(Routledge Revivals) Raymond Firth - Symbols - Public and Private-Routledge (2011)

Routledge Revivals
Symbols
This book first published in 1973 offers a broad survey of the study of
symbolic ideas and behaviour.
The study of symbolism is popular nowadays and anthropologists have

made substantial contributions to it. Raymond Firth has long been
internationally known for his field research in the Solomons and
Malaysia, and for his theoretical work on kinship, economics and reli-
gion. Here from a new angle, he has produced a broad survey of the
study of symbolic ideas and behaviour.
Professor Firth examines definitions of symbol. He traces the history of

scientific inquiry into the symbolism of religious cults, mythology and
dreams back into the eighteenth century. He compares some modern
approaches to symbolism in art, literature and philosophy with those in
social anthropology. He then cites examples in anthropological treat-
ment of symbolic material from cultures of varying sophistication.
Finally he offers dispassionate analyses of symbols used in contemporary
Western situations - from hair-styles to the use and abuse of national
flags; from cults of Black Jesus to the Eucharistic rite. In all this Pro-
fessor Firth combines social and political topicality with a scholarly and
provocative theoretical inquiry.
Symbols
Public and Private
Raymond Firth
§3 j Routledge
^ S m ,/ Taylor & Francis Group
First published in 1973
by George Allen & Unwin Ltd
This edition first published in 2011 by Routledge
2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Simultaneously published in the USA and Canada
by Routledge
711 Third Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 1973 George Allen & Unwin Ltd
All rights reserved. No part of this book may be reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in writing from the
publishers.
Publisher's Note
The publisher has gone to great lengths to ensure the quality of this reprint but
points out that some imperfections in the original copies may be apparent.
Disclaimer
The publisher has made every effort to trace copyright holders and welcomes
correspondence from those they have been unable to contact.
A Library of Congress record exists under ISBN: 0045730113
ISBN 13: 978-0-415-69466-7 (hbk)

ISBN 13: 978-0-203-14546-3 (ebk)
SYMBOLS
by the same author
We, the Tikopia
Social Change in Tikopia
Tikopia Ritual and Belief
Rank and Religion in Tikopia
{George Allen & Unwiri)
Primitive Economics of the New Zealand Maori

{Wellington Government Printer)
Elements of Social Organization

{Tavistock Publications)
Primitive Polynesian Economy

{Routledge)
RAYMOND FIRTH
SYMBOLS
PUBLIC AND
PRIVATE
London
G E O R G E ALLEN & U N W I N L T D
RUSKIN HOUSE MUSEUM STREET
First published in 1973
This book is copyright under the Berne Convention. All rights

are reserved. Apart from any fair dealing for the purpose of
private study, research, criticism or review, as permitted under
the Copyright Act, 1956, no part of this publication may be
reproduced, stored in a retrieval system, or transmitted in any
form or by any means, electronic, electrical, chemical, mech-
anical, optical, photocopying, recording or otherwise, without
the prior permission of the copyright owner. Enquiries should
be addressed to the publishers.
© George Allen & Unwin Ltd 1973
ISBN o 04 573011 3
This edition not for sale in the U.S.A.,

their dependencies or the Philippine Republic.
Printed in Great Britain

in 12 pt Fournier type by
Alden & Mowbray Ltd
at the Alden Press, Oxford
CONTENTS
Introduction page 9
I
1. An Anthropologist's Reflections on Symbolic
Usage 15
2. A Question of Terms: Scope and Meaning of
'Symbol' 54
3. Development of Anthropological Interest in
Symbols 92
4. Crystallization of Problems of Symbol Theory 127
5. Modern Anthropological Views of Symbolic
Processes 165
6. Private Symbols and Public Reactions 207
II
7. Food Symbolism in a Pre-Industrial Society 243
8. Hair as Private Asset and Public Symbol 262
9. Bodily Symbols of Greeting and Parting 299
10. Symbolism of Flags 328
11. Symbolism in Giving and Getting 368
12. Symbol and Substance 403
References 429
Index 457
INTRODUCTION
My anthropological interest in symbolism was aroused more than

forty years ago, among the Tikopia people of Western Polynesia.
Their pagan rituals embodied many symbolic actions and symbolic
statements of a vivid and complex kind, and interesting questions
were presented as they converted from paganism to Christianity.
But my earlier attraction to Romanesque art already bore on the
problem of relation between religious symbolism and structure of
society in a pre-industrial phase.
The aim of this book is to help to give perspective to the an-
thropological study of symbolic forms and processes and the
functions of symbolism. It is meant neither as a textbook nor as a
comprehensive general work. I have written about a range of ideas
and material that seem relevant to me in understanding the problems
of symbolism, and in what I have said of symbolism in art, litera-
ture, and religion I have no specialist knowledge. As well as giving
some review of the present state of knowledge in anthropological
studies of symbolism, I have tried to do three things. To provide
some time-dimension, I have examined some of the ideas on symbol-
ism put forward at the end of the eighteenth century and in the early
nineteenth century by mythologists, dream-philosophers, cult-
analysts - some of whom might be described as proto-anthropolo-
gists - and I have continued the study as anthropological interest
crystallized until the present day. To provide breadth I have brought
into discussion of the nature of symbolism hints of a few selected
contributions from other fields of knowledge, including philosophy,
since I think anthropologists need to be more aware of the depth of
such studies in symbolism. Thirdly, I have deliberately cited exam-
ples of symbolic behaviour and statements about symbols, taken
from newspapers and other ephemeral sources, because to my mind
9
10 INTRODUCTION
they show something of the richness of material open for investiga-

tion in modern industrial society.
In all this, while not, as I have once been labelled, an 'unqualified
empiricist', I have been very aware of the problem of evidence. I
think speculative reasoning in anthropology is stimulating and
necessary for the development of theory, but it is easy in the study
of symbolism to let it pass for fact. In anthropology, our imagina-
tion will have to be welded to honest craftsmanship if in the long
run it is to carry conviction.
In 1967, as the guest of Victoria University, New Zealand, I de-

livered the Chancellor's Lectures, beginning with 'Giving and
Getting', in which I put forward some of the ideas in this book.
When I was asked to become Sara H. SchafEher Lecturer at the
University of Chicago for the Fall Quarter of 1970,1 was stimulated
to expand those themes in the direction of specific studies of symbol-
ism, which seemed appropriate in view of the very lively interest in
symbolism in the Department of Anthropology there. By the
generous terms of the Lectureship, which was instituted by Mr
Joseph Halle Schaffner in honour of his mother, the University was
asked to give the lecturer such hospitality as would provide him with
a free informal atmosphere in which to work and mingle with all
members of all grades of the academic community. I am very grate-
ful to the University authorities, and especially to Bernard S. Cohn,
Chairman of the Anthropology Department, and my other colleagues
there, for all the help and hospitality they gave me then and later.
I have since added to the Schaffner Lectures, and have presented
parts of this book at various stages to audiences at Cornell, Case
Western Reserve, Columbia, McGill, Princeton, the London School
of Economics, the University of Hawaii, the University of British
Columbia, the University of Toronto, and the Graduate Center of
the City University of New York. I mention in particular also that a
version of material on the symbolism of exchange was given as the
Dr David B. Stout Memorial Lecture before the Undergraduate
Anthropology Club of the State University of New York at Buffalo,
in April 1970. From discussions arising from these lectures and
seminars I have profited much, and am grateful for the warmth
of my reception on these occasions. A substantial part of the mat-
INTRODUCTION II
erial and general argument of Chapter 9 has appeared in another

form as a contribution to a volume of essays in honour of my
colleague Audrey Richards— The Interpretation of Ritual, edited by
J. S. La Fontaine (London, Tavistock, 1972). I acknowledge with
thanks the agreement of editor and publisher to my reproduction
of this material.
More generally, I am indebted to many colleagues for stimulus
in preparing this work, particularly to Fred Eggan, Mervyn Meg-
gitt, Joan Metge, Jan Pouwer, David Schneider, Terence Turner,
Victor Turner, Nur Yalman, for interest and encouragement at
various stages. Also I should mention especially Barbara Babcock
and Judith Schapiro, whose many helpful thoughts included ref-
erence to Charles Feidelson's book on symbolism in American
literature, and Guy Michaud's study of the French Symbolists.
Finally, clarity of expression in the first part of this work, and the
thought of the whole, have benefited distinctly from talks with my
wife, who blends stimulus with criticism in a way I acknowledge
with gratitude.
London, August 1932 Raymond Firth
I
ERRATA
Page 196, line 9 read Heisenbergian not Heisenburgian

Page 265, line 16 from bottom read 1776, ii, not 1776, iii
Page 266, first line of footnote read Katharine not Katherine
Page 267, second line from bottom read {Letters, book II,
iv — 1637) not {Letters, book ii, iv - 1621)
Page 271, second line of footnote should now read: and was
used by Thomas Blount {Glossographia, 1656) - Hyacynthine,
of Violet or
Symbols
Chapter i
AN ANTHROPOLOGIST'S
REFLECTIONS
ON SYMBOLIC USAGE
Symbolization is a universal human process. But we still need to

understand much more about it, especially in its comparative aspects,
in different societies, different classes, different religions. Pervasive
in communication, grounded in the very use of language, sym-
bolization is part of the living stuff of social relationships. Western
literature is shot through with references which recall to us questions
of existence and identity in symbol terms. In an essay on The Poet,
Emerson wrote of the universality of the symbolic language: 'things
admit of being used as symbols because nature is a symbol' (but so is
culture) - V e are symbols and inhabit symbols'. In Sartor Resartiis
Carlyle held that in a symbol there is both concealment and revela-
tion. Oriental writings show analogous views. What is it in such
statements that some of us find so attractive? Is it truth or illusion
about human personality? And if these are not questions for an-
thropologists to answer, can we at least comment meaningfully
upon the forms of such statements, the conditions of their utterance,
and their social effects?
In intellectual circles, symbolism in literature, art and religion has
long been a subject of study; philosophers and linguists have scrutin-
ized the concept of symbol in its more abstract significance. I show
later why I think such treatment is of interest to anthropologists.
But anthropologists are also concerned with the ways in which
ordinary people think about symbols, behave symbolically in their
daily life as members of a society, and consciously interpret what
they do as having symbolic meaning.
The essence of symbolism lies in the recognition of one thing as
standing for (re-presenting) another, the relation between them
normally being that of concrete to abstract, particular to general.
The relation is such that the symbol by itself appears capable of
15
i6 SYMBOLS
generating and receiving effects otherwise reserved for the object

to which it refers - and such effects are often of high emotional
charge.
POPULAR RECOGNITION OF SYMBOLS

Now, what is very striking about the contemporary social scene is
the wide currency of the notion of symbolism, and the overt,
frequent use of the term symbol to describe objects, persons, actions,
relationships of public interest.* In recent years particularly, the
popular press gives many examples. I pick out a few almost at
random.
Countries are often spoken of as symbolized by their products. So
it has been said that in the early part of this century, peat, potatoes and
parish priests meant Ireland, and that these 'are still valid symbols'.
But a product of one country has been taken as a symbol of some
international organization because of some special quality of that
product. So a Chinese animal, the rare giant panda, has been made a
symbol of the World Wildlife Fund, and its image is embossed on a
plaque pledging furriers not to handle skins of other endangered
species. In the socio-political field, national military bases abroad
have long been £a symbol of world leadership'. Meetings of heads of
state have taken on a quality far exceeding the empirical content of
the occasion. A personal exchange of views of the President of
France and the Prime Minister of Britain about the Common
Market could, it was thought, serve a future historian of Europe as a
'symbolic turning point' in the twentieth century. President
Nixon's journey to Alaska to greet the Emperor of Japan - 4,000
miles for a talk of fifty minutes - in late 1971 was interpreted as 'an
essentially symbolic gesture' which it was hoped might ease some
tension in Japanese-American relations; the meeting was 'of enor-
mous symbolic importance'. The President's visit to the Peoples'
Republic of China was regarded in American quarters as 'an exercise
in symbolic diplomacy'-where the fact of his going to Peking
would be of more diplomatic importance than any substantive
agreements that might be reached while he was there. But since the
* The term has come also into the commercial field as a label, for example a London
biscuit firm (of high respectability as a supplier by appointment to a member of the Royal
Family) is called Symbol Biscuits Limited, with an elephant's head as its mark.
AN ANTHROPOLOGIST'S REFLECTIONS 17
trip was 'primarily symbolic' it was recognized that it might sym-

bolize different things in different Asian capitals, and be interpreted
otherwise in Tokyo from in Peking.
In dialogue between Peking and Washington a game of ping-pong
'has been made a symbol of "people to people" exchanges'. A
light-hearted if malicious act of militant Jewish youths in releasing
frogs and mice in Soviet commercial agencies in New York during
Passover week as part of attempts to obtain freedom of Soviet Jews
to leave for Israel was described as harassment by 'symbolic plagues'.
(This was thought to be parallel to the Biblical account of the efforts
of God to enforce the release of the Israelites by the Egyptians.)
And the recital of Psalm 83 during a Jewish service for their dead
who were killed in the war was described as having 'a symbolic
significance relating to the suppression of Jewish cultural and spiritual
rights in the Soviet Union'. On the other hand, when the Governor
of Illinois sat in his office at midnight with his aides to defy a bomb
threat, he explained that it was 'a symbolic act - we're trying to tell
something to the people who call and threaten and try to intimidate
government.'
Intellectuals may easily agree with the President of Columbia that
the university is 'the symbol and agent of the larger society' and that
as such should be open to criticism, provided that 'savage symbolic
attacks' do not cause it to collapse rather than reform. A more
sardonic note has emerged in a description of the new massive
Lyndon Baines Johnson Library at the University of Texas. 'Archi-
tecture as art and symbol is one of civilization's oldest games,'
and the architect of this monumental memorial has 'made a large
statement using the archives themselves for stunning symbolic and
aesthetic effect'. In development operations the jabbing of a shovel
into the earth 'in a symbolic earth-turning ceremony' by an Indian
dressed in buckskin jacket and eagle feather head-dress marked the
start of a new tribal money-making hotel-type convention centre,
with the ground to be actually levelled by bulldozer. By contrast, a
grand old Mississippi Gulf Coast hotel demolished to make way for a
store and shopping centre was described as 'a symbol of another day'.
Conceptualization of social change takes one form in the view that
people opposed to the legalization of marijuana see the prohibition
'as a symbolic barrier to a culture's disintegration' in the face of
i8 SYMBOLS
almost unmanageable pressures. It takes another form when Ameri-

can feminists argue that the demand that a woman specify her
marital status by the prefix Miss or Mrs is discriminatory and that it
should be replaced by Ms - 'Ms is really just a symbol - a gesture -
but it is important to many women.'
The range of objects and actions cited as symbols - from ping-
pong to a panda, from shovelling earth to prohibiting marijuana -
suggests that these things in themselves are not of primary signifi-
cance and that the key to an understanding of their symbolic
quality lies in circumstances to which they refer or of which they are
a part. It is not their particular nature but their relationships which
account for their selection as symbols. At the same time, one cannot
ignore completely the character of an object used as symbol (for
example hair, blood), because certain classes of objects tend to
stand in certain types of relationship to given situations.
This is perhaps especially the case when the object is regarded as
having some autonomy, some freedom of action of its own, and so
can be credited with intrinsic qualities deemed appropriate to a
situation. So, it is part of modern popular idiom that people can
easily become symbols. A priest is said to have 'a symbolic role of
being a "sign" of the church.' A devoted churchman with a social
conscience - such as the controversial Brazilian archbishop Dom
Helder - was described as being to his friends and admirers a saint,
or 'at the very least a symbolic man' whose courageous statements at
home and abroad called attention to the plight of the masses in his
country. Cardinal Mindszenty of Hungary who remained for
fifteen years in seclusion in protest against the regime was 'a symbol
of inflexible resistance of the cold-war era'. Description of political
leaders as symbols is common. P. C. Lloyd has pointed out in
reference to African societies that modern symbols of national
unity tend to be persons rather than objects or ideas (1967, 316).
The old-guard Nationalists of China, who reject violently any idea
of accommodation with the Communists, are 'symbolized' by
Chiang Kai-Shek; Lon Nol of Cambodia is described as 'a purely
symbolic premier' after a near-fatal stroke which immobilized him;
Sheikh Mujibur Rahman, of East Pakistan (now Bangladesh), was a
'symbol of Bengali resistance'. But the same phenomenon is recog-
nized further down the political ladder. Ernest Bevin, the British
trade-union leader, was 'the symbol of the industrial side of the

Labour movement' and when he was Foreign Secretary 'his roles in
the creation of NATO and in the spurning of the Schumann Plan
for the Coal and Steel Community were symbolic'. When ex-
premier David Ben-Gurion attended the mass Israeli Labour party
convention in 1971, his presence 'symbolized his rapprochement with
the party leadership' after more than a decade of ideological disputes
with it. Charles Evers, black Mayor of Fayette, Mississippi, was to
his friends 'not simply a man or a small-time mayor but a precious
symbol - a hero of the Southern black political renaissance'. Adam
Clayton Powell, first 'negro' from the East to serve in the United
States Congress, was for long 'a symbol of success to blacks'.
Dr Eric Williams, Prime Minister of Trinidad-Tobago and increas-
ingly under fire from the local black-power movement, was 'once a
symbol of the blacks' struggle against British colonialism'. And in
the economic field, a Scottish industrialist who was interested in
purchasing a bankrupt Clydebank shipyard 'emerged from com-
parative obscurity to become the symbol of hope for thousands of
shipyard workers on Clydeside'.
A striking example of the explicit incorporation of the notion of
symbol into the relationship of a person to an abstract entity is that
of the Emperor Hirohito of Japan. Previous to the defeat of Japan at
the end of the Second World War, the Emperor had been regarded
officially as a divine being, a direct descendant of the Sun Goddess.
Deified and worshipped publicly for many years, in an imperial
rescript of 1 January 1946 the Emperor renounced the idea of his
divinity 'based on the fictitious idea that the Emperor is manifest
god'- and became 'the Symbol of the state and of the unity of the
people'. By this act the Emperor translated himself from a sacred to a
secular being, from a god to a human being. In his role as an explicit
symbol of the state he appears to be acceptable to a large majority of
the Japanese people, and he seems also to be able to exercise a
considerable if subtle influence upon political decisions made by the
government. A person as political symbol may have active as well as
passive functions.*
* Note that some recent consideration has been given to re-definition of the Emperor's
position, to make him head of state, more in conformity with diplomatic protocol and
Japanese national pride. (Cf. New York Times, 21 September 1971, 26 September 1971,
31 October 1971.)
20 SYMBOLS
Modern popular views of law and morality are rich in recognition

of men as symbols. It is said that for citizens of the United States with
no monarchy and no established church, the Supreme Court
symbolizes the judiciary, and approaches the level of a sacred
institution. It has also been stated that J. Edgar Hoover, long-time
head of the Federal Bureau of Investigation, became 'symbolic of the
fight against evil-doers'. On the other hand, a former Black Panther
leader accused of assault on policemen and kept in jail by very high
bail was held by members of his local community to be 'a symbolic
victim of excessive bond' equivalent to preventive detention. Like-
wise, an Army officer convicted of the murder of civilian prisoners
in the My Lai massacre case was regarded by many people as 'a
symbol for all that was wrong with the Vietnam war'. In a general
conflict over what are variously regarded as the constraints or the
values of established norms, the leader of the rock group the Rolling
Stones, Mick Jagger, came for many 'to symbolize and personify an
entire rebellious generation with its drug-oriented, freewheeling
morality'. And a recent examination of the possibility of selecting a
black vice-presidential candidate for the United States elections of
1972 stressed very strongly the moral value of such a symbolic act.
It was argued that a black vice-presidential candidate would have 'a
tremendous symbolic impact on race relations in the United States'.
Man does not live by symbols alone, but man orders and interprets his
reality by his symbols, and even reconstructs it. A black moderate
running for Vice-President would be 'an extremely powerful symbol
of the possibility of achieving racial justice and harmony in American
society'. The symbol by itself would not create justice or harmony,
but it might create an atmosphere in which they might become more
possible. Thus a candidate who chose a black running-mate in the
name of reconciliation combined with social progress would be likely
to 'touch deep wellsprings of the American political symbol system'.
This spate of examples,* culled from only a few Press sources over
* Citations concerning Ernest Bevin are from Roy Jenkins, The Times, 9 June 1971; those
concerning the Johnson Library are from Ada Louise Huxtable, New York Times, 23 May
1971; those concerning a black Vice-President are from Andrew M. Greeley, New York Times
Magazine, 19 September 1971. Other material in this set of examples has been drawn from:
Chicago Daily News, 5 April 1971, 12, 15, 20, 26 May 1971; New York Times, 2, 12, 14, 18
April 1971, 23 May 1971, 2, 4 June 1971, 20, 21, 26, 27, 29 September 1971; Newsweek, 12, 26
April 1971, 10, 24 May 1971, 7 June 1971; Time Magazine, 17 May 1971, 20, 27 September
1971; The Times, 27 March 1971, 24 August 1971; Guardian, 24 August 1971.
a very limited period, illustrates the extent to which the notion of

symbol has come to permeate popular thought. Little wonder then
that an American publicist, Stewart Alsop, critical of modern
political party trends, has written about what he calls 'symbol-
think'. Members of the party 'did not have to think about the issues,
or argue about them. Instead, they reacted to symbols.' People who
did not react in the same way to the same symbols were not 'real'
party members and thus deserved to be cast into outer darkness.
Whereas formerly Republican 'symbol-thinkers' in passionate
opposition to the government once exercised power, now they have
been replaced by Democratic 'symbol-thinkers'... and so on
(Newsweek, 26 April 1971). The social awareness indicated by such
an opinion is reminiscent of the classical statement of £mile Durk-
heim: 'it is a well-known law that the sentiments aroused in us by
something spontaneously attach themselves to the symbol which
represents them.' (1926,219). What is remarkable about such modern
popular opinions is not their freedom of symbolic usage but the
open way in which they recognize the usage and call it symbolic.
Yet an overt use of the term symbol may not necessarily convey
clearly the kind of relation envisaged between the thing cited and
whatever it is intended to represent. The idea of a sex symbol seems
clear enough. But when a popular cinema star or singer is referred
to as a sex symbol, as often happens, just what is meant? An epitome
of sexual thoughts and longings, concretely represented in a particu-
lar body which suggests eroticism by subtle-or blatant-com-
binations of form and movement? Yes, but with qualifications:
a sex symbol for one sex only - a man for women, a woman for men;
for some people only - many are unmoved, apart from the complex
question of those who reject the symbolism with anger. And apart
from physical qualities, it seems as if the person so nominated usually
has that indefinable quality of personal attractiveness in spirit as well
as in body, so that the sex symbolism is not just the representation or
suggestion of sheer brute contact and coupling, but implies some
more emotional, even ethereal components as well. Perhaps it was
partly for such reasons that Marilyn Monroe, labelled one of the
greatest sex symbols of her time, is said to have commented that she
thought symbols were 'those things that clashed together'. Beneath
her wit, it may be, lay a sense of how vague such labels of symbol
22 SYMBOLS
really are. Complications in the concept show too in a feminine

protest at discrimination against educated women, by male em-
ployers with 'a secretary sex-symbol syndrome' (Time Maga^ne^
6 December 1971).
A symbol represents something else - but there may be several
levels of meaning involved. President Nixon's journey to Alaska to
meet the Emperor of Japan was called a symbolic gesture because it
was thought that there was no concrete object sought - no arrange-
ments to be concluded, no question of principle to be resolved, no
treaty to be signed, which demanded his presence. What the gesture
of long-distance travel stood for was presumably then a sign of the
amity which existed or was hoped to exist between the countries of
the two leaders. Indeed it was stated that the President's clear purpose
was to demonstrate a sentiment that had tended to be lost to sight
in recent controversy about economic relations between the United
States and Japan, the sentiment that mature alliance with Japan was
an important concern of American policy in Asia. It was also stated
that he intended to indicate a symbolic counterweight to his recent
friendly overtures to the Peoples' Republic of China. Yet it was one
writer's opinion that while both men and both governments were
'plainly caught up in the symbolism of the occasion', 'the symbolism
belonged to the past' (Max Frankel, New York Timesy 28 September
1971).
According to this view, there was considerable resentment among
the officials on both sides behind the polite fa$ade, so the greeting
symbol did not represent an existing state of affairs, but one that was
past, and that it was only hoped would return. This is an illustra-
tion of how a symbol can be detached from present reality and made
to stand for past reality or future contingency. Or putting it another
way, what was being 'said' by the ceremonial behaviour was in
effect: this greeting expresses not what we leaders - or our officials
and people - feel now, but the opposite of our present feelings, or a
minor not a major aspect. The idea of a symbol may then be be-
haviour which, if not devoid of meaning, is at least contradictory of
the actual state of affairs. The connotation of symbol here is super-
ficiality, not deepest reality. Moreover, the Japanese Emperor is by
imperial rescript the symbol of the state. So his 'symbolic stop-off'
for the meeting could be described as the symbolic behaviour of a
person already a symbol - a kind of symbol to the second power.

The idea that 'symbol5 is equivalent to 'not real' is brought out
by another example - though it also raises the question whether in
popular usage the term 'symbol' is often not misapplied. As part of
an attempt to help control the smuggling of antique paintings out of
Italy, the United States customs authorities in 1971 were stated to
have 'symbolically seized' the so-called Boston Raphael, about the
authenticity of which experts had been in conflict (The Times,
6 March 1971; New York Times, n September 1971). What is
meant by the word 'symbolically' in such context? What apparently
happened was that the customs authorities took charge of the paint-
ing and directed it to be removed from the wall of the Boston
Museum of Fine Arts, where it hung on public exhibition, to the
museum store-room. There it remained under customs seal on the
claim that it had presumably been smuggled into the country without
a permit. Where was the symbolism? The customs men probably did
not physically lay hands on the painting, and so did not actually
'seize' it; their act might therefore be described as a metaphorical
seizure, by which they obtained control of the picture without grasp-
ing it. To be properly described as 'symbolic' in conventional terms,
it might be argued that an act of seizure should have been meant to
represent something else, not that some other act represented a
seizure. The expression 'symbolic seizure' may of course have been
purely a newspaper invention. But from a public point of view it
may be that the 'seizure' was truly symbolic in that it was not in-
tended to grab that particular painting. The act may have been
meant to show disapproval by the customs authorities (or other
officials) of the practice which seems to have grown up of some
museums and galleries obtaining works of art from abroad without
going through the formalities of getting the appropriate licences - in
other words, smuggling what may well have been stolen goods.
In another example, the ambiguity of the symbolism is more
evident. When democratic elections were introduced after the war
in some countries where the level of literacy was low, pictorial
emblems were used to represent the different political parties seeking
votes. In 1969 the Indian Congress party split into two rival factions,
and a long legal battle ensued to determine which faction should
have 'the party's old election symbol, the two yoked bullocks'. The
24 SYMBOLS
issue was thought to be very important because it was held that

many voters being illiterate would recognize and vote primarily for
the symbol rather than for the party as such {The Times, 14 January
1971). When it seemed that the Supreme Court would deny this
symbol of yoked bullocks to either faction, Mrs Gandhi's party was
expected to adopt a symbol showing mother and child - but in the
event they used a cow and calf. Some other parties to the election
used analogous devices reminiscent of peasant activity - sickle and
ears of grain; woman at a spinning wheel; an oil lamp (Neville
Maxwell, New Society, 4 March 1971). Drawings of these things
were described as the symbols of the political parties. Now the
drawings were just conventionalized representations of bullocks,
grain, etc., that is signs for them. In saying that these sketch-outline
designs were 'symbols' of the various political parties it is presuma-
bly meant that they were intended to call to the minds of party leaders,
members and voting public generally the nature of each party
as a political entity, the ideals for which it stood, and its practical
programme. Presumably too, the selection of 'naturalistic' symbols
such as bullocks, cow and calf, mother and child, grain, spinning
wheel, simple lamp, was meant to suggest to a significant body of
voters that there was some connection between the specific party and
care for the peasant way of life which these things might represent.
As the legal battle over the yoked bullock device implied, it was
thought that values, including those of prestige, were attached to
these symbols. But what the device actually symbolized to voters
remained somewhat obscure. Did they associate bullocks with the
traditional Congress party so habitually that they would vote for the
bullocks thinking they were still voting for that party? Did they
realize that the bullocks might no longer represent the same party
but still vote for that device on the grounds that people who used
the bullocks as device automatically employed policies of which
voters approved - that the policy went with the label, not with the
party which might change label? Or did the voters not bother about
such issues, but simply vote for the device which they favoured on
much more literal grounds - that they preferred bullocks to a cow
and calf, or they liked the sentimentalism of a device of mother and
child? That this last is not at all out of the question is illustrated by
the comment of an illiterate Kelantan Malay woman to the an-
thropologist when confronted by local political party symbols of a

sheaf of rice, a fishing boat under sail, a star and crescent moon. She
said she voted for the sheaf of rice because to her as a woman rice was
her concern whereas a fishing boat was an affair of men, and the
heavenly bodies were far away. So to describe such devices as
'symbols' of political parties ignores a possible difference in level of
statement-in simplicity or directness as against complexity or
indirectness of representation. To put it another way - the associa-
tions of the device might be treated by some people as agricultural
and domestic, not state-wide and political as intended.
AN ANTHROPOLOGICAL APPROACH TO SYMBOLISM

I have shown the existence of a very wide range of symbolic
material - things called symbols and ideas about symbols - in the
current social milieu in which we all move. I have suggested too that
such material can be relevant to any general anthropological study
of symbols because of the problems of definition and image it
raises.
But what can be a specifically anthropological contribution to the
understanding of symbolism? What can an anthropologist do that
has not been done already by logicians, metaphysicians, linguists,
psychologists, theologians, art historians and the rest? Essentially as
I see it, the anthropological approach is comparative, observational-
ist, functionalist, relatively neutralist. It links the occurrence and
interpretations of symbolism to social structures and social events in
specific conditions. Over a wide range of instances, anthropologists
have observed what symbols people actually use, what they have
said about these things, the situations in which the symbols emerge,
and the reactions to them. Consequently, anthropologists are equip-
ped to explain the meanings of symbols in the cultures they have
studied, and to use such explanations as a means of furthering
understanding of the processes of social life. Victor Turner has said
of one of his studies - which have played a great part in modern
developments - that it is a demonstration of the use of rite and sym-
bol as a key to the understanding of social structure and social
process. Others have explicitly examined symbolic actions in their
social contexts to clarify the understanding of phenomena of political
B
26 SYMBOLS
or religious change. But I think that for many of us the prime rele-
vance of an anthropological approach to the study of symbolism is
its attempt to grapple as empirically as possible with the basic
human problem of what I would call disjunction - a gap between the
overt superficial statement of action and its underlying meaning.
On the surface, a person is saying or doing something which our
observations or inferences tell us should not be simply taken at
face value - it stands for something else, of greater significance to
him.
I take an illustration from my own experience in the Pacific,
years ago.* I remember seeing a Tikopia chief in pagan times stand
up in his temple and rub the great centre post of the building with
aromatic leaves drenched in coconut oil. Now you can oil wood to
preserve it or give it a polish, as decoration. And in the Pacific
you can oil your body and scent it with leaves, when you decorate
yourself, as for a dance. But as the chief did this rubbing he mur-
mured: 'May your body be washed with power.' Now scrubbing a
baulk of timber with a hunk of oily leaves is not a very elevated
intellectual act. But think of the timber as a body and of the fragrant
oil as a decorative medium. Think too not of a material body, but of
an invisible body - not necessarily with the shape of a post, but in
another context, an anthropomorphic body, of a spiritual being,
believed to control crops and fish and the health of men. Think too
of washing as cleansing, and cleansing as a preface to adornment,
and adornment as pleasing to oneself as well as to others. So you
can see this act as symbolizing the anointing of the body of a god
with fragrant scents to express the status relations and emotions of
worship - and to render the god more amenable to the requests of
his worshippers. This may seem a very faraway symbolism. Yet
think further of Christ's washing of the feet of his disciples; the
anointing of Christ by Mary of Bethany; the symbolic value to
Christians of the Cross, with its synonyms of the Wood, the Tree;
and think also of conceptions of the Eucharist (Chap. 12), of the
Mystical Body of Christ, of the Glorified Body of the Virgin. It is
not difficult to see that what we are dealing with in the Tikopia
case is a set of symbolic counters which though superficially very
• F o r more details and related ritual operations see Raymond Firth, 1967a, 209-11,
218-20, 234, 245; 1970a, 117-19, pi. I.
dissimilar to the Christian ones, share some of the same basic modes
of symbolic conceptualization and patterning. But the symbolic
arrangement is set in a social matrix of clans, chieftainship, modes of
bodily decoration, even of architectural design which need intensive
study for the symbolism to become fully intelligible.
The anthropological approach, fully applied, has as its objective
to provide a systematic description and analysis of such a symbolic
act in its verbal and non-verbal aspects; to distinguish those parts
of the action held to be significant from those which are incidental;
to mark the routine or standard elements as against those which are
personal and idiosyncratic; to get elucidation from actor, participants
and non-participants of the meanings they attach to the act; and to
set all this in its general conceptual and institutional framework, and
in the more specific framework of the statuses and group relation-
ships of the people concerned. This is a demanding task. But it has
been admirably done by many anthropologists - to mention here
only Audrey Richards, Monica Wilson and Victor Turner - whose
work I shall be examining in later chapters. Some anthropologists
have also studied change in symbolic idiom - as I myself have done
in the field of Tikopia religion (Firth, 1970a).
The study of symbolism, especially religious symbolism, is
fashionable now in social anthropology. There is a tendency to look
on this study as a totally new development, but in fact as I show in
Chapter 3, anthropological interest in symbols goes back at least
100 years, before the days of McLennan and Tylor. It is true that
until recently this interest was rarely intense, systematic or sustained,
and the modern interest is much more sophisticated, analytical and
highly focused.* I think there are several reasons for this delayed
development. Firstly, as a purely professional sequence of opera-
tions, systematic studies of symbolism have had to wait until a sub-
stantial measure of progress had been made in the more formal fields
of social structure, such as kinship and politics. Now that so much
groundwork has been laid we can build loftier constructions of
interpretation. Secondly, developments in the theory of communica-
tion and of semantics, of signs and their meanings, have focused
* In 1956 at the Philadelphia meeting of the International Ethnological Congress I myself
stated that little specific attention had been paid to symbolism by anthropologists. By con-
trast, Rodney Needham has recently written 'today, when there is an efflorescence of interest
in all forms of symbolism' (in Hocart, 1970, xxix).
28 SYMBOLS
attention on the interpretation of those elements of behaviour

where the meaning of the sign has often seemed most complex and
obscure. Thirdly, the growing interest in culturally-defined systems
of thought, and in concepts and thought-processes more generally,
has stimulated inquiry infieldssuch as symbolism, where the relation-
ships between elements seem above all to be of a conceptual kind.
All this is part of the relatively straightforward operations of
scholarship.
But I think two other reasons may be significant also. It is in
keeping with the general temper of our time to be attracted to studies
which concern themselves with the less rational aspects of human
behaviour, which tend to reject or criticize a positivist approach,
which make play with ideas of ambiguity, uncertainty, mystery. This
is probably in part a counter to or a relief from the demands for
rationality and precision of our industrial, machine-governed society.
The other reason is more personal. Some anthropologists (and I
think I should probably have to include myself here) find in working
out their position on symbolism a means of examining and stating,
perhaps resolving, some of their individual views about the nature
and determinants of human social relationships and activity. Here I
should say that while I am much impressed by a great deal of the
modern anthropological work on symbolism, I do not share all the
perspectives of some of its most distinguished exponents.
It seems to me to make sense, and to be relevant in the world
today, that anthropologists should try to interpret symbolical
language and symbolical behaviour and relate them to the range of
social forms and social values. In such study political symbols are
important. But I do not think that only those issues are relevant
which refer to political affairs - unless one conceives of the political,
as some of my colleagues do, as involving any kind of relations
between persons where power is concerned, irrespective of scale.
Religious symbols are important too, but I look upon them as
referring to the same order of reality as the rest, categorized by the
quality of attention given to them, not by the uniqueness of the
objects to which they refer. So while I include both political and
religious symbols in my examination in this book I deliberately take
in material from ordinary daily life - such as the symbolism of ways
of wearing the hair, of greeting and parting, of making and of accept-
ing gifts, of showing flags. I deliberately also try to consider private

as well as public aspects of symbolic behaviour and concepts, be-
cause I think that the inter-relationship between them has often been
neglected, by anthropologists as well as by other students of
symbolism. I think this relation between public and private, social
and personal symbols is important to consider because certainly
nowadays it seems that there are strong trends within society for
the rejection of traditional symbols and for the discovery, even the
invention, of new symbols - trends in which individual interests
and decisions are brought to bear upon the recognition of communal
symbolic forms.
Popular, unanalysed expressions of symbolism are of interest to
anthropologists because they are part of the raw material for com-
parative study of processes of human thought and action. They
reveal the direction and extent of peoples' involvement in social
processes of various kinds, and the quality of abstraction applied to
these processes. But at a more analytical level, specialized treatments
of symbolism also have their anthropological importance. Much
that philosophers, artists, art historians, literary critics, theologians,
have written about symbolism is not immediately germane to
anthropological studies. But I think it has distinct value for anthro-
pological purposes. Firstly, I find it a very proper satisfaction of an
intellectual curiosity to know at least the outline of the arguments
put forward about symbolism by specialists in other disciplines,
and the range of material they cover. Secondly, some of the illustra-
tions they give recall obliquely some of the data anthropologists
deal with, and suggest possible alternative lines of treatment.
Finally, some of the hypotheses they put forward about criteria for
identification of symbols, the relation between public and private
symbols, the relation of symbolization to expression and communica-
tion provide parallel or challenge to anthropological views. Yet
they often seem to lack that social dimension which is vital to an
anthropologist, and to make assertions which seem to an anthro-
pologist to be culture-bound, or 'ethnocentric'. So I think that no
systematic theoretical exploration of symbolism by anthropologists
should ignore the existence of such an interest by these other disci-
plines.
From this vast field of scholarly contributions I can only make a
SYMBOLS
3°
personal selection of a few examples which seem to me to have some
relevance for problems with which anthropologists are primarily
concerned. They may add to such limited interdisciplinary focus as
anthropologists have already sought.
E X P L O R A T I O N OF SYMBOLS I N LITERATURE
Literary interest in symbols can be traced far back in history. In

Chapter 3 I examine a few early-nineteenth-century examples.
Propositions about symbolism in literature during the late nine-
teenth century, and the often polemical examination of their signifi-
cance, did not precede anthropological references to symbols. But
if not more sophisticated, those in literature were at least of a more
highly focused quality. Unknowingly, they foreshadowed a present-
day anthropological concern for the relation between social reality
and models constructed to represent it. The parallel has occurred in
two ways: by the symbolists' rejection of naturalistic description as a
desirable even if achievable aim; and by their search for interpreta-
tion of deep forces of which the human operators themselves were
unaware. But the differences were also marked. In poetry, especially,
in the last decades of the nineteenth century, the aim of the sym-
bolist movement was to replace realism by the 'idea'. But the 'idea'
was evocative not analytical, emotional rather than intellectual.
All poetry relies greatly upon symbolism. In the French tradition,
for instance, Francois Villon, most vigorously direct and human in
so many of his ballads, often used a symbolic image. The poignant,
haunting if hackneyed ou sont les neiges d'antan? ('where are the
snows of yester-year?') evokes not merely the ladies of bygone days
of his poem's title, but also, one may think, the purity, calm and
beauty of all that one has held most dear. But this is very different
from the attitude of the proclaimed Symbolist.* The symbolism
of Mallarme and his followers aimed at the discovery of the pure
poetry arising from what was thought to be the non-logical,
* Cf. Guy Michaud (1947, 637) - in every age there has been symbolism in art and litera-
ture (du symholisme); but Symbolism (le Symholisme) really saw the light in France in the
second half of the nineteenth century. Cf. also £tiemble, who quotes Catulles Mendes to the
effect that symbol is as old as the world; refers to the many meanings of symbol and to the
wide variation in the definition of Symbolism; and gives an astringent critique of views which
treat Rimbaud as a Symbolist (1961, 67, 68, 89-90).
AN A N T H R O P O L O G I S T ' S R E F L E C T I O N S 31
irrational quality of language itself, trying, for example, to express

ideas by associations of form and colour. Verlaine said of Mallarme
that he was preoccupied with the intensity of beauty, and that
clarity was to him a secondary grace. Verlaine himself wrote, in
Poemes Saturniens:
Le Poete, L'Amour du Beau, voila sa foi,
L'Azur, son etendard, et L'Ideal, sa loi.
Here the cz{ar, the clear blue, represented in poetic symbolic language
the heaven, the infinite. But the influences involved were complex.
This French symbolism was a violent reaction against the classic
formalism of the earlier generations. It was also a reaction against the
frustrations which followed the military defeat of France two decades
before. But above all - and not unconnected with the foregoing - it
was, as Alfred Poizat has put it, 'the entry of the dream into litera-
ture, it was the turning back of the view from outside inwards'
(1924,143). Or as Arnold Hauser has summed it up: 'For symbolism,
the whole of empirical reality is only the image of a world of ideas'
(1952, II, 896). Mallarme was interested by 'the mystery of ideas';
he and his associates held that the only certain realities were our
ideas and consequently our dreams, our imaginings. We only know
the external world by images which it leaves on us and by the
reactions of our sensibility. Relationships between concrete and
abstract, thing apprehended and apprehender, could be created, it
was believed, without overt conceptual effort. Precise direct state-
ments of conscious formulation were regarded as less poetically
revealing than vague indirectly allusive utterances which could be
treated as products of unconscious impulse. Yet the romanticism of
the early part of the century, in the sense of a deliberate cultivation
of emotion, was rejected, especially when it used conventional meta-
phorical language. Shot through such claims for a purer aesthetic, in
which symbols would express the true nature of relationships more
adequately than would a set of descriptive terms, was a theme of
spirituality, the idea of an inner incomprehensible, incommensurate
quality of being of which the symbol stands as the outer representa-
tive. Hence Bowra (1943, 2-12) saw in retrospect the Symbolist
movement of the nineteenth century in France as fundamentally
mystical - in its intensity, its irrationality, its disregard for other
32 SYMBOLS
beliefs and its reliance on a world beyond the senses. So, while
Mallarme spoke of art as impersonal, by concentrating on his own
private visions the Symbolist stressed in effect the subjective charac-
ter of his art. Michaud, with his exhaustive documentation, has
pointed out the variety of Symbolist positions. But he has seen the
essential message of symbolism as providing a synthesis in the
simultaneous expression of different degrees or levels of reality, or
expressing truths which are valuable on different planes at once. So
the symbol is revelatory of an aspect of truth, of truth in its subjec-
tive aspect (1947, 414-20). And as Michaud has indicated, whatever
be the judgement as to the validity of the symbolist arguments, they
rendered a great service in bringing to the surface a consciousness of
the symbolic process.
This classical symbolist view in literature, which had its analogue
in painting (see later), made two kinds of statements about symbols
which in effect challenge much of the anthropological approach to
the subject. They asserted the primacy of the private recognition of
the symbolic; and they claimed that the referent or reality itself, can
be apprehended only through the symbol. An anthropologist is con-
cerned primarily with the public use of the symbolic, and his aim is
to separate symbol from referent so that he may describe the relation
between them. But while ostensibly about poetry or art in general,
such formulations have offered themes for stimulus and scrutiny to
anthropology.
The influence of the Symbolist movement was wide and enduring,
and in modified forms it attained a cosmopolitan status. Bowra
recognized its counterpart in England in the aesthetic movement,
with Rossetti, Pater and Oscar Wilde. But for Mallarme his 'great
idol' was Edgar Allan Poe, whom he translated 'admirably' and for
whom he wrote a lament - 'Le Tombeau d'Edgar Poe' (Verlaine,
1884, 54, 55; Bowra, 1943,9)-
The symbolism of nineteenth-century American writers such as
Poe, Melville and Henry James has been the subject of considerable
analysis by literary critics. Since various cultural assumptions are
built into such analysis, a confrontation between critics and anthro-
pologists might give it a greater flexibility and sensitivity, as well as
furnishing some new viewpoints for the anthropology of symbols. I
give here merely a hint of an anthropological tentative in this direction.
The author of one of the most important works in this field,

Charles Feidelson (1953), has argued that there is a 'family pattern'
in that phase of American literature which began with Hawthorne,
Emerson and Poe and ended with Melville and Whitman; that this
pattern is one of symbolism; and that the concept of symbolism is
not only a key to their situation but also a link between their situa-
tion and the present-day (Feidelson, 1953, 3, 5). Classed as roman-
tics-some as transcendentalists - they were regarded as minor
disciples of European masters; regarded as symbolists, even though
they themselves would not have recognized the title, they can be
credited with literary independence, looking forward to 'one of the
most sophisticated movements in literary history'. With the general
appropriateness of this classification an anthropologist is not con-
cerned - though he may note its faintly nationalist flavour. But it is
significant that in this literary context, to be credited with symbolism
in his operations gives a writer a mark of distinction. Some of the
studies in this field have analysed works which have been at least
marginal to anthropological study; others have considered works
with no direct anthropological bearing. But in both categories it is
the methods used by the literary analyst which are of interest - in
their identification and interpretation of symbols.
The symbolism of Herman Melville's Moby Dick has been a sub-
ject of study for half a century.* But Feidelson finds a latent sym-
bolism even in Melville's Typee, an allegedly descriptive book, which
has been used mildly for ethnographic reference on the Marquesas
Islands, where Melville spent a month in 1842. For Feidelson the
book is primarily a travelogue; its scene is the solid earth; and 'the
language does not often invite a symbolistic interpretation.' Yet
the stuff of Melville's experience seems to 'hover on the verge of
the symbolic expansion' it was to undergo in his later works. The
development of this idea is interesting from an anthropological
angle. Typee, opening with Melville's 'basic cartography' of the
division of earth and waters, stresses the circular pattern of the
island encompassed by the ocean. According to Feidelson, it shadows
forth the pattern of Melville's world, which in this literary critic's
view, is remarkably like the spherical universe concept of Emerson.
Now this raises a problem of attribution - of relation between
* It has been used for anthropological comparisons by Victor Turner, 1962.
34 SYMBOLS
symbolization and empirical reality - which has relevance for all

anthropologists interested in symbolism and not only those con-
cerned with Pacific islands. For Melville's Typee is not only one of
the semi-ethnographic sources for our understanding of early cul-
ture-contact conditions in Polynesia, it is also redolent of the ex-
perience of living in a simple Pacific island environment. I can myself
endorse some of his observations. As one who has lived for a long
period on a remote isolated Polynesian island and who faced day-
after-day on every side a blank horizon across a vast expanse of
water, I too have been impressed by the 'circular pattern' of my
island universe. With no thought of Melville I wrote of this island:
'in an abstract schematic way one may think of Tikopia as a circle of
land bounded by the wastes of the ocean, and just within the land
edge a circle of houses . . . ' (1936, 79; 1969, 65). But in this analogy
with Melville's conception, as interpreted by Feidelson, two points
emerge. One is that the identification of an abstract pattern in a
topographical field is not necessarily the same as the identification of
a symbol, unless the term symbol be taken in the very broad sense of
any representation of the field. (An abstract painter may strenuously
deny that the patterns he produces have any symbolic quality.) An
island microcosm in an oceanic macrocosm is not by itself a sym-
bolic image, though it may well have led Melville towards the formu-
lation of later symbols of such order. The second point is that while
Feidelson links Melville to Emerson in symbolic development,
mainly by contrasting the quality of their symbolic imagination, he
seems to be misled in stressing the resemblance between the geo-
metric patterns of their respective worlds: a circular pattern of island
and surrounding ocean is only superficially akin to the sphere which
was the unit of Emerson's formulation of experience. As I see it, this
example of latent' symbolism attributed to Melville can serve as a
kind of cautionary note for anthropologists concerned to identify as
symbolic, primitive forms of representing the natural and social
universe. But Feidelson's comments on literary symbolism in general
include the very pertinent statement that the centre of symbolism is
not in private feelings, but in the public medium of language
(Feidelson, 1953, 51). In other words, communication is the test of
a symbol that survives.
With Henry James the significance of literary symbolism for
anthropologists may be deeper. In The Turn of the Screw we are

given the raw material for a fascinating series of questions about the
nature of mental constructs, ideas of control of one personality by
another, the forms which symbolization of mental force may as-
sume, the relation of innocence to knowledge of evil. All of these
emerge in different shape as problems in the anthropological study
of witchcraft and related issues. But what Edmund Wilson has
termed the ambiguity of James may pose an interesting problem of
the relation between conscious and unconscious process in symbol
formation. Edmund Wilson has been admitted by Feidelson to have
been the first to remark on the affinity of nineteenth-century Ameri-
can writers with the symbolist aesthetic that produced much of
modern literature. More than a generation ago Wilson went to some
pains to elucidate the symbols in which as he saw it Henry James
presented his 'chronic inhibition' - some degree of sexual impotence.
One such symbolic presentation, according to Wilson, was that of
the maiden innocence of immature girls violated or destroyed. In a
striking example, Wilson identifies as 'one of the most curious sym-
bols' for James's condition the wax dummy in a Parisian hairdresser's
window, with which in one of the stories a man falls in love; he
finally buys the dummy and takes her home to live with him. Of high
significance in this interpretative analysis is the statement that the
dummy was cut off at the waist.* The interest of this example for
anthropologists is primarily methodological. That a sculptured
figure can be a surrogate for a human being and that sexual impo-
tence (being only 'half a man') can be represented by a figure lacking
sex organs and lower limbs, are not particularly novel ideas. Nor is
there much relevance for anthropology in tracing these ideas back
into Henry James's fantasy. But in the prevailing climate of anthro-
pological concern with symbolism the literary critic's methods of
identification of symbols are of interest. Clearly he is using very
much of an ad hominem argument. The conception or depiction of
the upper half of the body - head, bust and torso - as representative
of a person is obviously not in itself a sign of sexual impotence in an
artist - otherwise a whole host of sculptors, portrait painters and
* Edmund Wilson, The Ambiguity of'Henry James (1962,147,149^ the essay was originally
published in 1938). It may be coincidence that among the 'symbols of the subjection of
women* carried by Women's Liberation demonstrators in London in March 1971 was a
tailor's dummy, cut off below the waist (Observer, 7 March 1971).
36 SYMBOLS
society photographers would be implicated, as would even the pur-

chasers of what used to be called these 'likenesses'. The symbolic
character of a portrait is well-known. But in James's story the
dummy was not a portrait; the man fell in love with the figure, not
with a living person; and he exercised power over it by buying it,
and domesticating it. In the story the figure was a private symbol for
the chief character. What the critic has done is to make a trans-
ference and interpret it as a private symbol for the author. For
anthropologists, this raises the issue of the steps by which the
transference was made, of the criteria by which one particular item in
an author's product is picked out as symbolic of an aspect of his own
personality, and others left alone as simple imaginative creations not
implicating the personality of their author.
This problem is: How does one identify an object or an action as
symbolic, not only for the interpreter but also for the author of it,
when he provides no overt clue? It is an issue which should concern
anthropologists more than it has done heretofore. In Wilson's
otherwise sensitive essay I do not find much guidance on this
methodological issue, though it implies that the identification of
private symbols of public men can be of general interest. But one
thought arises by contrast - an anthropologist faced by a problem of
identification of private symbols can seek justification of his attribu-
tion because what he observes has consequences. Behaviour in
regard to the hairdresser's dummy ceases at the end of James's story
- 'what came after' is no more than a parlour game of guessing. But
if a man in the anthropologist's field of observation bears off a
female wooden figure to his home, a whole series of actions by the
man and his household are bound to occur, and reflect back in con-
firmation or contradiction on the anthropologist's original inference
about the symbolism involved. The anthropologist, though more
constrained than the literary critic in his symbolic attributions, has
more opportunity of testing them afterwards.
It has often been pointed out that symbolists in literature were not
united under any single banner, but ranged from those proclaiming
ideals of decadence through a Nietzschean brutal sensitivity to
occultism and Christian mysticism. And modern French studies in
this field, especially, are of great elaboration and sophistication - far
more than I can indicate here. But explorations of symbolism in
AN A N T H R O P O L O G I S T ' S R E F L E C T I O N S 37
literature among much of general intellectual interest have raised

several kinds of problem of concern to anthropologists: What
phenomena are classed as symbolic and what criteria are used for
identifying them? How far can the exploration go in the identifica-
tion of covert or latent symbols apparently unperceived by their
author? What is the relation between symbols of a public order and
private symbols of the poet or other writer? And is there an irrecon-
cilable opposition between the aesthetic, intuitive aspects of sym-
bolic recognition and an intellectual apprehension of symbolic form
and function?
SYMBOLS I N ART
Analogous processes of looking for symbols and discussing them

have been important in the study of painting and other graphic arts.
The interpretation of the symbolic qualities of Christian religious
art of the early mediaeval period in Western Europe can be used as
an illustration, since its lack of interest in naturalistic proportion and
its direct relation to ritual and belief provide analogy with the 'primi-
tive' art studied by anthropologists.
The mediaeval church was preoccupied by symbolism. The
church building itself was held in structure to represent the human
body, the chancel being the head and the transepts the arms. But it
also represented the Cross of Christ... and so on (Dunbar, 1929,
403-4). Romanesque painting and sculpture presented figures which
for the most part illustrated Biblical scenes. But these were not
simply translations into fresco or stone of written stories which an
illiterate population could not read; they also were meant to carry
messages of more general import. Old Testament scenes were in-
tended to prefigure New Testament events; animals as well as human
beings symbolized abstract qualities or spiritual powers. Take for
example the figure of the lion. In contrast to the leopard, which
appears rarely in the mediaeval sculptural field, and is the symbol of
cruelty, sin and hence of Antichrist, the Devil, the lion is very
common and generally represents a noble principle. (Later I discuss
the significance of Samson destroying the lion.) Following the
ancient bestiaries, the lion was regarded as the king of beasts, strong
and compassionate (Camerarius, 1668, II, 13). When shown fighting
3« SYMBOLS
a dragon or serpent (as on the Percy tomb at Beverley or a tym-

panum at Lieding in Carinthia), it represented the power of good
combating evil. According to the bestiaries, the lion had three
important characteristics, which though fantasy creations, were
elaborated by Christian exegesists in symbolic terms. The lion was
thought to sleep with its eyes open, and so could represent vigilance
- so, according to some views it was an appropriate guardian for
church doors, and hence appears on many pillars.* In particular, it
was an image of Christ (the Lion of the Tribe of Judah) who was
awake during the night in the tomb awaiting his resurrection. The
lion typified the resurrection of Jesus in another way. Its cubs were
said to be born dead, but watched over by their mother for three
days, after which time their father came and breathed over them, so
giving them life. (A boss in Canterbury Cathedral illustrates this
theme.) Again, it was held that when pursued, the lion obliterated
the traces of its footprints, its spoor, by sweeping the ground with
its tail - so symbolizing Christ veiling his Godhead in the Virgin's
womb (T. H. White, i960, 7-11).
But such interpretations varied according to context. When
Samson was depicted in Romanesque sculpture as overcoming with
nothing but his hands the young lion that had roared against him
{Judges 14: 5) this exploit of courage and strength was given various
symbolic meanings in which essentially Samson represented Christ
and the lion was the Devil. (This interpretation was reinforced by
Psalm 91: 13, by which the godly shall trample the young lion
underfoot.)
* M. D. Anderson (1938, 41-2), Emile Male (1947, 332), Novotny (1930, pi. 65), Gantner
and Pob^ (1956, pi. 80). But Male argues that lions in confrontation, at portals and on
capitals, for the most part were not symbolic but decorative. Calling on the authority of St
Bernard, who said they made no sense, Male holds that such lion pairs simply reproduce
designs of oriental fabrics. Such monsters on capitals were designed not to instruct but
please (Male, 340-1, 437). Similarly Male challenges the interpretation of a male figure seated
between two lions, and holding them round the neck, as being a symbolic portrayal of
Daniel in the lions' den. (For example Peter Meyer, 1945, pi. 5, for cathedral of Chur; Joseph
Gantner and Marcel Pobe\ 1956, 55, pi. 58, for Charlieu-Loire.) Male is sceptical of such
'Daniel' attributions except where the figure is engaged in prayer. He argues that Daniel was
not a lion-tamer, and that Romanesque artists represented the true Daniel praying between
lions not daring to approach him. So according to Male, the pseudo-Daniel on a capital of
Moissac cloister is really the Assyrian mythic hero Gilgamesh (1947, 352). The identifica-
tion of a man between lion and centaur on a capital at Serrabone as Gilgamesh, by Marcel
Durliat on the authority of an unpublished thesis by Mme Favre may also be due ultimately to
Male (Durliat, 1952,68-9). My main point in all this is to mark the uncertainty of many sym-
bolic attributions by later generations.
Of the many sculptured presentations of this theme* one of the

most spirited and aesthetically most satisfying is that on a tympanum
in Gurk Cathedral, Carinthia. Samson, with long hair and flowing
mantle, is towering over the lion. There is great aesthetic harmony in
the sweep of line, balance of mass and repetitive ornament of the
design. The sculpture gives an impression of great vitality, of force
momentarily arrested as Samson has opened the lion's jaws and
prepares to tear them apart while two rather astonished doves wait at
the sides. The combination of these artistic qualities with religious
values has tended to maintain interest in the symbolic character of
the sculpture. As late as mid-1971, 800 years after the execution of
the figures, I heard a sympathetic and informed interpretation given
by a young student guide, to the effect that Samson represents our
Saviour; the doves are souls waiting to be saved from the Devil, the
lion, who would otherwise devour them; the mantle flowing out
from Samson's shoulders represents power, that of the Saviour.
Here is thematic development and secondary symbolism. The
Samson story in the Old Testament symbolizes not just strength, but
the power of virtue - a power that was lost when virtue was later
seduced. The symbol of the power was Samson's hair, uncut since
his youth (see Chap. 8) but the significance of this has been extended
to his cloak, also flowing freely about him. And the symbolic pre-
sentation of strength through virtue of a human person has in turn
been used to symbolize the spiritual power of the supreme godly
person - the less symbolizes the greater.
Symbolism of this elaborate kind provided an important basis for
the developments of symbolic form in religion and art of Western
Europe for several centuries, and as I have shown, its effects have not
completely disappeared. But study of this kind of material reveals, in
addition to the rich and fascinating content and system of thought
involved, a couple of points of more methodological interest. One is
the quest for symbols which obviously operated in the Middle Ages
* For example on a capital of Autun Cathedral (Grivot and Zarnecki, 1961, 63, pi. 29;
Kiinstle, I, 297 ff.); on bronze door of Augsburg Cathedral (Leisinger, 1957, pi. 42); on apse
of Schongrabern (Novotny, 1930, pi. 5); on tympanum in Gurk Cathedral (Novotny, 1930,
79-81, 83, pi. 62); on tympana in Stretton Sugwas church, Herefordshire and Highworth
church, Wiltshire (A. W. Clapham, 1934, 138, 142, pi. 34; C. E. Keyser, 1927, figs 84, 85 -
who interprets the Highworth figure as David (Hi)).
For Samson as a type of Christ, see F. E. Hulme (1899, 126), who cited the lion (ibid, 4)
as an instance of how a symbol may mean two entirely different, opposing things.
4o
SYMBOLS
as it does today - a search for some concrete representation of what

is not evident to the senses but is felt to be of prime meaning. The
other point is the difficulty of symbolic transfer - of ensuring that
the object created or selected by one person as a symbol is identified
by other persons as having the same meaning. This difficulty is
increased when even though the creator of the symbol deliberately
set it up as such, he is long since dead and no check can be made of
the inferences as to his intentions. And it exists in another form when
the imputation is of unconscious symbolism, which needs most care-
ful collateral evidence to justify. It is evident that mediaeval painters
and sculptors meant more by their work than just illustration; this is
clear from the bestiaries and from contemporary theological writ-
ings. But whether our modern interpretations of the symbolism, or
even the contemporary interpretations of the clerics, actually con-
form to what the artists and craftsmen meant, seems often to be an
open question. The interpretation of Samson's mantle as power is an
example. As Paul Frankl has pointed out, in painting and sculpture
the symbolical meaning of the representations was usually present
already, in their creation; in architecture, however, it was often read
into the work afterwards. For instance, Suger, Abbot of St Denis,
ignoring structural considerations, argued that where there were
twelve columns in a church, this signified the Apostles. Again, in
church symbolism, the portal was interpreted as meaning Christ,
because Christ is the way to the Father-but every church must
have a door (Frankl, i960, 213).
In my view, such mediaeval art offers distinct parallels for anthro-
pological study. The world in which it was produced was before the
advent of a machine-based industrialization, with much greater
emphasis than today upon social relationships of a personalized
character. Much of the art was concerned to illustrate concepts in the
direct service of religion. Moreover, as in the 'primitive' societies of
the anthropologist's classical field, the producer and the consumer of
the art broadly shared the same set of values and interpretations of
the symbolic images created. The symbolic values of art were re-
garded as not alien to those of society - except in so far as they
portrayed ideals which all members of society, including the artists,
were believed to share, but to be incapable of attaining without
supernatural aid.
The avowedly symbolist movements in art in the late nineteenth

and early twentieth century were of a very different order. They were
not essentially religious in inspiration. Very loosely, they can be
seen as in part a rejection of the achievement of that technical pro-
gress in the arts which had been a preoccupation of artists since the
Renaissance, and a wish to seek achievement in other directions
thought to be less hampering to the creative imagination. Both the
stern discipline of the Impressionists' use of the prism in the painting
of light, and the analytical structuralism of a Cezanne were alien to
them. There was also an element of revulsion from the positivism
which had sought to depict objective reality; a search for presenta-
tion of truths hidden behind visible external appearance. There was
social stress too - overt or implied criticism of the matter-of-fact
urban middle-class society which surrounded the artists; and in
France a reaction to devastating political events such as the Franco-
Prussian war of 1870. Led by Odilon Redon and Gustave Moreau,
with others such as Puvis de Chavannes and Paul Gauguin loosely
associated, such painters emphasized ideational and subjective fac-
tors in their work. They attempted to give an emotional value of
suggestion to their use of colour, which they treated less analytically
than did the Impressionists. They also saw a symbolic character in
marked black outline, as in Gauguin's Le Christ Jaune - or later
in the work of Rouault.
Developments in the twentieth century became even more com-
plex, as attention focused more strongly on the significance of other
than purely descriptive or 'naturalistic' values in art. In Britain a
philosophic note was struck by R. G. Collingwood (1925, 27-8, 96),
who in traditional idiom saw art as the pursuit of beauty, which he
defined as imaginative coherence, but regarded this coherence as
qualitatively different from the coherence of an object of thought. It
was in his view an immediate or intuitive awareness of relations
between parts of the object, involving a 'symbolic vision' which is a
'premonition' of the truth explicitly reached by science and philo-
sophy. Independently of philosophers' formulations, many artists,
despite the diversity of their aims and styles, were working in ways
which corresponded broadly to such an interpretation. There were
some banal attempts to portray symbolically national virtues or
other idealistic motifs. But sterner, more austere or more wild
42 SYMBOLS
promptings seemed to animate most artists, whether their major

stress was laid upon abstraction - the depiction of the 'world without
objects'; upon the visions of the artist's own inner self; or upon
symbolic forms suggesting principles underlying nature and human
existence. For instance, the work of the painters of the 'Briicke group
was characterized by 'an atmosphere of catastrophe in the social
sphere, a division of the ego in the private sphere; brooding revolt
against the false, narrow morality and an ecstatic, almost religious
individualism: these are the essential marks of German Expression-
ism' (Thoene, 1938, 49). The more articulate artists of the later
'Blaue Reiter group, however, much as their work and their
formulations varied, also incorporated the notion of substitution of
a world of essences for a world of appearances. Destroyers of con-
vention, intent on revealing 'the hidden strata of existence' they
regarded themselves not as rejecting nature but as overcoming and
interpreting nature. They were trying, as Franz Marc wrote, to
'create symbols which could take their place on the altars of the
future intellectual religion', or to 'create symbols which reassure' the
mind, by corresponding to inwardly apprehended feelings, as Paul
Klee is quoted as saying. In line with such developments, art his-
torians and art critics began to interpret primitive art, Byzantine and
Romanesque art, oriental art, as not stigmatized by lack of propor-
tion and technical blemish, but as being more concerned with
symbolism than with naturalism.
Yet as with the more popular references to symbols, there has
been some obscurity or at least some complexity in the use of the
term in painting. On the one hand, are the symbolic associations of
the work of art in the sense of the suggestions that it offers of ex-
periences not directly portrayed but implied in the forms presented.
Yet such symbolization is often rejected as inappropriate in inter-
pretation of the art object. On the other hand, there is the conception
of symbolization as a process of reference, not to objects of the
external world as ordinarily perceived, but to some other reality -
whether in the mind of the artist, in that of the observer, or in the
innate quality of existence in general. And cutting through such
categorizations are the efforts of some modern artists to avoid all
symbolization whatever, to present their creative effort as a direct
confrontation with experience, in the attempt to provoke a dynamic
reaction and change the situation. 'In proportion as the artist is pure,
he is opposed to all symbolism/ wrote Roger Fry. Yet even abstract
painting, while rejecting the traditional symbolism of conventional
representational art, acquired a symbolic value in the quality of the
response evoked in the viewer. Though it may be claimed that non-
objective art has broken through the process of symbolization itself
by confronting the viewer with a 'direct experience' of the forces
involved in the creation of the painting, this claim has proved hard
to maintain in its entirety. The language of identification with
creative forces of nature in which some exponents of abstract art
clothed their arguments; the influence of systems of mystical thought
on some abstract painters (for example theosophy on Kandinsky and
Mondrian); the attempt to give personal significance to formal struc-
tures - this has tended to involve symbolic forms of expresssion at
some stage. Michel Seuphor states that to Mondrian femininity is
symbolized by the vast horizontal receptacle of the sea; masculinity
is symbolized by the wooden pilings against which the waves break
and which protect the dunes from the sea. He argues that this
fundamental dualism became the basis of Neo-Plasticism, an aesthetic
system founded entirely upon the principle of the right angle,
roughly prefigured in nature by the opposition between sea-and-
horizon and dunes (1962, 37).*
Even where it is held that the conformations of non-objective art
are 'symbolic only of themselves', and the term 'metasymbolic' has
been introduced to discuss their achievement in analytical style, it has
been argued that what has been involved has been a spiritual revolu-
tion, and 'the history of the destruction of the outer world of
appearance signifies a gradual spiritualization of art, for it leads to
ever more symbolic statements' (Fingesten, 1970, 113). In an early
statement on the issues Herbert Read distinguished between symbol-
ism in the ordinary sense, employing concrete imagery, and
* Cf. Kandinsky on the Spiritual in Art. 'There is no form, there is nothing in the world,
which says nothing.* See also Jakob Rosenberg (1967, 225) who says of Kandinsky's Kleine
Welten series of lithographs where: *allusion to natural form is abandoned and geometrical
shapes are patterned by colour or black and white' - one may speak of a symbolic image that
conjures up a little cosmos {kleine Welt) filled with energy yet controlled by laws of attrac-
tion and repulsion. Compare also an opinion of de Kooning's 'Light in August', a study in
white and black, as a 'symbolist abstraction'. Dissociated from a source in nature the organic
shapes carry emotional charges of the same order as mathematical signs There is visual
metaphor in which motifs released from specific objects are thus able to strike a broader
resonance of associations (Harold Rosenberg, 1965, 115-16).
44 SYMBOLS
symbolism which employs abstractions without parallel in visual

experience and operates by unconscious or intuitive process. One of
the most articulate movements of this last type, surrealism, made an
endeavour to utilize a dialectical process of artistic activity opposing
conscious and unconscious, reason and unreason, deed and dream.
So surrealism deliberately undertook manufacture of 'the object
functioning symbolically' as part of an attempt to multiply the ways
of reaching the most profound levels of the mental personality
(Andre Breton, 1936, 86)*
There are two significant points in all this. One is that for the
most part the creative symbolism of these art forms has been im-
plicit, not explicit; the process of exploration, identification and
labelling of symbolic patterns has often been done by observers and
interpreters, unacknowledged by and often unknown to the artists
concerned. Secondly, when such symbolic pattern has been recog-
nized in their work, it has been part of the modern canon to claim for
it a clearer autonomy, a greater dignity in the aesthetic scene. Sym-
bolic statement has tended to be viewed as preferable to realistic
statement because, it is argued, it is capable of conveying more
general and more profound meanings.
Problems of the relation of private symbols to public symbols are
raised especially in such art. How does the individual vision of the
artist become translated into the set of symbols which win public
acceptance? Does art arise from a fundamental paradox - the equal
insistence on the creative effort of an individual, and on the capability
of his product to be recognized and accepted by a body of other
individuals - a public? Does the artist's belief that he alone must be
the ultimate judge of the validity of his effort mask a parallel belief
that only if the result of his effort is acceptable to some other indi-
vidual can he himself accept it as valid?
A great range of views opens up here. If the artist sees art as
basically a means of communication he obviously must try for a code
by which what he has to 'say', that is paint or sculpt, can be inter-
preted by those to whom he wishes to communicate. He may try to
find this code in the field of publicly recognized symbols - which is
* The first Surrealist exhibition held in London in 1936 included a mask olmwai type from
the Sepik, New Guinea (cf. Bateson, 1958, pi. 28). For other material bearing on the discus-
sion above see Roger Fry, 1928, 16-38, 284-302; Herbert Read, 1933, 65-71, 128-44; T94^;
Arnold Hauser, 1952, II, 935; H. K. Rothel, i960; William Gaunt, 1962.
what I take it 'pop art' has tried to do. Or he may try to enforce the
recognition of a new code, perhaps of an order which putatively
belongs to some generic human understanding - which is what I
assume the various kinds of abstract art are trying to do.
But what some modern artists are attempting, if I construe them
rightly, is to eliminate as far as possible the element of personality
from the interpretation of their creations. The individual, personal
component is ineradicably there, in the selective integrative act of
creation. But it is the individual as creator-artist only that it is in-
tended to be recognized; not the individual as a particular person
with sex, temperament, social background. The characteristics of the
individual painter or sculptor may be regarded by the artist as
irrelevant to the consideration of the work, distracting to the
recognition of the value of the experience provided by it. So a self-
effacing painter such as Bridget Riley can treat her work as an
attempt to stimulate by visual means, through combination of line
and mass by circle, dot, stripe, in colour or black and white, a kind of
personal experience in the viewer. The artist is the conceiving mind
behind the creation - not necessarily even the executant, since such
work may have been done by assistants following out the instruc-
tions minutely laid down (Robertson, 1971). Analogous mathemati-
cally derived sculptures and drawings have been produced by the
American artist Sol Lewitt. This is an attempt to reduce to a mini-
mum the overt involvement of the artist in the interpretative effort of
the spectator. It seems to be successful in that knowledge of the
precise personality of the artist, what kind of social and tempera-
mental figure he or she represents, does not appear relevant to
interpretation of the painting as part of the spectator's field of
experience. Here there emerges what I feel like calling 'muted sym-
bolism5. The artist says in effect: interpret the painting in your own
way, in terms of your own experience; let the combinations of line,
mass and colour convey their own message to you - or more strictly,
let them suggest to you some stirring of the sensibilities which will
make for you a cognizable experience, an 'event'. Symbolic meaning
in the more figurative sense is not expected - may be even denied.
There is a belief in a direct relation between the physical object and
the appreciation of the viewer so that the forms of the painting do
not 'stand for' something else than themselves. They are expected to
46 SYMBOLS
evoke reaction without the mediation of other images. According to

the fashionable 'structuralist' phraseology, the art forms 'mediate'
directly, in a primary way, between the raw impulse-phenomena of
human nature and the culturally-defined position of the spectator.
There is also a further attempt to reduce the importance of the
material art object, in favour of the mental image - hence 'concep-
tual art', 'minimal art', 'hyper-realism' and other varieties of concern
to obliterate as far as possible the humanist elements in art.
But as I see it, the artist is not in fact eliminated as interpreter. It is
recognized that we are confronted by a personal aesthetic of the
artist. And even in the most advanced fields of modern art there is
still curiosity on the part of art critics and public as to what the artist
'intended' by the work. The artist himself often shows no particular
reticence in explaining what he has meant to do, sometimes in naive
theoretical terms. Josef Albers, a Bauhaus veteran who has worked
for years with monochromatic rectangular figures - for example his
'Homage to the Square' - has explained that the reason for aesthetics
is ethics (a reversal of a more common position); that the colours in
his work illustrate both independence and interdependence; that they
submit to, help, hate, marry one another in an ethical as well as an
aesthetic order. Critics too may react in symbolic terms - as one who
appreciated the conceptual clarity of some mathematically-based
images stated that they had come to seem uniquely valuable to him,
possessed of a 'gentle but insistent moral force' (Grace Glueck, New
York Times, 3 December 1971; Peter Schjeldahl, New York Times,
3 October 1971). Such explanations, whether by artists or by critics,
can vary widely. Some have a great deal of'overkill' of a theoretical
or speculative kind, in which one feels that the exaggerated language
is more concerned with the effect on the hearer or reader than with
strict interpretation of the work of art. But they help to demonstrate
the proposition which Gombrich has so elegantly defended in Art
and Illustration - that the work of art is a complex product in a
system of relationships in which artist and viewer are integrally
involved.
The notion of symbol itself as a product of an integral set of
relationships between fabricator and interpreter is congenial to an
anthropological inquiry. At the same time there is an aesthetic prob-
lem which is outside the anthropological range - of the nature of the
art forms which give the symbols their expressive power. The focus
here is not on what the symbols mean, for artist, for public; but on
what forms can be constructed so as to convey meaning most force-
fully, or most sensitively. (Presumably, 'pop art' could be defended
on the grounds that while its message may often be trivial, the forms
in which it is presented are powerful in suggestion.)
SYMBOLS IN RELIGION
Anthropologists may have been able to ignore what has been said on
symbolism by poets, artists or critics, but it has been harder for them
to leave aside discussion of symbols in religion. Some anthropolo-
gists indeed have directly used their interest in Western religious
symbolism to help them in the interpretation of primitive symbolism.
The symbolism of Oriental religions, especially Hinduism, is of
an unparalleled richness, and rests upon a very clear differentiation
between symbol and god, spirit being or cosmic force symbolized.
But for our preliminary inquiry here, a few reflections on the sym-
bolism of Christianity will suffice to illustrate the issues.
In the West the long and complex history of the term symbol itself
is much involved with religion. In Greek, originally sumbolos and
related words referred literally to the putting together of that which
had been divided. An example would be the production of two halves
of a token which had been broken and given to a pair of friends so
that they would share a mark of identification. Conversely, this token
served to differentiate them from other people who had no such
proof. It was in such sense that the early Christian church seems to
have adopted the term symbol. It came to be used for a formal
authoritative statement of religious belief, differentiating Christians
from non-Christians - a Creed. (It is suggested that it may have
also been used in mystery religions of the period, for an exterior sign
indicating the inner secrets of the cult.) In particular, the 'symbol'
applied to the confession of faith recited at baptism by a convert. In
the form of the so-called Symbol of the Apostles (Apostles' Creed)
this formal statement of belief was believed (apparently without
adequate historical evidence) to have been a joint profession of faith
drawn up by the disciples of Jesus after Pentecost. This 'symbol' be-
came then a kind of admission ticket to the new church. Now, not
48 SYMBOLS
only was the church struggling to establish itself against paganism

and against other Oriental religions, it had also to contend with its
own internal power struggles. There were differences of view about
who should exercise authority, about administrative jurisdiction, and
also about doctrinal matters; and such disputes were often expressed
in denunciations of heresy. Theologians and church leaders of the
second and third century A.D. used this baptismal symbol in the
name of apostolic legitimacy against gnostics and others who
diverged from their authority, denying them its use, or enforcing its
use against their beliefs. (For details see F. Kattenbusch, 1894,1900,
and review by Marcel Mauss, 1900-1, 298-301; Salomon Reinach,
1942, 262-3; A. C. McGiffert, 1953, 152-60; F. Creuzer, 1837-43,
iv, 503-17.) So nearly 2,000 years ago, this alleged symbol was not
just a sign standing for membership of a given religious body; it was
also an instrument of power, of definition and separation, a polemi-
cal weapon, whereby persons could be made to conform, or be
excluded from that body.*
As the concept of symbol developed in the early church it
acquired a clearer abstract usage, in which the notion of a material
token was linked with that of something unseen, immaterial. So the
material church, of wood or stone, became a symbol of the spiritual
church, the community of the saints and the faithful with the
Trinity. So too in the rituals of the church, baptism symbolized
purification, the first stage in the ascent of the soul to God, while the
eucharist, the sacramental taking of bread and wine, symbolized
the second stage, that of illumination. Such were solemn visible
means of grace, calling forth ideas of the invisible power and love of
God. In the fifth century, when such concepts of symbolism were
put forward, the writer purporting to be Dionysus the Areopagite
also held that owing to the transcendence of God, incomprehensible
and inaccessible to human understanding, what the Scriptures told
us of him could be only in symbolic language, suitable to our finite
* It remained so for a long period. Bishop Reginald Pocock, the first theologian to write in
English since Anglo-Saxon times, was condemned for heresy in 1457 because, in part, he had
expressed scepticism about the Apostles* authorship of the creed that bears their name
(Myers, 1966, 167-9). ^ n paintings and inscriptions much early Christian symbolism referred
to life beyond the tomb, using symbols such as the fish for mystical union with Christ or the
vine for the heavenly feast, of which the earthly pledge was the Eucharist. The cross appeared
relatively late in the fourth century, the crucifix about the eighth century. But side by side
with these came the development of symbolism of the regal authority of the church.
grasp. This theme, of symbol as an aid to understanding, penetrates

throughout the whole treatment of Christian theology and has
diffused more widely. It is assumed that symbols communicate
meanings at levels of reality not accessible through immediate ex-
perience or conceptual thought - hence it is argued, they are in some
sense revelatory. These meanings are often complex, and of dif-
ferent layers, as Dante defined the subject of his Divine Comedy as
not simple but 'polysemous'. So, the meanings are not apparent to
the uninitiated, hence instruction is needed to understand them.
Moreover, as Dionysus wrote, the images used were often mean and
contemptible, quite unlike what they represented, so there would be
no danger of the faithful confusing them with reality. Yet some
religious symbols, it is held, are 'natural symbols', because of their
associations. So elements such as the water in baptism, or the bread
of the eucharist are 'natural symbols' for spiritual cleansing and
nourishment, respectively, just as a key is a natural symbol for
authority because of its association with ownership or stewardship.
But religious symbols are not simply communication media: they
are held to be affectively charged, non-neutral in their emotional and
intellectual value. Moreover, they can also be envisaged as possessing
a spontaneous power, in themselves 'projecting the mind towards
the Absolute'. So, in the tradition of the mediaeval church, the mys-
tic interpretation of symbols at different levels 'simultaneously true'
led through literal description, allegorical instruction as to belief,
tropological instruction as to duty, to anagoge, elevation towards
the goal of union with the divine.*
In the modern religious field, a good illustration of a symbolist
approach is the fideism (or symbolo-fideism) of some Protestant
theologians in Paris at the end of the nineteenth century. The
central theme of this approach was in line with symbolist views in
literature and art rather earlier - that the essence of things escapes us
and that we know only their outward manifestations in the form of
figures, images, symbols. Hence no one can know what God is in
himself; all that it is possible to know are the ideas we have of Him
through our more or less anthropomorphized representations. So
* As illustration of the vast literature on this subject see Dunbar, 1929; McGiffert, 1953,
esp. Chap. 16; Somerville et al.y 1967. One of the fullest treatments is that of the Abbe* Auber
(1884, 4 vols), who examined the symbolism of the Scriptures, the church, church architec-
ture, decoration, ornaments, liturgy, etc., in great detail.
50 SYMBOLS
conceptions such as Lord, Father, Master and emblems such as Lion,

Rock, Sun embody some part of our symbolization of what we can
never fully comprehend (Eugene Menegoz, n.d., 151-2). These
theologians, obviously, were prepared to argue that such symbols
are without doubt the expression of a living reality of an extra-
human order, and they viewed their task as that of laying bare the
eternal truth from under its contingent manifestations and its his-
torical formulations.
Another view of religious symbolism was presented by Walter
Marshall Urban, an idealist philosopher of Yale (1939). He was
impressed by what he termed the 'intuitible' character of genuine
symbols. He saw the 'sense' of a symbol as a similarity in the way of
reflecting on the two things, symbol and thing symbolized - what he
termed 'community of subjective form'. (This brings to mind
mediaeval views on the doctrine of analogy, as well as Maurice
Godelier's emphasis on the criterion of analogy as basic to mythic
thought- 1971, 96-8.) Distinguishing strongly between sign and
symbol, Urban held that a 'mere sign' was a depersonalized symbol,
one that had lost its reference to the original object (1939, 413, 419).
He undertook a complex inquiry into the symbolism of poetry,
science and the Christian religion, and sought to formulate a general
theory of the symbolizing function. He distinguished between sym-
bolic and non-symbolic language, arguing that symbolic language is
found where the intuitive elements of language function for the
non-intuitive or conceptual. Drawing upon Jung, he held that any
'genuine' symbol in the primary sense must be the result of the
co-operation of conscious and unconscious factors. So a feature of
all symbolism is its dual character. Following Helen Flanders Dun-
bar, Urban classified symbols into: extrinsic or arbitrary symbols;
intrinsic or descriptive symbols; and 'insight symbols'. The first,
he thought, are particularly the symbols of science, the second, those
of art and religion. The third category, the 'insight symbols' -
really a sub-species of the second - do not really represent charac-
ters of relations but also lead to an understanding of them, as in
poetry and religion. Urban cited the continuous tradition from
Hellenic thought on through the Christian Fathers up to its com-
plete statement in mediaeval times, of the 'fourfold method' of
symbolic interpretation.
This idea of progressive interpretation, especially as it concerned

the moral function of symbols (topological), and their function in
elevation of the mind to higher things (anagogical), applied to
interpretation of statements of Scripture, including parables and
miracles. But in Urban's view it had also a universal significance.
He was concerned especially with religious symbols, and gave them
a very definite value. He argued that a humanist theory of religious
symbols was useless since it denied the objects for which the sym-
bols stood. A purely moral theory of religious symbols was also
untenable in his view, since they can only function morally by
being more than moral. Unless a religious symbol is given its ana-
gogic sense, unless it is metaphysically interpreted, its character as
a religious symbol is not properly understood. So, in his view, any
theory of religious symbolism must be a theory of religious cogni-
tion. If one were to accept his assumptions, one can see the force of
such argument. I think that echoes of something similar are to be
found in some anthropological writings on symbolism at the present
time. But Urban was more uncompromising than some of his col-
leagues. He argued that propositions that religious symbols have
human and social values are an attempt to retain the symbols without
the object which alone gives them significance as values (1939, 596,
606-8).
Urban also posed the question of verification. He held that while
the reference of a religious symbol is beyond our human moral and
value experience, its authentication can take place only within
that experience. This he saw as the element of truth in all moral
theories of the religious symbol. As part of his general theory
Urban pointed out that it is of the very nature of a symbol that it
contains both truth and fiction, both the real and the unreal. Along
this line he held that religious symbolism, much more than scien-
tific symbolism, is in its essence metaphysical in character, and
formed from the language of myth. It is always distorting the intui-
tion in order to suggest and represent the infinite and transcendental.
He called attention to the unique role which the 'element of nega-
tivity' plays in all religious symbolism, in that it is both the truth and
not the truth about the object symbolized (1939, 582-5).
This idealistic theory of symbolism raises some basic questions of
epistemology. But in its abstract conceptual approach, its emphasis
52 SYMBOLS
on a dualistic principle, and its analysis of material from myth and

religion, Urban's thought offers some parallels to the development of
anthropological interest in later decades. There is no evidence to
suggest that his ideas have had any marked influence on anthro-
pology. But they do bear upon the controversy which in religious
modernism has been called the distinction between truth of value and
truth of fact, and which has emerged in anthropology in discussion
about the meaning of stories of virgin birth.
So from theological writers and allied expositors, it is clear that
taking the Western Christian church as example, religious symbols
are regarded as largely institutionalized, as necessary aids to under-
standing and to action, and as a matter for intellectual and emotional
commitment. For the most part also they are held to be sacred in
that not only is their disturbance felt to be offensive, but the most
crucial symbols are regarded as having power in their own right to
lead the faithful to a more direct relationship with the divine. But
here arises a problem which is present in quite a different form in
literary or art symbolism - that of authenticity. With poet or painter
the question may be argued as to the validity of the symbols they
create - whether or not the symbols do the job claimed or hoped for
them, of evoking ideas, emotions, or stimulating experiences. It is
not ordinarily denied that the ideas, experiences, emotions can exist
in somewhat the form envisaged. But with religious symbols the
questions can take on a different aspect. The very existence of the
referent is not common ground among the commentators. Some
believe that there is an extra-human, divine entity or power, in-
visible, immaterial, even perhaps inaccessible, to be approached or at
least to be referred to by symbolic means. Others believe that this is
not so, and that the symbols purporting to make this reference are
actually referring to some other object-say, the operations of
human society or the character of human minds.
Anthropologists are in an ambiguous position here. Like a theo-
logian or other religious person an anthropologist has learnt to treat
religious symbols - anyone's religious symbols - with respect. He is
not expected to give them authenticity in their own terms - except
temporarily, perhaps by the people among whom he is studying
them. But some anthropologists believe firmly in the authenticity
of the symbols of one religious faith - for example Judaism, Chris-
tianity, Islam-while others are inclined to a kind of eclectic acknow-

ledgement that all religious symbols have some factor in their
referent which goes beyond the human sphere of comprehension.
Still others are avowedly humanist in their interpretations. There is a
kind of assumption of professional neutrality by anthropologists
towards the subject of investigation. Some have argued that in
studying religion, it makes a difference whether the anthropologist is
himself a religious person or not - though they have not specified
what this difference may be. For the problem of symbolic authenti-
city, the implications are still not clear in practice. No more than a
humanist, can a Christian, Jewish or Muslim anthropologist, if a
faithful believer, ascribe the same status to the pagan 'gods' of a
people he is studying as the people themselves do. But presumably
each will approach his study with sympathy and give inward as well
as outward respect to those religious concepts in which the people
express their deepest feelings. What should be understood is that,
contrary to views such as those of Urban, a humanist anthropologist
studying religious symbols is not alleging that they have no referent.
He may hold that the referent is different from that which religious
people attribute to the symbols, but that it lies in the field of human
desires, emotions, strivings, conceptualizations, institutional rela-
tionships.*
* On this point I would remark that one of the most profound and powerful analyses of
religious symbolism I have met is that by Gerschom Scholem, of the symbolism of the
Kabbalah. He points out, among much else, how Kabbalists based their interpretation of
Scripture on the assumption that every verse is a symbol of a stage in the divine process, an
impulse in the divine life; and how the Torah is conceived as a vast corpus symbolicum repre-
sentative of the hidden life in God. Scholem also expresses very lucidly in scholarly manner
the significance he sees in such studies (1955, 38-9, 205-35 assim).
Chapter 2
A QUESTION OF TERMS:
SCOPE AND MEANING
OF 'SYMBOL'
Most anthropological treatment of symbols has taken the meaning of

the term for granted, and focused on the interpretation of the rela-
tionship between symbol and object symbolized. While in general
this is justifiable, some exploration of what is meant by the word
'symbol' is needed, for several reasons.
Some anthropologists have complained that others writing about
symbols have caused confusion by failing to define their terms.*
More important, perhaps, is that as I have shown in Chapter i,
people in many other fields also write about symbolism in ways
which invite anthropological contact and comparison, but which
sometimes suggest that they are not always using the same cate-
gories. Without necessarily seeking agreement on all meanings, it is
useful to indicate common ground of statement and where dif-
ferences tend to arise.
In classical times Greek and Latin words related to symbol had
a variety of meanings, arranged round a notion of matching entities:
a sign or mark whereby one person gave another to understand
something; a token; a contribution of money to a feast; a share of
a reckoning; a commercial treaty between a pair of contracting cities
guaranteeing security and protection to their respective merchants.
The meaning of symbol ultimately developed into that of a concrete
indication of abstract values. In particular, the term became asso-
ciated in early Christianity with the set of beliefs forming the Creed,
in the canon known as the 'Symbol of the Apostles' (p. 47).
* For instance, Melford E. Spiro commented on a symposium entitled 'Forms of Symbolic
Action' that in none of the papers contributed was the term 'symbol* ever defined; the class of
symbols was never distinguished from the class of non-symbols. The papers were admirable
as contributions to anthropology, he thought, but it was not clear why they had been labelled
contributions to the study of symbolism, nor why their authors - including some of the most
distinguished names in this sphere-had imagined they were studying symbolism at all.
(This critique was subjoined to the papers - see Robert F. Spencer (ed.), 1969.)
54
A Q U E S T I O N OF TERMS 55
Historically, since the early period of Greek philosophy, epistemo-
logical problems have been concerned with the relation of particular
to universal, sense-perception to idea or 'form', finite to infinite; and
in the West have often taken a theological shape, as in theories of
analogy. Symbolism was one of the fields in which such problems of
knowledge were displayed, though as G. G. Coulton has pointed
out for the mediaeval period (1958, Chap. 13), symbolic interpreta-
tions were apt to be very much less systematic than those of re-
ligious doctrine in general. In such interpretation, the broad aim was
not only to accept the concrete as representative of the abstract but
to use it as a key in two ways - to explain the concrete by reference
to the abstract, the visible by the invisible; and to extract from the
concrete its hidden meaning for an understanding of the abstract.
As indicated in Chapter 1, such views were prominent in Christian
church art, but had a high development in some other religious
systems, notably that of the Jewish Kabbalah.
As part of symbolic inquiry, discussion of the relation between
names and things named has involved consideration of the nature of
words as symbols. This has meant, in Carnap's terms, a practical
distinction between language as an object of discussion and meta-
language, in which the discussion is carried out. But it has also led to
the invention of artificial languages, such as sets of symbols in
mathematics and science. The use of such symbols has allowed
operations to be performed with generality and brevity, in a way
that would not have been possible with ordinary language. 'Words
are the Signs and Symbols of Things,' wrote South in 1686 (OED).
But words, which can be flexibly handled in an almost infinite
variety of situations, are not precise enough to make fine distinc-
tions in 'the technical discussions associated with special activities in
a high state of development' (L. W. H. Hull, 1959, 101). Symbols
are not only more economical and more abstract than words; they
can also be made to conform to rules which allow of no irregulari-
ties. To quote Hull again: 'if the steps of an argument are symboli-
cally recorded, each step corresponds to a particular rearrangement
of the symbols.... We soon learn to recognize those symbolic
arrangements which correspond to valid reasoning processes. We
can then make such rearrangements automatically, without per-
petually considering their significance' (1959, 102).
56 SYMBOLS
But there are three points of importance about such symbols in

mathematics and science. They have been specifically invented to
allow certain kinds of operations to be performed. Their use is much
more limited than that of ordinary language. And, of special note,
they do not of themselves provide any information about the con-
ditions to which they refer. Such symbols are set in structural
arrangements, with given assumptions as to their values, and with
relations intended to be subject to variation. The resulting patterns
of relations are not regarded as giving an actual description of any
natural objects to which they may refer, but as suggesting that cor-
responding patterns exist for those objects. This is the function of
the use of symbols in say, economic models. There is analogy here
with religious symbols when they are used in a sophisticated way, as
in reference to a transcendental God. But as a rule users of religious
symbols tend to imply that such symbols do actually provide some
information about the object of belief, or have power to generate the
acquisition of such information. In the paintings in the Roman cata-
combs, for example, while the symbols of palm, ship, fish, shepherd
and the like were not meant for simple illustration or instruction they
were intended to portray by allusion the hopes of the Christians
concerned and even at times to suggest meanings that had otherwise
to be concealed. Much ordinary secular symbolization too is in-
tended as a means of communication of information and not just as
an instrument of analysis and exposition.
For anthropologists interested in the study of symbolism the work
of philosophers has been important because they above all have
devoted attention to questions of definition of symbols - the criteria
by which a symbol may be recognized, and the relation of the con-
cept of symbol to cognate ideas. From philosophy ultimately stems
the basic formulation in methodology - that our initial question
should not be 'What is a symbol?' but 'What is appropriate to
designate as symbol in the differentiation of processes of human
thought?' Criteria selected for classification of symbols may come
from the natural world, but 'symbol' is a cultural, not a natural
category. This point was made clearly a century and a half ago by
J. A. Dulaure, but it has been developed in this century especially by
Ernst Cassirer.
In his Philosophy of Symbolic Forms (first published in 1923) and
later in his Essay on Man (1944) Cassirer argued that symbolic
representation is an essential function of the human consciousness,
and is basic to our understanding of the operations of language, his-
tory, science, art, myth and religion. He distinguished the expression
function of symbolization, seen especially in myth (which he re-
garded as characteristic of the early stages of human culture) from
the intuition function, associated with the use of ordinary language,
and the conceptual function, manifest especially in the world of
science. In myth, sign and signified merge (as with Aquinas, exist-
ence and essence merge in God); in ordinary language and the world
of common sense they are differentiated, producing systems of ob-
jects or substances; and in science their differentiation produces
systems of relations. Cassirer postulated what amounted to a dialecti-
cal quality in symbolic patterns of culture - an opposition of ten-
dencies towards stabilization and towards disintegration of symbolic
relations.
But philosophical propositions about symbols may be of interest
to anthropologists not only because they offer frameworks for hand-
ling problems of symbol definition and classification; they sometimes
incorporate judgements about society, resting upon assumptions
which may be unstated, but which run counter to anthropological
experience. I take an example from a field in which anthropologists
have made few contributions but where they do have some know-
ledge. This is the symbolism of dance.
In a sensitive and stimulating theory of art, built around thenotion
of symbolism, Susanne K. Langer has discussed the evolution of
dance. Following Cassirer, she has regarded primitive phases of
social development as entirely dominated by 'mythic consciousness',
based on the symbolization of subjective feelings of potency, as
'potent Beings'. 'From earliest times, through the late tribal stages,
men live in a world of 'Powers' - divine or semi-divine Beings,
whose wills determine the courses of cosmic and human events
The first recognition of them is through the feeling of personal
power and will in the human body and their first representation is
through a bodily activity which abstracts the sense of power from
the practical experiences in which that sense is usually an obscure
factor. This activity is known as 'dancing'. The dance creates an
image of nameless and even bodiless 'Powers' filling a complete
c
58 SYMBOLS
autonomous realm, a 'world'. It is the first presentation of the world

as a realm of mystic forces . . . ' (1953, 189-90). Langer has argued
that dance is the most serious intellectual business of savage life - the
envisagement of a world beyond the moment of one's animal exist-
ence, the first conception of life as a whole. So the prehistoric
evolution of dancing is the very process of religious thinking, which
begets the conception of 'Powers' as it symbolizes them.
Now while anthropologists can learn much from Langer about
the theory of symbolism and the theory of art, and find fruitful
material for discussion in her views about the relation of 'public' to
artist, these statements about the symbolism of dancing will seem
superficial and romantic to many anthropologists. In the most
primitive societies we have studied, dancing is a highly social
activity, and not just an individual abstraction of a sense of power.
It is very often of grave religious import, but parallel to the ritual
dances are commonly other dances of plainly secular significance,
concerned with recreation, sexual alignment, social display, status
interest, property rights. And in the religious dances themselves, any
concern with 'nameless bodiless Powers' is often subordinate to very
detailed orientation towards personalized spiritual beings of com-
plex social affiliation. (See, for example E. E. Evans-Pritchard, The
Dance (1928) 1965a, 165-80; Raymond Firth, 1967, 281-372; A. R.
Radcliffe-Brown, 1922, 128-32, 134, 246-54; Roy A. Rappaport,
1968, 186-9; Marie Reay, 1959, 154—5.) What Langer has to say in
other contexts about the symbolism of the human body is very
illuminating, but as an illustration to the history of symbolic thought
her para-philosophic analysis of the dance is misleading.
An anthropological approach to the study of symbols should then
in my view be prepared to scan philosophical treatment of the sub-
ject, primarily for elucidation of the theoretical frame, but also to be
prepared to offer comment on its cultural content or assumptions.
A basic problem in the study of symbolism is the status of the
relation between a symbol and that which it represents. Some have
argued that there must be a natural or 'real' link between the thing
recognized as a sign (signans) and that for which it stands (signtfica-
tum); others have held that the relationship is only one of conven-
tion. Historically, the concept of 'natural' in a symbolic relationship
has been used by anthropologists in several different senses. For
Robertson Smith (1889, 180, 189) 'natural' symbols of divinity in
Semitic worship were rocks, trees, fountains - things of the external
environment which had independent existence unaltered by man, as
opposed to pillars or cairns of stones erected by human hands. But
the 'natural symbols' of Mary Douglas (1970) have been derived not
from nature in the wild but from the human body - the symbols
drawn from the physical personality, with itsflesh,blood, breath and
orifices, ingestions and excretions. From such symbols drawn from
external object or human body differ again those in the more analyti-
cal category envisaged by S. F. Nadel. For him, 'natural symbols'
were those where there is a closely observable 'correspondence'
between symbol and thing symbolized. Illustrations were: an 'iconic'
symbol as when a sculptured representation of an animal serves as a
totemic emblem (a meaning somewhat akin to that of C. S. Peirce, v.
infra); when the symbol is also a sign, as when weeping in a play
indicates mourning; and when a symbol is dynamically expressive,
as when darkness symbolizes mystery (1951, 261). Of related order,
though less abstractly conceived, was Edward Westermarck's view
of a Moroccan marriage custom in which a bride threw a slipper at
her husband, presumably to get power over him. Westermarck saw
the slipper as a symbol of power because the association between
slipper and domestic rule 'is so natural' that such a ceremony may
well have had from its beginning the meaning ascribed to it (1914,
256-7). The notion of'natural' in the examples of Westermarck and
Nadel involves the assumptions of a cultural frame, but rests also on
bases of generally observed human behaviour in situations of grief,
disorientation or aggression. It embodies relations which may be
deemed logical in terms of general inferences from the given assump-
tions.
Carl Jung has also used the notion of 'natural symbols', but for
him these have their roots deep in the human psyche. One such
'natural symbol' is the Cross, representing the basic principle of
order or stability, as opposed to the disorganized or chaotic charac-
ter of a formless crowd; so regarded, the Cross itself may be treated
as a symbol for the human body, as the self in extension. Another
'natural symbol' is the mandala, which the unconscious is stated to
produce spontaneously, and which in Jung's view appears not only
in Oriental art but also in rose windows and other circular forms in
6o SYMBOLS
mediaeval Christian art, and even in a palaeolithic disc pattern in

Rhodesia. In Jung's view such, perhaps all, symbols are 'natural5
because they reach down to and express the unconscious in primi-
tive fashion at the same time as they correspond to the highest
intuitions of consciousness. But they can be seen as 'natural' in
another way, which Mary Douglas's view shares, of being related to
the human body. 'The symbols of the self arise in the depths of the
body and they express its materiality every bit as much as the struc-
ture of the perceiving consciousness' (Jung, 1958, 138; also 318-26
and 1926, 605-7).
But on the whole in studies of symbolism the tendency has been
to emphasize the lack of 'natural' links between symbol and thing
symbolized - to view the symbolic attribution as a matter of cul-
tural determination, as conventional or even as 'arbitrary'. What is
implied by such expressions is that the range of possible representa-
tions of something, particularly of an abstract quality, is so great that
no exclusive choice of symbol is normally feasible by someone out-
side the system. The reason why a specific symbol then appears in
use, seems to depend upon some form of cultural condition; at the
worst, since cultural components in the relationship of symbol to
object are often hard to identify, the choice is termed inexplicable.
But in stressing the conventionality, the 'arbitrary' character of the
relationship - and in using such a criterion in the distinction of
symbol from sign or signal - it is the complexity rather than the
inexplicable nature of the link that is really being considered.
A firm opinion on the question of conventionality was expressed a
century ago by Charles S. Peirce, a forerunner in the study of sym-
bolic logic. A physicist and astronomer in the United States Coast
and Geodetic Survey, Peirce was also a philosopher of rugged
original mind. Though he held an academic post for only a brief
period, through his published papers and correspondence he exer-
cised a growing influence as a pioneer in the modern study of semio-
tics. (He called it semeiotics, the act of signifying - roughly, the
theory of meaning; the term is said to have been used by John
Locke.) Though Peirce said of himself that he thought in quite a
different system of symbols to words - he wrote much on logic from
a mathematical standpoint - he also said that he approved of invent-
ing new words for new ideas. And as he had, or believed he had,
many new ideas his writings about the classification of signs are
strewn with new words. (One such is ideoscopy, describing and
classifying the ideas that belong to experience without regard to
their validity or their psychology - a term that might serve anthro-
pologists engaged in 'cognitive anthropology' or 'ethnoscience'.)
Dodging the new words and much of the detailed analysis, one can
still find in Peirce several important points about symbols, first put
forward in a paper in the Proceedings of the American Academy of
Arts and Sciences in May 1867. There he defined logic as the doc-
trine of the formal conditions of the truth of symbols, that is the
reference of symbols to their objects. Later he saw that this involved
inquiry into all branches of the general theory of signs. Firstly, for
such an inquiry Peirce distinguished between index, icon and sym-
bol* An index in his view was a sign directly related in fact to what it
signified. If a hunter in pursuit of a lion sees a certain kind of foot-
print in the sand, this is an index to the passage of his game. A
proper name, a symptom of a disease, are indices in Peirce's sense.
An icon for him is a sign that represents its object by resembling it -
which is 'determined by its dynamic objective by virtue of its own
internal nature'. Peirce cited a curve of a distribution of errors; we
might regard the statue of a lion as iconic in having its form and
proportions determined by those of the animal. A symbol on the
contrary was defined by Peirce as a sign determined by its object
only in the sense that it will be so interpreted - an allocation depend-
ing on habit, convention or agreement, or natural disposition of the
interpreter. Following our example, a lion is a symbol of bravery by
convention. But as Peirce admitted, the characters of direct relation,
resemblance and convention are not completely exclusive. So an
icon - broadly equivalent to image - may have an element of con-
ventional recognition, as we may easily see in some of the mediaeval
figures termed lions at the porches of Romanesque churches.
A second point of interest made by Peirce was linked with his
pragmatism - it was his view that the meaning of a concept could be
completely understood only by discovering just what sort of general
habits of conduct a belief in the truth of the concept would reason-
ably develop. This was a doctrine that truth consists in future
* Peirce, 1958, 391, 395, 402-3; (C. Harteshorne and P. Weiss (eds)) 1931-5, vols II and
V; see also A. W. Burks, 1949.
62 SYMBOLS
serviceableness for our ends. So he held that the essential function

of a sign is to render inefficient relations efficient - a view which
seems to be echoed to some degree in Malinowski's treatment of
symbolism (p. 141). Peirce argued that the function of the sign is not
to set these relations into action, but to establish a general rule or
habit whereby they will act on occasion. (Such a dynamic view of
symbols seems a preferable form of statement to that which talks of
symbols themselves acting - cf. Mary Douglas, 'The symbols them-
selves lash back at people . . . ' (1970, xiv).)
A third point made by Peirce was that since symbol depends upon
convention, habit or agreement, it refers not to a single instance
alone, as does an index, but to a general class of instances. So for a
lion to symbolize courage, there must be a general idea of the look of
lions as a class and of the nature of courage as a virtue.* (Yet even
with an index such as a footprint or a proper name, one may argue
that for these to be meaningful there must be some general ideas -
that animals leave prints, that some kinds of verbal noises can refer to
persons.)
But Peirce's classification of signs, though it might have been use-
ful for anthropologists, lay unused by them. (The first anthropo-
logical reference to Peirce's ideas I know is by E. R. Leach, 1957,
121-2.) Moreover, it left several broad questions open. Peirce's use
of the notion of habit or convention in the definition of symbol
offered a social component but did not take it far; in particular the
problem of how the habit or convention was to be discovered and
defined was not faced. Nor was it clear how far Peirce regarded the
construction of symbols as an arbitrary procedure, as nothing more
than the routinization of the 'natural disposition' of the interpreter
by common agreement. Jean Piaget has recently criticized Peirce for
not using a contrast between individual and social signifiers in his
classification, and indicated his preference for Saussure's trichotomy,
which does so (1971, 115^). But Piaget's own equation of symbol
with individual signifier seems unduly restrictive (78-9). My own
view here is that one of the main tasks of anthropology is the reduc-
* In mediaeval funerary monuments a dead knight is often portrayed with his feet on a
lion, in contrast to his lady, whose feet are on a dog, symbolizing fidelity (A. R. Myers, 1966,
194). With change in symbolic idiom, were lions or knights thought not to be so brave, or was
bravery thought to be less important - and ladies or dogs not so faithful or fidelity less
important?
tion of arbitrariness as it appears in symbolic allocation; and that
linkage of individual with social elements in symbolism is a necessary
part of this task.
In the confrontation between 'natural' and 'conventional', 'in-
dividual' and 'social' as criteria in the definition of symbols, ques-
tions have arisen as to how far differences in consciousness or inten-
tion have been relevant to the process of representation. Opinion
has varied on this issue.
Susan Stebbing, the logician, and S. F. Nadel, the anthropologist,
both thought that for their own purposes they had best be concerned
in symbols with conscious, even designed, representation. But where-
as Stebbing was concerned with definition, Nadel was concerned
with evidence-and I examine his view later (p. 170). Long ago
Stebbing pointed out ((1930) 1948,10, i n , 508) how in situations of
everyday life our senses are being constantly stimulated by a variety
of impressions, among which we have learnt to pay attention to
some as being specially significant because they are signs of some-
thing else in which we are interested. When one thing signifies
another, there is between them a connection enabling us to pass in
thought from one to another. Stebbing cited waving a flag as a sign
of high spirits. She argued that in some contexts it is not possible to
draw a hard-and-fast line between 'sign' and 'symbol'; but made a
working definition that a symbol is a sign consciously designed to
stand for something. On this reading, an example of symbol would
presumably be a new national flag expressly designed to be flown by
representatives of a country hoping to be recognized as a political
entity and admitted to the United Nations (cf. Chap. 10). Another
example comes even closer to Stebbing's meaning. In 1955 a priest in
the church of Pellestrina in Italy glanced at the ceiling and saw to his
astonishment one of the angels in a fresco waving a red flag. This
might have been taken in Stebbing's sense as a sign of high spirits on
someone's part. But a young painter who had been recently hired to
freshen the fading colours of the ceiling was soon afterwards sen-
tenced to eight months in jail for 'vilifying the religion of the State'
{Chicago Daily Tribune, 16 April 1955). It was presumably not for
being just a sign of high spirits but for being a symbol with a
consciously-designed political meaning that this severe punishment
was awarded.
64 SYMBOLS
But definition of symbol in terms of conscious design has difficul-

ties, chief of which is given by the conception of consciousness itself.
If it is interpreted fairly narrowly as being equivalent to power of
formulation, or at least awareness of the nature of particular ex-
perience, then much of dream symbolism would be excluded. If it is
interpreted broadly as any sentient activity, equivalent to mind -
'a sort of distillate of all sensitive, teleological, organized function-
ing', as Susanne Langer has put it (1953, 127) then the creation of
symbols becomes a very primitive human activity. Indeed, Langer
has argued that the symbolic transformation of all perceptions is an
endowment of the human brain which allows it to handle stimuli
which otherwise would threaten the survival of the metabolic pro-
cess. Some such assumption of a basic 'primitive' symbolic func-
tioning of the human mind seems common to many anthropologists,
including Malinowski and Levi-Strauss, and the presence of un-
conscious elements in symbolization seems to be also admitted,
though evidence for such views has rarely been systematically
sought.
Another angle of approach to the problem of conscious apprecia-
tion of symbols is to consider their relation to what the symbols
represent - their significata or referents.* It has been argued, as by
Langer, that symbols differ radically from signs by their greater
articulation and presentation of concepts. Hence a sign is compre-
hended if it serves to make us notice the object or situation it be-
speaks; a symbol if it makes us conceive the idea it presents, f
Perhaps a better way of expressing this is to distinguish the referents
by their relative simplicity or complexity, since concept formation
would seem to occur in both cases. If one is driving a motor car, a
red flag in the middle of a road usually means an obstacle ahead,
a clear sign to slow down - and this is a convention well understood
internationally. We don't usually call this a symbol; the referent is a
simple idea. But a red flag on a Paris barricade a few years ago, like
* The term referent, which in the semantic field goes back at least to Ogden and Richards,
seems to me preferable to the more ponderous significatum. Stebbing has pointed out that
referent has another meaning in the logic of relations, as used by Bertrand Russell for the
term from which a relation proceeds. But Stebbing's own term referend has not found general
currency, and no confusion in the use of referent seems to have arisen. (J. Whatmough (1956),
a linguist, uses referend.}
t Langer, 1953, 26. Langer has explained that following Charles Morris, she has sub-
stituted signal for sign in her later work.
that of the Italian fresco painter, stood for a complex, not very
specific set of ideas and actions. In an immediate general sense it
stood for a simple attitude - defiance of established authority. But
behind this, or side by side with it, were much more elaborate,
much more vaguely delineated ideas of moral values and political
freedom. It was a political symbol. Moreover, as with many symbols,
strongly emotional attitudes were mingled with intellectual con-
cepts. Here, sign (signal) and symbol can be differentiated, not by
the thing doing the signifying - the same red flag was sometimes
used on the barricade as had been used earlier to warn of a hole in the
road - but by the degree of complication of relationships, and their
quality, especially of emotion or sentiment in the thing represented.
This kind of distinction is quite useful for anthropologists. So, before
a Tikopia religious performance, one could see columns of smoke
rising from houses where ovens were being prepared to cook the
food which would be later presented as offerings in the temple.
These smoke columns were signs of the particular activity of cook-
ing; they were also signs of which particular social groups were
contributing to the ritual. But they were not symbolic in that the
smoke did not represent any complex set of ideas about communicat-
ing with spirit powers - as in L. H. Morgan's account of a burnt
offering among the Iroquois (cf. p. 106). In Peirce's sense, it was an
index to a technical and social activity. But when the chief stood up
in the temple with his bunch of leaves drenched in oil, and rubbed it
on the post, this was a symbol for a complex set of ideas about spirit
powers and socio-political and ritual relationships. It did not take
me as an anthropologist very long to learn that a smoke column on
such an occasion represented not merely fire but also an oven for
food, or even to identify the social group involved. But to compre-
hend the significance of rubbing the post demanded long and arduous
exploration of Tikopia ideas. Interpretation of symbols is usually a
much more difficult matter than interpretation of signs.
The variant mode of classification suggested by Charles Morris
(1945), of using sign as the generic term (as did Peirce) and dis-
tinguishing under it symbol from signal has received much support.
For Morris a symbol is a sign that is produced by its interpreter and
that acts as a substitute for some other sign for which it is synony-
mous - as it might be, the wearing of black by a widow is a symbol
66 SYMBOLS
of some other sign of mourning such as weeping. In Morris's terms,

all signs not symbols are to be called signals - as it might be, a red
flag in the road is a signal to a motorist to slow down. I think it
might be argued that in ordinary usage the prime criterion of a
signal is that it is a sign demanding or expecting action - that is, a
more highly focused response is implied. Hence the difference in
terminology between road signs, such as directional plaques, which
are there for the motorist to take advantage of if he wishes, and road
signals, such as traffic lights, which he is expected to obey.
The sign/signal and symbol dichotomy has other qualifications.
What about flags, when used for communication by ships-are
these signals or symbols? They are consciously designed to stand for
something else, as in some definitions of symbol, and this something
else can be an idea of some complexity, or a concept involving
complex operations. But the hoisting of a Blue Peter flag by a vessel
in port, indicating that she is to sail in twenty-four hours, is or-
dinarily described as a signal. This can be justified in terms of the
character of the operations, which though involved, are primarily of
a technical order. If the Blue Peter were to be described as a symbol
then one could think of it as evoking ideas of a much vaguer order,
about people sailing away, leaving loved ones behind, going out to
new experiences - a symbol of parting, a symbol of adventure. So,
signal tends to connote some precision of technical consequences;
symbol, a much more imprecise, open-ended sequence of events and
experiences.
In considering the meaning of a symbol the position of the inter-
preter, as stressed by Peirce, Stebbing and Morris, is very important.
In the interpretation of signals, it is vital for efficient operation that
fabricator and interpreter are using a common code. In the case of
the Blue Peter, for instance, the flag of blue with a central white
square is also the letter P in the International Code of Signals. It is
essential for action that those concerned should interpret this P as
meaning 'sailing soon* and not, for example, as 'paying off'! But in
the interpretation of a symbol the conditions of its presentation are
such that the interpreter ordinarily has much more scope for exercise
of his own judgement - the alternatives in the situation may be much
less circumscribed. He may be left to 'get out of it' what he can by
the fabricator of the symbol, who may be concerned primarily with
A Q U E S T I O N OF TERMS 6?
his own mode of expression. Hence one way of distinguishing
broadly between signal and symbol may be to class as symbols those
presentations where there is much greater lack of fit - even perhaps
intentionally - in the attributions of fabricator and interpreter.
An aspect of great importance in consideration of all signs,
including symbols, is their connectedness, with that of their re-
ferents, in series or systems. This is obviously so with many
technical signs, such as the signal flags of the international code
and in the distribution of colour bands in national flags (cf. p. 337).
This is a matter of which anthropologists are well aware, from their
functional and structural training, and I give examples later, in par-
ticular from the work of Audrey Richards, Claude Levi-Strauss and
Victor Turner. But there can be argument about the perception of
system, as a matter of principle and as a matter of empirical fact.
Susanne Langer has held that while language, spoken or written, is a
symbolism^ a system of symbols, art symbols do not constitute a
symbolism. If I understand her view correctly, it is that each work of
art is a single, unique, indivisible symbol, itself highly articulated
but not capable of being resolved into components which can be
reassembled in another form to make another work of art, according
to given rules. But granting the autonomy, the non-repeatability, of
each work of art at a primary level of creation, it would seem that
we can nevertheless speak of the symbolism, say, of Romanesque art
in the sense of a common body of interconnected themes, patterns,
interpretations, using a set of rules provided by the sources from
which the artists drew their inspiration - the bestiaries, the Church
Fathers, the Scriptures. Even with modern artists trying to evade,
surpass or destroy convention, there are at least common under-
standings, shared with a sophisticated public, if not about particular
meanings of forms, about the general aims of manipulation of forms
and the general kinds of response to be expected. In this sense no
work of art stands completely alone, but is related to others in a
system of judgement and interpretation.
The question to what degree any given set of symbols is to be
interpreted systematically is a matter largely of empirical investiga-
tion, but may involve assumptions about regularity. The interpreta-
tion of the symbolism of colours offers good illustration. It has often
been pointed out in Western contexts how light colours are asso-
68 SYMBOLS
dated with pleasure, dark colours with sadness,* while in mediaeval

Japan there was a most sensitive awareness to the relationship of
colours to one another according to circumstance. In modern
anthropology, Victor Turner has made very effective use of colour
relations in his studies of Ndembu symbolism. He has demonstrated
how in the Ndembu system white stands for milk, purity, health,
good luck; black for faeces and other grim things; red for blood,
maleness, danger-all in complex symbolization (V. W. Turner,
1967, 59-92). Stimulated in part by Jung's notion of'archetypes' or
original models, Turner has interpreted the white-black-red triad as
representing the archetypal man as a pleasure-pain process. There is
much in Christian history to support such a view. For instance, in
the Middle Ages black, suggestive of material darkness, was sym-
bolical of the spiritual darkness of the soul; it was incorporated into
ideas of the Devil - the Prince of Darkness; and witchcraft - the
black art (F. E. Hulme, 1899, 28-9). Red and white, representing the
Resurrection and Transfiguration respectively, as robes of Christ,
were opposed to purple or other dark colour before these cosmic
events (G. G. Coulton, 1958, 266, app. 17). Comparative ethno-
graphic evidence of such similar symbolic colour systems can be
adduced, f But the significance of the individual colours is to be
interpreted primarily in terms of the relations envisaged among
them in each culture, not in terms of a colour's universal value. In
the West, the black shirts of the Italian Fascists and the brown shirts
of the Nazis were intended as symbols of glory in contrast to the red
banners of revolution; the interpretation of their colour values, as
with White Russians and the Red Army after 1917, depended upon
which side one was on. As we now know well, for those for whom
'black is beautiful', white has too often been the colour not of purity
and the milk of human kindness but of violence and oppression.
Colour terms themselves may lose their physical referent and be-
come symbolic: in Western countries the 'Blacks' are usually brown
and the 'Whites' are dirty-pink. And among the Konso of Ethiopia
black appears to be auspicious and white inauspicious (C. R. Hall-
pike, 1971, 2 8 0 - 2 ) .
* For complex association of colour with mood see Rimbaud's 'Voyelles* and 'Bateau
Ivre.
f A general analysis of material from antiquity, the Middle Ages and relatively recent times
has been given by F. Portal, 1857.
A QUESTION OF TERMS 69
Oriental colour symbolism tends to be markedly more systematic

than does occidental, and to apply in many situations where Western
colour values are relatively free of symbolic interpretation. Occiden-
tal conventions of dress until recently followed a few broad rules:
bright colours were appropriate to young people and subdued
colours to the elderly; black for male evening dress, with almost any
latitude to women; white for confirmation and bridal dress, and black
or purple according to stage for mourning. But this rough pattern-
ing, though highly sanctioned for some specific usages, was not
developed at all precisely. By contrast, Chinese major colours were
traditionally associated with the major compass points, the seasons
and the fortunes, and these ideas were reflected quite far into dress,
theatrical masks and the crises of life. Some aspects of this have been
brought out in an interesting study of the significance of variation in
dress colour at Chinese funerals by Arthur P. Wolf (1970; cf.
Martin C. Yang, 1947, 43; and more generally, Soame Jenyns, 1935,
125-6).
An observer may see mourners at a Chinese funeral wearing gar-
ments embodying white, red or blue. White is popularly regarded
in the West as the Chinese mourning colour, but the symbolism is
much more complex. Red is the colour of joy, shown in fire-
crackers, good luck charms, bridal dress, and its opposite is either
black or white. So at the death of the emperor, formerly, all shops
covered their red signs with white or black, brides travelled in black
not red sedan chairs, and so on. But red is also prophylactic, and is
presented to ward off evil; it may also be incorporated into the attire
of friends and kin of medium relationship when in mourning. White
worn by mourners does not imply deep grief; it is a neutral colour
worn as a courtesy by those who owe the deceased nothing more
than a token of respect; it implies social equality. So brothers of the
deceased who have been living in the same household wear white in
respect; brothers who have separated and formed households of their
own wear white too, but add a patch of red to their white headbands
as a protective device. The colour blue is halfway on the scale
between red and white, so is midway between the extremes of joy
and grief. It is the colour of scholars' robes and gowns of minor
officials, expressing dignified joy, a mild degree of attainment, but
also not inappropriate to sorrow. Contrasted with red, blue carries
70 SYMBOLS
negative connotations; contrasted with white, it is a positive colour.

Now the mourning gowns of a man's great-grandsons, if he has
been fortunate enough to have such descendants, are always dark
blue. A limited degree of joy at a funeral is permissible as an expres-
sion of respect that a man has lived to witness the birth of great-
grandsons and so has repaid his parents and grandparents in filial
duty; so his youngest descendants wear blue to symbolize a re-
strained joy. As Wolf emphasizes, the meaning of a colour in the
Chinese symbolic field depends not only on its place in the
Chinese spectrum, but also on the use of it in particular social
contexts.
Such patterning of similarity and contrast has its analogy in what
Mircea Eliade has called the 'internal logic of symbols', though he has
used this concept in a rather different way, in reference to the inter-
pretation of sequences of magical operations. For instance from
somewhat haphazardly selected ethnographic data he has indicated
the symbolism of knotting and untying knots as means of defence of
the human personality (1969, 37, n o ) . Such examples remind us of
the insistence with which men seek to make rational pattern out of
non-rationally selected material in their particular society-pre-
sumably with the aim of maintaining social communication through
the possibility of predicting behaviour. (The powerful analytical
instrument of structuralism, developed with great refinement by
Levi-Strauss for extracting meaning from systematic relations of
symbols, is referred to in Chap. 4.)
In considering the nature of symbols the notion of representation,
of one thing 'standing for' another, needs further consideration. If
there is no overt identifiable relation of a direct kind between symbol
and referent, no 'natural' link, how does a symbol manage to 'stand
for' another object? Susanne Langer's answer in essence is that there
is a formal congruence of logical structures between them, a com-
mon logical form (1953, 27). Granting this to be so, the question of
why specific elements have been selected out of a range of possible
logical relationships, still remains. Moreover, without pretending to
any very subtle psychological or philosophical implications, one can
see that other named categories such as simile, metaphor, emblem,
image, allegory all share with symbol some quality of the representa-
tional operations of language, and formal congruence between sig-
nans and significatum. Symbol is not alone in the linguistic and

logical field.
As regards the problem of choice among possible alternative
logical structures of congruence, a plausible postulate is that this is
determined by factors specific to the experience, both cultural and
personal, of the actor. It is part of the job of anthropologists to
identify such factors as closely as possible, particularly in the cul-
tural field.
In exploring the problem of representation it is useful to examine
briefly the categories allied to symbol, even without attempting any
very formal distinctions. For working purposes we can adopt some
such usage as the following. (I use as an illustration for contrast the
concept of 'key'.)
Simile is a kind of comparison, indicating perception of a com-
mon, often abstract quality shared by two objects, with overt state-
ment. The design of a carpet may embody a key-pattern, suggesting
the wards of a key. Metaphor also makes a comparison, but im-
plicitly, making a direct transfer of idea and substituting one term of
the comparison for the other. So, the statement that the control of
Gibraltar is the key to the naval strategy of the Mediterranean tacitly
refers to the function of a key in controlling passage through a door-
way. Emblem may show no discernible direct relation to its object,
but may rely on some past association of ideas. The Catholic saint,
St Zita, who has as her emblem a bunch of keys, is said to have been
a trusted household servant and so presumably controlled locked
doors. Image, used much in literature and public affairs* with a
carry-over from pictorial art, is concerned primarily with forming a
mental entity, giving shape in the 'mind's eye' to a set of qualities
perceived in or attributed to the object. Wordsworth's reference to
sonnet may serve as illustration: 'Scorn not the Sonnet. With this
key . . . Shakespeare unlocked his heart.' Allegory is often described
as a kind of sustained metaphor in narrative form, but its essential
features include the representation of one event or series of events in
a detached form (literally, 'putting side by side') to allow of separate
consideration of the implications, commonly of a moral order. In
the Middle Ages, the works of Virgil, like the Bible, were allegorized
* Fowler's Modern English Usage (1965) indicates the almost mystical value which the term
'image' has acquired in public relations.
72 SYMBOLS
and assigned hidden meanings not originally intended. But an open

allegory, The Romance of the Rose, used personification of abstract
figures to convey first ideas of romantic love, then of social criticism
(C. H. Haskins, 1958, 106; F. Heer, 1962, 369). Pursuing the 'key'
theme, an allegorical use of it occurs in Fitzgerald's rendering of
Omar Khayyam's Rubaiyat: 'There was a Door to which I found no
Key' and 'Of my base metal may be filed a Key/That shall unlock
the Door '
Symbol has links with all these; but the directness or likeness of
relationship to its referent is muted or attenuated. So is the relation
between symbol and referent often apparently arbitrary.* The
associations in a symbol are also often broader and more complex,
but the action-trigger of the symbol may be more immediate,
through the emotional involvement it often entails. Symbol has been
contrasted especially with allegory, on such grounds. Arnold
Hauser, for instance, has regarded allegory as the translation of an
abstract idea into a concrete image which, however, is but one of a
number of possible expressions of the idea. With a symbol, he
has argued, idea and image are fused into an indivisible entity. A
symbol can be interpreted in various ways (whereas an allegory
usually cannot) and this variability of interpretation, this apparent
inexhaustibility of the meaning of the symbol, Hauser maintained,
* Cf. J. Whatmough: *A symbol is a surrogate. But all these surrogates have one
feature in common. There is nothing in the nature of things that gives them the meanings
stated; that is something we have given them, by agreement or convention, so that the symbol
acquires a certain arbitrary character' (1956,19). Like Peirce and others, Whatmough separates
symbol from sign or index on these grounds - *a sign has a direct relation to its object.'
In the whole discussion of symbols there is a reminiscence of the mediaeval concept of
analogy as put forward by Thomas Aquinas. We predicate of God qualities to which human
qualities bear some resemblance, but this resemblance is always accompanied by some dis-
similarity, since Godhead is inconceivable and inexpressible in human terms - such is the
argument (E. L. Mascall, 1949, especially Chap. 5; F. C. Copleston, 1961, 93, 112; David
Knowles, 1962, 263, 305). The analogical mode of discourse, with 'application of a concept to
different beings in ways that are simply diverse from each other and are only the same in a
certain respect' has appeared in some anthropological treatment of religion, notably in Evans-
Pritchard's treatment of the problem of symbols (1956, Chap. 5).
Another kind of frame for discussion of the problem of representation is given by E. H.
Gombrich, who uses the concept of synaesthesia, what he describes as the 'splashing over' of
sense impressions from one modality to another, in his inquiry into the nature of symbolism
in art (1968,311-14). Referring incidentally to Baudelaire's Correspondences, and to Rimbaud's
assignment of colours to the vowels, he examines the translation of sense impressions into
visual forms over a rich field. Focusing on the notion of relationships in a structural matrix,
he argues that what we learn from a study of symbolism is the elasticity of definitions in any
attempt to indicate the nature of reality by reference to types of representation (op. cit., 85,
93). See also his discussion of ritualized gesture and expression in art (1966).
is its most essential characteristic. So, the allegory is the expression of
a static, the symbol that of a dynamic process of thought, setting
ideas in motion and keeping them in motion. He saw the allegory as a
kind of riddle, to be solved; but the symbol as capable only of inter-
pretation, not of solution (Hauser, 1952, 897). This is from the
standpoint of art. But from the standpoint of religion a not dissimilar
attitude has emerged. Gershom Scholem, as part of his massive
study of Jewish mysticism, has regarded allegory as 'an infinite
network of meanings and correlations' but all within the limits of
language and expression. Symbol radically transcends the sphere of
allegory. If allegory is a representation of something expressible by
another expressible something, symbol is an expressible representa-
tion of something which in itself lies beyond the sphere of expres-
sion and communication, a hidden and inexpressible reality (Scholem,
1955, 26-7). Whether one wishes to carry the distinction to such
lengths, it is clear that the notion of symbol commonly carries the
meaning of a more complex, more profound representation than
does that of allegory.
Referring again to our example of key image, in Christian
iconography St Peter is represented by two keys, often crossed -
they are his emblem, contrasted with the emblems of other saints, as,
for example with the winged lion of St Mark. Historically, the
emblems have at times been in political opposition, as the winged
lion of St Mark, the emblem of Venice, faced the keys of St Peter,
the emblem of the Papal states, in struggle for power and territory.
Sometimes the opposition was mainly symbolical, representing for-
mal rather than substantial positions. But the Christian keys can be a
symbol of a more intense kind. As Matthew's Gospel has it, in
return for his bold acknowledgement of Jesus as Christ the Son of
the Living God, Jesus conferred upon Peter the keys of the kingdom
of heaven. This metaphorical act of reciprocity has been popularly
interpreted as giving jurisdiction to let souls into Paradise or keep
them out. But in Catholic theology the 'giving of keys' means
granting of authority, and in the light of canonical and rabbinical
parallels it is held that the critical verse of Matthew's Gospel means,
not that St Peter is to be the 'gate-keeper of heaven', but that Christ
will confer on him vicarious authority over His household on earth,
that is the church. The keys that Christ will give to Peter are the
74 SYMBOLS
keys of the kingdom of heaven in the sense that Peter's authoritative

decisions on earth, as given by his apostolic successor the Pope,
will bind men in their conscience; so their entrance into the kingdom
of God for all eternity will depend upon their acceptance of Peter's
teaching of the Gospel and his directions for salvation. So the keys, a
favourite charge in ecclesiastical heraldry and appearing in many
Roman religious paintings (for example Perugino's fresco in the
Sistine Chapel), can symbolize not only Jesus's assertion of His
divinity and man's belief in His power, but also more personally,
the anxious problem of the fate of each individual soul in the after-
life - especially as determined by the representatives of Peter.*
Many Protestant scholars now argue that this whole passage was a
late interpolation in the Scripture to support the claims of Peter's
successors in the church. But this means that the symbolism is just
carried a step further, into the field of church politics and rivalries.
All this about signs and keys and modes of representation is only
one way, and a compressed way, of looking for criteria in use of the
term symbol, especially anthropological use. It seems to me that for
anthropologists the definition of symbols has to be primarily opera-
tional. Granted their representational quality, what can be regarded
as the effects associated with their use? Functional and structural
inquiry is needed (giving these terms a neutral connotation without
implication about the integrity of society). Inquiry is needed, too,
into what I would call organizational aspects of symbolic behaviour -
the ways in which use of symbols relates to interests and aims of
groups and of individuals; and how symbols are involved when
decisions are taken in interpersonal contacts.
From this point of view, on the question of terms, I would
advocate following a treatment which recognizes sign as a general
category, and differentiates types within it by relative emphasis
upon criteria rather than by trying for completely exclusive ele-
ments. In relation to social behaviour, then, under the general head
of sign may be differentiated:
Index - where a sequential relation is inferred, as of part to whole
or precedent to antecedent, or particular to general.
* See, for example Matthew's Gospel xvi: 16-19; Donald Attwater, 1965, 274-5; New
Catholic Encyclopedia ('Keys, Power of), 1967; F. E. Hulme, 1899, 145. For common
sexual symbolism of *key' in dreams see Freud, 1945, 336.
Signal- where the emphasis in a sign is upon consequential action;
and a relatively simple sign is a stimulus involving response of a
more complex kind. Signal may be regarded as a dynamic form
of index. Signature is a personal index which may be treated as a
signal if it is to be followed by further action.
Icon - where a sensory likeness-relation is intended or interpreted.
Change of scale or motion or dimension may be involved, since
an icon is constructed as a physical or imaginative representation,
suggesting a referent by a complex combination of elements.
One can presumably admit in music iconic sound - as in Vivaldi's
Four Seasons - as well as in pictorial art an iconic visual image.
Symbol- where a sign has a complex series of associations, often of
emotional kind, and difficult (some would say, impossible)
to describe in terms other than partial representation. The
aspect of personal or social construction in meaning may be
marked, so no sensory likeness of symbol to object may be
apparent to an observer, and imputation of relationship may seem
arbitrary.
Some rather formal distinctions along these lines seem necessary

for anthropological usage, if only because popular phraseology
seems often to ignore them. When newspapers and other public
media of communication use the terms 'symbol' and 'symbolic' all
that is meant seems often to be only 'index' or 'representation'. 'It
may be symbolic of trouble ahead' that one United States presidential
candidate got almost equal news coverage from telling a New York
newspaper casually of his plans as another did from an expensive
national telecast - 'indicative' would have been adequate here.
Stories of poverty cited in a newspaper's annual appeal for the
needy towards Christmas 'symbolized' hundreds of similar cases -
surely in ordinary language this could be 'represented'. A very great
Soviet investment in offensive strategic weapons in 1971 was de-
scribed as being 'symbolized'' by the missile silos ('holes') that the
Russians had been urgently constructing throughout the year-
'demonstrated' would seem a more appropriate term, or 'hinted at' if
the purpose of the silos was not clear {New York Times, 6 January
1972, 5 December 1971; Newsweek, 1 November 1971). In other
expressions emphasis is laid on the disjunction of sign from object,
76 SYMBOLS
and the unreality or lack of worth of the former. The contrast of

'symbol' with 'reality' comes out in the description of something as
being 'merely' or 'only' a symbol. A pair of slippers made from glass
in Cologne when it was a Roman colony in the third century A.D.
were described as 'merely symbolic by a museum director because
they couldn't be used for walking; what they were symbolic of
was not made clear {New York Times•, 17 January 1972). At the
accession of Princess Margrethe to the throne of Denmark her role as
Queen was termed 'a purely symbolic one', implying other non-
symbolic aspects of the role which she would not be exercising
{New York Times, 15 January 1972). From an earlier source, in
reply to an assertion that de Gaulle was the only possible head of
French resistance during the war it was alleged that de Gaulle had
not a great following but 'only a symbolic value'. The devaluation of
the U.S. dollar at a time when it had been divorced from its gold
basis was described as 'a mainly symbolic concession' {New York
Times, 12 January 1972). While the difficulties of trade relations
between Ottawa and Washington were such that it would 'take
more than symbols9 to convince Canadians that the United States was
not out to improve its trade at Canadian expense {Time Maga^ne,
18 January 1972). And the concept of symbol as being equivalent to
'idea' is evident in a statement that 'the importance of China is
being transmuted from symbol to actuality by the increasing power-
lessness of the West in Asia' (Ross Terrill, Observer, 2 January
1972). In such examples the term 'symbolic' is almost equivalent to
'notional' or 'fictional'; the suggestion is that whatever be the feelings
generated by the 'symbol' the appropriate action would not be forth-
coming. Though not necessarily a reversal of a common sociological
position in regard to symbols, this does imply a separability of
feeling and action which can be important in symbol interpretation.
In the light of such expressions I consider the concept of symbol
more closely in an action setting, from an instrumental point of view.
SYMBOL AS INSTRUMENT
A symbol is 'a device for enabling us to make abstractions', but with

some end in v i e w - a symbol has instrumental value. For brief
consideration here I look at symbols as instruments of expression,
of communication, of knowledge and of control.*
Expression
As instrument of expression symbols are to a supreme degree tools
of the artist. It is not surprising then that in her presentation of her
theory of art Susanne Langer has described symbol as 'a word
around which this whole book is built' (1953, x). In many contexts
an anthropologist too meets the symbolism of art - poetry, dancing,
sculpture - in many different types of society, clearly expressing
values regarded emotionally and intellectually as important by the
people who assert them. The art of Western societies poses a prob-
lem of the expressiveness of its symbolism which is not present to
the same degree in the societies normally studied by anthropologists,
namely, the gap that tends to exist between artist and public. Here
what the symbols express may be elite values, protest values, interest-
group values. Sometimes the expression provokes and leads a social
reaction, sometimes it seems to be tolerated as a kind of'radical chic'
which saves other members of the society from having to act more
positively themselves.
The instrumental nature of a symbol as a means of expression is
especially clear with political and religious symbols. Flag, national
anthem, church painting, scriptural text, national dress, even style of
headgear can evoke powerful emotions of identification with a group
and be used as rallying points for group action. This can be negative
as well as positive; symbols are convenient objects of hate as well as
of devotion (see Chap. 10 on Flags). An example of external criticism
has been the Palace of Culture in Warsaw, a building thirty-eight
storeys high which was a gift to Poland from Russia after the war. It
was intended to have both aesthetic and political significance.
Twenty years later, it was described in an American report as 'an
empty symbol', not because it was untenanted, but because of its
'hideous' outmoded Stalinist style {Time Magaîne^ 16 November
1970), and possibly because it no longer represents to the Polish
recipients the same sense of solidarity as before. Architecture easily
* Questions concerning the primary character of the symbolic function in human thought,
such as have been considered by Claude Levi-Strauss as part of his inquiry into 'untamed
thought' in La Pens&e Sauvage and elsewhere, are not immediately relevant to this analysis.
78 SYMBOLS
becomes an expressive symbol because of its public character.

Another controversial building, the Yale Art and Architecture
Building, by Paul Rudolph, completed in 1963, became 'a symbol and
cause celebre in the revolution of the consciousness of man' (Ada
Louise Huxtable, New York Times, 12 December 1971). It was first
much praised by the critics; then it became 'a supersymboP when it
burnt down in 1969 after being much hated by some students as an
archetype of the imposition of false values; then in reconstruction it
has been described in glowing terms as a material manifestation of an
ideal of human culture-'a symbol of rebellion and revolution has
been rehabilitated' {New York Times, 4 January 1972). In such cases
the primary focus seems to be on the symbol rather than on the
values it is taken to represent.
But the expressiveness of symbols can be intensified (or diminished)
by circumstances. In the stresses of the last war the playing of the
British National Anthem used to bring tears to the eyes of many who
ordinarily treated it as rather silly verses set to uninspiring music;
they now saw it as a symbol of freedom of nationhood and freedom
of person largely lacking in Hitler's Europe, and under threat in
Britain itself. Similarly theatres in the United States began per-
formances with the 'Star-Spangled Banner', and gave rise to feelings
of good prevailing over evil and hopes for the future {New York
Times, 31 October 1971). In such conditions personal scepticism
may be suspended, in a kind of complicity in recognition of a £ublic
symbol, in favour of the propriety of collective action. I think this is
a factor which sociologists writing about public symbols, including
Durkheim, have tended to overlook. (Some further issues about the
relation of private to public symbols are examined in Chap. 6.)
The element of individual variation in expressive symbols may
be illustrated by an experience of my own. Some time after the war
Queen Elizabeth the Queen Mother was installed as Chancellor of
the University of London, and delegates from many European
universities were among those assembled to honour the occasion.
Their formal dress was an astonishing mixture, from black morning
clothes to academic robes of brilliant colour, with headgear of much
diversity. Aesthetically, there was appeal in the colourful array, but
there were comic combinations of ruffs and gowns and trousers and
odd doctoral hats. Yet when I saw the gathering rise at the entry of
the Chancellor, a lump came in my throat. As a group, in their
academic robes, they became for m e - a n d I would imagine for
others too - a living symbol of Universities risen again after the
dark days of suppression, a collective symbol of the search for
knowledge and of the history of learning over 700 years, having
kept the lamps of scholarship burning despite all manner of attack. A
private symbol, created for myself out of my own interpretation -
yet the materials for it were assembled independent of my volition.
From other people's behaviour one might assume that they too had
analogous symbolic interpretations of the occasion. The expressive
value of what is called a collective symbol may lie in fact in a set of
variations on a common theme rather than in any uniform conceptual-
ization.
Communication
A major function of symbols is in facilitating communication.
Utterance of words - a basic form of symbolic action - allows us to
dispense with many kinds of manual and bodily actions in providing
stimulus or conveying meanings (for Malinowski's views, see
p. 145). In a ritual field, performance of a symbolic act allows ideas
to be shared and reformulated without use of words, or with minimal
verbalization. When the Tikopia chief was rubbing his temple post
with scented leaves, this was a coded sign which both conveyed ideas
of respect to his people and gave them a focus for their own religious
acts. Symbols also serve as stores of meaning in communication. In
pagan times a Tikopia temple post was a permanent reminder of
religious and economic values; it commonly outlasted the lifetime of
an individual worshipper, and for successive generations stood as a
material symbol of an immaterial spiritual being. Among Christian
Tikopia the Cross serves an analogous function.
To turn to economic symbolization, in a monetary economy
coins or banknotes provide a store of public and private meanings.
They symbolize past achievements and transactions; they stand for
potentialities of acquisition; they can dramatize petty conquests of
desire through non-spending; they are reference-points for much
family conversation. In most cases it is the values or amounts rather
than the actual coins or notes which are the object of interest; it is
8o SYMBOLS
the implication of relationships symbolized by the material items,

not their physical form and substance which are significant.
But the actual coins and notes, as objects of design, have symbolic
significance in another way: they communicate information of an
historical kind about concepts of the role of the state or its leaders in
the world at large. These concepts are sometimes of a political order,
sometimes of a moral order. The coins and notes of the United
States of America, for instance, apart from indication of their value,
portraits of presidents, national monuments and other symbols of
solidarity, and supplementary data, carry two affirmations: E
pluribus unum, and In God We Trust* Both mottos must be regarded
as expressions at a high level of generality; what these money tokens
communicate is a historical decision to make a public profession
of faith. Contemporary British banknotes, also having historical
portrayals, carry no moral or religious statements. The coins,
however, are more illuminating. For the last twenty years, as well as
details of denomination and of design such as lion or Britannia, they
have carried a portrait of the Sovereign, with two pieces of religious
information. One, conveyed usually in very abbreviated form, such
as D.G.REG. affirms that she is Queen by the grace of God; the other,
as FID.DEF. or more familiarly F.D. indicates that she is Fidei Defensor,
Defender of the Faith. The first is a pious assertion, meaningful to
Christians and perhaps to believers in other faiths, but in a literal
sense meaningless to her many sceptical or atheistic subjects, who
may interpret 'the grace of God' as equivalent to 'the will of the
people'. The second piece of information is a statement of an historical
event - the conferment of the title of Defender of the Faith on
Henry VIII by Pope Leo X as a reward for a pamphlet Henry wrote
in defence of the Seven Sacraments against Luther. After Henry
himself assumed the headship of the Church of England this title was
withdrawn from him by the Pope, but he wished to retain it, and it
was confirmed to him by Parliament in 1554. Kept on false pretences
from a Roman Catholic point of view, the title of Defender of the
Faith is now assumed by many British people to signalize the role of
the Sovereign as head of the Established Church of England!
* I seem to remember that about twenty years ago only coins and one dollar bills carried
the legend 'In God We Trust*; notes of higher value made fiduciary reference only to the
Federal Reserve Bank and the United States Treasury.
A QUESTION OF TERMS 8l
What this example demonstrates is the multiple character of the

referents of this symbol. A set of marks on pieces of metal represents
an historical event and its sequel; a royal claim to entitlement;
a view of what is important to preserve on coinage in political terms;
a belief in the religious leadership of the Sovereign by many of her
people. In a sense, this FID.DEF. lettering may be termed a 'con-
densed symbol' in that to many people the referent is incompletely
known, or their premises in regard to it are mistaken.*
Symbols as stores of meaning help to cope with problems of
communication over time, aiding recall and obviating to some extent
a need for reformulation of ideas. As such they are a cultural asset.
Condensation, the encapsulation of many forms, or many meanings,
in one symbol by processes of contraction, suppression, trans-
formation, can also facilitate communication by giving a common
reference point for a variety of originally disparate ideas. But it may
hamper communication by clogging the channels-by providing
too many alternatives for interpretation. If a symbol is to be an
effective instrument of communication, it is essential that it should
convey much the same thing to people involved - or that the range of
variation in their interpretations should not inhibit the action desired.
How is this brought about? In such connection various writers have
called attention to the ambiguous notion of what a symbol 'means'.
'There are different ways in which symbols can mean,' said Stebbing
(1948, i4n). Even ignoring the epistemological problems here,
anthropologists are very aware - perhaps more so than logicians -
how many symbols in the wide range of societies they study are
socially produced, if personally interpreted; that is, the meanings of
the symbols are inculcated by social processes of learning - as by
imitation and participatory experiment. For anthropologists, then,
determination of what may be termed the level of meaning is a very
important part of our inquiry.
* The notion of 'condensation symbols' in anthropology is not very clear. For Freud,
condensation was a process of compression of dream-thoughts, not just by contraction and
omission, but also by fusion, suppression, substitution, transformation (1945, 269-92), so
that the manifest symbol was a disguise for the latent dream-thought. The meaning given by
Sapir (1934) to his 'condensation symbols' seems to have been close to Freud's. But Victor
Turner's conception seems to rest primarily upon the concentration of many meanings in a
single symbol - 'many things and actions are represented in a single formation'. So what
Turner writes of ritual symbols - that their 'simplest property is that of condensation (1967,
28-9) is not necessarily true of dream symbols - or perhaps of symbols in general.
82 SYMBOLS
The concept of level of meaning in symbolic behaviour can

operate in anthropological inquiry in both an implicit and an explicit
way. Take again the illustration of the Tikopia chief rubbing the
temple post. Explicitly he explained the meaning of this (symbolic)
act as cleansing and decorating the body of his god, represented by
the timber, on analogy with other decorative acts applied to the
human body, and submissively making an appeal to the god by
giving him pleasure. But implicitly the chief was also making a
demonstration of power. He chose the time to perform the rite;
he stood up to his full height in front of his seated followers in a
temple so sacred that normally one should go on hands and knees.
He scrubbed the timber in an aggressive way, emphasizing by his
forceful actions that it was his privilege to do this. Implicitly the
chief was showing initiative, and claiming control of a political order
by a series of energetic physical behaviour patterns of a coded
kind designed to secure benefit in the non-human sphere of crop
fertility and the human sphere of health. This is my way of putting
it as an anthropologist. But this implicit meaning of the symbolic
action could have been got from the chief himself by more round-
about explanation - as indeed I got it piecemeal. On the other hand,
there were different levels of explicit meaning which varied according
to the status and knowledge of the member of the group concerned.
Some people could give much more coherent, more sophisticated,
more syncretist interpretations of the symbols than could others -
a kind of 'inner' meaning, though overtly expressed. It is tempting
to identify such more esoteric meaning as the 'true' meaning of the
symbols. But each level is valid, and must be aligned with the others
for a developed analysis of the place of symbolism in social process -
as Audrey Richards and Victor Turner, for instance, have shown.
Knowledge
A proposition that symbols are instruments of knowledge raises
epistemological issues which anthropologists are not trained to
handle. That symbolization helps us to know cannot I think be
easily denied. But what comes to be known thereby is another
question. What the process of symbolic representation presumably
does is to abstract some quality common to both referent and symbol
and allow one to perceive more clearly, more imaginatively, a
particular type of relationship, uncluttered by details of the referent,
or reduced in magnitude to comprehensible dimensions. That
symbolization is a way of knowing beyond this, a mode of know-
ledge in itself basically different from other ways of knowledge, is a
view I do not share. Eliade has argued - like some of the Symbolists
(p. 31)-that the symbol reveals certain aspects of reality, the
deepest aspects, which defy any other means of knowledge (1969,
12). But I think that this, like other assertions of similar order, is to
be looked at in the light of the general aesthetic and philosophical
position of the speaker.
Most anthropologists tackle the matter rather differently. They
concern themselves more with the knower than with the known,
with the social position of claimant and claim rather than with the
question of the objective reality of what is claimed. Yet one strand in
the complex modern interest in symbolism - even among anthro-
pologists - is a hope of identifying 'real' underlying phenomena in
an increasingly confusing world. From particle physics to personality
disorders come suggestions that the 'inner knowledge is symbolic
in character'. I think the issue for anthropologists here is not one of
Truth or Inner Reality, but of spheres of relevance, and of effects.
As an anthropologist I am sure that I am not entitled to overlook
the social context of such claims. Assertions that symbols provide a
unique way of knowing the truth seem to be often equivalent to
defence-mechanisms. A powerful way of arguing that 'what I say is
true' is to assert that 'I have a unique way of getting at the truth
which is inaccessible to ordinary knowledge.' This has been the
route of the mystic in all ages. A claim that symbolization offers a
unique path to truth not only has no validity in itself; it invites
consideration of why such a claim has been made. As anthropologists
we are bound to consider such a claim in its social context if we are
to comment upon its position in the theory of knowledge. And what
anthropologists have done for almost a century, from the studies of
James Mooney on the Ghost Dance for example, is to attempt to
contextualize such claims as have come within their purview
(cf. Chap. 12).
Control
Consideration of symbols as instruments of control, or more bluntly,
84 SYMBOLS
as instruments of power, has two main aspects. Symbols may be

used for reference and support when conduct is called into question;
they are appealed to as repositories of values, without being actually
regulated by the authority of those making the appeal. But invoking
a symbol for justification - such as the Bible, the Koran, the memory
of a dead parent - can be a powerful means of affecting someone
else's behaviour. The other type of control is when a symbol is
under the direct authority of, or capable of being manipulated by,
the person wishing to affect the behaviour of others. In the struggle
for a people to attain political autonomy, the selection and use of
coloured materials to make a new national flag gives an important
instrument for focusing and mobilizing the peoples' behaviour.
The emphasis in this instance is upon control of external behaviour
by use of symbols, and anthropologists, following Wright Mills
(1961) have specifically examined the significance of political
symbols in power relations. An interesting treatment of symbols as
action-media in this way is given by Schlomo Deshen (1971) in a
study of the relation of religion to politics in Israel. In the course of
an inquiry into national elections he attended a ceremony at which
representatives of one of the political parties formally presented a
scroll of the Law to a new synagogue. With the growth of settle-
ment this has happened fairly often in recent years, and is a very
solemn occasion. It also means great credit for whomsoever makes
the gift. This particular gift concluded with an address by a repre-
sentative of the political party concerned, and he ended his speech by
reciting the central credo of the Jewish faith, the Hebrew verse
'Hear O Israel, the Lord is our God, the Lord is One' (ehad). The
last word of the credo is written in the scroll with large characters,
and its final letter, D, is therefore of great significance. The election
symbol of this political party happened to be also the letter 'd'
(daled). So the representative concluded, after reciting the credo:
'This is the D, the great D, the D of the party! Vote D!' This was a
piece of political engineering, linking political with religious sym-
bolism by giving a homiletic interpretation of the Hebrew character
on this solemn emotional occasion. From the Mexican field an
incisive analysis of the mediating role of traditional ritual fiesta
symbols, and of the changes that took place in these as a reflection of
political pressures, has been given by Paul Friedrich (1966). In the
community he studied, changing political ideologies, in particular
as expressed through the channels of agrarian politics, resulted in the
decline of some of the more 'religious' symbolic practices and sub-
stitution of others with less Catholic, less general and more secular,
more local meaning, with consequent reactions on social and econo-
mic behaviour of the people.
But there is a rather different sense in which symbols can be said to
serve as instruments of control - that is, as instruments 'for trans-
forming subjective experience' (Nancy Munn, 1969). Here the
emphasis is not so much upon the way in which recognition of
symbols affects overt behaviour, as upon the way in which it
transforms or conditions the intellectual and emotional framework
or basis from which that behaviour proceeds. This approach,
clearly related to the more general interests of modern 'structuralist'
study, has resulted in subtle analysis, as Munn has interpreted Murn-
gin boys' circumcision myth and ritual in terms of reciprocal social
relations of men and women and their joint submission to the authority
of Murngin society. The drawing of blood from men and painting
it on their bodies in the ritual identifies them with women's menstrual
blood in myth, the symbol of female fertility; but whereas in myth
women are swallowed by a snake (sexual intercourse) in the ritual
it is the men who are swallowed and the women who are preserved
by exclusion from the ceremonial ground. By such series of identifi-
cations, reversals and transformations, it is held, there is a coding of
experiences in symbolic form which serves to organize the indi-
vidual's relation to the society, and provide him with 'communalized
forms of identity'.
Such type of analysis is valuable in providing fresh hypotheses
about the dynamic function of symbols. It is also valuable in
focusing attention upon the significance of individual perception of
symbolic form and possible differences in reaction thereto. But as
Abner Cohen has pointed out in a suggestive general statement on
the analysis of the symbolism of power relations (Cohen, 1969)
there is a constant danger that such consideration of subjective
elements in the symbol field may fail to bring the analysis of thought
structure to bear upon the dynamic intricacies of social organiza-
tion. To put it another way, the description of subjective experience,
of thought and feeling pattern, is inferential, and should be supported
86 SYMBOLS
by systematic reference to empirical observed behaviour. And I am

willing to assert that this must be so, if anthropology is to maintain
its claim to deal with symbolic process on a comparative basis.
The use of symbols as instruments of control is widespread.
Manipulation of symbols may be termed a domestic industry, in
that it is to be found in every household, in relations between hus-
band and wife, parent and child. Even in the relatively austere sphere
of exchange of ideas in academic lecturing symbols may have a
controlling effect. The philosopher J. N. Findlay has commented
(Findlay, 1963, 112-13) on the effectiveness of argument: ' . . . the
emotions covered by the word "logical" are concerned with symbols
and symbol structures, and this to the extent that such symbols
yield us mastery over the detail offered by the senses, and recon-
stituted in thought. And we may say, further, that they concern
such symbols as yield us another sort of mastery: that over all those
minds who have access to the same empirical material, and can
thereby share the same reference.'
But symbols as instruments of power and control are most pro-
minent in the public domain, not least in the advertising industry.
In the public domain the identification of what Wright Mills has
called master symbols - to which Victor Turner's dominant symbols
are roughly equivalent - becomes more than an academic exercise.
Wright Mills pointed to the central place of master symbols in
social analysis: those in authority attempt to justify their rule over
institutions by linking it, as if it were a necessary consequence, with
widely believed-in moral symbols, sacred emblems, legal formulae.
These central conceptions may refer to a god, the Vote of the majori-
ty', the 'will of the people', the 'divine right of kings'. Social
scientists, following Max Weber, have called such conceptions
'legitimizations' or sometimes 'symbols of justification'; and Mills
indicated how other thinkers have used other terms - Karl Mann-
heim, ideology; Durkheim, collective representations; Lasswell,
symbols of authority (Mills, 1961, 36-7). Mills cast his net wide, not
differentiating between major symbolic concepts such as the crown
in a monarchy, or the apostolic succession in the Catholic church;
and relatively minor symbolic tokens such as those of class distinc-
tion, powerfully seductive and mandatory as these may appear.
But when he stated that the relations of such symbols to the structure
A QUESTION OF TERMS 8?
of institutions are among the most important problems of social
science, one can agree. He argued that such symbols do not form an
autonomous realm within a society; their social relevance lies in
their use to justify or to oppose the arrangement of power and the
positions of the powerful within this arrangement; their psychological
relevance lies in the fact that they become the basis for adherence to
the structure of power or for opposing it. However, like Abner
Cohen's argument that in social anthropology the central theoretical
interest in the study of symbols is the analysis of their involvement
in the relationships of power (Cohen, 1969, 218) this kind of reduc-
tionism ignores a whole range of problems concerned with men's
conceptions of their social order in moral and aesthetic terms, and its
relation to their conception of the natural world.
An example of the complexity of the problem of dealing with
'master symbols' is given by rites concerned with the British
monarchy. Any modern sociological analysis can begin with the
assumption that we are not involved in any concepts of the order of
the 'divine right of kings', and many of us would regard any state-
ment that the Sovereign reigns 'by the grace of God' as a merely
symbolic affirmation. Yet many of us would also concede that a
monarchy even in present conditions can have useful social and
political functions, and is therefore entitled to a moderate respect,
even though the monarch be quite a commonplace person. Many
anthropologists would probably also concede that to assist in the
maintenance of these functions, some public rituals are advisable,
to mark the monarch's public role. From this point of view Kingsley
Martin's comment on the coronation rites for Queen Elizabeth II
is of interest. He wrote (Martin, 1963, 117-18): 'If there was some
reaction afterwards that was because the claims made for the Monarch
and the significance read into the ceremony were extravagant. There
is a wide gap between enjoying a national holiday on which a
popular young Queen is crowned, and sharing in a "national
Communion service" which in the Archbishop's [of Canterbury]
words, is to bring the whole country close to the kingdom of
heaven.' Martin pointed out that the coronation was an Anglican
service, and only a few of those who enjoyed it were actually members
of the Church of England; none but the erudite could be expected
to gather spiritual benefit from its 'obscure excursions into the
88 SYMBOLS
legendary past'. Yet the ceremony was widely appreciated, despite

the lack of close religious identification, and despite some critical
opinions. What Kingsley Martin showed, in effect, was that 'master
symbols' of the type Mills described are rarely simple. They are
complex combinations of many elements, each with a different
emotional and intellectual charge, appealing in different ways to
different kinds of people, and often imperfectly. Hence 'their use to
justify or oppose the arrangement of power in a society' as Mills puts
it, cannot be necessarily taken as a uniform process, predictable in
mass terms. The coronation was part of the supporting mechanism
of the monarchy; it employed a range of traditional symbols, both
political and religious, to enhance the significance of the 'master
symbol', the monarchy itself, personified in the Sovereign. The
manipulation of this master symbol by the power hierarchy, in-
cluding the Archbishop of Canterbury, undoubtedly had the effect,
with many people, of reconciling them with an existing power
structure to a greater degree. Yet some people were alienated, by
what seemed to them needless waste of resources, or mystic jargon,
or pandering to public sensationalism. The symbolic relevance of
the ceremony was not exhausted by interpreting it simply in power
terms; it had aesthetic and moral interest even for many who rejected
the assumptions of power distribution built into it.
The social relevance of symbols may indeed appear to be split
on occasion between their power correlates and the more diffuse
relationships that they may stimulate through their aesthetic and
moral qualities. Charles Morris, writing on what he called the social
pathology of signs, pointed out how religious symbols which are no
longer adequate to a society may be used by a sub-group to main-
tain its own privileged status. Not only this, members of the society
at large may resist changes in the symbols because of the reassurance
these give as to the status and destiny of the society as a whole.
This is in line with Wright Mills's general sociological formulations.
But as the illustration of the coronation ceremony indicated, even
in the somewhat acid reference by Kingsley Martin, while the
symbolic behaviour could be interpreted in terms of a power-
conservation hypothesis, some of it was undoubtedly maintained
in satisfaction of deep aesthetic interests, not only of heraldic and
ecclesiastical specialists, but of a great body of people who delighted
in the drama and colour of the spectacle. Moreover, it may be argued
that the people were not really fooled. They tolerated the symbolism
of the ceremony because they enjoyed it and had already accepted
the Queen. Much of the symbolism found response in a set of ideas
and sentiments of a diffuse moral kind about British social rela-
tionships, family life and institutional patterns focused upon and
epitomized in the person of the young Queen - such appears to be a
fair inference from their behaviour.
This line of argument can be reinforced by considering the dif-
ferent interpretations sociologists put forward. Shils and Young held
that the coronation ritual was a demonstration of the way in which
the Crown kept British society intact. In their view the Crown
symbolizes the authority system of British society by representing
generally the hierarchy of values of the society, and specifically
the benign aspect of the governing elites (Shils and Young, 1953).
This view was strongly contested by Norman Birnbaum, who
argued that the very absence of shared values in Britain accounted
for much of the attention paid to the coronation. It provided, he
thought, a surcease from conflict, and the object of fantasies and
identifications akin to the cult of adulation built up around certain
film stars. In essence this was a difference of opinion about the
acceptance of a 'master symbol' at face value. Both parties agreed
that the symbol was accepted in the sense of attracting favourable
attention in general. To Shils and Young the coronation was inter-
preted by the public as a justifying symbol, representing the posi-
tively valued aspect of the power structure -'an act of national
communion'. But to Birnbaum it was interpreted as a diversionary
symbol, a distraction from the realities of the power struggle. I do
not think there is enough evidence to come to a firm conclusion on
this issue. But it indicates the difficulty of projecting an idea, even
when drawn from a vast array of material of literary and historical
kind, as the 'subjective experience' of people in contact with ritual
performance and other presentations of symbols. It also indicates
the strong possibility that different interpretations of the validity
of the ritual have been current in the society, not merely among
sociologists, and that some of these may have been very alien to the
conceptions of those in authority responsible for organizing the
ritual performance. Also, by comparison with the historical accounts
D
SYMBOLS
9o
of the behaviour of the populace at earlier British coronations, all

this suggests the strong possibility of the attitudes of the people to
its symbolic value having changed over time.
An indirect comment on this view was given by Thomas Carlyle,
who also discussed an English coronation, that of George IV about
150 years ago. This too had attracted criticism. Carlyle commented
that the hereditary Champion of England, whose symbolic duty
was to defend his Sovereign, and who therefore in theory should be a
powerful knight, was so infirm on this occasion that he could
scarcely mount his horse without help. Carlyle described him as 'a
Symbol well nigh superannuated'. But in more serious vein he also
remarked that while Time adds much to the sacredness of symbols
he also defaces or desecrates them. Symbols wax old, Carlyle said,
though poets can shape new symbols. But 'we account him Legisla-
tor and wise who can so much as tell when a Symbol has grown old,
and gently remove it' {Sartor Resartus). Yet this is still a grimly
difficult issue since not only legislators but also the people whom they
serve are apt to cling desperately to old symbols even when their
original referent has radically altered.
In raising the questions -'what is a symbol?' and 'what does it
mean?'-1 have tried to show the complexity of the issues involved.
Our concept of what a symbol is depends on our view of the nature
of reality. We can take the view that reality, if not an illusion, is at
least undiscoverable, and that we are operating therefore only with
symbols. Or we can hold that for much of our life we deal with
reality, in our relations with people and things, both mental and
physical, and that symbolization is a mode of operation which is
basic and ubiquitous, but not the sole mode of dealing with reality.
Its functions are those of convenience and simplification, of giving
scope for imaginative development, of providing disguise for painful
impact, of facilitating social interaction and co-operation. This
latter view is of the kind which I myself hold. But while I have
gained much from other students of symbolism, I find I still need a
personal lens through which to examine the phenomena. I find it
necessary for instance to challenge the concept that all social inter-
action is symbolic interaction, because I think this blurs the distinc-
tion between direct co-operation and representative co-operation.
On the other hand, I have reservations about such labels as 'master
symbols' or 'dominant symbols' because they seem to assume too
great a uniformity in reaction of the people concerned. I would
prefer to speak of 'emphatic symbols' and 'non-emphatic symbols'
with reference to their being a focus of attention and an object for
manipulation, without assuming that they necessarily exercise
control.
The next three chapters review briefly the statements and findings
of anthropologists and other writers of allied interest on symbolism,
in the light of these issues.
Chapter 3
DEVELOPMENT OF
ANTHROPOLOGICAL INTEREST
IN SYMBOLS
The title of this chapter emphasizes both continuity and change.

There is an impression that the study of symbols is a modern
development in anthropology. It is true that explicit anthropological
focus on symbolism is recent. But the subject as a whole has a con-
siderable history, going back a century and a half, to a time of
intellectual ferment and questioning of traditional modes of expres-
sion.
Over a century ago literary and scientific interest in symbols was
part of the general intellectual heritage. Thinkers about the human
condition tried to distinguish ways in which men might have repre-
sented phenomena from the phenomena as they appeared to ex-
ternal observers. There were several components in this. There was a
need felt to separate truth from semblance (often called fiction).
There was recognition that some underlying realities demanded less
stark, more attractive, or more easily comprehended modes of ex-
pression. There was also a sense of historical change, a feeling that
certain forms of institution had survived the disappearance or altera-
tion of their content. And there was some excitement in puzzle-
solving, in seeing present-day forms as clues to the practices of a
vanished past. What came to distinguish the anthropological treat-
ment of symbols from that in theology, art or philosophy, especially,
was the cool analytical treatment of a wide range of comparative
instances specifically sought for their coverage of institutionalized
variation, particularly from fields of magic and religion outside
ordinary Western knowledge. Then and later, major anthropological
interest was in identifying objects or actions as symbolic, not literal,
explaining what they meant, and looking for systematic ways in
which such symbols were formed or transformed. In interpretation,
analogies were important; the meaning of objects or actions in one
92
DEVELOPMENT OF ANTHROPOLOGICAL INTEREST 93
culture was explained by their associations apparently present in

another. But the analytical development was at first shot through
with other strong interests.
S P I N - O F F FROM THE R O M A N T I C M O V E M E N T
Basic notions of symbols and their practical significance were not

unknown to the moral philosophers of the eighteenth century.
Adam Ferguson, for instance, in discussing the monarchy, pointed
out how the Sovereign owed a great part of his authority to the
'sounding titles and the dazzling equipage' which he exhibited in
public, and subordinate ranks owed much of their importance to the
constant display of 'the ensigns of their birth, or the ornaments of
their fortune'. These things marked out to individuals the relation
in which they stood to others, served as distinguishing features of
the various social ranks, and helped in the preservation of order in a
society otherwise disunited by ambition and interest (Ferguson,
1967, 104). However generally this be couched, it would be sur-
prising if an Edinburgh intellectual - and a Highlander - had not
been viewing somewhat ironically here the appeal of a Hanoverian
ruler to a Scots populace. But Ferguson did not turn his attention
directly to the subject of symbols in any general way. The temper of
the Enlightenment, that belief in the rational perfectability of man-
kind, was not much interested in symbolism.
It was as a phase of, or as related to, that complex set of challenges
to eighteenth-century rationalism known as the Romantic movement
that interest in symbolism began to develop overtly and systemati-
cally. The Romantic movement was by no means a mere sentimental-
ism. It had many diverse aspects. It included: a disbelief in ordered
progress and a feeling for change, even of a revolutionary order;
an interest in conflict and violence, in nature and in man; a return,
as it was seen, from the complexities of man's society, exemplified
especially in urban conditions, to the purity and simplicity of nature,
especially in the countryside; a refusal to accept the conventions of
society, and a renewed exploration of the values of the self; a rejection
of statements and solutions made on the basis of pure reason and a
recognition of the importance of feeling and emotion; a questioning
of the outward-seeming and a search for inner meanings. For such
94 SYMBOLS
conditions and interests, new symbols were required. They were

sought in man's inner invention, as in poetry, in myth and in dream,
or in the objects and forces of external nature.* A new interest also
developed in language. Novalis, one of the more influential figures in
the German Romantic field, expounded a symbolist position about
language quite strongly, in terms which call to mind some issues
concerning magic formulae. He held that if poesy, fantasy, is
capable of creating life it is on the condition of finding again the
dynamic and magical power of language, or more exactly, the Word,
of which the suggestive power by rhythm, cadence, harmony,
will be able to create the symbols of higher reality. Nature as a
whole, he said, could be looked upon as a vast symbol; and all,
between spirit and matter, is only correspondences. A symbol is
only the representation of another order of reality; a symbol pro-
vokes a spontaneous activity (Michaud, 1970, 23-4).
Towards the end of the eighteenth century many writers were
preoccupied with problems of the self in relation to society and the
external world generally. Among internal modes of expression of the
self, dreams became one focus of attention. The significance of
dreams as premonitory had long been accepted or been a matter for
inquiry, but now more positive views of their relation to the self
were being developed. The remarks of Lichtenberg, for instance,
indicate this, however briefly. Georg Christoph Lichtenberg,
physicist, rationalist with a mystical residue - Albert Schneider has
said he had 'a sort of nostalgia for belief 5 -had a keen analytical
mind. He was interested in the production of images in dreams, and
raised the problem of dream definition. The dream is a part of human
life, and dreams lose themselves in our waking life; one cannot say
where the waking life of a person begins. 'One can just as well dream
without sleeping as sleep without dreaming,' he argued. He held that
whatever dreams might reveal about the future, they had a present
significance. 'From the dreams of men, if they themselves reported
them exactly, perhaps much could be concluded. However, not just
* A 'nature symbol' of minor anthropological interest was the concept of primitive man as
'Noble Savage'. This picturesque idealized figure was not without its critics; but was taken
to represent simplicity, sensitivity and allied virtues, fresh, sincere and unsophisticated love,
freedom, mystery and passion in poetry. See the exhaustive study by Hoxie Neale Fairchild,
1928. For general treatment see, for example B£guin, 1939, and Sorensen, 1963, as guides in a
vast literature.
one, but quite a few would be needed.' 'When people tell their
dreams properly, that will allow their character to be divined sooner
than from their faces.' 'Dreams can be of use, in that they produce
unbiased results about our whole being, without the compulsion
of often feigned reflection. This thought deserves to be taken to
heart.' 'A dream often alters our resolution, assures our moral
foundations better than all teaching, as it goes by a roundabout way
into the heart When in dream I argue with someone, and he
refutes me and enlightens me, it is I who enlighten myself; so I
reflect. This reflection has come under the form of conversation.' It
is clear from such statements that Lichtenberg realized the expressive
symbolic function of dreams, and regarded them as having an 'auto-
diagnostic' value (Lichtenberg, 1764-70, in Grenzmann (ed.), 1949,
92-3,102-9; Beguin, 1939,14). Karl-Philipp Moritz had some analo-
gous views. Grappling with the rationalism of the Berlin school,
disturbed by interior conflict, a wanderer across the face of Europe,
a novelist and a friend of Goethe, Moritz tried through the study of
dreams and other symbolic images to come to terms with his own
inner tensions. Dying of tuberculosis in 1793 at the age of 36,
Moritz was much concerned with the prophetic quality of dreams,
and impressed by his disagreeable memory of them, since they
created disorder in his daily thoughts. Afraid of dreams, he never-
theless advocated exploration of them in order to understand better
what goes on in ourselves. A character in one his novels - specific-
ally termed 'psychological'-questions reality in a solipsist way by
virtue of dreams. Since his dreams were very realistic and vivid,
could he be dreaming in the full light of day? Could the people he
saw around him be purely creations of his imagination? Such sugges-
tions of the merging of dreaming and waking life, of the confound-
ing of inner sensation with external reality, clearly link with
Lichtenberg's views-and could be taken as forerunner to some
present day anthropological approaches to experience. More broadly,
Moritz's second major novel Andreas Hartknopf (1786) with its
sequel (both with portrait of a sphinx on the title page) have been
claimed as essentially symbolic in quality, in the way in which he
used the phenomena of nature to describe aspects of his self-analysis.
But Moritz himself described the novel as an allegory, and did not
generally use the term symbol (Langen, 1962; Schrimpf, 1968,
96 SYMBOLS
20-4). (Allegory and symbol had been contrasted in a wider

reference by Goethe and Schelling - Sorensen, 1963, 249.) Much of
the early discussion of dreams was not couched in the idiom of
symbolism, which did not assume overt importance until the Roman-
tic movement was gaining ground, and which was especially reserved
for the treatment of myth.
Towards the end of the eighteenth century and in the early part
of the nineteenth century a very lively intellectual interest in the
symbolism of myth became manifest. One of its mainsprings was a
reinterpretation of Greek mythology, by Clarke, Ernesti, Heyne,
Hermann, arguing that the classical gods, and some Homeric heroes
too, were symbolic representations of natural forces or abstract
principles, in a kind of primitive philosophy. The chaining of a god
meant strife between the elements; Apollo and Artemis were sun's
rays and moon's rays, signs of fertility, part of the age-old genius
of nature; the torch of Ulysses was the light of his wisdom, reflecting
in another form the golden lamp of his protecting goddess Athene,
the symbol of her clarity of understanding and so on. But there
were dissenting voices, or at least alternative treatments. K. P.
Moritz published a general account of Greek and Roman myths,
mainly for young people, which became a favourite school text-
book and ran into many editions. In this, myth was treated as a
development of folk belief, as a product of fancy, which should be
considered in its own terms, and should not be dissolved into
mere allegory, or made the subject of premature historical inter-
pretation. The symbolic themes were muted, and any grossness
eliminated.*
Very different was the treatment of Jacques Antoine Dulaure,
author of works on phallic cults and fetishism, first published in
1805 and reissued in 1825. Born in 1755, Dulaure had been trained in
architecture and engineering, but became involved in the politics
of the French Revolution, as a Girondin, anti-aristocrat but not
extremist. Forced to flee to Switzerland, after his return to France he
was an opponent of Bonaparte, but survived to the age of 80, to
* His Gotterlehre oder Mythologische Dichtungen der Alten, first published in 1791, had
reached a tenth edition by 1861. Already by 1830 there was a translation of the fifth edition -
'with improvements* - into English, published in the United States. This was entitled
Mythological Fictions of the Greeks and Romans, and was praised in an Advertisement (by
S. H. T.) issued with it, for the 'chasteness of its language' and its 'pure and elevated feeling'.
write various historical works.* In his work on symbolism he struck

a blow for behavioural studies. He criticized his predecessors for
having been much more attached in their interpretations to the
mythological fables than to the practices of the cult, to the idol than
to its attributes, to the personage of the role than to the actor. And,
one of their gravest errors, they considered only as symbol what in
its origin had been the divinity itself. In his own treatment Dulaure
emphasized the grossness, the materialism, of the classical cults. He
stoutly defended his study of phallicism, on the grounds that we
should be prepared to accept the facts as they are - we pronounce
without shame the names of instruments of death, such as dagger,
poison; we should not blush to utter the names of the instruments of
life! He argued that the priests of old had allegorized the great opera-
tions of nature to render them more venerable to the ordinary wor-
shippers. The classical myths were not the result of wise theories or
the discoveries of early natural science; the philosophical
commentators had given currency to primitive errors. A proto-
anthropologist, Dulaure adduced considerable data from India
and America to back up his contention that cults of the type
he described were well-nigh universal. His basic assumption, that
men in an early state of civilization were incapable of forming ela-
borate abstract ideas of the forces of nature and of rendering them
symbolically, was a hangover from the Enlightenment; it cannot be
* Dulaure's work seems little known. His name does not appear in the usual histories of
anthropology, nor (any more than that of G. F. Creuzer) in Slotkin's volume on early
anthropological materials. But an elaborate study by Marcellin Boudet, historian of Auvergne,
Les Conventionnels d*Auvergne: Dulaure (1874), gives much biographical detail. According to
this it was Dulaure's History of the Beard (Pogonologia, see infra, p. 266) which inspired him
with a taste for historical studies. Dulaure had a stormy political career, typically espousing
the Girondin cause only when it had been defeated. He was a deist, a Theophiloanthrope, who
rejected revelation and Catholic dogma, though he believed in the existence of God, the
immortality of the soul, and moral virtue. According to Boudet, Dulaure's research into
primitive religion was pushed forward by his interest in understanding how the Catholic
church, which he hated for its alliance with aristocracy, had survived the revolution - he
looked to discover a human explanation.
A new edition of Divinites Generatrices was published in Paris in 1905, with an editorial
explanation that it was issued in order to help put Christianity in its place among the diverse
religions which share the world. The book was to be regarded not as anti-Christian, but a
work of science though 'fort curieuse aussi'. It was also stated that it was originally conjoined
with Dulaure's other work on cults under the title of Histoire abregee de differens cultes in
order to deceive the harsh censorship of 1825. A complementary chapter by A. van Gennep to
bring Dulaure's work up to date did little more than give information about American and
Australian sex customs and interpret them in terms of sympathetic magic, relating to cor-
respondence between man and nature.
98 SYMBOLS
supported. But he did put forward an empiricist viewpoint which

is still relevant, even if it was largely a protest against symboliza-
tion by interpreters.
Moreover, Dulaure had some interesting general ideas about
symbolism. He said: symbols are not at all purely natural objects,
but are works of art. Let us define what we mean by symbol - it is
ordinarily the image of a representable object, which has evident
relationship with another object that cannot be represented. An
image is a work of art; hence a symbol must be so. More than that:
to conceive a symbol, one should presumably have perfect know-
ledge of the object that one wants to symbolize - without which the
symbol cannot be exact. This knowledge presupposes at first the
need of a symbol, then following, a certain instruction, the art of
combining ideas and of appreciating the relations which exist
between objects strange to each other. He argued that this need, this
illumination, these operations of the spirit do not belong to people
who are still savage. Then comes the point of his contention: when
certain writers have said of barbarous peoples that in their adoration
of sun, moon, mountains, flowers they considered these things as
symbols they have committed an error. What these writers have
done, he held, was to have loaned their ideas and their knowledge to
men who could not have had them because they were much more
practical (1825, I, viii, 24-6). In particular, Dulaure was critical of
his contemporary Dupuis's view that the most primitive form of
religion was worship of the heavenly bodies (Sabaism). Arguing
with his usual pragmatism that man worshipped the stars because he
needed them for his agriculture, Dulaure held that a cult of fetishes
preceded Sabaism, with an interest in the material objects for their
own sake. But man was a hunter before he was a cultivator, which
meant search for religious origins in Africa and America rather than
in Chaldaea. Dulaure was thus an interesting bridge between the
empiricism and rationalism of the Enlightenment and the symbolism
of the Romantic movement.
But there are still several problems about this period. If, as is
often said, symbolism was at the base of the Romantic movement,
in what relation to each other did they first emerge? I have not found
any clear account of these matters. But it would seem that a number
of disparate threads were involved: scholarship in the classics,
interest in philology, concern for historical and cultural connections,

curiosity about mental functioning, philosophical and aesthetic
explorations - all in a setting of a more speculative, more adven-
turous temper than before.
A very influential contribution specifically on the theme of
symbolism now followed, from (Georg) Friedrich Creuzer, Pro-
fessor of Philology and Ancient History at Heidelberg. In 1806,
with his Idee und Probe alter Symbolik he began a vast series of
studies, which in his own view were not intended to be of a philoso-
phical order, but rather philological and historical. Indeed, Creuzer
seems to have been responsible for much of the elucidation of the
historical meaning of the word symbol which appears unacknow-
ledged in later writers. Creuzer's treatment was distinctly compara-
tive, including material from India and Persia as well as from Egypt,
Greece and Rome. He also specifically entitled one section of his
huge Symbolik und Mythologie der alten Volker (begun in 1810,
with a second edition in two volumes in 1819-22, and a third edition
in four volumes, 1837-43) 'ethnographic observations' on the old
heathen religions (of Persia etc.). I find it impossible to summarize
Creuzer's general treatment of symbolism, but a few of his leading
ideas can be indicated. In his earlier work on Greek historiography,
which Momigliano has regarded as the first modern history of the
subject* Creuzer tried to set empirical history within the frame-
work of the new idealism, and his whole work on symbolism
gives the impression of wishing to weld together a concept
of the pragmatic significance of symbols as instruments of
learning with the idea of their mystical religious expressiveness.
Momigliano has described his work as an attempt to give a scientific
basis to the Neoplatonic interpretation of Greek mythology, but it
seems much more than that; apart from his involvement in the ways
of thought of the Romantic movement, Creuzer made a real attempt
at comparative generalization, and also put forward some pertinent
if provocative conclusions about the analogies of Christian festivals
with the course of the seasons in nature, and with Judaic and heathen
festivals based upon the annual cycle (Creuzer, iv, 72 5ff).The breadth
of his treatment is indicated in the title given to the French trans-
* Arnaldo Momigliano, 1946, 152-63. Momigliano holds that Creuzer did most valuable
work as an editor of Neoplatonic texts.
100 SYMBOLS
lation of his work — Religions de VAntiquite consideree principalement

dans leurs formes symboliques et mythologiques (Guigniaut, 1825-41).
In his general analysis Creuzer devoted much attention to the
many meanings of symbol and its allied forms found in Greek
literature, and he pointed out that writing about symbols specifically
could be traced back to Greek times. He cited Proclus to the effect
that those who speak of things divine by means of sensible signs
intuitively express themselves either in symbols and in myths, or in
figures (simple images). But those who announce their thoughts
'without veils' do it either by scientific method or by inspiration from
the gods. And he further cited the view that the exposition of divine
things by the way of symbols is orphic and proper to the authors of
'theomyths', whereas that which uses figures is pythagorean.
Creuzer himself went on to construct a diagram in which he dis-
tinguished these various modes of intuitive and discursive exposition,
with symbolic, mythic and iconic (mathematic) under the first head,
and scientific and inspirational under the second (Creuzer, iv, 497,
502, 503-17). Starting with the notion of symballein as meaning
basically bringing together that which has been parted, Creuzer
tended to emphasize constantly the unitive notion of symbol until it
reaches the theological idea of sacrament. He saw the symbolic as the
root of all imaginative expression, and traced it through a variety of
forms in art and ritual. And while he argued for the scientific signifi-
cance of symbols as means of clarification, he linked them very firmly
with mystical and religious expressions, and with myth. Indeed, he
puts as one of the conclusions of his study of religion and philosophy
the return and re-insertion of mysticism and symbolism into mytho-
logy and their continuing mastery there, at the same time as he
speaks of the 'pragmatic art of treating myths' (Creuzer, iv, 665-6,
670). I cannot claim to have mastered all his thought, but I see him
as trying to say that while symbols are dark and mystical by nature
(537) by their many-sidedness they help to clarify even more obscure
aspects of the relation between spirit and matter. In the Reformation
the symbolic as 'expression of religion's secrets' tended to disappear
(687), he argued.
But Creuzer's handling of symbolism, particularly his attempt to
demonstrate the influence of Oriental symbolism upon Christian
symbols, soon plunged him into controversy. A Protestant, his
DEVELOPMENT OF ANTHROPOLOGICAL INTEREST IOI
defence of symbols got him accused both of Catholicism and of

paganism. He was also accused of overestimating the value of
Oriental cults and of ancient mysteries - a controversy of which the
ripples were still perceptible a century later. In particular, Johann
Heinrich Voss, a scholar of an older generation, objected so strongly
both to what he regarded as maltreatment of the classical texts
generally, and to some of the religious implications of Creuzer's
work, that he published an Antisymbolik. In this, among much else,
he complained of the 'foreign-sounding labels' of myth and symbol
which had been attached to the classical tales, and to the transmuta-
tion of them into units of philosophical discourse, the 'philoso-
phemes' (a term which seems to have escaped modern itemizers).*
The pursuit of symbolic forms continued. In 1814 Gotthilf
Heinrich Schubert, nature philosopher and dramatic writer, pub-
lished Die Symholik des Traumes. This continued the interests of
Lichtenberg and Moritz in dreams but also combined philosophical
views of Herder and Saint Martin on the nature of man, and inter-
pretations of myth and poetry by Creuzer and A. W. Schlegel,
with Schubert's own somewhat mystical views on the character of
unconscious images. Much of the book was not about dreams as
such, but about a range of manifestations of the human spirit in
poetry and myth. But essentially what Schubert was arguing was
that the dream and its allied forms constituted a language of images
(Bildersprache\ unique in that it did not need to be learnt as ordinary
language - the prose of the conscious - must be. The dream then is
opposed to ordinary speech and with its related myth is the oldest
original tongue of man - a view which seems to have some echo in
ontogenetic theory in some branches of modern psychology. So
this dream symbolism can be a basis for new experiments in poetics.
Linked with this was the notion of art, including manifestations such
as poetry, as hieroglyphic! - a kind of symbolic mode in which the
* Voss's Antisymbolik (two vols, 1824-6) was published soon after Creuzer's second edi-
tion appeared. For fairly recent discussion see Ernst Howald, Der Kampf um Creuzer's
Symholik: Eine Auswahlvon Dokumenten, 1926. A denigrating reference to Creuzer is given by
P. Gardner — 'the fancy world of Creuzer, who traced the influence of the mysteries every-
where on Greek vases and Roman reliefs' {Hastings Encyclopaedia of Religion and Ethics,
New York, Scribner, n.d., article on 'Symbolism').
f There was an,interest in hieroglyphs at this period; the mediaeval idea that they were
mystic symbols still had currency, but their decipherment by de Sacy and Akerblad had
begun, and a few years later Young and, definitively, Champollion published their results.
102 SYMBOLS
spirit of man (or in some of Schubert's expressions, the mind of

God) declared its inmost quality. So dreams are an abbreviated
hieroglyphic form of speech (cf. Freud, 1952, 183, 209, 241),
corresponding more closely to the nature of our spirit than ordinary
words do; the images often go in much faster succession than do
those of speech, often reversing the situation of things as we know
them. On the one hand, such images are linked directly with the
system of nerve ganglions, Schubert thought; on the other hand,
they relate to our moral culture, and serve to awaken the soul to its
true character and needs. This book produced by this 'peaceable son
of the starry night', as one of his friends called him, was muddled,
and at the bottom was a Christian apologia. But as Albert Beguin
says (1939, 107), it remains the most original of all the theoretical
works consecrated to the Romantic myth of the Dream, because it
'looked into the inner world', its author sought in himself the solu-
tion of the 'universal enigma'. The book went through several
editions, but it is symptomatic that the anthropologist E. B. Tylor
made no reference to it in his discussion of oneiromancy, the sym-
bolic interpretation of things seen in dreams (1873, I, 121).
A little later attention was paid to the concept of the unconscious.
Already by the early part of the nineteenth century the contrast
between conscious and unconscious mind had been drawn, as by
the educationist J. F. Herbart (1824); now it was developed, in
particular by Carl-Gustav Cams. Physician, painter and romantic
philosopher concerned with the nature of personality, Cams held a
view of the world which was a variety of pantheism. He opened his
work Psyche (1846), on the history of the development of the spirit,
with the announcement: 'the key to the understanding of the nature
of the conscious life of the spirit lies in the region of the unconscious'.
He argued that whereas the conscious is an individual attribute, the
unconscious is a subjective expression for a general characteristic,
designating what objectively we recognize under the name of
Nature. Cams distinguished between a relative unconscious,
those aspects of the conscious mind in temporary retreat; and an
absolute unconscious, which remains inaccessible to consciousness.
As Beguin has indicated (1939,45,135,141), these correspond quite
closely to the notions of personal unconscious and collective un-
conscious put forward by Jung. For Cams, a dream was the activity
DEVELOPMENT OF ANTHROPOLOGICAL INTEREST IO3
of the consciousness in the soul returned to the sphere of the un-

conscious, and enfeebled by sleep. In the interpretation of dreams he
postulated an interior equilibrium disturbed, producing a sentiment
which evoked an image akin to a poetic symbol and by which the
interior betrayed itself (Cams, 1846, 218). But in Carus's view the
participation of the unconscious in a kind of pantheistic universal
life gave it a relation with near and far, past and future, which allowed
the dream to bear a predictive value - a mystical conception which
preserved the symbolic significance of prophetic dreams. In his
study of the symbolism of the human form {Symbolik der menschli-
chen Gestalt, 1853) Cams was less insightful. He was here concerned
with symbolism in the sense of the clues that the variant characters of
the human body might give to mental disposition and spiritual life,
and much of his treatment seems superficial; he also drew some un-
warranted inferences about racial psychology from racial physical
characters.* It is in his more direct psychological reflections that his
substantial contribution lies.
THE C O N T R I B U T I O N OF B A C H O F E N
The early interest in symbols and their meanings had been wide-
spread among intellectuals; but its specific developments had been
primarily on the part of classicists, allied with theologians and philo-
sophers. But the field of ideas expressed in myth attracted scholars of
many kinds, and took on a more definite comparative aspect. An
outstanding example here was J. J. Bachofen, a jurist by training and
a pupil of the very influential Savigny. From the ordinary view-
point of the history of anthropology Bachofen is usually regarded as
the author of a highly speculative, generally discredited work on
mother-right which nevertheless has the merit of having opened up
* A later work by Cams on 'life-magnetism' and magical operations generally is not of
great interest anthropologically (1857). Among his many commentators Johannes Volkelt
(1873) places him with Edouard von Hartmann as one of the few philosophers who had given
clarity of expression to the notion of the unconscious, and stressed its importance, especially in
its biological relationships; Christoph Bernoulli (1925) argued in similar strain. Erwin Wasche
(1933) held that while preoccupation with the unconscious was a mark of nearly all the
Romantics, Carus was perhaps the first to be so outspoken about it in non-poetic fashion.
Wolf Farbenstein (1953) wrote of Carus as a forerunner of Freud, while Artur Krewald
(1939) in a publication for the Wehrmacht Psychology centre, tried to bring out his interest
for racial studies.
104 SYMBOLS
the whole subject. But Bachofen was also, indeed primarily, con-
cerned with symbolism. He regarded himself basically as an historian
of antiquity, dealing with a period before the emergence of written
history, and therefore dealing with a different kind of material,
namely myth. He saw no breach of continuity between myth and
history (a position akin to that which for other reasons, Levi-Strauss
has advanced against Sartre). He stressed the importance of myth as
a guide to the truth of past conditions. But, he held, to understand
this guidance and grasp the historical meaning, myth had to be
analysed in its own terms; the scholar had to comprehend the spirit
of the times in which it was created. For Bachofen, the key to the
interpretation of the ancient myths was the theme of gynaecocracy,
the rule of women. Here lay the significance he attached to the idea of
symbol. He had been much impressed when in Italy by the care which
the ancients had given to the monuments of their dead, and in 1859
he published his Versuch uber die Grabersymbolik der Alten. In this
he interpreted the symbolism of the grave monuments as a most
ancient cult of fertility. He thought it presented the image of the
creative force residing in the earth, with the phallus, the symbol of
fructification, as a common design. He entitled his study of mother-
right 'an inquiry into the gynaecocracy of the ancient world, accord-
ing to its religious and legal nature', and he carried his symbolic
interpretations very far in pursuit of these aims. Even anterior to the
phallus Bachofen saw the egg (depicted on various monuments)
as primal, the beginning of all creation. In religion the egg was the
symbol of the material source of all things. The phallic god himself
springs from the darkness of the maternal womb; he stands son to
feminine matter; the bursting shell of the egg discloses the mystery
of phallic masculinity, hitherto hidden within it; the egg is the symbol
of matrimonial consecration. Bachofen also identified all bull forms
appearing in the classical mythology as symbols of maleness, of the
life-awakening side of the power of nature (1897, 39; 1925, 130).
He went into detail on the symbolism of the use of hands and fingers,
as in counting, and tried to relate symmetry and asymmetry of hand
use to indications of femaleness and motherhood (1925, 173-4). In
all this, though Bachofen did not apparently draw much directly
from Dulaure and Creuzer, he was following a recognized scholarly
tradition, but he gave it his own particular flavour.
He seems to have pursued his archaeological inquiries more widely

than did his predecessors, and his thesis of the primal significance of
mother-right was more sharply conceived and more seminal for
anthropology. Moreover, with Bachofen the Romantic tide came to
a flood. In his study of symbolism he went beyond the classical
conventions of his day in making much more articulate abstract
statements about the nature of his subject (1925, 46-8). He argued
that human speech is too poor to convey all the thoughts aroused by
such basic problems as the alternation of life and death and the sub-
limity of hope. Only the symbol and the related myth can meet this
higher need. The symbol awakes intimations; speech can only ex-
plain. The symbol plucks all the strings of the human spirit at once;
speech is compelled to take up a single thought at a time. Into the
most secret depths of the soul the symbol strikes its roots; language
skims over the surface of the understanding like a soft breeze. The
symbol aims inward; language aims outward. Only the symbol can
combine the most disparate elements into a unitary expression
Together with such almost lyrical statements Bachofen also developed
a position about the relation of symbol to myth. 'Myth is the exegesis
of the symbol. It unfolds in a series of outwardly connected actions
what the symbol embodies in a unity The quiescent symbol and
its mythical unfolding represent speech and writing in the graves;
they are the language of the tombs/ Historically, when he writes of
'the simple old symbolic faith, in part created and in part transmitted
by Orpheus and the great religious teachers of the earliest times' and
sees this 'resurrected in new form in the mythology of the tombs'
Bachofen's arguments do not stand the test of modern scholarship.
But from a literary point of view they have offered stimulus. He
argued that thoughts, sentiments, mute prayers can't be expressed
in words; only the symbol, by its calm and immutable silence, is
capable of making them come forth. When he speaks of 'the mys-
terious dignity of the symbol, which so eminently enhances the
solemnity of the ancient tombs' or pronounces the dictum: 'Words
make the infinite finite, symbols carry the spirit beyond the boundary
of the infinite, becoming, into the realm of the infinite being world',
one can realize how Bachofen might appeal to some German poets of
the 1920s as a kind of ancestral figure in a symbolist movement. By
some scholars (for example Alfred Baeumler) Bachofen was termed
io6 SYMBOLS
'the mythologue of romanticism' and H. G. Schenk (1969, 34, 43)

sets him quite simply among the Romantic writers, with a sophisti-
cated nostalgia for the past, and feelings of foreboding for the future.
Yet Adrien Turel has defended him from this charge, arguing that it
was the Germans who made Bachofen into a'Romantic5, and that the
turbulent times in which he lived-born in 1815, in Italy for the
revolution of 1848, and surrounded by unrest even in his native
Basel - were enough to account for the elements of stress and con-
flict by which Romanticism was identified in his writings. But I have
the clear impression that Bachofen was an enthusiast by tempera-
ment. The comparisons sometimes made between him and Nietzsche,
stimulated by the latter's Swiss experiences, and even Howald's
blunt, unfair characterization of him as 'a half-scientific dilettante'
(1926, 28) - Momigliano's 'erratic genius' is kinder - seem to me to
have some foundation. His views on symbolism, remarkable for
their time, are distinguished for their imagination rather than for
their sober scrutiny of evidence in a critical spirit. Hence, perhaps,
there is some basis for the opinion that he was a precursor of Carl
Jung. It was here that from an anthropological point of view some of
his contemporaries and immediate successors were more fruitful.*
PARALLEL INQUIRIES
Quietly proceeding parallel to the theoretical inquiries of Bachofen
and less anthropologically-oriented thinkers about symbolism were
the empirical studies of symbolic behaviour by the more perceptive
ethnographers. A good example of this is given in the work of
Lewis H. Morgan, who about the time Bachofen was studying
grave monuments in Italy was paying some attention to the sym-
bolism of sacrifice among the Iroquois. He did not call it symbolism,
but he was concerned to explain what he termed the 'true principle'
involved in a well-known Iroquois rite, the burning of a White Dog
at the celebration of the New Year, in mid-winter. A dog was
selected, pure white if possible and free from blemish, and ritually
* There is an extensive literature on Bachofen, much of it of little anthropological relevance.
I have drawn here mainly upon his two major works: Versuch iiber die Grdbersymbolik der
Alten (1859) 1925 edn; and Das Mutterrecht: Eine Untersuchung iiber die Gynaikokratie der
Alten Welt (1861) second edn, 1897; and upon the writings of Alfred Baeumler 1965, E. K.
Winter, 1928, Adrien Turel, 1938,1939. See also C. A. Bernoulli, 1924 and Joseph Campbell,
1967.
strangled, care being taken not to break the bones or shed the
blood. The dog was then spotted with red paint, ornamented with
feathers and white wampum (or in more modern times, vari-
coloured ribbons), hung up, and finally burned. Morgan argued that
earlier explanations of such sacrifice, in terms of expiation of sin, or
transference of guilt to a scapegoat, as among the Hebrews, were
erroneous. He explained that the burning of the dog had nothing
to do with the sins of the people, since Iroquois had no recognition
of atonement or forgiveness of sins - once done, an act was regis-
tered beyond the power of change. What the sacrifice represented
was a much simpler idea - to send up the spirit of the dog as a
messenger to the Great Spirit to convey thanks for the harvest and
other benefits of the year. For this a dog was peculiarly appropriate
because of his fidelity to man and his companionship in the hunt.
White was the emblem of purity and faith. The ornaments placed
upon the dog's body were voluntary offerings, each being regarded
as a gift for which the spirit would make return. The spirit of the dog
was believed to ascend in the flames, and the Iroquois used the
dog's spirit 'in precisely the same manner as they did the incense of
tobacco, as an instrumentality through which to commune with their
Maker' (Morgan (1851) 1962, 210-21).
I find several points of particular interest about this account. One
is the succinctness with which various points of theoretical interest,
including that of the reciprocity theme of offering, have been
incorporated into what is primarily a descriptive account. Another
is that the account clearly rests firmly upon empirical data, some at
least from personal observation. (Morgan mentions once in Feb-
ruary 1846 counting nine different-coloured ribbons on a white dog
hung up at Tonawanda by Senecas - though he does not say if this
was the only occasion he saw the rite.) Again, Morgan was cham-
pioning a communion theory of sacrifice, though he did so in con-
crete not abstract terms, and not involving commensality, as did
Robertson Smith more than a generation later. And in a way which
foreshadows Radcliffe-Brown's treatment, Morgan adopted a com-
parative interpretation of symbolic behaviour: he equated the flaming
consumption of the dog with that of the smoking of tobacco - using
'the incense of tobacco' - and saw them both as filling the same
ritual role as instruments of communication.
io8 SYMBOLS
In strong contrast to early anthropological studies of symbolism

by empirical observation of exotic rites or by pursuing the meaning
of images on classical vases and monuments - both of which could
be said to be in a Romantic tradition - was the impersonal analytical
exploration of the nature of symbolism in the mathematical and
philosophical tradition. As yet there seemed to be no rapprochement
between a general theory of signs and a theory of symbols. The work
of C. S. Peirce in particular (Chap. 2) attracted no attention from the
scholars interested in more literary approaches - partly perhaps
because of different channels of publication, but partly also because
of an apparent lack of an adequate social dimension for stimulus in it.
In the anthropological field, the social dimension of symbols con-
tinued to be developed on a more systematic basis. In opposition,
overt or implied, both to the romantically-oriented study of specific
symbols as universal types of imagery, and to the neutralist study of
symbols as abstractions, was the interest in symbolic behaviour of
a more pronounced institutional kind, and with a more specific
cultural definition. The difference was not clear-cut at first. Recovery
of the early history of the human race continued to be a preoccupa-
tion of McLennan and Tylor as well as of Bachofen and his pre-
decessors of the Romantic movement. But the Romantic movement
itself had been complex, assuming a variety of national forms. The
early Romanticism of the latter part of the eighteenth century had
been characterized by increasingly conscious repudiation of ideals
of rationalism and urban civilization, and by a growing questioning
of established bases of authority. It would be a distortion to hold
that the French Revolution was a translation of the idea of the
Noble Savage from sentimental literature into politics, but as Fair-
child has pointed out (1928, 140, 363) the revolutionary cult of
reason and the revolutionary cult of emotion were allied. Yet the
developments of the Revolution, leading to some shift of interest
from fixation upon nature to fixation upon man as symbol of hope,
also produced some disillusion, some preoccupation with interior
rather than with exterior states. And towards the mid-nineteenth
century, as empirical knowledge of exotic peoples accumulated, and
as interest in the time process of life on earth became systematized
in a plausible theory of evolution culminating in Darwin's authorita-
tive statement (1859) the Romantic treatment of symbols declined.
So the latter part of the century saw more careful consideration for
the immediate ethnographic context of symbols, more interest in
what the people who used the symbols might say about them, more
concern for a possible range of interpretation of a particular item of
symbolism. While curiosity about symbols was not necessarily
abated, it was taking a more disciplined form - and it was tending to
be satisfied within a broader framework of social studies, in which it
assumed a subordinate place. So to Bachofen's successors, mother-
right was more important as an institution than as a symbol. This
more mundane approach appears in the work of McLennan, and
even more so in that of Fustel de Coulanges and Tylor.
EXPLORATION OF INSTITUTIONALIZED SYMBOLS
Like Bachofen, John Ferguson McLennan was a lawyer by training,

but of a very different temper. His study of Primitive Marriage
(1865, reprinted in Studies in Ancient History', 1876) was one of the
first explorations of the symbolism of a specific social institution
from an explicitly comparative point of view. In this study McLennan
held that the chief sources of information regarding the early history
of 'civil society' were firstly, the study of races in their primitive
condition, and secondly, the study of the 'symbols' employed by
advanced nations in the constitution or exercise of civil rights. In this
he presumably had an eye on Adam Ferguson's work of a century
before. For McLennan, a symbol in this sense was essentially a
practice which had survived in form though its substance had
radically changed - a relic which indicated what the former custom
had actually been. As a lawyer he was much impressed by what he
termed 'legal symbolism', which he thought owed its origin to
reverence for the past. He regarded these legal symbols as 'fictions'
in that they did not correspond to the reality of behaviour. 'All
fictions, or nearly all, have had their germs in facts; they became
fictions or merely symbolical forms afterwards' (1876, 9). McLennan
argued, rather like Bachofen, that 'the symbolism of law in the light
of a knowledge of primitive life is the best key to unwritten history*
(1876, 6); and he found in marriage customs a most spectacular set of
materials for his thesis. The archaic Roman practice of coemption,
by which apparently there was a mutual fictitious sale of the spouses
no SYMBOLS
to each other, he classed as a 'legal symbol' but his interest was

attracted especially by the so-called 'marriage by capture'. He said:
'In the whole range of legal symbolism there is no symbol more
remarkable than that of capture in marriage ceremonies . . . nor is
there any the meaning of which has been less studied' (1876, 13).
Since McLennan's day it has become clear that the practice of 'cap-
turing' a bride is not in itself a form of marriage; and that its symbolic
significance lies not in the representation of a past custom, but
especially in presenting a formal expression of relationships between
kin categories and social groups involved.*
The influence of Fustel de Coulanges on the development of
thought in sociology and anthropology has been much more exten-
sive in the long run than that of McLennan, and possibly more
penetrating theoretically fhan that of Tylor. In the specific field of
study of symbols, however, his contribution was on the same
general level. A classical historian working with some archaeological
materials but mainly with literary sources, he demonstrated with
great elegance the significance of religion for the social and political
institutions of the Greeks and Romans. In contrast to many of his
predecessors - and to many other people with a classical education,
even to the present d a y - h e envisaged these ancient societies as
essentially exotic, not offering the direct models which people have
often tried to derive from them. This 'foreign' character was shown
especially in their religion, the basis of which was belief in and wor-
ship of dead ancestors, with the sacred fire on an altar as their major
symbol. The symbols of this ancient religion became modified in
time, as the gods became personified and the sacred fire of the altar,
* E. A. Westermarck, as his great History of Human Marriage developed (first edn, 1891)
first thought that marriage by capture was due to an aversion to close intermarriage over-
come by force. He wrote of it as 'either a reality or as a symbol' (second edn, 1894, 387-8).
But later he saw it as a custom expressing the sadness and grief of the bride at being parted
from her home and kin; representing - in a sham way - and accentuating antagonism between
the different social groups and reinforcing the solidarity of each; and representing also an
antagonism between the sexes (fifth edn, 1921,271-5). Clay Trumbull (1894, see later) was of
the opinion from personal observation that the element of 'capture' in Arabian marriage
ceremonies was not a 'survival' token of former violence, but a natural response of a groom to
overcome the reluctance of a bride who had not been consulted about her choice of a spouse
and shrank from the surrender of herself to a stranger. This immediate reaction may well have
formed a component in the struggle in some cases, but often the contest seems to have been
formal. For further discussion see E. Crawley, 1902, 367-70; R. R. Marett, 1933, 86-9;
Raymond Firth, 1936, 530-63 (from an empirical point of view); E. E. Evans-Pritchard,
1965a, 14.
DEVELOPMENT OF ANTHROPOLOGICAL INTEREST III
already 'a sort of moral being', was given a divine personality

((1864), 1955, 30-3,120). Like McLennan, Fustel de Coulanges saw
the marriage rite with simulated seizure of the bride as a symbol. But
he did not regard it as a token of past hostility, nor a recognition of
the modesty of the bride; he saw it as a religious ceremony, arising
from the need to transfer the bride from the jurisdiction of her own
family sacred fire to that of her husband, where from now on it
would be her duty to worship by his side. Fustel de Coulanges's
specific interpretation may not be valid, but in this characterization
of the symbolic significance of the pretended seizure he had grasped
the essential sociological point of the function of formal transfer of
the woman between groups.
The role which Fustel de Coulanges ascribed to religion in ancient
society has something in common with that given to it more
generally by Karl Marx, though as a mirror image. Like Marx,
Fustel de Coulanges saw religion as providing a basic order for
family, authority, class, and property relationships, giving them
pattern and consecration. But he treated this as a positive achieve-
ment, not as a means of fixing chains still further on an exploited
people. It is true that Fustel de Coulanges regarded the plebeian
revolutions as a redress of social justice, but he saw this attained in
the acquisition of a religion, not in its overthrow by the deprived
masses ((1864), 1955, 305-6). And in the end Christianity was vic-
torious because it occupied itself with the duties of men, not with
their interests; law, politics and morals had been freed from the
thraldom of the old religion, but religion itself was strengthened by
the divorce. In ideals if not in form Fustel de Coulanges was closer
to Tylor.
Edward Burnett Tylor is commonly regarded, with justice, as the
father of modern cultural anthropology. Considering the signifi-
cance that is now attached to the study of symbols, it is interesting
to see how Tylor treated the subject. In one sense, all Tylor's work
can be regarded as a study of symbols. Indeed Paul Bohannan,
introducing a new edition of Tylor's first major work Researches
into the Early History of Mankind (1865) calls it 'a book about the
history and processes of symbolization'. Evidently, if culture be
defined as in some modern views as a system of symbols, then the
two-volume Primitive Culture (1871) could be renamed 'Primitive
112 SYMBOLS
Symbols'. Yet Tylor is ancestor rather than immediate progenitor of

what is now coming to be known rather esoterically as 'symbolic
anthropology'. Bohannan himself seems to have realized this, since
he says that Tylor's 'mistake' (though later he admits Tylor was not
'wrong') was to tackle a problem such as that of images and names in
terms of 'processes of the human mind' instead of in terms of sym-
bolic representation, learning theory and social structure (1964, xi-
xii). If one does not try to cram all Tylor's work into a symbolic
box, but be selective along the lines of his own treatment, one sees
that he used the term symbol in a fairly restricted sense, took the
category for granted, and let his treatment of it arise out of his more
general studies. While he has some enlightening remarks about
symbols, he does not write much specifically about them. In con-
sidering why the treatment he gave was so much more measured than
that of some of his predecessors, for example Bachofen, one must
consider his background. As Radin has pointed out in an introduc-
tion to Primitive Culture, Tylor not only came from a Quaker
family, he also grew up along with the Benthamites and philosophi-
cal radicals of mid-nineteenth-century England; he was interested in
the same kinds of problems as Darwin and Spencer, Henry Maine
and John Stuart Mill. Born in 1832, he was growing into young
manhood when the impact of evolutionism struck English intellec-
tual circles, and he wrote as an historian, a humanist, and a rationalist
of broad views. He was not in tune with the Romantic movement of
the Continent, and so while many primary sources of travellers,
missionaries and voyagers are cited among his authorities, as well as
classical authors, few general works appear, and he has no mention
of the romantic explorations of symbolism by Creuzer, G. H.
Schubert or Bachofen. Tylor held that 'the turning of mythology to
account as a means of tracing the history of the mind' was a branch
of science scarcely discovered till the nineteenth century. When he
reviewed the ideas of 'older mythologists' to show through what
changes their study had at length reached a condition in which it had
scientific value, he did so in very general terms. He criticized Francis
Bacon for interpreting the classic myths of Greece as moral alle-
gories. Bacon makes Perseus symbolize war, said Tylor, and when
of the three Gorgons he attacks only the mortal one, this means that
only practicable wars are to be attempted. Tylor said that this was
DEVELOPMENT OF ANTHROPOLOGICAL INTEREST II3
only making one myth out of another - as any of us might do

according to fancy, by remote analogy. If political economy hap-
pened to be uppermost in our mind, we might with due gravity
expound the story of Perseus as an allegory of trade: Perseus is
Labour, and he finds Andromeda, who is Profit, chained and ready
to be devoured by the monster Capital; he rescues her, and carries
her off in triumph. To know anything of poetry or mysticism is to
know this reproductive growth of fancy as an admitted and admired
intellectual process; but sober investigation must rest on more
stringent canons of evidence, use the body of information now
available, be comparative in its survey. 'Scepticism and criticism are
the very conditions for the attainment of reasonable belief* (1873,1,
277-8, 280). 'The secret of mythic interpretation', almost lost, was
recovered from Aryan language, literature and folk-tale by the
brothers Grimm and by Max Miiller. With such guides, Tylor saw
primitive peoples displaying a myth-making faculty, with belief in
the animation of all nature as a foremost factor, rising at highest
pitch to personification. Animism looked at from an evolutionary
point of view was thus for Tylor the key to an understanding of the
symbolism of myth.
As part of his inquiry into the development of religious ideas
Tylor dealt extensively with the symbolic connections involved in
magical and religious rites. Regarding religious rites as in part
practical in intent - 'as directly practical as any chemical or mechani-
cal process' - he also saw them as in part 'expressive and symbolic
performances'. As if almost in anticipation of Edmund Leach's view
of ritual, Tylor wrote of rites as 'the dramatic utterance of religious
thought, the gesture-language of theology...' (1873, II, 362). A
search for the meaning of apparently obscure, bizarre items of
behaviour, which often turned out to be symbolic, infused much of
Tylor's inquiry. I think it probably no accident that his first sub-
stantial chapters in his prefatory book Researches into the Early
History of Mankind (1865) were devoted to the interpretation of the
near-symbolic gesture-language or the Language of Signs, as he
called it.*
* Symbolic or near-symbolic according to definition. The study of visual and aural signs as
substitutes for spoken language is not one of my primary concerns in this book. For an
interesting contribution, subsequent to Tylor, see for example W. P. Clark (of U.S. Army),
Indian Sign Language (Philadelphia, L. R. Hamersley, 1885). On the so-called 'Talking
114 SYMBOLS
But Tylor was not content merely to study the equation of symbol
and referent - to find out that what a represented was b; he also
wanted to know what kind of relation was believed to exist between
them. He was much concerned with religious images - 'idols'. He
pointed out that eidulos - 'the visible' in Greek - has come to be
restricted to images of spiritual beings. He argued as others have
done before and since,* that such an 'idol' enables the savage to give
a definite existence and personality to vague ideas of higher beings,
which his mind can hardly grasp without some material aid (1878,
109). As Marett put it much later, primitive thought 'finds it hard to
divorce the intelligible from the visible' (1935, n o ) . But, said
Tylor, an idol may tend to be confounded with the idea of which it
was the symbol, and thus become 'the parent of the grossest super-
stition and delusion' (1878, n o , 120). He pointed out that the line
between cases in which the connection between object and figure was
supposed to be real, and those in which it was known to be imaginary,
was very often difficult to draw. He cited the images of saints beaten
and abused for not granting the prayers of worshippers; and the
symbolical sacrifices of models of men and animals, including the
'economical paper-offerings' of the Chinese, as instances in which
symbol tended to be identified with reality (1878,121,122). Hanging
and burning in effigy, he thought, in civilized countries at least,
comes fairly out into pure symbolism. The idea that the burning of
a straw or rag body should act upon the body of the original per-
haps hardly comes into the mind of anyone who assists at such a
performance. But if Tylor had been writing in a modern context, say
after a campus demonstration, he would surely have added that
burning an effigy is a mode of action in its own right - a kind of
relief in the midst of frustration. While not conceived as affecting the
Drums' of Africa much has been written. For an illuminating brief article from a social
anthropological viewpoint see Robert G. Armstrong, 'Talking Drums in the Benue-Cross
River Region', Phylon, xv (1954), 355-63 (Atlanta University, Ga.).
* For example Salomon Reinach (1942, 9) cited G. F. Creuzer's statement that in very
remote ages a sacerdotal caste imbued with lofty religious and moral ideas thought to make
these more accessible to the multitude by disguising them under symbols, which were then
taken literally, as in Greek polytheism and the ancient mysteries. Paul Frankl has quoted
A. Wilhelm von Schlegel, lecturing in the University of Berlin in 1801, as stating 'the beauti-
ful is a symbolical representation of the infinite; because then it will be at the same time clear
how the infinite can become apparent in the finite . . . The infinite can only be made apparent
symbolically, in images and signs' (Frankl, i960, 452). A. F. Bernhardi wrote to similar
effect about that time; the same theme appears in Bachofen, as I have shown already.
body of the person caricatured, it may be hoped to influence his

mind, as a form of indirect disapproval by those doing the burning -
not so far from beating the images of saints!
In some passages Tylor displayed his simplicist evolutionary
limitations and a rather crude rationalism. He writes of superstition
in a pejorative sense, and he is scornful of the 'childish symbolism' of
Chiromancy - palm-reading. He betrays his value judgement also in
his view that man being the highest living creature that can be seen,
it is natural that idols should mostly be imitations, more or less rude,
of the human form. But as Marett pointed out in due course, we
need not treat it as an intellectual achievement of outstanding
importance to construct the likeness of a man and attach a symbolic
meaning thereto. Moreover, as I myself have shown for the Tikopia,
a society with quite developed religious concepts need have no
anthropomorphic idols (1970a, n 8-21 et passim). If a material ob-
ject is conceptualized as a representative of a spirit being, its shape
may be irrelevant; a stone or a house post will do as the spirit's
embodiment. Indeed, to cite Marett again, anthropomorphism in
imagery may run counter to the spirit of reverence - which is
exactly the view taken by orthodox Islam, which explicitly forbids
anthropomorphic religious sculpture or painting. The most de-
veloped symbolism may be the least naturalistic.
As far as I can see, Tylor did not define the term symbol; he just
used it as equivalent to representative (1873, II, 168). But following
Tylor, though in more restricted and often more penetrating way,
William Robertson Smith, in the Religion of the Semites, defined a
symbol as the 'permanent visible object, at and through which the
worshipper came into direct contact with the god' (1889, 151). This
was an action-oriented frame of reference, specifically directed to the
explanation of religious phenomena, and not intended as a general
categorization. But in his handling of the notion of symbolism
Robertson Smith did introduce a social dimension. In his earlier
work on Kinship and Marriage in Early Arabia he had paid much
attention to kinship groups, and in his analysis of institutions such as
sacrifice he considered the symbolism in terms of the relationship of
groups or sets of people in common behaviour to a given object. In
this conceptual frame, a symbol was sometimes a natural object such
as a tree or a fountain, sometimes an artificial erection such as a pillar
n6 SYMBOLS
or a pile of stones. Robertson Smith goes to some pains to refute the

suggestion - current long before Freudian theory, as Staniland
Wake has illustrated - that such a pillar or stone heap symbolizing
a god was necessarily a phallic symbol. He thought it was primarily a
convenient way of representing the entire god as a whole person
(1889, 194).
This notion of what I would call 'the empirical aptness of sym-
bols', a kind of pragmatic appropriateness in the choice of the simple
to represent the complex, is reminiscent of though not identical with
the notion of 'natural symbols' already discussed (Chap. 2).
SACRIFICE AND COVENANT
Robertson Smith explored a subject of much subsequent anthropo-

logical attention - sacrifice. A primary element in such a ritual act is
the deliberate initiative of undertaking actual material loss for pre-
sumed immaterial gain. Most sacrifices are not purely individual acts
but involve sets of people, so questions arise about the basis of their
association; what kind of gain is thought to be got; how is it con-
ceived to operate and how is it distributed. Already Fustel de
Coulanges had pointed out that in the 'ancient city' the principal
ceremony of the domestic worship was a repast, which was called
a sacrifice, and which satisfied the need of men to put themselves in
communion with the deity, who was invited to the feast and given
his share. 'Human association was a religion; its symbol was a meal,
of which they [the city members] partook together' (1955, 155-8).
He also indicated the expiatory function of sacrifice of animals, by
effacing all stain of evil from the city - for which purpose it was
necessary to have present all citizens, and only citizens, that no
uncleansed member might still contaminate the civic body. Robert-
son Smith followed much the same line, using ancient Semitic cus-
tom as his basic material. He saw sacrifice as a symbol of communion
rather than of slaughter, leading to the establishment of a deeper
more meaningful relation between man and his gods. Its basic
significance lay in commensalism, the 'common table', the eating
together by worshippers of a shared meal derived from the flesh of
a sacrificial victim identified in totemic fashion with the clan deity.
Although ethnographically inaccurate, and largely replaced by a
DEVELOPMENT OF ANTHROPOLOGICAL INTEREST H7
sacramental theory which stressed identity of worshipper with

victim,* this type of explanation had its importance in relating
symbolic behaviour to structure of society.
Another angle on the communion theory of sacrifice was given
by Henry Clay Trumbull, from a comparative theological if rather
unsophisticated point of view. His consideration of The Blood
Covenant (1885), which appeared just prior to publication of the
Religion of the Semites^ was originally delivered as lectures in the
summer school of Hebrew at Philadelphia. His main theme was that
blood was equivalent to life and the soul, that blood-transfer is soul-
transfer, and that blood-sharing secures a union of natures, divine as
well as human. Hence in the symbolic sacrifices in the Old Covenant
of God with his Chosen People, the Jews, it was the blood which
made atonement for the soul. So in the New Covenant ratified by the
crucifixion of Jesus, it was by the blood - the life of Christ - not by
his death and broken body, the Fractio^ that atonement and redemp-
tion came to man. Although from one point of view it was only
another unorthodox contribution in a long tradition of Biblical and
Oriental scholarship, and naive by our modern standards, Trum-
bull's work broke some new ground. Its breadth of comparative
instances embraced not only India and China as well as the Near
East but also Africa and Central America. This he justified by the
view - shared by Robertson Smith - that the Bible as an Oriental
book needed consideration in a Semitic context; but to this he added
the more original if more debatable assumption that since the Bible is
a record of God's revelation to the whole human race its inspired
pages can receive illumination from 'all disclosures of the primitive
characteristics and customs of that race everywhere'. So anthro-
pology and theology were brought together in a study of symbolism
which foreshadowed later more refined scholarly treatment. But
Trumbull's citation of primitive rite and myth was less of a con-
tribution, coming after Tylor, than his use of behavioural data. He
opened his analysis with a description of an actual blood-exchange
rite as witnessed by a native Syrian informant (that is, not cited from
a book). This he followed by a linguistic analysis to show that some
seemingly very diverse words might conceal a common concept. He
* See Hubert and Mauss, 1899, Evans-Pritchard, 1956, G. Lienhardt, 1961; cf. Raymond
Fkth, 1963; E. R. Leach, 1972, 266-8.
Il8 SYMBOLS
argued that the Arabic words for friendship, affection, blood, leech
or blood-sucker were but variations from a common root - a study
of concepts highly commendable in a Doctor of Divinity, even one
who had been a missionary! Trumbull pursued this blood symbolism
theme further in studies of threshold covenants and covenants with
salt, both of which he thought were based upon equivalences with
blood and the notion that a sharing of blood meant a sharing of life,
with approval from the gods. (He also produced a smug superficial
study of'Friendship, the Master-Passion'.) But his set of lectures on
Oriental social life (1894) contained some interesting personal
observations from his travels in Arabia, Egypt and Palestine, and
some pithy remarks on the sharing of food and drink as symbolic of
covenant, and on the symbolism of Oriental weddings and of pil-
grimage. An essay on 'the Oriental idea of father' illustrated very
well from empirical data the representative role of the head of a non-
kin related group of travellers. Trumbull's work serves to illustrate
the kind of contribution made to the study of symbolism by an
untrained but shrewd observer equipped with knowledge in a re-
lated field to anthropology, and fired by a ruling idea of a moral
order.
THE SECRET A N D THE P R O F A N E
In their studies of symbolism anthropologists have always tended to

be in an ambiguous position, either being invaded by amateurs or
themselves trespassing on fields already in occupation by other
disciplines. But at times they have also been acquainted with that
twilight world of protected knowledge in which secrecy, mysticism,
the occult, the forbidden all mingle, and where scientific curiosity
easily overlaps with, is even confused with, a search for sensuous
experience.
At an early period in proto-anthropological studies of symbolism
the notion of symbol as hieroglyph was alluded to, implying that it
was a sign of highly condensed meaning, the understanding of which
would place one in possession of a range of knowledge of an esoteric
kind. So dreams, and art, were hieroglyphic, giving clues to an
understanding of the real nature of man ordinarily concealed (cf.
Freud, 1952, 181). At the pragmatic level there was a desire to
DEVELOPMENT OF ANTHROPOLOGICAL INTEREST II9
penetrate the secrets of such organizations as those of Freemasonry,

not to gain understanding of symbolic processes as such, nor just to
know what were the practices hidden from the outside world, but
also in 'cracking the code' of the symbols used, to attain control of
whatever power they might have as explanatory principles of nature
and society. So an elaborate study in three volumes purporting to be
a comparative handbook of the symbolism of Freemasonry (Schau-
berg, 1861-3) excited a lively curiosity and reached at least three
editions.* In the latter part of the nineteenth century it was believed
by some people that the pyramids of Egypt, in their basic propor-
tions, held the key to significant general relationships of a mathe-
matical order applying to the operations of the world and discovered
at an early stage of human history. It was thought that proper
scientific investigation would reveal the 'secret' of the pyramids; and
part of the conclusion of Flinders Petrie's scholarly studies around
1880 was to disprove such speculations. Such aura of the mystical
and occult has accompanied anthropological studies of symbolism at
all stages, f
Yet recognizably anthropological studies of symbolism in the
secret and the occult have continued to be produced. Apart from
inquiries of an avowed administrative nature, into secret associations
* For a modern sociological study of Freemasonry in Sierra Leone see Abner Cohen
(1971). Some studies on secret societies begun with political interest have been assimilated to
the corpus of anthropological materials, for example Schlegel, 1866, Pickering, 1878-9, Ward
and Stirling, 1925-6, and Blythe, 1969, on Chinese secret societies of Malaysia.
f Some recent publications on Polynesian symbolism, while not by anthropologists, have
drawn upon anthropological material. For example Leinani Melville, Children of the Rainbow
(A Quest Book, Theosophical Publishing House, Wheaton, Illinois, 1969) claims to 'decipher
the esoteric code of the tahunas* (ritual experts). Symbols drawn on the sand by a Hawaiian
fisherwoman and seen again twenty years later 'on a rare old tapa cloth from Bishop Museum'
were stated to include the Cosmic Egg, the Seven Serpents of Wisdom...' etc. A recent
work in what is described as 'The Penguin Metaphysical Library - Religion/Anthropology'
(Joseph Epes Brown, The Sacred Pipe, Penguin Books, Harmondsworth, 1971) gives elabor-
ate symbolic explanations of the tobacco pipe, the spotted eagle, the buffalo by the Oglala
Sioux priest Black Elk. He is described as a 'man of vision', a 'holy man', upon whom destiny
in a time of cultural crisis had placed a heavy burden for the spiritual welfare of his people, and
whose account of ritual and beliefs 'could also be an important message for the larger world'.
This metaphysical library is stated to have the object of helping man on his struggle for self-
knowledge, by offering 'books that can recall in modern man the forgotten knowledge of
how to search for himself, knowledge he has lost in his haste to make himself comfortable in
the world'. In such views the study of symbolism, with which anthropology is overtly
affiliated, becomes a branch of an occult quest. Another very popular text, The Teachings of
Don Juan, by Carlos Castaneda (1969) is used primarily as material for analysis in cognitive
anthropology; but it appears to be treated by some students as a source of more generally
utilizable symbolic knowledge.
120 SYMBOLS
of a political type, which yielded data on symbolic construction and

relationships as a by-product, there have been ethnographic studies
of direct scientific intent. I mention only two, for illustration. A
somewhat pedestrian examination of symbolic form was Thomas
Wilson's study of the swastika (1894) presented as a report to the
United States National Museum. This dealt with the well-known
bent-arm cross (fylfot cross, croix gammee) later rendered notorious
by the German National Socialists but at that time regarded as a
widespread symbol of traditional mystical value. According to Wilson,
the swastika was one of 385 varieties of cross known to ornament
and heraldry, and had been identified variously as the symbol of
Zeus, Baal, Agni, Indra, Brahma, Vishnu, Shiva, Thor and Jupiter
Tonans. Wilson held too that it was the oldest Aryan symbol, to
have symbolized sun, sky, light, forked lightning, water; to have
been an emblem of the Great God, Maker and Ruler of the Universe,
and to have appeared in the footprint of the Buddha. While throw-
ing little light upon the meaning of the symbol in particular situa-
tions, this inquiry did suggest that one should be wary of accepting
any single identification of any widely spread symbol. A much more
penetrating inquiry of a very different order was that of Michel
Leiris into the secret language of the Dogon men's association
(1948).* Renowned in enthnographic history for their masked
dances and their complex philosophic concepts, the Dogon men have
had a special language in which initiates communicate during the
dances and funeral rites, and which is incomprehensible to ordinary
uninitiated members of the Dogon society. Very restricted, even
rudimentary, in its forms, the secret language is yet linked with the
use of ritual objects, with myths of origin, and with concepts of an
elaborate symbolic order. A gifted literary figure as well as anthro-
pologist, Leiris explained in carefully chosen words the system of
symbolic associations involved with this language - the interpreta-
tion of red, the symbolic mask colour, as linked with heat, sun, fire,
brilliance, ardour, force, danger, anger, power, command; yet the
mask linked also with ideas of freshness, humidity, beneficence,
peace, fertility. Leiris draws attention to the conjunction of the two
* This work is cited in this chapter because though its publication was post-war, the
original texts were collected, with commentary, in 1931 and 1935 and their interpretation
should be credited to this pre-war period.
opposed terms, heat and coolness (which seem most appropriate

translations in this context of ardeur-fraicheur\ linked with a male-
female dualism, which reappears again in concepts of fecundity.
Symbolized also by the mask, and by the terms which represent it in
the secret language, are the succession of generations of men, and the
concepts of death and reconstitution, all bound together in an inter-
woven series of ideas, myths and rites responding to the multiple
needs of the social life. In this study, the principle of binary opposi-
tion in much of the symbolic material is clearly demonstrated.
Anthropological studies of symbolism associated with secret
beliefs and practices have also had a vast field for inquiry in the areas
of magic and sorcery and especially witchcraft, where things said and
things done have commonly borne an evident symbolic significance,
and where secrecy was commonly regarded as a necessary adjunct if
not an actual symbolic component. Two points are especially
relevant here. One is the great success of anthropologists in breach-
ing the wall of secrecy. To judge from the internal evidence of such
classic field studies as those of Malinowski in his Argonauts of the
Western Pacific (1922) and Coral Gardens and their Magic (1935) or
Evans-Pritchard in his Witchcraft, Oracles and Magic among the
Aânde (1937) or from my own personal experience among Tikopia
and Kelantan Malay fishermen, persistent sympathetic inquiry in the
vernacular can yield a great deal of understanding of the symbolic
system involved. (Penetration of the symbolic apparatus of a specifi-
cally secret association is another matter; the anthropologist perforce
must often remain an external observer who picks up what he can.
If he actually becomes a member of the society, this also may have
its problems of preservation of the secrets.) The other point is that
for the most part anthropological studies of magic and witchcraft
until fairly recent years have not been presented in a symbolic idiom.
Evans-Pritchard, for instance, wrote of 'notions' of witchcraft,
treated the descriptions of witch behaviour as if they were valid in
their own terms, and stated that they were less an intellectual symbol
than emotional reactions to failure of a person's hopes of achieve-
ment.
Mingled in with many practices of secret societies and many con-
cepts of witch behaviour have been elements of sex symbolism. But
quite apart from this, ethnographic material about concepts of sex,
E
122 SYMBOLS
sex behaviour in and outside marriage, representations of sex in

dramatic and pictorial art, sexual aspects of religious cults, has
tended to form part of studies which attempt to give a fairly full
systematic account of any culture. Inevitably, some of this material
has dealt with beliefs and practices, realistic or symbolic, which have
seemed exotic by Western standards, and which have therefore
appealed to a section of non-scientific readers looking for sensual
gratification. Accordingly, some anthropological studies which were
primarily investigations of sex symbolism in ritual or aesthetic cul-
tural setting, or of the relation of sex norms to kinship structures and
marital institutions, have tended to rub shoulders in some bookshops
with literature of a more 'curious' kind or be relegated to locked
presses in libraries. Leaving aside more definitely pornographic
work, produced primarily to titillate, there is undoubtedly a con-
tinuum of a general kind here, in which classical sex manuals of
Kama Sutra type, literary and poetic exercises of Burton, Mardrus
and Powys Mathers in treatment of the 'Arabian Nights', grimly
encyclopaedic studies of woman by Ploss and Bartels, more sensi-
tively directed special inquiries by Albert Moll and Kraft-Ebbing,
broad comparative surveys by Edward Westermarck and Havelock
Ellis,* and profoundly penetrating clinical expositions of Sigmund
Freud, have all shared enough of a common sex theme to have been
treated with reserve at some time. The attraction and the danger of
sex, especially as presented in the moral codes of the nineteenth and
early twentieth century, gave writings about it a quality both secret
and profane, until in modern times more liberal views have pre-
vailed. How far anthropological writings have contributed to such
liberation it is impossible to say. But the work, for example, of
Malinowski {Sexual Life of'Savages', 1929) and Margaret Mead (Sex
and Temperament in Three Savage Societies, 1935) probably helped
appreciably in the pre-war period.
The stout inquiries of Dulaure and of Bachofen into phallic
symbolism have already been mentioned. This line of study con-
tinued through the nineteenth century, until it was absorbed into the
more broadly-based study of sex as part of the fabric of institutional
behaviour and values and concepts looked at as a whole by func-
* A considerable part of Havelock Ellis's massive work was devoted to 'erotic symbolism',
a term first used, apparently, by Eulenberg in 1895 in a technical sense.
tional anthropologists. An example is given by two papers read in

1870 before the Anthropological Society of London, one on phallic
worship by Hodder M. Westropp, the other on the influence of the
phallic idea in the religions of antiquity, by Charles Staniland Wake.
Nowadays such material would attract little attention. But then,
outside the anthropological sphere it could be subject to vulgar
criticism for its 'dubious' quality. So Alexander Wilder, contributor
of introduction, notes and appendix to a second edition of the two
papers, which had been published together, gave a stern rebuke to
possible critics. He pointed out that it was necessary to have respect
for the symbols of other religions - a view in line with anthropo-
logical attitudes but one which had produced trouble for Creuzer
half a century earlier. He defended the investigation of phallic sym-
bols by arguing, much in the spirit of Dulaure, that the ancients 'saw
no impurity in the symbolism of parentage to indicate the work of
creation. No man born of woman can with decency impugn the
operation of that law to which he owes his existence; and he is
impious beyond others who regards that law as only sensual' (1875,
vii). The main essay in the book was that of Wake, known otherwise
mainly for his work on kinship, who devoted particular attention to
the symbolism of the serpent. Applying comparative treatment to
the Bible story of the Fall of Man, he pointed out that the serpent
was at an early date a symbol of wisdom, of healing, of life, and he
linked this with its phallic significance. Following Bachofen (and
pointing forward to modern views of binary opposition) Wake held
that 'the doctrine of reciprocal principles of nature, or nature active
and passive, male and female, was recognized in nearly all the primi-
tive religious systems of the old as well as of the new world' (1875,
27 ff.).*
* As an example of later quasi-anthropological studies of sex symbolism Ettie A. Rout's
Maori Symbolism (London, Routledge, 1927) may be cited. Drawing partly upon literary
sources and partly upon some highly-coloured information supplied by an Arawa Maori
resident in London, this book attempted to interpret Maori myth, custom and art in symbolic
terms, especially of a phallic order. The author had an enlightened approach to sex education,
and had done valuable hygienic work with soldiers in the First World War. Assisted by her
husband, F. A. Hornibrook, she had developed some interesting ideas on the relation of
Maori dancing to bodily health. But she was less fortunate in her approach to anthropology,
and too uncritical of the ideas she wished to promote. A preface by Sir Arbuthnot Lane, a
surgeon of great skill, with eccentric ideas on disorders of the digestive tract, did nothing for
the reputation of the book in anthropological circles. (See review by 'Tangiwai', New
Zealand Herald ($ January 1927), headed 'Maori Symbolism: A Book of Surprises'.)
124 SYMBOLS
S T I M U L U S A N D M Y O P I A OF F R A Z E R
Another development of studies in symbolism with classical associa-

tion, which can be traced back through Bachofen at least to Creuzer,
but which was greatly stimulated by Wilhelm Mannhardt, was the
more explicit attention paid to cults of nature. Mannhardt, to whom
Sir James Frazer acknowledged his debt, had begun to publish in
1865, soon after Bachofen, but his major work, Antike Wald- und
Feldkulte, appeared in 1875 and 1877. It was a product of genuine
anthropological research, being based on the results of oral inquiry
and questionnaires as well as on a scrutiny of literature, and it was an
exhaustive exposition of agricultural and allied rituals and beliefs of
the peasantry of Northern Europe. Through complex cross-fertiliza-
tion these schemes of nature symbolism emerged to anthropological
attention particularly in the monumental volumes of The Golden
Bough. This, as Frazer himself summarized its aim, was a prolonged
inquiry primarily into the meaning of the concept of the slain god; or
as we may put it in more general terms, into the symbolism of
generation, change and decay in human beings and nature. Recogni-
tion of the significance of this concept Frazer owed to Robertson
Smith. Patiently unravelling the symbolic significance of the ancient
custom - or purported custom - of killing the priest of Nemi as he
began to lose bodily vigour, Frazer constructed his elaborate theory
of cults of vegetation. In this, the vivifying force of Nature was
represented by an anthropomorphic but divine or semi-divine per-
sonality whose birth, death and resurrection were paralleled by and
symbolized by the annual progress of the seasons and the cycle of
vegetal growth. A theme with an important symbolic content, it has
continued to hold the attention of anthropologists, even though their
interpretations have tended to depart from Frazer's views. An
example of a modern anthropological approach is that of Max
Gluckman (1963, i n ) . He has described how in myths Frazer saw
men handling dramatically the dying and resurrection of vegetation
with the changing of the seasons, and in ceremonies aiding and
reviving the dying hero and with him the vegetation. He held that
Frazer undoubtedly over-simplified the problem, but was interested
in the intellectual patterns which he believed must lie behind all
these customs. Moreover, while a modern anthropologist, basing his
analysis on detailed observation in the field, is concerned in greater

degree with ceremonial roles of persons, categories of persons and
social groups in relation to one another, Frazer could not have
pursued these problems for lack of relevant evidence.*
No one can deny the stimulus Frazer gave to the general anthro-
pological investigation of the symbolism of magic and religion.
Other scholars influenced directly by him pursued threads of sym-
bolism in classical, oriental, mediaeval and primitive fields. Much of
this work ran outside the mainstream of social anthropology. But
some, like that of Baldwin Spencer in Australia, resulted in rich
ethnographic data; others, like that of Jane Harrison and Malinowski
(who were also influenced by Durkheim) contributed theoretical
interpretations of importance to students of ritual. But the Frazerian
stimulus has been complex. Jessie L. Weston's analysis of the legend
of the Holy Grail, From Ritual to Romance (originally published in
1920, and classed by its modern publishers as 'anthropology') was
pursued in terms specifically suggested by Frazer's work. In its turn
it was said by T. S. Eliot to have inspired the basic symbolism of his
Waste Land. (Eliot also acknowledged his general debt to Frazer.)
Frazer himself was of considerable importance in the literary world
of his time, and this example is but one instance of a relation between
literary and anthropological interest in symbolism which has been
manifest sporadically in some periods, f
But the links between literature and anthropology around the turn
of the century were less remarkable than the way in which the impor-
tant literary issue of symbolism almost completely escaped anthropo-
logical attention. For some of this ignorance Frazer was responsible.
His own literary interests were primarily classical, and stylistic in
a broad sense; his anthropological interests were omnivorous, but
* For discussion of Frazer's theme of killing the king see E. E. Evans-Pritchard, 1948,
Raymond Firth, 1964,75-80. For more general analysis of the work of Frazer see I. C. Jarvie,
1967, and Edmund Leach, 1965.
t At the present day, one instance is Victor Turner's citation of Baudelaire's Correspond-
ances in preface to his appropriately named Forest of Symbols (1967). In the reverse direction
Jerome Rothenberg and some other poets have been interested in anthropological work of a
symbol character — see Technicians of the Sacred: A Range of Poetries from Africa, America,
Asia and Oceania, edited with commentaries by Jerome Rothenberg (New York, Doubleday,
1968); and the journal of 'ethno-poetics', Alcheringa, edited by Jerome Rothenberg and
Dennis Tedlock (New York, 1970, etc. — the name of this journal is taken from Australian
aboriginal ritual). In some cases the link has been an integral one, as with Nathaniel Tarn,
poet and anthropologist, and Michel Leiris, cited above.
126 SYMBOLS
primarily in the collation of facts. 'Theory' he distrusted, and he was

not greatly concerned with intricacies of philosophic thought. When
his friends pointed out that his ideas about magic preceding religion
were remarkably like those of Hegel, he made rather tart amends in
his next edition by remarking that he had never studied Hegel 'nor
attended to his speculations', and simply reprinting the relevant
passages as an appendix. Nor was he concerned with modern changes
in ritual and symbolic forms. (As I found out when meeting him for
the first time and, being asked about the Maori, I tried to tell him
about a modern religious cult service I had attended. His only com-
ment was: 'I am not interested in such things.') Frazer's aim, at
least in part, he stated, was to rectify the rationalization in the work
of writers on the origin of political institutions who discussed the
development of monarchy, and to see that due allowance was made
for what he called 'factors of superstition' in the history of sacred
kingship. As Malinowski has pointed out, Frazer had a kind of
romantic notion of the primitive - not an ignoble, but a magic-
ridden savage - whose symbols were capable of interpretation by
associational and allied principles. He made a great and original
contribution in the study of magic. Yet though his wife was French,
and the writings of the Symbolists were probably not unfamiliar to
him, Frazer apparently saw no links of any interest in their work,
not even in the views of Mallarme on language. Nor does he seem to
have examined the earlier views of Novalis on the power of words
(p. 94). What use Frazer might have made of Symbolist writings is
mere speculation. But a clearer view of the symbolic operations of
magic might have led him to separate off magic from science much
more distinctly on theoretical grounds, as Malinowski did later on
empirical grounds. But this was not Frazer's temper of mind; he was
not concerned with the more abstract issues of symbolism but with
concrete practices in magic and religion.
Chapter 4
CRYSTALLIZATION OF
PROBLEMS OF
SYMBOL THEORY
The nineteenth century saw considerable development in the study

of symbols, by anthropologists and writers in other fields. The
specific recognition of symbols as a subject of study had begun in the
classical historical area but soon broadened out in comparative study
in Oriental and other exotic societies, of myth, religious concept and
ritual behaviour. Symbols were recognized as constructed objects,
human art products, needing instruction to identify and interpret,
but subject to rule or principle. There was a notion of symbols as
constituting a language, and language as having a symbolic power.
Symbols were regarded as corresponding to an 'inner world' of
reality not otherwise comprehensible. This inner world was primarily
not one of conscious conceptualization; it had intellectual com-
ponents but a strong base of emotion and feeling. So it was not long
before the symbolism of dreams was recognized as reflecting com-
plex facets of the personality.
But there were many aspects of the study of symbols. In the
growth of the Romantic movement attention was focused on
identification and interpretation of symbols already existing, in
nature or referring to nature (for example myth, dream); whereas the
Symbolists focused on the invention and construction of symbols
out of human experience. They adopted a theoretical, questing mode
of approach. So also did the philosophers, for example C. S. Peirce.
But with Peirce, as a forerunner in symbolic logic, the prime concern
was not the content of the symbols he studied, but the relations -
between symbols, and between symbol and content. The content in
itself was not important for him; it could be supplied according to
the nature of the problem. Now for the Romantics, as for the
Symbolists, the question of the content of their symbols was impor-
tant, even disturbing. Exactly what did their symbols represent - the
127
128 SYMBOLS
forces of nature, the inner personality of man, the mind of God?

Moreover, they were for the most part looking for a system, a
unified set of indices which would represent major facets of ex-
perience.
Such interests were not in the forefront of the nineteenth-century
anthropological study of symbols. Anthropologists were concerned
with understanding the significance of particular symbols, and the
nature of symbolic process in an institutional context, largely as a by-
product of other studies. Anthropological interest in symbols re-
mained relatively unconcerned with what may be called the final
content of symbols studied. Nor did it attempt to construct any
thorough-going theory of symbolism, any further than exploring
the linkage between different symbols in a given society or between
different forms of the same symbol in a range of societies.
ETHNOGRAPHIC DEVELOPMENTS
Around the turn of the century specific anthropological interest in

symbols began to crystallize more clearly, as ethnographic studies
became more systematic, and representational properties of objects
and of behaviour came to be more sharply distinguished from their
pragmatic quality. Such identification and interpretation of symbols
in specific cultural contexts can be seen in the work of Edward
Westermarck. Between 1898 and 1930 he had paid about thirty visits
to Morocco, totalling about nine years in the country, and had
acquired in vernacular contexts a vast amount of data on a great
range of ceremonial and ritual institutions, rich in examples of
symbolic usage, noted as such and annotated (1914, 1926, 1930).
But the most obvious field for symbolic study was primitive art,
and here the work of Franz Boas is an outstanding example. The
results of Boas's thinking about primitive art symbolism did not
reach the public for over a quarter of a century (Boas, 1927) but he
taught and stimulated research in this field from about 1900. Boas
himself stated that Ehrenreich was the first to observe the highly-
developed symbolism of the North American Indians, publishing
about the Cheyenne in 1899. Based on these results more extended
study of the symbolism of American Indian art was undertaken by
A. L. Kroeber among the Arapaho, Roland B. Dixon among Cali-
PROBLEMS OF SYMBOL THEORY 129
fornian tribes, Clark Wissler among the Sioux and Blackfeet, and
H. H. St Clair among the Shoshone. One of the contributions of
such work was to show in detail how abstract ideas such as a battle,
or a path to a destination, could be indicated by geometric patterns
in colour. Illustrative of some phases of this work was Kroeber's
doctoral thesis on decorative symbolism of the Arapaho (1901) in
which he put forward a strongly-worded argument against pre-
suming either realistic pictography or symbolic conventionality as
the origin of the art themes. With his characteristic caution, Kroeber
was against the search for origins and isolated causes; but he was in
favour of the study of those general tendencies which he thought to
be inherent in the mind of man as a social and cultural being. Prob-
lems of the relation of meaning to form were preoccupying many
students of primitive art and received evolutionary solution - as
A. C. Haddon showed in tracing geometrical shapes in New Guinea
art back to realistic forms such as crocodile or frigate-bird. Boas
himself was more eclectic. He noted the range of interpretation that
could be obtained about the meaning of a particular design, even
among the people of one community, and he made suggestions for
systematizing such data in field inquiries. He also injected into his
treatment of primitive art some remarks on symbols more generally.
Referring to the symbolism of national flags, he pointed out that
they are not only ornamental but also possess a strong emotional
appeal. They call forth feelings of national allegiance and their values
cannot be understood on a purely formal basis; they are founded on
the association of form with definite fields of our emotional life.
Boas also referred to the swastika (cf. p. 120) as a symbol of anti-
semitism, and to the Star of David as a Jewish symbol in Germany
at that time (before the rise of the Nazis to power!). He pointed out
that these symbols had a very definite political significance, and were
apt to excite most violent passions when used for decorative pur-
poses. Owing to their strong emotional value such symbols, as also
military insignia and students' emblems, tended to be restricted to
special classes of objects or reserved for privileged classes or indi-
viduals (1927, 100-1). Here, though briefly, Boas was making a
significant general point about collective symbols of a sacred or
quasi-sacred order, akin to the view put forward by Durkheim.
I think this is worth noting because in general Boas seems to have
I30 SYMBOLS
discussed problems of this type in an idiom other than that of

symbolism. In his introduction of 1910 to the Handbook ofAmerican
Indian Languages (1911) Boas advances towards questions of sym-
bolization but never actually gets there. He mentions linguistic
similarity in Chinook mythology - where the Culture Hero dis-
covers a man in a canoe who obtains fish by dancing and tells him he
must not do so but catch fish with the net, a tale based on the homo-
phonic similarity (Boas calls it identity) of the words for dancing and
catching with a net. Boas was concerned to show how similarities in
linguistic form could lead to speculation of a religious or allied
kind, and he thought that this might show an important characteris-
tic in the mental development of different branches of mankind. He
was inclined to believe, he said, that the 'frequently occurring
image5 of the devouring of wealth had a close relation to the detailed
form of the winter ritual among North Pacific Coast Indians, and
that the 'poetical simile' in which a chief was called 'support of the
sky' had been taken literally in the elaboration of mythical ideas.
But Boas did not follow up such leads.
THE D U R K H E I M I A N LEGACY
Around the turn of the century two theoretical viewpoints of major

significance for the study of symbolism began to be defined, and in
due course exercised a profound influence upon anthropological
studies in this field. Durkheim and Freud represented very different
traditions, both in intellectual discipline and in national background
of thought, and they affected anthropology in very different ways.
Their views may be contrasted briefly, without qualification and
almost without distortion, as complementary opposites. Durkheim
was concerned primarily with the symbolism of groups, as an ab-
stract exercise of interpretation; Freud was concerned primarily with
the symbolism of individuals, in a clinical setting, as a clue to the
solution of pragmatic problems. Durkheim was interested in sym-
bols that were expressive of consonance, of solidarity of the indi-
vidual with his society; Freud saw symbols as expressive of dis-
sonance, of the individual's lack of harmony with his society as
represented especially by his nearest kin. And while for Freud sym-
bols tended to be looked upon as disguises to avoid confrontation
PROBLEMS OF SYMBOL THEORY I3I
with a painful reality, for Durkheim they were means whereby one
could more fully embrace a satisfying reality. Both attached great
importance to the significance of such symbolic representations, but
the difference of their objectives led each to stress a different aspect
of the symbolic process. In the anthropological study of symbols it
has been the viewpoint of Durkheim that until recently has carried
more weight, in line with the effect of his more general social theory.
But despite the protestations of some anthropologists about the
need to avoid psychology, the views of Freud, supplemented by their
transformation in the work of Carl Jung, have affected the anthropo-
logical treatment of symbols to a growing degree.
There is no need to dwell on the magnitude of Durkheim's con-
tribution to the understanding of symbolic behaviour. In strong
contrast to the nineteenth-century studies of symbols, which tended
to treat them as discrete entities, having their meaning hidden within
themselves or sharing it with symbols of the same class on a com-
parative basis, Durkheim advocated the need to relate symbol to
social factors. In arguing that not nature, only society, offered
sufficient basis for the development of a religious symbolism Durk-
heim firmly turned his back on the last vestiges of the Romantic
movement. His positivism alienated him from interest in symbolism
of the poetic kind. He was not concerned with the claims of the
Symbolists to attain an inner reality; nor did he attempt too seriously
to unravel the genesis of symbols. He looked primarily at symbols
as modes of expression, and wished to examine their effect upon
other members of a society. For this purpose the study of religious
symbols seemed most suitable to him. His study of religious sym-
bols took the rather surprising form of an examination of the
'elementary' class of totemic emblems, in Australia. Pushing much
further the idea put forward by Tylor, that it is simpler to consider
a concrete object than an abstract idea, Durkheim demonstrated very
clearly the relation between symbol, religious sentiment and society.
He drove home the idea of society as a system of active forces
involved in and conditioned by the symbolizing process. In con-
sidering the problem of ritual, defined by things sacred, he concluded
that the relation between sacred things and their source was sym-
bolic, not intrinsic. Without symbols, the social sentiments could
have only a precarious existence. Enduring symbols were needed to
I32 SYMBOLS
give continuity to memories of social experiences. So, concluded

Durkheim, 'social life, in all its aspects and at all the moments of its
history, is possible only thanks to a vast symbolism' (1912, 331). In
fact, Durkheim wrote ethnographically of emblems rather than of
symbols, or occasionally of emblematic symbols, but his message
was clear. In the view of Talcott Parsons, who in general was critical
of Durkheim's positivism, Durkheim's use of the concept of sym-
bolic relationship opened the door to two great lines of thought. It
led to a clearer view of the relation of religious ideas to non-
empirical aspects of reality; and it emphasized the significance of
active attitudes in this relation. Parsons held that Durkheim, in his
explicit discussion of the role of symbolism in ritual, has added a
whole new normative category to the structure of action (1949,
466-7). This puts the study of symbolism in direct relation to the
metaphysical problem of the nature of reality and what methods can
be adopted to apprehend it - a problem which as I have shown was
already the concern of Bachofen and others long before. But if
stress be laid upon the social dimension then the claim is justified.
By comparison with Le Bon, Trotter, McDougall, all of whom
contributed to understanding of social process in group behaviour,
Durkheim's interpretation, using symbolism as an important instru-
ment, was outstanding (see also Chap. 10, p. 339).
Parsons, a sociologist, is in many ways the most subtle interpreter
of Durkheim. But anthropologists have shared the same general
view. Even Robert Lowie, who had some scathing remarks about
Durkheim in his history of the development of anthropology, did
praise him for his excellent treatment of symbolism, in showing
convincingly how a symbol may succeed in concentrating upon it-
self all the fervour that properly belongs only to the ultimate reality
it represents (Durkheim, 1912, 219-21; Lowie, 1937, 211). Stanner's
critique is more measured and more penetrating (1967, 217-39). He
asks the question: To what extent did Durkheim succeed in his main
purpose-'to go beneath the symbol to the reality which it re-
presents and which gives it meaning'? Stanner notes Durkheim's
early dependence upon Robertson Smith for an understanding of the
role of religion in social life, and the degree to which despite Durk-
heim's fascination for what seemed the primitivity of the Australian
aborigines, he used the material on totemism as an illustration rather
than as a proof of his theories. From his great field experience in
Australia, Stanner notes certain shortcomings in Durkheim's treat-
ment - his confusions about clan structure; his too rigid distinction
between sacred and profane realms; his insistence on giving religious
symbols an empirical, concrete reference. In Stanner's view, in line
with that of Parsons, there was a basic flaw in Durkheim's analysis
of symbolism, namely the failure to appreciate the complex con-
ceptualization processes involved in an act of symbolization. By
holding that 'non-empirical symbolic referents must be distorted
representations of empirical reality' Durkheim failed to realize the
multiple incidence of symbols - 'the cognitive aspects of the feelings,
values, and aspirations which as well as the conceptualizations go
with the symbols'. He consequently assumed, unjustly, that the
Australian aborigines were incapable of any very systematic abstract
thinking about the basic issues of life and the human experience. So
Durkheim's argument that Australian totems symbolized concrete
social entities such as the clan is not acceptable; these elements may
colour or mediate what is symbolized, but it is broader facets of
experience that are represented.
From Stanner's criticism it is clear that Durkheim's assumptions
about the aborigines' level of symbolic thinking, and about the con-
crete collectivity as the prime referent of the symbol must be
seriously revised. I would add a few criticisms of my own. Durkheim
points out that it is the figurative representations of a plant or
animal, the emblems and totemic symbols, which are the source of a
'religiosity' of which the objects represented receive only a pale
reflection. The totem is above all a symbol, a material expression of
something else - but of what? he asks. Of two things, he replies, the
totemic principle, the god; and the social group, the clan. Then
Durkheim proceeds to a staggeringly simplicist conclusion. 'If then
it is at the same time, the symbol of the god and of the society, is it
not that the god and the society make one? How could the emblem
of the group become the figure of that quasi-divinity, if the group
and the divinity were two distinct realities?' And Durkheim puts the
matter quite bluntly: 'the god of the clan, the totemic principle,
cannot then be anything else than the clan itself, but hypostatized
and represented to the imagination under the sense-perceptible
{sensible) species of vegetable or animal which serves as the totem'
134 SYMBOLS
(1912, 294-5). So in the last resort the symbols are symbols of

society, or at least of sectors of society. Here is a wide logical gap.
That one symbol may stand for two things indicates that they
probably have some element in common, but not necessarily that the
two things are the same! The reason for this identification in
Durkheim's view is twofold. Society in a general way has all that is
needed to awaken the sensation of the divine, because of the control
it exercises over its members. And, in particular, society 'maintains
in us the sensation of a perpetual dependence', and so becomes an
object of true respect. But what Durkheim has given is a magnificent
series of assertions. It is essentially an imaginative creation, betrayed
as such by the way in which he slips from statements about Aus-
tralian clans, vegetables and totems into statements about the sensa-
tions which society has for us. The blandness of such assertions can
be seen when one considers the inference he draws from this idea of
our dependence upon society: it leads to a sensation of respect, and
a concept of the divine. Not for a moment does he consider that in a
clash of individual with social interest it might lead for example to a
sense of revolt. Little wonder that Malinowski reacted against such
propositions on logical as well as on more empirical grounds.
Yet I think a more positive comment may also be made. Durkheim
was not writing for anthropologists only, but for the broad spectrum
of social studies. What he did for the anthropological study of
symbolism I do not find easy to describe. But essentially, it seems to
me, he focused our attention upon the significance of a symbol for
the corporate character of human conceptualization and sentiments.
Whether identified in nature or of human design and workmanship,
the object chosen as symbol was regarded not as representing some
cosmic principle or deity by direct reference, but as having been
produced by and as having influenced the conduct of an association
of men. Since this association, by implication, could be contrasted
with other such associations, and that symbol with other symbols,
questions of defining characteristics, of boundaries, of structures,
were foreshadowed, and stimulated more refined concepts and more
precise field research. (Stanner's own remark may be noted here -
reflecting 'on some particular questions that arose in my own field-
work, which was much affected by Durkheim's approach' - 1967,
217). Irritating as Durkheim's misconceptions may be to an Aus-
tralian specialist, and his dogmatism on points of theory to a general
anthropologist, what he did was to provide a novel, daring set of
hypotheses with an elaborate arrangement of supporting evidence in
carefully and intricately argued form which drove people to seek
theoretical and empirical bases for confirmation or rebuttal.
While the work of Durkheim and his colleagues of UAnnee
Sociologique - Hubert, Mauss, Hertz et al. - impressed upon anthro-
pologists the vital importance of the social dimension of symbolism,
it was to Radcliffe-Brown and Malinowski, borrowing in part from
this inspiration, that British social anthropologists looked for de-
velopment of these ideas in empirical research.
SYMBOLS AND MEANING: RADCLIFFE-BROWN
Radcliffe-Brown's explicit treatment of symbolism was somewhat

spasmodic. In 1922, in the Andaman I slanders^ he specifically dis-
cussed questions about processes of symbolic thought. He was con-
cerned to explain Andamanese symbolism so that it should not
appear strange to his readers. He pointed out that when Andamanese
painted themselves with clay as a protection against becoming what
might be translated as 'hot' this term was used symbolically to
indicate danger arising from energy produced by power.* Among
other 'symbolic representations' of forces affecting the social life he
cited fire and red paint. In Andamanese colour symbolism red was
pre-eminently the colour of blood and of fire. Blood was associated
with the warmth of the body and with life, fire with activity and
mental excitement. So, as a symbol of a condition of well-being, red
paint was applied to a sick person, to a dancer, and to a homicide
(for purification). Radcliffe-Brown suggested that symbolism of red-
ness, blood and fire, which he thought might be universal, could
have a psycho-physical basis of a 'dynamogenic' kind (1922, 308-10,
318; cf. Lowie, 1924, 286). Radcliffe-Brown also explained the sym-
bolic significance of Andamanese myths in terms of their expression
of interests and relationships of importance to the society. His
interpretation of specific symbolic concepts and behaviour in this
* A modern social analogy not available to Radcliffe-Brown then is the concept of a
national economy becoming 'overheated' by relaxation of controls on the use of resources.
Note that Reo Fortune (1932, 295-6) acknowledges the stimulus of Radcliffe-Brown for his
collation and interpretation of analogous ideas about heat in Dobu (cf. Leiris, infra).
136 SYMBOLS
way was ingenious and convincing. Moreover, though his presenta-

tion was very didactic, his method of comparative scrutiny was
appealing and his reference to social states suggestive.
The relation between symbol and society was very clear to Rad-
cliffe-Brown in his own formal terms. Of course, it owed much to
Durkheim. Generally, he argued, the needs of society require that
sentiments regulating the conduct of individual members in ac-
cordance with those needs are given collective expression on appro-
priate occasions by ceremonial customs. Such collective expression
constitutes symbols. But he did not examine the concept of symbol
very closely. He used the term as a synonym for 'expression'. When
in an Andamanese marriage the couple put their arms round each
other, and the groom finally sits in the bride's lap, 'the social union
is symbolized or expressed by the physical union of the embrace'
(1922, 236). Radcliffe-Brown also linked the notion of symbol
directly with that of meaning, observing of the Andamanese peace
ceremony that the 'meaning' of the dance is easily discovered, the
'symbolism' of the dance being obvious. Yet he suggested that the
symbolical was only one kind of meaning by noting that a screen of
fibres employed at the dance 'had a peculiar symbolic meaning'
(1922, 237). Analytically, such treatment is not very satisfactory;
some categories ordinarily separated seem at times to be confused.
I would hold that 'symbolizing' and 'expressing' are not necessarily
identical: expressing is an observable token of invisible attitudes,
whereas symbolizing is a particular form of expression, or has a
component of expression. To say of a man that he 'expressed his
pleasure' by a smile - implies a direct response to a stimulus; he
'symbolized his pleasure' by some action - such as opening a door
he had previously kept locked - implies an indirect representation of
his feelings. There are difficulties also about identification of the
referent of the Andamanese symbols. Did they express actual senti-
ments or (as in the essay on taboo) apparent sentiments or the social
significance of events or relationships or values? Again, in stating
that the Andamanese seemed to associate with the idea of heat all
conditions of mental activity or excitement Radcliffe-Brown wrote
that there was good ground for thinking that 'all such associations or
symbolisms (sensory metaphors)' have a physiological basis (1922,
309). He seems here to have intended to indicate only a particular
type of symbolism. Throughout he seemed more concerned with
interpretation than with definition of symbolic phenomena. But if
symbol did relate to sentiment, was the appropriate sentiment
expressed by the symbolic ceremonial behaviour, evoked by it, or
alluded to in it - or all of these? The marriage embrace 'by the power-
ful emotions it evokes5 Vividly impresses' on the young couple what
their marriage means. But is this what it really does - or what it
ought to do?
It may have been in part because of such obscurities in Radcliffe-
Brown's handling of the concept of symbolism that a kind of hiatus
developed in his use of the term. In 1926 and again in 1931 he wrote
of the rainbow serpent in Australia - as a large mythical reptile
found very widely as a concept identified with the rainbow, asso-
ciated with waterholes and rain-making, and playing a part in initia-
tion rites (cf. Elkin, 1954, 216, 220-1). He characterized it as 'the
most important representation of the creative and destructive power
of nature' - but he did not call it a symbol, as he did in his Frazer
Lecture on taboo later. Again, neither in an article of 1929 on the
sociological theory of totemism nor in his classic study of the social
structure of Australian tribes (1930) did he mention symbolism
specifically; he used the term 'representation' or 'emblem' instead.
In his discussion of taboo (of 1939), however, he was considering
the relation of ritual values to the essential constitution of society,
and here he introduced the concept of symbolism once again. Ritual
acts, he held, differed from technical acts in having some 'expressive
or symbolic' element in them. The equation between expression and
symbol which he used in his Andamans study re-appears. He also
gave a further twist to his argument by stating: 'I am using the words
symbol and meaning as coincident. Whatever has meaning is a
symbol, and the meaning is whatever is expressed by the symbol'
(1952, 143). Such formalistic statement is very shaky. If anything
that has meaning is a symbol, and if ritual acts are to be distinguished
from technical acts by their symbolic element, then by inference,
technical acts, being non-symbolic, have no meaning - which is
surely not what Radcliffe-Brown intended. And if symbol and
meaning are coincident, then how can one infer the meaning of a
dance from its symbolism, or speak of 'symbolic meanings', which
implies that there are other kinds of meaning which are non-
i38 SYMBOLS
symbolic? RadclifFe-Brown seems in fact in his definitions to be

oscillating between a view about all signs (all meanings) and a view
about a particular class of signs (with what he regarded as expressive
meanings, that is not primarily concerned with change in physical
relationships). His discussion of how to determine the meaning of
'rites and other symbols' seems to show that it was really the latter
with which he was concerned. He regarded avoidance of personal
names (as in the Andamans) as a 'symbolic recognition' of a fact
that a person was not at the time occupying a normal position in the
social life, but had an abnormal ritual status - hence the name avoid-
ance was an expression of a change in the total personality. The
taboos kept by an expectant father when a child was about to be
born, RadclifFe-Brown interpreted as a suitable symbolic expression
of his concern, in the general ritual idiom of the society. It is felt
generally, he argued, that a man in that situation ought to carry out
the symbolic or ritual actions or abstentions. (RadclifFe-Brown was
at pains to state that he was not trying to explain ritual in terms of
human psychology - but here as elsewhere his references to 'feelings
of concern' have a personal psychological component.)
RadclifFe-Brown's general view of symbols did not emerge as any
very systematic theory, though he did envisage such a theory. In his
essay on taboo he put forward the thought that in human society
'social coaptation' depends upon the efficacy of symbols of different
kinds, and referred for the broader implications to A. N. White-
head's observations (1952, 152). Elsewhere he took a more general
stand. In his attempt to formulate what he called a 'natural science of
society' he drew attention to the existence of common symbols in
culture, with common meanings, rendered necessary by the need
for communication. He cited words, gestures, works of art, rituals
and myths as examples. He defined a symbol, again, rather vaguely,
as 'anything that has meaning'. He made certain broad distinctions:
between three levels of symbols - universally human, socially
specific, and individual. He held that not all symbols, especially
personal ones used by 'psychically unbalanced people', necessarily
imply communication, and that such non-communicatory ones are
not 'culture symbols' or 'social symbols' which communicate a
meaning recognized by a group. But his usage was not very precise,
especially when he tried to separate words as signs - 'representations
of phenomenal reality - from words as symbols, involving com-

munication of meaning or regular association with meaning (1957,
99-103). His attempt to use meaning exclusively as the defining
quality of symbol leads to difficulties in his linguistic treatment,
since taken literally he seems to be implying that in many circum-
stances direct, non-meaningful relations could exist between words
and the things they indicated. His failure to examine what kind of
meaning symbols might have - except his rather vague references to
expressiveness and communication - was evident in his lack of
consideration for the problems of how and in what conditions
symbols can change their meaning.
Yet Radcliffe-Brown's statements were important for the develop-
ment of anthropological studies of symbolic behaviour. What he
did in particular was to link theoretical ideas about symbolic process
with actual field observation. Durkheim had interpreted in stimulat-
ing general terms the symbolism of sacred religious objects, and had
done it with the primitive materials with which anthropologists
were familiar. Radcliffe-Brown did likewise, but he added a be-
havioural dimension to the material and a freshness of first-hand
observation which carried conviction. So when he suggested that
the ceremonial behaviour of the Andamanese was a symbolic
expression of basic social sentiments, and that to explain a given
instance one had to look around for analogies over a range of social
situations with a common theme, this formulation (which L. H.
Morgan had sketched in an incidental descriptive illustration) was
immediately put to use by other anthropologists confronted by
formal procedures in a variety of social institutions.
SYMBOLS A N D M E A N I N G : MALINOWSKI
About the same period Bronislaw Malinowski was also concerned

with problems of symbolism. Though he shared many of Radcliffe-
Brown's basic ideas about the nature of social institutions and the
methods to be used in studying them, Malinowski operated in very
different territory. Like Radcliffe-Brown, he tended to take symbol
as a given term or counter in his arguments. But unlike Radcliffe-
Brown, he had great ability as a linguist, was much interested in
problems of language at an analytical level, and had already acquired
140 SYMBOLS
facility in several languages other than his mother-tongue - notably

in French, German, English and Trobriand. His treatment of
symbols was more deeply rooted in the study of language than that
of Radcliffe-Brown, and also developed into a more widely-ranging
theoretical construction.
Malinowski's views about symbols were essentially bound up
with his theory of meaning. His semantic interests were exemplified
by his early contribution to the study by C. K. Ogden and I. A.
Richards on The Meaning of Meaning (1923 - Malinowski's article
was reproduced in a posthumous collection of essays, 1948). On
the basis of a diagram produced by Ogden and Richards, Malinowski
distinguished three forms of symbol: symbol in active relation to
referent (used to handle it); symbol in indirect relation to referent;
and symbol in mystically assumed relation to referent. The first was
concerned with speech in action; the second with an act of imagery
in narrative speech; and the third with the language of ritual magic
based on traditional belief. In this essay Malinowski did not define
symbol explicitly - he merely contrasted it with referent, the Ogden-
Richards term for the object symbolized (1923, 260). Yet he pointed
out that since a significant symbol is necessary for man to isolate
and grasp an item of reality, there is no definition of a thing without
defining a word at the same time (262). He stressed his view that
words are symbols-'the Ethnographic view of language proves
the principle of Symbolic Relativity as it might be called, that is
that words must be treated only as symbols'. (He was arguing here
against the notion still then apparently prevalent, that words could
have meaning in themselves.) 'A psychology of symbolic reference
must serve as the basis for all science of language' (242). Malinowski
argued that 'in the normal use of words, the bond between symbol
and referent is more than a mere convention' (258). He stressed the
significance of speech as a mode of action rather than as a simple
means of communication. He wished to define the meaning of words
by what he called the context of situation, which he identified
broadly with Ogden and Richards's sign-situation.* In his later
linguistic work in the second volume of his study of Trobriand
* This seems reminiscent of the argument of Alexander Bryan Johnson (in a Treatise on
Language) in the mid-nineteenth century, that words have no inherent signification, but as
many meanings as they possess applications to different phenomena. The phenomenon to
which a word refers constitutes the significance of the word (see Kretzmann, 1967, 393). For
agriculture Malinowski used the term symbol hardly at all. A cryptic

reference to 'sos-symbols* appears in the table of contents, but the
word 'symbol' is replaced by 'signal' in the text (1935, Vol. II, 56).
But I doubt if his viewpoint had changed;* it was rather that with
each new presentation he turned to another angle of vision upon his
subject. In Coral Gardens he dwelt upon his idea of human belief in
the mystic and binding power of words, relating this to the develop-
mental use of language both in the relation of the child to its parents
and in that of the member of society to his fellow members. So, in
what was like an echo of Novalis, the early German Romantic, the
pragmatic effectiveness of speech was to be correlated with the
mystical character of the spoken word. But Malinowski took occa-
sion to reaffirm here his debt to Durkheim, saying that the sociologi-
cal explanation to the belief in the mystical power of words is
obviously a reinterpretation of Durkheim's theory that mysticism is
but an expression of belief in man's dependence on society. But
Malinowski regarded himself as reinterpreting Durkheim in empirical
terms while at the same time broadening the Durkheimian analysis.
In particular he criticized the emphasis which Durkheim laid on the
'phenomena of material symbolization'-the churinga, the national
flag, the cross - as generating religious attitudes. Malinowski made a
very important contribution here as elsewhere in his publications,
in demonstrating the profound organizing functions of the symbolic
behaviour of magic, including magical language.
There is no doubt that by modern linguistic standards Malinow-
ski's theory, of meaning of utterance being completely contained
within context of situation, is unacceptable. I think that commenta-
tors have tended to take his specific formulations too literally and
comment on Malinowski's semantic theory see J. R. Firth, 1957; and for general criticism see
T. D. Langendoen, 1968.
* There is some difference of opinion about this. Langendoen has argued that the 'later
Malinowski', a much more radical behaviourist, attached overweening importance to the
contextual theory of meaning, to the detriment of all other determinants. Moreover, he has
stressed what he sees as a shift in Malinowski's position on magical language: in the Argonauts
Malinowski described magical language as a special kind of language unlike ordinary narra-
tive style, whereas in the Problem of Meaning article he regarded it not as a special kind of
language use but an exemplification of the primary use of language, as a mode of action. This
opinion is counter to that of Jack Berry, who in his introduction to the Indiana University
Press reprint of Coral Gardens and their Magic holds that the view of language Malinowski
presented in this work was substantially the same as that contained in his earlier works, with
little or no development (1965, xi). My impression coincides with this view.
142 SYMBOLS
exclusively, and to ignore the value he attached to the 'process of

moulding'-'the effect of traditional cultural norms and modes upon
the growing organism' (1935, ii, 236). I am sure that Malinowski
would have agreed that by this 'context of presumptions', to use
Berry's term, much of the meaning of a word is already determined
before its use in any specific situation. He said so in various ways. As
I show later, he specifically mentions the significance of the group
in transmitting to the individual the elements of symbolism and the
understanding of symbols. 'Society and its component groups are
the carriers of verbal - that is, symbolic - tradition' (1939, 964). 'By
knowledge we mean the whole body of experience and of principle
embodied in language and action...' (1939, 958). By this, as I
understand him, he subsumed the significance of all past contexts in
the determination of the meaning of utterance, and did not for a
moment think that meaning was determined uniquely by present
context as represented by the personal experience of the actors.
What he would have argued, I think, is that meaning is updated by
that experience, given an additive, and so by implication, rendered
liable to modification. What I am sure he would not have accepted is
a statement of the kind made critically by Berry: that since as part of a
'common sense' view of language we can recognize utterances 'out
of context' as meaningful, and can find contexts for them, this seems
to suggest that the meaning of an utterance is not determined by its
context, but rather is 'autonomous'. That an utterance could have an
autonomous meaning would have seemed to Malinowski not com-
mon sense but nonsense - this is precisely what his 1923 article was
arguing against. Malinowski's recognition of the significance of
past contexts in determination of meaning - the traditional, trans-
mitted elements - has been noted by Langendoen, who complains
however that this broader notion of context as he finds it in J. R.
Firth's writings, is unusable - 'nothing can be said about it in any
relevant way' (1968, 45). Linguistically, perhaps - but Langendoen
makes no reference to Malinowski's article of 1939 where he dis-
cusses some related general ideas. What Malinowski made no allow-
ance for, perhaps because he regarded them as too speculative, were
determinants of meaning arising from phonological characters and
elements in basic patterns of thought. I can see that to linguists his
theory of meaning was too specific at one level, too general at
P R O B L E M S OF SYMBOL T H E O R Y 143
another, with his formulations too loosely expressed to serve as
a basis for more sophisticated analysis. As Berry points out, Malinow-
ski usually was dealing with speech rather than with language, and in
rejecting a distinction between them tended to confuse levels of
analysis. Berry calls attention to Malinowski's lack of linguistic
training, and in a sense he could be described, I think, as a parlist
rather than a linguist.
Yet in conceding these linguistic criticisms, I think there is still
something to be said about Malinowski's semantic theory. In the first
place, some of the critique springs from a basic difference of assump-
tion about the nature of language; in the conflict between behaviourist
and mentalistic views some of the arguments over Malinowski have
been decided in advance, in what is a kind of cyclical movement.
Then, Malinowski's views about meaning in language are in line
with his views about history, which have been misrepresented on
occasion, as being a denial of the significance of the historical dimen-
sion in the interpretation of social facts. Here again comes a question
of assumption - as to the relative significance of an event as a
unique occurrence or as one of a class of occurrences, the problem
which has bedevilled historians and philosophers through the
ages. Malinowski, in the light of his reading of anthropology and
philology, and of his own personal experience, saw how greatly the
interpretation of linguistic material had been impoverished by not
considering the verbal framework of utterance (such as the alterna-
tion of dialogue), the circumstances which led to the utterance, and
the non-verbal behaviour associated with it. It was not that he wanted
to deny the significance of historical factors; he wanted to insist that
they be given their full value at the moment of exposure. Malinow-
ski viewed each occasion of utterance as involving a creative act,
behaviour in which the past was transformed into the present; so
merely to describe the past was not enough to interpret the trans-
formation. There is no doubt that despite its weakness in linguistic
theory, Malinowski's conception has been of great use to field
anthropologists.
As regards symbolism, however, it is rather different. Berry has
argued that it is in keeping with Malinowski's ethnographic theory
that he should have emphasized the pragmatic functions of language
at the expense of the symbolic. But while his theory of meaning was
144 SYMBOLS
inadequate as regards the determinants of the significance of symbols,

I doubt if it is enough to argue, as Leach has done, that Malinowski
undervalued the significance of symbolic statement and therefore
the significance of communication. I think rather that Malinowski
gave all verbal communication a symbolic status (all words are
symbols), but in strongly emphasizing the expressive value of
utterance, failed to make a sufficient difference between the kinds or
levels of symbolic statement.
But Malinowski's interest in symbolism was not so much in
symbolic forms as such, their classification and their interpretation,
as in the process of symbolization in its bearing on the formation and
operations of culture. Again he approached this through language,
exemplified by his treatment of myth (1926). Following an explicit
Trobriand classification (1926, 35) he saw myth as sacred story told
with justificatory purpose. He flatly denied that such myths had a
symbolic character, because this suggested to him the romantic
theories of the nature-mythology German scholars, such as Ehren-
reich or Frobenius, who, as he saw it, made primitive man construct
'symbolic personified rhapsodies'. His examination of myths in
their social context brought out the way in which themes of impor-
tance to the workings of Trobriand society were expressed. He
exaggerated the opposing position, and he was unduly suspicious
of'symbolic interpretations' in terms of'hidden realities' (1926, 79).
He was also inconsistent, in denying to the language of myth the
term symbol which he had applied to words in general. What he
was concerned to emphasize was the overt practical social referent
of myths. But he recognized that they could represent at the ap-
parently trivial level of human error and guilt, or of mischance,
great elements of fate and destiny - as the Trobriand myths of death
and the after-life had a 'deeper metaphysical reference' than the
myths of clan origins (1926, 103). He also stressed their coherence
as a unit.
Malinowski's relatively late article on the group and the individual
in functional analysis contains a more general explicit statement of
his theory. He assumed the symbolic character of language: 'since all
rules and all tribal tradition are expressions in words-that is,
symbols-the understanding of social organization implies an
analysis of symbolism and language' (1939, 940). In discussing the
routine procedures of eating, involving table, kitchen etc., he began
'the symbolic expressions here used - table, kitchen....' In this
Malinowski simply picked up his much earlier analysis of meaning
in language. But he went on to give what he called 'the cultural
dimensions of symbolism5. This was a very general statement.
'Symbolism is a component of human culture, with language as its
prototype.... Symbolism must make its appearance with the earliest
appearance of human culture. It is in essence that modification of
the human organism which allows it to transform the physiological
drive into a cultural value' (1939, 955). Malinowski argued that the
first discovery and use of an implement already implied the birth
of symbolism. Incorporation of the implement into the human
sphere of interest and transmission of the idea of its use implied
recognition of value; the deferred indirect satisfaction of need in-
volved the mechanism of symbolization. The 'transference of the
physiological urge on the secondary reality', as the urges of hunger,
sex or security on to objects which could satisfy them, was in its
essence symbolic. The social setting is indispensable, because it is
the group which maintains and transmits the elements of symbolism,
which trains each individual and develops in him the knowledge of
technique, the understanding of symbols, and the appreciation of
values (1939, 956). But Malinowski went further-he not only
traced the symbolic process to the origins of culture, but also gave
reasons why symbolism should occupy such a basic position. Early
human beings used language and symbolism primarily as a means of
co-ordinating action or of standardizing technique and providing
rules ('imparting prescriptions') for industrial, social and ritual
behaviour (1939, 957). So, symbolism was the means in the very
beginning of human civilization, of mastery of implements, produc-
tion of goods, and incorporation of achievement into a permanent
tradition (1939, 964).*
* Just after the war M. M. Lewis, a philosopher of education with a behavioural approach
and interested in the social nature of language, stressed the importance of contextual inter-
pretation. Broadly in line with C. K. Ogden and I. A. Richards, Lewis specifically caught up
Malinowski's ideas, using his data from the Trobriands and generalizations from A Scientific
Theory of'Culture, with other ethnographic materials, to illustrate an analysis of group com-
munication. Lewis held that 'mind is behaviour mediated by symbols'. He stressed that in
group behaviour 'group-remembering, group planning, group feeling and willing - all these
are modified by the existence of some form of symbolic communication within the group*
(1947, 94). In further consideration of the significance of symbols Lewis pointed out the
growth of what he called 'allusive power' in common verbal symbols, through the spread of
146 SYMBOLS
Malinowski's outline reconstruction of the role of symbolization

as a basic process in culture formation seems to me quite plausible.
Moreover, I think that much modern anthropological treatment of
notions of basic human thought makes much the same kind of general
assumption about 'symbolic function'. But it is a speculative theory,
and a mere sketch. Then on the more critical side I think one might
challenge Malinowski's statement that 'symbolism from its very
inception had to be precise' in the sense that it had to provide a
correct formula for the permanent incorporation and transmission of
the cultural achievement, that it had to be effective in transferring
the drive of the physiological need to the given object which could
satisfy it (1939, 956). This seems to miss out the possibility of that
element of ambiguity characteristic of much symbolization - and in
which lies risk of failure as well as function of stimulation. The ex-
position of Malinowski's theory of symbolization given in his
A Scientific Theory of Culture (published posthumously) added
little to the corpus of his ideas. Again the text argued very strongly
that symbolism is an essential ingredient of all organized behaviour,
that it is a modification of the original organism which allows the
transformation of a physiological drive into a cultural value. It was
argued too that a symbol could be physically recorded, described
and defined in terms of its 'pragmatic context' (1944, 23-4,132). The
postulate was also stressed that a verbal or other symbolic act only
becomes real through the effect it produces. On this point Malinow-
ski has been criticized, with some justification, by Talcott Parsons
(1957, 61) for taking altogether too weak a view of the significance
of symbolic interaction as a determinant of this context, in the
general theory of the social system. Malinowski was relatively un-
interested, also, in problems of symbolic forms and their trans-
formations, and of the comparative structure of symbolic systems,
which have become increasingly prominent in social anthropology.
In brief, in the recognition of the significance of symbolic
behaviour, both verbal and non-verbal, both Radcliffe-Brown and
Malinowski made specific anthropological contributions. Both
patiently worked out from their field materials an interpretation of
modern communications; and in a chapter on language and social conflict he discussed the
position of Blacks in the United States, including the 'condensation', 'displacement* and
'allusion' in verbal symbols for Negro.
symbolic concepts and behaviour - RadclifFe-Brown those of myth
and life crises among the Andamanese, Malinowski those of myth,
kinship and ritual exchange among the northern Massim. They
demonstrated the use of different methods - RadclifFe-Brown that of
comparative inference, Malinowski that of collation of speech and
behaviour in complex sequences of operations. And more generally,
whereas RadclifFe-Brown focused his attention on problems of
symbolism in ritual, Malinowski pioneered in more speculative
theory of the role of symbolization in language and the formation of
culture.
DEPTH PSYCHOLOGY AND ANTHROPOLOGICAL STUDIES

OF SYMBOLISM
Side by side with the development of ethnographic study of symbols

in ritual, art and social institutions, and with refinement of theoretical
insight into the sociological significance of symbols came the
emergence of a specifically psychological focus on the subject. This
had a long ancestry, to mention only the attention paid to the inter-
pretation of the symbolism of dreams by G. H. Schubert (1814),
K. A. Schemer (1861) and J. Volkelt (1875). But in the beginning of
the present century it was the approach of Sigmund Freud that began
to attract most interest in intellectual circles. Anthropologists for the
most part maintained a stout resistance to Freudian ideas about
symbolism - Frazer, as Malinowski noted, being one of these. But in
Britain and in the United States some began to champion or at least
to experiment with such novel ideas.
Freud's contribution to the understanding of symbolism was as
startling as it was novel. As an important part of his work in the
study of neuroses he wished to interpret - as he said, discover the
hidden meanings of- the symbolic forms in which his patients ex-
pressed their anxieties. Dreams were very important in this. As he said,
dreams are not in themselves social utterances, not a means of giving
information, but his aim was to turn them into utterances of value, to
treat them as fully valid psychical acts. He pointed out that the use
of the symbolism of dreams was not the discovery of psychoanalysis,
priority in modern times being due to K. A. Schemer (1861) who
particularly stressed the significance of symbolic representations of
148 SYMBOLS
the whole human form (1937, 221, etc.; 1952, 159, 166).* What
Freud did in psychoanalysis was to expose particularly the relation
of the dream symbolism to the latent dream-thought and the
operations of the unconscious mind. His technique of dream inter-
pretation included a review of the chronological order of dream
events, of the 'day's residues', and of markedly striking events in the
dream; as well as a patient examination of the 'free associations' of
the dreamer. Then if in proceeding to an interpretation of the latent
dream-thoughts resistance was met, this was a sign of conflict
which had to be investigated. The interpretation as Freud admitted,
rested on certain postulates about the meaning of dream symbolism
and its associations, but while more cautious than many of his
critics allowed, he held that most of these postulates were justified
by the analytical evidence. For instance, he held that the number
of things represented symbolically in dreams is not great-the
body, parents, children, siblings, birth, death, nakedness, sexual
organs and processes - and that a large part of the symbolism is an
unconscious expression of the sex impulses. But he pointed out that
he had never put forward the thesis that all dreams are of a sexual
nature. And while he admitted that why certain objects have become
male sexual symbols is not easy to understand, he maintained that it
was unquestionable that hats and cloaks were such - and even argued
that this was backed up by some experimental data! Freud also
pointed to the widespread nature of symbolic usage, in myth, jokes,
poetry, religion, art and language in general. He held that light could
be thrown on such phenomena, especially myth, by the interpreta-
tion of the symbolism of dreams, but he emphasized that the inter-
pretation is by no means confined to sexual themes.
Freud's specific interpretations of symbolism were clearly culture-
bound. Obviously, in the ethnographically 'primitive' fields, um-
brellas and revolvers could not be expected as penis symbols, nor
apples and peaches as breast symbols; nor is dying likely to be sym-
bolized by travelling on a train. Freud never really faced this anthro-
pological dilemma, as is evidenced by the way he blithely cites a
Beduin marriage rite with a cloak as evidence which 'I hope will
* Schemer examined general characteristics of dreams, for example as a 'field for strife', and
as reproductive, a source of originality. Dividing dreams into eleven classes, he considered
what he regarded as their basic symbolic formations. Schubert and Volkelt are also cited by
Freud.
impress you5 that in a (Western) woman's dream a cloak stands for
a man (1964, 24). But Freud was always less interested in the specific
symbol than in the mental process it represented. So, it could be
argued, the equivalent pointed, rounded, covering objects could
occur in any culture to serve as sex symbols. Moreover, Freud
himself emphasized in various contexts the major importance he
attached to the study of the 'dream-work'- the process by which the
latent dream-thoughts are transformed into the manifest dream. He
isolated processes of condensation, displacement and secondary
elaboration as especially relevant for understanding of the symbolism.
But he looked back continually at the clinical problems so the sym-
bolism was significant for him as a factor in dream-distortion, which
in turn gave clues to the nature of neurosis. Indeed he sometimes
went so far as to say that the mechanism of the dream-work is a
kind of model for the formation of neurotic symptoms, or even that a
dream itself is a neurotic symptom, occurring in all healthy people. In
the early part of the twentieth century, Freud's views on symbolism
were among the most articulate, forceful, systematic and theoretical,
so it became difficult for the more thoughtful anthropologist to ignore
them, however much he might criticize their sweeping assumptions,
lack of cultural sophistication and dogmatic presentation.
One can only indicate, not summarize, the effect of Freud's
views upon social anthropologists' concern with symbolism in
the early part of this century, and I mention only a few points.
Concepts of deep, unconscious motivation existed before Freud, but
it was his brutal insistence upon the force and the generality of such
motives that led to an almost unwilling recognition of their signifi-
cance. Concepts of manifest and latent content, resistance, am-
bivalence and repression exemplified the need to interpret behaviour
with an eye to more than its superficial qualities. The frank, overt
discussion of sexual elements in dream and neurosis reinforced the
rather apologetic examination of phallic cults, premarital intercourse,
ritual defloration and other exotic material which anthropologists
had felt it necessary to present in order to convey a realistic picture
of the cultures they studied. The focus on nuclear family as the core
of social action and basic beliefs sent anthropologists back again to a
scrutiny of detailed behaviour and ideas over a range of institutions
from child-training to the role of the mother's brother.
150 SYMBOLS
In the British field the anthropologists first most affected by

Freudian views were W. H. R. Rivers and C. G. Seligman. In
addition to their early work with Haddon in the Torres Straits,
both Rivers and Seligman had done field research independently
elsewhere; although they had devoted much attention to rites and
ceremonies, their interest in symbolism as such had been slight.* But
their interest in Freudian concepts was aroused primarily by their
practical concern with neurotic ('shell-shocked') patients in hospital
during the First World War. Rivers became convinced of the signifi-
cance of unconscious elements of the mind in understanding the
condition of these soldiers and setting up therapy for them. He
carried this interest through into some theoretical examination of
dream and other mental process and discussed briefly some aspects
of symbolization in Melanesia (for example 1918; cf. 1920, 159-69).
Seligman continued this approach, and after the death of Rivers in
1922 he took up the task of advocating psychological and particularly
psychoanalytical methods of inquiry in what were considered con-
ventional anthropological fields. He applied his ideas especially to
the interpretation of the symbolism of dreams, on a comparative
basis, and stimulated other anthropologists in such studies.f From
his medical training, Seligman was able to appreciate implications of
Freud's work which could escape anthropologists in general. On
the other hand he adopted a somewhat uncritical attitude towards
psychoanalytical theory, including the central problem which faced
anthropologists interested in it, namely the translation from indi-
vidual to social dimension.
For while psychoanalysis offered a powerful instrument for
anthropologists in the study of symbolism, it also raised difficulties.
* For example Seligman discussed in detail the former Vedda custom of a groom giving a
lock of hair to his bride, questioned its ornamental value, stressed how the bride prized it, but
did not examine the symbolism (1911, 98-100). Rivers likewise described the Toda bow-and-
arrow ceremony at a woman's first pregnancy, to establish the social fatherhood of the child,
but did not explore the symbolic aspects. His only mention of symbolism I can find among the
Todas refers to the swinging of a corpse over a fire, which 'would be symbolic of its destruc-
tion by fire', with the 'symbolic burning* having the advantage of not destroying valuable
property (1906, 319-22, 363). (Seligman's reference to an arrow symbol is a quotation from
an earlier writer.)
f Jackson Steward Lincoln, psychologist and anthropologist, who had also been psycho-
analysed, drew considerably upon Seligman's ideas in his The Dream in Primitive Cultures,
1935. My own study of dreams in Tikopia was in response to Seligman's interest (Firth,
1967b) but was not along lines directly suggested by him.
First, if all symbolism is unconscious, as orthodox theory seems to
require, in order to interpret it, an anthropologist is involved in
much more manipulation of spoken evidence and more imputation of
motivation, than he ordinarily regards as justified. Then there is the
question of sexuality. Few field anthropologists would wish to deny
the great importance of sexual factors in the whole field of symbolic
behaviour they study, from poetry to magic and myth recital. But
as Rivers argued, they tend to identify referents of symbols in social
relationships as well as in sexual relationships, and even when in
sexual relationships, they see other than genital elements as of prime
relevance. Again, in his Totem and Taboo (1912) Freud had put
forward views that were naive and distorted by most anthropological
opinion: not only had he taken up the discarded primal horde
hypothesis of J. J. Atkinson, but he also argued that there was
analogy between much of the social, especially ritual, behaviour of
primitive peoples and the individual behaviour of neurotics. Taken
in conjunction with Levy-Bruhl's parallel assertion of the pre-
logicality of the primitive,* there is little wonder that anthropologists
reacted vigorously against this last view.
In this turbulent intellectual situation Freud's views on symbols
had defenders and interpreters in the anthropological sphere. For
instance, working from literary sources, J. S. Lincoln regarded the
Freudian theory of symbolism as a most useful and accurate means
of approaching latent dream meanings, and interpreted Amerindian
data accordingly. A major, if idiosyncratic, figure here was Geza
Roheim, by training a psychoanalyst, who also carried out field
research of anthropological type in Central Australia and Nor-
manby island (near the Trobriands) and among the Yuma of Cali-
fornia. R6heim made a special study of shamans or medicine-men,
including their dreams. He was at pains to emphasize that most of
these men whom he met were not neurotic; on the contrary they
were healthy, capable hunters and leaders, whose symbolism might
be conscious as well as unconscious. But their dreams revealed to
R6heim's interpretation of their latent content some of the classic
analytical elements. When a Ngatatara would-be medicine-man
* Freud used Le Bon's Psychologie des Foules (1895) as his source for identification of
group mind with primitive, both characterized by lack of conflict in the presence of logical
contradiction. He went on to equate this with aspects of unconscious mental life of individuals,
including neurotics (Freud, 1965, 15).
152 SYMBOLS
begins by eating his totem, a yam, that is 'the oral introjection of the
father'. Since the dream that followed was very like the conception
dreams of women (in which the latent content was held to be coitus
with the father), except that mouth was substituted for womb, 'we
may therefore suspect a female attitude with regard to the father as
latent in the mental make-up of the medicine man'. Similarly,
R6heim saw the 'latent significance' of agriculture, in terms of
Trobriand and Normanby island myth and magic, as being symbolic
body destruction and restitution fantasies connected with the mother.
He acknowledges that anthropologists will probably object very
strongly to all this, saying inter alia that yams are cultivated because
of their usefulness, and these are only the unconscious aspects of an
economical, purposeful activity. But he sternly refutes such argu-
ments, holding that all the occupations in the 'seemingly practical
core of group-living' are more or less distorted or projected equiva-
lents of the infantile situation (Roheim, 1971, 4-6, 69-76, 93, 127).
Roheim's general conclusion, a modification of Freud's view, is that
culture is really composed of psychic defence systems against
anxiety, and that therefore specific cultures are structurally similar
to specific neuroses (1971, 106). But while many anthropolo-
gists have not accepted the general propositions resulting from
R6heim's filial piety, some have found his specific interpretation of
symbols in their manifest field context as very suggestive.
In the inter-war period other anthropologists such as Kroeber,
Sapir, Goldenweiser, Prince Peter of Greece and Jules Henry en-
gaged with psychoanalysis, either as a personal experience or as an
intellectual problem, to a marked degree, and some of their findings
emerged in reflections upon symbolism. Others, such as Linton and
Malinowski, were more guarded. In The Study of Man (1936), a
work which effectively began in anthropology the fruitful study of
status and role, Linton pointed to the widespread though not univer-
sal development of symbols in society, and mentioned the flag, the
cross and the swastika as examples. He stressed the culturally-
patterned character of symbols and their frequent lack of intrinsic
importance. But he adopted a curiously limited viewpoint in holding
that when we find in uncivilized societies symbolic meanings
attached to animals, objects or natural phenomena we usually call
them totems - ignoring a great range of religious and political
symbols to which the totemic label was not attached (1936, 424-6).
Linton stated specifically that he had not been impressed by the
orthodox Freudian explanations of cultural data he had read in the
period between his expeditions to the Marquesas and to Madagascar
in the 1920s; he had adopted instead Abraham Kardiner's view of
'basic personality structure' as giving a focal point for social inte-
gration (1939, xviii). By implication then, Linton was mainly
concerned with other aspects of psychological process than the
symbolic.
Malinowski's reaction was of a different order. Intrigued intellec-
tually by the complex arguments about mental process, in so many
ways an advance in sophistication on the psychology of Wundt,
and even of Shand, with which Malinowski was well acquainted, he
was repelled by the naivety and grossness of some of the Freudian
interpretation in the cultural field. Malinowski was convinced of
the significance of unconscious factors in mental process; he also
accepted the view that such factors came to overt expression in
symbolic forms such as ritual and myth. But his concern for the
empirical evidence, his belief in the importance of cultural condi-
tioning, and his own theses about functional inter-relation of ele-
ments of culture made him question the universality of some of the
cardinal themes of the Freudian argument. He held that the symbolic
form in which the fundamental conflicts of family relationships would
emerge must be directly correlated with the particular structure of
family, itself an outgrowth of complex historical conditions. In his
own previous work on the family among the Australian aborigines
he had spent considerable analytical skill in disproving notions of
indiscriminate or communal sharing of wives and in bringing out the
significance of the forces of co-operation in marital and filial rela-
tions. His Trobriand fieldwork had reinforced these ideas. So in the
Trobriand matrilineal descent system, with denial (he called it
ignorance) of physiological paternity, and overt assumption of
authority by mother's brother, he saw the fundamental struggle
within the family as between mother's brother and sister's son, not
between father and son, and incestuous longing not between son and
mother but between brother and sister. He indicated that this came
to overt expression in an appropriate type of myth. In The Father in
Primitive Psychology and more extensively in Sex and Repression in
F
154 SYMBOLS
Savage Society (both published in 1927) Malinowski indicated his

basic acceptance of the significance of the unconscious in the develop-
ment of symbolism in myth. But he also argued stoutly for the rele-
vance of social factors in the channelling of the expression of the
unconscious. The exponents of psychoanalysis, led by Ernest Jones,
rejected these arguments on the ground that the data Malinowski
presented could be interpreted more simply as displaced versions
of Oedipal material. Despite the authority of the mother's brother, it
was the child's father that was sleeping with its mother, and there-
fore it was upon him that the male child's hostility would be primarily
directed. Perhaps orthodox psychoanalysis could not have held
otherwise. But it seems to me now, as it did during the controversy,
that psychoanalysis lost an opportunity to explore in more subtle
detail the relation between sexual (including genital) factors and
social factors in a situation where admittedly the structure of family
relationships was regarded as being fundamental to the develop-
ment of the individual psyche, with all its symbolic expressions. But
lacking such encouragement Malinowski dropped this particular
line of symbolic interpretation which became linked with 'revision-
ist' argument in psychology.*
Edward Sapir's treatment of symbolism came from another angle.
His interest in linguistics combined with his receptivity to Freudian
theory and his own great sensitivity gave his views more penetra-
tion than those of most of his colleagues. In early papers and in his
concise but suggestive review of the field in the Encyclopedia of
Social Sciences (1934) he put forward views which have served many
later anthropologists as a source of ideas. Taking a broad scope for
that period Sapir argued that all culture is heavily charged with
symbols, as is all personal behaviour. (He did not go so far as to
hold that culture is a system of symbols.) Even comparatively simple
forms of behaviour are far less directly functional than they seem to
be (a protest against Malinowski's influence), but include symbolic
elements. Even an elaborate, well-documented scientific theory may
be little more than a symbol of the unknown necessities of the ego.
* A thoughtful account of Malinowski's debt to psychoanalysis is given by Meyer Fortes

(1957) - though I think he overdoes the argument in maintaining that it was the notion of
Oedipus complex that gave Malinowski inspiration for the main features of his kinship
theory. See also the stimulating analysis by Anne Parsons, 1969, 3-63.
'Scientists fight for their theories not because they believe them to
be true but because they wish them to be so/ he wrote with ironic
exaggeration (1949,566-7). Beginning with a discussion of problems
of referential and expressive symbolism in language, he made an
attempt to prove that there was an unconscious or intuitive logic
which makes phonetic systems constellate. He pointed out that a
symbol has two constant characteristics. It is always a substitute for
some more closely intermediating type of behaviour, so all sym-
bolism implies meanings which cannot be directly derived from the
contexts of experience. It also expresses a condensation of energy, its
actual significance being disproportionate to the apparent triviality
of meaning suggested by its outward form. So Sapir distinguished
two major types of symbols: referential symbols and condensation
symbols. In referential symbols, a primary feature was their economy
for purposes of reference - such as national flags, or indeed oral
speech. Condensation symbols were equally economical, but their
main feature was that they were substitute behaviour for direct
expression, so allowing easy release of emotional tension in con-
scious or unconscious form, and diffusing this emotional quality
to types of behaviour or situation seemingly far removed from the
original meaning of the symbol. The distinction is rather crude, and
deals basically with the difference recognized between the need to
handle objects of the external world, and the need to handle the
problems of the self. But the notion of condensation symbols having
an unconscious spread of emotional quality, and of peculiar potency,
even danger (with an obvious debt to Freud's theory of dreams)
has clearly found continuity in the concept of condensed symbols
used by Victor Turner and Mary Douglas.
Psychoanalysts with anthropological training seem to have found
little difficulty in adapting Freudian views on symbolism to their
needs (for example Bott, 1971). But for anthropologists looking to
interpretations of symbols in terms of unconscious elements Freud's
treatment has often seemed inadequate. In orthodox Freudian
theory symbols are the product of unconscious repressed
attitudes of identification, though modification of such views has
been put forward (for example by Roheim). But Freud from his
clinical standpoint tended to speak of symptoms rather than of
symbols, and he was interested primarily in the 'symptom-forma-
156 SYMBOLS
tion' rather than in the character of the symbol and its object.*
In a sense, since symbols were a product of repression, and Freud's
aim was to relieve people as far as possible from the suffering of
repression by leading them to understand its roots, he wished to
free them from the tyranny of their symbols. In the history of depth
psychology Carl Jung took up Freud's concepts and made a virtue
out of necessity; he embraced symbolism. As Rieff has put it, Jung
transposed the forms and figures of earlier systems of symbolism
into psychology. So in the very symbols from which Freud wanted
to free mankind, Jung saw the principle of man's salvation (Rieff,
1963, 17). Freud himself saw this inversion very clearly-and
bitterly. In his History of the Psychanalytical Movement he discusses
how the 'symbolism in the language of dreams' slowly became
clear to him, and how Jung distorted the interpretation of dreams by
representing them as means for producing attempted solutions of
'the life-task'. He objected strongly to what he saw as Jung's concept
of 'symbolic' as equivalent to 'having no real existence', and to
replacement of 'all that is disagreeable in the family complexes' by
bland abstract ideas of a 'merely symbolic' order. Freud accused
Jung of 'approximation to the demands of the multitude' and of
creating a new religious-ethnical system, ignoring the sexual roots
of the symbolic forms he studied (Freud, 1963, 53-4, 92-8). But he
did not foresee how far in modern times Jung's views on symbolism
would 'justly claim to be a liberation for youth' (Freud, 1963, 92).
Indeed a considerable vogue for Jung in student circles, including
anthropologists, would seem to be linked with a general search for
reassurance in personal and social life rather than with a specific
intellectual curiosity as such about symbolism and analytical proce-
dures.
Jung's contributions to the study of symbolism have been mani-
fold, stimulating, doctrinaire, sometimes portentous and obscure.
His Psychological Types', with its notions of introversion and extra-
version, was an important and analytically novel work in the 1930s
for anthropologists, though there was much disagreement with it.
Jung's 'analytical psychology' attempted to reach deep layers of the
human mind by focusing less on logical formulation than on sym-
* For example Freud (1921) 1965, 17, 49, 95; cf. also his view of people's apparently pur-
poseless, chance acts as 'symptomatic actions' (1914) 1963, 117.
bol, and by regarding the latter as the product of a range of rela-
tively impersonal factors, not of personal family relations in the
Freudian manner. For Jung the symbol was always extremely
complex, neither rational nor irrational but with one side that accords
with reason and another that is inaccessible to reason; stimulating
sensation as well as thought. In the creator of a new symbolism the
highest mental achievement is demanded as well as the lowest and
most primitive motions of the psyche. Jung was prepared to admit
consciously-designed material as symbolic - even a scientific theory -
but he focused on the referent of the symbol as being relatively (or
at times, he said, essentially) unknown. A symbol for Jung was the
best way of expressing a thing, the nature of which is withheld
from present knowledge. So for him the Cross as a representation of
Divine Love is not a symbol but semiotic (meaningful directly),
because Divine Love is an apt description of the fact; whereas that
description of the Cross is symbolic which puts it above all imaginable
explanations, regarding it as an expression of an unknown and as
yet incomprehensible fact of a mystical or transcendent character
(1926, 601-10). Important in such a scheme is the well-known
notion of the archetype - a symbolic image of great power, part of
the psychic content that has not yet undergone conscious treatment,
but which can emerge from the unconscious with dynamic force,
and be used as part of the redemptive process which the mind fashions
for itself when the proper conditions are present. The archetypes
are the content of the collective unconscious, and as such are uni-
versal, not individual in character. Jung has argued: 'a symbol is not
an arbitrary or intentional sign standing for a known and conceivable
fact, but an admittedly anthropomorphic - hence limited and only
partly valid - expression for something suprahuman and only partly
conceivable* (1958, 152). Such a characterization of symbol can
appeal to those of a religious or mystical turn of mind, as Freud
saw, and it has drawn criticism from those psychologists who feel
safer with a personal unconscious than a collective unconscious.
As J. A. C. Brown has put it, they regard it as 'certainly unorthodox
in science to describe the partly known in terms of the wholly
unknown' (1961, 44-5). But Jung's influence has grown neverthe-
less.
One of the earliest anthropologists to be influenced by Jung was
158 SYMBOLS
John Layard, who worked closely with Jung in Zurich. Layard

previously had done field research on Atchin and Vao, two small
islands of Malekula in the New Hebrides, in 1914-15, under the
stimulus of W. H. R. Rivers; he had made an elaborate study of the
ritual of the local graded societies, in which he saw evidence of the
spread of megalithic culture, probably from Egypt. He interpreted
intricate sand drawings as labyrinth designs, and re-birth symbols,
from this point of view. But he also made some interesting observa-
tions on the symbolism of the rites, pointing out how height was
symbol of aspiration, both social and spiritual, and was often asso-
ciated with light, of sun and stars as male representatives. One of
the major symbolic figures in the rites was the hawk, also a symbol of
aspiration. Men imitated the fluttering of hawks on critical occasions,
hawks' feathers were used as symbols of rank, and giant hawk
banners were set up in the form of kites (1942, 732, 734). Layard's
opinion that these rites were comparable with the 'mysteries' of
classical antiquity, and of the Christian church, links this study with
earlier analyses by Jane Harrison and other Cambridge scholars. But
Layard's practice as an 'analytical psychologist' led him further in
the study of symbolism along Jungian lines. In The Lady of the
Hare: Being a Study in the Healing Power of Dreams (1944) he
examined a series of dreams of a patient, regarding them from a
therapeutic point of view as positive achievements. Having dis-
covered in one the critical sacrificial role of the hare, he then engaged
in a comparative scrutiny of data on the symbolic (archetypal) role
of the hare in China, Africa, etc. With the hare as the major
example., Layard's conclusion was that the dream-mind is a
storehouse of ancient symbolic wisdom, unsuspected by the most
learned scholars of our time2 and also quite unknown to the
dreamer (1944, 101). This is a generalization directly in keeping
with Jung's views, some aspects of which will be raised further in
the next chapter.
INTEREST IN SYMBOLIC F U N C T I O N A N D PROCESS
As overt interest in symbolism developed in anthropology more

attention began to be devoted to two themes: the nature of the
symbolization process, and its relation to non-symbolical thought;
and the functions performed by symbols with particular reference to
individual expressiveness and social requirements.
Study of how symbols function in society was tackled in breezy
fashion by R. R. Marett, whose contributions seem somewhat
undervalued nowadays.* Marett operated with symbolism from a
distance, his field observations having been minimal. Enjoying the
Oxford anthropological scene between the wars, Marett had a
scholarly and ingenious mind, though rather unfocused and over-
given to imagery. ('Surf-riding on metaphors' has been my own
image for some of his argument.) Marett had become known for
his concept of animatism, or pre-animist religion, in which man was
concerned with his relation to impersonal forces; this was a supple-
ment or counter to Tylor's definition of religion in minimum terms
as belief in spiritual beings. Marett was also well acquainted with
the work of Durkheim and others of the school of VAnnee sociolo-
gique, as well as with that of Levy-Bruhl. But his own interests
developed in a more pragmatic way. He concentrated on the view of
religion as a kind of simple action-system, and on a major function
of symbolism as a means of action for coping with superhuman
forces. 'Symbolism, in a word, is in its essential function a spiritual
lightning conductor it does some primitive community stricken
with smallpox a world of good to carry out an image of the disease-
demon and douse it in the nearest stream' (1935, 90). Or again, a
symbol allows for substitute action, as against death. 'So long as the
ritual of custom is enacted with dramatic thoroughness, the will is
taken for the deed, and passion exhausts itself on the symbol'
(1935, 188). Sceptical but dynamic; grounded in introspective
common sense rather than in concrete observation, his ideas helped
to carry forward the concept of the pragmatic value of the symbolic
process, though they soon lost momentum through lack of empirical
analysis. That symbols may be used with cathartic effect, however,
is a notion which emerges in various ways in much subsequent
work.
* Another writer often neglected, but whose merits have recently been brought out by
Rodney Needham, is A. M. Hocart. But his observations on symbolism were more closely
allied to the traditional comparative type of study than to the developing social anthropology.
Hocart wrote of the symbolism of settlements, and of'idols', and of symbolic action in ritual,
such as the role of the herald. His observations were pertinent, but obscured for social theory
by his commitment to quasi-historical explanations (1970 (1936)).
i6o SYMBOLS
Other contributions about the same period emphasized the emo-

tional component of symbolic expression. Robert Lowie, in a
chapter on association of ideas in his study of primitive religion
(1936), discussed how colour may come to acquire a symbolic
value, in a manner reminiscent of Radcliffe-Brown's concept of
sensory metaphors. (He cited Radcliffe-Brown's remarks about the
colour red, in the Andamanese context, but warned against a sim-
plicist interpretation.) Lowie referred to the 'invaluable indivisible
emotional experience' which makes a religious symbol intelligible.
Yet though he asserted the unanalysable nature of the symbolic
experience he dismissed the notion of a pre-logical mentality with
which some writers, notably Levy-Bruhl, had linked it. But Levy-
Bruhl himself, after various works on primitive mentality and the
soul of the primitive had established his position, published in 1938
another work directed especially to problems of symbolism. In it
he reiterated his assertion of a primitive type of mind - which he
withdrew soon after, in his (posthumous) Carnets (1949). For the
primitive, Levy-Bruhl argued, the difference between experience and
belief, so sharp for us, becomes almost imperceptible. The primitive
spirit can seek to apprehend this reality, at once visible and invisible,
only through symbols (1938, 170). 'Prefiguration' of the results of
desire, and magical formulae, are forms of symbolic action by which
the primitive translates and attains his object. Is it surprising that
in the eyes of the primitive the symbolic actions have the virtue of
making real what they prefigure? (1938, 294-5). This thesis accoun-
ted for the intimate link between belief in the validity of mystical
experience and confidence in the efficacy of symbolic actions. But
the criticism which it drew was widespread, and I add only a few
comments. First, in some respects the thesis was a highly literary
product; it was very much the view of a European intellectual
writing in vivid prose about the ideas of people whom he had never
seen - they were essentially his ideas of how the minds of'primitives'
worked. His assertions about their mode of mental functioning had
therefore for many field anthropologists an air of unreality - one was
reminded of Ezra Pound's reported comment on James Joyce's
Ulysses -'the book is all language'. Secondly, the theory was really
a statement about a general mode of human thought, not about the
minds of a cultural sector of mankind. As Levy-Bruhl himself came
to realize towards the end (and as some anthropologists had tried to

tell him all along) what is valid in his thesis can apply to civilized as
well as to primitive man. The dividing-line between logical and pre-
logical (non-logical in the sense of uncritical) thinking is to be looked
for, not between civilized and primitive men, but between different
kinds of conditions and wants to which thought is applied. To define
these conditions and wants, not an easy task, is part of the anthro-
pological problem. A third point is that the degree of acceptance of
Levy-Bruhl's propositions has seemed to be to some extent a
reflection of the particular metaphysical position of the various
anthropologists involved - they appear in another guise in formula-
tions of the 'anthropology of experience' which stresses personal
involvement in situations being described or analysed. Finally, what
I think Levy-Bruhl helped to do, even if in a somewhat oblique way,
was to focus the attention of anthropologists on the significance of
symbolic thinking and symbolic behaviour more forcefully than
before. Hence, it would seem, the support which Evans-Pritchard
has tended to give to Levy-Bruhl's views in his examination of
theories of primitive religion (1965, 78-92).
The period of the 1930s saw publication of the first substantial
results from the fieldwork of the generation of social anthropologists
who, coming after Malinowski, had gained much from his technique.
Though they wrote little in the specific idiom of symbolism they had
a lively interest in problems of symbolic behaviour.
Functions of symbolic behaviour in specific social conditions were
demonstrated, for instance, with great penetration by two anthro-
pologists working in very different fields and with different back-
grounds: Edward Evans-Pritchard, an Oxford historian who had
his anthropological training in London under Seligman, and
Gregory Bateson, a Cambridge zoologist trained in anthropology
under Haddon. Both owed much to Malinowski (Bateson in particu-
lar acknowledged this, in his foreword to Naven) but each developed
an individual approach. In early studies both exemplified the signifi-
cance of symbols as paradox. Evans-Pritchard's study of witchcraft
(1937) showed how the incidence of natural forces impacting upon
individuals in Zande society was translated into symbols of personal
attack, believed to operate at an invisible level, with a logic intelli-
gible to ordinary human understanding, once the premises were
162 SYMBOLS
granted. This study has provided a starting-point for many subse-

quent studies of witchcraft and allied phenomena. Bateson's study
of ritualized relations between mother's brother and sister's son
among the latmiil of the Sepik River in New Guinea (1936) was
analytically much more complex, and its value took longer to be
appreciated. But it embodied a great deal of material on the sym-
bolism of role reversal, with an involved theoretical disquisition on
this and cognate subjects. Neither author paid much attention to
problems of symbolism in specific terms. Evans-Pritchard assumed
a sufficient framework of ideas about the subject for his purpose;
Bateson, while taking less for granted, was preoccupied by the
explanation of symbolic behaviour in terms of symmetrical or
asymmetrical (complementary) pattern of social relationships.
Bateson referred in passing to symbols of obvious sexual type -
fishtrap and eel as symbols of vulva and penis - and to symbols which
he thought could be so interpreted - when a great man dies, a
figure is set up to represent him decorated with spears to the number
of people he has killed and baskets to the number of wives he has
had, these being symbols of his achievement. But Bateson then
clearly regarded the 'fascinations of symbol analysis' as a 'pitfall'
distracting him from his major aim of constructing a theoretical
framework for the explanation of the whole of latmiil social pheno-
mena. In his epilogue of 1958 he pointed out that Naven was written
'almost without benefit of Freud' and that he had ignored some very
obvious examples of sexual symbolism in the rites he had been
studying (1958, 282-3). But in another field, that of mental disorder,
Bateson's later studies of symbolism have been of cardinal impor-
tance.
Explicit concern with symbols, in continuity of the Durkheimian
tradition in anthropology, was indicated by Fortes and Evans-
Pritchard in their basic introduction to the essays on African
Political Systems (1940), a work which has had a deservedly marked
influence on developments in political anthropology. They laid
great emphasis on the importance to African society of common
interest in symbols. They held that attachment to these symbols,
more than anything else, gave to African society its cohesion and
persistence. They argued that in the form of myths, fictions, dogmas,
ritual sacred places and persons these symbols represented the unity
and exclusiveness of the groups which respected them. But, said

Fortes and Evans-Pritchard, such things are regarded not as mere
symbols but as final values in themselves. To explain them socio-
logically, such symbols must be translated into terms of social
function and the social structure which they serve to maintain.
Africans have no 'objective knowledge' of the forces involved -
indeed it might well be that if they did understand the 'objective
meaning' of the symbols these would lose their power. The power of
the symbols lies in their symbolic content and in their association
with the nodal institutions of the social structure, such as the king-
ship (1940, 17-18).
In the light of present views it would be easy to criticize this
presentation. There was no definition of what was meant by symbols.
So the inclusion of dogmas as well as myths could blur demarcation
lines. The reference to 'fictions'-which recalls McLennan and
Jeremy Bentham - is also unclear, though presumably it means such
items as fabricated genealogies. There seems to be tautology in
saying that the power of symbols lies in their 'symbolic content'.
And if these myths, dogmas, sacred places and persons are not re-
garded by Africans as 'mere symbols' but as 'final values', what
precisely is their 'objective meaning' which Africans do not under-
stand? What is the evidence that they do not understand that attach-
ment to their symbols serves to support the structure of their society?
But I think a more positive attitude is justified. Fortes and Evans-
Pritchard give no bibliographic references for their statements about
symbolism, but one does not suppose they invented all these ideas
afresh - in fact, debts to Durkheim, RadclifFe-Brown and Malinow-
ski seem clear, and Fortes (1936) had already written on ritual
symbolism of political structure. What the authors did was, in a
necessarily compressed statement, to affirm a series of general
propositions about the character and functions of symbols in a way
which specifically enmeshed them very closely with one particular
aspect of institutions in society, the political. They offered symbols as
a starting-point, particularly in this relatively unexplored field of
political structure and organization, for further investigation. But
they also re-opened some dilemmas - notably that of ascertaining
where the 'objective meaning' of symbols does lie. Can the observer
maintain his patronizing claim to more subtle understanding than the
164 SYMBOLS
symbol-user? Can the symbol-user be fully aware of the sign-func-

tion of what he is using and still treat it as a 'final value5, or must his
awareness of symbolic function destroy his faith and his capacity for
action? It is towards such problems of a dynamic kind that some
modern trends of anthropological interest have turned.
Over more than a century then, the study of symbols, by pre-,
proto- and proper anthropologists had developed greatly: from
descriptive to theoretical treatment; from empirical to abstract
definition; from elucidation by comparative similarity to that by
scrutiny of the social context; from interpretation in terms of indi-
vidual symbols to that in terms of symbol-system. Basic questions
had been raised - about quality of relationship of symbol to referent;
about the validity of distinction between sign and symbol; about the
relative significance of intellectual and emotional elements in symbol
recognition; and of cognitive and expressive features of symbolic
activity. The contribution of other disciplines had made itself felt,
notably of art, philosophy, psychology and sociology. Even if what
had been taken from each by anthropologists had been fragmentary,
enough of a framework had been erected to allow of more searching
inquiries into symbolism on an institutional basis.
Chapter 5
MODERN ANTHROPOLOGICAL
VIEWS OF SYMBOLIC
PROCESS
Is modern social anthropology engaged in a retreat from empirical

reality? We are concerned with 'deep structure' rather than with
content; with models rather than with behaviour; with symbols
rather than with customs. Leaving aside the question of how far this
is a reaction to fundamental changes in the degree to which men feel
they can control their social environment, I think at least three
elements can be identified in this anthropological focus. There is a
search for abstraction as a heuristic device, to eliminate idiosyncratic
variation and perceive general principles more easily. There is the
recognition of observer-effect, by which what purports to be a
record of fact is admitted to be a record of opinion, or at least to
incorporate some quantum of personal interpretation. There is also a
challenge to positivism, whereby from a relatively neutral position
of criticism or from some specific commitment the autonomy, even
priority, of the non-empirical is insisted upon. In all this the language
of symbolism, and inquiry into the nature of symbolism, have come
to the forefront.
A SYMBOLIC IDIOM
What I call the symbolic idiom is widespread in post-war anthro-

pology. In 1949, as an appropriate tribute to the author of works on
primitive symbolism, Maurice Leenhardt wrote of the posthumously
published notebooks of Levy-Bruhl that 'their simple appearance
and richness of content might stand as a very modest symbol of the
philosopher's personality' {Carnets^ Preface, v). At the other end of
the scale of social complexity, Claude Levi-Strauss, in his Introduc-
tion to the work of Marcel Mauss, gave a clue to much subsequent
development in the passage beginning: 'All culture can be con-
sidered as an ensemble of symbolic systems in the first rank of which
165
i66 SYMBOLS
are placed language, matrimonial rules, economic relationships, art,

science, religion ' (1950? xix). In the first Malinowski Memorial
Lecture, warningly entitled 'Rethinking Anthropology', Edmund
Leach propounded the heresy - as he termed it - that the organiza-
tional ideas present in a society should be thought of as a set of
mathematical patterns; that there is a fundamental opposition ideolo-
gically between relations of incorporation and of alliance in kinship
and marriage groupings; and that relations of incorporation are
distinguished symbolically as those of common substance while
those of alliance are viewed as mystical or metaphysical influence
(1961a, 9,10, 21). In an analysis of the nature of kinship in American
society David Schneider begins with the hypothesis that sexual
intercourse is the symbol in terms of which the family members are
defined and differentiated, and grounds this view on the assumption
that American kinship as a cultural system is a system of symbols
(1968, 1, 31). More summarily - just for illustration - Ronald
Berndt, Mary Douglas and Terence Turner have examined aspects of
the symbolism of the human body; Nur Yalman has studied the
symbolism of food and the symbolic equation of eating and sex in
caste relations in Ceylon. Victor Turner has written of the 'forest of
symbols' in Ndembu ritual; Hortense Powdermaker looked at
symbolic meanings in the Hollywood 'dream factory' of the
cinema and in world views of American college youth; Leslie Hiatt
has given an analysis of literary symbolism in Nabokov's Lolita^
which he calls 'a "Freudian" cryptic crossword'.
In all this, symbolic representation has attained an identity and a
dignity lacking before. Symbols have become important, not for
what they represent, but for what they themselves are thought to
express and communicate. They have come to be regarded not as
surrogates or evasions of reality, but as a kind of higher form of
reality. Some partisan interests have been involved in this trend, but
the general mood has been to give symbolism a distinct place in
anthropological study.
As a result, certain basic questions have been sharpened. In the
range of anthropological materials, where does one look for sym-
bolic forms? How does one recognize symbols, by what criteria of
identification? What is the meaning of symbols, in relation to that
which they symbolize, and in relation to one another? What is the
MODERN ANTHROPOLOGICAL VIEWS 167
nature of the symbolizing process as a mode of thinking? And what is

the relevance, and the effectiveness, of symbolic behaviour in rela-
tion to general behaviour at the individual and the social level?
To some extent answers to these questions tend to overlap in the
literature, so I begin with some of the earlier statements.
R E C O G N I T I O N OF SYMBOLIC FORMS
Modern social anthropology has been characterized by a much more

explicit recognition than formerly of the symbolic nature or the
symbolic qualities of the phenomena studied. A keynote for a period
was struck by Edmund Leach in his crisp expression that in his view
ritual action and belief were alike to be understood as forms of
symbolic statement about the social order (1954, 14). Ritual in its
cultural context is a pattern of symbols, to be matched with another
pattern of symbols operated as technical terms by the anthropologist;
their common structure refers to the system of socially-approved
relations between individuals and groups in the community under
study. What ritual performances do is to recall and present in symbol
form the underlying order that is supposed to guide the members of
the community, in their social activities.
Some of the groundwork for such statements can be found in
earlier views. George Homans, in The Human Group, a work which
used much anthropological material and had considerable influence
upon anthropologists of the immediate post-war period, pointed
out as Sapir had done, that a great deal of behaviour in the 'internal
system5 of groups, that is in the expression of the sentiments of
group members towards one another, is symbolic. But - for technical
brevity, he said - he used the terms symbol and symbolic only for
physical objects, spatial relationships and verbal behaviour; he
did not apply them to 'ongoing behaviour'. So, in such terms, a gift
is a symbol of friendship, but the actual process of making the gift is
an expression of the friendship (1950, 137). For most anthropolo-
gists, however, an act of giving, in its formality, its deprecatory signs,
its status consciousness, represents a complex set of social positions
in manner which can fairly be called symbolic.
Already Meyer Fortes, in an elaborate structural study which was
analytically very productive, had given his conception of social
i68 SYMBOLS
symbols (as exemplified in Tallensi totemism) in terms of material

objects set in a complex formation together with their associated
beliefs, attitudes and conventional actions. To this notion of symbol
Fortes attached other criteria: that it had no self-evident utilitarian
or logical significance to the observer or even to its users; that it
carried a strong emotional charge and had the value of a direct moral
imperative; and that its meaning could be inferred from the attitudes
and actions of the symbol users (1945, 136). Certain animals were
endowed with ritual significance, were symbols of ancestors and so
of the moral relations of men in their local connection with the
tracts of ground with which the ancestors had been associated. By
tying together discrete principles of descent and locality, such ritual
symbols also tended to emphasize the functional differentiation of
social groups. 'Totemic and other ritual symbols are the ideological
landmarks that keep the individual on his course (1945, 69,142-4).*
My own treatment of the symbolic significance of ritual and belief
was of the same general order, though less pointed. From rather
casual reference to symbols in pre-war writing (for example Maori
heirlooms as material symbols of exchange of goodwill (1929, 346);
Tikopia kin terms, even hair circlets, as symbols for behaviour
(1936, 184, 271); Tikopia ritual acts as symbols for social rules
(1940, 19)) I had given more systematic general consideration to
symbols in Elements of Social Organisation (1951). As well as referring
to symbolism in art, money, name change, dreams, funerary practices,
millenary movements, I defined ritual as 'primarily symbolic in
character'. I designated religion as a symbolic system, exemplifying
this by an analysis of a Tikopia communion feast of worship - a
kind of primitive Eucharist, a 'willing offering'. In general, I saw
symbolism as having two major connotations. In a broad sense,
symbols may be said to occur when some components of the mind's
experience elicit activity and values ordinarily associated with other
components of experience. Not an original viewpoint, this did
present a criterion of cardinal importance for the definition, that of
mental experience. Almost all language and art is symbolic in this
sense. In a narrower sense, symbols are objects or actions that repre-
* Levi-Strauss (1962,105-12) gave qualified approval to this interpretation but put Tallensi
totemic symbolism, together with that of Tikopia, on a much more general plane. For com-
ment on his view see Fortes, 1966, 9-14, Firth, 1970b, 280-94.
MODERN ANTHROPOLOGICAL VIEWS I 69
sent other entities in virtue of some arbitrarily-assigned conceptual

relation between them (1951a, 176). Note that this definition, unlike
that of Homans but like that of Fortes, included 'actions' as possible
symbols. Note too that the 'arbitrariness' of assignment of symbolic
quality, a feature recognized by many writers on symbolism from at
least Peirce onwards, was not claimed as an absolute or random
relation, but was seen as arising from a lack of shared experience by
the observer or interpreter. The job of an anthropologist studying
symbolic behaviour is to attempt to replicate the relevant experience,
or indirectly to gain an understanding of it.
One of the most powerful and scholarly thinkers of that period,
S. F. Nadel, was more interested in problems of symbolic behaviour.
In his study of the Nupe kingdom (1942) he had discussed symbols
of class distinction, pointing out how turbans, swords, gestures of
greeting, gifts, even language and music, were used to mark posi-
tion on the social scale. (He noted too how Nupe had 'kept abreast of
the times'; a novel symbol of status was an 'alarum clock' in a
little wooden case, which body-servants of men of rank carried after
their masters! (1942, 128-30).) Nadel's concern for symbols was
very largely because of what he regarded as their 'diacritical' role
in culture - demonstrating the differential position of individuals in
the group and so the range of social tasks consequent upon that
position. In his general study The Foundations of Social Anthropology
(1951) he explored this role quite deeply. His viewpoint, though
expressed in abstract terms, was robustly pragmatic. Following
primarily John Dewey and Charles Morris and in line with a common
distinction, he used 'sign' for a representation having an 'intrinsic',
a 'natural' connection with the thing it signified, while the representa-
tive capacity of a symbol is 'artificial, valid only under prescribed
conditions, and determined by rules of usage' (1951, 6jn). He did
admit certain symbols as being more 'natural' than others (see above,
p. 59) but he saw them all as being employed by individuals in
accordance with the cultural canon, their meaning being decided by
existing cultural norms. For Nadel then as for other anthropologists,
the arbitrariness of what he termed 'artificial' symbols was only
relative and temporary; the correspondence of symbol to object
lay not in a single reduplication of event or situation but in the
systematic arrangement and rules of usage in regard to its position in
170 SYMBOLS
a set of symbols. Although fully capable of adopting a 'mentalistic'

outlook - as the last part of his book, on 'psychological explana-
tions', including 'mental energy', demonstrates - he sturdily argued
that the logical consistency of a symbolism does not preclude the
intervention also of a purposive and utilitarian nexus (1951, 263).
He was concerned with behaviour, and he was concerned with
communication. An idea not communicated, he argued, is beyond
examination; once it is communicated, by means of some compre-
hensive sign or symbol, it provokes action. Nadel was not a be-
haviourist in the technical sense, but he was much impressed by
questions of evidence. 'Accepting consciousness behind behaviour
is one thing, identifying its operations, another. Here the anthro-
pologist's position is precarious. He is not of the group which he
studies; he observes individuals whose motivations, thoughts and
feelings may differ widely from those which are to him habitual and
self-evident . . . the sentiments, thoughts, and other motives
behind behaviour are accessible only within set limits' (1951,64,71).
He set those limits wide, admitting in his study of incest in the Nuba
mountains an approach to consciousness through incest dreams.
When therefore Nadel stated that in his view 'uncomprehended
symbols' (of an unconscious nature) have no part in social inquiry; if
they indicate nothing to the actors they are from an anthropological
point of view irrelevant and indeed no longer symbols (1954,108), he
was stressing not the impropriety but the difficulty of their inter-
pretation. I think he made his statement too strong. But in fact he
qualified it immediately by distinguishing between 'strict (and com-
prehended) symbolism' and 'ceremonials' which 'point beyond them-
selves' in a manner analogous to signs or symbols, expressing im-
plicitly support for principles of social structure or social ideals, or
compensatory mechanisms to social control (1954, 110-13), a n ^
which it is legitimate for the anthropologist to interpret.* Nadel's
exclusion of 'uncomprehended symbols' from the purview of the
anthropologist has often been quoted, and criticized. But his
qualifications tend to be omitted. Victor Turner argues with strength
that a social anthropologist can justify his claim to be able to inter-
pret a society's ritual symbols more deeply and comprehensively
than the actors themselves, by being more detached and by having
* A somewhat similar point was made by Evans-Pritchard, 1956, 232-3.
studied the inter-relationships in the society concerned in a structural

manner (1967, 26-7). Such a statement, with which I agree, is I
think in line with NadeFs attitude towards 'border zones of sym-
bolism in ritual', where the interpretation is established by percep-
tion of a relevant parallelism between the form of ritual and the
norms governing social relationship (1954, n o ) .
THE PROBLEM OF IDENTITY
So far I have shown how in recent work in social anthropology

symbols have been sought and recognized, especially in the ritual
field. But it is clear that even at that stage there were problems of
identification. Granted that some kind of representation of an
oblique, culturally determined kind was occurring, just where was it
to be identified; what kind of isolate could be made? Answers of
a rather different kind from those already indicated were given by
anthropologists who focused less on the behavioural or action side
of the phenomenon and more on the mental, cognitive side. In an
exploration of Nuer ideas Evans-Pritchard considered problems of
symbols from this angle. In his investigation of what he called 'a
relation of Spirit to persons through things' he was much concerned
that it should be realized that in many of the Nuer symbolic equa-
tions there was an implicit third term. So, in the equation of a cucum-
ber with an ox, or of a twin with a bird, the relation is not conceived
of by Nuer as a material one, but is an ideal equivalence, conceived
of in relation to God. Broadly, such inferences about Nuer modes
of thought, based upon Nuer exegesis, were borne out by observa-
tion of Nuer behaviour - though as I have pointed out (Firth, 1966a)
this behaviour was not always consistent with their ideal norms.
The problem of Nuer religious symbolism was pursued by Evans-
Pritchard in terms of a scholastic analysis of the word 'is' in expres-
sions of the order of 'rain is God' and 'a crocodile is Spirit', dis-
tinguishing an elliptical statement about manifestation from a sym-
bolic statement about representation.* Evans-Pritchard rejected
Levy-Bruhl's view that a primitive assertion of a symbolic relation-
* The discussion is reminiscent of A. J. Ayer's examination of the problem of ambiguous

symbols, also using 'is', involving existence, or class-membership, or identity, or entailment
(1946, 62-3).
172 SYMBOLS
ship implied a prelogical thinking in terms of mystic participation.

His own view was that the problem was one common to all religious
thought - that a religious symbol has always an intimate association
with what it represents; that which brings to the mind with what
it brings to the mind (1956, 134, 140-2).
An opinion of somewhat the same kind was advanced about the
same time quite independently by Dorothy Lee. Evans-Pritchard's
formulations may have been clarified philosophically by Dorothy
Emmett as he suggests (1956, ix), but Dorothy Lee's inquiries into
the behaviour of the self in relation to the cultural codification of
experienced reality had other philosophical roots. Yet in her com-
ments on symbolization and value (1959, from 1954) she too empha-
sizes the notion of what one may call the integrative identity of
symbol and thing symbolized. Intellectually, several elements may
be combined in this approach, though she might not acknowledge
their direct relation: 'field' theory; Levy-Bruhl's theory of participa-
tion; and Whorf's view of the influence of language upon perception.
Lee has argued that a symbol is part of a whole, a component of a
field which also contains the so-called thing, as well as the process of
symbolizing, and the apprehending individual. A symbol, she holds,
is not a thing but rather a point in a creative process of symboliza-
tion, whereby the physical reality is transformed into the thing, the
experienced reality. The symbol conveys the meaning of the situation
in which it participates, and has no existence and no meaning apart
from this situation. And language is an important system of symbols
through which the individual transforms physical reality into ex-
perienced reality. So giving something a name gives it recognition
and status in the categories of experienced reality (1959, 79, 80, 83).
In some respects this seems quite plausible. Much general dis-
cussion of symbolism has held that the relation between symbol and
referent is conceptual - the association between them is a mental one.
As Fortes has emphasized, the approach has been an 'actor-centred'
(or as I would prefer 'actor-oriented') one, concerned with the mean-
ing of symbols from what can be inferred as to the participant's
standpoint. I concur in such a formulation as 'true symbols...
acquire valid existence and value only through participation in
meaningful situations' (Lee, 1959, 88). But while Dorothy Lee's
more trenchant holistic approach is persuasive, I think it needs some
reservations. When I was once talking to some Tikopia about

seabirds, and asked about their names for the larger ocean-going
petrels, they replied: 'They have names, but we don't know them.'
Here surely was the idea of name and object being inseparable, of
the name attached to or conjoined with the bird floating out some-
where over the ocean. Yet three conceptual elements were involved:
birds; names; we knowing, and these could be separately handled.
Even though their specific names were not known (thus, it was im-
plied the birds had such names independent of the knower) the birds
could be spoken of and described; moreover, even without specific or
general names they could be observed, taken captive if they came
ashore, etc.-they had separate recognition and status in the cate-
gories of experienced reality. As symbols, bird-names could be con-
ceptually integrated with or conceptually separated from the objects
to which they were attributed. The strength of Lee's argument seems
to me to lie in the emphasis she puts upon the significance of con-
textualization, upon the meaning of a symbol not being simply a
matter of arbitrary assignment but as arising from its use in concrete
situations, in which it has participated and which it has helped to
create (1959, 79).
But an idealistic view of symbols must take care not to be dis-
ingenuous. Clearly, any statement of the order of 'the bear' or 'the
crocodile' is a symbol of such-and-such a clan must be referring not
just to a material object but to a type or species, that is, to an idea.
But it is not just an idea that is involved; this has an empirical corre-
late of some sort. Fortes makes this point in comment on Evans-
Pritchard's argument that since a Nuer may rarely if ever see his
totem, 'it is no longer a question of a material object symbolizing an
idea but of one idea symbolizing another' (1956, 135). As Fortes
puts it, without some objective extra-personal reference it would be
impossible for such an observance to be held in common by a
number of people (1966b, 13).*
M E A N I N G OF S Y M B O L S FOR S O C I A L RELATIONS
The recognition of symbolism, and inquiries into the meaning of

* Even unicorns, which Evans-Pritchard cited in proof of his argument, as having never
existed, do in fact have an elaborate material base in literature and pictorial art, as well as a
putative referent in narwhal and rhinoceros (see T. H. White, i960, 43-4).
174 SYMBOLS
symbols, have been forced upon social anthropologists working in

the field (for example, Firth, 1951, 24-6). For the most part they
have been content to accept the representative quality of objects
and actions, taking as indices the information they acquired about
associations with other objects and actions and emotional involve-
ment of the actors, without pursuing questions of symbolic identity
very far. In the Australian aboriginal field, for instance, bull-roarer
and other sacred objects, clan patterns carved on trees around an
initiation ground, stones said to be transformed heroes, Wondjina
painted figures on rock walls (for example, Elkin, 1954, 176, 219)
could be clearly classed as symbols. At the same time these material
things, used in systematic elaborate behavioural sequences, were
important means of portraying, implementing and even changing
social relationships.
Nadel (1951, 262) indicated three main forms of symbolic be-
haviour which though not at all exhaustive, can introduce a dis-
cussion. The first he cited were emblems, badges and other 'diacriti-
cal' signs; whether material objects or gestures or other behaviour,
their display indicates group membership or other social relation-
ship and provides expectations for corresponding 'real' behaviour.
The second category of symbolic behaviour mentioned was all
forms of social nomenclature, including, for example, classificatory
kinship terminology; from the use of these too, expectations as to
other conforming behaviour are derived. Nadel's third category of
symbolic behaviour was the 'dramatizations' pervading primitive
cultures, instanced by the initiation rites of Van Gennep's classical
analysis. Nadel was concerned with a wider problem - the nature of
explanation in anthropology - which is not of immediate relevance
to us here. But he made two points of direct significance to the inter-
pretation of symbolic behaviour: that symbolic patterns may vary
in consistency, both internally and in their relation to the scheme of
social relationships; and that for interpretative purposes the observer
is justified in going beyond the actor's conception of the given
nexus, provided he can perceive a rational relation between the
modes of action (1951, 263-6).
I will expand a little on the last point. In the interpretation of
symbolic acts one may have direct contextualization - such linkage
with other actions, verbal and non-verbal, that meaning emerges
from the set of operations in concert - or parallelism - whereby

elements separated in time and space and belonging to different
concerted sets may be brought together in the same frame of seman-
tic operations. Take the Tikopia example of anointing the temple post
(see Chap, i, p. 26). The meaning of the symbolic act of rubbing
the post with a bunch of aromatic leaves is given partly by the
immediate context - the chief as officiant, his fresh clothing and
decorations, the formula he recited for welfare, etc. But it is also
given in part from comparison with parallel operations elsewhere -
washing of sacred stones, bathing of people, use of oil and fragrant
leaves in dancing. Some linguistic bridge may be obtainable - the
Tikopia term for anointing the post (the same for anointing a
canoe) is linked etymologically with that for swimming, and equated
also with oiling. But even when there is no linguistic bridge or
overt explanation, collateral evidence can be derived from common
experience of ritual occasions. Watching the care, even fear, with
which people go about their formal procedures, conforming one-
self to signs of warning or approval, and seeing other people act
predictably in accord with one's expectations, all help to indicate
that one is participating in acts of symbolic quality, with meaning for
the social group. What meaning? Politically and economically, the
rite drew together representatives of the constituent lineages of the
clan in orientation towards their chief; it set a positive stamp upon
solidarity. But why anoint the central post of the temple? Architec-
tural observation could show that only major clan temples had such a
centre-post, so reinforcing the supportive interpretation. In such
external behavioural terms, the anthropologist can go a long way in
his attribution of meaning. What he could not do without more
specific explanation from the people themselves was to understand
that rubbing the post with aromatic leaves and oil was not like
painting a door - a touch of adornment to a temple to give pleasure
to a god - but was believed to be adorning a material representation
of the god himself. The post was not an anthropomorphic image -
the Tikopia did not believe their god looked like a baulk of timber.
Although they called it by a word translatable as 'body', it was the
god's symbol, the concrete object to which they could direct physical
actions of worship which the god himself could observe in spiritual
invisible form from elsewhere. All this was given added meaning
176 SYMBOLS
when considered in context of a long complex cycle of observed

ritual performances, with much explanation from participants and
other members of the society, which in itself threw light on social as
well as religious matters. From such data I as analyst was able to
perceive an order in the pagan Tikopia religious symbolism, and to
assign it meaning in the Tikopia system of social relations. But some
of this meaning was definitely inferential. I described an initial rite
of burning through a stick of firewood in the middle so that it fell
in two halves, as 'a symbolic act of parting, of separating the sacred
from the profane period'. But I pointed out that I could get no
definite Tikopia statement of this general interpretation; I had to
infer it from other Tikopia statements and the general context of
behaviour, in which the response to the rite was to stop all dancing
and cast an aura of sobriety over the whole land (1967a, 49-52).
So, in assigning meaning in social terms to symbolic behaviour an
anthropologist is usually not concerned with 'proof5 in a definitive
sense, but with degrees of plausibility of inference. General cate-
gorizations of rites as expressive, communicative, supportive, mani-
pulative, etc., in the symbolization of social relations, are normally
inferential to a considerable degree.
SYMBOLISM OF RITUAL
Ritual is a symbolic mode of communication, of 'saying something'

in a formal way, not to be said in ordinary language or informal
behaviour. This idea of'not to be said' in an ordinary way means that
a special character of ritual is its reserve, its apartness, its 'sacred'
quality.* Its 'grammar' is different from that of ordinary language.
What is said in ritual may refer to individual, personal states and
actions, but on the whole tends to refer to social states and inter-
personal relations, f This kind of statement may represent several
elements in a situation. The symbolic behaviour of ritual may indi-
cate a 'shortfall' in verbal statement - expressions of interest,
* The 'sacred' in my view is not a necessary criterion in the definition of ritual, but rather a
quality created as a protective device for rituals which on other grounds are regarded as of
prime importance. There can be secular rituals, as there can be secular symbols, rituals varying
in the degree of their sacredness. (This is an alternative concept to retention of 'ceremony',
which has tended to disappear as a technical term of weight in anthropology.)
f For valuable brief discussion of these issues see Beattie, 1966, 66-8.
respect, anxiety, desire which are put into ritual form because they
are thought not to be capable of expression in other form. But ritual
may also be regarded as a symbolic understatement of the 'real*
situation, going so far as the human officiant can go in terms of his
capacity, but recognized as inadequate, a kind of'token communica-
tion'. It may also be seen as a kind of symbolic disguise (see Fortes,
1966a, 409-22), a statement which alludes indirectly to 'reality' not
by exposing it but by evading it or representing it as other than it is.
Or again, the symbol idiom of ritual may allow statements to be
made in terms less brusque, more protracted, more tempered by
involvement with other acts, than in ordinary language (cf. Firth,
1967a, 21-5).
Ritual has therefore been one of the great fields for the study of
symbolic behaviour, and many modern social anthropologists have
contributed to its interpretation. Their primary search for the mean-
ing of symbolism in ritual has focused upon social relations. But it is
to be noted that in two of the major areas of theoretical concern -
what can be reasonably inferred about unstated aims and functions;
and what is the relation of public to private, or social to personal
meanings - the treatment and conclusions of different analysts have
varied considerably.* I take here for consideration mainly British
studies, with which I am more familiar.
An enlivening essay on the symbolism of ritual came from Max
Gluckman in 1954 (1963). Using the vivid and impressive account
by Hilda Kuper (1947) of the Swazi Incwala, often termed a first-
fruits ceremony, Gluckman examined the symbolism of political
catharsis in the songs and dances of hostility which form part of the
complex ritual. Under the title which by now he has made well known,
of 'rituals of rebellion', he pointed out that the ritual was organized
to exhibit the co-operation and conflict which make up the political
system. In the traditional Swazi system the political and social order
itself was not questioned, but communal interest, including that
* For contrast may be noted a recent overt preoccupation of some artists with *ritual*,
in which they take as subject some common theme that characterizes human behaviour, and
in which they believe they can discern a kind of basic human experience of moral or didactic
value. So, Romare Bearden has used as theme what he has termed T h e Prevalence of Ritual'
to express in painting and collage deep emotion about life's problems, especially in the life
style of American Blacks. For illustration and brief discussion of a Museum of Modern Art,
New York, exhibition see Carroll Greene, Romare Bearden; The Prevalence of Ritual
(Museum of Modern Art, New York, 1971).
178 SYMBOLS
exemplified in the person of the sacred king, could conflict with

individual and segmental group interest, which was thus allowed
formal expression in the ritual. So 'ritual rebellion' could be en-
joyed as a social blessing, with traditional sanction. Granted the
general cogency of Gluckman's argument and the suggestive hypo-
theses it involved, this example illustrated the difficulties of inter-
pretation the anthropologist faces with such symbolic material. That
alternative inferences as to meaning of symbolic behaviour are
possible even with highly equipped social analysts is shown by the
difference of view of Kuper and Gluckman about a rite in which the
king is driven into his sanctuary - whether he is being entreated to
migrate with the royal clan or being denounced by them (Kuper,
1947, 216-18; Gluckman, 1963,124, 258).
But to a non-Africanist like myself there is another more general
question: is the title 'rituals of rebellion' not a misnomer? The songs
cited do not seem to express the singers' hatred of the king, but
other peoples hatred -'they hate the king'; 'O King, they hate thee';
'You hate the child king' Several interpretations are possible
here. Psychologically, the 'you' and 'they' may be projections,
equivalent to 'we'; this seems to be Gluckman's inference, though he
does not say so. But taken more literally, they may be accusatory
though without specifying anyone; 'they' hate the king, but 'we'
are loyal - and the rite thus may be termed supportive, not rebellious
in tone. Again, the king himself supplied a clue to a possible further
interpretation. 'With sociological insight', as Kuper commented
(1947, 224), he explained that the warriors danced and sang at the
Incwala and so did not fight, although they came from all parts of
the country and were proud and jealous. 'When they dance they feel
they are one and they can praise each other.' Part of the social
mechanism for helping their unity could be to join in asserting that
their king is threatened (by persons unknown) both mystically and
physically, in 'few, mournful, and tremendously moving' words
(1947, 206). My point in traversing these arguments is to show that
while inference is often necessary to supply meaning to symbolic
behaviour in ritual, caution is advisable in ascribing to it basic
themes of either support or conflict in the social system.
The complexity of interpretation of ritual symbolism is well
illustrated by the phenomena of sacrifice, which have had a long
line of distinguished analysts, from Trumbull, Robertson Smith,

Hubert and Mauss, to Evans-Pritchard, Middleton, Goody and
Godfrey Lienhardt. To identify the slaughter of an animal as a
sacrifice, that is a ritual act of dedication, may pose problems for
an anthropologist.* He sees a bull being killed - is this symbolic
or is the beast being butchered for meat? The empirical indicators
are usually obvious - as when in Kelantan in 19401 saw a white cloth
being held as a canopy over the animal when its throat was cut as it
was held down on the ground. I could quickly establish that this was
not to shield the Malay actors from the sun, but to indicate that the
bull was dedicated to Allah, as a prayer confirmed. There were other
religious associations of the act - but there were also economic and
social referents. A bull is a costly animal, much more so than a goat
or a sheep. But as I was informed, the smaller animal would not do as
a sacrifice on this occasion because in the local conceptualization, the
donor of the sacrificial animal would have the privilege of riding it
before the Lord in the grand parade on the Last Day, the Day of
Judgement, and to appear mounted on a sheep or goat would be
ridiculous. But few men could afford a whole bull at a time, so
elaborate annual purchase of shares in a co-operative venture - in
sevenths - enabled a man to accumulate a bull-equivalent in the end.
The whole transaction, most carefully calculated, also took into
consideration the sharing and consumption of the meat - no non-
sense about a burnt-offering for these Malays! Here then deter-
minants at the economic level of poverty then obtaining among those
Kelantan Malays affected the form of symbolic conception, which
was trimmed to allow a wide range of participation in status behaviour
as well as in ritual practice (Firth, 1943, 203). If the symbolism had
not allowed of sacrifice on the instalment plan, only the most
wealthy would have been able to conform to the religious ideal.
(Elsewhere I have pointed out other implications of economic fac-
tors for the theory of sacrifice (1963, 12-23). ^ course it would
have been possible for these Malays to have done as Nuer or LoDagaa
have done, and to have used sheep or goat as a substitute for a bull
by calling it such. But this does not seem to have occurred to them,
perhaps because sacrifice plays no commonly repeated part in their
* See, for example, Goody, 1962, 162-3 f°r distinction between sacrifice and beast offered
as a funeral honour to the deceased.
i8o SYMBOLS
ritual life and problems of scarcity have therefore different applica-

tion for them. (For Lugbara attitude also see Middleton, i960, 88.)
In a caustic comment R. R. Marett (1935, 148) gave a warning
against over-interpretation of the meaning of sacrificial practice.
When the grosser forms of symbolism survive into the higher
religions, as typically in the case of sacrifice, he said: 'a crude affair of
shedding blood and appropriate only to those who do their own
butchering' - it needs much verbal reinterpretation to allegorize the
primitive act. Without accepting this implied criticism, it is fair to
note that anthropological interpretations of the symbolism of sacri-
fice have been affected both by religious commitment and by pre-
occupation with factors of unconscious mind. There is ample
demonstration in the literature of the socially integrative functions of
sacrifice: as focus for religious activities of groups; as medium of
expiation for individuals to remedy non-fulfilment of obligation or
other offence against social rule; as means of social control. There is
abundant evidence too that the symbolism of sacrifice embodies
complex sets of ideas and emotions, including, it may be, social self-
consciousness, tension and release, joy, thanksgiving, apprehension,
propitiation - in relation to systems of belief about spirit powers and
forces of man and nature. Evans-Pritchard has asserted, very
plausibly, that when Nuer give their cattle in sacrifice they are in a
very intimate way giving part of themselves, an identification
especially easy because what they surrender are living creatures, the
most precious of their possessions (1956, 279). But a part of oneself
is not the whole of oneself; so it is a further step to argue not only
that in sacrifice some part of a man dies with the victim, but also that
sacrifice is a self-immolation, the substitution of a life for a life. This
complete personal identification of man with victim, which the
Nuer themselves do not seem to have clearly acknowledged, and the
idea that the life can only be given by its liberation through death,
seem to me to be related to Evans-Pritchard's view of the Nuer
conception of God, which he treats as a mystery. To say that a
sacrifice is a dramatic representation of a spiritual experience is an
expression that can probably go unchallenged by most anthro-
pological opinion. To say that its meaning depends finally on an
awareness of God and man's dependence upon him - at which point
the theologian takes over from the anthropologist - is to give a very
MODERN A N T H R O P O L O G I C A L V I E W S l8l
personal interpretation (1956, 322). From a very different angle, that

of psychoanalysis, interpretations of Nuer sacrifice have taken a new
turn. Beidelman has argued (1966, 464) that Evans-Pritchard has
ignored the element of sexuality, which he himself suggests is at the
very core of Nuer moral ideas. The victim in the sacrifice, an ox,
represents by its castration the moral subordination of Nuer male
sexuality to social values; the instrument, the spear, 'a socialized
penis', represents the aggressive element which subordinates women
and through its power of death, initiates the transformations in-
volved in the ideal aspects of agnatic values. So in spearing an ox a
Nuer 'expresses a kind of transfiguration, through immolation, of
his sexual self and an anticipation of his own transformation, through
death, into the agnatic ideal person which his own living, domestic
sexual self cannot wholly be and, indeed, cannot wholly accept'.
Whether this qualifies as Marett's 'allegorization' or not, it is clear
that in such type of analysis many assumptions are being made about
mental functioning of Nuer, which are derived from a general body
of theory not based - as is psychoanalysis in the West - on a body of
clinical evidence. Beidelman does not claim conclusive proof of his
argument from Nuer data, but holds that the issues of symbolic
interpretation he raises demand attention. Granted the significance
of the problem of Nuer sexuality in relation to their identification
with cattle, this type of interpretation of ritual symbolism points to
a need for much more systematic collection of empirical data about
people's mental associations in fields such as sacrifice.
SYMBOLS IN T R A N S I T I O N RITES
The interpretation of transition rites has been a stock activity of

social anthropologists for more than half a century, especially since
the publication of Arnold Van Gennep's Les Rites de Passage in
1909. (An English translation with introduction by Solon T. Kimball
was issued in i960, and a series of essays by Gluckman and others
in 1962 gave point to modern commentary.)
But in the post-war period probably the most sustained and pene-
trating analysis of a type of such rites was given by Audrey Richards
in Chisungu (1956), an examination of a girls' initiation ceremony
among the Bemba. Traditionally this consisted of a long elaborate
182 SYMBOLS
succession of ritual acts including singing, mime, dancing and

handling of sacred emblems, lasting more than three weeks, and
ending in a wedding ceremony. (Formerly analogous rites were
practised in many parts of East, Central and South Africa, but they
have all undergone radical change in modern times.) Audrey
Richards points out that whereas there are many accounts of male
initiation rites in the literature, those of female rites are rare. She
sees in such rites a double significance: they celebrate the attainment
of sex maturity, hence the possibility of physical parenthood; and
they also celebrate social maturity with its responsibilities and
privileges. In this twofold aspect Richards sees the possibility of
differential stress, and hence of local variation. But with a female
there is additional complication: the onset of her menstrual flow is
linked with emotionally tinged ideas about the power of blood -
negatively associated with danger and mystical damage; and posi-
tively with creative fertility. So linked with rites of protection and
purification are also those of promotion of child-bearing and domes-
tic efficiency. But within this general framework of assumptions
Richards sees many difficult problems of interpretation. Yet it is here
that her most distinct contribution lies.
She argues that an anthropologist struggling to interpret a com-
plex ceremony has to use a number of approaches. She distinguishes
the expressed or formulated purposes of the rite from what the
observer infers, but also separates the short-term specific aims from
the long-term more generalized ones. The former are apt to be more
clearly conceived and easily expressed - V e do this to make child-
birth easy'; the latter may be vaguer - V e do this to take anger from
men's hearts'. But secondary ends may also be recognized and
expressed - to gain prestige from sponsoring a rite, or to bind kin to
return an economic service. She therefore makes the point that by
the nature of ritual there should be multiple explanations for most of
its symbolic behaviour - with the corollary that complete explana-
tions of every item are unlikely to be obtained from any one member
of the community. But since rites are necessarily an effort to do
something, the performers can always give some explanation. Yet
she holds that what an observer infers (she has written 'deduces')
may be an essential part of the interpretation. Indirect evidence as to
the meaning of a rite may be: linguistic statements which reveal some
common stress or interest; on-the-spot emotional reactions of con-

cern, tension or boredom which show the relative importance of a
symbolic act; refusal to omit or shorten a rite, as contrasted with
readiness to abandon it, as an index to its critical place in the system.
Audrey Richards also cites Radcliffe-Brown's analysis of Anda-
manese beliefs (cf. p. 135) as support for inferential attribution of
symbolic meaning from other rites in which apparently identical or
similar acts occur (1956, 115).
Unlike some of her colleagues at that time, Richards was not
averse to psychological interpretation of symbols. She noted the
prevalence of serpents and other universal-type symbols, and the
mimetic representations of journeys through dark and difficult
places to reach knowledge and safety. But while linking these with
dream presentations and general human experience, she regarded
them as primarily a matter of interest to people in allied disciplines,
rather than to anthropologists, whose major interest is in the clues
which such imagery may give to the emotional expression of ele-
ments in a particular social structure (1956, 153-4). A cautionary
note, however, is expressed. Rites which seem to cut across the
normal institutional modes of a society, especially by display of
aggression against established positions, have been interpreted by an-
thropologists as symbolic outlets for repressed hostility or symbolic
compensation for traditionally subordinate roles. Citing analyses by
Bateson and Gluckman of women's aggressive behaviour so con-
strued, Richards describes these as unproven conjectures, since
evidence was lacking as to whether the people concerned felt the
tensions ascribed to them. Rather sardonically, she points out in
effect that the existence of the thing signified is really derived from
the presence of the symbol; a valid interpretation must depend on
independent evidence of the psychological states postulated. This
criticism, generalized, has a much wider application.
Richards makes very clear the difficulties faced by an anthro-
pologist in seeking valid interpretations of ritual symbolism, and she
frankly admits that she has been unable to resolve some of the in-
congruities of the chisungu. What she does, however, in a construc-
tive way is to relate the ritual symbols to structural situations as well
as to emotional attitudes. She shows how in these girls' initiation
rites and the marriage rites which follow them it was not the chastity
184 SYMBOLS
of the girl but the virility of the bridegroom which was emphasized;
and she indicates how the symbolism fitted the matrilineal structure
of the society. The bride belonged to the matrilineage and the groom
was allowed access to her to make her fertile - to produce a further
asset for her group. In daily life he was submissive to his in-laws and
worked under their orders; in the chisungu he appeared symbolically
as a roaring lion, a lion-killer, a crocodile, a hunter, a warrior, a
chief. So symbolic status as a procreator and impressive figure was
a compensation for the pains of everyday life.
The symbolism of these rites expressed themes of significance
both for the society and for individual initiates. Richards holds that
the rites symbolized the values and beliefs of the group, expressed,
reinforced and taught tribal norms, promoted social cohesion - all in
a complex way. So mortuary ritual could symbolize the performers'
duty to their chief or the values set on economic activities, as well as
grief for the dead man. But she makes two important qualifications.
The correspondence between the total value system of a group and
its symbolic expression in ritual is never exact - some basic values
are dramatically represented, others not. And some weighty rites are
just as often an occasion of group division as of union — a statement
which she refers back to her analysis of social differentiation at Bantu
meals long before (1932, 71-2; 1956, 117-18).
As a brief digression I would observe in this connection that the
notion of'group division' in relation to ritual symbolism can refer to
several types of situation. One type is that of social differentiation in
the sense of Richards's Bantu meal analysis; she points out how from
an early age males and females, seniors and juniors eat in separate
groups though their food may come from a common household.
This is a recognized expression of socially approved norms of struc-
tural kind, in direct symbolic form. Another type of symbolic ex-
pression of group division is where the rites do not tend to maximize
but to minimize the social differentiation. Segmentary groups, each
with its own competing interest, are brought together in a com-
munity of action which may either have the effect of promoting
co-operation and positive sentiments of unity or at least of'papering
over the cracks' of disunity to allow of some broad achievement of
common policy. Another type of symbolism concerned with divi-
sion is the ritual expression of hostility which may be roused because
of asymmetric allocation of resources or power, but be inhibited

from ordinary overt display because of moral or other values; this is
a common anthropological interpretation of much symbolism of the
'conflict' category. Finally, there is the type of group division which
arises quite overtly when competing social units struggle for con-
trol of the same symbols of unity - as Christian sects do around the
'Church of the Rock' in Jerusalem. Here the shared symbolism of
the rite neither marks nor masks social differentiation, but acts as a
focus for it.
One of the most interesting aspects of the symbolism of the
chisungu is the inter-relation shown between the position of the
initiates and that of their kin. Most studies of initiation have focused
on what the initiates learn, how their status is changed, etc. In my
own study of Tikopia initiation (1936, 418-67) I emphasized the
significance of the rites in bringing home to the initiates their
dependence upon their kin, and the meaning of kinship obligations
and categories. But Richards goes further in pointing out how the
tests or ordeals which the candidates undergo are also ordeals for
their parents and relatives - who are anxious until they see that the
young people have proved their competence or at least have been
allowed to pass through. Bizarre though some of these tests may
seem, passing through them lets the kinsfolk see that their novices
have received the recognition of society in their change of status.
Richards notes that European accounts of such initiation rites
have tended to emphasize either their obscenity or their educa-
tional function. She corrects this: there is sex but it is not lewd, and
there is education but it is not formal. What Bemba girls gained was
not technical but social and symbolic knowledge, and much of that
indirectly - a deeper knowledge of the social attitudes and symbols
concerned with women's functions in sex, marriage, wifehood and
childbirth. They weren't given lectures on marriage guidance, but
got a cumulative process of experience, often in apparently dis-
connected acts, as in jumping over a hurdle, singing obscure old
songs and handling odd crude pottery figures of snakes, lions, naked
brother-and-sister. The basic point is that these seemingly odd and
individually meaningless experiences were pulled together by setting
them in a frame of symbolic ideas and values related to the structure
of Bemba society.
G
l86 SYMBOLS
Finally, Richards insists, as Victor Turner has done, on the mani-

fold associations of such symbols. 'The interpretation of the songs . . .
shows that there is in every case at least one double meaning for each
symbol. The hoe represents the woman's gardening duties but it
also represents the husband who makes his wife fertile' (1956, 164).
Some interpretations show how 'as in dream life the symbols of the
chisungu evoke a variety of linked associations, verbal and other . . .
the curious figurines stand for certain emotional associations com-
mon to the whole community, but they may also acquire special
meanings in the light of that particular ceremony, or even in terms of
individual experience.' Other interpretations 'show how the same
symbol can at one and the same time represent the thing that must be
done, and the thing that must not be done'. 'All symbolic objects
make it possible to combine fixity of form with multiple meanings,
of which some are standardized and some highly individual' (1956,
165). In these and analogous generalizations the themes of 'am-
bivalence' and 'multivalence' which appear so markedly in much
later anthropological study of symbols, emerge quite clearly.
SEARCH FOR A SYMBOLIC PATTERN
An important aspect of Audrey Richards's treatment of the sym-

bolism of ritual was the breadth and care of its methodological
approach, explicitly formulated. The contribution of Monica Wilson
has had some analogy with this, though she has been perhaps rather
more conservative in her acceptance of the validity of psychological
hypotheses and techniques. Her major examination of symbolism
was in two works (1957, 1959), the first concerned with the rituals
celebrated by kinsmen at death, at a girl's puberty and marriage, at
birth and in misfortune; the second concerned with communal
rituals celebrated especially in public misfortune such as plague or
famine, or at succession of a chief or of younger generation to elder.
Using Radcliffe-Brown's principle of similar significance of symbols
recurring in different ritual situations, Wilson stresses her view that
Nyakyusa rites reveal a symbolic pattern which has an intellectual
system of interpretation. She holds that it is foolish to imagine that
an individual schooled in one culture can understand the symbolism
of another without instruction (1957, 6). Hence she is very careful to
cite her vernacular authority for her attribution of symbolic mean-

ings, the more so since like Audrey Richards she regards anthro-
pological literature as being full of symbolic guessing, the ethnog-
rapher's interpretation of the ritual of other people. At the same time
she points out that not all the Nyakyusa with whom she dealt were
conscious of all the interpretations she gave. While a few symbols,
such as association of different types of banana with male and female,
were known to everyone, knowledge of others was restricted to
specialists or was purely individual formulation. But if individual
formulations were used 'it was at least a Nyakyusa subconscious, and
the interpretation in terms of the culture'. Trenchantly she argued
that 'any analysis not based on some translation of the symbols used
by people of that culture is open to suspicion' (1957, 6-7). So she
reckoned to deal with common, not individual symbols; with socio-
logical, not psychological analysis. It is true that Monica Wilson
does define symbol very loosely -'something which typifies or
represents something else' (1957, 9) - which allows her a wide range
of descriptive materials to draw upon. Moreover, she does occasion-
ally go beyond her brief and write of the 'implicit' meanings of
Nyakyusa symbols; and suggest that 'perhaps' certain rites are
symbols of marriage union, etc. She states that while 'the doctor'
interpreted water dropping on mourners and nubile girls running in
and out of a doorway as a symbol of their receiving spirit-shades
back on their bodies, 'we' interpret it, rather more widely, as a sym-
bol of rebirth (1957, 204-5; *959> 211). But all this is well within the
plausible, making use of only a modicum of modern psychological
lore. She is very modest in her claims. She admits that interpretations
of even the most self-conscious Nyakyusa do not reveal the whole
truth about their rituals. She says it is probable that certain Nyakyusa
symbols are universal in that they express the same ideas in all
societies: spear as penis; house with its doorway as womb (1957, 66,
74,117; 143, 205). She also postulates that many symbols not always
explicitly recognized by Nyakyusa are emotionally intelligible to
Westerners when interpreted.
The results of such analysis have been impressive. More than most
other anthropologists, Monica Wilson has striven to present the
symbolic pattern in the Nyakyusa conception of reality as a total
system, with all its range of emphases, its inconsistencies, its blank
i88 SYMBOLS
areas - not just as a few selected illustrations to prove some dramatic

theme. The evidence is carefully marshalled. Rite after rite is de-
scribed, then discussed for overt purpose and symbolism, with
anthropologist's questions and answers of informants cited in detail.
When eating is labelled a symbol of sexual intercourse it is explained
that while everyone knew that a banana stood for a penis, fewer
people realized what eating it meant for a bride in her ritual; but
some were quite definite - she eats the banana means that she has
intercourse with her husband. When it is stated that blacksmithing is
felt to be like coition the analogy is given explicitly: 'The man is the
smith.... The woman is the bellows.' Question: 'The woman is a
bellows?' 'Yes, because when she is having intercourse she gasps
"He, he, he!" . . . ' So the recurrent symbolic pattern of the rites is
unfolded, and the connections made plain. Wilson argues in all this
that contrary to a commonly held theory that ritual is always more
constant than interpretation, in Nyakyusa family rituals at least there
was variation, whereas the conceptions expressed varied very little.
On the other hand in contrast to the symbolism of ritual, common
and obligatory, was the symbolism of medicine, individual and free.
Nevertheless, the constraints of a common symbolic system operated,
and the imagination of the local healers worked within the general
symbolism of the people. In medicine as in ritual one type of banana
stood for a male and another for a female; a banana sucker repre-
sented a kinsman; a sprouting shoot, conception; hair stood for
growth and a bald pate a devastated field; pythons, crocodiles,
leopards and lions stood for power and majesty; spears and thorns
for attack; vomiting and emptiness for innocence; spitting for for-
giveness. But in Nyakyusa medicine, associations were much less
predictable than in ritual, with much more individual interpretation
by specialists. Yet some general principles can be extracted - such as
'Men mime that which they wish to avoid, as well as that which they
desire; like things are antagonistic as well as sympathetic' (1959,153).
Two other features in Wilson's treatment of symbolism stand out:
her concern for what is meant by the trite expression 'people of the
culture' investigated: and her interest in change in a symbolic sys-
tem. In all her analysis she distinguishes closely the interpretations
of specialists from those of local laymen, headmen from commoners,
men from women. She also examines the attitude of Christian
MODERN A N T H R O P O L O G I C A L V I E W S 189
Nyakyusa to the rituals performed by their pagan kinsmen and

neighbours - their fulfilment of social obligation but rejection of the
religious ideology; the effects of their scepticism or partial belief; the
parallel substitution of church rites for the pagan confession, sacri-
fice and communion with the shades of the dead. To my mind her
whole analysis succeeds admirably in bringing out the significance
of a symbolic system with respect for it yet without imputing to it
any rhetorical mystical quality.
'A F O R E S T OF S Y M B O L S '
Named from an image in a poem of Baudelaire, and dedicated to

Monica Wilson, Victor Turner's studies which appeared in The
Forest of Symbols (1967) commingled an aesthetic and a scientific
tradition in a manner quite new in anthropology. His analyses of
symbols in Ndembu ritual, first introduced at a meeting of the
Association of Social Anthropologists of the Commonwealth in
London in 1958, and continued in a massive series of interpretations,
have been one of the most vigorous and theoretically suggestive
statements on symbolism in modern social anthropology. If at
times, like the incense of Baudelaire he seems to 'sing of the trans-
ports of the spirit and of the senses' this is done with a sincerity and
a power which compel respect. I am concerned mainly here, as with
Richards and Wilson, with the general methodological aspects of
Turner's analysis, which like theirs is focused very largely upon the
understanding of symbolism in ritual.
Turner regards ritual as a patterned process in time, the units of
which are symbolic objects and serialized items of symbolic be-
haviour. Ritual performances are themselves phases in broad social
processes, the span and complexity of which are roughly propor-
tional to the size and degree of differentiation of the groups in which
they occur. To a high degree ritual helps to correct or anticipate
deviations and conflicts in such groups, though its precise goals will
vary according to social circumstances. Hence to give an adequate
explanation of the meaning of a particular symbol it is necessary to
examine the widest 'action-field context' in which the rite is simply
a phase; the context of the specific ritual; and the behaviour directed
towards the symbol. Turner distinguishes three levels or fields of
190 SYMBOLS
meaning, essentially by their different contexts. The kind of meaning

derived from indigenous interpretation (which with a respect for the
vernacular informant he terms exegetical meaning) is paralleled by
the operational meaning derived from observation of the symbol in
use, noting the types of personnel involved and the affective quality
of their reactions. In turn, these levels of meaning are to be con-
trasted with the positional meaning of the symbol, derived from its
relation to other symbols in a patterned totality. Turner also empha-
sizes as a very important property of symbols their capacity for a
single symbol to encapsulate many meanings - to represent many
different things, in different contexts or at different levels of under-
standing in the same general context. This Turner refers to as the
property of polysemy or multivocality - which he seems to equate by
a kind of reverse definition, with condensation.*
Much of this way of looking at symbols is I think common
ground among social anthropologists. But Victor Turner, drawing
as we all do on ideas from a variety of sources, has welded them
together into an explicit systematic method of approach. He has also
introduced some modifications. Audrey Richards drew attention to
the positive and negative aspects of a symbol - how at one and the
same time it can represent both that which should and that which
should not be done. This was a viewpoint with many congeners,
for example in the more philosophical statements of Thomas Car-
lyle or W. M. Urban. She stressed the significance of symbolism to
express the accepted and approved as well as the hidden and denied,
the rules of society and the occasional revolt against them, the com-
mon interests of the whole community and the conflicting interests
of different partsof it (1956,169). She saw the useof symbols in ritual
as securing some kind of emotional compromise giving satisfaction
to most individuals in a society and supporting its major institutions.
Turner also recognizes a polarization in the meaning of symbols. But
he stresses the 'ideological pole' and the 'sensory pole' of his 'dominant
symbols' - one set of referents being the principles of the moral and
social order, and the other being items of desire and feeling, even of
'gross, frank, physiological order' (1967, 28). At the sensory pole
* Turner, 1967, 45-52. Cf. 'By these terms [polysemy or multivocality] I mean that a
single symbol may stand for many things' (1967, 50); 'The simplest property is that of
condensation. Many things and actions are represented in a single formation' (1967, 28). As
already noted (p. 81) this use of'condensation' differs from that of Freud.
M O D E R N A N T H R O P O L O G I C A L VIEWS 191
come particularly those involvements of symbolism with aspects of

the human body which other anthropologists too (for example Mary
Douglas) have found so significant.
Another way in which Victor Turner has developed the theory of
symbolism is in relating symbols, especially what he has termed
'dominant symbols', more closely to the theme of conflicts in society.
As well as showing how the symbolism of ritual helps to mitigate or
conceal the more overt elements of hostility, or to give them cathar-
tic expression, he indicates situations of what he has called 'blocked
exegesis', occurring where there is sharp conflict between norms, or
between norms and wishes. (To some extent he is inclined to identify
with this his occasional inability to get an explanation of a piece of
symbolic behaviour from the people themselves.) But much more
generally, Turner sees the operation of major symbolic forms in
society as founded if not in conflict at least in contradiction. He
argues that any major ritual that stresses the importance of a single
principle of social organization does so only by blocking the expres-
sion of other important principles. He also holds that when a
dominant symbol is regarded as a unit of the total symbolic system,
and not interpreted simply in terms of 'official' explanation, there
may be discrepancy and even contradiction in meanings given by
informants. This he regards as a 'quintessential property of the
great symbolic dominants in all religions' (1967, 40, 43).
On the issue of 'uncomprehended symbols' Victor Turner ex-
presses firm opinion. Taking as illustration the Ndembu 'milk tree'
symbolism, he contrasts the overt interpretations of it in girls'
puberty ritual as standing for the unity and continuity of Ndembu
society, with the unacknowledged discriminatory behaviour shown
around it, when women separate from and taunt men as they sing
around it, or take part in generation demarcations or matrilineage
conflicts. He takes this discrepancy, and the inability or unwilling-
ness of Ndembu informants to acknowledge it, as a standpoint from
which to argue that a social anthropologist can justify his claim to
interpret a society's ritual symbols more deeply and comprehensively
than the actors themselves. Turner is critical of Nadel and Monica
Wilson on this debatable issue, though I think that they both might
be ready to allow the interpretation of those kinds of non-verbal
behaviour which are integrated into a systematic ritual scheme even
192 SYMBOLS
though no verbalized explanation is given. But in more positive

terms, I think Turner makes out a strong case, by demonstration
even more than by protestation. In particular, one may underline his
argument, for which he provides ample empirical validation, of the
possibility and value of analysing symbols 'in a context of observed
emotions' when the investigator is well acquainted with the common
idiom in which a society expresses such emotions (1967, 39).
In the richness of his treatment there is room for disagreement.
I personally find some of the language in which Victor Turner
clothes his ideas somewhat ornate. His use of 'orectic' presumably
derives not from mediaeval philosophy but from the psychological
equivalent to 'affective-conative' suggested by Aveling and some
other British psychologists a generation ago (V. Turner, 1967, 54;
1970, 176; M. M. Lewis, 1947, 2on., 195, etc.). But this 'semantic
pole' with referents of 'a grossly physiological kind' seems to have
much in common with the 'sensory pole' used in other contexts.
Exegesis for explanation and hermeneutics for interpretation sound
rather scriptural, while latinisms such as personae, stigmata and
liminars (1969, 99, 121, 143), with communitas^ though possibly
more precise than persons, signs, threshold figures and communion,
seem rather formidable. But I take it that their purport has been to
invest the study of African and other symbolic systems alien to
Western readers with the status of data proper to technical study by
experts in other fields of symbolism, particularly religious symbol-
ism. They may have emerged, however, in the course of a deeper
commitment. Leaving aside questions of religious allegiance, Turner
has adopted specifically some of Carl Jung's concepts, with the
mode of their phraseology. His critique of Nadel and Monica Wilson
uses Jung in support, citing with approval Jung's definition of
symbol - 'always the best possible expression of a relatively un-
known fact, a fact, however, which is none the less recognized or
postulated as existing' (V. Turner, 1967, 26). As a definition for
working anthropologists this is very unsatisfactory. The notion of
'best possible expression' may conceal a circularity - how is any
symbol to be judged as anything but the best, unless some external
criteria are given? But if the referent is 'relatively unknown', how
precisely is it known that the symbol applies to it and not to some
other referent? . . . and so on. (Taken literally, the Jungian formula-
tion would seem to cover, say, the most careful scientific meteoro-
logical description of the state of the weather at the North Pole, a
relatively unknown fact, but certainly existing on any ordinary
plane of conceptualization.) In this connection I think the point of
'symbol' may be, not the element of the unknown in the referent,
but the degree to which the meaning of the sign which refers to it is
capable of intellectual, aesthetic and emotional development.
But I think that in the Jungian formulation, and in Turner's use of
it, we are back to the classical antithesis in another form - the idea
of the known as key to comprehension of the unknown, the material
as key to the immaterial. It would seem that it has had some of this
value for Victor Turner, when he discusses Ndembu symbolism.
'There are religious depths here that cannot be fathomed by the
analysis of observational data. The symbols I have discussed have a
fathomless lucidity of meaning which men of every grade of cultural
complexity can grasp intuitively if they wish' (1962,172). This is the
language of faith. In such statements, as in others about the 'numi-
nous simplicity' of presentation of sacra (1967, 108), one is led
towards a religious dimension of symbolism. Yet it seems as if
Turner does not wish to obtrude this but rather bring down the
argument to the level of the difficult and complex relationships
between the overt and the submerged, the manifest and latent pat-
terns of meaning, as interpreted in terms of behaviour in social
contexts. If we can agree to understand him in such sense, then
we can accept his series of vivid formulations about the positive
nature of symbols in social situations. With a permissible amount of
reification, he points out how symbols, with their complex meanings,
unite the organic with the socio-moral order, over and above the
conflicts within that order. Powerful drives, associated with human
physiology, especially of reproduction, are endowed by the ritual
process with a normative quality, and reinforce this, so being made
to appear obligatory. So symbols are both resultants and instigators
of this process and encapsulate its properties (1969, 52-3). At the
same time they have important structural relationships, both as
series and as representing elements of harmony, integration, tension
or conflict, in the operations of the society.
There is one aspect of Turner's methodology which I think is
open to question, not in its results so much as in its claims: I refer to
194 SYMBOLS
the concept of 'social drama', examined by means of the 'extended

case method'. Gluckman is right to contrast this with evidence from
'apt and isolated illustrations', and appropriately credits Clyde
Mitchell and Victor Turner with having utilized the extended-case
method to great advantage in their Central African social and politi-
cal studies. But to view social relations 'through a longish period of
time', seeing how various parties operate mystical beliefs to serve
their own interests, analysing the conflicting pressures of discrepant
principles and values as generations change and new persons come
to maturity (Gluckman, 1965, 235-8), as in Turner's study of
struggles around the succession to headmanship of a Ndembu vil-
lage, is no such novelty as Gluckman thinks. Granted some dif-
ference in scale, though not in conception, Ian Hogbin's analysis
'The Father Chooses His Heir: A Family Dispute over Succession in
Wogeo' (1940-41), lays bare the roots of a protracted political issue
in dramatic terms, with the richness of perspective and structural
appreciation required. The concept of 'social drama' related to this,
in its use of sequences of disputes (inter-connected 'eruptions of
conflicts' as Turner has called them (1957,91)) provides a convenient
focus for a body of data for analysis of social process. But it has been
rather unsophisticated theoretically, in two ways. Generally, it
raises the question of what is meant by calling a crisis or conflict a
drama. This suggests a scheme of role-playing, with possibilities of
assumption and abandonment of social positions in response to
different types of interest - a plausible but not the only way of
looking at social process. More specifically, it calls for comparison
with the notion of 'dramatism' in social interaction, a notion pro-
pounded especially with great vigour by Kenneth Burke, followed
by H. D. Duncan, and applied much more widely than to situations
of conflict. Depending upon the double meaning of the term 'act',
the dramatistic theory of action (with its 'dramatistic pentad' of act,
scene, agent, agency, purpose) purports to be basic to an under-
standing of the nature of symbolic action, especially in verbal form.
Analyses in this frame of reference tend to be somewhat arid by
comparison with the vivid sinewy handling of concepts and material
by Turner, but the whole notion of social drama seems most effec-
tive when used deliberately in a metaphorical sense, not as a methodo-
logical tool. The 'processional form' which Turner has identified in
MODERN ANTHROPOLOGICAL VIEWS I95
his 'social drama* (1957, 91-2): breach; crisis; redressive action;

re-integration or recognition of schism - seems not to be intrinsic to
the notion of drama, but to be very much in line with Radcliffe-
Brown's type of analysis of restoration of equilibrium after com-
mission of an offence.*
A P L E T H O R A OF INTERPRETATIONS
The image of a 'forest of symbols', with Nature as a temple, and the

symbols observing the passer-by with a familiar (? 'old-fashioned')
look, is attractive. But Baudelaire was concerned with more than the
properties of symbols as factors in social action, as positive forces in
an activity field. He was concerned implicitly, in poetic form, with
the correspondence by means of symbols between the material world
and the spiritual world (cf. Michaud, 1947, 721). I have the impres-
sion that something of this attitude is shared by some anthropo-
logists. But this manifestation of the new Symbolism or recrudes-
cence of the Romantic movement, though shared by many people
with different kinds of commitment and though reminiscent of the
wave of enthusiasm for the study of symbols of a century and a half
ago, is really of a new order. The modern attitude is not Romantic in
the popular sense of being absorbed by concern for the sentimental
or the exotic. It is Romantic in a more technical sense, concerned
with stress and conflict in the personality, impressed by ambivalence
and ambiguity, emphasizing confusions between appearance and
reality, holding that problems of existence are not ultimately soluble
- or at least not soluble by human effort. In such a view, symbol, if
not mystery, is allied to mystery. In some interpretations, art and
religion may be closely intertwined with the scientific pursuit. In
some, symbol seems implicitly to be equated with revelation. Symbol
conveys truth not otherwise accessible, a truth which intellectual
approach is too rational to apprehend.
My own reaction to this is to concur that symbol may embody
mystery, and may embody some measure of revelation. But I see
that mystery and that revelation as belonging to the realm of human
* For symbolism in what seems more appropriately-termed social drama - for example
possession cults-see Leiris, 1958; Firth, 196yd; in theatre proper, see Peacock, 1968 (with
reference to Burke).
196 SYMBOLS
imagination and human comprehension. Men in their social exis-

tence construct and inherit the intellectual and emotional frames in
which they set and attribute meaning to the details of the external
world. The problem of which is the 'reality'; the world outside or
the frame through which we perceive it, is not relevant for our pur-
poses. As anthropologists, we treat the phenomena we study as
being in principle capable of elucidation. They may evade our
understanding because of their complexity. And there may be a kind
of Heisenburgian element in the situation of study, which may
prevent us from finally apprehending separately and completely the
position both of the observer and of that which he observes. But we
make continuing efforts to reduce the area of the unknown. From
this point of view, while stress and conflict are not denied, ambiguity
is sought for clarification, not for its obscurity. Confusion between
appearance and reality are seen as common, but as part of human
fallibility and error rather than as part of the nature of things. In all
this, anthropologists are aware of the social dimension of symbolic
behaviour. Moreover, imagination, intuition, creative capacity, aes-
thetic sensibility may all be pressed into service in the task of symbol
interpretation. And we hope to use the insight we gain from symbol
interpretation as we use what we gain from aesthetic appreciation-to
deepen our sensitivity of relationships and provide us with fresh
ways of conceiving of meaning.
This is one way of looking at a Romantic revival in modern
anthropological study of symbols. But other trends in modern
'symbolic anthropology' do not conform to such a pattern. Some
are marked by an intense analytical interest in what Victor Turner
referred to as 'positional meaning' of symbols and has been more
generally characterized as structuralism. Others have displayed a
much more pragmatic concern for the place of symbols in systems of
action.
In a broad way one can see a series of dichotomies in the applica-
tion of anthropological effort, partially overlapping but having their
separation to a considerable degree in differences of values and
philosophical, even aesthetic approach. There is an inclusive view, in
line with the thought of G. H. Mead and Talcott Parsons, and
represented for example by David Schneider in anthropology, that
all relations of people to one another are mediated and defined by
systems of culturally structured symbols. On the other hand, with-

out necessarily denying this, many anthropologists have preferred
to focus on a study of symbols in more explicit convention, dis-
tinguishing them from more pragmatically oriented behaviour.
Parallel to this division up to a point is a division on interpretation,
between those who, like Levi-Strauss, are interested primarily in the
structural relations and transformations of symbols internally within
the system; and those who argue that the validity of such interpreta-
tion is very limited and that such meanings make very little sense
without consideration of the social uses to which symbols are put.
(The situation here is sometimes like that of the religious contrast
- in salvation being attained by attention to thought rather than to
works, with the claim that the former is both more revealing and also
intellectually more rewarding.) There are divisions between those
who regard anthropological analysis as concerned primarily, even
solely, with the understanding of collective symbols which are
socially shared, and those who think that this understanding must
remain very incomplete without taking account of individually-
developed symbols. There is also a dichotomy of sphere, if not of
interest, recognized between the symbol patterns of tightly knit
groups, concerned for instance with protection of interests and
transmission of codes, and those of looser social aggregates, con-
cerned more with expressive forms of symbolism which allow their
members more self-realization. And then, very forcefully, the moral
theme offers a dichotomy which some would translate into other
terms: should 'symbolic anthropology' be an academic study de-
voted to elucidation of meanings as a tribute to knowledge, or
should it be more bluntly involved in unravelling the knots in social
problems and contributing to change in society, or at least to an
understanding of change? This last point of view takes several
forms. For Mary Douglas, one of the gravest problems of our day is
the lack of commitment to common symbols; the whole history of
ideas should be reviewed in the light of the power of social struc-
tures to generate symbols of their own, deceptive symbols, which
purport to separate spirit from matter and lead to alienation; it
behoves the anthropologist then to interpret alienation (1970, 1,
151). For Abner Cohen, the challenge to social anthropology today
is the analysis of the dynamic involvement of symbolism, or of
198 SYMBOLS
custom, in the changing relationships of power between individuals

and groups (1969, 219).
Here I can illustrate only very briefly a few of these trends,
catching up some of the references I have made earlier.
First, a reference to the structuralist approach, led pre-eminently
by Levi-Strauss. Since comment upon his work has now become
almost an industry, it is enough here to indicate simply its position in
the body of anthropological studies of symbolism. In his discussion
of his Nambikwara field material, Levi-Strauss evidently found little
of a symbolic character to interest him; indeed he stated that
Nambikwara rites did not seem to be in any way symbolic (1948,
126), being rather pushed to physical extremes and showing sincere
spontaneous sentiment. But deeply concerned with the problem of
the universality of human nature, holding that the subject matter of
social anthropology is the communication of man with man by
means of signs and symbols-a kind of semeiology - he has become
preoccupied with questions of variation in symbolic form. Led partly
by Marx and Freud to believe that if inquiry be pushed deeply and
widely enough, variation can be seen to be historically and logically
determined, Levi-Strauss embarked on the gigantic and laborious
undertaking of interpreting the vast mass of material of symbolic
variation which appears on the one hand in the relations of kinship
and marriage, and on the other, in myth. To do this he has had to
make certain assumptions of simplification, each open to challenge.
The first assumption is that he is dealing primarily with structures,
not functions, and can therefore handle the variations and the sys-
tems of which they form part, in their own terms. Secondly, the
structures are ideal or model versions, not behavioural versions.
Thirdly, they are conceived as intellectual structures, both for pur-
poses of analysis and also as representative of human thought pro-
cess. To be prepared to make such drastic and far-reaching assump-
tions in the modern climate of social anthropology was in itself a
daring, creative act; to follow out their implications so rigorously has
demanded an even greater tenacity. The results are well-known. The
study of the form and meaning of symbols has been greatly enriched
by the method of consideration of them in terms of their structural
relations within defined systems, irrespective of their relation to
external factors. Even if at times it assumed the shape of an elitist
anthropology, structuralism has had pervasive effects, and con-

tributed indubitably a great deal to our understanding of symbolism.
In the most conventional field of anthropological studies of
symbolism, that of ritual and associated beliefs, much analysis and
argument still proceeds along what have become classical lines. From
earlier ideas of witchcraft beliefs, for instance, as symbolic ways of
representing explanations of misfortune, anthropological inquiry
has delved deep into structural correlates of witchcraft accusations,
in terms of kinship and status relations, and economic and political
inequalities. Jean La Fontaine has said of the ideology of witchcraft
in Bugisu that it is both a symbolic statement of conceptions about
society and an active force influencing behaviour (1962, 213). Mary
Douglas, in comment on an impressive series of modern essays, says
that the symbols of what we recognize across the globe as witchcraft
all build on the theme of vulnerable internal goodness attacked by
external power - though the symbols vary according to local social
structures and local patterns of meaning (ed., 1970,26). In analogous
fashion, the study of spirit medium cults (possession cults) has con-
centrated in recent years very much on the way in which the actions
of the medium or possessed person, and his or her dialogue with the
other participants, have symbolized their status relationships, and
the tensions arising among members of the community in their daily
life. Following on Turner's work on 'cults of affliction' among the
Ndembu, Richard Werbner has shown how among the neighbour-
ing Kalanga, illnesses believed to arise from breach of domestic
morality are treated by women believed to be possessed by guardian
demons. These women symbolically act the part of lions; they
assume male, elder symbols (and Werbner here recalls Bateson's
Iatmiil analysis (1936) where the adoption of male paraphernalia by
women serves to symbolize authority in the specific cultural idiom). As
lions, the women grapple ritually with selfishness which cheats and dis-
posseses close kin of what should be theirs; they tell one another home
truths through the cult (Werbner, 1971, 312, 322, 324). Moreover,
Werbner indicates that such cult performances are used as symbolic
means of status competition and manipulation - that they are used for
preservation of small group identity and not only as Turner suggested,
to give dominant emphasis to unity of the whole society (1971,321).*
* The symbolic significance of behaviour in spirit medium and possession cults has not yet
200 SYMBOLS
A special field of anthropological interest in post-war times has

been the 'cargo cults' and other millenarian movements on which
now there is a vast literature. Their interest for symbolic study has
varied, but among the more sensitive interpreters the problem of
their symbolism has always stood high.* With many of these cults
the account given of their symbolism, if complex, has been primarily
ethnographic. In Weston La Barre's elaborate study of peyotism, 'an
essentially aboriginal American religion, operating in terms of funda-
mental Indian concepts about powers, visions and native modes of
doctoring' the symbolic interpretation is very intricate. For example,
he points out that the Huichol ritual paraphernalia is 'heavily sym-
bolized'. With his eagle and hawk plumes the singing shaman can
see and hear anything anywhere, cure the sick, transform the dead,
even call down the sun. The plumes symbolize the antlers of deer,
which in turn symbolize peyote and the chair of'Grandfather Fire',
the greatest shaman of them all, whose flames are his plumes. Deer
antlers also symbolize arrows, the arrow being par excellence the
symbol of prayer... and so on (1969, 32). Among much else, the
development of an elaborate symbolic system of such order means
that a social premium is put on knowledge of the symbolic relation-
ships, with obvious status implications. But some anthropologists
studying millenarian movements have adopted a standpoint which
received adequate systematic treatment on any scale. Much detail can be obtained, however,
from works such as those of Eliade, 1964, Beattie and Middleton, 1969, and I. M. Lewis,
1971, also Marcelle Bouteiller, 1950, and the fascinating study of voodoo by Alfred Metraux,
1959. For emphasis on the importance of symbolic communication in such cults, see Firth
196yd, i97od.
* For example Kenelm Burridge, i960, 1969. Cf. John Beattie, 1966, 71. Beattie argues,
much as I have done, that what is primarily important in 'cargo cults* is that they are symbolic
ways, akin to ritual drama, of doing something about situations otherwise felt to be unen-
durable. 'They are recourses, in times of stress, to the consolations of rite and drama; in a
very fundamental sense to the consolations of make-believe.'
Criticism has been raised against my own use of the term 'fantasy' in reference to some
'cargo cult' behaviour, as by Burridge (1969, 120,123-4). I think there is misconception here.
Some modern anthropologists shy away from any suggestion of 'fantasy' or 'irrationality'
as if it was a kind of pollution. Clearly, by themselves 'fantasy' and 'irrationality' are no
explanation, nor have I offered 'fantasy' as such. There are also more important things to say
about 'cargo cults', as I have indicated. But if the accounts are correct, in some of these cults
the people engaged in a simulation of technical procedures which experience must have
shown them by comparison were inadequate in themselves to produce the desied result.
Their belief that nevertheless they had power to induce aeroplanes to land or ships to call with
'cargo' I term fantasy - an attempt to bridge a felt gap between ends and means by imaginative
construction. I see no reason for the anthropologist to abdicate completely his function as
external observer of such a situation, in giving such labels. But what is significant is the use
made of such fantasy, in socially structured situations.
raises a basic problem in our whole inquiry, significant for the general
study of religion: the manner in which the beliefs and practices of the
cults can be regarded as symbolic at all. Are these beliefs and prac-
tices in fact not 'representations' but direct expressions of a reality
which the anthropologist can do no more than accept? There is, in
my view, a certain amount of dodging around this issue. Burridge,
for instance, in a very thoughtful general review of the theory of
millenary activities, is dissatisfied with descriptions of them as
'symbolic means' of dealing with problems - 'true, if it is not wholly
a truism', and 'ethnocentric'. He allows that Monsignor Knox's
'authoritative' study of eighteenth-century millenarian movements
referred the historic events to 'his own and his subjects' view of the
symbolic system' and adopted much the same approach as many
anthropologists, particularly Stanner. But Burridge still stresses the
inadequacy of 'ethnographic' explanations and in the end plumps for
a 'Hegelian' approach which attempts to make use of a total ex-
perience to explain itself as well as other kinds of social order,
experience or tradition. Such a 'Hegelian' explanation 'admits the
operation of a transcendent power'. In recognizing the existence of a
force whose nature we do not yet understand, a 'Hegelian' explana-
tion 'is clearly exploratory and therefore potentially fruitful' (1969,
120-37). Such putative suspension of judgement about the symbolic
quality of the cult phenomena presumably corresponds to a personal
commitment of a broader kind about the nature of reality (cf. I. M.
Lewis, 1971, 28).
The study of Australian aboriginal religion by W. E. H. Stanner
examined ritual symbolism in an equally penetrating, but more
positive, way. Agreeing with Nadel that 'uncomprehended' symbols
are not of concern for the anthropologist, Stanner makes the impor-
tant point that if this means 'unaccompanied by intellectual concep-
tions' it cannot be simply inferred from wordlessness. Moreover, a
methodical search for congruence between ritual facts of different
orders may show that an implicit and apparently uncomprehended
symbolism of one order is formulated explicitly in another order.
But even then, some 'going beyond the facts of observation' is
intrinsic to the act of study in every anthropological field, and not
even theoretically separable from it. So, in the several levels of
awareness of ritual symbolism, a few symbols are clearly recognized
202 SYMBOLS
as such, but the vast majority of rites are practised without clear
recognition of their symbolical character. Yet the symbolisms are
constituents of collective acts of mutuality, with a logical structure,
a detectable range of meanings, and an aesthetic appeal as well as a
'premial' place in the social development of individuals. Stanner
pointed out that the rites involve anything from scores to hundreds
of men, in unguided co-ordination, with no master of ceremonies as
such; hence the spatial patterns of ritual are of cardinal importance.
At the place of congregation, Stanner recognized four systems of
symbolism in congruence - spatial configuration, gesture, language,
music - and was impressed by the practical, logical and expressive
efficacy of the form (i960, 61-4). With a combination of field
observation, theoretical rigour and imaginative insight Stanner was
able to demonstrate and interpret cogently and systematically the
development of symbolic forms in aboriginal ritual. His broad con-
clusion was that the Murinbata aborigines among whom he worked
were in the given rite expressing outwardly a complex sense of their
dependence on a source outside themselves - that by an 'inner
paradigm' of setting apart, destruction, transformation and return,
much of it unverbalized, they were intimating a mystery 'of good-
with-suffering, of order with tragedy' (i960, 70, 77). And yet Stan-
ner was prepared to argue that a full understanding of the religious
symbolism was to be gained only by a thorough morphemic analysis
of the whole language, and while this remains unfulfilled we have not
penetrated 'the true inwardness of the stuff of symbolism'.
With this analysis may be compared that of Nancy Munn. In an
early study of Walbiri graphic signs she meticulously interpreted
aboriginal totemic designs as devices for conveying narrative mean-
ing, and has had a general interest in iconography. In a study of the
effectiveness of symbols in Murngin rite and myth (1969) she drew
inferences about the way in which collective symbolic forms could
transform subjective experience. Though working along very dif-
ferent lines from those of Stanner, with reported, not observed
material, she also emphasizes the relation between narrative code and
system of ritual action, and in somewhat parallel fashion, sees the
myth as conveying body destruction images which the ritual con-
verts into feelings of well-being. But whereas Stanner is much con-
cerned with the symbolic significance of actual spatial forms as
vehicles for conception and expression of the meaning of things,

events or conditions, Munn uses the concept of what she terms
'symbolic space', referring to the degree to which the individual
experience of the ritual performer is identified with the experience of
the persons described in the myth. Following a lead from Levi-
Strauss's analysis of shamanism (1949) Nancy Munn argues that the
myth-ritual complex acts as a mechanism of social control by pro-
viding an external, regulatory system for states of bodily feeling, and
indicates diagrammatically a series of postulated cultural codes for
this process. Her conclusion is that analyses of symbolic structures
should not simply be directed at abstracting underlying conceptual
systems, but should be concerned with symbols as mechanisms which
regulate the orientations of actors to each other and to common
situations.
The symbolism of myth is very often studied side by side with
that of ritual, as in the analyses mentioned of Munn and Stanner. But
whether influenced by or separate from the structuralist position of
Levi-Strauss, studies of myths alone, in particular of classical myths,
have been offered as contributions to the understanding of symbolic
forms. I mention here only analyses by Leach (1961b) and by
Terence Turner (1969b). In an essay with the intriguing title of
'Levi-Strauss in the Garden of Eden' Leach points out that the
study of myth has always had a central place in anthropological
studies, but distinguishes 'symbolists' from 'functionalists' in such
sphere. The functionalists are represented by the 'charter' theory of
Malinowski; the symbolists by Frazer, with Freud and Cassirer in
support. On symbolic lines, myths are interpreted as explaining
(away) basic human problems such as the origin of the world or of
death by manipulating symbolic representations of these facts. Hold-
ing that Levi-Strauss had reopened what seemed to be an almost
closed argument, Leach reviewed his analysis of the Oedipus tale,
and went on to examine the creation story as told in Genesis. With-
out suspending his critical judgement of the structuralist thesis,
Leach pointed out that adoption of the view that a myth could be
studied as a thing in itself allowed for scrutiny of'literary myths' for
which social context is largely lacking. In a later work (1966a) Leach
pursued his structural investigations into the theme of the succession
of Solomon, with particular reference to the occurrence of patterned
204 SYMBOLS
contradiction as a presentation of historical facts. Terence Turner's

essay takes up Levi-Strauss's pioneering analysis of the Oedipus tale
and develops it in two ways. On the one hand, he argues that
Levi-Strauss paid insufficient attention to the patterning of the narra-
tive sequence of the myth - that the form of the 'story' is itself
significant for an understanding of its symbolic significance. Myth in
Terence Turner's view is able to present a diachronic, unique sequence
of events as a synchronic model only because of its 'story' develop-
ment. On the other hand, he argues that myths in their narrative
quality provide models which a listener can use to manipulate his
tensions about social conflicts, and so help to control them.
In such work one may detect both the strong influence of the
structuralist position of Levi-Strauss, and a moving away from that
position towards greater contextualization of the symbolic forms.
That with Munn and Terence Turner, for example, this has involved
some fairly abstract postulates about the effects of symbolic presenta-
tion, and conceptualization upon individual action, is perhaps in-
evitable.
Finally, in this compressed review I mention two lines of inquiry
which both seem to offer prospect of great ethnographic develop-
ment, even if the theoretical implications are not so clear. One is
what I may call an autologic study of symbols, which in our Western
field means the kind of examination which Mary Douglas has given
to our concepts of purity and defilement (1966) or to some social
experiences expressed in symbols based on the human body (1970);
which Lloyd Warner gave to the meanings and functions of Chris-
tian symbolic life in America (1961); or David Schneider has given
to the basic symbols of American kinship and their relation to one
another (1968). The recent studies of Schneider and of Douglas,
offering comprehensive theoretical frameworks for study of sym-
bolism of very different kinds, both take as object lesson the sym-
bolic quality assigned to the biological properties and functions of
the human body. On a much lesser scale but more immediately
empirically grounded is an analysis by Vieda Skultans (1970) of the
symbolic significance of menstruation and the menopause in a Welsh
community. She discovered that the concept of the menopause was
a cultural rather than a biological one, and was used to reflect ideas of
woman's social role as sexual partner, mother and housewife. In par-
ticular, in a manner conformable generally to Van Gennep's theory

of rites de passage the menopause for some women was envisaged as
marking the transition from one role to another. (A similar point
of view was put by Marion Crawford, New Society, 29 October 1970,
to the effect that in symbolic terms the menopause is a kind of im-
posed disengagement from the central role in a woman's life, that of
mother.) From some of this autology, this scientific study of our-
selves, certain conclusions of practical application may be drawn, for
use in social operations concerned with families and with problems
of personality.
More broadly, the emphasis given by both Mary Douglas and
Victor Turner to the significance of experience carries some didactic
force. Turner indicates the importance of experience in dialectical
terms (1969,97,127), of alternating exposure of individuals to struc-
ture and to the communion of fellowship {communitas). He holds
that communitas has an existential quality involving the whole man
in relation to other whole men. Mary Douglas too makes the concept
of social experience the basis of her programmatic approach to the
problem of differences in ritual and symbolic interest, both as regards
persons and societies. But at this point they seem to diverge: Turner
to look with approval - as I interpret him - at applications of com-
munitas or anti-structure, seeing it as a necessary creative phase in
the dialectical process of society; and Douglas to regard anti-
ritualism disapprovingly as an ineffective way of humanizing society.
In her view, 'it would be more practical to experiment with more
flexible institutional forms and to seek to develop their ritual expres-
sion (1970, 155).
The second line of inquiry I mention is also autologic to a con-
siderable degree, but starts from a different base: it is concerned with
the political function of symbols, and is represented for example by
Abner Cohen (1969). Cohen argues that there can be no science of
symbolic behaviour in any general sense. Symbolic forms are the
product of creative work and their study is partly in the field of the
sociology of art. Many men keep their creative symbolization to
themselves, while others externalize and try to share it. All social
behaviour is couched in symbolic terms. But since nearly all social
behaviour has a power dimension, the symbolization of power rela-
tions is an exceedingly important feature of social life. Indeed,
206 SYMBOLS
Cohen goes so far as to say that in social anthropology we are

interested in symbols mainly in so far as they affect and are affected
by power relations, I think this is a reductionist argument - that a
great range of expressive symbols at life crises, for instance, are not
power symbols; and that status symbols, which are equally a con-
cern of the anthropologist, should not be merely equated with power
symbols. But power symbols are indeed a critical, under-studied
field; and social anthropology can make a basic contribution to their
analysis. Such analysis in Cohen's opinion should be of value to
political science in the contemporary scene, as well as forming an
essential constituent of social anthropology itself.
So in the modern anthropological study of symbols, though there
are clear intellectual satisfactions, it is not all simply an intellectual
exercise. Some symbol analysts hope to contribute to a further
understanding to basic processes of human thought. Others hope to
throw light on facets of human experience which can in a general
way give guidance and encouragement to people as they face life's
problems. Others still hope to expose more directly and more force-
fully the operations whereby symbols become social instruments in a
struggle for control.
Behind much of this work two major questions still stand. What
is meant by saying that symbols or symbol systems have an existence
of their own, instigate action, enter social relations as actors; is this
more than a metaphorical kind of statement? And what is the nature
of collective symbols, in relation to the symbolization of individuals
in a society? This latter question will be discussed in an anthropo-
logical setting in the next chapter.
Chapter 6
PRIVATE SYMBOLS AND
PUBLIC REACTIONS
In the study of symbolism by social scientists there are two fairly

clearly recognized domains. In one, the province of sociologist and
anthropologist, symbols are taken as being characteristic of sets or
groups of people, of institutions or of types of situation. So in
anthropology we speak of the symbolism of the Nyakyusa or the
Ndembu; the symbolism of sacrifice or of peyote or of witchcraft.
Sometimes the word symbol is given such general currency that it is
equated with 'custom' (Abner Cohen, 1969,222); it is the symbolism
of collectivity - of myth, of ritual, of social structure - with which
anthropologists are mainly occupied. As against this generalizing
viewpoint there is the broad psychological domain in which the
study is made not exclusively but basically of symbolic forms pre-
sented by individuals, often not shared with other people, and cor-
responding essentially to personal interests, claims, stresses. The
symbolism of dream, hallucination, prophetic revelation or drug-
induced experience belongs in this domain; and so does much of the
initial creativity of poetry and the visual arts. A great deal of this
personal symbolism is private in the sense that it is intended to be, or
construed to be exclusive to the individual concerned, offering his
own particular solution to his problem of adaptation to some aspects
of his immediate environment or to his conception of the world. To
a person who is dissociated or schizophrenic, the shapes, colours,
voices of his vision represent reality in a way which marks him off
from ordinary people. This tends to pose special problems to those
who wish to be in touch with him or help him. In a notable account
of his own psychosis, John Perceval (in a narrative edited by Gregory
Bateson, 1961, 271) pointed out how many of his delusions con-
sisted in mistaking a figurative or poetic form of speech for a literal
one: when his Voices' told him to wrestle with his keeper, or to
207
208 SYMBOLS
suffocate himself upon his pillow, he tried to do this physically, not

realizing, as he later saw, that it was intellectual struggle and stifling
of his feelings that were meant, by another part of his personality.
At first sight the relevance of the phenomena, and the problems of
interpretation, do seem very different in these two domains. An
anthropologist is concerned to find out what a given symbol cor-
responds to in the general understanding and operations of a body
of people. He seeks if not consensus at least the highest common fac-
tor in referents. He does this because he is looking at symbols as
bases for or expressions of common action; the element of com-
munication of meaning is dependent upon the possibility of shared
understanding. A psychologist - using this term widely - may be
interested in this too, but he is also interested in the departure of the
symbol from consensus, with the clues it offers to individual de-
velopment, and in the way in which social or physiological stimuli
have given rise to specific reactions in the individual personality.
Edmund Leach in characteristically brisk manner has characterized
the aim of public symbolism as being communication, and of private
symbolism as expression; the former he regards as a subject for
anthropological study, the latter for psychological study.
As an example of division of spheres take the very rich symbolic
field explored by R. M. Berndt in his description and analysis of the
Kunapipi cult of north-east Arnhem Land (1951). In this area the
name Kunapipi expresses a dual concept: it refers to a Fertility
Mother or Mothers, and to the great Rainbow Snake (cf. p. 137).
'This is the symbolism of the Uterus and the Penis, natural instru-
ments of fecundity' (1951, 12). In myth, in ritual and in decorated
material objects, this basic symbolism is given form, transmitted to
new generations, and validated by ceremonial sexual relations among
Kunapipi participants. The elaborate symbolic procedures, accom-
panied by initiation of novices, co-operation of local groups, intri-
cate kin exchanges, demarcation of roles between men and women,
dancing and aesthetic display, are clearly of public concern, and of
direct anthropological interest. On the other hand, many men who
participated in Kunapipi had dreams which expressed their deep
involvement in the ritual, as well as some degree of individual
identification with Kunapipi figures - from being ordinary people,
in their dreams they become participants in the myth, ancestral or
PRIVATE SYMBOLS AND PUBLIC REACTIONS 209
totemic beings. Now dreams, as expressions of the personality, are

clearly the province of the psychologist; they are series of private
symbols.
The attitude of many British social anthropologists towards
psychology has been ambivalent. Twenty years ago, Evans-
Pritchard, following RadclifFe-Brown and Durkheim's view (cf.
Murphy, 1971, 68) stressed the essential difference as he saw it
between social anthropology and psychology, and his argument for
independent pursuit of their research has been quoted frequently. Its
effect has been seen in part in the attitude complained of by loan
Lewis - that if there is one thing that unites most British social
anthropologists it is their fierce antagonism towards psychology and
psychiatry, and their 'complacent contempt' for the psychological
aspects of the phenomena they study (1971,178). Whether this is to
be taken literally or not, I have myself argued that British social
anthropologists who abjured 'psychology' still made statements
about thought and feeling - often on little evidence; that the distinc-
tion often made between study of 'the individual' or 'mental sys-
tems' and that of 'groups' or 'social systems' was inadequate; and
that anthropologists must come to terms with psychology (Firth,
1951b, 487).
Many American anthropologists have done so, with varying re-
sults - one of the most impressive, scientifically, being the series of
essays produced by Anne Parsons before her death in 1964, com-
bining high theoretical grasp with a range of comparative observa-
tional data in open and clinical conditions (1969). In the British field,
with varying degrees of acceptance of psychological and psychiatric
concepts, many anthropologists have independently illumined their
interpretations of symbolism by such means - for example Margaret
Field, Elizabeth Bott, T. T. Hayley, Audrey Hayley, J. B. Loudon,
Geoffrey Gorer, R. M. Berndt, Victor Turner, Meyer Fortes,
Audrey Richards. Problems of integration of findings, of translation
of particular results into general propositions, of the kind of assump-
tions that are most useful to make about mental functioning, still
remain. In their bearing upon symbolism, the methodological issues
for a social anthropologist can be seen in three main areas: relevance;
access; typicality.
Relevance is a matter of implication for interpersonal behaviour.
2IO SYMBOLS
Thoughts, dreams, emotions of people in the expression and develop-

ment of their own personality are not an anthropologist's concern
but they become so when translated into actions affecting the con-
duct of other people. To dream of a tiger, to identify one's boss with
a tiger in a phase of mental disturbance, are matters not necessarily of
anthropological interest - though they may have social components,
for example of status resentment. But if in a trance a medium por-
trays or engages in battle with a tiger spirit as part of diagnosis and
therapy of an illness - as can happen in north-east Malaya (Firth,
1967c!) - the ideology of this and the reactions of the patient,
spectators and religious functionaries are a proper subject of anthro-
pological study. Information about a person's thoughts, emotions
and images is relevant for anthropological interpretation if these have
social effect, particularly if they affect people's group relationships.
Access is a more difficult matter. To work with verbalized material
may offer unfamiliar linguistic problems, and poses basic questions
of relation of word to thought and emotion. But the general prob-
lems involved are confronted by anthropologists in most of their
work. Non-verbalized beliefs, thoughts, feelings are another matter.
These may be formulated but unspoken; unformulated but aware to
the person's mind; or he may be unaware of them, revealing them
only by his actions. The techniques of the anthropologist are slender
to cope with such material, and anthropological opinion has differed
considerably on whether to admit such material for symbolic study.
But in varying combinations of assumption and inference, most
anthropologists have committed themselves to some statements
about human mental process in general or particular thought and
emotional constructs of people, without having been told such by
anyone. My own view is a rather cautious concurrence in such
methods of inference, provided the empirical basis of the generaliza-
tions is made clear.
The question of typicality deserves more attention than it usually
gets from anthropologists. The Durkheimian legacy in social anthro-
pology has encouraged neglect of individual variation. In study of
local grouping, kin units, and ritual such variation speedily became
patent through observation of behaviour. But in statements about
concepts, beliefs, emotions it is all too easy to assume typicality from
sheer lack of breadth of evidence about degree of variation. State-
P R I V A T E SYMBOLS A N D P U B L I C R E A C T I O N S 211
ments about attitudes towards 'master symbols' or 'dominant sym-

bols' often show disregard for this question of typicality, and so may
miss important developments in social reaction and symbol forma-
tion.
My basic subject of inquiry in this chapter is - granted precautions
to be taken in the study of mental functioning, what is the relevance
for anthropologists of what can be initially be described as private
symbols?
An overlap in content may be of some significance, in that themes
familiar to anthropologists in a general, category sense can be seen
operating in a personal, private context. So, when John Perceval's
Voices' told him repeatedly that such and such persons were his
mother, sisters and brothers, and he later conceived that this was a
spiritual idea or that there was some other resemblance (ed. Gregory
Bateson, 1961, 274) this is reminiscent of phenomena well known to
anthropologists, of men conceiving themselves as kin to others with
whom no biological tie can be traced. This implies a basic question
as to what is the nature of such concepts, what is their symbolic force.
The problem of significance of private symbols touches the interest
of social anthropologists from another angle - that of psychological
studies of religion. For psychologists, while focusing on individual
behaviour, have often related this to institutional patterns. In his
Varieties of Religions Experience (1902) William James held that he
was examining 'personal religion', which was more fundamental
than theology or ritual. Gordon Allport, in The Individual and His
Religion (1950, 119) cited James to the effect that in religious think-
ing we make use of such 'poor symbols' as our life affords, though it
is unfortunate that they must be the same symbols as do duty for
common sense and for scientific discourse. So confusion arises;
personalities are infinitely varied; the symbols that appeal to one
person may be meaningless and repulsive to another. Yet doubters
who cavil at the unclarity or unacceptability of any particular symbol
are speaking only for themselves; a re-forming of religious symbols
to give greater designative fidelity for one person would merely
throw others into a state of doubt. Allport agrees that various re-
ligious sects are defined by having approximately common sets of
symbols for their members. But he argues that the cosmic conditions
pointed to in religious language (the referents of the symbols) are
212 SYMBOLS
not demonstrable, not knowable in their entirety, and therefore not

accurately signifiable. So individuals are justified in effect in match-
ing their symbolism to their aspirations, self-imposed ideals. Where
they concentrate on the constituent images and symbols rather than
on the general intent of the whole act of worship, they tend to fall
into doubt. This, which may be called a case for a private symbolism
in religion, contrasts to some extent with Thouless's view (in the
Psychology of Religion^ 1961, 114, 147). Thouless held that if an
image has a fairly uniform meaning for different people it is called
a symbol. Unconscious thinking tends to use symbols, and this is
why symbolism plays such a large part in religion. Some of the pro-
cesses of religious thinking, he argues, show an alogical and infantile
character, as shown by the dependence upon symbolism. So it is
difficult to translate religious ideas into the exact language of directed
thinking, which is a voluntary activity, using mostly words and is
a function acquired late in evolution.
So, from such psychological angles, private symbolism is regarded
as relatively free-floating, even at times a kind of dwarfed or under-
developed intellectual activity, less significant than a reasoned
examination of the personal meanings of religious ideas. But from
the other side, sociological and anthropological approaches to re-
ligion have tended to subsume private symbolism under public
symbolism, and to give public symbolism a significant, respectable
place in social action. Even more, public symbols have been regarded
as having power to regulate individual behaviour, to express per-
sonal sentiments, and to dictate forms in which private symbols
present themselves.
CARE I N CLASSIFICATION
As an anthropologist I share this approach. But there is a simple

question of field methodology which must be cleared before we can
consider the relation of public to private symbols more closely.
How is the categorization of a symbolic form as 'public' arrived at?
Anthropologists do not deny that the symbolic formulations current
in a society about the meaning of ritual objects and actions are ex-
pressed in personal terms, often idiosyncratically. The symbolic con-
cepts of a group may have concrete representation in flags or
statues, but their meaning is described by individual people; it passes

through a personal lens. Yet the process by which the anthropologist
takes these personal statements and combines them into a formula-
tion about the public meaning of a particular symbol sometimes
remains obscure. I take an example from one of the most sophisti-
cated studies of an African religion, that of the Nupe by S. F. Nadel.
He describes a masked dance performed to cleanse a village of evil
influences, succeeded about a fortnight later by a sacrifice in public
before a pair of bronze figures, of a man and a woman, the rite being
accompanied by music of rattles and drums. The officiant priest has
the two rattle-players on his right and left. 'The place on the priest's
left, opposite the female figure, signifies "birth", the place on his
right, opposite the male figure, "health"; which symbolism of right
and left is unique in Nupe religious practice.' Later a white cock is
killed and some meat offered to the two bronze figures, with a prayer
for health for the towns and plenty of children (1954, 93-5). Now
there is no particular reason to doubt this interpretation, which
makes good sense in the general social and religious context. But in
the light of discussion about the relative inferiority and inauspicious-
ness of the left side in symbolic systems in Africa and elsewhere (for
example Hertz, i960; Needham, 1967a; Beattie, 1968) the question of
whether we are concerned with a generally acknowledged Nupe
symbolism or only a personal interpretation is significant. The
offerings and the prayers for health and children were only partially
public rites; some of the performance took place before a meal in
which only the elders shared. Now one might postulate that the
'right and left' aspect of the ritual might have been connected with
a faintly outlined division of the people into two ritual moieties. But
Nadel was sceptical of the opinions he got about the exclusiveness of
these moities, particularly since he was told of the rule about one
moiety by only one old man, albeit of high rank. He did not see the
rules in action and suspected that they were mere 'theory' - 'an
attempt to lend neatness and logic to an unconvincing and haphazard
division' (1954, 86-7). But now who told Nadel about the 'right and
left', 'health and birth' symbolism of the bronze figures? Though it
was a 'unique' symbolism, he does not say. Could this interpretation
too have been an attempt to 'lend neatness and logic' to ritual posi-
tions by a reading backwards from the wishes expressed in the
214 SYMBOLS
prayers - a case of pragmatic interests giving rise to structural inter-

pretation? And a bit of private symbolism, perhaps by the priest,
perhaps by one of the elders, to explain something about which there
was no public opinion at all? It could even be that the 'right and left*
interpretation was meant to support some private view about status.
But this we are unlikely ever to know.
Even with Victor Turner's careful description of his methods of
elucidation of the meaning of Ndembu symbols, to get 'reasonably
reliable* interpretations (that is, as he puts it, mutually consistent
interpretations) the relation of public to private symbolism oc-
casionally remains obscure. What Turner gives is what he calls the
'standardized hermeneutics' of Ndembu culture. In such general
interpretations he is not overtly greatly concerned with the 'free
associations' or 'eccentric views' of individuals. What he is after is
the regularities which can be built up into a general pattern (1969,
9-10). This is normal anthropological procedure, and I single it out
here only as an example of the problems that confront us all in
symbol interpretation. Turner quotes individual opinions in support
of his generalizations. But what is it that makes one individual's
opinion standard and another's a free association or eccentric and so
to be discarded - when it is discrepant from the general pattern?
Much information is given to show how the Ndembu regard the
mudyi, the 'milk tree' which is central to girls' puberty rites and
other rituals. Cognitively the Ndembu give the tree many attributes,
associating it with breast development, breast-milk, mother-child
relationship, matriliny, the Ndembu tribe as a whole. They 'speak
and think' about the milk tree as a unity, almost as a unitary power.
Yet while Turner was convinced that his informants genuinely
believed that the milk tree represented only the linking and unifying
aspects of Ndembu social organization, he was equally convinced
that in action situations the tree served as a focus for group opposi-
tion. Much of the behaviour observable in connection with it re-
presents a mimesis of conflict within the matrilineal units or rela-
tionships (1967, 20-5, 52-5). Could it be that more attention to
discrepant, eccentric, private symbolism might have thrown some
light on this apparent refusal to match ideology to action? Or could
it be that the more abstract aspects of the milk tree symbolism were
themselves derived from a few informants only, and that this was
really the more private, even eccentric symbolism? For instance, the
explanation that the milk tree is not only mother's milk but also the
matrilineage, 'brought out most clearly in a text (I) recorded from
a male ritual specialist' (1967, 21) occurs also in a text cited from the
ritual specialist Muchona. But Turner adds the possibly significant
note that this man's accounts and glosses were always fuller and
internally more consistent than those of other specialists (1967, 134-
5). He had evidently pondered long on the mysteries of his profes-
sion, says Turner, collating material given to him in the course of his
own instruction, in critical review. Here then it seems pretty clear
that public and private symbolism have got fairly well intertwined.
I repeat, this is a dilemma which confronts all field anthropologists.
But the serious problem for social anthropologists is not so much
to demonstrate how one arrives at a general statement of any piece of
public symbolism, as to work out how public and private symbol-
isms impact upon each other. How far do public symbols really
condition the forms of private symbolization? What kind of con-
tribution does private symbolism make to the formation or modifica-
tion of public symbols? In what way, if at all, does private symbolism
have an effect in social action? How far does the existence of clusters
of private symbols, if known or suspected, disturb the community
when they do not seem capable of public utilization but imply social
action with which the community may have to deal? Such questions
are relevant, though often hard to answer.
A very pertinent field for the study of the inter-relation of private
and public symbolization is art. Creativity in an artist is the display of
a personal vision, in which symbolization may play an important,
perhaps vital part. The symbols must be personal, individual,
unique, stamped with the artist's own imaginative power, if he is to
generate positive reaction in other people. (As modern examples
Paul Klee or Pablo Picasso clearly qualify.) Yet if the symbolism
remains purely private, unrecognized, the stimulus of the creative
act is lost to the community. Communication is the keynote. There
must somehow be enough communication between artist and public
for the initial recognition of something of the artist's vision to be
caught, absorbed, generate emotional and intellectual reaction, and
stimulate further aesthetic reaction. Private symbolism must be able
to be communicated to become public symbolism. A symbolic form
2l6 SYMBOLS
which is completely obvious may be banal. But a symbolic form

which has meaning which stays completely locked up in the artist's
private world is not an object of art in any socially-significant sense.
In primitive society the relation is ostensibly reversed - an artist
is concerned to portray symbols already recognized by the com-
munity rather than to invent new symbolic forms which the com-
munity has to construe. But the reversal is apparent rather than real.
For the artist still makes a personal contribution which expresses his
own rendering of material from the common symbolic stock, just as
in our society the artist in fact draws heavily upon already conven-
tionalized symbolic images. But in primitive society the initial com-
mon ground is more openly recognized - the symbols of primitive
art are shared from the outset by groups of people; there is not the
same initial dichotomy between artist and public (cf. Firth, 1951a,
177).
A most interesting analysis in this field.is the study of Abelam
phallic symbolism by Anthony Forge (1965). He shows how the
main symbolic forms are related: the ridge-pole of the ceremonial
house, standing for violence and warfare and associated with the
spear; the penis as generative, nutritive organ, in paintings on the
house fagade and in carvings of clan spirits; the long yams, cultivated
by individual growers with great care and skill, and publicly dis-
played in presentations of rivalry and hostility which confer prestige
on grower and community. Here in the pictorial art, public symbols
are periodically re-created, by private artists, working with almost
a minimum of conscious interpretation. The communication is
effected by the work itself, carried out in common in a broad context
of common meanings. In our society, however, we have the spectacle
of artists often operating a private symbolism and striving for its
public recognition, but becoming frustrated and angry at the failure
of the community to understand what they are trying to convey.
And while it often does occur, the conversion of private symbolism
into public symbolism may take a considerable period of time.
INTERPRETATION OF D R E A M S
Dream is one of the most obvious areas for investigation of the

relation between private and public symbols. A dream is a highly
personal thing, and its symbolism is private in the sense that the
ultimate clues to its meaning depend in some part upon the peculiar
circumstances and mind of the dreamer. Yet to some extent its
symbols are public in that many of them are clearly derived from
shared social experiences, and are recognizably of common form, in
'type-dreams'. In Edmund Leach's categorization, their aim is ex-
pressive for the individual dreamer, as representations of unconscious
mental process; but it is also communicative, since many dreams are
verbally described to other people and indeed are known primarily
by such mechanism. Moreover, their symbols become public in that
they are often the source of social action. It is not surprising then
that anthropologists even before Tylor paid attention to the subject
of dreams and recorded instances which seemed to throw light on
mental process, belief and individual and public reaction. As I have
shown in Chapter 3 (p. 101) study of the symbolism of dreams by
others than anthropologists, with attention to unconscious process,
was a serious subject for much of the nineteenth century.
Many questions about dream symbols may be raised in psycho-
logical anthropology, which do not concern us here.* But apart from
establishing norm and variation in dream content and interpretation
in different types of society, social anthropologists have been con-
cerned with three main questions in relating these private symbols to
the public sphere. First is the significance of dream material as an
index to the degree of social commitment, or social involvement.
I have referred already to Ronald Berndt's material on dreams of
aboriginal men participating in the Kunapipi cult (p. 208). Berndt
collected hundreds of dreams from about thirty men, and found that
while they ranged over a wide variety of themes, many were con-
cerned with indigenous religion, and were regarded as very impor-
tant by the people themselves. Men 'dreamed out' the meanings of
their sacred ritual, coming into contact spiritually, they thought,
with their Ancestral Beings. In dreams they performed ritual, used
sacred emblems in unconventional ways, and dreamed additions to
their totem designs and rites. This private symbolism has public
effect in that, Berndt points out, many sacred designs are altered
because of dreams, and sacred songs too are influenced thereby. In
* For example the relation between manifest content and latent content, see M. J. Field,
i960.
H
2l8 SYMBOLS
a very different context, I myself have indicated how in Tikopia,

dreams have portrayed symbolically social situations of tension such
as friction between chiefs, or chiefs' feelings of responsibility for the
condition of the people (1967b, 166; 1970a, 49n.). Again, Tikopia
incest-dreams have revealed a mechanism for removing some of the
guilt of incest-wishes (1936, 328) and given some reflection of com-
mitment to a social norm. And some dreams symbolized dramatically
the struggles of Tikopia preoccupied with problems of conversion
from paganism to Christianity, and helped to induce them to take
overt action (1970a, 325,327,391,394-5). So, granted the possibility
of distortion and condensation of meanings, dream symbolism has
still seemed to offer evidence about the extent to which individuals
have felt themselves committed to ritual or other social situations -
with implications for the intensity and continuity of their activities
therein, in the public field.
Allied to this is the question of how far the occurrence of certain
kinds of symbolic presentations in dream leads to social action, not
only by the dreamer but also by others. These others may either
share the dreamer's belief in the significance of his dream, thus con-
verting a private symbol into a public symbol, or so respect his
belief that even without sharing it they are willing to give effect to
what he regards as its implications. Evans-Pritchard has recorded of
the Zande that their bad dreams were commonly interpreted in
terms of witchcraft - not as symbols of witchcraft but as actual
experiences of it. Although they usually did not consult oracles about
such dreams, on occasion a man would ask his blood-brother, kins-
man or friend to make oracular inquiry on his behalf, and if witch-
craft was confirmed, would pursue the matter more publicly. Tradi-
tional Iroquois attitudes appear to have been more drastic. Unlike
the Zande, they seem to have regarded dreams as predictive, with the
evil effects capable of being warded off by evasive or cathartic
action. L. H. Morgan noted that dreams were often regarded by the
Iroquis as 'divine monitions', that a dreamer would wander about
inviting others to guess his dream, and that when interpreted, its
injunctions were followed 'to the utmost extremity' (1851, 214-15).
He recorded the case of the sachem Cornplanter, whose influence
had for some years been on the wane with his people because of his
friendly relations with the whites. As the result of interpretation by
another man of a dream of Cornplanter's that he should relinquish

his chieftainship and abolish his ties with whites, he gave up his
office and removed from his house or destroyed all presents he had
received from George Washington and other white leaders; the only
exception was his tomahawk, which he used in unorthodox fashion
to nominate his successor. Similar material has been discussed by
Anthony Wallace, who distinguishes two major types of Iroquois
dreams, symptomatic dreams and visitation dreams, the former in
particular being followed by much public acting-out, often of
cathartic type. Iroquois dreams are not to brood over, they are to be
told or hinted at, and other people rally round with gifts and ritual.
'The dreamer is fed; he is danced over; he is rubbed with ashes; he is
sung to; he is given valuable presents' (1958, 246-7).* Yet despite
Morgan's phrase of following out the dream's injunctions 'to the
utmost extremity', a prudent response to the symbolization is some-
times evident. Cornplanter used his tomahawk, a gift from whites,
to ensure his successor. A man whose dream manifested a wish to
kill either stops short of immolating his real-life victim provided, or
is satisfied with a coat allegedly taken from a dead enemy. The private
symbol is not necessarily or always completely incorporated into the
public domain.
The third kind of question relates to the degree to which the
symbols of a dream are regarded as not merely pertinent to the fate
of the dreamer, but are absorbed by the social body and seen as a
basis for their collective action. Private symbol becomes identified as
public symbol. Historically, millenarian movements, indigenous
churches and analogous attempts of people to forge their own
religious instruments for coping with their destiny have tended to
use symbols provided by dreams. Individual dreams have been
regarded as stimulants and guides to action for the corporate body;
symbol of one has meaning for all. In the New Guinea 'cargo cult'
field, both Lawrence and Burridge have pointed to the common
belief in the validity of information derived from dreams (Lawrence,
1964, 167; Burridge, i960, 179-82). Burridge in particular has
emphasized the notion of imperative contained within a dream, and
the use made of this notion by dreamers in trying to persuade others
* Wallace's claim that the Iroquois independently anticipated Freud's theory of dreams
seems to me far-fetched; it was a very superficial theory of the unconscious.
220 SYMBOLS
to a particular line of action or point of view. So for the Tangu,

dreams can inform, direct and prophesy; they are not simple fan-
tasies woven in sleep; they are a normal technique for solving prob-
lems. Dreams 'tend to pull a future into current, sensible reality;
they give definity to hope, adding faith'. When dreams are so
believed to have an authority surpassing that of the person who
dreams them, they can easily be accepted as sanction for political
action. So cult leaders have employed dreams to give their private
symbolism a public referent, to make it appear that the individual
incidents of a dream have symbolic meaning for all the people, and
should be translated into action.
A similar inter-relationship between private and public symbols in
dreams is excellently illustrated by Sundkler's study of Bantu
prophets in South Africa (1948, 273-4). In the African Zionist
churches dreaming in accordance with pattern is a common and
accepted way of demonstrating group loyalty; the members of a
church 'dream what their church expects them to dream'. Testimony
about dreams is produced in Sunday services, and listened to for
hours because people believe that Jehovah has revealed Himself in
this way. The stereotyped dream is the true, prophetic dream and has
great integrative strength. In Sundkler's view the sect thus exercises
control over the individual's 'subconscious mind'. It has been noted
by various authorities how frequently in these dreams having to do
with the position of a person in a church, symbolic figures of
authority clad in shining light garments appear. The concept of
stereotyped or standardized dreams has been adapted, as Sundkler
explains, from Malinowski, who distinguished between ordinary
spontaneous dreams arising in Trobriand life from everyday ex-
perience, and dreams prescribed and defined by custom, and expected
and hoped for. Malinowski explicitly stated (1932, 327) that the
distinction between free and standardized dreams was not formu-
lated or recognized by the Trobrianders. But it does lay stress on the
obligatory nature of the social component entering into the make-up
of many dreams, depending upon the cultural patterns, and it is
significant that such standardized dreams are normally followed by
socially recognized action. So a Trobriand gardening or fishing
magician, in dreams inspired by ancestral spirits, learns of impending
drought or rain, or of a coming shoal of fish, and will give advice
and orders accordingly. In this way susceptible individuals are used as

channels of communication and crystallizers of opinion for the social
body, be it an African church or a Trobriand village; their dream
symbols are both dictated and received by the social body, which can
draw strength from the process.
But while the stereotype dream may be the true and prophetic
dream, the social body may not always be able to mould the individual
into its own image. The symbolism of some people's dreams may not
fit the accepted pattern. One young woman belonging to a Bantu
mission church had a hysterical illness which doctors and herbalists
tried in vain to cure. In a dream she saw a group of people in white
garments, with white veils on their heads, each carrying a long stick
in his hand. She had not heard of Zionists before, but after this dream
she joined one of their groups, who indeed did wear white clothing
for ritual performances. But after a while she discovered that the
sticks carried by this group of Zionists did not correspond to the
sticks she had seen in her dream. So she left them and was fortunate
enough to find another Zionist church that used the right kind of
stick. Here the private sign served as guide in the dissatisfied search
for health, a kind of symbolic fulcrum for change of group allegiance.
Her private symbolism took public shape in a recruitment decision.
Here the social effect was minimal. But where it is not follower but
leader who has the aberrant dream the results may be disintegrating.
The prophet of one Zion church had no children. He told his congre-
gation he had seen in a dream two beds on one of which he himself
was lying and on the other a young girl from the congregation; the
Spirit, he said, obviously wanted him to take this girl as a second
wife. The congregation were in a dilemma. They all believed that
dreams were the channel of revelation from Jehovah, but some
thought the prophet was mistaken in his interpretation. After a
stormy church meeting the prophet was told that his dream would
be accepted as true and prophetic only if his first wife corroborated
it by an identical dream. Demand for this pragmatic test split the
church. The prophet himself carried on with a small group who
accepted polygyny, basing their authority on the Biblical examples of
Abraham and Solomon. So, the private symbolism of the leader
became a public stumbling-block.
This raises an important problem, namely, by what public
222 SYMBOLS
criteria is the acceptability of a symbolic experience judged? What

constitutes a standard interpretation? What kinds of personal
symbolism can be treated as creative orthodoxy, and what must be
regarded as heresy? What symbols are meaningful to the public at
large, and what are irrelevant?
Sometimes a fairly automatic standard is applied - as when if a
dream conforms to Biblical pattern it is inspired by God's angels,
but if it is in contradiction to the Bible it has been inspired by the
angels of Satan. But even here argument can arise about what is
'contradiction'. Very broadly, I would argue that the criteria for
judgement tend to be taken on the whole not from within the
immediate framework of symbolic concepts, but from outside it, from
the power structure and other institutional make-up of the society.
Personal symbolism is accepted as socially relevant and valid when
though novel, it does not appear to threaten vested interests of those
called upon to judge. When it may appear so to threaten those
interests, the interpretation of the symbols or the consequent action
following their interpretation, may be tailored to the prevailing
norms. So, as I noted with the Iroquois examples, the dreams sym-
bolizing repressed aggression were translated in practice into action
dealing with surrogates for human victims. Judgement on the public
validity of initially private symbols may of course change as external
circumstances alter. Peter Lawrence cites a case of a New Guinea
cult leader, a young woman who claimed that an angel had warned
her in a dream of the imminence of a Second Flood to destroy the
'wicked' and herald the coming of 'cargo' for the 'elect'. She built
a house on tall posts and advised other people to leave their homes
and join her. Some destroyed their pigs and other property 'to
shame God and the ancestors with their poverty' and did so. When
it began to rain the prophetess announced this as a sign of the Flood,
and ordered people to sit outside in it to be washed as if in baptism.
But as the weeks passed without further events the people grew
tired and disbanded, 'the immigrants going back somewhat sheepishly
to their homes' (Lawrence, 1964, 162).
Anthropological interest in the interpretation of dreams is directed
primarily to their public significance. A dream is initially a private
matter, a communication of one part of the self to another, an
expression of deep-seated, unconscious or only part conscious
feelings. Like myth, a dream commonly has a sequential structure,

even a logic of its own. But apart from difficulties of recall, a dream is
often so fragmented that interpretation of elements in terms of their
internal relations within the dream may not be possible. The anthro-
pologist ordinarily has neither the time nor the training to undertake
a systematic analysis of the dream associations which might reveal
the underlying significance of the dream in terms of the personality
of the dreamer. Nevertheless, as I have shown, the anthropologist
can point to the social components which appear in the dream
account, and the social issues it reflects; he can also indicate the ways
in which the private experiences of the dream can assume symbolic
quality for the society and provide a basis for social action.
SYMBOLISM OF PERSONALITY DISORDERS
Over the variety of cultures studied by anthropologists it is common

to find that dream experiences are attributed to experiences with
spirits. The characteristic mode of expression here is for the dreamer
to believe that he has encountered a spirit while he was asleep.
But other forms of encounter with spirits are also commonly
recognized, of what we would term visionary or hallucinatory
character. In many of these, the symbolic forms of the encounter
portray the recognition of some social norm which has been breached.
In a highly technical study of 'ethno-psychiatry' in rural Ghana,
with the suggestive title of Search for Security^ M. J. Field, an an-
thropologist with medical training, has examined nearly 150 cases of
people with mental illness who were treated by confession at local
shrines. Field demonstrates how the range of private symboliza-
tions generated in situations of personal stress conformed in general
to the publicly held concepts of the society, including the super-
natural determination of mental illness. Interpreting the symbols in
such terms, the shrine priests were able to give relief in many cases.
An example (i960, 232-3) is that of a man who had hallucinations of
being attacked by people sent by a deity to kill him; some tried to
make him eat lethal food, others came after him with clubs and cut-
lasses, and he was in a state of extreme terror. Treated by the shrine
priest and making confession he was able to explain that when
224 SYMBOLS
sweeping the shrine yard he had picked up a small gold trinket and
kept it instead of handing it in; so he had incurred the god's anger.
After a ritual purification he was discharged, free of disturbance.
There is a view which has received some popular support that
madness is created by society, in that society determines the criteria
by which certain types of behaviour of individuals are categorized
as madness, and treated accordingly. In a formal sense this is correct.
But I doubt if many anthropologists would agree with the formula-
tion in a substantial sense. They see a high degree of individual
reaction to social circumstances and social norms, varying with
complex combinations of personality elements. They see behaviour
disorder for the most part proceeding from the initiative of the
person concerned, not enforced upon him by society. To the sufferer
from mental illness the traps to personality set by the social order
no doubt appear unavoidable, just as 'odd' behaviour seems a
natural reaction to them. But to other members of the society the
symbolic presentations of the mentally ill person are only one of a
number of alternative avenues out of a circumscribing social situa-
tion. This is the basis of the treatment - to find out where the social
pressures are and what alternatives exist, so that the patient can
make another behavioural choice. In the case just cited, the patient
had the initial choice of handing in the found object. In other cases,
often of domestic difficulty, the choices may not be so simple, but
some resolution is often open for suggestion. Even with the fantasies
described by schizophrenics, which may have no apparent basis in
objective reality, their private symbolic expressions can often be
interpreted in terms of more general social theses.
What is suggested by some of the literature, and what I would
support from personal field experience, is that so-called primitive
and non-industrial societies are often more tolerant of the private
symbolism of mental illness than are our own societies, more willing
to try and enter a dialogue with the patient in comparable symbolic
terms, and more successful in alleviating a condition of mental
strain.
Another type of behaviour, which may be classed as an alternative
order of personality rather than a personality disorder, is that of
trance, usually identified in psychological terms as dissociated
personality. The complexities of this phenomenon are considerable,
likewise the anthropological literature upon it is great.* But inter-

preted by the societies concerned as spirit possession and spirit
mediumship, it has proved a very convenient mechanism for trans-
lating private symbolism into a guide to public action. As Field has
expressed it: 'dissociation is a mental mechanism, a technique, a
vehicle for conveying both the convictions of the prophet and the
tenets of contemporary thought and ethics' (i960, 86).
There are two highly significant but related points about such
dissociation in many cultures. The first is that the alter ego of the
dissociated personality is regarded as either having a special spiritual
quality-as with Old Testament Hebrew prophets-or as being
the actual expression of a different, new spirit personality descended
upon the human body of the medium. The spirit encounter is con-
ceived as not just meeting with spirits, as it might be in a vision, but
as being physically controlled by a spirit, even to the extent of one's
body being inhabited by an autonomous alien spirit. Hence the
second point - that the communications of the medium are not simply
those of the ordinary man; they have an extra-human authority. So
the significance of spirit mediumship is not just that society gets a
personal view on its problems from a specially-endowed individual;
it is that with this individual as channel, usually vocal channel,
society gets an opinion which is credited with authority precisely
because it is not regarded as the individual's own.
It is in this sense generally that the term 'symbolic' is appropriate
here. What the medium says, unprompted or in reply to questions, is
often of a very pedestrian order. As those with experience of trance
performances will confirm, few mediums utter lofty thoughts or
grand prophecies, and their language is not necessarily highly
figurative. But why it can be regarded as symbolic is because the
individual's behaviour stands for more than itself; to it are assigned
meanings of a complex kind of which the individual himself is by all
evidence unconscious or only partly conscious, f In the original
Greek meaning he is prophetes, a speaker for another, the conception
* Cf. I. M. Lewis, 1971, for a valuable synoptic account. An interesting analysis from a
medical as well as anthropological point of view is given by Margaret Field, i960, 55-86.
t The restriction of the term 'symbolic', as a description of the behaviour of prophets, to
conscious, rational gesture-'mimed parables consciously invented and acted out by the
prophets in order to enhance the effect upon their auditors of the message they delivered'
(G. Rosen, 1968, 44) is a misapplication of the term.
226 SYMBOLS
of the 'other' being of an entity whose full mind and powers are
unknown. And what gives the symbolism much of its importance is
that to some degree it is 'private'; there is an element of unpredict-
ability in it precisely because it stems from an individual, with his
own personal background. If what a medium or prophet said was
automatic, with no element of uncertainty, much of its value would
disappear. But if it was purely personal, it would be meaningless. In
the institutions of spirit mediumship and allied phenomena it is the
combination of public and private symbolisms that gives them force.
S Y M B O L I S M OF THE B O D Y
Use of the human body as basic symbolic material has been dis-
cussed by anthropologists and analysts in allied disciplines, from
Birdwhistell and Gombrich to Mary Douglas and Mircea Eliade.
From this and much other treatment one can perceive at least four
kinds of body symbolism in vogue.
As a symbolic instrument, a person may use his own body as a
means of communication, to indicate by bodily action or reference
some more abstract idea. When a person kneels in prayer he is
symbolizing his humility before what he conceives as a higher
personality, a god; when he says to someone else 'I bow to your
opinion', he symbolizes his deference to authority. These are simple
instances, but there are many more subtle ways of using one's body
to express a social relationship. As Mauss put it: 'The body is the
first and most natural instrument of man' (1950, 372). In Chapter 8 I
show how people put their bodies to symbolic as well as pragmatic
use in situations of greeting and parting.
A more general kind of body symbolism refers not to any concrete
entity but to a set of abstract constituents. A man is the head of a
family; the backbone of his team. A social unit is conceived as
analogous to a human body, in some of its major parts. Bodily parts
in themselves can represent the whole man, or his abstract qualities.
He can be weak-headed or swollen-headed, indicating not physical
condition but character defects; he can have no backbone or plenty
of backbone, indicating similarly degree of resolution. Size of
bodily parts can convey other qualities: big-hearted for generosity;
big-mouthed for volubility; broad-shouldered for responsibility
and endurance. (There is an equivalent symbolism, apparently just

as fictional, about male virility.) Translation of bodily parts into
other inanimate material can also be used to describe human qualities.
If a man is generous he has a heart of gold; if very strong he has
arms of steel; if eloquent he is silver-tongued; if loud-voiced he has a
throat of brass; if discovered to have grave defects after having been
admired he has feet of clay. The symbolism is selective - a stupid
man has a wooden head, but traditionally a brave constant man had
a heart of oak. Functional associations also vary in their descriptive
validity: a heart bleeds in sympathy, and a mouth waters in desire or
envy; salivation does empirically relate to desire for food, but the
heart's blood has no relation in modern physiology to emotional
flow. Some of these usages might be described as simply meta-
phorical; others have the complex allusiveness of symbolic quality.
Another kind of symbolic usage treats associations of people -
'corporate' social units - as a more literal type of body, not neces-
sarily human, but with human embellishments. So we speak of the
body politic, its head and its members; of the Christian church,
especially if we are Catholic, as the Body of Christ, its members
being in mystical union with Him. As the All Saints' Day Collect in
the Book of Common Prayer has it, the elect are knit together 'in one
communion and fellowship, in the mystical body of thy Son Christ
our Lord'.
This particular example of body symbolism is of interest from the
subtle distinctions drawn between the physical body of Christ,
which was animate on earth and ascended into heaven; the eucharistic
body of Christ which is received in Holy Communion; and the
mystical body of Christ which is the inner life of the church.
In the Catholic church the concept of 'mystical body' has been
examined and elaborated by many authorities, including Pope Pius
XII. In this symbolic concept, derived by analogy from St Paul, the
church is likened to a body with constituent members each of which
functions only through relation to the others in integral form. But as
it is explained (for example Brantl, 1967,106-16) what distinguishes
the Catholic notion of 'mystical body' from legal, philosophical or
sociological notions is not that the mystical body is conceived as
having both an invisible and a visible aspect - such can be true of
almost any philanthropic or political association - but that the in-
228 SYMBOLS
visible aspect is constituted essentially by a 'life of grace'. The head

of the body is Jesus Christ as God-Man; the members of the body are
all the baptized except those who have cut themselves off by schism,
apostasy or heresy; the vital principle of the body is the Holy Spirit,
Third Person of the Trinity, who informs the body through grace.
As the life-principle of the mystical body, grace, the gift of God to the
soul, is equivalent to participation in the life of God himself. The
church distinguishes between transient or 'actual' grace, given to
strengthen a person in his acts, and habitual or 'sanctifying' grace, a
supernatural quality inherent in the soul, meaning that the soul
partakes of the divine nature and life. So, by virtue of habitual grace
the Catholic church, it is asserted, is a mystical body in that it is
believed to be united essentially through the supernatural endowment
of a special quality by God himself. Hence the 'mystical body of
Christ' can claim freedom from human control. So the concept of
the human body assumed for Christ has received a double extension:
of his physical substance, in form of bread and wine transmuted into
his body and blood; and of his metaphorical substance, in form of all
members of the church allegedly founded by him. In this case the
symbolism is seen as not contingent on human perception, but as
intrinsic, absolute, sacred in the sense of not permitting contact with
worldly explanations. Indeed, it is argued, not symbol but reality.
In a further kind of body symbolism the object of interest is not
the actor's own body, or a postulated metaphorical body, but what is
believed to be an actual part of the body of another person. Simple
instances are the lock of hair which in Victorian times a person
might cherish as a reminder of a beloved; or a tooth of a parent which
a Tikopia traditionally might wear on a string round his neck as a
token of filial sentiment (Firth, 1957, pi. 7; 1963, 256). Such objects
are symbols because they are not just reminders of the existence of
the person to whom they belonged, but stand for a complex set of
emotional and intellectual dispositions. These tend to be private
symbols, with a very restricted range of meaning, much of it not
accessible to ordinary external observation. But some actual or
putative physical residues of people become symbols of a public
order. In this category may be ranked objects such as a presumed
tooth of the Buddha preserved in Ceylon, a presumed hair of the
Prophet Mohammed preserved in Pakistan, and the undoubted
body of Lenin, preserved in Russia. All of these are cult symbols,

the object of elaborate public rituals of respect, each representing
in its own way the values and history of the religious or political
system within which it has been singled out for attention.
Symbols of this public kind tend to have special qualities associa-
ted with them. One is the quality of sacredness, having a virtue of a
non-empirical kind which demands special precautions of a formal
order in any operations connected with the symbol. In theory this
sacredness has been derived from the person of whose body the
symbol once formed a part; in practice the attribution of sacredness
to the symbol has often been a later cult development. In the West,
the most common example of sacred bodily symbols are the relics
of saints, persons whose lives have been dedicated to God to an
exceptional degree and whose bodily remains have therefore an
exceptional value as reminders and stimulants for the religious
congregations. But another quality apt to be associated with these
public symbols is power of a supernormal kind. Muslim saints can
be wonder-working while alive (see Gellner, 1969) but Christian
saints, thanks to a bureaucratic process of canonization, must wait
usually for miracles until they are dead. But in Catholic Christian
belief the power of body symbols and associated relics of saints is
often regarded as great indeed.
The early mediaeval Christian church had a vast history of deal-
ing with relics of saints, motivated in part by magical belief that by
securing a part of a saint's body his protective powers were secured
for the local altar and congregation.* But the folk-beliefs of Catholic-
ism still preserve many examples, demonstrating a marked difference
in the level of symbol appreciation between ordinary congregations
and the more scholarly faithful. The symbolic value of the blood of
St Januarius is a case in point. Nothing is known for certain of the
life of this saint, traditionally a martyr bishop of the early fourth
century, though a legendary account of his death, referred to by
Bede, was translated into Greek in the tenth or eleventh century.
After some transfers of his remains a portion of his blood said to
have been collected after his martyrdom was given in a reliquary
to the Bishop of Naples, and is preserved as a dark mass that half
* For data on transfer of relics and beliefs in their powers see Hirn, 1912, 39-47; Haskins,
1958, 233-6.
230 SYMBOLS
fills a hermetically sealed small glass container. The special property

of the blood is that it liquefies eighteen times during the year, on
formal occasions calendrically specified, when the reliquary is held
by the priest close to the altar, on which is located what is believed to
be the martyr's head. While the people pray, 'often tumultuously',
the priest turns the reliquary up and down in the full sight of the
congregation until the blood liquefies. Thereupon he announces 'the
miracle has happened' and the Te Deum is chanted by clergy and
people. Various thermal experiments and spectroscopic analysis have
been applied to the contents of the reliquary, but it is held by the
authorities that the phenomenon still eludes natural explanation.
The blood with its periodic liquefaction is a symbol of the miracle-
working powers of the saint.* But the authorities display some re-
serve on the matter. They point out that while the happening has
been recorded for the past 500 years there is no record of it before
A.D. 1389, over 1000 years after the presumed death of St Januarius.
Moreover, similar miraculous claims are made for the blood of other
holy personages, nearly all in reliquaries in the neighbourhood of
Naples, and some of the relics, notably the blood of St John the
Baptist, are stated to be 'manifestly spurious' (Attwater, 1965, 184).
The implication is then that some degree of cult imitation has taken
place, a borrowing of symbolic thinking. The essence of this type of
body symbolism however is the focus on the overcoming of natural
process. Blood dries, and it is against nature that it should again
liquefy without human aid. Analogous phenomena have occurred
with the preservation of the bodies of saints in lifelike conditions
after death, with sweet savour, apparently antithetical to the processes
of natural decay. The human body, or portions of it, are used to
symbolize the power of supernatural forces to overcome natural
law.
SYMBOLISM OF THE SACRED HEART
Mary Douglas is one of the most recent of those who have suggested
that the human body is a symbolic medium which is used to express
particular patterns of social relationships (1970, xiii). She has en-
* For discussion on a comparable miracle, that of the bleeding Host of Bolsena, see
Malinowski, 1936, 48, with some general remarks on relation between physical event and
supernatural force.
couraged the examination of Christian ritual and symbolism from

this point of view, in line with similar interests of Edwin Smith,
E. O. James or Bronislaw Malinowski. I want to pursue this theme
briefly, not in regard to bodily style, boundaries and apertures, as
Douglas has done, but in regard to one organ, the heart, which
gives some significant illustration of relations between private and
public symbolism.
For many centuries in the Western world the heart was regarded
as the seat of the emotions (elsewhere stomach or liver were often
so identified). Long ago, Trumbull pointed out that in more than
900 instances in our common English Bible the Hebrew or Greek
word for 'heart' is applied to man's personality, as if it were in a
sense synonymous with his life, his self, his soul, his nature. A person
can be hard-hearted, tender-hearted, heartless; his heart can be
warm, cold; it can be in the right place, in his mouth, at rest
In every phase of man's character, needs, experiences, says Trum-
bull, 'heart' is employed by us as significant of a man's innermost
realest self.
So it was natural that the church at an early period included
reference to the heart of Jesus (and that of Mary his mother) as
objects of special attention. Christ's Sacred Heart especially was
taken as the fountain that dispensed the Spirit from the Saviour's
wounded side. There was a gradual transition in patristic theology
from the idea of the wound in Jesus's side as a source of grace to the
preaching of the Sacred Heart itself as the express object of a more
personal devotion. Various religious authorities, including Jesuits,
became advocates of this form of devotion, and in the seventeenth
century one in particular, St John Eudes, a priest of the French
Oratory, though concerned also with the veneration of the Immacu-
late Heart of Mary, was prominent in bringing devotion to the
heart of Jesus into public worship in the church. In such worship the
heart of Jesus was specifically the symbol of his love.
But the cult of the Sacred Heart was focused and energized by the
visions of a nun, Marguerite Marie Alacoque (1647-90) in the Visita-
tion convent at Paray-le-Monial. What the Catholic Encylopaedia
describes as 'the private revelations made to her by our Lord' con-
sisted of her visions of reposing on Christ's bosom for a long time,
where He discovered to her 'the marvels of His love, and the in-
232 SYMBOLS
explicable secrets of His Sacred Heart' which could no longer con-

tain within itself theflamesof its ardent charity. Jesus said He needed
the help of Marie to introduce to the faithful the 'precious treasures'
of the physical organ. Christ then demanded her heart, put it in His
own, where it was set on fire, and then returned it glowing to her
bosom. Later Christ appeared again to her, with His five wounds
'radiant like suns' and complained of the ingratitude of men; He
ordered His devotee to communicate on the First Friday of each
month, and to prostrate herself for an hour late at night, begging inter-
cession for sinners. Some of this enthusiasm may have been difficult
for her fellows to bear. A later commentator has described her
ecstasy rather unkindly as a sexual urge disguised as piety. But with-
out accepting such a psychological explanation, even such a relative-
ly favourable authority as Attwater has termed her 'exemplary but
perhaps somewhat humourless'. Certainly her willingness to expiate
the shortcomings of her companions must have been rather trying.
But while not the source of the official cult, her experiences gave a
great impetus to publicizing the devotion to the Sacred Heart
and shaping its practices.
There was opposition within the church. Theological contro-
versy arose - stimulated by the role of Jesuits in the cult - over the
precise object of the officially approved devotion. Was the 'heart'
to be understood as physical, metaphorical or symbolic? But by
official statement it is now generally admitted, following Pope
Pius XII, that the physical heart is included in the object of the cult,
and that without the physical heart the public devotion authorita-
tively approved and prescribed for all Catholics by the Church is
not realized. In the upshot Marguerite Alacoque was beatified by
Pope Pius IX in 1864 and canonized in 1920.
The official devotion to the symbol has emerged in various con-
crete forms. Ritually, a feast is celebrated on the Friday following the
second Sunday after Pentecost. Holy Hour, votive Mass and Com-
munion of Reparation are observed on the first Friday of the month.
Consecration of families, nocturnal adoration (on the model of
St Marguerite) and enthronement of the Sacred Heart may occur in
the home. The Enthronement is a recently developed rite which,
begun in 1907 in the Chapel of Apparitions at Paray-le-Monial, the
seat of the original visions, has become an officially recognized
public crusade. The cult, begun by Father Mateo Crawley-Boevy,

S.S.C.C, and preached by him in many parts of the world, consists
essentially in a solemn installation of the Sacred Heart in a promi-
nent place in the home, and consecration of it, though it can be
carried out in any institution. What it means is a dedication to a way
of life in which love is fostered by the sharing of family interests
with Christ and Mary by frequent renewal of the consecration to the
Sacred Heart in union with the Immaculate Heart of Mary, and by a
fuller liturgical life at home and in church. In 1917 Pope Benedict
XV gave formal approval to the Enthronement in a private letter to
Father Mateo, and this was renewed in 1948 by Pius XII, who also
issued an encyclical on the subject. It has been estimated that since
1915 over 2 million families in the United States of America alone
have enthroned the Sacred Heart. It is the official view that through
Enthronement and Night Adoration in the home the Congregation
of the Sacred Hearts has contributed to family life in the United
States.
On the organizational side, many sets of people have devoted
themselves to promotion of the idea of the Sacred Heart or have
enrolled under its symbolic label. These include: the Society of the
Sacred Heart; the Sacred Heart Brothers; the Sacred Heart Mission;
various Sacred Heart Colleges; the Sacred Heart University (founded
in 1963 at Bridgeport, Connecticut, with completely Catholic
personnel). Before the war at least, copra bags in New Guinea might
be seen bearing the mission trading label 'Sacred Heart of Jesus,
Ltd.' The basilica of the Sacre Coeur in Montmartre, begun in 1875
and approved as a work of public utility stands, as Salomon Reinach
has put it, as 'a monument of Jesuit theology, and of the illimitable
credulity of the human mind' (1942, 419). Books, a 16 mm. film and
a phono-record of the Enthronement have helped the cult. (For
general data on the cult see New Catholic Encyclopaedia; and A. D.
Howell-Smith, 1950, 540-5.)
Iconographically, there have been innumerable representations of
the Sacred Heart, with varying degrees of realism. According to
Mrs James (1900, xxiv) the Flaming Heart is 'the rather vulgar and
commonplace emblem of Divine Love', not to be met with in any of
the very early pictures except that of St Augustine. The heart
crowned with thorns is given to St Francis de Sales; impressed with
234 SYMBOLS
IHS it is given to the 'Jesuit saints' and some others. The popular
form, depicting a heart with a wound, encircled with a crown of
thorns and a small cross above, the whole radiating light, did not
appear till the end of the seventeenth century. Since then, says the
Catholic Encyclopaedia fastidiously, 'statues and paintings of dis-
putable taste, often of vulgar sentiment repulsive to educated sensi-
bilities' proliferated from the beginning of the nineteenth century.
Now for some general points, on this recognition of a bodily organ
as a representative of human and divine love, in terms of public and
of private symbolism.
The first point is that the object of devotion (let us call it worship)
is conceived as something quite material in origin. 'In the devotion
to the Sacred Heart the special object is Jesus's physical heart of
flesh as the true natural symbol of his threefold love ' says the
Catholic Encyclopaedia. Now assuming that there was an historical
Jesus, and so a fleshly heart, it is obvious that unlike many of the
relics of the saints, there is certainly now no concrete object of
adoration; the physical, fleshly heart is only a conceptual object,
which has been re-translated into material form by the icono-
graphers. It would seem a reasonable suggestion that it needed the
ecstatic private experience of the nun to provide that dimension of
concreteness, that illusion of physicality desirable to win acceptance
of the heart of Jesus as vehicle of a public symbolic message. Again,
the symbolism is conceived to have begun in terms of physical
communication. Marie Alacoque saw her heart being received into
that of Jesus, illumined there by the flames of its charity, and after
these celestial pyrotechnics, returned again to her breast - experience
clearly generative of acute physical sensation.*
Secondly, the heart of Jesus is very much of a condensation
symbol, in Victor Turner's use of the term. The material object, of
red cloth or paint on canvas or what not, is a symbol of a symbol -
it stands for an abstract, conceptual heart, which in turn represents a
* An interesting analogy was provided by the principal leader of the Ghost Dance among
the Caddo, a man known as John Wilson, about 1890. When Mooney met him he explained
that an amulet on his breast, consisting of a polished end of a buffalo horn, surrounded by a
circlet of downy red feathers within another circle of badger and owl claws, was the source of
his prophetic and clairvoyant inspiration. The buffalo horn was 'God's heart', the red feathers
contained his own heart, and the circle of claws represented the world. When he prayed for
help, his heart communed with 'God's heart' and he learned what he wished to know (Mooney,
1965, 162).
wide spectrum of meanings, of human and divine love, of family

solidarity. And we may note that the meanings are intensified by
action - as the institutionalization of the Enthronement provides
avenues for emotional expression and development of ideas about
family relations.
Thirdly, the meaning of the symbol in a general and public
sense is not identical with the message associated with it by its
private promoter. It is clear that Marie Alacoque had convictions
about the inadequacy of the worship accorded to Jesus: He com-
plained to her about the ingratitude of men; He ordered her to beg
for intercession for sinners - among whom were numbered ap-
parently some of her own convent. The manifestation of the five
radiant wounds - devotion to which was much practised in mediaeval
times - and of the Sacred Heart, was not simply to request increased
attention but to protest against neglect. It was a message of accusa-
tion as well as of invitation, and Marie was the self-designed bearer
of it. Hence the initial view of some of her fellows, not lacking in
self-interest, that she was suffering from delusions.
Now it is relevant to this personal interpretation and manipula-
tion of the symbol by its promoter that somewhat analogous condi-
tions appear to have developed in connection with the apparition of
Our Lady to children near Fatima between May and October 1917.
As narrated in works on the subject, the children were told by the
introductory angel: 'The Holy Hearts of Jesus and Mary will be
attentive to your prayers and have great designs of mercy on you'.
And when the Virgin Mary appeared to them she announced as part
of her message that Jesus wanted to 'establish the Devotion to my
Immaculate Heart in the world. I promise salvation to those who
practise i t . . . ' One of the children told how they saw a heart pierced
by thorns before the palm of Our Lady's right hand, which they
took to indicate the Immaculate Heart of Mary so offended by the
sins of mankind. A commentator noted the 'amazing significance'
of the words: 'the Mother of God tells a child of ten years, who is
totally illiterate, that God Himself wants her to work for the purpose
of getting the world to know better and love better the Immaculate
Heart of Mary.' The emphasis in most of the visions of the children
was on the conversion of sinners, for which the children were invited
to undergo sacrifice and suffering; and much later in 1929 Our Lady
236 SYMBOLS
appeared to the survivor asking for the consecration of Russia,

which was full of error, to Her Immaculate Heart. In this case the
physical image was not so prominent, but the message of conversion
and devotion was associated with the concept of a physical seat of the
emotions, an immaculate or sacred heart, as in the case of Marie
Alacoque.
But conversion of symbol from private to public use needs
a process of organization. As so often in such cases, the initial
visionary acts as a trigger for the cult-one might almost say a
catalyst since though changed by the experience personally, he or she
often does not participate in the cult developments, but retires from
the scene. The system of organized ritual practices which is erected
on the basis of the vision is controlled by other hands. In the case of
Marie Alacoque, the developmental role was played by her Jesuit
confessor (who in some quarters was credited, apparently wrongly,
with having invented or suggested the mode of devotion). And two
centuries later the formulation of the Enthronement procedure by
Father Mateo reinforced the cult. Students of symbolism, even
anthropologists, sometimes write as if the significance and importance
of symbols are somehow self-evident to those who believe in and
act upon them, or at least as if 'society' somehow dictates their
recognition. I maintain to the contrary that there is a process of
selection of symbols in which personal, private preferences may be
of critical importance. There may be competition between forms of
symbolic representation in which the interests of promoters come
to expression. In one and the same 'society' there may be conflict
of opinion as to the appropriateness of a symbol to the conditions
specified, or indeed as to whether the object selected can be properly
regarded as a symbol at all. The accusations of 'cardiolatry' levelled
from within the church itself in the eighteenth century at the pro-
moters of the Sacred Heart devotion testify to lack of enthusiasm
initially for what the highest authority later declared to be the duty
of every Catholic to acknowledge. Moreover, there may be sym-
bolic development on a considerable time-scale. In this case it took
several centuries for the public and private symbolism to coincide.
A further area of possible divergence between public and private
symbolization lies in the processes of material reproduction of the
symbol. In this example, the conception is of an original fleshly
heart of Jesus, long since disappeared and hence retained only as an

ideal form. Material images have been made of this - of plaster, of
cloth, of wax, of stone, of metal, of paint on canvas, in many variants.
Now aesthetically some of these variants have been unpleasing to
some followers of the cult, especially to those of 'educated sensi-
bilities'. Yet by what criterion are these judged? Aesthetically,
there are canons of taste in representation, debated though they may
be. Also, it has been maintained that the difference between allegory
and symbol is that while allegory is moral, symbol is aesthetic. But
as representations of an organ which is itself representative, con-
ventionally if inaccurately, of human and divine love, what is the
basis of evaluating one picture or object as better or worse than
another, if each provokes the appropriate behaviour of devotion?
De facto, aesthetic and intellectual criteria are introduced in judge-
ment on a religious symbol. Clearly, the Sacred Heart of Jesus, or the
Immaculate Heart of Mary, are public symbols of devotion in ths
sense of general rather than abstract concepts, the content of which
can vary greatly in the people who share it. The public symbol is a
general idea; the private symbol is its working out.
THE V A L I D A T I O N PROBLEM
Material given in this chapter on dream, dissociation, vision has

illustrated the intricacy of the relationship between private and public
symbolism. In the interpretation of dreams by the uncovering of un-
conscious wishes, a technique in which Freud above all showed his
mastery, success is based on an assumption that there is a certain set
of experiences, rooted in family life, which have been common to all
men and women, and which find their reflection in times of stress in
recognizably similar patterns of dreaming. The nature and ubiquity
of those experiences has been disputed, as also the techniques by
which they can be recovered from individual evidence. Levi-
Strauss, who acknowledges his debt to Freud for the pursuit of
phenomena involving the most fundamental structure of the human
mind, emphasizes that the family experiences may have been sym-
bolic, not historic. He even goes so far as to accuse the analytic
method of dealing with patients as the invention of myth to explain
238 SYMBOLS
the facts - the analyst conniving with the patient instead of diagnos-
ing or interpreting him (1949, 610-11; 1969, 490-2). Yet in his own
essay analysing the efficacy of a Cuna shaman's treatment of a sick
woman he argues 'that the mythology of the shaman does not corres-
pond to an objective reality does not matter. The sick woman believes
in the myth and belongs to a society which believes in it' (1958,217
1963, 197). I am not concerned with the question of the efficacy of
myth; I merely wish to make the point again that the private ex-
periences of dream, sick person's fantasies, shaman's trance, each
with symbolic components, all embody recognizably socially-
engendered material, upon which their interpretation mainly de-
pends.
But the problem of interpretation of the symbolism may have a
double significance. The symbols may be regarded as means whereby
the individuals simply express their own experiences or feelings and
attempt to communicate these by comprehensible images to other
people. Any question about the validity of the symbols then relates
to the degree to which they are thought to reflect accurately the
experiences and feelings signified. The meaning of the symbols is
the clue they give to the state of mind of the person who expresses
them. Quite another significance of the private symbols may be
thought to lie in the degree to which they express experiences or
feelings of what may be called the audience. The dream, trance
presentation or other symbolic form is regarded as relevant not
only to the individual's situation but also to that of those to whom it is
communicated. It is thought to correspond to their feelings, be
explanatory of their circumstances, even be predictive of future
events. Other types of semi-private symbolic formation, engendered
in discussion and controversy, such as a new scientific theory or a
new theological statement, may also be regarded from a similar
standpoint.
A central problem here is what secures public acceptance of the
validity of private experience conveyed symbolically? To some ex-
tent, conformity with fresh increments of experience - as with many
scientific theories, up to a point, and with many dream and vision
symbolisms. We may recall the case of the New Guinea prophetess
cited by Peter Lawrence (p. 222) whose supporters trickled home
again after the dream-symbolized flood did not occur. But other
kinds of symbolic statement either cannot be subjected to such

empirical test, or the test is rejected as unsuitable. This is true of
many statements of mystical experience, of many theological state-
ments and indeed even of many scientific theories of the more ab-
stract kind: they are symbolic constructs, often alternatives to
others in their field, deriving their acceptance from their capacity to
include a wide range of phenomena in their explanatory scope, but
not directly testable by external observers. For public acceptance of
symbols privately created, a consistency in the symbolic arrange-
ment, a degree of logical fit, seems necessary. But also, I would hold,
there is a prime element of aesthetic concern, a positive interest in
patterning which is not simply reducible to logical principles. If one
looks at some of the modern theological statements, say of pro-
tagonists in the so-called 'Death of God' movement, the level of
personal symbolic content is high, as also is that of logical argument.
But in the last resort the appeal of concepts such as the kenotic or
self-emptying process of self-annihilation of God, the divine self-
transformation from transcendent to immanent, from Spirit to
flesh, is an aesthetic (or aesthetic-moral) appeal. The radical theolo-
gian's concept of God as dialectical process rather than as individual
being has been joined with Nietzsche's poetic phrase that God is
dead, and Blake's symbolic view of God's self-annihilation. But
contemporary preoccupations also help in acceptance - as the con-
junction of the dialectic with the idea that Christ is (and pre-
sumably ought to be) in the world - clearly reflects the deep com-
mitment of the 'Death of God' protagonists to the problems of
modern society. A final word on the more personal side. Some private
symbol formations seem to receive public acceptance because of the
special gifts of the person who communicates them. From this point
of view one may define charisma as the ability to get private symbols
publicly accepted.
There is a further problem in the translation of a personal sym-
bolism into a public symbolism. With a truly private symbolism the
social problem is one of action only in the sense of whether the pri-
vate symbols inhibit or enhance the operation of the personality
in its external contacts. If so, then questions of communication,
identification of the symbols, interpretation of their significance
arise in relations with the person concerned. But with some other
240 SYMBOLS
types of personal symbolism the problem is more forcefully posed.

It is characteristic of many mystics that they have felt a need - almost
a compulsive need - to communicate their experience and to assert
its validity. Not only again do they claim their symbolic presentations
as true; they claim this truth as having more than a personal applica-
tion. So, commonly the personal symbolic experience of a mystic
has resulted in a call for a commitment by others, for a recognition
of the social validity of his symbolic order. And this in turn has often
meant a need for judgement by the established forms of authority
in the society. (The personal symbolism of a poet is usually easier
to handle; except perhaps in a dictatorship, his market is apt to be
uncluttered by power considerations.) Any attempt to get private
symbolism translated into public symbolism may involve the social
body in strain; it may mean organization, mobilization of resources,
reactions upon existing social and political structures. There may be
always a danger that the innovative activity of a mystic and his
followers may threaten the established order. Hence in established
religious bodies a careful watch is apt to be kept by the authorities over
personal statements of a symbolic kind related to doctrine by their
members. Divergent symbolic attributions have often been a form of
expression for cleavages of a more substantial organizational kind,
concerned with struggles for power. Historically in Christianity, as
in Judaism and Islam, the mystic is apt to operate on a knife-edge -
his imaginative symbolic creations lift the expression of his religious
body to a new height; but they may also claim an authority which the
established religious body is not willing to concede. For a personal
symbolism in the religious field to be both creative and successful it
would seem to have to reflect some unsatisfied demand in the power
structure.
II
Chapter 7
FOOD SYMBOLISM IN
A PRE-INDUSTRIAL
SOCIETY
If every society or culture has its own set of symbols, as anthro-

pologists have demonstrated, one may expect the symbols of a pre-
industrial society to differ very markedly from those of a society of
industrial kind. The case for some symbolic universals has been
made - or at least for some symbolic general ideas - as in the status
significance of up-and-down or right-and-left; the sexual significance
of water or hair, or pointed and cleft objects; the emotional signifi-
cance in the contrast of light and dark colours and the stimulus of
red. Examples of all of this can be found in nearly any pre-industrial
society that has been reasonably well studied. But a pre-industrial
society tends to have a smaller range of man-made objects to get
symbolic about. In the relative absence of machinery its members
tend to be less concerned with symbolic problems of alienation of
man from the products of his labour. With one major product then,
food, should we expect to find it embedded to a high or to a low
degree in the symbolic system of the society?
Claude Levi-Strauss has already shown in great detail and with
consummate skill how in apparently crude empirical categories
connected with food, such as raw and cooked, fresh and decayed,
boiled and roasted, there can be identified an intricate set of abstract
ideas, of a kind that may be called symbolic, and that constitute in
one of his expressions 'a logic of sensible qualities' (1964, 9). This
logic has been exposed, by exercise of great analytical power, over
more than 800 myths, many of which contain only incidental re-
ferences to the subject of food, but which bear in one way or another
on the basic structural principles being examined. In his analysis
Levi-Strauss is concerned only with the internal structure of the
system, conceived in terms of intellectual process, for reasons which
he stoutly defends: 'Authentic structuralism seeks to seize above all
243
244 SYMBOLS
the intrinsic properties of certain types of order'; and 'it is these

(intellectual) operations alone that we can claim to explain, because
they participate in the same intellectual nature as the activity which
attempts to understand them' (1971, 561, 596; cf. 1964, 172 ff.).
In this brief chapter I cannot attempt an adequate analysis of
the internal structure of any substantial part of the system of sym-
bolic relationships to be discerned in a pre-industrial society. In this
respect I can do no more than indicate a few features of what Victor
Turner has termed the 'positional analysis' of some symbolic items
related to food, together with an examination of statements and be-
haviour in regard to them. The society I take for consideration is that
of Tikopia in the Western Pacific, which when I first knew it more
than forty years ago had almost no literacy, was ignorant of the use of
money, had very rare contact with the outside world, and maintained
a highly traditional social system almost untouched by modern
political forces.
A small isolated island in a vast ocean space, with no food im-
ports and few imports of any other kind, traditional Tikopia was
preoccupied, one might almost say obsessed, with problems of its
own food supply. Drought, hurricane, population increase, famine
were recurrent fears, and much thought was given by responsible
leaders to coping with these problems.* Planning for the avail-
ability of some types of food took place months, even years ahead -
in the planting of long-term crops, the fallowing of land, and the
construction of fishing canoes. A major daily concern of all adults
and most children was getting food. No one who has lived only in a
society where food can be bought in shops, can be stored in re-
frigerators or in tins, comes pre-packed, prepared, almost pre-
digested, can conceive fully of a life where all food has got to be
obtained raw, direct from nature, uncleaned; where storage facilities
are very few and cannot be applied to many kinds of foods; where
people must go out to seek nearly every meal afresh. In an industrial
society getting a meal is an interval or a conclusion to the day's
work; in a society such as Tikopia, getting a meal is the day's work.
Some people go inland, digging up taro or yams, plucking bread-
* For data on these problems, and on the place of food in Tikopia life generally, see my
various monographs, especially Firth, 1936, 94-116, 451-3, 546; 1939 passim; 1940, passim;
1959, Chaps 3 and 4; 1970a, 227-30, 248-50. For an analysis of social and political symbolism
of food in a Massim society see Young, 1971.
FOOD SYMBOLISM 245
fruit or coconuts, or culling forest products; others go out on the
reef, sweeping it for fish; others (only men) take canoes out to sea
for long-lining by day or flying-fish netting by night. With the
insignificant exception of a few occasional birds, traditional Tikopia
knew no animal meat. Their cookery also was limited since they had
no pottery or metal vessels that could take fire; mostly they baked
food with hot stones in an earth oven. Apart from a little roasting
of fish in leaves over an open fire, and occasional boiling by sliding
hot stones into liquid in a wooden bowl, their cooking patterns were
determined by the oven technology - which also gave indicators
externally for social and ritual purposes (see p. 65). But despite its
limitations their food technology was shrewdly adaptive and diversi-
fied, they commanded a fair range of food ancillaries, and had some
quite imaginative food recipes (1936, 103-10).
For individual Tikopia, getting a main meal, usually between
about noon and mid-afternoon, was one of the most important
features of the day's business. But it was not just an individual search
for and preparation of the day's food; it was a highly organized,
planned co-operative activity in which not only division of labour
but also transfer of food played a significant part, particularly among
households of close kin, if only to fill gaps in immediate supplies.
From all this it should be obvious that much of the relationship of
people to food in Tikopia is strongly pragmatic, empirical. They
want to eat it, they are anxious about the supply of it, they organize a
great deal of their activity around getting it and making it ready for
eating. They also are very interested in the idea of food, intellectually
and emotionally. They talk a lot about food; they enjoy their own
foods cooked in their own way; they are very hospitable in pressing
food upon visitors; and very pleased when visitors enjoy it too. With
all this the Tikopia have quite an elaborate set of symbolic concepts
in which food figures - either being symbolized by other things or
itself symbolizing activities and relationships.
I want here to make an important point: strictly speaking there are
no symbolic objects - there are only symbolic relationships. To
speak of food in Tikopia as a symbolic object is often a convenient
short-hand term, but it is the conceptualization of the object in a
given relationship that is significant. This bears on the notion of
'master symbols' or 'dominant symbols' which have been regarded
246 SYMBOLS
as key features in ritual (cf. p. 86). Victor Turner has argued that
Ndembu dominant symbols refer to values that are regarded as ends
in themselves, axiomatic values. He also has pointed out that there
may be discrepancy, even contradiction between many of the
meanings given by informants to a dominant symbol when it is
regarded as a unit of the total symbolic system; and that such dis-
crepancy is a quintessential property of the great symbolic domi-
nants in all religions (1967, 20, 43). Now food in general has
certainly an axiomatic value to all Tikopia, and was a prime symbolic
element in the traditional religion. But I do not think there was any
basic 'quintessential' discrepancy between the views of Tikopia
informants as to its basic significance in the symbolic system. I
suspect that there may be variation in this as between different kinds
of symbols, if not among different kinds of ritual or symbolic
systems. But for Tikopia I think this problem does not emerge if one
looks upon the issue of dominance in a somewhat different light, as
referring to relationships, not objects. I doubt if it would be ade-
quate to speak of food as being a dominant symbol in Tikopia, though
as I shall show it enters into every ritual situation; I think it prefer-
able to speak of the symbolization of food as a dominant - or at
least prominent - theme. At the same time the symbolic relation-
ships of food in Tikopia society can function side by side with non-
symbolic, quite pragmatic relationships. Food is not wasted in
Tikopia, as one may imagine from the care taken to conserve it; so
food which serves a symbolic purpose also serves the purpose of a
meal. The symbolic and the non-symbolic relationships are inter-
twined.
LANGUAGE OF FOOD IN TIKOPIA
Tikopia has a rich series of linguistic expressions to describe food

in general, in specific forms, and in social relationships. The basic
Tikopia term is kai> which according to preceding particle can mean
food in general, foods in plural, a particular food, etc.; also the act of
eating, in its various verbal modifications. In contrast to the concept
of 'the land' as a whole, meaning the whole population, the expres-
sion te kai can mean the total food resources of the island. Within
this general category there is a great variety of names for many kinds
FOOD SYMBOLISM 247
of food. (It can be noted that quite a number of the staple vegetable
foods have simple two-syllable names: niu for coconut; ufi, yam;
taro (like potato); mei, breadfruit; futi, banana.) There are also series
of linguistic categories of the order of: raw, cooked; unripe, ripe;
fresh, rotten; tender, tough-with connotations outside the food
field. But I want to focus on certain discriminations of a more strik-
ing character, which lead more directly into the field of symbolic
relationships.
The Tikopia term for food, kai, is often contrasted much as in
Western society, with the word for water, vai. Not only are generally
food and water bracketed together as requirements for life; they are
specifically linked in complementary fashion in a Tikopia meal.
Eating in traditional Tikopia style a person first takes food from the
leaf platter set before him, then is handed a water bottle (traditionally
made from a coconut - nowadays a cup serves instead) and drinks.
So kai and vai, solid food and liquid, are the twin components of a
meal. Vai is ordinarily water-as vai tai, sea water; vai marie,
fresh water. But vai is also liquid more generally and can refer to
liquid food, as to soup-like foods made by sliding hot stones into
wooden bowls of watery mixtures. The reference here is to the
consistency of the food, and the adjective vaivai in usual reduplica-
tive form means watery. But there is an anomaly: when mourners are
confined to the house in which a dead person has been ritualized, they
are fed for three nights by kinsfolk and neighbours who have been
free to collect food from the cultivations, and who enter brusquely
when the sun has set, wail with the mourners and then insist on their
partaking of the food they (the visitors) have brought. Tradi-
tionally they bring with them waterbottles to assuage the thirst of
the mourners, who have been fasting all day, and the drink is offered
as soon as the wailing ends. But not only the drink but also the solid
food is termed vai, and to urge the mourner to eat the expression
used is 6Inu se vai mou, * Drink (a) water for yourself. The Tikopia
are quite aware of the inconsistency of calling solid food liquid, and
explain it first by pointing to the association of the food basket with
the water bottle. But when pushed further they say that the gift is
called vai and not kai because to speak of it as kai would imply that
one was boasting about it (1939, 374). It is for this reason that
periodic gifts of food from commoners to men of rank, especially
248 SYMBOLS
chiefs, are described as vai. Such gifts to men of rank are recipro-
cated (tongoi, the appropriate category term for a food-to-food
transaction) later on in no particularly prescribed style. But mourners'
vai are reciprocated on the last night of the feeding by non-food
goods (koroa, in the category fahapenu) and the obligation of the
transaction is carried on until in due course another funeral offers an
opportunity of repayment. The reason for this double transaction
would seem to be that the mourner's vai is not just a gift of food to
someone who can go out and get his own food; it is the service of a
meal to someone who has been fasting and inhibited from going
abroad to sustain himself. Now what underlies this nomenclature of
calling solids 'liquid' is the symbolic concept that in status situa-
tions where food is an instrument of social policy liquid rates lower
than solid. But it is also more complex. The recipients know they are
getting solid food - and incidental to my argument here, though not
to Tikopia, the solid food should be of high quality, containing
pudding and ideally fish. With people of acknowledged rank, such
as chiefs, the power recognition is evident - commoners are afraid
of seeming to boast lest they suffer for it. But with mourners the
situation is in reverse - they are the disadvantaged ones, and the
bringers of food do not wish to seem to boast lest the mourners be
shamed. The motivational part of my interpretation follows Tikopia
explanations, but the rating of solid over liquid is an inference which
the Tikopia did not overtly cite.
In the symbolic field where food is made to serve social relation-
ships, then, the concept of kai as solid food opposed to vai as liquid
food can be intellectually manipulated and emotionally supported
in a distortion.
An even more startling distortion occurs in the field of edible
and non-edible. Kai is the word for 'edible' in all general contexts.
One asks of fish, fruit and many other things e kai} and one is told
e kail or sise e kai, edible or inedible accordingly. There are several
categories of inedible objects recognized by the Tikopia. There are
objects or substances which are not edible by their physical nature,
because they cannot be masticated or cannot be digested or their
flavour is antipathetic: earth, wood, grubs, some marine fauna, etc.
Then there are others which are inedible because of their social
nature - certain fine fish and other good foods which a person in
FOOD SYMBOLISM 249
mourning 'cannot' eat because of restrictions conventionally imposed
or voluntarily assumed. (This is the condition in which a person is
known aspali to such foods - Firth, 1966b.) Finally, there are other
objects which are inedible because of their spiritual nature - most
birds and certain fish, in particular eels, which traditionally have been
regarded as material representations or emblems of spirits. Such
objects, some of which in other circumstances might be considered
as food, are tapu, taboo. Linked with this last category are other
things ordinarily very definitely in the food category, such as coco-
nuts, which are tapu and therefore theoretically inedible, because
they have been temporarily reserved for consumption by a chief or
other man of rank. ('Theoretically' inedible because such tapu was
not infrequently broken in time of food shortage.)
Now high in the inedible category of the first order, physically
antipathetic, is human excrement. It is regarded as disgusting, and
though excretion on the beaches is common, or in the sea, this is
treated by the Tikopia as a sanitary measure not calling for remark
but not lessening their distaste. Some of the most filthy - and most
frequent - Tikopia curses invoke the father or other kinsman of the
cursed to eat excrement, to eat the inedible. Excrement is taey and
favoured epithets begin kai tae In metaphorical terms the cursed
one is invited to bring together the incompatible ideas of eating and
non-edible, a conjunction supposed to imply frustration, shame and
degradation.
What then is one to make of the many Tikopia ritual invocations
to gods and spirits which begin with an announcement by the priest:
'I eat your excrement' or more fully 'I eat ten times your excrement'.
In an address to the gods above all, on a most formal occasion, when
people sit around in strained attention in a temple heavy with sacred-
ness, why should the priest bring together eating and the non-edible
in such a clashing, putatively disgusting way? The first thing to re-
mark is that whereas a curse by ordering to eat excrement is taken
laughingly or regarded as offensive, this formal statement in a prayer
was simply accepted, without comment, as part of the ritual proce-
dure. I have sat through many temple rites and seen the congrega-
tion pay no more attention to it than does say a Christian congrega-
tion to the opening words of the Lord's Prayer. When asked to
explain, both officiants and congregation members made little of it;
1
250 SYMBOLS
they said it was deprecatory speech, supplicating speech, pleasing to

the spirits in emphasizing the lowly character of their worshippers.
One of my most thoughtful informants, pushed hard, said that the
origin of this kind of speech was not certain; it was ritual speech,
known effectively only to the chiefs. But to the chiefs themselves it
was clearly only traditional ritual phraseology; they could explain
its meaning but not its genesis. But the meaning of such excremental
speech was reinforced by two analogies. On the one hand, it was
part of a general set of deprecatory phrases, in which the priest
described himself and his people in terms of poverty, said they were
orphans (in need of food and protection), thieves (because they
lacked food of their own). As it was explained, the whole object of
such prayers was to turn the mind of the spirits favourably towards
the priest and his people, to grant rain, crops, fish and health, which
were believed to be dependent upon spiritual power. The second
reinforcement of the 'eating excrement' meaning came from chants
of thanks ceremonially shouted to a chief who had just given a great
feast to the people - an official chanter might incorporate phrases
of eating excrement, addressed to the host, as 'Excrete hither for
me to eat, for me to drink....' This also was a status affirmation,
placing the beneficiaries from the feast in a very lowly position
before their chiefly host. It was also a message of thanks, acknow-
ledging the bounty given by the feast.
But there is still a further element in the interpretation of this
phraseology. 'Eating excrement' in this context is not like eating
dirt; it is a metaphorical way of expressing the eating o(foody in the
nutritional sense and not just the sense of something consumed by
the mouth. In the Tikopia ideology the notions of human abasement
and god-bounty were combined in the idea that god's excrement was
human food in the full, rich sense of the term. The shoals offish in
the sea, the breadfruit, taro, coconut and other products of cultiva-
tion were conceived at this level of expression as being the excretions
of the gods. So in some prayers gods were asked to produce excre-
ment for their orphaned land, or to excrete into the sea to make shoals
of fish. So in such solemn religious context the inedible becomes
edible, the disgusting becomes pleasurable, the voided is once more
absorbable - one of the type of reversal-transformations with which
Levi-Strauss has made us familiar.
FOOD SYMBOLISM 251
So far the interpretation of such symbolic language about food

has followed Tikopia explanation fairly closely.* But there are
further implications which Tikopia did not overtly draw, but which
are latent in their usage. I have shown that in different contexts solid
food can be equated linguistically with water and with excrement.
But the context is very relevant. In behavioural terms, it is specified,
transactional, physical food that is equated with water; and unspecified,
non-transactional, conceptual food that is equated with excrement.
Using the colloquial English terms of today, when a mourner is
presented with a funeral meal he is told:' Drink the water I've brought
for you'-not 'Eat the shit I've brought'! But the priest said to his
god 'I eat your shit', not 'I drink your water', when reciting his
prayer. Yet when he made the actual food offerings to the gods he
called it neither 'shit' nor 'water' but just 'food'. Here the status and
transactional elements are of prime relevance. Superior to grossly
inferior (god to priest) is alleged to give 'shit'. Situationally superior
to social superior or relative equal (commoner to chief, funerary
provider to mourner) is alleged to give (and in part does give)
'water'. Grossly inferior to superior (priest to god) gives 'food'. It is
from this point of view that an apparent exception to the rule that
only conceptual food is 'shit' fits in. When the official chanter
gives thanks to a chief host by 'eating shit' in the form of his bounty,
this is physical food. But in Tikopia ideology, a chief is in quite a
different category from a commoner-he has mystical qualities.
Since commoner is to chief as chief is to god, it is linguistically and
symbolically quite appropriate that the food from the chief's feast
should be described as 'shit'. But it would be unthinkable for a
commoner to label the food which he presented to a chief as 'shit'; it
is 'water', f
I think it will be clear from what I have said that expressions of the
* For details see Firth, 1970a, 151, 225-6, 256-8.
f A comparative point may also be made, again in colloquial terms. Tikopia curse by
commanding to eat the inedible. In some societies people curse by commanding to 'fuck the
unfuckable'- to copulate with mother or sister, for instance. But note that the 'unfuckable* is
only ideologically so; physically mother or sister can be copulated with satisfactorily, and
there may be suppressed desire to do so; whereas coprophagy is a rare condition. Tikopia do
not command people in ordinary curses either to eat eels, or to copulate with their sisters;
that is, they avoid swearing by the physically quite feasible though ideologically incompatible.
Only at white heat are incestuous curses hurled, and fighting results (see Firth, 1936,314-21).
(In some Tikopia contexts eating and the sex act are equated linguistically, as when the organ
of one partner is said to 'eat' that of the other in intercourse.)
252 SYMBOLS
order of 'I eat your excrement' are symbolic statements of a highly

elliptical kind, the ultimate referents being food and status. Set out
in ordinary language, the statement is equivalent to: 'I acknowledge
my gross inferiority and believe in your power to provide food if
appealed to in humility/ It will also be clear that the Tikopia
equation of excrement with food has much in common with Freudian
notions. This is reinforced by Tikopia dream interpretation, in
which a dream of coming into contact with ordure is regarded as a
sign that the dreamer will catch fish (Firth, 1967b, 168-9). What I
hope to have shown is that such symbolic expressions are part of a
relatively systematic set, operating on principles which are intelli-
gible, a logic to which the non-rational is subject, once the initial
assumptions are understood.
For this purpose I have taken primarily linguistic expressions
about food in general. But there is also Tikopia linguistic symbolism
about particular foods, also conforming to principles of logical
arrangement. For instance, in a 'totemic' system which allocates
one of each of four major vegetable foodstuffs to each of the four
clans which compose the society, each foodstuff is identified with
the major clan god. Coconut and breadfruit are each the 'head' of a
god, and taro and yam are each the 'body' of a god. The symbolism
is based upon a rough physical analogy: the first two are round and
sizable; the latter have thin brown skin. But a feature which is not
so apparent is that the symbolism is part of a symmetrical arrange-
ment linked with the relative status of the clans. The symbolism,
which reaches out into economic and ritual procedures, is part of the
process of expressing responsibility and control over foodstuffs in a
segmented fashion at both human and spiritual level (see Firth,
1966a, 11; 1967b, 228-32). On the other hand, reference to a particu-
lar type of food can serve as a symbolic expression for a type of
emotion. There are no highly formalized incest prohibitions in
Tikopia, except for parents and children or siblings; but unions
between other close kin tend to be the object of scornful remark if
not actually opposed. The dislike of such a union, putative rather
than actual, was expressed in the phrase of 'eating soi\ a bitter fruit
(actually an aerial yam) which takes a great deal of steeping in water
to render palatable (Firth, 1936, 549). Or again, names for some
kinds of food can have a symbolic value which has significance
FOOD SYMBOLISM 253
according to context. An areca nut much used as masticant has
several names, one of which is very suggestive of the word for
testicles. Accordingly, when relatives by marriage are together,
people between whom only 'proper' speech should pass, the sexually
suggestive name for the nut is avoided.
TRANSACTIONAL SYMBOLISM OF FOOD

In linguistic expressions relating to food, sometimes 'food5 and its
congeners such as 'eat' are symbolized by other terms, and some-
times they themselves are used as symbols. But in the transactional
sphere, food is normally the symbolizer, not the symbolized. One of
the few exceptions was in a special ritual offering known as Manava-
o-te-Kere (Firth and Spillius, 1963, 21-2; Firth, 1970a, 253). The
name which may be translated 'Belly of the Land' was applied to a
small food basket in which offerings to two principal gods of
Tikopia were placed at the outset of each great seasonal cycle of
ritual performances for fertility and welfare. Ordinarily, the basket
held food. But when famine struck the land a stone was put in as
well. The symbolism of this offering the gods 'not bread but a stone'
was not that of insult or lack of sympathy; it was an indirect way of
advising them that the land was grievously short of food and asking
for their help. The stone was both a surrogate for food and a sym-
bolic request for food, a vividly mute appeal. An analogous sym-
bolic statement, with a slightly different rationale, was to put a small
package of stones in with a small basket of food when formal gifts
were exchanged between chiefs in famine periods. Here also the
stones symbolized lack of food, and were intended to inform the
gods and stimulate crop fertility. But they also were a tacit state-
ment about the desirability of co-operation between chiefs - even
if there is no food to give, something must be sent to acknowledge
the relationship. A tacit statement of another kind about poverty
was made in some rites by making the offerings to the gods quite
small, in the hope that this would be taken by the gods as a hint to
increase the food supply from which offerings were drawn.
The Tikopia use food as a symbolic instrument to express ideas
of social co-operation and social status, by many subtle variations
in amount, composition and style of presentation. Every type of
254 SYMBOLS
formal social occasion tends to have its own type of food transaction,
often with a special name. I have described many types of such
transaction in my various publications on Tikopia, and select only a
few here to demonstrate some main points about food symbolism.
But first a general theme running through all these transactions -
that food is the major mechanism whereby the kinship ties which are
basic to the structure of Tikopia society are given concrete expres-
sion. Since food is so important pragmatically to Tikopia one of the
best ways in which kinship can be shown to be meaningful is to
help with food - to assist in providing the raw materials, to help
with labour in preparing it, to give supplies of food ready to eat.
One of the most stringent and well-kept rules of the Tikopia social
system is that which requires men who have married women of a
lineage to attend any formal ceremony of the lineage as cooks. They
are known collectively as 'firewood' or 'oven stones', and their job
is apt to be hard, hot and dirty. Each man comes with his bundle of
firewood and bunch of coconuts, while his wife carries their raw
food contribution, a backload of taro, bananas, etc. This raw food is
the Jiuri, a term applied to any contribution of green food brought
or sent to a formal oven. Its normal reciprocity is a share of the
proceeds, some eaten as a meal on the spot and some borne off in a
basket to be consumed at home. Fiuri and its normal reciprocal,
taumafa - raw contribution and cooked share - symbolize the net-
work of kin relations of Tikopia society.
A special relationship, political and formerly ritual as well, is that
with the chiefs of the clans, who receive acknowledgement when-
ever any man of rank performs an important ceremony, and at all
major life crises such as initiation or marriage. Such a present of
food to a chief, known as fakaariki in reference to the chief's title
(ariki), is much larger than the share of an ordinary commoner, and
usually has a supplement of coconuts, both dry and sprouting, and
perhaps some bundles of raw taro as well. That the gift symbolizes a
social and political relationship and is not just a package of food
supplies is shown by the fact that commonly a man from another clan
than that of the chief is selected to bear the food to him, so tacitly
stating the importance of breadth in co-operation (Firth, 1936,
453). Formerly other special food gifts of large size went from each
lineage to their clan chief on the occasion of re-consecration of their
2
FOOD SYMBOLISM 55
clan temple and their sacred canoe. That these symbolized political
and religious ties of major import was illustrated by the manner in
which the baskets from the canoes were topped by bark-cloth
ritual vestments dedicated to the gods concerned; and the baskets of
both types of gift were ritually offered by the chief to the gods when
he received them.* Formal food exchanges between chiefs also
symbolized their political unity, and a special relationship believed to
have been created by an ancestral marriage about eight generations
before was annually celebrated by an enormous food exchange, with
bearers staggering along a path between two temples in conditions
of high ritual tension (1967a, 131-8, 249).
The general symbolic function of raw food contributions has
been mentioned, as representing social co-operation. But special
types of raw food served also as ritual symbols. Freshly plucked
green coconuts were associated with canoe and fishing rites, in one
of which a coconut was smashed by the chief and thrown out to sea
in a symbolic dismissal of the sea deities to their work (1967a,
85-7). In another rather touching piece of ritual symbolism, raw
food from the cultivations of a dead man was stood on his grave,
as a last farewell to him - 'it is announced to the man who is dead;
it is his severance from the middle of the orchard; that is the parting
of his hand from the woods'. This was not just a piece of senti-
mentalism. It was believed that if no such symbolic gesture was made,
the spirit of the dead man would take umbrage and afflict the cultiva-
tions with pests (1970a, 249).
In the Tikopia scheme of presentation of food after it has been
cooked, great attention is paid to the manner in which it is packaged,
and also to the term by which this package is called. The usual
method is to wrap the food in banana or giant taro leaves-the
equivalent of paper in industrial society - and then put the parcels in a
basket. These baskets, made from strong leaf strips, vary in size,
fineness of workmanship and durability, and normally serve dif-
ferent purposes. Tanga, small fine bags, hold betel materials and
* I have observed an apparent contradiction on a small point in my various accounts of this.
In one general passage I state that these gifts were reciprocated by the chief by a basket of
cooked food (1939, 374); in more specific passages I state they were not (1967a, 109, 224,
249). The latter is technically correct. But as I point out in the same context, there tends to be
indirect reciprocation by redistribution of food by the chief, so substantially there is some
basis for the first statement.
256 SYMBOLS
other objects, not usually food. Longi are medium-sized very open
baskets to hold household food; raurau are of the same type but of
finer quality. Kete are rectangular baskets to hold raw fish and other
things when people are at work. Popora are rough open-work
baskets meant to hold already packaged food or coarse food only,
and to last only for a few days. All ordinary gifts of food are taken
in popora, and given a name associated with their particular institu-
tional function or just called 'food'. But some gifts, though carried
in popora, are named tanga, longi, or kete - it is as if bread and meat
carried in sacks were to be labelled 'a handbag', 'a briefcase', etc.
In traditional Tikopia social life there were not many of these sub-
stitute labellings, and I cannot suggest in all cases why they should
have been given. But some seem to refer to rather critical phases of
ritual. The general meaning of the special labels was deprecatory, like
calling 'food' 'water'; in each case a large mass of food was spoken of
as a small mass. The idea of calling a large gift a small one is of
course well known in many societies, and is linked with ideas of
acknowledgement of status. But it is significant that this is really an
acknowledgement of the symbolic value of language. No one
actually believes that the gift is small - the large containers are there
for all to see; but the polite forms are observed. Clearly it is in
the selectivity of such appelations that their value lies. If all large
gifts were termed small, the reduction process would have to be
reconsidered. It is the particular social context that gives the cate-
gorization its meaning.
I give a couple of examples, one from marriage and the other
from initiation (Firth, 1936, 558, 452). In the series of reciprocal
presentations between bride's kin and groom's kin on a Tikopia
marriage, a very large basket of cooked food, prepared by the groom's
kin, is carried by the bride in particular to her kin. The package is so
heavy that it may tax her strength to take it. Early the next day its
reciprocal, also a large mass of cooked food, is sent over to the
bridegroom's kin, some of it being carried by the bride too. Nor-
mally a transaction is ended by the return gift. But on this occasion
when the bride returns again to her husband she takes with her what
is termed a tanga, literally a little bag, of food. Actually it is quite a
big basket, and it is intended to provide a meal for the immediate
household of the newly-married pair. I was at first unable to explain
FOOD SYMBOLISM M7
why such a derogatory name should be given to a substantial
gift, and Tikopia were no help on this specific point. On further
consideration I think two structural elements are involved. By
ordinary social rules the transaction is finished before the tanga is
given, so this is hors de la serie - it can be described appropriately
then as 'just a little something' about the size of a betel bag, and so
not start up another round of exchanges. The second element is that
by intending the food for the household of the newly-weds, the
groom's kin are being a bit more restrictive than usual in such pre-
sentations. This is quite in order since masses of food have gone to a
wide range of other kin. But in calling this gift in effect 'just a little
bag', of the kind to contain betel materials or tobacco, they empha-
size its domestic quality, diminish its social significance and so fend
off possible difficulty in exchange. These are my inferences, but they
fit general Tikopia attitudes.
My second example refers to the longi from which boys are fed on
the morning after their circumcision-type operation. Longi are essen-
tially household food containers, semi-personalized, used normally
for carriage and storage of meals within the domestic sphere. Now
the food on this occasion, though called longi, consists of large
packages in a popora. Significantly, it is brought from the adjacent
oven-house to the more secular 'profane' inland side of the dwelling-
house where normally formal food displays do not go. The reason
for this is that the food is designed not just as a meal for the initiated
boys but also as a gift for their mother's sisters and mother's brothers'
wives, who have spent the night sleeping as a kind of guard of
honour at their feet. The women eat too, but then carry off the
remainder of the food to their own dwellings, while the men at the
ceremonial site are catered for separately. The food of the longi is
brought in to the more secular side of the house because that is
associated with women, and since the formal context is female, the
domestic food basket is an appropriate symbolic title for the food
presentation. Again, this my inference, but it accords closely with
other Tikopia explanations.
As in any Western society, differences in the quality of Tikopia
food can be indices to recognition of social status or attribution of
putative status as a matter of etiquette. There is a vast difference
between offering someone a few green bananas or yams roasted over
258 SYMBOLS
a fire, and an elaborately prepared pudding made with sago flour

and coconut cream. In traditional Tikopia society formal rules
prescribed very closely what types of food should be served on
different kinds of occasions and to whom. And a variety of symbolic
statements could be made with fish!
In concluding this empirical analysis I refer to two types of
symbolic use of food in which though quality of the food is impor-
tant, the mode of its handling is critical.
The first type is the practice of bringing a parcel of cooked food
as a household meal to a communal affair such as a formal dance
festival taking place during the daytime. Ordinary Tikopia informal
dancing, primarily a recreation for young people, takes place
at night, and daytime dancing is apt to be a formally organized
event. People go in family and lineage groups, taking their lunch.
But this is not a casual meal. They sit according to social divisions,
these depending somewhat upon how the affair has been conceived
by its initiators; and the food has been carefully prepared, with a
proper balance of delicate components. Moreover, units do not
consume what they have themselves brought; a system of exchange
of contributions ensures that everyone consumes someone else's
food. The lines of exchange are partly traditional, partly organized
on the spot, but essentially the consumption of the food by the
communal gathering after the exchange is a symbolic statement of
social unity. In the traditional Tikopia religious rites this principle of
exchange of food contributions at a dance was developed to a high
degree. Both women and men were involved, in separate series of
exchanges at different stages, those of the women being expressly
the more important. Huge amounts of food were prepared, requir-
ing the labour of many more people than the nominal principals, and
though termed longi in typical Tikopia deprecatory style the baskets
were no domestic food-holders but great popora containers. The
exchange was conducted with formal procedures, the baskets of the
chiefs being assigned a special precedence. There was also a special
role signalized by an especially large food contribution, for novices.
The whole intricate system, focused upon food, was a symbolic
mode of emphasizing some basic principles of the social structure:
clan and lineage groups; the chieftainship; the complementary
roles of men and women; the social significance of first experience in
FOOD SYMBOLISM 259
an institution; the meaning of kinship ties. And yet, in addition to all
the necessary technology and economic organization, an occasional
optional pragmatic note crept in. If a very efficient household got its
food cooked by early afternoon, it might fill its basket and send the
contribution along to its exchange partner without waiting for all
the formality of public allocation. It was described then as having
been 'sent in by the back way'-'that it may arrive for their hunger'!
(1967a, 316-19, 361-4).
In a second type of Tikopia symbolic use of food the emphasis is
not upon exchange of contributions, but upon the meal itself. This
is the kaifakamavae, the farewell meal. Tikopia are prone to formal-
ize and sentimentalize partings - as for instance the pagan Tikopia
chiefs performed a religious rite of farewell to their traditional gods
before they officially converted to Christianity (Firth, 1970a, 391). As
part of the ceremonies for marriage, a set of young bachelors and
unmarried girls who have been friends of the married pair gather
near the young couple at the stage when they are about to have their
first formal meal together, and then all eat from the same basket of
food. This is the 'food of parting from the unmarried state'. It is a
symbolic farewell, reminiscent of their carefree days and nights of
work and dance and song and sexual adventure together. Eating in
common, from the same food container, is the last intimacy they will
share before the married pair take up the new responsibilities and
new stereotypes for action which will put them in a different cate-
gory of social beings. This type of meal is a symbolic farewell, since
all participants still continue to be members of the same society and
may meet daily. But nowadays the Tikopia have adapted the con-
cept to meet the frequent case of people going overseas to work or
to school. (I was given such a farewell meal when I last left Tikopia,
an affecting occasion at which speeches were made in the best
banquet tradition.) Here the act of eating together symbolizes the
last association before a physical parting as well as a social parting,
and tends to carry a considerable affective load.
GENERAL CONSIDERATIONS
For an illustration of an anthropological analysis of symbolism I

have taken the theme of food. In contrast to the more common
260 SYMBOLS
anthropological studies of symbolism in myth or ritual I have de-

liberately chosen as material and pragmatic a field as one can get.
Granted that meaning is ultimately 'in the head', in this case as
Tikopia would probably agree, head and stomach are closely allied.
Reality may lie in the world of ideas but illusion demands that the
body be fed. Yet food and eating have a distinct symbolic quality
for the Tikopia. Concrete object and act represent abstract complex
ideas of events and processes relating to basic social and personal
elements. The pragmatic and the symbolic are closely intertwined -
as Leach pointed out about totemism, things which are good to
think with may also be good for eating.
Tikopia food symbolism has a recognizably specific mode of
expression, conformable to the particular vegetables and fish in the
diet* and to the structural demands of the society. Some principles
of general application can be seen: that for symbolic purposes
rough, hard foods are inferior to smooth, soft foods; hot, dirty
labour around the oven is degrading - it is forbidden to chiefs even
though they come in the affinal category of'cooks'. But for the most
part, though the interpretation of the food symbolism can be made
intelligible to an outside inquirer, it needs an apparatus of knowledge
of the social context to seem meaningful in any very precise terms.
Moreover, the symbolic spheres are often quite limited. Anthro-
pologists sometimes have a tendency to write as if once a symbolic
equivalence has been arrived at in one context then it is available for
application in all other social contexts. I do not believe this to be the
case. As I have shown, because food can be equated with excrement
in some Tikopia ritual contexts, it cannot be so equated in other
contexts, where such equation if attempted would be offensive. One
is not entitled to say that the Tikopia word tae is used purely equivo-
cally when applied to ordure on a beach and to the bounty of a god
in a prayer. Both have an excremental connotation. But the sym-
bolic equation, excrement = food, is of restricted range; it serves
certain purposes in prayer, and can be otherwise discarded. I suppose
that one might describe Tikopia food symbolism from this point of
view as a series of limited analogies, but such a scholastic descrip-
tion would not seem to add anything to its meaning.
What I think is clear from my analysis is that there is a structure
* Cf. the symbolic 'decipherment' of a Western meal by Mary Douglas, 1972.
FOOD SYMBOLISM 26l
of meaning in the field of Tikopia food symbolism, but that it

is not a completely systematic structure. Not only do symbolic
categories sometimes contradict pragmatic categories - food is
excrement; big baskets are termed small. Symbolic categories some-
times contradict one another or at least cannot be fitted into the
same scheme at the same time. I have shown how food can be equated
both with excrement and with water, with a rationale for both; but
both symbolic statements cannot be made in the same context. One
of the reasons for this limitation of symbolic spheres is that symbols
have work to do. They are not just static expressions of social
relationships or ideas about the meaning of the world; they are
instruments in an on-going process of social action. They are
linked with ideas of labour and resources, status and power, and are
focused accordingly. Hence over the symbolic landscape the various
features show differences of clarity, intensity, diversity.
Finally, in lighter vein, one may compare food in its symbolic
significance with another highly important symbol of social rela-
tionships which Levi-Strauss has done so much to elucidate-
women. In contrast to words, which have wholly become signs,
woman, states Levi-Strauss (1969,496) has remained both a sign and
a value. The same may be said of food. Women as signs to be com-
municated have the advantage of durability, and of auto-mobility -
they last, and they can be sent to bear their own message; food soon
perishes, and it must be carried. Women are also instruments of
production in the way that most food is not-though Tikopia
occasionally present seedling foods which can be grown, as a social
message. On the other hand women are, objectively considered, of
less variety than food for symbolic transfer, though in more per-
sonal terms they are capable of almost infinite symbolic transforma-
tion. At times also they are capable of disconcerting initiative, such
as running away and spoiling the message, as against the inert
quality of food. But perhaps the quality which is of most importance
in the symbolic values of food, is its divisibility. Women can be
shared but they cannot be divided, whereas food can be almost
infinitely portioned out without loss of quality. Ultimately too, there
is no threat of a food liberation movement which may destroy its
symbolic values and seek to make man rather than woman the bearer
of the social communication.
Chapter 8
HAIR AS PRIVATE
ASSET AND PUBLIC
SYMBOL
A few years ago an anthropologist addressing a group of people on

a social studies course confessed that before meeting them he nearly
went and had his hair cut - that he did not was due to the length of
the queue (line) at the barber's. He used this as an illustration of how
bodily characters, such as length of hair, are used as symbols, create
expectations about conduct and provoke social reactions (Harre,
1968). Put this together with Hallpike's recent generalization (1969)
that among men, wearing long hair is equivalent to being outside
society while short hair is equivalent to social control - and that in
an attenuated form the same principle can be extended to the hair of
women. Add to it the Elizabethan John Heywood's proverb 'long
hair, short wit', or the French 'tongues cheveux, courte cervelle\ We
can then ask what is there about hair, especially in its length, that
makes it of such social, even symbolic interest?*
A person's hair is a biological accessory, a very personal, private
thing, growing and changing with his bodily condition, and capable
of only very limited voluntary regulation by him. Hair is a horny
product of the epithelial or skin tissue, and is a character peculiar to
mammals. Man is remarkable for the general scantiness of his hair,
except on the scalp, where though inferior in hairiness to gibbon and
gorilla, he has about twice as many hairs per sq. cm (312) as the
orang-utan and nearly three times as many as the chimpanzee.
Among mammals man is unique not for his general lack of hair -
whale, elephant and hippopotamus are more naked, area for area -
but for his complete absence of tactile hairs, such as a cat's whiskers,
which are organs of great delicacy. Despite metaphors of his hair
* A general study, Hair, by Wendy Cooper (London, Aldus Books, 1971), appeared after
this chapter had been prepared.
262
H A I R AS PUBLIC SYMBOL 263
'standing on end' with shock or terror, man's hair is relatively inert.
According to physical anthropologists, hair is apparently a true racial
character in man, hereditary and unaffected by environment. Racial
variables include length of head hair; hairiness of body exclusive of
scalp hair; form, texture and colour of hair. In evolutionary terms,
an important function of hair has been for protection, especially in
serving as an insulator to retain heat, but for human beings this has
long ceased to be of much significance (Hooton, 1946, 41, 192-9,
469-75, 483-8; Howells, 1947, 33, 214; Turney-High, 1949, 22).
Hair is not only perishable, a wasting asset; it can be lost completely
with only social, not physiological disability, or at least minimal
physical discomfort.
It is striking to note how out of this sluggish, physiologically
almost functionless appurtenance of his body, man has imaginatively
created a feature of such socially differentiating and symbolic power.
But in contrast to other bodily appurtenances hair has a number of
qualities which recommend it as an instrument for social action.
Though personal in origin, it is multiple, any single hair of a person
tending to be like any other. It is detachable, renewable, manipulable
in many contexts, so to some degree can be treated as an independent
object. Yet there is some variation in texture and colour, so it offers
scope for social differentiation. And it is associative, tending to call
up important social ideas, especially concerning sex.
Consider first the general association of hair with ideas of multi-
tude and fineness. Before modern technology had developed pro-
cesses of counting and measuring to present refinement, hair re-
presented in Western thought an exemplar of extreme number and
extreme delicacy. The Biblical statement that 'the very hairs of your
head are all numbered' (Matt, x: 30), whether metaphor or not, was
intended to emphasize the interest of the Omniscient in humanity by
his having a knowledge that it was regarded as absurd for any human
being to attempt. (According to the Jewish Encyclopaedia the num-
ber was supposed to be one billion seven hundred thousand, though
actually, authorities report that the human head has about 120,000
hairs). Terms such as hairs-breadth, hair-spring, hair-trigger, hair-
splitting, all indicate extreme fineness and delicacy, whether in
mechanics or in the dialectic of argument. When Samuel Butler
wrote of his Presbyterian Knight:
264 SYMBOLS
He was in Logich a great Critick

Profoundly skill'd in Analytick
He could distinguish and divide
A Hair 'twixt South and South- West side
(Hudibras, I, Canto 1)
he meant to indicate, if ironically, the making of minutest distinc-

tion. Now with computer and electron microscope, use of hair as a
measure (only 1/150 to 1/1500 inch in thickness) has become out-
dated, even ludicrous in its grossness. But metaphorically, hair has
been able to hold its place, presumably because of its close relation-
ship to human personality.
In some contradistinction to the notion of hair as indicating fine-
ness is the idea of its wild quality. Man has often been differentiated
from other animals by his lack of hair ('the naked ape'), so people
with an unusual amount of hair may be given feral qualities. Tradi-
tional notions of man living in a state of nature have represented him
as shaggy, in a manner shown by early histories of mankind - and by
old English inn-signs (cf. Hallpike, 1969, 261-3 for other data).
Wildness or lack of refinement are specified in the contrast between
the Biblical Esau as a rough hairy man and Jacob as a smooth-skinned
as well as a smooth-tongued man. 'Hairy' as a colloquial expression
for rough, miserable, uncomfortable {Webster s Dictionary gives it
as a slang equivalent for 'rugged', 'trying') is probably related to this
notion of wildness. 'Hair-shirt' as a synonym for penitential suffer-
ing has a direct material referent in roughness of texture. 'Hairy-
heeled' as a synonym for coarse would seem to refer to the difference
between the rough hocks of a cart-horse and the finer ones of hack or
hunter, and so by implication to a difference in social quality between
them. But when an English society girl reported that when living in
Paris she ended up in 'a hairy place on the sixth floor with no lights
on the stairs', this was clearly a metaphor, probably involving a
social as well as a physical judgement.
An extended use of this meaning of 'hairy' links it with danger, a
development apparently associated with flying aeroplanes. In the
American 'parlance of the pilots' ready rooms' the kind of encounter
between them and aircraft of other armed forces engaged in recon-
naissance can be 'pretty hairy' if they get close (for example New
York Post, 19 May 1964; New York Times Maga{ine, 22 November
1970). In a clearly related usage, a new model automobile was
launched some years ago as 'the hairiest looking sporty car in
America, even at the risk of scaring some people off', with an instru-
ment panel 'that may make you feel more like a pilot than a driver'!
H A I R AS MALE A N D AS FEMALE SYMBOL
But in a further related association hairiness in a man has been

treated as a sign of his virility. Differential distribution of hair on the
body as between the sexes, particularly in facial and thoracic hair,
has given a basis for the link between hairiness and roughness to be
focused on the male, and to have some association with sexual
aggressiveness. 'A hairy body and arms stiff with bristles, give
promise of a manly soul' (Juvenal, Satires, ii, 11); and in de Sade's
Justine, Roland, one of his symbols of exaggerated male sexuality,
was described as 'covered in hair like a bear'. In Western countries,
growth of hair on the face, physiologically a sign of maturation, has
often been put into a social framework and linked with courage and
experience. In the late eighteenth century Demeunier noted that
when women have seen only men who are bearded, they feel aver-
sion and repugnance at the first sight of a shaven chin (1776, iii,
186). 'He that hath a beard is more than a youth, and he that hath
no beard is less than a man', said Beatrice (Much Ado About Nothing,
Act II, Sc. 1). Again, Shakespeare differentiated military from civil
usage in As You Like It (Act II, Sc. 7) in the soldier - bearded like
the pard; and the justice - with beard of formal cut.
Yet this is a sphere where cultural differences may be very marked.
Starting from the purely physical basis of racial differences in male
facial hair-less common among Mongoloid peoples than among
Caucasoid or Australoid (cf. the 'Hairy Ainu' and the Japanese) -
some societies such as the Chinese have traditionally attached more
importance than others to beard development. But with similar
beard potential some societies have categorized beards positively or
negatively in terms of cultural demarcation. Jews are said to have
worn beards proudly through the days of traditional Egyptian
bondage while the Egyptians were shaven. While Greeks and
266 SYMBOLS
Romans mostly went shaven they stigmatized beard-wearing 'bar-

barous', people as outside the pale of polite society. Historically in
the West beards have been sometimes an index of social class, or a
badge of a profession; sometimes an object of sumptuary laws - as
when Peter the Great taxed them. Sometimes worn by soldiers, they
were often prohibited to the military, ostensibly to preserve uni-
formity and discipline. And the transitional state may be socially
disapproved - for an ordinarily clean-shaven man to appear ob-
viously unshaven is apt to be regarded as a mark of disorder, even
an affront, or at the mildest, as an indulgence permitted on holidays.*
In short, hairiness on the faces of men has often been in an ambiguous
position between manliness and savagery - perhaps because conven-
tions of manliness may entail some conception of physical brutality.
This is exemplified in the West by a curious facet of literary his-
tory - a series of works on the wearing of beards, semi-informative,
semi-facetious, from the sixteenth century onwards, by Johan
Valerian, Antoine Hotman (with Hervet, Calmet, Oudin also), and
culminating in J. A. Dulaure's Pogonologia of the end of the eigh-
teenth century. Translated from the French with the subtitle 'A
Philosophical and Historical Essay on Beards' this last work had an
elaborate discussion of fashions in bearded chins, false beards, golden
beards, whiskers, beards of priests, etc., citing ethnographic evidence,
and bringing out the social significance of the practice. 'In ripe age,
the beard is a sign of physical powers; in old age the symbol of
veneration.' There was a joking apology for long beards, and the
work appears to have been composed in half-serious, half-joking
mood for the author's circle of friends (Boudet, 1874, 40-1; see also
above, Chap. 3, p. 97). It is in such sense presumably that one
should take his theme - that to suppress the work of nature is an
outrage on nature of which depraved societies alone are capable;
that people who shave are those where woman has rule! But what
Dulaure does bring out are two other themes of sociological signifi-
cance: the great variation in estimation of the beard in the same
society at different periods; and the very strong affective reaction
associated with such estimation. He remarked that the beard was
worn and highly respected at some periods and despised at others -
* That such a view may still obtain in a modern family see Katherine Whitehorn, Observer,
7 May 1972.
that by the Roman church, for instance it was successively con-
sidered as an odious heterodoxy or the symbol of wisdom and
Christian humility. 'Like objects of great worth, the beard never
excited petty quarrels; both its enemies and its partisans were vio-
lent. . . . ' And he asked, now that it was a whole century since his
society men wore beards, had they gained by the change? (1786, iv,
51,64).
In the West, until modern times, the ambiguity of the male beard
was not matched by any parallel oscillation in women's hair growth.
Facial hair was apt to be reckoned as a disadvantage to a woman,
but in general presented no serious problem. While there was great
variation in head-hair style, long hair was the convention for
women. So in a kind of polarization, beard intermittently and long
hair regularly have come to epitomize man and woman. Darwin's
theory that by process of selection the male with the longest beard
and the female with the longest hair would acquire choice of mates by
mutual recognition of superior sexual attraction, has not received
much modern anthropological support - at least by Ernest Hooton
(1946, 197). But traditionally in the West a woman's long hair has
been an ultimate token of femininity, as hackneyed references to
'woman's crowning glory' or to Milton's phrase from Lycidas-
' sporting with the tangles of Neaera's hair' indicate. Yet there are
paradoxes in this. So far has poetic licence gone in the West that in
metaphor, by the strength of feminine attraction, one hair alone is
enough to hold a man captive. Both Dryden and Pope used this
theme:
She knows her man, and when you rant and swear
Can draw you to her with a single hair
(Persius, satire v, 246)
Fair tresses man's imperial race insnare,

and beauty draws us with a single hair
(Rape of the Lock, can. ii, 27)
But I prefer the earlier Howell
One hair of a woman can draw more than a hundred pair of oxen
(Letters, book ii, iv-1621)
So female hair, head hair in particular, is made to be the symbol of
268 SYMBOLS
contrast in character, and reversal of qualities: physical weakness of a

woman and strength of her femininity; frailty of a single hair and yet
its power to move male desire and love.
But this poetic imagery and the symbolism it represents are essen-
tially contingent upon social conditions. While they have had con-
siderable historical continuity, fashion, class relationships, moral
ideas, have tended to affect the mode of expression. In an essay on
Womankind'in the latter part of last century Charlotte M. Yonge was
all in favour of sobriety and refinement. Hair is the woman's glory,
but she ought to reduce it to Veil-ordered obedience' - 'not wasting
time in needless elaboration, but obtaining the fresh sensation of a
head thoroughly brushed, and securely and neatly arranged. Tumble-
down hair, falling dishevelled on the shoulders, sounds grand in
fiction, but it is disgusting in real life The associations of the
loose unkempt locks of Sir Peter Lely's portraits are not those of
pure and dignified maidens or matrons.' Simplicity was to be the
watchword, so 'refinement, as well as truth, will forbid her eking out
her own tresses with other people's, or changing the colour. This is
finery - that very different thing; though it is one of the great
difficulties to draw the line between the two, especially when dealing
with classes below us . . . ' (1889, 112). Moreover, while a woman's
long hair is a symbol of her femininity in general, the mode of wear-
ing it may be an index to the quality ascribed to that femininity. By
nineteenth-century convention a girl wore her hair long over her
shoulders until late adolescence, and then 'put it up' in a roll on the
top of her head; this was a sign of maturity and specifically in a
young woman, a sign of marriageability. After this transition, to
allow the hair once more to float freely over the shoulders took place
only in intimate personal circumstances - as in a bedroom.* The
simple technology of putting up the hair daily was delegated to
maids by upper-class women. Hence the advice of Charlotte Yonge
- t o middle-class girls of small income-'when the melancholy
moment of "turning up the hair" has come, no girl whose life is to be
spent without a maid should be content till she has learned to make
her edifice firm, and as graceful as nature will permit'.
* Note that the stress is upon informality, not on the bedroom - in Singapore it was custo-
mary for Chinese prostitutes to let down their hair as a sign that for the time being they were
unable to receive their clients (Ward and Stirling, 1925,1, 26n.).
All such conventions have become much confused and overlaid
by modern developments. Apart from the multiplicity of rapidly
changing women's hair styles in which length becomes subordinate
to fashion, the traditional imagery in which the femininity of a
woman's hair was expressed has lost much of its point through the
rivalry of the advertising profession. If one compares 'a cascade of
lustrous curls' with 'the loose train of your amber-dropping hair',
Milton (Comus, 859) is only marginally ahead of Madison Avenue.
But in the sphere of sex differentiation, hair symbolism is focused
primarily on hair of the head, and nearly all literary statements (until
recent years at least) have concerned this. Hair on other parts of the
body, though also allowing some degree of differential develop-
ment, has in the West been treated with much more reserve. Hair on
the chest, normally an attribute of men, may be held to connote
strength and sexual vigour, but also tends to have an association of
lack of refinement. Men who devote time and money to careful
dressing of hair on scalp and face may make no attempt to barber
their hair on the chest - though I have been told by a colleague that
'chest toupees' have appeared on the Costa Brava! Axillary hair,
common to both sexes, is commonly regarded as neutral on a man,
but unfashionable on a woman, as also with hair on legs. Both can be
in the category of 'superfluous hair' recognized by purveyors of
depilatories. This 'superfluous' is a complex social category. It can
include hair of legs and armpits, conventionally unaesthetic, to be
concealed or removed. But what about pubic hair? In the West and
fairly generally elsewhere this should be concealed from public view
- except in vulgar nudity shows, of which it is a high point, or in
some modern ballet or other stage presentations in which the
aesthetic focus is on the nude body as a whole and not specifically on
the sex organs. But the pubic hair is not normally removed by either
sex, presumably because of its private erotic interest - leaving aside
the fullsome descriptions of pornographic literature. But this con-
cealment, yet preservation, of pubic hair is a Western custom. Other
societies vary, from taking nakedness and open display of pubic
hair completely for granted, to complete depilation of the genital
area. So, in Western society, hair on the upper part of the body,
especially the face for men and the scalp for women, has been
traditionally an object of admiration and solicitous care, whereas
270 SYMBOLS
hair on other parts has been a matter of private interest, indifference

or embarrassment, according to the particular social context.
Historically in Western society there seems always to have been a
considerable range of variation in the treatment of women's head-
hair, particularly in upper-class 'polite5 society. What is remarkable
in modern conditions is the variety of such treatment, and its exten-
sion through all grades of society. Hair conditioners of many kinds
claim to keep the hair soft, silky and manageable, shining clean, with
beautiful sheen, glowing and alive.... Diversity of style allows
matching of features, temperament, occasion. The names of the
styles themselves invite semantic inquiry - to take a few haphazard-
ly: Pony Tail; French Knot; Veil; Bangs; Helmet; Puff; Bird;
Cornrow; Poodle. A vast flourishing industry of hairdressers, with
training schools, advanced academies, magazines, guilds, festivals
and world championships - and a comically serious belief in its own
importance - helps to support the technological, social, almost mysti-
cal values involved.
The development of new modes and the change of standards of
evaluation is shown in hair colour. Physically, variation in hair
colour depends on amount and quality of pigment, brown-black or
red-gold, to be found in the hair, and on the extent of unpigmented
spaces in it that reflect light. The result is a range of variation in two
distinct series, from black through brown to blond; and from intense
red to golden red; with a mixed red-brown series as well. In addition
there is the greying or white appearance of hair in the ageing, due to
growth of hair lacking in pigment. Traditionally, social values have
been given to different pigmentation variants. Broadly, the extremes
have received the clearest value-ratings - as when red hair is looked
upon as fearsome in an unsophisticated black-haired population. In
Western society it may no longer be true that 'gentlemen prefer
blondes', but very light hair colour has long seemed to have attracted
favourable attention. Use of precious metal terms 'golden' and
'silver' for yellow and white hair show this.* But jet-black hair has
been found seductive, and red hair equated with a fiery temperament.
In poetic treatment some odd shades have been identified. So Milton
* Salomon Reinach (1930, 154-5) cites Procopius, Ovid and Virgil for instances of life-
principle being believed to be contained in golden hair. Cf. George Peele 'his golden locks
hath time to silver turned* {Polyhymnia)', Alfred Tennyson 'sweet girl-graduates in their
golden hair* {Princess, Prologue, 141).
described Adam with 'hyacynthine locks', a hue which some dic-

tionaries define as 'light violet to moderate purple' {Paradise Lost,
iv, 300),* and which clearly is meant as a term of approval. But in the
modern world of hair bleaches, dyes and rinses such evaluations of
hair colour have largely lost their meaning. Dyeing the hair to con-
ceal effects of ageing, or to offer specific sexual attraction, has long
been known but the advances of modern chemistry have allowed a
freedom of treatment not possible before. As a consequence, the line
between nature and culture has been blurred for the observer's
judgement, the aesthetic range has been widened, and discrimina-
tion on moral grounds of the kind suggested by Charlotte Yonge is
rendered much more difficult.
As a broad impression then it could be said that in addition to
physiological differences in the hair of men and of women, social
differences tend to be created and maintained, with symbolic value.
But a question arises as to how far in modern conditions these social
differences and their symbolism tend to be lost in the almost bewilder-
ing variety of forms of hair treatment available and utilized.
NORMS AND STATEMENTS WITH HAIR
In all cultures, it appears, hair has social as well as personal signifi-

cance. There is material from all over the world to demonstrate how
styles of wearing the hair of the head are used as indicators of social
difference, varying according to age, sex, marital and other status and
crisis situations. What is particularly important is that these styles
are often not just aesthetic modulations or a matter of individual
preference; they are strongly conventionalized, given moral approval
and used as instruments of social expression and social control. They
are not just signs, they may be symbols too. For example, it has been
common among Amerindian tribes for a woman's mode of wearing
her hair to indicate whether she was a maiden or married, but often
the hair style represented also a complex set of ideas and beliefs about
* The expression has equivalent in Homer's Odyssey -'locks like the hyacynthine flower';
and before Milton was used by Blount (Glossograph, 1656)-Hyacynthine, of Violet or
Purple colour. Pope also has it in his version of the Odyssey -'His hyacynthine locks descend
in wavy curls'; and it occurs as late as 1874 with Lowell, who describes Agassiz as 'shaking
with burly mirth his hyacynthine hair'. But in such modern usage it seems equivalent to
reddish-brown or tawny.
272 SYMBOLS
the particular condition of her female powers. So, recently Terence

Turner has reported (1969a) of the Tchrikin, one of the least known
peoples of the central Brazilian wilderness, that recognition of a
woman's reproductive powers is symbolized by her being allowed
to wear her hair long only upon the birth of her first child.
Yet even in the less developed, remote societies, hair styles can
change rapidly when new stimuli are presented. More than forty
years ago, when I first visited Tikopia, their hair styles were the
exact reverse of what then obtained in Western culture - women
wore their hair cropped short, and men wore theirs long, down their
backs. Women also wore their hair plain whereas men, especially
young men, often bleached their hair with lime to a golden colour,
giving them a most spectacular appearance. There were customary
variations - young boys grew a tuft of hair on each side of the head;
marriageable girls had a fringe at neck and brow, cut off at their
wedding. But the contrast to Western custom was clear. Nowadays
this is no longer so. For as Western young men have grown their
hair long (to be like the Tikopia?) Tikopia young men have cut theirs
short, to be like the West. And Tikopia young women, formerly
with shaven scalps, now sport crops of curly or wavy locks, bleached
or parti-bleached in modish fashion. Only in one respect have
Tikopia men tended to be conservative: they still cut their hair at
mourning for a close relative. (And some women still wear head-
rings made up from the shorn hair of their menfolk, as a token of
sentiment and propriety - as a Western woman used to wear a hat
when she went out publicly.) On the other hand, an exception to the
shearing process was the chiefs, whose hair was taboo and was not
sacrificed at the death of a relative; until quite recent times they did
not cut their hair with the rest of the men.*
In the wearing of the hair one may recognize custom, or standard
modes of behaviour, and fashion, or an oscillation of focus on par-
ticular styles. But in addition to these norms, the manner of wearing
the hair has sometimes been used to make a more personal statement.
This has been so even in Tikopia, where the traditional conventions
were strong. Among the generation of women who were staid
mothers of families when I first visited Tikopia, were a few who as
girls, unmarried, had scorned convention and grown their hair long.
* Firth, 1936, 503-9, and Plates VI, VII, IX; 1967b, PI. 1, 96.
HAIR AS PUBLIC SYMBOL 273
This had been regarded, probably correctly, by the young men as
something of a challenge to their privilege; and it fitted in with the
spirit of sexual display and antagonism which tended to characterize
relations between young people. It was remembered especially be-
cause the girls' long hair had been a subject for the jeering dance
songs through which each sex both taunted and invited the other
(Firth, 1936, 507-9). Another kind of personal statement by hair
used to be made by young American college girls. Upwards of
twenty years ago a teen-ager 'going steady' with a young man could
let this be known by various signs - such as a single ear-ring 'for one
man'. One such sign was the wearing of the hair in a single braid,
in contrast to two braids or none at all if she was free. A more
recent kind of statement, indicating not a personal relationship so
much as a personal commitment, has been the wearing of'Afro' hair
styles by black American women. Appearing in the mid-1960s as a
manifestation of black pride, with its suggestion of African, not
American origins and independence, the 'Afro' became a symbol of
ethnic identity and as such a political statement. So it could be
noticed that a young woman student arriving from the south at a
northern university might soon abandon her straightened, wavy hair
style for the frizzled, heightened style which affirmed her solidarity
with other blacks on the campus. Now that black identity has been
made much more clear and some political advance made, it seems,
the 'dramatic spherical cloud5 of the Afro is less commonly worn.
The reasons also include the pressures to change any fashion, and the
fact that the Afro demanded special treatment in combing and con-
ditioning which could not only be more costly than some other
styles but also could affect the quality of the hair. Yet it has been
pointed out that the Afro has been important in helping to free
black women from the problems of dealing with kinky hair in a
straight-haired society, and stimulate them to adopt a wider variety
of bold hair styles. In this context it is revealing to read what E. A.
Hoebel wrote at an earlier period: 'A definitely New World sym-
bolism has arisen among American Coloreds in the matter of hair
form. The passion for hair-straightening and kink-removing com-
pounds among American Coloreds reflects an identification of non-
kinky hair with social status of whites' (1966,283; cf. Time Magazine,
25 November 1971). How faded and distorted does this opinion now
274 SYMBOLS
seem, however accurate it may then have looked! More generally,

just as some Tikopia young women asserted their individuality and
a challenge to men by growing their hair long, so in the West some
women have done likewise by cutting theirs short. For a woman to
wear her hair short in Europe or America has often been interpreted
as a sign of masculinity, and may well have sometimes been an
accompaniment of her homosexuality. But in many cases it has surely
been a statement of egalitarianism with men. Women who rightly
believe that they should be given equal opportunities with men
occupationally, may tend to emphasize their attitude by adoption of
a hair style associated with greater freedom of action, and by reject-
ing the conventional image of the long-tressed feminine weaker role.
Hair is a very personal thing, in two ways. It is very much
attached to the body, associated with personal growth and maturity.
But it is also a personal asset, detachable in large part from the body,
transferable to other persons, even on a commercial basis (see later).
With the manner of treating it personal statements can be made,
either of fairly simple sign order (as when an adolescent college girl
wears a braid to show she is emotionally engaged and so theoretically
not socially available) or as more complex symbols (in expression of
sex challenge or of ethnic identity). Such personal statements may be
more or less group-promoted, but what is especially interesting is
the keenness with which they are group-judged, and the strength of
the expressions of approval or disapproval which they evoke.
In recent years this has emerged particularly in regard to the
wearing of long hair by men.
M O D E R N S I G N I F I C A N C E OF M E N ' S L O N G HAIR
In the West until recently, norms of hair style were based on the
convention that while curls or ringlets might be permissible to
children of both sexes, in adolescence and maturity the hair of
females was worn long and that of males was short. Style might
indicate social progression, as when a girl symbolized her maturity by
'putting up' her hair into a roll or 'bun' on the nape of the neck or
top of the head, equated with 'growing up'; or in parallel, a boy
began to shave. For men, short scalp hair had been the custom for
more than a century. With men's facial hair more variation was
allowable, and fashion saw changing styles, from clean-shaven to
combinations of beard, whiskers and moustache. But less than
twenty years ago the view could be expressed that shaving of the
beard continued in all ranks of life in England, though the mous-
tache, in vogue for many years, had become less common {Pears
Cyclopaedia, 1958-9). As for scalp hair, Hoebel in the third edition of
his Anthropology (1966, 282) could remark that 'in America, the
culture pattern with its attendant symbolic quality is still so strong
that short-haired women are considered mannish and long-haired
men effeminate5, though he added cautiously - ' a t least by most
individuals over 30'.
But such opinions have been overtaken by events. In recent years,
much more tolerance has been extended to male facial hair. What in
England we thought were the occupational beards of submariners
during the war turned out to be merely the advance guard of an
army. Beards are now so common among men of all ages as to attract
little attention, and indeed have become almost standardized pattern
in some circles.* Many men well over the age of 30 can be seen with
long-haired coiffure, from shaggy strands to duckstail or other neck
fringe. Seasonal factors may enter-as a Prime Minister known
socially as rather a trend-setter was reported to be having his hair
shortened for the summer after having it long during the winter.
Essentially, many men have been using their hair for aesthetic experi-
ment with their personalities.
But as well as personal aesthetic and conformity to fashion, it
seems clear that deliberate statements are often being made by such
practices. Quite generally, in line with modern values of self-
expression, the wearing of long hair by men, from hippy to univer-
sity student, can express an attitude of laissez-faire - avoidance of
the trouble and expense of hair-cut, letting nature take its course,
and so on. There may be also an attitude of sexual awareness,
though in a contrary direction to that put forward by Darwin in his
theory of the value of hair in sexual selection. When I once asked a
graduate student why he wore his hair long, he replied, laughing,
because his wife considered it more sexy;f she, listening, only
* There is some point to a New Yorker cartoon (23 May 1970) of two clean-shaven business
men standing on a commuter platform filled with an array of wearers of beards and whiskers;
one saying to the other: *I feel like a damn fool\
t For an elaborate examination of the sexual significance of head hair see E. R. Leach, 1958.
276 SYMBOLS
smiled. Complex elements of identification may also operate between

a young man and young woman who are socially and sexually
associated, and both of whom wear their hair long - though each
may adopt a different mode. Again, there may be a definite rejection
of the hair patterns of an older generation, which in their demands
for cleanliness, 'tidiness', etc. represent the authority of the 'estab-
lishment' which must be defied.
In a more general sense, the wearing of long hair by young men
came to take on the character of a quasi-political symbol. Mingled
with anti-establishment views were other values, including those of
some degree of political commitment to the idea of a different type of
society with institutions of another order than those current in the
West. Whatever be the particular form of commitment, long hair
came to become a recognition signal among young people, of some
sharing of common values and differentiation from the rest of
society. But this situation became blurred with the passage of time,
as the fashion of men wearing their hair long spread. Long-haired
people often turned out to be 'fad-followers' in the eyes of those
whose convictions they did not share. As one student commentator
on this subject expressed it to me: 'You used to be able to trust long-
haired people, but not any more. Now some of the people you can
trust are short-haired; there are "freaks" among short-haired people
too.'
Now modern Western society has become more permissive about
many kinds of social identification - a striking instance being the
toleration now extended in Britain to private behaviour of homo-
sexuals, and to public representations of homosexual attitudes in the
theatre. While some of this permissiveness has extended to modes of
the wearing of hair, much resentment is still visible on the part of the
public to the long hair of young men. Symbolically, the former
individual idiosyncratic behaviour has become stereotyped: while in
fact there is great variety of style in which modern young men wear
their hair long, in the eyes of the public it is all 'long hair', and to
many people equally obnoxious.
What is noteworthy is the widespread reaction of criticism, from
ridicule to high indignation and even physical violence, which this
male wearing of long hair has aroused. If its effects were not often so
serious, the popular response of outrage would be comic to an
external observer. To many middle-aged people in Europe and in
America, a canon of absolute value has come to be attached to short
male head-hair, associated with cleanliness, efficiency and masculinity.
Ignorant of the widespread practices of wearing male hair long,
accepted and esteemed, in other cultures; and in seventeenth- and
eighteenth-century society among men of the highest status and
fashion, modern outbursts against male long hair illustrate not only
the changeability of fashion but also the relativity of the moral
judgements so expressed. Reasons for public expressions of outrage
seem various. Long hair of young men in its more extreme forms is
equated with wildness, calling up suggestions of a feral animal
nature, and rationalized by allusions to untidiness and dirt. It is seen
as a confusion of symbolic categories, not only suggesting a potent
feminine symbol, but also blurring actually at times a casual identifi-
cation of man from woman. It is also a breach of the code of uni-
formity, indicating - since hair length is a matter of personal control
- that the wearer does not wish to be subject to the ordinary norms
of social style. But what seems to exacerbate public reaction almost
to a frenzy is a realization that long male hair may be a symbol for a
deliberate protest against society, or those sections of society which
represent authority. And what is unfortunate is that public reactions
do not always distinguish between long hair as a statement of per-
sonality in aesthetic choice - a personal symbol of primarily indi-
vidual reference - and as an act of recoil from the style of the estab-
lishment - a communication of meaning to the wider society.
One type of public reaction has been the use of information about
male long hair for its incongruity, as news items, often with a touch
of ridicule. So, a society wedding is noted in which the spouse with
the flowing, shoulder-length locks was the groom, while the bride
had her short hair done up in ringlets. News articles on hairstyling
either note the decline of barbers - 'an endangered species in New
York State' - or the development of salons for men which provide
services similar to those which women have been having for years -
to have their hair 'stripped, bleached, dyed, streaked, "relaxed"
(straightened), curled, and teased'. Such salons, which cater not only
for homosexuals, report that men are more fussy about their hair-
styling than women, and some are beginning to have their hair
'done' for special occasions, just as women have done. Such items
278 SYMBOLS
draw attention to sex in an indirect way by emphasizing the ap-

proach of men to feminine patterns (for example Chicago Tribune,
7 November 1970; Chicago Daily News, 22 April 1971; New York
Times, 13 January 1972).
Another type of reaction has been of an official kind, on the basis
of long hair as an offence against public morals. Until recently, male
travellers passing through Yugoslavia were scrutinized at the border
and if their hair was thought to be unduly long, they were made to
cut it before being allowed to enter the country. Similar action was
taken at various times by the governments of South Vietnam,
Panama and Singapore. In Malacca a magistrate warned that he
would not hear evidence from long-haired male witnesses, as it was
unbefitting the dignity of the court {Province, B.C., 25 May 1970;
Chicago Daily News, 7 November 1970; New York Times, 2 Decem-
ber 1971; The Times, 1 March 1972). An interesting analogy is
given by the treatment of transvestites who play female roles in
Javanese popular plays, and who formerly let their hair grow long.
In the name of modern progress an attempt has been made to
clean up the transvestites' sexual image, to discourage their homo-
sexuality offstage, and to develop the idea that their role is adopted
simply as a means of serving the national cause through the
political propaganda of the plays. Hence most transvestite players
have responded to pressure and have cut their hair (Peacock, 1968,
206-7).
In this connection, in his study of social and symbolic aspects of
Indonesian proletarian drama, James Peacock has raised an interest-
ing theoretical point. He holds that for these transvestites the haircut
seems to have been a particularly traumatic part of the Revolution,
and adds that perhaps long hair had an especially potent meaning for
transvestites since it is the only gross male body-part that can be
made more womanly by the natural process of letting it grow.
Other womanly qualities were artificial - powder, padding, rouge,
- but hair could not be removed; it is peculiarly between body and
culture, having attributes of both. Thus long hair may have signified
a more sustained commitment to the womanly role than did interest
in the more external feminine trappings. This suggestion ignores the
possible parallel of long finger-nails, also a symbol of femininity
intermediate between body and culture. But it is very plausible, if it
be remembered that it can be valid only in a culture in which long
hair is a recognized female symbol.
Linked with the reaction of moral indignation at male long hair is
the view that it is prejudicial to discipline, for instance in schools and
colleges, or on athletic teams. It was reported recently from Croy-
don, London, that 100 fourth-formers walked out and held a protest
meeting after their headmaster had banned boys from having long
hair. The banning of a boy from a Sunderland school until he got his
hair cut provoked a public controversy in which artists' representa-
tions of Christ with shoulder-length hair were countered by a
quotation from St Paul to the effect that if a man have long hair, it is
a shame unto him (The Times, 10-17 February 1971; 19 January
1972). Various American university football and swimming coaches
established dress codes which specified that the hair of male athletes
must not touch the collar, and threatened offenders with suspension,
against protests from student leaders and campus newspapers (for
example Chicago Sun Times, 25 September 1970; Chicago Daily News,
31 October 1970; 1 November 1970; Chicago Today, 22 November
1970). With a motto of'discipline, dedication, duty' the coach of the
professional football team, the Kansas City Chiefs, imposed a fine of
$500 on any of his men who let their side whiskers grow longer than
his, and forbade moustaches and beards (New York Times, 22
November 1970). But it is significant to note that such reactions
were by no means universal, and that in some cases suspensions were
challenged as being an interference with personal freedom, and taken
to the courts. The father of a high school student who was suspended
for defying the school dress code brought suit in the district court
for his son to be reinstated and damages paid, on the ground that his
son had the right to wear his hair as he wished. But the legal, like the
moral issues, seem by no means clear. A United States Court of
Appeal, faced by cases from Colorado, Utah and New Mexico op-
posing hair-length regulations for public school boys, gave the
opinion that the Constitution and statutes did not impose on federal
courts the duty and responsibility of supervising the length of a
student's hair. The Court, despite claims of freedom made, did not
recognize the existence of any federally protected right as to the
wearing of hair (New York Times, 29 November 1971; Time Maga-
zine, 1 November 1971). And in a recent London case of a schoolboy
28o SYMBOLS
barred from school till he had his hair cut, an Education Department
spokesman said there was a school rule which stated that hair must
be of 'reasonable' length - which obviously leaves room for a great
deal of interpretation {Evening Standard, 12 May 1972).
The complications of the male long-hair issue emerge still more
when questions of employment arise. The managing owner of a
Chicago taxicab company posted a notice warning his drivers that
those who did not soon get their hair cut would not be allowed to
drive. He argued that passengers had complained about their long
hair, and unless he conformed he would lose business and money.
'We gave them [the drivers] a choice/ he said, 'either get haircuts or
don't drive. It's their choice.' The retort of the drivers was to collect
signed statements from passengers who had no objection to long
hair. Another manager, in Honolulu, defended his long-haired
drivers against criticism on the grounds that if he did not hire such
men they would be standing in the compensation lines for unem-
ployment benefit {New York Times, 29 December 1970; Chicago
Tribune, 17 November 1970). But the benefit issue itself was not
clear, since if prospective employers were known not to be interested
in long-haired men, these might be thought to be unwilling to take
work if they refused to cut their hair. So on several occasions in
California unemployment benefit has been refused on this score. The
manager of the state unemployment office for the Monterey penin-
sula was reported in May 1970 to have said that men whose hair,
beards or sideburns were long and scraggly could not collect un-
employment insurance on the peninsula. He said that after a survey
of 900 employers it was clear that the overwhelming majority of
them would not accept long hair on male employees - hence 'we
feel that those who persist in wearing it are voluntarily restricting
their availability'. Such men had only to shave or trim their hair and
they would receive insurance payments. At the beginning of 1972
the California Unemployment Insurance Appeals Board ruled that
jobless men with long hair were not eligible for unemployment
compensation benefits in towns where most of the employers rejected
long-haired applicants {The Province, B.C., 29 May 1970; New York
Times, 9 January 1972). Note that in this type of case, as with the
taxi-drivers, the operative factor was public opinion, real or alleged.
In some employment fields orders to employees to cut their hair
HAIR AS PUBLIC SYMBOL 28l
have been successfully challenged. The policy of the New York
City Housing Authority for some time had been against the wearing
of beards by their patrolmen, and a patrolman who insisted on
wearing a beard had been discharged for insubordination. But pre-
sumably it was later thought that public sentiment did not provide
sufficient sanction for such action, and the Authority rescinded its
'no-beard' rule, on the ground that 'where it is not clear that the rule
has anything to do with the proper performance of duty' it was 'un-
wise' and 'an unwarranted infringement on personal freedom' to
enforce it (New York Times, 23 April 1970; 16 May 1970). It may be
that legal sanctions also were not thought strong enough. For
instance, a volunteer fireman of Long Island was suspended for a
year for not conforming to the rules about hair; he brought a case
against the Board of Fire Commissioners of his district, and the state
Supreme Court upheld the constitutional right of a public employee
to determine the length of his hair, sideburns, moustache and beard.
When the fireman promised to obey the department rules, the judge
said, he did so not in waiver or violation of his constitutional rights
(New York Times, 3 December 1971).
But with more direct organized sanctions, the reaction against
long hair has tended to be firmer. On various occasions the United
States Army and Navy have issued directives about grooming of
hair and beards, and made it plain that they do not condone long
unkempt hair. In December 1969 it was reported that a black airman
was convicted of refusing to obey an order to trim his Afro-style
haircut, and was sentenced by special court-martial to three months'
confinement, demotion, and $60 a month fine for three months. The
offence was not wearing his hair long but refusing to obey a military
order, but the issue was clear (Life, 8 May 1970; cf. Newsweek, 23
November 1970; 29 May 1971; Chicago Daily News, 3 December
1970; New York Times, 28 November 1971).
The most extreme forms of reaction against male long hair in the
name of discipline have occurred in prisons, where haircuts have
been imposed by force. Two kinds of rationalization have been
invoked in such cases-that the shearing has been for 'sanitary
reasons', or to destroy a possible hiding-place for forbidden goods.
Both the United States and Britain have had recent instances of this,
with offenders who have been involved with the law primarily on
K
282 SYMBOLS
moral or political grounds, and in some cases local public support

was shown for the authorities. In February 1970, members of the
'Chicago Seven' had their long hair cut in Cook County jail, and the
sheriff was reported to have shown pictures of them to an amused
Republican gathering as an earnest of energetic action. In May of the
same year a local branch of the Barbers Union awarded an honorary
barber's degree to a Michigan sheriff who had ordered the hair of a
group of students cut while they were being held in the local jail for
bond to be posted. The sheriff was congratulated by the barbers for
'following our own barber book which says long hair is not good'
(in this case the sheriff had a suit for $200,000 filed against him,
arguing that he had violated the constitutional rights of the students)
(New York Times, 25 February 1970; 2 May 1970). In July 1971 the
three editors of the underground London magazine 0{, awaiting
sentence on obscenity charges, had their shoulder-length hair cut in
Wandsworth jail, it was said at the request of prison staff. In subse-
quent public discussion it was pointed out, with some truth, that this
act would be interpreted by many young people as a desire to
humiliate the defendants for being unconventional and disrespectful
to society (The Times, 31 July 1971; 1 August 1971).
To sum up the material in this field-it appears that modern
Western reactions to the wearing of long hair by men are diverse,
but that there is in conservative circles a widespread if sporadic type
of censorious judgement. It tends to be emotional, and on occasion
to take aggressive form, and it offers various rationalizations, princi-
pally those connected with order and discipline and with femininity.
The sanctions involved are not always clear, but a basic issue tends
to be the degree to which an individual has the right, against what
are interpreted as the interests of society, to determine his own
physical appearance.
But there are ambiguities in the situation. In the name of order,
long growth of both facial and scalp hair may be stigmatized. But
there is an opposition involved here: both beards and long head-hair
may be wild; but beards are manly, while long hair is feminine. So
rules which would make men both shave and cut their hair are to
some extent at cross purposes. Moreover, long hair in itself is not
disordered; if bound up on the top of the head it is not so, and in fact
it seems to be loose flowing men's hair to which major objection is
taken. But the interpretation of constraint or freedom in the treat-
ment of men's hair is largely a cultural matter. With traditional
Tikopia, in everyday affairs in house or village a man normally wore
his hair floating freely. But in work, for convenience, he usually tied
it at the nape of his neck or in a roll on his head. This was an indi-
vidual choice. But if he was appearing formally before his chief, or
was taking a critical ritual role, he had to loosen his hair and allow it
to hang free. So there was equation between physical constraint on
the hair and freedom of movement in work, and physical freedom of
the hair and constraint in ritual observance. This conceptualization,
tacit only, as far as I know, seems to have been that lack of constraint
of hair indicated an openness to direction and control, a submission
to authority - of chief or patron god - as opposed to a man's own
exercise of control in binding his hair. To loosen one's hair for a
formal occasion was a symbolic gesture of submission* - almost the
exact reverse of a conservative modern Western view that loose, long
hair is a 'symbol of lawlessness and subversion'.
LONG HAIR AND BEARDS IN RELIGION
When the symbolism of scalp and facial hair is not simply part of a
diffuse set of social norms, but is integrated with an organized set of
religious norms, the issues take a different form.
Many religious communities have incorporated regulations about
hair into their body of rules for the faithful, involving both cutting
or shaving of hair and growth of hair, according to circumstance.
Judaism, which tends to equate male hair with strength and vitality,
has rules for Nazarite-vows, focusing upon a man's growing his hair
for thirty-day periods (or multiples thereof), ritual cutting and
burning of it, and presentation of animal offerings in celebration.
Detailed specification of such vows includes a distinction between
one kind of lifelong Nazarite, who may lighten his hair with a razor
if it becomes too heavy; and another kind 'the like of Samson', who
may not so lighten his hair (Mishnah, Nazir - Danby, 1933, 281). In
orthodox Hinduism a shaving of a child's head leaves only a tuft of
hair, a topknot, 'universally' recognized as the distinctive mark of a
* But note that in other cultures the symbolism may take a contrary form. Before a Toda
man entered the dairy, a sacred place, for ritual performance, he tied up the straggling hair at
the back of his head (Rivers, 1906, 92, 221).
284 SYMBOLS
Hindu. Rationalizations are that it is a hygienic measure, or meant to

keep the head cool, or more esoterically, that it removes impurities
contracted in the womb before the child had any conscious existence.
Some modern sophisticated Bengalis cut off the topknot, while others
wear a small lock which they keep inconspicuous generally but dis-
play in orthodox company. The special status of ascetics is indicated
by their lack of adherence to rule: some have their heads clean-
shaven, others have their hair braided and coiled up on the head,
often with artificial braids added, and others have long dishevelled
locks. As religious devotees, they are above conformity (O'Malley,
1935, 116, 207-8). In these two major religious systems, then, a
man's long hair is not an object of suspicion and disapproval, but is
used as a defining characteristic, either generally or in relation to
some specific commitment. Moreover, the greater the religious
intensity of commitment, the more freedom allowed to the individual
to determine his own hair length.
But conflict can take place when the dictates of religion are not in
keeping with the norms of society. This has occurred in Britain in
recent years with Sikhs, whose custom of men wearing long hair and
beard has clashed with job requirements or with public opinion. Some
of these people have shaved and cut their hair accordingly. But the
majority have not, and a running controversy has taken place in one
city, with Sikh bus conductors protesting against the Corporation
employer's decision that they should abandon their turbans when on
duty, cut their hair and wear regulation caps.* The direction of
popular prejudice (even if mingled with personal animosity) was
indicated by the case of a Sikh in England who brought an action at
law against people who, he claimed, had invited him into a house and
there cut off his beard and moustache. A leading Sikh commented:
'Cutting off a Sikh's beard is worse than killing him. It is a symbol of
his religious identity' (Times, 21 August 1970). But the police,
though sympathetic, seemed uncertain about how to classify the
offence. The point about some of these cases is that while the right of
* An analogous issue was involved in the decision of the New York Stock Exchange to
recognize a Jewish employee's right to wear a religious skull cap (yarmulke) during working
hours - a practice it had refused to countenance until threatened by a complaint of religious
discrimination with the City Commission on Human Rights. The claim was made on grounds
of Jewish orthodoxy and granted by the Exchange on grounds of 'established religious re-
quirement'.
individuals to freedom in the disposal of their personal hair is at
stake, and is not seen publicly as an unambiguous right, some con-
cession is made to religious symbolism. There is a tendency to see
long hair or beard not as just personal idiosyncrasy or protest, but as
conformity to a recognizable set of sanctions which can be treated
seriously even if they are not accepted fully.
In some religious contexts, however, there have been internal
controversies. In Eastern Christianity beards have traditionally been
held appropriate for priests, but this has not been the view of the
Western church, where there has been considerable divergence of
opinion. The long-standing controversies in this field are illustrated
by a work on clerical beards by Johan Piers Valerian, published in
Latin in Paris in 1533, and in London in the same year, in transla-
tion: A Treatise Written by John Valerian a greatte clerke of Italie,
which is intitled in latin Pro Sacerdotum barbis translated into Eng-
lysshe. Valerian, a Roman, addressed his essay to Cardinal Ippo-
lito Medici, whose cultured aesthetic interests and dislike for the
ecclesiastical life might have predisposed him to sympathy for the
argument in defence of priests' beards. Valerian argued that - pre-
sumably as part of the re-ordering process of the Counter-Reforma-
tion - pressure was being put upon the Pope to bring forward a law
about beards, previously common, and clergy were therefore being
advised to shave. He held that this was against Roman tradition,
that the ancient Roman gods had beards, that only children, women
and gelded men went beardless, and that both Hebrews and Christ
and his disciples wore beards. 'The Beard is a garment for manly
chekes given of nature for comfort and health ' Half a century
later Antonius Hotoman produced a treatise entitled Pogonias [in
Greek] sive De Barba, republished in several editions in the early
seventeenth century. In argument similar to Valerian he held that
without a beard, men get confused with women; a beard is a sign of
maturity; it was sacrosanct in ancient France . . . etc. A major point
in such controversy, from which the work of Dulaure on beards
mentioned earlier (p. 266) was a development, is the significance of
the beard as a symbolic demarcator. The problem involved implicitly
was what should be the categories of prime concern: the distinction
of men from children and women, tacitly less mature and less re-
sponsible; the distinction of priests from other men, of less ritual
286 SYMBOLS
aura; and possibly the distinction of Western priests from Eastern

priests, who could be thought to be of less fine grain, since they not
only wore beards but in the lower ranks could take wives?
The sociological significance of long hair as group demarcator in
a religious sectarian context has been brought out in J. A. Hostetler's
examination of Amish symbols (1964). From the sixteenth century
the Anabaptists passed a resolution forbidding the trimming of the
beard and hair 'according to worldly fashions', though the mous-
tache came to be opposed because of its military associations. For
Amish men beards are expected to appear by baptism, and though
during courtship young men manage to keep their beards very short,
after marriage they no longer dare to trim them. The hair of Amish
men is worn bobbed, cut below the ear or slightly above the ear
lobes, depending on their local community. Hair-length is an index
to the conservatism of the community, and an Amish man makes
himself subject to the sanctions of his church by wearing his hair too
short. The hair of Amish women also must be long and uncut,
parted in the middle only, and combed down the sides. From
infancy, girls' hair is braided and with adolescence is put up at the
back.
In a religious context, long hair has sometimes been assigned a
specific symbolic character, not as group demarcator but as indica-
tive of moral quality. So in Christianity, as in Judaism, long hair
worn by men may sometimes symbolize strength, in allusion to the
story of Samson.* It is held too that since in ancient times unmarried
women wore their hair loose and long, virgin saints are frequently
portrayed in art with long, flowing hair. J. J. Bachofen, linking hair
with sex themes, argued that the sacrifice of a woman's long hair was
a symbolic offering of her chastity (1925, 326 ff.). And early Rabbis
held that a woman's long hair was such a powerful augmentation of
beauty that a married woman was recommended to hide it; and the
cutting of it was taken as a sign of mourning or of degradation
{Jewish Encyclopedia, Hair). Yet in Renaissance art loose, flowing
hair was also a symbol of penitence, in allusion to the woman who
washed the feet of Jesus with her tears and wiped them with her hair.
This Biblical story, it is alleged, led to the custom of hermits and
* Cf. 'the virility of horses is extinguished when their manes are cut' according to a
Bestiary (T. H. White, i960, 86).
others doing penance, letting their hair grow long (George Ferguson,
1961, 47, 135, 137). Interpretation varied according to context.
L O S S OF H A I R
Considering the positive values attached to hair, one can under-

stand that loss of hair can also be invested with social significance.
Natural hair loss, leading perhaps to baldness, whether temporary or
permanent, is usually deplored - hence the many remedies against
'falling hair', which are a tribute to hope rather than to scientific
insight. The negative value attached to baldness is illustrated by
many classical observations, from the story of Elisha and the jeering
children to Ovid's reference to a head without hair as a field without
grass.
Unlike almost all other parts of the body, hair is detachable with
relative ease while growing, and several strongly contrasted atti-
tudes tend to be associated with hair purposely detached, as con-
trasted with hair shed involuntarily. Hair shed involuntarily tends to
be treated like other human exuviae and regarded as unpleasing.
That which was tolerated, even admired, while it was an integral part
of the human personality now becomes an object of disgust. To find
a scalp hair in one's food or a pubic hair in one's bath may cause
acute revulsion. She may, as the poets say, be able to draw you to her
with a single hair, but not if the hair is in the soup! Such reaction
does not seem to be merely a matter of incongruity, the 'dirt as
matter in the wrong place' argument which Mary Douglas has used
effectively in her study of pollution. Nor do I think it is because of
specific sex associations, as psychoanalysts might argue (see below).
It seems to have the double suggestion of intrusion of another per-
sonality into one's own intimate operations of eating and cleansing;
and also of some association with the other person's eliminations,
that is the rejected part of his personality. In my view it is this status-
laden symbolic confrontation, this giving for incorporation into
one's own personality something which another person has dis-
carded, which lies at the root of the strongly emotional rejection of
hair exuviae. It is not surprising then that the use of such hair takes
on a more elaborately symbolic form in ideas and practices of
sorcery (see later). On the other hand, hair intentionally severed
288 SYMBOLS
from the head has often been prized and used as a symbol for
manipulation, especially for memorial purposes. In the eighteenth
and nineteenth centuries aristocratic and middle-class circles in
Europe and America made great play with lock, tuft, curl, tress or
ringlet of hair-as reminiscence of her child's babyhood for a
mother, token of affection for a lover, or sign of mourning in a close
relative. An exotic example of this kind of interest is the recent sale of
a lock of George Washington's reddish-brown hair, which his wife
Martha had given to a New York society belle, for $550 {Chicago
Tribune, 2 November 1970; Chicago Daily News, 22 November
1970). Analogous customs have existed in many other cultures - as
my example from Tikopia has indicated (p. 272).
In many societies, however, it is the act of severing a person's hair
rather than the fate of the hair which attracts prime attention. In
ordinary modern Western conditions the focus is upon the skill of
the operator and the result, for which he is highly rewarded, but in
some other cultures the relationship between the parties may be the
main object of concern. A striking instance of how cutting a per-
son's hair may imply a special relationship is given in traditional
Dobu society. Care of hair is a reciprocal service between husband
and wife, and is interpreted as associated with sexual intercourse. So
an adulterer will openly cut the hair of a woman with whom he has
committed adultery if he wishes to make the matter public and defy
the woman's husband (Fortune, 1932, 50).
In other conditions it is the shorn state of the person that is the
most significant element. Simulated baldness, as by shaving the pate,
has been used to give relief from heat or to cope with scalp infection
or - formerly - as part of treatment in some fevers. Socially, head
shaving has been used for dramatic effect, as by the 'soul' singer
Isaac Hayes {Ebony, 1970, 86; New York Times, 18 April 1971); or as
a demarcation sign, by British 'skinheads', teenagers opposed to
long-haired 'hippies' and bearded 'Pakistanis' {New York Times,
29 March 1970).
But on the whole, deliberate shaving of the head, or close cutting
of the hair, has taken on a ritual quality, intended to mark a transi-
tion from one social state to another, and in particular to imply a
modification in the status or social condition of the person whose
hair is so treated. At many periods women have cut their hair short
H A I R AS P U B L I C SYMBOL 289
to demonstrate if not an equality with men at least a repudiation of

the woman's traditional sex role.* But cutting the hair close or
shaving the head often denoted mourning. In a soliloquy on hair in
Gryll Grange Thomas Love Peacock, quoting Aristotle, wrote: 'In
mourning, sympathizing with the dead, we deform ourselves by
cutting off our hair,' and he added the reflection that: 'A woman's
head shaved is a step towards a death's head'. A recent exemplifica-
tion of this was the act of four young women associated with the
notorious principal in a murder trial; they appeared outside the
courtroom with their heads shaven and vowing self-immolation if
the man was sentenced to death (Chicago Daily News, 29 March
1971). Again, where the initiative does not lie with the person
primarily concerned, an enforced cutting or shaving of hair may
convey contempt and degradation, an extreme reduction in status.
In some countries which had been occupied by the Germans during
the war, after the liberation there was some shaving of heads of
women accused of collaboration with the invaders, especially in
sexual favours. Recently, in the tragic events in Northern Ireland,
this practice has been revived by Catholic 'patriots' against girls
wishing to marry British soldiers or thought to have given informa-
tion (New York Times, 11-14 November 1971). It is in circum-
stances such as these especially that the statement of Antonius
Hotoman applies - that the cutting of hair is a sign of tristitia -
sadness!
The symbolism of shaving the head becomes more complex when
it is used as one of the marks of participation in a religious ritual, or
entry to a religious order. An example of the former, occurring as
part of a great ancestral festival of the Dogon of West Sudan, was
the shaving of the heads of participants so that symbolically they
resembled new-born infants, and had their dependence upon their
ancestors forcefully demonstrated by a reduction of status. A
well-known example of shaving the head as a ceremony of initiation
required for reception into a religious order is the tonsure of the
Catholic church. The tonsure, described by the New Catholic
Encyclopaedia as a sacred rite, is said to be the outgrowth of an
* A fourth-century Christian mystic condemned marriage and encouraged the celibate life
for both sexes, inducing many women to leave their husbands, cut their hair, adopt male
attire and enter monasteries. This form of asceticism was condemned by the Council of
Gangra, in A.D. 340 (Margaret Smith, 1931, 43).
290 SYMBOLS
Eastern custom of cutting the hair of slaves (essentially a mark of

reduction of status); and to have been adopted first by the monastic
orders and then by the secular clergy for its symbolic value in
manifesting the dedication of the cleric to the service of God. By
general law all Catholic clergy are bound to wear the tonsure, the
most usual being the Gallican mode of shaving only a small circle at
the top of the head.
I have pointed out that in most general terms shaving the head is
a sign of tristitia, as Hotoman put it, of diminution of the self,
whether in terms of status or of relation to the world and human
affairs.* The hair of the head is an intimate element of the per-
sonality, and to remove it by intention is in effect a reduction or at
least a change in the personality. Yet there is a kind of dialectical
relationship here. For reduction of the personality in one direction
may allow of its growth in another. This is illustrated particularly by
the values attached to hair and to shaving in Indian myth and re-
ligion. The classical description of Shiva shows him with tresses long
and matted, partly streaming, partly stacked in a kind of pyramid -
the hair of the model yogi of the gods. Supra-normal life energy,
amounting to the power of magic, resides in such a wilderness of
hair untouched by scissors. Shiva does not shave or shear his hair,
which loosens out as a halo around him when he performs his cosmic
dance. But then there is the contrast. To enter upon the spiritual
path of absolute asceticism one must be shaved. When the Buddha
set out on his path of Enlightenment he severed his topknot. Ascetic
hostility to the hair of the human organism is such that Jains tolerate
no hair at all on the person of an ordained holy man. Part of the
ritual of ordination consists in a thorough weeding out of every hair
growing on the head and body. So in this symbol-system, long,
unshorn hair, and completely shorn, no-hair, are both conditions of
power (see Heinrich Zimmer, 1946, 157 ff.). In such a complex
subtle ideology and symbolism, as I interpret it, long hair symbolizes
the cosmic power of growth; no hair symbolizes the power of sub-
jugation of the self to social rule. (In this context, Hallpike's thesis,
* A seventeenth-century voyager to Batavia observed of the Chinese there that their hair, in
which they were very neat, they valued at the highest rate, since it is the last thing they would
stake at play, gambling away their wives and children first. But when a man had lost his hair
by gambling, he lost with it all his credit and reputation, and was looked upon as a slave, to
work for others (Fryke, 1929. 29).
H A I R AS PUBLIC SYMBOL 29I
referred to later, has a limited validity.) So, broadly, loss of hair

symbolizes destruction of personality - either in the course of nature,
by ageing; or artificially, by act of man - or alteration of personality.
It generally implies a lowering of status, though the lowering of
status may be only temporary, a reduction of the individual to
order in the name of a collectivity or an extra-human being with
which association gives the individual a special new status.
WIGS AND PERSONALITY
Since loss of hair is commonly regarded as lowering to the per-

sonality, it is not surprising that in many cultures efforts have been
made to remedy the deficiency, by wigs. According to the dic-
tionaries, 'wig' is an abbreviation of periwig, which in turn is an
adaptation of the French perruque, from which the German Periicke
is also derived; this illustrates the cultural range of the wearing of
wigs in European polite society from about the sixteenth century.
But with modern technology and mass fashion the wearing of wigs
has become much more widespread and the class association has
tended to disappear. In modern times too male wigs have tended to
conceal baldness while female wigs have primarily been additions to
a normal head of hair to meet the changing dictates of fashion.
Current usage gives wigs, wiglets; toupees for men and 'falls' for
women; and more elaborate descriptions speak of 'quality hair-
pieces' or 'hairpieces of distinction'. Not so long ago a 'switch' was
not only a tuft of long hairs at the end of a cow's tail but also a heavy
strand of hair added in some coiffures to a person's own hair.
The nature of the mass appeal is indicated by newspaper advertise*
ments: 'For every woman who says "I can't do a thing with my
hair" we have a wig that says you can.' Wigs are described as
exclusive, coming in all colours, having a wonderfully natural look,
adjustable to any head size, soft to the hand-and the cheek-
romantically realistic, washable, stretchable, sensuous, lightweight,
comfortable Not only are they worn professionally by fashion
models - they are even made for female children. 'It was bound to
happen: wigs for little girls. Practically every fashion idea these days
filters down, crosses over age and sex lines.' Wigs for children
started apparently as playthings, but became more serious fashion
292 SYMBOLS
items, as for parties (illustrated in Chicago Daily News, 21 October

1970).
A special development has occurred in modern wigs for men.
Apart from concealing loss of hair, wigs for judges, barristers and
actors have long been in vogue, for defining their roles. But a strik-
ing use of auxiliary hair has occurred recently in some United States
military circles, of getting a long-haired wig to cover up the normal
short-haired military style when a soldier is off-duty among his
friends. Reports indicate that social pressures about their short hair
brought young men into wig shops on such pretexts as that they ran
a music shop or played in a band and needed long hair as an occupa-
tional badge - whereas what was really worrying them was the
opinion of their girl-friends. So in March 1970 a wig-shop near a
well-known military establishment was selling thirty male wigs a
week, at a price of $25 for human hair and $20 for artificial hair -
which was not so popular. But a reverse quirk was the wearing of
short-hair wigs by army reservists when they had to go into train-
ing, covering their long civilian hair. Realizing that the prejudices of
army sergeants and commanding officers might order them to the
barber, some ingenious men hoped to pass muster with short wigs
for the training period and keep their normal long-haired image in
civilian life (New York Times, 4 March 1970; Newsweek, 29 Novem-
ber 1971).*
In the religious field, similar problems can arise as with the shav-
ing of hair or the growth of beard, in the wearing of wigs. In the
history of the Catholic church there has been argument about the
propriety of wigs for clerics, illustrated by the His wire des Perruques
by Jean-Baptiste Thiers, a work printed at his own expense in Paris
in 1609. Thiers, a cleric and doctor of theology, argued that many
ecclesiastics of his day wore wigs, thinking that such was not inter-
dicted by the church - but they were in error! Whereas wigs then
were common in France, formerly only kings had a right to long
hair (but cf. Wallace-Hadrill, 1962) - until the mid-twelfth century.
Wigs of ecclesiastics were condemned by the doctrine of St Paul,
who held that the heads of men should be uncovered while praying;
* An alternative to a wig for men has been a hair transplant of scalp grafts from sides to top
of the head {Chicago Daily News, 20 April 1970). Analogous to wigs as enhancement of
personality have been false beards for men and false eyelashes for women.
and the Fathers condemned disguises, but wigs disguised the clerics
who wore them Thiers proposed various measures, including
a papal bull, to stop such abuse. That his arguments had some effect
is seen by the publication of his work in Italian nearly a century
later, with episcopal authority from Benevento.
An interesting question about the concept of personality is raised
by the wearing of wigs. Ordinarily, contact with another person's
discarded hair is apt to be repulsive, if there is no positive affective
tie, as with a lock of hair from a relative or lover. Yet human hair in
a wig is acceptable, even preferred to artificial hair. How can this be
so? It is clear that many external elements from living or dead
things can be associated with the human personality without strain -
leather, wool and meat from animal tissue for shoes, clothing and
food. Even elements from another person's body may be incor-
porated nowadays by blood transfusion, corneal grafts and organ
transplants. But it is a delicate issue. Apart from susceptibilities
about food taboos on pork, or beef, or meat generally, there may be
religious objections to some surgical transfers, as in Muslim debates
about the propriety of eye grafts.
What is interesting about wigs made from human hair is that they
seem to arouse no serious objection. A few people seem relieved to
know they are wearing man-made plastic, but most seem happy to
sport 'real first-grade human hair'. This is so even when ethnic
boundaries are crossed - for much hair made up into Western wigs
comes from Oriental heads. Two conditions here are relevant. The
first is that people incorporate other people's hair into their own
personality on their own responsibility - they are not just con-
fronted with it at random; it is under their own control. The second
condition is that it is de-personalized. Once on the commercial
market, human hair is treated and sold just as if it were a fibre of
animal or vegetable origin. This point can be emphasized in another
way, by contrast of custom. I mentioned that Tikopia women with
close-cropped heads have customarily worn, not a wig, but a ring of
their menfolk's hair as a kind of everyday obligatory ornament. In
the West, however, whether women cut their hair or not, men until
recently have done so. Yet I have never heard of a woman's tresses
being made into a wig for her father or brother, or a man's locks for
his sister or wife; this would probably be regarded as eccentric if not
294 SYMBOLS
disgusting. One may wear or treasure a lock of the beloved's hair,

but not have the locks made up into a wig. Yet to wear some other
man's or woman's hair on one's head is quite proper, so long as it has
come through the market process, and by definition the previous
owner has not been known. So in the West personal association but
not incorporation is appropriate for the residues of someone with
whom one has identified social contact; for incorporation one needs
anonymity. Fusion of physical personalities is frowned upon if the
personalities are related; one must dehumanize or depersonalize the
body tissues in order to absorb them. (A main exception to this is
kidney grafts, presumably from the gravity of the case.) In strong
contrast is the use made of wigs by men of the Mount Hagen area,
New Guinea; they combine their own hair, hair collected from other
men, their own wives and other women - whom they pay for the
service - to make elaborately decorated wigs, with symbolic mean-
ing (Andrew and Marilyn Strathern, 1971, 22, 65, 84-94).
MAGICAL SYMBOLISM OF HAIR
The intimate relation of the hair of the head with its owner's bodily
appearance and movement, and his personality, makes it an obvious
object of emotional association and stimulus, so that it may be said
to have force or power as a symbol for the personality, as in poetry
or sentimental conservation. But ethnographic literature gives a
wide range of examples where the power of hair is regarded as not
simply emotional but magical. The hair is believed to have in itself
some quality of affecting either the person from whom it has been
obtained, or the person with whom it is newly put in contact. It is
not uncommon in traditional sorcery for hair of an intended victim
to be secured as an object for spells, a vehicle to convey the mystical
power. On the other hand, hair is sometimes used as if it had thera-
peutic or prophylactic qualities, such as a cure for snakebite, an aid
to recovery after circumcision, or a reinforcement of a warrior's
virtue.*
In cases of negative magic or sorcery using hair of a projected
victim, what seems to be dangerous to the owner is not the involun-
* For data of classical type see for example Seligman (Veddas), 1911, 197; Rivers (Todas)
1906, 257, 267; Tylor, 1878, 127-30; Crawley, 1902, 107-8, 202. For ideas of sacredness of
organic tissues see Durkheim, 1926, 137-8.
tary detachment of the hair - loss of 'soul-stuff' in the older anthro-
pological terminology - but loss of control over it. Hair detached
but under owner's control cannot be used against him. So until
recently old Maori men who might go to a barber to have their hair
cut often made a contract with him to have all the hair ends gathered
up; the owner would then go and hide them away to avoid the
possibility of having them used for sorcery against him. Conversely,
a controlled hair of one's own could be launched against someone
else. There is a story of a Maori in quite recent times who had begun
to drink a glass of beer in a public house when he noticed a hair in
the liquid. He reacted swiftly, not aesthetically in disgust but ritually
in fear. Supposedly plucked from a sorcerer's head, the hair was a
sign of evil power against him; he vomited, ejected the hair and so,
he thought, escaped death. What such a sorcery interpretation does
is to translate a kind of careless aggression on the physical plane -
leaving one's hair lying around - into intentional aggression on the
conceptual (putatively spiritual) plane.
But as Leach has pointed out in his subtle analysis of 'magical
hair' (1958) such symbolic conceptions are not simply an outgrowth
of individual theories about personality; they are highly patterned
social responses, elaborately integrated into a system of beliefs about
protection and pollution, sacred and profane. As he indicates, too,
there are many types of situation where it is not just individually
identified and associated hair that is the object of magical interest,
but any hair as an object in its own right, infused with attributes of a
mystical kind. I can illustrate this from a Malay example.
At the end of 1939 a woman in the village where my wife and I
were living fell ill, in her delirium trying to distribute the rice from
her household and making sexual advances to men. One of our
neighbours, who had been an object of such advances, said that
earlier she had learned magic and had a tiny familiar spirit (pelesit)
which attacked people. Now it had turned on her. She was given
relief by spirit medium performances. Subsequently I found out that
one of the most respected and feared spirit medium performers (who
had not on this occasion been called in to the woman) had a lock of
female hair attached to his rebab, the fiddle of Arab type which is a
prime instrument of the master of spirits. He said the fiddle was 'a bit
potent' because the lock of hair was the repository of a pelesit, and
296 SYMBOLS
that in fact he had obtained it from the woman in question. I gathered

that she herself had given it to him some years before, after the death
of her first husband, who was thought to have been attacked by the
familiar spirit. She was alleged to have bought the hair in the first
place and used it for black magic. The spirit master said he had been
offered $15 for it but had refused; his intention was to throw it
away at sea some time. In the meantime his idea in attaching it to
his fiddle was apparently to nullify its evil properties while at the
same time making use of its power to increase the potency of his
fiddle in therapeutic action. This was a clear case of 'anonymous'
hair being regarded as having special properties other than those of
simple pollution. I was also told that people would sometimes steal
hair, presumably for magical purpose, since when caught with it they
would not explain why; it was cut from the head of a sleeper, or
from a corpse when the watchers were asleep. But empirically, there
is a kind of fusion here between an interpretation of private fetishism
in terms of a public conception of magical intention, and actual intent
to practise sorcery along lines suggested by the public image of it.
The problem of why hair as such occupies an important place in
the magical scheme, as well as being involved in ritual transfers
among kin, and ritual sacrifice at life crises such as funerals, has been
answered in psychological terms by its unconscious sexual associa-
tions. This issue has been very adequately dealt with by Leach, who
points out that while sexual themes are very common in ethno-
graphic data on use of hair, the symbolism is often quite overt.
Moreover, the significance of hair often seems to lie not in its
narrowly sexual reference but, as earlier anthropologists indicated, in
its usefulness as a manipulable representation of the entire person.
So, when used as a ritual instrument, its symbolism was to be
interpreted as referring to relations between persons, or categories or
groups of persons, who were defined socially, specified, demarcated
or united by such symbolic mechanism.
HAIR SYMBOLISM AND SOCIETY
Most anthropological discussions of hair have been ethnographical,

concerned mainly with stylistic differences and ritual use, and their
symbolization of social status and social progress. A more theoretical
treatment, following Leach's pioneering essay, is that of Hallpike
(1969). Hallpike argues that the ritual uses of hair are of such widely
varying types that no theory of hair in ritual can reduce them all to
sexual symbolism. But he puts forward a generalization regarding the
contrast between long hair and short hair, with particular reference
to the reduction of one to the other. Briefly, he holds that long hair is
associated with being 'outside society' either wholly or in part, and
the cutting of hair symbolizes re-entering society or living under a
particular disciplinary regime in society. 'Cutting the hair equals
social control' (1969, 260-1).
Now this thesis has some plausibility, especially if one confines it
to Western and Oriental society. For instance, the shearing of re-
cruits to the army or prisoners in a jail, or the tonsuring of candidates
for admission to a religious order are symbolic of submission to
social control, just as the long hair of modern 'freaks' is an affirma-
tion of a wish to be free from control. But the phenomenon is not
uniform if a wider range of ethnographic evidence is taken. Dif-
ferent forms of social control may demand different forms of hair
treatment, even in the same society. Hallpike argues that in Western
society the long hair of intellectuals, rebels and women is an indica-
tion of their being in some respects less subject to social control than
the 'average man'. But what about Tikopia, where as I pointed out
earlier (p. 272) men traditionally wore their hair long and women
wore theirs short: can one really suggest that men there were less
subject to social control than women? If it is contended that woman
exercised less power in political and ritual fields - were the takers,
not the makers of the rules - then cannot the same be said of our
Western society, still largely male-dominated in these fields? But
Tikopia men cut their hair off in mourning for a dead relative -
subject to social control - whereas women having much less to lose
made only little reduction; yet both sexes wailed equally, in response
to the same controls. And since a Tikopia man appearing before his
chief or performing a key ritual role had to let his long hair flow
loose instead of tying it back or up on his head, different types of
hair mode corresponded to different types of social control. Much
other ethnographic data could be cited to show that in many
societies the wearing of long hair, by adults of either sex, is fully
consonant with social responsibility and amenability; hair cutting
298 SYMBOLS
does not mean re-entry to social control, but immediate transfer

from one form of social control to another or emphasis on the con-
trols of a particular type of situation. Indeed, when modern Western
women cut their hair it is very often not to 're-enter society' as a
symbol of submission to its controls, but rather to demonstrate their
wish to be free of many of the restrictions thought to attach to the
image of the long-haired mother-wife-drudge, who is only too
acutely controlled by 'society'. Moreover, the long hair of members
of sub-groups or categories, from Amish to 'hippies', is not an
index of being outside 'society' in any very subtle sense; their long
hair is a response to very definite social canons within a particular
sector of society. The jurisdiction within which the norm operates is
a very important element in the interpretation.
As anthropologists have amply demonstrated, the essence of
interpretation of such symbols, hair or other, lies not in attributing
empirical significance to the symbolism of each item in itself, but in
recognizing the symbolism of the conjoined likenesses and contrasts,
in systematic arrangement. Man is an ingenious creature, making
much social capital of small physical resources. So, if women have
long hair, men may have it short; if women wear theirs short, men
may grow theirs long. And if both sexes wear their hair at much the
same length then they differentiate by style of dressing it. Refine-
ments within the system can mark out also stage of social progression
and social status. The cutting of hair can be made a formal process,
and a symbolic process - but not simply of sexual loss, as Berg
would have it, or entry into social control, as Hallpike would have it.
It may signify social loss - of a display feature of the personality, an
abasement of the personality, a sacrifice. When one shears one's
head for a person who has died, or on entry to a religious order,
sentiments both of loss and of submission may be thought to be
involved. But anthropologically, attention is directed to the marks of
personal status that are sacrificed with the locks of hair, or more
simply to the social definition of a change in relationship. Such hair
symbolism means that men and women in specific kinds of society at
specific periods are using their own physical raw material in terms of
the social norms to provide indices to their personality and make
statements about their conception of their role, their social position
and changes in these.
Chapter 9
BODILY SYMBOLS OF
GREETING AND PARTING
Greeting is the recognition of an encounter with another person as

socially acceptable. Parting, in social sense, is the recognition that the
encounter has been acceptable. Both concepts involve a postulate of
a positive social quality in the relationship. Encounter of a physical
kind may take place without such social relationship - as by two
persons rubbing shoulders in a bus. They recognize the physical
presence of each other, but the encounter is not socially acceptable;
they do not speak; the existence of each is not incorporated into the
social universe of the other. It needs some exchange of signs, as by a
word or a nod, to create a social relationship. Forms of greeting and
parting are symbolic devices - or signs if they are just specifically
descriptive — of incorporation or continuance of persons in a social
scheme. A greeting or parting sign is often represented as conveying
information or expressing emotion - an announcement that one has
come or is about to go, a statement of pleasure at someone's arrival
or sadness at his departure. Granting that this may often be so, the
informational or emotional content of the sign may be highly
variable, even minimal. What is of prime relevance is the establish-
ment or perpetuation of a social relationship, the recognition of the
other person as a social entity, a personal element in a common
social situation. This is indicated by reverse behaviour, what used to
be called 'cutting' a person who is already known but found objec-
tionable - the refusal of a greeting to him on passing him in the
street or meeting him in society. This refusal is a tacit denial of him
as a social entity in what would normally be a shared situation. So
too, when two people are 'not on speaking terms' they do not greet
each other and so reduce the area of their common social relation-
ship to as small a compass as possible. As with all social relationships,
299
SYMBOLS
3°°
reciprocity is important; an expectation in greeting is that it will
elicit social recognition in return.
GENERAL NATURE AND FUNCTIONS

Great variety of custom in greeting and parting behaviour has been
observed across the world. But this variety occurs through the range
of a relatively few, simple sets of words and non-verbal actions,
involving usually speech organs, head, hands and body. Limited as
these components are, with few cultural accessories, their expressive
power is considerable and their social implications of great sensi-
tivity. Greeting and parting signs, though highly conventionalized,
are not merely formal empty recognition procedures. Perception of
them can be instrumental in modifying the behaviour of the person
to whom they are directed. There is usually no suggestion that the
procedures in themselves have any independent effect, though his-
torically some accompaniments of greeting or parting behaviour
have stated or implied a peculiar virtue of their own, as in the giving
of a blessing.
Study of greeting and parting procedures has behind it some con-
siderable fields of inquiry. One is the study of personal encounters
developed particularly by Erving Goffman, and itself using results
obtained from work on language, on small groups and on psycho-
therapeutic relationships. Another is the study of body movement
and the use of the body as an instrument of communication, asso-
ciated especially with the work of E. T. Hall and of Ray Bird-
whistell. Linked also, in contrast as well as in similarity, are the
studies in animal behaviour conducted by ethologists under the
general head of ritualization (for example Julian Huxley et al^ 1966).
While each of these types of inquiry can supply fruitful suggestion
about the significance of bodily behaviour, they cannot be trans-
ferred completely to the problems of greeting and parting as studied
by a social anthropologist. These demand knowledge of a broader
social context, an institutional matrix, often lacking in the more
individually focused human studies. In the ethological field, the lack
of a developed language in animals means not necessarily a less
sensitive interpretation of signals among participants, but the ab-
sence of a medium for diversification and amplification of com-
SYMBOLS OF GREETING AND PARTING 3OI
munication which is so fundamental to human society that it can be

regarded as constituting a different dimension of behaviour. If one is
considering only non-verbal human behaviour of greeting and part-
ing, however, the formalized use of body and limbs often strongly
suggests analogies in animal behaviour of a routine communica-
tive kind, with display elements, signalling relevant information on,
for example, sexual accessibility or territorial interest. But such
animal ritualization is normally characteristic of all members of a
given species in the given situation, and is regarded by ethologists as
being adaptive, with positive implications for survival. Seemingly
corresponding behaviour in man has a much greater range of varia-
tion, not only within the species but also within a society, and its
adaptive possibilities are often obscure.
Leaving aside possible analogies with ritualization in animals,
human formal behaviour at greeting and parting is often termed
ritual. If by ritual is meant symbolic action in relation to sacred
objects, the term is often inappropriate for the actions when people
meet or part. But in a broader sense it is relevant since the actions are
commonly formal, institutionalized, and credited with moral, some-
times almost mystical value. They are communicative, but the
information they convey refers to the control or regularization of a
social situation rather than to some descriptive fact. For instance, an
explorer greeted by indigenous people may be given signs about
food and water, but more commonly is welcomed by expressions of
friendship, which may be termed symbolic of social relationship. In
general, greeting and parting conventions may be regarded as a
mild variety of Van Gennep's rites de passage - what Elsie Clews
Parsons characterized as crisis ceremonialism, 'ceremonial to signalize
or allow of the passing from one stage of life to another' (1916, 41).
Following her lead, one might coin the term teletic rites, from the
Greek concept oitelesis, putting off the old and putting on the new.
One can apply this term to greeting and parting behaviour, where
the major stimulation is provided by the arrival or departure of a
person from the social scene.
In the human field, forms of greeting and parting vary not only
according to differences of culture, which might be regarded as sub-
specific areas, but also within individual cultures, according to
relative status of persons involved and type of social situation:
302 SYMBOLS
individual or group confrontation; formality or informality of the

occasion; prior acquaintance or unfamiliarity of the parties; conven-
tional emotional quality ascribed to the occasion; face-to-face or at-
a-distance communication. Even between people well known to one
another, the relative formality of the occasion may affect the style of
greeting or parting. At a public reception the structure of the words
and acts of hosts and guests tends to be rigid and highly prescribed,
in contrast with the easy exchanges of welcome or farewell at a week-
end party or any other ordinary social gathering. Even if an occasion
is not classed as a formal one by the nature of the invitation, the
presence of merely casual acquaintances among one's friends may
have an inhibiting effect on greeting or parting behaviour. American
or British people who might exchange a kiss in private greeting may
refrain from such intimacy in public. But this is a highly cultural
matter - a Frenchman in office may bestow a kiss on another on a
formal public occasion when he would not do so at an informal
private meeting.
In ordinary social intercourse formality may be enhanced when
the person to be signalized in greeting is previously unknown to the
signalizer. I myself have observed a simple form of this in travel in
the United States. When two men previously unknown to each
other are seated side by side in an aeroplane on a long trip, informal
social relationships of an elementary order may be established in
getting to and from seats, remarks about the weather, etc. But after
casual exchanges may come formal identification. T m James Brown,'
says one man, extending his hand, whereupon the other normally
follows suit with his own name and handclasp. Search may be made
for further points of contact-home town, business, common
acquaintance. But presumably gratifying, this is not of primary
relevance. What is of main significance is the act of identification,
which has been sealed by the clasp of hands. Indeed, once the name
has been uttered the hand gesture is almost automatic; as an experi-
ment, one may disconcert such a casual acquaintance by uttering
one's name but withholding the hand movement! Yet in materialist
terms the act is almost meaningless - the men are seated side by side,
have already talked and could continue to talk; they will probably
never meet again; neither wants anything of the other except his
temporary companionship, which is already available. The name
SYMBOLS OF GREETING AND PARTING 303
given may even have been false. But what matters is that some name
has been given, and some personal manual contact made. The
accidental travel neighbour has been socially pin-pointed, and the
handshake is the formal symbol of a social relationship established,
of the reduction of an unknown to a (putatively) known social
position. The handshake has been interpreted as equivalent to a
disclaimer of aggression, as a residual pledge against resort to force
of arms. But in such modern conditions the risk of physical assault
is minimal (these words were first drafted before air hijacking had
been invented). If a threat is conceived it is of interference with the
personality, not with the person, and the handshake has a much more
subtle function of serving to reduce social uncertainty.
In such a situation the social identification has been provided
by the two parties immediately concerned. The relationship is
casual, the identification nominal. But it is not automatic; if neither
party takes the initiative no formal greeting takes place. Moreover,
it is status-regulated. Handshake in such travel conditions is pri-
marily behaviour between male equals or those making a show of
equality. My guess is that it occurs much less frequently between
men and women; and between women no handshake at all may be
exchanged - though other identificatory signs may be. Between
adults and children the handshake of casual acquaintance is nearly
always lacking. When children shake hands, it is commonly a con-
cession to adult, usually parental, direction, and is primarily a symbol
of relations between the adults involved - the child is a kind of
instrumental extension of adult greeting.
Modern anthropological idiom stresses both the communicative
and the expressive functions of ritual. In rituals of greeting and part-
ing the inter-relation of these elements is quite complex. There is
conveying of information - of social recognition and acceptance,
of status and general quality of the relationship. But there is also
an implication, in popular estimation in Western society, of emo-
tional involvement, if even low-keyed. Greeting behaviour is
expected to express approval of the encounter, even pleasure;
parting behaviour is expected to express the opposite, or at least a
recognition that the severance of the encounter is necessary. Yet
it is a matter of common knowledge that the reverse may be the
case, or that the parties may be indifferent. Popular jokes make play
3°4 SYMBOLS
of the incongruity between outward behaviour and inward feeling

on such occasions. But while such incongruity is openly admitted
out of the immediate context, there is general demand that the forms
of emotional interest be preserved, and to omit them gives offence.
Politeness consists in not allowing incongruity to be overtly per-
ceptible, though it is possible to so shade wording, tone of voice,
mode of gesture and even posture in greeting or parting behaviour
as to allow indifference or even negative attitudes to be inferred.
In reflective comment, then, forms of greeting and parting can be
regarded as 'empty', as 'meaningless'. But while their expressive
function may not correspond to their superficial form, greeting and
parting behaviour is still significant as a communicative device.
'Pleased to meet y o u . . . ' on introduction, 'How good of you to
have c o m e . . . ' at parting, as overt expressions of satisfaction may
be directly contradicted by more private behaviour. But what they
express by implication and communicate explicitly is a willingness to
enter into, or to continue a social relationship. Their simulation of
emotional involvement or moral approval may be only part of the
conventional 'small change' of social intercourse, but it has an
emollient quality. It is a 'softener' of the social relationship because by
convention it sets the relationship in a status frame, implying respect
by the speaker to the personality of the recipient.
The significance of such communicative function is brought out
by forms of greeting or parting in societies which use no ex-
pressive modes implying affect, but confine themselves primarily
to descriptive phrases of acknowledgement or invitation. 'You have
come'; 'Enter (the house)'; 'You are going' are typical expressions
of greeting or parting in many societies. It emerges clearly from
such evidence that the primary factor of importance is not what is
said but that something is said. The action of verbal utterance
signifies recognition of the situation in social terms; potentially
threatening silence in which intention is uncertain is broken; the
physical movement of coming or going is put into a social frame,
and the parties are in communication.
In greeting and parting rituals verbal and non-verbal behaviour
is in close relation. Utterance and bodily action are often simul-
taneous or juxtaposed - as saying 'good morning' with a smile and
handshake, or calling 'hi' or 'hullo' with a wave of the arm. Much
of this is reinforcement behaviour, intensification of what is ex-

pressed and communicated; but sometimes it may be alternative, as
when one waves a greeting to a person too far off to hear any words.
In greeting, an interesting combination of verbal and non-verbal
behaviour occurs in what is known conventionally in English as
'an introduction'. Traditionally, in sophisticated European circles,
greeting between two persons not previously acquainted should be
prefaced by an introduction by a third party previously known to
both; without having been 'introduced' the two people should not
address each other in a formal social situation. This rule was especially
severe when a man and a woman were concerned. Though the rule
has now been relaxed except in the most formal circumstances,
an intermediary may be still employed. 'Will you introduce me
to * can still be an intelligible and appropriate request in some
social situations. The role of the third party here is twofold: he
(or she) is a social bridge, a mediator who facilitates the social
contact of the other two people; he may be also an ostensible guaran-
tor of their social identity, and up to a point, of their reputation.
In such introduction the words 'May I introduce/present...' are
commonly accompanied by changes in orientation of the head,
movements of the hands, inclination of the body, which relate by
metaphorical links the two persons who are being introduced, and
give signals to them to begin their own greeting patterns. But the
concept of introduction is not simply a relating of persons socially;
it embodies the notion of one party being displayed, led forward,
'presented' to the other; it is a sequential operation. Hence the ele-
ment of relative status tends to emerge - it becomes important to
indicate who is presented to whom. By convention the latter, who is
the recipient of the introduction, is deemed to be of higher status
than the former who as the person being introduced is in a depen-
dent position. (In ordinary social introductions both parties are
assumed to be of equal status, and the traditional rule has been for
the introducer to repeat the names of the parties, in reverse order,
so that each in turn has the status of recipient.) Rules of formal
introduction follow general rules of social precedence: a younger
man is introduced to an elder man; a man to a woman (a 'gentle-
man' to a 'lady'); an unmarried woman to a married woman; a
commoner to a member of a Royal Family. But this rule may be
306 SYMBOLS
overridden, according to context, when immediate role may be more

significant than general status, or than rank. So, an elderly clerk may
be introduced to a younger man who is manager of his firm, a
member of a Royal Family to the head of a college where she is
becoming enrolled as a student. But as already noted, such cases are
usually narrowly defined by the immediate role situation. So, in a for-
mal university degree-conferring ceremony, an officiating dean may
introduce an honorary graduand to the Chancellor, with mutual
acknowledgement of tipping of hats in salute, while in private life all
three may know one another very well and be on first-name terms.
A general point of theoretical significance is the parallelism which
obtains between much greeting and parting behaviour. Some
societies, for instance the Japanese, have an identical term to cover
both meeting and parting salutations. Formerly, in Western
European social circles, a gentleman raised his hat upon taking leave
after an open-air encounter just as he did when the encounter took
place. In Tikopia an expression of what sounds like dismay (aue)
is uttered in greeting a friend not seen for a long time, just as when
he departs; and pressing of noses takes place on either occasion.
In such cases the parallelism of greeting and parting ritual can be
regarded as an expression of formal recognition of change in the
social situation. In the marked behaviour of the Tikopia example,
there is communication of affect, but what is being expressed overtly
is not goodwill primarily, but emotional disturbance and modifica-
tion of social relations. The arrival is represented as a moving event,
analogous in this sense to a departure. Emphasis is placed not on the
contrast between the joy of greeting and the sorrow of parting, but
on their similarity. Both arrival and departure of a friend alter social
routine, and are disruptive of ordinary emotional routine; regret for
past severance and regret for future severance are aligned in be-
haviour (cf. Radcliffe-Brown, 1922, 239-42). This parallelism in
greeting and parting ritual brings out the symbolic quality of the
behaviour-it is not just an immediate reaction to the arrival or
departure of a social acquaintance - a kind of register of fresh pre-
sence or impending absence. It embodies elements of less obvious
significance, relating to the social implications of the movement of
persons, in which implicit reference to the time dimension may be
involved, and to values associated with presence and absence.
S Y M B O L S OF G R E E T I N G A N D P A R T I N G 307
In the remainder of this analysis I am concerned primarily with

non-verbal aspects of greeting and parting behaviour.
S Y M B O L I C USE OF THE W H O L E B O D Y I N
GREETING AND PARTING
Metaphorical use of concepts associated with the physical human

body to refer to some types or aggregates of social relations is a
common ethnographic phenomenon, and Mary Douglas (1970)
has made an interesting study of some aspects of this. Among other
themes she has emphasized the argument of Marcel Mauss, that the
'social body' constrains the way the physical body is perceived.
Granting the general relevance of this, there is a great variety of
circumstances in which the physical structure and powers of the
human body are perceived in direct relation to performance, irre-
spective of the character of the social body. Simple mechanical
actions such as lifting a pot or paddling a canoe are interpreted cross-
culturally in similar terms, with recognizable common evaluation of
achievement. Perception of arm and body in movement, and con-
ception of their function, would seem to involve minimal social
determination. But the context of associated ideas soon becomes
important. When in northern Nigeria I saw men raise their arms by
the roadside and shake clenched fists at me I soon realized that this
was not a threat but a greeting; the form of the body movement and
its interpretation were constrained by conceptions of a social pattern
(Firth, 1951a, 23). Here I consider briefly not the social relevance of
general concepts of body, but the social interpretation, in contexts of
greeting and parting, of use of body movement.*
The human body is far more of a social instrument than is often
suspected, in the greeting-parting syndrome. Such instrumentality
is mainly symbolic, the element of physical performance being
* Underlying much of the modern anthropological approach to these problems is a rejec-
tion of the Cartesian dualism of mind and body. An important set of essays on the difficult
metaphysical issues involved has been assembled by Stuart F. Spicker under the title of The
Philosophy of the Body (1970). In his studies of body motion communication Ray L. Bird-
whistell, a pioneer in the 'kinesics' field, points to the misconceptions in vogue about so-
called 'natural' gestures, stating that after fifteen years' research he and his associates have found
no body motion or gesture which has the same social meaning in all societies (1970, 81). For
other relevant analyses see Macdonald Critchley (1939), D. Efron (1941), E. T. Hall (1961,
1968). A popular treatment of 'body language' by Julius Fast (1970) attempts some practical
inferences about meaning of communication by bodily posture and movement.
308 SYMBOLS
usually small. A greeting movement may occasionally be of primarily

physical import, as when a host opens a door or unlatches a gate for
a guest, but even here the physical act of admission may also involve
a symbolic act of welcome, a throwing wide of defences in a way
which has other obvious analogies. The body as a whole can be a
greeting or parting symbolic instrument in three main respects:
by maintaining a distance gap between the parties; by adopting an
overall posture; and by movement in meeting or dismissing the
other party.
Degree of spatial distance kept between the parties is apt to be
broadly an index to degree of social distance between them. In
Western societies this operates unsystematically as a rule (military
regulations offer contrary instances). But in societies with a developed
caste system, especially where notions of pollution by contact occur,
bodily relationships may be strictly regulated by distance, even to a
measurable index, and meeting and greeting behaviour structured
accordingly. A. Aiyappan has recorded (1944, 38-9, 46) that in
traditional Malabar society a member of the Irava caste (the highest
of the 'castes that pollute from a distance') polluted a Namputiri
(Brahman) from a distance of thirty-two feet, and had no access to
the houses, temples and wells of the higher castes, and no freedom
to use roads or footpaths in the presence of members of these castes.
To obtain communication with a member of a higher caste an Irava
might then have to use an intermediary while himself remaining at
bodily distance.
Bodily posture is important in many greeting conventions. One
mode of showing respect is by sinking to the ground, conveying a
depreciation of the self and symbolizing humility and recognition of
superior status. Hence in ordinary social intercourse a mutuality of
esteem is expressed by mutual body lowering. S. F. Nadel has noted
(1942,129) how Nupe men of equal rank greet each other by sinking
low (his word is 'cowering' which seems hardly apt) for half a
minute or longer while exchanging salutations. If they are well
acquainted, they stretch out their hands several times in succession
and lightly touch each other's fingers. A man of lower rank when
meeting his social superior will bow very low or kneel down, and
only if difference of rank is slight or obviated by personal intimacy
will he venture to offer his hand. If he is on horseback he will dis-
SYMBOLS OF G R E E T I N G A N D P A R T I N G 309
mount, if he is wearing sandals he will take them off, though it may

be on the road, and put them beside him. The man of higher rank
will stand or sit still, make a short perfunctory gesture of bowing,
and hardly move his arm to meet the other man's hand. Eugene
Mangin has given an equally-detailed description of the traditional
form of greeting known as the 'Mossi salute' (1921,18-19). In order
to give such a greeting a Mossi would put down his load, hat and
sword, remove his shoes, sit on the ground, fold his legs slightly and
lean the upper part of his body forward. The elbows rested on the
ground, as far apart as possible; the forearms, folded in front of the
chest with the hands half-closed and the thumbs raised, struck the
ground together three or four times. In such humble posture, the
man concerned did not look the person being greeted in the face.
Father Mangin noted that though complicated, the Mossi traditional
salute was rendered easily and gracefully, not in the clumsy awkward
manner in which Mossi attempted to imitate the European military
salute. Such a form of greeting appears to have been used in all the
Black empires of the Sudan, especially Ghana, Mali and Gao. In
such type of greeting, the lower the status of the person initiating
the action, the more the disturbance to which he subjects himself- a
principle which is of very general application. To greet by a distur-
bance of the whole body to such a degree that one sinks to the ground
is alien to Western notions of everyday behaviour - though not to
some notions of ceremonial behaviour - and Westerners have often
reacted with strong ethnocentrism at the sight of it. Mangin ob-
served how foreigners on first entering Mossi country were shocked
to see men thus lower themselves before others. A generation later,
in Nigeria, I myself heard the labels of 'undemocratic' and 'degrad-
ing' applied to an analogous Hausa greeting convention in which
men mutually sank to the ground in a flowing movement and lightly
touched hands. In early relations with Far Eastern societies, exam-
ples of body-lowering in greeting, such as the Chinese 'kow-tow',
provoked shocked, contemptuous interpretation from Westerners,
who ignored the occasional practices of kneeling, crouching and
even prostration in some Western institutionalized contexts. (Some
study of these is given in my consideration of postures and gestures
of respect (1970c), in some ways a companion piece to this chapter.)*
* For early discussion of this subject see Tylor, 1878, 46-8. As recent additional examples
3 io SYMBOLS
Even where status differences are not highly marked in ordinary

social intercourse (as in the West in modern times) and etiquette
demands equalization of posture in greeting between social equals,
some modulations are still observable. Commonly, a man sitting in a
room rises to his feet to greet another man who has come to see him.
But in social conventions which still have some currency, it is 'pro-
per5 for a lady to remain seated while greeting a man who has entered
the room and who stands before her. Yet a man who is manager of a
firm may not rise in saying good morning to his woman secretary
('lady' and 'woman' here being status indices). Such differences in
greeting posture correspond to recognized if minor status differen-
tials. Yet ambiguity can occur, especially if the categories are mixed.
However, modern practice allows considerable personal discretion,
so a range of variation is found. A manager may rise to greet his
woman secretary when she first comes in, in the morning, in recogni-
tion of the general sex rule that a gentleman should not remain
seated when a lady is standing; or he may stay in his chair in accord-
ance with his interpretation of the professional situation. Or he may
be guided by some rule-of-thumb - remain seated if the secretary
is a young girl or rise if she is a mature woman, in response to a cross-
cutting age rule; or he may rise when she first comes in as a general
acknowledgement of the sex rule, and remain seated thereafter in
conformity with the professional rank rule. What is of special signifi-
cance here is how, in what is characterized as an ordinary practical
business situation, other codes are still allowed to intrude. But the
point of most general interest is how in such situations of'manners',
which are supposedly matters of delicacy, we use our bodies in a
lump, as it were, in crude mass, as expressions of social relationship.
may be cited: a woman kneeling to be ordained as deacon by a Bishop of New York Diocese;
the Duchess of Windsor curtseying to Queen Elizabeth who was taking leave after a visit;
200 candidates for the priesthood in the Philippines lying prostrate in front of the altar as
part of their ordination ritual by the Pope (New York Times, 14 November 1971; 4 January
1972; The Times, 19 May 1972). An elaboration of a greeting gesture took place when Queen
Elizabeth visited France. Having been told that the Queen liked horses the French introduced
a white horse which had been trained to kneel, and which at the appropriate moment went
down on its knees 'in a dutiful manner' on a pile of sand provided for the purpose. In contrast
to such proceedings was the attitude of Mrs Martha Mitchell, wife of a high United States
official, who remained upright when presented to Queen Elizabeth at a garden party. 'I feel
that an American citizen should not bow to foreign monarchs', Mrs Mitchell wrote in explana-
tion - curtseying apparently being optional for guests who are not British subjects (Time
Magazine, 29 November 1971).
Moreover, we tend to do this as a matter of course, as if manipulation

of the body mass is the recognized equivalent of an agreed articulate
language. A further point is that in a very rough way, the amount
of bodily displacement engaged in by each party is in inverse pro-
portion to his status - the lower the status the more the bodily
movement. The material act of expenditure of energy is translated
into a symbolic act of status acknowledgement.
What I have just considered is mainly vertical body movement.
But a similar broad symbolic interpretation is given to horizontal
body movement in greeting and parting. In Western society only in
diplomatic protocol and analogous ceremonious exercises, such as
those of a university graduation, a Royal Court or a religious initia-
tion, is much formal expression given to the significance of body pro-
gression in greeting or parting. But tacitly it is fairly well understood
in higher business or professional circles. To be conducted into a
businessman's office by a secretary is usual, but gradations of being
farewelled - at your host's desk, at the door of his inner sanctum,
in the outer office, at the door of the lift (elevator), down at the
front entrance of the building - are recognized status indicators.
As with the Nupe and many other peoples, the degree of dis-
arrangement of the one party indicates the relative status of the other.
In traditional Chinese officialdom such gradations were expressed
with much greater finesse, and had much more specific status signifi-
cance.
SYMBOLIC GESTURES IN GREETING AND PARTING
Gestures are commonly regarded as movements of the hands and

face in particular, accompanying speech for purposes of emphasis -
a kind of italicized speech, as Macdonald Critchley has put it
(1939, 11). But some gestures may serve as substitutes for speech,
especially when they act as a general means of communication or
expression for which no simple speech equivalent can be found, or
when speech would be inappropriate, as during a solemn church
service. In the translation of movement of parts of the body into
symbolic gestures of greeting or parting the range is wide. If one
includes what may be called the 'receptor-parts', which are involved
as platforms or bases against which the action is performed - for
3 I2 SYMBOLS
example breast to which another person is clasped-very many

major external features of the body have been called into service in
different cultures. Western societies confine their greeting and
parting gestures very much to the upper parts of the body.* But
in areas of highland New Guinea, as I myself have seen, a greeting
embrace between men is accompanied by mutual patting of each
other's buttocks - a kind of 'pat on the back' which has slipped
down. Even the feet, which one is inclined to associate primarily
with gestures of rejection or contempt (spurning with one's foot;
stamping in disapproval), may serve as receptor, as in the Oriental
gesture, simulated or actual, of 'kissing the feet'. But I confine myself
here to consideration of the symbolic use of head and hands, and
features of these.
As a prime sensory centre the head may perhaps be expected to
serve as a major medium for gestures of greeting and parting, which
involve recognition of another person, usually by visual means.
In fact, some movement of the head, linked with eye movement, is a
characteristic recognition and greeting signal in many societies,
either with or without verbal utterance. But cultural nuances occur.
Commonly in Western societies an upward nod is a recognition or
greeting signal, whereas a downward nod is a signal of assent in
reply to a question. By convention in some circles an upward lift of
both eyebrows is a recognition signal to someone in a crowd
whereas a raised single eyebrow is a quizzical comment not articu-
lated. (The wink, a lowering of one eyelid, is a wordless gesture
directed at a person to indicate a familiar relationship of complicity.
It may be used in greeting, especially where a more formal action
might disturb a gathering.) t But among Melanesians of the Solomon
Islands an upward lift of both eyebrows is a common sign of assent.
Generally, the greeting or parting function of the eye is expressed by
the look, which would seem to be produced primarily by subtle
local muscle movements around the eye and not by the eye itself.
* But cf. a fictional incident in the Battle of Britain, in Elleston Trevor's novel Squadron
Airborne (1962): a male corporal pinched one of the Women's Auxiliary Airforce on her
bottom; she protested but walked away amicably 'mincing her hips in an aircraftwoman's
farewell'.
f Charles Darwin studied nodding and face movement in terms of signs of affirmation or
negation, and concluded that despite some cultural variation the forms were too general to be
regarded 'as altogether conventional or artificial'. But he was looking for innate responses
(1872, 274-7).
(Modern psychological studies have thrown much light on the

significance of eye movement, though in somewhat restricted social
contexts.) Values attached to changes in modes of treating the hair,
as symbols of greeting or parting, have been discussed already in
Chapter 8.
In Polynesian societies the nose was the organ traditionally used
in greeting, and is so even nowadays, though 'nose rubbing' has
been supplemented by handclasp (for details see Firth, 1970c, 196,
199-201). The tongue seems an unlikely greeting medium in its
own right, since use of it outside the mouth inhibits speech. But
putting out the tongue symbolically as a greeting gesture has been
well authenticated for Tibetans (where it apparently has no overt
aggressive significance) and for Maori (where it forms part of their
superficially aggressive welcoming dances for visitors). Although
the cheek is a relatively immobile part of the face it plays quite an
important role as the site for delivery of a large proportion of kisses,
and as a lip-substitute in cheek-to-cheek greeting gestures of
familiarity which are not so intimate as to require use of mouth.
Only the ears, as far as I can judge, seem not to be involved sym-
bolically in any greeting or parting gesture. Though not less mobile
than the nose, and in modern civilized life far more involved in the
recognition of objects, their position at the sides of the head away
from the eyes has seemed to disqualify human ears from any signifi-
cant cultural use as greeting signals. (The ears of most animals seem
to fill such a function much more directly.)
Apart from the importance of the mouth as a focus of interest for
verbal greeting, its use in greeting gesture is twofold. In many
human societies the curious gesture of drawing back the lips and
baring the teeth - the smile - is one of the major signs of welcome.
The significance of this in non-human primates has been extensively
studied, and also in children. But as far as I know no systematic
study has yet been undertaken of smiling by adults, in different
cultural settings of greeting and parting situations. Ray Birdwhistell,
who probably has come nearest to such a position, has some very
pertinent observations on the differences in the frequency of smiling
among the inhabitants of different areas of the United States, on
smiling as a learned response, and on the complexity of behaviour
involved in smiling and of the concept given this name (1970,
L
314 SYMBOLS
29-39). At a crude level of differentiation it seems clear that the

baring of the teeth when they are tightly clenched is a sign of anger
or of other strain, not of pleasure, and an inference sometimes drawn
is that smiling as a greeting sign may be related developmentally to
baring of the teeth in defence. But this is hypothetical. It would seem
that any subtle interpretation of the meaning of a smile demands
attention to other muscular behaviour of the face - around the eyes,
at the corners of the mouth - as well as to type of lip position and
movement, and degree to which the teeth are displayed. Subtle
differences of this kind account largely in Western society for our
reading of mouth gestures as: the fixed smile, indicating stress or
underlying unease; the secret smile, indicating private thought; the
dubious smile, indicating only partial acceptance of what has been
said; the sneering smile, indicating contempt; as well as the welcoming
smile. Anthropologists familiar with particular alien cultures will
probably agree that granted a reasonable margin of error, they can
interpret the behavioural clues of smiling in those cultures. The evi-
dence for this view would probably be the conformity of what
happened later to their expectations from the smile. Though the
published evidence is slim, and systematic research on the subject
largely lacking, superficial experience does suggest that the smile
is one of the most characteristic mouth-greeting signs in very many
societies. But how far the smile is a formalized sign of acceptance
of a greeting situation rather than a spontaneous response to the
advent of a visitor is a more difficult problem.
KISS OF GREETING
Certainly not all mouth gestures are universal. A type of gesture

which is thoroughly built into the conventions of Western society
is the kiss. Primarily a lip gesture, the kiss is susceptible of a great
many modifications. Linked with the sensitivity of lips is their
function in erotic contact, which introduces a possible ambiguity
into their use in gestures which are intended to signalize social but
non-erotic greeting. It is presumably for this reason that many
non-Western societies ignore the kiss as a conventional mode of
greeting and confine themselves to its analogue in the lip-play of
mother on babe, or of lovers. Some societies, as for instance India,

even regard the kiss with distaste and are apt to treat it as immoral.
It appears that Indian films do not involve kissing scenes so popular
in Western films, and censors have been required to cut out kissing
scenes from foreign films before passing them for local distribution.
This is not a revulsion from the symbolic presentation of sex
themes, but rather from the public display of non-symbolic sex
contact. By contrast, in many overt Indian ritual contexts the
phallic symbols oilinga and yoni appear with no disapproval shown,
and historic sculptures to which symbolic meaning is given have
much erotic content. By contrast with the tender associations of the
kiss in Western eyes, reinforced by many literary references, one
can cite specific counter-views from other cultures, especially in
Africa. Henri Junod reported from the Thonga that kissing was
formerly unknown to them, and laughing at Europeans Thonga
would say: 'Look at those people! They suck each other! They eat
each other's saliva and dirt!' (1927, i, 352; for analogous Chinese
ridicule see Doolittle, 1876, ii, 375).
Even when established as a greeting, the kiss is subject to many
variations. In England over the last century the normal convention
has been that women may kiss women, but men do not kiss men,
except in intimate family circumstances such as greeting or parting
between father and son - and even then it is optional. In circum-
stances of some familiarity, such as close kinship, men and women
may greet one another by kissing. Both men and women may kiss
young children of either sex, possibly with no prior tie at all (for
example the classic 'kissing of babies' by a parliamentary candidate).
In the past the rules have been fairly clear, and at the periphery could
be invoked to justify familiarity, as when a man might claim a kiss
from a girl on the ground that she was a cousin. But in modern
times these rules have been considerably relaxed. In circles that I
know, kissing between men and women friends has become much
more common, in line with general relaxation of social norms in
favour of more individual freedom, and less perturbation about
possible erotic implications. In Britain the idea still seems to hold
that except in special circumstances such as those of family life,
kissing between men is a sign of effeminacy. But it is well recog-
nized also that in some other Western countries, for instance France
316 SYMBOLS
and Russia, kissing between men is a definite part of social conven-

tions of greeting, especially on public occasions. But an accom-
paniment to kissing in many social circumstances is the embrace by
the arms, and it is this arm-breast contact rather than the lip-lip or
lip-cheek contact that appears to be of prime significance. Certainly
the greeting-hug is as much a public demonstration of amity as the
greeting-kiss.*
The kiss itself presents several forms, graded roughly according
to type of social relationship. Lovers (commonly) and spouses
(often) greet each other with a lip-to-lip kiss of varying duration and
intensity. Parents and children may greet lip-to-lip, but other kins-
men and close friends usually greet lip-to-cheek - normally on one
cheek but sometimes on both if moved by affection or continental
experience. But while in members of the immediate family and among
close kin or friends the erotic implications of a kiss are expected to be
absent or subliminal, they may have more overt interpretations
between people of more distant relationship. (I have noted a recent
letter to an American newspaper columnist in which a young woman
asked for guidance-she had greeted her husband's father with a
kiss on the cheek but he had insisted upon a kiss on the lips, which
shocked her.) So, in some circles there is a custom of what may be
called cheek-to-cheek kissing - a facial contact without use of mouth
at all. This delicate gesture of greeting may have an emotional charge,
but it is removed from the more erogenous zones. (Such a cheek-to-
cheek greeting is treated as a completely appropriate gesture by
Westerners, whereas a nose-to-nose gesture is treated as odd -'nose-
rubbing'!) In some sections of European continental society another
delicate if more socially distant usage is that of man taking a lady's
hand and bringing it up before his lips as he bends over it. 'Kissing
the hand' need not involve any actual contact of lip with hand;
though the simulated gesture of brushing the lip over the lady's
fingers can be made to serve a variety of purposes from respectful to
erotic messages. A further simulacrum, in which the respect motif
* As a demonstration of cultural difference, compare English football players who have
just won the much-coveted Cup hugging each other in delight (but not kissing), with the
Russian greeting of returning cosmonauts when Leonid Brezhnev gave each 'a bear-hug and a
kiss*. Recently Brezhnev was described as sometimes going 'the traditional Soviet bear-hug
one better by kissing men and women alike smack on the mouth' {Newsweek, i November
1971).
of greeting is uppermost, is to substitute for the actual kissing

motion a verbal statement of the 'I kiss your hand, Madame'
type.
When greeting takes place outside the domestic circle or circle
of friends, the implications of a kiss may include quite complex
political considerations. In general, the familiarity of the kiss,
ordinarily reserved for intimate acquaintance, signifies a compli-
ment, implying that the recipient is worthy of a more intimate social
relationship, of positive sentimental content. In Hawaii, the custom
of greeting visitors with a lei and a kiss has been turned to account
in public receptions as a symbol of amity between the parties. On a
recent visit of Queen Elizabeth to France the President greeted her
with a warm handshake, whereas his predecessor 'the avuncular
President Coty' had earlier greeted her 'with a decisive kiss on both
cheeks'. A political explanation that after monarchical governance by
de Gaulle had ended, the institution of monarchy may have lost
some of its allure, may be far-fetched; but it would not be surprising
if political as well as personal undertones could be found in the
difference of greeting (see Observer', 21 May 1972).
The significance of the lips as erogenous zones has given kissing
a kind of danger-quality as a greeting. It is regarded in Western
society as a form of greeting for which above all permission may have
to be asked. To kiss someone is apt to be held as an intrusion on the
personality more intimate than that of any other form of greeting.
To ask 'May I shake your hand?' apparently sometimes occurs, as a
frank signal of respect, though my impression is that it has rather a
formal literary quality. But no one asks 'May I clap you on the
shoulder?'-this would seem ridiculous. So, kissing on special
occasions such as birthday or wedding, or in congratulation for
some honour gained, takes on a special quality of exchange of
personalities, in which the sub-erotic qualities of lip-contact may be
a significant component. (Hence, one can imagine the press interest
in such incidents.) So too, granted that in Western society kissing is
an acceptable form of greeting between given categories of persons,
a number of associated patterns have been developed to mitigate its
side-effects and to reduce to a minimum any suggestion of erotic
elements in the exchange. What is of particular interest is the way in
which Western societies are prepared to skate upon thin ice in
3i8 SYMBOLS
juxtaposing so closely an erotic and a merely social gesture, while

most other societies have avoided this problem.
In some circumstances a kiss is overtly a respect signal rather than
a greeting signal. Such instances as kissing a bishop's ring, kissing the
toe of St Peter's statue and certain other effigies of saints in Rome,
'kissing hands' of the Sovereign before taking up an overseas
diplomatic appointment, exemplify this. But the gesture also conveys
a sense of greater intimacy than others - some element of self-
involvement of an emotional kind in the relationship. A point here
is that the involvement is one-way: in such contexts the gesture is
not reciprocated, hence the emphasis is on acknowledgement of
superiority, on respect for status, rather than on greeting as such. It
is from this point of view that the most notorious kiss in history - or
mythology - that of Judas Iscariot, carries its shock effect. The
gesture of respect shown in the kiss, and shown intimately, was
incompatible with the repudiation of respect and of intimacy shown
in the betrayal. This example also shows a significant function of the
kiss in some circumstances - as a marker of social position. A
critical social indicator may be provided by who is greeted and who
greets in a group, by kissing.
H A N D A N D ARM I N G R E E T I N G A N D P A R T I N G SYMBOLISM
Comparable in mobility with the mouth as a greeting instrument is

the hand with the arm. Here too great variety of gesture is shown,
both cross-culturally and within any single society. Broadly, greet-
ing and parting gestures may be conveyed by salutation with one
hand, by joining one's hands, or by joining hands with the other
party. Avoidance of physical contact between the parties in the one
case and seeking it in the other may be symptomatic of deeper
cultural attitudes about the nature of personality and the degree to
which it can be compromised or reinforced by intrusion of another
personality upon it. One-handed greeting may be made with open
or closed fist. In the West we have been accustomed to regard the
clenched fist as primarily a sign of aggression and defiance, and it has
been conventionalized on such a basis into a political communist
greeting and display. But as I mentioned earlier (p. 307) the shaking
of clenched fist in northern Nigerian pagan society may be an
ordinary greeting, a token of amity. Open-handed salutation may

vary from the casual wave of greeting to a passing acquaintance or
the more regular flapping of the hand on an uplifted wrist in farewell,
to the precise military salute, formalized and designed expressly to
indicate and support differences in rank.
Traditionally, in the East, a person has joined his own hands in
greeting, either clasped in front of him in Chinese style, or set palm
to palm in front of the face, as in India and many parts of South-
East Asia. In some of these societies the height at which the hands are
held or to which they are raised is a measure of the social status of
the person greeted. As with posture (Firth, 1970c) this is one of the
simple physical indices which can be made of great symbolic im-
portance by setting them within a social framework. In the West
the tradition has been for each person to proffer a hand to the other,
and the greeting has consisted of a handclasp or handshake. An
Oriental variant of this is 'hand-joining', with both parties using two
hands. One, normally the junior, places his hands palm to palm
within the similarly-held hands of the other. In Malaya, for instance,
such hand-joining has had rules governing who by sex, age and
difference of rank may properly join hands with whom. Only
elderly women, not young women, should by tradition join hands
with men; this should be in respect and motherly or sisterly affection,
and by custom a clean layer of cloth should be placed between the
pairs of hands (Zainal-'Abidin, 1950, 47). Another form of two-
handed salutation, clapping, is significant partly because it is com-
monly non-reciprocal - though in some countries it is customary to
join in applause given to oneself as a gesture of appreciation of the
compliment paid. But it is also significant because as a form of
percussion, it is one of the simplest accessories to bodily display in a
field which uses a variety of noise-making instruments of salutation
(cf. Needham, 1967b).
Some other societies than those of the West have used the single
handclasp in greeting. Rattray has observed that handshaking as
a salutation appears to have been an Ashanti custom before the
advent of Europeans (Rattray, 1916, 43; 1929, 103. Cf. Speke,
1912, 168, 169). In such case hand had to lock in hand -'five must
lie within five' was the Ashanti way of expressing the full grasp.
To give the tips of the fingers only was discourteous - as in England
320 SYMBOLS
it has also tended to be, except in special circumstances of formal

reception by a person of high rank. Where the handclasp has been
in vogue as a greeting, it has often been customary to use only the
right hand; the left hand, especially in Oriental countries, has often
been associated with the toilet and is unclean. But the reversion of
this rule may mark a special type of social relationship, as with the
Boy Scouts of Britain, or the Ogboni society of the Yoruba. Accord-
ing to Ajisafe (1924, 91) the latter use the left hand 'in a peculiar
way known only to members of the cult'. Such a focus on the right
hand in ordinary greeting is clearly part of the much more general
polarity of right and left examined by Robert Hertz (1909, trans.
i960) and attributed to mystical religious categorization. Hertz did
not discuss the predominance of right-handedness in greeting, but it
is important to note that this applies to the handclasp rather than to
waving - in other words, the mystical value of right-handedness is
relevant primarily in personal physical contact. This illustrates the
complexity of symbolic values in salutation - relatively superficial
indication of amity in the handclasp as such, and deeper less aware
aesthetic and moral significance of use of the right hand.
Since the handshake has been the traditional greeting of Western
countries, with the spread of Western technology and consumer
habits the practice of handshaking has become much more common.
Many societies where it was formerly unknown have now adopted it,
as an ancillary to or substitute for their own traditional greeting. So,
in modern Polynesia, a clasping of hands may be accompanied by a
leaning forward and pressing of noses. But the introduction of the
handclasp has sometimes been cautious, because of other possible
ritual values involved. Junod has recorded how among the Thonga
it was formerly taboo to shake hands with a chief, owing to the
belief that contact with the body of such a person of ritual quality
was dangerous. However, even more than forty years ago Junod
observed that the ruler Muhlaba, on the adoption of Christianity,
was somewhat reluctantly accepting the hands held out to him by
the most modest of his subjects, even by children, on Christmas
Day (1927, i, 383) — the new faith presumably being regarded as
protecting from or nullifying the ancient powers.
Like other forms of salutation the handclasp can be used as a
status differentiator. It has been reported of the Bambara that
SYMBOLS OF GREETING AND PARTING 32I
traditionally a man saluted his superior by extending his palm

upwards, whereupon the superior put his own hand palm downwards
over it. Inversely, when a man greeted an inferior he extended his
hand with palm down; but for an equal his palm was held perpendicu-
lar to the ground (again, notions of upwards and downwards
physically involved with ideas of status). But a most significant
status indicator has been the demarcation of categories of people
with whom one does not clasp or shake hands. Among the Bam-
bara, according to Paques (1954, 118) a woman should never shake
the hand of a man in order to greet him. In Western society of the
'polite' order it was formerly proper to greet servants verbally,
but never to shake them by the hand, which would have been an
admission of equality. There are still occasions of a very formal kind
in which it is appropriate to bow before Royalty or other people of
high status but not to offer to shake hands. Towards the end of
President de Gaulle's official life it was recorded that at presidential
receptions at the Elysee a blue triangular mark at the top left-hand
corner of an invitation card indicated guests who would not be
introduced to the President and guests of honour. 'Restricting the
number of handshakes at Elysee functions is a relatively new practice
which a French newsmagazine has seen as an indication that the
76-year-old general now tries to conserve his energy' {Atlantic
Monthly, May 1967, 116). Clearly the presence or absence of
handshake entitlement was a status indicator of considerable note.
ACCESSORIES IN GREETING AND PARTING SALUTATION

In enhancement of bodily movement and posture in greeting and
parting rituals, any one of three main types of accessory may be
involved: use of items of dress such as hat or gloves; use of elements
attracting attention, by sight or sound, such as waving flags or
firing guns; presentation of gifts such as greeting cards or scarves.
I deal here briefly only with types of accessory most nearly related
to bodily movement.
In greeting and parting behaviour items of bodily covering which
are easily removable, especially from extremities, are often treated
as ritual elements. Their removal is conceptualized as a symbol of
respect to the person greeted, in accordance with the implicit
322 SYMBOLS
principle that lower status is equated with disarrangement of the

person. (This is in line with the custom of showing respect to
sacred places, as by removal of hats by men going into church, or
of shoes on going into a mosque. Conversely, that which is bare
should be covered, as the heads or arms of women in church.) The
removal of sandals or shoes in greeting has already been described
for Nupe and Mossi. Westerners do not remove their shoes in
greeting, but in the period when hats and gloves were worn out-
doors by gentry, gentlemen greeted ladies at a distance by taking off
their hats or at least raising them briefly from the head; and when
shaking hands, they removed the right glove. A lady, of putatively
higher status, was expected to keep on her hat and gloves in greeting.
For more formal occasions in polite society, variations of glove-
wearing and glove length were associated with differences of greet-
ing.
Visual signals to celebrate greeting or departure are manifold. As
extensions of the arm, flags, green branches, garments have com-
monly been waved, sometimes with specific messages as well as
general salutation meaning. The use of some other accessories may
combine visual display with presentation of gifts according to a
complex code. On formal occasions in Fijian society the presenta-
tion of a whale's tooth is still a most important element in the rituals
of greeting, in an elaborate context of prescribed bodily movement
and speech. (In 1970 I myself took part in such greeting ritual in
Beqa, under the guidance of the late Dr Rusiate Nayacakalou.) In
Central Asia traditionally a similar function was performed by the
silk scarf. Owen Lattimore has described how in Inner Mongolia a
scarf is exchanged in greeting between friends, sent with messages
and used in all kinds of ceremonies. Its symbolism is specified closely.
It is folded three times lengthwise; it should be held on the palms of
the outstretched hands, with thumbs hooked over the top; the open
fold or 'mouth' must be towards the front, towards the receiver - if
reversed towards the giver it is discourteous and unlucky, since the
virtue of the gift will 'spill out' (Lattimore, 1941, 38-9). Tibetan
exchange of scarves in greeting used to observe gradations of rank
scrupulously. Sir Charles Bell has noted that if the recipient was of
much higher rank than the giver, he remained seated while the
other person laid the scarf at his feet. If the recipient was only a
little superior the presentation scarf was placed on a table in front
of him; where both were equal in rank, they stood and laid their
scarves over each other's wrists. Should the giver of the scarf be of
higher position he laid the scarf over the neck of the recipient, who
bowed to receive it (Bell, 1928, 100, 253). Another extension of
bodily greeting procedures is the practice of drinking 'toasts' to
guests, involving arm and head movement, focus of attention upon a
glass held up, and appropriate words. Where reciprocal toasting on a
personal basis is part of a formal dinner, status considerations may
enter; it is a mark of respect to invite another person to drink a
mutual toast, but it is the role of the senior to take the initiative.
GENERAL OBSERVATIONS
Greeting and parting behaviour is often treated as if it were a simple

acknowledgement of arrival or departure, or a spontaneous emo-
tional reaction to the coming together or separation of people. The
social message carried is regarded as overt. But sociological observa-
tion, in line with findings of kinesics and ethology, suggests that
such behaviour is highly conventionalized, and culture-specific,
not universal. In a broad sense greeting and parting behaviour may
be termed ritual since it follows patterned routines, it is a system of
signs which convey other than overt messages and so may be
categorized as symbols, and it is sanctioned by strong expressions of
moral approval.
But though human societies show a wide range of variation in
greeting and parting behaviour, everywhere they tend to use a
comparatively small set of basic materials for the purpose, primarily
involving the human body. Ancillary cultural instruments are often
utilized - hats, gloves, scarves, guns, flags - but for the most part
these are extensions of an individual's physical apparatus, his body
and limbs. They serve to supplement his waving and shouting, the
alteration of his features and movements of his limbs, his getting up
and sitting down. This instrumental use of the human body, with
its attachments, for social purposes, is a marked feature of greeting
and parting behaviour, which so far has not been given the systematic
study it deserves. Functionally it involves the participants directly
324 SYMBOLS
in the operations, and promotes the establishment and working of

social relationships. It helps to insinuate new individuals into social
situations, and to fill the gap when known individuals depart; at the
same time small differences in procedure are standardized and made to
serve as carriers of significant social features.
In greeting and parting behaviour involving use of the body as an
instrument I see several major social themes, each connected with the
concept of personality. The first theme is that of attention-producing.
A primary object of much greeting or parting activity is to attract
the attention of the other party - by directed glance or out-thrust
hand if close by or by oscillation of the hand (waving) if at a dis-
tance. By focusing attention on the personality of each participant a
sign is given that further communication is desired. A second theme
is that of identification. The people concerned are differentiated as
persons entering or continuing individually the social relationship.
One function of much greeting and parting behaviour is in providing
a framework within which individuals can identify one another as
preliminary to further action. A third theme is that of reduction of
uncertainty or anxiety in social contact, particularly between persons
who are not previously known to one another. Confrontation with-
out communication is threatening. Even the most casual greeting
gesture tends to remove an element of uncertainty from the en-
counter. To nod or say a brief word to a stranger is more than a
token of friendliness; it puts him in a social context, within which
further communicative action can follow. Salutation at parting
serves in parallel fashion to put a definite point to the departure,
to establish the severance as a social and not merely a physical fact -
not leaving the relationship hanging in the air, so to speak, as an
unresolved issue.
An important element in much greeting and parting behaviour
is status demonstration. Relative posture and gesture, especially
in degree of elevation, are used in very many cultures symbolically
to indicate the relative status of the parties engaged. In this a kind of
rough logic is displayed, whereby bodily elevation is correlated
with social elevation. I would argue that a common, even basic
function of greeting and parting rituals is in creating occasion for
establishment of relative status positions, against the possibility of
future social action. Another basic function is in providing a code
(a Vocabulary') in which status relations can be simply and con-

cretely expressed. Such rituals may also provide conditions for ex-
ploration or assertion of changes in status, as Junod's Thonga
example of Christian handshaking indicates.
Of special interest in all this is the way in which some simple
physical actions involving the body as a whole are given symbolic
significance. Forms of greeting or parting in which a person lowers
himself, sets himself at a distance, or removes articles of clothing
indicate inferior status to other persons who do not behave in this
way, or do so only to a modified degree. The visible lowering,
distancing, stripping of the body - all acts which metaphorically
if not physically leave the individual relatively unprotected - em-
phasize by contrast the other person who is protected by height,
distance and covering. So social inferiority is expressed symbolically
by a simulacrum of physical defencelessness.
Analogies with animal behaviour here are clear. With animals
social identification is linked directly with personal security. The
kind of contact that is made, often in exploratory way, has a function
of establishing the freedom of the individual animal from immediate
danger, and in some cases of promotion of sexual co-operation. By
visual, aural, tactile and olfactory means animals establish identity
and social relationship. Human beings do not use the physical
range of the senses so widely or so acutely - smell is rarely used in
greeting behaviour, though sometimes it is discriminatory. But
the bodily means of establishing relationship by touching - with
hands, lips, noses - in such context is most marked, and would seem
directly relatable to animal behaviour.
But the analogies between man and other animals cannot be
pushed very far. Ethologists assume that ritualization in animals is
adaptive for the species, and broadly of the same general order for
each species. Greeting and parting rituals in human societies are of
very wide variety, and their adaptive function is discernible only
in broadest outline. Known aspects of human conceptualization of
time period give a quality or dimension to greeting and parting
behaviour which animals seem largely to lack. Anticipation of
greeting as distinct from generalized precautions against meeting
seems absent from animal behaviour. Ability of an animal to dif-
ferentiate length of time during which another will be absent seems
326 SYMBOLS
fairly limited, whereas human parting conceptualizes a future to

which a time scale can be assigned and by which ritual at parting can
be modified accordingly. (For instance, it would seem ridiculous to
farewell with the same intensity a member of the family who is
going round the corner to shop, as one who is going abroad for
several years.) Then, whereas for animals greeting rituals are vital
on personal security grounds, and in relation to territory and mating,
their parting rituals seem empirically scant and theoretically less
significant. Human greeting and parting rituals have developed
elaborations: use of intermediaries to provide introduction - animal
mothers do this only to limited extent; separation of formal from
informal roles, so that a person can be greeted in two very different
ways, even on the same overall occasion; adoption of representative
status, so that a person may be greeted or farewelled not simply for
his own sake but as a symbol of a group of others not present. Much
of this is completely alien to animal society. Then overwhelmingly,
the flexibility which human speech allows in greeting and parting,
and the associated development of concepts gives a completely
different dimension to the whole set of operations. As part of this
flexibility the specification of personal identity by use of personal
names is a marked characteristic of human greeting and parting
rituals, allowing a high degree of manipulation of the situation.
Finally, a feature which marks off human greeting and parting
rituals from those of animals is the relative ease with which they may
change. Apart from modifications associated with changes of fashion
in dress - men who do not wear hats cannot raise them, and must
bow or wave a hand instead-what appear to be deeply-rooted
modes of behaviour seem able to be changed quite rapidly. The
comparatively recent spread of the Western handshake is a case in
point. There is a paradox here. At any one period in any one society
conventional forms of greeting and parting tend to be fairly limited,
and to be regarded as of aesthetic and moral significance. Failure to
give the 'correct' signals tends to be criticized, even in children, as
insulting to the party affected. Lucy Mair has reported from her
experience among the Ganda thirty years or so ago that children
began to be taught phrases of greeting and farewell almost before
they could speak, and were drilled in the correct gestures-to
kneel and put their hands between those of the stranger. 'Refusal to
do this is one of the few reasons for which I have seen a child beaten'
(1934, 65).
Yet despite the moral sanctions in vogue for carrying out the
'correct' behaviour it is remarkable how easily such patterns have
altered. The more elaborate formal procedures of many African
and Asian societies - including the Ganda-have tended to be
abandoned in modern times as familiarity with Western patterns has
spread and as Western economic, educational and religious institu-
tions have affected traditional status alignments. I would put status
considerations at the core of the symbolism of greeting and parting
rituals.
Chapter 10
SYMBOLISM OF FLAGS
Like the symbolism in the manner of wearing the hair, and in be-
haviour at greeting and parting, the symbolism offlagslies primarily
in their display. But whereas hair is basically a personal symbol, a
flag is basically a social symbol. Not an attached, growing, shearable
part of a human personality, it is a manufactured object, often made
to a formula, detachable from maker or user and meant to show to
others rather than to be an ornament to oneself. And yet a flag may
be an object of strong sentiment, its symbolism may be deeply felt,
and as such it may serve as a symbol for the unity of a large body of
people. Now when it is said in Durkheimian fashion that a flag is a
symbol to which sentiments spontaneously attach themselves, and
that this intensifies social solidarity, what is meant by such state-
ments?
I begin with a piece of personal observation. A couple of years
ago, from my room overlooking the quadrangle of an eastern
university in the United States, I heard one day the sound of a
kettledrum in even taps, at the same time as I saw the statue of one
of the founding fathers of the university, a seated bronze figure,
being draped in a plastic robe. Then a procession came into view,
headed by a single male figure with head-band and long hair, slowly
carrying an uplifted American flag on a pole. He was followed by
about twenty young men, some carrying a large chair. To the sound
of the drum the procession advanced to the plinth of the statue, the
chair was set down, and a black student was seated temporarily in it.
Then the student was 'rescued' and the chair was set on fire. A
speech was given, inaudible to me and apparently to other people at
a distance - some lying on the grass reading, some talking with backs
turned. From later report the speech gave some explanation of these
events.
328
SYMBOLISM OF FLAGS 329
The interpretation was briefly that it was a ritual parody, con-

cerned with the agitation for freeing a member of the Black Panther
group who was then being held by the United States authorities
for trial. The parody was an illustration of the thesis that, as alleged
in various posters, the Fascists of the government had decided
that this particular man must die in the electric chair. The seated
bronze figure in robes was therefore meant to represent the judge in
court; the black student represented the prisoner who ought to have
been rescued; the chair which was burned represented the punish-
ment seat which ought to be destroyed. The whole event was de-
signed to dramatize the Black Panther trials, to decry the justice
of the United States courts, and to mobilize support for protest
movements against them. What now was the place of the United
States flag in this performance? It was clearly a symbol, intended to
serve as a hallmark of the proceedings, a leading feature, represent-
ing the court of the people of the United States. But the use of this
flag in parody raises some general issues about the symbolism of
flags which are worth discussing analytically from a more theoretical
sociological standpoint.
ETHNOGRAPHY OF FLAGS
There is some comparative ethnographic data about the nature and

use of flags, particularly in display, but little precise examination of
their functions.
Simple forms of flag have existed in many cultures. In the Pacific
region, for instance, among pagan Tikopia forty years ago a streamer
of white bark-cloth was used as a flag to celebrate the completion
of yam-planting in a sacred cultivation during the great seasonal
ritual cycle (Firth, 1967a, 185). According to Tikopia tradition the
early occupation of the neighbouring island of Anuta by a Tikopia
chief was signalized by his setting up a pennant of bark-cloth; when
Tongans later came there they took his pennant down and sub-
stituted their own (Firth, 1954, 121). In the first of these cases the
bark-cloth streamer, compared by Tikopia to European flags which
they had seen on vessels, was said to stimulate the growth of yams,
and to warn people to keep clear of the taboo area. In the second
case it was intended as a token of ownership. In northern Luzon a
330 SYMBOLS
flag of another kind, presumably combining both ritual and status

components, was the piece of bark-cloth dipped in the blood of a
newly-taken head and hung up in a house after a successful head-hunt
(DeRaedt, 1964, 315). But for flags to be fully developed a more
efficient textile in the form of woven cloth has been historically
necessary, and many ethnographical references concern calico or
similar material. Nearly a century ago Colonel Richard Dodge was
exploring among Sioux Indians when, crossing a fresh trail, he saw
on a hill a pole to the top of which was fastened a streamer of white
cloth. Under a small cairn of stones at the foot of the pole was a
bundle of sticks and a cotton cloth with totemic hieroglyphs, and a
pouch containing tobacco and corn. These things were a message
telling the number of men in the party, who they were and where
they were going and that they intended to eat and smoke in peaceful
company. The white streamer was to call attention to the message
and invite their friends to follow them (Dodge, 1883, 411-12). But
in small-scale societies with much face-to-face contact, the need for
flags as signals would be small.
While preceded in the West historically by insignia of various
kinds, such as the eagle standard of the Roman legions, carried
especially in war,* flags of cloth in modern form seem to have been
an Oriental invention, transmitted to Europe probably by the
Saracens. The function of flags in display is illustrated by countless
instances in descriptions of Oriental culture of their use in horse or
bull racing, in the travel of princes or high officials. Such flags often
had two important functions in addition to general ostentation and
attraction of attention: they might embody ritual values in their
colour - as in the yellow royal standard of an Emperor; and they
might have inscribed on them identificatory devices such as the
'cognomen' of a lord. Such specific marking of the individuality of
the person whose banners were carried bore a more direct relation
to social group structure than the inscriptions of the 'welcome' or
'happiness' character which were also very frequent. The Western
use of flags as heraldic devices is well known. An interesting parallel
was the use of the Mongol 'banners' which gave rise to a technical
term in English. As a flag the Mongol banner was a cloth bearing the
* The Latin vexillum, a military standard, from which allied English words have been
derived, is related to vehere, to carry.
SYMBOLISM OF F L A G S 331
emblem of a prince, and by English commentators the term 'Banner'
has been used to indicate the social group related to this-the
tribal following of the prince. By a further complication, as Latti-
more points out, the Manchus took over the Mongol concept, but
developed their Banner as a regimental formation, applying to mili-
tary but not to general political units (Lattimore, 1934, 146-52).
Apart from decoration and display, expression of social unit
solidarity, or specific individual identification, flags have often
served to mark out personal life crises in a more dynamic, even drama-
tic way. By followers of the Sapilada religion among the Igorot
of Luzon a white flag was formerly characteristic. When a young
couple got married such a flag was tied to a long pole set up in front
of their house. It was taken down on the third day of the wedding
ceremonies, then put up again after another three days. When the
cult was revived during the last war red flags were used as well as
white, possibly in imitation of Japanese troops (Eggan and Pacyaya,
1962,95-113). In Japan, so much was the use offlagsassociated with
holidays that as Embree records, national holidays were indicated
in the calendar by little symbols of crossed flags. But flags also
served sterner purposes. In Suye Mura flags were not only set up for
holidays, boys' ceremonies and completion of house framework -
they also marked funerals and memorial services for the dead. They
also indicated the drafting of young men into the army. Before the
war, when a youth was selected to serve as a soldier, a tall bamboo
was cut and stripped to a topknot of leaves. Below this leaf cluster
a national flag was fastened and the flagpole was erected in the house
yard. The flag was left in position while the son of the house was
away in the army, and those houses which had soldiers in training or
overseas could be told by the location of the flags (Embree, 1946,
94, 147, 149, 199, etc.). Differentiation of symbolic function was
indicated by the fact that while marking the absence of the soldier
was presumably a form of mourning, the use of the national flag and
not just any coloured banner, showed respect for the nation whose
cause he was serving.
The symbolism of status combined with an implicit comment on
marital relations is illustrated by M. G. Smith's description of a
Carriacou wedding. In this Caribbean ceremony, there is a fight
betweenflagbearers representing groom and bride, until the groom's
332 SYMBOLS
flag defeats the bride's flag and is crossed over it. Then the two
flags are hoisted above the house, the groom's again on top. The
symbolism here is obvious in one sense, but whether it represents
male dominance in actuality or only in wish-fulfilment is not clear.
The flags bear various devices, such as the Union Jack, and em-
broidered mottoes such as 'Long Life', 'Prosperity' or (a rather
depressing comment on the future of marriage) 'In God We Trust'
(Smith, 1962, 127-8).
F L A G S AS SIGNALS
Whatever be the theoretical argument about use of the terms sign,

symbol and signal, it is convenient operationally to separate the
use of flags as signals from their use as symbols. In the one case they
are intended primarily to convey information, in the other they
are meant to express ideas or emotions, often of quite complex
order.
A simple instance of a flag signal is when at a country cottage a
flag is put on a pole in the hedge at the roadway to attract the atten-
tion of travelling milkman or newspaperman. The message is -
milk or newspapers are wanted; it is imprecise, has minimal content
and almost no emotional loading. Moreover, the flag is non-specific.
Shape, colour and design are immaterial, or nearly so - the flag
should not be red. For an analogous signal in such a country lane, a
red flag means by convention that men are at work on the road ahead,
perhaps out of sight round a bend. Here too the signal is imprecise,
with little emotional loading beyond a mild anxiety until the reason
for the signal is made clear. But while the response to both signals is
left voluntary, that for milk or newspapers is a request, and that for
caution against men ahead is virtually a command. The sanctions for
disregard of each are of very different order, those of the latter being
definitely punitive. Within limits the size, shape and exact position of
the red flag of warning are irrelevant, but its colour is not. The red
fabric has been deliberately chosen because whatever may be its
physiological stimulus, by social convention red in such contexts of
road travel and in social movement generally is recognized as a
signal of danger (cf. the redflagflownby a vessel carrying explosives,
or hoisted on an artillery range when firing is in progress).
S Y M B O L I S M OF F L A G S 333
The complex use of flags as signals has been developed to a high
degree as a means of communication between vessels at sea, especially
since the eighteenth century. These maritime signals have used
combinations of the cardinal elements of shape, colour, pattern and
position to achieve a very elaborate code, which has attained the
status of an international convention. In a general way, simple
coloured flags have been used at sea as on land to convey stock types
of message. A yellow flag is a sign of quarantine, or of infectious
disease aboard; or more broadly, of a health condition needing
clarification and permission to proceed. A white flag in time of war
indicates a wish for a truce, or a surrender, or at least absence of
hostile intent.* But the international signalling code, dating from
1857, is much more specific. The yellow health flag is the letter Q in
the international code, whereas a yellow flag with a black ball in the
centre is the letter I; a flag with alternate yellow and blue stripes is
the letter G, one with blue cross on white field is the letter X - and
so on with different colours and designs for all the alphabet. Nearly
all these signal flags are rectangular, but a series of number pennants
are distinguished in part by their tapering shape. So messages may
be spelled out in detail over any visual distance by hoisting series
of such flags. Yet an alphabetical message is cumbrous to send and
read, so for economy small groups of flags have been given more
elaborate if still specific significance. So while yellow with black
ball in centre, and blue cross on white field mean independently flag
letters I and X, the two-letter group IX in combination, flown from
the halyards, means 'I have received serious damage in collision'. So
on for a large set of combinations - said to be 78,000 in all.
Here the meaning of the signal is to be obtained from the flag
in its structural context: the particular pattern, shape and colour
combination cannot be construed in isolation, but only in association
with other flags, according to their order and precise position on the
display field of the vessel. The significance of this structural dimen-
sion is seen by the manner in which the colour pattern flag signals
of this maritime international code can be translated into the
equivalent codes of semaphore signalling, and of morse. Semaphore
* In the troubles in Belfast in mid-1970 a man with a white flag emerged from a housing
block engaged in exchange of fire with the police, asking for medical attention for injured
people. After guarantees of mutual safety, this was arranged (Daily Telegraph, 26 June 1972.
Cf. Hulme, n.d., 25; also on use of black and of blood-red flags).
334 SYMBOLS
is essentially a visual mode of signalling; while flags have commonly

been used, their relative position, not their colour, is the critical
index. (Blue and white is the convention, but is not obligatory.)
Hence wooden or other arms can substitute for flags, as does a
central post for the human body which is the elementary reference
point and source of mobility in the original semaphore system. (If
flags are lacking a semaphore signaller can use his own arms alone.)
In this system it is the position of the arms and/or flags in angle to
the body, individually or in combination that gives the key to the
interpretation. So colour in context is translated into angle in con-
text. With morse the transformation is made from position to
sequence; the meaning is given by the differential flow of a series of
short and long pulses. These can be given by flags, waved rapidly
in a short arc or slowly in a long arc; but vision can be replaced by
sound or by inaudible electrical impulse. This possibility of trans-
formation underlines the contextual, structural nature of the code
significance of flags as signals.
Conventions of more widespread order allow the use of almost
any flag as a signal of limited kind by variance of its normal posi-
tion. By dipping a flag, that is lowering it slowly and raising it
smartly, a salute can be given in respect. A flag set at half-mast is
taken internationally as a sign of mourning. A flag lowered and not
flown again can be a sign of abdication of control - as in the cere-
monial ending of British rule when a colony becomes independent.
A flag flown upside down may be a sign of distress, and is commonly
so interpreted at sea. But the meaning of the signal must be read
according to circumstances. The British union flag (commonly
called Union Jack, though it is rarely flown from the jack staff of a
vessel) is often flown by ordinary citizens for special celebrations of
a national order, and sometimes in error appears upside down, with a
narrow instead of a broad stripe uppermost near the staff. This is
not interpreted as a sign of distress, even by those who realize the
inversion. (For special interpretations of inverted national flags see
later, p. 361.)
F L A G S AS SYMBOLS
Many flags in ordinary use must be regarded not primarily as signals

conveying information, but as symbols representing ideas or objects
of value, often in a very general and emotional manner. In maritime
history the white flag of surrender often seems to have meant much
more in value terms than a simple message of intention to stop
fighting and hand over the ship to the enemy. The black flag
popularly, if apocryphally, attributed to Spanish Main pirates stood
for a whole attitude of mind towards property and sovereignty,
and was not just a signal of aggression. The modern red flag of
revolution also represents a complex set of ideologies and be-
haviour.
In addition to their flag signals, the military establishments of
most countries have tended to use a considerable amount of flag
symbolism. 'Colours' carried on formal occasions by regiments of
many nations, and displaying their battle honours, stand for the
reputation and solidarity of the unit. In Britain, such symbolism is
illustrated specifically when the Sovereign or other dignitary pre-
sents new colours to a regiment. The use of ensigns by ships also
has symbolic function. In the Royal Navy the flag of an admiral is a
white square bearing a cross of St George in red. When flown by a
vice-admiral theflagbears one red ball in the upper canton next to the
staff, and when flown by a rear-admiral it has two red balls in the
upper and lower cantons next to the staff. As signals, such ensigns
denote the presence of an officer 'of flag rank* aboard, but they are
also status symbols with elaborate rights, privileges, duties and
rules behind them. The degree to which display of such flags goes
beyond mere signalling is shown by the ceremonial manner in
which they are hoisted and lowered, on occasion with appropriate
bugle calls, or other formal recognition.
Broadly we can distinguish symbols of situation, such as a white
flag of surrender, from symbols of status, such as an admiral's
flag, though they often tend to merge. But a more relevant distinc-
tion is between symbols of office status and symbols of personal
status. So, an admiral in the Royal Navy is entitled to his status
symbol of the flag of St George while he remains in that office, but
he must relinquish it when he retires to private life. But if he is of
armigerous family, he remains entitled to his personal heraldic
symbols as long as he lives. But even official and personal status
may merge in the use of symbols such as flags. Historically, a great
number of symbols of personal status have been supplied byflagsof
336 SYMBOLS
heraldry - banners, bannerets, banderoles, standards, oriflammes-

which displayed the particular devices of the nobles and knights on
occasions of collective assembly, especially at court or in war. As
signals they indicated the position of their owners and served as
markers or rallying points, but as symbols of personal status they
represented or evoked a complex set of sentiments relating to
prowess in battle, wealth in land, strength of followers. Yet many of
these flags also symbolized the official position of the bearer's lord,
as head of an important and powerful social unit. As such, a flag of
this kind represented a collectivity, not just a person.
NATIONAL FLAGS: THEIR COMPOSITION
The symbolic character of national flags illustrates most clearly the

sociological significance of the status theme. Consider first the major
features of their composition.
National flags vary greatly in ancestry, some being of very old
design, others being quite contemporary as a function of admission
to the United Nations. But all have been constructed very much in
terms of differentiating signs, not autonomously conceived, but
having much of their meaning from the contrast that each presents
to others of the same general class.*
All national flags except one so far (Nepal) are rectangular in
shape (in contrast to pennants for signalling and other uses). Colour
is a distinctive attribute of them all, and most are multi-coloured.
Historically, presumably due in part to the Muslim aversion to
depiction of the creatures of Allah, Islamic flags tended to be of
single colour without device, but many modern flags of Muslim
countries now bear coloured devices or are multi-coloured, f
* Detailed systematic information on national flags is given in many publications, for
example E. M. C. Barraclough, Flags of the World, 1959; British Admiralty, Flags of All
Nations, 1965. A valuable content analysis of colour combinations and devices, with some
examination of their symbolic meanings, has been made by a sociologist, Sasha R. Weitman
(1971) quite independently of my own inquiry for this chapter. Weitman also has discovered
the existence, for sociologists, of the Flag Research Centre which conducts studies of
'vexillology' with great energy, from Lexington, Massachusetts.
t Black (alleged to be the colour of vengeance) is supposed to have been the colour of
Mohammed's banner; it was adopted by the 'Abbasid Caliphs in the eighth century, and is
still used by Shi'ites, as in Iraq. By contrast, Omayyad Caliphs took white, and the Khawarij
took red (such a plain red flag has been kept to modern times by the Sultanate of Muscat and
Oman). But green, the colour adopted by the Fatimid dynasty, eventually became the most
Many national flags are distinguishable only by their specific
colour variations, their design of cross, vertical or horizontal stripes
being identical. So, the green, white and orange of Eire parallels the
green, white and red of Italy. The flag of Poland has a white hori-
zontal section above a red one, while that of Indonesia has the re-
verse. Denmark has a white cross on red, Sweden a yellow cross on
blue; Switzerland has a white cross inset in a red field, while, though
not a national flag, the International Red Cross has adopted the
reverse style. Western Germany has black, red, yellow in horizontal
bars, as against Ethiopia's green, yellow, red. Some national flags
have the colours identical, but in different arrangement: Belgium
shares the black, red, yellow with Western Germany, but in vertical
as against horizontal bars. The defining character of each flag is
provided by the uniqueness of the combination as part of a group of
related colour masses. What this means, inter alia, in these days of
modern mass media, is that only those flags with a distinctive device
- such as a pattern of stars, or a hammer and sickle - are capable of
being reproduced easily in black and white form, as in newspaper
cartoons. But while the colour range of national flags is considerable,
it does not cover the entire spectrum equally. Orange, indigo and
violet are much more rare than red and blue. Of the mixed colours,
green is popular, but brown seems to appear only on flags of Bhutan
and a few other small countries, and almost no use is made of special
shades of the 'kingfisher blue5 or 'shocking pink' type. White and
black, on the other hand, are commonly used as if they were colours.
In all this selection, historical factors are important, but so also is
clarity of perception.
The problem of a natural basis for colour selection has provoked
much argument. Forty years ago Radcliffe-Brown wrote of the
'stimulating dynamogenic power of sensations of redness' in explain-
ing the use of red paint by Andamanese (1922, 318), an idea to which
Lowie gave only guarded approval. Recently Levi-Strauss and
Charbonnier discussed the matter briefly in another context. Like
Radcliffe-Brown, Charbonnier pointed out that red is a source of
commonly recognized colour of Islam. (That a green flag in other contexts can have secular
meaning is shown by the custom of 'topping-out' in the building trade when the frame of a
building is finished. On completion of the new London Stock Exchange building, it was
reported that 'in accordance with tradition' a green flag was hoisted, with a sprig of evergreen;
for this ceremony the chairman of the Stock Exchange presided (Timesy 16 July 1969).)
338 SYMBOLS
physical and physiological excitement, and inferred that it is para-

doxical to choose red for 'Stop' signs, as in traffic lights. Levi-
Strauss commented that red might be a sign of heat and communica-
bility, just as green might be a chilling and rather venomous symbol.
But he argued that this underlying meaning was extremely weak -
not completely arbitrary but so to a large extent (Charbonnier, 1969,
116). In the symbolic context of national flags this 'dynamogenic'
significance of red can be broadly seen, but it is by no means uni-
versal (see later).
Sometimes the pattern of a national flag is intended to symbolize
linkage as well as differentiation. Flags of Australia and New Zealand,
for instance, bear the 'Union Jack' in an upper corner to indicate the
continuing connection with Britain and the Commonwealth, while
adopting individual arrangements of the stars of the Southern Cross
constellation to mark their autonomy. But the linkage may not
necessarily mean any acknowledgement of former political authority.
I was intrigued to see when I lived in Hawaii that the state flag bears
the 'Union Jack' in one corner - not that the British ever owned
Hawaii, but before it was annexed by the United States, when it was
an independent Polynesian kingdom, it had a friendly relation with
Britain, and marked this symbolically in the flag.* So also the cres-
cent and star on flags of Egypt, Libya, Malaysia, Pakistan indicates
Muslim relationship, though historically not all of them acknow-
ledged the suzerainty of the Ottoman Turks who originated this
symbol on a national scale. (Cf. also the Red Crescent substituted
for Red Cross by some Muslim nations, on the grounds that the
latter is a Christian symbol.)
GENERAL SYMBOLIC VALUE OF NATIONAL FLAGS

'Explaining' the symbolism of a national flag in terms of its history
is a common school exercise in some countries. But it is not so much
the individual histories that are of sociological interest as the ques-
tion of what values are attached to these pieces of coloured cloth, and
what social acts are related to these.
* In 1776 the new Continental Army of the American colonies had a flag of red and white
stripes with the British union flag in the canton - the Grand Union Flag, before independence
was sought. Not till 1777, a year after the Declaration of Independence, did Congress decide
on the Stars and Stripes, and no official flag of this pattern seems to have actually appeared
till 1783.
A key problem was referred to by Thomas Carlyle, when he noted
the small intrinsic value of such flags. 'Have I not myself known 500
living soldiers sabred into crows' meat, for a piece of glazed cotton
which they called their Flag, which, had you sold it at any market-
cross, would not have brought above three groscheri, he wrote in
Sartor Resartus, on 'Symbols'. But in sociological terms it was
Durkheim who made the classic statement which I quoted at the
beginning of this book. In connection with his analogy that the
Australian totem was the 'flag of the clan' he put forward a series of
propositions. It is a well-known law, he held, that sentiments aroused
in us by something spontaneously attach themselves (or are com-
municated) to the symbol which represents them. The idea of a thing
and the idea of its symbol are closely united in our minds. The
soldier who dies for his flag dies for his country, but as a matter of
fact, in his own consciousness, it is the flag that has first place. It
sometimes happens that this even directly determines action. Whether
one isolated standard remains in the hands of the enemy or not does
not determine the fate of the country, yet the soldier allows himself
to be killed to regain it. He loses sight of the fact that the flag is only
a sign, and that it has no value in itself, but only brings to mind the
reality that it represents; it is treated as if it were this reality itself
(1926, 2 1 9 - 2 0 ) .
This statement, classic though it be, needs some comment. As my
quotation from Carlyle has shown, the idea of the discrepancy
between the market or 'intrinsic' value of the flag and its symbolic
value was no novelty, though Durkheim's statement was more
neutral in tone. Durkheim's statement has its defects. Note the
superb arrogance of the remark that with the soldier who dies for his
flag it is as a matter of fact the flag that has first place in his con-
sciousness (or 'value recognition', depending on which meaning is
given to conscience). This is either a tautology or an unwarranted
assertion. Assuming there is historical basis for such an imaginary
illustration, the soldier may have been prepared to die 'for his flag'
deliberately as an act of leadership to induce others to follow and
capture the position, with all that may have implied; the flag may
have been quite secondary in his judgement of the military issues
involved. But Durkheim's statement calls attention to certain basic
elements in the symbolic use of flags. First, the flag, a specific
340 SYMBOLS
material object, is taken as the representative of a very general

object, a country, of abstract as well as of material character.
Secondly, the material symbol becomes in itself an object of senti-
ment, which is transferred from the object represented. Whether
such transference is 'spontaneous' as Durkheim argued, or has been
a much more complex process, including elements of indoctrination,
can be a matter for argument. In some countries quite overt con-
scious efforts have been made to impress such transference upon
children. But such transference of sentiment, or development of
sentiment for symbol as well as for original object, raises the ques-
tion of incongruity to which Carlyle and others have drawn atten-
tion. Thirdly, attitudes to the symbol are not merely intellectual and
emotional; they also tend to take shape in action. Whether attitude
to the symbol and attitude to that which it signifies can ever be com-
pletely separated may be an academic question; but certainly with
national flags the symbol is a common and highly important be-
havioural focus, often treated as an object of value in its own right.
And lastly, national flags tend to be placed in a special category.
Even more than regimental colours, school banners and other unit
symbols, national flags tend to be assigned a quality of special
reserve, removing them from the more sordid aspects of common
handling. They represent 'society' much more, in its broadest
political aspect. Use of the term 'desecration' for behaviour not
regarded as appropriate to handling of a national flag indicates how
the notion of reserve is related to the Durkheimian concept of the
sacred. A representative in the United States Congress has recalled
how reassuring it was to him as a child when the teachers opened the
day with a pledge of allegiance to the Flag and a prayer to the
Supreme Deity (Newsweek, 22 November 1971).
The discrepancy between attitudes which might be expected
towards simple objects taken alone and the same objects taken as
symbols seems to have been a continual matter of remark, even
among sophisticated sociological analysts, when they came to the
example of flags. Edward Sapir shared this view. He wrote: 'By
gradual extensions of meaning the terms symbol and symbolism
have come to include not merely such trivial objects and marks as
black balls, to indicate a negative attitude in voting, and stars and
daggers, to remind the reader that supplementary information is to
be found at the bottom of the page, but also more elaborate objects
and devices, such as the flags and signal lights, which are not ordin-
arily regarded as important in themselves but which point to ideas
and actions of great consequence to societies— .In the case of a
national flag or a beautiful poem, a symbolic expression which is
apparently one of mere reference is associated with repressed emo-
tional material of great importance to the ego' (1934). The interest of
this statement does not lie so much in the hint of latent wonder that
such mediocre objects should arouse such strong sentiments - which
one would expect to be a lay reaction rather than a professional one in
such a sensitive sociologist. It lies rather in the notion of the symbol
evoking emotional responses of a repressed kind. It is not clear just
what Sapir thought this repressed material to be in the case of flags.
But the statement is an echo of Sapir's psychoanalytical interests, and
a pointer to some later anthropological ideas about symbolism.
The Durkheimian sociological viewpoint about flags has been
repeated by many writers, including Boas, Linton and Levi-Strauss.
It is epitomized in the kind of statement made by J. J. Honigman
about the rituals of the United States Memorial Day - that the
emotions canalized towards flags, images or dead heroes intensify
solidarity (1959, 515).
QUALITIES OF FLAGS USED AS SYMBOLS
But what I have not found in sociological writings about the use of
flags as symbols is any systematic examination of why flags should
have attained such prominence. It is true that other national symbols
have been of great importance, especially national anthems and
national emblems. Indeed the government of India has been so
aware of this that a few years ago they issued a series of brochures
explaining the origin, meaning and use of these symbols of the Indian
nation. 'The National Flag, the National Anthem and the National
Emblem are the three symbols through which an independent
country proclaims its identity and sovereignty and, as such, they
command instantaneous respect and loyalty. In themselves, they
reflect the entire background, thought and culture of a nation/
This might be called an exercise in practical Durkheimianism.
342 SYMBOLS
Postage stamps, seals, coins also serve as national symbols of

importance.
But the special qualities offlagswhich make them prime choice for
symbolic use are availability and variability. Flags can be made of
any fabric. Bunting, a particular kind of woollen cloth, has become
conventional for many official flags because of its durability, but
cotton, silk, satin and man-made fibres have also been used; primi-
tive flags such as Tikopia pennants have been made from simple
bark-cloth. A flag is often cheap and can be hastily constructed. The
history of national flags has many examples of flags having been put
together from materials at hand by hard-pressed warriors or their
inventive ladies. When such tales pass into legend, as that of Betsy
Ross in the United States, this is another way of associating the
national symbol with the ordinary people. Flags are also capable of
being greatly varied in shape and design, by use of different patterns,
colour combinations and specific motifs. The adaptability of flags in
display means that their symbolic value can appear over a great range
of ritual occasions. Again, their basic characteristics can be recog-
nized at a distance, and they are as effective among illiterate as among
literate populations. Finally, simple actions with a flag can implicate
complex themes. Because a flag is cheap to make or buy, easy to
manipulate, observable by numbers of people at once, it is a prime
vehicle for conveying attitudes towards a social unit of which one is
a member, or expressing other sentiments. Hence there is oppor-
tunity for personal identification with the symbol which can give
added force to its use.
So, on occasions of national rejoicing, individual citizens fly
flags, waveflags,in ways which express their direct involvement with
the occasion and their social group.
TRADITIONAL SENTIMENTS TOWARDS NATIONAL FLAGS

AS I D E A S
Before examining behaviour towards concrete examples of national

flags, I consider briefly some attitudes traditionally expressed about
the idea of a flag in the abstract. Among the best indicators here are
nineteenth-century poets, among whom national enthusiasm rode
high.
It is a fair assumption that poets have expressed both their own
personal sentiments and those of a wider public in those verses which
have passed into the literary and educational heritage of the countries
whose national flags have been celebrated. In the English-speaking
world a parallelism can be clearly seen in British and American
writers, who have lauded their respective national symbols in what
have by now become hackneyed quotations. (The examples which
follow have been taken from common anthologies.) Take a couple
from Thomas Campbell, in the early nineteenth century:
Ye mariners of England
That guard our native seas
Whose flag has braved a thousand years
The battle and the breeze...
(Ye Mariners of England)
This is a touching sentiment - if one ignores the fact that the flag
commemorated has not been the same flag for the whole period. But
one cannot ignore the passage of history so easily when reading
With Freedom's lion-banner
Britannia rules the waves . . .
(Ode to the Germans)
Nor is the claim advanced by Rudyard Kipling to be conceded with-
out a pang nowadays:
Never was isle so little, never was sea so lone,
But over the scud and the palm-trees an English Flag was flown.
(The English Flag)
American sentiments have been analogous, while tailored to the
circumstances of the development of the United States. There is the
invocation of the flag in Whittier's Barbara Frietchie and in Emer-
son's hymn sung at the completion of Concord Monument (with the
classic phrase of the 'shot heard round the world'). There is Fitz-
Greene Halleck's poem named The American Flag:
Forever float that standard sheet!
Where breathes the foe but falls before us
With Freedom's soil beneath our feet,
And Freedom's banner streaming o'er us.
344 SYMBOLS
There is Francis Scott Key's Star-Spangled Banner (of 1814):
Praise the Power that hath made and preserved us a nation!

Then conquer we must when our cause it is just,
And this be our motto, 'In God is our trust!'
And the star-spangled banner in triumph shall wave
O'er the land of the free and the home of the brave.
To these patriotic sentiments, which seemed very justifiable in the

nineteenth century, may be added the more sardonic contemporary
comment of James Russell Lowell:
We've a war and a debt, an' a flag; an' ef this

Ain't to be inderpendent, why, whut on airth is?
{Biglow Papers, 2nd series, no. 4)
Strong sentiment has often been expressed in the United States on

the use of the flag in schools, to promote csane Americanism'.
Evidence given to a 1937 Senate hearing was in favour of furnishing
a national flag by the government to the funeral of each man and
woman who was formerly in the armed services, and giving the flag
afterwards to the kin. It was argued that 'it is a patriotic thing to
provide that the symbol of the unity of our country, the national
flag, should be placed upon the casket of every deceased man who
has served The expense involved would be an investment in
patriotism which will bring back huge dividends to our beloved
Nation in the future.' It was pointed out in comment that a flag of the
specified size and materials would cost $5.47, and that the numbers
of people entitled would be considerable - so the 'dividend' was
judged to be too low (S. 947, Hearings of Committee on Military
Affairs of U.S. Senate, Government Printing Office, Washington,
1937). In the regional field, some massive publications dealing with
local regimental flags of the United States reinforce the attitudes
towards the national flag by describing how the flags were a 'beacon
star', consecrated by the blood of the state's noblest sons, carried
amid toilsome marches, bloody strife, heroic achievements and
so on. (At the same time the records often speak of the difficulty of
preserving such flags, though money had been set aside for the pur-
pose, with the implication that people's interest in the symbols had
tended to wane when once the crisis period had passed.)
Though the British literature is less voluminous, analogous views
emerge. I cite here simply an instance, like some American ones, of
manufacture of a national flag during siege. At the siege of Chitral of
1895, a British union flag was made of red and white cotton and a
blue turban, from the design found on a Navy-Cut tobacco tin. This
design was in fact incorrect - it showed white stripes of equal width,
whereas the union flag has unequal white stripes. Nevertheless, while
technically 'wrong' the symbol performed its function. The leader of
the force, Sir George Robertson, reported that the flag was an
inspiration to all in the fort - 'it helped us greatly. It cheered our
hearts and stiffened our backs' (M'Millan, n.d., 215-16).
The more patriotic of these American and British effusions now
have a pathetic quality, with the shrinking and change of status of
the British Empire to Commonwealth, and the loss of much of
Britain's international political role; and the fading of the American
dream of linking the ideals of liberty, democracy and equality with
the brute facts of economic and political power and material pros-
perity. Yet, tattered like historic flags, the ideals still remain, and
with them their symbols. Both in Britain and the United States the
national flag remains a symbol of high importance for a great
majority of the citizens. Moreover, it remains also a symbol for
citizens of other countries - who can be moved to action against it
or treat it with respect, according to circumstances. Indeed, the
expression 'showing the flag' is still used metaphorically for a
demonstration of the power, influence and reputation of a country
abroad - the expression deriving from the practice of naval vessels
entering foreign waters on visits of courtesy or aggressive warning,
flying their country's ensign. Most standard literature about flags
contains some rather sententious statements of positive sentiment
about those of major political significance. The Encyclopaedia Ameri-
cana, for instance, asserts straightforwardly: 'A national flag is the
symbol of the nation.... A nation's flag is reverenced by its citizens
and respected by the citizens of friendly nations. An insult to the
flag is considered as an insult to the nation' (1968, 308).
Analogous treatment can be found in a range of literature dealing
with other national flags. For example, the centenary of the attain-
M
346 SYMBOLS
ment of Italian unity was celebrated by a survey of the history of the

Italian flag - 'il simbolo piu significativo della Nazione: la Bandiera
Italiana'. Allusion to the colours which have become incorporated in
the Italian tricolour was traced back to Dante's Purgatorio (the
whole analysis denying by implication the thesis sometimes ad-
vanced that the Italian tricolour is merely a Napoleonic adaptation of
the French tricolour). Linked with the spirit of independence his-
torically demonstrated, the flag is 'the living symbol of the country'
(Colangeli, 1965). The history of the national flag of the West
German Republic (Friedl, 1968, 23 etc.) shows poignant association
of the colours of the flag with ideas of democracy. An analytical
treatment of this whole matter begins with the blunt statement
'symbols are realities', and considers Paul Tillich's distinction be-
tween 'discursive' and 'representative' symbols. Adopting this latter
idea, of the symbol as indicating what is not immediately grasped in
historical consciousness, Friedl traces the history of the German
flag in terms of its meaning for German nationhood. He cites a
declaration of Scharnhorst of 1808 - 'The war must be pursued for
the freeing of Germany by Germans. This must be expressed in the
banners of the local militia.' Friedl points out that after the capitula-
tion of Germany in May 1945 there was no more German flag.
German ships had to fly an international flag (the code flag for C,
blue-white-red-white-blue - colours which were not politically sig-
nificant for the German people). Four years later the flag of the
former Weimar Republic was decided upon by the Federal Republic
as West Germany's new flag. The colours were laid down by law,
and the spokesman said: 'The tradition of black-red-gold is Unity
and Freedom; or perhaps I may say better - Unity in Freedom. This
flag shall serve us as a symbol, so that the idea of freedom, the idea of
personal freedom shall be one of the bases of our future State '
As this latter example demonstrates, it is not just a particular form
of flag which has symbolic value: it may be primarily the colour
combination, if it be sufficiently distinctive. So, the black-red-gold of
Germany, the blue-white-red tricolour of France, tend to stand for
national interests. So also with certain distinctive designs. In black and
white drawing, the Union 'Jack' of Great Britain, the Stars and
Stripes of the United States, the Soviet hammer and sickle, are easily
recognizable and can have symbolic force.
C R E A T I O N OF NEW N A T I O N A L FLAGS
It is a mark of the power of flag symbols that the process of their

creation still continues. A new national flag is a potent symbol, a
highly condensed focus of sentiment which emphasizes the indepen-
dence of the newly created political unit. One need not subscribe to
the cynical view that along with a flag and an anthem the symbols of
nationhood all too often include a prestigeful but money-losing
national airline. But it is significant that the entry of the many new
states to the United Nations has always been accompanied by a
cheaper symbolization - the display of their new flags. Even a
change in type of government may be symbolized by abandonment
of the old flag and creation of a different one. This happened at the
French Revolution and at the Russian Revolution. A recent instance
is the making of a new Cambodian flag. When towards the end of
1970 Cambodia became a republic under the government of Lieu-
tenant-General Lon Nol, the republican flag was created from red,
white and blue cloth. On a blue ground a red rectangle was sewn in
the upper left-hand corner. A white cut-out image of the main
temple of Angkor was placed on the red, and three white stars on the
blue. The Western press identified the red, white and blue in terms
of the French tricolour or the American flag. The temple image
represented the most famous national monument. The three stars
seemed more difficult to place: according to report the principal of
the school where the prototype flag was made said they represented
the north, centre and south of Cambodia; but no one else had a clear
idea - except some people who assumed they stood for the three
stars on General Lon Nol's uniform collar. What did catch the
imagination of the press reporters was the alleged requirement that
according to Cambodian tradition a new flag must be made by
Vestal virgins' - who seemed to be most appropriately represented
by the pupils of a girls' high school. Nine students accordingly
sewed the flag, sitting in a circle on the floor, under the supervision
of their principal, and following the instructions of the Cambodian
parliament. While the sewing proceeded a Buddhist altar was set up,
silver bowls of fruit were placed at its foot, and tall candles lit. The
flag was furled and unfurled by a bevy of young girls specially
dressed, while monks chanted, and showered jasmine blossoms over
348 SYMBOLS
the girls, who had also knelt with lotus flowers. The rite was said to
have been improvised for the occasion (so much for the 'tradition'!).
But clearly it involved Buddhist elements of chant, lotus symbolism
and offering, with concepts of purity and more general symbolic
behaviour of kneeling and other signs of respect. It was both a
dedication of the new national flag to public state uses, and also a
respectful recognition of its importance as a sacred symbol (New
York Times, 9 October 1970).
The creation of a national flag is so much part of modern political
symbolism of nation-making that a people may even proceed to the
recognition of a flag before they attain nationhood. So in the move-
ment of East Pakistan towards autonomy as Bangladesh, the Bengal
Nation, when militant students and workers began to demand com-
plete independence, the green, red and gold flag of Bangladesh was
unveiled in anticipation (New York Times, 29 March 1971). Many
Blacks in the United States have adopted their own liberation flag -
black, with three horizontal bars of red, black and green. (These
were colours popularized by the late Marcus Garvey.) According to
the chairman of the Pan-African Congress, U.S.A., the red bar
stands for the blood shed by black people to achieve the Black
Nation. The black bar represents the 'black race' which, though
fallen on evil days, will rise again to take its rightful place in the
world. The green bar is symbolic of land and nationhood, for land is
the basis of power and freedom. The creation of this flag is regarded
as based on the black American's 'passionate belief that the true
meanings of the nation's patriotic symbols and ceremonies have
never been extended to include him'. The broad truth of this state-
ment can hardly be questioned. But one may surmise that the crea-
tion of this flag, and the recognition of a 'black national anthem' as
well, are perhaps symbols of compensation, substitutes for more
tangible independence in terms of political and economic power
(L. F. Palmer, Jr, 1970).*
An interesting light is thrown on the relation between symbolic
* This black-green-red flag of 'Negro nationalism' was raised over the coffin of one of the
convicts killed in retaking the Attica prison in September, 1971. It was also waved in an
African-American Parade in Harlem in the same month. In Vietnam a 'Black Powerflag*dis-
played 'Black Unity' in black letters at the top, and a black fist in the middle, on a red ground.
In August 1971, the Black-owned and Black-operated Amsterdam News of New York
adopted a combination of red and green ink with its black print. As the editor stated: 'This
was our symbolic way of expressing typographically and in our format our commitment to
creation and considerations of expediency by the history of the
wheel symbol in the national flag of India. It has been said that the
chakra-dhvaja or the Wheel Flag of India is the symbol of her
civilization as evolved through the ages. When the national leaders
of India were fired with the spirit of asserting the country's indepen-
dence they naturally felt the need of a national flag. Various designs
were used from 1906 onwards, commonly with red, yellow and
green. The Home Rule movement of 1917 devised a flag which
incorporated the Union Jack, since they were aiming at local
autonomy, not complete independence. 'The people, however, did
not take kindly to it'! Finally, partly on Mahatma Gandhi's sugges-
tion, a tricolour was accepted, of saffron above, green below, with
white in the middle, in which was set a spinning-wheel sign of navy-
blue. With the coming of Independence in 1947 it was felt necessary
to modify the spinning wheel for the sake of simplicity, to avoid
directional confusion; if the symbol appeared identically on both
sides of the flag it would be pointing different ways. So, on a resolu-
tion moved by Pandit Sri Jawaharlal Nehru, a simple wheel was
substituted. He described the flag as 'a flag of freedom and symbol of
freedom'. He pointed out that in the white previously there was the
chakra which symbolized the common man in India, which sym-
bolized the masses of people, which symbolized their industry. But
in considering a simplified revision, minds went to the wheel of the
Asokan column. 'That wheel is a symbol of India's ancient culture;
it is a symbol of the many things that India had stood for throughout
the ages....' The theme of the symbolic significance of the wheel in
Indian history was elaborated by others also, and the wheel was
described as the Wheel of the Law of Dharma, denoting motion,
representing life, symbolizing righteousness, etc. (Vasudeva, 1964;
Y. G. Naik, 1957). So here a contemporary symbol, the spinning
wheel, associated with the mass of the peasantry, and with Gandhi's
simplicist but powerfully appealing mode of life, was converted to
the historic symbol, the wheel of Asoka, associated with much more
complex religious sentiment, at a deeper level. Though motivated by
expediency, the translation gave an increment of symbolic power.
The circumstances of the decision on the symbolism of the Indian
what it means to be black in this country today*. (See for example New York Times, 13-19
September 1971; 17 November 1971.) For action in schools see later.
35° SYMBOLS
flag appear quite clear. But as with the national flag of the United
States, establishing the origin of some other national flags has been
a subject of some controversy, often of political significance. This
seems to have been the case in Argentina, where the history of the
precise colours of the national flag has been open to debate, and
where a military circular was issued on the antecedents of the
'national symbols' (no. 635/41; cf. Luis Canepa, 1943, 1953).
MORAL S I G N I F I C A N C E O F F L A G COLOURS
Running all through discussion of the symbolism of national flags

is the vague but persistent theme that not only the design but also
the colours of the particular flag of a country may have more than an
accidental significance. It is often held that irrespective of the cir-
cumstances of their selection to compose the national flag they
represent objects or qualities of significance to the nation, or to
mankind more generally. These are always of a positive, morally
approved character.* I give only a few examples. The flag of the
Republic of Eire has green and orange, to represent the two major
religious components of its population, with a band of white, repre-
senting peace, in between. The white and gold of the Vatican City
are likewise symbolic of peace; here the avoidance of 'strong'
colours has led to a breach of the heraldic rule of not ordinarily
juxtaposing 'metals' - the white being equivalent to silver. In the
flag of Afghanistan the royal colour, red, is stated to symbolize wars
fought against aggressors, while the bordering strips of black and
green stand for the dark past and the bright future of the country.
Such moral overtones of colour symbolism appear especially in the
newly-created African states. In the flag of Kenya, which attained
independence only after a bitter struggle, black is said to stand for
the people, red for their blood, green for their land, and white for
unity. In the Dahomey flag, which has a common African colour
combination of green, red and yellow, the green stands for palm
groves, the yellow is for the northern savannah, and the red repre-
sents the struggle for freedom. But in Chad, which has a tricolour
with blue not green, the blue stands for agriculture, the yellow for its
* The statistical survey by Sasha Weitman (1971) examines the meanings officially assigned
to the major colours occurring in about seventy national flags.
deserts, and red for unity between the regions. Of the Nigerian flag,
decided upon after a competition, the designer himself spoke of the
green as a symbol of agriculture and the white as the sign of amity
and peace. Some of these attributions are of a fairly obviously
descriptive kind - red for blood, green for agriculture. Others are
more abstract, as red for unity, overlapping with white in this
respect. In some of these newer flags the symbolic meaning has been
conceived in terms of a combination of colours, such as green and
yellow representing the different regions of the country and their
juxtaposition, national unity. But while there is a common thread
running through many of these colour symbolisms, there is no clear
predictive value to be seen in the use of any particular colour.
The arbitrary nature of flag colour symbolism appears even more
clearly with national flags of long-standing where usually there is no
official interpretation of meaning by colour. Among the many
accounts of the origin and development of the national flag of the
United States, some severely scholarly, others more popular and a
few apocryphal, there are various statements of the positive moral
values attaching to this symbol. In one straightforward eulogy: 'The
three colours used in our flag, of which we lustily shout our approval
when we sing " Columbia's the gem of the ocean - three cheers for
the red, white and blue" - mean something. The red signifies valour,
the white purity and the blue truth The flag is the symbol of our
national sovereignty and of the union of the states, and the birth,
progress and growth of the nation are told upon its folds. There is
perhaps no national flag more beautiful, and none other can be so
beautiful in the eyes of a true American.'
N A T I O N A L F L A G S AS SYMBOLS OF S T A T U S A N D POWER
So far the poetic statements, the speeches, the histories of these

national symbols all seem to speak the same language - and rather
surprisingly, to speak the same language as the sociologists and
anthropologists. Carlyle, Durkheim, Sapir, Nehru, Honigman all
emphasize the significance of flags as reinforcing solidarity, having
power to evoke sentiment and lead to self-sacrificing action. Such
moral attitudes evidently persist in contemporary society.
National flags are prominently displayed on national holidays and
352 SYMBOLS
special occasions of public rejoicing, by ordinary citizens as well as

by public offices. In some countries government offices regularly fly
the national flag - a useful mark of identification - and as a token of
loyalty, some private establishments may also fly the flag. In the
United States such enthusiastic use of the national flag is common.
The Chicago Tribune, a newspaper long noted for its strident
Americanism (it describes itself as 'The American Paper for Ameri-
cans') bears the device of a United States flag waving (now in
colour) over its headline, on every issue. Every day, for nearly a
decade this newspaper published a photograph of 'Today's Flag'
flying outside a commercial office or private residence. Towards the
end of 1970 it had been calculated that 3,000 such photographs had
so far been published, of 2,300 separate flags, and that the items of
print, placed end to end, would be more than 1,000 ft high! This
kind of patriotic obsession may have no direct parallel elsewhere, but
the attitudes it expresses can be found in many other contexts and
countries. The value of a national flag as symbol is evidenced by the
discreet use of it in political campaigning in many countries. Its
solemn associations are brought home by its use as a shroud for a
coffin or bier of a soldier killed in the course of military duty, or as a
memorial sign in a military cemetery, or as borne in procession on a
national memorial day.
The significance of the flag as a symbol of solidarity and national
virtues may be indoctrinated through the educational system. In the
United States, Flag Day Assembly and similar patriotic occasions
intensify attitudes of respect inculcated by daily pledges of allegiance
to the flag. Teachers and athletic coaches may have the obligation
specifically laid upon them to instruct their pupils appropriately.
(A handbook of the Inter-Suburban Association of a number of
schools near Chicago 'tells the coach to instruct players in the
proper method of paying respect to the colors' - Chicago Tribune,
10 October 1970). Unofficial pressures such as neighbours' gossip or
remarks of business colleagues can help to reinforce conformity to
the conventional attitudes towards the national symbol.*
But there is another side to the picture. A national flag, as a sym-
* Cf. the New Yorker cartoon of a business executive addressing an employee: 'Naturally,
X, the company doesn't care whether its employees have little flags on their desks or not. It's
purely a voluntary thing. We just wondered why you happen to be the only person here who
hasn't got one!' (24 October 1970).
bol of state power, has its composition and internal use in the society
controlled by authority. By legal prescription the shape, size, cloth
and design of most national flags are controlled. A citizen's error in
such detail may be ignored by the state or left to the informal con-
trol of public opinion in which some citizens score off others by
display of expertise. But there is sometimes an uneasy balance
between official authority and private initiative in the public display
of national flags. When a foreign dignitary arrives in the more
democratic countries the display of flags is left primarily to the
spontaneous feelings of the citizens, over and above the official
decking of buildings. The quantity of flags shown is some measure
of the popular respect and support the visitor enjoys. (When the
Emperor Hirohito visited the Netherlands during his world tour the
lack of flags in display was remarked upon in the press, as indicative
of the resentment still felt by many Dutch people against the war
conduct of the Japanese - New York Times, 9 October 1971.) Under
a dictatorship, however, explicit or implicit orders for display of
national flags means that the amount of display is a direct reflection
of the official view of the occasion.
Externally, display of a national flag on vessels of war, on aircraft
of a national airline, on embassies and other representative buildings,
symbolizes the status of the country concerned. Hence the 'protocol'
involved in the use of national flags at the United Nations. (When
the People's Republic of China became a member of the United
Nations great care was taken to obtain the correct placing of the flag
at the front of the U.N. building - after ascertaining the govern-
ment's wishes it was inserted at C (for China) and not at P.) This
external status is linked with power, in that the presence of the flag
symbol implies a direct possibility of action to enforce compliance
with the policy of the state concerned. Internationally, a national
flag can be used as a warning of armed force to follow, and in effect
as a substitute for armed force. But a national flag may have a rather
different function in symbolizing rights conceded by international
convention, without reference to powers of enforcement. The 'flags
of convenience' adopted by some shipping lines are legal symbols,
internationally recognized as entitling the carrier, if properly en-
dowed, to very definite rights and privileges pertaining to the
country whose flag is carried, irrespective of the national status of
354 SYMBOLS
the owner of the vessels. Countries such as Liberia and Panama, who
are prepared to register vessels under their jurisdiction but who are
less concerned than many other countries about regulation of crew
wages and employment conditions, standards of seaworthiness and
taxation levels, present operating advantage to shipping companies
which some regard as more than offsetting protection from their
own national flag. The system has been shown to have advantages
too at a national level for a country such as the United States, which
wishes for strategic reasons to keep in being a considerable merchant
fleet but finds difficulty in manning it on American standards.*
Such a device of transfer of status involves much more than
merely flying an appropriate national flag; complex legal instruments
are necessary to secure and implement the status which the flag
symbol indicates. In other conditions the use of a national flag may
be a more formal device to indicate not so much temporary transfer
of sovereignty as a claim to privilege or an avoidance of responsi-
bility, with the implied consent of the countries concerned. In the
field of international relations some national vessels engaged on
United Nations business, as F.A.O. fishing research vessels, may fly
a United Nations flag in addition to or as substitute for their own
flag. A practice, which has as much an element of disclaimer as of
claim in it, is that reported of submarines in the Adriatic manned by
Chinese crews but flying the Albanian flag - though Albania was not
known to possess submarines.
In these examples the national flag is a symbol of the political
entity, and it is displayed officially as a symbol of the status, often
also of the power, of the nation concerned. By implication too it is
expected that the symbol will receive the respect ordinarily paid to
more substantial tokens of the nation's strength. So, the expression
'showing the flag', even if metaphorical, means staking a claim to
national interest. So when a Soviet ship opened a passenger run
between South-East Asia and Australia, this new development was
interpreted by foreign diplomats as a matter of Russian policy -
'another Russian step to show her flag in a new area'. When the
United States acquired a naval base on Bahrain this was treated as
* For brief discussion of economic and political questions raised by 'flags of convenience',
see Roger Eglin, Observer, 7 March 1971; Michael Baily, Times, 3 June 1972. Official dis-
regard of a 'flag of convenience' by Cuba was reported in tension with the United States
(New York Times, 19 December 1971).
'a flag-showing operation to manifest United States interests in the
area'. That a physical flag was flown in such circumstances was
almost irrelevant; in each case the political implications of the
physical operations were indicated by the verbal use of the symbol of
the flag. So when the President or Prime Minister of a country makes
an official visit abroad he may be said to be 'showing the flag' merely
by advancing his country's interests.
S T A T U S SYMBOLS I N R E V E R S E
But the symbolism of national flags also operates in reverse. If the

flag represents its country's interests, it can be used as a concrete
instrument of protest against these interests or against conduct
related to them. The symbolic significance of the national flag is
dramatically demonstrated by treatment in which it is physically
maltreated and defiled. When in mid-1970 the British Conservative
government moved towards the sale of arms to South Africa, amid
the many protests from the Commonwealth was one from Malaysian
students in Kuala Lumpur. The British union flag was laid down in
a street and a motor-car forced to drive over it; the flag was later
burnt, together with an effigy of the British Prime Minister. When
President Nixon announced that he was sending American combat
troops into Cambodia, demonstrations against the United States in
many foreign cities included the burning of the Stars and Stripes.
A leftist Filipino demonstration against the government of Presi-
dent Marcos was also against the influence of the United States in
the Philippines, and ended by burning an American flag. Later in
1970 the United States Embassy in Sierra Leone was stoned and the
American flag ripped from its staff, in a demonstration against the
alleged involvement of a member of the embassy unit in a plot to
overthrow the Sierra Leone government. In September 1971 mili-
tant Jewish youths in protest against the Soviet Union's refusal to let
Jews migrate to Israel stole a Soviet flag from an international dis-
play, diverting police by burning a Soviet plastic flag a short dis-
tance away. In Amsterdam Dutch demonstrators burned Japanese
paper flags in protest against the visit of the Emperor Hirohito. In
New York Nationalist Chinese burned a Chinese People's Republic
flag in front of the hotel of the first communist delegation from
356 SYMBOLS
Peking to the United Nations. (For some details see: Times, 24 July
1970; New York Times, 25 May 1970; 15 September 1971; Time
Maga^ne, 18 May 1970; Chicago Tribune, 16 October 1970.) The
major point about such demonstrations is that while they show
resentment against a national power, they substitute action against
the symbol for action against more integral constituents of the
nation such as the person of its citizens. So, an American report on
the Filipino incident of burning the Stars and Stripes described the
demonstration as 'noisy but peaceful' since no one was assaulted. It
is noteworthy that such incidents have now come to be accepted as
part of the modern way of life by the powers whose flag symbols are
maltreated. Assuming that formal diplomatic protest is lodged, there
nevertheless seems to be tacit recognition that in the prevailing
climate of criticism of economic and political power as exercised by
the larger nations, these had better tolerate abuse of their symbols
than threaten reprisals. In this sense the symbol is treated as a surro-
gate, on which moral and physical force can be allowed to spend itself
with minimal harm.*
FLAGS IN D E C O R A T I O N AND DESECRATION
The symbolism of national flags is not completely covered by the

conventional treatment of essayists, politicians and sociologists. It
is not correct that in all cases a national flag is reverenced by its
citizens, and intensifies their solidarity. There are discordant notes.
Not only do Malaysians run automobiles over the Union Jack or
* But the formal destruction of aflagof political significance has more gravity. Historically,
flags captured from an enemy in battle have been preserved as trophies - which in itself
indicated a kind of respect for the symbols so acquired. But in the Jacobite struggle against the
English Hanoverian kings, after the battle of Culloden, which effectively ended the High-
land resistance in 1746, the rebel standards captured were treated with ignominy. They were
carried by the chief hangman of Edinburgh and by chimneysweeps, with an escort, and laid
in the dust, while a proclamation was read explaining why they were to be burnt by the public
hangman. Each standard was then held over the flames, while the senior herald named the
Scottish clan that had marched behind it to battle. This was deliberate disrespect, with sym-
bolic modes of contempt: trailing in the dust; handling by executioners and men associated
with black soot; consumption by fire. Yet when standards were preserved in what might seem
superficially to be analogous conditions of disrespect, the interpretation was very different.
The standards of the Appin Stewarts and of Clan Chattan were torn from their poles by
clan members, who wrapped them round their bodies, ran from the battlefield and saved
them. While intentional contact with dirt and fire was regarded as degrading to the symbols,
contact with sweaty human bodies was regarded as meritorious, or at least as not defiling (see
John Prebble, Culloden, Pelican, 1967, 93, 99, 190-1).
Filipinos burn the Stars and Stripes - some British citizens use the
image of their national flag on their shopping bags, and some United
States citizens have worn images of their own flag on their trousers.
Such acts imply lack of reverence, and some of them have indubitably
provoked dissension rather than solidarity among citizens. Has
something gone wrong with the argument? If the flag is a rallying
point for the unity of all members of the society, how is it there can
be internal as well as external maltreatment of it? Clearly the stock
Durkheimian view can only be regarded as a statement of a majority
position or a general tendency; the image of society as something
that is revered, with symbols respected and held sacred, is incom-
plete. I would suggest indeed that while the national flag is generally
recognized as the symbol of the society, the sentiment postulated
towards the symbol may be more complex and have a much wider
range of variation than the Durkheimian position allows.
There can be variation in intensity of appreciation of such national
symbols. In both Britain and the United States, for instance, there
are opinions which are highly critical of the state of the society, and
which protest against what they regard as its oppressive character.
But on the whole in the British scene there has been much less focus
on the national flag as an instrument of protest. In Britain the
national flag as symbol of the society is held in looser association
than in the United States. I would argue that one of the mechanisms
by which the United States authorities have helped to identify sym-
bol and society very closely, the enactment of laws which set out in
detail how the national flag should be treated and how offenders
should be punished, has to some extent backfired. In Britain an
alternative symbol, the Monarchy, has been available to take some of
the pressure of protest off the flag.
In Britain until recently the Union Jack was used in sober fashion
by ordinary citizens, for holiday and memorial display and occasional
discreet commercial advertisement. The commercial aspect has con-
tinued, as when one of our major airlines has displayed the flag on
books of matches, ashtrays and cushions. The Apple and Pear
Development Council illustrate the union flag in colour on paper
bags to hold fruit, with the encouraging injunction 'Pick an English
pear'. In recent years, with the development of freer English fashion
and the influence of Carnaby Street, the Union Jack has been seen
358 SYMBOLS
not only on shopping bags but also on some items of clothing. In

mid-1970 newspapers had illustrations of exuberant women with
dresses, hats and umbrellas made from Union Jacks cheering a Pro-
testant parade in Belfast; and of hostesses for a world cycling
championship in Leicester with the design of the Union Jack on the
fronts of their skirts and their umbrellas. (See The Times, 14 July
1970; 7 August 1970.) In none of these cases was the use of the
national symbol regarded as an offence against the law, though it
might have been classed as an offence against taste - vulgar but not
illegal.
In the United States the situation has been very different. Towards
the end of the nineteenth century the Daughters of the American
Revolution advocated a law providing penalties for mutilation,
commercial or political use of the national flag, and many states
passed such laws. Most states also decreed that the flag should be
displayed at all schoolhouses while school was in session and that
pupils should pledge allegiance to the flag. Some states also pro-
hibited the display of any red or black flag along with the national
flag in any procession or public hall. The Congress of the United
States was slower to act, but in 1942 a federal flag code was enacted,
becoming public law no. 623. This specified in great detail how the
national flag was to be displayed and respected in a variety of cir-
cumstances. {Inter alia, it was clear that the national flag should
never be used on a receptacle, as the Union Jack has been used on
British shopping bags.) By a new flag law, which received the
presidential signature in 1968, a substantial fine or imprisonment for
a year, or both, could be applied for publicly mutilating, defacing,
defiling, burning or trampling the national flag. In this way the
American national flag has been treated as a sacred symbol much
more overtly than has the British national flag.
When the flag of another country is insulted the act is usually
brief and violent - trampling, tearing or burning being the most
common methods of showing disapproval. Similar methods may be
used by protesters who are citizens of the country concerned. When
the independence of Bangladesh was being first proclaimed people
at a mass rally in (then) East Pakistan burned the Pakistan flag as a
symbol of their attitude. In Israel, a group of ultra-Orthodox Jews,
Neturei Karta, 'Guardians of the Gate', are reported to hate the
Zionist state, to refuse to carry identity cards issued by the govern-
ment, and to burn Israeli flags. In the United States in the spring of
1970 a 17-year-old schoolgirl was sentenced to six months in jail in
Massachusetts for burning an American flag in a high-school pro-
test; on appeal she was given a lecture on patriotism by the judge and
did penance by carrying a large American flag on a three-mile
march through the streets of Cambridge. Later in the year two people
were arrested for burning an American flag outside a hotel where
Vice-President Agnew was making a speech. In riots in Georgia, a
peaceful parade culminated in the burning of the Georgia state flag
torn from the city hall staff, and of an American flag taken from a
nearby funeral home {New York Post, 22 September 1971; New York
Times, 29 May 1971; 3 April 1970; 13 May 1970; Chicago Tribune,
7 October 1970; Time Maga^ne, 30 November 1970). But internal
protest in the United States has sometimes taken more personal and
more subtle forms. Various cases have been reported in the Press
over the last few years of men arrested, charged or penalized for
using the American flag on their clothing or in other idiosyncratic
ways. One wore the flag over his shoulders, another wore a shirt
patterned like the flag, another had an American flag upside down
on the right leg of his trousers, and still another had a flag sewn to
his trouser-seat. Others again were charged with 'contemptuous
display' because they had painted their automobiles with stars and
stripes of appropriate colour.
On the other hand, while various federal and state laws can be
invoked against 'defilement' or 'desecration' of the national flag, it
has not always been clear exactly what those terms mean, and
whether they can always be applied with certainty to some specific
cases. When characters in the anti-Establishment rock musical
'Hair' played with an American national flag on stage, a New York
State law official who investigated held that while the conduct of the
actors was at times irreverent, there was no violation of the law.
(This opinion was hotly contested by the president of the U.S. Flag
Foundation, a body formed in 1889 for the purpose of honouring
the American flag and preserving it from desecration.) An American
magazine once asked the question: 'Does it desecrate the American
flag to sit on it?' The answer, in mock solemnity, was - not pre-
sumably when the sitter is a famous film-star. This problem arose
360 SYMBOLS
because some young men, charged in court with having had a picnic
on the American flag, defended themselves by producing a photo-
graph of the actress clad rather scantily in some arrangement of the
Stars and Stripes. The judge, inquiring rhetorically: 'Is she glamour-
izing the flag or desecrating it?', dismissed the case. A New York
City Transit Authority patrolman, who had been charged with
wearing an unauthorized emblem on his uniform, was found not
guilty on the grounds that this was done out of patriotism, and that
President Nixon himself had supported similar action in another
case, writing: 'We need to encourage Americans in pride of country'.
From such incidents it may be inferred that associating the national
flag with the person in unorthodox ways may be interpreted as
decoration or desecration depending on what may be thought as to
be the wearer's motivation.* In this context it is significant that the
action under scrutiny had a public character - wearing or using the
symbol so that it could be seen generally. Whether for patriotic
decoration or for protest, the symbol is expressly utilized as a
medium of communication of an attitude towards the society it
represents.
Other protest methods of handling national flags may be even
more open to ambiguity in that it may be hard to say whether they
are acts of desecration or respectful usage to express personal views.
In the 'hard hat backlash' of May 1970 in New York, when massed
building-construction workers demonstrated with Stars and Stripes
against the frequent anti-war demonstrations, there was no doubt:
they were carrying the national flag in defence of what they regarded
as the national interest. The national flag painted on their construc-
tion helmets was taken as positive criticism and fully permissible
(New York Times, 16, 21 and 25 May 1970; Time Maga^ne, 25 May
1970; 1 June 1970). But how about the man who flew the American
flag upside down and at half-staff in front of his house, with the
flagpole draped in black, in protest against the conviction of an Army
lieutenant for murder of civilians in Vietnam? The protester's argu-
ment was that such a conviction was not in the best national interest;
it was 'killing our flag'. The local police did not accept this attitude,
* For detail, see Honolulu Star-Bulletin, 2 September 1968; New York Times, 1 February
1970; 24 and 25 February 1970; 16 April 1970; Chicago Tribune, 3 December 1970; Ithaca
Journal, 14 May 1970; Time Magazine, 2 November 1970; The Times, 25 March 1971.
saying: 'You can't put the flag that way; you're not in distress/
There is also the case of an American housewife who flew the
national flag upside down to protest against the involvement of the
country in the Vietnam war, and who was arrested and charged with
violating a state law which prohibited showing 'contempt, either by
word or act, upon the flag'. The judge held that an act of turning a
flag upside down did not indicate a dishonourment or defilement of
the flag, and that what was being expressed as a personal opinion of
distress was a legitimate use of the flag. Yet when a group of high
school students petitioned their authorities to allow the school
American flag to be lowered to half-staff as a token of mourning for
the notorious shooting of Kent State University students, the judge
in this case refused the request. His judgement was that to fly the
national flag at half-staff was to commemorate the death of a per-
sonage of national or state standing or of a contributor to the local
community service. It should not be so flown to involve 'an expres-
sion of a political concept' (see New York Times^ 18 March 1970;
2 June 1970; Chicago Tribune^ 31 March 1971). On the other hand,
a group of demonstrators who flew the United States flag upside
down from the head of the Statue of Liberty for two days, as a
distress signal against the Vietnam war, were allowed to leave with
impunity in compliance with a court order for their eviction {Time
Maga{iney 4 January 1972).
Why these differences of treatment? To some extent they were due
to differences in state law, and in judicial interpretation. But broadly
they indicate that a symbol is essentially not an object but a relation-
ship; that with acceptance of the object goes also acceptance of
notions of its 'normal' or 'proper' use. Hanging out or waving -
even wearing - of a national flag symbol in normal 'correct' position
is a socially acceptable act, irrespective of whether the motivation
that inspires it is loyalty and respect or a scornful concession to
popular opinion. But to put the symbol in an unusual context causes
confusion, and focuses attention on motivation. Yet the situation is
ambiguous because most of the evidence about motivation often comes
only from the act of flag display itself. Hence the variation in judicial
and other public interpretations of 'contempt' towards the symbol -
leaving aside the debatable issue of the right of 'free speech' which
may be involved in personal use of a national flag. The 'distress
362 SYMBOLS
issue is especially interesting. The well-known conventional way of

expressing 'distress' at sea is to fly aflagupside down. In the language
of flag signals the effect is to link disturbance in the order of the
symbol with disturbance in the order of the vessel concerned. But
the signal refers to disturbance in the physical order - severe damage
to structure or equipment - or possibly to such disturbance in the
social order as to hamper the functioning of the physical order -
sickness or death of crew, mutiny. Any idea of attitude of mind of
the signaller is inferential, and on the whole irrelevant to the
immediate purpose. But the display of a national flag upside down by
an individual citizen belongs to a different order of communication:
it refers primarily to his attitude of mind, and not to physical con-
ditions. Hence the analogy claimed is not exact, and the authorities
have some basis for intervention on this ground, quite apart from
the question of the right of the citizen to use a national symbol on a
personal basis.*
A parallel problem of ambiguity arises when the national flag is
taken as a basis for modification in pattern or colour. Some of these
modifications have been patently of a political order. In mid-1970
a man in upstate New York was sentenced to ten days in jail on a
charge of defiling the American flag; he had hung a replica of the
flag, with a 'peace' symbol in the blue field, from his front porch.
On the other hand, a policeman arrested a university student of
Dayton, Ohio, with a United States flag bearing green 'ecology'
stripes instead of the normal blue, and charged him with desecrating
the flag. When other students complained the head police official
(described by one of his aides as 'unbelievably flexible') said 'we
were wrong' and returned the banner. {Post Standard, Syracuse,
3 June 1970; Newsweek, 9 November 1970.) Modifications in the
name of art have caused even more difficulty. A painting 'Three
Flags' by Jasper Johns in 1958, a simple superimposition of Stars
and Stripes, has apparently caused no concern, and so also with wall
* Note that the issue tends to be affected by the design of the flag. British people not un-
commonly fly the Union Jack upside down out of ignorance on occasions of public rejoicing,
since broad and narrow white stripes are not easily distinguished. But from the position of
the stars it is very clear when the United States flag is flown the wrong way up. (Yet the
Town Council of Boston, England, proudly displayed an American flag for nine years,
presented by an American television company and seen by an estimated 5,000 American
tourists, which had white stars on a red, not a blue ground, without its being discovered as an
error-New York Times, 29 November 1970.)
rugs depicting the American flag with more stripes and fewer stars.
But a New York art dealer who exhibited works of art including a
phallic symbol wrapped in the American flag was convicted of
violating a state law against flag desecration and sentenced to sixty
days in jail or $500 fine. The State Court of Appeals upheld this
conviction, ruling that the right of free speech did not permit the use
of the American flag in 'dishonourable' ways as a form of protest.
This particular construction was deemed 'insulting', and several
artists who displayed work in a 'People's Flag Show' in support of
the dealer were arrested and charged with desecrating the flag on
their own account (see New Yorker', 14 May 1970; New York Times,
19 February, 10 April and 15 November 1970). The majority
judicial view on appeal was that while the art dealer may have had a
'sincere ideological' point of view he should have found other ways
to express it, and whether he thought so or not, 'a reasonable man'
would consider the wrapping of a phallic symbol with the flag an act
of dishonour. Yet a minority dissenting opinion, that of the Chief
Judge, held that the challenged 'constructions' were in effect only
political cartoons, and that they had been singled out for prosecution
only because of their political message.
This political element seems to be the nub of the issue in such
cases of alleged desecration of the flag in the American scene. The
possibility of aesthetic values, personal sentiment, freedom of speech
are all admitted, but the flag is not regarded as the appropriate
symbol for expressing them if its use can be given a political inter-
pretation which runs counter to established norms. This is one of the
dilemmas of democracy: the national flag is a symbol for the society;
but it is a symbol for the society in its political as well as its cultural
dimension. As such, it can serve as a focus for negative as well as
positive sentiment for the political actions of the society. So a ques-
tion of practical import is: how far can the established organs of
society tolerate private idiosyncratic use of national symbols such as
the flag to promote ideas which run counter to immediate national
policy as expressed by the governing authorities?* In the con-
* This issue was raised in another form in December 1971 by a decision of Newark, New
Jersey Board of Education to permit the hanging of the 'Black liberation flag' of red, black and
green (cf., above) in every school classroom containing a majority of black students, for
example in all but a very few schools. The decision seemed legal, but was strongly resisted by
local White citizens and by responsible Black authorities as divisive in nature, likely to arouse
3<$4 SYMBOLS
servative view, if flag as symbol represents society, it should express

consensus, not dissent.
F L A G AS SYMBOL OF D I S S E N T
I can now return at a more theoretical level to the incident described

at the beginning of this chapter, where the American flag was used to
introduce a parody of a courthouse scene. The flag here was not
desecrated - it was carried solemnly in procession, and it wTas the
'victim's' chair that was burned. But the flag was not displayed as a
symbols of solidarity, focusing the sentiments of participants and
observers in loyalty towards their society. It was used as the
representative sign of one of the basic institutions of the society, and
so as a symbol of the society itself. But it was used as part of a
criticism of the society, as a symbol to mobilize protest against what
were regarded by the participants as the injustices of their society.
It was used in apparent respect, but in a context of irony which
inverted its political meaning. On the other hand, while some of the
potential audience seemed impressed, others ignored the proceedings.
They reacted neither favourably towards the demonstration which
used the flag in protest, nor unfavourably to challenge this unortho-
dox, critical use of the national symbol. They behaved as a kind of
non-Durkheimian body, uninfluenced by the use made of one of
their most prestigious national symbols.
From the data I have presented, drawn from the raw noted
observation of many scattered instances and not from the introspec-
tion of sociologists, it is clear that there are no monolithic sentiments
towards flags as national symbols in a modern politically differen-
tiated society. Attitudes towards the symbol can be very flexible,
within the structure of authority as well as in the general public
body. Durkheim's idea of the essential unity of society and symbol is
meaningful only at a very general level of abstraction. I have taken
other partisan activity. The sponsor of the resolution argued that it would stimulate Black
consciousness and pride without lessening allegiance to the American flag; but critics held
that it was an unwarranted introduction of political symbolism into the educational system.
(New York Times, 2,3 and 5 December 1971; Time Magazine, 13 December 1971). The issue
was high-lighted by an earlier attempt in a Florida school to use a Confederate battle flag
pattern as a school banner (New York Times, 14 October 1971) which outside observers saw
as equally subversive.
the United States flag as a prime example because superficially in the
overt sacralization of its national symbol that country has gone
further than many others, and so invited more direct challenge by
those of its citizens who wished to protest. The United States posi-
tion in this respect may be a reflection of several circumstances: the
initial revolutionary character of the society; its development by
federation of a large number of discrete units; and its rapid absorp-
tion of many immigrants from diverse linguistic, cultural and
national sources. All this would make sense of concentration on some
easily identifiable symbols of national unity, and protection of them
from abuse. The United States flag has been described as 'a protec-
tive symbol' but it has also needed protection.
The Durkheimian type of proposition about national flag as focus
for sentiment about society, with sacred character, is valid to only a
limited degree. The 'sentiment5 has usually been interpreted as
positive, involving consensus. But in writing about Australian
aboriginal society, with totem as 'flag of the clan', Durkheim omitted
the political dimension. For totemic belief and behaviour the assump-
tion of positive sentiment and consensus was reasonable, though not
unchallengeable. But to cite as a parallel the national flag, or flag of
a military unit, was to ignore the significance of the power com-
ponent. A national flag of a modern state is an officially defined
symbol, not simply a symbol of informal public choice or traditional
development. As such, its 'sacredness' is an officially imputed quality.
Hence 'desecration' becomes a legal matter, to be judged in the light
of official pronouncements, either codified or expressed in the com-
mon law.
What is at stake in most of these cases of unorthodox use of a
national flag is the issue of legitimacy. The national symbol is mal-
treated as an instrument of protest against authority, to deny the
legitimacy of what that authority has done. It may express dissent
from a single type of action - such as involvement in overseas con-
flict - or it may represent basic disagreement with the structure and
ideology of the society as at present constituted. But in essence the
national symbol is manipulated to assert moral value over existing
power value. (That some of the issues may have been misconceived
is beside the point.)
Such critical use of national flags as symbols bears also upon the
366 SYMBOLS
Durkheimian concept of the sacred. Durkheim has argued that

sacred things evoke respect, and are separated from profane things
by an opposition which may become antagonism. This division of
the world into two domains, as Durkheim has put it, has been shown
to be inadequate. In an analysis of Australian aboriginal concepts,
W. E. H. Stanner has shown (1967, 217-40) that a concept of an
intermediate sphere is needed, the 'mundane', the affairs of the
everyday world in which the sacred is not regarded as antagonistic
but ignored. But if Durkheim's sphere of the profane needs weaken-
ing at one side it needs strengthening at the other. In the use of
national flags as symbols, the sphere of the profane is shown in the
United States by the interdicted areas of display - commercial adver-
tisement or political ideology. The sphere of the mundane (in
S tanner's terms) is shown in Britain by the use of the Union Jack on
dresses or umbrellas - they treat the flag neither with reverence nor
despite. But the intentional destruction of his national flag by a pro-
tester is profaning it in a more intensified way than Durkheim meant.
To Durkheim, the sacred thing was par excellence that which the
profane should not touch and cannot touch with impunity. It is the
profane which tends to be punished or damaged by the contact - as
a person who touches a sacred relic may be stricken by blindness. But
when a Union Jack or a Stars and Stripes is burned, it is the flag not
the protester who receives the damage; the sacred is desacralized.
And it is the profane, not the sacred, which carries the moral value
charge. This is very different from what Durkheim noted as intro-
ducing the sacred temporarily to the profane as a means of stimulat-
ing its powers - as throwing stones into a sacred pool to rouse the
immanent powers or whipping a sacred image to wake it up to
perform its duty. The national flag is not symbolically stimulated,
not symbolically ignored, not symbolically polluted, but symboli-
cally violated. The sacred representation is not just contaminated by
careless contact with non-sacred things; it is actively treated in
symbolic fashion as a counter-instrument to what it represents. In
the general ritual field the nearest analogy to such symbolism of
destructive behaviour is the Black Mass, the significance of which is
the inversion of customary procedures with the sacred. So also with
the burning of national flags by political dissidents; the flag stands
for the political authority of the Establishment, and its destruction
has definite symbolic value, of integrating effect for the group con-
cerned.*
Durkheim's use of the term profane incorporated two notions,
that of common or secular, and that of defiling or contaminating.
But the concept includes also the notion of violating, in contrast to
the inviolate. I do not propose another category of phenomena to
describe this field of action, but I think it deserves attention. In
essence it is an inversion of the sacred, using the symbol in a ritual
manner, but in a contrary sense to normal usage, and attaching anti-
thetical values to the referent, in a positive moral way. Durkheim
and most other commentators have seen the significance of national
flags as symbols in serving as rallying points for social solidarity, as
foci and even as generators of sentiments of a positive kind. But a
symbol is a double-edged instrument. When as is the case with
nationalflags,it is associated with a structure of power and authority,
it becomes an officially defined representation, with the possibility of
dissent from the values it is intended to convey. The symbol still
stands for the society, or for the controlling power in society, but
the 'sentiment' component is diversified, and the controlling power
may take steps to try and correct this accordingly. So a society does
not merely choose its symbols; it also may choose the level of reaction
it will endure against them. This operation of a feed-back principle
indicates a complexity in behaviour towards flag symbols which I
have not seen examined hitherto in systematic analysis.
* It has been reported {New York Review of Books, 22 April 1971) that 'underground'
groups in the United States may start their clandestine meetings by burning the American
flag, with powerful emotional reaction.
Chapter n
SYMBOLISM IN GIVING
AND GETTING
As we walked down one evening from a little Umbrian hill town to

the local railway station my wife gathered some wild-flowers, of
which she is very fond. While we waited at the station we passed the
time of day with the station master's wife, who was tending her
garden, and praised her flowers. Just before the train arrived she
came up and smilingly presented my wife with a bunch - unusual
behaviour to a couple of complete strangers, passing tourists. As an
act of courtesy, a pure gift, it touched us deeply. How far was it a
symbolic act?
In ordinary social and economic life in industrial society there is a
marked disjunction between two major types of behaviour: that of
market operations, of buying and selling; and that of presentation, of
making gift of goods or services. These are sharply opposed, struc-
turally, materially and morally. A market sale is a two-way transac-
tion, a gift is one-way. A market sale is for business, impersonal,
hoping for profit, competitive, contractual; a gift is friendly, per-
sonal, not seeking profit, contributory, free. In both spheres one
seeks to hand over what is judged appropriate by the other party.
But in the market the giver reckons the acceptable by its reciproca-
tion; in the gift he reckons it by the recipient's satisfaction. In the
market sphere behaviour is seen as sensible and careful with re-
sources, while gifts are apt to be thought emotionally dictated and
wasteful. In the social sphere outside the market, the market be-
haviour is often regarded as greedy, sordid; while gift-making is
held to be generous and is highly praised. These are the conven-
tional stereotypes.
Yet as anthropologists and sociologists have realized since the
work of Thurnwald, Malinowski and Mauss about fifty years ago,
368
S Y M B O L I S M IN G I V I N G A N D G E T T I N G 369
the opposition between these two spheres of transaction is by no

means so clear-cut. Empirically, in market operations buyer and
seller do not always try to exact the last shred of profit. They may
use definitely personal trading links, traditionally of long-standing in
some businesses. Even the conventions of haggling may involve
courtesies of an elaborate social order.* In applying provisions of a
contract strong views often exist as to what is ethical, with criticism
for a man who is 'too sharp' in taking advantage of the law, and
praise for one who is 'fair-minded'. Conversely, the implications of
a gift are rarely exhausted by a single transfer; the notion of recipro-
city is often near the surface. If there is little calculation of profit to
be got from exchange of gifts there is much of estimation of
equivalences. Malinowski demonstrated from analysis of the Tro-
briand kula how the elaborate ritualized gifts of these Melanesian
valuables were based on a firm code of obligation to make counter-
gifts of at least equal value. And he acknowledged a minor criticism
from Marcel Mauss, who pointed out that even what Malinowski had
called the 'pure gift' from Trobriand husband to his wife was in
effect a return for sexual and other domestic services. In more general
context Malinowski was fond of asserting that 'reciprocity is the
basis of social organization'. Stimulated by such findings, Mauss
raised discussion of the gift to a more abstract plane by formulating
three neat rules of obligation: the obligation to give; the obligation
to receive; and the obligation to repay. Many others, including Karl
Polanyi, Barry Schwartz, Marshall Sahlins, Michel Panoff, as well as
myself, have taken up these themes and showed how in societies
lacking overt market institutions, customs of gift-exchange have
served to facilitate transfer of goods and services, as well as to sym-
bolize social relations, f
Identification of relevant process has been carried beyond the
traditional field of the primitive. Classical societies of Greece, Rome,
India have been shown to have recognized the implications of the
gift. As Mauss quoted the Edda to the effect that a gift always looks for
recompense, so Peter Blau cited Aristotle to the effect that all men or
* See for example the account by T. F. Mitchell (1957) of the language of buying and selling
in Cyrenaican shops and markets.
t For a recent study of ceremonial exchange among New Guinea 'big-men* see A. Strathern,
1971.
370 SYMBOLS
most men wish what is noble but choose what is profitable; and
while it is noble to render a service not with an eye to receiving one
in return, it is profitable to receive one (Blau, 1964, 88). It was sug-
gested by Mauss but demonstrated by M. I. Finley (1962, passim)
that antiquity indeed knew a period when gift-exchange was the
dominant mode of transaction. In modern times it has often been
pointed out that the tradition of 'one good turn deserves another'
underlies many of our social institutions, from sending of Christmas
cards to wedding invitations. Manning Nash has even gone so far as
to say that middle-class social life in the United States approximates
to the reciprocal-exchange systems of the Melanesian archipelago;
that among families of equal status, dinner invitations in number and
quality have all the essentials of exchange equivalents; and that if the
balance gets too badly disturbed there may be some show of aggres-
sion or the invitations may cease (Nash, 1966, 31).
In an even broader approach George Homans and Peter Blau
have sought to show, not just that giving is a part of exchange but
that the structure of social relations in areas right outside the market
field is an exchange structure. Homans has seen social behaviour as
an economy, with interaction between persons as an exchange of
non-material as well as material goods, and a pay-off in terms of cost
and reward. Blau is much more concerned with the polity of social
life, the status and power differentials which are the object of recipro-
cal social behaviour. (Cf. Ehrmann, 1970, for a striking analysis in
this field.)
In all this the theory of the gift has been brought into line with the
theory of market exchange. The gift relationship appears as a kind of
market relationship, in that a kind of cost-benefit analysis applies but
in different operational channels - less open to alternative offering,
a minimum of haggling, and an almost explicit admission of in-
tangible, non-material elements into the exchange situation. From
different angles, both Mauss and Homans have indicated this. Mauss
has stated (1925, 34) that the market is a human phenomenon
familiar to every known society. Primitive societies do not lack the
'economic market' - though their exchange patterns are different
from ours. And Homans has argued that his theory of social inter-
action as exchange is quite compatible with 'the special conditions of
exchange that economics chooses to confine itself to' (1961, 79). He
SYMBOLISM IN GIVING AND GETTING 37I
has said in his charmingly blunt style that he is out to rehabilitate the
'economic man', and that by broadening the range of values of
'economic man' he gets a greater realism into his propositions; that
the new economic man is just 'plain man'.
In the terms of such a theoretical approach the gift of flowers to
my wife, which I mentioned earlier, can be regarded as not just a
simple expression of goodwill to a stranger, but as an item in an
on-going series of social exchanges. My wife and I made the effort to
initiate a social contact. We used the occasion for a mild exercise in
Italian; and we diverted ourselves during the wait for a train - in
a very minor way we obtained social benefit from the encounter. But
so also did the station master's wife - she had momentary relief from
the fatigue of gardening; a fleeting acquaintance with the world out-
side to give food for gossip; and praise of her flowers to support her
self-esteem. Her present of a bunch of flowers could be seen then as
not just a courtesy but also as an acknowledgement of social service.
One can even introduce a principle of balanced opposition into the
analysis - her gift of garden flowers was the response of culture
and nurture to nature and casual exploitation.
Yet while one may have much sympathy with this type of inter-
pretation the monolithic exchange theory seems inadequate. One
need not accept the postulate of Karl Polanyi and George Dalton,
that the distinction between market behaviour and gift or gift-
exchange behaviour is so radical that ordinary concepts of economics
are useless in interpretation of 'primitive' economic systems (cf.
Firth, 1972). But the patterned differences between market process
and the passage of gifts are considerable. Among the criteria generally
involved are: a wider field of alternatives in choice in market rela-
tions; much less personalized ties as a rule in buying and selling
than in gift; a different range of sanctions for reciprocity, with much
more specific legal sanctions in the market;* and much more closely
specified equivalents in market exchange.
* Legal sanctions have not necessarily been absent from the field of the gift. In Britain
formerly it was an implied condition that the gift of an engagement ring was returnable in the
event of the marriage not taking place, unless it was the man himself, the giver, who refused
to marry. By an Act of January 1971, however, gifts made to each other by engaged persons
will normally be returnable only if they were made on the specific assumption that there would
be a marriage. A purely personal present such as an engagement ring is therefore not recover-
able - though conventional wedding presents are, since they are thought of as made on the
condition that the wedding takes place (see The Times, 6 June 1970).
372 SYMBOLS
C O N C E P T OF THE G I F T
The notion of the gift is complex, in several ways. The basic element
of a gift is an outgoing from the self, but this implies an input to
another. In judging the implications of a gift then, the effect upon
the other has to be taken in conjunction with the effect upon the self.
So, as F. G. Bailey and others have emphasized (Bailey, 1971,
23-4) all gifts, like all exchanges, have both competitive and co-
operative elements. A gift commonly involves the transfer of a
material object, or the performance of a service over time which
involves the displacement of material objects. But it is also com-
monly regarded as implying some immaterial quality, of positive
sentiment, of goodwill. In many conditions the material good or
service is regarded as not only the token but as even the measure of
the sentiment. Gifts are often judged not primarily as a contribution
to resources, but as an index to the attitude of the giver. In this
sense that which is seen, the gift, is a symbol of what is unseen, the
concepts and emotions of the giver. But it may not be a reflection of
a simple state: the giver may feel he is acting under pressure and
react later accordingly; he may be using the gift as a 'sprat to catch a
mackerel', to benefit by a larger counter-gift; he may be focusing on
the social rather than on the material aspects of the transfer and
looking to intangible benefits; he may be concerned with his own
reputation and only marginally with the recipient; he may envisage
his gift primarily as a statement of a moral position. The donor may
even see the gift as just an expression of himself in a general bene-
factor role. Max Beerbohm's remark that mankind is divisible into
two great classes, hosts and guests, might be translated into terms of
donors and recipients, and be regarded as valid in some people's
conceptualization.
By definition a gift is freely made. Any notion of duress is alien
to the concept. As Blackstone put it in his Commentaries 200 years
ago: 'gifts are always gratuitous; grants are upon some considera-
tion or equivalent'. A gift is a voluntary, non-contractual transfer of
property without any valuable consideration - in particular, without
any material equivalent being stipulated.
Each of these criteria, however, is capable of further interpreta-
tion. The notion of gratuitous transfer may be used as a disguise for
SYMBOLISM IN GIVING AND GETTING 373
what is in effect a contributory factor in an exchange. Examples of

such eccentric so-called 'gifts' are the extra merchandise offered to
housewives when they buy So-and-so's soap or cereal. Premiums of
this kind have now become a vital part of the market philosophy in
such fields, and in economic terms have considerable aggregate
value - an annual value of £200 million in Britain alone in the last
few years has been given as an estimate.* According to market
research findings - not necessarily the last word on the subject — the
operative word which attracts the housewife, the only word which
advertisers find really matters, is the word 'free' and its equivalents.
A well-known airline has enticed its customers with: 'Begin your
trip to Hawaii aboard a . . . Fan-Jet. There are orchids on every tray;
complimentary champagne and steak dinner. For every lady, a
beach bag - a gift from ' Yet obviously in such a purely com-
mercial relationship the notion of such items being 'free' gifts can
be taken in only a special, fantasy sense; they are part of the service
for which a market price is paid. But it is the notion of gift that is
thought to bait the market hook. My argument at this stage is not
that a gift is not free, but that the illusion of freedom is used to label
a market service as a gift because the idea of voluntaristic transfer
is attractive.
But the categorization of voluntariness depends upon the sanc-
tions. A transfer of property without equivalent, if done as a result
of force, is not a gift: if done under compulsion from a properly
constituted authority it is confiscation; if otherwise it is robbery, t
In the case of taxation the sanction of force may be remote, but in
any case the one-way transfer can be regarded as being ultimately
compensated for by diffuse benefits.
In some of the elaborate series of transactions involved in market
relationships the criterion of freedom in the gift is sometimes used as
a demarcator. Where in the taxation of business firms, certain kinds
of business expenses are deductible but contributions to political
campaign funds are not, the business company may sometimes
represent such contributions as actual ordinary company expenses
for services rendered to the company. Effort is made to avoid the
* John Davis (1972) has estimated the economic magnitude of some other categories of
gifts in the United Kingdom.
t For relation between gift, taxation and robbery in nineteenth-century African conditions
see Speke, 1912, 107-62.
374 SYMBOLS
concept 'gift', which would not be tax deductible, in favour of

expense, on grounds that the payment was not voluntary and free
because it was necessary to meet a service rendered. (Empirically, it
is merely a substitution of long-term putative benefit for short-term
calculated benefit, with a wider choice for making the payment in
the former case.) In a reverse type of instance, a claim to have made a
transfer of property by gift is made in order to divest oneself of
responsibility for taxation. In countries such as the United States,
where genuine gifts of a charitable order can attract tax deductions
of considerable magnitude, a person with a high income has actually
been able to save money by giving away property to charity. But
questions of interpretation have arisen in respect of the 'family
foundations' in which rich men or women endowed members of their
families with shares of their business. In the ensuing argument it has
emerged that the essential feature of true gift in such cases has been a
complete relinquishment of rights in the property, including right of
future disposal. In discussions about the establishment of such
family foundations as a mode of tax avoidance one study commented
that such foundations 'lack the finality which characterizes a true
parting with property'. But even with final relinquishment of
interest in that which is given on the part of the donor, the frame-
work of society may impose restrictions on freedom of gift, as I
show later.
The ordinary concept of gift carries with it an association of
positive moral value. This is linked with the notion that together
with the transfer of the material object or service an element of the
self is also offered - not in the rather crude sense put forward by
Mauss, of a detachment of part of the donor's personality, which can
injure or be injured thereby, but in the more abstract sense of a
commitment of a personal kind to the implications of the action.
A gift, a relinquishment of rights to another, is a commitment of the
self to the interests of another. It is this emphasis which at first
sight receives primary attention in the classics of our literature which
enshrine and express so many of our values.
The New Testament is outspoken: 'It is more blessed to give than
to receive', wrote St Paul (Acts, 20: 35). 'Freely ye have received,
freely give', wrote St Matthew (10: 8). And as a high-minded
Christian moralist, Wordsworth rammed the lesson home: 'Give
all thou canst; High Heaven rejects the lore, of nicely calculated
less or more'-in the Ecclesiastical Sonnets -'Tax Not the Royal
Saint5. But perhaps the best example of poetic expression in this
line comes from a confessed pagan, Walt Whitman: 'Behold I do
not give lectures or a little charity - When I give, I give myself
(Song of Myself).
But commitment of the self is not necessarily without thought.
A text in the Douai Bible (Ecclesiasticus, 11) is cautious: 'Do good
to the just, and thou shalt find great recompense, and if not by him,
assuredly of the Lord.' Even in the New Testament there are dif-
ferences of emphasis. According to St Luke: 'Give and it shall be
given unto you; good measure, pressed down and shaken together
and running over shall men give into your bosom' (6: 38). So one
should give not because one has received, as St Matthew says, but
in order to receive - with forethought of the advantage to be gained,
which sounds just like the 'nice calculation' which Wordsworth so
scorned! Coleridge put it more coolly: 'We receive but what we give'
- from a poem perhaps appropriately named Dejection. And in his
Complaints Edmund Spenser listed the results of lack of calculation
more sadly - 'to spend, to give, to want, to be undone'.
There is clearly a paradox in this traditional literary field - giving
is good, but getting is not to be ignored. I am not concerned to
explain discrepancies in Biblical and other standard texts, but
literary wisdom, and presumably popular wisdom too, bear out the an-
thropological conception of reciprocity attaching to a gift. I examine
this aspect in more detail later. But if one distinguishes the symbolic
quality of a gift from its pragmatic quality, it is evident that insistence
on a pragmatic return may detract from the symbolic value of the
transfer. Underneath the paradox about giving and receiving, then,
may be a dilemma as to how far pragmatic components may be
supported, in the light of symbolic aims. This is where the formalism
of the gift has an important function.
F O R M A L I T Y OF GIFT-MAKING
It is clear that what is important in making a gift is not just the

transfer of goods or services in a putatively one-way action, and
that the name of the 'gift' or 'present', though significant, is not
376 SYMBOLS
completely acceptable at face value. The conditions of the gift, the

implication of equivalent desired or not, and the attitudes of giver
and recipient are all relevant in interpretation.
Sociologically, a striking aspect of gifts of material objects is that
they are very rarely just handed over. They are usually given with
some formality. Broadly speaking, the degree of formality in gift is
commensurate with the degree of publicity, that is with the degree
to which the gift is of general social interest. Indeed, the term presenta-
tion, in contrast to present tends to be reserved for gifts made in
fairly formal circumstances on public or semi-public occasion.
A 'presentation' of a medal to someone for distinguished service
implies an assembly and some formal statement; a 'present' of a
medal to him might be a private gift. But private gifts too, even
in the domestic circle, may have their own formal touches, especially
when a number of people are involved. There are families who when
their children are young, celebrate Christmas by piling up all their
gifts under a Christmas-tree and having some ceremonial distribu-
tion of them by some adult disguised as Father Christmas, or some
other elder such as a grandfather. Birthday presents are not just put
at the disposal of the recipient ('presentee' is apparently a good
dictionary term); they are laid beside the plate at the breakfast
table on the birthday morning. Christmas and birthday presents are
apt to be wrapped in some formal way, perhaps in paper specially
bought for the purpose, often with a written message of greeting.
Why such formality, when the wrapping paper is immediately torn
off, and the last thing one wants in a family usually is to be sending
written messages to one another? The reason is well understood -
that with family presents as with public presentations, the formality
celebrates not so much the gift as the social relationship which the
gift symbolizes. The transfer of the material thing is a recognized
expression of the importance of the immaterial relationship between
the persons, and this is enhanced by removing the gift out of the
sphere of everyday transactions by diacritical signs - as by removing
the price tag. Such symbolization of intangible social ties by gift
seems to be a universal phenomenon.
The formality of gift commonly carries with it a display of posi-
tive social interest. The passage of a gift implies that friendly rela-
tions should exist between the parties. Even if such do not exist the
popular spirit of the gift is that they ought to exist, in a kind of

public statement of a viable social relationship. In some circum-
stances a gift is overtly taken to serve the function of converting a
hostile relationship into a friendly one. Marshall Sahlins has said
epigrammatically: 'If friends make gifts, gifts make friends' (1965,
139). Yet such display may be at a purely formal level; the parties
may hate each other. This too may be well understood by all con-
cerned. What the gift stands for in such a case then is a social rela-
tionship at two levels - at a superficial level of amity; and at a deeper
level of viability. To paraphrase Sahlins - gifts do not make friends;
they make people behave as friends behave, that is maintain an on-
going relationship.
This symbolic significance of the gift is manifested by the common
convention of formal physical contact between the parties. The
object itself, or a material token of it, should be involved in direct
personal relationship. In Western societies, on public occasions of
presentation especially, the gift is physically handed over from
donor to recipient. This act, often done with the left hand, leaves
the right hand free for a handshake. The high point of the ceremony
is the hand-to-hand touch, which formally seals the transaction.
(A smile by the donor is nearly always obligatory.) Procedures vary
according to circumstances. Sometimes the gift is kept in the back-
ground until the critical moment arrives, and is then put by others
into the donor's hand to make the actual presentation. Sometimes the
actual thing never appears, but some token of it is used in the presen-
tation instead - a cheque, some title deeds to land, the key of a
building. Status considerations may enter into these more public
formal occasions: some gifts are handled by servants or functionaries
of lesser status until the moment of actual presentation arrives,
when they are passed to the hands of the person of high status who
is to be the actual donor; other gifts may not be handled at all by
the donor, who involves himself in the presentation act by some
word or gesture. But in some societies where there is no full hand-to-
hand transfer of objects it is the recipient who signifies his acceptance
of the gift by touching it and so linking it with his personality
(compare Malinowski, 1922, 352). In Western domestic circles,
whether or not a gift formally changes hands, the transfer may be
sealed with an even more intimate gesture, a kiss. In all such ways
N
378 SYMBOLS
the physical contact - of hand with hand; of hand with object; of

lip with cheek - symbolizes a social relationship established or
maintained. But it is to be noted that the symbolic pattern varies -
according to the particular culture, the nature of the social unit
concerned, the relative status positions of the participants. In general
we are on the threshold of the kind of primitive thinking to which
Mauss was pointing-the relation between physical and con-
ceptual personality, with the latter envisaged in a social dimension.
Another aspect of formality in gift-making is its verbal signaliza-
tion. The physical transfer of the gift is often accompanied, especially
on public occasions, by utterance in a conventional form of words
of what the English call rather oddly '<z speech'. This is often only a
duplication in words of what may be seen in deed, but it allows the
symbolic character of the transfer to be underlined, through state-
ments as to the value - possibly the emotional meaning - of the
relationships involved. But the verbal accompaniment to a gift may
be complex in intention. One of its functions may be to avoid em-
barrassment by disguising the emotional aspects of the transfer.
But a contrary function may be to allow the recipient to underline
the emotional significance, the bonding value, of what has been
presented to him. Superficially, in many societies, there are two
sharply contrasting modes of behaviour here, maximization and
minimization. In maximization the donor draws attention to the
value of his gift, praises its quality and leaves the recipient in no
doubt as to his view of the importance of what is being given and,
by implication, the importance of the linkage thus effected. We like
to think that maximization of a gift is not a British or American
habit. Our technique of signalization goes the other way, to under-
praise, in order (we say) that the recipient shall not feel embarrassed.
But under-praising is just as much a convention as over-praising. It
is implied that the gift is so evidently valuable that it can 'speak for
itself as we say. Moreover, the convention prescribes that the more
the gift is underplayed by the donor, the more should the recipient
make it clear that he is not accepting this statement at face value, but
realizes the worth of what he is being given. In effect, the words
spoken, casual though they may often seem, are part of the total
gift situation, and are taken into account by the recipient in his
acknowledgement, which is or is part of his form of reciprocity.
But in some cases the technique of minimization appears to take

idiosyncratic forms, as when a donor refuses even to identify him-
self. This is not a disclaimer of the value of the gift, but an avoidance
of public linkage of gift with person of the donor, and so a dis-
claimer of public credit. This may be no mere subterfuge. In recent
years a very generous donor of funds for students' hostels for
colleges of the University of London laid it down as a condition
that if any attempt were made to find out his identity the gift would
be cancelled. One can only speculate on the motivation of such
anonymous acts. The desire to do good by stealth may correspond to
a noble and selfless devotion to the welfare of others in need. It may
indicate a shrinking from the formal conventions of public gift and
acceptance in British society. It may be in conformity to another
convention, such as the religious maxim of Judaism, that secret
charity is a higher form of giving than open charity. But it may arise
from a wish to remain free from involvement and responsibility and
to avoid further calls upon the donor's wealth. It may even cover
a delight in being the unseen object of public attention, all the more
to be savoured because only the donor knows to whom it is directed
- a kind of voyeur in the field of philanthropy. Such interpretations
may sound ungenerous. But part of my analysis in this chapter is to
show how complex can be the motivation of much gift-making, and
even more, to point out how the accepted social conventions codify
and standardize the expression of these motives, so that gifts assume
regular symbolic patterns. From this example of the anonymous
donor it can be seen how such idiosyncratic behaviour poses prob-
lems of interpretation and enforces unfamiliar responses upon
recipients. It also raises a more theoretical question about the charac-
ter of reciprocity when no donor is available to receive recompense.
The formality of gift-making emerges in another way, that of the
appropriateness of the gift. Theoretically, the donor decides what to
give - his act is free. But in practice most gifts are fairly rigorously
circumscribed, by type and amount according to occasion. For this,
recognized conventions operate in every society, with associated
symbolism at various levels. At a Tikopia funeral there are three
main sets of formal participants, each with a service to perform,
and each the recipient of gifts. The immediate family of the de-
ceased, with his father's kin, form the chief mourners - with sub-
380 SYMBOLS
sets according to whether the deceased was male or female, old or

young, married or single. Their prime task is to wail for the dead
person for the prescribed period, to celebrate his virtues and mourn
his loss. Such wailing, irrespective of its emotional base, is a con-
ventional expression, developing into a funeral dirge of symbolic
significance; whether or not a mourner 'feels' grief he must 'utter'
grief. The representatives of the mother's family of the deceased are
mourners too, but whereas the deceased's father's kin stay secluded
in the death house, the job of his mother's kin is to bury him, and as
such they are in a separate category. The third main set of partici-
pants is made up of neighbours, friends, more distant kin, who are not
prime mourners but providers of food for the mourners. They visit
the mourning house for three successive evenings, with water
bottles and baskets of food, and when all have wailed together, they
induce the mourners to eat. And at the end, like the burial party,
their services are compensated for by gifts of property (see Firth,
1939, 325-31; and above p. 247). Now at every stage the gifts are
specified as to type, and largely also as to amount, by tradition.
Cooked food is brought for the household mourners to eat, and the
providers are compensated with gifts of wooden bowls and coconut
sinnet cord. Pandanus mats and bark-cloth sheets, bowls, sinnet cord
and food are presented to the burial party; they must be reciprocated
in turn by bowls and sinnet alone, but of the same quality. A great
range of subsidiary transactions in raw and cooked food and bark-
cloth takes place. Some of the basis for this is utilitarian - food for
the hungry mourners, wrappings for the corpse; but much is con-
ventional, its appropriateness being socially not pragmatically
dictated. Some of the items are directly, obviously, symbolical - as
when raw food from a man's cultivations is stood on his grave to
represent the work of his hands and indicate the readiness of his
successors to dedicate it spiritually to him (Firth, 1970a, 249). But
others are only symbolical in a more general sense, representing
social bonds between the parties and not any specific thought. The
wooden bowls given to the food providers are overtly intended to
express gratitude for the feeding, and recognition of the status factors
involved, but not any explicit symbolism of food preparation for
which the bowls are used; they are primarily presented as valuable
property.
In a Western society, analogous social specification of gifts

exists. For a couple who are newly married, a present of kitchenware;
for a family member with a birthday, an item of personal clothing;
for an anniversary after twenty-five years of marriage, a piece of
silver - each of these is appropriate in its context and violently out
of place in some other contexts. But the freedom of choice still
permissible within such conventional limits has given rise in modern
times to a further convention - that for an occasion such as a wed-
ding the projected recipients draw up a list of desirable gifts, from
which the donor may choose, so combining utility with symbolic
value.
The symbolic value of a gift is particularly attested by the limita-
tions on the disposal of it. A gift may be of small practical use - if,
for instance, it is a duplicate of something already owned - and it
may have market value; but to sell it may be insulting, or at least
wounding.* As Barry Schwartz (1967) has pointed out, to make a
gift is self-defining; it is an index to idea of personality of giver and
recipient. Hence to sell a gift or even to change it for another item is
to reject or modify by implication the donor's conception of the
personalities and the relation between them.
EARNEST AND TOKEN
Part of the ordinary convention in many societies is that gifts should

be broadly proportionate to resources of donor and recipient, not
only as a pragmatic matter but also as an indication of their symbolic
significance. Either meanness or ostentation maybe equally criticized,
in terms of a lack of fit between size and style of gift and quality of
social relationship it represents.
But there are circumstances in which a gift is recognized by both
donor and recipient to be very disproportionate to the resources
which could have been expended - perhaps only a small fraction of
what ordinarily might be expected - and yet it is acceptable. Such a
gift may have several functions.
It may be an earnest, an indication of what is further to come, or
* I have been told that in the United States, Christmas gifts are often taken back to the
store where they were originally bought, and exchanged for something more suitable of
equivalent value, on the excuse that they are Vrong size', 'wrong colour', etc. But to treat
a family gift in this way may need prior negotiation.
382 SYMBOLS
what may further come if certain conditions are met by the recipient.
A Trobriand example of this is what Malinowski has described as a
'solicitary gift' (1922, 354), by which a man who hopes to secure a
valuable shell ornament in the kula distribution tries to make an
initial present of some smaller item to the immediate owner, with
the implication that he is ready to make the appropriate large gift
if he is successful. Interpretation of the donor's attitude may be
difficult - he may be supplicating from an inferior position for the
grant of the valuable, or he may be virtually issuing a command that
he expects it to be given to him. But in terms of socio-economic
relationship, the offer of the earnest has set up a bond of obligation
which the recipient must take into account even if he does not
accede to the donor's request. In Western society a small gift is
sometimes made as an earnest because a larger gift which is intended
or promised is not yet ready for presentation. Here the accent of the
earnest is on maintaining the character of the social relationship,
lest the recipient may think the donor promises but does not perform,
and modify his attitude accordingly.
Of the same general order as an earnest is a token gift. The two
terms are often merged, but as I indicate later, the concept of token
can also carry a more critical negative connotation. In common
usage one function of a token gift is to serve symbolically in overt
fashion. A small presentation is made, commensurate neither with
the resources of the donor nor with the needs of the recipient, but as
index of commitment. In itself this may be encouraging, as support
for the recipient's interests or conception of his own personality.
But in public affairs in Western society a token gift is often designed
as a means to stimulate other gifts, more substantial, or other com-
mitments which can lead to effective action. In a recent election
campaign in the state of Michigan, the chairman of a noted auto-
mobile firm sent $900 to aid the cause of a Democratic candidate. The
amount was small by comparison with other contributions, but it
was noted in a New York newspaper with the comment that 'in his
case, the symbolism was the main thing' {New York Times, 8
November 1970). Looked at as a serious campaign contribution
from such a multimillionaire source, the gift was ludicrous; but the
knowledge that this particular donor had engaged himself on the
candidate's side was presumably of great value to the campaign
managers. But it is to be noted that in such circumstances the effect

of the token gift is dependent largely upon the publicity it receives;
a secret gift of a million dollars without the symbolism would have
achieved a different effect.
But a token gift may be symbolic not of commitment but of a
wish to avoid commitment. Historically, tokens have been coin-
shaped pieces of metal circulating at less than their nominal value,
and the notion of token can carry an idea of substitute or inferior
counterpart. So the modern colloquial word 'tokenism' has de-
veloped, to indicate behaviour which assumes the form of, but falls
short of, true commitment. What is symbolized by a token gift in
this sense is the handing over of some inconsiderable item in the
form of a gift but without intention to follow this up by any sub-
stantial action of the kind required by the situation. In the philan-
thropic climate of modern Western society it is not uncommon to
buy a badge or a button in a street or campus campaign, not just
from a generous wish to help the cause, but also to make one's
sympathy plain to other people, or even to protect one from being
troubled further. Such crude 'tokenism' in giving is socially dis-
approved, though probably many people in modern cities practise it
to some degree. But more insidious, and less stigmatized, is the token
gift of money as a symbolic substitute for giving one's time, with
its commitment of thought and energy.
The idea of the gift as a generous voluntary gesture by which a
person divests himself of his rights in something, to transfer them
for the benefit of someone else, is description at a superficial level.
The gift may be complicated by elements of formality, publicity
and appropriateness. It may embody obscure psychological notions
of the definition of personality and of the validity of personal con-
tact, and more obvious sociological notions of the symbolic signifi-
cance of things transferred and contacts made.
GIFT AND COUNTER-GIFT
I now return to the idea of counter-gift, the notion of reciprocity

which is recurrent in anthropological literature. It has been simple
for anthropologists to demonstrate empirically the prevalence of
384 SYMBOLS
reciprocity in gifts, in the societies they have studied, and also to

draw attention to parallels in Western society, especially in the field
of family ceremony. But the treatment has gone much wider than
the field of the gift in the strict sense.
The theme of reciprocity has been found in proverbs. Among the
Tswana of Botswana traditionally a man had a special relation to
some one of his sisters. He had to buy her clothing occasionally and
give her other presents. Whenever he slaughtered an ox he had to
give her a flank. In turn this sister looked after him before he was
married - cooking, washing, mending his clothes-and after he
married she helped his wife in domestic tasks. Most of these obliga-
tions have not been compulsory in the sense that the courts would
not enforce them. But it would be said that an offender did not show
his relatives honour and respect. Ill-feeling resulted, and his rela-
tives might refuse to help the offender materially when he needed
them. Symbolically, the Tswana put this ideal of mutual help as a
proverb: 'Hands, like sheep, bump against each other'- putting their
concept of reciprocity in a pastoral idiom (Schapera, 1955, 187).
(Note again the image of touching hands as symbolizing co-opera-
tion and social ties.)
Myth too has its illustrations of the reciprocity theme. One such
touching symbolic form of the representation of implications of gift
has been cited by Denise Paulme (1967, 48-61) from West Africa.
It concerns the way in which mankind came to die. Originally,
Death lived in the bush and did not come into the villages; he killed
game and man did not die. One day a hunter went into the bush.
He found Death, who gave him some meat. He thanked Death and
brought the meat back to the village, but he did not know that he
was in debt. One day Death came to the village to ask for his pay-
ment. The man said: 'Why, was not the meat a gift?' Death said:
'I was in the bush when you came and collected all my meat. Today,
you must repay me.' So the hunter gave him one of his children,
whom Death took away, and men then began to die. So symbolically,
in this story, if no exchange takes place, no return gift is made, a
gift received means death for the debtor or his kin. At the least, a
gift made means power, of donor over recipient, and material and
immaterial elements are closely interwoven.
In the stimulus to provide counter-gift two main themes may be
involved in varying degree. One is the theme of recoupment, of

compensating original donor for his loss. The other is the theme
of re-assertion, of establishing original recipient once more on a
level of equality. Both have material expression, but the latter can
more easily assume a symbolic significance, and as such neglect
equivalence in favour of over-compensation.
In the anthropological theory of exchange the central notion of
reciprocity is still a subject of debate, and I take it up later. But it is
convenient to traverse Mauss's initial formulations by the way,
since they have been of cardinal importance in the analysis. Loosely
phrased and ethnographically inaccurate as some of his generaliza-
tions were,* and tinged with almost a mystical flavour, they did bring
into sharper focus than before some of the basic features of this
aspect of exchange. Mauss showed that a gift is not a free individual
act, but is highly institutionalized; that it is not simply a voluntary
gesture, but can be a response to firm obligation; that it is not just a
one-way transfer, but tends to involve reciprocity, even sometimes
by a return-gift of the same form; that it is accordingly in a symbolic
way a transfer of personality and a linking of personalities; that it
incorporates ideas of relative status, and of power; and that it may be
an important mechanism for group action and for linking groups. So,
Mauss gave his epitome in terms of: the obligation to give; the
obligation to receive; and the obligation to repay.
Mauss's own formulations have not gone unchallenged. C. S.
Belshaw, for instance, while acknowledging the significance of the
concept of prestation as indicating the contextualization of gift, has
argued that Mauss has overstressed the social imperatives in presta-
tion at the expense of individual choice and entrepreneurial ele-
ments (1965, 48). Along somewhat similar lines I would argue that
though Mauss did introduce some qualification to the precision of
* Michel Panoff(i970,60-1) concedes that Mauss may have misread ethnographic accounts
at some points, but holds that this does not impair the theoretical results of his study of gift-
giving. But while it is true that Mauss's most general results are unaffected, his theory of the
relation of the gift to the personality of the giver is certainly distorted by his ethnographic
errors. Sahlins (1970) traverses Mauss's Maori ethnography and commentaries upon it, and
comes up with an ingenious (though I think misconceived) suggestion of his own about the
concept of hau which is central to this part of Mauss's work. Sahlins's rendering of the hau of
the forest as its fecundity, in my view, mistakes an indirect expression of a relationship for a
direct expression. It reminds me of my own concrete interpretation of the Tikopia concept of
manuy but the evidence is of a different order (cf. Firth, 1929, 412-15; 1967b, 174-94).
386 SYMBOLS
meaning he gave to concepts of obligation, these are by no means as

firm or rigorous as he generally maintains.
THE O B L I G A T I O N TO G I V E
In his examination of the obligation to give, Mauss focused largely

on the potlatch of north-west Amerindian tribes, stressing in particu-
lar the significance of status estimation and rivalry in inducing
large-scale presentations.* In a very different context, that of gifts
made by anthropologists to people who have been their informants
and friends in the field, I have indicated the importance of status
considerations, but also the complexity of the situation. Recognition
of obligation is by no means rigorous and automatic but on the
contrary may involve choices of a meaningful, even painful kind
(Firth, 1967c, 11-14). The symbolic significance of a gift is not
merely restrained in the interests of a more utilitarian calculus; it
must be estimated against other symbolic considerations.
Here I examine how some of the symbolic aspects of gift may be
restricted or conditioned by some other types of modern pressure.
In his account of the San Bias Cuna Indians of Central America,
the late David B. Stout noted a growing conflict between traditional
norms of hospitality and sharing patterns, and the new competitive
and individualistic values. Anyone with a temporary surplus of food
or goods was beset by friends and relatives, wishing him to share out
his property. They would make reference to the old precept that a
Cuna would give away or share whatever a fellow Indian asked of
him. But even a generation ago young Cuna men would make the
rationalization that they did not work harder or attempt to accumu-
late more property because if they did they would only have to give
it away. If they did not so give it, they would be labelled as stingy
(1947, 77). Here new wants and opportunities, rising expectations
and changing values reduced the pressure of obligation to give, and
tended accordingly to modify the symbolic patterns of social
intercourse. (For analysis of complications in Zaire kin relations see
Lux, 1972.)
Limitations upon the obligation to give may be even firmer from a
* For a recent statement on exchange structures of potlatch type see Paula G. Rubel and
Abraham Rosman, 1970.
more fully developed legal system, in which control of property

in the wider interests of society may be held to supersede the narrower
interest of the individual, whether symbolic or not. The simple
conventional notion of gift, that a person disposes freely of his
own property, is subject in a modern state to various over-riding
demands.
The basic notion of gift is the alienation of property from oneself
with complete surrender of all rights over it to another person. Yet
in the eyes of the law, a man may not be allowed to divest himself at
will of all interest in his property. The attempt at surrender may not
be recognized as complete. We are familiar with rules whereby gifts
inter vivos are still reckoned in certain circumstances as being part of
a deceased person's estate for death-duty purposes if he had died in
less than a given period (earlier five, later seven years) after making
his gift. The state, in the form of the law, asserts a right not to limit
the gift but to consider the transfer as incomplete for a statutory
period. It may be argued that this is simply a device for maintaining
revenue. But even so it has the effect of placing an impediment in the
way of achieving the symbolic as well as the practical aims of the
gift. In some other conditions the law may refuse to recognize the
existence of any obligation to give, and may disallow what a person
attempts to hand over as a gift. A record of an English law case
some twenty years ago notes a challenge to a man's legacy of
£1,000 to the vicar and churchwardens of a church, the income to be
spent on 'seasonable food and drink' to be distributed by them in the
name of the testator among twenty communicants and women of the
parish on Christmas Eve. The judge held that the bequest was in-
valid. Its motive, he said, was vanity, there was no express note of
benefit to the deserving poor, it was therefore not charitable and
failed accordingly. With regard to the suggestion that this bequest
might have been for a religious purpose his lordship said he was not
prepared to hold that the provision of plum pudding was for the
advancement of religion {The Times, 20 March 1954).
Muslim law is particularly revealing on the subject of the obliga-
tion to give, and its symbolic implications. Gifts of a charitable
nature are expected from Muslims as part of their religious duty.
The Koran enjoins believers to give as charity without material
return, what they can spare. God knows, and God will repay - in
388 SYMBOLS
fact, if there be any good deed, He will repay it doubly, says the
Koran, provided that it should not be given in order to be seen of
men. (Here is one form of reciprocity for an anonymous giver!) Yet
while approving gifts in general, Muslim law lays down quite
stringent conditions about them, particularly where they affect the
family (see Seymour Vesey-Fitzgerald, 1931, 201-2, 206-23).
Although a gift is gratuitous, the Muslim law regards it as always
made with an object in view - it is not a simple discard of property.
Gifts made to acquire merit in the sight of God have as object a
recompense in the next world. Worldly gifts are made to ingratiate
oneself with other human beings. A leading Muslim legal treatise
says of these that the object of a gift to a stranger is to get a return.
(It is lack of specification of return, and lack of contract for return
that makes the transfer a gift.) It is the custom to send presents to a
person of high rank that he may protect the donor; to a person of
inferior rank that the donor may obtain his services; and to a person
of equal rank that the donor may obtain an equivalent. All these
gifts are legitimate.
Gifts to acquire merit, if made without thought of worldly recom-
pense, may be given to people of one's own surroundings, even to
one's own kin. In fact, it is held that charity literally begins at home,
and properly so; gifts to members of one's own family are highly
approved by Islam. But here the interest of the law begins to emerge.
A gift to acquire merit cannot be used as a cloak for an act unpleasing
to God and the Muslim courts will reject it.
Moreover, to support the interests of the family, Muslim law
severely restricts freedom of testation. A person can leave by will
only one-third of his property, and legacies to heirs are severely
restricted in the interests of the equitable division of the property,
which is a cardinal principle of Muslim law. On the other hand,
Muslim law makes allowance for a special type of religious gift known
as waqf. In theory, this means putting an object in the category of
divine property so that the donor's right in it is extinguished and it
becomes the property of God to be used for the advantage of His
creatures. Such would be a piece of land donated for a religious
foundation. But since marriage and family life are the religious
obligation of every adult Muslim, a man's duty to his family takes
precedence over all other objects of his generosity. He can then make
an endowment for his family, the assumption being that on the

failure of his descendants it would go to the poor. To make such
an endowment a person must be free, sane and master of his own
affairs. He must not make it for anti-social purposes - such as
avoiding payment of his just debts. But the contest which goes on
continually between the rule of law and the interest of the individual
emerges in many such cases. These inalienable family endowments
came to be treated in some Muslim countries as opportunities for a
man, in effect, to distribute the income among his heirs at his dis-
cretion, instead of abiding by the normal legal and religious rules of
inheritance. It has been a matter of common knowledge that, as
Gibb has said (1947, 91), this practice has been 'the cause of much
moral corruption and economic loss'. Hence modernists in Muslim
countries have for long agitated to remove such private family
provisions from legitimate protection by religion. The reform has
distinguished between genuinely charitable waqf intended for the
endowment of religious and philanthropic institutions, and private
endowments intended to benefit a particular family or individual.
Their aim was primarily utilitarian, but it was in the direction of
protecting the general symbolic values of the gift from financial
manipulation. In Egypt and in Syria religiously backed family
endowments have now been abolished for some time, and in other
Muslim countries they have tended to be more strictly limited.* So
Muslim reformers, while not wishing to reject the sacred law, have
dared to reinterpret it radically, stripping it of what they have re-
garded as the human accretions that jurists have put upon it during
its long history. In so doing, while not restricting the 'obligation to
give' they have channelled it into forms which could be given most
broadly-based social approval, and which had most obvious
symbolic value in representing the most abstract religious concepts.
THE OBLIGATION TO RECEIVE
The symbolic qualities of the gift come out especially in the obliga-
tion to receive it. Even if on the material side a person may not need
* For further details see J. N. D. Anderson, 1959, 78-9; Snouck Hurgronje, 1906, II, 321;
J. Schacht, 1955, 82-3; Gibb and Bowen, I, pt 2,165-78; E. C. C. Howard, Minhaj et Talibin,
230-3.
39° SYMBOLS
the object given to him, he may be 'constrained', as Mauss puts it, to

receive it; to reject it could be considered a refusal of friendship and
intercourse. If in Tikopia, a man comes with a basket of food to
one's house as a present, one does not think of refusing it though
one may have half-a-dozen baskets already, much more than ample
for household needs. The correct thing is to accept it with thanks,
and then later if need be to avoid waste, give it away to some other
household. That the obligation to receive may entail a further
obligation to give, in a kind of chain reaction, is quite in accord with
Tikopia ideas of hospitality and social linkage. As Mauss has noted,
there is an association in Polynesian society between: hospitality,
food, communion, peace, exchange, law.
But there is a danger that such statements will be overgeneralized,
since Mauss was not careful enought to distinguish the kinds of
social situations where they were applicable from others in which
they were not. Without undertaking a systematic study I note that
the obligation to receive a gift is conditioned by the structure of the
society concerned, operating in particular social situations.
In non-industrial societies of type commonly studied by anthro-
pologists, where individual roles tend to be less sharply differen-
tiated than in industrial, more complex societies, the obligation to
receive a gift tends to be more general. In the more complex in-
dustrial type of society, with high degree of differentiation of role,
there are codes of refusal for gifts just as there are codes of acceptance.
It may be argued that if giving is part of the nobler side of life,
associated with friendliness, generosity, the contribution of the self
to society, why this reluctance to receive, this rejection of participa-
tion in the philanthropic act? There is sometimes much ambivalence
about acceptance of a gift. There may be pragmatic reasons: that
the recipient does not think the donor can afford the gift; that he
fears he himself cannot easily afford to pay it back. There may be
more subtle considerations: in modern society, however much he
might wish and be able to repay, the diffuse patterns of occupation
and residence might give him no opportunity to do so. Even further,
granted that conventionally 'gifts make friends', the intended
recipient may not wish to 'make friends'; the diversity of modern
society allows people not only to choose their own friends - even a
primitive society allows this-but also to refuse social ties with
others who might like to be their friends. Hence the various strata-
gems to evade gifts from people with whom one does not want to
intensify social relations. The obligation to receive is inhibited then
in the light of the further obligation - the need to repay, it being
understood that repayment may involve intangible as well as tangible
considerations.
From this point of view quite stringent limitations may arise on
the obligation to receive gifts, when differential status of the parties
involves the higher status person as recipient. There are two possible
interpretations of gift in such circumstances: that it is complimentary,
an acknowledgement of status as a kind of return for the general
symbolic role of the recipient; or that it is establishing a platform for
future benefit in goods, services or reputation from having social
contact with the person of high status. When gifts to modern
Western heads of state are in question, both interpretations may
apply, but it would seem that the danger of the second has led to
restriction. It is understood, for example, that presents sent to
members of the British Royal Family by private citizens not known
to them are normally returned, though presents from other heads of
state, public bodies, and their personal friends are not. While no
implication of reciprocity may be involved with the act of the private
citizen, the acceptance of the gift might be treated as the creation of a
personal link which might be invidious as regards other citizens.
The obligation to receive is not acknowledged because it runs coun-
ter to another type of obligation - not to discriminate among citizens
in what should be equality of general relationship to the Sovereign
and Royal Family.
With a head of state the status factor is of prime importance, since
in modern times no one is likely to think of any service that can be
secured by a gift to a member of the Royal Family. But in the broader
sphere of public role the obligation not to receive a gift may be much
more pointed. By the Civil Service regulations of most advanced
countries a civil servant is required not to accept gifts from people
with whom he may have technical relationships in the course of his
duty. Acceptance of a gift, or even of lavish hospitality, has at
times imperilled a politician's or a civil servant's reputation, led to
his resignation or dismissal from public office, or possibly brought
him to the brink of criminal prosecution.
392 SYMBOLS
In the United States, allegations of bribery are made from time to

time and though usually denied, arouse public disquiet. In 1970, a
Federal grand jury investigating reported conflicts of interest by
members of Congress indicted a veteran representative on charges
that he accepted a bribe of $25,000 in return for seeking to block
a Justice Department prosecution of a business firm for irregularities
in house improvement proceedings. In the same year, widespread
allegations were made against members of the police department of
New York, ranging from accusation that gamblers, narcotics dealers
and businessmen made illicit payments of millions of dollars to New
York policemen, to charges that small weekly payments were made
to police by many shopkeepers so that they could operate on Sun-
days in violation of state regulations. Public officials became greatly
concerned. In the middle of the year the state legislature passed new
laws tightening up rules about acceptance of gifts by all municipal
employees and requiring all local governments to issue a code of
ethics prescribing the standards of conduct 'reasonably expected' of
employees. The intention was to prohibit the acceptance of gifts to
secure governmental favours, and the definition of gift was a broad
one, pertaining to 'money, service, loan, travel, entertainment,
hospitality, thing or promise, or in any other form ' Later still
the New York Police Commissioner proclaimed a stand against the
soliciting or receiving of gifts by policemen, in any form, even in-
cluding externally financed Christmas parties (see, for example
New York Times, 1 and 25 April 1970; Post Standard (Syracuse),
3 June 1970; New York Post, 9 December 1971).
In Britain, such corrupt practices are generally regarded as being
rare,* partly because of the different structure of national and local
government, partly because of more careful public audit, but largely
because of a different tradition of public service official behaviour.
But allegations of bribery of police are occasionally made, and public
concern sometimes arises over payments made to politicians or to
public servants, for which no clear legitimate consideration can be
shown. Some years ago it was reported that a clerk of works in a
Midland county was sentenced to four months' imprisonment under
the Prevention of Corruption Act for accepting small sums of money
* That is nowadays; they were common in the seventeenth and eighteenth centuries, for
example in the career of Samuel Pepys, or of Robert Walpole.
on a council housing estate. He said rather naively: 'I thought it

was perks for the clerk of works' (for some details see The Times,
2 July 1967; 13 February 1971; 14 July 1972, New York Times,
30 March 1970). In other countries too, periodically, where the
Press is allowed to publish details, cases of alleged bribery are re-
ported, though the degree of public anxiety about the probity of
bureaucratic and other figures in public life varies considerably.
In Mexico, for instance, a group of peasants appealed to the governor
of their state to stop illegal depredations by timber merchants upon
the forests; they complained that forest guards did nothing to inter-
vene, being silenced by bribes {The Times, 7 August 1970). In
Zaire the system of bribery has been said to be ubiquitous among
government officials, being necessary to expedite most ordinary
services, such as passing building materials at reasonable speed
through customs formalities {Newsweek, 22 November 1971).
The view that private gift does not accord well with public
performance seems to have a long ancestry. In sixteenth-century
England 'gift' in one sense meant something given with a corrupt-
ing intention, a bribe. And the early seventeenth-century Authorised
Version of the Old Testament has repeated warning against the accept-
ance of gifts because of their untoward influence upon judgement.
'Thou shalt take no gift; for the gift blindeth the wise, and perverteth
the words of the righteous' (Exodus, 23: 8 and Deuteronomy, 16:19;
cf. II Chronicles, 19:7; Ecclesiastes, 7:7). And an Apocrypha version
reads: 'Presents and gifts blind the eyes of the wise, and as a muzzle
on the mouth, turn away reproofs' (Ecclesiasticus, 20 (Rev. v.)). The
'obligation to receive' is here quite clearly stated in reverse.
The notion of acceptance of a bribe is in some societies associated
with eating, often with an unpleasant association. In Malay the
expression for taking bribes is makan duit, to eat money, or makan
suap, to eat a mouthful, as in handfeeding (with idea of gobbling).
In German a corrupt judge, one that takes bribes, has been termed
Gabenfresser, an eater of gifts, using the word for eat that is ordinarily
applied to eating by animals, devouring.
Yet the economic and moral classification of gift in such contexts
is not entirely simple. Anthropologists have long recognized that in
some Oriental and African countries there is a thin line between
bribe and status-gift. Traditionally, if one appears before a superior
394 SYMBOLS
one should not go empty-handed; one should acknowledge his

rank by a gift of introduction. Muslim law specifically allows this
(see above). Many years ago William Robertson Smith noted that:
'It was the rule of antiquity, and still is the rule in the East, that the
inferior must not present himself before his superior without a
gift "to smooth his face" and "make him gracious"' (1889, 328).
To omit such a gift is ill manners. Conversely, superiors or equals
may be expected to recognize the social standing and the worth of
other people with whom they have business relations, by gifts either
unilaterally transferred or exchanged. In Chinese communities
payment of small amounts of 'tea money' to inferiors has been
common, as also exchange of more substantial gifts of'lucky money'
or analogous articles between friends and relatives before the Chinese
Lunar New Year. In the domestic circle and between private citizens
such gifts are commonly status indicators and symbolic tokens of
social relationship. But from private citizens to public officials they
can so easily assume the character of bribes that in recent years the
government of Hong Kong specifically forbade its officials to receive
'lucky money' {Honolulu Star-Bulletin^ 29 January 1969), even
though it might be distributed in the red packets which traditionally
expressed the symbolism of good fortune.
Disentangling elements in a gift-bribe-corruption syndrome is not
an easy task, since a critical issue is how far the mind of the recipient
may be thought to have been influenced towards partiality by the
gift, and even he himself may not be able clearly to decide. Cyril
Belshaw, who has been much concerned with problems arising from
adaptation of traditional economic systems to modern conditions,
has taken a rather cautious view of this question. He has pointed out
that bribery and corruption carry moral connotations in modern
Western society which are not necessarily applicable to the con-
texts of public affairs in developing countries. He uses the criterion
of whether or not a gift effects alteration in the recipient's judgements
and actions, particularly in the direction of seeking more gifts. He
would probably not class the gift as a bribe if it is the material sym-
bol of a relationship, the sealing of a contract, the payment for a
service openly rendered, one element in a complex of continuous
exchanges, or part of a stock which the recipient uses to maintain
further social obligations. He adds very reasonably that 'the accusa-
tion of bribery should be used most sparingly in contexts of de-

velopment, at least until it has been demonstrated without doubt
that the transactions were not part of an indigenous exchange net-
work' (1965, 46). This is perhaps in part a counsel of expediency.
The fact that a gift was a payment for a service openly rendered would
not seem to remove it from the category of inducing an official to
partiality against other claimants to service. But where all claimants
are expected to use similar measures, it could be argued that it is
secrecy which destroys at least part of equality of opportunity to
obtain service. The humourous gag 'I believe in bribery but not in
corruption' covers an ethical notion that it is theoretically possible
to separate acceptance of a gift made with intention to get special
treatment, from actual furnishing of such special treatment. In such
view the element of corruption lies in perversion - in performing for
one party a service which would have been performed equally for
another party had there been no gift, or would have not been per-
formed for any party with the same speed or energy. An ingenious
analyst in Thailand, said to be a senior government economist,
has argued that the system of 'under-the-table' payments common
throughout Asia is not corruption in the true sense of the word; it is
a system of supplementing sub-standard incomes. Granting that
realistically the payments can be so viewed, his classification of
corruption is somewhat narrow, being restricted to actions outside
the law. In this scheme, corruption implies an illegal transaction;
and these bribes are not corruption because the officials who take
them are being given them to do what they are supposed to do
legally, in any case {Honolulu Star-Bulletin^ 25 March 1969). (This
distinction, however, is not supported by the dictionaries.)
In the complex administrative conditions of a modern state it
seems hardly possible to make such fine distinctions. To separate
status-gift from bribe in public life and avoid bureaucratic corrup-
tion seems a counsel of perfection. Hence the great care taken in the
scrutiny of any payments to officials personally from private citizens,
and the denial of their obligation to receive gifts.
THE OBLIGATION TO REPAY
In so far as a gift in material terms is a one-way transfer, it creates an

396 SYMBOLS
asymmetrical relationship socially. In general, among those of

broadly equal status, the very act of giving is a symbolic enhance-
ment of the status of the donor. This appears in literary form in
various contexts. 'They give to get esteem', wrote Goldsmith (in
The Traveller). The implication here is that something intangible is
exacted in return for the gift - an element of social credit is trans-
ferred from recipient of the gift back to the donor. In one type of
thinking, then, the acceptance of a gift means weakening one's
own personality in return. So the hero of a recent novel (Cecil
Brown, The Life and Loves of Mr Jiveass Nigger) is made to say:
'The only way to keep your strength is to give; never accept any-
thing from anybody'. It is in this field of the symbolic significance
of giving and receiving for the personality that some of the most
sensitive issues of interpretation appear. So if a recipient cannot
refuse a gift because of the social sanctions he may seek for some
equivalent to give back and so recover his status. It is not only more
blessed to give than to receive - it is also more comfortable!
There is a kind of compensatory principle at work here: material
versus immaterial consideration. A gift by itself yields status, or
helps to maintain status. But a gift which is followed by or implies a
demand for a material return may have to be content with less
immaterial return in status, or less of that concession to status which
is the social expression of gratitude. This is brought out in tangential
fashion by a comment of Graham Greene's about bribery. The point
about a bribe is that it is a gift the purpose of which is understood by
both parties to be to secure some material service not ordinarily to
be expected. In his novel The Comedians - concerning Haiti-
Greene has one of his characters say: 'I have often noticed that a
b r i b e . . . changes a relation. The man who offers a bribe gives away
a little of his own importance; the bribe once accepted, he becomes
the inferior, like a man who has paid for a woman.' In other words,
by the passage of the bribe, with implicit stipulation of a material
favour in return, the giver has traded status for service.
But as I have just indicated, in social relations the acceptance
of a gift may be difficult precisely because no material return is
envisaged or possible in the circumstances. The hardest of all ways
of reciprocating a gift may be for a person to yield up elements of
his status, his pride, his personal dignity. The true symbolic con-
cession of the self, the acceptance of another's will or view, can be

the most refractory. Here the institutionalized ways of yielding ob-
served by anthropologists in some less complex societies than our
own (for example by use of middle-men), with very different social
and ritual sanctions, offer interesting contrasts to ours.
An attempt has been made by Wilton Dillon to utilize some of
these anthropological findings in an essay on international relations,
under the title of Gifts and Nations: The Obligation to Give, Receive
and Repay (1968). The author has set out to adapt Mauss's concept
of reciprocity to an explanation of the strained relations between
Gaullist France and the United States: the United States was obliged
to assist France with Marshall Aid after the war, and France was
obliged to receive it; but inadequate channels were made available
by the United States for repayment to be made, hence French con-
cepts of their own status suffered and relations between the two
countries worsened. Dillon has couched his argument partly in
terms of a somewhat fanciful contrast between kula and potlatch
behaviour. Treating kula as alternate giving and receiving in
harmonious equalization, and potlatch as desperately competitive
humiliation of partners by reiterated giving without accepting
return, Dillon holds that the relationship between France and the
United States developed along potlatch instead of kula lines. But as
Michael Thompson has pointed out, the distinction is not valid:
kula is as competitive as potlatch and a more continuous status
struggle. 'In the international kula America is the "big-man" and
exchange relationships will not normally be balanced. The blocking
of counter-gifts from the inferior partners is not so much un-
fortunate as inevitable' (1969; cf. Berreman, 1970). Dillon's rather
simplicist treatment also tends to overlook the significance of
the distinction between the circulation of traditional valuables in the
kula, the 'agonistic' destruction of some types of valuables in the
potlatch, and the reproductive use of technological resources which
was stimulated by Marshall Aid - though some intricate problems
of economic relationship are embedded therein.
A complex issue also relates to the moral evaluation of the gift
in such situations of greatly disparate resources at the command of
the parties. Recognition of the differential may lead to a very dif-
ferent type of reaction than that postulated by Dillon - a claim to
398 SYMBOLS
receive resources which will be given without reciprocity. This is in

fact the type of demand which is increasingly being put forward by
'Third World' nations in seeking aid from the more developed
nations. It is true that reciprocity theory is invoked to some extent
in that it is argued that modern aid without repayment would be only
a just recompense for former exploitation. But a major basis for the
claim is not the equivalence of exchange theory but equalization in
the general name of human rights and international viability.
From all this, however, one significant thought emerges - that
transactions of the gift order are apt not to stand as single items but
to be part of continuing series of relationships of immaterial as well
as material kind. This is the kind of phenomenon Mauss indicated
by his term prestations totales, though the ethnographic code he
gave to it was too dogmatic, despite his claim to 'have access to the
minds of the societies through documentation and philological
research'. But one inference from this continuity is the difficulty of
making precise statements about what is reciprocity for what, since
each item can be referred forward as well as back. It is a common
experience in modern Western society to hear that someone has not
expressed 'proper' gratitude for a gift, that is, in the view of the
donor the recipient has not provided the immaterial repayment he
regards as due to him. But the recipient may argue that no gratitude
is due because the gift is itself an equivalent for some earlier service.
President Harry S. Truman's pungent philosophy 'There is no
gratitude for things past; gratitude is always for what you're going to
do for people in future' may be too extreme. But reciprocation of a
gift may envisage favours to come as much as favours past. In much
of social life it may be convenient to match transfers, to behave as if
transactions were one-to-one. But the concept of a flow often guides
the behaviour of the participants, and it is their different interpreta-
tions of the point which the flow has reached which is responsible for
much manipulation and argument.
In line with such thoughts Marshall Sahlins has analysed the
concept of reciprocity. He has pointed out that there has been a
popular tendency (deriving in part from Mauss) to view reciprocity
as balance, as one-to-one exchange, whereas if considered as material
transfer, reciprocity is often not that at all. It is through scrutiny
of departure from balanced exchange that the interplay between
reciprocity, social relations and material circumstances can best be

seen. Sahlins has accordingly made a distinction between 'balanced
reciprocity' by direct exchange in gift form, with social relations
sustained by the material flow; and 'generalized reciprocity', with
much less evidence of material recompense, the material flow resting
upon prevailing social relations without exactly matching trans-
actions. I would prefer another way of expressing this point - to
view transactions of reciprocity in terms of their diffuseness or
specificity; their symmetry or asymmetry; and the degree to which
they involve elements which can be equated, as being of the same
or different orders. The problem of asymmetry and reciprocity has
been taken up especially by Takie Sugiyama Lebra, in a sensitive
study of the Japanese concept of on (1969). She has argued that the
concept of reciprocity is powerful as an analytical tool, but difficult
to deal with because of the dynamic interlocking between the sym-
metrical strain in the abstract concept and the general asymmetry
manifest in actual social transactions and ideals. Hence in the Japanese
field of obligation represented by o/z, balancing mechanisms are
continually at work in conditioning and constraining the official,
traditionally recognized patterns of asymmetry ascribed to it. It
would not be difficult to see a dialectical process in operation in such
phenomena. But what is very evident is that the 'obligation to repay'
is not an automatic force behind the reception of a gift; it offers a wide
range of possible conduct, with very complex factors involved.
It is evident also that the possibilities of combination of different
elements of reciprocity - material goods, services, intangible
attributions of reputation and prestige - taken in conjunction with
variation in the time factor, allow for accumulation and manipula-
tion of power. In the field of politics, reciprocity for a service may
come in the form of a tacit influence upon policy - intangible,
non-measurable, perhaps almost unidentifiable, yet a most prized
asset.
But the idea that the concept of reciprocity is basic to that of
gift has not gone unchallenged. That a gift not only evokes recipro-
city but may have been made with the notion of reciprocity in mind
may seem offensive. There can be revolt against the notion of
calculation, of self-interest, and a stress upon the altruistic, volun-
tary non-contractual, non-reciprocal aspect of the act. From this
40o SYMBOLS
point of view Richard Titmuss has put forward a powerfully

argued case for the provision of blood for medical care as a species
of gift not treated as a market consumption good. Calling on Mauss
and Levi-Strauss for evidence on the nature of gift-transactions
Titmuss has entitled his book The Gift Relationship: From Human
Blood to Social Policy (1970). He recognizes the significance of the
concept of reciprocity for the gift, but is impressed more by its
altruistic character, its moral component.lt is the Mauss of the'moral
conclusions' more than the Mauss of the 'obligation to return' who
is his model. Unlike the gift-exchange in traditional societies, Tit-
muss points out, 'there is in the free gift of blood to unnamed
strangers no contract of custom, no legal bond, no functional deter-
minism, no situations of discriminatory power, domination, con-
straint or compulsion, no sense of shame or guilt, no gratitude
imperative...' (1970, 239; cf. E. R. Leach (review), 1971). But
what he also points out is how strongly what one may be tempted to
interpret as moral insight and selfless care for one's fellow men is
conditioned by the specific structural arrangements of a particular
form of society.
The complexity of the issues here was brought home to me by an
experience of my own more than forty years ago. On my way through
the Solomon islands to Tikopia I had to rely for transport and
hospitality on the Melanesian Mission, and for some weeks was the
guest of the head of the Mission, Bishop Steward, on the Mission
yacht Southern Cross. As we travelled together among the islands we
discussed many problems of human relationship in the island com-
munities. Malinowski had only recently published his book Crime
and Custom in Savage Society in which he stressed the importance of
reciprocity as a force of binding obligation in Melanesian social
organization. He argued that there was an inner symmetry in all
social transactions and that this involved a reciprocity of services
without which no primitive community - and, he implied, no com-
munity at all - could exist. The Bishop borrowed the book from me,
read it, and strongly disagreed. He argued vehemently that Melane-
sians, like other people he said, performed many acts for others
freely and without thought for return. Giving, not reciprocity, was
the prime motive of service, he held, and he denied the implication
of self-interested action in this field. We argued amicably about this
S Y M B O L I S M IN G I V I N G A N D G E T T I N G 4OI
and other themes, and I think came to respect each other - partly
perhaps because being more detached I could question his views
more stoutly than could his clergy. At last the time came for him to
land me on the beach of Tikopia and leave me to my fate. He had
shown me many kindnesses, which I could not repay. This was his
last trip on the Southern Cross; he was retiring from the Mission after
many years and we both knew it was unlikely we should ever meet
again - and we never did. As he said goodbye, leaving me alone in
this remote community he shook me firmly by the hand, said
gruffly 'No reciprocity!', turned his back and walked off down the
beach to the boat. This was his way of hiding his emotions with a
joke - but his words were also a reaffirmation of a moral viewpoint.
The Christian ethic in its basic postulates stresses the significance
of vicarious giving, of lack of thought for the interest of the self.
All major religions indeed include in their precepts some positive
approval of giving without thought of return. Of course this may be
wishful thinking. Notions of merit so acquired may be built into
the ideology. Ideas of reaping after sowing, and casting one's bread
upon the waters (or in a new version, shipping one's grain across the
sea) suggest fairly direct interest in output-input analysis on a
spiritual plane. As I have argued, one counterpart to giving in this
sphere is the maintenance of the dignity and integrity of the self.
While such compensatory theory may find little backing in modern
theology, it is clear that empirically, consistent selfless gift-making
is no more of an operative principle than is reciprocity. Moreover,
pragmatically, we tend to oppose any attempt to translate the
philosophy of complete self-sacrifice into ordinary social terms.
Anyone who might try to put into practice the principles of the
Sermon on the Mount would be regarded as eccentric; 'sell all you
have and give to the poor' is regarded as metaphor, not an injunction
to be taken literally. But such principles are not to be completely
dismissed from a sociological analysis.
Each society defines differently what acts should conventionally
be followed by a 'return'. In the West we have isolated par excellence
the economic, commercial field of buying and selling, and applied
it to most transfers of goods and services. From this range of
transactions is normally excluded, and termed gift, those which
operate especially in the domestic field, in kinship and friendship
402 SYMBOLS
relations and in religious behaviour. Yet in such relations there is

much half-concealed reciprocity, with subtle compensations.
It might be argued that in such spheres the moral approval to
gift-making without repayment is part of the struggle to free the
individual from the demands of the pervasive economic system. But
also it might be argued that historically, the Western isolation of
commerce from moral considerations* is a means of protection of
the individual from the logic of the self-effacing Christian principle.
The history of attempts by church and state to enforce the anti-
usury laws is a comment on this relationship. But while the issues of
giving and getting are apt to be presented in Western society in the
form of a dialogue between self-interest and religious principle, they
can be set in a secular as well as a religious frame, and are so set in
many societies. What is clear is that whether the concept of giving
and getting be set in a frame of altruism and gratitude, or in a frame
of obligation and reciprocity, the procedures in many aspects are
symbolic instruments used for maintenance or alteration of social
relationships, in the interests of both self and society.
* For example a claim that the law of supply and demand ignores and (by implication
should ignore) morality, as made by Lord Cole, Chairman of Unilever in an annual report
address (The Times, 3 May 1967).
Chapter 12
SYMBOL AND
SUBSTANCE
In this book I have been examining some of the general ideas about
symbols held by anthropologists, or held by others and of possible
use to anthropologists. I have exemplified this examination by a
series of illustrations from some fields where public and private
symbolism seems to be closely inter-related: food; hair; flags; giving
and getting; greeting and parting. Each of these objects and actions
is meaningful symbolically to individuals personally as well as
collectively and socially. Unlike some of my colleagues, I argue that
important problems of interpretation and clues to understanding lie
in analysis of such intricate conjunction between the individual and
the collective symbolization.
The kind of illustrations I have taken bear for the most part on
issues of status. This was deliberate because I am convinced that
what I have called status-involvement is basic in a large number of
social arrangements, and comes to expression in highly sensitive,
elaborately patterned symbolic forms which are both individually
meaningful and socially validated. It has been said often enough that
the nature of life in society can be grasped only in symbolic forms,
and that people must use symbols to handle their problems of rela-
tionship with one another. For their actions to be effective, that is for
them to lead a viable social existence, the symbols they use must be
individually meaningful as well as collectively recognized. But
differences of interest lead to lack of agreement on symbolic mean-
ings as well as to manipulation of symbols in partisan fashion.
Broadly speaking, all the symbols of the kinds I have considered
may be said to represent the social order and the individual's place
in it. Food exchanges symbolize basic social relationships; mode of
wearing the hair makes a statement about the wearer's personality
and his attitude to established authority; hanging out a national flag
403
404 SYMBOLS
symbolizes identity with the political community; a particular pattern

of greeting expresses a view about relative social position... and
so on. The fundamental question of 'symbols of what?' is fairly
easily answered: these are all symbols of social living - of society in
relation to its members and their relation to one another. But there
are other fields of symbolism where the answer is more open, among
anthropologists themselves as well as between anthropologists and
users of the symbols. This is so with many religious symbols, which
represent not simply sets of social relationships of people, but also
relationships of the people to mystical entities and forces removed
from everyday experience. Some political symbols such as national
flags may have sacred quality attributed to them; but the political
community to which they refer and whose authority they legitimize,
however abstractly conceived, is grounded in a body of people.
But a religious symbol such as the Cross stands not only for the
Christian community, the church, with its doctrine and ritual, but
also for the concept of the redeeming sacrifice of Christ, with its
ultimate referent in the person of God himself. And whereas the
church has various manifestations of empirical existence, God, a
concept of another order, can be regarded in many ways, from a very
real Supreme Being and controller of the universe to an imaginative
human construct of a set of ideal values.
In the interpretation of any kind of symbols there is always room
for difference of view on the theoretical framework adopted to
express the relationships. But with religious symbols there may be
basic disagreement even among anthropological interpreters about
what is being comprehended by the symbols, where is the true locus
of the power claimed to give significance to the symbols. It may be
argued that this is not a problem about which anthropologists should
be concerned - that their job is to make clear what the people who
use the symbols think they mean, and how they use the symbols,
and not to try and identify the referent as such. Granted that for the
most part this is so, there are still two reasons which draw the atten-
tion of anthropologists to the question -'symbols of what?'. One
reason is that, if somewhat obscurely, the question is being put to
anthropologists as a general issue. In considering problems of
African conversion to Islam and Christianity Robin Horton refers
to what he terms the 'symbolist' orthodoxy - the view that the
SYMBOL A N D SUBSTANCE 405
spiritual beings of the African cosmologies can be understood only
if they are seen as symbols. Regarding this as barren, he argues for an
intellectualist position which takes traditional systems of religious
belief 'at their face value', as theoretical systems intended for the
explanation, prediction, and control of space-time events. Horton's
complaint is that anthropologists - even Victor Turner - have not
dealt with the problem of why men should have felt constrained to
have invented symbols with such attributes as unobservableness and
omnipresence, that is spiritual beings. What lies behind the symbols?
The other reason is that the anthropologist's view of what the sym-
bols actually represent may affect his whole treatment of the problem.
A generation ago Sir Hamilton Gibb, a noted Islamic scholar,
opened a discussion of modern trends in Islam with the statement
that 'the metaphors in which Christian doctrine is traditionally en-
shrined satisfy me intellectually as expressing symbolically the
highest range of spiritual truth which I can conceive, provided that
they are interpreted not in terms of anthropomorphic dogma but as
general concepts, related to our changing views of the nature of the
universe' (1947, xi). Obviously, his view of meaning of the symbols
of Islam would not be identical with that of a Muslim of parallel
scholarly interests.
The problem of the 'inner reality' to which religious symbols
correspond has been approached from another angle, that of the
principles according to which the relations of the symbols to one
another are organized. In terms of such 'structuralist' analysis, of
which the most distinguished modern exponent is Claude Levi-
Strauss, the symbols of any religious system conform to a grand
logical design, of which those who use and believe in the symbols are
unaware, and which cannot be perceived by ordinary observational
methods of anthropology alone. In such a theoretical scheme the
'power' of the symbols lies in the basic determinate structure of their
relationships, irrespective of what maybe thought about their genera-
tion from a mystical source.
It is clear that heuristically, structural analyses of this kind have
great force. But methodologically they depend upon assumptions
about identity and difference of human verbal and non-verbal be-
haviour, which are not always made explicit, and which are some-
times open to question. At times they are apt to denigrate less
406 SYMBOLS
highly abstract constructions as lacking an appreciation of 'inner

reality'. With the less sophisticated exponents of the structuralist
method this view is rather like the attitude of a Malay magical prac-
titioner towards the 'secret', the knowledge of the principles of the
inner processes of nature which gives him superiority over his rivals.
My aim in this chapter is less ambitious. Various anthropologists
have suggested that as part of investigation of religious symbolism,
Christian belief and practice are appropriate for scrutiny, and I
myself have given attention to them in brief studies of religious
mysticism and concepts of God (Firth, 1964,1968). Here I consider
what kind of statement is being made in a series of Christian sym-
bolic usages of a descriptive and ritual order: the concepts of the
ethnicity of Jesus; of the 'Sacred Heart' of Jesus (already referred to
in Chapter 6); and of the Eucharist. My prime concern in such in-
quiry is the substance of the symbolic expression - what is overtly
asserted by the description of the event or situation in a symbolic
way; the logical implications of this kind of assertion compared with
other assertions of a non-symbolic kind within the same general
context; the patterning of behaviour in respect of the symbolic
frame; and the sociological inferences that can be drawn from what is
said and done.
ETHNICITY OF JESUS
The problem of what I have called the ethnicity of Jesus is not

simply modern, though it has received most explicit attention in
modern times. It is essentially this: Jesus, as Saviour and Redeemer
of all mankind, according to the Christian statement, may be re-
garded as in some sense sharing in the characteristics of every dif-
ferent kind of man. Symbolically, He is the conceptual representa-
tion of every individual's humanity, with its fleshly weaknesses and
lusts, transmuted into a spiritual nature and so rendered capable of
participating in salvation through God's mercy. Jesus as God
symbolizes anthropomorphically the love, compassion and mercy
which man feels to be necessary to save him from his baser self; He
also symbolizes the protection against Divine judgement - which in
turn symbolizes the verdict of man's own inner logic in reflecting on
his shortcomings. But Jesus as Man represents the object of these
emotional and moral attentions. So, many times in the history of

Christianity there have been tendencies to portray Jesus as man of a
kind like unto those men who worshipped Him.
In the attempt to establish as direct a relation as possible between
Jesus and individual men, the language in which He is held to have
spoken is rendered into the vernacular. This was so even before
the Catholic church allowed the vernacular Mass; there was exposi-
tion in the local languages. It is intelligible then that the physical
form in which Jesus is conceived to have lived should also have
been rendered into local ethnic terms.
In the West, for much more than 1,000 years, from early Christian
Roman mosaics to Manet's Mocking of Christ, Jesus has been de-
picted as Caucasian in physical type. The problem of His actual
physical semblance has been learnedly debated-was He fair or
dark, or as some would have it, red-haired? Was He bearded?*
Serious scholarship does not claim any authentic portraits of Christ,
and an ingenious suggestion by Sir Wyke Bayliss reinforces the
cogent reasons for the lack. Bayliss argued that in the days of perse-
cution in the early Christian church it was not safe for Christians to
declare their faith openly by bearing upon their persons the portrait
of their Master. So 'the natural alternative was symbol'. A picture of
a youth carrying a lamb across a stream represented Christ the Good
Shepherd; a beardless Orpheus playing a lyre in the midst of wild
beasts and attracting them was a reminiscence of the power of Christ
to sustain the martyrs in the arena. 'These were the symbols - safe
yet intelligible. But the essential condition of them was that they
should not bear the Likeness' (1905, 68). However that may be,
artists were free to symbolize Jesus in human form as in other forms,
and conventionalized Him in vaguely Mediterranean or even northern
European type, with pale skin colour, straight or wavy hair and
thin nose and lips.
This seems to have been the image which pictorially and other-
wise Christian missionaries offered to Asiatic, African and other
populations distant from the Mediterranean. But this image has been
if not rejected at least supplemented by artists working in such dis-
* Considerable literature has accumulated on this problem: for example P. Doncoeur, Le
Christ dans V'Art Franpais, 1939-1948; E. Senior, Portraits of Christ, Harmondsworth, 1940;
Engelbert Kirschbaum, S.J., Christus, Christusbild, in Lexikon der Christlichen Ikonographie,
*> 355-454, Rome, 1968.
408 SYMBOLS
tant lands and striving to express in representations of Jesus some

basic principles of common humanity, by giving them local facial
features. So, in fifteenth- or sixteenth-century bronzes of the
Crucifixion from the lower Congo, it is said that 'the face of the
Christ generally presents the negroid type'. Pictures and carvings of
the Son of Man by Indian, Chinese and African artists often show
Jesus with ethnic features appropriate to the artist's own group. A
well-known East African Crucifixion by Job Kekana depicts the
Christ with African features; a Chinese picture of Jesus as shepherd
by Tsui Hung-I shows Him with Chinese features. Of the painter
Lu Hung Nien it was said that his 'desire as a Christian artist has
been to depict our Lord as a Chinese Saviour surrounded by Chinese
people', when with those of other Chinese artists, his pictures were
shown at a pre-war exhibition at the Roman Catholic University of
Fu-jen in Peking. The justification for such treatment of Jesus in
portraiture is clearly its symbolic value. In a general defence of
Chinese form in Christian religious art Cardinal Faulhaber cited the
action of the eighth-century (Second) Council of Nicaea providing
that images should be venerated though not adored. Since it is the
spirit and not the appearance that is significant in Catholic ideology,
by implication the visage of the Christ may 'speak the language of
the time' provided that it expresses a genuine religious faith. In
analogous style Bishop Carey, introducing a volume of illustrations
of such portrayals, stated that the Incarnate Christ 'is ever expressing
Himself in beauty, truth, goodness and love, in human apprehensions
and hearts. He is expressing himself from within . . . Artists pierce
through incidents and trappings to the soul underneath... In
portraying incidents they catch a glimpse of Him who underlies all.
In this book they reveal Him, and we understand and adore'.*
But if Jesus was ever actually a human being, He cannot have dis-
played the physical traits of all races; He cannot have been both yellow-
skinned and black-skinned, had straight as well as frizzly hair. Some
of the portrayals must be metaphorical, representing Jesus con-
ceptually but not naturalistically; they cannot all at the same time
pretend to show what He actually looked like.
If it is only symbols that are being dealt in, then the ethnic image
* For material and opinions see: Cardinal Faulhaber, 1932; Walter J. Carey, 1939; S.P.G.,
The Life of Christ by Chinese Artists, 1948; Rob. L. Wanijn, 1950, 45.
of Christ may be allowed to vary according to the audience by
whom the representation is needed. But Western Christians, as
indeed the church itself after the Council of Nicaea, have been
slow to admit Christ-figures of all ethnic types to the same level of
iconic authenticity. Hence as well as aesthetic and religious statements
of Christ with the lineaments of Oriental or African Man but the
representative of common humanity, there has arisen the political
statement of an ethnic Christ. What may be called religious national-
ists, stringently determined to insist on the immediate relevance of
Jesus to their lives in a physical as well as a metaphorical sense, have
developed the 'Black Jesus' thesis, which rejects the validity of any
other ethnic type of Saviour for Blacks.
To many Christians in the Western world-black as well as
white - the assertion that Jesus is black has seemed naive, absurd,
perhaps shocking. It contravenes not only the pictures of Christ
traditionally available in mosaic, fresco, oils, etc.; it also is incon-
gruent with the setting of Christ's life given in the New Testament
story.* Granted the debatable nature of the evidence, assuming that
Jesus was an historical figure He must have been a Jew - though
Wagner in his Heldentum tract seems to have denied that Jesus was of
Jewish stock and claimed Him as a kind of archetypal Aryan. If
Jesus had been a Roman citizen it is presumably faintly possible that
He could have been of negroid stock, but most unlikely. So the choice
is between ethnicity and historicity: if Jesus is to be black, then it is
the metaphorical, symbolical Jesus, not the actual historical Jesus
who is being portrayed. But if Jesus is simply a conceptual figure -
as many people would maintain - then white Christians have no
monopoly in His physical type. The problem is much simplified if
Jesus is recognized as a symbolic figure who can be given any com-
plexion desired. This is what the black poet Countee Cullen did
forty years ago in his poem The Black Christ: he represented Jesus
taking the place of his brother at a lynching, assuming the form of the
threatened man as part of His Saviour's role in His daily Crucifixion.
In this moving poem it makes much more sense indeed for Jesus,
* The parallel problem has not arisen in Islam, as far as I know. Not only has religious
iconography, including portrayals of the Prophet, been severely restricted by Muslim theo-
logy; the few paintings depicting the Prophet show him as of broadly Arab or Persian type,
whereas a black face is given to Bilal (the first of the Muslims appointed to give the call to
prayer) because of his Abyssinian birth (T. W. Arnold, 1928, 97 and pi. XXII).
O
4io SYMBOLS
the representative of suffering humanity, to be black than white. It

would be logical too for this theme to be projected further, and Jesus
to be accepted as black in the modern Christology, since over the
last few centuries it is probably Blacks who might claim to have en-
dured more of human suffering than have Whites.* Unless of course
the Crucifixion should be regarded as a kind of spiritual tokenism, a
sacrifice on account of those who have sinned on an ethnic basis.
Then Blacks might think that since Whites have sinned most, as their
representative, Jesus should properly be white!
Leaving aside aesthetic representations, cults of Black Jesus have
taken two main forms. One is the quasi-historical assertion that
Jesus Himself, the original founder of the cult, was black. In the early
1920s under Bishop McGuire, colleague of Marcus Garvey, the
African Orthodox church had Black Madonna and child as a stan-
dard picture in the homes. The worship of a Black Christ was
openly advocated, and that of a White Christ spurned (C. Eric
Lincoln, 1961, 62). In a recent version, that of the Black Christian
Nationalists, it is held that the original Israelites were also black.
With this provocative view has been linked the idea of Jesus as the
Black Messiah, sent by God as a revolutionary leader to rebuild the
black nation Israel and liberate black people from white exploita-
tion. Shrines of the Black Madonna serve as temples for the cult {New
York Times^ 6 April 1970). The other form of the cult is a more
mystical one, of identifying a present-day black religious leader as
Jesus Himself. The former thesis conserves the historical dimension
at the expense of spiritual immediacy; the latter thesis is timeless and
ahistorical, but brings godhead into the midst of the congregation.
Each in a different way is a statement of need - that the religious
inspiration and fount of morality should be symbolically expressed
in a human form near to that of the worshippers.
But this perpetuates a dilemma for white Christians. God the
Father might conceivably materialize in any colour. But fixed into
humanity as God the Son, it is more difficult to admit alternatives to
Him. So orthodox Christians, assuming they believe in the historicity
* This seems to have been a line along which the thought of some black artists has de-
veloped. Since Christ symbolically represents the sufferings of mankind, he must have been
black to have had the qualities attributed to him. Hence in the attempt to create images with
which black people can identify, artists have included portrayals of Christ as a black man. To
some, 'Christ is a purely symbolic black voice*.
of Jesus, have a delicate choice: they might like to acquiesce on

symbolic and sympathetic grounds in the blackness of Jesus, but this
would mean flying in the face of tradition and the environmental
features as they are known; but if they stick to the whiteness of Jesus
all they can do is to understand the blackness of Jesus as metaphori-
cal, non-historical, a kind of amiable concession to rising black
nationalism. This can be even more exasperating as patronage to
earnest Christian Blacks, who may feel it intolerable to have to
accept a white Saviour - 'an effeminate white Christ'. In practice,
presumably many white Christians do not accept the blackness of
Jesus as either literal or metaphorical, and are shocked by the
claim.
This example illustrates two important points. Historically, many
religious symbols have been selected, consciously or not, for politi-
cal reasons. And the question whether something is to be classed as
a symbol or not is not just an academic one, but can be a matter of
great practical concern. The Black Christian Nationalists, it appears,
do not believe that Christ was black metaphorically, but literally; for
them His blackness was real, not just symbolic. I would argue that
some of the highest emotional loading can occur not where a thing is
admittedly only symbolic, but where its symbolism is either denied
or believed to be shared with a real participation in the thing sym-
bolized.
THE S A C R E D HEART
A most interesting example of merging of literal and metaphorical in

a religious symbolic presentation is in the so-called 'Sacred Heart
of Jesus' cult (for general description see Chapter 6). Here is a case
where something essentially personal, an intimate internal organ of
someone believed once to have lived as a man, has been given world-
wide currency as a symbol of love.
For a long while in the history of the cult a question of critical
significance was the exact nature of the organ to which devotion was
to be paid. The official view of the church at first was that it was the
symbolic character only of the Heart to which reverence should be
paid, not its physical character. The Cordicolism of those attached to
this cult of the physical Heart was attacked, especially by the
412 SYMBOLS
Jansenists, and defended, especially by the Jesuits. Special con-

fraternities arose, mainly in France, Germany and Poland to propa-
gate the cult of adoration of the actual heart of the Saviour. As the
iconography depicting the physical heart in various forms con-
tinued to develop, Pius XII clarified the issue by pronouncing the
physical heart to be authoritatively included in the object of the cult.
So conceptual or metaphorical heart and physical heart - usually
shown as a red pear-shaped object - were worshipped together as
symbolic objects, symbolic of the redemptive love of Christ, both
human and divine.
But a further issue is also of anthropological interest - the concern
of theologians for the appropriateness of the heart as a symbol for
this love. We know nowadays that the heart is not the seat of the
emotions. But there is little doubt that in the seventeenth century
Marie Alacoque believed that it was, and the cult grew up on this
assumption. Yet within the church there were doubts about ac-
cepting this widening gap between physiology and faith. At an
early date a liturgical feast proposed in Rome failed to gain accept-
ance because the devotion was presented as based upon the heart as
the principle and organ of love. A petition of 1765 was accepted only
after it omitted what was termed the 'objectionable explanation'. In
the end faith - and the cumulative force of popular enthusiasm - won
out over physiology. The physical heart of Christ was not only
accepted as the symbol of his love, but also accepted as the proper
symbol. The New Catholic Encyclopaedia states authoritatively:
'Although every part of Christ's sacred humanity is worthy of the
strict adoration due to God alone, the heart is singled out for a special
devotion because of its inherent symbolism'. The natural fleshly
heart of Jesus is the 'true natural symbol' of his love.
The authorities do seem a bit uneasy on this point. It is recog-
nized that intellectually and emotionally, as distinct from blood
circulation, the heart no longer is correctly regarded as the seat of the
inner life of man. Yet it is still held that the special object involved in
every form of cult rendered to Christ's humanity has somehow a
correspondence on the spiritual side which is 'proper' to i t - a
sensible element that has some intelligible connection with the
psychological element, as the Encyclopaedia puts it. So Pius XII
called the physical heart of Christ a 'natural symbol' of His threefold
SYMBOL AND SUBSTANCE 413
love. And to the adoration of this redemptive love are added devo-
tional acts that it is said 'spring from the character of this special
object: imitation of the virtues of Christ's heart...'. The argument
is in fact a bit apologetic. It is admitted that neither the Scriptures
nor the Fathers of the church expressly refer to the physical heart of
Jesus as the symbol of his love. But, it is argued, they do explicitly
declare that Christ has a true and integral human nature, and hence
(presumably by implication) a heart upon which his entire affective
life exercises a real physical influence. 'This real connection between
the physical heart and the affective life provides the basis for the
natural symbolism of the heart in respect to love.' But as anthro-
pologists we should notice that this is really a reversal of the classical
view of the relationship: the symbolism is treated as natural not
because the heart is the seat of the emotions, but because emotions
affect the heart.
But this form of argument could have unexpected results if
logically pursued. In effect what we are seeing is the momentum of
a symbolic recognition* - the Catholic church has been unable to
make the switch from heart to brain as symbol, as physiological
knowledge has altered, so it is still the Sacred Heart and not the
Sacred Head that is worshipped. (Polynesians have been more dis-
cerning; they have made the head and not the heart taboo.) What we
really have is a case of concordance between liturgy and popular
thought - the church's symbolism of the Sacred Heart of Jesus is in
line with the popular metaphors which still talk of hearts aching and
hearts breaking, and loving people with all one's heart. It is in line
with Western popular distribution of symbolic meanings attaching
to the body, that other organs of Christ have gone uncelebrated, f
There is no cult of the Holy Tongue, though this was the organ
chiefly responsible for proclamation of Jesus's message. If Christ
really had a 'true and integral human nature' there are other organs
as well as his heart upon which his affective life might have exercised
a real physical influence. One commentator (A. D. Howell-Smith,
1950, 542) has bluntly observed that Catholics would be horrified at
the adoration of the divine genitals; but notes that the worship of
* For analogy in the field of emotional relations see D. M. Schneider, 1968, 115.
t Otherwise too, considering the value placed upon Mary's virginity, one might have
expected a cult of adoration of the Holy Womb - the Holy Hymen could perhaps be sym-
bolized by the veil.
414 SYMBOLS
the litigant, the conventionalized symbol of the phallus of Shiva, by

pious Hindus is deemed perfectly natural. I think one can detect in
some recent Catholic writings a move to make sex more religiously
respectable, as part of a general 'with-it' trend in the church. (It has
been suggested, for instance, that nuns, in their physical purity,
might serve as sex symbols of an idealized kind. But a recent Vatican
decision to allow lay Catholic women not only to take a vow of
virginity but also to be consecrated as virgins by a bishop in the
same way as cloistered nuns, if they wish to dedicate their lives to
God, has been described as a re-affirmation of the significance of
chastity in a world obsessed with sex.) But so far neither tongue nor
phallus have appeared overtly in the Christian iconography; the
church has chosen to locate the sympathetic life of the Saviour in His
Heart alone. It is a clear demonstration of the social character of a
religious symbol, even when it is put forward as a 'natural' repre-
sentation. There is a definite element of arbitrariness in the selection
of religious symbols, but their arbitrariness is part of a system, with-
in which the operations of individuals may have galvanizing force,
and may be determined in part by relation to a wider, secular sym-
bolic system. Moreover, in the selection of themes for symbolic
representation, in an elaborately organized institution such as the
Catholic church, the emergence of a symbol into liturgical use is by
no means only a 'natural' process but a matter of prolonged con-
sideration, including manipulation by pressure groups, using
strategies akin to those in vogue in the secular world of politics.
THE SYMBOLIC RITE OF 'GIVING THANKS'
Before turning to the symbolism of the Christian Eucharist I con-

sider briefly a parallel from a more conventional ethnographic field.
The Tikopia in pagan times had a communion feast, known as the
'Sacred Food' - a term which nowadays they apply as Christians to
the Holy Communion. Male members of the clan assembled in their
temple, led by their chief, and celebrated their yam harvest by each
man eating ritually a cooked yam tuber straight from the oven. The
rite was also called the Hot Food because it was in part a com-
petitive display to see who would be the first to bite off and swallow
a piping hot morsel of the vegetable. The rite had peculiar signifi-
cance for two reasons. One was that the yam tubers were regarded in
totemic alignment as symbolizing the supreme god of the clan -
indeed, the vegetable was termed the 'body' of the god. The other
reason was that the god himself was believed to attend the rite in
person, to be actually present in the body of the presiding chief -
looking out through human eyes at the performance which he was
believed to have instituted in times long past. The man who first
swallowed the scalding hot morsel of yam was thought in a vague
way to stand well in the graces of the god for the ensuing season (see
Firth, 1967a, 152-60; 1971 (1951a), 227-32).
The physical act of eating the yam was regarded as consuming the
body of the god in the presence of the god himself, and this was
regarded as symbolical of a wider set of relationships between man
and the forces and objects of nature. The Tikopia did not engage in
any elaborate analysis of the exact processes thought to be involved in
consuming both the yam and the body of their god. They regarded
identification of yam with god's body, and chief with god's presence,
as conforming to what they believed were the capacities of their
deities, not calling for searching explanation. Questions of timing -
when the god actually entered or left the chief's body - or of dis-
tribution of qualities - how god and vegetable could co-exist in the
same material form - did not seem to worry the Tikopia.
Some exploration of this problem in another context has been
given in an inquiry of mine into problems of identification in primi-
tive religious thought (Firth, 1966a); and also by Audrey Hayley
(1968) in an examination of the 'symbolic equation' in the Nuer
substitution of a cucumber for an ox in sacrifice - which I have also
referred to in a study of organizational problems of offering and
sacrifice (Firth, 1963). In the course of her analysis Audrey Hayley
has distinguished two meanings of symbol: as 'standing for', 'being
a representation of; and as equivalent to the concept of 'is', and
identified with the referent. As an example of the latter is taken
Evans-Pritchard's citation of the Nuer view that the crocodile is not
just a representation of God; the creature is God. As he put it,
'symbol and what it symbolizes are fused'. But if to the Nuer the
crocodile was a god, or a portion of the personality of God, in all
circumstances, then it was not a symbol. It was the anthropologist
who by classing it as a symbol, invented the fusion. But if, as seems
416 SYMBOLS
plausible, the Nuer believed the crocodile to be God in some cir-

cumstances and not in others, or God in some respects or qualities
and not in others, symbol and object were conjoined, and fused in
only a limited sense. That a symbol comes to take on some of the
qualities of what it symbolizes - for example sacredness - is a com-
mon phenomenon, as also is the attribution to a material object of
being both itself in physical qualities and something else in spiritual
qualities. Leaving aside the question of whether we should best
speak of such phenomena as identification, conjunction, fusion or
(in Levy-Bruhl's term) as mystic participation, I would argue that
these do not constitute a logically different class from those of
symbolic representation - that the differences are those of degree and
aspect of identification.
Neither Tikopia nor Nuer intellectualized such a symbolic equa-
tion process. But we do have a more intellectualized parallel in the
Christian religious rite of Communion, the sacrament of the Euchar-
ist. The name Eucharist applied to the consecration and distribution
of the elements of bread and wine in the Christian rite refers to a
thanksgiving, an acknowledgement of favour, which was also the
Tikopia attitude in their pagan rite of consuming the body of their
god in the yam festival. Anthropologists and psychologists, for
example B. Malinowski, E. O. James, C. G. Jung, Lloyd Warner,
Mary Douglas, have already examined various aspects of the
Eucharistic ritual and concepts. But I want to focus particularly on
the ideas about relation of symbol to substance found therein. In the
development of the cult of the Sacred Heart of Jesus, Pope Pius XII
was at some pains to stress that the faithful should pay their devo-
tions to the physical as well as to the metaphysical organ - or since
no physical heart of Jesus exists, to the idea of the physical organ. In
some ways a parallel concern with the physical has marked the
Catholic theory of the Eucharist in the Mass, though in this case the
material substances exist and are of human manufacture. In both
cases it would seem that the material symbol is retained in the in-
terests of emotional involvement in the cult or rite, and related by an
elaborate analytical construction to the spiritual referent, in the
interests of intellectual demands for an acceptable reasoned statement
on the phenomenon.
In the Catholic Mass, as is well known, the overt rite includes the
offering of bread (the 'Host', from kostia, a sacrificial victim) and
wine (in the chalice) at the altar, purification of them with incense,
and consecration of them with appropriate invocations. Since the
later Middle Ages in the Roman Catholic church, those of the
congregation who are ritually entitled to do so then partake of the
bread, the priest alone drinking the wine. (Other Christian churches
have modified forms of this Holy Communion service; in most the
congregation as well as the priest partake of the wine.) The ritual
paradigm for this communion is the Last Supper of Jesus with his
disciples. As recorded in the Scriptures (for example John, vi; I
Corinthians, xi) on the night before he was betrayed, Jesus took
bread, gave thanks, broke it and said 'This is my body for you; do
this in remembrance of me', and took wine likewise, describing it
metaphorically as his blood, and gave both to the disciples to be
consumed. There seems to be some doubt whether Jesus's words as
reported should include the actual injunction 'do this in remem-
brance of me' (cf. E. O. James, 1937, 127). But for all Christians the
general symbolism of the communion rite is the same: it com-
memorates the Last Supper; and it symbolizes the sacrifice of Christ,
His death on the Cross as the representative of humanity in the per-
son of God the Son, His redemptive love, and the personal link of
every Christian with Him, periodically renewed in solemn gathering
of the faithful. The symbolic interpretation of specific parts of the
rite varies considerably in the different sectors of Christianity, and in
the more liturgical-minded groups can be very elaborate indeed. So,
the breaking of the bread (Fractio) can be interpreted as symbolic of
Christ's death on the Cross; the mixing of water with the wine can
be seen as the symbolic commingling of divinity and humanity in
Christ's nature. The conjunction of elements, the Commixtio, when
bread is dipped in the wine, has been interpreted as a symbolical
representation of the Resurrection, when body and blood were
again reunited - or in another symbolism, the congregation, the
body of Christ, as the bread, is infused with the spirit of Christ, the
wine. Almost any action in the rite can be turned to use in the
symbolic system.
For many Christians the Eucharistic rite is primarily a com-
memorative one; it is done in memory of Jesus, on whom they think
solemnly and with gratitude while they are partaking of the com-
418 SYMBOLS
munion elements. For many others it is not only commemorative

but also symbolic, in that the elements themselves stand for the
bodily substances of Jesus and therefore allow the congregation to
absorb metaphorically part of the person of their Saviour and there-
fore to share in his sacrifice and in His eternal life. For them, the
words of the Gospel of St John are very meaningful, when Jesus
said that he was the 'living bread' that came down from heaven, and
if any man eat of this bread he shall live for ever. William Tyndale's
translation, now more than 400 years old, gives the sense of the
passage very vividly, as very many Christians have known it, as it
continues: 'And the breed that I will geve, is my flesshe, which I will
geve for the lyfe of the worlde. And the Iewes strove among them
selves sayinge: How can this felowe geve vs his flesshe to eate? Then
Jesus sayde vnto them: Verely, verely I saye vnto you, except ye
eate the flesshe of the sonne of man, and drinke his bloude, ye shall
not have lyfe in you . For my flesshe is meate in dede; and my
bloude is drynke in dede. He that eateth myflessheand drynketh my
bloude, dwelleth in me and I in h i m . . . ' (1534, Cambridge edn,
1938, 200). Protestants generally leave the matter at this level: when
Jesus died, his body was broken and his blood was shed; by eating
the bread and drinking the wine of the Communion they both re-
enact the Lord's Supper, and in simulacrum make contact with his
body and blood from the Crucifixion, with its spiritual implications.
Many of them are aware of the more general ritual significance in
other contexts of breaking of bread and eating together as a sign of
friendship, and of mutual drinking of wine as a pledge. Some too
may know that the realism of the Gospel language of eating probably
refers to early ideas of flesh and blood as vitalizing agents in mystery
cults, to ideas of feeding upon Christ as a necessary means of union
with the substance of the Saviour as a means of escape from the
material world (cf. James, 1937, 126 ff.; McGiffert, 1932, 32, 47;
Loisy, 1948, 251-2; Warner, 1961, 314-39).
C O N C E P T OF S A C R A M E N T A L SUBSTANCE
But for Catholics the interpretation is more complex, and for

anthropologists even more interesting. Catholics believe, and by the
rules of their church are bound to believe, that the Mass is not simply
a memorial, not simply a symbol, but a divinely efficacious sacra-
ment. The elements of bread and wine do not remain unchanged
throughout the rite, as Protestants hold, but are changed at the
moment of consecration into the actual flesh and blood of Christ -
'the sacramental presence of Our Lord's Body with Its natural
dimensions in the Holy Eucharist'. This transubstantiation, or
change from one substance into another, of the Eucharistic elements
has been part of Catholic belief for about 1,000 years, though a sub-
ject of bitter controversy until made an article of faith in 1215, and
re-affirmed against Protestant heterodoxy by the Council of Trent
three centuries later. Moreover, it is the teaching of the church that
not merely parts of Christ's bodily substance thus appear at the
Mass - His body and blood under the species of bread and wine -
but Christ Himself is involved as a whole personality, renewing His
sacrifice on the Cross, hence the doctrine of the 'Real Presence'. 'The
same Christ is contained and is immolated in an unbloody manner,
who, on the altar of the Cross, offered Himself once in a bloody
manner' - in the words of the Council of Trent.
All this has been the subject of a great deal of subtle metaphysical
and mystical interpretation by Catholic theologians and apologists.
It has also been given more sociological study, as I noted earlier.
E. O. James interpreted the Eucharistic mystery as a primitive life-
giving rite of sacrifice, akin in its basic notion of a dying to live, to
the pagan mysteries which preceded it. As part of the 'age-long
struggle for eternal life by sacramental means' he saw the Eucharist
in St Ignatius's terms as 'a medicine of immortality', parallel to the
conception of the Vedic sacred beverage soma (James, 1937,149-52;
cf. McGiffert, 1932,43). C. G. Jung (1958) saw the Mass as anthropo-
morphic symbol of a mystery which relates the human spirit to the
eternal; wherein nature, man and God are all sacrificed in the unity
of symbolic gift under the forms of bread and wine. The sacrifice of
God was seen by Jung as a kind of punishment. Ruthlessly pushing
his argument beneath the surface of the church's explanations, Jung
saw the 'natural logic' of the punishment in guilt, and God's guilt in
the fact that as creator of the world he was inadequate and therefore
had to transform himself by ritual slaying from a concrete to a
spiritual lord. Malinowski used the Mass (1936,47-50) as an illustra-
420 SYMBOLS
tion of the relation between myth and miracle. In the transubstantia-

tion of the sacrament, the miracle, supernatural realities are created
by ritual acts. In the stories of the origin of the Mass, the myth, there
is the affirmation of primeval miracles. For Mary Douglas, too, the
doctrine of the Eucharist is a demonstration of belief in efficacious
power, the rite is an effective field of change and instrument of
change (Douglas, 1970,46-9). The crux of the doctrine is that a real,
invisible transformation has occurred when the priest has recited the
sacred words, and that the eating of the consecrated host, in which
Christ the Lord is present, has saving efficacy, a redemptive force as
a channel of grace. But whereas Malinowski despite his Catholic
upbringing, was discussing the Eucharist as an external observer,
Douglas, seemingly as a matter of personal conviction, holds that its
full meaning involves magical and sacramental efficacy, and is scorn-
ful of her co-religionists who are insensitive to this meaning.
I am not here concerned with these problems. What I want to do
is to consider particularly some aspects of the logical forms in which
the symbolism is couched, and what in effect is being asserted by
what is being said.
In the traditional Catholic doctrine, the Mass is a true sacrifice of
Christ which performed once in blood on the Cross, is perpetuated
without blood in mystical manner on the altar, and communicated to
the faithful under the appearance of bread and wine. Anthropologi-
cally, the use of bread and wine is intelligible as it was to John of
Damascus in the eighth century-that familiar objects give reas-
surance, whereas the unfamiliar and unusual would be repellent (see
McGiffert, 1932, 324). And from the point of view of the solemnity
of the rite, the Catholic doctrine of transubstantiation and the Real
Presence may be more telling than the Protestant idea of simple
analogy. To imagine that one has on one's tongue a portion, how-
ever minute, of the flesh of the living God, who is present at the
rite, is an awful thought, which should have its effect upon conduct.
As a matter of ritual organization, then, the transubstantiation doc-
trine makes good sense; it aims at bringing the sacred home per-
sonally to every worshipper. With Catholic Christian as with pagan
Tikopia, the belief in ingestion of sacred substance reinforces the
respect for the rite.
But the traditional Catholic doctrine has certain logical implica-
tions to which theologians have given attention. To assert that

bread becomes flesh, and wine becomes blood, not by digestive pro-
cess slowly but by ritual process instantly, when a rite of sacraliza-
tion is performed by a priest, calls for explanation by any thoughtful
person. That symbol has become referent introduces a concept
which has to be fitted into the system of ideas. Malinowski implied
that the Eucharistic explanation of identity of symbol and referent
is a case of inference from general assumption, and not of a separate
mode of thought. He has pointed out that a theologian would not
give the title of miracle to this process, since it is regarded as falling
under the law of the nature of the supernatural, and not occurring in
terms of a tangible, concrete sensuously appreciable working of
supernatural force. The theological explanation indeed tries very
carefully to distinguish what it is that actually alters, in order to
present a plausible form of argument to the faithful who may ques-
tion the meaning of the rite. So a popular exposition of the mystery
of the Eucharist by a Dominican (traditionally, guardians of doc-
trine) states: 'As God, Christ is everywhere; but His Sacred Body
and precious Blood are not everywhere. They are in the Blessed
Sacrament under the phenomena of bread and wine But the
phenomena or sensible qualities of bread and wine undergo no
intrinsic change when the substances underlying them pass, at the
words of consecration, into the Body and Blood of Christ. They
remain as they were before consecration' (O'Neill, 1933, 64-5). So,
the communicant does not feel the taste of blood and meat upon the
tongue. Hence, the accusation of cannibalism, with analogies of
totemic feasting, is rejected as superficial. So too are notions of
aesthetic shock avoided. This is done by drawing a distinction
between accident, the external appearance and properties, and sub-
stance, the inner part. This distinction between externals and inner
reality is of course common in many contexts, and in Western
religious and philosophical thought has been identified going back
through Thomas Aquinas to Aristotle. But what is uncommon, said
to be unique, is that the inner conceptual reality should suffer a
physical and not merely a conceptual change. Transformation is
indeed not the correct word, since it is substance and not form which
is believed to alter. And Catholic theologians insist on the complete-
ness of the change. 'Transubstantiation means the conversion of the
422 SYMBOLS
bread and wine into the Body and Blood of Christ, in such a com-
plete manner that none of the substance of the bread or wine remains,
but only the species thereof... It implies t h a t . . . Our Lord's
Sacred Body comes, and is truly and really t h e r e . . . ' (compare
O'Neill, 39-40). The concept means basically that the accidents or
qualities of the bread and wine, which are perceptible to the senses,
continue to exist apart from, and without the support of, the sub-
stance of the bread and wine, which ceases and is replaced by the
substance (but not the sensory qualities) of Christ's body and
blood.
Various anthropological issues of interest arise from this concept.
It has certain parallels over a wide ethnographic field. Pagan Tikopia,
for instance, believed that every religious offering had in addition to
its material stuff, visible and palpable, an immaterial counterpart,
invisible and impalpable. They regarded this as in part a matter of
demonstration, as when fresh taro plants set out as offering on a
grave wilted; they argued that the immaterial counterpart of the
plants had been taken away by the spirits, leaving the solid part of
the offering behind. In ordinary terminology one might speak, as
I have been accustomed to do, of the visible part of the offering as
the substance, and the invisible counterpart as the essence. But in
Catholic terminology one might speak of the visible part as the
appearances (or accidents or species) and the invisible part as the
substance. Now both the pagan Tikopia and the Catholic theologians
have been faced by the same problem: how if the substance (essence)
be withdrawn can the appearance sustain itself? The Tikopia view
was less abstract - the material stuff suffers - but they bothered little
about the problem, attributing the whole phenomenon to the work
of spirits. The Catholic view has been to regard the persistence of
the bread and wine when deprived of their essential substance as a
matter altogether above the order of nature - 'part of the mystery
of the Real Presence which we do not pretend to fathom', as our
popular account puts it. The explanations are very similar.
In keeping with the sacred character of the Eucharistic rite, a
number of ritual precautions have been traditionally observed in
regard to it. The fitness of the priest to consecrate the Host, and the
fitness of the celebrant to receive it have been naturally the subject of
considerable regulation. And since partaking of the Eucharist means
communing not only with Christ but also with fellow Christians in
a common body, views have been expressed on who should be
allowed to share in the celebration. More specific ritual procedures
have also been closely specified. Traditionally, the wafer of un-
leavened bread has been placed by the priest on the tongue of the
celebrant; it has been taboo for the celebrant to touch it with his
hand. In some modern Catholic churches it has now been allowed for
a celebrant to take the host in his hand and place it in his mouth, and
an apostolic constitution issued by the Pope in 1953 relaxed some of
the rules for fasting before Communion. But while the sacred
proprieties seem capable of modification, an attempt is made to
preserve the sensory message of the Eucharist. Some modern
Catholic circles find it very important that celebrants shall still feel,
taste and smell bread and wine; that no violence be done to the
senses. A recommendation in a new Roman Catholic missal was that
the Host received by laymen during Holy Communion should
'really appear to be food'. In the New York metropolitan area in
1970 a new order of the Mass tried to bring the sacred rite closer to
the people inter alia by using real unleavened bread instead of the
former round small communion wafers that quickly dissolved on the
tongue without chewing. The idea of the real bread was to emphasize
that the rite was not only a memorial to the sacrifice of Christ but
also a communal meal. But there were technical difficulties. The
church requires the bread to be unleavened, and without yeast it is
hard to get the bread to rise. A suggestion to use Jewish matzoth as
in Passover, understandably, met with little favour. For some time
convents which made communion host experimented. Wafers made
from whole-wheat flour were found to be more tasty, but to
crumble easily; and the production change-over required to make
them needed new equipment that could be expensive for convents on
tight budgets. A large commercial concern in this field began
marketing 'unleavened altar bread', tiny baked cubes that could be
chewed; but some priests complained that these 'looked like break-
fast food' (patent commercial cereal). (See New York Times, 23
March 1970; Chicago Daily News, 16 November 1970.) So the search
for new recipes may still proceed. In parallel style, outside the
Catholic range, another concession to sensory elements occurred in
some Nonconformist circles which stressed teetotalism. Spurred on
424 SYMBOLS
by tales of reformed alcoholics who relapsed at their first sip of

communion wine, some generations ago Methodists and others
began to use unfermented grape juice to symbolize the blood of the
Saviour.
With all such attention to pragmatic detail, as part of sacred ritual,
anthropologists are quite familiar. The exegetical apparatus of the
Catholic Eucharist is, however, rather special in its care for the
logical proprieties. The doctrine of the sacramental presence of
Jesus 'under the veils of bread and wine' has produced not merely a
distinction between two kinds of reality - that which is superficial
and that which is underneath - but also between two kinds of per-
ception corresponding to these. Moreover, the validity of the thesis
of transformation of inner substance, while held to rest ultimately on
faith and revelation, is defended vigorously by reason as far as
possible. It is argued: that the doctrine of transubstantiation has
foundation in Scripture; that the church, divinely endowed, has the
hereditary title, which she can prove, to pronounce authoritatively
upon the subject; that those who question the doctrine are trying to
measure Christ by a human standard, and cannot see that they are
substituting their individual judgment for His; that it is ludicrous to
suppose that the whole Christian world has erred for 2,000 years in
the interpretation of what Christ meant. It is claimed not only that
the doctrine is true, but that it is also reasonable and consistent;
that Catholic philosophers and theologians have subjected it to tests
of severe intellectual analysis in order to discover its relation to the
eternal principles of thought reflected in the human mind. So
Catholic theologians draw as legitimate the conclusion that 'the
actual separation of an existing material thing from the phenomena
which naturally accompany it, is neither contradictory in conception
nor impossible in fact'. Though it may be difficult to conceive just
how the processes take place, the mind can conceive of the sub-
stantial presence of a material thing in a place without being subject
to the conditions of place. As imaginative constructs, such assertions
about the independent nature of 'substance' and 'accidents' are of
great interest. But to translate them into statements of fact requires
assumptions about divine power of another order of reality, accept-
ance of which requires faith and not simply an appeal to reason.
I now summarize the position of the symbolism of the Eucharist,
as seen by an anthropologist not a participant believer in the
cult.
It is clear, from historical research, by E. O. James and others,
and from comparative ethnographical evidence, that the Eucharistic
ideas of a communion obtained by partaking of the bodily substance
of the god are in conformity with a wide range of symbolic concepts
which unite sacred with secular in food, as it is consumed (cf.
Fortes, 1966b, 21). They belong too to that sector of concepts in
which it is not the food which becomes sanctified, but the sanctified
which becomes food. The special ritual values attaching to blood are
important here, in line with the magical significance of blood in many
religious systems.* But the blood and the body of Christ are pre-
sented in the Catholic Eucharist in an unusual way, as part of a very
complex and intricately organized symbolic system. They are pre-
sented as occurring without sensory evidence, in invisible, impal-
pable 'substance' (in another context one might have said 'essence')
only, beneath the 'veils' of bread and wine - that is, empirically,
simply as assertion. The assertion is defended elaborately, by appeal
to reason as well as to faith and authority. It is offered also, however,
as a challenge to reason, a demand for faith, as a mystery which is not
and cannot be fully comprehended. But by its very form, by main-
taining the continuity of appearance of bread and wine to the senses,
it avoids the more blatant contradiction of the senses, as well as the
aesthetic shock of oral confrontation with flesh and blood. The
Eucharistic doctrine gives an example of what may be called a super-
symbol. The wafer and the wine symbolize Jesus and the Crucifixion,
atonement and redemption, the union of the faithful in the body of
the church. But as Mary Douglas has maintained, symbolizing is not
understood in the Catholic tradition to have exhausted the meaning
of the Eucharist, which also is believed to have sacramental efficacy,
providing a channel of grace for redemption. The doctrine of the
Eucharist is then a statement about power and the source of power.
It is an assertion that the symbol is not merely about power, but in
its inner substance is power.
Propositions of this kind are not capable of proof or disproof.
* Cf. the notion of the redemptive power of Christ's blood. Also the Church of England
prayer * Almighty God, our heavenly Father, who hast purchased to thyself a universal
church by the precious blood of thy dear Son . . . ' (Communion Collect in Ember Week,
Book of Common Prayer).
426 SYMBOLS
But they have sociological implications, including those of defining

group allegiance. The doctrine of transubstantiation has served the
Catholic church as a critical indicator of definition on various
occasions. So Berengar of Tours had to do penance for the scandal
of his heresy in the eleventh century after he had stressed the sym-
bolism of the Eucharist in his attempt to submit the mysteries of
faith to logical treatment by dialectic (Knowles, 1962, 94-5). And
the Council of Trent in the sixteenth century reasserted an earlier
article of faith about transubstantiation, and redefined the term, in
its effort to support the church in the Counter-Reformation. So
statements purportedly about bread and wine in their symbolic
relationships are also statements about social structures and institu-
tions. They are assertions in defence of established positions or
claims to attain such positions, on a pragmatic social level as well as
on a conceptual level. Primitive societies have such symbol markers
too, but are commonly less dogmatic about them; their critical tests
are more pragmatic.
CONCLUSION
For a variety of reasons anthropologists have recently become very

interested in symbolism. One of our major jobs has been identifica-
tion of alien symbols and interpretation of them. We have tried to
show the coherence, the 'logic' of such symbol systems, as concep-
tual entities, as systems of thought. I have followed this line, but
only part way. I have tried to show also how symbol systems are not
always monolithic, not always uniformly coherent; how thinking
about them is often inconsistent; how it may be personal as well as
conventional, fiercely defensive or experimental and progressive,
with new implications being seen and faced, and new attempts at
definition being made. An institution such as the Catholic church,
with nearly 2,000 years of history offers a most interesting field for
study of such processes.
In such symbol systems a lack of coherence, with gaps, modifica-
tions and inconsistencies, can occur because of the continual in-
trusion of pragmatic issues into the kinds of functions symbols are
made to serve. Despite the metaphorical language of some commen-
tators I do not think symbols can rightfully be described as actors,
operating in their own right; I think people invent them, acquire

them by learning, adapt them, use them for their own purposes. And
there tends to arise a gap between the results of symbolic action and
those of pragmatic or empirical action. It is not a matter of which is
real; it is a matter of sorting out the different implications of each.
In a New Guinea cargo cult, for instance, people may build an air-
strip, or an aeroplane, as a symbolic means of obtaining the desired
goods, the 'cargo' from the white man's world. Their activity may
be classed as no less real to them than is our technological construc-
tion to us. But the results are of different order. To expect a material
aeroplane to land on a symbolic airstrip, or material rice and calico
to come out of a symbolic aeroplane, is a confusion of implications.
But such confusion of implications of symbolic action can occur,
and out of it can come much dissatisfaction and disturbance of social
relations.
Symbolic thought offers great advantages in speed and ease of
communication. And because of the allusiveness and indefiniteness
of symbols, possibly their ambiguity, they allow some greater
understanding of complex entities and action with them. Symbols
also allow flexibility in individual handling and interpretation. But
they may also have liabilities. The ambiguity of a symbol can be
made part of a defence mechanism, to support an argument about
the quality of knowledge it conveys. By allowing a range of alterna-
tives, a symbolic form of statement may inhibit thought about the
implications of what is being said. The operation of relationships at
symbolic level can lead to an avoidance of decision on a pragmatic
level. (It may be simpler to burn the national flag - or to charge
someone with desecrating the national flag - than to think out and
act upon the substantive issues at stake.)
This seems particularly so with political and religious symbols,
which tend to be credited with absolute value, to be treated as if they
were empirical referents. Symbolic relationships have sometimes
been described as having truth of value if not truth in fact. This is a
distinction which may not be capable of being maintained in meta-
physical analysis, and it is often denied by those who believe in the
significance of the symbols. But the attempt to claim a physical,
pragmatic reality for things which an outside observer would regard
as only symbols not only blurs relationships in the orders of reality;
428 SYMBOLS
it also has historically been a potent trigger for social conflict. To

put the point from another angle, problems arise not merely in the
vagueness of symbols but also in the denial of symbolic quality, or
claim of more than symbolic quality, to what is claimed as substance.
Symbolic concepts merit respect as a framework for organizing
experience, as a way of apprehending the world around one,
especially the world of human relations. Whether or not the asser-
tion that symbolism constitutes a unique and verifiable mode of per-
ception is justified (a claim I myself am not prepared to concede),
symbolism surely yields results that are aesthetically satisfying and
often operationally viable. To criticize a symbolic mode of approach
as 'unreal', 'unverifiable', seems to me then to be beside the point.
But what does seem extremely relevant and to be criticized, is the
demand often made for acknowledgement of a symbolism as uniquely
real. Such a demand for acquiescence in the absolute validity of the
symbol is a political demand, even when it speaks in the name of a
religious system. In all such contexts the primary problem for an
anthropologist is not to pronounce on 'ultimate reality'. It is to
examine the forms of symbolic statement, to try and understand the
system of ideas they express, the order of that system, and the effects
associated with the use of such symbolic concepts.
References
AIYAPPAN, A. 1944 Iravas and Social Change, Bull. Madras Gov.

Museum, n.s. gen. sec. V, 1. Madras: Government Press
AJISAFE, A. K. 1924 The Laws and Customs of the Yoruba People.
London: Routledge
ALLPORT, GORDON w. 1950 The Individual and His Religion.
New York: Macmillan
ANDERSON, J. N. D. 1959 Islamic Law in the Modern World. New
York: Universities Press
ANDERSON, M. D. 193 8 Animal Carvings in British Churches.
Cambridge: University Press
ARNOLD, SIR THOMAS w. 1938 Painting in Islam. Oxford: Claren-
don Press
ATTWATER, DONALD 196 5 The Penguin Dictionary of Saints.
Harmondsworth: Penguin
AUBER, L'ABB£ C. A. 1884 Histoire et Theorie du Symbolisme
Religieux. 4 v. Paris: Fechoz and Letouzey
AYER, ALFRED JULES 1946 Language, Truth and Logic, 2nd edn.
London: Gollancz
BACHOFEN, j . j . 1897 Das Mutterrecht. Eine Untersuchilng iiber

die Gynaikokratie der alten Welt nach ihrer religiosen und rechtlichen
Natur (1861), 2nd edn. Basel: Benno Schwabe
1925 Versuch iiber die Grabersymbolik der Alten (1859), 2n< ^
edn, foreword C. A. Bernoulli; appreciation Ludwig Klages.
Basel: Helding & Lichtenhahn
(See also Campbell, Joseph)
BAEUMLER, ALFRED 1965 Das Mythische Weltalter: Bachofens
Romantische Deutung des Altertums. Miinchen: C. H. Beck
429
43° REFERENCES
BAILEY, F. G. (ed.) 1971 Gifts and Poison: The Politics of Reputa-

tion. Oxford: Basil Blackwell
BATESON, GREGORY 193 6 Naven: A Survey of the Problems sug-
gested by a Composite Picture of the Culture of a New Guinea Tribe
Drawn from Three Points of View. Cambridge: University Press
(2nd edn with Epilogue, 1958)
(ed., introd.) 1961 Perceval's Narrative: A Patient's
Account of his Psychosis 1830-1832. Stanford: University Press
BAYLISS, SIR WYKE 1905 Rex Regum: A Painter s Study of the
Likeness of Christ from the Time of the Apostles to the Present Day.
London: S.P.C.K.
BEATTIE, JOHN 1966 'Ritual and Social Change/ Man, i, 60-74
1968 'Aspects of Nyoro Symbolism,' Africa, 38, 413-42
BEATTIE, JOHN and MIDDLETON, JOHN 1969 Spirit Mediumship
and Society in Africa. London: Routledge & Kegan Paul
B£GUIN, ALBERT 1939 VAme Romantique et le Reve, 2nd edn.
Paris: Jose Corti
BEIDELMAN, T. o. 1966 'The Ox and Nuer Sacrifice: Some
Freudian Hypotheses, 5 Man, 1, 453-67
BELL, SIR CHARLES 1928 The People of Tibet. Oxford: Clarendon
Press
BELSHAW, CYRIL s. 1965 Traditional Exchange and Modern
Markets. Englewood Cliffs, N.J.: Prentice-Hall
BENT, CHARLES N., s.j. 1967 The Death-of-God Movement. West-
minster, Md: Paulist Press
BERNDT, RONALD M. 1951 Kunapipi: A Study of an Australian
Aboriginal Religious Cult. Melbourne: F. W. Cheshire
1952 Djanggawul: An Aboriginal Religious Cult of North-
Eastern Arnhem Land. London: Routledge & Kegan Paul
BERNOULLI, c. A. 1924 Johann Jakob Bachofen und das Natur-
symbol. Basel: Benno Schwabe
1925 Die Psychologie von Carl Gustav Carus und deren
geistesgeschichtliche Bedeutung. Jena: Eugen Diederichs
BERREMAN, GERALD 1970 Review of Dillon, op. cit., American
Anthropologist, 72, 867-9
BERRY, JACK 1965 Intro, to B. Malinowski, Coral Gardens and
their Magic, 2, vii-xiii. Bloomington: Indiana University Press
REFERENCES 431
BIRDWHISTELL, RAY 1952 Introduction to Kinesics. Louisville:
University of Louisville Press
1970 Kinesics and Context. Philadelphia: University of
Pennsylvania Press
BIRNBAUM, NORMAN 1955 'Monarchs and Sociologists: A Reply
to Professor Shils and Mr Young/ Sociological Review, n.s. 3,
5
~23
BLAU, PETER 1964 Exchange and Power in Social Life. New York:
Wiley
BLYTHE, WILFRED 1969 The Impact of Chinese Secret Societies in
Malaya. London: Oxford University Press
BOAS, FRANZ 1927 Primitive Art, Inst, for Samm. Kultur-
forskning, ser. B., viii. Oslo: H. Aschehong
BOHANNAN, P. 1964 Intro, to Tylor, Researches into the Early
History of Mankind (1865), abr. from 3rd edn. Chicago: Univer-
sity Press
BOTT, ELIZABETH 1971 'Psychoanalysis and Ceremony' (1968),
in The Interpretation of Ritual, ed. J. S. LaFontaine. London:
Tavistock, 205-37
y
BOUDET, MARCELLIN 1874 Les Conventionnels d Auvergne: Du-
laure. Paris: Auby
BOUTEILLER, MARCELLE 1950 Chamanisme et Guerison Magique.
Paris: Presses Universitaires
BOWRA, B. M. 1943 The Heritage of Symbolism. London: Macmillan
BRANTL, GEORGE (ed.) 1967 Catholicism. New York: Washington
Square Press
BRETON, ANDR£ I936 What is Surrealism! London: Faber
BROWN, J. A. c. 1961 Freud and the Post-Freudians. Harmonds-
worth: Pelican
BURKS, ARTHUR w. 1949 'Icon, Index and Symbol/ Philosophy
and Phenomenonological Research, 9, 673-89
BURRIDGE, K. o. L. i960 Mambu: A Melanesian Millennium.
London: Methuen
1969 New Heaven New Earth: A Study of Millenarian
Activities. Oxford: Basil Blackwell
CAMERARIUS, JOACHIM (YOUNGER) I 668 Symbolorum et Emblema-

tum Centuriae Tres. Moguntiae: L. Bourgeat
432 REFERENCES
CAMPBELL, JOSEPH 1967 Intro, to Myth, Religion and Mother

Right. Selected Writings of J. J. Bachofen, trans. Ralph Manheim,
Bollingen ser. 84. Princeton: University Press
CANEPA, LUIS 1943 Antecedentes Hisotricos y Tradicionales de los
Simbolos Argentinos. Buenos Aires: Talleres Graficos Linari
1953 Historia de los Simbolos Nationales Argentinos.
Buenos Aires: Albatross
CAREY, BISHOP WALTER J. 1939 Intro, to Son of Man: Pictures
and Carvings by Indian, African and Chinese Artists. London:
S.P.G.
CARUS, CARL GUSTAV 1846 Psyche: Zur Entwicklungsgeschichte
der Seele. Pforzheim: Flammer & Hoffman
1853 Symbolik der Menschlichen Gestalt: Ein Handbuch %ur
Menschenerkenntnis. Leipzig: Brockhaus
1857 Ueber Lebensmagnetismus und iiber die magischen
Wirkungen uberhaupt. Leipzig: Brockhaus
CASSIRER, ERNST 1953-7 Philosophy of Symbolic Forms, 3 v.,
trans. Ralph Manheim. New Haven: Yale University Press
1944 An Essay on Man: An Introduction to a Philosophy of
Human Culture. New Haven: Yale University Press
CASTANEDA, CARLOS 1969 The Teaching of Don Juan: A Yaqui
Way of Knowledge. New York: Ballantine
CATHOLIC UNIVERSITY OF AMERICA 1967 New Catholic Encyclo-
paedia. New York: McGraw-Hill
CHARBONNIER, G. (ed.) 1969 Conversations with Claude Levi-
Strauss, trans. John & Doreen Weightman. London: Cape
CLAPHAM, A. w. 1934 English Romanesque Architecture after the
Conquest. Oxford: Clarendon Press
COHEN, ABNER 1969 'The Analysis of the Symbolism of Power
Relations/ Man, 4, 215-35
1971 'The Politics of Ritual Secrecy,' Man, 6, 427-48
COLANGELi, ORANZO 1965 Simboli e Bandieri nella Storia del
Risorgimento Italiano. Bologna: Patron
COLLINGWOOD, R. G. 1925 Outlines of a Philosophy of Art.
London: Oxford University Press
COPLESTON, F . c. 1961 Mediaeval Philosophy (1952). New York:
Harper
REFERENCES 433
COULTON, G. G. 1958 Mediaeval Faith and Symbolism (1928).
New York: Harper
CRAWLEY, ERNEST 1902 The Mystic Rose: A Study of Primitive
Marriage. London: Macmillan
CREUZER (GEORG) FRIEDRICH 1806 Idee und Probe alter Symbolik.
1810 Symbolik und Mythologie der alten Volker besonders
der Griechen (2nd edn, 2 v. 1819-22; 3rd edn, 4 v. 1837-43).
Leipzig: C. W. Leske
CRITCHLEY, MACDONALD 1939 The Language of Gesture. London:
Edward Arnold
DANBY, HERBERT 193 3 The Mishnah. London: Oxford Univer-

sity Press
DARWIN, CHARLES 1872 The Expression of the Emotions in Man
and Animals. London: Murray
DAVIS, JOHN 1972 * Gifts and the U.K. Economy/ Man, n.s. 7,
408-29
DE RAEDT, JULES 1964 'Religious Representations in Northern
Luzon/ St Louis Quarterly', 2, 315
DESHEN, SCHLOMO 1970 'On Religious Change: The Situational
Analysis of Symbolic Action/ Comparative Studies in Society and
History, 12, 260-74
DILLON, WILTON S. 1968 Gifts and Nations: The Obligation to
Give, Receive and Repay. Hague: Mouton
DODGE, COL. R. I. 1883 Our Wild Indians. Hartford, Conn.:
A. D. Worthington
DONCOEUR, P. 1939-48 Le Christ dans VArt Franfais. Paris: Plon
DOOLITTLE, REV. JUSTUS 1876 Social Life of the Chinese, 2 v.
New York: Harper Brothers
DOUGLAS, MARY 1966 Purity and Danger. London: Routledge &
Kegan Paul
1970 Natural Symbols. London: Cresset
1972 'Deciphering a Meal', in Myth, Symbol and Culture,
Daedalus, Winter 1972, 61-81
DOUGLAS, MARY (ed.) 1970 Witchcraft Confessions and Accusa-
tions, A.S.A. Mon. 9. London: Tavistock
[DULAURE, J. A.] 1786 Pogonologia: or a Philosophical and His-
torical Essay on Beards, trans, from Fr. Exeter: R. Thorn
434 REFERENCES
DULAURE, J.-A. 1805 Des Divinites Generatrices ou du Cuke du

Phallus. Paris: Dentu
1805 Des Cultes qui ont Precede et Amene Vldolatrie ou
VAdoration des Figures Humaines. Paris: Fournier
1825 His wire Abregee de Differ ens Cultes, 2 v. (combined
2nd ed. reissue from 1805). Paris: Guillaume
1905 Des Divinites Generatrices che^ les Anciens et les
Modernes - avec un chapitre complementaires par A. van Gennep.
Paris: Societe de Mercure de France [A privately printed Eng.
trans, (by A.F.N.) was issued by Panurge in New York, 1933.]
DUNBAR, HELEN FLANDERS 1929 Symbolism in Mediaeval Thought.
New Haven: Yale University Press
DURKHEIM, £MILE 1912 Les Formes Elementaires de la Vie
Religieuse. Paris: Alcan
1926 The Elementary Forms of the Religious Life, trans.
J. W. Swain. London: Allen & Unwin
DURLIAT, MARCEL 1952 La Sculpture Romane en Rousillon, 2nd
edn, 1. Perpignan: La Tramontane
EFRON, D. 1941 Gesture and Environment. New York: King's

Crown Press
EGGAN, FRED and PACYAYA, ALFREDO 1962 'The Sapilada
Religion: Reformation and Accommodation among the Igorots of
Northern Luzon,' Southwestern J. of Anthrop., 18, 95-113
EHRMANN, JACQUES 1970 'Structures of Exchange in Cinna,9 in
Introduction to Structuralism, ed. Michael Lane. New York: Basic
Books, 223-47
ELIADE, MIRCEA 1964 Shamanism: Archaic Techniques of Ecstasy,
Bollingen ser. 76, trans. Willard R. Trask. New York: Pantheon
1969 Images and Symbols (1952), trans. P. Mairet. New
York: Sheed & Ward
ELKIN, A. P. 1954 The Australian Aborigines, 3rd edn. Sydney:
Angus & Robertson
EMBREE, JOHN F . 1946 A Japanese Village: Suye Mura. London:
Kegan Paul
£TIEMBLE, REN£ 1961 Le Mythe de Rimbaud, 2: Structure du
Mythe, new edn. Paris: Gallimard
REFERENCES 435
EVANS-PRITCHARD, E. E. 1937 Witchcraft, Oracles and Magic
among the Aânde. Oxford: Clarendon Press
1948 The Divine Kingship of the Shilluk of the Nilotic
Sudan. Cambridge: University Press
1956 Nuer Religion. Oxford: Clarendon Press
1962 Essays in Social Anthropology. London: Faber
1965a The Position of Women in Primitive Society —
London: Faber
1965b Theories of Primitive Religion. Oxford: Clarendon
Press
FAIRCHILD, HOXIE NEALE 1928 The Noble Savage; A Study in

Romantic Naturalism. New York: Russell & Russell
FARBENSTEIN, WOLF 1955 Romantische Einfliisse in der Geschichte
Psychologie im deuthschen Sprachgebiet: Carl Gustav Carus und
die neuere Psychologie. Zurich: Kammerdruck
FAULHABER, CARDINAL 1932 'L'Eglise et L'Art Religieux. Art
Chretien Chinois,' Collectanea Commissionis Synodalis, 5, Peiping,
496-508
FEIDELSON, CHARLES JR 19 5 3 Symbolism and American Literature.
Chicago: University Press
FERGUSON, ADAM 1767 An Essay on the History of Civil Society.
Edinburgh: Kincaid & Bell
FERGUSON, GEORGE 1961 Signs and Symbols in Christian Art.
FIELD, MARGARET J. i960 Search for Security: An Ethno-
psychiatric study of Rural Ghana. London: Faber
FINDLAY, j . N . 1963 Language, Mind and Value. London: Allen
& Unwin
FINGESTEN, PETER 1970 The Eclipse of Symbolism. Columbia,
S.C.: University of S. Carolina Press
FINLEY, M. 1. 1962 The World of Odysseus. Harmondsworth:
Pelican
FIRTH, j . R. 1957 'Ethnographic Analysis and Language with
reference to Malinowski's Views,' in Firth, Raymond (ed.), 1957
93 Xl8
~
FIRTH, RAYMOND 1929 Primitive Economics of the New Zealand
436 REFERENCES
Maori. London: Routledge (2nd edn, 1959. Wellington: Govt

Printer)
1936 We, The Tikopia: A Sociological Study of Kinship in
Primitive Polynesia. London: Allen & Unwin (repr. 1957)
1939 Primitive Polynesian Economy. London: Routledge
(2nd edn, 1965)
1943 'The Coastal People of Kelantan and Trengganu',
Geographical J., 101, 193-205
1951a Elements of Social Organisation. London: Watts
(3rd edn, 1971, Tavistock)
1951b 'Contemporary British Social Anthropology',
American Anthropologist, 53, 474-89
1954 'Anuta and Tikopia: Symbiotic Elements in Social
Organization/ J. Polynesian Society, 63, 87-131
1959 Social Change in Tikopia. London: Allen & Unwin
1963 'Offering and Sacrifice: Problems of Organization/
J. Royal Anthrop. Inst., 93, 12-24
1964 Essays on Social Organisation and Values, L.S.E.
Mon. Soc. Anthrop., 28. London: Athlone
1966a 'Twins, Birds and Vegetables: Problems of Identifi-
cation in Primitive Religious Thought,' Man, n.s. 1, 1-17
1966b 'The Meaning of Pali in Tikopia,' in In Memory of
J. R. Firth, ed. C. E. Bazell et al., 1966, 96-115. London: Long-
mans
1967a The Work of the Gods in Tikopia (1940), L.S.E.
Mon. Soc. Anthrop. 1 and 2. London: Athlone
1967b Tikopia Ritual and Belief. London: Allen & Unwin
1967c 'Themes in Economic Anthropology: A General
Comment,' in A.S.A. Mon. 6, 1-28. London: Tavistock
i967d 'Ritual and Drama in Malay Spirit Mediumship,'
Comp. Studies in Society and History, 9, 190-207
1968 'Gods and God: An Anthropologist's Standpoint,' in
The Humanist Outlook, ed. A. J. Ayer. London: Pemberton,
31-44
1969 'Tikopia Social Space,' Bijdragen, tot de Taal- . . . ,
125, 64-70.
1970a Rank and Religion in Tikopia. London: Allen &
Unwin
REFERENCES 437
1970b 'Reflections on Tikopia "Totemism",' Oceania, 40,
280-95
— — 1970c 'Postures and Gestures of Respect/ in Echanges et
Communications: Melanges offerts a Claude Levi-Strauss, ed.
Jean Pouillon and Pierre Maranda. Hague: Mouton, 188-209
1972 'Methodological Issues in Economic Anthropology/
Man, n.s., 7, 467-75
FIRTH, RAYMOND (ed.) 1957 Man and Culture: An Evaluation of
the Work of Bronislaw Malinowski. London: Routledge
FIRTH, RAYMOND and SPILLIUS, JAMES 1963 A Study of Ritual
Modification, Occ. Paper 19. London: Royal Anthrop. Inst.
FORGE, ANTHONY 1965 'Art and Environment in the Sepik,'
Proc. Royal Anthrop. Inst., 23-31
FORTES, MEYER 1945 The Dynamics of Clanship among the
Tallensi. London: Oxford University Press
1957 'Malinowski and the Study of Kinship,' in Raymond
Firth (ed.), 1957, 157-88; repr. as 'Malinowski and Freud',
Psychoanalysis and the Psychoanalytic Review, 1958, 45, 127-45
1966a 'Religious Premisses and Logical Technique in
Divinatory Ritual,' Philosoph. Trans. Royal Soc, B, 251, 409-22
1966b 'Totem and Taboo,' Proc. Royal Anthrop. Inst., 5-22
1970 'Ritual Festivals and Social Cohesion in the Hinter-
land of the Gold Coast' (1936), Time and Social Structure . . .
L.S.E. Mon. Soc. Anthrop., 40, 147-63. London: Athlone
FORTES, M. and EVANS-PRITCHARD, E. E. (eds) 1940 African
Political Systems. London: Oxford University Press
FORTUNE, R. F. 1932 Sorcerers of Dobu. London: Routledge
FRANKL, PAUL i960 The Gothic: Literary Sources and Interpreta-
tions through Eight Centuries. Princeton: University Press
FREUD, SIGMUND 1937 The Interpretation of Dreams, trans.
A. A. Brill, 8th German edn. London: Allen & Unwin
1952 A General Introduction to Psychoanalysis, trans. Joan
Riviere, intro. Ernest Jones. New York: Liveright
1963 The History of the Psychoanalytic Movement (1914),
intro. Philip Rieff. New York: Macmillan
1964 New Introductory Lectures on Psychoanalysis, trans.,
ed. James Strachey. New York: Norton
1965 Group Psychology and the Analysis of the Ego (1921),
438 REFERENCES
trans. James Strachey; intro. Franz Alexander. New York:

Bantam
FRIEDL, ALOIS 1968 Deutsche Staatssymbole. Frankfurt:
Athenaum
FRIEDRICH, PAUL 1966 'Revolutionary Politics and Communal
Ritual/ in Political Anthropology, ed. M. J. Schwartz et al.
Chicago: Aldine, 191-220
FRY, ROGER 1928 Vision and Design. London: Chatto & Windus
FRYKE, CHRISTOPHER 1929 Voyages to the East Indies, intro. C.
Ernest Fayle. London: Cassell
FUSTEL DE COULANGES [1955] The Ancient City (1864). Garden
City, N.Y.: Doubleday
GANTNER, JOSEPH and POB£, MARCEL 1956 Romanesque Art in

France. London: Thames and Hudson
GAUNT, WILLIAM 1962 Dictionary of Pictorial Art. London: Dent
GELLNER, ERNEST 1969 Saints of the Atlas. London: Weidenfeld
& Nicholson
GENNEP, ARNOLD VAN i960 The Rites of Passage (1909), trans.
Manika B. Vizedom and Gabrielle L. Caffee, intro. Solon T.
Kimball. London: Routledge & Kegan Paul
GIBB, H. A. R. 1947 Modern Trends in Islam. Chicago: University
Press
GIBB, H. A. R. and BO WEN, H. 1950-7 Islamic Society and the
West, 1 and 2. London: Oxford University Press
GLUCKMAN, MAX 1963 Order and Rebellion in Tribal Africa.
London: Cohen & West
GLUCKMAN, MAX (ed.) 1962 Essays on the Ritual of Social Rela-
tions. Manchester: University Press
GODELIER, MAURICE 1971 'Myth and History,' New Left Review,
69,93-112
GOFFMAN, ERVING 1967 Interaction Ritual. Garden City, N.Y.:
Doubleday
GOMBRICH, E. H. 1966 'Ritualized Gesture and Expression in Art,'
see Huxley, 393-401
1968 Art and Illusion, 3rd edn. London: Phaidon
GOODY, JACK 1962 Death, Property and the Ancestors. London:
Tavistock
REFERENCES 439
GORE, CHARLES 1901 The Body of Christ: An Inquiry into the
Institution and Doctrine of Holy Communion, 2nd edn. London:
Murray
GRENZMANN, WILHELM (ed.) 1949 G. Chr. Lichtenberg: Gesam-
melte Werke, 2 v. Frankfurt a/M: Holle
GRIVOT, DENIS and ZARNECKI, GEORGE 1961 Gislebertus Sculptor
ofAutun. London: Trianon
GUIGNIAUT, J. D. 1825-41 Religions de VAntiquite Considerees
Principalement dans leurs formes symboliques et Mythologiques
(trans. F. Creuzer, Symbolik und Mythologie, 2nd edn, rearranged,
developed and clarified). Paris: Treuttel & Wiirtz
HALL, EDWARD T. 1961 The Silent Language (1951). Greenwich,

Conn.: Fawcett
1968 'Proxemics,' Current Anthropology, 9, 83-95
HALLPIKE, c. R. 1969 'Social Hair,' Man, n.s. 4, 256-64
1971 The Konso of Ethiopia: A Study of the Values of a
Cushitic People. Oxford: Clarendon Press
HALSTED, JOHN B. (ed.) 1965 Romanticism: Problems of Defini-
tion, Explanation and Evaluation. Boston: D . C. Heath
HARRE, JOHN 1968 'A Matter of the Length of Hair: Beliefs and
Symbols in Social Studies,' National Education [New Zealand]
HASKINS, CHARLES HOMER 1958 The Renaissance of the Twelfth
Century (1927). New York: Meridian
HAUSER, ARNOLD 19 5 2 The Social History of Art, 2 v. New York:
Knopf
HAYLEY, AUDREY 1968 'Symbolic Equations: The Ox and the
Cucumber,' Man, n.s. 3, 262-71
HEER, FRIEDRICH 1962 The Mediaeval World, trans. Janet
Sondheimer. New York: Mentor
HERBART, j . F . 1824 Psychologie als Wissenschaft. Konigsberg
HERTZ, ROBERT i960 'The Pre-eminence of the Right Hand: A
Study in Religious Polarity,' in Death and the Right Hand, trans.
Rodney and Claudia Needham. London: Cohen & West
HIATT, L. R. 1967 'Nabokov's Lolita: A "Freudian" Cryptic
Crossword,' American Imago, 24, 360-70
HIRN, YRjo 1912 The Sacred Shrine: A Study of Poetry and Art of
the Catholic Church. London: Macmillan
440 REFERENCES
H0CART, A. M. 1970 Kings and Councillors, new edn, intro.
Rodney Needham. Chicago: University Press
HOEBEL, E. ADAMS0N 1966 Anthropology: The Study of Man.
New York: McGraw-Hill
HOGBIN, H. IAN 1940-1 'The Father Chooses His Heir: A Family
Dispute over Succession in Wogeo,' Oceania, n , 1-39
HOMANS, GEORGE c. 1950 The Human Group. New York: Har-
court Brace
1961 Social Behaviour: Its Elementary Forms. New York:
Harcourt Brace
HONIGMAN, J. J. 1959 The World of Man. New York: Harper
HOOTON, E. A. 1946 Up From the Ape, rev. edn. New York:
Macmillan
HORTON, ROBIN 1971 'African Conversion/ Africa, 41, 85-108
HOSTETLER, JOHN A. 1964 'The Amish Use of Symbols and their
Function in Bonding the Community,' / . Royal Anthrop. Inst.,
94, 11-22
HOTOMAN, ANTONIUS 1628 Pogonias Sive De Barha . . . (1586).
Rostochii: ex officina Ferberiana
HOWALD, ERNST 1926 Der Kampfum Creu\ers Symbolik: Eine
Auswahl von Dokumenten. Tubingen: J. C. B. Mohr
HOWARD, E. c. c. 1914 Minhaj et Talibin: A Manual of Muham-
medan Law according to the School of Shafi, by Mahiudin Abu
Zakaria Yahya ibn Sharif en Nawawi, trans, from Fr. of L. W . C.
Van den Berg. London: Thacker
HOWELL-SMITH, A. D. 1950 Thou Art Peter: A History of Roman
Catholic Doctrine and Practice. London: Watts
HO WELLS, WILLIAM 1947 Mankind So Far. New York: Doubleday
HULL, L. w. H. 1959 History and Philosophy of Science. London:
Longmans
HULME, F. E. n.d. The Flags of the World: Their History, Blazonry
and Associations. London: Frederick Warne
1891 The History, Principles and Practice of Symbolism in
Christian Art. London: Swan Sonnenschein
HURGRONJE, c. SNOUCK 1906 The Achehnese, 2 v. Ley den: Brill
HUXLEY, SIR JULIAN 1966 'A Discussion on Ritualization of
Behaviour in Animals and Man,' Philosoph. Trans. Royal Soc, B,
No. 772, vol. 251, 247-526 .
REFERENCES 441
JAMES, E. o. 1937 Christian Myth and Ritual. London: Murray

JAMESON, MRS 1900 Legends of the Monastic Orders as represented
in the Fine Arts. London: Longmans
JARVIE, I. c. 1967 The Revolution in Social Anthropology. London:
Routledge & Kegan Paul
JENYNS, SOAME 193 5 A Background to Chinese Painting. Lon-
don: Sidgwick & Jackson
JUNG, c. G. 1926 Psychological Types, trans. H. Gordon Baynes.
London: Kegan Paul
1958 Psyche and Symbol, ed. Violet S. de Laszlo. New
York: Doubleday
JUNOD, H. A. 1927 The Life of a South African Tribe, 2 v. London:
Macmillan
KANDINSKY, v. 1946 On the Spiritual in Art (1912), trans. Hilla

Rebay. New York: Guggenheim Foundation.
KATTENBUSCH, F. 1894-1900 Das Apostolische Symbol, 2 v.
Leipzig: Hinruchs
KEYSER, c. E. 1927 A List of Norman Tympana and Lintels, 2nd
edn. London: Elliot Stock
KIRSCHBAUM, ENGELBERT, s.j. 1968-71 Lexikon der Christlichen
Ikonographie, 3V. Rome: Herder
KNOWLES, DAVID 1962 The Evolution of Mediaeval Thought.
London: Longmans
KRETZMANN, NORMAN 1967 'Semantics, History of,' Encyclo-
paedia of Philosophy, 7. New York: Macmillan
KREWALD, ARTUR 1939 Carl Gustav Carus seine philosophischen
psychologischen und characterologischen Grundgedanken (Die Lehre
von der praktischen Menschenerkenntnis, 3, 2). Berlin: Bernard
& Graefe
KROEBER, A. L. 1901 'Arapaho Decorative Art,' American Anthro-
pologist, n.s. 3, 308-36
KUNSTLE, KARL 1928-36 Ikonographie der Christlichen Kunst, 2 v.
Freiburg LB.: Herder
KUPER, HILDA 1947 An African Aristocracy: Rank among the
Swaî. London: Oxford University Press
p
442 REFERENCES
LA BARRE, WESTON 1969 The Peyote Cult (enlarged edn). New

York: Schocken
LA FONTAINE, j . s. 1963 'Witchcraft in Bugisu,' in Witchcraft and
Sorcery in East Africa, ed. John Middleton and E. H. Winter.
London: Routledge & Kegan Paul
LANGEN, AUGUST 1962 K. Ph. Moritz 'Weg zur symbolischen
Dichtung,' Z.f dt. Philologie, 81, 169-218, 402-40
LANGENDOEN, TERENCE D. 1968 The London School of Linguistics:
A Study of the Linguistic Theories of B. Malinowski and J. R.
Firth. Cambridge, Mass.: M.I.T. Press
LANGER, SUSANNE K. 195 3 Feeling and Form: A Theory of Art.
New York: Scribner
LATTIMORE, OWEN s. 1934 The Mongols of Manchuria. New
York: John Day
1941 Mongol Journeys. New York: Doubleday
LAWRENCE, PETER 1964 Road Belong Cargo. Manchester: Univer-
sity Press
LAYARD, JOHN w. 1942 Stone Men of'Malekula, I. Vao. London:
Chatto & Windus
1944 The Lady of the Hare. London: Faber
LEACH, E. R. 1954 Political Systems of Highland Burma. London:
Bell
1957 'The Epistemological Background to Malinowski's
Empiricism,' in Firth (ed.), 1957, 119-38
1958 'Magical Hair,'/. Royal Anthrop. Inst., 88, 147-64
1961a Rethinking Anthropology, L.S.E. Mon. Soc. Anthrop.
22. London: Athlone
1961b 'Levi-Strauss in the Garden of Eden: An Examina-
tion of Some Recent Developments in the Analysis of Myth/
Trans. New York Acad. Sciences, ser. II, 23, 386-96
1965 'Frazer and Malinowski,' Encounter, Nov., 24-36
1966a 'The Legitimacy of Solomon: Some Structural
Aspects of Old Testament History,' European J. Sociology, 7,
58—101
1966b 'Ritualization in Man in Relation to Conceptual and
Social Development,' Philosophical Trans. Royal Soc, B, 251,
403-8
REFERENCES 443
1971 'The Heart of the Matter' (review of Titmuss, op.
cit.), New Society, 23 December
1972 'The Structure of Symbolism/ in The Interpretation
of Ritual: Essays in Honour of A. I. Richards, ed. J. S. La Fon-
taine. London: Tavistock, 239-75
LEBRA, TAKIE SUGIYAMA 1969 'Reciprocity and the Asymmetric
Principle: An Analytical Reappraisal of the Japanese Concept of
" O n " / Psychologia, 12, 129-38
LEE, DOROTHY 1959 Freedom and Culture. Englewood Cliffs,
N.J.: Prentice-Hall
LEIRIS, MICHEL 1948 La Langue Secrete des Dogons de Sanga
{Soudan Franfais). Trav. et Mem. Inst. d'Ethnologie, L. Paris:
Institut d'Ethnologie
1958 La Possession et Ses Aspects Theatraux che% les
Etkiopiens de Gondar. Paris: Plon
LEISINGER, HERMANN 1957 Romanesque Bronzes. New York:
Praeger
L£VI-STRAUSS, CLAUDE 1948 'La Vie Familiale et Sociale des
Indiens Nambikwara//- Soc. des Americanistes, n.s. i, 37,1-132
1949 Les Structures Elementaires de la Parente. Paris:
Presses Universitaires
1950 'Introduction a l'Oeuvre de Marcel Mauss/ Sociologie
et Anthropologic par Marcel Mauss. Paris: Presses Universitaires,
ix-lii
1957 'The Principle of Reciprocity/ in Sociological
Theory, ed. L. A. Coser and B. Rosenberg. New York: Mac-
millan, 74-84.
1958 Anthropologic Strueturale. Paris: Plon
1962a Le Totemisme Aujourd'hui. Paris: Presses Universi-
taires
1962b La Pensee Sauvage. Paris: Plon
1963 'The Efficacy of Symbols/ Structural Anthropology.
New York: Basic Books
1964 Le Cru et Le Cuit. Mythologiques, I. Paris: Plon
1969 The Elementary Structures of Kinship, ed. Rodney
Needham, trans. J. H. Bell, R. H. von Sturmer, rev. edn. Boston:
Beacon
1971 VHomme Nu: Mythologiques, IV. Paris: Plon
444 REFERENCES
L£VY-BRUHL, L. 1938 VExperience Mystique et les Symboles

chei les Primitifs. Paris: Alcan
— — 1949 Les Carnets de Lucien Levy-Bruhl, pref. Maurice
Leenhardt. Paris: Presses Universitaires
LEWIS, 1. M. 1971 Ecstatic Religion: An Anthropological Study of
Spirit Possession and Shamanism. Harmondsworth: Penguin
LEWIS, M. M. 1947 Language and Society. London: Nelson
LICHTENBERG, GEORG CHRISTOPH (see Gren^mann, Mautner)
LIENHARDT, GODFREY 1961 Divinity and Experience: The Re-
ligion of the Dinka. Oxford: Clarendon Press.
LINCOLN, c. ERIC 1961 The Black Muslims in America. Boston:
Beacon
LINCOLN, JACKSON STEWARD 193 5 The Dream in Primitive
Cultures. London: Cresset
LINTON, RALPH 193 6 The Study of Man. New York: Appleton-
Century
1939 Intro, to The Individual and His Society (with
Abraham Kardiner). New York: Columbia University Press
LLOYD, P. c. 1967 Africa in Social Change. Baltimore: Penguin
LOISY, ALFRED 1948 The Birth of the Christian Religion, trans.
L. P. Jacks. London: Allen & Unwin
LOUDON, j . B. 1965 'Social Aspects of Ideas about Treatment,'
in Transcultural Psychiatry, ed. A. V. S. de Reuck and Ruth
Porter. Ciba. London: J. & A. Churchill, 137-60
LOWIE, ROBERT 1924 Primitive Religion. New York: Boni &
Liveright
1937 History of Ethnological Theory. London: Harrap
LUX, ANDRE 1972 'Gift Exchange and Income Redistribution . . .
Yombe Rural Wage-Earners,' Africa, 42, 173-91
MCGIFFERT, ARTHUR CUSHMAN 1953 A History of Christian

Thought, I. New York: Scribner
MCLENNAN, J. F. 1865 Primitive Marriage, repr. in Studies in
Ancient History. London: Quaritch
M'MILLAN, WILLIAM n.d. Scottish Symbols: Royal, National and
Ecclesiastical... Paisley: Alexander Gardner
MAIR, LUCY P. 1934 An African People in the Twentieth Century.
London: Routledge
REFERENCES 445
MALE, EMILE 1947 VArt Religieux du XIle Siicle en France.
Paris: Armand Colin
MALINOWSKI, BRONISLAW 1922 Argonauts of the Western Pacific.
London: Roudedge
1926 Myth in Primitive Psychology. London: Kegan Paul
1927a The Father in Primitive Psychology. London: Kegan
Paul
1927b Sex and Repression in Savage Society. London:
Kegan Paul
1932 The Sexual Life of Savages in North-Western
Melanesia, 3rd edn. London: Roudedge
1935 Coral Gardens and Their Magic, 2 v. London:
Allen & Unwin
1936 The Foundations of Faith and Morals, Riddell Mem.
Lectures, 7th ser. London: Oxford University Press
1939 'The Group and the Individual in Functional Analy-
sis,' Amer. J. Sociology, 44, 938-64
1944 A Scientific Theory of Culture. Chapel Hill: Univ. of
N. Carolina Press
1948 Magic, Science and Religion. . . , intro. Robert
Redfield. Glencoe, 111.: Free Press
MANGIN, EUGENE 1921 Les Mossi. Paris: Auguste Challamel
MANNHARDT, WILHELM 1875-7 Antike Wald- und Feldkulte . . .
Berlin: Gebriider Borntraeger
MARETT, R. R. 193 3 Sacraments of Simple Folk. Oxford: Clarendon
Press
1935 Head, Heart and Hands in Human Evolution.
London: Hutchinson
MARTIN, KINGSLEY 1963 The Crown and the Establishment.
Harmondsworth: Penguin
MASCALL, E. L. 1949 Existence and Analogy. London: Longmans
MAUSS, MARCEL 1925 'Essai sur le Don, Forme Archaique de
PEchange,' V Annee Sociologique, n.s. i (trans. Ian Cunnison,
The Gift. London: Cohen & West, 1954)
1900-1 Review of Kattenbusch, VAnnee Sociologique, v,
298-301
MAUTNER, FRANZ and HATFIELD, HENRY 1969 Lichtenberg:
Aphorisms and Letters. London: Cape
446 REFERENCES
MEAD, MARGARET 1963 Sex and Temperament in Three Savage

Societies (1935). New York: Morrow
MELVILLE, HERMAN 1938 (1846) Typee: Narrative of a Four
Months Residence ... Harmondsworth: Penguin
M£N£GOZ, EUGENE n.d. 'Symbolo-Fideism,' Hastings Encyclo-
paedia of Religion and Ethics. New York: Scribner
M£TRAUX, ALFRED 1959 Voodoo in Haiti, trans. Hugh Charteris.
London: Deutsch
MEYER, PETER 1945 Schwei%erische Milnster u. Kathedralen des
Mittelalters. Zurich: Atlantis
MICHAUD, GUY 1947 Message Poetique du Symbolisme. Paris:
Nizet
MIDDLETON, JOHN i960 Lugbara Religion. London: Oxford
University Press
MILLS, c. WRIGHT 1961 The Sociological Imagination. New York:
Grove
MITCHELL, T. F. 1957 'The Language of Buying and Selling in
Cyrenaica: A Situational Statement,5 Hesperis: Archives Ber-
beres ... 32—71
MOMIGLIANO, ARNALDO 1946 'Friedrich Creuzer and Greek
Historiography,* J. Warburg and Courtauld Inst., 9, 152-63
MOONEY, JAMES 1965 The Ghost Dance Religion (1896), intro.
A. F. C. Wallace. Chicago: University Press
MORGAN, LEWIS H. I 8 5 I League of the Ho-de-no-sau-nee, Iroquois.
Rochester N. Y.: Sage [reprod. with introd. William N. Fenton
1962, New York: Corinth]
MORITZ, KARL PHILIPP 1830 Mythological Fictions of the Greeks
and Romans, trans from 5 th edn, with improvements, by Charles
Frederick William Jaeger. New York: A. & C. & H. Carvill
1861 Gotterlehre oder Mythologische Dichtungen der
Alten. 10th edn, intro. Dr Friedrichs. Berlin: F. A. Herbig
1968 Andreas Hartknopf eine allegorie 1786; Andreas
Hartknopfs predigerjahre 1790; Fragment aus dem Tagebuche
eines geistersehers, 1787 (facs). Stuttgart: J. B. Metzler
MORRIS, CHARLES 1945 Signs, Language and Behavior. Chicago:
University Press
MUNN, NANCY 1969 'The Effectiveness of Symbols in Murngin
Rite and Myth/ in Forms of Symbolic Action, ed. Robert F.
REFERENCES 447
Spencer, Amer. Ethnological Society. University Washington

Press, 178-207
MURPHY, ROBERT E. 1971 The Dialectics of Social Life. New
York: Basic Books
MYERS, A. R. 1966 England in the Late Middle Ages, rev. edn.
Harmondsworth: Pelican
NADEL, s. F. 1942 A Black Byzantium: The Kingdom of Nupe in
Nigeria. London: Oxford University Press
1951 Foundations of Social Anthropology. London: Cohen
&West
1954 Nupe Religion. London: Routledge
NAIK, Y. G. [1957] Indian National Emblems: Their Significance.
Bombay: Visual Education Society
NASH, MANNING 1966 Primitive and Peasant Economic Systems.
San Francisco: Chandler
NEEDHAM, RODNEY 1967a 'Right and Left in Nyoro Symbolic
Classification,' Africa, 37, 425-52
1967b 'Percussion and Transition/ Man, n.s. 2, 606-14
N0V0TNY, FRITZ 1930 Romanische Bauplastik in Osterreich.
Vienna: Benno Filser
OGDEN, c. K. and RICHARDS, I. A. 1923 The Meaning of Meaning.
London: Kegan Paul
O'MALLEY, L. S. S. 1935 Popular Hinduism - The Religion of the
Masses. Cambridge: University Press
O'NEILL, A. M., O.P. 1933 The Mystery of the Eucharist. Dublin:
M. H. Gill
PALMER, L. F. JR 1970 'Black America's Other National Anthem,'
Chicago Daily News, 26-27 August 1970
PALMER, PAUL F., s.j. 1953 Mary in the Documents of the Church:
London: Burns & Oates
PANOFF, MICHEL 1970 'Marcel Mauss's The Gift Revisited/
Man, n.s. 5, 60-70
PAQUES, VIVIANA 1954 Les Bambara, International African Inst.
Survey of Africa, Fr. ser. pt I. Paris: Presses Universitaires
PARSONS, ANNE 1969 Belief Magic and Anomie: Essays in
Psychosocial Anthropology. New York: Free Press
448 REFERENCES
PARSONS, ELSIE CLEWS 1916 'Holding Back in Crisis Ceremonial-

ism,' American Anthropologist, 18, 41-52
PARSONS, TALCOTT 1949 The Structure of Social Action (1937),
2nd edn. Glencoe, 111.: Free Press
1957 'Malinowski and the Theory of Social Systems,' in
Firth (ed.), 1957, 53-70
PAULME, DENISE 1967 ' T w o Themes on the Origin of Death in
West Africa,' Man, n.s. 2, 48-61
PEACOCK, JAMES L. 1968 Rites of Moderniiadon: Symbolic and
Social Aspects of Indonesian Proletarian Drama. Chicago:
University Press
PEIRCE, CHARLES S. 1931—5 Collected Papers of Charles Sanders
Peirce, ed. C. Harteshorne and P. Weiss. Camb., Mass: Harvard
U.P.
1958 Values in a Universe of Chance, selected writings, ed.
P. P. Wiener. New York: Doubleday
PIAGET, JEAN 1971 Structuralism, trans, and ed. Chaninah
Maschler. London: Routledge
PICKERING, w. A. 1878-9 'Chinese Secret Societies,' / . Straits
Branch Royal Asiatic Soc, 1, 63-84; 3, 1-18
POIZAT, ALFRED 1924 Le Symbolisme, de Baudelaire a. Claudel,
2nd edn. Paris: Bloud & Gay
PORTAL, FR£D£RIC I 8 5 7 Des Couleurs Symboliques dans VAntiquite,
le Moyen-Age et les Temps Modernes. Paris: Treuttel & Wiirtz
POWDERMAKER, HORTENSE 1950 Hollywood The Dream Factory:
An Anthropologist Looks at the Movie-Makers. Boston: Little,
Brown
PREBBLE, JOHN 1967 Culloden. Harmondsworth: Pelican
RADCLIFFE-BROWN, A. R. 1922 The Andaman Islanders. Cam-

bridge: University Press
1930-1 'The Social Organization of Australian Tribes,'
Oceania, Mon. 1. Melbourne: Macmillan
1952 'Taboo,' in Structure and Function in Primitive
Society. London: Cohen & West
1957 A Natural Science of Society, ed. M. N. Srinivas.
Glencoe, 111.: Free Press
REFERENCES 449
RADIN, PAUL 1958 Introduction to E. B. Tylor, Primitive
Culture
RAPPAPORT, ROY A. 1968 Pigs for the Ancestors: Ritual in the
Ecology of a New Guinea People. New Haven: Yale University
Press
RATTRAY, R. s. 1916 Ashanti Proverbs {The Primitive Ethics of a
Savage People). Oxford: Clarendon Press
1929 Ashanti Law and Constitution. Oxford: Clarendon Press
READ, HERBERT 193 3 Art Now. London: Faber
1936 Introduction to International Surrealist Exhibition
Catalogue. London
1948 Introduction to Klee {i8yc^-ic^4o). London: Faber
Gallery
RE AY, MARIE 195 9 The Kuma: Freedom and Conformity in the
New Guinea Highlands. Melbourne: University Press
REINACH, SALOMON 1942 Orpheus^ A History of Religions', new
edn, trans. Florence Simmonds. New York: Liveright
RICHARDS, AUDREY I. 1932 Hunger and Work in a Savage Tribe.
London: Routledge
1956 Chisungu: A Girl's Initiation Ceremony among the
Bemba of Northern Rhodesia. London: Faber
RIEFF, p. (See Freud, 1963)
RIVERS, w. H. R. 1906 The Todas. London: Macmillan
1918 Dreams and Primitive Culture, Bull. John Rylands
Library (Manchester), 4, nos 3-4
1920 Instinct and the Unconscious. Cambridge: University
Press
1924 'Mind and Medicine' (1919), in Medicine^ Magic and
Religion. London: Kegan Paul
ROBERTSON, BRYAN 1971 Bridget Riley', Paintings and Drawings
1951-7*1 intro., biographical note. London: Arts Council of
Great Britain
ROHEIM, G£ZA 1969a Psychoanalysis and Anthropology (1950).
New York: Internat. Universities Press
1969b The Gates of the Dream (1952). New York: Internat.
Universities Press
1971 The Origin and Function of Culture. New York:
Doubleday
450 REFERENCES
ROSEN, GEORGE 1968 Madness in Society: Chapters in the Histori-

cal Sociology of Mental Illness. London: Routledge & Kegan Paul
ROSENBERG, HAROLD 1965 The Anxious Object. London: Thames
& Hudson
ROSENBERG, JAKOB 1967 On Quality in Art. London: Phaidon
ROTHEL, H. K. Modern German Painting, trans. Desmond and
Louise Clayton. London: Eyre & Spottiswoode
ROUT, ETTIE A. 1927 Maori Symbolism. London: Routledge
RUBEL, PAULA and ROSMAN, ABRAHAM 1970 'Potlatch and Sagali:
The Structure of Exchange in Haida and Trobriand Societies,'
Trans. New York Academy of Sciences, ser. II, 32, 732-42
SAHLINS, MARSHALL 1965 ' O n the Sociology of Primitive

Exchange/ in The Relevance of Models for Social Anthropology,
ed. M. Banton. A.S.A. Mon. i , London: Tavistock, 139-236
1970 'The Spirit of the Gift' Echanges et Communications.
Melanges offerts a Claude Levi-Strauss, ed. Jean Pouillon and
Pierre Maranda. Hague: Mouton, 998-1012
SAPIR, EDWARD 1934 'Symbolism,' Encyclopaedia of the Social
Sciences, repr. in Selected Writings of Edward Sapir . . . ed. David
G. Mandelbaum. Berkeley: University California Press, 564-8
SCHACHT, JOSEPH 195 5 ' T h e Law,' in Unity and Variety in
Muslim Civilisation, ed. G. E. von Griinebaum. Chicago:
University Press, 65-86
SCHAPERA, I. 1955 A Handbook ofTswana Law and Custom, 2nd
edn. London: Oxford University Press
SCHAUBERG, JOSEPH 1861-3 Vergleichendes Handbuch der Sym-
bolik der Freimauerei, 3 v. Basel: Benno Schwabe
SCHENK, H. G. 1969 The Mind of the European Romantics. New
York: Doubleday
SCHERNER, KARL ALBERT 1861 Das Leben des Traumes - Ent-
deckungen aufdem Gebiet der Seek (Erstes Buch). Berlin: Schindler
SCHLEGEL, GUSTAVE 1866 Thian Ti Hwui: The Hung-League or
Heaven-Earth League. Batavia: Lange (repr. 1958, Singapore:
Govt Printer)
SCHNEIDER, ALBERT 1954 G.-C. Lichtenberg: Precurseur du
Romantisme; VHomme etVOeuvre. Nancy: Societe d'Impressions
Typographique
REFERENCES 451
SCHNEIDER, DAVID M. 1968 American Kinship: A Cultural Ac-
count. Englewood-Cliffs, N.J.: Prentice-Hall
SCHOLEM, GERSHOM 1955 Major Trends in Jewish Mysticism,
3rd edn. London: Thames & Hudson
SCHRIMPF, HANS JOACHIM 1968 (See Moritz, 1968)
SCHUBERT, GOTTHILF HEINRICH 1814 Die Symbolik des Traumes,
Bamberg: C. F. Kunz
SCHWARTZ, BARRY 1967 T h e Social Psychology of the Gift,'
American J. Sociology, 73, 1-11
SCHWENK, KONRAD 1851 Die Sinn-Bilder der alten Volker.
Frankfurt a/M: Sauerlander
SELIGMAN, c. G. 1932 'Anthropological Perspective and Psycho-
logical T h e o r y , ' / . RoyalAnthrop. Inst., 62, 193-228
SELIGMAN, C. G. and B. z. 1911 The Veddas. Cambridge: University
Press
SENIOR, ELIZABETH 1940 Portraits of Christ. Harmondsworth:
Penguin
SEUPHOR, MICHEL 1962 Abstract Painting. London: Prentice-Hall
SHILS, EDWARD and YOUNG, MICHAEL 1953 'The Meaning of the
Coronation,' Sociological Review, n.s. 1, no. 2, 63-81
SHURMER, PAMELA 1971 'The Gift Game,' New Society, 23
December 1971
SKULTANS, VIEDA 1970 'The Symbolic Significance of Menstrua-
tion and the Menopause/ Man, n.s. 5, 639-51
SLOTKIN, J. s. 1965 Readings in Early Anthropology, Viking Fund
Pub. Anthrop., 40. Chicago: Aldine
SMITH, MARGARET 1931 Studies in Early Mysticism in the Near
and Middle East. London: Sheldon
SMITH, M. G. 1962 Kinship and Community in Carriacou. New
Haven: Yale University Press
SMITH, w. ROBERTSON 1885 Kinship and Marriage in Early
Arabia. Cambridge: University Press
1889 Lectures in the Religion of the Semites. Edinburgh:
Black (new edn, 1907)
SOCIETY FOR PROPAGATION OF THE GOSPEL 1948 The Life of
Christ by Chinese Artists. London: S.P.G.
SOMERVILLE, J. et al. 1967 'Symbol,' New Catholic Encyclopaedia.
New York: McGraw-Hill
452 REFERENCES
S0RENSON, BENGT ALGOT 1963 Symbol und Symbolismus in den

aesthetischen Theorien des 18 Jahrhunderts und der deutschen
Romantik. Kopenhagen: Munksgaard
SPEKE, JOHN HANNING I 9 I 2 Journal of the Discovery of the Source
of the Nile, Everyman edn. London: Dent
SPENCER, ROBERT F. (ed.) 1969 Forms of Symbolic Action, Proc.
Ann. Spring Meeting, Amer. Ethn. Soc. Seattle: Univ. Washing-
ton Press
SPICKER, STUART, F. 1970 The Philosophy of the Body. Chicago:
Quadrangle
STANNER, W. E. H. i960 ' O n Aboriginal Religion,' Oceania, 31
1967 'Reflections on Durkheim and Aboriginal Religion, 5
in Social Organisation: Essays Presented to Raymond Firth,
ed. Maurice Freedman. London: Frank Cass, 217-40
STEBBING, L. s. 1948 A Modern Introduction to Logic (1930).
London: Methuen
STOUT, DAVID B. 1947 San Bias Cuna Acculturation: An Introduc-
tion, Viking Fund Pub., 9. New York: Wenner-Gren Foundation.
STRATHERN, ANDREW 1971 The Rope of Moka: Big Men and
Ceremonial Exchange in Mount Hagen New Guinea. Cambridge:
University Press
STRATHERN, ANDREW and MARILYN 1971 Self Decoration in
Mount Hagen. London: Duckworth
SUNDKLER, BENGT G. M. 1948 Bantu Prophets in South Africa.
London: Lutterworth Press
THIERS, JEAN-BAPTISTE 1609 Histoire des Perruques. Paris: at

author's expense
THOENE, PETER 193 8 Modern German Art, trans. Charles Full-
man. Harmondsworth: Penguin
THOMPSON, MICHAEL 1969 Review of Dillon, op. cit., Man, n.s.
4, 664-5
THOULESS, R. 1961 Psychology of Religion. New York: Macmillan
TITMUSS, RICHARD M. 1970 The Gift Relationship: From Human
Blood to Social Policy. London: Allen & Unwin
TRUMBULL, HENRY CLAY 1885 The Blood Covenant. New York:
Scribner
REFERENCES 453
1894 Studies in Oriental Social Life. Philadelphia: John
D. Watts
1896 The Threshold Covenant or the Beginning of Religious
Rites. New York: Scribner
1899 The Covenant of Salt. New York: Scribner
TUREL, ADRIEN 1938 Introduction to J. J. Bachofen, Regne de la
Mere au Patriarcat, pages choisis. Paris: Alcan
1939 Bachofen-Freud: Zur Emancipation des Mannes von
Reich der Mutter. Bern: Huber
TURNER, TERENCE s. 1969a 'Tchrikin: A Central Brazilian Tribe
and its Symbolic Language of Bodily Adornment,' Natural
History', 78, 50-70
1969b 'Oedipus: Time and Structure in Narrative Form,'
in Forms of Symbolic Action, ed. Robert F. Spencer. American
Ethnological Society: Univ. Washington Press, 26-68
TURNER, VICTOR w. 19 5 7 Schism and Continuity in an African Society.
Manchester: University Press
1962 Chihamba, The White Spirit (Rhodes-Livingstone
Paper 33). Manchester: University Press
1966 'The Syntax of Symbolism in an African Religion,'
see Huxley, 295-304
1967 The Forest of Symbols: Aspects of Ndembu Ritual.
Ithaca: Cornell University Press
1969 The Ritual Process: Structure and Ann-Structure.
Chicago: Aldine
TURNEY-HIGH, H. H. 1949 General Anthropology. New York:
Thomas Y. Crowell
TYLOR, EDWARD BURNETT 1873 Primitive Culture, 2 v., 2nd edn.
London: John Murray
1878 Researches into the Early History of Mankind and the
Development of Civilisation. 3rd edn. London: John Murray
URBAN, WALTER MARSHALL 1939 Language and Reality: The

Philosophy of Language and the Principles of Symbolism. London:
Allen & Unwin
VALERIAN, JOHAN PIERIUS 1533 A Treatise written by Johan

Valerian a greatte clerke of Italic, which is intitled in latin Pro
454 REFERENCES
Sacerdotum barbis translated into Englysshe. London: T h o .

Berthelet (orig. pub. 1533; Paris: Christianus Wechelus)
VASUDEVA, s. AGARVALA (AGRAWALA) I 964 The Wheel Flag of
India: Chakra-Dhvaja. Varanasi: Prithivi Prakashan
VERLAINE, PAUL 1884 Les Poetes Maudits. Paris: Leon Vanier
1939 Poemes Saturniens. Paris: Editions de Cluny
VESEY-FITZGERALD, SEYMOUR 1931 Mohammedan Law: An
Abridgement. London: Oxford University Press
VOLKELT, JOHANNES 1873 Das Unbewusste und der Pessimismus.
Berlin: F. Henschel
1875 Die Traum-Phantasie. Stuttgart: Meyer & Zeller
1876 Der Symbol-Begriffin der neuesten Aesthetik. Jena:
Hermann Dufft
VOSS, JOHANN HEINRICH 1824-26 Antisymbolik, 2 v. Stuttgart:
Metzler
WAKE, CHARLES STANILAND and WESTROPP, HODDER M. 1875

Ancient Symbol Worship. Influence of the Phallic Idea in the
Religions of Antiquity, 2nd edn, intro., notes, app. Alexander
Wilder. New York: J. W . Bouton
WALLACE, ANTHONY F . c. 1958 'Dreams and the Wishes of the
Soul,' American Anthropologist, 60, 234-48
WALLACE-HADRILL, J. M. w. 1962 The Long-Haired Kings and
Other Studies in Prankish History. London: Methuen
WANIJN, ROB. L. 1950 'Ancient Religious Insigniae in Bas-
Congo,' in Arts in Belgian Congo and Ruanda- Urundi [Vatican
Exhibition]. Brussels: C.I.D., 41-55
WARD, j . s. M. and STIRLING, w. G. 1925 The Hung Society or the
Society of Heaven and Earth, 3 v. London: Baskerville
WARNER, W. LLOYD 1961 The Family of God: A Symbolic Study
of Christian Life in America. New Haven: Yale University Press
WASCHE, ERWIN 193 3 Carl Gustav Carus und die romantische
Weltanschauung. Diisseldorf: G. H. Nolte
WEITMAN, SASHA R. 1971 'The Flags of All Nations: First
Results of a World-Wide Statistical Survey,' Paper presented to
the 66th annual meeting of the American Sociological Assn,
2 September 1971
WERBNER, RICHARD P . 1971 'Symbolic Dialogue and Personal
REFERENCES 455
Transactions among the Kalanga and Ndembu/ Ethnology, 10,
311-28
WESTERMARCK, E. A. 1891 The History of Human Marriage.
London: Macmillan (subsequent edns, for example, 5 th edn,
3 v., 1921)
1914 Marriage Ceremonies in Morocco. London: Macmillan
1926 Ritual and Belief in Morocco, 2 v. London: Macmillan
1930 Wit and Wisdom in Morocco: A Study of Native
Proverbs. London: Routledge
WHATMOUGH, JOSHUA 1956 Language: A Modern Synthesis.
New York: Mentor
WHITE, T. H. (ed.) i960 The Bestiary: A Book of Beasts (1954).
New York: Capricorn
WHITEHEAD, A. N. 1927 Symbolism, Its Meaning and Effect.
New York: Macmillan
WILSON, EDMUND 1962 The Triple Thinkers. Harmondsworth:
Pelican
WILSON, MONICA 1957 Rituals of Kinship among the Nyakyusa.
1959 Communal Rituals of the Nyakyusa. London: Oxford
University Press
WILSON, THOMAS 1894 The Swastika, report of the U.S. National
Museum, 757-1030
WINTER, E. K. 1928 'Bachofen-Renaissance/ Z. f die gesamte
Staatswissenschaft, 85, 316-42. Tubingen
WINZEN, DAMASUS 1957 Symbols of Christ. London: Longmans
Green
WOLF, ARTHUR p. 1970 * Chinese Kinship and Mourning Dress,'
in Family and Kinship in Chinese Society, ed. Maurice Freedman.
Stanford: University Press, 189-207
YALMAN, NUR 1969 ' On the Meaning of Food Offerings in Ceylon/

in Forms of Symbolic Action, ed. Robert F. Spencer. American
Ethnological Assn: University of Washington Press, 81-96
YANG, MARTIN C. I 947 A Chinese Village: Taitou, Shantung
Province. London: Kegan Paul
YONGE, CHARLOTTE M. 1889 Womankind, new edn. London:
Walter Smith & Innes
456 REFERENCES
YOUNG, MICHAEL D. 1971 Fighting with Food: Leadership,
Values and Social Control in a Massim Society. Cambridge:
University Press
ZAINAL- 'ABIDIN BIN AHMAD 1950 'Malay Manners and Etiquette/

J. Royal Asiatic Society, Malayan Branch, 23, 43-74
ZIMMER, HEINRICH 1946 Myths and Symbols in Indian Art and
Civiliiation, Bollingen, 6. Washington, D.C.: Pantheon
Addenda (for chs. 8 and 9):

D£MEUNIER, J. N. 1776 L9 Esprit des Usages et des Coutumes des
differens peuples. 3 v. London/Paris: Pissot
TEGG, WILLIAM 1877 Meetings and Greetings: The Salutations,
Obeisances and Courtesies of Nations. London: William Tegg &
Co. (This compendium, for notice of which I am indebted to
T. M. Ciolek, covers a wide range of ethnographic data. Though
superficial, it is apparently the first general work in the field.)
INDEX
Abelam 216 Archetype 157, 158; see also J u n g

Abstraction 4 2 - 4 , 54-5, 60, 7 1 , 76, Aristotle 289, 369, 421
96, 97, 105, 108, 131, 133, 165, Armstrong, R. G. i i 4 n .
204, 213-14, 234, 243; see Accident Arnold, T . W . 409m
Accident, and substance 421-4 Art 23, 32, 57, 58, 77, 100, 118, 138,
Aesthetic 88, 99, 237, 239, 275, 277, 148, 164, 207, 217, 363; and
326, 421; see Art communication, 215-16; primitive,
Aggression 182-3, I 9 I > 222
> 345? 37° 128-9; an<
^ ritual, i77n.; symbols
Aiyappan, A. 308 in, 37-47
Ajisafe, A. K. 320 Atkinson, J. J. 151
Akerblad, J. D . i o i n . Attwater, D . 230, 232
Albert, J. 46 Auber, Abbe 49m
Alienation 197, 243 Authority i n , 225-6, 240, 276, 277,
Allegory 49, 7 0 - 3 , 9 5 - 7 , 181, 237 283
Allport, G. 211 Autologic 204, 205
Alsop, S. 21 Aveling, F . A. P . 192
Ambiguity 28, 35, 118, 186, 195, Ayer, A. J. 1 7 m .
196, 282, 310, 360, 362, 427
Ambivalence 149, 195 Bachofen, J. J. 103-6, 108, 109, 112,
Amish 286 11411., 122, 123, 124, 132, 286
Anagoge 49 Bacon, F. 112
Anology 50, 51, 55, 70, 72, 82, 92, Baeumler, A. 105, io6n.
139, 151, 260, 362, 420, 421 Bailey, F . G. 372
Andamanese 135-8, 160, 183, 337 Baily, M. 35411.
Anderson, J. N . D . 389m Barraclough, E. M. C. 336m
Anderson, M. D . 38m Bartels, M. 122
Animatism 159 Bateson, G. 44m, 161, 183, 199,
Animism 113 207, 211
Anti-structure 205 Baudelaire, C. P. 72n., i25n., 189, 195
Apostle's Creed 4 7 - 8 , 54 Beard 97m, 265-7, 275, 279, 283-6;
Aquinas, T . 57, 72, 421 see also Hair
457
458 INDEX
Bearden, R. 17711. Brand, G. 227

Beattie, J. 17611., 20011., 213 Breton, A. 44
Beerbohm, M. 372 Bribe 392-3
B£guin, A. 9411., 95, 102 Brown, C. 386
Beidelman, T. 181 Brown, J. A. C. 157
Bell, C. 322 Brown, J. E. 119m
Belshaw, C. S. 385, 394-5 Burke, K. 194,195m
Bemba 181-6 Burridge, K. L. 20on., 201, 219
Bentham, J. 163 Burton, R. 122
Berg, C. 298 Butler, S. 263
Berndt, R. M. 166, 208, 209, 217
Bernhardi, A. F. 11411. Calmet, Dom 266
Bernoulli, C. A. 10311., io6n. Camerarius, J. 37
Berreman, G. 397 Campbell, J. io6n.
Berry, J. 14m., 142, 143 Campbell, T. 343
Bible 203, 221, 222, 225, 231, 263, Carey, W. J. 4o8n.
264, 286, 287, 374-5, 393, 401, 417 Cargo cult 200, 219, 222, 238, 427
Birdwhistell, R. 226, 300, 30711., 313 Carlyle, T. 15, 90, 190, 339, 340, 351
Birnbaum, N. 89 Cams, G. V. 102-3
Black Christ, 410; Blacks in U.S. 19, Cassirer, E. 56-7, 203
20, i46n., I77n., 273, 363^; flag, Castaneda, C. ii9n.
348; see also Colour symbolism Catholicism 73-4, 80, 85, 86, 97n.,
Blackstone, W. 372 101, 227-37, 267, 289-90, 292,
Blake, W. 239 407-26
Blau, P. 369, 370 Cezanne, P. 41
Blood 18, 59, 67, 85, 117, 118, 135, Champollion, J. F. ioin.
182, 229; see also Body, Colour Charbonnier, G. 337
symbolism, Eucharist Charisma 239
Blount, T. 27 m. Chief 26, 79, 82, 175, 184, 186, 218,
Blythe, W. 119m 219, 248, 250-5, 258, 272, 283
Boas, F. 128-30 China, Chinese 16, 17, 18, 22, 76,
Body 175, 185, 203, 204; mystical, 114, 158, 265, 268n., 290, 319, 353,
26, 228; as social instrument, 355, 394; greeting, 309, 311, 315;
307-11,323-5; types of symbolism, mourning, 69-70; secret societies,
226-30; see also Blood, Sacred u 9m
Heart of Jesus Chiromancy 115
Bohannan, P. 111-12 Christ, Jesus 26, 37, 38, 40, 68, 73,
Bott, Elizabeth 155, 209 117; body, 227-8; ethnicity, 406-11;
Boudet, M. 97n., 266 heart, 230-7, 411-14; see also
Bouteiller, Marcelle 2Con. Christianity, Eucharist, Sacred
Bowra, M. 31, 32 Heart
INDEX 459
Christianity 50, 52, 55, 56, 102, m , Cooper, Wendy 262m
189, 204, 218, 249, 259, 267, 286, Copleston, F. C. 72m
320, 325, 401-2; art, 37-9, 60; Coronation 87-90
colour symbols, 67-8; see also Corruption 392-3
Blood, Christ, Coronation, Mon- Coulanges, F. de 109, n o - 1 1 , 116
archy, etc. Coulton, G. G. 55,68
Church 18, 73-4, 219, 227, 311, 322; Covenant n 7-18
Zionist, 220-1 Crawley, A. E. non., 294m
Clapham, A. W. 39n. Creuzer, F. 48, 97n., 99-101, 104,
Clark, W. P. ii3n. 112, i i 4 n . , 123, 124
Clarke 96 Critchley, M. 307n., 311
Coaptation 138 Cross as symbol 26, 37, 48n., 59,
Coemption 109 79, 141, 152, 157,404
Cognitive anthropology 61, 164, 171 gpCulture 151-2, 153, 160, 164, 220
Cohen, A. 85, 87, ii9n., 197, 205-6,
207 Dalton, G. 371
Colangeli, O. 346 Danby, H. 283
Coleridge, S. T. 375 © Dance 57-8, 77, 120, i23n., 130,
Collingwood, R. G. 41 136, 137, 175, 176, 178, 182, 208,
Colour symbolism 67-70, 72n., 84, 213, 258, 259, 273
120, 135, 160; flags, 331-8, 347, Dante Alighieri 49,346
350, 358; hair, 270-1 <0 Darwin, C. 108, 112, 267, 275,
Commitment 383 3i2n.
Communication 34, 56, 65, 70, 73, Davis, J. 373n.
107, 138, 140, 144, 145, 166, 170, Deism 97
176, 217, 261; in art, 4 4 - 6 , 215-16; Demeunier 265
dream, 222; symbolic function, De Raedt, J. 330
79-82, 208, 238, 239-40, 277, 427; Deshen, S. 84
in spirit mediumship, 225; 'token', Dewey, J. 169
177 Dialectic 44, 57, 205, 239, 290
Communion 87, 89, 107, 189, 414; Dillon, W. 397
see also Eucharist Disjunction 26, 75
'Communitas' 192, 205 Displacement I46n., 149, 154
Condensation 81, i46n., 149, 155,0 Dissociation see Trance; Spirit med-
190, 218, 234 iui
Conflict 89, 93, 95, 148, 177, 178, Dixon, R. B. 128
185, 189, 190, 191, 193-4, 195, 196, Dobu 288
204, 214, 216, 236, 284, 428 Dodge, R. 330
Contextualization 25, 173, 174, 189, Dogon 120, 289
192, 203, 260, 300 Doncoeur, P. 407^
Contradiction, patterned 204, 222,246 Douglas, Mary 59, 60, 62, 155, 166,
460 INDEX
Douglas (cont.) Ernesti, J. 96

191, 197, 199, 204, 205, 226, 230, Ethology 300, 323, 325
287, 307, 416, 420 Etiemble, R. 3on.
Drama 20on.; see Social drama Eucharist 26, 48, 168, 227, 416-26
Dream 44, 118, 127, 158, 170, 183, Eulenberg, A. 122
186, 207, 252; history of study, Evans-Pritchard, E. E. 58,72n., 1 ion.,
94-5, 101-3; personal and social i i 7 n . , 121, i25n., 161-3, i7on., 171,
relevance, 208-9, 210, 216-23; in J
73> J79> 1 8 0 , 2 0 9 , 2 1 8 , 4 1 5
psychoanalysis, 147-9, 150, 156 Evers, C. 19
Dry den, J. 267 Evolution 108, 112, 115, 129, 263
Dulaure, J. A. 56, 9 6 - 8 , 1 0 4 , 1 2 2 , 1 2 3 , Exchange 147, 168, 208, 258; as
266, 285 general principle, 370; see also Gift,
Dunbar, Helen 37, 49m, 50 Reciprocity
Duncan, H. D . 194 Experience 4 2 - 3 , 4 5 , 7 1 , 8 5 , 8 9 , 9 5 , 1 1 8 ,
Dupuis, C. F. 98 155, 160, 161, 168, 183, 185, 201,
Durkheim, E. 21, 78, 86, 125, 129, 204, 205, 237, 240, 428
130-5, 136, 139, 141, 159, 162, Expression, by symbols 73, 7 7 - 9 ,
163, 209, 210, 294m, 339-40, 341, 131, 136, 137, 176, 208, 217, 260,
351/357, 3 6 4~7 303, 379
Durliat, M. 38n.
Fairchild, H. N. 94n., 108
Economics 79, 85, 175, 179, 182, Family relations 235, 237, 258, 29on.,
184, 280, 293, 370-1, 393, 401 326, 38m., 388; see also Freud,
Efron, D . 307n. Kinship
Eggan, F. 331 0 Fantasy 20on.
Eglin, R. 354n. Farbenstein, W . i03n.
Ehrenreich, P. 128, 144 Fast, J. 307m
Ehrmann, J. 370 Faulhaber, Cardinal 4o8n.
Eliade, M. 70, 83, 20on., 226 Feidelson, C. 11, 33
Eliot, T. S. 125 Feminism 18, 22, 35n., 261, 274, 289
Elkin, A. P. 137, 174 Ferguson, A. 93, 109 - '
Ellis, Havelock 122 Ferguson, G. 287
Embree, J. 331 Fetishism 96, 98
Empiricism 10, 98, 106-7, l i o n . , 0 Fiction 23, 51, 92, 109, 162, 163
116, 117, 133, 153, 159, 181, 192, Fideism 49
204, 210, 227, 239, 298 Field, Margaret 209, 217m, 223-4,
Emblem 70, 71, 132, 137, 174, 182, 225m
233 Findlay, J. N. 86
Emerson, R. W . 15, 33, 34, 343 Finley, M. I. 370
Emmett, Dorothy 172 Fingesten, P. 43
Enlightenment, The 93, 97, 98 Firth, J. R. 14m., 142
INDEX 461
Firth Raymond 27, 34, 58, iion., Gesture 138,304,311-23

117m, i25n., i5on., 168, 171, 177, 0Ghost Dance 83, 234m
195m, 20on., 216, 307, 313, 328, Gibb, H. 389, 405
385^,386,406,415 Gift 29, 84, 107, 167, 169; character-
Fitzgerald, E. 72 istics, 368-71; concept, 372-5;
Flag, as symbol 29, 63, 64, 65,66,77, formality, 375-81; countergift, 383—
84, 129, 141, 152, 155, 212, 321, 6; earnest and token, 381-3; obli-
403; in ritual parody, 328-9, 364; gation to give, 386-9; to receive,
ethongraphy, 329-32; national, 336— 389-95; to repay, 395-402
41, 347-55; in protest and dissent, Gluckman, M. 124, 177-8, 181, 183,
355-67 194
Food, symbolism 166, 184, 243-61, Glueck, Grace 46
380, 403 God 86, 96, n o , 115, 118, 124, 133,
Forge, Anthony 216 226, 250, 253, 283; and Church,
Fortes, M. I54n., 161-3, 167, i68n., 228; death, 239; dependence on,
169, 172, 173, 177, 425 80, 180; existence, 57, 72m; referent
Fortune, R. 13 5n., 288 of symbol, 128, 404, 415; transcen-
Fowler, H. W. 7m. dence, 48, 56
Frankel, M. 22 Godelier, M. 50
Frankl, P. 40, ii4n. Goethe, J. W. von 95, 96
Frazer, J. G. 124-6, 147, 203 Goffman, E. 300
Freemasonry 118-19 Goldenweiser, A. 152
Freud, S. 8in., 102, io3n., 116, 122, Goldsmith, O. 396
130, 147-56, 162, 166, 198, 203, Gombrich, E. H. 46, 72n., 226
2i9n., 237, 252 Goody, J. 179
Friedl, A. 346 Gorer, G. 209
Friedrich, P. 84 Greeley, A. M. 2on.
Frobenius, L. 144 Greene, Carroll i77n.
Fry, R. 43, 44n. Greene, Graham 396
Fryke, C. 29on. Greeting and parting symbolism 28,
Funeral 69, 70, 246-7, 251, 255, 272, 169, 299-327, 404
359, 379-80; see also Mourning Grenzmann, W. 95
Grimm, J. and W. 113
Gantner, J. 38n. Grivot, D. 39
Gardner, P. ioin. Guigniaut, J. D. 100
Garvey, M. 410 Gynaecocracy 104
Gauguin, P. 41
Gaunt, W. 44m Haddon, A. C. 150, 161
Gellner, E. 229 Hair 18, 28, i5on., 168, 188, 228,
Gennep, A, van 97m, 174, 181, 205, 262-5, 296-8, 313, 403; male and
301 female symfc>ol? 265-71; mens'
462 INDEX
Hair (cont.) Hooton, E. A. 263, 267

long, 274-83; beard, 266-7, 2 ^3 - 7; Horton, R. 404
loss, 287-91; magical, 294-6; and Hostetler, J. A. 286
social control, 296-8; statements Hotman, A. 266, 285, 289, 290
with, 271-4; see also Wig Howald, E. 10in., 106
Hall, E. T. 300, 307n. Howard, E. C. C. 389^
Halleck, F. G. 343 Howell, J. 267
Hallpike, C. R. 68, 262, 264, 297-8 Howells, W. 263
Hand, right and left 213, 243, 320; Howell-Smith, A. D. 233, 413
clasp, 302,319, 321; shake, 303,304, Hubert, H. H7n., 135, 179
317, 319-21, 325, 326, 377; touch, Hull, L. W. H. 55
384 Hulme, F. E. 39n., 68, 333n.
Harre, J. 262 Hurgronje, S. 389^
Harrison, Jane 125, 158 Huxley, J. 300
Hartmann E. von i03n. Huxtable, Ada 2on., 78
Haskins, C. H. 72, 229n.
Hauser,A. 31, 4411., 72, 73 Icon 61, 74, 100
Hawthorne, N. 33 Iconography 202, 233, 409
Hayley, Audrey 209, 415 Ideoscopy 61
Hayley, T. T. 209 Image 46, 49, 70, 71, 72, 94, 100, 101,
Heer, F. 72 102, 104, 108, 112, 114, 115, 140,
Hegel, G. W. F. 126, 201 183, 195, 210, 212, 216, 221, 238,
Heisenberg, W. K. 196 269
Henry, J. 152 Incest 170, 218, 25 m., 252
Herbart, J. F. 102 Index 61-2, 72n., 75
Herder, J. G. von 101 Immaculate Heart of Mary 231, 233,
Hermann, J. G. J. 96 235? 237
Hertz, R. 135, 213, 320 Iroquois 65, 106, 218-19, 2 2 2
Hervet, G. 266 Islam 53, 115, 293, 336, 338, 387-9,
Heyne, C. G. 96 404, 409n
Heywood, J. 262
Hiatt, L. 166 James, E. O. 231, 416, 417, 418, 419,
Hieroglyphs ioin., 102, 119 425
Hinduism 47, 283-4, 290, 315, 414 James, Henry 32-6
Hirn, Y. 229^ James, William 211
Hocart, A. M. 27n., 159 James, Mrs 233
Hoebel, E. A. 263, 267 Januarius, St 229
Hogbin, H. I. 194 Japan 16, 19, 22, 265, 306, 353, 399
Homans, G. C. 167, 169, 370 Jarvie, I. C. i25n.
Homer, 271 Jenkins, R. 2on.
Honigman, J. J. 341,351 Jenyns, S. 69
INDEX 463
Johnson, A. B. i4on. La Fontaine, Jean 199
Joyce, J. 160 Langen, A. 95
Judaism 52, 73, 84, 99, 129, 284m, Langendoen, T. D . 141, 142
286, 358, 379, 423 Langer, Susanne K. 5 7 - 8 , 6 4 , 6 7 , 7 0 , 7 7
Jung, C. G. 50, 59-60, 68, 102, 106, Language 31, 34, 50, 51, 55, 56, 57,
131, 156-8, 192,416,419 70,73,79,94,101,105,113,117-18,
Junod, H. 315, 320, 325 120, 126, 127, 130, 139, 140-3, 147,
Juvenal 265 148, 154, 155, 160, 166, 168, 169,
172, 175, 177, 182, 202, 210, 225,
Kabbalah 53n., 55 2
4 < H 3 , 2 5 6 , 3 ^ , 4 0 7 , 418
Kandinsky, V. 43 Lasswell, H. 86
Kardiner, A. 153 Lattimore, O. 322, 331
Kattenbusch, F. 48 Layard, J. 158
Kelantan, Malaysia 24-5, 121, 179, Law 20, 86, 271, 279-82, 387, 388,
210, 295-6, 406 395, 402; legitimacy, 365; legiti-
Key as symbol 7 2 - 1 , 7 3 - 4 mization, 86
Key, F. S. 344 Lawrence, P. 219, 222, 238
Keyser, C. E. 39m Leach, E. R. 62, i i 7 n . , i25n., 144,
Kimball, S. T. 181 166, 167, 203, 204, 217, 260, 275 n.,
Kinesics 3 0 7 ^ , 323 295, 297, 400
Kinship 27, 69, 84, 86, n o , 115, 118, Le Bon, G. 132, 15 m .
122, 130, 147, 149, 152, 153, 161, Lebra, Takie S. 399
166, 174, 181, 182, 185, 186, 188, Lee, Dorothy 172-3
198, 199, 204, 208, 211, 254, 259, Leenhardt, M. 165
^93, 3*5, }i*> 319, 379, 3*4, 3 ^ n . , Leiris, M. 120, i25n., i35n., i95n.
402 Leisinger, H. 39n.
Kipling, Rudyard 343 LeVi-Strauss, C. 67,70,77n., 104,165,
Kirschbaum, E. 407n. i68n., 197, 198, 203-4, 237, 243-4,
Kiss 302, 312, 313, 314-18, 377 250, 337-8, 341, 400, 405
Klee, P. 42, 215 LeVy-Bruhl, L. 151, 159, 160-1, 165,
Knowles, D . 72n., 426 171, 172, 416
Kooning, W . D e 43 Lewis, I. M. 20on., 201, 209, 225n.
Kraft-Ebbing, R. 122 Lewis, M. M. 145n., 192
Kretzmann, N. i4on. Lewitt, S. 45
Krewald, A. 103 m Lichtenberg, G. C. 9 4 - 5 , 101
Kroeber, A. 128, 152 Lienhardt, G. i i 7 n . , 179
Kiinstle, K. 39n. Lincoln, C. E. 410
Kuper, K. 39m Lincoln, J. S. i5on., 151
Kuper, Hilda 177-8 Linton, R. 152, 341
Lion as symbol 37-9, 50, 61, 62, 184,
La Barre, W . 200 185, 188, 199
464 INDEX
Lloyd, P. C. 18 Mascall, E. L. 72n.

Locke, J. 60 Mathers, W . Powys 122
Loisy, A. 418 Mauss, M. 48, 117, 135, 165, 179, 226,
Loudon, J. B. 209 307, 368, 369, 370, 374, 378, 385,
Lowell, J. R. 2 7 m . , 344 386, 390, 397, 398, 400
Lowie, R. H. 132, 135, 160 Maxwell, N. 24
Lux, A. 386 Mead, G. H. 196
Mead, Margaret 122
McDougall, W . 132 Melville, H. 32-4
McGiffert, A. C. 48, 49n., 418, 419, Mendes, C. 3on.
420 Menegoz, E. 50
McLennan, J. F. 2 7 , 1 0 8 , 1 0 9 - 1 0 , i n , Metraux, A. 20on.
163 Meyer, P. 38n.
M'Millan,W. 345 Michaud, G. 11, 3on., 32, 94, 195
Magic 70, 94, 103m 113, 121, 126, Middleton, J. 179, 180, 20on.
140, 151, 160, 220, 229, 406, 420; Mill, J. S. 112
hair, 294-6 Mills, C. W . 84, 86, 88
Maine, H. 112 Q Milton, J. 267, 269, 270
Mair, Lucy 326 Mitchell, C. 194
Male, E. 38n. Mitchell, T. F. 3 6 9 ^
Malinowski, B. 62, 79, 121, 122, 125, Moll, A. 122
126, 134, 139-47, i5 2 ~4> i ^ i , 163, Model 30, 56, n o , 149, 165, 198, 204,
166, 203, 220, 23on., 231, 368, 369, 290
377, 382, 400, 416, 419-21 Momigliano, A. 99, 106
Mallarme, S. 30-2, 126 Monarchy 87-90, 93, 126
Malay see Kelantan Mondrian, P. 43
Mangin, E. 309 Monroe, Marilyn 21
Mannhardt, W . 124 Mooney, J. 83, 234m
Mannheim, K. 86 Morality 2 0 , 5 1 , 6 5 , 8 0 , 8 6 , 8 8 , 9 5 , i n ,
Maori i23n., 126, 295, 313 118, 168, 197, 277, 278, 279, 304,
Marc, F. 42 350, 365, 367, 371, 393, 400-2
Mardrus, J. C. V. 122 Moreau, G. 41
Marett, R. R. i i o n . , 114, 115, 159, Morgan, L. H. 65, 106-7, 139, 218,
180, 181 219
Market concept 368-70,381 Moritz, K. P. 95-6, 101
Marriage 59, 69, 118, 122, 137, 166, Morris, C. 64n., 65-6, 88, 169
182-5, 186, 187, 198, 268, 272, 2 7 7 , ^ . Mother-right 103-5
331; by capture, 110-11; exchanges Mourning 69, 286, 289, 297, 331; see
at, 254, 256, 259; of priests, 286 also Funeral
Martin, K. 87-8 Miiller, M. 113
Marx, K. i n , 198 Munn, Nancy 85, 202, 204
INDEX 465
Murngin 85, 202 Oedipus complex 153-4,203 - 4
Murphy, R. 209 Ogden, C. K. 64n., 140, i45n.
Myers, A. R. 48n., 62n. O'Neill, A. M. 421, 422
Mystery 180, 195, 419, 422, 425; Oneiromancy 102
cults, 158,418 Orphic 100, 105
Mysticism 7 m . , 94, 100, 103, 113, Oudin, F. 266
118, 140, 141, 157, 160, 166, 172, Ovid 27on., 287
178, 182, 189, 240,270,320,416; in
art, 31, 43; Christian, 36, 49, 227, Pacyaya, A. 331
289^; Jewish, 73 PanofT, M. 369, 385n.
Myth 51, 57, 120, 151, 207; anthro- Pantheism 102, 103
pological interpretation, 85, 135, Paques, Viviana 321
138, 144, 147, 152, 153, 162, 163, Parsons, Anne i54n., 209
198, 202-4, 2 ° 8 , 2 2 3, 243; Freudian Parsons, Elsie C. 301
interpretation, 148, 237; historical Parsons, T. 132, 133, 146, 196
study, 94-106, 112-13, 117, 127 Parting see Greeting
Pater, W . 32
Nadel, S. F. 59, 63, 169-71, 174, 191, Paulme, Denise 384
192, 201, 213,308 Peacock, J. L. 195m, 278
Nash, M. 370 Peacock, T. L. 289
Nature 93-6, 102, 108, 127, 128, 137, Peele, G. 270m
271; and culture, 277, 278, 291, Peirce, C. S. 59, 6 0 - 2 , 65, 66, 7m.,
293-4, 371; naturalism rejected, 30, 108, 127, 169
42; natural science of society, 138 Perceval, J. 207-8, 211
Nayacakalou, R. 322 Personality disorder 207, 223-6; see
Ndembu 68, 166, 189-93, 199, 207, also Trance
214-15 Perugino P. 74
Needham, R. 27n., i59n., 213, 319 Peter, Prince of Greece 152
Nehru, Pandit J. 349, 351 Petrie, Flinders 119
N e w Guinea 294, 312, 3 6 9 ^ , 427 Peyote 200, 207
Nietzsche, F. W . 36, 106, 239 Q Phallic cult 96-7, 104, 116, 149,216,
* Noble Savage' 94n., 108; cf. 126 315, 363
Nose-pressing 313,316,320 Philosophy and symbols 56-8, 6 0 - 3 ,
Novalis (von Hardenburg) 94, 126, 108, 164; philosophemes, 101
141 Piaget, J. 62
Novotny, F. 38n., 39m ^Picasso, P. 215
Nuer 171, 179, 180-1, 415 Pickering, W . A. H9n.
Nupe 169, 213, 308, 311, 322 Ploss, H. H. 122
20
Nyakyusa 187-9, 7 Pobe, M. 38n.
Pocock, R. 48n.
Occult 118-23 Poe, E. A. 32-3
466 INDEX
Poetry and symbol 30-2, 94, 101, Radcliffe-Brown,A.R. 58,107,135-9,

113, 131, 195, 207, 240; illustration, 146-7, 160, 163, 183, 186, 195, 209,
264, 267, 270, 343 306
Poizat, A. 31 Radin, P. 112
Polanyi, K. 369, 371 Rainbow serpent 137, 208
Politics 108, 126, 177-8, 206, 240, Rappaport, R. A. 58
278, 411, 428; political symbols, Rationality 70, 93, 94, 95, 98, 108,
16, 19-25, 28, 7 7 - 8 , 120, 129, 162, 157, 195, 212
229, 273, 276, 318, 427, see also Rattray, R. S. 319
Flags; religion and politics, 8 0 - 1 , Read, H. 43, 44n.
82, 84, i n , 175; see also Power Reay, Marie 58
Pollution 27on. Reciprocity 107, 248, 254, 25 5n.,
Polygyny 221 2 5 6 - 7 , 368, 370, 378, 380, 395-402;
Polysemous 49, 190 see also Gift
Positivism 28, 41, 131, 132, 164 Redon, O. 41
Pope, A. 267, 27 m. Referent 52,53, 64, 70, 72, 81, 82, 83,
Portal, F. 68n. 114, 140, 157, 211, 220, 404, 421;
Potlach, 386., 397 referend 64m
Pound, E. 160 Reinach, S. 48, ii4n., 233, 27on.
Powdermaker, Hortense, 166 Religion 100, 110-11, 123, 126, 159,
Powell, A. C. 19 160, 168, 171, 180, 193, 201-2,
Power 82, 188, 198, 205, 222, 240, 2ii-ii;see also Christ, Christianity,
248, 252, 261, 345, 353, 365; in Cross, God, Hair, Judaism, Hin-
language symbolism, 127, I45n.; duism, Sacred Heart, etc.
mystical, 39, 57-8, 200, 229, 290, Representation 70, 7 1 - 5 , 82, 92, 94,
425; of symbolic instrument, 83-91, 98, 112, 114, 128, 136, 137, 138,
212, 405 166, 169, 201, 415, 418-26; repre-
Pragmatism as philosophic doctrine, sentative status, 326
61; pragmatic context, 146, 214, Richards, Audrey 27, 67, 82, 181-6,
221, 246, 260; pragmatic treatment 187, 189, 190, 209
of myth, 100; pragmatic significance Richards, I. A. 64m, 140, I45n.
of symbols, 99, 116, 141, 143, 159, Riley, Bridget 45
169, 426 Rimbaud, J. N. A. 3on., 68n., 72n.
Prebble, J. 356n. Ritual 65, 79, 85, 107, 113, 124, 147,
Proclus 100 167, 168, 170, i76n., 199, 2 0 1 - 3 ,
Procopius 27on. 207, 208, 210, 213, 249, 288, 297,
Psychology 147-58, 164, 170, 183, 301, 321, 326; idiom, 138, of
186-7, 207-10, 313, 426, 427 rebellion, 117; rites de passage, 205,
301; teletic, 301; values, 137:
Race relations 20, 103, 203; see also ritualization, 300, 325; see also
Black Coronation, Sacrifice, etc.
INDEX 467
Rivers, W. H. R. 150,151,158,283^, Savigny, F. K. von 103
294n. Schacht, J. 389n.
Robertson, B. 45 Schapera, I. 384
R6heim, G. 151-2,155 Schauberg, J. 119
Role 390; reversal, 162 Schelling, F. von 96
Romanesque 37, 42, 61, 67 Schenk, H. G. 106
Romanticism 31, 33, 141; Romantic Schemer, K. A. 147, i48n.
Movement, 93-103, 105-6, 108, Schjeldahl, P. 46
127, 131, 195; Revival, 196 Schlegel, A. W. 101, ii4n.
Rosen, G. 225n. Schlegel, G. ii9n.
Rosenberg, H. 43 Schneider, A. 94
Rosenberg, J. 43 Schneider, D. M. 166, 196,204,4i3n.
Rosman, A. 386n. Schrimpf, H. J. 95
Rosetti, D. G. 32 Scholem, G. 53n., 73
Rothel, H. K. 44n. Schubert, G. H. 101-2,112,147, i48n.
Rothenberg, J. i25n. Schwartz, B. 369, 381
Rouault, G. 41 Secret association, ii9n., 121; see also
Rout, Ettie A. i23n. Freemasonry
Rubel, Paula G. 386n. Seligman, C. G. 150, 161, 294m
Rudolph, P. 78 Semeiology 198
Russell, B. 64m Semeiotics (Semiotics) 60
Seuphor, M. 43
Sabaism 98 Shaman 151, 200, 203, 238
Sacred 86,90,129,131,133,144,176n., Shand, F. A. 153
182, 193, 217, 228, 229, 255, 294m, Shils, E. 89
295, 340, 348, 366, 404, 416, 420, Sign 27, 28, 58, 79, 113, 299; signansy
422, 425; sacralization, 365 58, 64, 71; sign function, 164; sign-
Sacred Heart of Jesus cult 230-7, situation, 140; and symbol, 50, 59,
6
411-14 °-3> 65> 72n., 74, 108, 169-70
Sacrifice 107, 116-17, 178-81, 207, Signal 64n., 65-6, 67, 75, 141, 300;
213, 235, 272, 298, 415, 417, 418, see also Flag; Greeting
419-20 Simile 70, 71, 130
Sacy, A. I. S. de ioin. Senior, Elizabeth 407n.
Sade, D. A. F. de 265 Sex 204, 208, 259, 263, 273, 369;
Sahlins, M. 369, 377, 38511. 398. symbolism, 21, 35, 85, 121-3, 162,
St Clair, H. H. 129 166, 181, 182-5, 188, 232, 243, 253;
Saint Martin 101 see also Freud, Hair, R6heim
Samson 37-9, 40, 283, 286 Skultans, Vieda 204
Sapir, E. 8in., 152, 154,167, 340, 351 Smiling 313-14
Sartre, J. P. 104 Smith, E. W. 231
Saussure, F. de 62 Smith, M. G. 331
468 INDEX
Smith, Margaret 28911. 25-30, 5i-3, T7> 58, 62-3, 71, 74,
Smith, W. Robertson 59,107,115-16, 75, 83, 86, 90-1, 92, 124, 146, 161,
124, 132, 179, 394 170, 426-8; classification, 50, 174,
Social, context, see Contextualization; 212-16; definition, 15, 25, 34, 46,
drama, 194-5; structure, 25, 112, <>i, 75> 9°> 9 8 , " 5 , 137-8, 163,
163, 170, 184, 191-3, 197, 200, 187,192; dominant (master), 8 6 - 9 1 ,
207, 214, 240, 252, 254, 258, 260, 191, 211, 245-6; history of term,
390, 400, 426 47, 54; 'natural* symbolsl, 25, 49,
Somerville, J. 49m 58-60, 63, 98, n 5 , 169, 413-14;
Sorensen, B. A. 94n. logic* of symbols, 243, 252, 324,
Speech 143, 378; see also Language 419, 420-4; meaning for social
Speke, J. H. 319, 373m relations, 81, 82, 133-4, 163, 173-6,
Spencer, B. 125 190, 402, 427; systems, 67, 70, 146,
Spencer, H. 112 164, 202, 246, 252, 290, 298, 414;
Spencer, R. F. 54m validation, 237-40, 424; variation,
Spicker, S. F. 307m 214, 246, 323, 325
Spillius, J. 253 Symbolist movement 3 0 - 2 , 83, 105,
Spirit 26, 47, 58, 65, 79, 82, 114, 159, 126, 127, 131
171, 216, 217, 223, 250, 405, 422; Synaesthesis 72n.
medium cult 199, 210, 225, 295
Spiro, M. E. 54n. Taboo 136, 137, 138, 253, 272, 293,
Stanner, W. E. H. 132-3, 134, 201-2, 320, 329, 413, 423
366 Tarn, N . 125m
Status 88, 138, 152, 167, 169, 179, Tedlock, D . i25n.
185, 199, 200, 206, 210, 214, 248, Tennyson, A. 27on.
250-9, 261, 403; see also Flag, Theomyth 100
Gift, Greeting, Hair Thiers, J-B. 292
Stebbing, Susan 63, 64n., 66, 81 Thoene, P. 42
Stirling, W. G. ii9n. Thompson, M. 397
Structure see Social Structure; Symbol Thouless, R. K. 212
systems Thurnwald, R. 368
Structuralism 46,70,85,121,171,196, Tikopia 26, 34, 65, 79, 82, 115, 121,
198, 199, 243; see also Levi-Strauss i5on., 168, 173, 175, 185, 218, 228,
Stout, D. B. 386 306, 329, 342, 379, 3 8 5 ^ , 390,
Strathern, A. 294, 369m 414-16, 420, 422; food symbolism,
Strathern, Marilyn 294 244-61; hair, 272-3, 288, 293, 297
Suger, Abbot 40 Titmuss, R. 400
Sundkler, B. 220-1 Todas i5on., 2 8 3 ^ , 294m
Surrealism 44 Totemism 59, 131-4, 137, 152-3,
Swastika 120, 129, 152 168, 202, 209, 217, 252, 260, 330,
Symbol, anthropological approach 339, 421
INDEX 469
Trance 224, 238; see also Spirit Wannijn, R. 4o8n.
medium cult Ward, J. S. M. ii9n., 268n.
Trevor, E. 3i2n. Warner, W. L. 204, 416, 418
Trobriands 140, 144, I45n., 151-3, Wasche, E. io3n.
220, 221, 369; kula, 382, 397 Weber, M. 86
Tropology 49, 51 Weitman, S. 336m, 35on.
Trotter, W. 132 Werbner, R. 199
Trumbull, H. C. iion. 117-18, 179, Westermarck, E. 59, non., 122, 128
Weston, Jessie L. 125
Turel, A. 106 Westropp, H. M. 123
Turner, T. S. 166, 203-4, 272 Whatmough, J. 64n., 72m
Turner, V. W. 25, 27, 33n., 67, 68, White, T. H. 38, 173m, 286n.
8in., 82, 86, i25n., 155, 166, 170, Whitehead, A. N. 138
186, 189-95, 196, 199, 205, 209, Whitehorn, Katharine 266n.
214-15? 2 34, 244, 246, 405 Whitman, W. 33, 375
Turney-High, H. H. 263 Whittier, J. G. 343
Tylor,E. B. 27,102,108, n o , 111-15, Whorf, B. 172
131, 217, 294^, 309^ Wig 291-4
Wilde, O. 32
Unconscious 40, 60, 101-3, 149, 151- Wilder, A. 123
3, 170, 212, 217, 2i9n., 222, 225, Wilson, Edmund 35, 36
237; collective, 157 Wilson, Monica 27, 186-9, I 9 I > I92
Urban, W. M. 50-3, 190 Wilson, T. 120
Winter, E. K. io6n.
Valerian, J. 266, 285 Wissler, C. 129
Values 20, 28, 51, 56, 77, 79, 84, 89, Witchcraft 35, 68, 121, 161, 199, 207,
136, 145, 163, 164, 172, 184, 185, 218
196, 246, 371 Wolf, A. 69-70
Van Gennep, A. see Gennep Wordsworth, W. 71,375
Verlaine, P. 31, 32 Wundt, W. 153
Vesey-Fitzgerald, S. 388
Villon, F. 30 Yalman, N. 166
Virgil, 71, 27on. Yang, M. C. 69
Vivaldi, A. 75 Yonge, Charlotte 268, 271
Volkelt, J. I03n., 147, I48n. Young, Michael 89
Voss, J. V. 101 Young, Michael D. 244
Wake, S. 116, 123 Zainal-'Abidin 319

Wallace, A. 219 Zarnecki, G. 39n.
Wallace-Hadrill, J. M. 292 Zimmer, H. 290

(Routledge Revivals) Raymond Firth - Symbols - Public and Private-Routledge (2011)

Uploaded by

Copyright:

Available Formats

(Routledge Revivals) Raymond Firth - Symbols - Public and Private-Routledge (2011)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Routledge Revivals) Raymond Firth - Symbols - Public and Private-Routledge (2011)

Uploaded by

Copyright:

Available Formats

Routledge Revivals

The study of symbolism is popular nowadays and anthropologists have

Professor Firth examines definitions of symbol. He traces the history of

A Library of Congress record exists under ISBN: 0045730113

ISBN 13: 978-0-415-69466-7 (hbk)

Primitive Economics of the New Zealand Maori

Elements of Social Organization

Primitive Polynesian Economy

This book is copyright under the Berne Convention. All rights

© George Allen & Unwin Ltd 1973

This edition not for sale in the U.S.A.,

Printed in Great Britain

My anthropological interest in symbolism was aroused more than

they show something of the richness of material open for investiga-

In 1967, as the guest of Victoria University, New Zealand, I de-

erial and general argument of Chapter 9 has appeared in another

Page 196, line 9 read Heisenbergian not Heisenburgian

Symbolization is a universal human process. But we still need to

generating and receiving effects otherwise reserved for the object

POPULAR RECOGNITION OF SYMBOLS

trip was 'primarily symbolic' it was recognized that it might sym-

almost unmanageable pressures. It takes another form when Ameri-

trade-union leader, was 'the symbol of the industrial side of the

Modern popular views of law and morality are rich in recognition

a very limited period, illustrates the extent to which the notion of

really are. Complications in the concept show too in a feminine

person already a symbol - a kind of symbol to the second power.

issue was thought to be very important because it was held that

thropologist when confronted by local political party symbols of a

AN ANTHROPOLOGICAL APPROACH TO SYMBOLISM

attention on the interpretation of those elements of behaviour

ing gifts, of showing flags. I deliberately also try to consider private

Literary interest in symbols can be traced far back in history. In

irrational quality of language itself, trying, for example, to express

The author of one of the most important works in this field,

symbolization and empirical reality - which has relevance for all

anthropologists may be deeper. In The Turn of the Screw we are

society photographers would be implicated, as would even the pur-

literature among much of general intellectual interest have raised

Analogous processes of looking for symbols and discussing them

a dragon or serpent (as on the Percy tomb at Beverley or a tym-

Of the many sculptured presentations of this theme* one of the

as it does today - a search for some concrete representation of what

The avowedly symbolist movements in art in the late nineteenth

promptings seemed to animate most artists, whether their major

symbolism which employs abstractions without parallel in visual

evoke reaction without the mediation of other images. According to

only was the church struggling to establish itself against paganism

grasp. This theme, of symbol as an aid to understanding, penetrates

conceptions such as Lord, Father, Master and emblems such as Lion,

This idea of progressive interpretation, especially as it concerned

on a dualistic principle, and its analysis of material from myth and

tianity, Islam-while others are inclined to a kind of eclectic acknow-

Most anthropological treatment of symbols has taken the meaning of

But there are three points of importance about such symbols in

autonomous realm, a 'world'. It is the first presentation of the world

mediaeval Christian art, and even in a palaeolithic disc pattern in

serviceableness for our ends. So he held that the essential function

But definition of symbol in terms of conscious design has difficul-