d e d a s b e i n g in a permanent
state of mort al sin. Such are t he pitfalls o f e xc e s s i ve legalism!
On t he ot her hand, t he doct r i ne of usus pauper clearly impugned
t he church' s view of the l egi t i mat e a u t h o r i t y pos s es s ed by a religious
superior. If t he head of a Fr anci s can h o u s e , o r e v e n province, were
to use his discretion and per mi t for p r a g ma t i c o r st ruct ural reasons
t he use of considerable quant i t i es of g o o d s , t h e Spi ri t ual friars, in
terms of their own doct ri ne of usus pauper, m i g h t we l l feel themselves
Communitas: Model and Process
153
in no way bound to obey him, thus setting the vow of poverty at
odds with the vow of obedience. In fact this tacit challenge to t he
hierarchical structure of the church was one of the major factors in
the ultimate extirpation of the Spirituals from the order by the
severe measures of Pope J ohn XXI I in a series of bulls backed by
the sanctioning power of the Inquisition. Nevertheless, their zeal
was not altogether in vain, because later reforms of the Franciscan
Order were infused by t he spirit of poverty they had defended so
obstinately.
APOCALYPTIC COMMUNITAS
In considering the early history of the Franciscan Order, it becomes
clear that social structure is intimately connected with history, be-
cause it is the way a group maintains its form over time. Structureless
communitas can bind and bond people together only momentarily. In
thehistory of religions, it is interesting to observe how often communi -
tas-type movements develop an apocalyptic mythology, theology, or
ideology. Among the Franciscan Spirituals, for example, even the
rather dry theologian Olivi, appointed lector at Santa Croce in
Florence, was strongly addicted to the millenarianism of the Joach-
imites. Indeed, Olivi equat ed Babylon, t he great whore, with the
papacy, which was to be destroyed in the sixth age of the world,
while the Spiritual Franciscans, in their absolute poverty, constituted
the true church founded by St. Francis and his twelve companions.
If one is looking for structure in the communitas of crisis or catas-
trophe, one must find it not at the level of social interaction but in a
Levi-Straussian way, underlying the lurid and colorful imagery of
the apocalyptic myths generated in the milieu of existential com-
munitas. One finds, too, a characteristic polarization in movements
of this type between, on the one hand, a rigorous simplicity and
poverty of elected behavi or"naked unaccommodat ed ma n "
and, on the other, an almost febrile, visionary, and prophetic
poetry as their mai n genre of cultural utterance. Ti me and history,
154
The Ritual Process
however, bri ng s t r uct ur e i nt o t h e i r soci al life and legalism into their
cultural out put . Oft en, wh a t wa s o n c e seen as a literal and universal
i mmi nence of cat as t r ophe b e c o me s i nt erpret ed allegorically or
mystically as t he d r a ma of t h e i n d i v i d u a l soul or as t he spiritual fate
of the t rue chur ch on e ar t h or a s pos t pon e d to t he remotest future.
Communi t as not i ons ar e b y n o me ans always associated with
visions or theories of wor l d c a t a s t r o p h e . I n t ri bal initiations, for
example, we find i mpl i ci t l y a t l e a s t t he not i on of absolute poverty
as a mar k of l i mi nal be ha vi or ; b u t w e do not find t he eschatological
ideas of mi l l enari an move me n t s . Never t hel es s , we very often do find
t hat the concept of t hr e at or d a n g e r to t he gr oupand, indeed,
t here is usually real da nge r i n t h e f or m of a circumciser' s or cica-
trizer' s knife, many or deal s, a n d s e ve r e disciplineis importantly
present. And this dange r is o n e o f t h e chief i ngredi ent s in the pro-
duction of existential c o mmu n i t a s , l i ke t he possibility of a " bad t r i p"
for t he narcot i c communi t as of c e r t a i n i nhabi t ant s of a modern city
t hat bears St. Franci s' s n a me . I n t r i b a l i ni t i at i ons, too, we find myths
and their ri t ual enact ment s i n l i mi n a l i t y t hat rel at e to divine catas-
trophes and crises, such as t h e s l a y i n g or self-immolation of import-
ant deities for t he good of t h e h u ma n communi t y, which locate
crisis in t he living past if n ot i n t h e i mmi nent future. But, when
crisis tends to get pl aced be f or e r a t h e r t han after or within con-
t emporary social exper i ence, w e h a v e al r eady begun to move into
t he order of st r uct ur e a n d t o r e g a r d communi t as as a moment of
transition r at her t han a n e s t a bl i s h e d mo d e of bei ng or an ideal soon
to be per manent l y at t ai ne d.
TH E SAH AJ I YA MO VEMENT OF BENGAL
Not all communi t as, howe ve r , i s t h e communi t as of crisis. There
is also t he communi t as of wi t h d r a wa l a n d r et r eat . Sometimes these
genres converge a n d over l ap, b u t us ual l y t hey evince distinct styles.
Th e communi t as of wi t h d r a wa l is n o t so closely linked with belief
in an i mmi nent end of t he wo r l d ; r a t h e r , it involves a total or partial
Communitas: Model and Process
155
withdrawal from participation in the structural relations of the
world, which is, in any case, conceived of as a sort of permanent
"disaster area. " This kind of communitas tends to be more exclusive
in membership, disciplined in its habits, and secretive about its
practices t han the apocalyptic genre just discussed. Although ex-
amples of it may be found in the Christian religion and in the
secular Utopian movements that in so many ways derive from the
Judaeo-Christian cultural tradition, perhaps it is within Hi ndui sm
that some of the clearest examples of the communitas of withdrawal
may be found. I shall confine myself once more to the discussion of
a single movement, that of the Vaisnavas of Bengal, described by
Edward C. Dimock, J r . (1966a, 1966b). Dimock is a highly com-
petent and perceptive Bengali scholar who has published elegant
translations of Bengali tales "from court and village," and his dat a
and conclusions must be regarded with respect.
POETS OF RELIGION: CAITANYA AND FRANCIS
Dimock's work is concerned with a movement that was in some ways
complementary to and in others divergent from the great bhakli (devo-
tional) religious movement t hat "swept across Nort hern Indi a in the
Fourteenth to Seventeenth Centuries and the older bhakti movements
of the s out h" (1966b, p. 41 ) . Since we have already considered one
Christian communitas-type movement in terms of an outstanding
founder, it would be worth repeating this approach with the
Vaisnavas of Bengal and begin our case history with Caitanya (1 486-
1 5 3 3 ) , " t he most significant figure of the Bengal movement . " Just
as in the previous instance we compared Franciscan doctrine with its
practice, let us first consider Caitanya' s teachings and then the
history of the movement he inspired. Dimock tells us t hat Caitanya
was the "revivalist, " not the originator, of Krishna-bhakti (intense
devotion) in eastern Indi a. Vaisnavite movements had been known
in Bengal since the eleventh or twelfth century A.D. i.e., for at least
three centuries before Caitanya' s time. Like St. Francis, Caitanya
156
The Ritual Process
was not himself a theologian. He left a t ot al of ei ght v e r s e s i n his life-
t i me of a devotional, not theological, nat ur e. He r e a g a i n t h e parallel
wi t h Francis' s canticle to "Br ot her S u n " is s t r i ki ng. Cai t anya' s
devotion, too, like Francis' s, was fostered by i mages a n d identifi-
cations, in his case wi t h t he principal actors in t he g r e a t Vai snava
sacred texts, especially t he Bhagavata. Th e pr i nci pal t h e m e of these
texts is t he childhood, boyhood, and yout h of Kr i s h n a , who was
t hought to be an avatdra (incarnation) of t he god Vi s h n u . I n his turn,
Cai t anya was t hought by many to be an avatara of Kr i s h n a or , rather,
a j oi nt i ncarnat i on of Kri shna and his especially b e l o v e d mi l kmai d
Radha, human completeness being represent ed i n b i s e x u a l form
t ranscendi ng all cul t ural and social sex di st i nct i ons.
Th e central episode of Kri shna' s early career wa s h i s l ove for a
group of gopis, t he cowherdesses of Vr ndavana. He h i ms e l f h a d been
reared as a cowherd in this sacred place, and after p l a y i n g all kinds
of t ender and erotic tricks on t he gopis, when he c a me t o manhood
he charmed t hem beyond caring by t he sound of h i s flute in the
forest, so t hat they left their homes, husbands, a n d f a mi l i e s and fled
to hi m in the night. I n one celebrated i nci dent , Kr i s h n a d a n c e s with
all t he gopis in such a way t hat each regards h i m as h e r part i cul ar
lover. Thi s is sometimes represented in I ndi an ar t as a r i n g of girls
bet ween each of whom is t he blue and beautiful for m o f t h e i r divine
lover. I n later Bengali elaboration Ra dh a becomes t h e part i cul ar
object of Kri shna' s love, and in a sense she e pi t omi ze s al l t he rest.
Gai t anya was ent ranced by Kri shna' s dance a n d s u b s e q u e n t love-
maki ng with t he gopis, and by his pr eachi ng he i ns pi r e d s o powerful
a revival of devotional religion t hat " dur i n g his l i f et i me a n d shortly
after his deat h it encompassed t he great er par t of E a s t e r n I ndi a"
(Dimock, 1966b, p. 43 ). One of t he mai n ent husi ast i c pr a c t i c e s that
he stressed was an ar dent medi t at i on in whi ch t he wo r s h i p e r identi-
fied himself successively wi t h t he various relatives, f r i e n ds , a n d lovers
of Kr i shna: for example, his foster par ent s, wh o h a v e for him
parent al affection; his brot her, who considers hi m w i t h b o t h frater-
nal love and a comrade' s loyalty; and, most i mp o r t a n t l y , t he gopis,
for whom Kri shna was lover and beloved. He r e s oci al r el at i onshi ps
were regarded as nat ur al points of depar t ur e for a d e v o t i o n t hat was
Communitas: Model and Process
157
regarded as supernatural in character. The highly erotic tenor of
the texts and devotions apparent l y presented similar problems to
later Vaisnavite theologians as have confronted Jewish and Christian
exegetes of the Song of Songs of Solomon. But the ritual solution of
the Sahajiyas, as Caitanya' s movement was called, was rat her diffe-
rent from that of such Christian mystics as St. John of the Cross and
St. Teresa of Avila, who regarded the erotic language of Solomon' s
Canticles as purely metaphorical. The central ritual of the Sahajiyas
was an elaborate and protracted series of liturgical actions, inter-
spersed with the repetitive recitation of mant ras which culminated
in the act of sexual intercourse between fully initiated devotees of
the cult, a man and a woman, who simulated in their behavior the
love-making of Kri shna and Radha. Thi s was no mere act of sensual
indulgence, for it had to be preceded by all kinds of ascetical prac-
tices, meditations, and teachings by accredited gurus. It was essenti-
ally religious in nat ure, treating the act of sex as a kind of sacrament,
"an outward and visible sign of an inward and spiritual grace. "
What is sociologically interesting about this ritual is that, just
like the gopis, the female part ners of the male Sahajiya initiates had
to be married to other men (see also De, 1 961 , pp. 204-205). Thi s
was not regarded as adul t ery but , as Dimock points out, more after
the fashion of the Courts of Love in medieval Europe, in which true
love is regarded as "l ove in separation [of which] the logical exten-
sion is love apart from marri age [for] in marri age there is always a
touch of the carnal. The son of the t roubadour, says De Rougemont,
'quickens with noble emotions love outside marri age; for marriage
implies no more t han physical union, but " Amor " t he supreme
Erosis the transport of the soul upwards to ultimate union with
light' " (1966a, p. 8). St. Francis sang of his Lady Poverty in much
the same way, incidentally, as a t roubadour of his far-off lady,
wedded to anot her earthly mat e.
My own view is t hat what is now being considered, in both
sixteenth-century Bengal and twelfth-century Europe, as a love that
is both divine and faintly illicitas contrasted with licit, marital
loveis a symbol of communitas. Communi t as is the link between
the gopis, t he blue god between each milkmaid. Communi t as is also
i
5
8
The Ritual Process
the friar's relationship to My Lady Poverty. In terms of t he symbol i c
opposition between romant i c love and marri age, marri age is homo-
logous with property, just as love in separation is homologous wi t h
poverty. Marri age, therefore, represents structure i n this t heol ogi -
cal-erotic language. The notion of personal possession or owner shi p
is also antithetical to the kind of communi t as-l ove epi t omi zed by
the relationship between Kri shna and the gopis. Di mock, for ex-
ampl e, cites a later Bengali text that "embellishes a story from t he
Bhdgavata." It seems t hat the gopis told Kri shna they were filled
with love for him, and then began the dance. "But dur i ng t he dance
Kri shna disappeared from t hem, for into t he mi nds of all t he Gopi s
had come the t hought ' he is mi ne, ' and in t he t hought ' he is mi ne , '
paraki ya [i.e., true love in separation] cannot remai n. . . . But wh e n
longing again rose in the Gopis' minds, Kri shna again appear ed to
t h e m" (1966a, p. 1 2) .
Sahajiya doctrine differed fromVaisnava orthodoxy in t hat t he l at t er
prescribed sacramental union between spouses, whereas t he followers
of Cai t anya, as we have seen, prescribed ritual intercourse bet ween
a devotee and the wife of anot her. Cai t anya himself had such a r i t ual
mat e, " t h e daught er of Sathi, whose mi nd and body were devot ed
to Cai t anya. " It is interesting to note t hat t he ritual part ners of t he
Gosvamins, the original compani ons of Cai t anya and t he expositors
of Sahajiya theology, were " wome n of. . . casteless groups, washer -
women or women of other low cast es" (ig66a, p. 1 27) . I ndeed, t he
gopis themselves were cowherdesses, and hence not of t he hi ghest
caste. Thi s communi t as quality of failing to recognize hi erarchi cal
structural distinctions is in fact qui t e typical of Sahajiya, and of
Vaisnavism as a whole.
TH E CLEAVAGE BETWEEN
DEVOTIONALISTS AND CONSERVATIVES
Cai t anya, then, like St. Francis, was a poet of devotional rel i gi on,
humbl e and simple, living his faith rat her t han t hi nki ng about it.
But his six Gosvamins were theologians and philosophers, who set
Communitas: Model and Process
159
up an as'rama (a school of religious instruction) for Vaisnavas where
the formal doctrine of their sect could be elegantly forged. Three of
these Gosvamins were members of a single family. This family,
though reputedly of Brahman origin, had lost caste by virtue of their
high positions at the court of the Muslim ruler of Bengal at the time.
They continued, in fact, to hold dialogue with certain Sufis, a group
of Muslim mystics and poets who had strong affinities with t he
Sahajiyas themselves. These six scholars wrote in Sanskrit and
"pl ayed the major role in t he codification of the doctrine and ritual
of the sect " (1966b, p. 45 ). But once more a devotional movement
was doomed to founder on the rock of doctrinal formulation. After
Caitanya' s death, his followers in Bengal split into two branches. One
branch followed the lead of Caitanya' s friend and intimate com-
panion, Ni t yananda, known as the "casteless Avadhut a" (the
Avadhutas were ascetics); the other followed Advaita-acarya, an
early and leading devotee of Caitanya, a Brahman of Sant apur.
There are certain affinities between Ni t yananda and the Spiritual
Franciscans. He was not only casteless himself; while he "st ayed
with Si i dras" (1966b, p. 5 3 ), and was "apost l e to the Banyas"
(both Sudras and Banyas were low-caste Hi ndus), he also allowed
thousands of Buddhist monks and nuns into the Vaisnava fold. One
of Caitanya' s biographers has him say to Ni t yananda: " I t is my
promise, made with my own mout h, t hat ignorant and low-caste
and humbl e people will float upon the sea ofprema [love] . . . you can
set t hem free by bhakti" (1966b, p. 54). Bhakti, or salvation t hrough
personal devotion to a deity, did not recommend itself to Advaita-
acarya, who reverted to the " pa t h of knowl edge" of orthodox
monists, who in Indi a had always taken mukti, release from the
cycle of rebirths, as their pri mary concern. Advaita, being a Brah-
man, did not make light of this fact. It was consistent with this caste
affiliation t hat he should revert to the doctrine of mukti, for release
from rebirth, in orthodox Hinduism, is very much dependent upon
a person' s regular fulfillment of his caste duties. If he fulfills these
duties, he may expect to be reborn in a higher caste; if, in addition,
he lives a holy and self-sacrificial life, he may ultimately escape from
j 5
0
The Ritual Process
H OMOLOGIES BETWEEN
SAH AJIYA AND FRANCI SC AN I SM
Ni t yananda and his rival Advai t a r e p r e s e n t e d r e s pe c t i ve l y the
principles of normat i ve communi t as a n d s t r u c t u r e a t t h e level of
group organi zat i on; their branches we r e h o mo l o g o u s respect i vel y
suffering and t he power of maya, or t h e i l l u s o r y phenomenal world.
Monists, like Advaita, believed t hat final r e l e a s e could best be
secured by dispelling illusion t hr ough k n o w l e d g e of the single
reality known as dtman-brahman. I n ot he r w o r d s , for t hem salvation
was t hrough gnosis, not devotion, a n d it i n v o l v e d accept ance of the
social structure in its present formfor al l e x t e r n a l forms were
equally illusory and void of ul t i mat e r eal i t y. B u t Ni t y a n a n da did not
share this passive social conservatism. Be l i e v i n g a s he did t hat every
man, regardless of caste or creed, coul d o b t a i n s a l v a t i o n by personal
devotion to Kri shna and Radha, he e mp h a s i z e d t he missionary
aspect of Vaisnavism.
Cai t anya and Ni t yananda convert ed m a n y Mus l i ms a n d thus
ant agoni zed t he Musl i m reigning p o we r a n d de l i be r a t e l y broke a
a number of ort hodox Hi ndu religious l aws . F o r e x a mp l e , " Caitanya
rejoiced when he had persuaded Vas ude va t o t a k e prasadaremnants
of food offerings t o t he dei t ywi t hout h a v i n g f i r s t wa s h e d his hands.
' Now, ' said Cai t anya, ' you have t rul y b r o k e n t h e ties wi t h your
b o d y ' " (1966b, p. 5 5 ) . Thi s saying r e mi n ds u s o f ma n y of those of
Jesuse. g. , t hat t he Sabbat h was ma d e for m a n a n d not man for
t he Sabbat h, and t hat t he t r ut h shall ma k e y o u f r e e . For Caitanya
and t he Ni t yananda br anch of his fol l ower s, bhakti emanci pat ed
t hem from laws and convent i ons: " t h e y d a n c e d ecst at i cal l y and
sang; they were as if ma d " (1966b, p. 65 ) . I t i s h a r d t o t hi nk that
t here is not hi ng in common bet ween t h e e c s t a t i c c ommun i t a s of
Dionysos and t hat of Kri shna. I ndeed, Ov i d ' s puer aeternus came
from "adusque decolor extremo qua cingitur India Gauge" ( " Da r k Indi a
girdled by t he farthest Ganges, " Metamorphoses, I V , l i n e 2 1 ) .
Communitas: Model and Process
161
with the Spiritual and Conventual Franciscans. In bot h the Euro-
pean and Indi an cases the successors of the founder had had to cope
with the problems of group continuity and of theological definition.
The founders, Francis and Caitanya, were poets of religion; they
lived out the colorful religious imagery t hat filled their meditations.
In the case of the Vaisnava-Sahajiyas, it was the group of Gosvamins
who assumed the task of defining the central concepts of the sect.
Whereas the Franciscans had located their Archi medean point in
the notion of poverty, and then gone on to discriminate between
dominium and usus wi t h regard to property, and had finally been led
into factionalism around the doctrine of usus pauper, t he Sahajiyas
had centered their controversies on anot her aspect of possession, in
this case sexual possessionthough, as we have seen, for t hem sexual
union took on a sacramental character.
The Vaisnavas' sacred books, the Bkagavata and t he Gila Govinda,
are full of the imagery of passion; they tell of the love of the gopis for
Krishna. But, as Dimock points out, " t h e idea of trysting wi t h t he
wives of other men is not one acceptable to most of Indi an society"
(1966b, p. 5 5 ), despite, one mi ght add, its traditional religious toler-
ance, even t hough this tolerance depends on no Second Amend-
ment! Thus t he Vaisnava exegetes, and especially the Sahajiyas,
had many problems. Vaisnava doctrine had always borrowed freely
from Sanskrit poetic theory, and one of the salient distinctions i n
this theory was to divide women into two classes: svakiya or sviya,
she who is one' s own, and parakiya, she who is anot her' s. Parakiya
women can be those who are unmarri ed and those who are another' s
by marriage. I n t he Bhagavata text the cowherdesses were clearly
of the latter kind. The first exegetical at t empt by a Gosvamin, Ji va
by name, was to deny t hat this could be meant literally. For one
thing, st andard poetic theory did not recognize parakiya women as
acceptable for pri mary roles in dr ama; therefore the gopis, who were
heroines, could not really be parakiya. Again, the gopis had never
really consummated their marriages. "By the mdya-power of
Krishna [his power to fabricate illusions], shapes like the gopis, not
the gopis themselves, had slept with their husbands. Furt hermore t he
The Ritual Process
gopis are really s'aktis [i.e., powers emanat i ng from a dei t y c o n c e i v e d
of as goddessesthus t he sakti of t he god Shiva is t he godde s s Ka l i
or Dur ga] of Krishna, essential to and in some way i de nt i cal wi t h,
h i m" ( 1 9 66b, p. 5 6) . Therefore, they fall into t he class of svakiya
really his own; they are only seemingly parakiya, t he w o m e n
Q
f
others.
Gosvamin Ji va' s kinsman, Rupa, accepted t he parakiya i n t e r p r e t a -
tion, which strains the sense of the original texts less, but a r g u e d t h a t
ordi nary human ethical yardsticks could hardl y be appl i ed t o ' ' t h e
ruler of all t hat is to be rul ed. " Thi s ar gument has been r e s o r t e d t o
in Judaeo-Chri st i an exegesis to explain some of t he s t r a n g e r a c t s
and commands of Jehovah, such as the command to A b r a h a m t o
sacrifice Isaac. In t he Bhdgavata itself, someone asks how K r i s h n a ,
described as the "uphol der of pi et y, " could have i ndul ged i n l o v e
play with the wives of others, and receives t he r epl y: " F o r t h o s e
who are free of egoism there is no personal advant age her e b y m e a n s
of proper behavior, nor any disadvantage by means of t he o p p o s i t e . "
Thi s view accords well with the attitudes of a sect who felt t h e y w e r e
beyond t he bounds and st andards of ordi nary, s t r uct ur ed s o c i e t y .
A similar freedom pervades the beliefs of many ot her m o v e m e n t s
and sects that stress enthusiastic or devotional communi t as a s t h e i r
basic principle; one mi ght mention the Hussites of P r a gue o r t h e
Onei da Communi t y of New York State.
RADH A, MY LADY POVERTY, AND COMMUNITAS
But t he later exegetes came to accept as ort hodox t he l i t e r a l v i e w
t hat t he love of t he gopis for Kri shna was consistent wi t h t h e i r
parakiya conditionand t hat this condition made it s o me h o w m o r e
pur e and real. For, as Dimock notes, "svakiya leads to kdma, t o d e s i r e
for t he satisfaction of t he self; only parakiya results in t he prema, t h e
intense desire for the satisfaction of the beloved, wh i c h i s t h e
characteristic, to be emul at ed by the bhakta [the de vot e e ] , o f t h e
love of the gopis. It is because t he love of the gopis is a parakiya. l o v e
Communitas: Model and Process
that it is so intense. The pain of separation, possible only in parakiya,
a n
d the resultant constant dwelling of the minds of the gopis on
Krishna, is their sal vat i on" (1966b, pp. 5 6-5 7). One is reminded
again of certain passages in the Song of Songs and of the verses of
St. John of the Cross, in which the soul pines for the absent lover,
who is God. However, in the Sahajiya sect, such pining is not for-
ever; after the "discipline of the sixty-four devotional acts, " which
involves "act i vi t y, repetition of mant ras, physical discipline, in-
tellectual knowledge, asceticism, medi t at i on" (ig66a, p. 1 95 ), t he
Sahajiyas depart from Vaisnava orthodoxy by entering on the stage
of the sexual ritual of vidhi-bhakti. In this the partners are bot h
initiates and are regarded as gurus, teachers or spiritual guides of one
another, and here as sacramental expressions of Kri shna and Radha
themselves. The partners are considered to be "of one t ype " (1966a,
p. 220), in which case "t her e can be uni on" (p. 21 9) , and this type
is the highest of their respective sexes. Clearly, the motives for this
act are not domi nant l y sensual, for a rich erotic literature attests to
the abundance of secular practices available to the Indi an volup-
tuary of the period wi t hout any need at all for a long preliminary
training by ascesis.
In the era of dept h psychology, we must of course be attentive for
signs of the Oedi pus complex in a love t hat is powerfully idealized
and at its noblest at a distance. Again, the Jungi ans would have
much to say about a union with a Great Mot her archetype as a
symbol of the union between conscious and unconscious components
of the mi nd t hat precedes the wholeness of "i ndi vi duat i on. " But
these " de pt h s " may be socially and culturally "superficial" if our
focus of attention is upon modalities of social relationships. The
Sahajiyas seem to be intent upon utilizing various cultural and
biological means to attain a structureless state of pur e social com-
munitas. Even in the sexual ritual t he aim is to uni t e not merely a
male with a female but the male and female within each individual;
thus, as Cai t anya himself was said to be, each devotee would be an
incarnation simultaneously of Kri shna and Radha, a complete
human being. Symbolically, however, the tie of marri ageand with
164
The Ritual Process
it the family, t hat basic cell of social st ruct urewas dissolved by
parakiyd love. At its very source, therefore, in a society largely
structured by kinship and caste, structure was rendered nugat or y,
for t he lovers broke all caste rules, too. Th e Franciscans deni ed
themselves property, one pillar of social structure, t he Sahajiyas
marri age and t he family, anot her major pillar. It is significant t hat
the anthropologist Edmund Leach, who gave t he influential Re i t h
Lectures on the BBC' s Thi r d Pr ogr amme in 1967, should also have
ret urned to the attack on the familyregarding it as t he source of
all neuroses and ment al cri ppl ement onl y in order to prai se
collectives and communities, like t he Israeli ki bbut zi m, wi t h t hei r
creches. Dr. Leach is very familiar wi t h Sout h I ndi an as well as
Singhalese literature. Perhaps there is a Tant r i c echo in his at t ack.
At any rate, he seems to be striking a blow for communi t as!
BOB DYLAN AND TH E BAULS
Th e successors of Cai t anya failed because Advai t a' s gr oup be came
absorbed in the caste system, and Ni t yananda' s gr oup, exclusivist
and full of missionary fervor, encountered persecution and gradual l y
lost heart . Historically, t he tide of Sahajiyaism seems slowly t o have
ebbed in the seventeenth and eighteenth centuries, t hough Vai snav-
ism is still an active force, Dimock notes, in Bengal. For exampl e,
t he sect of musicians known as Bauls, who play a "pr i mi t i ve but
haunt i ng one-stringed i nst rument , called ek-t ara, " and sing "songs
as gentle and as stirring as t he wind whi ch is their home , " cl ai m to
be "madde ne d by the sound of Kri shna' s flute, and, like a Gopi ,
caring not hi ng for home or for t he respect of t he world, they follow
i t " (1966a, p. 25 2). A fascinating exampl e of t he convergence, under
modern conditions of transportation and communi cat i on, of West ern
and Eastern liminars and communi t as-bearers may be found i n
many music shops today. Th e cover of a recent recordi ng of Bob
Dyl an' s songs shows t he Ameri can folk singer and spokesman of t he
structurally inferior flanked by Bauls, these musical vagabonds of
Communitas: Model and Process
165
Bengal: guitar and ek-tara have come together. It is even more
fascinating to consider how often expressions of communitas are
culturally linked with simple wind instruments (flutes and har-
monicas) and stringed instruments. Perhaps, in addition to their
ready portability, it is their capacity to convey in music the quality
of spontaneous human communi t as that is responsible for this. The
Bauls, like St. Francis, were "t roubadours of God, " and it mi ght
be fitting to close this chapt er with one of their songs, which clearly
indicates how the spirit of Vaisnava communitas has persisted into
the world t oday:
Hi ndu, Muslimthere is no difference,
Nor are there differences in caste.
Kabi r the bhakta [devotee] was by caste a Jol a,
but drunk with prema-bhakti [true love, best expressed,
as we have seen, by extramarital love]
he seized the Black Jewel' s feet [i.e., Krishna' s feet].
One moon is lantern to this world,
and from one seed is the whole creation sprung
(1966a, p. 264).
This is t he aut hent i c voice of spontaneous communitas.
5
Humility and Hierarchy:
The Liminality
of Status Elevation
and Reversal
RITUALS OF STATUS ELEVATI ON
AND STATUS REVERSAL
Van Gennep, t he father of formal processual anlysis, u s e d t w o s et s of
t erms to describe t he three phases of passage from o n e c u l t u r a l l y
defined state or status to anot her. Not only did he use, w i t h p r i ma r y
reference to ri t ual , t he serial terms separation, margin, a n d reaggrega-
tion; he also, wi t h pr i mar y reference to spatial t r ans i t i ons , e mp l o y e d
t he terms preliminal, liminal, and postliminal. Wh e n h e di s c us s e s hi s
first set of terms and applies t hem to dat a, van Ge nne p l a y s e mp h a s i s
on what I would call t he " s t r uc t ur a l " aspects of pa s s a ge . Wh e r e a s
his use of the second set indicates his basic concer n w i t h u n i t s of
space and time in which behavior and symbolism ar e mo me n t a r i l y
enfranchised from t he norms and values t hat govern t h e p u b l i c lives
of i ncumbent s of st ruct ural positions. Her e l i mi nal i t y b e c o me s
central and he employs prefixes at t ached to t he adj ect i ve " l i mi n a l "
to i ndi cat e t he peri pheral position of st r uct ur e. By " s t r u c t u r e " I
mean, as before, "soci al st r uct ur e, " as used by t he ma j o r i t y of
British social anthropologists, t hat is, as a mor e or less di s t i n c t i ve
ar r angement of specialized mut ual l y dependent i ns t i t ut i ons a n d t h e
1 6 6
Humility and Hierarchy
167
institutional organization of positions and/ or of actors which they
imply- I am not referring to " s t r uct ur e " in the sense currently
made popular by Levi-Strauss, i.e., as concerned with logical cate-
gories and the form of the relations between t hem. As a mat t er of
fact, in the liminal phases of ritual, one often finds a simplification,
even elimination, of social structure in the British sense and an
amplification of structure in Levi-Strauss' s sense. We find social
relationships simplified, while myt h and ritual are elaborated. That
this is so is really quite simple to underst and: if liminality is regarded
as a time and place of wi t hdrawal from normal modes of social
action, it can be seen as potentially a period of scrutinization of the
central values and axioms of the culture in which it occurs.
In this chapt er t he pri mary focus w'
1 1
hp on liminality,_.as bot h
phase and state. I n complex large-scale societies, liminality itself, as
a result of t he advanci ng division of labor, has often become a
religious or quasi-reTigious state, and, by virtue of this crystalliza-
tion, has tended to reenter structure and acquire a full complement
of structural roles and positions. Instead of the seclusion lodge, we
have the church. More t han this, I wish to distinguish two mai n
types of liminalitythough many others will undoubt edl y be dis-
mveredC-first, ^the liminality t hat characterizes rituals of status
elevation, in which t he ritual subject or novice is being conveyed
irreversibly from a lower^to a higher position in an institutionalized
system of such positions^Secondiy) t he liminality frequently found
in cyclical and calendrical ritual, usually of a collective kind, in
which, at certain culturally defined points in t he seasonal cycle,
groups or categories of persons who Ji abi t ual l y occupy low status
positions in t he social structure are positively enjoined to exercise
ritual authority over their superiors; and they, in their t urn, must
accept with good will their ritual degradat i on. Such rites may be
described as rituals of status reversal. They are often accompani ed by
robust verbal and nonverbal behavior, in which inferiors revile and
even physically mal t reat superiors.
A common vari ant of this type of ritual is when inferiors affect
the rank and style of superiors, sometimes even to the extent of
The Ritual Process
arrayi ng themselves i n a hi er ar chy mi mi c k i n g t h e secul ar hierarchy
of their so-called bet t ers. Briefly put , on e mi g h t cont r as t the limin-
ality of t he strong (and get t i ng s t r onger ) w i t h t h a t of t he perman-
ently weak. Th e l i mi nal i t y of t hose goi n g u p usual l y involves a
put t i ng down or humbl i ng of t he novi ce a s i t s pr i nci pal cultural
const i t uent ; at t he same t i me, t he l i mi n a l i t y of t he permanently
st ruct ural inferior contains as its key s oci al e l e me n t a symbolic or
make-believe elevation of t he r i t ual s ubj ect s t o pos i t i ons of eminent
aut hori t y. Th e st ronger ar e ma d e we a ke r ; t h e we a k act as though
they were strong. Th e l i mi nal i t y of t he s t r o n g is socially unstruc-
t ured or simply s t r uct ur ed; t hat of t he we a k r e pr e s e n t s a fantasy of
st ruct ural superiority.
LIFE CRISIS RI TES
AND CALENDRI CAL RI TES
Now t hat I have put my cards on t he t abl e , s o t o s peak, I will supply
some facts to support these assertions, b e g i n n i n g wi t h t he traditional
ant hropol ogi cal distinction bet ween life-crisis r i t e s and seasonal or
cal endri cal rites. Life-crisis rites ar e t hose i n wh i c h t he ri t ual subject
0subj ect s move, as Ll oyd Warner_lxQ' SQ) h a s p u t it, from " a fixed
pl acent al pl acement wi t hi n his mot he r ' s w o mb , t o his deat h and
ul t i mat e fixed poi nt of his t ombs t one a n d final cont ai nme nt in his
grave as a dead or gani s mpunct uat e d b y a n u mb e r of critical
moment s of t ransi t i on whi ch all soci et i es r i t ua l i z e and publicly
mar k wi t h sui t abl e observances to i mpr e s s t h e significance of tEe""
i ndi vi dual and t he gr oup on living me mb e r s of t he communi t y.
These ar e t he i mpor t ant t i mes of bi r t h, p u b e r t y , mar r i age and
d e a t h " (p. 3 0 3 ) . I woul d add t o t hese t h e r i t e s t h a t concern entry
into a hi gher achi eved st at us, whe t he r t hi s b e a pol i t i cal office or
member shi p of an exclusive cl ub or secr et s oci et y. Thes e rites may
be ei t her i ndi vi dual or collective, but t h e r e is a t e n de n c y for t hem to
be performed mor e frequently for i ndi vi dual s . Cal e ndr i cal rites, on
t he ot her hand, al most al ways refer t o l a r ge g r o u p s a n d qui t e often ^
Humility and Hierarchy
embrace whotesori eri js^Oft en, too, they are performed at well-
delineated points in the axmual productive cycle., and attest to the
passage from scarcity to plenty (as at first fruits or harvest festivals)
or from plenty to scarcity (as when the hardships of winter are antici-
pated and magically warded against). To these also one should add
all rites de passage, which accompany any change of a collective sort
from one state to another, as when a whole tribe goes to war, or
a large local community performs ritual to reverse the effects of
famine, drought, or plague. Life-crisis rites and rituals of induction^
into office are almost always rites of status eTevatioriTcalendrical
rites and r i t e s e DFj j r our ^
reversal.
^fhave written elsewhere (1967, pp. 9 3 - 1 1 1 ) about those symbols
of liminality that indicate the structural invisibility of novices
undergoing life-crisis ritualshow, for example, they are secluded
from the spheres of everyday life, how they may be disguised in pig-
ments or masks, or rendered inaudible by rules of silence. And I
have shown above (p. 108) how, to use Goffman's terms (1962, p. 1 4) ,
they are "leveled" and "stripped" of all secular distinctions of
status and rights over property. Furt hermore, they are subjected to
trials and ordeals to teach them humility. One example of such
treatment should be sufficient. In the Tsonga boys' circumcision
rites, described by Henr i Junod (1962, Vol I, pp. 82-85 ), .
t n e
b
v s
are "severely beaten by the shepherds . . . on the slightest pretext"
(p. 84); subjected to cold, they must sleep naked on their backs all
night during the chilly months of June to August; they are absolutely
forbidden to drink a drop of water during the whole initiation;
they must eat insipid or unsavory food, which "nauseates them at
first" to the point of vomiting; they are severely punished by having
sticks introduced between the separated fingers of both hands while
a strong man, taking both ends of the sticks in his hands, presses
them together and lifts the poor boys, squeezing and half crushing
their fingers; and, finally, the circumcised must also be prepared to
die if their wound does not heal properly. These trials are not only,
as Junod supposed, to teach the boys endurance, obedience, and
170
The Ritual Process
manliness. Manifold evidence from other societies suggests t h a t t h e y
have the social significance of renderi ng t hem down i nt o s o me k i n d
of human prima materia, divested of specific form and r e d u c e d t o a
condition t hat , al t hough it is still social, is wi t hout or b e n e a t h a l l
accepted forms of status. The implication is t hat for a n i n d i v i d u a l
to go higher on t he status ladder, he must go lower t han t h e s t a t u s
l adder.
STATUS ELEVATION
Th e liminality of life crisis, therefore, humbl es and ge ne r a l i z e s _th<^
aspi rant to higher-structural status. Th e same processes ar e p a r t i c u -
1arly~vividly exemplified in many African installation r i t ua l s . T h e
future i ncumbent of t he chieftainship or he admans hi p i s f i r s t
separated from t he commonal t y and t hen must unde r go l i mi n a l
rites t hat rudel y abase hi m before, in t he reaggregat i on c e r e mo n i e s ,
he is installed on his stool in final glory. I have al r eady d i s c u s s e d
t he Ndembu installation rites (Chapt er 3) where t he c h i e f - t o - b e
and his ri t ual wife are abased and r epr i manded dur i ng a n i g h t ' s
seclusion in a small hut by many of t hei r future subjects. A n o t h e r
African exampl e of t he same pat t er n is vividly pr es ent ed i n Du
Chaillu' s (1868) account of t he election of " a ki ng i n Ga b o o n . "
After a description of t he funerary rites for t he old ki ng, Du Gh a i l l u
describes how t he elders " of t he vi l l age" secretly choose a n e w k i n g ,
who is himself " ke pt i gnorant of his good fortune t o t he l a s t . "
It happened that Njogoni, a good friend of my own, was elected. T h e
choice fell on him, in part because he came of a good family, but chi efl y
because he was a favourite of the people and could get the most vot es. I d o
not think that Njogoni had the slightest suspicion of his elevation. As h e
was walking on the shore on the morning of the seventh day [after t he d e a t h
of the former king] he was suddenly set upon by the entire popul ace, wh o
proceeded to a ceremony which is preliminary to the crowning [ and mu s t
be considered as liminal in the total funerary installation complex of r i t es ]
and must deter any but the most ambitious man from aspiring to t he c r o wn .
Humility and Hierarchy iji
They surrounded him in a dense crowd, and then began to heap upon him
every manner of abuse that the worst of mobs could imagine. Some spat in
his face; some beat him with their fists; some kicked him; others threw
disgusting objects at him; while those unlucky ones who stood on the out-
side, and could reach the poor fellow only with their voices, assiduously
cursed him, his father, his mother, his sisters and brothers, and all his
ancestors to the remotest generation. A stranger would not have given a cent
for the life of him who was presently to be crowned.
Ami d all the noise and struggle, I caught the words which explained all
this to me; for every few minutes some fellow, administering a specially
severe blow or kick, would shout out, "You are not our king yet; for a little
while we will do what we please with you. By-and-by we shall have to do
your will."
Njogoni bore himself like a man and prospective Ki ng. He kept his
temper, and took all the abuse with a smiling face. When it had lasted about
half an hour they took him to the house of the old king. Here he was seated,
and became again for a little while the victim of his people's curses.
Then all became silent; and the elders of the people rose and said,
solemnly (the people repeating after them),"Now we choose you for our
king; we engage to listen to you and to obey you."
A silence followed, presently the silk hat, which is the emblem of royalty,
was brought in and placed on Njogoni's head. He was then dressed in a red
gown, and received the greatest marks of respect from all who had just now
abused him (pp. 43 - 44)-
This account not only illustrates the humbling of a candi dat e in a
rite of status elevation; it also exemplifies the power of structural
inferiors in_a rite..ostajjisJreveri^in
r
a"cycle of political ri tuals. Tt is
one of those composite rituals t hat contain aspects ofst at us eleva-
tion along with aspects of status reversal. In the first aspect, an
individual's permanent structural elevation is emphasized; in the
second, stress is laid upon the temporary reversal of the statuses of
rulers and ruled. An individual' s status is irreversibly changed, but
the collective status of his subjects remains unchanged. Ordeals in
rituals of status elevation are features of our own society, as the^
"Eazings in fraternity and military-academy initiations attest. _P_ne
modern ri tual pf status reversal at least comes to myr ni nd. In the
172
The Ritual Process
British Army on Christmas Day, pr i vat es a r e w a i t e d on at dinner
by officers and N. C. O. ' s . After this ri t e t h e s t a t u s of t he privates
remains unchanged; indeed, t he s e r ge ant - maj or m a y bawl them out
all t he more stridently for havi ng been m a d e t o r u n about with
turkey at their behest. Th e ri t ual , in fact, has t h e l on g- t e r m effect of
emphasizing all t he more _trenchjmriy ^t he_s oci al definitions~oTTne~
group.
STATUS REVERSAL: TH E MASKI NG FUNCTI O N
I n West ern society, t he traces of rites of a g e - a n d sex-rol e reversal
persist in such customs as Hal l oween, w h e n t h e powers of the
structurally inferior are manifested in t he l i mi n a l d o mi n a n c e of pre-
adolescent children. Th e monst rous masks t h e y of t e n wear in dis-
guise represent mai nl y cht honi c or e a r t h - d e mo n i c powerswi t ches
who blast fertility; corpses or skeletons from u n d e r g r o u n d ; indigen-
ous peoples, such as I ndi ans ; t rogl odyt es, s uch a s d wa r v e s or gnomes;
hoboes or ant i -aut hori t ari an figures, such as p i r a t e s or traditional
West ern gun fighters. These tiny ear t h powe r s , i f n o t propi t i at ed by-
treats or dainties, will work fantastic a n d c a p r i c i o u s tricks on the
aut hori t y-hol di ng generat i on of h ous e h ol de r s t r i c ks si mi l ar to those
once believed to be t he work of ear t h s pi r i t s , s u c h as hobgoblins,
boggart s, elves, fairies, and trolls. I n a s e ns e , t o o , t hese children
medi at e between t he dead and t he l i vi ng; t h e y a r e not long from
t he womb, whi ch is in many cul t ures e q u a t e d w i t h t he t omb, as
bot h are associated wi t h t he ear t h, t he s our ce o f f r ui t s and receiver
of leavings. Th e Hal l oween chi l dren e x e mp l i f y s ever al . liminal
motifs.Lj:heir masks insure t hem, anonymi t y, f or n o o n e knows just
whose part i cul ar children thev ar e. But , as w i t h mos t rituals of
reversal, anonymi t y here i s f or purposes of a g g r e s s i o n ^ n o t . humili-
at i on. Th e child' s mask is like t he h i g h wa y ma n ' s ma s k a n d , indeed,
children at Hal l oween often wear t he masks of b u r g l a r s or execution-
ers. Maski ng endows t hem wi t h t he power s of f e r a l , cr i mi nal auto-
cht honous and supernat ural bei ngs.
Humility and Hierarchy
173
In all these respects t here is something of the character of t heran-
thropic beings in primitive myt h, for example, t he mal e and female
jaguars of t he " f i r e " myths of the Ge-speaking Amazoni an peoples
described by Levi-Strauss in Le Cru el le Cuit ( 1 9 64) . Terence Tur ner ,
of the University of Chicago, has recently reanalyzed t he Ge myths
(in press). Fr om his precise and complex analysis of Kayapo myths of
the origin of domestic fire, he concludes t hat t he j aguar form is a
kind of mask t hat bot h reveals and conceals a process of structural
realignment. Thi s process concerns t he movement of a boy from t he
nuclear family to t he men' s house. The j aguar figures here represent
not merely t he statuses of father and mot her but also changes in t he
boy's relationships to each of these parentschanges, moreover,
that involve the possibility of painful social and psychical conflict.
Thus, the mal e j aguar of t he myt h begins by being genuinely terri-
fying and ends as benevolent, while t he female j aguar , always
ambivalent, ends as malevolent and is slain by t he boy on t he advice
of the male j aguar .
Each j aguar is a multivocal symbol : while t he mal e j aguar repre-
sents both t he pains and t he joys of specific fatherhood, he also
stands for fatherhood in general. Ther e is in fact among the Kayapo
the ritual role of "sur r ogat e father, " who removes t he boy from t he
domestic sphere at about t he age of seven to assimilate hi m into the
wider male moral communi t y. Symbolically, this appears to be
correlated wi t h the " d e a t h " or extirpation of an i mport ant aspect
of the mother-son relationship, which corresponds with the mythical
account of t he slaying of t he female j aguar by t he boywhose will
to kill has been fortified by t he mal e j aguar. Clearly t he mythical
account is not concerned wi t h concrete individuals but wi t h social
personae; yet, so delicately interwoven are structural and historical
considerations t hat t he direct representation, in human form, of
mother and father in myt h and ritual may well be situationally
blocked by t he powerful affects always aroused in crucial social
transitions.
There may well be anot her aspect of the masking function bot h
in American Halloweens and in Kayapo myths and ritualsand
174
The Ritual Process
in many other cultural manifestations as well. An n a F r e u d h a s had
much t hat is illuminating to say about t he frequent pl ay i d e n t i f i
c a
.
tion of children wi t h fierce animals and other t hr e at e ni ng mo n s t r o u s
beings. Miss Freud' s argument whi ch derives its force, a d mi t t e d l y ,
from the theoretical position of her own mi ght y f at he r i s c ompl e x
but coherent. What is being given ani mal guise in chi l d f a n t a s y is
t he aggressive and puni t i ve power of t he par ent s, p a r t i c u l a r l y the
father, and especially wi t h regard to t he wel l -known p a t e r n a l cast r a-
tion t hreat . She points out how small children ar e qui t e i r r a t i on a l l y
terrified of animalsdogs, horses, and pigs, for e x a mp l e n o r ma l
fear, she explains, overdetermined by unconsci ous fear of the
menacing aspect of t he parents. She then goes on t o a r g u e t h a t one
of the most effective defense mechanisms utilized by t h e e g o agai ns t
such unconscious fear is to identify with t he t erri fyi ng o b j e c t . In
this way it is felt to be robbed of its power ; and pe r h a ps p o w e r may-
even be drai ned from it.
For many dept h psychologists, too, i dent i fi cat i on a l s o means
replacement. To draw off power from a st rong be i ng is t o we a ke n
t hat being. So, children often play at being tigers, l i ons , or cougar s ,
or gunmen, Indi ans, or monsters. They are t hus, a c c o r d i n g t o Anna
Freud, unconsciously identifying themselves wi t h t h e v e r y power s
that deeply t hreat en t hem, and, by a species of j uj i t s u, e n h a n c i n g
their own powers by t he very power t hat t hr e at e ns t o enfeebl e
t hem. Ther e is in all this, of course, a traitor-like q u a l i t y u n c o n -
sciously one aims " to kill t he thing one l ove s "and t hi s is pr eci s el y
the quality of behavior t hat generalized par ent s mu s t e x p e c t from
generalized children in the customs of t he Ame r i c a n Ha l l o we e n .
Tricks are played and propert y is damaged or ma d e t o l ook as
though it has been damaged. In the same way, i d e n t i f i c a t i o n wi t h
the j aguar figure in t he myt h may indicate t he pot e n t i a l f a t h e r h o o d
of the initiand and hence his capacity to r epl ace s t r u c t u r a l l y his
own father.
It is interesting t hat this relationship bet ween t h e r a n t h r o p i c
entities and masks and aspects of t he par ent al rol e s h o u l d b e ma d e
both at rituals of status elevation and at cul t ural l y d e f i n e d poi nt s
Humility and Hierarchy
175
of change in t he annual cycle. One mi ght speculate t hat feral repre-
sentation of t he parent s concerns only those aspects of t he total
parent-child relationship, in its full longitudinal spread, t hat pro-
voke strong affects and volitions of an illicit libidinal, and particu-
larly aggressive, character. Such aspects are likely to be structurally
determined; they may set at odds the child' s apergu of his parent ' s
individual nat ure and the behavior he must direct toward and expect
from his parent in terms of cultural prescription. " Fat he r , " he must
think, "i s not acting like a human bei ng, " when he acts in accord-
ance with aut hori t ari an norms rat her t han with what is usually
called " humani t y. " Therefore, in terms of subliminal appreciation
of cultural classifications, he may be t hought to be acting like some-
thing outside humani t y, most frequently an ani mal . "And if, as an
animal, rat her t han t he person I know, he exercises power over me,
then I may borrow or drai n that power if I too assume the culturally
defined attributes of t he ani mal I feel hi m to be . "
Life crises provide rituals in and by means of which relations
between structural positions and between t he i ncumbent s of such
positions are restructured, often drastically. Seniors take t he respon-
sibility for actually maki ng t he changes prescribed by custom; they,
at least, have the satisfaction of taking an initiative. But juniors, with
less understanding of t he social rationale of such changes, find t hat
their expectations with regard to t he behavior of seniors toward
them are falsified by reality during times of change. From their
structural perspective, therefore, the changed behavior of their
parents and other elders seems t hreat eni ng and even mendacious,
perhaps even reviving unconscious fears of physical mutilation and
other punishments for behavior not in accordance with parent al
will. Thus, while the behavior of seniors is within t he power of t hat
age groupand to some extent t he structural changes they promot e
are for t hem predi ct abl et he same behavior and changes are
beyond the power of juniors either to grasp or to prevent.
To compensate for these cognitive deficiencies, juniors and infe-
riors, in ritual situations, may mobilize affect-loaded symbols of great
power. Rituals of status reversal, according to this principle, mask
176
The Ritual Process
t he weak in st rengt h and demand of t he st r ong t h a t t h e y b e passive
and pat i ent l y endure t he symbolic and even r e al a g g r e s s i o n shown
against t hem by st ruct ural inferiors. However , i t is n e c e s s a r y here
to revert to t he distinction made earlier be t we e n r i t u a l s of status
elevation and rituals of status reversal. I n t h e f o r me r , aggressi ve
behavior by candi dat es for hi gher status, t h o u g h o f t e n present
tends to be mut ed and const rai ned; after al l , t h e c a n d i d a t e is
"goi ng u p " symbolically, and, at t he end of t h e r i t u a l , wi l l enjoy
mor e benefits and rights t han heretofore. But , i n t h e l a t t e r , t h e group
or category t hat is permi t t ed to act as if it wer e s t r u c t u r a l l y super i or
and in this capacity to ber at e and bel abor its p r a g ma t i c super i or s
is, i n fact, perpet ual l y of a lower status.
Clearly, bot h sociological and psychological mode s of e x p l a n a t i o n
are pert i nent here. What is st ruct ural l y "vi si bl e " t o a t r a i n e d a n t h r o p -
ological observer is psychologically " unconsci ous " t o t h e i n di vi dua l
member of t he observed society; yet his orect i c r es pons es t o s t r uc t ur a l
changes and regularities, mul t i pl i ed by t he n u mb e r o f me mb e r s
exposed to change generat i on after gener at i on, ha ve t o b e t a k e n into
cultural, not abl y ri t ual , account if t he society is t o s u r v i v e wi t h out
disruptive tension. Life-crisis rites and ri t ual s of r e ve r s al t a k e these
responses into account in different ways. Th r o u g h s uc c e s s i ve life
crises and rites of status elevation, i ndi vi dual s as cend s t r uc t ur a l l y .
But rituals of status reversal make visible i n t hei r s y mb o l i c and
behavi oral pat t erns social categories and forms of g r o u p i n g t h a t are
considered to be axi omat i c and unchangi ng bot h i n e s s e n c e a n d in
relationships to one anot her.
Cognitively, not hi ng underlines r egul ar i t y so wel l as a b s u r d i t y
or par adox. Emotionally, not hi ng satisfies as muc h as e x t r a v a g a n t
or t emporari l y permi t t ed illicit behavi or. Ri t ual s of s t a t us r ever s al
accommodat e bot h aspects. By maki ng t he l ow hi gh a n d t h e hi gh
low
T
they reaffirm t he hi erarchi cal pr i nci pl e. By ma k i n g t h e l ow
mi mi c (often to t he poi nt of cari cat ure) t he behavi or of t h e hi gh,
and by restraining t he initiatives of t he pr oud, t hey u n d e r l i n e t he
reasonableness of everyday culturally pr edi ct abl e be havi or b e t we e n
t he various estates of society. On this account , it is a p p r o p r i a t e t h a t
Humility and Hierarchy
177
rituals of status reversal ar e often located either at fixed points in
the annual cycle or in relation to movable feasts t hat vary within a
limited period of time, for structural regularity is here reflected i n
temporal order. It might be argued t hat rituals of status reversal
are also found contingently, when calamity threatens the total
community. But one can cogently reply by saying t hat it is precisely
because the whole communi t y is threatened t hat such countervailing
rites are performedbecause it is believed that concrete historical
irregularities alter t he nat ural balance between what are conceived
to be permanent st ruct ural categories.
COMMUNITAS AND STRUCTURE
IN RITUALS OF STATUS REVERSAL
To return to rituals of status reversal. Not only do they reaffirm the
order "of st ruct ure, - they aIso~~restore relations between t he actual
' historical individuals who occupy positions in t hat structure. All
human societies implicitly or explicitly refer to two contrasting social
models. One, as we have seen, is of society as a structure of j ur al , _
f i i i i Wt f . ' '
political, and ^economic positions, offices, statuses, and roles, in
which the individual is only ambiguously_gr^rjgdJie]iixicLthe_so.cial
persona. The^ol her is of society as a communi t as of concrete idio-
syncratic individualsTwho, t hbugrTcl i fl ermgTn^^si cal and ment al
endowment, are nevertheless regarded as equal in terms of shared
humanity. The first model is of a differentiated, culturally structured,
segmented, and often hierarchical_system of institutionalized posi^
tigns, The secqnd presents_society as an undifferentiated, homo-
geneous whole, qri^hjehJrdividuals confront one anot her integrally,
and not as " s e g me n t a t e d " into statuses and roles.
In the process of social life, behavior in accordance wi t h one model
tends to "drift a wa y " from behavior in terms of t he other. Th e
ultimate desideratum, however, is to act in terms of communi t as
values even while pl ayi ng structural roles, where what one culturally
does is conceived of as merely i nst rument al to t he ai m of at t ai ni ng
1
7
8
The Ritual Process
and mai nt ai ni ng communi t as. Seen from t hi s p e r s p e c t i v e , the
seasonal cycle may be regarded as a measure of t h e d e g r e e of drift of
st ruct ure from communi t as. Thi s is par t i cul ar l y t r u e o f t h e rel at i ons
between very high- and very low-ranked social c a t e g o r i e s a n d groups,
t hough it holds good for relations bet ween i n c u mb e n t s o f a n y rank
or social position. Me n use t he aut hori t y vest ed i n t h e i r office to
misuse and abuse t he i ncumbent s of lower pos i t i on s a n d confuse
position wi t h its i ncumbent . Ri t ual s of st at us r e ve r s a l , e i t h e r pl aced
at strategic points in t he annual circle or g e n e r a t e d b y disasters
conceived of as bei ng the result of grave social si ns, a r e t h o u g h t of
as bri ngi ng social st ruct ure and communi t as i n t o r i g h t mut ual
relation once agai n.
TH E APO CEREMONY OF TH E AS H ANTI
To illustrate, I quot e a familiar exampl e from a n t h r o p o l o g i c a l
l i t erat ure concerning t he Apo ceremony of t he n o r t h e r n As h a n t i of
Ghana. Thi s ceremony, whi ch Rat t r ay (1 923 ) obs e r ve d a mo n g t he
Teki man peoples, takes pl ace duri ng t he ei ght days i mme d i a t e l y
preceding t he Teki man new year, whi ch begins on Apr i l 1 8 . Bos ma n
(1 705 ), t he early Dut ch historian of t he Coast of Gu i n e a , des cr i bes
what Rat t r ay calls "undoubt edl y one and t he s a me c e r e mo n y "
(p. 1 5 1 ) in the following t er ms: t here is " . . . a Fe as t of e i g h t days
accompani ed with all manner of Singing, Ski ppi ng, Da n c i n g ,
Mi r t h, and Jol l i t y; in whi ch t i me a perfect l a mp o o n i n g l i b e r t y is
allowed, and Scandal so highly exalted, t hat t hey ma y f r e e l y say
of all Faults, Villainies, and Frauds of t hei r Super i or s , a s we l l as
Inferiours wi t hout Puni shment or so much as t he l east i n t e r r u p t i o n "
(Bosman, Letter X) .
Rat t r ay' s observations abundant l y confirm Bos man' s c h a r a c t e r -
ization. He derives t he t erm Apo from a root me a n i n g " t o s pe a k
roughly or harshly t o, " and points out t hat an a l t e r na t i ve t e r m for
t he ceremony ahorohorua is possibly derived from t he v e r b horo, " t o
wash, " " t o cleanse. " Th a t t he Ashant i make a posi t i ve c o n n e c t i o n
Humility and Hierarchy
between frank, rough speech and purification is demonstrated by
the words of t he old high priest of the god Ta Kese at Teki man as
told to and literally translated by Rat t r ay:
You know that every one has a sunsum (soul) that may get hurt or knocked
about or become sick and so make the body ill. Very often, although there
may be other causes, e.g., witchcraft, ill health is caused by the evil and the
hate that another has in his head against you. Again, you too may have
hatred in your heart against another, because of something that person has
done to you, and that, too, causes your sunsum to fret and become sick. Our
forbears knew this to be the case, and so they ordained a time, once every
year, when every man and woman, free man and slave, should have free-
dom to speak out just what was in their head, to tell their neighbours just
what they thought of them, and of their actions, and not only to their
neighbours, but also the king or chief. When a man has spoken freely thus,
he will feel his sunsum cool and quieted, and the sunsum of the other person
against whom he has now openly spoken will be quieted also. The King of
Ashanti may have killed your children, and you hate him. This has made
him ill, and you ill; when you are allowed to say before his face what you
think you both benefit (p. 1 5 3 ) .
It can be seen at once from this indigenous interpretation t hat
leveling is one of the principal functions of the Apo rites. The high
must submit to being humbl ed; the humbl e are exalted through the
privilege of plain speaking. But there is much more to the ritual
than this. St ruct ural differentiation, bot h vertical and horizontal,
is the foundation of strife and factionalism, and of struggles in dyadic
relations between i ncumbent s of positions or rivals for positions. In
religious systems that are themselves structuredmost commonly
by the intercalated segmentations of the solar and l unar year and
by climatic nodal points of changequarrels and dissensions are
not dealt with ad hoc as they emerge, but in generic and omnibus
fashion at some regularly recurrent point in the ritual cycle. The
Apo ceremony takes place, as the Ashanti say, " whe n the cycle of
the year has come round " or when " the edges of the year have met . "
It provides, in effect, a discharge of all the ill-feeling t hat has accum-
ulated in structural relationships duri ng the previous year. To purge
The Ritual Process
or purify structure by pl ai n speaking is to reani mat e t he spi ri t of
communi t as. Here the widespread sub-Saharan African belief t h a t
grudges nourished in t he head or heart physically har m bot h t hose
who hold t hem and those against whom they are di rect ed oper at es
to insure t hat wrongs are ventilated and wrongdoers refrain from
taking reprisals against those who procl ai m their misdeeds. Si nce
it is more probabl e t hat persons of high rank wrong those of l ow
rank t han t he reverse, it is not surprising t hat chiefs and ari st ocrat s
are regarded as t he typical targets for publ i c accusation.
Paradoxically, t he ri t ual reduct i on of st ruct ure to communi t as
t hrough t he purifying power of mut ual honesty has t he effect of
regenerat i ng t he principles of classification and orderi ng on whi ch
social structure rests. On t he last day of t he Apo ritual, for exampl e,
j ust before t he new year begins, t he shrines of all t he local and s ome
of t he nat i onal Ashant i gods are carried in procession from t hei r
local temples, each wi t h an ent ourage of priests, priestesses, a n d
ot her religious officials, to t he sacred Tano River. Ther e t he shri nes
and t he blackened stools of deceased priests are sprinkled and pur i -
fied wi t h a mi xt ure of wat er and powdered white clay. Th e pol i t i cal
head of Teki man, t he chief, is not personally present. Th e Que e n
Mot her attends, however, for this is an affair of gods and pri est s,
representing t he universal aspects of Ashanti culture and soci et y
r at her t han of chieftainship in its more narrowl y st ruct ural aspect .
Thi s universal quality is expressed in t he pr ayer of t he pri est l y
spokesman of one of t he gods as he sprinkles the shrine of T a Kesi ,
t he greatest of t he local gods: " We beg you for life; when hunt er s
go to the forest, permi t t hem to kill meat ; may t he bearers of chi l d-
ren bear chi l dren: life to Yao Kr amo [the chief], life for all hunt er s ,
life to all priests, we have taken t he apo of this year and put it i n
t he r i ver " (pp. 1 6 4 - 1 6 6 ) . Wat er is sprinkled upon all t he stools a n d on
all those present, and after cleansing t he shrines, everyone r et ur ns t o t he
village while t he shrines are replaced in the temples t hat ar e t hei r
homes. Thi s solemn observance, which ends such a Sat ur nal i an
ritual, is in reality a most complex manifestation of Te ki man As hant i
cosmology, for each of t he gods represents a whole const el l at i on of
Humility and Hierarchy
181
values and ideas and is associated with a place in a cycle of myths.
Moreover, t he ent ourage of each replicates t hat of a chief and bodies
forth the Ashanti concept of structural hierarchy. It is as t hough
structure, scoured and purified by communitas, is displayed white
and shining again to begin a new cycle of structural time.
It is significant t hat t he first ritual of the new year, performed on
the following day, is officiated over by t he chief, and t hat no women,
not even the Queen Mot her, are allowed to be present. The rites
take place inside t he t empl e of Ta Kesi, the local god; the chief
prays to hi m alone and then sacrifices a sheep. This stands in marked
contrast to the rites of t he previous day, which are attended by
members of bot h sexes, held in t he open air by the waters of the
Tano River (i mport ant for all Ashanti), involve no bloody sacrifice,
and entail t he exclusion of t he chief. Communi t as is t he solemn not e
on which t he old year ends; structure, purified by communitas and
nourished by the blood of sacrifice, is reborn on the first day of the
new year. Thus, what is in many ways a ritual of reversal seems to
have t he effect, not only of temporarily inverting t he "pecki ng
order, " but of first segregating the principle of group unity from the
principles of hierarchy and segmentation and then of dramatically
indicating t hat t he uni t y of Teki manand, more t han Teki man, of
t he Ashanti state itselfis a hierarchical and segmentary unity.
SAMH AIN, ALL SOULS, AND ALL SAINTS
As noted, the emphasis on the purificatory powers of the structurally
inferior and the connection of such powers with fertility and other
universal human interests and values precede the emphasis on
fixed and particularistic structure in xh&Apo case. Similarly, Hallow-
een in Western culture, with its emphases on the powers of children
and eart h spirits, precedes two traditional Christian feasts t hat
represent structural levels of Christian cosmologyi.e., All Saints'
and All Souls' . Of All Saints' Day, t he French theologian M. Olier
(quoted in Attwater, i g6i ) has sai d: " I t is in some sort greater
The Ritual Process
t han the Feast of Easter or of the Ascension, [for] Christ is perfected
in this mystery, because, as our Head, He is only perfect and ful-
filled when He is united to all His members t he saints (canonized
and uncanoni zed, known and unknown) . "
Here again we meet wi t h the notion of a perfect synthesis of
communi t as and hierarchial st ruct ure. It was not only Dant e and
Thomas Aqui nas who pi ct ured heaven as a hierarchical st ruct ure
with many levels of sanctity and, at t he same time, as a luminous
unity or communi t as in which no lesser saint felt envy of a great er
nor great er saint any pri de of position. Equal i t y and hi erarchy
were there mysteriously one. All Souls' Day, whi ch follows, com-
memorat es t he souls in purgat ory, emphasizing at once their lower
hierarchical position to t he souls in heaven, and t he active commun-
itas of the living, who ask t he saints to intercede for those under -
going liminal ordeal in purgat ory and the saved dead both in heaven
and in purgat ory. But it would appear t hat , as in t he "l ampooni ng
l i ber t y" and status reversals of t he Apo ceremony, t he rude power
t hat energizes bot h t he virtuous hierarchy and t he good communi t as
of t he Saints and Souls of t he calendrical cycle is derived from pr e-
Christian and aut ocht honous sources t hat are often given infernal
status at t he level of folk Christianity. It was not until t he seventh
century t hat November i began to be observed as a Christian festival,
while All Souls' Day was brought into t he Roman Ri t e only in t he
t ent h century. I n Celtic regions, some aspects of t he pagan wi nt er
festival of Samhai n (our November i) were at t ached to these
Christian feasts.
Samhai n, which means " s umme r end, " accordi ng to J . A.
MacCul l och (1948) "nat ur al l y poi nt ed to t he fact t hat the powers
of blight, typified by winter, were begi nni ng their reign. But it may
have been part l y a harvest festival, while it had connections wi t h
pastoral activities, for t he killing and preserving of animals for food
for winter was associated wi t h it. . . . A bonfire was lit and repre-
sented t he sun, t he power of which was now wani ng, and t he fire
would be i nt ended to strengthen it magically. . . . I n dwellings t he
t he fires were extinguished, a pract i ce perhaps connected wi t h t he
Humility and Hierarchy
seasonal expulsion of evils. Branches were lit at the bonfire and
carried into t he houses to kindle the new fires. There is some evidence
t hat a sacrifice, possibly human, occurred at Samhain, laden as the
victim would be with the ills of the community, like the Hebrew
scapegoat " (pp. 58-59)-
Here, too, it would appear that, like the Apo ceremony, Samhain
represented a seasonal expulsion of evils, and a renewal of fertility
associated with cosmic and chthonic powers. I n European folk
beliefs, the midnight of Oct ober 3 1 has become associated with
gatherings of the hellish powers of witchcraft and t he devil, as in
Walpurgisnacht and Ta m o' Shanter' s near-fatal Halloween. Subse-
quently, a strange alliance has been formed between the innocent
and the wicked, children and witches, who purge t he community
by the mock pity and terror of trick or treat and prepare the way
for communitas feasts of sunlike pumpki n pieat least in the
Uni t ed States. Somehow, as dramatists and novelists well know, a
touch of sin and evil seems to be necessary tinder for the fires of
communi t asal t hough elaborate ritual mechanisms have to be
provided to t ransmut e those fires from devouring to domestic uses.
There is always a felix culpa at the heart of any religious system t hat
is closely bound up with human structural cycles of development.
TH E SEXES,
STATUS REVERSAL, AND COMMUNITAS
Ot her rituals of status reversal involve the supersession by women
of masculine authority and roles. They may be held at some node
of calendrical change as in the case of the Zulu Nomkubulwana cere-
mony, analyzed by Max Gl uckman (1954) where " a domi nant role
was ascribed to the women and a subordinate role to the men at
rites performed in local districts in Zululand when the crops had
begun to gr ow" (pp. 4 - 1 1 ) . (Similar rites, in which girls wear men' s
garments and herd and milk the cattle, are found in many southern
and central Bantu societies.) More frequently, rituals of this t ype
184
The Ritual Process
may be performed when a major t erri t ori al d i v i s i o n of a tribal
society is t hreat ened by some nat ur al cal ami t y, s u c h as a plague of
insects or famine and drought . Dr . Peter Ri gby ( 1 9 68 ) has recently
published a detailed description of women' s r i t e s of this variety
among t he Gogo of Tanzani a. These rites have b e e n elaborately dis-
cussed elsewhere by such aut hori t i es as Ei l e e n I Cr i ge, Gluckman,
and J unod. Thus , I shall poi nt out only t hat i n a l l t he situations in
which they occur, t here is a belief t hat t he m e n , some of whom
occupy key positions in t he social st ruct ure, h a v e s ome h ow incurred
t he displeasure of t he gods or ancestors, or, a l t e r n a t i ve l y , have so
altered t he mystical bal ance bet ween society a n d nat ur e t hat dis-
t urbances in t he former have provoked a b n o r ma l i t i e s in the latter.
Put briefly, st ruct ural superiors, t hr ough t h e i r dissensions over
particularistic or segmental interests, have b r o u g h t disaster on the
local communi t y. I t is for st ruct ural inferiors, t h e n ( i n the Zulu
case, young women, who are normal l y un de r t h e patria poteslas of
fathers or t he manus of husbands) , r e pr e s e nt i ng communi t as, or
global communi t y t ranscendi ng all i nt ernal di vi s i on s t o set things
ri ght agai n. They do this by symbolically u s u r p i n g for a short while
t he weapons, dress, accout erment s, and be h a vi or a l s t yl e of structural
superiorsi. e. , men. But an old form n ow h a s a new content.
Aut hori t y is now wielded by communi t as i t sel f mas quer adi ng as
st ruct ure. St ruct ural form is divested of selfish a t t r i b u t e s and puri-
fied by association wi t h t he values of c o mmu n i t a s . Th e unity that
has been sundered by selfish strife and c o n c e a l e d ill-feeling is
restored by those who are normal l y t hought of as be n e a t h the battle
for j ur al and political status. But " b e n e a t h " h a s t wo senses: it is
not only t hat whi ch is st ruct ural l y inferior; i t i s al s o t he common
basis of all social lifethe eart h and its fruits. I n o t h e r words, what
is law on one social dimension may be basi c on a n o t h e r .
I t is per haps significant t hat young mai de ns a r e often the mai n
prot agoni st s: t hey have not yet become t h e mo t h e r s of children
whose st ruct ural positions will once mor e p r o v i d e bases for opposi-
tion and compet i t i on. Yet, i nevi t abl y, r ever s al i s ephemeral and
transitory ( "l i mi nal , " if you like), for t he t wo mo d e s of social inter-
Humility and Hierarchy
relationship are here culturally polarized. For girls to herd is a
paradox for classification, one of those paradoxes that can exist only
in the liminality of ritual. Communi t as cannot mani pul at e re-
sources or exercise social control without changing its own nat ure
and ceasing to be communitas. But it can, through brief revelation,
"bur n out " or "was h away"what ever met aphor of purification
is usedthe accumulated sins and sunderings of structure.
STATUS REVERSAL IN
" TH E FEAST OF LOVE" IN VILLAGE INDIA
To summarize our findings so far on rituals of status reversal: t he
masking of the weak in aggressive strength and the concomitant
masking of the strong in humility and passivity are devices that
cleanse society of its structurally engendered " s i ns " and what
hippies might call "hang- ups . " The stage is then set for an ecstatic
experience of communitas, followed by a sober return to a now purged
and reanimated structure. One of the best "i ns i de" accounts of this
ritual process is provided in an article by the usually sober and dis-
passionate analyst of Indi an village society, Professor McKi m
Marriott ( 1 9 66) . He is discussing the Holi festival in the village of
Kishan Garhi , "l ocat ed across the J uma n from Mat hur a and
Vrindaban, a day' s walk from the youthful Krishna' s fabled land
of Vraja. " Indeed, the presiding deity of the rites was Krishna, and
the rites described to Marri ot t as " t h e feast of l ove" were a spring
festival, the "great est religious celebration of the year. " As a green
field worker, Marri ot t had been plunged into the rites the previous
year, inveigled into drinking a concoction containing marijuana,
smeared with ochre, and cheerfully drubbed. In t he intervening
year, he reflected on what might be the social function, a la Radcliffe-
Brown, of these turbulent rites:
Now a full year has passed in my investigations, and the Festival of Love was
again approaching. Again I was apprehensive for my physical person, but
was forewarned with social structural knowledge that might yield better
understanding of the events to come. This time, without the draft of
The Ritual Process
mari juana, I began to see the pandemoni um of Holi falling i nto a n extraordi-
narily regular social ordering. But this was an order preci sely i nverse to the
social and ritual principles of routine life. Each riotous act at H ol i implied
some opposite, positive rule or fact of everyday social org ani zati on in the
village.
Who were those smiling men whose shins were bei ng mos t mercilessly
beaten by the women? They were the wealthier Brahman and J a t farmers of
the village, and the beaters were those ardent local Radhas, the "wi ves of
the village," figuring by both the real and fictional i ntercaste system of kin-
ship. The wife of an "elder brother" was properly a man' s j oki ng mate
while the wife of a "younger brother" was properly remov ed from him by
rules of extreme respect, but both were merged here wi th a man' s mother-
surrogates, the wives of his " father's younger brothers," i n one revoluti onary
cabal of" wives " that cut across all lesser lines and links. Th e boldest beaters
in this veiled battalion were often in fact the wives of the f armers' low-caste
field-laborers, artisans, or menialsthe concubines and ki tchen help of the
victims. " Go and bake bread!" teased one farmer, eggi ng hi s assailant on.
" Do you want some seed from me ?" shouted another flattered victim,
smarting under the blows, but standing his ground. Si x Br ahman men in
their fifties, pillars of vi llage society, li mped past in panti ng fli ght from the
quarter staff wielded by a massive young Bhangi n, sweeper of thei r latrines.
From this carnage suffered by their vi llage brothers, all daughters of the
village stood apart, yet held themselves in readiness to attack any potential
husband who mi ght wander in from another, marri ageable vi l l age to pay a
holiday call.
Who was that " Ki ng of the H ol i " riding backwards on the donkey? It
was an older boy of high caste, a famous bully, put there by hi s organized
victims (but seeming to relish the promi nence of his di sgrace).
Who was in that chorus singing so lustily in the potters' l ane ? Not just
the resident caste fellows, but six washermen, a tailor, and three Brahmans,
joined each year for this day only in an idealistic musical c ompany patterned
on friendships of the gods.
Who were those transfigured "cowherds" heaping mud and dust on all
the leading citizens? They were the water carrier, two young Brahman
priests, and a barber' s son, avi d experts in the daily routi nes of purification.
Whose household temple was festooned with goats' bones by unknown
merrymakers? It was the temple of the Brahman wi dow who had constantly
harassed neighbors and kinsmen wi th actions at law.
Humility and Hierarchy
187
In front of whose house was a burlesque dirge being sung by a professional
ascetic of the village? It was the house of a very much alive moneylender,
notorious for his punctual collections and his insufficient charities.
Who was it who had his head fondly anointed, not only with handfuls of
the sublime red powders, but also with a gallon of diesel oil ? It was the
village landlord, the anointer was his cousin and archrival, the police head-
man of Ki shan Garhi .
Who was it who was made to dance in the streets, fluting like Lord
Krishna, with a garland of old shoes around his neck? It was I, the visiting
anthropologist, who had asked far too many questions, and had always to
receive respectful answers.
Here indeed were the many village kinds of love confoundedrespectful
regard for parents and patrons; the idealized affection for brothers, sisters,
and comrades; the longing of man for union with the divine; and the rugged
lust of sexual matesall broken suddenly out of their usual, narrow channels
by a simultaneous increase of intensity. Boundless, unilateral love of every
kind flooded over the usual compartmentalization and indifference among
separated castes and families. Insubordinate libido inundated all established
hierarchies of age, sex, caste, wealth, and power.
Thesocial meaning of Krishna' s doctrine in its rural North Indi an recension
is not unlike one conservative social implication of Jesus' Sermon on the
Mount. The Sermon admonishes severely, but at the same time postpones
the destruction of the secular social order until a distant future. Kri shna
does not postpone the reckoning of the mighty until an ultimate Judgment
Day, but schedules it regularly as a masque at the full moon of every March.
And the Holi of Kri shna is no mere doctrine of love: rather it is the script
for a drama that must be acted out by each devotee passionately, joyfully.
The dramati c balancing of Holithe world destruction and world
renewal, the world pollution followed by world purificationoccurs not
only on the abstract level of structural principles, but also in the person of
each participant. Under the tutelage of Kri shna, each person plays and for
the moment may experience the role of his opposite; the servile wife acts the
domineering husband, and vice versa; the ravisher acts the ravished; the
menial acts the master; the enemy acts the friend; the strictured youths act
the rulers of the republic. The observing anthropologist, inquiring and
reflecting on the forces that move men in their orbits, finds himself pressed.to
act the witless bumpkin. Each actor playfully takes the role of others in
relation to his own usual self. Each may thereby learn to play his own
The Ritual Process
routine roles afresh, surely with renewed understanding, possi bl y witjj
greater grace, perhaps with a reciprocating love (pp. 2 1 0 - 2 1 2 ) .
I have one or two small cavils wi t h Marri ot t ' s ot her wi s e a d mi r a b l e
and empat het i c account . It is not t he biological dr i ve of " l i b i d o "
t hat "i nundat es all established hierarchies of age, sex, c a s t e , we a l t h ,
and power, " but the liberated experience of communi t as , w h i c h , ^
Blake mi ght have said, is " a n intellectual t hi ng" i . e . , i t i nvol ves
total cognizance of anot her' s human total. Co mmu n i t a s is not
merely i nst i nct ual ; it involves consciousness and vol i t i on. St at us
reversal in t he Holi festival liberates t he ma n (and w o ma n ) from
the status. Under certain conditions this can be a n " e c s t a t i c "
experience, in t he etymological sense of the i ndi vi dual ' s " s t a n d i n g
out si de" his st ruct ural status. " Ec s t a s y " = " e xi s t e nce . " Ag a i n , I
would not altogether derive t he "r eci pr ocat i ng l o v e " s e n s e d by
Marri ot t from the actor' s t aki ng t he role of an alter. Ra t h e r , I woul d
regard this mock role-playing merely as a device to de s t r oy all roles
and pr epar e for t he emergence of communi t as. But Ma r r i o t t has
well described and grasped t he salient characteristics of a r i t u a l of
status reversal: t he ri t ual domi nance of st ruct ural i nfer i or s , t hei r
bl unt speaking and rough doi ng; t he symbolic humi l i t y a n d act ual
humiliation of their status superi ors; t he way in whi ch t h o s e st r uc-
turally " be l ow" represent a communi t as t hat floods across s t r uc t ur a l
boundaries, t hat begins wi t h force and ends wi t h l ove; a n d , finally,
t he stressing, not t he overt hrowi ng of t he principle of h i e r a r c h y (i.e.,
of graded organization), undoubt edl y purifiedeven, pa r a doxi c a l l y,
by the breach of many Hi ndu pollution rul est hrough r e ve r s al , a
process whereby it remains t he st ruct ural vert ebra of vi l l age life.
RELIGIONS OF H UMILITY
AND OF STATUS REVERSAL
I have thus far been discussing liminal rites in rel i gi ous systems
belonging to societies t hat are highly st ruct ured, cyclical, a n d repet i -
tive. I would like to cont i nue by tentatively suggesting t h a t a dis-
Humility and Hierarchy
tinction similar to t he one we have been maki ng between the liminality
of rites of status elevation and the liminality of rites of status reversal
may be found, at least in their early stages, in religions of wider
than tribal scope, especially duri ng periods of rapi d and unpre-
cedented social changewhi ch themselves have liminal attributes.
In other words, some religions resemble t he liminality of status
elevation: They emphasize humility, patience, and t he uni mport -
ance of distinctions of status, property, age, sex, and other nat ural
and cultural differentiae. Furt hermore, they stress mystical union,
numinosity, and undifferentiated communi t as. Thi s is because
many of t hem regard this life as itself constituting a liminal phase
and the funerary rites as prepari ng for t he reaggregation of initiands
to a higher level or pl ane of existence, such as heaven or ni rvana.
Other religious movements, on t he contrary, exhibit many of t he
attributes of tribal and peasant rituals of status reversal. Th e limin-
ality of reversal did not so much eliminate as underl i ne structural
distinctions, even to t he point of (often unconscious) caricature.
Similarly, these religions are conspicuous for their emphasis on
functional differentiation in t he religious sphere, and/ or t he religious
reversal of secular status.
STATUS REVERSAL IN
SOUTH AFRICAN SEPARATISM
One particularly clear example of a religion of status reversal can
be found in Sundkler' s study of Bant u separatism in South Africa
( 1 9 61 ) . As is well known, t here are now well over a thousand more
or less small African-organized churches and sects in South Africa,
which have broken off either from white mission churches or from
one another. Sundkler, who studied African independent churches
in Zululand, has this to say about " t h e reversed colour bar in
heaven":
In a country where some irresponsible Whites tell the Afri can that Jesus is
only for the Whi te man, the Afri can takes his revenge by projecting the
colour bar right into heavenly places. The colour-complex has painted their
190 The Ritual Process
PSEUD O H I ERARCH I ES
IN MELANESIAN MI LLENARI ANI SM
Al t hough t he l i t erat ure on religious and s emi r el i gi ous movements
does not lend complete suppor t to t he view I h a v e b e e n taking, and
many problems and difficulties r emai n, t her e i s nevert hel ess strong
very heaven black, and die Black Chri st has to see to t hat . Shembe [
a
famous Zulu prophet] at the gates turns away the Whi tes, because they, as
the rich man, have already in their lifetime received thei r g ood thing, and
he opens the gate only to his faithful followers. The fate of the Afri can who
belonged to the Whi te mission churches is lamentable: " O n e race cannot
enter by the gate of another race," on the arrival at the Wh i t e people's gate
they are turned away. . . . The colour-complex takes the parabl es of Jesus
into its service. Here is one to whi ch I have heard references i n some Zionist
churches: "There were ten virgins. And five of them we r e Whi te, and five
were Black. The five Whi tes were foolish, but the five Bl acks were wise, they
had oil in their lamps. All ten came to the gate. But the five Whi te virgins
received the same answer as the rich man recei ved: Be c aus e the Whites
rule on earth, the Blacks do so in Heaven. The Whi tes wi l l g o a-begging to
dip the tip of their finger in cool water. But they will get as a repl y:' Hhayyi
(no)nobody can rule twi ce"' (p. 290).
I t will be not ed t hat here status reversal is n o t p a r t of a total
system of rituals, t he final effect of whi ch is to p r o mo t e reconciliation
between t he different st rat a of t he st ruct ural h i e r a r c h y . We are not
dealing with an i nt egrat ed social system i n wh i c h s t r uc t ur e is per-
vaded by communi t as; t hus, we have only t he r e v e r s a l aspect em-
phasized, wi t h every hope t hat this will be ma n ' s ul t i mat e state.
Nevertheless, t he exampl e is instructive in t h a t i t s ugges t s t hat reli-
gions t hat stress hi erarchy, whet her direct or i n v e r t e d , as a general
at t ri but e of religious life, are generat ed from t h e mi d s t of the struc-
turally inferior in a socio-political system t h a t r es t s as much upon
force as on consensus. I t rnight be wor t h poi nt i ng o u t her e too that
many of these Sout h African sects, small t h o u g h t h e y are, have
elaborate clerical hierarchies, and t hat women oft en occupy import-
ant ritual roles.
Humility and Hierarchy
evidence t hat religious forms clearly at t ri but abl e to the generative
activities of structurally inferior groups or categories soon assume
many of the external characteristics of hierarchies. Such hierarchies
may merely reverse secular ranking, or they may altogether replace
the secular framework either in the ecclesiastical structure of the
movement or in its eschatological beliefs. A good example of a
movement that, in its organizational form, at t empt ed to replicate
the European social structure may be found in Peter Lawrence' s
Road Belong Cargo ( 1 9 64) . I n the program of Yali, one of the Madang
Melanesian prophet s:
The people were to give up living in hamlets and come together in large
"camps," which were to have their houses built along streets, and to be
beautified with flowers and shrubs. Each " camp " was to have a new " Rest
House," which was no longer to be called a haus kiap, but a hausyali. It was
to be used by Yali when he visited the people in his capacity as an Admini-
stration Officer. Each " camp " had to have proper latrines, and new roads
had to be cut throughout the area. . . . The old headmen would have to be
superseded by " boss boys," who would supervise the work of reconstruction
and also see that Yali's orders were carried out. Monogamy was enjoined,
second wives would be divorced and married to bachelors (p. 160).
Ot her features imitating European administrative structure and
material and religious culture were introduced into this "Car go
cult. " Many other Cargo cults, of course, have similar organizational
features, and, in addition, hold to the belief that the Europeans will
be driven out or destroyed, but t hat their own ancestors and living
prophets will govern t hem in a pseudo-bureaucratic structure. It is
not at all certain, however, t hat the liminal-religious generation of
pseudohierarchies is solely t he outcome of structural inferiority. The
factor of status reversal is, I am convinced, correlated with permanent
structural inferiority. But, it may well be that elaborately ranked
ritual or ceremonial hierarchies represent the liminality of secularly
egalitarian groups, regardless of t he rank of such groups in t he wider
society. One could instance the Freemasons, the Rosicrucians, t he
Elks, the Sicilian Mafia, and other kinds of secret societies and
I Q 2
The Ritual Process
brot herhoods, wi t h el aborat e ri t ual and c e r e mo n i a l , a n d with gener-
ally a strong religious tinge. Th e member s hi p of s u c h groups is often
dr awn from socio-political communi t i es of s i mi l a r l y r anked persons,
wi t h shared egalitarian values and a s i mi l ar l e ve l of economic
consumpt i on.
I t is t rue t hat in these cases t her e is an as pe ct o f reversal , toofor
secular equality is cont radi ct ed by l i mi nal h i e r a r c h y b u t this is
not so much a reversal of r ank order wi t hi n a p a r t i c u l a r structural
system as t he substitution of one t ype of s ys t em ( a hi erarchi cal one)
for anot her (an egalitarian one) . I n some cas e s , a s wi t h the Mafia,
t he Ku Kl ux Kl an, and some Chinese secret s oc i e t i e s , liminal hier-
archy acquires i nst rument al political val ues a n d funct i ons and loses
its "pl ay- act i ng, " fantastical qual i t y. Wh e n t h i s happens, the
directed, purposive charact er of political or quas i - mi l i t ar y action
may well find t he hierarchical form congeni al t o i t s organizational
requi rement s. Th a t is why it is so i mpor t ant , w h e n studying such
groups as t he Freemasons and t he Hel l ' s An g e l s mot or cycl e gangs
of California and compari ng t hem wi t h o n e a n o t h e r , to specify
what phase they have reached in t hei r d e v e l o p me n t a l cycle and
under what social field conditions t hey c u r r e n t l y exi st .
SOME MODERN EX AMP L ES
OF REVERSAL AND PS EUDO - H I ERARCH Y
It may be objected t hat in these l i mi nal mo v e me n t s hierarchical
organization necessarily develops as t he n u m b e r of members in-
creases; however, many examples show t h a t s u c h movement s have a
mul t i t ude of offices but a small numbe r of me mb e r s . For example,
Allan C. Speirs, of Cornell University ( u n p u b l i s h e d thesis, 1 9 66) ,
describes how t he Aaroni t es' communi t y o f Ut a h , a Mor mon
separatist sect number i ng not many mor e t h a n t wo hundred souls,
nevertheless possessed " a complicated h i e r a r c h i c a l st ruct ure some-
what similar to t hat of Mor moni s m . . . h a v i n g such positions as
First Hi gh Priest, Second Hi gh Priest, P r e s i d e n t , First Vice-Presi-
Humility and Hierarchy
193
dent, Second Vice-President, Priests of Branches, Bishops of Coun-
cils, Teachers, and Deacons " (p. 22) . A rat her different kind of
group, described in several published articles and unpublished
manuscripts by R. Li ncol n Kei ser, of t he Universityof Rochester,is
the Conservative Vice Lords, a gang or " c l u b " or " n a t i on " of
adolescent Negro youths in Chicago. Mr . Kei ser generously gave me
access to the colorful aut obi ography of " Te ddy , " one of t he leaders
of the Vice Lords. The Vice Lords had a number of ceremonial
activities, such as a " Wi n e Cer emony" for their dead and those in
penitentiaries, and on these and other occasions they wore black
and red capes as ceremonial dress.
What is particularly striking about t he Vice Lords and such other
gangs as t he Egypt i an Cobras and t he Imperi al Chaplains is the
complex and hierarchical nat ure of their organization. For example,
the Vice Lords were divided into "Seni or s, " "Juni or s , " and " Mi d-
gets," dependi ng on the time of joining, and into territorial branches,
the sum of which constituted the "Vi ce Lord Nat i on. " " T e d d y "
describes t he organizational structure of the St. Thomas br anch:
"Everybody in t he group at St. Thomas when they first started had
some kind of position. The officers were President, Vice-President,
Secretary-Treasurer, Chief War Councilor, War Councilor, and
they had Sergeant s-at -Arms" (p. 1 7) . In t he main, t he behavior of
the gang members was fairly casual and egalitarian, when they
were not fighting among themselves over the control of territory.
But their structure in formal and ceremonial situations was the
reverse of egalitarian. Ther e was a strict pecking order, while
branches t hat sought to become i ndependent of the original " c l u b "
were swiftly brought into line.
Another contemporary example of the tendency for structurally
inferior categories to have hierarchical liminality is provided by the
young motorcycle riders of California known as the Hell' s Angels.
Hunt er S. Thompson (1966) claims t hat most of t he members are
sons of people who came to California before World War I I hi l l -
billies, Okies, Arkies, and Appalachians (p. 20 2) . Today the men
are "l ongshoremen, warehousemen, truck drivers, mechanics, clerks,
i94
The Ritual Process
and casual laborers at any work t hat pays quick wages a n d t h a t
requires no allegiance. Perhaps one in ten has a steady j ob a n d a
decent i ncome" (pp. 73 - 74) . They call themselves t he o n e - p e r c e n -
ters, " t h e one percent that don' t fit and don' t c a r e " (p. 1 3 ) . T h e y
refer to members of the " s t r ai ght " world as "ci t i zens , " w h i c h
implies that they themselves are not. They have opt ed out o f t h e
structural system. Nevertheless, like the Negro Vi ce Lor ds , t h e y
constitute a formal organization with complex initiation c e r e mo n i e s
and grades of membershi p emblematized by badges. Th e y h a v e a
set of bylaws, an executive committee, consisting of pr es i dent , v i c e -
president, secretary, treasurer, and sergeant -at -arms, a n d f o r ma l
weekly meetings.
Among t he Hell' s Angels we find replication of t he s t r u c t u r e of
secular associational organization, rat her t han status r ever s al . Bu t
we do find elements of status reversal in their initiation c e r e mo n i e s ,
duri ng which t he Angel recruits bri ng clean new Levis a n d j a c k e t s
to the rite, only to steep t hem in dung, urine, and oil. Th e i r d i r t y
and ragged condition, " r i pe n e d" to the poi nt of di s i nt e gr at i on, is
a sign of status t hat reverses t he " ne at and cl e an" s t a n d a r d of
"ci t i zens " t rapped in status and st ruct ure. But, despite t hei r p s e u d o -
hierarchies, bot h the Vice Lords and the Angels stress t he v a l u e s of
communi t as. The Vice Lord " Te ddy , " for exampl e, sai d o f t h e
general publ i c: "And then pretty soon they said we had a n o r g a n i -
zation. But all we t hought , we just buddy- buddy. " ( Ke i s e r , 1 9 6 6 ) .
Thompson, too, frequently stresses the "t oget herness g r o u p " c h a r -
acter of the Hell' s Angels. Thus pseudost ruct ure does not a p p e a r t o
be inconsistent wi t h real communi t as. These groups are p l a y i n g t h e
game of structure rat her t han engaging in t he s oc i oe c on omi c
structure in real earnest. Thei r st ruct ure is " expressive " in t h e ma i n ,
t hough it has instrumental aspects. But expressive st r uct ur es o f t hi s
type may under certain circumstances be converted into p r a g ma t i c
structures, as in the case of Chinese secret societies, such as t h e T r i a d
society discussed in Gustaaf Schlegel' s The Hung League ( 1 8 6 6 ) .
Similarly, t he ceremonial structure of t he Poro society of Si e r r a
Leone was used as t he basis of a politically rebellious o r g a n i z a t i o n
in t he Mende Rising of 1898 (Little, 1965, passim).
Humility and Hierarchy
195
RELIGIONS OF H UMILITY
WITH H IGH - STATUS FOUNDERS
There are many examples of religions and ideological and ethical
movements t hat have been founded by persons of high, or, if not
high, of solidly respectable, structural status. Significantly, the basic
teachings of these founders are full of references to t he stripping off
of worldly distinctions, property, status, and the like, and many of
them stress the "s pi r i t ual " or "s ubs t ant i al " identity of male and
female. In these and in many other respects the liminal religious
condition they seek to bring about , in which their followers are
withdrawn from the world, has close affinities with t hat found in the
liminality of seclusion in tribal life-crisis ritesand, indeed, in other
rituals of status elevation. Abasement and humility are regarded
not as t he final goal of these religions but simply as attributes of
the liminal phase through which believers must pass on their way
to the final and absolute states of heaven, nirvana, or Utopia. It is a
case of reculer pour mieux sauter. When religions of this type become
popular and embrace the structurally inferior masses, there is often
a significant shift in the direction of hierarchical organization. In a
way, these hierarchies are "i nver t ed"at any rat e in terms of the
prevalent belief systemfor the leader or leaders are represented,
like the Pope, as "servant s of the servants of God" rat her t han as
tyrants or despots. Status is acquired through t he stripping of
worldly authority from the incumbent and the put t i ng on of meek-
ness, humility, and responsible care for members of the religion,
even for all men. Nevertheless, just as in t he South African Separa-
tist sects, the Melanesian Cargo cults, the Order of Aaron, Negro
adolescent gangs, and the Hell' s Angels, the popul ar expansion of a
religion or a ceremonial group often leads to its becoming hierarchical.
In the first place, there is the probl em of organizing large numbers.
In the secondand this is seen in small sects with complex hier-
archiesthe liminality of the poor or weak assumes t he trappings of
secular structure and is masked in parent al power, as we saw earlier
in the discussion of ani mal and monstrous disguises.
19.6
The Ritual Process
The Buddha
As examples of structurally superior or we l l - e nt r e nche d religious
founders who preached t he values of humi l i t y and c o mmu n i t a s , one
mi ght cite t he Buddha, St. Francis, Tolstoy, and Ga n d h i . T h e case of
Jesus is less clear-cut: while Mat t he w and Luke t r ace t h e descent of
his pater Joseph to Ki ng Davi d, and while t he i mp o r t a n c e a n d status
of a carpent er are high in many peasant societies, J e s u s is usually
considered to be " a ma n of t he peopl e. " Th e Bu d d h a ' s fat her was
reported to be an i mpor t ant chief among t he t r i be o f t h e Sakiyas,
while his mot her, Ma h a Maya, was t he daught er of a nei ghbor i ng
king in a region to t he southeast of t he Hi mal ayas. Ac c o r d i n g to the
received account, Si ddhart ha, as t he pri nce was k n o wn , l ed a shel-
tered life for 29 years behi nd t he prot ect i ve walls of t h e r oy a l palace,
waiting to succeed his father. Next comes t he c e l e br a t e d t al e of his
three ventures into t he world beyond t he gates wi t h hi s coachman
Channa, duri ng whi ch he encount ered successively a n old man
worn out with labor, a leper, and a rot t i ng corpse, a n d s aw at first
hand t he lot of st ruct ural inferiors. After his first e xpe r i e n c e of death,
on his ret urn to t he pal ace, he was met by t he s ound of mus i c cele-
brat i ng t he arrival of his first-born son and h e i r a s s ur a n c e of the
structural continuity of his line. Far from bei ng d e l i g h t e d , he was
disturbed by this further commi t ment to t he domai n of a ut h or i t y and
power. Wi t h Channa he stole away from t he pal ace a n d wandered
for many years among t he common peopl e of I ndi a, l e a r n i n g much
about t he realities of t he caste system. For a while he b e c a me a severe
ascetic with five disciples. But this modal i t y of s t r uc t ur e , t oo, did not
satisfy hi m. And when he entered his cel ebrat ed me di t a t i on for
forty days under t he Bo tree, he had al ready c ons i de r a bl y modified
t he rigors of t he religious life. Havi ng at t ai ned e n l i gh t e n me n t , he
spent t he last 45 years of his life t eachi ng what was i n effect a simple
lesson of submission and meekness t o all people, i r r e s pe c t i ve of race,
class, sex, or age. He did not pr each his doctrines for t h e benefit of a
single class or caste, and even t he lowest Pari ah mi g h t , a n d some-
times di d, call himself his disciple.
Humility and Hierarchy
197
I n t he Buddha we have a classic case of a "s t r uct ur al l y" well-
endowed religious founder who underwent initiation into communi -
tas t hrough stripping and equalizing and put t i ng on t he behavior of
weakness and poverty. I n I ndi a itself, one could cite many further
examples of structural superiors who renounced weal t h and position
and preached holy poverty, such as Cai t anya (see Chapt er 4) ;
Mahavl ra, t he founder of Jai ni sm, who was an older contemporary
of the Buddha; and Nanak, t he founder of Sikhism.
Gandhi
In recent times, we have had the impressive spectacle of t he life and
mart yrdom of Mohandas Kar amchand Gandhi , who was at least as
much a religious as a political leader. Like t he others j ust men-
tioned, Gandhi came from a respectable segment of t he social hier-
archy. As he writes in his aut obi ography ( 1 9 48 ) : " T h e Gandhi s . . .
for three generations from my grandfather . . . had been pri me mi n-
isters in several Kat hi awad St at es" (p. 1 1 ) . His father, Kaba Gandhi ,
was for some time Pri me Minister i n Rajkot and t hen i n Vankaner.
Gandhi studied law in London and afterward went to South Africa
on legal business. But soon he renounced wealth and position to lead
the South African Indi ans in their struggle for greater justice,
developing t he doctrine of nonviolence and "t r ut h- for ce" into a
powerful political and economic i nst rument .
Gandhi ' s later career as mai n leader of t he Nat i onal Independence
movement in I ndi a is well known to all. Here I woul d merely like
to quote from his aut obi ography ( 1 9 48 ) some of his t hought s on t he
virtues of stripping oneself of propert y and maki ng oneself equal to
all. Gandhi was always devoted to t he great spiritual guide of
Hinduism, the Bhagavad Gita, and i n his spiritual crises he used t o
t urn to "t hi s dictionary of conduct " for solutions of his i nner
difficulties:
Words like aparigraha [nonpossession] and sambhava [equability] gripped me.
How to cultivate and preserve that equability was the question. How was
one to treat alike insulting, insolent and corrupt officials, co-workers of
The Ritual Process
yesterday raising meaningless opposition and men who had al ways been
good to one ? H ow was one to divest oneself of all possessions ? Was not the
body itself possession enough? Were not wife and children possessi ons? Wa s
I to destroy all the cupboards of books I had ? Was I to gi ve up all I had
and follow H i m? Straight came the answer: I could not follow H i m unless
I gave up all I had (p. 3 23 ) .
Eventually, and part l y t hr ough his st udy of English l aw ( n o t a b l y
Snell' s discussions of t he maxi ms of equi t y), Gandhi came t o u n d e r -
stand t he deeper teaching of nonpossession to mean t hat t h os e wh o
desired salvation "shoul d act like t he trustee, who, t h ough h a v i n g
control over great possessions, regards not an iota of t h e m a s his
o wn " (p. 3 24). It was thus, t hough by a different rout e, t h a t Ga n d h i
came to t he same conclusion as t he Cat hol i c Chur ch i n its c on s i de r -
ation of t he probl em of Franci scan pover t y: a j ur i di cal di s t i n c t i on
was made between dominium (possession) and usus ( t r us t e e s hi p) .
Gandhi , t rue to his new conviction, allowed his i nsur ance p o l i c y t o
lapse, since he became certain t hat " Go d , who creat ed my wi f e a n d
children as well as myself, woul d t ake care of t h e m" ( p. 3 2 4 ) .
Christian Leaders
I n t he Christian tradition, too, t her e have been i n n u me r a b l e
founders of religious orders and sects who came from t h e u p p e r hal f
of t he social cone, yet preached t he style of life-crisis l i mi nal i t y as t he
pat h of salvation. As a mi ni mal list, one mi ght cite Sai nt s Be n e di c t ,
Francis, Dominic, Clare, and Teresa of Avila in t he Ca t hol i c s p h e r e ;
and t he Wesleys, wi t h their " pl ai n living and hi gh t h i n k i n g , "
George Fox, founder of t he Quakers, and (to quot e a n Ame r i c a n
example) Alexander Campbel l , l eader of t he Disciples of Chr i s t ,
who sought to restore pri mi t i ve Chri st i ani t y and e s pe ci al l y t he
primitive conditions of Chri st i an fellowship, in t he P r o t e s t a n t
sphere. These Protestant leaders came from solid mi ddl e cl as s back-
grounds, yet sought to develop i n t hei r followers a s i mpl e, un os t e n -
Humility and Hierarchy
1 99
tatious life-style without distinctions of worldly status. That their
movements subsequently succumbed to " t h e wor l d"and, indeed,
as Weber shows, throve in itin no way impugns their pristine
intents. In fact, as we have seen, the regular course of such move-
ments is to reduce communitas from a state to a phase between in-
cumbencies of positions in an ever developing structure.
Tolstoy
Gandhi was strongly influenced, not only by aspects of Hinduism,
but also by t he words and work of the great Christian anarchist and
novelist Leo Tolstoy. The Kingdom of God Is Wilkin You, wrote
Gandhi (1948), "overwhel med me and left an abiding impression
on me " (p. 1 7 2) . Tolstoy, who was a wealthy nobleman as well as a
famous novelist, went t hrough a religious crisis when he was about
50 years old, in the course of which he even contemplated suicide as
an escape from the meaninglessness and superficiality of life among
the upper class and intellectuals and esthetes. It came to him then
that " i n order to understand life I must understand not an excep-
tional life such as ours who are parasites on life, but the life of the
simple labouring folkthose who make lifeand the meaning which
they at t ri but e to it. The simplest labouring people around me were
the Russian people, and I t urned to t hem and the meani ng of life
which they give. That meaning, if one can put it into words, was as
follows: Every man has come into this world by the will of God.
And God has so made man that every man can destroy his soul or
save it. The aim of man in life is to save his soul, and to save his soul
he must live ' godl y' and to live ' godl y' he must renounce all the
pleasures of life, must labour, humbl e himself, suffer, and be
merciful" (1940, p. 67 ) . As most people know, Tolstoy made
strenuous efforts to replicate his beliefs in his life, and lived in
peasant fashion until his life's end.
2 0 0
The Ritual Process
SOME PROBLEMS
OF ELEVATION AND REVERSAL
Enough has been said to underl i ne, on t he one hand, t he affinity
bet ween t he liminality of rituals of status elevation and t he religious
teachings of structurally superior prophet s, saints, and teachers, and,
on t he other, t he affinity between t he liminality of calendrical or
nat ur al crisis rituals of status reversal and t he religious beliefs and
practices of movements domi nat ed by structural inferiors. Crudely
put , t he liminality of t he strong is weaknessof t he weak, strength.
Or again, t he liminality of weal t h and nobility is poverty andpauper -
ismof poverty, ostentation and pseudohierarchy. Clearly, t here
are many problems here. Why is it, for instance, t hat in t he intervals
between occupying their culturally defined socioeconomic positions
and statuses, men, women, and children should in some cases be
enjoined and in others choose to act and feel in ways opposite to or
different from their st andardi zed modes of behavi or ? Do they under-
go all these penances and reversals merely out of bor edom as a
colorful change from daily routines, or in response to resurgent
repressed sexual or aggressive drives, or to satisfy certain cognitive
needs for bi nary discrimination, or for some other set of reasons ?
Like all rituals, those of humility and those of hi erarchy are im-
mensely complex and resonate on many dimensions. Perhaps, how-
ever, one i mport ant clue to their underst andi ng is t he distinction
made earlier between t he two modalities of social interrelatedness
known as communi t as and st ruct ure. Those who feel t he burdens
of office, who have by bi rt h or achievement come to occupy control
positions in structure, may well feel t hat rituals and religious beliefs
t hat stress t he stripping or dissolution of structural ties and obliga-
tions offer what many historical religions call "r el eas e. " I t may
well be t hat such release is compensated for by ordeals, penances,
and other hardships. But, nevertheless, such physical burdens may
well be preferable to t he ment al burdens of giving and receiving com-
mands and acting always in t he masks of role and status. On t he other
Humility and Hierarchy 2 0 1
hand, such liminality may also, when it appears in rites de passage,
humbl e the neophyte precisely because he is to be structurally
exalted at the end of t he rites. Ordeals and penances, therefore, may
subserve antithetical functions, on t he one hand punishing t he
neophyte for rejoicing in liminal freedom, and, on the other, temper-
ing hi m for the i ncumbency of still higher office, with its greater
privileges as well as more exacting obligations. Such ambiguity
need not by now surprise us, for it is a property of all centrally
liminal processes and institutions. But, while the structurally well-
endowed seek release, structural underlings may well seek, in their
liminality, deeper involvement in a structure t hat , t hough fantastic
and simulacral only, nevertheless enables t hem to experience for a
legitimated while a different kind of "r el eas e" from a different
kind of lot. Now they can lord it, and "s t r ut and stare and a' t hat , "
and very frequently t he targets of their blows and abuse are the
very persons whom they must normally defer to and obey.
Both these types of rituals reinforce structure. In t he first, the
system of social positions is not challenged. The gaps between the
positions, t he interstices, are necessary to the structure. If there were
no intervals, there would be no structure, and it is precisely the gaps
t hat are reaffirmed in this kind of liminality. The structure of the
whole equat i on depends on its negative as well as its positive signs.
Thus, humility reinforces a just pri de in position, poverty affirms
wealth, and penance sustains virility and health. We have seen how,
on t he other hand, status reversal does not mean " a n o mi e " but
simply a new perspective from which to observe structure. Its topsy-
turviness may even give a humorous war mt h to this ritual viewpoint.
If t he liminality of life-crisis rites may be, perhaps audaciously,
compared to tragedyfor both imply humbl i ng, stripping, and
pai nt he liminality of status reversal may be compared to comedy,
for bot h involve mockery and inversion, but not destruction, of
structural rules and overzealous adherents to t hem. Again, we mi ght
regard t he psychopathology of these ritual types as involving i n the
first case a masochistic set of attitudes for t he neophytes, and, in t he
second, a sadistic component .
2 0 2
The Ritual Process
As regards t he relationship of communi t as, t her e a r e t hos e who,
in t he exercise of daily aut hori t y or as r epr es ent at i ves of major
structural groupings, have little oppor t uni t y to deal wi t h t he i r fellow
men as concrete individuals and equals. Perhaps, i n t h e liminality
of life crises and status changes, t hey mi ght find a n oppor t un i t y to
strip themselves of all out war d tokens and i nwar d s e nt i me nt s of
status distinction and merge wi t h t he masses, or even t o b e symbolic-
ally at least regarded as t he servants of t he masses. As f or t hose who
are normally at t he bot t om of t he pecki ng or der a n d experience
t he comradeship and equality of j oi nt subor di nat es, t h e liminality
of status reversal mi ght provi de an oppor t uni t y t o e s c a p e from the
communi t as of necessity (which is therefore i n a ut h e n t i c ) into a
pseudost ruct ure where all behavi oral ext r avagances a r e possible.
Yet, in a curious way, these bluff communi t as - be ar e r s are able
t hrough jest and mockery to infuse communi t as t h r o u g h o u t the
whole society. For here too t here is not only r ever sal b u t leveling,
since t he i ncumbent of each status wi t h a n excess of r i gh t s is bullied
by one wi t h a deficiency of rights. Wh a t is left is a k i n d of social
average, or something like t he neut r al position i n a g e a r box, from
whi ch it is possible to proceed i n different di rect i ons a n d a t different
speeds in a new bout of movement .
Both types of rites we have been considering seem t o b e bound up
with cyclical repetitive systems of mul t i pl ex social r el at i ons . Here
t here appears to be an i nt i mat e bond of r el at i ons hi p bet ween an
institutionalized and only slowly changi ng st r uct ur e a n d a part i cul ar
mode of communi t as whi ch tends to be localized i n t h a t part i cul ar
kind of st ruct ure. Undoubt edl y, in large-scale c ompl e x societies,
with a high degree of specialization and division of l a bor , and with
many single-interest, associational ties and a gener al weakeni ng of
close corporate bonds, t he situation is likely t o b e ve r y different.
I n an effort to experience communi t as, i ndi vi dual s wi l l seek mem-
bership of woul d-be universal ideological move me nt s , whos e motto
mi ght well be Tom Pai ne' s " t h e worl d is my vi l l age . " Or , they will
j oi n small-scale " wi t h dr a wa l " groups, like t he h i ppi e and digger
communities of San Francisco and Ne w York, wh e r e " t h e village
Humility and Hierarchy
203
[Greenwich or otherwise] is my worl d. " The difficulty t hat these
groups have so far failed to resolve is t hat tribal communitas is t he
complement and obverse of tribal structure, and, unlike the New
World Utopians of the eighteenth and nineteenth centuries, they
have not yet developed a structure capable of mai nt ai ni ng social and
economic order over long periods of t i me. The very flexibility and
mobility of social relations in modern industrial societies, however,
may provi de better conditions for t he emergence of existential com-
munitas, even if only in countless and transient encounters, t han
any previous forms of social order. Perhaps this was what Wal t
Whi t man meant when he wrot e:
One's-self I sing, a simple separate person,
Yet utter the word Democratic, the word En- Masse.
One final comment : Society (societas) seems to be a process rat her
t han a t hi nga dialectical process with successive phases of structure
and communi t as. Ther e would seem to beif one can use such a
controversial t erma human " n e e d " to participate in bot h modali-
ties. Persons starved of one in their functional day-to-day activities
seek it in ritual liminality. The structurally inferior aspire to symbolic
structural superiority in ri t ual ; the structurally superior aspire to
symbolic communi t as and undergo penance to achieve it.
Bibliography
APTHORPE, RAYMOND.
I 96 I. Introduction to C. M. N. White, Elements in Luvale beliefs and
rituals. Manchester: ManchesterUniversity Press. Rhodes-Livingstone
Paper no. 3 2.
ATTWATER, DONALD (ED.).
1 9 61 . A Catholic dictionary New York: Macmillan.
BAUMANN, H., and D. WESTERMANN.
1 9 48. Lespeuples el les civilizations de I'Afrique. Pari s: Payot .
BOEHMER, HANS.
1904. Analecten zur Geschichte des Franciscus von Assisi. Leipzig:
Tubingen.
BOSMAN, WILLEM.
1 7 0 5 . Coast of Guiana. London.
BUBER, MARTIN.
1 9 58. land Thou. (Trans, by R. G. Smith). Edi nbur gh: Clark.
1 9 61 . Between man and man. (Trans, by R. G. Smi t h) . London and
Glasgow: Fontana Library.
1966. Paths in Utopia. (Trans, by R. F. C. Hul l . ) Boston: Beacon
Press.
COHN, NORMAN.
ig6i. The pursuit of the millennium. New York: Ha r pe r Torch Books.
204
Bibliography
205
DE, SUSHIL JUMAR.
i g 6i . The early history of the Vaisnava faith and movement in Bengal.
Calcutta: General Printers and Publishers.
DEARDORFF, MERLE H .
1 9 5 1 . Handsome Lake. In Wi lli am N. Fenton (Ed.), Symposium of local
diversity in Iroquois. Washington, D. C. : U.S. Government Printing
Office.
DIMOCK, EDWARD C. , JR.
i g66a. The place of the hidden moon. Chi cago: University of Chi cago
Press.
1 9 66b. Doctrine and practice among the Vai snavas of Bengal. In
Mi lton Singer (Ed.), Krishna: myths, rites, and attitudes. Honolulu:
East- West Center Press.
DOUGLAS, MARY.
1966. Purity and danger. London: Routledge and Kegan Paul.
Du CHAILLU, PAUL B.
1 868. Explorations and adventures in Equatorial Africa. New York:
Harper.
ELWIN, VERRIER.
1 9 5 5 . The religion of an Indian tribe. London: Oxford University Press.
EVANS-PRITCHARD, EDWARD E.
1 9 56. Nuer religion. Oxford: Clarendon Press.
1 9 65a. The position of women in primitive society. London: Faber and
Faber.
1 9 65b. Theories of primitive religion. Oxford: Clarendon Press.
FENTON, WILLIAM N.
1 9 41 . Tonawanda longhouse ceremonies: ninety years after Lewis Henry
Morgan. Bureau of Ameri can Ethnology, Bulletin 1 28, Anthropology
Paper No. 1 5 .
FIRTH, RAYMOND.
1 9 5 1 . Elements of social organization. London: Watts.
FORTES, MEYER.
1949. The web of kinship among the Tallensi. London: Oxford University
Press.
1 9 50. Kinship and marriage among the Ashanti- In A. R. RadclifFe-Brown
and C. D. Forde (Eds.), African systems of kinship and marriage
London: Oxford University Press.
1 9 62. Ri tual and office. In Max Gluckman (Ed.), Essays on the ritual
of social relations. Manchester: Manchester University Press.
GANDHI. MOHANDAS K.
1 9 48. Gandhi's autobiography: the story of my experiments with truth.
(Trans, by Mahadev Desai.) Washington, D. C. : Public Affairs
Office.
206
The Ritual Process
GENNEP, ARNOLD VAN.
1909. The rites of passage. (Trans, by Moni ka B. Vi zedom and
Gabrielle L . Caffee.) London: Routledge and Ke g a n Paul.
GLUCKMAN, MAX.
1 9 54. Rituals of rebellion in South-East Africa. Manchester: Manchester
University Press.
1 9 5 5 . Custom and conflict in Africa. Oxford: Bl ackwel l .
1 9 65. Politics, law and ritual in tribal society. Chi cag o: Al di ne Publishing
Company.
GOFFMAN, ERVING.
1 9 62. Asylums. Chi cago: Aldi ne Publishing Company.
GOULD, J., and W. L . KOLB (Eds.).
1964. A Dictionary of the social sciences. London: Tav i stock.
HILLERY, G. A.
1 9 5 5 . Definitions of communi ty: areas of agreement. Rural Sociology,
vol. 20.
JUNOD, HENRI A.
1 9 62. The life of a South African tribe. New H yde Park, N. Y . : University
Books. 2 vols.
KE3 SER, R. LINCOLN.
1966. Autobi ography of the Vi ce Lord, " Te ddy . " Unpublished
manuscript, Department of Anthropology, Uni versi ty of Rochester.
KRIGE, EILEEN.
1968. Nomkubulwana ceremonies of the Zul u. Africa, vol. 3 8, No. 2.
LAMBERT, MALCOLM D.
1 9 61 . Franciscan poverty. London: Allenson.
LAWRENCE, PETER.
1964. Road belong Cargo. Manchester: Manchester Uni versi ty Press.
LEVI-STRAUSS, CLAUDE.
1964. Le cru et le cuii. Pari s: Plon.
1 9 67 . The savage mind. Chi cago: University of Chi cag o Press.
LEWIS, IOWAN M.
1 9 63 . Dualism in Somali notions of power. Journal of the Royal
Anthropological Institute, vol. 9 3 , Part 1 .
LITTLE, KENNETH.
1 9 65. The political function of the Poro. Africa, vol . 25 , No. 4.
MACCULLOCH, JOHN A.
1 9 48. The Celtic and Scandinavian religions. London: Hutchinson's
University Li brary.
MAIR, LUCY.
1960. The social sciences in Afri ca south of the Sahara: the British
contribution. Human Organization, vol. 1 9 , no. 3 .
Bibliography
207
MARRIOTT, MCKIM.
1966. The feast of love. In Milton Singer (Ed.), Krishna: myths, rites
and attitudes. Honolulu: East-West Center Press.
MORGAN, LEWIS HENRY.
1 8 7 7 . Ancient society. Chicago: Charles H. Kerr.
RATTRAY, R. S.
1 9 23 . Ashanti. Oxford: Clarendon Press.
1 9 27 . Religion and art in Ashanti. Oxford: Clarendon Press.
RESEK, CARL.
. 1960. Lewis Henry Morgan, American scholar. Chicago: University of
Chicago Press.
RICHARDS, AUDREY I.
1 9 56. Chisungu. London: Faber and Faber.
RIGBY, PETER
1968. Some Gogo rituals of purification: an essay on social and moral
categories. In E. R. Leach (Ed.), Dialectic in practical religion. Cam-
bridge : Cambridge University Press.
ROSCOE, JOHN.
1924. The Bagesu and other tribes of the Uganda protectorate. Cambridge:
Cambridge University Press.
SABATIER, PAUL.
1905. The life of St. Francis. (Trans, by L. S. Houghton.) New York:
Scribner's.
SCHLEGEL, GUSTAAE.
1866. The Hung league. Batavia: Lange.
SINGER, MILTON (Ed.).
1966. Krishna: myths, rites and attitudes. Honolulu: East-West Center
Press.
SPEIRS, ALLAN C, JR.
i g66. Village in the desert: the Aaronite community of Eskdale.
Unpublished B.A. thesis, University of Utah.
SUNDKLER, BENGT.
1 9 61 . Bantu prophets in South Africa. London: Oxford University Press.
THOMPSON, HUNTER S.
1966. Hell's angels. New York: Ballantine.
TOLSTOY, LEO.
1940. A confession: the Gospel in brief, and what I believe. (Trans, by
Aylmer Maude.) London: Oxford University Press.
TURNER, TERENCE.
n.d. The fire of the jaguar. Chicago: University of Chicago Press. In
press.
208
The Ritual Process
TURNER, VICTOR W.
1 9 5 7 . Schism and continuity in an African society. Manchest er: Manchester
University Press, for the Rhodes-Livingstone Inst i t ut e.
1 9 61 . Ndembu divination: its symbolism and techniques. Manchester:
Manchester University Press. Rhodes-Livingstone Pa pe r no. 3 1 .
1 9 62. Chihamba, the white spirit. Manchester: Manchest er University
Press, Rhodes-Livingstone Paper no. 3 3 .
1 9 67 . The forest of symbols. Ithaca, N.Y.: Cornell University Press.
1 9 68. The drums of affliction. Oxford: Clarendon Press.
WARNER, LLOYD.
1 9 59 . The living and the dead. New Haven, Conn. : Yal e University
Press.
WEINSTOCK, STEVEN.
i g68. The vagabond and his image in American society. Unpublished
paper delivered at the Society for the Humanities Semi nar, Cornell
University.
WILSON, GODFREY, and MONICA WILSON.
1 9 3 9 . The study of African society. Manchester: Manchest er University
Press. Rhodes-Livingstone Paper no. 2.
WILSON, MONICA.
1 9 54. Nyakyusa ritual and symbolism. American Anthropologist, vol. 56,
no. 2.
1 9 5 7 . Rituals of kinship among the Nyakyusa. London: Oxford University
Press.
Index
All Saints, I 8 I
All Souls, 1 8 1 , 1 8 2
Angelo da Clareno, 1 5 1
Apthorpe, Raymond, 5, 204
Aquinas, Thomas, 1 8 2
Ashanti, 46
Apo ceremony, 1 7 8 - 1 8 1
community in kinship based societies,
1 2 0 - 1 2 5
Attwater, Donald, 1 0 7 , 1 8 1 , 20 4
Bachofen, Johann J . , 2
Bantu societies, 1 1 , 48, 1 8 3
Barnard, Helen, 84n.
Bauls musicians, 1 64
Baumann, H., 46
and Westermann, D., 46, 204
Beidelman, T. O., 40
Bemba; See Rites of the Bemba
Bengal
the Sahajiya movement of, 1 5 4 - 1 5 5
Vaisnavas of, 1 5 5
See also Francis and Sahajiya
Bergson, Henri, 1 1 0 , 1 28 , 1 3 2
Blake, William, 1 3 2, 1 4 1 , 1 8 8
Boas, Franz, 3
Boehmer, Hans, 1 44, 1 45 , 1 47 , 20 4
Bosman, William, 178
Boys' circumcision rites, 1 6, 1 7 , 25 , 3 5 ,
4> > 5 2. 65 , 9
6
. i 7
and St. Benedict, 1 0 8
Buber, Martin, 1 3 2, 20 4
on community, 1 26, 1 27 , 1 3 6, 1 3 7 ,
1 42, 1 43
Buddha, as a leader, 1 9 6- 1 9 7
Buddhism; See Zen Buddhism
Caitanya, 1 5 5 - 1 6 5 , 1 9 7
See also Francis
Campbell, Alexander, 1 9 8
Cargo cults, Melanesian, 1 9 1 - 1 9 2 , 1 9 5
Central Africa, population, 9
See also Ritual studies in Central
Africa
Chokwe people, 4, 1 4
Circumcision; See Boys' circumcision
rites
Chekhov, Anton, 1 1 0
Chief Ikelenge, 7, 8
Christian leaders, 1 9 8 - 1 9 9
Clare; See Christian leaders
Cohn, Norman, i n, 204
Colson, Elizabeth, 5
209
2 1 0
Index
" Communitas," (community), 9 6, 9 7 ,
1 0 5 , 1 0 9 ff.
apocalyptic, 1 5 3 - 1 5 4
developmental cycle, 9 7
and " hippies," 1 1 2 , 1 3 8
ideological and spontaneous, 1 3 4, 1 40
in kinship based societies; See
Tallensi; Nuer; Ashanti
and liminality, 9 5
liminality and low status, 1 2 5 - 1 3 0
modalities of, 1 3 1 - 1 3 3
musicians in; See Bauls
and the sexes, 1 8 3 - 1 8 5
and structure in rituals of status
reversal, 1 7 7 - 1 7 8
and symbolic thought, 1 4 1 - 1 4 5
See also Bengal; Martin Buber;
Franciscan, conceptualization
and structure; Franciscan
poverty; Franciscan, rights over
and consumption of property
Congo; See Suku of Congo
Cults; See Women's cults
Dante, A. , 1 8 2
Deardorff, Merle H., 1 , 2, ' 20 5
De Rougement, D., 1 5 7
De, Sushil Jumar, 1 5 7 , 20 5
Developmental cycle; See " Com-
munitas," developmental cycle
Dieterlen, Germaine, 3
Dimock, Edward C , J r. , 1 5 5 , 1 5 6, 1 5 7 ,
1 5 8 , 1 6 1 , 1 64, 20 5
Dominic; See Christian leaders
Dostoevsky, Fyodor, 1 1 0
Douglas, Mary, 1 0 9 , 20 5
Du Chaillu, Paul B., 1 7 0 , 20 5
Durkheim, Emil, 3 , 5 3 , 1 3 2
Dylan, Bob; See Bauls musicians
Eister, A. W., 1 25
Eliade, M. , 3 8
Elwin, Verrier, 1 1 6 , 20 5
Evans-Pritchard, Edward E. , 2, 46,
9 2, 1 1 9 , 20 5
"Feast of Love, The," 1 8 5 - 1 8 8
Fenton, William, 2, 20 5
Fernandez, James, 1 0 3
Firth, Raymond, 1 2 6 , 20 5
Fortes, Meyer, 9 7 , 9 9 , 1 1 4 , 1 1 5 , 1 1 6 ,
1 1 7 , 1 1 8 , 1 2 0 , 1 2 1 , 20 5
Fowl, meaning of white and red; See
Rites with fowl
Fox, George, 1 9 8
Francis, 1 4 0 - 1 4 5 , 1 5 5 , 1 9 6, 1 9 8
and Cai tanya, 1 5 5 - 1 6 0
and permanent liminality, 1 4 5 - 1 4 7
and Sahaji ya, 1 6 1 - 1 6 2
Franoiscan, 1 3 3
conceptualization and structure,
1 4 7 - 1 5 0
poverty, 1 9 8
poverty and " communitas," 1 40 - 1 41
right over and consumption of
property, 1 5 0 - 1 5 3
See also " Communi tas" and symbolic
thought
Frazer, Sir J ames George, 3
Freud, Si gmund, 3 , 1 3 0
Freud, Anna, 1 7 4
Gandhi, Mohandas K. , 1 3 3 , i g6, 1 9 9 ,
20 5
as leader, 1 9 7 1 9 8
Gennep, Arnold van, 3 , 1 4, 48 , 9 4,
1 66, 20 6
Ghana; See Tallensi ; Ashanti
Ginsberg, Allen, 1 1 3
Girls' puberty ri tes, 7, 2 1 , 41 , 46, 5 2,
63 , 6 5 , 6 6
Gluckman, Max, 5 , 7 8 , 8 1 , 1 0 9 , 1 8 3 ,
1 8 4, 20 6
Goffman, Ervi ng, 1 0 8 , 1 28 , 1 69 , 206
Golib Festival, 9 9 , 1 1 6
Gould, J . , and W. L. Kolb, 1 25 , 206
Griaule, M. , 3
Halloween, Ameri can, 1 7 2 - 1 7 4
Hegel, G., 8 3
Herz, R., 3 , 40
Hierarchies, pseudo
modern examples of reversal and,
1 9 2 - 1 9 4
and secret societies, 1 9 0 , 1 9 1
Hillery, G. A. , 1 2 6 , 20 6
"Hippies"; See "Communi tas"
Hobbes, Thomas, 1 3 1
Hockett, Charles, 2
Index
2 11
Homans, G., 155
Hubert, H., 3
Humbu
Kafwana, headman of, 98- 101
See also Mbwela people
Hume, David, 1 1 1
Humility; See Religion of humility
Hunters' cults, 8, 35
Installation rite, 97 ff.
"The Reviling of the Chief-Elect,"
100 ff.
Iroquois, 2, 4
Isoma, 9- 43, 96
aims of, 18- 20
classificatory structure: Dyads, 38
classificatory structure: Triads, 3 7 -
38
curative process, 3 3 - 3 7
medicines, collecting of, 24- 27
the name, 1 5 - 1 6
planes of classification, 41 - 42
preparation of site, 20
processual form, 1 3 - 1 4
reasons for performing, 1 1 - 1 3
ritual symbolism, 42- 43
situation and classification, 41
symbols of, 1 4- 1 5
See also Twinship rite
John of Parma, 151
Jung, Carl, 163
Junod, Henry, 169, 184, 206
Kafwana; See Humbu
Kalahari, Bushmen of, 46
Kanongesha, senior chief, g8, 99, 100,
102
and medicines of witchcraft, 98
Katanga; See Lunda of Katanga
Keiser, R. Lincoln, 193, 194, 206
Kolb, W. L. , See Gould, J .
Krige, Eileen, 184, 206
Krishna, 1 5 5 - 1 65
Kumukindyila; See Installation rite
Lamba, 5, 109
Lambert, M. D., 206
on Franciscans, 1 41 , 144, 1 45, 146,
147, 148, 149, 150, 152
Lawrence, Peter, 1 9 1 , 206
Leach, Edmund, 164
Levi-Strauss, Claude, 3, 20, 3 1 , 42, 69,
106, 126, 127, 1 3 1 , 1 3 3 , 1 53, 1 7 3.
206
L^vy-Bruhl, Lucien, 3
Lewis, Iowan M., 99, 206
Little, Kenneth, 206
Livingstone, David, 9
Lowie, Robert, 3
Luchazi, 4, 14
Lunda of Katanga, 4, g8, 99, 100
Luvale, 4, 14
MacCulloch, John A., 182, 206
Mair, Lucy, 5, 206
Malinowski, Bronislaw, 3
Marriott, McKi m, 185, 188, 207
Marx, Karl, 83
Mauss, M. , 3
Mbwela people, 98, 99
Medicines, 41
collecting of, 53 ff.; See also Isoma;
Wubwang'u
hot and cold (life and death), 27 - 3 1
Mitchell, Clyde, 5
Morgan, Lewis Henry, 130, 207
and religion, 1 - 4
Lectures for 1966, 1
Muchona (informant), 69
Mukanda; See Boys' circumcision rites
Nadel, S. F., 9
Namoos people, gg
Needham, J . , 40
Ndembu
divorce, 12
forms of duality among, 91
marriage, 1 2, 82
as matrilineal society, 8
obscenity, 92
religious characteristics, compared to
Christianity and other religions,
107
residential filiation (village), 21
woman's role of, 11
See also "Communitas"; Rituals; Rites
Nietzsche, F., 1 1 0
Nkang'a; See Girls' puberty rites
Nkula rite, 7, 18
2 1 2
Index
Nuer, 1 3 0
community in kinship based societies,
1 1 9 - 1 2 0
of the Nilotic Sudan, 46, 47
Nyakyusa people, 6
rites, 48
Oedipus complex, i 63
Olier, M. , 1 8 1
Olivi, 1 5 1 , 1 5 3
Paine, Thomas, 202
Parker, Ely S., 2
Prince Philip, 9 7
Pope Gregory IX, 1 5 1
Pope John XXI I , 1 5 3
Radcliffe-Brown, A. R., 3 , 1 8 5
Rattray, R. S., 46, 1 20 , 1 2 1 , 1 22, 1 23 ,
1 24, 1 7 8 , 1 7 9 , 20 7
Reith Lectures, 1 64
Religions, 1 1
of humility with high status founders,
1 9 5
of humility and status reversal,
1 8 8 - 1 8 9
of status reversal and separatism, 1 8 9
See also Rituals; Rites
Rhodes-Livingstone
Institute for Social Research, 5
Paper, 9
Richards, Audrey, 1 0 3 , 20 7
Rigby, Peter, 6, 40 , 1 8 4, 20 7
Rilke, R. M. , 1 3 9
Rites
of the Bemba, 1 0 3
divining, initiation into, 3 5
with fowl, 3 1 - 3 3
funerary initiation, 3 5
hunters'; See Hunters' cults
ilembi, 21
installation; See Installation rite
life crisis and calendrical, 1 68 , 1 7 0
masks and myths, 1 7 2 - 1 7 7
" of separation," 21 ff.
See also Rites of the River Source;
Boys' circumcision rites; Girls'
puberty rites
Rites of the River Source, the, 5 3 - 60 ,
7 1 , 8 4
the stream and the arch, 63 ff.
Ritual
of status elevation, 1 66- 1 68 , 1 7 0 - 1 7 2
of status reversal, 1 7 7 - 1 7 8 , 1 8 3 - 1 8 5 ;
See also "Feast of Love";
Religions of humility
studies of, in Central Africa, 4- 6
symbols of, 5 2
women's 1 1
See also Ndembu; Twinship
Robertson-Smith, William, 3
Roscoe, John, 48 , 20 7
Rousseau, J . J . , 1 3 6
Sabatier, Paul, 1 43 , 20 7
St. Benedict, 1 0 7 , 1 9 8
monks of, 1 0 8
See also Boys' circumcision rite
St. Bonaventura, 1 5 1
Samhain, 1 8 1 - 1 8 3
Schapera, I., 45
Schneider, D., 1 1 5
Secret societies; See Hierarchies and
secret societies
Seneca Indian, 2
Separatism, South African; See Religion
of status reversal
Sexes
contest of, 7 5 ft, 84
equality between, 81
See also " Communitas "
Shakespeare, William, 1 3 4, 1 3 6, 1 40
Schlegel, Gustaaf, 1 9 4, 20 7
Singer, Milton, 20 7
Soja; See Uganda
Speirs, Allan C, 1 9 2, 20 7
Spencer, Herbert, 3 , 1 25
Suku of the Congo, 48
Sundkler, Bengt, 1 8 9 , 20 7
Symbolism, ritual; See Isotna, ritual
symbolism; Twinship rite
Symbols; See Ritual; Isoma, symbols
Tale, 9 9
land and community, 1 1 9
Tallensi, 1 28 , 1 3 0
community in kinship based societies,
1 1 3 - 1 1 8
of northern Ghana, gg
Tanzania, Gogo of, 1 8 4
Tekiman people, 1 7 8 , 1 8 0
Index
213
Teresa of Avila; See Christian leaders
Thompson, Hunter S., i g3, 194, 207
Tolstoy, Leo, 1 3 3 , 196, 207
his influence and beliefs, 199
Tongo, chief of, gg
Trobriand Islanders, 12
Turner, Terence, 1 7 3 , 207
Turner, Victor, W., 18, 25n., 40, 59,
81 , 85, 94, 1 23, 143, 208
Twain, Mark, 1 1 0
Twinship rite, 7, 44- 93
and isoma, difference, 85
medicine collecting for, 5 1
mystery and absurdity, 84, 85
and Ndembu view of, 86
recognition of duality, 91
ritual symbols, 52
shrine in the village, 69
Tylor, E. , 3
Ubertino, 151
Uganda
Soja of, 48
Gisu of, 1 1 9
Warner, Lloyd, 168, 208
Weber, Max, 3, 6, 198
Weinstock, Steven, 208
Westermann; See Baumann
Whitman, Walt, 203
Wilde, Oscar, 1 1 7
Wilson, Godfrey, 5, 6, 208
Wilson, Monica, 6, g, 1 1 , 48, 102, 208
Witchcraft, ig, 28, 38, 40
anti, 64
See also Kanongesha, medicines of
witchcraft
Woman
and miscarriages, 16, 20
unfruitful, 26
See also "Rites of separation "
Women
cults, 14, 21
rituals; See Isoma
Wubinda rite, 7
Wubwang'u rite; See Twinship rite
Wundt, W., 3
Wuyang'a; See Hunters' cults
Zen Buddhism, 1 1 3
Zulu, 183, 184
land, churches in, 189
Bilkent University
Li brary