The Voice of Prophecy: And Other Essays 9781785335570

Table of contents :
Contents
Foreword – Edwin Ardener’s Prophetic Vision
Introduction
Acknowledgements
1 Social Anthropology and Language (with editorial preface)
2. The New Anthropology and its Critics
3 Language, Ethnicity and Population
4 Belief and the Problem of Women
5 Some Outstanding Problems in the Analysis of Events
6 ‘Behaviour’ – a Social Anthropological Criticism
7 Social Anthropology and Population
8 The ‘Problem’ Revisited
9 The Voice of Prophecy – Further Problems in the Analysis of Events
10 ‘Social Fitness’ and the Idea of ‘Survival’
11 Comprehending Others
12 The Problem of Dominance
13 Social Anthropology and the decline of Modernism
14 ‘Remote Areas’ – some Theoretical Considerations
15 Witchcraft, Economics and the Continuity of Belief
16 Social Anthropology and the Historicity of Historical Linguistics
17 Edward Sapir, 1884–1939
18 The Construction of History: ‘Vestiges of Creation’
Postscript 1 – The Prophetic Condition
Postscript 2 – Towards a Rigorously Empirical Anthropology
Appendix: Edwin Ardener – a Bibliography
Notes
References
Index

Citation preview

The Voice of Prophecy And Other Essays

E D W I N A RDEN ER

Foreword by MICHAEL HERZFELD Harvard University

Edited and with an Introduction by MALCOLM CHAPMAN

Postscripts by MAR YON MCDONALD and KIRSTEN HASTRUP

Second and Expanded Edition

berghahn NEW YORK • OXFORD www.berghahnbooks.com

Published in 2007 by Berghahn Books www.berghahnbooks.com

© 2007, 2018 Shirley Ardener Original edition © Blackwells 1989 Second and expanded edition published in 2018

All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without written permission of the publisher.

Library of Congress Cataloging-in-Publication Data A C.I.P. cataloging record is available from the Library of Congress

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library

ISBN 978-1-78533-769-7 hardback ISBN 978-1-78533-557-0 ebook

Contents

Foreword – Edwin Ardener’s Prophetic Vision by Michael Herzfeldvii Introduction by Malcolm Chapmanxv Acknowledgements by Malcolm Chapmanxxxvii The Ardener Papers   1 Social Anthropology and Language (with editorial preface)

1

  2. The New Anthropology and its Critics Appendix 63

45

  3 Language, Ethnicity and Population

65

  4 Belief and the Problem of Women

72

  5 Some Outstanding Problems in the Analysis of Events

86

  6 ‘Behaviour’ – a Social Anthropological Criticism

105

  7 Social Anthropology and Population

109

  8 The ‘Problem’ Revisited

127

  9 The Voice of Prophecy – Further Problems in the Analysis of Events

134

10 ‘Social Fitness’ and the Idea of ‘Survival’

155

11 Comprehending Others

159

12 The Problem of Dominance

186

13 Social Anthropology and the decline of Modernism

191

vi  Contents

14 ‘Remote Areas’ – some Theoretical Considerations

211

15 Witchcraft, Economics and the Continuity of Belief 

224

16 Social Anthropology and the Historicity of Historical Linguistics

236

17 Edward Sapir, 1884–1939

255

18 The Construction of History: ‘Vestiges of Creation’ Editorial Note 272

264

 he Prophetic Condition Postscipt 1 – T by Kirsten Hastrup274 Postscript 2 – Towards a Rigorously Empirical Anthropology by Maryon McDonald279 Appendix: Edwin Ardener – a Bibliography

282

Notes286 References315 Index337

Edwin Ardener’s Prophetic Vision

Michael Herzfeld Edwin Ardener’s was a prophetic voice in precisely the sense of prophecy that he lays out in these pages. Yet it was also muted, to use a term to which his thought gave new precision, by an academic politics that, today more than ever, risks surrendering to pseudo-populist calls for greater simplicity, less theory, and more ‘relevance’ – all claims that would have prompted Ardener to ask what ideological and other motivations lie behind attributions of triviality and obscurantism. Ardener’s style of argument was never either gratuitously trivial or deliberately obscure. Nevertheless, his attacks on the absurdity of a clearcut division between the material and the symbolic (he insisted that there was ‘no distinction between the material and the ideal’, p. 184) must have made his insights seem fundamentally unpalatable and even dangerous to those of positivistic bent. Those battles continue today, and his voice is as deeply needed as ever. To an extent far greater than that noted by his three distinguished former students whose essays grace the present collection, Ardener anticipated numerous central issues in the social sciences today. His ‘blank banners’ speak to the ‘hidden transcripts’ celebrated in the work of James C. Scott (1990), while the ‘muted’ discourses of ‘englobing’ marginals offers a way around the limitations of the more bludgeon-like concept of ‘resistance’. His recognition that anthropologists are as liable to the fear of categorical contamination as any remote non-European population extends and amplifies Mary Douglas’s (1966) original analyses of pollution and taboo and lucidly foreshadow her own later attention to ‘modern’ institutions (e.g. Douglas 1986). And he saw, long before Johannes Fabian (1983) braved the ire of established scholars, how anthropological ways of representing cultural ‘others’ as ‘synchronic unities’ (p. 254, n. 13) are exercises in the pre-emptive, ideological manipulation of time that Fabian calls ‘allochronism’. He also anticipated (p. 156) Fabian’s important rediscovery of the evolutionist bias that persisted in the work of evolutionism’s most ardent critics, the functionalists. Moreover, although not the first to work

viii

Edwin Ardener’s Prophetic Vision

in a European society, Ardener also quietly inserted himself into that genealogy through his work on Gaelic culture and thereby furthered the newly-begun process of weaning the discipline from a deeply rooted exoticism (although his musings on the idea of remoteness also temper the current thrust to dismiss all studies of rural and tribal societies as irrelevant to the fashionable focus on modernity). Only these many years after his death has Europeanist ethnography perhaps at last begun to realize what he clearly had already seen as its potential to illuminate the Eurocentric biases of the discipline itself. Even at what were superficially, at least, his most structuralist moments, Ardener never embraced received wisdoms uncritically. Thus, his elaboration of the distinction between paradigmatic structures and the syntagmatic chains of events that give them existential reality permits a way out of the structuralists’ rationalistic universalism – a stance that renders Lévi-Strauss’s thought, in particular, speculative and programmatic rather than empirical and ethnographic. Ardener’s move permits us to see such underlying structures as both particular to specific social groups and yet also subject to re-elaboration, change, and border-crossing – a far more flexible instrument than Lévi-Strauss’s grand vision could entertain. Here Ardener’s profound knowledge of the history of linguistics also freed him from a slavish dependence on the thought of Saussure and from a too-easy acceptance of others’ invocation of that reconstructed old master. For this reason, too, he was equipped to engage with American and British socio-linguistics and the ethnography of speaking. He did so, moreover, long before most of his local colleagues saw fit to look with anything other than amused condescension on the importance of language as a cultural instrument, finding in the pursuit of such unlikely topics as etymology, not only a basis for the critique of anthropological terminology, but also a source of traces that, as J.L. Austin [1971:99–100] similarly foresaw, could point us back in the direction of reading sedimented significance in the inchoate (those blank banners again!). He listened carefully for the ‘muted’ voices of groups disenfranchised in their home societies and in anthropology alike. That quality emerges most famously in his work on what he rather slyly called ‘the problem of women’. Although a man (and conscious that this might be problematic for the debate), he also anticipated much feminist work in anthropology with his justly famous pair of essays on this topic, opening up questions of hegemony that went far beyond gender to highlight concealed but extremely powerful structures that oppress and exclude in many domains of social life. Like many feminists since his time, he saw the exclusion of women as both unjust in its intentions and effects and symptomatic of these larger hierarchies. He expressed a sympathetic understanding of the fact that many of his female colleagues were no more willing than men to probe these inequalities. But he could also rarely resist the opportunity for gentle irony; when he describes Bakweri women performing a men’s ritual as ‘dames in an order of chivalry or girls at Roedean [a famous English high school] . . . performing a male scenario’ (p. 85), it seems not unreasonable to suspect that he hoped some of those female colleagues might spot their own foibles in that ethnographic mirror. His rare ability to see the common humanity of scholar and subject also meant that he saw his informants as engaged in theory-building operations,

Edwin Ardener’s Prophetic Vision

ix

both constrained by such social boundaries as those of gender and class. In this, Ardener anticipates the practice theorists’ later recognition of all human beings’ capacity to theorize their social situations, but without the implicit reservation in Pierre Bourdieu’s (1977: 18) rather condescending nod to ‘semi-theoretical dispositions’, and without Anthony Giddens’s (1991) curious insistence on a fixed historical moment for the emergence of modernity. Indeed, Ardener offers a refreshingly ironic view of modernist claims to totalizing collective achievement and originality (pp. 192–193); he goes on to show how the modernist project, more recently dissected by Scott (1998) in his exploration of the modernist state, encompasses the architectural functionalism of Le Corbusier and the anthropological functionalism of Malinowski alike (p. 199), but also that it perpetuated some of the very biases that it claimed to have eradicated. That critical perspective allowed Ardener to challenge the comforting boundary that modernism places between itself and the exotic. In particular, he recognized the continuity between local and so-called scientific knowledge (p. 140) in a way that uncannily foreshadowed the elegant demonstration of this point by Akhil Gupta (1998). Most important of all, perhaps, is the way in which his willingness to grapple with the inchoate and the implicit prefigures (a blank banner in its own right?) much of the present, practice-based concern with ‘indeterminacy’ – what he called, in the context of theory-building, ‘admirable provisionality’. In that sense, he rejected the false legibility that, Scott assures us, is the dominant goal of modernist statecraft, and that Ardener himself identified as the positivists’ strange disregard for inconvenient or inchoate facts. Ardener was not uninterested in systems, but saw them as products of contingency rather than as pre-existing verities. He was thus a practice theorist, and a theoretical practitioner, literally (if I may so put it) avant la lettre. If Ardener had done nothing more than anticipate a great deal of modern theory, the present collection would have deserved the investment of a reprint edition. There is much to be learned from pondering paths rediscovered or ideas reinvented. My hope, however, is that this publishing event will achieve something much more significant still: a long-overdue recognition that Ardener not only forged ahead of today’s mainstream but bequeathed a legacy of ideas that can regenerate and redirect anthropological thought today. Among the conceptual challenges he still poses for anthropology, his recasting of the peculiar relationship between anthropology and linguistics remains both prophetic and salutary. Virtually alone among British social anthropologists of his time, although with an increasing coterie of followers today, he saw that language was a more than a necessary field skill and more, too, than simply the ‘classificatory terminology’ so beloved of structuralists and componential analysts alike. Moreover, because he remained firmly committed to a social anthropological perspective, he avoided the tendency, powerfully emergent in American scholarship at that time, to cocoon the anthropological discipline in the trappings of a separate ‘subfield’ and so, unintentionally but destructively, to shield the study of linguistic and other cultural dynamics from productive mutual engagement. He also anticipated the trend toward a more reflexive idiom of anthropology. More specifically – and in ways that probably influenced my own work (e.g.

x

Edwin Ardener’s Prophetic Vision

Herzfeld 1987) more than I realized at the time – he saw that anthropology itself was a cultural artifact, and that Mary Douglas’s (1966) insights into the fear of categorical confusion could therefore be usefully deployed to understand why some anthropologists shied away from ‘cultural relativism’ (p. 11). Today, a very similar fear appears to grip the exponents of various public ideologies of ethnic, religious, and scientific purity – exponents for whom anthropology itself has become a dangerous source of ‘anomaly’ in this sense. Perhaps some such as yet unrealized critique was another of Ardener’s own blank banners; his writing is not overtly political in the more obvious sense of that term, yet it is deeply imbued with a fundamentally political determination to root out the residue of older ethnocentrisms at the core of a discipline too prematurely convinced that it had already cleansed itself. Especially with the greater specificity that hindsight can now lend our reading of Ardener’s work, this campaign is as worth pursuing in our own time as it was in his. Ardener was not against statistics or science, as Malcolm Chapman points out in his very useful introductory essay; he was, however, locked in a fierce battle against scientism. He saw that the indeterminacy of human action – note the etymologically erudite and sociologically elegant rejection of ‘behaviour’ as the relevant term here – introduced anomaly into the most rule-governed domains of social activity, leaving analysts in what, with delicious irony, he calls the ‘scientifically undignified position’ (p. 48) of having to consult the circumstantial in preference to the definitional. Here, clearly, the attention of such linguistic philosophers as Austin and Wittgenstein to the conceptual priority of use over reference was a significant influence, one that we can see especially well explored in the work, cited by Ardener, of another of his students, the late Malcolm Crick (1976). Ardener’s own statistical work on demography had led him to a critical (and indeed almost Whorfian) understanding of the impact of categorical choices – such as the implication of the term ‘population’ – on the meanings that analysts attributed to statistical information. One cannot really understand such events as the Biafra conflict without taking into account the prior processes of reification whereby ethnic identities gained salience. In those processes, colonial statisticians had played a key role. The close attention Ardener paid the semantic lability of ethnic labels was more than a display of linguistic erudition. Like Whorf, he was a practical person; and, again, like Whorf, he saw clearly that the choice of terminology played a determining role in the direction of future action, whether in the immediate environment of social life or on the world stage of wars and nation-building. This critical perspective on the use of language had profound implications for Ardener’s battle to achieve some form of cooperation between demography and the kind of social anthropology that he respected and practised. He saw this, too, as a practical matter, calling for considerable caution in the face of apparent methodological progress. His observation (p. 117) that improvements in statistical method actually concealed or refigured the local realities they were intended to address, for example, has urgent implications for the current passion for reducing human experience to market surveys. Here again, Ardener’s work represents more than a present-day version of Da Vinci’s flying

Edwin Ardener’s Prophetic Vision

xi

machine—a clever idea that was never translated into material action. To the contrary, it offers a template, to use another of his favoured terms, for a political critique of the rhetoric of ‘excellence’ that Marilyn Strathern and others have attacked under the label of ‘audit culture’ (Strathern, ed., 2000). Indeed, it is hard to imagine that he would not have participated, with his familiar stance of delicate gusto, in that important work of social criticism. His appreciation of the relationship between language and other forms of social action is central to any understanding of his critical labour. For him, terminology was the surface realization of a larger difficulty: the tendency of social science to ignore its own cultural entailments in pursuit of an entirely spurious notion of objectivity. Most notably in his remarks on the effects on ethnographic reporting of the fact that in many societies it was primarily men who offered commentary on the roles and attitudes of both men and women – the key effect of which was to occlude the actual opinions of the women – he saw the social practices of the people studied and the scholarly habits of those who did the studying as liable to mutual reinforcement. Thus, claims to objective knowledge were made possible by creating a categorical separation between ethnographer and subject and thereby obscuring the collaborative relationship and the real insights that it generates. For Ardener, therefore, ‘admirable provisionality’ was the only acceptable alternative – a stance that made of every categorical certainty a semiotic worthy of critical attention in itself. It was not a stance calculated to gain the affection of those for whom crunching numbers was sufficiently scientific on its own merits. But it is one that honesty should compel us to adopt today if we are serious in our claims of recognizing indeterminacy as a key aspect of the human condition. Ardener was an original and provocative thinker. His anthropology not only anticipated some present concerns, but actually addressed them in ways I have tried to sketch in this brief prefatory note. Whether we will revert to the language of syntagmatic and paradigmatic structures may be dubious, but the critical attention that he drew to the distinction between event and programme should remain in the forefront of analytical concerns. So, too, should our entailment in the forces that have often obscured that distinction. If we want to practise reflexivity, turning to Ardener’s legacy of guidance is a good place to begin. I would like, in that spirit, to conclude in a somewhat personal vein. I was not, strictly speaking, an Ardener student, except in the sense that in those days virtually all of the students at the Institute of Social Anthropology at Oxford went to his lectures and tried to absorb his complex and humane view of the discipline. But I might never have had a career as an anthropologist without his benign recognition (or so it was related to me) that, while my undergraduate career had been disastrous, I might be worth admitting to the Institute since I had meanwhile managed to pursue my research interests in language and symbolism despite that initial discouragement. Was this prophecy? If so, was it prophecy in the sense that I understand Ardener to mean by that term – an ability to identify in the inchoate (and I was all of that!) something that might eventually take on a more identifiable meaning? Was his intervention, and the intellectual excitement to which it led me, a prophecy of only the self-fulfilling

xii

Edwin Ardener’s Prophetic Vision

sort? It would be unprofitably self-indulgent to speculate further about this, but I mention it because, in a very concrete but characteristically indirect way, Ardener’s insights touched my life and intellectual development as few others have done, even while his influence often seemed nebulous and inchoate at the time; and because I suspect, and hope, that the re-publication of this book will allow many others to benefit from his radical understanding of the human condition in the same, productively roundabout way. His intellectual and personal generosity, which in these pages he extended with striking consistency to at least one distinguished detractor of his work, was as great as his intellectual originality. Interest in Ardener’s ideas has waxed and waned over the years, as it did in his lifetime. He sometimes seemed unnecessarily obscure, apparently wrapping his ideas in an unfamiliar algebra of neologisms and puns that were themselves gold-mines of insight. His soft, self-effacing mannerisms masked sharply ironic jabs at the self-important and the portentous: good fun, but perhaps, given the forces he was attempting to confront, a provisionally self-defeating combination of gentle ivory-tower erudition and provocative challenge. Today, I suspect, he has become easier to read than in the past; this, it must be said, is partly because his students – including the three who contributed to the original publication of this book – have been both personally loyal and intellectually stimulating in their promotion of his ideas and have demonstrated some of the rich possibilities of those ideas in their own work. Beyond doubt, the internationalization of the discipline and its radical engagement with the limits of positivistic social theory has also created a more receptive atmosphere for Ardener’s thinking, while global politics may have made his message of ironic critique increasingly urgent. This new edition will allow a new and more receptive audience to come to grips with Ardener’s distinctive mode of analysis and understanding, bringing it more clearly into the mainstream of anthropological thought not only as a historical contribution but also, and especially, as a source of new reflections. The intellectual bequest of a brilliant and compassionate human being surely deserves no less; and the world has sore need of the quizzical honesty that is the enduring trademark of Edwin Ardener’s life and work.

R E F E R E N CE S

Austin, J.L. 1971[1956–57]. A Plea for Excuses. In Colyn Lyas, ed., Philosophy and Linguistics (London: Macmillan), pp. 79–101. Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Trans. by Richard Nice. Cambridge: Cambridge University Press. Crick, Malcolm. 1976. Explorations in Language and Meaning: Towards a Semantic Anthropology. New York: John Wiley/Halsted. Douglas, Mary. 1966. Purity and Danger: An Analysis of Concepts of Purity and Taboo. London: Routledge & Kegan Paul. ———. 1986. How Institutions Think. Syracuse, NY: Syracuse University Press. Fabian, Johannes. 1983. Time and the Other: How Anthropology Makes its Object. New York: Columbia University Press.

Edwin Ardener’s Prophetic Vision

xiii

Giddens, Anthony. Modernity and Self-Identity: Self and Society in the Late Modern Age. Cambridge: Polity. Gupta, Akhil. 1998. Postcolonial Developments: Agriculture in the Making of Modern India. Durham: Duke University Press. Herzfeld, Michael. 1987. Anthropology through the Looking-Glass: Critical Ethnography in the Margins of Europe. Cambridge: Cambridge University Press. Scott, James C. 1990. Domination and the Arts of Resistance: Hidden Transcripts. New Haven: Yale University Press. ———. 1998. Seeing Like a State: How Certain Schemes to Improve the Human Condition have Failed. New Haven: Yale University Press. Strathern, Marilyn, ed. 2000. Audit Cultures: Anthropological Studies in Ethics, Accountability, and the Academy. London: Routledge.

STATE OF NEW YORK

________________________________________________________________________

10322--A

IN ASSEMBLY March 15, 2006 ___________

Introduced by M. of A. FIELDS, EDDINGTON, SWEENEY, RAMOS, ALESSI, ENGLEBRIGHT, DINOWITZ -- Multi-Sponsored by -- M. of A. THIELE -- read once and referred to the Committee on Alcoholism and Drug Abuse -- reported and referred to the Committee on Codes -- committee discharged, bill amended, ordered reprinted as amended and recommitted to said committee

AN ACT to amend the mental hygiene law and the social services law, in relation to a pilot program in Suffolk county regarding recovery homes for rehabilitation of persons with chemical dependence

The People of the State of New York, represented in Senate and Assem______________________________________________________________________ _________________________ bly, do enact as follows: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Section 1. Legislative intent. The legislature finds that there is a significant need for housing options for persons recovering from chemical dependence. One such option is the "recovery home" which is an unsupervised, community-based group home for persons who have progressed enough in their recovery that they can live in the community, but still require the support of others in the same or similar situation. The legislature further finds that while many recovery homes are well run and offer effective programs, some recovery homes are not therapeutic and can in fact be injurious to the recovery of the occupants and the neighborhoods in which the homes are located. The legislature therefore finds that recovery homes are in need of government oversight for the protection of both the occupants and the communities in which the homes are located. The office of alcoholism and substance abuse services is hereby directed to establish a pilot program in Suffolk county to test these findings. § 2. Section 1.03 of the mental hygiene law is amended by adding a new subdivision 56 to read as follows: 56. ______________________________________________________________________ "Recovery home" or "recovery housing" means any facility registered with the office of alcoholism and substance abuse services that ________________________________________________________________________ provides an unsupervised residence for four to fourteen unrelated ________________________________________________________________________ persons who are assessed or diagnosed with chemical dependence, as ________________________________________________________________________ defined in subdivision forty-four of this section, that are referred ________________________________________________________________________ either by a provider licensed by the office of alcoholism and substance ________________________________________________________________________ EXPLANATION--Matter in italics _______ (underscored) is new; matter in brackets [ ] is old law to be omitted. LBD15458-05-6

AB

STATE OF NEW YORK

___________________________________________________________________

10322--A

IN ASSEMBLY March 15, 2006 ___________

Introduced by M. of A. FIELDS, EDDINGTON, SWEENEY, RAMOS, ALESSI, E BRIGHT, DINOWITZ -- Multi-Sponsored by -- M. of A. THIELE -- read and referred to the Committee on Alcoholism and Drug Abuse -- rep and referred to the Committee on Codes -- committee discharged, amended, ordered reprinted as amended and recommitted to said co tee

AN ACT to amend the mental hygiene law and the social services la relation to a pilot program in Suffolk county regarding recovery for rehabilitation of persons with chemical dependence

_________________________________________________________________ The People of the State of New York, represented in Senate and A _________________________ bly, do enact as follows: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Section 1. Legislative intent. The legislature finds that there significant need for housing options for persons recovering from c cal dependence. One such option is the "recovery home" which is an pervised, community-based group home for persons who have progr enough in their recovery that they can live in the community, but require the support of others in the same or similar situation legislature further finds that while many recovery homes are well and offer effective programs, some recovery homes are not therap and can in fact be injurious to the recovery of the occupants and neighborhoods in which the homes are located. The legislature ther finds that recovery homes are in need of government oversight for protection of both the occupants and the communities in which the are located. The office of alcoholism and substance abuse service hereby directed to establish a pilot program in Suffolk county to these findings. § 2. Section 1.03 of the mental hygiene law is amended by adding subdivision 56 to read as follows: 56. "Recovery home" or "recovery housing" means any facility r _________________________________________________________________ tered ___________________________________________________________________ with the office of alcoholism and substance abuse services provides ___________________________________________________________________ an unsupervised residence for four to fourteen unre persons ___________________________________________________________________ who are assessed or diagnosed with chemical dependenc defined ___________________________________________________________________ in subdivision forty-four of this section, that are ref either ___________________________________________________________________ by a provider licensed by the office of alcoholism and subs

EXPLANATION--Matter in italics _______ (underscored) is new; matter in bra [ ] is old law to be omitted. LBD15458

AB

15 Witchcraft, Economics and the Continuity of Belief

This paper1 is about the interrelations of the witchcraft beliefs of a certain African people and the various social and economic pressures that they were subject to over a period of fifty years or more. The story will show that the content of witchcraft beliefs may be subject to fashions: the beliefs change, as the world to which they refer changes. Among this people, the particular and unusual form of the beliefs was associated with economic stagnation, and many of their neighbours blamed the stagnation on the failure of the people to move in some sense into the modern world. Yet, when the opportunity for economic expansion arrived, there was a skilful, if unconscious, adjustment of their beliefs which made it possible for the people to take advantage of the expansion, without shaking the basis of the beliefs themselves. In telling this story, and it is a very interesting one in itself, we shall come upon some sidelights on the nature of witchcraft beliefs, and also perhaps draw some lessons on how far one can blame a people’s economic failure upon its cherished beliefs. Such lessons may even have relevance for populations of much greater size. The question of continuity in the structure of belief will also be raised. The Bakweri2 of West Cameroon are Bantu-speaking – the Bantu languages reach with them almost their farthest northern and western extension. Their population is not much greater than 16,000. Before the German conquest of 1894, they were living in scattered settlements round the southern slopes of Mount Cameroon. There are very few mountains in West Africa, and none is as high as this (more than 13,000 feet). It is also unusual in that it stands right on the coast, descending through a maze of foothills to the sea. At four degrees north of the equator, it is not quite high enough for permanent snow. Instead, winds loaded with moisture from 4,000 miles of Atlantic travel precipitate copious rains, and

Witchcraft, Economics and Belief 225

swathe the slopes in mist and drizzle for many months of the year. Inside the clouded summit is an active volcano from which new craters burst out every few decades. The rain and the volcanic soil have made the mountain area one of the most fertile in Africa, and forest covers the mountain up to 6,000 feet. The Bakweri lived (and live) in the thickest concentration in a belt of villages between 1,500 and 3,000 feet above sea level, but they occupied the whole base of the mountain below this very thinly, as far as the sea. With the advent of German rule, the fertility of the soils was quickly recognized, and the area was developed as plantations – initially for tobacco and cocoa, and then (with the failure of these) for various other crops of which bananas, rubber, oil-palm products and tea remain important to the present day. The acquisition of plantation land began through negotiation with village heads and the like, but after 1894, with the conquest of the mountain villages by the laissez-faire governor, von Puttkamer, alienation increased rapidly. In the 1900s, the process came under some official restraints, and reservations were established for the Bakweri on a fixed number of hectares of land per hut. Before the establishment of the plantations, Bakweri settlements had been dispersed through the mountain forest in clusters of bark-walled huts occupied by close patrilineal kinsmen and their families. These clusters were grouped territorially into what may be called ‘villages’, but on the lower slopes of the mountain, the Bakweri desire for elbow room resulted in some of these villages occupying large territories. One, for example, occupied fifty square miles – the same size as Berlin, as one German noted – but with a population of only a few hundred. On the upper slopes in the more densely inhabited belt, the settlements were more compact and had less room for manoeuvre. The village was the normal political unit, under the control of the village head and lineage elders, sometimes aided, or even superseded, by regulatory associations containing most of the male membership of the village. The Bakweri cultivated a number of tubers, together with the coarse variety of banana which in West Africa is generally known as the plantain. In addition, the villagers in the upper belt of settlement were able to climb to the grassy higher slopes of Mount Cameroon, to hunt antelope and small game. The forest itself supported elephant, which were trapped in pits. While most agriculture was women’s work, the originally staple plantain crop was a male concern. By 1890, the Bakweri economy had already changed considerably from what it must have been some forty years before. For one thing, the plantain as a source of food had been completely overshadowed by a new food crop, the Xanthosoma cocoyam, an American plant that had been introduced at the coast in the period after 1845 by a small missionary settlement from the neighbouring island of Fernando Po.3 The xanthosoma flourished in Bakweri country like Jack’s beanstalk. The rich soils and the absence of any drought made it throw up huge stems, and leaves under which a tall man could shelter. The ‘mother’ corm in the soil was not dug up annually as it is in drier conditions. It continued to throw out knob-like cormels from itself for a period of years, and these were cut off regularly while the main plant still grew. This women’s crop solved the problem of food supply, and diminished the incentive of Bakweri men to farm. Large quantities of small livestock flourished on the rich vegetation. Pigs grew fat on the cocoyam waste.

226 Witchcraft, Economics and Belief

Between 1850 and 1890, the Bakweri became rich in other ways. By trading foodstuffs to the coast, and blocking the way of expeditions into the interior, they had acquired considerable trade goods and an armament of flint-lock guns. The largest village, Buea (later less than 1,000 in population),4 had begun to instigate small raids which the German government began to advertise as a threat to the coastal establishments. By 1891, the Bakweri were at a peak: in that year Buea defeated a German expedition. The misty, forested mountain was a barrier that reduced the Germans to small importance in Bakweri eyes. The German colonial government was starved of funds at this time.5 All this was shattered by a single blow in December 1894.6 A well-mounted expedition found the Bakweri now unprepared; the commander, von Stetten, looked round in disbelief for the source of the threat. He saw a timid people living in small huts in the forest surrounded by fences to keep the flourishing livestock away from the even more flourishing cocoyams. In the next decade, the Bakweri were systematically tidied up. Scattered huts were grouped in lines and lands alienated for plantations. Von Puttkamer enthused over the neat settlement he had made out of a village called Soppo: all that is now required, he said with full Prussian sentiment, is the village postman going from door to door.7 The plantations found it difficult to find adequate local labour. The Bakweri were neither numerous enough nor used to the work. So began an influx of labourers from outside, who came, over the decades, to outnumber the Bakweri by three or more to one. The predominance of males among the migrants began a drift of Bakweri women into concubinage and prostitution, which became a byword. The Germans were succeeded in 1914 by the British, who did not dismantle the plantation industry. They did pursue for several decades a ‘pro-Bakweri’ policy – reserves were enlarged and after the Second World War a land policy was suggested whereby plantation lands might be excised for Bakweri use. By then, however, the Bakweri had acquired a reputation as unprogressive. The term then in public use was ‘apathetic’. Far from requiring more land (it was said), they wasted what they had; they let it out to strangers at a profit, or (worse) they were even too apathetic to make a profit. They did not respond to government exhortations. They were said by one observer to be ‘unindustrious but grasping’. The women ran away from the men because they had to do all the work. The Bakweri men were too apathetic to control the women. They were about to die out. They lived with pigs, in huts that were falling down. This view of the Bakweri was heartily endorsed by the large numbers of industrious migrants working in the plantation industry – which after 1949 was nationalized and feeding its profits to the government. The territory itself became in 1954 partly autonomous, with an elected legislature. There was to be no putting back the clock, and the present Federated State of West Cameroon needs every franc it can get from its only major industry. But in the later 1950s, the Bakweri suddenly stopped being apathetic, and made fortunes in peasant banana-growing (Ardener, Ardener and Warmington 1960: 329–32). So what happened? It is in this context that one must turn to the witchcraft beliefs of the Bakweri. First of all, in the whole Cameroon forest zone in which the Bakweri live, there is an ancient and general belief in witchcraft. This kind is known under the name liemba, and the term itself goes back to a common Bantu

Witchcraft, Economics and Belief 227

word-form (it may therefore date from the Bantu dispersion). The belief in liemba was deeply seated just before the German conquest, we know, because the first German expedition, which the Bakweri defeated in 1891, was believed by them to have been sent because they had just hanged two women for liemba. The names of these two women were given to me over sixty years later. Of course, coastal Christians, both black and white, were known to object to witch-hangings. It was common knowledge among the Bakweri that the missionary settlement of Victoria at the coast had gained its convert population from escaped witches and other malefactors from their own villages. Liemba is a ‘classical’ form of witchcraft. It is generally regarded as inborn, although it can sometimes be passed on to a person who is without it. Witches may be of either sex and are said to leave their bodies at night and ‘eat’ people so that they become ill and die. What is eaten is the elinge: the word means ‘reflection’ and ‘shadow’. Many sicknesses would be attributed to witchcraft, but the essential diagnosis was always made by a diviner. In the milder cases, a treatment was prescribed that would defeat the witch. In serious cases, the suspected witch would be named and made to drink sasswood medicine. If the suspect vomited, he was innocent; if not, he was guilty and was hanged. Every village had a witch-hanging tree (Ardener 1956: 105).8 The Bakweri were not greater believers in liemba than were their neighbours; indeed, they lacked some of the elaborations of the forest peoples, in which there was a post-mortem examination in order to determine, from the position of blood clots, the actual kind of witchcraft from which a person died. However, the Bakweri belief partook somewhat of their misty mountainous environment. With plenty of room to move in, they came to believe that they were moving into scattered hamlets to avoid witches. It was well known that you should never live too near your patrilineal relatives for they suffered from that chief Bakweri vice: inona. This word may be translated as ‘envy’, but the flavour is of the most ignorant, ill-wishing envy. Where inona was, there too was liemba. As among the Tiv, there were grounds for the belief that close patrilineal relatives would be the most envious, for relations with these would be influenced by the rules of inheritance by which a man’s brothers and half-brothers had a substantial claim over his property at his death, prior to his own sons. The Bakweri were very conscious of inona. The accumulation of property – chiefly in goats, pigs and dwarf cattle – was the chief means of establishing status in this jealously egalitarian society. Yet it was collected mainly to be destroyed. At a potlatch ceremony known as ngbaya, performed only by the very rich, hecatombs of goats, fowls and cows of a tsetse-free dwarf breed would be killed and distributed among those attending. To receive a great share at ngbaya was a mark of status, and also a severe blow, for the recipients would be expected to ruin themselves even more splendidly. As a fascinating by-product, it was only at a ngbaya ceremony that boys with certain mysterious headaches could be cured. The cured boys often went on to be medicine men. The extraordinary psychic energies bound up in the acquisition of property, and the twin emotion of inona, give Bakweri beliefs an interesting flavour of their own. The envy of relatives was further stimulated and assuaged by making any riches in livestock that survived a man’s death the subject of another ngbaya-like ceremony, known as eyû. Here again the goats stood trembling, tethered in rows, their heads to be cut off one by one. The last prize sacrifice

228 Witchcraft, Economics and Belief

(if possible, a slave) was the subject of a special dance called motio, and the head was to be felled at a blow (Ardener 1956: 76–7, 89–90). It seems, then, that the Bakweri attitudes to property, to envy and to liemba witchcraft were closely bound up in a highly emotional knot. One should imagine an essentially proud, but rather inward-looking people, on average rather slight in build, shivering in a damp and foggy climate and dwelling in bark-walled huts, but having also enjoyed, for many decades, an environment where livestock and the cocoyam thrived, where extra food could be obtained from hunting on the isolated mountain top, and where trade goods arrived from the coast – a strange mixture of deprivation and riches. The conquest and the establishment of the plantations must have come as a great shock. The first carpet-bagger Bakweri interpreters and other servants of the Germans moved in, and many quaint iron houses were built in the villages by the new rich. Most of the Bakweri in the villages presented a dull, lifeless impression to the German administrators, the women now clad in colonial cotton frocks, the men timid and withdrawn. We now embark upon that period of Bakweri apathy to which reference was to be made so often for fifty years of this century. I first went to Bakweri country in 1953 and by then they had for decades worried themselves with the problem of how to control the entry of their women into prostitution and concubinage with the large migrant population. They were convinced that they were dying out. Villages were dwindling. The spread of venereal disease from the plantation centres to the villages must have been responsible for the undoubted degree of reproductive disturbance among them, possibly exacerbated by environmental factors (Ardener 1962). Anyone entering a Bakweri village in those years could not but be impressed by the lack of those hordes of children which are so typical of West Africa, and by the empty tin houses of the early Bakweri carpet-baggers. Empty, that is, except for zombies. For in the intervening years the Bakweri had come to believe in a new kind of witchcraft compared with which the old liemba was regarded as almost a harmless trifle. This new kind was called nyongo. It was believed to have been brought in from outside by ‘wicked people’ at about the time of the First World War. It took a peculiar form. A person with nyongo was always prosperous, for he was a member of a witch association that had the power of causing its closest relatives, even its children, to appear to die. But in truth they were taken away to work for their witch-masters on another mountain sixty or seventy miles to the north: Mount Kupe in the territory of the Bakossi people. On Mount Kupe, the nyongo people were believed to have a town and all modern conveniences, including, as will be seen, motor lorries. Nyongo people could best be recognized by their tin houses, which they had been able to build with the zombie labour force of their dead relatives. How this belief grew up, and by what processes the association of dying children and the ownership of tin houses became so firmly fixed, cannot easily be traced. But by 1953 the belief had taken such a hold that no one would build a modern house for fear of being accused of possessing nyongo. Dying people who were being taken in nyongo were expected to be able to see the witch and to name him. I knew an old tailor, trained in German times, who lived in a plantation camp while his tin house in his home village was shunned, and people were dying with his name on their lips. A nyongo witch himself was not safe, it was believed, for one

Witchcraft, Economics and Belief 229

day his fellow nyongo people would come to him and take him too, and his house would stand empty. In this atmosphere, any conspicuous material success became suspect. Men with no deaths in their families, who built their houses visibly with their own hands, could hope to still the worst accusations, but most people were slow to exhibit the fruits of success for fear of suffering (or being accused of) nyongo. The ancillary ornaments to the belief in nyongo were rich and circumstantial. When a person was taken in nyongo, he was not really dead; the nyongo witch, acting as one of an association, would insert a dead rat under the body so that the corpse would appear to smell badly and would be quickly buried. Then he would invisibly abstract the body from the grave. Wherever the nyongo witches and their victims were, they manifested themselves as vekongi: zombie spirits. They occupied the tin houses, causing lights to be seen and noises to be heard. There would be unexplained knocks on doors in the night. A nyongo witch when dead was known to leave his grave himself as vekongi. To obviate this, known nyongo witches were buried face downward so that they would move deeper into the earth. Sometimes the head was severed. In 1953, the belief in nyongo was deep-seated in all classes of the population, and there is no doubt that all ‘economic’ initiative was much affected by the climate of that belief. The non-Bakweri origin of some of its content can be confirmed. The name (=nyungu) is a foreign word, meaning ‘rainbow’ in the language of the neighbouring Duala people, among whom persons of unusual prosperity are supposed to have captured the magical python which manifests itself in the rainbow. Yet the Bakweri did not take over this belief in the rainbow source of the nyongo power. Again, the localization of the zombie town in another tribe, on a mountain they may not have heard of before the colonial period, tends to confirm the Bakweri belief that nyongo was a new witchcraft. It is, however, interesting that although the belief in zombies is found in East and Central Africa, it is not commonly found in West Africa. Its occurrence in the West Indies has been supposed by some to be a spontaneous growth. It is of some linguistic interest that the Bakweri word for the act of giving a relative to nyongo is sómbà, ‘to pledge’ or ‘pawn’, and in the tense used takes the form sómbî. For example, à mò sómbî ô nyòngò: ‘he has pledged him in nyongo’. Possibly we should look somewhere in our area for the African transmission point of the Caribbean belief. Even if the circumstantial realien of the nyongo witchcraft were new to the Bakweri, an underlying belief in the possibility of witchcraft ‘pawns’ or ‘pledges’ may long have existed – an idea that could have become flesh on either side of the Atlantic. Once more, in any event, the propertygathering connotations of the ‘pledge’ may be noted.9 The whole tale in its extraordinary detail must be viewed as an exaggeration of those trends already marked in Bakweri belief: the powerful ambivalence towards riches and property; the sudden breach of the isolation of the society, accompanied, as is all too commonly the case with the victims of decline in power or status, by a sense of collective guilt; the low fertility, and the fear of dying out. Perhaps all this turned against those who were thought to have benefited by the events that had caused so much damage. Envy, disaster, property and witchcraft were once more in close association. So much for Bakweri apathy. To anyone who knew them well, the quiet exterior seemed to cover a dangerously explosive mixture.

230 Witchcraft, Economics and Belief

In 1954 things began to happen. After many years of rejecting all avenues of economic profit, the Bakweri had begun to take up commercial banana-farming. Things began very slowly at first, the stimulus coming from a group of educated Bakweri and from government officials. The idea was that individuals should not corner any economic markets, but that the villages themselves should work as cooperative units. There would be an office and organization; the old and the young would all gain. The commercial banana was a crop not dissimilar to the traditional plantain. The first steps were shaky and not free from inona. What clinched the operation was the immediate and disproportionate monetary return at the end of the first full year (1953). An embarras de richesses poured in: sums of the order of £100 per farmer entered Bakweri villages, with their tumbling huts and empty tin houses.10 By common tacit consent, the making of small improvements, such as cementing a floor, began to be taken as not necessarily of nyongo origin. Then, in 1955 and 1956, a masked figure called Obasi Njom began to glide about the villages. Bakweri villages had used over £2,000 of the first banana revenue to purchase the secrets of a witch-finding association from the remote Banyang tribe. With the new advanced ritual technology, they began to clear the nyongo witchcraft from the villages. The extraordinary logic of this behaviour makes one almost feel that it could have been done only by conscious reasoning. Jarvie (1963) would doubtless contend that this was so. For with the nyongo threat removed, obviously the next revenue could be spent on self-advancement, and yes, even on tin houses. Yet the change came like this: On 9 January 1955, in the village of Lisoka, a young man named Emange Isongo was found dead at the foot of a palm tree which he had climbed to tap palm wine. His climbing rope, it was said, was unbroken. The village was shocked. Soon after, the story spread that Emange’s father was one of the nyongo people, and that it had been his turn to supply a victim. Emange, it seemed, had fought the nyongo people at the foot of the tree, but his own father had clubbed him to death. It was Emange’s sisters and a surviving brother who came back from a diviner with this story. Somehow, this particular event greatly upset the youth of the village, and they held meetings, at which the proverb was often quoted, mèfondo mekpâ mεεsɔ mìikìsεnε: ‘When the hair falls from the top of the head the temples take over’. This means that it was natural for the old to die before the young, whereas the reverse had been happening. Prominent in the succeeding period was a middle-aged traditional doctor named Njombe, who dealt with dangerous nyongo cases. As he put it to me in his own words in the local Cameroon creole: So Emange died. So everyone vex and the young men say they want to run out for town as he be young man too. So we begin to give advice to all young men: they can’t run outside, so they begin cool their temper. From January, going February, March, April, dreams for sleep. People begin cry, time no dey, and man die on waking.

In June new events began to happen. On a Sunday, 3 June 1955, a youth named Njie Evele was attacked in broad daylight by vekongi zombie spirits who were hiding in the village Presbyterian chapel. Njombe, the nyongo doctor, was called in (as Njie put it) to ‘doct him’. Njie was in a delirious state. What did he see? asked the

Witchcraft, Economics and Belief 231

doctor. He saw thirty-six vekongi, he said. Who led them? Njie called out ‘Efukani!’ Njombe put it rather charmingly: ‘When he called Efukani, we say who is Efukani as we no get any Efukani’. Njie said this was the secret name of a man called Mbaki. Now Mbaki was a man with a large inguinal hernia which made it difficult for him to get about. But, Njie said, he had actually been driving the nyongo lorry, which was to take him away. Mbaki and three others who were also named were dragged to Njie’s bedside, where they were ordered to let Njie go. The accused old men meekly did as they were ordered by the doctor Njombe. Njie made an excellent recovery. The next day, another young man, Manga ma Vekonje, suffered a similar experience. He gladly dictated to me his account. He had had a primary school education. I was working under Ekona Costains. Then it was payday of 4 Juni [sic] of 1955. When I received my pay, when coming back to Lisoka I saw some little men of three feet high, black with white shirts and white trousers. They were plenty and I couldn’t count them. They said we should go to Mpundu and sing elonge [a kind of part-song].

But Manga managed to get back to his mother’s hut. She gave him a dish of mashed cassava meal. The vekongi were still outside, climbing about in a small mango tree in front of his mother’s door. (It is an interesting sidelight that probably under the influence of the local schoolteacher he called these vekongi ‘eskimos’.) Manga was eating, but the little men ran in and ate up his cassava meal, and took him off to Mpundu and made him dance. They ran away at the approach of Njombe the nyongo doctor. Manga woke up on his bed, naming a suitable Lisoka adult who had been with the ‘eskimos’, and who duly confessed, after being found hiding under the bed in his own hut. Manga recovered with no ill effect except that the vekongi had escaped with his money and a long bar of washing soap. On the following day, 5 June, there was a big meeting in Lisoka to protest against the conditions in the town. Njombe was prominent in the debate. The need for a really good medicine was stressed. ‘So dey make a price 1/-, 1/- general, man, woman, big man, small boy, 1/-. We was collect on that day £17 14s 0d.’ Two days later, the decision was made that only among the Banyang people, 150 miles away, did an appropriate medicine exist. After a further whip round, three delegates left for Banyang country with £30. On the 16th, they returned with the news that the Banyang men would come, but that their price was £100. The town meeting nevertheless agreed; part of the banana fund was subscribed. And so, on 10 August, the Banyang people came with their medicine and their masked dancer, Obasi Njom. The marvels of the succeeding days surpassed expectation. Witches were flushed out in large numbers. The father of the dead Emange of the palm tree was pointed out by the medicine. He confessed that he had killed his son, but swore repentance. He was made to dance a rhythm at the request of the Banyang visitors, which they identified as a nyongo dance, which he clearly knew quite well. Other revelations followed. A by-product was the exposure of an ordinary (liemba) witch: a woman, Namondo, who was accused of having sexual intercourse with other women in witch form. Before the Banyang people left, they trained thirty doctors capable of performing Obasi Njom. They also left a powerful fetish in the bush to protect the village.

232 Witchcraft, Economics and Belief

They left on 7 September. It was like a great load off the village’s mind. By 1956, the news had spread from village to village. Emissaries flowed to Banyang country from a long roll-call of Bakweri settlements. Approximately £2,000 of the new banana money, as I have said, went to Banyang doctors. The newly constituted Obasi Njom lodges continued their anti-nyongo work for a year or more. Through 1956 and 1957, they came into conflict with the Churches and the law. In Wova, eight people were fined £5 a piece for digging up the floor of a wooden chapel, when the Obasi Njom masked figure indicated that the evil witchcraft medicine would be found there. As indeed it was: like all the others it was a latex rubber ball of nail-pairings and witchcraft symbols. Meanwhile, the villages resounded with hammering as the new houses went up. On a visit at this time, I could hardly hear what Njombe the nyongo doctor was saying, for the noise of the tin going on his enormous new house. We must leave this happy scene and reluctantly not pursue more of the details of the exorcising of nyongo from Bakweri land. Suffice it to say that from now on, the new banana income was spent on village ‘betterment’; the zombies retired and, as Njombe said, ‘From there we no get trouble. They no die the same die. Plenty women conceived’. Nyongo was gone; there was Obasi Njom now is the time for all good men to come to the aidinstead.11 DI SCUSSI ON

The general points I want to make are as follows. We clearly have three main elements in this situation: the questions of (1) a change in morale among the Bakweri; (2) a change in economic circumstances; and (3) a change in the supernatural situation. Of these, only one, the economic change, was documentable or measurable by any ‘objective’ criteria. The economic change moved, indeed, from strength to strength, and began to show a downturn only about 1960 when, as we shall see, because of the weakening of the banana market, combined with certain political and economic factors that affected Cameroon particularly, the peasant cooperative movement in bananas began to lose money. The other two changes, the change in morale and the change in the supernatural situation, were, in rather different ways, what development economists would regard firmly as ‘subjective’. Any Bakweri would, of course, have judged the change in the supernatural situation to be more important than either of the others. The change in the economic situation merely provided the means to rectify the supernatural situation. The change in morale followed this. The logic is perfect. There are of course other logics. Here was a people who associated economic ambition with powerful forces of destruction that were supposed to be destroying the youth and fertility of the people. Suddenly, the attempt at communal betterment through cooperative exporting succeeds richly. Was not the previous body of hypotheses disproved? Such might be the neo-Popperian view. As Professor Evans-Pritchard has taught us to see (see Evans-Pritchard, 1937), witchcraft beliefs express numerous situations of conflict and tension. While these conditions exist, and especially while individuals are thought to be able to project their wishes on reality and thus to change it, witchcraft beliefs are largely immune to disproof. The Bakweri hypotheses were not disproved. When the people were shaken by the shocks of fifty years, the pre-existing witchcraft beliefs took on a darker and more

Witchcraft, Economics and Belief 233

morbid form. When the ‘objective’ circumstances changed, they did not abandon the beliefs: they acted as we have seen according to the logic of those beliefs; they exorcized the zombies but the zombies were, to them, no less real: they were now powerless.12 When one reflects on this material, certain coincidences seem to be worth exploring. The experience of the Bakweri in 1954 of the influx of cash from commercial banana-farming must have been of the same kind as the earlier experience in the second half of the nineteenth century. Then, too, there was a striking rise in material wellbeing with the new xanthosoma cocoyam, and the flow of trade goods into an otherwise isolated economy. We can thus recognize the following phases of economics and belief. Phase I II III IV

Date Pre-1850 1850–1894 1894–1954 1954–1961

Economy Pre-xanthosoma; isolated Xanthosoma; trade goods Marginal to plantations Banana boom

Belief ? No nyongo Nyongo Nyongo controlled

Some indication of a possible pattern has emerged since the decline of the banana boom. After 1961, the massive profits of cooperative banana production were reduced. The Bakweri who, during part of Phase IV, had even supplied a Premier to West Cameroon, suffered political setbacks in addition. They had generally been lukewarm to the unification of the two Cameroons. By 1963, the peasant economy was set back to a lower level than that of the end of the 1950s, but not back to the level of Phase III. In that year, a rumour spread from Bakweri villages on the mountain that the elders had ordered that no money should be picked up from the ground, since it was being scattered as a lure to entice men to the waterside. There, ‘Frenchmen’ would use them to work as zombies on a new deep-sea harbour, or use them to appease the water-spirits. For a number of months, it was commonplace to see coins and even low-value notes lying about the streets of the capital. In June of that year, the Chief of Buea pointed out to me with a significant gesture a disturbingly shiny 50 franc piece, lying in the garden of the British Consulate, during a gathering of notables for the celebration of the Queen’s birthday. These interesting events merit more than this brief mention. It is enough here, however, to note the revival of the zombie theme with totally new realien.13 We may then set up tentatively: Phase 5

1961+

Economic setback

Nyongo-like resurgence

The old nyongo spirits still remained exorcised: Bakweri were firm on this. Since 1963, no more striking events have occurred. It now seems to me, however, that Bakweri statements about the novelty of the zombie witch beliefs are to be taken to refer to the incidental, ‘syncretistic’ dress in which they appeared. Taken together with the linguistic evidence, which suggests a longer history for the sómbà witch-pledge than the Bakweri are aware of, it seems likely that the ‘template’ for

234 Witchcraft, Economics and Belief

the new versions of this belief is permanently in existence. Replication occurs, perhaps, when general prosperity (inona low) is replaced by times in which individuals of property begin to stand out and to attract inona. Perhaps this speculative graph14 may contain some truth:

+ Nyongo-like phenomena

? –

Phase I

? Phase II

Phase III

Phase IV

Phase V

Perhaps the ambiguity of Phase V derives in part from the decline of the potlatch controls which may have helped to stabilize the property-versus-inona conflict during the great Phase II period of prosperity. The first (and still I think the only) eyû mortuary potlatches for a generation were performed in 1955 and 1960 during Phase IV, but no attempt has been made to revive ngbaya. There has been some discussion among psychologically minded economists as to whether some societies have innovating ideologies and others not. Such a distinction becomes unreal when one sees that the ‘non-innovating’ Bakweri have at least twice been supremely ‘innovative’: when they accepted the xanthosoma and when they accepted cooperatives.15 I have used the idea of the ‘template’ to express the persistence of certain themes in belief, from which ‘replication’ occurs only when other elements in the social and physical environment combine to permit this. The realien, the circumstantial details of ‘content’ through which the replicated element is expressed, may be different on every occasion – assembled, it may be, by that unconscious process of bricolage to which Lévi-Strauss (1962) has drawn our attention, and by that resolution of opposites upon which he (1964, 1966b) as well as Needham (e.g. 1967) and others have worked to such effect.16 When the Bakweri beliefs were ‘live’ during the critical months of 1955, the creation of new realien could be observed. For example, the novel and rather peculiar form of the zombies as lively little men of multi-coloured appearance is in striking agreement with a syndrome known as ‘Lilliputian hallucinations’ (a term invented by Leroy in 1909). It is found in Europe in combination with many states ranging from schizophrenia to relatively mild conditions – including measles.17 In this connection, it is interesting to note that the doctor Njombe had told me independently that the zombie spirits were twelve feet high. When Manga, an eye-witness, reported them as small, Njombe accepted this as evidence that they could change size. Manga, at least, I take to be a susceptible individual whose Lilliputian hallucinations were, in the prevailing excitement, assimilated to the general terms of Bakweri belief. Field (1960) shows how Lilliputianism is commonly reported by patients in terms of cultural beliefs, such as, for example, that in fairies. Many will be struck by a fellowship in literary atmosphere of the Bakweri beliefs with those of the more remote parts of central Europe. The emergence of

Witchcraft, Economics and Belief 235

nyongo witches from their graves, and their general preying on the young, recall the whole complex of vampire beliefs. And the heroic position of nyongo-fighting doctors like Njombe, who braved the thick of the fray, rescuing victims at great spiritual risk to themselves – enabled perhaps to see the invisible conflict by tying a seed of Aframomum, the so-called ‘Grain of Paradise’, to the brow – all this has the sound of Tolkien’s Lord of the Rings. Such men knew themselves to be fated to succumb at death to the evil. Nyongo doctors too were specially treated at burial, for there was the danger that they themselves would rise from their own graves. Njombe died a few years ago, in his own view, and in that of the people, fatally weakened. He had only one eye, a disability that was taken as a wound. Witchcraft beliefs are all too often ignoble and sordid. On the contrary, the mundane troubles of the small and insignificant Bakweri people gained a certain grandeur from their projection into the spiritual conflict.

16 Social Anthropology and the Historicity of Historical Linguistics

This paper approaches the relations between social anthropology and linguistics from a relatively novel direction. I hope the reason for the choice will emerge. In discussing the relations between two disciplines, we run the risk of not realizing that certain specialist approaches must always remain of little mutual relevance, because of the technical requirements of their disparate subject matters. Others, when stripped of their technicalities, turn out to be genuinely of dual reference. Here we consider certain implications of historical linguistics (born in the same period as social anthropology itself) for some common problems of the two modern disciplines.1 In so doing, we shall set out from a careful consideration of the historicity of the Neogrammarian model of language. I

There was a time when ‘historical linguistics’ (or ‘comparative philology’ as it was long almost exclusively called) was all the systematic linguistics that existed, or at least that was recognized. Since at least the advent of Saussure, that time has ended. More conservative practitioners of the old comparative philology have in some universities hardly recognized that this is so. The result is that many who are concerned with general linguistics still face a certain isolation from philologists, especially from those working with the European languages. Almost a fortiori, then, the kinds of linguists who are likely to be interested in social anthropology or ‘sociolinguistics’ are likely to be least involved in historical linguistics. Now, if this is so, it is a pity, because it is the question of the very ‘historicity’ of historical linguistics that is of especial interest to social anthropologists. Historical or ‘historical’ linguistics raises a problem inherent in the consideration

The Historicity of Historical Linguistics

237

of models based on clearly expressed operations and rules. Such models are in themselves ‘timeless’, or neutral in regard to time – achronic. Ethnographic models, almost above all others, have since Malinowski exhibited this timelessness. The ‘ethnographic present’ is eternal. The timelessness has been, however, perceived in conflicting ways. The rejection of ‘history’ by Malinowski was, in its best expression, the rejection of models falsely purporting to offer ‘historicity’. (Among such were some affected, as it happens, by the ‘historicity’ of historical linguistics: Evans-Pritchard 1965c.) The ‘timelessness’ was not understood at all by many functionalists who seemed to think that their models pertained to a real ‘present’, to which indeed notional dates were often ascribed. This confusion was a result of a misinterpretation of the practice of fieldwork. The social anthropologist had, of course, obtained his material at certain chronological dates. Nevertheless, the structures he imposed upon the data were ‘timeless’, precisely through the paradox that they had implications for notional ‘states’ of the society other than the state at the time of the visit. The model of a segmentary lineage system of the type described by Fortes (1945, 1949) for the Tallensi thus appears to make statements about the past relationships of the ancestors of present Tallensi one to another. The historical plausibility of such past states having actually occurred in such systems is low or nil (Ardener 1959: 116–17). Lévi-Strauss (1963a: 283–9) refers to models of this kind as ‘mechanical’. He had in mind those called ‘Newtonian’ by Wiener (‘mechanical’ from the system of mechanics): ‘… in any such theory the future after a fashion repeats the past. The music of the spheres is a palindrome, and the book of astronomy reads the same backward as forward’ (Wiener 1948, 1961 edn: 31; see also Postscript, below). Functionalist ethnographic models ‘generate’ states of society that may never, almost certainly will never, occur. It was, indeed, the failure of such models to ‘predict’ (that is to make only ‘right’ predictions) that precipitated some of the dissatisfaction that began to characterize post-Malinowskian social anthropology. A common expression of such dissatisfaction lay in arguments that functionalism could not express change (Pocock 1961: 103; Jarvie 1963), which had been much aided by arguments that history was not foreign to the work of social anthropology (Evans-Pritchard 1950, 1961a). The dubbing of functionalist models as synchronic, in opposition to supposed diachronic models that would build in the dimension of time, did not however succeed in producing true diachronic models in social anthropology. Those that were attempted were merely more satisfactorily explicit about the significance of states generated by the model for ‘true’ states of society. They represented a higher (more explicit) state of functionalism. Thus, Leach (1954b) produced an ‘oscillatory’ or ‘cyclical’ model for the states of Kachin society, which is also timeless; it is an orrery working in its own eternity. In this respect, it is not inferior to that devised by Toynbee for material of broader scope. The use of the term diachronic thus states merely an expectation of certain models, which is usually neither clearly expressed nor truly realized. Of a truly diachronic model it is required that the workings of the model shall generate, not possible ‘pasts’ or ‘futures’, but a ‘real’ past and a ‘real’ future; the issue may be avoided somewhat by an acceptance of ‘pasts’ and ‘futures’ that shall be ‘probable’.

238 The Historicity of Historical Linguistics

The kind of model that might generate time has been talked about for many years. In social anthropology, Lévi-Strauss opposed ‘statistical’ models to ‘mechanical’ models (1963a: 283–9). In this, he again followed Wiener. The terminology has caused misunderstanding; it is not about models based on quantified material, but models of the type devised in ‘statistical’ mechanics: probabilistic, yes, but concerned with that movement towards more and more randomization known as entropy. Leach (1961b) shows that this idea of time (in itself, our only sure appreciation of its passage) is usually rendered symbolically by society in quite other ways that are essentially cyclical or repetitive, and, thus, mechanical. In this respect, then, the search for true diachronic models tends to be abortive for others as well as social anthropologists.2 Nevertheless, through the problem of historicity, social anthropologists stemming from functionalist backgrounds have come upon similar ground to that with which linguists have become familiar through the generative grammarians after Chomsky (1957). General linguists, on the other hand, have frequently turned their backs on historicity, because it has for them been over-examined. As a result, many non-linguists believe that it has been already solved for language. For linguistics, the terminology synchronic and diachronic was firmly developed by Saussure (1916). In thus labelling two approaches, he was misled by that very historicism which he set out to reject. He was concerned with two modes of analysis which in themselves were no more concerned with time than were those of social anthropologists. That synchronic models are timeless (that they have nothing to do with a chronologically ‘present’ or momentary state) has become self-evident since Saussure. At least, most people on reflection will concede that the progress from Saussure’s ‘synchronic’ linguistics, to ‘structural’ linguistics, and to ‘transformational’ or ‘generative’ linguistics, has been a real progression away from a special concern with the ‘present state’ of a language. The ‘timelessness’ of the diachronic linguistic model is by no means as self-evident and needs some demonstration. The diachronic linguistics of Saussure (1916) was the comparative philology of Indo-European in which he was bred, and which he helped significantly to establish. The system in its full maturity, as bequeathed by Brugmann and the Neogrammarians,3 provides a model which in principle reduces the multiplicity of documented forms in the set of languages it is concerned with to a dictionary of ‘starred’ or hypothetical forms, plus a set of rules whereby the original forms (‘reflexes’) may be built up again from them. The form of the model is totally ‘generative’, and requires no historical interpretation for its operation. For example: *ul.quˆ os generates Sanskrit vr.kah., Lithuanian vilkas, Gothic wulfs, and Old Bulgarianˆ vluˇkuˇ, through rules (for example phonemic ones) of the type: *l. → Sanskrit r. Lithuanian ir, il Gothic ur, ul Old Bulgarian luˇ The rule books for what *ul.quˆ os will generate are much more complex than this ˆ simple example at a phonemic level would indicate. Any starred form in IndoEuropean generates intermediate phonemic sequences which themselves generate

The Historicity of Historical Linguistics

239

further sequences, until ultimately *ul.quˆ os generates, for example, English wolf, ˆ Russian volk. Even linguists are sometimes puzzled by ‘generative’ models; social anthropologists tend to be even more bemused because of the specialist guise in which they have been discussed by linguists. All models are ‘generative’: the trouble is that many so-called models are not models of formal systems. The generative terminology is that of mathematics; for example: In any formal system one encounters the following constituents: (i) Certain constant symbols… (ii) An alphabet of variables… (iii) A set of rules, which state which strings of variables and constants constitute sensible statements: the so-called well-formed formulae… (iv) A set of well-formed formulae, known as axioms, from which others known as theorems are to be deduced… (v) A set of rules of inference which enables the deduction of theorems from other theorems or from axioms… (Kilmister 1967: 70)

For a formal system to provide a model for a corpus of given data, it must generate all and only the forms in the corpus: these are the well-formed items. Neogrammarian linguistics provided a model of precisely that kind. The generativeness is protected by: (a) The rule of analogy: e.g. Middle English wa + c [c → a non-velar consonant] → wɔ [e.g. /wɔz/ ‘was’, /swɔn/ ‘swan’]. The rule does not generate swam. The ‘exception’ is generated by the analogy rule in some such way as: ‘sit’:‘sat’ [sæt] → ‘swim’:‘swam’ [swæm] (Ross 1958: 34).4 (b) The rule of loan-effect: e.g. I-E *guˆ → Latin w. The rule does not generate Latin bo¯s (‘cow’). It does generate Oscan bo¯s, which is thus defined as a ‘loan-word’ in Latin. (c) An exclusion rule: items that the model does not generate are outside the model (usually expressed: ‘phonetic laws admit no exception’). This necessary tautology is of importance to us only because of the question of historicity. The transformation of the idea of etymology by Neogrammarian linguistics made it possible to say that a word had ‘no’ etymology: ‘Not every word has an etymology; thus MnE girl and much of the Hungarian vocabulary are totally without one’ (Ross 1958: 39). Such statements should be noted as clear confirmation that the model is not one of ‘history’. The definition of the ‘corpus’ of Neogrammarian linguistics is thus that corpus that the model has generated or will generate. The excellence of the Neogrammarian model is revealed precisely because of its approximation to a total formal system. The model as such is timeless. Its status is already that to which generative grammar merely aspires. To claim that *ul.quˆ os existed at a chronologically earlier time ˆ would be as if to claim that the elements in ‘deep structure’ existed in the past before the presented elements in ‘surface structure’. The question may not always be

240 The Historicity of Historical Linguistics

irrelevant (see now King 1969), but generative grammar is not thereby ‘historical’. The starred forms ‘exist’ in the known corpus: Old Bulgarian, Sanskrit, Modern English, Russian or the like. *ul.quˆ os ‘is’ vr.´kah., wulfs, … plus certain rules. ˆ Then wherein lies the apparent historicity of the model? (a) The corpus to be generated is already ordered a priori by chronologically marked elements. (b) In addition, all ‘chronologically’ marked sequences are required to be generated sequentially by the model. The model does not, however, generate only these chronologically marked sequential forms. Other sequences are generated as part of the calculus: they are the source of ‘surplus historicity’ in that they may perhaps have happened; they are a kind of guess, which may turn out to be well or ill substantiated according to other criteria outside the model (e.g. certain notions about universals).5 This may seem tediously self-evident, but that it is not is demonstrated by the common belief that the model generated (‘predicted’) forms not previously attested. In particular, that it ‘predicted’ Hittite. In fact, that would be an exaggeration of its achievement. A part only of the Hittite vocabulary and morphology was defined by the model as Indo-European, and the shears of the ‘rule of analogy’ were liberally applied. The situation was confused by the supposed confirmation by Hittite of the hypothetical laryngeals, which precisely did not belong in the true Neogrammarian model. The model for any post-Hittite ‘Indo-European’ is in fact a revised model, and the ‘historicity’ of such revisions has been seriously questioned by comparativists themselves.6 The preceding argument is concerned only with the most rigorous interpretation of historicity in Neogrammarian linguistics, and thus with the confusion in the model itself. The looser historical expectations based on the model have always been precarious. The discovery of Tokharian, for example, brought ideas of the dialectal distribution in Indo-European into flux (Meillet 1950). Forms like *ul.quˆos do have a double life; they are formulae expressing rules of generation in ˆ model. They are also phonological test-words for a theory of history (‘derithe vation’). The confusion has lasted well into our own days. Thus, the well-formed model for a Neogrammarian etymology outlined by Ross (in terms of symbolic logic) is perfectly satisfactory in its aims save in this particular: he interprets the two axioms that he permits himself as statements of an existential past (1958: 28–42).7 II

Not only linguists must be concerned with this problem, as it is intuitively felt that the Neogrammarian model generates more ‘history’ than it puts in. Neogrammarian linguistics is indeed the model of all models for a diachrony of social phenomena. Other disciplines, including history, archaeology and forms of anthropology, have even used the cart-before-the-horse term ‘reconstruction’ in the firm belief that at least linguistic reconstruction has taken place already. Once more we may start from Saussure. In his chess analogy, he was able to illustrate synchrony in language by the state of the board at successive moves of the game, and diachrony by the rules of movement of the different pieces from one state to another. The notion of value was used to denote the ‘synchronic’ relationship of any one piece at any one state of the board to all other pieces on the board at that state. In so far

The Historicity of Historical Linguistics

241

as a synchronic value is an expression of a diachronic rule, how far can a study of synchronic values enable the diachronic rules to be deduced? The recovery of the rules of chess from individual states of the board is in part possible. Certain rules are not deducible until the last state of the board (e.g. the rule that mate closes the game). Past existential states of the board cannot be deduced regularly in this way. Saussure was content to point out that the rules of synchrony and those of diachrony belong in two separate fields: that one set was not reducible in practice to the other. But recently Vendler in another context has said: Suppose that while watching a game of chess I see two Pawns of the same colour standing in the same column. Then I say: ‘One of them must have taken an opposing piece in a previous move’. Is it sufficient to say that in all chess games we ever witnessed this correlation held? No, given the rules of the game, two contingent historical states of affairs appear to be necessarily connected. Moreover, obviously any ‘game’, or, in a larger context, almost any rule-governed activity, will be the source of such propositions. And this domain may range so far as to include mathematics or the rules governing the synthesis of the manifold of experience. (Vendler 1967: 18)

Some of the ‘surplus’ historicity of sequences generated by the Neogrammarian model derives from rules of the two-pawns-in-a-column type. For example, the Germanic sequences: 1. p t k → f ƿ χ 2. b d g → p t k Transformation (2) will have occurred after transformation (1); otherwise all p t k from b d g would fall together with ‘original’ p t k and generate only f ƿ χ. Many traditional comparativists honestly believe that if all chronological markers were stripped from their corpus of items, and thus were not built into the model, they would still be able to order the material into sequences that would be ‘chronological’. This is partly a misunderstanding. It is not sufficient to ensure that the analyst will be presented with merely undated material from Middle High German, Gothic, Old English and the rest. The material is chronologically marked by more than the matter of carrying a literal date, the least of all requirements. The markers and the material are indissoluble. As will be seen in the next section, the real absence of chronological markers is no trivial empirical problem. The case of two-pawns-in-a-column sequences is more important, for critics of the position expressed here retire to a level that implicitly states that these logical sequences will always have chronological implications. In fact, the matter is more like this: sequences that were chronologically marked reveal logical sequences on analysis, which, in so far as the marked chronology was ‘correct’, have other chronological implications. To take yet another analogy: in the so-called ‘painting by numbers’, 1 = red, 2 = blue and the like, on a set of outlines drawn on a plain sheet. The painted configurations revealed on the canvas when the numerical rules are obeyed are, of course, autonomously open to interpretation by a ‘reality codebook’. ‘It is a weasel’, or even ‘A weasel is fighting a dog’. But a revision of the numerical markers is quite sufficient to dissolve all these other analyses, which derive their cogency or force from more general rules of interpretation, which are

242 The Historicity of Historical Linguistics

quite separate from the rules defining the system. So it is with this ‘surplus historicity’ of the Neogrammarian model: it is a revelation of the full implications of chronologically marking one’s material. This is exactly why it is a good model: like all such, its apparent novelties are logical implications. The Neogrammarian model of language is always interpreted as if its rules really were those of a game that had been actually played. A sequence in the calculus is always interpreted as a necessarily chronological sequence in the languages of the corpus. Its achievement is tautologous, as we have seen: (1) the corpus includes prior established chronological sequences; (2) it defines as chronological all new sequences in the calculus; (3) the rules exclude from the corpus all of language to which the rules do not apply (as if there were spare pieces on a chessboard to whose operation the rules of chess did not apply). Such are the ‘shears of Brugmann’. III

The application of ‘comparative philology’ beyond the Indo-European ‘family’ brought with it the establishment of other similarly defined corpora of linguistic data. Certain of these contained chronologically marked material of as much variety as did the Indo-European. Some, however, had very little or no chronological marking. We may take, for example, the Bantu family. Here, starred forms generate regularly all and only forms designated Bantu (Guthrie 1948). The model is, however, jejune compared with Neogrammarian Indo-European. There is no regular ordering of the sets generated; there are no intermediate forms equivalent to asterisked ‘Celtic’ or ‘Germanic’, or even to ‘Middle English’ or the like. Possible orderings are from time to time suggested, but no rules exist compelling their acceptance. The starred Bantu forms generate at the model level (and that is the only level at which ‘Bantu’ exists) all the extant reflexes, without intermediate grouping. The absence of chronological markers is clearly the reason for this. The Bantu model fails historically in much the same way that the Indo-European model fails with the so-called Indo-European dialects (Italo-Celtic and the like) and the centum/satem division, where the entry of Hittite and Tokharian found the Indo-European model precisely at its weakest (Meillet 1950). The historicity that was built into Neogrammarian models was, as has been argued, the amount that they truly yielded. The rich documentation that obscured this for IndoEuropean did not exist for Bantu. Thus, the discovery that diachronic Bantu linguistics did not yield history was easier to make. A firm hold on the logical bases of Neogrammarian linguistics would have helped Bantuists to avoid the historical controversies that surround their subject. Their most austere representative (Guthrie 1948, 1953) has attempted to employ rigorously the Brugmannian shears to protect the generativeness of the Bantu model. Yet their operation has not been accepted as readily as in Indo-European studies, especially that of the rule of exclusion: that phonetic laws admit of no exception. The position of Guthrie is undoubtedly formally correct. Its effect is to define as non-Bantu a range of languages which others (like Greenberg) wish to call Bantu (e.g. Tiv). For the social anthropologist in the area in which these lan-

The Historicity of Historical Linguistics

243

guages occur, the Bantu line (that historico-linguistic figment) swings back and forth bewilderingly over vast stretches of country, fortunately to no great danger to the inhabitants (Richardson 1956–57; Greenberg 1963b; Crabb 1965; Ardener 1967: 294–5). The Brugmannian model for Indo-European shows similar ambiguities, in the Anatolian zone. Is Hittite any closer to ‘reconstructed’ Indo-European than Tiv is to ‘reconstructed’ Bantu? Would Hittite not have been called ‘semi-Indo-European’ by the adherents of ‘semi-Bantu’? The Indo-Hittite hypothesis (Sturtevant 1942) echoes several hypotheses in African linguistics, but there is no firm party line to put them finally down. The Greenbergian hypotheses (1963) for the interrelations of the African languages are bold, but (or: and therefore) they are not rigorously Neogrammarian.8 They resemble the activities of the adherents of ‘internal reconstruction’, or of students of relations between Basque and Caucasian, or of those attempting to establish a ‘Mediterranean’ family. All these interesting preoccupations are proscribed by traditional historical linguistics even today. Having declared for freedom, however, the adherents of the new formulations cannot guarantee at the same time the old Brugmannian rigour. Murdock (1957) has mistakenly used Greenberg’s scheme, as if it contained even the historicity of the Indo-European, in order to build a ‘conjectural history’ of Africa, which has gone on to influence historians. On the other hand, Guthrie (1962) has attempted by an injection of statistics to raise the generative powers of his essentially Neogrammarian model. In so doing, he has departed from that model completely. His model is in effect a new one: based upon the geographical distribution of the percentages of known reflexes of starred Bantu reported in the known vocabularies of a fixed number of Bantu languages. The ordering of such percentages produces prima facie geographical possibilities. That this model of Bantu is no longer self-defining in a Brugmannian way is demonstrated by Guthrie’s inclusion of a ‘non-Bantu’ controversial language (Tiv). IV

The statistics of the new Bantuists have not yet received the critical execration that was heaped upon the glottochronologists. The latter were accused essentially of publicly practising vices that are the private predilections of all naïve historical linguists. It is unnecessary to refer to the detailed literature in view of Hymes’s invaluable survey (1960), which covers in effect the Blüteperiod of the theory. In brief, Swadesh and others in various publications (e.g. Swadesh 1950; Lees 1953) attempted to derive a measure of the rate of change of vocabulary over time. To do this, Swadesh took a control group of languages with long-documented histories (in the order of millennia). He devised two word-lists (of 100 words and 200 words) which set out to represent the area of vocabulary what would be most resistant to change (the ‘core vocabulary’). The percentage of each list retained per millennium was purported to be constant for the control group. Much refinement of assumptions and method was attempted, and for a time the glottochronological

244 The Historicity of Historical Linguistics

method was cited at length in American textbooks of linguistics (Gleason 1955a, 1955b; Hockett 1958). The chief value of the method was hoped to be the provision of chronological dates, rather like carbon dates, for the separation of languages, apparently lexically related but without documentation. The method was another reaction, in fact, to the unsatisfactory historicity of the Neogrammarian model, when applied to exotic (here originally Amerindian) languages.9 As is well known, historical linguists did not react with much satisfaction to this attempt to quantify the rate of linguistic change. The scholarship and skill with which the theory was demolished evoked much admiration (e.g. Bergsland and Vogt 1962). The question whether glottochronology survives or not may be debated, but it is true to say that it does not exist as the source for a simple rule for the quick determination of chronological dates of linguistic separation, except with the most extreme qualifications. It should, however, be noted that the assumptions about comparable lexical units (lexemes of cognate form and meaning) that enabled a glottochronological statement to be made, and about a ‘core’ vocabulary, were merely crudely explicit versions of those that lay behind the ‘genetic’ view of linguistic relationship. Thus, what is it that makes it possible for a historical linguist to say that English ‘continues’ Old English and not Old French? This question was solved by the Neogrammarians in characteristic manner: a second row of protective conventions lay beyond the protective rules of analogy, loan-effect and exclusion. These were: (1) ‘there are no mixed languages’; (2) ‘there are no substratum effects’. On reflection, it will be apparent that convention (1) preserves the generative model from generating contradictions in the ‘same’ language; it is merely a statement that in such a case recourse must be made to the rule of loan-effect. Convention (2) similarly excludes the possibility of a language demonstrating change, except through the principle of phonetic laws admitting of no exception. These I have termed ‘conventions’ because, although they are mere corollaries of existing rules, they were not readily perceived as such by many historical linguists who brought them from time to time into debate. The Neogrammarian view is that in all cases of doubt, the correct relationship will be determinable by examination. At a conference in Oxford as recently as 1967, one eminent visitor said that a ‘mixed’ language could occur only in the case (stated rightly to be absurd) of one language being exactly composed of 50 per cent of one language and 50 per cent of another. The naïveté of this view quite equals the naïveté of the more extreme exponents of the idea of mixed languages, against which it is directed. The problem has been recognized, however, as a real one: for example, the ambiguity of classification for a long time of Albanian and of Maltese. In the end, the rules about loan and analogy do not take into account even all examples of these two phenomena. As Meillet said of Sorabian and German in Lusatia: ‘le parler sorabe n’est souvent, avec de formes slaves, qu’un calque de l’allemand’ [‘spoken Sorabian is often, using Slavonic forms, a mere calque upon spoken German’] (1933: 168). In the case of some modern spoken Welsh, it can be said to be, at the level of phrase and sentence, a calque upon English: a one-to-one code. In the terms of generative grammar, the ‘deep structure’ is shared in part with English. Here the findings of the modern descendant of ‘synchronic linguistics’ clash with those of the traditional historical linguistics. For

The Historicity of Historical Linguistics

245

what is more ‘historical’ about Modern Welsh? Its English connections? Or those with Irish, Breton and epigraphic Gaulish? The idea of ‘core’ vocabulary is also not far from the minds of historical linguists. There is in any language classified by an Indo-Europeanist what amounts to a minimum ‘register’ generated by the model. An essential part of any such register might be (for example) numerals, some kin terms, some bodily activities, but probably not (for example) house building ‘registers’, or those pertaining to trade. It is, in effect, assumed that all registers but one (or one small set) can be borrowed without ascribing the language to a different ‘genetic’ family. The intuitive basis of the term ‘genetic’ is that this smallest register will equal that learnt at the mother’s knee. The lesson for historical linguistics from ‘exotic’ languages, then, was not glottochronology, but a more serious one about the bases of the initial assumptions of any ‘genetic’ historical model of language. Thus, the problem of Hittite is hardly solved by its triumphant attribution to Indo-European, when the very adverbial suffix used to mean ‘in the language of ’ (-ili) is not generated by the model. Nor can the problem of why such an ‘old’ language looks so much less ‘conservative’ than the younger Sanskrit or Greek (Meillet 1937, 1964 edn: 56; Hauschild 1964: 36) be solved without some consideration of processes, of which creolization is an extreme case, that ‘speed up’ linguistic change.10 V

So far we have discussed only the model of ‘Neogrammarian historical linguistics’ rather than the methods by which the model was built up. This has been necessary for reasons which have been partly made obvious. The system of historical linguistics is received as a system, and as one that is historical. For many of its traditional adherents, the bases in method have become subsumed in the aims of the system. The method is usually referred to as one of ‘reconstruction’, although it is rather a method of ‘construction’. It is a ‘comparative’ method, indeed the original method to be so called, and one that strongly influenced social anthropology in the nineteenth century (Evans-Pritchard 1965c: 15; Henson 1971). Since those days, the method of comparison has become commonplace, but it has become clear that it does not in itself produce historical generalization. That it should briefly have been thought to do so results from the historical concerns of comparative philologists and Darwinian biologists. The method of ‘reconstruction’ in historical linguistics is in principle not very different from the same procedures used in other branches of linguistics. The same kind of approach abstracts ‘phonemes’ from the corpus of speech sounds, reducing the latter to series of simpler constructs which generate the original corpus by means of rules. In the heyday of their ‘discovery’, phonemes, when constructed, caused the same existential problem as did *ul.quˆ os and the like. If they ˆ they been? A historical could be constructed, where were they? Or, where had explanation was not seized upon, as might have occurred in another age, although it lurks behind such ideas as that of the ‘archiphoneme’. It is, however, interesting that the development of laryngeal theory in Neogrammarian Indo-European was in effect a conventional internal phonemic analysis applied to the ‘reconstructed’

246 The Historicity of Historical Linguistics

Indo-European vowel system. The historical interpretation of the proposed laryngeal phonemes produced a debate such that quite recent introductory manuals of Indo-European philology still do not deal with them.11 The arbiter for the ‘well-formedness’ of constructed phonemes in some approaches was the ‘native speaker’. The question of the relationship of an analyst’s model of human activity to the interpretation made by the people involved is an old one in social anthropology. That historical linguistics implies a similar ‘native speaker’ problem has been obscured by the circumstance that comparative philology grew up in opposition exactly to ‘native speakers’ in the field of etymology. The native speaker was defined out, and in much of historical linguistics he still is. The very success of this operation is of some interest to social anthropology because of its undoubted dependence on him. The transformation of etymology by Neogrammarian linguistics can be illustrated by these two sequences (Ross 1958: 42): 1) I-E *guˆ o¯us → Common Germanic *kwoˉz → Old English cu¯ → Modern ˆ English ‘cow’. u 2) I-E *g ˆ o¯us → Oscan boˉs → [by borrowing] Latin boˉs, accusative boˉ vem ˆ accusative stem] Anglo-Norman bœ – f → [by borrowing] Middle → [by the English be¯f → Modern English ‘beef ’. These are etymological ‘sames’. Yet in English, as is well known, ‘cow’ and ‘beef ’ are not the same. No folk-etymology of the two words would turn up a suggestion more apparently improbable at first sight. For Leach (1964), the two words clothe categories of great importance in the English symbolic system. These categories have, however, no place in any etymology deriving from Brugmann. Vendryes (1933), in his discussion of the tasks of ‘la linguistique statique’, understood the problem as one of those deriving from the ‘synchronic section’ of Saussure’s langue. As he presciently put it: C’est au point de vue de l’homme qui parle que le linguiste staticien doit se placer. La seule étymologie qui compte pour lui ne peut être que l’étymologie dite populaire… [The ‘synchronic’ linguist must see from the perspective of the native speaker. The only etymology which counts for him, is that which we call ‘folk’ etymology] (1933: 176)

In the absence of chronological markers, however, the clear distinction between etymology and folk-etymology cannot be so crisply maintained. The field of folk-etymology is close to the zone of dead metaphor. For example, Chinese: hwoˇ (‘fire’), che¯ (‘cart’), yields hwoˇ che¯ (‘train’), but ‘electric train’ is dyànlì-hwoˇche¯ (dyànlì, ‘electric power’), not, for example, dyànlì-che¯ (Colby 1966: 10; Hockett 1954: 111). Similarly, ‘steamroller’ in English was for many speakers incorporated in ‘diesel steamroller’. The possibility of recovery of a dead metaphor is recognized by native speakers. Often, they will say that the ‘real’ meaning of a word is ‘x’ or ‘y’, referring to the process of pointing out such a dead metaphor. A serious discrepancy between Neogrammarian and folk-etymology soon became marked, however. For example, ‘asphalt’ is a material which among English builders is commonly pronounced ‘ashfelt’. This situation produces two etymologies:

The Historicity of Historical Linguistics

(1) asphaltos (Greek) a ‘negative’

sphallo ‘cause to slip’

‘non-slip’

(2)

247

ashfelt (English)

ash (of the set containing e.g. ‘potash’)

felt (of the set containing e.g. ‘roof felt’)

‘mineral waterproof material’

‘Ashfelt’ is to the untutored user a form of the same status as ‘steamroller’. While the latter is a dead metaphor for the drivers of ‘diesel steamrollers’, ‘ashfelt’ is a dead metaphor that never existed. Indeed, it might better be considered a ‘living’ metaphor calqued on ‘asphalt’! The essential neutrality of the Neogrammarian model in relation to history, in the sense of the generation of ‘real’ sequences, is clear when we consider that if ‘ashfelt’ were declared to be ‘English’ by some consensus of usage, a Neogrammarian etymology of the elements ‘ash’ and ‘felt’ could be supplied, and might be thought necessary. The true historicity of the sequence asphalt → ashfelt is a separate matter: it is a sequence given by a chronological record. It is the independent record, not (of course) the etymology, that validates it.12 The same problem occurs for the social anthropologist in the study of nonlinguistic symbolism. He is offered ‘folk-etymologies’ by the native speaker; he attempts ‘etymologies’ of his own. Try as he may, however, the historical kind of etymology cannot be surely constructed without the historical records. Thus, in the Oxford degree ceremony there are striking passages in which the two proctors march up and down, and then sit down again. The folk-etymological faculty is continually exercised by spectators, who produce interpretations such as ‘they are pretending to think’, ‘they are pretending to open the doors’.13 We are told by the historians of the ceremony that the proctors are providing an opportunity for any member of Congregation who wishes to object to the presentation of the degree to do so by plucking a proctor’s gown. The operation is linked by the historians with taking the vote in a normal assembly of Congregation (Wells 1906). Is it possible to ‘reconstruct’ the etymology from the present ceremony? By hindsight, no doubt certain approaches are possible. One might note that an opportunity for dissent also occurs at marriage rites. In the two rites there is, then, a symbolic space which may be filled by walking up and down, in one case, and by a formal utterance by the celebrant, in the other. The corpus of all rites in the society might be analysed and an alphabet of the possible symbolic spaces might be prepared. The mode whereby such a symbolic space is realized in any given rite (‘walking up and down’, ‘an utterance’) would be a separate level of study, drawing upon further comparative material for its elucidation. Once we have got so far in trying to replicate a ‘historical’ etymology of a feature of a certain rite, we discover that (as the previous argument would lead us to expect) we are, in the absence of the given

248 The Historicity of Historical Linguistics

historical documentation, replicating post-Saussurean linguistic methods: the basic method of structural linguistics, of historical linguistics and of social anthropology turns out to be the same. Once more: there is no historicity in the method as such. (For similar conclusions from a different standpoint, see Ellis 1966.) There is a further problem in the attempt at an etymology of symbolism which is inherent in the method of reconstruction. If the symbolic spaces to be compared have identical ‘realizations’, little more can be said. If in the wedding service the clergyman had walked up and down, made no admonitory utterance, we should be no further forward. In such a case, the ‘native speaker’ may be our only guide in new directions. The societies that social anthropologists study frequently have formal or informal bodies of lore-guardians who characteristically operate at the level linguists would refer to as folk-etymology: collecting such ‘etymologies’, creating new ones. Their function may be seen as one of continually blocking up the fortuitous gaps in the symbolic sets. The remains of the old systems may be seen propping up bits of the new. These bricoleurs (Lévi-Strauss 1966a: 16–17) turn out indeed to be village Brugmanns with a clear grasp of the principle that a model should generate only what it has been designed to generate.14 It is no surprise that the narrow ‘historicity’ aimed at in the method of reconstruction falls into relative insignificance in front of problems of a social anthropological type. History is continually re-created in the wake of society in such a way that no one model of historicity can do more than hint at a possible aspect of a possible past. Say that ‘ashfelt’ and ‘silverside’ (a certain cut of beef) both exist in a dialect of English, and that folk interpretations ‘ashfelt’ = ‘ash’ + ‘felt’, and ‘silverside’ = ‘silver’ + ‘side’ are made. The latter is a ‘real’ dead metaphor; the former is created to look like a dead metaphor: the distinction has here been rescued from the very teeth of oblivion. The ability of human beings to rearrange Saussure’s chessboard to look as if a game has been played so far which did not occur makes the quest for narrow ‘historicity’ almost meaningless. The vexed question of whether Adam had a navel led Edmund Gosse’s father to make the answer, in effect, that the evidences of man’s past were created with him.15 VI

The purpose of this paper has been to show that a consideration of Neogrammarian historical linguistics yields insights for social anthropologists which are helpful both in approaching their own problems and in appreciating the directions in which recent developments in ‘non-historical’ linguistics have moved. The grandeur of the Neogrammarian model for historical linguistics literally left nothing more to be said. This grandeur lay in its perfect generativeness. It did not, however, generate history. The Chomskyan generative model is in comparison incomplete, but its aims are more ambitious. Social anthropologists have had models which on their small scale were ‘generative’. They are likely to waste their time in trying to find analogies with particular details of the Chomskyan system. They are likely to repeat the failure that attended Lévi-Strauss’s early effort to find social ‘phonemes’: a confusion of the peculiar data of linguistics with the broad ‘semiological’ principles (Lévi-Strauss 1963a: Chapter II).

The Historicity of Historical Linguistics

249

The words of Hjelmslev (1963: 14–19), originally written in 1943, may now be translated into the terms of my argument: He speaks of two factors in his use of the word theory: 1) ‘A theory … is in itself independent of any experience. In itself, it says nothing at all about the possibility of its application and relation to empirical data. It includes no existence postulate…’ (Thus we define the model of a well-formed system.) 2) ‘A theory introduces certain premises concerning which the theoretician knows from preceding experience that they fulfil the conditions for application to certain empirical data. These premises are of the greatest possible generality and may therefore be able to satisfy the conditions for application to a large number of empirical data.’ (These are the ‘reality conventions’ for the interpretation of the model.) Hjelmslev goes on: ‘The first of these factors we shall call the arbitrariness of a theory…’ (this is the consistency of the model, and thus its generativeness), ‘the second we shall call its appropriateness’ (this is the fit of the model). Finally, for what is to be generated (the corpus), his wording is of striking familiarity: For example, we require of linguistic theory that it enables us to describe self-consistently and exhaustively not only a Danish text, but also all other given Danish texts, and not only all given, but also all conceivable or possible Danish texts, including texts that will not exist until tomorrow or later, so long as they are texts of the same kind, i.e. texts of the same premised nature as those heretofore considered. (1963: 16)

The generative views of Hjelmslev, of the Neogrammarians and of Chomsky are all expressed in a stubbornly linguistic guise, which obscures their general relevance, and in a stubbornly individual manner, which disguises from the untutored the essential similarity of their approaches. Of all these, Hjelmslev has been the least understood. He had a message for synchronic and diachronic linguistics when few ears were ready to hear, even had he not published in wartime, and in Danish (he himself says somewhere in Essais linguistiques, 1959, ‘to write in Danish is to write in water’). It is no surprise that he was regarded by Bally as the most Saussurean of Saussure’s successors (Hjelmslev 1959: 31). The great advance of the new linguistics for non-linguists is in making explicit the idea of ‘transformation’. In this direction, however, Lévi-Strauss has already independently made the necessary first steps for social anthropology (1962b, 1964, 1966b, 1968). These more ambitious models generate sequences some of which no doubt will be of historical significance. But the problem of historicity is of course ultimately the problem of history itself. In discussion with working historians, one sometimes notes in them a reluctance to concede that historiography offers a choice of pasts – even though their own theoreticians have long said no less. Each historian’s model generates the data it was made to generate, with the operation of an exclusion law: the guillotine of relevance. Historians are reluctant because ‘propaganda’, ‘legend’ and ‘myth’ share these procedures, a fact that to social anthropologists is of especial interest (Leach 1969: 25–30 and notes). The

250 The Historicity of Historical Linguistics

lesson of historical linguistics is that good models can be made, but that there can be no winning outright in the attempt to generate ‘the past’. P OSTSCR I P T

Any social anthropologist setting out under the watchful walls of both historical linguists and the transformational generative grammarians will be fortunate to escape back into his own trenches from the hail of fire. It has been particularly encouraging, therefore, that although this paper has occasioned debate both at the conference in Sussex at which it was first delivered (it was written in this form in December 1968), and before historical linguists and some transformationalists at Oxford in 1970, its main conclusions have remained standing. Consequently, I have left the text substantially as it was, save for the paragraphs on ‘surplus historicity’ in section II, which are essentially an expansion of the argument already summarized by my quotation from Vendler (1967). Some further remarks are required, however, in order to make clear what I am not saying, and also to explain my use of ‘historicity’. Furthermore, we now have R. D. King’s book on Historical Linguistics and Generative Grammar (1969), received only shortly before this piece was finalised. I should like to add a few words upon it here also. Finally, not all the anthropologists were in agreement with me about the relevance of any of this to them. I shall take these three groups of points in order. It should first of all be self-evident that this paper is not ‘anti-historical’ and not ‘anti-historical-linguistics’. So much so, indeed, that some have justly seen it as strongly ‘pro-Neogrammarian’. The paper is about the nature of formal systems, a subject that at least the writings of Lévi-Strauss, Leach and Needham (e.g. 1962) have made well known to social anthropologists. Although the title refers to ‘historical linguistics’ (in part for euphony and brevity), it should be clear that the model I discuss is that of Neogrammarian historical linguistics in its most ‘German’ heyday (which for many is not over). This model was almost totally generative. Suggestions that one should look towards other more ‘liberal’ exponents of the general approach miss the point: the latter have ceased to work as if they had a totally generative model. Rather, there are many partial models but none fully articulated. Lehmann and Malkiel (1968), in an important compilation, and especially Weinreich, Labov and Herzog (1968) therein, now present a rich programme for historical linguistics, which builds on Saussure and the Neogrammarians ‘without leading to positions which disregard the achievements of past theory’ (p. 20). Nevertheless, one can say that where historical linguistics is still understood to be equivalent to comparative philology, as in some quarters of my own university, the Neogrammarian model remains substantially unamended. In stressing the generativeness of the Neogrammarian model in its most rigorous form, I have also said something about generativeness itself for non-mathematical people in the humanities tradition who, in this country at least, think there is a holy mystery in it. Furthermore, it should also be clear that generativeness is not a sign of modernity. Awareness of its nature possibly is. For literary people, if a model is rigorous it is ‘too rigid’; they believe all models are of reality (which is not rigid), instead of being models on reality, which must be rigorous. The principle

The Historicity of Historical Linguistics

251

must be: ‘let a hundred flowers blossom’ – or ‘let a hundred properly articulated formal systems contend’. When new masses of data arrive, tinkering with the old models (if they are any good) simply destroys their formal properties. A new model or set of models is required. While these are being created, there is rarely any doubt of whether they work: usually they chew up facts for a generation before their implications are exhausted. The emphasis on the neutrality of model-building principles means that each model must be used with a set of ‘reality’ conventions. The ‘historicity’ with which Neogrammarian philology is concerned derives from the ‘book of conventions’ for the interpretation of the model. For this reason, ‘extreme’ statements such as those of Ross have real integrity because of their appreciation of this. Historical linguistics, in so far as it is not Neogrammarian, is not a gloss on Neogrammarian history: it is based on, or hopes for, different models. Yet another comment must be made: the Neogrammarian model was not the only apprehension of language the Neogrammarians, as people, had. But an apprehension is less easily taught (see Osthoff and Brugmann 1878, now easily accessible in Lehmann 1967: 197–209). This brings me to a final misunderstanding of my paper from the ‘traditional’ side. What is true of one model of history is true of all: the ‘surplus historicity’ it generates, over and above the chronological markers that are fed in, derives from the possibility of establishing necessary logical sequences which will be expressible by the reality conventions for the model as if they had actually happened in time. We act as if we generate reality, whereas not only is the ‘truth’ of the whole structure dependent on the values given to the essential minimum of chronologically marked documentation, but (and this is where social anthropologists can place the problem in broader perspective) the evidence is not (as Kingsley thought of geology) ‘written by God upon the rocks’. I cannot express this part of my case any more clearly than in Section V above. For people in the old humanities, the lesson of modern social anthropology is that their ‘science’ is no worse than that of the scientists; it is merely more difficult. To turn now to the generative grammarians. My audiences have tended to be afraid of them: for good reasons. When a group of academics acquires a powerful formal system, woe betide any naïve argument in its path. Mine here is self-evident: the terms ‘generative’ and ‘transformational’ are signals that the transformationalists care about the formal properties of models. When I say that the Neogrammarian model is generative, I do not mean either that I have done a detailed rewrite of it in the modern transformationalist notation (that is obvious) or that a modern generative model for historical linguistics would take the form of the Neogrammarian model, in whatever notation it is expressed. King (1969) provides an account of what such a modern generative model might be, while partially rewriting the Neogrammarian one. Many of the conclusions of the Neogrammarian reconstructions are not dislodged – they are described by different rules, and by a much more sophisticated notion of a rule. As the differences between synchronic dialects are expressed by the transformationalists in terms of the presence or absence of specific rules, so between chronologically ordered dialects the differences can be expressed in terms of the

252 The Historicity of Historical Linguistics

acquisition or loss of rules. In simple terms, the phonological and other developments of historical linguistics are, in the model, a set of statements of rule-acquisition and rule-loss. Two further principles operate also: those of rule-reordering and of simplification. For the first time, therefore, the operational elements in a synchronic and a diachronic comparison can be expressed in the same terms. The transformationalist approach to historical linguistics does not as yet produce a totally generative model as we have considered it here. It is not logically necessary, for example, in the establishment of their model, to have also a theory of the psychological or social mechanisms of rule-acquisition or rule-loss, but reference to such theories is (as usual) made. As in the ‘synchronic’ field, the generative grammarians are aiming at including all the phenomena that determine ‘competence’ in language. This aim has not been achieved, although it is a grand one, worthy of their best endeavours. The narrow ‘historicity’ of the Neogrammarian view has not, however, been retained: [T]he proper historical phonology of a language is clearly much more than a set of rules that derive the sounds of let us say, West Germanic from proto-Indo-European. Even if these rules are made as simple as possible in terms of the distinctive features involved, there is not the slightest reason to suppose that they correspond meaningfully to historical reality. Historical reality includes restructuring, and a simple enumeration of the innovations in a language need not bear any resemblance to what happened historically if the grammar has been restructured. (King 1969: 104)

And yet it seems that there remains a residual belief that a real ‘history’ is still demonstrable from a model: ‘A proper historical phonology is the history of the grammars of a language, of the competences of successive generations of speakers’ (ibid.). The strength of the transformationalist movement lay initially in its attempt at self-awareness at the model level. It should not be forgotten, then, that the grammars of the transformationalists are (as they are usually themselves aware) ‘only’ models of the competence of individuals. Indeed, the term ‘competence’ is a Saussurean ‘signified’ – itself an abstraction. While the code-book for the interpretation of the Neogrammarian model was labelled ‘history’, that of the transformationalists is labelled ‘competence’, to which their historical interests add only an appendix. The Chomskyan shears are sharper than those of Brugmann, for the aim is even more ambitious. Mechanical and Statistical Models For social anthropologists, first of all, a digression upon the derivation of LéviStrauss’s terms, which some have questioned. Nutini (1965, 1970 edn: 85) says: ‘Lévi-Strauss’s most distinctive and important contribution to the theory of social structure was his dichotomizing of models into mechanical and statistical’. Nutini rightly points out the ambiguities in Lévi-Strauss’s own usage, and the different interpretations of it of, for example, Nadel and Leach. He also notes: ‘Models of the same scale as the phenomena are called “mechanical”; those on a different scale are called “statistical”. Lévi-Strauss never explains what he means by “on the same scale as”’.

The Historicity of Historical Linguistics

253

The sources of the terms are, however, quite clear. Lévi-Strauss’s explicit and laudatory references to Wiener occur in his 1951 paper (Chapter III of LéviStrauss 1963a), and the remarks ‘on the same scale’ derive from Wiener’s discussion of the Maxwell’s Demon problem (Ardener, E. 1971a, pp. liv, lxxxiii; also, this volume, pp. 33 and 43). In the article on ‘Social Structure’ (1963a: Chapter XV) the linkage with Wiener occurs as follows (p. 284): The theory of a small number of physical bodies belongs to classical mechanics, but if the number of bodies becomes greater, then one should rely on the laws of thermodynamics, that is, use a statistical model instead of a mechanical one…

And further (p. 286): ‘Anthropology uses “mechanical” time, reversible and noncumulative … On the contrary, historical time is “statistical”; it always appears as oriented and non-reversible’. In Wiener (1948/1961), Newtonian mechanics and Newtonian time are discussed in Chapter I, and statistical mechanics in Chapter II. Gibbsian irreversible time, and the movement towards increasing entropy, is compared with Bergsonian time. Wiener is rich in sources of the imagery of Lévi-Strauss and Leach (entropy, topology, transformation groups; Ardener 1970). Wiener again: The succession of names Maxwell-Boltzmann-Gibbs represents a progressive reduction of thermodynamics to statistical mechanics: that is, a reduction of the phenomena concerning heat and temperature to phenomena in which a Newtonian mechanics is applied to a situation in which we deal not with a single dynamical system but with a statistical distribution of dynamic systems; and in which our conclusions concern not all such systems but an overwhelming majority of them. (1961 edn: 37)

Lévi-Strauss does not get the images quite right, and no doubt (as Nutini says) there was an association with Durkheim’s ‘mechanical’ and ‘organic’ solidarity to confuse the issue. Lévi-Strauss also slipped over into the use of ‘statistical’ to refer to notionally quantifiable averages – the ‘statistical norms’ of Leach. Beattie in Other Cultures (1964) understands the matter similarly. Wiener plays the role, in the genesis of Lévi-Strauss’s view of models, that Trubetzkoy and Jakobson play in the ‘phoneme’ stage of his structural view (see above, pp. 31–32). Remember the pun that anthropologie is entropologie (Lévi-Strauss 1955). Nevertheless, a careful reading of Wiener does elucidate Lévi-Strauss, despite the later accretions. Wiener, indeed, puts it very clearly in his final chapter on ‘Information, Language and Society’: It is in the social sciences that the coupling between the observed phenomenon and the observer is hardest to minimize. On the one hand, the observer is able to exert a considerable influence on the phenomena that come to his attention. With all respect to the intelligence, skill, and honesty of purpose of my anthropological friends, I cannot think that any community which they have investigated will ever be quite the same afterward … On the other hand, the social scientist has not the advantage of looking down on his subject from the cold heights of eternity and ubiquity … Your anthropologist reports the customs associated with the life, education, career, and death of people whose life scale is much the same as his own … In other words, in the social sciences we have to deal with short statistical runs, nor

254 The Historicity of Historical Linguistics can we be sure that a considerable part of what we observe is not an artefact of our own creation … There is much we must leave, whether we like it or not, to the un-‘scientific’, narrative method of the professional historian. (1948, 1961 edn: 163–4)

What Lévi-Strauss has enabled us to add (through his own later work) to Wiener’s far from negligible insights is that the method of the anthropologist and of the professional historian (as well as that of the linguist) is also a method of models. Finally: this paper has departed from its simple task of suggesting a few parallels in the modern trends of thought in social anthropology and linguistics, by looking back on a branch of linguistic theory that was the product of a revolution in method and conceptualization, and which has passed through many stages of growth, decay and reordering. Even if my treatment is distorted, or in some real linguistic senses wrong, the case is clear. The newer social anthropology and the newer linguistics both have similarly wide-ranging aims. The linguists have so far been more successful in making their point, because their body of common material is relatively more copious and slightly less difficult to obtain, but we should not retire simply because ‘the task may be too much for us’ (Lévi-Strauss 1963a: dedication).

17 Edward Sapir, 1884–1939

How shall we approach Sapir in an introductory lecture?1 There is really no need nowadays for an introduction to this important anthropological linguist. The student has all the materials readily to hand, or so it would seem. There is first of all his original famous work, entitled Language, published in 1921 (referenced by his memorialists quite frequently as his only real book, as opposed to papers and monographs). Then there is that compendious work of piety, the Selected Writings, collected by David G. Mandelbaum (1949).2 Its 617 pages largely fill the gap between his ‘real book’ and his life’s output – although even this volume does not contain the exhaustively complete works. Then there is the centenary volume, Edward Sapir: Appraisals of his Life and Work, edited by Konrad Koerner of the University of Ottawa and published recently (1984). This latter is what in English Departments would be called the ‘Critical Heritage’. It contains nine obituaries and memorials, dated 1939–52, reviews, mostly early, of Language (1921) and of the monograph Time Perspective in Aboriginal American Culture (1916), several reviews of the Selected Writings (1949), and finally various re-appraisals dated 1956–80. One begins to expect the centennial volume to be reappraised in the centennial volume by some version of the Russellian paradox! We may add to all of this a small paperback of selections from the Selected Writings (Mandelbaum 1956), the early memorial volume edited by Spier and others in 1941, and many more. There is, therefore (as I said), no special need for an introductory lecture. This man is the very stuff of Introductions, you may well think! One is best employed in a gathering like this in presenting a personal view (yet another one) that will bring out points which are not immediately apparent to the reader of the critical apparatus: the ‘dogs which did not bark in the night’, perhaps. Let us begin with this fact. In 1971, I referred in writing to Saussure’s rejection of the view of language as a mere labelling device. Soon afterwards, I received a letter from Professor C. Vœgelin of Yale, asking, in all seriousness, whether Saus-

256 Edward Sapir, 1884–1939

sure was a misprint for Sapir. I must confess to having been slightly taken aback. I repeated my references to the Cours (Saussure 1916, 1922 edn: 34, une nomenclature), but I could not help being somewhat impressed that the Atlantic was wide enough then for Sapir to be more salient than Saussure on this point. My own view at the time was quite other. In Saussure’s systematics, the rejection of any ‘nomenclature’ view is quite basic to the development of his whole later argument (such as we have it). Even if Sapir had said it, it was to me then as if I had referred to Newton’s falling apple and someone had seriously asked if this was a slip of the tongue or pen for William Tell! The personal stature of Sapir in the United States and especially at Yale has, then, to be taken as an important fact. In Vœgelin’s sketch of 1952, in the centennial volume, he refers to his ‘brightness’ among the ‘giants’ of American anthropology (‘with a brightness we associate with youth and poetry and innocence’ – Vœgelin, in Koerner 1984: 33). Kroeber, the great American anthropologist (1876–1960) and pupil of Boas, made an even more remarkable statement in 1959, within a year of his own death. Comparing Sapir directly with Boas, he said: I have always felt that Boas was an extraordinary person for his dynamism, for the energy, intellectual and ethical, which he could and did develop, for the output of his work, his range of interests, and so on. But I asked myself when I was doing one of the obituaries on him whether he was by the ordinary understanding of the term a genius or not. And I came to the conclusion that while he was a great man, he lacked the quality of genius of the sort that Sapir did exemplify.

He added: Edward Sapir, I should say, is the only man that I have known at all well, in my life, whom I should unreservedly class as a genius. (In Koerner 1984: 131)

Sapir was an almost exact contemporary of the anthropologist Malinowski. Both were born in the same year (1884). Sapir died in 1939, Malinowski in 1942, both still at the height of their powers. Both were foreign-born incomers to the countries they made their own (the United States and England, respectively). Sapir was born in Lauenberg in Germany, Malinowski in Krakow, Austrian Poland. Sapir was, however, a very young immigrant (only five years old) compared with the 26-year-old Malinowski. In other respects, their backgrounds were also different, and yet there is another, more intellectual aspect of their biographies which they share, as we shall see in a moment. Sapir’s family was Jewish and lived in New York, where his father was a synagogue cantor. These factors combined in his first choice of university (Columbia) and of academic subjects (‘Germanics’ and ‘Semitics’). On his graduation in 1904, he had already moved beyond these towards the Amerindian languages and his anthropological future. Nonetheless, his philological training was of the first importance in his theoretical development, while to the end of his life he continued to make important contributions to Semitic and Judæo-Germanic linguistics. In pursuit of his new specialism, however, he joined the staff of the University of California in 1907–08, where he worked on the language of the Yana Indians. Then at the University of Pennsylvania he

Edward Sapir, 1884–1939

257

studied the Paiute language. He received his doctorate at his home university of Columbia in 1909. In 1910, a totally new and formative period began, lasting until 1925. He went to Ottawa as Chief of a new Division of Anthropology of the Geological Survey of the Canadian National Museum. He married and had three children. During this time, he studied many Amerindian peoples and their languages, and he became well known for his contributions on the Nootka of Vancouver Island, the Athapascan languages and many more. By 1925, however, his wife had died, and he came back to the United States and to the University of Chicago. The memorialists speak of Sapir’s intellectual frustration during this period. The milieu of the Survey was not a conventionally academic one, but the experiences and the intellectual problems were new. This sort of enforced retirement to think in action reminds one rather of a feature of the biographies of a number of outstanding innovators in various disciplines of the period. A digression on this point may be in order here. Three scholars of the Modernist period (see Ardener 1985) share in their development the interesting feature that each, after an early partial success in conventional terms, spent some years in a practical milieu, during which their ideas and speculations were able to acquire a new kind of focus outside the purely academic mainstream. Freud, after the cocaine debacle, spent nearly a decade out of the public eye, before the publication of The Interpretation of Dreams. Einstein’s years at the Zurich Patent Office are famous for the history of his work on relativity. Nearer in status and discipline to the subject of this lecture is (once again) Malinowski who, after embarking on what was essentially an extension of the Melanesian survey work of his teacher Seligman, stayed for a complex of reasons on the single island of Kiriwina in the Trobriands. There his experience led him to a series of academic innovations, which resulted in his virtual refoundation of the British School of Anthropology. There is something very similar in its atmosphere to Sapir’s long period with the Canadian Geological Survey. It was indeed after eleven years with that body that the work appeared which established his new reputation. If there are such parallels, however, certain differences in the results of this productive isolation for Sapir’s branch of his discipline, compared with the obvious effects of the other thinkers on theirs, will require consideration later. Preston, in a valuable discussion, says that during the Ottawa years Sapir experienced a clear change in direction, away from his ‘Boasian’ orientation. He says: He was twenty-six years old and had recently received his PhD when he obtained, apparently to his surprise, an offer from the director of the Geological Survey, on Boas’ recommendation. Evidently Boas got him the job as a research position from which Sapir could get to the task of collecting ‘salvage’ data on the Indians of Canada. Boas had by that time three other students located in museum research positions – Kroeber at Berkeley, Speck in Philadelphia, and Lowie at the American Museum in New York … It has been speculated that Ottawa was also in part a kind of exile; Sapir was such an intensely brilliant colleague that Boas found comfort for himself in keeping Sapir at a distance. In any event, Ottawa did become an exile, a place where intellectual isolation and personal difficulties threw Sapir very much on his own personal resources. The result was a period of great productivity, followed by a period marked more by profound rethinking than by research activity. (In Koerner 1984: 179)

258 Edward Sapir, 1884–1939

A strange refraction of Sapir’s analytical mind is revealed in a poem he wrote at the time. Perhaps the discomfort of his mentors was justified: You sit before me as we talk Calmly and unafraid. Calmly and unafraid I sink my net into your soul, That flows before me like a limpid stream. I draw forth many lovely things That you had thought were hid; I draw forth many ugly things That you had thought were pure, That you had never thought to hide.3

His famous book Language (1921) was a product of this period, and it was recognized immediately as fresh in its approach, and written in excellent style. The volume exemplifies a particular combination of the skills of Indo-European historical linguistics, and of a general European background, with an encyclopaedic control of Amerindian languages. The effect of this union of skills was publicly to turn exotic languages into tools of general scholarship. Some of his later papers combine materials from Amerindian and Indo-European studies – and at their most technical, too, for example his 1938 paper ‘Glottalized Continuants in Navaho, Nootka and Kwakiutl (with a note on I.E.)’. Sapir wrote important papers on Tokharian (its relation to Tibetan, for instance) and Hittite (e.g. 1936). His appreciation of Hittite laryngeals was illumined by his awareness of Amerindian phonologies. From the time of Language onwards, this particular ‘worldwide’ linguistic scholarship became a characteristic feature of American anthropological linguistics. We take it for granted (or did – for it has begun to wither a little of late). It is a feature of Bloomfield’s great contribution (also called Language) of 1933. It is in Swadesh, Greenberg, Pike and many others. Yet Sapir was taken by them to be the model for them all. The main explicit theoretical points that people remember from Language concern the notion of drift as the motor of linguistic change. Drift is the name that Sapir gave to a kind of pattern dynamic that drove languages of common genetic origin when separated to continue to change in parallel ways. This is a typical Sapir quasi-theory – ‘quasi’ because it is almost impossible to make precise. There are quantities of illustrative cases that can be cited, however. One might be the independent development in south-western Irish Gaelic and in north-western Scottish Gaelic of a diphthong before velar nasals and laterals (e.g. ann/auN/ - ‘in it’; geall/g’auL/ - ‘promise’, where /L/ and /N/ are velarized phonemes). These changes occur, therefore, at the opposite and long-separated ends of the Gaelic realm (O’Rahilly [1932] 1972: 50–51, 122). Similar discussions have occurred, from time to time, in the context of the relationships between English at various early dates and Old Saxon, on the one hand, and Frisian on the other. For Sapir, however, the notion of drift developed (as later papers hint) from the problem of applying the historical methods derived from the analysis of texts of various dates to orally living material. He says somewhere (about establishing stages of a

Edward Sapir, 1884–1939

259

common language) that Common West Germanic cannot be ‘reconstructed’ with the phonemes /hl/ and /hr/ by utilizing the modern West Germanic languages alone, as those languages have ‘lost’ (that is, show no trace of) /h/ combinations of that sort. Faced only with languages still living today, we should, therefore, be permitted by the rules to restore /r/, /l/ only. When West Germanic was ‘Common’, argues Sapir, all the Western Germanic languages must have had such forms: the common absence is a post-separation phenomenon. They all lost the missing phonemes later, by linguistic drift. Sapir developed this view because of a genuine problem in setting up fixed intermediate stages in reconstructing the earlier interrelationships of Amerindian languages, using standard historical linguistic methods. Restored stages cannot be regularly and consistently fixed to specific linguistic features. Nowadays this has a modern ring. A Celticist recently declared that when we refer to Common Celtic, we may be referring to a period of several thousand years. To unpack the meaning of such non-Brugmannian heresies would need a full consideration of wherein lies the historicity of ‘historical reconstruction’ (cf. Ardener 1971b; also above, chapter 16, pp. 236–254), but for our immediate purposes it is sufficient to note that experience of Amerindian ‘reconstruction’, without historical documents, reflected upon in Ottawa, contributed to Sapir’s own early insight. Language is also marked by a concern for linguistic typology and for classifying previously inadequately analysed language groups. The Sapir reclassification of Amerindian languages was bold and has been subsequently built upon. Its methods certainly inspired Greenberg in his own reclassification of African languages in the 1950s (Greenberg 1955 and subsequent publications). That classification does not resemble the close-knit, densely documented classification of, for example, the Indo-European languages, backed by regular tables of sound changes. The classification of Niger-Kordofanian, in which the Bantu family found itself redefined as a greatly swollen subset, is much looser, more statistical – with a Heisenbergian element of uncertainty. Sapir had a very strong influence here. All very good, but possibly not riveting for a non-specialist? In fact, it is quite misleading to look in Sapir’s writings merely for technicalities – he is deeply informative on these – or for an orderly theory. Nothing can be set up in the way of an organized exposition of a theory of language along the lines of, say, Jakobson or Saussure. Yet among his contemporaries it is clear that ‘Sapirism’ was a total cast of mind. We saw how Professor Vœgelin couldn’t (or didn’t) accord certain priorities to Saussure over Sapir. In this connection, Bloomfield’s review of Sapir’s Language, in 1922, a year after its publication, is also of interest: Dr. Sapir in almost every instance favors those views which I, for one, believe to be in accord with our best knowledge of speech and of the ways of man … As Dr. Sapir gives no bibliography, one cannot say how much of his agreement with scholars who have expressed similar views is a matter of independent approach. For instance … the author develops what he justly calls an ‘important conception’, – the ‘inner’ or ‘ideal’ phonetic system of a language: it is exactly the concept of distinctive features developed by the school of Sweet, Passy, and Daniel Jones…

(Bloomfield fails to mention the clear terminological echoes of the innere Sprachform of von Humboldt.)

260 Edward Sapir, 1884–1939 The same concept was developed (independently, I think) by Franz Boas (Handbook of American Indian Languages…) and by de Saussure (Cours de Linguistique Générale [Paris 1916]). It is a question of no scientific moment, to be sure, but of some external interest, whether Dr. Sapir had at hand, for instance, this last book, which gives a theoretic foundation to the newer trend of linguistic study. (Bloomfield 1922, in Koerner 1984: 47)

A question, we may think, of some ‘scientific moment’. It is interesting that there were really two routes to the ‘discovery’ of the phoneme (to which Bloomfield was referring): through the vaguely German inner forms of language, and through precise phonetic analysis. Bloomfield lumps together both approaches, but there is no doubt that he himself belongs more with the latter and Sapir with the former. In my own view, however, the important point is that the day of the new synchronic systematics had dawned, and from now on the contribution of Sapir would appear intuitive and unorganized in comparison. The discourse of Sapir takes for granted a holistic view of language. It was an irony that the Chomskyan purity squads, in their heyday years after his death, dubbed him mentalistic, and thus allowed him into the pantheon of the Cartesian predecessors of transformationalism, a movement with which surely Sapir would have had little sympathy. As a final word about Language before we leave it, there is no doubt that it suffered from the appearance only a year later of the vastly successful second edition of Saussure (so often reprinted). As we have already mentioned, Bloomfield’s own Language (surely the title must have been a deliberate if not a provocative echo?) appeared in 1933 and almost immediately ushered in the American phase of structuralism; which, because of its immense vogue, to some extent obscured Sapir, who would in six years be dead – on the very eve of that war which strengthened (through the famous military crash courses) the general awareness of the contribution of what for long was thought of as ‘Bloomfieldian structuralism’. It is quite clear, of course, that that school – characterized as ‘objectivist’, ‘anti-mentalist’, and which proposed ‘discovery procedures’, based on the concept of an infinite corpus of observable data – certainly does not belong in the same theoretical world as Sapir’s. Nevertheless, his work was not totally overwhelmed, as we have seen. The American school never lost its respect for his cast of mind. Let us look at the nature of that cast of mind. He had, first of all, a broad view of the idea of culture. He explicitly included in it ‘higher culture’, such as literature and poetry. As we have seen, he aspired to verse himself. Furthermore, he included mental phenomena (his so-called ‘mentalism’), as is illustrated by papers such as ‘Cultural Anthropology and Psychiatry’ (1932), writing in terms that Evans-Pritchard (who recommended Sapir as reading) might in part have used: The true locus of culture is in the interactions of specific individuals and, on the subjective side, in the world of meanings which each one of these individuals may unconsciously abstract for himself from his participation in these interactions. (In Mandelbaum 1949: 515)

In 1934, we have, as a development of this, his famous article on ‘Symbolism’ in the Encyclopaedia of the Social Sciences, much leaned on by Victor Turner (1964: 30ff; 1967). He distinguishes there between ‘referential symbols’ (including ‘oral speech, writing, the telegraphic code’, etc.) and ‘condensation symbols’ (‘a highly condensed form of substitutive behaviour for direct expression, allowing for the

Edward Sapir, 1884–1939

261

ready release of emotional tension in conscious or unconscious form’). He notes, however, that language, although at one level referential, has deep condensation roots. This point was lost subsequently, when the rise of communication/information theory tended to overstress the aspect of language as a ‘code’. In an interesting article in Psychiatry in 1938, called ‘Why Cultural Anthropology Needs the Psychiatrist’, he refers to being rather shocked when, reading J. O. Dorsey’s Omaha Sociology, he would find an account of a cultural practice, followed by some such phrase as ‘Two Crows [an Omaha informant] denies it’ (the title now of a recent thorough treatment of the Omaha by Dr R. H. Barnes [1986]). It was, says Sapir, as though Dorsey had not squarely met the challenge of his source material and given the kind of data that we, as respectable anthropologists, could live on. It was as though he ‘passed the buck’ to the reader, expecting him by some miracle of cultural insight to segregate truth from error. We see now that Dorsey was ahead of his age … The truth of the matter is that if we think long enough about Two Crows and his persistent denials, we shall have to admit that in some sense Two Crows is never wrong … The fact that this rebel, Two Crows, can in turn bend others to his own view of fact or theory or to his own preference in action shows that his divergence from custom had, from the very beginning, the essential possibility of culturalized behaviour. (In Mandelbaum 1949: 567, 572)

It is not surprising that the linguists did not see in this much more than ‘mentalism’ at this time. What we have here is a perception that wherever human beings are, total consistent systems cannot be. There would be four decades of structural and transformational theory to get through before the idea of language as a total system began to come under general attack. As for the anthropologists – among them, the idea of consistency in social systems had almost as long a life. Perhaps one of the great puzzles about Sapir is why he should have been remembered by most linguists and anthropologists today for something known as the ‘Sapir-Whorf hypothesis’. I have often discussed this matter from Whorf ’s point of view (Ardener 1971a, 1982; also above, chapters 1 and 11), and the question of attribution has considerable complexity. I have repeatedly argued that the hypothesis as commonly summarized (that language determines the view of reality) does not really square with Whorf ’s views, and that the extreme relativism implied was a product of his professional colleague’s interpretations and his own occasionally hasty phraseology in semi-popular writings. It is the more surprising that Sapir should have been made, as it were, to take the prior responsibility for what was to become a rather controversial thesis. It seems likely that the ‘Sapir-Whorf hypothesis’ received its public baptism from Harry Hoijer’s contribution (of the same title) to a conference of 1953 in Chicago, which led to a major review of these ideas, published under his editorship as Language and Culture, in 1954. In that volume, Hockett (‘The Whorfian Theses’) and Fearing (‘Conceptions of Benjamin Whorf ’) refer only to Whorf in their papers. In the ‘Discussions’, which form half of the book, all save Hoijer refer to the ‘Whorf hypothesis’ or the like, and by the end so does he (‘the Whorf hypothesis’, p. 263). Hoijer’s own ‘Preface’ confirms the matter (1954 edn: ix) with its references to ‘the Whorf hypothesis and its problems’, and to ‘what Whorf [not Sapir] actually said’. The Hoijer volume remains one of the best discussions of Whorf. At the

262 Edward Sapir, 1884–1939

time, it was widely perceived as a major attack on ‘the hypothesis’. In fact, it is still required reading on the wider problems, which are still with us. To return to Sapir’s possible role in all this, there is, of course, a famous citation of Sapir by Whorf, as we shall see in a moment, and Whorf also refers respectfully to his studies with Sapir. However, Whorf describes important hypotheses on the subject that he developed before he met Sapir, which are very much his own and in his very recognizable mental style – particularly those linking the materiality of action with linguistic conceptualization. It is as if, when Whorf ’s approach had its main post-war impact, which came after a posthumous publication (1952), it was important to some that Sapir’s role should be given priority and his name provide a certain professional respectability to it. I am not aware that he himself ever declared anything so scientifically grand as a ‘hypothesis’. The Whorf-like passage, quoted by Whorf, is from Sapir’s paper, ‘The Status of Linguistics as a Science’, published in 1929, having begun life as a public address the year before: Language is a guide ‘to social reality’. Though language is not ordinarily thought of as of essential interest to the students of social science, it powerfully conditions all our thinking about social problem and processes. Human beings do not live in the objective world alone, nor alone in the world of social activity as ordinarily understood, but are very much at the mercy of the particular language which has become the medium of expression for their society. It is quite an illusion to imagine that one adjusts to reality essentially without the use of language and that language is merely an incidental means of solving specific problems of communication or reflection. The fact of the matter is that the ‘real world’ is to a large extent unconsciously built up on the language habits of the group. No two languages are ever sufficiently similar to be considered as representing the same social reality. The worlds in which different societies live are distinct worlds, not merely the same world with different labels attached. (In Mandelbaum 1949: 162)

Some large claims are certainly made in this paragraph, which is, however, one among many in a somewhat discursive paper covering the relationships between language and human life. Furthermore, there is another passage (Whorf-like in a slightly different way) in an earlier paper of 1924, with which it is not entirely consistent: It would be absurd to say that Kant’s ‘Critique of Pure Reason’ could be rendered forthwith into the unfamiliar accents of Eskimo or Hottentot, and yet it would be absurd in but a secondary degree. What is really meant is that the culture of these primitive folk has not advanced to the point where it is of interest to them to form abstract conceptions of a philosophical order. But it is not absurd to say that there is nothing in the formal peculiarities of Hottentot or of Eskimo which would obscure the clarity or hide the depth of Kant’s thought – indeed, it may be suspected that the highly synthetic and periodic structure of Eskimo would more easily bear the weight of Kant’s terminology than his native German.

Whorf might have written that last clause.) Further, to move to a more positive vantage point, it is not absurd to say that both Hottentot and Eskimo possess all the formal apparatus that is required to serve as a matrix for the expression of Kant’s thought. If these languages have not the requisite Kantian vocabulary, it is not the languages that are to be blamed but the Eskimo and the Hottentots themselves. The languages as such are quite hospitable to the addition of a philosophical load to their lexical stock-in-trade. (In Mandelbaum 1949: 154)

Edward Sapir, 1884–1939

263

Now all that may be true or false, but it is quite conventionally universalistic in tone. The subject is Whorfian but the conclusions are somewhat different. On the other hand, later in the same paper (ibid.: 157), he is quite aware that ‘innocent linguistic categories may take on the formidable appearance of cosmic absolutes’, and he speaks of language producing ‘spurious entities’. On the whole, Sapir’s lack of formal theorizing makes it difficult to see him as a caricature SapirWhorfian. Nevertheless, much that is sound in discussions of the relations between language, thought and reality is to be found in his writings. How shall we evaluate Sapir? There is no doubt that his concern with what he kept seeing as Psychology of Culture diverted his energies in the later years. Unlike Malinowski, with whom he has points in common, he did not locate his insights in either a general theory of language or in a general theory of society. Sapir died in 1939. His widow has written: ‘Edward died with the feeling that he had an important point to make that he hadn’t managed to get across’ (see Koerner 1984: 192). There was a failure of sorts. His perception of the links between life and language were taken too literally as a matter of linguistics alone. It is not surprising, as we have said, that Sapir should have found his ‘important point’ eluding him, when it is realized that both anthropology and linguistics were at the height of their modernistic, scientizing obsessions. In addition, his best insights require a knowledge of historical linguistics which is rare in general anthropologists. The path into psychology was trodden not only by him, but (of course) by Benedict, Mead, Bateson, Nadel, Fortes and many others (including Audrey Richards). The then available psychologies were, however, either too individual, too positivistic, too culturally loaded or too totalitarian in their claims (psychoanalysis managed to be all these at once!) for this to be other than a blind alley. It has taken most of the post-war period for anthropology itself to work through the Mind/Society/Language relationship – the seventeen years to date of the Journal of the Anthropological Society of Oxford alone probably provide sufficient illustration of that! The case of Sapir is a good illustration of the way that an apparently successful professional life can be lived with powerful intellectual winds against it. Even if we recognize certain intuitions of Sapir’s that are now more subtle than we first thought, the appropriation of Sapir by the ‘Sapir-Whorf ’ discussion makes it very difficult to put him forward as the ancestor of post-Whorfian discussions, especially when we note the largely ad hoc nature of many of Sapir’s views. One thing is certain: both he and Whorf have suffered from misinterpretations of quite serious, though not identical kinds. Let us remember, however, that they, like Rosencrantz and Guildenstern, ‘did make love to this appointment’. They accepted conventional ways out, oversimplified, went with the grain of their times, and perhaps they could not always even remember what they had originally wished to say. They would not be the first, nor the last, to fall into these traps! Sapir had, and still has, great exemplary value in this and other ways – especially for humanists. Without his memory, certainly American anthropological linguistics (and much anthropology of language in general) would not have so sturdily weathered (as it did) the mortal storm of Chomsky’s transformational formalism that left so much else in ruins for a generation.

18 The Construction of History ‘Vestiges of Creation’

A theory of history has long eluded social anthropology. Even Evans-Pritchard seemed to see it, at one level, as a series of synchronic sections (1950, 1961a), at least when discussing practical fieldwork method. At another level, he clearly saw history as a matter of retrospective interpretation (as in his own reinterpretation of the history of social anthropology). Lévi-Strauss (1958, trans. 1963a) saw it as the uncovering of unconscious processes (Structural Anthropology). It is not, therefore, entirely an objective matter. Yet the absence of a sound theoretical protection against the dangers of total arbitrariness is evident. In 1971, I drew attention to the problem of the chess puzzle. Some chess problems show a state of the chessboard that has as a ‘history’ previous real states of a game that has really been played. Others are ‘constructed’ states that might have been reached in a game, but were not. These constructed states are passages in games that have never been played. Yet we cannot tell the difference. Phillip Gosse, the father of Edmund Gosse, and a natural historian, was a committed nineteenth-century creationist. He was unable to bring himself to reject the Biblical chronology of Archbishop Ussher in the light of the tremendous geological and palaeological advances of his period. Darwin’s great work was a considerable shock and challenge to him. He argued that Adam’s body had been itself created bearing the evidences of a non-existent biological growth and development. It was not enough to posit that he must have lacked a navel (a common creationist argument!): the problem was more fundamental than that. In the same way, he argued, the earth was created together with its fossil history of millions of years. The fossils were for Gosse (1907) likewise mere ‘Vestiges of Creation’, which itself took place only a few thousand years ago (Ardener 1971b: 227; also above, chapter 16, p. 248). Thus, this anti-evolutionist Plymouth Brother stumbled on a

The Construction of History

265

real question, lost though it was in the preposterousness of its particular polemical application. For, in a sense, all baselines of history are conceptually in this situation: real histories are, in the absence of total documentation (what would total documentation be like?), rearranged by changes in the infinite sequence of successive presents, producing, as with the chess puzzle, histories that did not happen. If we could artificially start the game, somewhere, we might trace the shape of the changing historical space. Partial approximations are possible. Hastrup’s account of the trajectory of the internal categories of the Icelandic Free State, from its foundation in about 800 (by settlement) to its total redefinition in the eleventh century (by absorption into the Norwegian state), is the most advanced we have in social anthropology (Hastrup 1985). It incorporates a model of catastrophic rearrangement over time, that takes into account the continuity of the human rearrangers (Ardener 1975; also above, chapter 5). Yet no historical space is truly independent. The Free State was part of the wider Atlantic space of its own period, and now forms part also of our own historical (or historiographical) space, while Hastrup’s own study is itself part of our social anthropological space. It is thus not a matter of spaces nesting in hierarchical sequence, relatively tidily. On the contrary, we may argue, all such spaces englobe each other anti-, or better, a-hierarchically. The ordering feature is the frame. Perceptually and cognitively the framing ‘rules’ (fluctuating and ambiguously evidenced as they frequently are) are the source of the very concept of the space. We may say the frame and the space are just the internal and external aspects of the same concept. Where is the real world among these possible worlds? Is it worth asking this question? Some have seemed happy enough to do without it – hence the extreme idealist tendencies of many approaches that have otherwise been ready to undertake the specification of the cognitive aspects of human social life. The approaches loosely seen as structuralist were ultimately subject to this criticism. In their strong form they are totally nihilistic in their contemplation of the question of reality. Nevertheless, in a ‘post-structuralist’ world, we can attempt to unpack this anomaly in structural analysis, for its nihilism is technically most evident in the matter of history. In the structuralist frame, history is subject to the same structuring as other narrative, and is ultimately reducible to text. This achievement (which was anticipated and paralleled outside anthropology) is a genuine one. The study of mythology is now seen to encompass at one level all history, or at least all historiography. It seems, then, that structuralism is ‘Gossian’ in its view of the past. Any ‘historical’ elements (in the older, securer sense) in a narrative are mere ‘vestiges of creation’. Although likely to be hotly rejected, even by structuralists in their heyday, this is a strong position (as I have already argued) and not something to be ashamed of. Nevertheless, this ‘flatness of vision’ in structuralism is a result of its essential ‘textuality’, the demonstration of which I will not repeat here (Ardener 1985; also above, chapter 13). Edmund Leach was early involved in this controversy through his analysis of Old Testament sources (Leach 1969). This is a field in which the documents are seen as both text and truth by many experts. I recall that, in a public lecture, he extended his analysis to the New Testament, and showed Christ and John the Baptist to be structural transformations of each other. A theologian rose and said: ‘To some of us present these were real people’. That is: the text was indeed his-

266 The Construction of History

tory. I do not intend to go into all the possible discussion. With Biblical data, all neoanthropological analyses produce versions of this reaction. For example, Mary Douglas’s analysis of Leviticus (1966) produced parallel objectivist discussions. The Biblical objections merely raise in relatively explicit doctrinal form the question of whether the very possibility of structuralism denies the possibility of history, save perhaps by a version of Piaget’s inadvertently appropriate remark (cited in a previous discussion) that ‘God is a structuralist’ (Piaget 1971, cited in Ardener 1985: 53; also above, chapter 13, p. 197). Nevertheless, the answer to the implicit question, like what songs the sirens sang, ‘is not beyond all conjecture’. It is interesting to compare two classes of text with this question in mind. Leach (1969: 81) cites memorable features of the rulers of the Tudor dynasty of England. Henry VIII and Elizabeth I appear as structural transformations of each other (broadly as male/overuxorious: female/underuxorious). They and other royal figures associated with the dynasty also appear in structural opposition to each other. Although intended light-heartedly (but not less revealingly), this sort of thing (or rather, the very possibility of this sort of thing) inevitably raises questions about Lévi-Strauss, just as the ‘proof ’ that Max Müller was a Sun-deity did concerning that other great mythologist. Let us suppose that such analyses are possible on ‘historical’ narrative, and there is plenty of evidence that they are; how could they come about? (1) For oral history, ‘traditional’ history or the like, we may simply argue that the memory of events has been totally restructured. They have been turned into narrative, and obey the structuring processes of narrative. In effect, this is the position taken about the majority of texts that tell a story about the past – before the advent of ‘professional’ history, or alongside it. It is the uncomplicated structuralist position. In cases like the Tudors, however, the documented accounts of the period itself have strong narrative features, which give it a certain ‘epic’ quality. If these derive from retrospective restructuring, does it occur particularly in particular periods? Is it possible that certain historical conjunctures are prone to crystallize historiographically in such ways? ‘Happy the country whose history is boring’, the saying goes. Why, if history is simply (unconsciously) restructured as narrative, is any history boring? We may have been able to conceive such a question since the ‘objective’ recording of poorly structured times and places has become possible (Ardener 1987; also above, chapter 14). In oral history there are, indeed, actual blanks, which suggest the failure of the structuring process. (2) A second approach would be to admit the aforementioned, but to propose instead that sometimes events arrange themselves, and individuals fall into relationships, that have resemblances to the structure of more mythological narrative. This solves the problem of embarrassingly ‘fictional’-looking stretches of history and keeps such questions as the historicity of Jesus and John the Baptist (as well as of Henry VIII and Elizabeth I!) satisfactorily open. This is essentially a concession to the existence of professional history, and it is not without value as accounting for the greater narrative richness of certain stretches of even conventional narrative. This is, however, only another way of conceding in advance that structuralist patterns are coin-

The Construction of History

267

cidences in ‘real’ history – which does not help the structuralist analysis of any text, which someone, somewhere, is prepared to declare historical. Furthermore, when we are dealing with oral or traditional statements, we still cannot separate, by inspection, any fortuitous stretches of structuralized history from the hypothesized total restructuring or ‘mythologizing’ that is expected to mark a piece of traditional text in any case, and which was after all the original point of structural analysis. (3) So we would be better placed to argue alternatively that structural oppositions are built into history as it happens. There are, indeed, plenty of grounds for saying that the ‘memory’ of history begins when it is registered. It is encoded ‘structurally’ as it occurs. The structuring, by this view, is actually part of the ‘registration’ of events (Ardener 1978; also above, chapter 5). Then we can say that since not all events do survive, but only ‘memorable’ or ‘significant’ events (Ardener 1987; also above, chapter 14), the structural processes are not necessarily retrospectively imposed, but are synchronic – all part of the very nature of event-registration (with or without any specific physical recording). It is what I have referred to as the ‘slaughter of the event’, the creation of the ‘dead stretches’ of the registering process. Such a conclusion is uncontroversial enough nowadays. Retrospective restructuring simply continues the process, since each restructured event is freshly objectivized and there could be an infinite sequence of rememorization. (4) Finally, we may go further and say that the relations between persons, and their mutual definitions, actually embody ‘structuralist’ processes. The relations between parent and child, between lineages, between affines, are self-defining relations of symmetry and opposition. Not only does this recognition properly embody the structuralism of Les Structures Elémentaires and its successors into the historical process (which Lévi-Strauss never did), it also recalls the Freudian processes of identity construction in the family. Henry VIII (uxorious, male) and Elizabeth (non-uxorious, female) then ceases to be a structuralist whimsy, and becomes part of an intergenerational psychoanalytic drama. Freudianism and structuralism are metaphysics of a similar order, as I have argued before (Ardener 1971a; also above, chapter 1). Nevertheless, we begin at this point to part with structuralism as such, although we are near the world of Lacan and revisionist Marxist-Structuralism. The important point is that event-specification begins in one important aspect, in the structuring of relations between persons. So far in the language of ‘structure’ I have elicited four levels, and a fifth by implication, which I will summarize not in the order of unpacking but in reverse order. a) Structures of personal relations, or structures inherent in the mutual selfdefinition of persons. b) Structures by which relationships are registered or perceived (‘events’). c) Structures through which registered events are remembered. d) Structures ‘imposed’ on events, retrospectively (restructuring). e) Structures of text. This is, for our historical purposes, the implicit fifth level, which is critical for our discussion; for narratives can be structured de novo

268 The Construction of History

in the absence of all the preceding levels. Mythologies, pseudo-histories, are precisely the core materials which originally lent themselves to structural analysis and which are notionally without history at all. This summary unpacks further the implications of a fully structuralist discussion of history as fairly as possible, although it has never been previously articulated. It is, as a result, no longer structuralist at all, since it has been possible to articulate it only by revising structuralism in the light of history itself. And, even so, we still cannot dissolve the problem with which we started; for if we have no historical documentation, only a narrative, the levels are obliterated and structuralist analysis cannot ‘recover’ them. All resemble level (e). R E STAT E M E N T

Memory has been an evident theme in the preceding discussion. Sperber (1975) has been responsible for emphasizing this theme, although Leach had early on linked structure to memorability. The structures of structuralism, with their pluses and minuses, work at such a relatively unconscious level, that the apparently banal observation that myths (for example) are structured as they are for their memorability is an explanation of honourable simplicity. Sperber cites Bartlett’s pre-war experiments, whereby messages are modified in transmission in the direction of deterioration modified by reconstruction. We may recall the comic case, immortalized by the late Will Hay, in which the message ‘Am going to advance: send reinforcements’ ultimately became ‘Am going to a dance: send three and fourpence’. The most stable thing about the message is the structural pattern. What, however, is memory when moved from the individual to the social, from its organic context to a collective simulacrum of itself ? It seems that it is mainly a way of labelling that stability of pattern. Once more, the term has moved up a level: the ‘story’ behaves like a memory, but it is the story that is now memorized, not the events it purportedly embodies; and, it must be repeated, we still cannot on structural grounds tell the difference. This is the material basis of the movement of ‘life into text’ (Ardener 1985; also above, chapter 13), which eventually accounts for the perennial existence of the problem of the actual confusion of life and text. When regarding the aforementioned summary of levels, we see that they clearly partake of what I call a ‘simultaneity’. The movement from (a) to (b) is a movement from individual experience to social experience; the passage through (b) and (c) and (d) is a passage through the social space. The distinction between (b) and (c): whether historical structuring is laid down as the events occur, or is ‘adventitiously’ perceived retrospectively, is not a difference in principle. Restructuring is a continuous revision of previous states. Again, the distinction between (c) and (d): the perceiving of structural coincidences in past events, or the complete reconstruction of the past, is similarly ambiguous. We can speak of (d), complete reconstruction, only in the absence of knowledge about a continuous series of intermediate stages. What then about (e) or mere myth or story, complete unhistory? How do we know that it actually occurs? By now it seems clear that we have here a total analogue of the theoretical problem presented by sets of unit categories; on the surface they picture a totally rel-

The Construction of History

269

ativistic universe. At the level of the social space itself, they produce a historically relativistic universe. It is my argument that as materiality is perceived through the concept of ‘semantic density’ in the linguistic aspect, and of variable ‘event density’ in the synchronic aspect of the social space, so are its effects revealed as an underlying feature of the space as a whole, in the form of ‘historical density’. The problem of the ‘flatness’ of the Gossian world – the inability to decide how much of it was created yesterday – is the macrocosmic aspect of the ‘flatness’ of category sets. ‘Historical density’ is the trace left by previous reconstructions of the space, each of which forms in some sense a barrier to perceiving the realities of their predecessors by wiping out the ‘structures’ that expressed them. Charles Kingsley, in his response to Gosse, exclaimed that he could not believe that God (the structuralist?) had written ‘an enormous lie upon the rocks’. We may ourselves ask how the shadow of the ‘vestiges of creation’ would appear to us in the search for real past states. SOM E CA SE S

In Northern stories to be discovered in Scandinavian, Old English and German sources appear the names of peoples and persons who are documented on the continent between the years 360 and 500. During those years, Gothic dominance expanded from the Baltic to the Black Sea under Ermenaric, only to be shattered by the Huns. The Western Roman Empire succumbed under the resulting pressures and was replaced by successor German states including that of Theodoric the Great. A minor event was the destruction of the Burgundian settlements on the Rhine, casually swept aside by Attila, now leader of the Huns, and a mass of subject forces, on their great push into Gaul to eventual stalemate and defeat on the Catalaunian Fields. The destruction of the Burgundians led to the death of their king Gundahar (Gunthaharius, Gunnar, Guthhere, Gunther) in 437. He, together, with Ermenaric (Jormunrekr, Eormenric, Ermenrich) who died in about 370, Attila (Atli, Aetla, Etzel) who died in 453, and Theodoric (Thjodrekr, Theodric, Dietrich von Bern) who died in 526, all appear as contemporaries in the common stock of tales. What picture do we receive? There is no trace of the Roman Empire. The scene is set in a world of royal or chiefly lineages, in which Attila and the Huns are indistinguishable from the Germans. The marital problems of Gunnar dominate the action. Two queens, Gunnar’s wife Gudrun (Kriemhild) and Brynhildr, the wife of his friend or vassal Sigurdr, quarrel over precedence when bathing. Gudrun blurts out a statement that leads to the unravelling of a skein of relations by alliance or contract, and to the death of all concerned. The beneficiary, and the cause, is Atli, who marries the widowed Gudrun/Kriemhild, who eventually is the cause of his death. At this point we may note that the ‘real’ Attila died, according to Roman sources, after spending the night with a Germanic slave girl Ildico (Hildiko, little Hildr), who it seems was absolved from blame. The stories are enriched and expanded in numerous recensions, but most of these were transmitted not in the Rhenish area of history, but in some Norwegian world that never participated in the original events, finally being preserved and elaborated in a mass of material (the Poetic Eddas and related prose sagas) re-

270 The Construction of History

dacted in the North Atlantic, in Iceland. Meanwhile, in mediaeval Germany, the tale was differently restructured in romantic ways as the Nibelungenlied. The Volsungasaga and the Nibelungenlied belong now in different social spaces. Hardly an event can be identified as ‘historical’, yet the Norse recensions are perceived as patriotic records. Rising German nationalism in the late nineteenth century utilized the stories for new purposes. The Wagnerian recension belongs to no tradition; it is a new recension, a restructuring in its own right. We meet here some remarkable continuities in names and genealogical relationships, among the debris of a historical period. Every level of structuring treated earlier appears in it. Every conventional analysis, thematic or textual, leads deeper and deeper into certain cruces: the central constellation of characters swing round the peculiar quarrel between the two queens, leading to the death of one and involving the other in the collapse of the alliance with Attila. To regard the Northern Tales as mythology is perfectly justified. We find almost every feature of fifth-century reality completely removed. Motivations for the characters are contradictory and obscure. They are never ‘artistically’ resolved. The basic political themes have dissolved. Other themes have been pressed into service. Not even its status as a Burgundian epic survives. Its time and place have evaporated. Yet certain personal configurations and an anomalous unresolved conflict have everywhere been transmitted. If names and relationships are in some way ‘real’, but the events have vanished, or have been transformed or recontextualized, can history in some way have survived? I am reminded of a letter by L. Sprague de Camp in Astounding Magazine (February 1939: 154–5), which raises this question in relation to the historicity of King Arthur: Consider this hypothetical case: I’m writing a story of the super-titanic-colossal type, involving a war between the galaxies, with the fighters shooting planets at each other for missiles … I want a name for my hero, so I look on the obituary page of the New York Times, and discover the obit of a certain Platypus McDandruff, who was a corporation lawyer specializing in railroad legislation. I like the name, so my hero becomes Platypus McDandruff. Would a future historian, comparing my story with the obit of the real Platypus, be justified in saying that [my hero] and the lawyer were ‘the same’?

Nevertheless (pace Astounding Magazine) we know, by the still unoverturned theory of bricolage, that all structures are put together from ‘events’ of some kind (the fashionable, but old process of deconstruction in which we are now involved is thus débricolage). It is a hypothesis that existing structural relationships of a strong kind are not lightly abandoned. These are the flotsam and jetsam of real life. In the Burgundian case, it is a possibility that unresolved anomalies, always located in the same area in the various surface texts, evidence the core of the traumatic events in Western Europe as seen, not by the Romans, but by those upon whom the effects were like an extraordinary explosion and expansion of reality: the Germans of the period. It would be as if we were looking at the scattered remains of a supernova. I argue that the fragmented remains of the transforming event are still perceivable, preserved in the ambiguous relations at the heart of the story. The destruction of that family and the Attilan events were closely related, and were seen as a collapse in the internal relations of contract, service and alliance. A small domes-

The Construction of History

271

tic scandal unravelled their world at a critical moment, and so many worlds were similarly overwhelmed together that this one, with its ambiguous but clearly familial causation, became the model for them all. We should not be surprised that that ultimate deluge was drawn into the Norse prediction of Ragnarokr: the collapse of the cosmological order; or that the fateful scandal should be tied to even greater mythological precursors, involving gold and dragons, and divine interventions. In the case of Brynhildr, she is the centre of some sort of rebellious, headstrong claim. She has status but is under no consistent family control, save that of her husband Gunnar. She claims her sister-in-law’s husband. She is in some dependent relationship to Atli. Gudrun/Kriemhild is, in structural contrast, locked into family relations at all turns. It is, again, not a surprise that the separate story of the Valkyrie Sigrdrifa and her fire should be tacked on to account for the disruptive Brynhildr’s ill-omened autonomy – at least in the North. In the Nibelungenlied she is an heiress addicted to athletic sports! The case of Maelbetha, a Scottish king of the pre-Norman period of Britain, exemplifies a similar situation in some ways. All early Scottish historical record is perceived through layers of political restructuring, and a passage through a linguistic redefinition, from Gaelic to English. In the course of the latter, his name was twisted into the form ‘Macbeth’, which passed through several Scottish sources into Shakespeare’s play. The earliest record, which survives significantly in a geneaological tract, shows his story to be motivated by the destruction of the regular rotation of royal succession between two collateral lineages. ‘Young’ king Duncan is the beneficiary who inherits out of turn, succeeding his grandfather Malcolm directly. The claims of the disinherited line survive in the female line, in the persons of the lady Gruoch and her son Lulach. She marries Maelbetha. He himself has weaker claims through female links, as well as representing even more ancient Northern claims of Moray to the crown. For Shakespeare, following centuries of restructuring by Scottish historians, we have a tale of usurpation by Macbeth based on fantasies of legitimacy (a version before Shakespeare’s has only a vision of the three witches) personified as evil. Yet somehow Lady Macbeth is to blame. Gruoch’s bloodstained claim emerges anomalously in the sanitized Stewart tale. She ‘has given suck’ she states, but where is the infant with those ‘toothless gums’ to which Shakespeare makes her allude? Where is Lulach, her son, who succeeded her husband briefly after his seventeen years of apparently fairly harmonious rule? In 1603, Macbeth must be childless: that is the point. His line and his claim must have died out. Macbeth belongs to a world before the Stewarts. The restructuring is orderly, however: young upstart Duncan (minus) killed in fair fight (plus), now transformed into good old Duncan (double plus) treacherously killed in bed by his host (double minus!). The lady Gruoch’s relationship, stripped of its legal context, remains an ‘anomaly’ in the heart of the story. The deficiencies of structuralism include a too ‘cognitive’ view of narrative. Gruoch is, at the level of the flat surface of the later redactions, a peculiarly marked element in the narrative. The principle of ‘density’, which reshaped our view of categorical sets by shadowing realities underlying the arbitrariness of classification, elicits, when applied to the structure of the Macbeth narrative, an area of relative structural silence round Lady Macbeth, which is semantically dense,

272 The Construction of History

and potentially event-rich (Ardener 1987). The excellent principle of ‘the dog that did not bark in the night’. I have been able only to sketch lightly a few implications of these two examples – themselves only two out of a range of available examples. My basic conclusion is that certain ‘event-related’ structures do not restructure easily. It is my hypothesis that a certain clustering of anomalous features will show the trace of the survival of a structure from ‘life’. Pure text were it to exist, would, in contrast, present almost no barriers to quite arbitrary restructuring. This would be excellently exemplified, indeed, by Sprague de Camp’s fantastic case. Finally, it was not my purpose to carry as a subtext to this paper a further demonstration of the theory of muting. If the strangely unarticulated characters of Gruoch, Brynhildr and Gudrun/Kriemhild are all women, we should not be surprised by that. They are not the only centres of event-richness in the mythological domains – not by far (Atli for example and his absent Huns). Nevertheless, it is as well to remember that the tradition is, in both cases, a male one, in that the dominant restructuring has gone through males. However, if these texts also suggest that deeply structured areas in the traditions involve areas at the heart of key familial relations, then we shall obviously expect the ‘evidences of creation’ to lie, in part, in the history of women. E DI TO R I A L N OT E TO E . W. A R D E N E R : T H E CON STRUCTI O N O F H I S TO RY

This paper is published here in the same form as presented to the ASA conference, Easter 1987. Mr Ardener died on 4 July of the same year, and the paper was unrevised at the time of his death. He would certainly have wished to add to this paper, to amplify some of the argument, and to include discussion of points raised during the conference. The paper, however, stands without this. The conference paper was distributed with one final note: ‘the notes and references to this paper will be available later’. Unfortunately, nothing was found in Mr Ardener’s papers to suggest that the notes and references existed even in partial form at the time of his death. The conference text included various author/date references, which are filled out below (and it is reasonably clear in all instances to which works he was referring). The section relating to old Germanic sources is unreferenced, and the original contained the sentence, ‘The details of this case will appear in the fuller version of this paper’. No attempt has been made to add this detail, but some probable bibliographical suggestions can be made. It seems possible that Mr Ardener relied to some extent upon the work of H. M. Chadwick (see particularly 1912; also 1940), who provides detailed discussion of appropriate sources, events, characters and historiographical traditions. For the Nibelungenlied, see Mowatt 1962, and Andersson 1987. For the Edda and the Volsungasaga, see, for example, Dronke 1984. For the early sources of the Macbeth legend, see A. O. Anderson 1922, and M. O. Anderson 1973. Shakespeare’s main sources were probably Hector Boece’s early sixteenth-century Latin reconstruction of Scottish History, Scotorum Historiae, disseminated in the Scots translation of 1531 by John Bellenden, The Chronicles of Scotland, and passing through later recensions at the hands of George Buchanan and Raphael Holinshed.

The Construction of History

273

Mr Ardener suggested, during the conference, that he was probably the first person to cite Astounding Magazine in an ASA conference paper. This seems probable. Out of respect for the historiographical imponderable presented by the case of Platypus McDandruff, no attempt has been made to check this source. Malcolm Chapman, May 1988.

Postscript 1 The Prophetic Condition By Kirsten Hastrup

In the Munro Lecture given at Edinburgh in 1975, at which I was present, Edwin Ardener identified the prophetic condition. It is a condition, both for structures and individuals, of being, as it were, between two worlds. While situated in the ‘old’ world, prophets give voice to a ‘new’ one. Often the voice of prophecy is not heard; it seems incomprehensible beforehand; afterwards it may seem trivial: when the new world has become commonplace the prophet’s voice cannot be distinguished from general speech. When the new world fails to happen, or when it remains latent for an extensive period, prophets may not be registered at all. When the epistemological gap between the worlds is too great, the structural condition is not in favour of prophecy. Prophets may be discarded along with other anomalies in the social. The privileged prophetic condition obtains at that singular point in time when a discontinuity is beginning to register, but when it is still not capable of expression. I do not propose to repeat what Edwin Ardener has lucidly expressed elsewhere in this volume, which derives its title, ‘The Voice of Prophecy’, from the Munro Lecture of 1975. It goes without saying, also, that my thoughts are not patented interpretations of the master’s voice. The concept of the prophetic condition, however, serves as an apt starting point for the present thoughts on the work of Edwin Ardener. Quite apart from the obvious qualities of the concept in our dealing with discontinuities in social spaces, the prophetic condition may serve as an allegory for Edwin Ardener’s work, and ultimately for the entire anthropological project. Prophets do not predict the future, in the terms of the present. Rather, they foretell a present reality before it has been accommodated in the collective representations, and in language. Prophets, therefore, have to create a new language

The Prophetic Condition

275

which in turn helps define the new reality. This was very much the case of Edwin Ardener, whose vision of a new anthropological reality could be expressed in the current discourse only with difficulty, and whose conceptual universe only gradually could be disclosed as categories were developed. Although there is a remarkable continuity in Edwin Ardener’s work, there is also a gradual reshaping of language. What was only dimly perceived in the 1960s could be expressed more simply in the 1980s. The language expanded as new conditions were realized. Once the prophecy is fulfilled a new structural condition obtains, in which the language must of necessity expand still more. Prophecy links language, space and time. This is why the prophetic condition gives privileged access to the anthropological reality. In the following I shall use these three aspects to label separate entries into this reality, as defined by Edwin Ardener (and as subsequently discovered by this apprentice). Since my aim is not exegetic I shall refer explicitly only to a few of his papers. Given the continuity of his thought, all his papers reflect a singularly integrated view of the world, however, and the range of reference is immaterial. The concepts I single out are chosen for their exemplary qualities and for their short-circuiting of the wide-ranging discussions of ‘the real’ – to which I shall return in the final section of this paper. L A N GUAGE : SE M A N T I C D E N S I T Y

The position of language has long been a major concern in Edwin Ardener’s anthropology. Language has a dual relationship to the social, metonymical and metaphorical. In the first sense it is part of the social; it is one feature among others. In the second sense it ‘expresses’ the social; language represents the world, so to speak. This dual definition of language makes it of singular significance in our determination of the separate realities lived by various peoples. Language is much more than a linguistic superstructure upon a material reality – language and materiality are not apart, allowing some anthropologists to forget language, and others to forget materiality. Language is closely linked with the material; in the social experience they form a simultaneity. The match between them is virtually tautological, although not in the simple sense of categories shaping thought and experience. The link between language and materiality, as perceived by Edwin Ardener, is most forcefully expressed in his notion of ‘semantic density’ (see above, chapter 11, part 2). It concerns the interrelationship between the category system and ‘reality’ – the statistical and material features of categories, of which certain meanings are socially more significant than others. An example will illustrate this: to an international academic community ‘Oxford’ has a semantic density around the university; the meaning of the term has a centre of gravity there which make the industrial outskirts and the supermarkets seem peripheral. In one sense, everything in that particular urban area of Southern England is equally ‘Oxford’. Within the meaning, however, there is an uneven spread of significance. All categories may potentially yield similar density patterns; this is not solely a feature of famous locations. Categories always coexist with a statistical feature, which is part of their materiality. Language itself does not reveal this; experience does. This implies that no reality can be exhaustively described by its own cate-

276 The Prophetic Condition

gories. In the social experience and in practice categories have unequal densities which lock them into the material world. The categorical and the material form an experiential simultaneity. Density is related also to frequency, that is to the question of which meanings will be implied more often than others when particular categories are invoked. This is aptly illustrated by identity categories, where peoples are named and defined according to selected criteria, reflecting a semantic density centring around only a minority of the actual population. Such densities can be identified in anthropological fieldwork. Through experience of an ‘other’ reality, the fieldworker transcends the language of the obvious and internalizes its material implications. Since densities are not immediately acknowledgeable in language, they cannot be translated. They must be recreated in a separate language, whose creation is the anthropologist’s task. SPACE : E V E N T R I CH N E S S

At a more comprehensive level of analysis the social space as a whole is marked by similar densities. The counterpart of the semantic density of particular categories is the event richness of certain social spaces. Some spaces or some periods seem to generate more events than others. This is not solely a mensurational feature, but also more importantly a feature of registration. For events to be registered as such, they have to be significant. Although events, as happenings, do have objective properties, it is not these properties as such that give them their significance. Significance is projected from a cultural scheme of a different order. Some social spaces are more event rich that others. This means that more happenings are registered as events, and more behaviours interpreted as actions. Even richness is a feature of ‘remote areas’ in particular (see above, chapter 14). In ‘remote areas’ all singularities of the social space are continually reinforced in response to the intruding outer world. The individuals and their actions are defined by the social space, but they are also definers of the space. Consequently, event richness is both a defining feature and a consequence of definition. The material and the definitional merge. In order to register events and comprehend their significance, one has to have recourse to a larger whole – a world. We may define events as relations, between a happening and a structure of meaning. Worlds, then, are relations between relations and societies are their embodiment. While relations can be inferred from material facts, relations between relations have no direct manifestation. This is comparable to the notion that the meaning of categories cannot be heard. Again, fieldwork is a necessary precondition for knowing the empirical. We will never know the unequal experiential densities of categories and social spaces through hearing and seeing, because the social is no mere text but a lived materiality. The dual relationship between language and the social has a parallel in the relationship between individuals and their world. There is a continuity between the social space and the individuals that constitute it – a continuity which is also of a dual nature. The individuals are defined by the world of which they are part, but they are simultaneously the definers of the space (see above, chapter 14). Thus,

The Prophetic Condition

277

the tautological match is enriched by a feature of unpredictability, individualism and manipulation. While individuals generally experience and express only what their world singles out for registration, some of them may also step out of it and experiment with either action or expression (see above, chapter 9). Thus, the two-faced continuity between individuals and their world is the fundamental precondition for history, containing both continuities and changes; it is also the fundamental precondition for prophecy. TI M E : H I STO R I CA L DEN S I T Y

History is always in the making. Individual actions either reproduce or transform the structure of the social space. Thus history is always adjective to the world under study. However, the story of the past is a selective account of the actual history. The selection is not accidental – it is the corollary of a structured memory. Not all events survive in the memory of people. For events to be memorized and to become part of ‘history’, they have to be or to have been experienced as significant. This apparently self-evident point covers a profound truth: the structuring of history, or the selective memory, is not solely imposed retrospectively. Although the universe created and memorized in history is relative to a series of successive presents, memorability is identified synchronically. Contemporary event registration is the basis for the material left to us as history. The logic of the relationship between history as reality and history as story, is akin to the relationship between the world and its language, or between the empirical and the definitional aspects of reality. The dimension of space has been replaced by the dimension of time, but even this distinction is more or less an analytical artefact belied by experience. The world and the semantic densities unfold in time, which cannot therefore be separated from the space. When speaking of long-term history, we may still distinguish between synchronic and diachronic perspectives, but in fact even they are mutually defining. In the interest of comprehending the nature of history I shall retain the distinction for the moment, however. The effect of semantic density and of the concurrent event richness, identified in the synchronic dimension, is a generation of a particular historical density in the representation of the diachronic. Historical density, or relative memorability if you wish, is the trace left by previous (re-)constructions of the space. Although the story is always at one remove from history itself, it is no arbitrary reconstruction. The retrospective recasting of events is a continuation of the contemporary process of event registration. History, then, is a sequence of rememorizations of events originally perceived as such. The structure of the memory, the way in which the past is presented, are very much particular features of the world of which they are a part. The modes of registration and the conventions of representation apply equally to densities in space and time. In that sense, history is as culture-specific as are the meanings of particular categories. The relativization of history is a mode of objectivization (see above, chapter 14). We cannot deal objectively with history without first acknowledging the

278 The Prophetic Condition

profound relativity of memory and story. Histories are stories of the past behaving like memories, but as time passes it is the story itself that is memorized. However, memory relates to a material experience of densities and events which is reflected in the reconstruction that occurs through time. History, therefore, is never, unlike myth, mere story. Like the world itself, history is not reducible to text. T H E A N TH RO P OLO GI CA L CON D I T I O N

At several points in the preceding discussion I have alluded to the prophetic qualities of anthropology. I will now make this explicit. Edwin Ardener’s vision of the anthropological reality implies a simultaneity of discovery and definition. The anthropologist discovers new worlds while they are defined. In that sense, anthropologists may ‘speak’ new worlds into being. The gradual or abrupt development of the discursive space of anthropology reflects a changing reality. Usually, anthropologists give voice to contemporary but spatially separate realities. We write other cultures. This is not different from the voice of prophecy, either in principle or in fact – even though prophets are generally thought of as perceiving a temporally separate reality within the same space. The point is that the prophetic condition is principally defined, not by a linear time referent, but by a multiple referent of time, space and language. The prophet foretells a different time-space; in language he defines a new world. The anthropological condition is an equivalent of this; in narrating and writing the other, the anthropologist is author to a reality. This principal correspondence between the prophetic and the anthropological condition has a practical implication at the level of anthropological categories. Through our perception of separate worlds, we sense the inadequacy of local categories. In our devising of new categories, different realities may be discovered. In the anthropological discourse, discovery and definition still merge. Not all anthropologists are equally perceptive, of course. Nor are they all equally creative in the shaping of new language. Only a few are prophets by gift. Edwin Ardener was, in my view, one of these.

Postscript 2 Towards a Rigorously Empirical Anthropology By Maryon McDonald

Many students of anthropology have found themselves intrigued but confused by the work of Edwin Ardener. No doubt the density and economy of his prose account in part for this. There is more, however, to the apparent opacity of Ardener’s ideas. The structure of his work, the metaphors he uses, and the nature of his critique of positivism, are not always those most familiar to, and expected of, post-1960s’ anthropology. The dominant schools of thought in British social anthropology were fairly simple until the 1970s. The common rote, which most students of anthropology acquired, ran through evolutionism and diffusionism, functionalism and structural functionalism, and then structuralism. And then what? Structuralism came and went, the old securities had gone, but somehow little had changed. The various versions of post-structuralism or post-modernism which followed provided a happy lifeboat for some, but often within the structures of the same securities from which they imagined themselves to be sailing free. Ardener remained apart from these securities and excitements, seeing them as the stuff of ethnography rather than the means of analysis, and consequently muddled readers and listeners who wished, by means of them, to pin him down. Most monographs and introductory texts of the 1960s (and many present-day introductory courses in social anthropology) still tend to assert solidly empirical domains such as politics, economics and kinship, and then to add a distinct domain called something like ‘beliefs’ or ‘systems of thought’. Relation classifications – social structure and culture, or social relations and ideas, were common, self-evident and acceptable. Such structures were, in part at least, responsible for

280 Towards a Rigorously Empirical Anthropology

structuralism’s appeal, but they subsumed it. In general terms, structuralism fitted well with many of the moral and political ideals and promises of the 1960s, sitting (as was required) in synonymy with opposition to the Establishment and materialism. Within social anthropology more specifically, structuralism appealed as an invitation to study domains – of values, beliefs, symbolism, myth and ritual – which functionalism had helped to make expressive, representational and secondary. It was precisely in domains of this kind that the 1960s found the meaning of life; when structuralism gave priority to these areas, it appeared open to the charge of idealism, and to lend it credence. A division of labour developed in anthropology around these dichotomies. This division of labour, which structuralism encouraged, and in which it was caught, encouraged new specialisms such as symbolic anthropology and cognitive anthropology – specialisms which seemed distinct from the concerns of the empirical man. Marxian ideas reappeared for a while, accusing these specialisms of floating well above the ground. Structuralist, symbolic and cognitive anthropologists seemed, in the meantime, to do much to confirm the justice of the accusation. Politics, economics, kinship and social structure were often left untouched; even, perhaps, virtually scorned as the stuff of empiricist reaction. The empirical and the symbolic, the real and the ideal, and a host of similar dichotomies, gathered themselves around the issues of quantitative and qualitative research, as if around two naturally separate enterprises. The insights that structuralism could have brought were lost in an apparent divide between ‘rigour’, on the one hand, and ‘(mere) interpretation’ on the other. Much of the anthropology which is now self-consciously post-structuralist or post-modern has been trying to pull itself out of this. If structuralism, however, which did, after all, have the apparently objective (if fictive) underpinning of the ‘human mind’, was able so easily to fall into the jaws of the very structures it should have been examining, then it is perhaps not surprising that many of the anthropologies that have come since have, within these same structures, been eaten alive. Edwin Ardener’s work inevitably shares with that of other anthropologists many of the new influences that came into the discipline in the 1960s and 1970s. An older tradition had to be made sense of somehow, but this could not be done satisfactorily through accretion, simply tacking the new ideas on to the old in the bulging category of ‘beliefs’ or in new specialisms. Linguistics and linguistic models, theories of meaning, phenomenology, hermeneutics and much else besides, generally persuaded anthropologists that positivism was not for them. However, much contemporary anthropology is still celebrating the idea that the discipline is not a natural science. This is often done with a dated notion of what scientists think and do, accompanied by embarrassing forays into literary criticism, resulting in much textualism and poetics, in texts which themselves sometimes show extremes of selfblindness, or of tortured and self-indulgent reflexivity. Ardener’s critique of positivism was of a completely different order. His long-standing interest in linguistics gave him a grasp of Saussurean ideas which other anthropologists, struggling with Saussure through a veil of structuralism, did not have. He also had a background of many years of detailed empirical work in the field in Africa, part of which was studiously quantitative. Moreover, in his case, the context in which re-thinking was effected was not simply social anthropology, but the

Towards a Rigorously Empirical Anthropology

281

human sciences more generally. All these factors helped to give Ardener’s work its own peculiar stamp, and enabled him to step over the traps that structuralism laid for the less wary. This meant that his prose was not always attractive to those seeking, in their eschewal of science, the titillation of the arts. When Ardener used the analogies of text and translation (see above, chapter 11) it was to show, through detailed technical examples demanding of the reader, the problems and limitations of such models of the anthropological endeavour, and to send wide ripples of such problems through literature, archaeology, ethology and biology. Ardener’s critique of positivist anthropologies – of functionalism, say – was not one which criticized them for their neglect of text or poetics, or for their lack of concern for ‘meaning’ (as if neglecting the soul). Rather, he criticized those who believed themselves to be the champions of empirical rigour for being neither rigorous nor empirical enough (see above, chapter 2). To those still celebrating the joys of 1960s’ idealism, this was as incomprehensible as it was to many of the ‘old-school’ empiricists. Some of the generalities of Ardener’s early critique were filled out in later papers on specific themes. His reflections on population, demography and ethnicity, for example, illustrate in a fairly straightforward way the impossibility of separating qualitative from quantitative concerns. All mensuration is inevitably categorybased, as is the apprehension and generation of social reality more generally. Reflections on the old quantitative/qualitative divide are not simply, therefore, reflections on the necessity of knowing the categories through which, as it were, the counting is done. These are reflections which, as they encounter this divide, or any of the other divides mentioned early, take us back each time to a classificatory base-line of human cultural life. The cultural worlds which anthropologists study – whether the anthropologists themselves define their focus as economics, politics, kinship, or the more exotic symbolism, myth or ritual – are composed of self-defining worlds of categories in action. As Ardener makes clear, classification is not always simply contained in, or backed up by, language; if the anthropological endeavour is that of ‘translation’ between different category systems, then understanding is a question of ‘disequation rather than of equation’ (see above, p. 183), and this ‘disequation’ is an inevitable feature of the relative autonomy of these systems as they confront one another. This is in itself apt comment on the excitement, and also the confusion and misunderstanding, which some of Ardener’s own work has had the capacity to generate. A focus on classification has been too easily tidied back into the dichotomies I mentioned earlier. However, which some features of reality may feel more material than others, they are none the less conceptual for that. There is no ‘more’, no ‘something else’ requiring analytical attachment to the underbelly of classification in order to ground it; no ‘real’ world of ecology or economics innocent of cultural perception; no analytically distinct area, whether motor or residual, of ‘social change’; no gap between classification and reality. Ardener’s work leads us, by a difficult but rewarding route, to the kind of self-aware empiricism that many anthropologists are now seeking. Only in an older language would one feel that the world had been turned on its head because social structure or economics had been pulled, at last, into the domain of ‘beliefs’.

Appendix: Edwin Ardener – a Bibliography

1952 ‘A Socio-economic Survey of Mba-Ise’ (Ibo, Eastern Nigeria), West African Institute of Social and Economic Research, typescript (bound). 1953 ‘The Origins of Modern Sociological Problems Connected with the Plantation System in the Victoria Division of the Cameroons’, West African Institute of Social and Economic Research, conference proceedings, sociological section, March, pp. 89–105. 1953 ‘A Rural Oil-palm Industry: 1) Ownership and Processing’, West Africa, 1909, 26 September, p. 900. 1953 ‘A Rural Oil-palm Industry: 2) Opposition to Oil Mills’, West Africa, 1910, 3 October, pp. 921–3. 1954 ‘The Kinship Terminology of a Group of Southern Ibo’, Africa, 24, April, pp. 85–99. 1954 ‘Some Ibo Attitudes to Skin Pigmentation’, Man, 54 (101), May, pp. 71–3. 1954 ‘Democracy in the Cameroons’, West Africa, 1932, 6 March, p. 203. 1956–9 Various sections in Annual Reports on the Cameroons under United Kingdom Administration (unattributed), HMSO, London. 1956 Coastal Bantu of the Cameroons (The Kpe-Mboko, Duala-Limba and Tangayasa Groups of the British and French Trusteeship Territories of the Cameroons), London, International African Institute (Ethnographic Survey of West Africa, 11). 1957 ‘Sociological Investigations of the West African Institute of Social and Economic Research in the Southern Cameroons – Digest of Principal Findings’, WAISER, duplicated. 1957 ‘Cameroons Swing to Tribalism’, West Africa, 2090, 4 May, p. 411. 1957 ‘Numbers in Africa’, Man, 226, November, p. 176. 1958 Various sections in Victoria Southern Cameroons 1858–1958 (unattributed), Victorial Centenary Committee, Victoria, Southern Cameroons (now West Cameroon), London, Eyre and Spottiswood. 1958 ‘Marriage Stability in the Southern Cameroons’, Nigerian Institute of Social and Economic Research, conference proceedings, mimeographed.

Edwin Ardener – a Bibliography

283

1958 ‘The People’, in Introducing the Southern Cameroons (unattributed), Federal Information Service, Lagos, pp. 17–21. 1958 ‘Wovea Islanders’, in Nigeria, 59, pp. 309–21. 1958 ‘The “Kamerun” Idea’, West Africa, 2147 p. 544 and 2148 p. 559. 1959 ‘Lineage and Locality among the Mba-Ise Ibo’, Africa, 29 (2), pp. 113–34. 1959 ‘Cameroons Election Aftermath’, West Africa, 2185, February, p. 195. 1959 ‘The Bakweri Elephant Dance’, Nigeria, 60, pp. 31–8. 1959 (with D.W. MacRow), ‘Cameroon Mountain’, Nigeria, 62, pp. 230–45. 1960 ‘The Linguistic Situation in the Southern Cameroons’, Nigerian Institute of Social and Economic Research, mimeographed. 1960 ‘A Note on Intestate Succession’, Nigerian Institute of Social and Economic Research, mimeographed. 1960 (with S. Ardener and W.A. Warmington), Plantation and Village in the Cameroons, Oxford, Oxford University Press. 1961 ‘Duala’, in Encyclopaedia Britannica. 1961 ‘Kpe’ (Bakweri), in Encyclopaedia Britannica. 1961 ‘Historical Research in the Southern Cameroons’, third conference on African history and archaeology, School of Oriental and African Studies, mimeographed, July. 1961 ‘Crisis of Confidence in the Cameroons’, West Africa, 2306, 12 August. 1961 ‘Cautious Optimism in West Cameroon’, West Africa, 2313, 30 September, p. 1071. 1961 ‘Social and Demographic Problems of the Southern Cameroons Plantation Area’, in Southall, A. (ed.) Social Change in Modern Africa, pp. 83–97, London, Oxford University Press. 1962 ‘The Political History of Cameroon’, The World Today, 18 (8), pp. 341–50. 1962 Divorce and Fertility – an African Study, Oxford, Oxford University Press. 1963 ‘Imperialism and the British Middle Class’, West Africa, 2391, March, p. 357. 1965 Contributions to La Population du Cameroun Occidental (unattributed), Société d’Etudes pour le Développement Economique et Social, Paris. 1965 Historical Notes on the Scheduled Monuments of West Cameroon, West Cameroon, Government Publication. Translated into German 1969 as Anmerkungen zur Geschichte de geschuetzten Denkmaeler West Kameruns, Buea, Government publication. 1965 (with S. Ardener) ‘A Directory Study of Social Anthropologists’, British Journal of Sociology, 16 (4), pp. 295–314. 1965 Review of ‘The Revolution in Anthropology’, Jarvie, I.C. 1963, Man, 65, p. 57. 1966 Comment on ‘Frazer and Malinowski: the Founding Fathers’, Leach, E.R. and Jarvie, I., Current Anthropology, 7 (5), December, p. 570. 1966 Contributions to the English version of La Population du Cameroun Occidental (unattributed), Société d’Etudes pour le Développement Economique et Social, Paris. 1967 Comment on ‘Competence and Incompetence in the Context of Independence’, Colson, E. Current Anthropology, 8 (1–2), February–April, pp. 101–3. 1967 ‘The Nature of the Reunification of Cameroon’, in Hazelwood, A. (ed.) African Integration and Disintegration, London, Oxford University Press. 1968 ‘Documentary and Linguistic Evidence for the Rise of the Trading Polities between Rio del Rey and Cameroons 1500–1650’, in Lewis, I. (ed.) History and Social Anthropology, ASA Monograph 7, London, Tavistock. 1970 Kingdom on Mount Cameroon: Documents for the History of Buea, 1844–1898 (forthcoming).

284 Edwin Ardener – a Bibliography 1970 Review of Custom and Politics in Urban Africa: A Study of Hausa Migrants in Yoruba Towns, Cohen, Abner, Oxford Magazine, 7, Hilary, pp. 199–200. 1970 ‘Witchcraft, Economics, and the Continuity of Belief ’, in Douglas, M. (ed.) Witchcraft Confessions and Accusations, London, Tavistock, pp. 141–160. 1970 ‘Galileo and the Topological Space’, Journal of the Anthropological Society of Oxford, 1 (3), pp. 125–30. 1971 ‘The New Anthropology and its Critics’ (Malinowski Lecture, 1970, with appendix), Man, 6 (3) pp. 449–67. 1971 (ed.) Social Anthropology and Language (ASA Monographs 10), London, Tavistock. 1971 ‘Introduction’ to Social Anthropology and Language, Ardener, E. (ed.) 1971, London, Tavistock, pp. ix–cii. 1971 ‘Social Anthropology and the Historicity of Historical Linguistics’, in Ardener, E. (ed.) Social Anthropology and Language, London, Tavistock, pp. 209–41. 1972 ‘Language, Ethnicity and Population’, Journal of the Anthropological Society of Oxford, 3 (3), pp. 125–32; reprinted in Beattie J.H.M. and Lienhardt, R.G. (eds), 1975. 1972 ‘Belief and the Problem of Women’, in La Fontaine, J. (ed.) 1972, The Interpretation of Ritual, London, Tavistock; also in Ardener, S. (ed.) 1975, Perceiving Women, London, Dent. 1972 Introduction and Commentary to reprint of Specimens of Dialects: Short Vocabularies of Languages: And Notes of Countries and Customs in Africa, Clarke, J. 1848, Farnham, Gregg International. 1973 ‘Behaviour: a Social Anthropological Criticism’, Journal of the Anthropological Society of Oxford, 4 (3) pp. 152–4. 1974 ‘Social Anthropology and Population’ (Wolfson Lecture, 1973, ‘Population and “Tribal Communities”’), in Parry, H.B. 1974, Population and its Problems, Oxford, Clarendon Press, pp. 25–50. 1975 ‘The Cosmological Irishman’, New Society, 14 August. 1975 ‘Language, Ethnicity and Population’ (see 1972), in Beattie, J.H.M. and Lienhardt, R.G. (eds) 1975, Studies in Social Anthropology: Essays in Memory of E.E. Evans-Pritchard, Oxford, Clarendon Press, pp. 343–53. 1975 ‘The Problem of Women Revisited’ in Perceiving Women, London, Dent; USA, Wiley, pp. 19–27 (also contains ‘Belief and the Problem of Women’, pp. 1–17, first published 1972, in La Fontaine, J. (ed.). 1975 ‘The Voice of Prophecy: Further Problems in the Analysis of Events’ (Munro Lecture); see above, chapter 9. 1976 ‘“Social Fitness” and the Idea of “Survival”’, Journal of the Anthropological Society of Oxford, 7 (2), pp. 99–102. 1977 ‘Comprehending Others’, paper given to the Wenner-Gren Symposium, part 2 published as ‘Social Anthropology, Language and Reality’, in Parkin, D. (ed.), 1982, and in Harris, R. (ed.) 1982; part 1 published as ‘Ethology and Language’, in Harré, R. and Reynolds, V. (eds) 1984. 1977 Introduction to Social Anthropology and Language, Ardener, E. (ed.), 1971, was issued in three parts in Spanish translation, in the series Biblioteca de Linguistica y Semiologia (Paidos: Buenos Aires), 6–8, 6, Antropologia Social y Lenguaje, 7, Multilinguismo y Categoria Social, 8, Antropologia Social y Modelos de Lenguaje. 1978 ‘Some Outstanding Problems in the Analysis of Events’ (ASA Conference Paper, 1973), in Schwimmer, E. (ed.) 1978, Yearbook of Symbolic Anthropology, London, Hurst, pp.

Edwin Ardener – a Bibliography

285

103–20; reprinted in Foster, M. and Brandes, S. (eds) 1980, Symbol as Sense, New York, Academic Press. 1979 ‘Social Anthropology’, in A New Dictionary of Sociology, Duncan Mitchell G. (ed.), London, Routledge and Kegan Paul. 1980 ‘Ten Years of JASO’, in Journal of the Anthropological Society of Oxford, 11 (2), pp. 124–31. 1981 ‘The Problem of Dominance’, Journal of the Anthropological Society of Oxford, 12 (2), pp. 116–21; reprinted in Dubew, L. Leacock, E. and Ardener, S. (eds), 1986, Visibility and Power: Essays on Women in Society and Development, Oxford, Oxford University Press. 1982 ‘Preliminary Chronological Notes for the South of Cameroon – Rapport de Synthèse’, Contributions de la Recherche Ethnologique à l’Histoire des Civilisations du Cameroun, 2, pp. 563–77, ed. Claude Tardits, Centre National de la Recherche Scientifique, Paris, 1982. 1982 ‘Social Anthropology, Language and Reality’, in Parkin, D. (ed.), Semantic Anthropology, New York, Academic Press; reprinted in Harris, R. (ed.) Approaches to Language, Oxford, Pergamon Press. 1983 ‘The ASA and its Critics’, Royal Anthropological Institute Newsletter, 56. 1984 ‘Ethology and Language’, in Harré, R. and Reynolds, V. (eds) The Meaning of Primate Signals, Cambridge, Cambridge University Press; Paris, Editions de la Maison des Sciences de l’Homme. 1985 ‘Social Anthropology and the Decline of Modernism’, in Overing, J. (ed.) Reason and Morality (ASA Monographs 24), London, Tavistock. 1986 ‘Barbara Pym and the Social Anthropologists’, Sunday Telegraph, 6 July. 1987 ‘“Remote Areas”: Some Theoretical Considerations’, in Jackson, A. (ed.) Anthropology at Home, London, Tavistock. 1987 ‘Edward Sapir (1884–1939)’, Journal of the Anthropological Society of Oxford, 18. 1987 Japanese translations of ‘Belief and the Problem of Women’ and ‘The Problem of Women Revisited’, respectively pp. 33–58 and 121–34, of Women, Nature and Culture, Tokyo, Shobun-sha. 1989 ‘The Construction of History: “Vestiges of Creation”’, in Tonkin, E., McDonald, M. and Chapman, M. (eds) History and Ethnicity, London, Routledge (first delivered as a conference paper, ASA conference, 1987).

Notes

N OT E TO I N T RO DUCT I O N

1. A variety of appreciations followed Ardener’s death. The following is not a complete survey, but see: The Journal of the Anthropological Society of Oxford (1987, XVIII: 2), for pieces by Anthony Boyce, Malcolm Chapman, David Parkin, Zdzislaw Mach and Andrzej Paluch; St John’s College Notes, for pieces by Freddy Beeston and Anthony Boyce; Anthropology Today (1987, III, 4), for a piece by Malcolm Chapman and Maryon McDonald; West Africa (3660, 5/10/88), for a piece by Martin Njeuma. There were also unattributed obituaries in The Times (9/7/87), The Oxford Times (12/7/87), The Oxford Mail (10/7/87), The Jericho Echo, and Jericho News. N OT E S TO CH A P T E R 1

1. (Editorial note) The papers contained in the volume Social Anthropology and Language, to which chapter 1 of this book was an introduction, were: H. Henson, ‘Early British Anthropologists and Language’, pp. 3–32. R.H. Robins, ‘Malinowski, Firth and the “Context of Situation”’, pp. 33–46. D. Hymes, ‘Sociolinguistics and the Ethnography of Speaking’, pp. 47–94. J.B. Pride, ‘Customs and Cases of Verbal Behaviour’, pp. 95–120. W.H. Whiteley, ‘A Note on Multilingualism’, pp. 121–8. E. Tonkin, ‘Some Coastal Pidgins of West Africa’, pp. 129–56. N. Denison, ‘Some Observations on Language Variety and Plurilingualism’, pp. 157–85. D. Crystal, ‘Prosodic and Paralinguistic Correlates of Social Categories’, pp. 185–208. E.W. Ardener, ‘Social Anthropology and the Historicity of Historical Linguistics’, pp. 209–42. G.E. Milner, ‘The Quartered Shield: Outline of a Semantic Taxonomy’, pp. 243–70. C. Humphrey, ‘Some Ideas of Saussure applied to Buryat Magical Drawings’, pp. 271–90.

Notes

287

2. (Editorial note) Supplementary text, taken from original, pp. lxviii–lxix: It was through the German-born Max Müller at Oxford that the comparative linguists had their chief effect. Müller developed his views in the more speculative phase of linguistic studies of the age of Bopp and Schleicher, before the rise of the Neogrammarians. He remains, however, the only theorist with anything remotely like a modern approach to myth. Tylor’s interest in deaf-and-dumb and sign languages prefigures some of the proposed semiology of Saussure, but he held firmly to an evolutionist view that early linguistic signs were ‘motivated’. The use of ‘native’ categories like mana, totem, and taboo did not at this earlier period (nor, indeed, much later) lead on to a consideration of the relation of category to language. Nevertheless, the collection of ‘comparative’ material under these heads did lead to important advances in the hands of other theorists (totemism: Lévi-Strauss 1962a; taboo: Freud 1913; Steiner 1956; Firth, R., 1966: 109–13; mana: Mauss 1950: 101–115; Firth, R., 1940; Milner 1966). 3. If not never, then hardly ever. The correspondence involving Bohannan (1956, 1958a, 1958b) with Beals (1957) and Taylor (1958, for example, is curiously muffled on the subject. Bohannan seems to argue that linguistics does not help in the learning of languages. This brief interchange comes from a milieu that was nevertheless exceptional in retaining an interest in language (see above, p. 36). 4. See also Hjelmslev 1943: 49; trans. 1963: 53; Malmberg 1964: 128; Capell 1966: 39 (where it is not correctly demarcated). Hjelmslev also alludes to other systems: differentiation of siblings by sex and age as between Magyar, French and Malay (see Hjelmslev 1957: 104); differentiation of ‘tree-wood-forest’ between French, German and Danish (1957: 106; 1943: 50; trans. 1963: 54). See also Ullman (1951). 5. Newton (December 1675) thought the seven colours would correspond to the seven intervals in our octave: For some years past, the prismatic colours being in a well darkened room cast perpendicularly upon a paper about two and twenty foot distant from the prism, I desired a friend to draw with a pencil lines across the image, or pillar of colours, where every one of the seven aforenamed colours was most full and brisk, and also where he judged the truest confines of them to be, whilst I held the paper so, that the said image might fall within a certain compass market on it. And this I did, partly because my own eyes are not very critical in distinguishing colours, partly because another to whom I had not communicated my thoughts about this matter, could have nothing but his eyes to determine his fancy in making those marks (correspondence in I.B. Cohen 1958: 192; Turnbull 1959: 373–7).

Berchenshaw wrote of Newton’s system (10 February 1676): That the natural genuine, and true reason of the excellency and fullness of the harmony of three, four, five, six and seven parts, may clearly de discerned by the system of seven parts (Cohen ibid: 226).

6. GPC (1968), s.v. glas, divides the colour referents into (1) blue, azure, sky-blue, greenishblue, sea-green; (2) green, grass-coloured, bluish-green, light-blue, pale-blue or palegreen, greyish-blue, slate-coloured, livid, pallid, pale, grey. A further puzzle could not be elucidated without the structural diagram in figure 1.2: glas, finally, can sometimes have the same referent as llwyd, ‘grey’, ‘holy of clerics’, which is explicable because of their neighbouring positions at the point where the Welsh axes of hue and brightness join. 7. E. Ardener (1954). It was Miss M.M. Green (a linguist and anthropologist) who first mentioned the characteristics of ocha to me. In the functionalist terms of the day I

288 Notes

8.

9. 10.

11.

12.

expressed the ocha/ojii antinomy in terms of ‘attitudes’. A simplified orthography has been used here. The ‘African Alphabet’ renderings, where they differ from those in the text, are as follows: ɔcha, ɔbara ɔbara, uhyɛ uhyɛ, ahehea ndɵ, akwɵkwɔ ndɵ. The theory of a historical order in the succession of types of colour classification comes from Berlin and Kay (1969). In their view, systems may contain (1) ‘black’ and ‘white’ only; (2) ‘black’ plus ‘white’ plus ‘red’; (3) ‘black’ plus ‘white’ plus ‘red’ plus ‘yellow’ or ‘green’; (4) ‘black’ plus ‘white’ plus ‘red’ plus ‘yellow’ and ‘green’; (5) ‘black’ plus ‘white’ plus ‘red’ plus ‘yellow’ plus ‘green’ plus ‘blue; (6) ‘black’ plus ‘white’ plus ‘red’ plus ‘yellow’ plus ‘green’ plus ‘blue’ plus ‘brown’; and so on. Thus Hanunóo would be in phase 3, Ibo and early Welsh in phase 4. The translation presents some difficulties, but ‘the crepuscular light’ is attested in other Celtic sources: ‘terram pulcherrimam obscuram tamen et aperto solari lumine non illustratam’ (Loomis 1956: 165). These cases support Kroeber (1909) and what I take to be the present position of Needham (see Needham 1971). Von Humboldt (1836/1967) is the intellectual ancestor of the field theory, although his ‘mother tongue mysticism’ is not always attractive. It dates in its modern form from 1910 with R. Meyer’s analysis of military terminology. Weisberger, Trier, Porzig, Jolles and Ipsen (who first used the term ‘field’) are the chief names (full references in Ullmann 1951: 152–70; see also Ullmann 1963: 250). (Editorial note) The following example has been removed from the main text here (original, p. xxix): The Banyang and Bangwa are two neighbouring peoples of West Cameroon. Among the former the linguistic term ngo refers to both ‘gun’ and ‘fire’. Among the latter the word ŋwo, borrowed from the former, means ‘gun’, while emɔ means ‘fire’. The implication that the Bangwa first received firearms from the Banyang direction is useful, since it was at least possible, on general grounds, that they received them peoples on the other side. So far, then, linguistic data have suggested a historical implication. However, the Banyang themselves received the gun from the Efik via the Ejagham. In each case, the artifact was exchanged without the Efik word. Yet Efik (and Ejagham) also label ‘gun’ and ‘fire’ by one term (Efik ikaŋ, Ejagham ngon). The Banyang accepted both the gun and (through translation) its identification with fire. This identification did not, as we see, survive the onward transmission to the Bangwa. The problem we now face is the explanation of the different kinds of linguistic contact between Efik (and Ejagham) and Banyang, and Banyang and Bangwa. We note, however, that physical contact between Banyang and Bangwa is interrupted by a high escarpment. The further analysis of these differences and similarities would lie in both linguistics and social anthropology, and in social anthropology for its own sake – not simply for the assistance (if any) this may yield to linguists. (Editorial note) The following text has been removed from the main text here, and slightly amended (original, pp. xxxi–xxxii): The process of the reconstruction of Saussure’s lectures is itself of keen anthropological interest: All those who had the privilege of participating in his richly rewarding instruction regretted that no book had resulted from it. After his death, we hoped to find in his manuscripts, obligingly made available to us by Mme de Saussure, a faithful or at least an adequate outline of his inspiring lectures. At first we thought that we might simply collate F. de Saussure’s personal notes and the notes of his students. We were grossly misled. We found nothing – or almost

Notes

289

nothing – that resembled his students’ note-books. As soon as they had served their purpose, F. de Saussure destroyed the rough drafts of the outlines used for his lectures. In the drawers of his secretary [read ‘secrétaire’ or ‘writing-desk’!] we found only older outlines which, although certainly not worthless, could not be integrated into the material of the three courses (Saussure 1916, 1922 edn: 7–8; trans. 1964: xiii).

So it was that the notebooks of seven students were pooled, and the courses constructed: The problem of re-creating F. de Saussure’s thought was all the more difficult because the re-creation had to be wholly objective. At each point we had to get to the crux of each particular thought by trying to see its definitive form in the light of the whole system. We had first to weed out variations and irregularities characteristic of oral delivery, then to fit the thought into its natural framework and present each part of it in the order intended by the author even when his intention, not always apparent, had to be surmised (1922: 9; trans. 1964: xv).

13.

14.

15. 16.

Thus was compiled and published the Cours de Linguistique générale, and with it was founded the Geneva school of linguistics which Bally and Sechehaye carried on in succession to the master until 1945, dying in 1946 and 1947 respectively. Since then all the sources have been published and critically analysed (Gödel 1957; Engler 1967, 1968). It is somehow appropriate that the Cours and Saussure should co-exist like signifier and signified in one of his own linguistic signs! I give the page references both of the Cours (1922 pagination) and of the translation (1964 edition), but generally quote the latter, despite its detailed inadequacies, so that the flow of the English text may not be broken up by frequent passages in French. Nevertheless, I have amended the translation in various plaices where noted, since sometimes it is seriously misleading, and at least once unintentionally comic. ‘Social anthropology in Britain (to speak only of that country where it has acquired most renown in recent decades) had been inspired by certain general ideas, subtly derived from the early French sociologists, which have had a singular theoretical influence, and much of the progress is to be attributed to them. They are analytical notions such as “transition”, “polarity” (opposition), “exchange”, “solidarity”, “total”, “structure”, “classification”. Now these are not theories but highly general concepts; they are vague, they state nothing. At first sight there is nothing to be done with them, and certainly they cannot be taught as elementary postulates in introductory courses of social anthropology. Indeed, their significance is only apprehended after arduous application to the task of understanding social phenomena; the less one knows about human society and collective representations the less they appear to mean. Yet they have proved to possess a great and perennial analytical value, such that it may be claimed that it is they which are essentially the “theoretical capial” of social anthroplogy’ (Needham 1963: xlii–xliii). Thus, Von Wartburg (1969: 194). Collinder (1968: 183) says that Schuchardt expressed the notion of la coupe verticale and la coupe horizontale of language in 1874. For Collinder: Das Panorama, das im Cours de linguistique générale aufgerollt wird, ist kein getreues Bild der wirklichen Sprachwelt. Dieses système où tout se tient ist nicht einer urwüchsige Landschaft ähnlich; es gleicht viehmehr einem altmodischen zugestutzten französischen Schlosspark (p. 210).

We see here the wrong-headed but common complaint that a formal model does not generate ‘reality’.

290 Notes 17. With reference to Wittgenstein’s Philosophical Investigations (1953), Ullmann says: There is an unmistakable affinity between some of his ideas and contemporary linguistic thought – an affinity all the more remarkable as Wittgenstein does not appear to have been familiar with books on linguistics (1959: 303).

18. For example: Bogatyrev 1931 (which I have been unable to consult) and 1935. 19. I am aware that a ‘panchronic linguistics’ has been several times previously announded (Ullmann 1951: 258–99). It is no coincidence that it should have had an important part in a vision of semantics. Nevertheless, in discussing the diachrony/synchrony distinction of Saussure, we must recognize that this exists at the level of models of formal systems (see Ardener 1971c; also above, ch.16). Similarly, the ‘panchrony’ of Saussure must be realizable at the model level: so far only the transformationalists have credibly demonstrated, if only partially, the possibility of such a model. It is interesting that ‘panchrony’ was generally ignored in exegesis of Saussure by structural linguists, from whom his message was a charter for synchrony, and for whom even diachrony was of lesser import (e.g. see Wells 1947, in Joos 1957). 20. The term semiotic goes back to Locke as ‘the doctrine of signs’. Its use was developed by C.S. Pierce. Morris, Carnap and Hjelmslev helped its modern vogue. Margaret Mean coined the term semiotics for the study of ‘patterned communication in all modalities’ during a discussion at the Indiana Conference of 1962, which is reported in Sebeok, Hayes and Bateson (1964, see pp. 1–7, 275–6). Semiotics thus lies close to Saussure’s sémiologie (closer indeed than does Barthes’s sémiologie). It may be useful to retain semiology to describe the study of semiotics, used as the plural of semiotic. In its turn, a semiotic is a sign system. The coexistence in society of large numbers of semiotics means that any useful description must be made through models of systems, abstractions, ideal systems. 21. McLuhan (1970: 39) speaks of chairs ‘outering’ the human body, leading then to tables, and a restructuring of the human environment. His ‘pop’ usages sometimes curiously, but inadequately, reflect modern ‘structuralist’ trends. 22. The brief discussion of the phoneme included here is obviously selective, and might be omitted were it not that the term as discussed by Lévi-Strauss still has a mysterious aspect for some of his readers. Their questions are not necessarily directly answered by turning to standard works on linguistics. I include the section, aware of its European bias. This may be balanced by consulting Fries (1963) on the Bloomfieldians. He notes that ‘the strong stress upon the procedures and techniques of analysis … did not stem directly from Bloomfield’ (p. 22). In general, see Mohrmann, Norman and Sommerfelt (1963); Mohrmann, Sommerfelt and Whatmough (1963); and Hockett (1968: 9–37). 23. (Ardener’s note, supplemented by text from original, pp. lxix–lxx): Shaw says of Sweet’s polemic Oxford reputation: With Higgins’s physique and temperament Sweet might have set the Thames on fire. As it was, he impressed himself professionally on Europe to an extent that made his comparative personal obscurity, and the failure of Oxford to do justice to his eminence, a puzzle to foreign specialists in his subject … although I well know how hard it is for a man of genius with a seriously underrated subject to maintain serene and kindly relations with the men who underrate it and who keep all the best places for less important subjects which they profess without originality and sometimes without too much capacity for the, still, if he overwhelms them with wrath and disdain, he cannot expect them to heap honours on him (Pygmalion, preface, 1941 edn: 8–9) (see also Jakobson 1966).

Notes

291

Just as German comparative philology was not a good exemplar for anthropologists of the day, British phonetics was not in a position to be very helpful. Sweet, who was at that time, at Oxford, had to fight (as did many of his successors) for a place for a version of linguistics other than that enshrined in the humanities syllabus. His polemical reputation may have convinced our grandfathers that the synchronic problem of language was essentially a technical matter of transcription. Nevertheless, the early anthropologists, with few exceptions, were hardly aware even of the ‘phonetic’ problems. It may well be that the native genius for the exotic expressed through deep linguistic study was almost totally absorbed in the study of the classics, on the one hand, and in imperial duties, on the other. There is every reason indeed to look back past the early pre-emptors of the name ‘anthropology’ to Sir William Jones, sometime President of the Asiatic Society. His famous Discourse of 1789, delivered in Calcutta, which by common consent first clearly asserted the relationship of Sanskrit to the Classical languages and Gothic and Celtic, is otherwise more of an ethnographical disquisition that a linguistic one ( Jones, 1799; no in Lehmann 1967: 10–20). Like Sweet he was a polemical figure: In the parliamentary election of 1780, as a candidate for the University of Oxford, his detestation of the American War and of the slave trade were too strongly expressed to be agreeable to the voters, and he was forced to withdraw from the context. In the same year he failed to secure election as Professor of Arabic in the University for similar reasons (Firth, J.R. 1957a: 161).

24.

25. 26. 27.

Certain precursors of Koelle (1854) in Africa might be mentioned for their ethnographic as well as their linguistic contributions. Latham has been mentioned by Hymes (1964: 3), to whom may be added Clarke (1848). Jarvie (1963) no doubt was justifiably reacting against this. The present writer reviewed Jarvie’s book in a critical vein (Ardener, E. 1965b) because it seemed unaware of the important developments outside the Malinowski tradition, to which I refer. My own concern is with the excessively long time that recognition of the obvious changes in the climate of thought in the subject, and in the skills required, has taken to percolate through. We are virtually forced to fall back on ad hominem explanations, in a small subject like social anthropology. Malinowski’s impatience with contrary opinion was accepted as fair exchange for scientific advance. His failure to recruit and keep many students of a sceptical bent from the mainstream of the European tradition must surely account for his neutral intellectual legacy. The death of Nadel (in Australia) was a loss. Whatever Radcliffe-Brown’s faults, the existence of his works enabled the dissidents from Malinowskian anti-intellectualism to find a temporary alternative stimulus, if only through attack (e.g. Evans-Pritchard and Leach). For some reason, women anthropologists in the Malinowski tradition also maintained a lively presence, and continue to do so. (Editorial note) The text from the beginning of this paragraph to the end of the section (p. 40), has been transposed from its original position, pp. lxx–lxxiv of the original. I accept Tambiah’s point (1968: 203) that Malinowski’s views were at times closer to those of Evans-Pritchard than we might expect. (Editorial note) The following text has been taken from the original at this point: Similarity of terminology can, however, bring confusion. The Chomskyan system is characterized by precision of expression, where Lévi-Strauss is programmatic. Detailed point-by-point comparisons are not to be recommended. Thus, the antinomy

292 Notes between ‘deep’ and ‘surface’ structures occurs in Lévi-Strauss, as in Chomsky (‘Ainsi l’analyse structurale se heurte à une situation paradoxal, bien connue du linguiste: plus nette est la structure apparente, plus difficile devient-il de saisir la structure profonde…’ – 1958). The ‘deep structure’ of Chomsky, as applied, for example, to sentences with ambiguous surface structures, is revealed through clearly stated sequences of transformations within one model, say that of English (see Hymes, D. 1971: 53). The ‘competence’ of the transformationalists can, of course, be seen as a kind of generalized ‘deep structure’, or a generalization of the base rules for the set of all deep structures of a language. The semantic component of the language is tied to the deep structures. Lévi-Strauss’s deep structures in the analysis of myth, on the contrary, are derived from units already ascribed a conventional meaning. The transformations of inversion, sign reversal, and the like, operate to demonstrate, through the differences or contradictions in surface meaning between related myths, the nature of the mythlogic itself (1964, 1966b, 1968). Compared with Lévi-Strauss, Chomsky is (paradoxically for a proclaimed ‘rationalist’) more ‘empiricist’ in style. There is generally recognized to be a difference in tone and aims between the Chomsky of before Aspects of the Theory of Syntax (1965) and the subsequent Chomsky. Pre-Aspects Chomsky still shows signs of his explicit concern with exact models: as his system was received at the time, it appeared as a rebuttal of simple ‘left-to-right’ generated models of language, of a ‘finite-state’ type. Such models derived ultimately from the original work of Shannon in communication theory (1948), and with Chomsky the main wave of direct application of that theory to language studies. The finite-state model is expressed in ‘box-and-arrow’ form as in computer studies: so are the alternative, more powerful models of Syntactic Structures (1957), and of Current Issues (1964). Post-Aspects Chomsky has turned from prime and only concern with the output of his model – the corpus of utterances – to a more difficult problem, which in a sense was left over from the destruction of the finite-state model with its implied statistical probabilities. This was: how does a child acquire the model of competence (the generative grammar) for his language? From this point of view, one can describe the child’s activities as a kind of theory construction. Presented with highly restricted data, he constructs a theory of the language of which this data is a sample (and, in fact, a highly degenerate sample, in the sense that much of it must be excluded as irrelevant and incorrect – thus the child learns rules of grammar that identify much of what he has heard as ill-formed, inaccurate and inappropriate). The child’s ultimate knowledge of language obviously extends far beyond the data presented to him. In other words, the theory he has in some way developed has a predictive scope of which the data on which it is based constitute a negligible part (1969: 63).

Chomsky maintains, therefore, that the organism has ‘as an innate property’ a structure that will account for this mode of acquisition – put frivolously by McNeill (in Lyons and Wales 1966: 116): ‘Metaphorically speaking, a child is now born with a copy of Aspects of the Theory of Syntax tucked away somewhere inside.’ Chomsky has turned to Descartes, Leibnitz, and the rationalist philosophers of innate ideas for a philosophical charter for his approach; these thinkers being opposed to Locke and the empiricists, whose most extreme descendants are taken to be the psychological behaviourists. The philosophical basis of Chomskyan mentalism is a subject of disputation. In effect he gives an ontological status to what behaviourists see as a ‘capacity’ or a ‘capability’ for language. It would be out of place here to attempt a detailed discus-

Notes

293

sion (see, for example, Cohen 1966: 47–56; Hook 1969; Lyons and Wales 1966; Lyons 1970). It is sufficient to note that for our purposes that the Chomskyan system began with a transformational generative grammar, with the characteristics of a well-defined system. Is it ‘present’ as an analogue of a full write-out of a transformational analysis, or is it expressed in some other form – in the way that, for example, the cogs of a clock only indirectly enact what we know about the movements of the sun and earth from the Newtonian Laws of motion. 28. (Editorial note) The following has been taken from the main text here (original, pp. lxv–lxvii): We can transpose these remarks in terms of Hockett’s (1968) critical review of Chomsky’s theory. He sets up a summary formulation of Chomsky’s system (at 1965) in nineteen points, which were largely endorsed by Chomsky himself. The case is then argued with considerable skill and documentation through the volume that no physical system, and in particular language, is well defined. His attack is upon the original programme, therefore, for its arbitrariness. But it is by now self-evident that a model of a formal system (which is well defined) says nothing of the ‘well definition’ of the natural order. Truly generative models are models; they are, of course, less than the phenomena they help to explain. Nevertheless, Hockett’s criticisms confirm in a different way our awareness of the ambiguity of the Chomskyist movement. We have said that what seems to have begun in 1957 as a conscious application of model-building to postwar structural linguistics – its ‘generative’ and ‘transformational’ terminology is quite clear on this point – has now outgrown its early phases. The well-known capacity of Chomsky for fresh and creative development has disguised the fact that his total system is no long itself well defined. Hockett’s critique thus paradoxically falls on two contradictory grounds: If Chomsky’s model of a system were formal and thus well defined, it would not be counter to the system to say that language as a natural phenomenon is not well defined; and, in so far as Chomsky’s system is not well defined, Hockett has no criticism. In fact, Hockett’s discomfort probably derives from an intuition of the contradiction between the formalism of the transformationalist terminology and the lively and speculative accretions of the transformationalist world-view. In my paper (1971c; see above, chapter 16), I suggest that the neogrammarian model of comparative philology was totally generative. Its basic ‘inextensibility’ should illuminate both the power of truly generative models and the dangers of forgetting their functions. The dissatisfaction of Chomsky with his earlier aim, and the extension of the search to meaning and beyond, have been highly productive, but no total formal system has yet been set up for this. Chomsky is both Bopp in level of achievement of his programme, and Brugmann in his search for precision. It is this desire for a totally formalistic presence that falsely sets him apart from Lévi-Strauss, who disguises his own formalism in literary metaphor. These are presumptuous remarks coming from a social anthropologist. The reason why I feel impelled to make them is precisely because the generous aims of the transformationalists and those of the new social anthropology are, within their disciplines, rather similar. It is interesting that their thinking covers some of the same ground. It seems sometimes that the transformationalist approach would benefit from a more careful consideration of non-behaviourist social anthropology, which would in turn no doubt gain much from the encounter. The kinds of criticism made of the latter by the survivors of Malinowski (who occur in all age groups) resemble those made by the

294 Notes

29.

30.

31.

32.

American post-Bloomfieldians of the transformationalists. The older social anthropology finds the newer variety ‘incredible’, precisely for its apparent indifference to a particular positivist view of the natural order. The neo-anthropologists are also aked to provide the equivalent of ‘discovery procedures’, and they too seem to regard the aim as only of subsidiary interest, although in fact large amounts of ‘empirical’ data have been analysed (I refer specifically to their work in kinship and symbolism). They too began with the establishment of elegant and simple models of formal systems. They too have grown out of these earlier aims in the direction of theories of wide-ranging scope. They too are prepared to consider the existence of universals, beyond the scope of ethnographic solipsism. The only critique of stylistic interest comes indeed from the American Marvin Harris (1969), a ‘plague-on-all-your-houses’ cultural materialist. He speaks with reluctant if ironical admiration of ‘professional idealists, as distinct from eclectic American amateurs who have rubbed shoulders with logical positivism and behaviourism too long to know how to really get off the ground’ (p. 505). See also below, note 32, and chapter 2. I use the nonce-term ‘neo-anthropology’ to cover post-functionalist movements of a creative type, not all of which would accept the term ‘structuralist’. Some of them clearly have a good deal in common in subject matter with the so-called ‘new ethnography’ of the United States (Sturtevant 1964). Leach, a senior exponent, still ambiguously claims to be a ‘functionalist’. The newanthropologists are recognizably different in interests and style from the majority of the British profession in characteristic ways, but since they do not necessarily agree even with each other (and may refuse to be linked together) they lack the earmarks of a school. In this respect they have remained isolated and divided vis-à-vis the relatively united ‘Palaeo’ group. Barnes (1963), Freedman (1963), Goody (1966), Maquet (1964), Worsley (1966) made valuable points. Leach may have inadvertently set off the fashion (1961), but his language was not properly understood. As long ago as 1954 (pp. 92–3) Leach wrote, in reviewing Pocock’s translation of Durkheim (1951), of a ‘general revival of interest in ideas and ideals for their own sake, in contrast, for example, to the extreme empiricism of Malinowski’. In the 1960s hardly an issue of the main international professional journals failed to contain some discussion of the views of Needham and his pupils. Hardly a literary journal lacked some exegesis of Lévi-Strauss. Douglas (1966) did much to draw the attention of social anthropologists in general to some of the important themes. The activity of Evans-Pritchard’s colleagues and pupils (other than Needham) has been mentioned. Yet the significance of none of this was truly taken note of until the end of the decade by the representative professors of the subject. For a justified pessimism, see now Needham (1970). (Editorial note) At this stage in the original text, the section entitled ‘The Present Volume’ began. The last few paragraphs of this section, which have an ‘end of piece’ feel to them, have been given the title ‘Inarticulate Rationalities’, and kept as a conclusion to the present chapter. Of the remainder of ‘The Present Volume’ section, some, where it seemed appropriate, has been incorporated either into the main text or into its notes. What follows in this note is a further piece from ‘The Present Volume’ which, while it is specifically concerned with papers not presented here, also contains characteristic ideas and asides which are of some importance, concerning socio-linguistics, transactionalism, linguistic typologies, pidgins and creoles, diatypes and so forth:

Notes

295

Professor Hymes shows, in his comprehensive review (Hymes 1971), what is meant in sheer scope and method when we refer to American anthropological linguistics. His essentially undoctrinaire writings provide a mine of varied material, to which all interested in ‘socio-linguistics’ from a wide range of disciplines are indebted. His book of readings alone (1964) is in itself, because of his commentaries and scholarship, an original work. There is, nevertheless, a special consistency of view that emerges in his contribution below, as in other writings (1962; Gumperz and Hymes, 1964), which derives from his firm hold upon the ‘ethnography of speaking’. The idea is effectively a realization of the social anthropology of la parole. Hyme’s encyclopaedic approach may superficially appear to be irreconcilable with those deriving from the continental linguistic schools, for which this introduction has argued equally consistently. This impression would be mistaken. Hymes directs our attention to the plane at which language is generated in society – in this respect he is close to what many working linguists and anthropologist ideally demand of a ‘socio-linguist’. Any analysis of material acquired from this standpoint has, however, to be organized through models, and through less conscious organizing systems, set up by speakers and actors, by social anthropologists and linguists and socio-linguists – or by ethnography of speaking. The fecundity of Professor Hymes’s insights comes from the fact that so many structural statements made from different logical premises must meet at the ‘plane of articulation’. Whether it be Chomsky’s ‘performance competence’, or Bernstein’s ‘restricted : elaborated codes’, they are all open to revision, illustration, confirmation, or comment at this meeting-place. The plane of the ethnography of speaking may thus be placed diagrammatically at right angles to their plane, in the same conceptual relationship as syntagm to paradigm. As there is a choice of ‘paradigms’, so there is a choice of ‘syntagms’, although this is commonly less clearly realized (for the use of the terms paradigm/syntagm, see above, chapter 2; also Ardener, E. 1971b, p. lxxxviii, note 26). Professor Hymes exercises this choice with great freedom. There is hardly a branch of linguistics and social anthropology in which the ethnography of speaking may not appear. Thus Colson and Gluckman write on gossip, but where is the gossip delineated? ‘Ethnographic accounts are rife with terms that in fact denote ways of speaking, though they are not always recognized to be such’ (Hymes 1971: 77). Hymes’s basic approach is formally consistent, despite the variety of paradigmatic systems it cross-cuts, which makes his paper in itself an introduction to socio-linguistic writings. Had post-war functionalism developed Malinowski’s own linguistic insights, it might well have extended ethnographically in the plane of Hymes’s own interest. ‘Context of situation’ itself belongs to that syntagmatic plane. On these grounds Professor Pride’s rejection of the view that contexts of situation are necessarily ‘below the level of a general abstract theory’ (Pride 1971: 96) seems particularly convincing. As a socio-linguist coming from the direction of linguistics, he places himself at the plane of linguistic transactions: thus linking it with Barth’s model of social anthropological analysis. The interest of Barth’s model is that it too falls in the syntagmatic plane, with a clear definition of the elements of the model: the notional ‘transactions’. In social anthropology it is sometimes complained of as ‘rigid’, ‘partial’, ‘mechanistic’ – all, as we shall agree by now, the honourable stigmata of a model of a formal system. If fully articulated it could probably be shown to be truly ‘generative’ in the formal rather than the metaphorical sense. Gluckman’s interchange

296 Notes with Paine on gossip (Paine 1967, 1968; Gluckman 1968), where Paine takes a ‘transactional’ view, is based in part on a failure to see that syntagmatic models are not in the same plane as paradigmatic models (Gluckman’s ‘transactions between individual persons cannot explain institutional structures’, is thus a truism). In social anthropology, ‘transactional imagery may be described as a part of the ‘highest stage of functionalism’. That is: a functionalism become aware (or about to become aware) that the field of behaviour or action, even when arbitrarily isolated from the ideological programme that determines its meaning, must itself be structured by the observer before it can be ‘observed’. The interest to the socio-linguist of this approach pinpoints certain differences in the histories of anthropological and linguistic enquiry. Functionalist anthropology was (in loosely Saussurean terms) concerned with the social as parole. The most recent developments have led social anthropology to be concerned with the social as langue (the diachronic and synchronic versions, as well as the ‘structural’ and ‘transformational’ views, differed until very recently only in emphasis in this particular respect). It is natural that now socio-linguists, in seeking to study language as parole, should either use functionalist approaches or find those of functionalist social anthropology converging on the same area. The relative lack of formalism in the old functionalist world-view will undoubtedly be amended by this, but there is still an uncertainty in the newer developments. ‘Theory’, to the functionalist, has long meant the confusion of statements based on models of a syntagmatic type (to which the stress on observation binds him) with paradigmatic statements. The confusion became the worse confounded because the truly paradigmatic statements of writers like Evans-Pritchard were interpreted as syntagmatic ones. I have already referred to the notion of ‘opposition’ in The Nuer (paradigmatic) being reinterpreted as ‘conflict’ (syntagmatic). Douglas (1970a: xiv–xxii) now shows how Witchcraft, Oracles and Magic, which was about ‘cognitive structure’ (paradigmatic), was reinterpreted as about ‘social control’ (syntagmatic). I have developed some of these points elsewhere (see above, chapter 2). It is necessary to touch on them lightly here however, in order to suggest that there are two approaches to socio-linguistics which parallel those of functionalist and post-functionalist (‘neo-anthropological’) social anthropology. They may be expressed diagrammatically as follows: Paradigmatic Social Anthropology A (‘structuralist’ of Lévi-Strauss; ‘neoanthropological of Needham, Leach, Douglas) Syntagmatic Social Anthropology B (‘functionalist’, ‘neo-functionalist’; transactions, networks, etc.)

Socio-linguistics A

Socio-linguistics B

Socio-linguistics as generally described is essentially a Socio-linguistics B. It is that of Malinowski, of Hymes in his most characteristic phases, and of Pride at his most analytical, as well as of Whiteley, Denison and others. Socio-linguistics A is essentially the approach developed above: which some may possibly consider only an epistemological raid from Social Anthropology A, for the writers who have set it out from the linguistic side are not normally thought of as socio-linguists. The names that would be cited (Saussure, Jakobson, Hjelmslev, the later Firth, the German semanticists, Sapir, Whorf and their American exemplifiers) are leading names in general linguistics. Furthermore, the transformationalist approach (for some the very antithesis of a

Notes

297

socio-linguistics) has its nearest anthropological congener in Social Anthropology A. At the 1969 ASA Conference much debate was generated between A-type social anthropologists and B-type social anthropologists, as well as between A-type social anthropologists and B-type linguists and socio-linguists. Some B-type social anthropologists rejected linguistics of any type. In such circumstances of confusion it is wisest not to give hostages to new labels, and certainly not to new ‘disciplines’, built an any outmoded nineteenth-century style: with their nation-state apparatus of buildings, professorships and degree structures. It looks likely that Socio-linguistics B will for some time to come appear to be the major bearer of the label ‘socio-linguistics’, which Socio-linguistics A will be apprehended as a kind of social anthropology, a kind of linguistics, or a kind of philosophy, according to the point of view of the practitioner. The existence of multilingual situations has long provided a field for the comparison of models of language and of society. The loosening of rigid terminological distinctions between ‘language’, ‘dialect’, ‘register’, ‘code’, and the like has greatly extended the range of relevance of such studies. We now see that the problems presented by the coexistence of several varieties of ‘one language’ are not different in kind from those presented by the coexistence of ‘different languages’ in a multilingual speech community. ‘Multilingual’ situations provide a genuine field for empirical work in which both syntagmatic and paradigmatic approaches may be brought to bear. So far most of the work that has been carried out has been of a syntagmatic type, but there is no lack of hints for ‘paradigmatists’. Dr Tonkin has attempted ‘a social history’ of the succession of pidgins, ending in Pidgin English, on the Guinea coast (see Tonkin 1971). This is a ‘diachrony’ of the social situation not of the linguistic situation – for which there is only scattered phonological or other documentation. Yet such studies bring into doubt certain supposedly ‘linguistic’ typologies, by revealing the sociological assumptions that lie behind them. In particular, the distinction creole/pidgin disappears. The supposed difference lay essentially in the presence (in creoles) or absence (in pidgins) of monolinguals in the language type, and of transmission between generations. More deeply, perhaps, it derived from conflicting analytical reluctances: on the one had, from a reluctance to award the full status of ‘a language’ to what might be thought to be ‘invented’, or ‘limited’, jargons; and, on the other, from a reluctance to deny it to forms of speech which had become mother tongues to some, if not all, of their speakers. Pidgins have been a stumbling-block to all the great schools of linguistics. Both pre-Saussurean and post-Saussurean linguistics have usefully operated with models in which the basis of the diachronic or synchronic system was the single language. We have seen that the treatment of languages as if they were well-formed systems has led to great advances. An unsatisfactory treatment of pidgins was part of the price paid for those advances, for most of the efficient models of language have ‘snipped out’ pidgin phenomena, with the very shears that demarcated their field of operation (see Ardener, E. 1971c: 222; above, chapter 16). We may note, further, that West African Pidgin English has in reality always been a part of a multilingual or plurilingual context in combination with English or with one or more African languages. Where it has become supposedly ‘creolized’ it has become in fact diglossic – primarily with forms of standard English. ‘Available evidence suggest that most of Africa has been multilingual for a long time, even if the domains of such behaviour were characteristically restricted, e.g. to trade or hostility’ (Whiteley 1971: 22). Dr Denison’s detailed analysis

298 Notes of the linguistic diatypes of Sauris – a ‘trilingual’ community in the Carnian Alps, in which a variety of German coexists with Friulian and Italian – is thus concerned with a situation that is also common in Africa, and throws light on other possibilities in the past of Europe (see Denison 1971). One may perhaps note in socio-linguistic thought so far a certain prejudgement in the attribution of a ‘social status’ to certain varieties of language. We should not lose sight of the fact that much work in this field is still in the classifying phase. To simplify a little: at the first stage it may well appear intuitively (or from statements and observations) that certain diatypes have ‘high’ or ‘low’ status. From there we may proceed with C.A. Ferguson (1959) and Fishman (1968b) to ‘H’ and ‘L’ divisions of diatypes. We may then appear to discover that H diatypes are being used in L social contexts, or the like. Theory here is now at the dangerous phase: the original source of the differentiation (which was an ad hoc assessment of the status of contexts of sue) begins to pass out of recognition, and we are all set form many comfortable years of exemplification of ‘H-ness’ and ‘L-ness’, until the basis of the typology is again revised. As Dr Denison rightly points out, the Sauris situation, in any event, appears to require an ‘M’ diatype (for ‘middle’). The various diatypes are in effect in opposition to each other. We first need a model of the structure of these oppositions in the whole linguistic context. We require, likewise, a model of the structural oppositions with the society at large. We may attempt to map these two models (or sets of models) upon each other; it may be, as we hope, that there will be some transformational links between them. Thus, we avoid prejudging the basis of the mapping; for the term ‘status’ itself ceases to have any particular privileges in such an analysis. Furthermore, the possibility of negative and inverse transformations (perceived as ‘contradictions’ or ‘exceptions’) becomes a normal expectation and can be examined as such. The language of paradox and of incipient nominalism (some H settings have L diatypes) gives way to the language of structure – here derived from social anthropology, despite its resemblance to that of transformational linguistics. Denison’s material illustrates this well with the oppositions Italian/Friulian, Italian/German, Friulian/German, Friulian of Sauris/Friulain of Udine, German of Sausis/standard German, on the linguistic side, together with the no doubt more delicate ones that he is able to observe. On the social anthropological side one can already detect some crude relevant oppositions – Udine/ Sauris, rural/urban, home/school, adult/child/ architect/foreman, foreman/workman and so on. The rich material in Denison 1968 (e.g. pp. 584–5), would suggest many more. One might suspect a temporal structure also, cyclical perhaps, in which the tourist season (more Italian spoken?) may be opposed to the rest of the year. It is also possible that the young, who speak more Italian, will not retain this tendency when their social prospects are firmly assured. This point is worth making because of the question of prediction. Is Sauris German dying out (Denison 1968: 589)? To take an example: it has been said through most of our lives that congregations in Soviet churches consist of persons over the age of fifty, and that that organized religion is ‘dying out’. It is clear, however, that the fifty year olds of today were adolescents in the 1930s. The congregation seems to have acquired a pattern of recruitment by age. The possibility of such a structure over time means that the opposition youth/age may outlive the present occupants of the ‘age-slots’ concerned. All of these comments are highly speculative, and take the excellently documented case merely as a convenient example. A full so-

Notes

299

cial anthropological study would start from hints such as these, and even if not expressed in this terminology would proceed to the further finer distinctions that a field-study would yield. The division H and L would be subsumed in such an analysis, while the middle term M could be dispensed with. The results might not turn out to look very different but we should have avoided an ad-hoc terminology (an avoidance that is, indeed, part of the spirit of Ferguson’s original analysis – see Hymes 1964: 431) (see Ardener, E. 1971b: xxxviii, note 27, for further comparison.) These remarks are relevant to some social anthropologists who find it difficult to visualize how a structural analysis in the newer sense can be the subject of ‘empirical’ study (I am referring, of course, to a social anthropological study without overt linguistic aims.). The Sauris-type situation could never, however, be easily handled by a social anthropologist without Dr Denison’s linguistic skill. It would be possible to imagine a social anthropological study, nevertheless, which produced models of Sauris society that would assist the linguistic. The ‘diatypes’ are, we may guess, symbolically realized at more levels than one. 33. To contrast a statistical study with a symbolic study of the same ethnographic phenomena: see Ardener, E. 1962 (marital instability) and 1972b, above, chapter 4 (symbolism and women); and Ardener, E., Ardener, S. and Warmington, W. (economics), and Ardener, E. 1970c (see above, chapter 15) (belief). N OT E TO CH A P T E R 2

1. Saussure himself actually used the term ‘série-associative’ for the paradigmatic relations (1922: 170–84). N OT E S TO CH A P T E R 3

1. To distinguish them from the distant Batanga of the South Cameroon coast. 2. Under the name of Romby – Ardener, E. 1968, 1972c. 3. For a full-scale treatment of this problem see chapter 7. N OT E S TO CH A P T E R 4

1. This paper was read at Dr Kaberry’s seminar in University College London in late 1968. In presenting it for Dr Audrey Richards’s festschrift, I acknowledged my debt to her for the main part of my early anthropological training. Her astringent humour and basic open-mindedness are qualities that I have respected ever since. I also thanked Dr Jean La Fontaine for her appreciative remarks on the paper, and for entering into the spirt of the analysis in her comments as editor (see La Fontaine, ed. 1972). 2. This version was given in 1929 by Charles Steane, a Bakweri scholar, to B.G. Stone (MS 1929). 3. Moto, eto, and ewaki are the ordinary words for ‘person’, ‘rat’, and ‘ape’. Mojili or mojele is to the coastal Bakweri a spirit. For inland Bakweri his name is a euphemism for ‘ape’. It is likely that the term belongs to the animal world, but is borrowed from the fishing peoples. Possibly it is the manatee. 4. When the term is used in isolation the spelling liengu will be used (not, that is, the ‘Africa’ alphabet spelling liɛngu, nor the occasional spelling with orthographic subscript liengu).

300 Notes

5. 6.

7.

8. 9.

10.

The belief appears to be of coastal origin. There it is concerned with men, fishing and the dangers of the deep. This paper is concerned with the liengu belief as utilized by the Bakweri. Elements of content are differently combined even between the coast and the mountain. Ittmann’s rich material (1957) is to be used with caution because it combines several different systems. The pidgin translation for water-spirt is ‘mammy-water’. The ‘mammy-water’ has wide currency in West Africa in urban contexts. The ambiguity of the position of women in African towns makes this secondary elaboration of the belief very appropriate. See also Ardener, E. (1956). Various forms cited by myself (1956) and Ittmann (1957) are closer to ‘fattening room’ seclusion rites of the Cross River area in form and content. Their assimilation to the liengu belief is explicable because the latter belief most clearly organizes the women’s world-view for the Bakweri. Here is a subtle case of identical content yielding different meaning. The Duala mer-people hate European objects, but the maengu are often male. There they symbolize men’s domination of the deep; they particularly detest paper (conceived of as the bible). For the liengu language, see Ardener (1956) and Ittmann (1957). It is a code calqued upon Bakweri with vocabulary from various sources. Dr La Fontaine commented on this paper that men plus wild = death, destruction; women plus wild = agriculture, fertility. She, a woman, thus expresses that faith in the female civilizing mission shared by so many reflective members of her sex! For some unresolved puzzles of a new woman fieldworker see Bovin (1966). For a resolution through literature see Bowen (1954). N OT E S TO CH A P T E R 8

1. The paradox is that the studies of the cultural relativity of ideas of ‘women’ should seem to increase rather than reduce the tendency to see this as a ‘women’s’ subject. 2. I think the date was 1966. 3. For those familiar with this terminology, the following diagram will suffice: Dp

mp D = dominant

model of specification

m = muted S1

S2

S3

The ‘reality’ configurations (s-structures) are generated from p-structures. Dominant p-structures generate s-structures relatively directly. Subdominant p-structures generate only indirectly – through the mode of specification of the dominant structure. 4. Since the ‘wild’ is always ‘symbolic’ it is not surprising that women do sometimes see themselves as part of it (cf. Ortner 1973). Her approach, by a greatly different route, complements mine.

Notes

301

5. I should like to add here that I find myself in general agreement with much of Mathieu’s own position. In misreading mine she has really done my paper too great an honour: she has judged it by the standards of length of a monograph. To push through an argument in a short paper one begins with certain common-sense categories in order to dissolve them. The terms are redefined between the beginning and the end. I accept then that there are ‘generalizations’. But I do not think that the many examples of ethnographers to whom the women have ‘spoken’ (among whom I am after all one!) touch the central point that within social anthropology ‘no one could come back from an ethnographic study … having talked only to women and about men, without professional comment and some self-doubt. The reverse can and does happen constantly.’ I have not replied to all points, not because they are without interest or are too compelling, but because the central charge of biologism is so improbable that it distorts all her presentation. As a final exemplification of it I quote from her last words: ‘Vouloir rendre la parole aux classes inarticulées en allant rechercher “aux niveaux les plus profonds” du symbolisme ce que, tels des schizophrènes, ils tenteraient d’exprimer, présente le même danger en ethnologie que l’explication constitutionnaliste de la schizophrénie en psychiatrie…’. It is astonishing that here the Laingian approaches to the ‘meaning’ of schizophrenia, with which my approach is most comparable, are interpreted as ‘constitutionalist’ psychiatry. N OT E S TO CH A P T E R 9

1. Particularly in chapters 5, 7 and 8. There are further details in chapters 2, 3 and 6. 2. ‘That the founder of a new literature and the founder of a new state should both be saturated in the legends of Cú Chulainn was something peculiar enough to suggest that both of them were being used by some force outside themselves. It is true that both were fortunate in their periods. Yeats because his coincided with a general disillusionment regarding the values of Victorian life and literature, Pearse because the time was ripe for a whole series of national independence movements.’ (O’Connor 1967: 7). 3. Feadag and Gobag also mean ‘whistling’ and ‘biting’ of winds, as in the saying ‘Feadagan e tuilleadh gu Féill Pàdruig – whistling and biting winds on to St Patrick’s day’. But we also find A’Ghobag explained as the day before Féill Brìde, Candlemas day (2 February) which itself gives its name to February month in some areas, including Ireland (Nicholson 1881: 414; Dwelly 250: 511). 4. This period has parallels in the Welsh y marwfis (‘the dead month’) and y mis du (‘the black month’), with parallels too in Welsh scholarly attempts to fix once and for all whether they referred to ‘January’ or ‘December’ or even ‘October’ (Parry 1939: 40– 2; Richards 1950: 204–5). There are exact Gaelic equivalents and Am mios dubh (‘the black month’) is ascribed by the undaunted Nicholson to November, and Am mios marbh (‘the dead month’) to December/January. 5. Faol ‘wolf ’ is assumed here to be the etymology lying behind Carmichael’s adjective ‘ravenous’. The other words in brackets are the literal meanings of the ‘month’ names – and the interpretations resemble Nicholson’s. 6. The rather ad hoc names for the summer months are similarly related to the fact that Samhradh was reserved for canonical ‘summer’. The terms for ‘spring’ (An t-Earrach)

302 Notes

7.

8. 9.

10. 11.

and ‘autumn’ (Am Foghar: ‘harvest’) look rather out of place, the one being already preempted by the Am Màrt or An Céitean periods, and the latter by the vague An t-Sultainn, and by the important An t-Samhainn, ‘Hallowtide’. January, February, March, April, July and December are named after the Latin months or Christian feasts. The three old ritual epochs give their names to May (An Bhealtaine), August (mí na Lúnasa), and November (mí na Samhna). June was named ‘mid-summer’ (an Meitheamh) and the vague September and October named ‘mid-Autumn’ (Meán Fhómhair) and ‘end of Autumn’ (Deireadh Fómhair). Quoted in Ornstein 1975, which is a useful discussion of some of the ‘psychological’ aspects of time. The world that Lowes began to explore was that of Coleridge’s mind, but Lowe’s essentially modern contribution was to locate that mind in its textual and historical sources, without reducing that mind to its sources. Elsewhere we recognize the problem of the anthropologist in the study of events, when he says: ‘above all…, it may not be forgotten that we are disengaging the strands of an extremely complex web. It is, however, one of the limitations of our finite minds that we are compelled to consider in succession things which are in reality simultaneous’ (1927: 55). Mr Joseph Deacon’s remarkable autobiography is now published (Deacon 1974). It took twelve months to dictate. The simple definition space works of course as a static diagram. For definition of conjugality, for example, let us define conjugality by criterion ω (say ‘sexuality’) specified by non-sexual parameters, for example α´, ‘duration of relationship’, and α˝ ‘jural rights’. If we reduce the α to vanishing point, then the defining parameters become sexual (ω´ and ω˝) and the previous parameters are collapsed in the new space. There is no reason why this transformation in a definition space should reality homologues, but it happens that the resulting tranformation in this case may resemble definitions of prostitution or the like. Parameter collapse – marriage stablity α´

ω marriage

girl concubinage friends

ω = sexuality α = other defining features

ω´

α prostitution

Notes

303

N OT E TO CH A P T E R 1 0

1. See Barltrop 1975: 48–50. This interesting case ran as follows. In 1914 a member of the Peckham branch, Mr Wren, violated the SPGB’s ‘Hostility Clause’ by signing a petition to a Liberal MP. On orders from Executive Committee (EC) the Branch expelled Wren by fourteen to seven. The minority of seven were then expelled (by a poll of all party members) by 103 to 27. The 27 were then pursued. Ten members voted against the final expulsion and EC demanded that these also should be expelled, but branch secretaries and members were becoming elusive and the matter petered out in 1917. Barltrop asks (p. 190) ‘what is there to be said for persistent membership of a small party whose electoral returns are absurdly small, whose influence is restricted, and which will not change its mind? Above everything else, the SPGB remains the only custodian of the vision of socialism.’ N OT E S TO CH A P T E R 1 1

1. For an account of the traditionally weak relations of British social anthropology to formal linguistics, and the particular nuance of its concern with language, see chapter 1 and Henson 1974. For a detailed discussion of the search for meaning in social anthropology, see Crick 1976. Bibliographies embodying my intellectual sources will be found in my other publications (see now also Hastrup 1978). The West African examples derive, unless stated, from my own field work. 2. It will be obvious that this section is not addressing the wider questions of whether language-like symbolizations can occur among non-human primates, whether primate communication occurs, and whether these things can be studied. It is not argued that the usages I discuss are typical of primatological research, nor in any case do I use them for any but illustrative purposes. 3. For the development of the term ‘behaviour’ in scientific usage see chapter 6. The term always meant ‘socially-ordered activity’. Its appearance in scientific use in the 1850s represented a demonstration that nature was orderly. ‘Behaviour’ was subsequently also used to denote an activity for which the demonstration of orderliness was only an aspiration – thus the paradox of ‘random behaviour’, and the desocialization of the term. 4. See Ellen and Reason (1979) for recent contributions, and bibliographical references. Durkheim and Mauss (1963) was a founding study. 5. The Ibo live in South-Eastern Nigeria, the erstwhile Biafra, and number some 7 million. The language is conventionally rendered Igbo. This spelling is not used here. The gb phoneme is not a labiovelar but an imploded bilabial. The spelling Ibo, although strictly non-phonetic, thus leads to a less misleading English pronunciation than Igbo, which is often mispronounced Ig-bo. Ibo is also the normal Nigerian-English spelling, in which certain vowels are distinguished by subscript dots, nor is it an IPA rendering. In the examples, nye has an open e, the vowels of mkpisi are both i with a subscript dot (a close e), okhu has undotted o (approximately as in not) and a dotted u (roughly as in put), ukwu has dotted u in both syllables. Tones are not marked here. Aka and ukwu both have two high tones. See also Green and Igwe (1963). 6. When English butchers classify the carcasses of (dead) animals a series of separate usages appears, including a distinction between front and rear limbs. Thus:

304 Notes Beef shin leg Mutton shoulder leg Pork hand leg The shift of human categories is very notable in butchery terms. The ‘hand’ of pork is not strictly like an aka of dog, because the ‘hand’ of pork excludes the ‘pig’s feet’ or ‘trotters’ which are attached to it. 7. The body has different categories for different classes of classifier. The “polite” body has many fewer subdivisions than the “sexual” body. The “medical” body may have more divisions than either and can be ambiguously ‘polite’ or ‘sexual’ (see chapter 1, p. 14). Victorian polite ‘throat’ included much of the female trunk, and ‘limb’ replaced ‘leg’. Traces of such phenomena can still be discovered in the history of body labelling. 8. Callan (1970: 43) argues that the categories ‘greeting’, ‘rank’, ‘hierarchy’, ‘dominance’ and the like are applied to animals by what she calls the ‘Aha!’ reaction: ‘what one tends to get is a double thrill of recognition… – “aha! here’s an animal being territorial (or dominant)”, “aha! human beings are territorial (or dominant) as well”… Concealed premises: that we know and can recognize territoriality or dominance in animals without having drawn on our own workaday model of society in the first place.’ 9. This presentation is not the occasion to take the matter further, but a hint of the dynamism of the processes concerned may be further exemplified. Nwa nna (see above) is contrasted with nwa nne (‘mother’s child’ or ‘full-sibling’). This is also used for ‘patrilineal relative’ when closeness of relationship is emphasized. These usages are totally context dependent, but in a regular manner (see Ardener, E. 1954a and 1959b). We have then two unit categories each with its own frequency gradient, transferred as a pair to represent relative ‘nearness’ and ‘distance’ of patrilineality. At this new ‘meta’-level of categorization, the old frequency gradients are now symbolic only. Put oversimply the close associations of nwa nna with ‘half-sibling’ and of nwa nne with ‘full-sibling’ are reduced to degrees of patrilineality defined by other criteria, in differenct contexts. Here again we catch a category peeling away from materiality with no indication in the linguistic terminology to warn us. 10. Cf. G. Steiner (1975: 316): ‘the codified strangeness of most translations from the Persian, the Chinese, or the Japanese haiku’. 11. Anthropology’s use of ‘native’ terms (totem, taboo, mana) is related to this discussion. Henson (1974: 25–30) has a good account. See also F. Steiner’s Taboo (1956) where totemism ‘demarcated … a solid block of otherness’. 12. N. MacAlpine (1832), E. Dwelly (1920) and J. Munro (no date) all contain lists of the names. There are some variants (MacAlpine’s Iomhar = Evander), but the lexicographers do not have any doubts that ‘translations’ exist. Gilleasbuig means ‘Servant of the Bishop’. Gille and Maol (‘tonsured servants’, literally ‘bald’) were interchangeable in religious dedication names, of which this was one. The element bald (from old English beald, ‘simple/bold’) in Archibald (from old English Eorcenbeald) may have influenced the ‘translation’. Perhaps the ‘arch’ of Archibald suggested ‘Archbishop’, kindling associations with the ‘easbuig’ (Gaelic for ‘bishop’) of Gilleasbuig. Ruaraidh is not related to Roderick or Derek. Tearlach is the Anglo-Irish Turlough. Tormoid was a name of the Campbells. The English equivalent Norman may refer to the Norse element the name exhibits. The women’s names present more puzzles than can be discussed here.

Notes

305

13. I choose Danish because a conventional translation works rather easily – thus, Hobbiton = Hobbitrup, Baggins = Saekker. Although Westron often becomes Danish in this manner, however, the Danish translator does not reshape the nomenclature of Rohan into Old Norse (my thanks to J. Ovesen for the text). 14. Thus mam for Northern English dialects may be derived from a different Middle English source, reinforced or influenced from Welsh or other Celtic usage. Mom is from momma which may be a restressed [mǝma:]. Mum may be similarly derived. Effects of initial and final labial nasals may be involved. Any such results still show complexities. The OED is not helpful. 15. Or, as he translates: ‘to complain of having no cudgel during a duel which arises in the forest is preposterous.’ Do we feel that his register is a little inappropriate? 16. Cf. G. Steiner (1975: 407): ‘to demonstrate the excellence, the exhaustiveness of an act of interpretation and/or translation is to offer an alternative or an addendum.’ N OT E S TO CH A P T E R 1 2

1. See Hudson, D. Munby, Man of Two Worlds: The Life and Diaries of Arthur J. Munby 1828– 1910, London, Murray, 1972, for a biography of A.J. Munby, who was psychologically obsessed by working women. Nevertheless, his accounts of their way of life are of great value, and very revealing. 2. See above chapters 4 and 7; also Ardener, S. 1975a, 1975b and 1978. The argument of this paper special relevance to the theory of ‘muting’ and the way groups are defined. 3. For work on life-trajectories, see Ardener, S. 1978: 40–3. N OT E S TO CH A P T E R 1 3

1. My thanks to, amongst others over the years, Jonathan Webber and Herminio Martins who invited me to speak on the theme of modernism at a seminar in early 1983. For ‘intellectuals’ many ‘structuralist’ themes, and ideas like ‘deconstruction’, are being treated as ‘post-modern’ novelties. From social anthropology we can see that this is a misconception. They are, of course, in a sense novelties outside social anthropology, but I shall be saying in the end that the whole structuralist movement is the fiery decline of modernism itself. This paper is then on a small theme related to larger matters. Much of what I have said is an illustration of approaches in other papers. 2. My Malinowski lecture (see chapter 2) should be compared with the present exposition. There the ‘new anthropology’ was shown to have suffered a failure of nerve, but it was hoped that structuralism would develop out of its early phase. This did not happen. 3. See Kuper 1973; Harris 1969. The varied uses of the term ‘structuralism’ form the most confusing element. 4. Romanticism reflected the revolutions themselves. The image was of the devil unchained (Blake, Byron, Brontë) in the form of the unpredictable individual of almost cosmic scope and creative power. In those days it was not unfashionable to seek to attribute the industrial and social explosion to someone, rather than to a process. The romantic here, like the entrepreneur, is the individual embodying process. Modernism made process the hero. The only personal heroes recognized by modernism were the

306 Notes

5.

6.

7. 8.

9.

10.

11.

12. 13.

founders of the modernist movement. These founders became virtually immune from personal criticism, while the romantic heroes were ruthlessly exposed as smaller than life, with ‘feet of clay’. Yet, because the social and cultural development was slower in various countries, the romantic expression co-existed chronologically with modernism in different places. Despite the implacability of the dominant successor, it has shown power to survive it. I may have improved the grammaticality of J. Kennedy. My citation of Ortega does not mean that he was a modernist. Modernism redefined the past, and provided only one acceptable language for critics of the past. Ortega attempted to find a language ‘between the lines’. (See further, ‘Some responses to questions’.) The special foreword to the third edition of The Sexual Life of Savages (1932) was the source of these remarks. A typescript of the 1926 contribution to the Encyclopaedia Britannica is lodged in the Library of the Institute of Social Anthropology, Oxford. There are many corrections in his own hand. This discussion at times seemed to suggest that it was incorrect to criticize Malinowksi only by his writings (e.g. Leach, Jarvie and Gellner et al. 1966: 565). My warm thanks to Dr Hodder, of Cambridge, Dr D. Miller of London University, Dr Henrietta Moore and many others of the New Archaeologists, who allowed me to be present at their early symposia. I hope they do not disagree seriously with these words, and that they will recall that I had urged them to place their excellent sources figuratively in the hold, as ‘Not wanted on the voyage’, and to develop the New Archaeology from its own problems. As far as I know, that is what they are doing. For some of these ‘transformational’ characteristics, see Ardener 1971b: lxi–lxviii. Transformational generative grammar (Chomsky 1957, 1965 and elsewhere) was the linguistic version of late modernism, and had the same or greater impact. It could not be called structuralism because linguists have already called their own period of ‘functionalism’ (synchronic, descriptive) structuralism! The linguistic movements were always less blurred, and more theoretically explicit than the anthropological ones. Nevertheless, it is always confusing when neighbouring disciplines borrow ‘last year’s clothing’ from each other. The same situation prevails in social anthropology now as it did then. The argument was that the language of ‘structure’ was exhausted. The post-structural position was one which lacked any other specification than that it was non-structuralist. It was argued that a new trajectory must already be in existence. Our bad experience of the premature expropriation of the term does not change the argument. It accounts, however, for why one does not wish to name any post-modernist possibilities in social anthropology, lest another ‘nerveless [terminological] arrow miss its mark’! There is no space to go over Central Europe’s place in the terminological spectrum. We have here ‘functional-structural’ preceding ‘structural’, before Malinowski was even dead. Were they ‘mere’ labels? At one level, yes, but ultimately nothing more expresses the unity of modernist terminology than the appearance of these combinations all over our time period. To ‘blame’ Chomsky on Saussure, as some now do, can raise eyebrows. Yet in linguistics the progression from 1916 has a certain unity. By ‘misplaced concreteness’ I mean that the ‘synchronic space’ is timeless, not located in real space and time. Our residual primitivism makes it possible for us to speak as if ethnographies represent ‘really’ synchronic unities.

Notes

307

14. One study (Ardener, E. and Ardener, S. 1965) shows that the tiny demography of the social anthropological profession in 1961 had nevertheless been greatly enlarged by the graduation classes of 1949–52. 15. It is possible in the mid–1980s for a graduate student to ask in all innocence, ‘who were the New Anthropologists?’. This suggests that many of their views are now quietly assimilated. 16. Imagine at any other time after 1920 being asked ‘where is the avant garde?’ and being unable to point to some recognizable person or tendency. Yet since 1975 or so, we would have to reply, ‘They were with us when we set out but we have not seen any lately.’ This vacuum occurs all over the humane disciplines (we may ignore the embarrassing party going on in criticism around the corpses of structuralism and its congeners). The notion of an avant garde is of course the very product of modernists. They claimed to live in the expropriated future. The disappearance of the avant garde is one of the evidences of the end of modernism, for without that future, where will it reside? 17. As a profession we have a propensity to lose touch with the common language of our time. This, which is one source of our strength, is also the source of our failure to influence general thought, save at very rare intervals. N OT E S TO CH A P T E R 1 4

1. This is a paper of some degree of abstraction. It is not an account of the Western Isles, but it should not, despite the terminology, be other than obvious to Gaels. It takes a great deal of explanation, they will be aware, to state the facts to those outwith. 2. Adam Ferguson wrote, in An Essay on the History of Civil Society (1767: 31): ‘the titles of fellow citizen and countryman unopposed to those of alien and foreigner, to which they refer, would fall into disuse, and lose their meaning’. This had a great influence on Evans-Pritchard (see Pocock 1961: 78; and chapter 1, p. 36). Despite this, an ESRC correspondent referred to it as a recent and untried theory. 3. See above, chapter 1, p. 7. Thus: ‘even the most exemplary technical approach to language would not in fact have solved the basic problem of communication. The anthropological “experience” derives from the apprehension of a critical lack of fit of (at least) two entire world-views, one to the other.’ 4. There is endless useless confusion between relativity and relativization on the one hand, and a chimera called (usually by non-anthropologists) ‘cultural relativism’ on the other. Like many contemporaries (cf. Gellner 1983; Edwards 1985) I am not a ‘cultural relativist’. The very act of the comparison of cultures implies the existence of appropriate canons of comparison. By those canons judgements can be made. The relativity of social worlds is a mere fact, beyond all judgements: they are constructed differently, not equally. It is, of course, inappropriate to charge a culture with inferiority because it has few hue terms, or does not separate arm from hand terminologically. Judgements may, however, be made about the ‘adequacy’ of a terminological system. It is sufficient evidence to support this assertion to point out that judgements of inadequacy are daily made, even within a culture. Thus doctors devised anatomical terms, and artists construct colour charts. It is no great step further to assert, if we want to: ‘cultures are extremely unequal in their cognitive power’ (Gellner 1968: 401). The sentence remains, of course, a sentence in our own language.

308 Notes 5. ‘Brazil’ was a red dye-wood; later an imaginary Atlantic island was so named in maps; even after it was localized in South America, a non-existent ‘Brazil Rock’ remained on British Admiralty charts until the second half of the nineteenth century. California was taken from a story of 1510, published in Madrid; it was near the Indies and the terrestrial paradise. India: variously placed, particularly in Indonesia and the Antilles. Libya: once Africa. Africa: one Tunisia. Ethiopia: once any African land occupied by people with ‘burnt faces’. 6. See Loomis (1956: 61–76) for the Arthurian Antipodes, and once more the terrestrial paradise. 7. Barbara Pym included known anthropologists and African linguists in several of her novels, in particular Less then Angels (1955), or as composite characters (‘Everard Bone’ and the like). 8. The Highland Clearances were already under way at the time of Samuel Johnson’s visit to the inner isles in 1776. The sagacious doctor greatly blamed the landlords for encouraging emigrations. In some sense they are still going on. The period for which the term is notorious, some time between 1790 and 1860, was marked as such precisely because of its ideological nature. The Duchess of Sutherland’s commissioner, James Loch, wrote: ‘it was one of the vast changes which the progress of the times demand and will have, and I shall feel ever grateful that I have had so much to do with (these) measures’ (cited Richards, 1982: 185). At ground level the Morayshire agricultural entrepreneur, Patrick Sellar, with his colleague William Young, provided a practical sense of purpose to the implementation of the fashionable ideas after 1809. ‘It was during these removals’ (in Strathnaver) ‘that Patrick Sellar was alleged to have set fire to houses and barns, and caused the deaths of several people, including a nonagenarian woman called Chisholm. He was brought to trial and acquitted in 1816’ (Richards 1982: 312). Derick Thomson writes, in his well known poem, ‘Srath Nabhair’: Agus siud a’bhliadhna cuideachd a shlaod iad a’chailleach do’n t-sitig, a shealltain cho eòlach ‘s a bha iad air an Fhirinnn oir bha nid aig eunlaith an adhair (agus cròthan aig na caoraich) ged nach robh àit aice-se anns an cuireach i a ceann fòidhpe. In his own translation: ‘And that too was the year / they hauled the old woman out onto the dung-heap, / to demonstrate how knowledgeable they were in scripture, / for the birs of the air had nests / (and the sheep had folds) / though she had no place in which to lay down her head’ (in Macaulay 1976: 153). 9. The line is from Iain Mac a’Ghobhainn: ‘Có sgrìobh mi? Có tha dèanamh bàrdachd shanas-reice de mo chnàmhan? Togaidh mi mo dhòrn gorm riutha: ‘Gàidheal calma le a chànan. ‘Who wrote me? Who is making a poetry / of advertisements from my bones? / I will raise my blue fist to them: / “A stout Highlander with his language” (in Macaulay 1976: 179). Derick Thomson writes:

Notes

309

Cha do dh’aithnich mi ‘m bréid Beurla, an lìomh Gallda bha dol air an fhiodh, cha do leugh mi na facial air a’ phràis, cha do thuig mi gu robh mo chinneadh a’ dol bàs.

10. 11. 12.

13. 14. 15.

16.

17.

‘I did not recognize the English braid, / the lowland varnish being applied to the wood, / I did not read the words on the brass, / I did not understand that my race was dying.’ (in Macaulay, 1976: 157). The Glasgow and Edinburgh reviewers of Chapman’s book were unnecessarily outraged, but see the careful consideration, in two long articles by James Shaw Grant, in the Stornoway Gazette (1978), and the appreciative review by Parman in Man. The tendency for publicists to react ambiguously to those using the threatened language in a non-private way is comprehensible. See Holt and Pym (1984) and its reviews. It is not thought odd that the London regional television programme should have the Scots presenter interviewing local representatives with Northern accents. The suddenness of city explosions, when they occur, suggests that there are pockets of remoteness within these blank spaces! Adamnan’s Life of St Columba is a medieval classic of remote area studies. The Baptist settlement of 1858 had an ‘improving’ philosophy; the German annexation of 1884 led to the establishment of plantations. HIDB friends will not be offended; they read worse every day in the press. Ardveenish may yet take off. Lord Leverhulme’s ambitions for Lewis and Harris were a benign form of paternalism. Round a crofthouse in Lewis were the following items, according to the writer Derek Cooper (one of the most sensitive reporters of the Hebrides): ‘5 cwt van (circa 1950s); Ford tractor minus one wheel; fragment of pre-Great War reaper; upright piano; 37 blue plastic fishboxes; 7 green lemonade crates; 2 chimney pots; a sizeable pyramid of sand; a pile of cement blocks; 7 lobster creels; assorted timber; 2 bales of barbed wire (rusted); broken garden seat; Hercules bicycle frame; piece of unidentifiable machinery (loom?); a sofa’ (Cooper 1985: 192). This is a carefully fictionalized picture, and several islands are combined. Tamara Kohn has pointed out that Hebrideans nevertheless are used to pulling apart composite pictures and painstakingly reassembling them. In any case, there are no prizes! N OT E S TO CH A P T E R 1 5

1. An earlier version of this paper was given at Dr Mary Douglas’s seminar in London on 26 October 1967. An earlier version still was given for the Oxford University Delegacy of Extra-mural Studies on 27 January 1965. 2. The following can be amplified from Ardener (1956, 1961, 1962) and Ardener, Ardener and Warmington (1960). The Bakweri call themselves Vakpe and (with the removal of the Bantu prefix) they are also referred to in the literature as Kpe. I have now published (1970) the history of the mountain Bakweri in the nineteenth century. 3. Baptist missionaries landed at Fernando Po in 1841 and founded settlements at Bimbia (1844), Douala (1845) and Victoria (1858), finally evacuating Fernando Po itself. They often proudly claimed to have introduced the new crops. Saker (Missionary Herald, 1871: 56–7) recalled that harvests on the coast had formerly provided only for

310 Notes

4.

5.

6.

7. 8. 9.

10.

three months’ consumption. The xanthosoma bears in Bakweri and Duala the same name as the island of Fernando Po (Bk. Likao, Du. Dikabo); the Duala plural form makabo was later borrowed by inland peoples together with the crop. Bakweri population of Buea (1953): 914. Preuss (DKB, II, 1891: 517) assessed the population in 1891 at 1,500 of whom 600 were men capable of bearing arms, 400 being provided with guns. Later, 600 guns were said to have taken part in the battle of Buea. See DKB, III, 1892: 14–18; Ardener 1956: 24–5. Von Gravenreuth, von Schuckmann, von Stetten, Dr Richter and the gardener Pfeil, 150 African mercenaries, 10 Kru policemen and a Maxim gun were halted at a barricaded ravine. Gravenreuth was shot dead, the Maxim gun jammed, and the expedition returned to the coast bearing its buried leader’s head and heart. DKB, VI, 1895: 134, 321, 382–3. Dominic (1901: 105–6) describes the unearthing of the rest of Gravenreuth’s bones, finding them still clad, he says (as if echoing some grotesque advertisement), ‘in den wohlerhaltenen gelben, festen, nägelbeschlagenen Schnürschuhen’ [in his well-polished, dun-coloured, nail-studded boots]! The remarkable obsequies of von Gravenreuth illustrate the peculiar muffling effect of belief on behaviour between foreign peoples. With Gravenreuth dead, the German belief system sought as if by antennae for the belief system of the Bakweri. The latter, as ‘savages’, were expected to prize a victim’s head and heart: the Germans spared their leader’s body the ignominy of losing those parts to the Bakweri by removing them themselves. Any explanation of the German behaviour must be found, then, in German belief, not in the belief of the Bakweri, for whom the events, had they been aware of them (and possibly they were), would merely have suggested (on the contrary) that it was the Germans who prized the heads and hearts of the dead. By such a dialectic, no doubt, colonizers and the colonized may come in time to produce a body of shared beliefs that differ sharply from either of the parent systems. For the possibility that the Germans shot Gravenreuth by accident, see now Ardener 1970. Most of this report appears in DKB, X, 1899: 513. The sasswood mixture was not thought of as in itself killing the witch. For some of the closely comparable elements to be found in the beliefs of peoples neighbouring the Bakweri, see, for example, Talbot 1912; Ittmann 1953. The Haitian vocabulary of West African origin (Huxley 1966: 237–41) seems to contain some other coastal Cameroon elements. We may compare Bakweri lova, ‘god’, ‘sky’, Haitian loa, ‘gods’, ‘spirits’; more securely: Bk. nganga, ‘doctor’, Ha. gangan, ‘magician’; Bk. gbanga, Du. bwanga, ‘medicine’, Ha. wanga, ‘a magical charm used for selfish ends’; Bk. mwana, Du. muna, ‘child’, Ha. ti moun, ‘a child’; Bk. ngole, Du. ngokolo, ‘millipede’, Ha. gangolo, ‘centipede’; Bk. maese, Du. mawasa, Ha. marassa, ‘twins’. For a detailed history of the Bakweri Cooperative Union of Farmers, see Ardener, S. (1958) and Ardener, Ardener and Warmington (1960: 329–32). The roots of the Union lay in a meeting of 5 August 1951. In 1952, it had an initial membership of only 73, yet in that year it shipped 8,000 stems of bananas valued at £2,500. There was a dramatic increase in membership. In 1953, the first full year of operations, it shipped 34,000 stems valued at £13,410. In 1958, the co-operators shipped 1,350,000 stems valued at nearly £1 million. In that year, the villagers on the mountain received a net income of £300,000, about £150 per member. The 182 members at Lisoka (a village that looms large in this account) received £39,000 gross: an average of £216

Notes

11. 12.

13.

14.

15.

16.

311

gross each, about £75 net (Ardener, Ardener and Warmington 1960: 330 and Table 81). For Obasi Njom among the Banyang themselves (Basinjom), see now Ruel 1969: 210–13. It is noteworthy that the cooperative nature of the venture enabled individual ambition to be achieved through collective endeavour. The higher and the more general the level of prosperity, the less likely may be the stimulation of inona. Whether all societies that lay great stress upon economic ‘justice’ tend to react in a Bakweri way (that is, with the sensitivity to individual success increasing when the economy is stagnant, thus increasing the tendency to stagnation) is a good question. Or a new arrangement or realien. The water-spirits are part of a different section of Bakweri belief: in liengu (pl. maengu), appropriate to the deep-sea wharf environment (Ardener 1956: 93–4, 98–100). A surprising linkage of all the themes occurs, however, in the account of Bakweri folklore given to the administrator B.G. Stone, whose MS report (1929) leans heavily on the work of the late Mr Steane, a Bakweri teacher of Victoria. He says (paragraph 110) that there is a powerful spirit ‘at the bottom of the lake near the summit of Mount Kupe in Kumba Division. In the latter lake there is a great spirit market and spirits from all over the world are thought to meet there and barter their goods. The new Nigerian coinage and paper money were said first to have been introduced in this spirit market, and to have been distributed thence to the people’ (my italics). Mount Kupe is, of course, the nyongo mountain. Phases III–V are documented. Phase II (no nyongo) rests on the statement of the Bakweri, the relatively modern realien of the nyongo belief, and the saliency of the old liemba at the time of the colonial occupation. Phase I is merely a symmetrical speculation, aided by my guess that sómbà (sómbî) = Caribbean zombi. The transmission would have to occur before the almost total cessation of the slave trade from this area after 1840 (the peak was from c.1785 to 1830). The pre-xanthosoma economy is assumed not to be expanding, and to be based on the former plantain staple cultivated by men. There are some ritual hints of a yam staple before the plantain, based on the D. dumetorum species (Ardener 1956: 46). A shadowy economic ‘Phase minus I’ is perhaps remotely conceivable. The Bakweri ‘innovations’ were group borrowings that differed from ‘non-innovative’ borrowings mainly in their unforeseen effects. Yet the second innovation is ideologically connected with the first: it restored the place of the banana (man’s crop), which the xanthosoma (woman’s crop) had overwhelmed. Whatever the reinforcing effect of the ‘jackpot’ consequence, the earliest steps towards acceptance of export banana-farming were aided by its being a ‘man’s’ crop. Lévi-Strauss, particularly in his latest studies (1964, 1966b), analyses brilliantly the constant rearrangements of themes and motifs, and transformations in myth. He uses the term bricolage (1962b) to indicate the impression of new structures being ‘bodged up’ out of pieces of other structures. In his terminology, every so often, by a process of bricolage, a new Bakweri myth is created out of old and new elements embodying a statement involving economics with zombie-like phenomena. Something is repeated and revived over time through these successive replications. Whatever ‘it’ is, it is to be considered at a different level of analysis from that used in even the sophisticated analysis of content. I call it the ‘template’. Its simple meaning of a ‘form’ or ‘shape’

312 Notes used for copying is self-evident. I have in mind, however, the word’s use in molecular biology, in the description of the process whereby the genetic material (DNA) continuously replicates itself from the chemical materials that are presented to it. The analogy I wish to bring out is that the molecules that replicate do so because their structures are logically limited in such a way that the chemical reactions they can take part in always compel the same end: the repetition of the original structures. They are like puzzles whose lengthy solution is finally in the form of the same puzzle. I suggest, merely on the loosest analogy, that in systems of belief over time certain configurations continually recur. Across time, ‘synchronically’, they may from time to time be absent. The metaphor of the ‘template’ is a way of visualizing that diachronic continuity perceived in the phrase: plus ça change, plus c’est la même chose. So we see the ideological template of the father-despot replicated in Russian history through totally different (even ‘reversed’) realien in Tsarist and Stalinist times. In French ideology, two templates in particular replicate with monotonous regularity: one summarized by ‘l’état c’est moi’, and one by ‘the barricades’. 17. ‘[T]he patient sees very small, perfectly formed figures, usually active and mobile, gaily coloured, and pleasant to look upon.’ In a case of scarlet fever, a child reported them as tiny ‘clowns’ who seemed to be moving across and under his bed. They reappeared at fifteen-minute intervals (Leroy 1922; Savitsky and Tarachow 1941). N OT E S TO CH A P T E R 1 6

1. The writings of Professor L. R. Palmer (in particular his The Latin Language, 1954) have long been one stimulus of my interest in this subject. I should like to acknowledge this here, although I have reason to believe that this paper does not carry his complete approval. Professor Guthrie’s writings on Bantu linguistics and the problem of Cameroon linguistics provided another and more empirical stimulus. The debate on this paper at the ASA Conference was very lively and enjoyable, and I should like to thank Dr E. R. Leach and Dr G. Milner, who took up positions in support, as well as Professor Dell Hymes, who raised cogent points of disagreement. 2. For an amplification, see the Postscript to this paper. 3. The basic statements were Brugmann and Delbrück (1886–1900) and Paul (1880) in various editions. See also, in declining complexity: Brugmann (1902–04), Pedersen (1931), Meillet (1937), Krahe (1962–63), Bloomfield (1933). An early version of the Neogrammarian creed (their term) is Osthoff and Brugmann (1878: iii–xx); see also Brugmann (1885). 4. Only a simplified notation is used. See Chomsky (1957, 1964, 1965b, 1966a, 1966b); Chomsky and Halle (1965). 5. See Greenberg (1963a, 1966). ‘Universals’ belong, if attested, in the axioms of the model. Neogrammarians often treated them as if they were generated by the model. The uncertain status of some ‘universals’, however, makes them best regarded as part of the conventions of interpretation. ‘Our work as historical linguists is narrowly constrained by our judgements of what is and what is not a universal property of natural language, and we can expect the progress of historical linguistics to be closely connected with the search for linguistic universals’ (King 1969: 16). 6. See Crossland (1951). Meillet (1937) expressed a certain disappointment with Hittite: ‘Le nombre, encore assez petit, des textes interprétés, les obscurités d’une notation

Notes

7.

8. 9.

10. 11.

12.

13.

313

qui n’est pas faite pour une langue indo-européenne, les incertitudes qui subsistent sur bien des points, le degré avancé d’évolution de la langue à certains égard, font que le hittite ne rend pas tous les services qu’on souhaiterait’ [the number, still rather small, of interpreted texts, the obscurities of a notation not designed for an Indo-European language, the uncertainties which surround many points, the advanced degree of development of the language in certain respects, have meant that Hittite has not given us all the help for which we hoped] (1964 edn: 56). The discredited ‘Indo-Hittite’ remains in the 1964 edition of Hockett (1958). Leroy (1967: 115) states concisely the methodological flaw: ‘It consists in comparing the newly-deciphered languages with this Indo-European that was reconstructed on the basis of all the other Indo-European languages – i.e. excluding the new ones – and indeed in some respects, on the basis of two or three languages [especially Sanskrit and Greek] that are quite arbitrarily considered as being especially typical. So it is not in the least surprising that characteristically Hittite features, for example, do not appear in this reconstructed Indo-European … It is obvious that if Hittite and Tokharian had been discovered fifty years earlier, Brugmann’s Indo-European would have looked very different’. Thus: ‘Axiom I. “Two languages are related if, and only if they were once one language”’ (p. 28). Ross’s attempt is merely a sketch. The systematization by Hoenigswald (1960), in terms of American post-Bloomfield structural linguistics, is much more interesting, since he is quite explicit that the analysis can begin only after the corpus is chronologically marked. This is the burden of his remarks about the palaeographical preliminaries required (p. 3). The treatment of Bantu as a greatly dispersed sub-category of a Niger-Congo family is particularly attractive. Haas (1969), whose basic position is unashamedly Neogrammarian (in that non-pejorative sense that my argument implies), notes rightly that the full power of the method in its classical form has not yet been fully exploited among historical linguists of exotic languages (p. 16). She falls into the error of confusing model with reality with statements like ‘Every protolanguage was … once a real language’ (p. 32). Hauschild (1964) even uses the term Mischsprache. See also Cowgill (1963: 105). ‘Die “Laryngaltheorie” kann … weder in ihrer Substanz noch in ihrer Methodik als gesichert gelten’ [‘Laryngeal theory’ cannot be considered secure in either its substance or its methods] (Krahe 1962–63, I: 101). The field of historical linguistics that most closely touches on these interests of social anthropologists is associated with the names of Trier (1931), von Wartburg (1943, 1969) and, in this country, Ullmann (1951, 1959, 1963). This is discussed in the introduction to the present volume. The whole question of loan, analogy, and the like is discussed by Lyons (1968: 36–8). It can be argued that ‘asphalt’ never existed in the speech of the users of ‘ashfelt’. For the older folk-etymology of the ‘Proctor’s walk’, see Wells (1906: 9–10): ‘The Proctor’s walk is the most curious feature of the degree ceremony; it always excites surprise and sometimes laughter. It should however be maintained with the utmost respect, for it is the clear and visible assertion of the democratic character of the University … But popular imagination has invented a meaning for it, which certainly was not contemplated in its institution; it is currently believed that the Proctors walk in order to give any Oxford tradesman the opportunity of “plucking” their gown and protesting against the degree of a defaulting candidate … There is a tradition that such a protest

314 Notes has actually been made within living memory and certainly it was threatened quite recently…’. 14. Linguistic folk-etymology links directly with the mythopoeic faculty. The Mycenaean Greek place-name Aptarwa (Ventris and Chadwick 1956), of dubious origin, is rendered in later Greek Aptera. Stephanus of Byzantium, like Isidore a true bricoleur, explained that it was there that the Sirens, after being defeated in a contest with the Muses, were so disgusted that they threw their wings from their shoulders, turned white and jumped into the sea: hence Aptera, ‘wingless’, and the name of the nearby islands, Leukai, ‘white’. (I am indebted to Miss G. Hart for this reference.) 15. Gosse (1907, 1949 edn: 87–8). Of P. H. Gosse’s Omphalos, Charles Kingsley could only cry that he could not ‘believe that God has written on the rocks one enormous and superfluous lie’. For those who doubt the real problem, onomastic studies are a good introduction. I have offered a detailed example (Ardener 1968). N OT E S TO CH A P T E R 1 7

1. Delivered in the series General Linguistics for the Sub-Faculty of Linguistics, Oxford University, May 1987. 2. All the papers of Sapir’s mentioned below are to be found in this volume. 3. Cited in Koerner 1984: 181.

References

Adler, I. 1958 The New Mathematics, New York, Day. Althusser, L. 1969 For Marx, London, Allen Lane. Anderson, A. O. 1922 Early Sources of Scottish History, Edinburgh, Edinburgh University Press. Anderson, M. O. 1973 Kings and Kingship in Early Scotland, Edinburgh, Scottish Academic Press. Andersson, T. M. 1987 A Preface to the Nibelungenlied, California, Stanford University Press. Ardener, E. 1954a ‘The Kinship Terminology of a Group of Southern lbo’, Africa, 24, pp. 85–99. ––– 1954b ‘Some lbo Attitudes to Skin Pigmentation’, Man, 54, pp. 71–3. ––– 1956 Coastal Bantu of the Cameroons, London, International African Institute. ––– 1957 ‘Numbers in Africa’, Man, 226. ––– 1958 ‘The “Kamerun” Idea’, West Africa, 2147 and 2148. ––– 1959a ‘The Bakweri Elephant Dance’, Nigeria, 60, pp. 31–8. ––– 1959b ‘Lineage and Locality among the Mba-Ise Ibo’, Africa, 29 (2), pp. 113–34. ––– 1961 ‘Social and Demographic Problems of the Southern Cameroons Plantation Area’, in Southall, A. (ed.) 1961. ––– 1962a Divorce and Fertility – An African Study, Oxford, Oxford University Press. ––– 1962b ‘The Political History of Cameroon’, The World Today, 18 (8), pp. 341–350. ––– 1965a Historical Notes on the Scheduled Monuments of West Cameroon, West Cameroon, government publication. ––– 1965b Review of Jarvie, I. C. 1963, Man, 65, p. 57. ––– 1967a ‘The Nature of the Reunification of Cameroon’, in Hazelwood, A. (ed.) 1967. ––– 1967b ‘The Notion of the Elite’, African Affairs. ––– 1968 ‘Documentary and Linguistic Evidence for the Rise of the Trading Polities between Rio del Rey and Cameroon 1500–1650’, in Lewis, I. M. (ed.) 1968. ––– 1970a Kingdom on Mount Cameroon: Documents for the History of Buea, 1844–1898 (forthcoming). ––– 1970b ‘Galileo and the Topological Space’, Journal of the Anthropological Society of Oxford, 1(3), pp. 125–30.

316 References ––– 1970c ‘Witchcraft, Economics, and the Continuity of Belief ’, in Douglas, M. (ed.) 1970, pp. 141–60. ––– 1971a (ed.) Social Anthropology and Language (ASA Monographs 10), London, Tavistock. ––– 1971b Introduction to Social Anthropology and Language, Ardener, E. (ed.) 1971, pp. ix–cii. ––– 1971c ‘Social Anthropology and the Historicity of Historical Linguistics’, in Ardener, E. (ed.) 1971, pp. 209–41. ––– 1971d ‘The New Anthropology and its Critics’, (Malinowski Lecture 1970, with appendix), Man, 6 (3), pp. 449–67. ––– 1972a ‘Language, Ethnicity and Population’, Journal of the Anthropological Society of Oxford, 3 (3), pp. 125–32; reprinted in Beattie, J. H. M. and Lienhardt, R. G. (eds) 1975. ––– 1972b ‘Belief and the Problem of Women’, in La Fontaine, J. (ed.) 1972 (also in Ardener, S. (ed.) 1975). ––– 1972c Introduction and commentary to reprint of J. Clarke, Specimens of Dialects, Clarke, J. 1848, Farnham, Gregg International. ––– 1973 ‘Behaviour: a Social Anthropological Criticism’, Journal of the Anthropological Society of Oxford, 4 (3), pp. 152–4. ––– 1974 ‘Social Anthropology and Population’ (Wolfson Lecture, 1973, ‘Population and “Tribal Communities”’), in Parry, H. B. 1974, pp. 25–50. ––– 1975a ‘The Problem of Women Revisited’, in Ardener, S. (ed.) 1975, pp. 19–27. ––– 1975b ‘The Voice of Prophecy: Further Problems in the Analysis of Events’, (Munro Lecture). Now chapter 9 in the current volume. ––– 1976 ‘“Social fitness” and the idea of “Survival”’, Journal of the Anthropological Society of Oxford, 7 (2), pp. 99–102. ––– 1977 ‘Comprehending Others’, paper given to the Wenner-Gren Symposium, published in part as ‘Social Anthropology, Language and Reality’, in Parkin, D. (ed.) 1982, and in Harris, R. (ed.) 1982. Now chapter 11 in the current volume. ––– 1978 ‘Some Outstanding Problems in the Analysis of Events’ (ASA conference paper, 1973), in Schwimmer, E. (ed.) 1978, reprinted in Foster, M. and Brandes, S. (eds) 1980. Now chapter 5 in the current volume. ––– 1979 ‘Social Anthropology’, in A New Dictionary of Sociology, Duncan Mitchell G. (ed.), London, Routledge and Kegan Paul. ––– 1980 ‘Ten Years of JASO’, Journal of the Anthropological Society of Oxford, 11 (2), pp. 124–31. ––– 1981 ‘The Problem of Dominance’, Journal of the Anthropological Society of Oxford, 12 (2); reprinted in Duke, L., Leacock, E. and Ardener, S. (eds) 1986. ––– 1982 ‘Social Anthropology, Language and Reality’, in Parkin, D. (ed.) 1982 and in Harris, R. (ed.) 1982. ––– 1984 ‘Ethology and Language’, in Harré, R. and Reynolds, V. (eds) 1984, pp. 111–15 (first delivered as part of Ardener, E. 1977). ––– 1985 ‘Social Anthropology and the Decline of Modernism’, in Overing, J. (ed.) 1985. ––– 1987a ‘“Remote Areas” – some Theoretical Considerations’, in Jackson, A. (ed.) 1987. Now chapter 14 in the current volume ––– 1987b ‘Edward Sapir 1884–1939’, Journal of the Anthropological Society of Oxford, 18 (1), pp. 1–12. Now chapter 17 in the current volume. ––– 1987c ‘Evidences of Creation’, ASA conference paper, in Tonkin, E., McDonald, M. and Chapman, M. (eds) History and Ethnicity, London, Tavistock. Now chapter 18 in the current volume.

References

317

Ardener, E. and Ardener, S. 1965 ‘A Directory Study of Social Anthropologists’, British Journal of Sociology, 16 (4), pp. 295–314. Ardener, E., Ardener, S., and Warmington, W. A. 1960 Plantation and Village in the Cameroons, Oxford, Oxford University Press. Ardener, S. 1958 ‘Banana Co-operatives in the Southern Cameroons’, Nigerian Institute of Social and Economic Research: Conference Proceedings, pp. 10–25, Ibadan. ––– 1968 ‘Eye-witnesses to the Annexation of Cameroon, 1883–1887’, Buea: Government publication. ––– 1975a (ed.) Perceiving Women, London, Dent. ––– 1975b Introduction to Ardener, S. (ed.) 1975a. ––– 1978 (ed.) Defining Females, London, Croom Helm. Armstrong, R. A. 1825 A Gaelic Dictionary, London, J. Duncan. Banaji, J. 1970 ‘Anthropology in Crisis’, New Left Review, 64, pp. 71–85. Banton, M. (ed.) 1966 Anthropological Approaches to the Study of Religion (ASA Monographs 3), London, Tavistock. Barltrop, R. 1975 The Monument – The Story of the Socialist Party of Great Britain, London, Pluto Press. Barnes, J. A. 1949 ‘Measures of Divorce Frequency in Simple Societies’, Journal of the Royal Anthropological Institute, 79, pp. 37–62. ––– 1963 ‘Some Ethical Problems in Modern Fieldwork’, British Journal of Sociology, 14 (2), pp. 118–34. Barnes, R. H. 1984 Two Crows Denies It: A History of Controversy in Omaha Sociology, Nebraska, University of Nebraska Press. Barnett, C. 1972 The Collapse of British Power, London, Eyre Methuen. Barth, F. 1966 Models of Social Organisation, occasional paper, 23, London, Royal Anthropological Institute. ––– 1969 Ethnic Groups and Boundaries, London, Allen and Unwin. Barthes, R. 1964 Elements de Simiologie, Paris, Le Seuil. ––– 1967 Elements of Semiology, trans. of Barthes 1964, London, Cape. Bates, R. H. 1970 ‘Approaches to the Study of Ethnicity’, Cahiers d’Etudes Africaines, 10; pt. 40, pp. 516–61. Bateson, G. 1936 Naven, Cambridge, Cambridge University Press. Bazell, C. E., Catford, J. C., Halliday, M. A. K. and Robins, R. H. (eds) 1966 In Memory of J. R. Firth, London, Longman. Beals, R. L. 1957 ‘Native Terms and Anthropological Methods’, American Anthropologist, 59, pp. 716–17. Beattie, J. H. M. 1957 ‘Nyoro Personal Names’, Uganda Journal, 21, pp. 99–106. ––– 1960 ‘On the Nyoro Concept of Mahano’, African Studies, 19, pp. 145–50. ––– 1964a Other Cultures, London, Cohen and West. ––– 1964b ‘Kinship and Social Anthropology’, Man, 64, pp. 101–3. ––– 1968 ‘Aspects of Nyoro Symbolism’, Africa, 38 (4), pp. 413–42. ––– 1970 ‘On Understanding Ritual’, in Wilson, B. (ed.) 1970. ––– 1971 ‘Has Social Anthropology a Future?’, Journal of the Anthropological Society of Oxford, 2 (3), pp. 111–20. Beattie, J. H. M. and Lienhardt, R. G. (eds) 1975 Essays in Social Anthropology – Essays in Memory of E. E. Evans-Pritchard by his Former Colleagues, Oxford, Clarendon Press. Beck, B. E. F. 1969 ‘Colour and Heat in South Indian Ritual’, Man (n.s.) 4 (4), pp. 553–72.

318 References Beidelman, T. O. 1964 ‘Pig (Guluwe): an Essay on Ngulu Sexual Symbolism and Ceremony’, Southwestern Journal of Anthropology, 20, pp. 359–92. Benedict, B. 1972 ‘Social Regulation of Fertility’, in Harrison, G. A. and Boyce, A. J. (eds) 1972. Benveniste, E. 1939 ‘La Nature du Signe Linguistique’, Acta Linguistica, 1, pp. 23–9, reprinted in Benveniste 1966, pp. 49–55. ––– 1966 Problèmes de Linguistique Générale, Paris, Gallimard. Bergsland, K. and Vogt, H. 1962 ‘On the Validity of Glottochronology’, Current Anthropology, 3, pp. 115–58. Berlin, B. and Kay, P. 1969 Basic Colour Terms, their Universality and Evolution, Los Angeles, University of California Press. Bernstein, B. 1958 ‘Some Sociological Determinants of Perception: an Inquiry into Subcultural Differences’, British Journal of Sociology, 9, pp. 159–74, reprinted in Fishman (ed.) 1968. ––– 1960 ‘Language and Social Class’, British Journal of Sociology, 11, pp. 271–6. ––– 1961 ‘Aspects of Language and Learning in the Genesis of the Social Process’, Journal of Child Psychology and Psychiatry 1, pp. 313–24, reprinted in Hymes (ed.) 1964. ––– 1964 ‘Elaborated and Restricted Codes: their Social Origins and some Consequences’, in Gumperz and Hymes (eds) 1964, pp. 55–69; and in Smith (ed.) 1966, pp. 427–41. ––– 1965 ‘A Socio-linguistic Approach to Social Learning’, in J. Gould (ed.) 1965, pp. 144–68. Berry, J. 1966 Introduction to second edition of Malinowski, 1935, vol. 2, pp. vii–xvii, London, Allen and Unwin. Blacking, J. 1963 Review of Pocock 1961, African Studies, 22 (4), pp. 194–5. Blanc, R. 1960 Manuel de Recherche Démographique en Pays Sous-développés, CCTA, Paris. Bloch, M. 1977 ‘The Past and Present in Present’ (Malinowski Lecture, 1977), Man (n.s.) 12, pp. 278–92. Bloomfield, L. 1922 ‘Review of Edward Sapir’s Language’ in Koerner K. (ed.) 1984. ––– 1933 Language, New York, Holt, Rinehart and Winston; London, Allen and Unwin, 1935. Boas, F. 1913 The Mind of Primitive Man, Lowell Institute Lectures 1910–11, New York. Boece, H. 1536 The Chronicles of Scotland, translated into Scots by John Bellenden (1531) from Scotorum Historiae, Paris, 1527. Republished in 1941 for the Scottish Texts Society, Edinburgh, Blackwood, Batho, E. C. and Winifred Husbands, H. (eds). Bogatyrev, P. 1931 ‘Prispevek k Strukturalni Etnografii’, Slovenska Miscellanea, Bratislava. ––– 1935 ‘Funkcno-strukturalna Metoda a ine Metody Etnografie a Folkloristiky’, Slovenske Pohl’ady, 51, pp. 550–8. Bohannan, L. 1952 ‘A Genealogical Charter’, Africa, 22, pp. 301–15. Bohannan, P. 1956 ‘On the Use of Native Language Categories in Ethnology’, American Anthropologist, 58, p. 557. ––– 1958a ‘On Anthropologists’ Use of Language’, American Anthropologist, 60, pp. 161–2. ––– 1958b Rejoinder to Taylor 1958, American Anthropologist, 60, pp. 941–2. Borges, J. L. 1971 ‘Tlon, Uqbar, Orbis Tertius’ in Borges, J. L. Ficciones, Madrid, Allianza Emece. Bovin, M. 1966 ‘The Significance of the Sex of the Field Worker for Insights into the Male and Female Worlds’, Ethnos, 31 (supp.), pp. 24–7. Bowen, E. S. 1954 Return to Laughter, London, Gollancz.

References

319

Braithwaite, R. B. 1960 Scientific Explanation, Cambridge, Cambridge University Press. Brown, R. W. and Lenneberg, E. H. 1954 ‘A Study in Language and Cognition’, Journal of Abnormal and Social Psychology, 49, pp. 454–62, reprinted in Saporta (ed.) 1961. Brugmann, K. 1885 Zum Heutigen Stand der Sprachwissenschaft, Strasburg, Carl J. Trubner. ––– 1902–4 Kurze Vergleichende Grammatik der Indogermanischen Sprachen, Strassburg, Carl J. Trubner. Brugmann, K. and Delbruck, B. 1886–1900 Grundriss der Vergleichenden Grammatik der Indogermanischen Sprachen, Strassburg, Carl J. Trubner. Buea (ms) 1926 Assessment and Intelligence Reports on Ngie, Ngwaw, Mogamaw and Ngemba, Cameroon archives, Buea. Burling, R. 1964 ‘Cognition and Componential Analysis: God’s Truth or Hocus Pocus’, American Anthropologist, 66, pp. 20–8. ––– 1969 ‘Linguistics and Ethnographic Description’, American Anthropologist, 71 (5), pp. 817–27. Buyssens, E. 1943 Les Langues et le Discours: Essais de Linguistique Fonctionelle dans le Cadre de la Sémiologie, Brussels, Lebegue. Callan, H. 1970 Ethology and Society – Towards an Anthropological View, Oxford, Clarendon Press. Campbell, Lord Archibald 1889 Waifs and Strays of Celtic Tradition, vols 1–5. Campbell, J. F. 1872 Leabhar na Féinne: Heroic Gaelic Ballads collected in Scotland chiefly from 1512 to 1871, London, Eyre and Spottiswoode (photo-reprint 1972, Shannon, Irish University Press). Campbell, J. G., 1900 Superstitions of the Highlands and Islands of Scotland, Glasgow, MacLehose and Son. Capell, A. 1966 Studies in Sociolinguistics, The Hague, Mouton. Carmichael, A. 1900 Carmina Gadelica, Edinburgh, Oliver and Boyd. Carnap, R. 1937 The Logical Syntax of Language, London, Kegan Paul, Trench, Trubner. Carroll, J. B. (ed.) 1964 Language, Thought and Reality; Selected Writings of Benjamin Lee Whorf, Cambridge, Mass., MIT Press. Casagrande, J. B. 1963 ‘Language Universals and Anthropology’, in Greenberg (ed.) 1963, pp. 279–98. Cassirer, E. 1933 ‘La langue et la construction du monde des objets’, Journal de Psychologie, (numéro exceptionnel) 30 (1), pp. 18–44. ––– 1953 Philosophy of Symbolic Forms, vol. 1, New Haven, Yale University Press. ––– 1954 An Essay on Man, New York, Doubleday. Chadwick, H. M. 1912 The Heroic Age, Cambridge, Cambridge University Press. ––– 1945 The Nationalities of Europe, Cambridge, Cambridge University Press. Chadwick, H. M. and Chadwick, N. K. 1940 The Growth of Literature, Cambridge, Cambridge University Press (reprinted 1986). Chagnon, N. A. 1972 ‘Tribal Social Organization and Genetic Microdifferentiation’ in Harrison, G. and Boyce, A. (eds) 1972. Chalmers, N. 1977 ‘The Heterogeneity of Play’, paper for the Wenner-Gren Symposium, July. Chapman, M. 1977 ‘What Science is Saying about the Celts’, Journal of the Anthropological Society of Oxford, 8 (2), pp. 84–94. ––– 1978 The Gaelic Vision in Scottish Culture, London, Croom Helm. ––– 1982 “Semantics” and the “Celt”’, in Parkin, D. (ed.) 1982.

320 References ––– (with M. McDonald) 1987 ‘E. W. Ardener – Obituary Notice’, Anthropology Today, 3 (6), pp. 21–3. Chapman, M., McDonald, M. and Tonkin, E. (eds) 1989 History and Ethnicity (ASA Monographs 27), London, Routledge. Chilver, E. M. 1966 ‘Zintgraff’s Explorations in Bamenda, Adamawa and the Benue Lands, 1889–1892,’ Buea, Government publication. Chilver, E. M. and Kaberry, P.M. 1968 Traditional Bamenda, Buea, Government publication. Chomsky, N. 1957 Syntactic Structures, The Hague, Mouton. ––– 1964 Current Issues in Linguistic Theory, The Hague, Mouton. ––– 1965 Aspects of the Theory of Syntax, Cambridge, Mass., MIT Press. ––– 1966a Topics in the Theory of Generative Grammar, The Hague, Mouton. ––– 1966b Cartesian Linguistics, New York, Harper and Row. ––– 1968 Language and Mind, New York, Harcourt. ––– 1969 ‘Language and Philosophy’, in Hook, S. (ed.) 1969. Chomsky, N. and Halle, M. 1965 ‘Some Controversial Questions in Phonological Theory’, Journal of Linguistics, 1, pp. 97–138. Chowning, A. 1970 ‘Taboo’, Man, (n.s.) 5, pp. 309–10. Churchill, W. S. 1956 A History of the English-Speaking Peoples, vol. 1, London, Cassell. Clarke, J. 1848 Specimens of Dialects: Short Vocabularies of Languages; and Notes of Countries and Customs in Africa, Berwick-on-Tweed. Cohen, Abner 1969 Custom and Politics in Urban Africa, London, Routledge and Kegan Paul. Cohen, Anthony (ed.) 1982 Belonging, Manchester, Manchester University Press. Cohen, I. B. (ed.) 1958 Isaac Newton’s Papers and Letters on Natural Philosophy and Related Documents, Cambridge, Cambridge University Press. Cohen, L. J. 1966 The Diversity of Meaning, second revised edition, London, Methuen (first edition 1962). Cohen, M. 1948 Linguistique et Matérialisme Dialectique, Paris, Perches; also in Cinquante Années de Recherches Linguistiques, Ethnographiques, Sociologiques, Critiques et Pedagogiques 1955, pp. 38–52, Paris, Klincksieck. Colby, B. 1966 ‘Ethnographic Semantics: a Preliminary Survey’, Current Anthropology, 7 (1), pp. 3–32. Collinder, B. 1968 Kritische Bemerkungen zum Saussure’schen Cours de Linguistique Générale, Uppsala, Acta Societatis Linguisticae Upsaliensis, Almqvist and Wiksell. Conklin, H. C. 1955 ‘Hanunóo Color Categories’, Southwestern Journal of Anthropology, 11, pp. 339–44, reprinted in Hymes, D. (ed.) 1964. Cooper, D. (1985) The Road to Mingulay – a View of the Western Isles, London, Routledge and Kegan Paul. Cowgill, W. 1963 ‘Universals in Indo-European Diachronic Morphology’, in Greenberg (ed.) 1963a. Cox, J. C. 1910 The Parish Registers of England, Parish Register Society, Public Records Office, London. Crabb, D. W. 1965 Ekoid Bantu Languages of Ogoja, Cambridge, Cambridge University Press. Crick, M. 1973 ‘Some Reflections on the ASA Decennial Conference’, Journal of the Anthropological Society of Oxford, 4 (3), pp. 176–9. ––– 1976 Explorations in Language and Meaning: Towards a Semantic Anthropology, London, Dent. Crossland, R. 1951 ‘A Reconsideration of the Hittite Evidence for the Existence of ‘Laryngeals’ in Primitive Indo-European’, Transactions of the Philological Society, pp. 88–130.

References

321

Crystal, D. 1971 ‘Prosodic and Paralinguistic Correlates of Social Categories’, in Ardener, E. (ed.), 1971a. Deacon, J. J. 1974 Tongue Tied, London, National Society for Mentally Handicapped Children. De Bono, E. 1969 The Mechanism of Mind, Harmondsworth, Penguin. Delancey, P. and Schraeder, M. 1986 Cameroon – World Bibliographical Series Vol. 63, Oxford, Clio Press. Denison, N. 1968 ‘Sauris: A Trilingual Community in Diatypic Perspective’, Man, (n.s.) 3 (4), pp. 578–92. ––– 1971 ‘Some Observations on Language Variety and Plurilingualism’, in Ardener, E. (ed.) 1971a. Deutsches Kolonialblatt (DKB), Berlin (series). Devereux, G. 1955 A Study of Abortion in Primitive Societies, New York, Julian Press. Dominic, H. 1901 Kamerun: Sechs Kriegs und Friedensjahre in Deutschen Tropen, Berlin. Doroszewski, W. 1933 ‘Quelques Remarques sur les Rapports de la Sociologic et de la Linguistique: Durkheim et F. de Saussure’, Journal de Psychologie, (numéro exceptionnel) 30(1), pp. 82–91. Douglas, M. 1966a Purity and Danger, London, Routledge and Kegan Paul. ––– 1966b ‘Population Control in Primitive Groups’, British Journal of Sociology, 17 (3), pp. 263–73. ––– 1968 ‘The Social Control of Cognition: Some Factors in Joke Perception’, Man (n.s.) 3, pp. 361–76. ––– 1970a Natural Symbols, London, Cresset. ––– (ed.) 1970b Witchcraft Confessions and Accusations (ASA Monographs 9), London, Tavistock. Dresch, P. 1975 ‘Marxist Analyses and Social Anthropology’, Journal of the Anthropological Society of Oxford, 6 (3), pp. 213–17. ––– 1976 ‘Economy and Ideology: an Obstacle to Materialist Analysis’, Journal of the Anthropological Society of Oxford, 7 (2), pp. 55–77. Dronke, U. (ed., intro., and trans.) 1984 The Poetic Edda, Oxford, Oxford University Press. Duke, L., Leacock, E. and Ardener, S. (eds) 1986 Visibility and Power – Essays on Women in Society and Development, Oxford, Oxford University Press. Dumond, D. E. 1965 ‘Population Growth and Cultural Change’, Southwestern Journal of Anthropology, 21, pp. 302–24. Dumont, L. and Pocock, F. F. 1957–66 contributions to Indian Sociology Series, The Hague, Mouton. Dumont, L. 1975 Preface to the French edition of The Nuer, in Beattie J. H. M. and Lienhardt, R. G. (eds) 1975. Durkheim, E. 1912 Les Formes Elémentaires de la Vie Réligieuse, Paris. ––– 1951 Sociology and Philosophy, translated by D. F. Pocock from French edition of 1898, London, Cohen and West. Durkheim, E. and Mauss, M. 1963 Primitive Classification, translated with an introduction by R. Needham, London, Cohen and West. Dwelly, E. 1920 The Illustrated Gaelic-English Dictionary: Containing Every Gaelic Word and Meaning Given in All Previous Published Dictionaries, and a Great Number Never in Print Before, 2nd (standard) edition (reissued by Glasgow, Gairm). Edwards, J. (1985) Language, Society and Identity, Oxford, Basil Blackwell. Ellen, R. F. and Reason, D. 1979 Classifications in their Social Context, New York, Academic Press.

322 References Ellis, J. 1966 Towards a General Comparative Linguistics, The Hague, Mouton. Engler, R. 1967 Edition Critique of Saussure 1916, Wiesbaden, Harrassowitz. ––– 1968 Lexique de la Terminologie Saussurienne, Utrecht/Antwerp, Spectrum. Epale, S. J. 1985 Plantations and Development in Western Cameroon, 1885–1975: or, ‘A study in Agrarian Capitalism’, New York, Vantage Press. Epstein, A. L. 1958 Politics in an Urban African Community, Manchester, Manchester University Press. ––– (ed.) 1967 The Craft of Social Anthropology, London, Tavistock. Evans-Pritchard, E. E. 1934 ‘Imagery in Ngok Cattle Names’, Bulletin of the School of Oriental and African Studies, 7 (3), pp. 623–8. ––– 1937 Witchcraft, Oracles and Magic among the Azande, Oxford, Clarendon Press. ––– 1940 The Nuer, Oxford, Clarendon Press. ––– 1948 ‘Nuer Modes of Address’, Uganda Journal, 12 (2), pp. 166–71, also in Hymes (ed.) 1964. ––– 1950 ‘Social Anthropology, Past and Present’ (Marett Lecture), Man, 50, pp. 118–24. ––– 1954a ‘Zande Texts’, Man, 54, p. 164. ––– 1954b ‘A Zande Slang Language’, Man, 54, pp. 185–6. ––– 1955 ‘Zande Historical Texts I’, Sudan Notes and Records, 36, pp. 123–45. ––– 1956a ‘Zande Historical Texts II’, Sudan Notes and Records, 37, pp. 20–47. ––– 1956b ‘Sanza, A Characteristic Feature of Zande Language and Thought’, Bulletin of the School of Oriental and African Languages, 18, pp. 161–80. ––– 1956c Nuer Religion, Oxford, Clarendon Press. ––– 1957 ‘Zande Historical Texts III’, Sudan Notes and Records, 38, pp. 74–99. ––– 1960 Preface to Hertz 1960. ––– 1961a Anthropology and History, Manchester, Manchester University Press. ––– 1961b ‘A Note on Bird Cries and Other Sounds in Zande’, Man, 61, pp. 19–20. ––– 1962a ‘Three Zande Texts’, Man, 62, pp. 149–52. ––– 1962b ‘Some Zande Texts, pt. 1’, Kush, 10, pp. 289–314. ––– 1962c ‘Ideophones in Zande’, Sudan Notes and Records, 63, pp. 143–6. ––– 1962d Essays in Social Anthropology, London, Faber and Faber. ––– 1962e ‘The Divine Kingship of the Shilluk of the Nilotic Sudan’, in Evans-Pritchard 1962d. ––– 1963a ‘Some Zande Texts, pt. 2’, Kush, 11, pp. 273–307. ––– 1963b ‘Meaning in Zande Proverbs’, Man, 63, pp.4–7. ––– 1963c ‘Sixty-one Zande Proverbs’, Man, 63, pp. 109–12. ––– 1964 The Oxford Library of African Literature series (co-editor with W. Whiteley and R. G. Lienhardt), Oxford, Clarendon Press. ––– 1965a The Position of Women and Other Essays in Social Anthropology, London, Faber and Faber. ––– 1965b ‘The Position of Women in Primitive Societies and in Our Own’, in EvansPritchard, E. E. 1965a. ––– 1965c ‘The Comparative Method’, in Evans-Pritchard, E. E. 1965a. ––– 1965d Theories of Primitive Religion, Oxford, Clarendon Press. ––– 1970 ‘Social Anthropology at Oxford’, Journal of the Anthropological Society of Oxford, 1 (3), pp. 103–9. ––– 1973a ‘Some Reminiscences and Reflections on Fieldwork’, Journal of the Anthropological Society of Oxford, 4 (1), pp. 1–12.

References

323

––– 1973b ‘The Intellectualist (English) Interpretation of Magic’, Journal of the Anthropological Society of Oxford, 4 (3), pp. 123–42. Ferguson, A. 1767 An Essay in the History of Civil Society, London. Ferguson, C. A. 1959 ‘Diglossia’, Word, 15, pp. 325–40, also in Hymes (ed.) 1964. Field, M. J. 1960 Search for Security, London, Faber and Faber. Field, M. 1969 ‘The Prime Minister’s Lodge, Buea’, Buea, Government publication. Finnegan, R. 1969a ‘Attitudes to the Study of Oral Literature’, Man, (n.s.) 4 (1), pp. 59–69. ––– 1969b ‘How to do Things with Words’, Man, (n.s.) 4 (4), pp. 537–52. ––– 1970 Oral Literature in Africa, Oxford, Clarendon Press. Firth, J. R. 1934 ‘Linguistics and the Functional Point of View’, English Studies, 16 (1), pp. 18–24. ––– 1957a Papers in Linguistics, 1934–1951, Oxford, Oxford University Press. ––– 1957b ‘Ethnographic Analysis and Language with Reference to Malinowski’s Views’, in Firth, R. (ed.) 1957 and Palmer (ed.) 1968. ––– 1957c ‘A Synopsis of Linguistic Theory 1930–1955’, in Studies in Linguistic Analysis, Oxford, Basil Blackwell, pp. 1–32. Firth, R. 1940 ‘The Analysis of Mana: an Empirical Approach’, Journal of Polynesian Society, 49, pp. 483–510. ––– 1951 ‘Contemporary British Social Anthropology’, American Anthropologist, 53 (4), pp. 474–89. ––– (ed.) 1957 Man and Culture, an Evaluation of the Work of Malinowski, London, Routledge and Kegan Paul. ––– 1966 ‘The Meaning of Pali in Tikopia’, in Bazell et al. (eds) 1966, pp. 96–115. Firth, Rosemary 1971 ‘Anthropology within and without the Ivory Towers’, Journal of the Anthropological Society of Oxford, 2 (2), pp. 74–82. Fishman, J. A. (ed.) 1968a Readings in the Sociology of Language, The Hague, Mouton. ––– 1968b ‘Sociolinguistic Perspectives on the Study of Bilingualism’, Linguistics, 39, pp. 21–49. Fortes, M. 1945 The Dynamics of Clanship among the Tallensi, Oxford, Oxford University Press. ––– 1949 The Web of Kinship among the Tallensi, Oxford, Oxford University Press. ––– 1970 Time and Social Structure, London, Athlone Press. Fortes, M. and Dieterlen, G. (eds) 1965 African Systems of Thought, Oxford, Oxford University Press. Foster, M. and Brandes, S. (eds) 1980 Symbol as Sense, New York, Academic Press. Fox, R. and Tiger, L. 1972 The Imperial Animal, London, Secker and Warburg. Frake, C. O. 1961 ‘The Diagnosis of Disease among the Subanun of Mindanao’, American Anthropologist, 63, pp. 113–32. Frazer, J. G. 1910 Totemism and Exogamy (4 vols), London, Macmillan. Freedman, M. 1963 ‘A Chinese Phase in Social Anthropology’, British Journal of Sociology, 14 (1), pp. 1–19. Freeman, D. 1962 Review of Leach, E. R. Rethinking Anthropology in Man, 62. Freud, S. 1913 Totem and Taboo, translated and published 1950, London, Routledge and Kegan Paul. Standard edition, vol. 13, London, Hogarth. Fries, C. C. 1963 ‘The Bloomfield “School”’, in Mohrmann, Sommerfelt and Whatmough (eds) 1963, pp. 196–224. Garvin, P. L. (ed.) 1957 Report of the Seventh Annual Round Table Meeting on Linguistics and Language Study, Washington, Georgetown University Press.

324 References Gellner, E. 1968 ‘The New Idealism: Cause and Meaning in the Social Sciences’, in Lakatos, I. and Musgrave, A. (eds) 1968. ––– 1983 Nations and Nationalism, Oxford, Basil Blackwell. ‘Gertrude’ 1978 ‘A Postface to a few Prefaces’, (translated by McDonald, M.) Journal of the Anthropological Society of Oxford, 9 (2), pp. 133–42. Gibbon, E. 1787 The Decline and Fall of the Roman Empire, vol. 3 (reprinted 1954), London, Dent. Gladwin, T. and Sturtevant, W. C. (eds) 1962 Anthropology and Human Behaviour, Washington, Anthropological Society of Washington. Glass, D. V. and Eversley, D. E. C. 1965 Population in History, London, Arnold. Gleason, H. A. 1955a An Introduction to Descriptive Linguistics, New York, Holt, Rinehart and Winston. ––– 1955b Workbook in Descriptive Linguistics, New York, Holt, Rinehart and Winston. Gluckman, M. (ed.) 1964 Closed Systems and Open Minds, Edinburgh, Oliver and Boyd. — 1968 ‘Psychological, Sociological and Anthropological Explanations of Witchcraft and Gossip: a Clarification’, Man, (n.s.) 3, pp. 20–4. Godel, R. 1957 Les Sources Manuscrites du Cours de Linguistique Générale de F. de Saussure, Paris, Droz. ––– (ed.) 1969 A Geneva School Reader in Linguistics, Bloomington, Indiana University Press. Golding, W. 1955 The Inheritors, London, Faber and Faber. Goodenough, W. H. 1956 ‘Componential Analysis and the Study of Meaning’, Language, 32, pp. 195–216. ––– 1957 ‘Cultural Anthropology and Linguistics’, in Garvin, P. L. (ed.) 1957. Goody, J. 1958 The Development Cycle in Domestic Groups, Cambridge, Cambridge University Press. ––– 1966 ‘The Prospects for Social Anthropology’, New Society, 13 October, pp. 574–6. Gosse, E. 1907 Father and Son, London, Heinemann (reprinted Harmondsworth, Penguin 1949). Gould, J. (ed.) 1965 Penguin Survey of the Social Sciences 1965, Harmondsworth, Penguin. GPC, 1964 Geriadur Prifysgol Cymru, pt. XVIII, Cardiff, Gwasg Prifysgol Cymru. GPC, 1968 Geriadur Prifysgol Cymru, pt. XXII, Cardiff, Gwasg Prifysgol Cymru. Grant, I. F. 1961 Highland Folk Ways, London, Routledge and Kegan Paul. Graunt, J. 1662 Natural and Political Observations upon the Bills of Mortality, London, reprinted 1964, Journal of the Institute of Actuaries, 90, pt. 1 (no. 384), pp. 4–61. Green, M. M. and Igwe, G. E. 1963 A Descriptive Grammar of Igbo, Oxford, Oxford University Press. Greenberg, J. H. 1948 ‘Linguistics and Ethnology’, Southwestern Journal of Anthropology, 4, pp. 140–7. ––– 1955 Studies in African Linguistic Classification, New Haven, Compass. ––– (ed.) 1963a Universals of Language, Cambridge, Mass., MIT Press. ––– 1963b The Languages of Africa, The Hague, Mouton. ––– 1966 ‘Synchronic and Diachronic Universals in Phonology’, Language, 42, pp. 508–17. Gumperz, J. J. and Hymes, D. (eds) 1964 The Ethnography of Communication, Menasha, Wis., American Anthropological Association. Guthrie, M. 1948 The Classification of the Bantu Languages, London, International African Institute.

References

325

––– 1953 The Bantu Languages of Western Equatorial Africa, London, International African Institute. ––– 1962 ‘Some Developments in the Pre-history of the Bantu Languages’, Journal of African History, 3, pp. 273–82. Haas, M. R. 1969 The Prehistory of Languages, The Hague, Mouton. Hammel, E. A. (ed.) 1965 Formal Semantic Analysis, American Anthropologist, special publication 67 (5), pt. 2. Hamp, E. P. 1958 ‘Consonant Allophones in Proto-Keltic’, in Lochlann, 1, Oslo, Oslo University Press pp. 209–17. Hardman, C. 1973 ‘Can there be an Anthropology of Children?’, Journal of the Anthropological Society of Oxford, 4, pp. 85–99. Harré, R. and Reynolds, V. (eds) 1984 The Meaning of Primate Signals, Cambridge, Cambridge University Press. Harré, R. and Secord, P. 1972 The Explanation of Social Behaviour, Oxford, Basil Blackwell. Harris, M. 1969 The Rise of Anthropological Theory, London, Routledge and Kegan Paul. Harris, R. (ed.) 1982 Approaches to Language, Oxford, Pergamon Press. Harris, Z. S. 1951 Methods in Structural Linguistics, Chicago, Chicago University Press. Harrison, G. A. and Boyce, A. J. (eds) 1972 The Structure of Human Populations, Oxford, Clarendon Press. Hart, H. L. A. 1961 The Concept of Law, Oxford, Clarendon Press. Hastrup, K. 1978 ‘The Post-structuralist Position of Social Anthropology’, in Schwimmer, E. (ed.) 1978. ––– 1982 ‘Establishing an Ethnicity: the Emergence of the “Icelanders” in the early Middle Ages’, in Parkin, D. (ed.) 1982. ––– 1985 Culture and History in Medieval Iceland, Oxford, Oxford University Press. Hastrup, K., Ovesen, J. and Ramlov, K. 1975 Den Ny Antropologi, Copenhagen, Borgen/ Basus. Hauschild, R. 1964 Die Indogermanische Völker and Sprachen Klienasiens, Berlin, Akademie-Verlag. Hawthorn, G. 1970 The Sociology of Fertility, London, Macmillan. Hayes, E. N. and T. (eds) 1970 Claude Lévi-Strauss: The Anthropologist as Hero, Cambridge, Mass., MIT Press. Hazelwood, A. (ed.) 1967 African Integration and Disintegration, Oxford, Oxford University Press and RIIA. Heelas, P. 1970 ‘Meaning for Whom?’, Journal of the Anthropological Society of Oxford, 1 (1), pp. 1–10. Henson, H. 1971 ‘Early British Anthropologists and Language’, in Ardener, E. (ed.) 1971a. –––– 1974 British Social Anthropologists and Language: a History of Separate Development, Oxford, Clarendon Press. Hertz, R. 1960 (first published 1909) Death and the Right Hand, translated by R. and C. Needham, London, Cohen and West. ––– 1973 ‘The Pre-Eminence of the Right Hand: A Study in Religious Polarity’, translated and reprinted in Needham, R. (ed.) 1973. Highland Society of Scotland, 1828 Dictionarum Scoto-Celticum: A Dictionary of the Gaelic Language (2 vols), Edinburgh, Blackwood. Hjelmslev, L. 1943 Omkring Sprogteoriens Grundlaeggelse, Copenhagen, Munksgaard. ––– 1957 ‘Pour une Sémantique Structurale’, in Hjelmslev (ed.) 1959, pp. 96–112.

326 References ––– (ed.) 1959 Essais Linguistiques, Copenhagen, Nordisk Sprogog Kulturforlag. ––– 1963 Prolegomena to a Theory of Language (translation by F. J. Whitfield of Hjelmslev, 1943; first edition, 1961) Menasha, Wis., University of Wisconsin. Hocart, A. M. 1937 ‘Kinship Systems’, Anthropos 22, pp. 345–51; reprinted in Hocart, 1970. ––– 1970 The Life-giving Myth, edited by R. Needham (2nd edition), London, Methuen. Hockett, C. F. 1954 ‘Chinese versus English: an Exploration of the Whorfian Hypothesis’, in Hoijer, H. (ed.) 1954. ––– 1958 A Course in Modern Linguistics, London, Macmillan. ––– 1968 The State of the Art, The Hague, Mouton. Hoenigswald, H. M. 1960 Language Change and Language Reconstruction, Chicago, Chicago University Press. Hoijer, H. (ed.) 1954 Language and Culture, Chicago, Chicago University Press (second edition 1971). Hollingsworth, T. H. 1969 Historical Demography, London, Hodder and Stoughton. Holt, H. and Pym, H. 1984 A Very Private Eye – the Diaries, Letters and Notebooks of Barbara Pym, London, Macmillan. Homans, G. and Schneider, D. M. 1955 Marriage, Authority and Final Causes, New York, Free Press. Hook, S. (ed.) 1969 Language and Philosophy, New York, New York University Press. Householder, F. W. 1952 review of Jones, D., ‘The Phoneme’, International Journal of American Linguistics, 18, pp. 99–105. ––– 1957 ‘Rough Justice in Linguistics’, in Garvin (ed.) 1957, pp. 153–65. ––– 1970 review of Hockett, 1968 Journal of Linguistics, 6 (1), pp. 129–34. Hudson, D. 1972 Munby, Man of Two Worlds: The Life and Diaries of Arthur J. Munby 1828– 1910, London, Murray. Humboldt, W. von 1836 Über die Verschiedenheit des menschlichen Sprachbaues und ihren Einfluss auf die geistige Entwicklung des Menschengeschlechts, Berlin, Königliche Akademie der Wissenschaften. Photo-reprint, Bonn, Dummlers Verlag, 1967. ––– 1969 De l’Origine des Formes Grammaticales suivi de Lettre à M. Abel Rémusal, Bordeaux, Ducros. Humphrey, C. 1971 ‘Some Ideas of Saussure applied to Buryat Magical Drawings’, in Ardener, E. (ed.) 1971a. Hurault, J. 1969 ‘Eleveurs et Cultivateurs des Hauts Plateaux du Cameroun: la Population du Lamidat de Banyo’, Population, 5, pp. 963–94. Huxley, F. 1966 The Invisibles, London, Hart-Davis. Hymes, D. 1960 ‘lexicostatistics so far’, Current Anthropology, 1 (1), pp. 3–44. ––– 1962 ‘The Ethnography of Speaking’, in Gladwin, T. and Sturtevant, W. C. (eds) 1962. ––– (ed.) 1964 Language in Culture and Society, New York, Harper and Row. ––– 1971 ‘Sociolinguistics and the Ethnography of Speaking’, in Ardener, E. (ed.) 1971a. IPA 1949 The Principles of the International Phonetic Association, London, International Phonetic Association. Ittmann, J. 1953 Volkskundliche und religiöse Begriffe im nördlichen Waldland von Kamerun, Afrika u. Übersee, Beiheft 26, Berlin. ––– 1957 ‘Der Kultische Geheimbund djengu an der Kameruner Kuste’, Anthropos, 52, pp. 135–76. Jackson, A. 1987 Anthropology at Home (ASA monographs 25), London, Tavistock.

References

327

Jackson, K. H. 1953 Language and History in Early Britain, Edinburgh, Edinburgh University Press. Jacquard, A. 1968 ‘Evolution des Populations d’Effectif Limité’, Population, 2, pp. 279–300. Jakobson, R. 1960 ‘Why “Mama” and “Papa”?’, in Kaplan, B. and Wapner, S. (eds) 1960. ––– 1966 ‘Henry Sweet’s Path towards Phonemics’, in Bazell et al. 1966, pp. 242–54. ––– 1968 Child Language, Aphasia and Phonological Universals, The Hague, Mouton. Jakobson, R. and Halle, M. 1956 Fundamentals of Language, The Hague, Mouton. Jarvie, I. C. 1963 The Revolution in Anthropology, London, Routledge and Kegan Paul. Jenkins, T. 1975 ‘Althusser’s Philosophy’, Journal of the Anthropological Society of Oxford, 6 (1), pp. 1–17. ––– 1977 ‘The Death of Marx: a Media Event’, Journal of the Anthropological Society of Oxford, 8 (3), pp. 116–24. Jerschina, J. 1983 ‘Polish Modernism and Malinowski’s Personality’, unpublished symposium paper, Krakow, 27 September. Johnson, S. and Boswell, J. (1775/1785) Journey to the Western Islands and A Tour to the Hebrides (edited by R. W. Chapman, 1930, and subsequent editions), Oxford, Oxford University Press. Jones, D. 1962 The Phoneme: Its Nature and Use, revised edition, Cambridge, Heifer. ––– 1964 The History and Meaning of the Term ‘Phoneme’, London, International Phonetic Association. Jones, G. and Jones, T. J. 1949 The Mabinogion, London, Dent. Jones, Sir W. 1799 ‘The Third Anniversary Discourse on the Hindus’, The Works of Sir William Jones 6 vols, London, Robinson and Evans; reprinted in Lehmann (ed.), 1967. Joos, M. (ed.) 1957 Readings in Linguistics I; The Development of Descriptive Linguistics in America 1925–1956 (reprinted 1967), Chicago, Chicago University Press. Kaberry, P. M. 1939 Aboriginal Woman: Sacred and Profane, London, Routledge and Kegan Paul (reprinted Gregg International, 1970). ––– 1952 Women of the Grassfields, London, HMSO (reprinted Gregg International, 1970). ––– 1974 ‘A Glimpse of Malinowski in Retrospect’, Journal of the Anthropological Society of Oxford, 5 (2), pp. 104–8. Kaplan, B. and Wapner, S. (eds) 1960 Perspectives in Psychological Theory, New York, International Universities Press. Kardiner, A. and Preble, E. 1961 They Studied Man, London, Secker and Warburg. Kilmister, C. W. 1967 Language, Logic and Mathematics, London, English Universities Press. King, R. 1969 Historical Linguistics and Generative Grammar, Englewood Cliffs, Prentice Hall. Koelle, S. W. 1854 Polyglotta African, London, Church Missionary Society. Koerner, K. 1984 Edward Sapir: Appraisals of his Life and Work, Amsterdam and Philadelphia, John Benjamins. Kondo, S., Kawai, M. and Ehara, A. (eds) 1975 Contemporary Primatology, Basel, Karger. Krahe, H. 1962–3 Indogermanische Sprachwissenschaft, vol. 1, 1962, vol. 2, 1963, Berlin, De Gruyter. Krapf-Askari, E. 1969 Yoruba Towns and Cities, Oxford, Clarendon Press. Kroeber, A. L. 1909 ‘Classificatory Systems of Relationship’, Journal of the Royal Anthropological Institute, 39, pp. 74–84. Kuper, A. 1973 Anthropologists and Anthropology: The British School 1922–72, London, Allen Lane. ,

328 References La Fontaine, J. (ed.) 1972 The Interpretation of Ritual, London, Tavistock. Laine-Kerjean, C. 1942/3 ‘Le Calendrier Celtique’, Zeitschrift fur Celtische Philologie, 23, pp. 249–84. Lakatos, I. and Musgrave, A. (eds) Problems in the Philosophy of Science, Amsterdam, North Holland. Langendoen, D. T. 1968 The London School of Linguistics: A Study of the Linguistic Theories of B. Malinowski and J. R. Firth, Cambridge, Mass., MIT press. Leach, E. R. 1954a review of Durkheim, 1951, Man, 54, pp. 92–3. ––– 1954b Political Systems of Highland Burma, London, Bell. ––– 1957 ‘The Epistemological Background to Malinowski’s Empiricism’, in Firth, R. (ed.) 1957. ––– 1958 ‘Concerning Trobriand Clans and the Kinship Category Tabu’, In J. Goody (ed.) 1958. ––– 1961a Rethinking Anthropology, London, Athlone Press. ––– 1961b ‘Two Essays Concerning the Symbolic Representation of Time’, in Leach, E. R. 1961a. ––– 1962 ‘Classification in Social Anthropology’, Aslib Proceedings, 14 (8), pp. 239–42. ––– 1964 ‘Anthropological Aspects of Language: Animal Categories and Verbal Abuse’, in Lenneberg, E. H. (ed.) 1964. ––– 1969 Genesis as Myth and Other Essays, London, Cape. ––– 1970 Lévi-Strauss, London, Fontana/Collins. ––– 1971a ‘More about Mama and Papa’, in R. Needham (ed.) 1971. ––– 1971 b Review of M. Douglas, Natural Symbols, New York Review of Books. Leach, E. R., Jarvie, I. C., Gellner, E., et al 1966 ‘Frazer and Malinowski: on the Founding Fathers’, Current Anthropology, 7 (5), pp. 560–76 (earlier, in part, in Encounter, October, 1965). Lees, R. B. 1953 ‘The Basis of Glottochronology’, Language, 29, pp. 113–27. Lefebvre, H. 1966 Le Langage et la Société, Paris, Gallimard. Lehmann, W.P. (ed. and trans.) 1967 A Reader in Nineteenth Century Historical Indo-European Linguistics, Bloomington, Indiana University Press. Lehmann, W. P. and Malkiel, Y. (eds) 1968 Directions for Historical Linguistics, Austin, University of Texas. Leitner, G. W. 1874 discussion in Journal of the Royal Anthropological Institute, 4, pp. 212– 14. Lenneberg, E. H. (ed.) 1964 New Directions in the Study of Language, Cambridge, Mass., MIT Press. Lenneberg, E. H. and Roberts, J. M. 1956 The Language of Experience, Baltimore, Waverly Press, Indiana University Publications in Anthropology and Linguistics 13; reprinted in Saporta (ed.) 1961. Leopold, J. 1973 ‘Tylor’s Solar Sixpence’, Journal of the Anthropological Society of Oxford, 4 (1), pp. 13–16. Leroy, M. 1967 The Main Trends in Modern Linguistics, Oxford, Basil Blackwell. Leroy, R. 1922 ‘Syndrome of Lilliputian Hallucinations’, Journal of Nervous and Mental Disease, 56, p. 325. Leroy-Gourhan, Z. 1964 Le Geste et La Parole, Paris, Albin Michel. Le Vine, V. and Nye, R. 1974 Historical Dictionary of Cameroon (African Historical Dictionaries, no. 1), Metuchen, New Jersey, The Scarecrow Press.

References

329

Lévi-Strauss, C. 1949 Les Structures élémentaires de la Parenté, Paris, Presses Universitaires de France (2nd edition 1967). ––– 1953 ‘Social Structure’, Anthropology Today (ed. Kroeber), Chicago, Wenner Gren Foundation (see also Structural Anthropology, 1963) ––– 1955 Tristes Tropiques, Paris, Plon. ––– 1958 Anthropologic Structurale, Paris, Plon. ––– 1962a Le Totemisme Aujourd’hui, Paris, Presses Universitaires de France. ––– 1962b La Pensée Sauvage, Paris, Plon. ––– 1963a Structural Anthropology (translation of Lévi-Strauss, 1958), New York, Basic Books. ––– 1963b Totemism (translation of Lévi-Strauss, 1962a, by R. Needham), Boston, Beacon Press. ––– 1964 Mythologiques 1: Le Cru et le Cuit, Paris, Plon. ––– 1966a The Savage Mind (translation of Levi-Strauss, 1962b), London, Weidenfeld and Nicolson. ––– 1966b Mythologiques 2: Du Miel aux Cendres, Paris, Plon. ––– 1966c Introduction to third edition of Mauss, 1950, Paris, Presses Universitaires de France. ––– 1967a The Scope of Anthropology (translation of Leçon Inaugurale Faite Le Mardi 5 Janvier 1960), London, Cape. ––– 1967b Les Structures Elémentaires de la Parenté (2nd edition) Paris, Presses Universitaires de France. ––– 1968 Mythologiques 3: L’Origine des Manières de Table, Paris, Plon. ––– 1969 The Elementary Structures of Kinship, translated and edited by R. Needham (translated from revised edition (1967) of Levi-Strauss, 1949), Boston, Beacon Press. ––– 1970 The Raw and the Cooked (translation of Levi-Strauss, 1964), London, Cape. Lévy-Bruhl, L. 1910 Les Fonctions Mentales dam les Sociétés Inférieures, Paris. Lewin, K. 1935 A Dynamic Theory of Personality, New York, McGraw-Hill. ––– 1936 Principles of Topological Psychology, New York, McGraw-Hill. ––– 1952 Field Theory in Social Science, London, Tavistock. Lewis, H. 1943 Yr Elfen Ladin yn yr laith Gymraeg, Cardiff, Gwasg Prifysgol Cymru. Lewis, I. M. (ed.) 1968 History and Social Anthropology, London, Tavistock. ––– 1971 Ecstatic Religion, Harmondsworth, Penguin. Lienhardt, R. G. 1961 Divinity and Experience, Oxford, Clarendon Press. Loomis, R. S. 1956 Wales and the Arthurian Legend, Cardiff, University of Wales Press. Lorimer, F., et al. 1954 Culture and Human Fertility, UNESCO, Paris. Lounsbury, F. G. 1956 ‘A Semantic Analysis of the Pawnee Kinship Usage’, Language, 32, pp. 158–94. ––– 1965 ‘Another View of Trobriand Kinship Categories’, in Hammel, E. A. (ed.) 1965. ––– 1969 ‘Language and Culture’, in Hook (ed.) 1969. Lowes, J. L. 1927 The Road to Xanadu: A Study in the Ways of the Imagination, London, Constable. Lyons, J. 1968 Introduction to Theoretical Linguistics, Cambridge, Cambridge University Press. ––– 1970 Chomsky, London, Fontana/Collins. Lyons, J. and Wales, R. J. (eds) 1966 Psycholinguistic Papers, Edinburgh, Edinburgh University Press. MacAlpine, N. 1832 Pronouncing Gaelic-English Dictionary, reissued 1973, Glasgow, Gairm. Macaulay, D. (ed.) 1976 Nua-Bhardachd Glulidhlig (Modern Scottish Gaelic Poems), Edinburgh, Southside.

330 References McCall, G. 1970 ‘£.s.d. Varna’, Journal of the Anthropological Society of Oxford, 1 (3), pp. 131–6. McDonald, M. E. 1978 ‘Language “At Home” to Educated Radicalism’, Journal of the Anthropological Sociey of Oxford, 9 (1), pp. 13–34. ––– 1982 ‘Social Aspects of Language and Education in Brittany’, Oxford, unpublished D. Phil thesis (forthcoming, 1989, London, Tavistock). Macdonald, S., Holden, P. and Ardener, S. (eds) 1987 Images of Women in Peace and War, London, Macmillan. Macintosh, D. 1819 Collection of Gaelic Proverbs and Familiar Phrases, Edinburgh, Stewart. McLuhan, M. 1970 Counterblast, London, Rapp and Whiting. Maguire, M. 1974 ‘Criminology and Social Anthropology’, Journal of the Anthropological Society of Oxford, 5 (2), pp. 109–17. Mair, L. 1972 ‘Recent Writings on Witchcraft’, Journal of the Anthropological Society of Oxford, 3 (1), pp. 33–41. Malinowski, B. 1926a ‘Anthropology’, Encyclopaedia Britannica, 13th edition s.v. ––– 1926b ‘Anthropology’, typescript of 1926a presented to E. E. Evans-Pritchard, Tylor Library, Oxford. ––– 1932 special foreword to The Sexual Life of Savages, 3rd edition, London, Routledge and Kegan Paul. ––– 1935 Coral Gardens and their Magic, vol. 2, 2nd edition 1966, with an introduction by Berry, J., London, Allen and Unwin. ––– 1944 A Scientific Theory of Culture, University of North Carolina Press. Malmberg, B. 1964 New Trends in Linguistics, Stockholm, Lund. Mandelbaum, D.G. (ed.) 1949 Selected Writings of Edward Sapir, Los Angeles, University of California Press. ––– 1956 Edward Sapir: Culture, Language and Personality, Los Angeles, University of California Press. Maquet, J. J. 1964 ‘Objectivity in Anthropology, Current Anthropology, 5 (1), pp. 47–55. Mathieu, N.-C. 1973 ‘Homme-Culture et Femme-Nature? ’, L’Homme, July–Sept., pp. 101–13. Mauss, M. 1950 Sociologie et Anthropologie, 3rd edition, 1966, Paris, Presses Universitaires de France. ––– 1954 The Gift, translated by I. Cunnison, with an introduction by E. E. Evans-Pritchard (corrected edition, 1969), London, Cohen and West. Maxwell, J. C. 1871 The Theory of Heat, 3rd edition, 1872, London. Maybury-Lewis, D. 1965 ‘Prescriptive Marriage Systems’, Southwestern Journal of Anthropology, 21, pp. 207–30. ––– 1967 Akwe-Shavante Society, Oxford, Clarendon Press. Meek, C. K. 1931 Tribal Studies in Northern Nigeria, London, Kegan Paul, Trench, Trubner. Meillet, A. 1933 ‘Sur le Bilinguisme’, Journal de Psychologie, 30, pp. 167–71. ––– 1937 Introduction à l’Etude comparative des Langues indo-européennes, reprinted 1964, Alabama, Alabama University Press. ––– 1950 Les Dialectes indo-européens, Paris, Champion. Meyer, R. 1910 ‘Bedeutungssysteme’, Kuhns Zeitschrift fur Vergleichende Sprachforschung, 43, pp. 352–68. Milner, G. B. 1954 Review of J. Perrot, La Linguistique, Man, 54, p. 172. ––– 1966 ‘Hypostatization’, in Bazell et al. 1966, pp. 321–34. ––– 1969 ‘Siamese Twins, Birds and the Double Helix’, Man, (n.s.) 4, pp. 5–23.

References

331

––– 1971 ‘The Quartered Shield: Outline of a Semantic Taxonomy’, in Ardener, E. (ed.) 1971. Missionary Herald, Baptist Missionary Society series, London. Mitchell, J. C. 1966 ‘Theoretical Orientations in African Urban Studies’, in The Social Anthropology of Complex Societies Banton, M. (ed.), London, Tavistock. ––– 1967 ‘On Quantification in Social Anthropology’, in Epstein, A. (ed.) 1967. Mohrmann, C., Norman, F. and Sommerfelt, A. 1963 Trends in Modern Linguistics, Utrecht/ Antwerp, Spectrum. Mohrmann, C., Sommerfelt, A. and Whatmough, J. (eds) 1963 Trends in European and American Linguistics 1930–1960, Utrecht/Antwerp, Spectrum. Moore, O. K. and Olmsted, D. L. 1952 ‘Language and Professor Lévi-Strauss’, American Anthropologist, 54 (1), pp. 116–19. Morris, C. 1946 Signs, Language and Behavior, Englewood Cliffs, Prentice Hall. Mowatt, D. (ed. and trans.) 1962 The Nibelungenlied, London, Dent, Everyman. Munro, J. (no date) Gaelic Vocabulary and Phrase Book, reissued Glasgow, Gairm. Murdock, G. P. 1951 ‘British Social Anthropology’, American Anthropologist, 53 (4), pp. 465–73. ––– 1957 Africa: Its Peoples and Culture History, New York, McGraw-Hill. ––– 1965a Culture and Society, Pittsburgh, University of Pittsburgh Press. ––– 1965b ‘Cross-language Parallels in Parental Kin Terms’, in Murdock, G. P. 1965a. Nadel, S. F. 1957 The Theory of Social Structure, Cohen and West, London. Needham, R. 1954 ‘The System of Teknonyms and Death-names of the Penan’, Southwestern Journal of Anthropology, 10, pp. 416–31. ––– 1958 ‘A Structural Analysis of Purum Society’, American Anthropologist, 60, pp. 75–101. ––– 1960a ‘Descent Systems and Ideal Language’, Philosophy of Science, 27, pp. 96–101. ––– 1960b ‘The Left Hand of the Mugwe: an Analytical Note on the Structure of Meru Symbolism’, Africa, 30, pp. 20–33. ––– 1962 Structure and Sentiment: A Test Case in Social Anthropology, Chicago, Chicago University Press. ––– 1963 Introduction to Durkheim and Mauss 1963, London, Routledge and Kegan Paul. ––– 1967 ‘Right and Left in Nyoro Symbolic Classification’, Africa, 37 (4), pp. 425–52. ––– 1970 ‘The Future of Social Anthropology: Disintegration or Metamorphosis?’, in Anniversary Contributions to Anthropology: Twelve Essays, Leiden, Brill. ––– (ed.) 1971 Rethinking Kinship and Marriage (ASA Monographs 11), London, Tavistock. ––– (ed.) 1973 Right and Left: Essays on Dual Symbolic Classification, Chicago, University of Chicago Press. Nicholson, A. 1881 A Collection of Gaelic Proverbs and Familiar Phrases, based on Mackintosh’s Collection, Edinburgh, Stewart. Njeuma, M. 1987 ‘Pioneer in Cameroon Studies – E. W. Ardener’, obituary notice, West Africa, 5 October, (3660), pp. 1963–5. Nutini, H. G. 1965 ‘Some Considerations on the Nature of Social Structure and Model Building: a Critique of Claude Lévi-Strauss and Edmund Leach’, American Anthropologist, 67, pp. 707–31; also in Hayes E. N. and T. (eds) 1970, pp. 70–107. O’Connor, F. 1967 The Backward Look, A Survey of Irish Literature, London, Macmillan. Ojo, G. 1966 Yoruba Culture: a Geographical Analysis, London, University of London Press. O’Rahilly, T. 1972 Irish Dialects Past and Present, Dublin, Institute for Advanced Studies (1st edition 1932).

332 References Ornstein, R. 1975 ‘Temporal Dimensions of Consciousness’, New Scientist, 27 March, pp. 772–3. Ortega y Gasset, J. 1921 El Tema de Nuestra Tiempo, Madrid. ––– 1961 The Revolt of the Masses, London, Allen and Unwin. Ortner, S. B. 1973 ‘Is Female to Male as Nature is to Culture?’, Feminist Studies, 1 (2); also in Rosaldo M. Z. and Lamphere, M. (eds) 1974. Osthoff, H. and Brugmann, K. 1878 introduction to Morphologische Untersuchungen auf dem Gebiete der Indogermanischen Sprachen, 1, pp. iii–xx, Leipzig, Hirzel. Overing, J. (ed.) 1985 Reason and Morality (ASA Monographs 24), London, Tavistock. Ovesen, J. 1978 ‘Maurice Godelier and the Study of Ideology’, Journal of the Anthropological Society of Oxford, 9 (1), pp. 1–12. Paine, R. 1967 ‘What is Gossip About? An Alternative Hypothesis’, in Man, (n.s.) 2, pp. 278–85. ––– 1968 ‘Gossip Transaction’, Man, (n.s.) 3, pp. 305–8. Palmer, F. R. (ed.) 1968 Selected Papers of J. R. Firth 1952–1959, London, Longman. Palmer, L. R. 1954 The Latin Language, London, Faber and Faber. Parkin, D. (ed.) 1982 Semantic Anthropology (ASA Monographs 22), London, Academic Press. Parry, H. B. (ed.) 1974 Population and its Problems, Oxford, Clarendon Press. Parry, T. 1939 ‘Y marwfis, y mis du’, Bulletin of the Board of Celtic Studies, 9 (1), pp. 40–2. Paul, H. 1880 Prinzipien der Sprachgeschichte, Tübingen, Niemeyer (8th edition, 1968). Pedersen, H. 1931 Linguistic Science in the Nineteenth Century, translated by J. Spargo, Cambridge, Mass., Harvard University Press; reissued Bloomington, Indiana University Press, 1962. Piaget, J. 1971 Structuralism, London, Routledge and Kegan Paul. Pike, K. L. 1954, 1955, 1960 Language in Relation to a Unified Theory of the Structure of Human Behaviour, 3 vols, Glendale Summer Institute of Linguistics. ––– 1956 ‘Towards a Theory of the Structure of Human Behaviour’, reprinted in Hymes, D. (ed.) 1964. Pocock, D. F. 1961 Social Anthropology, London, Sheed and Ward. ––– 1973 ‘The Idea of a Personal Anthropology’, ASA conference paper (unpublished). ––– 1974 ‘Nuer Religion: a Supplementary View’, Journal of the Anthropological Society of Oxford, 5 (2), pp. 69–79. Popper, K. R. 1957 The Poverty of Historicism (originally published 1944–5; paperback 1961), RKP, London. Pride, J. B. 1971 ‘Customs and Cases of Verbal Behaviour’, in Ardener, E. (ed.) 1971a. Pym, B. 1955 Less than Angels, London, Macmillan. Radcliffe-Brown, A. R. 1957 A Natural Science of Society, Glencoe, Free Press. ––– 1977 ‘Two Letters from Radcliffe-Brown to Evans-Pritchard’, 8 (1), pp. 49–50. Rees, A. and Rees, B. 1961 Celtic Heritage, London, Thames and Hudson. Reyher, R. H. 1952 The Fon and his Hundred Wives, New York, Doubleday. Reynolds, V. 1975 ‘Problems of Non-comparability of Behaviour Catalogues in Single Species of Primates’, in S. Kondo et al. (eds) 1975, pp. 280–6. Richards, A. I. 1956 Chisungu: A Girls’ Initiation Ceremony among the Bemba of Northern Rhodesia, London, Faber and Faber. Richards, E. 1982 A History of the Highland Clearances – Agrarian Transformation and the Evictions 1746 – 1886, London, Croom Helm.

References

333

Richards, M. 1950 ‘Y mis du’, Bulletin of the Board of Celtic Studies, 13 (4), pp. 204–5. Richardson, I. 1956–7 Linguistic Survey of the NW Bantu Borderland, 2 vols, London, International African Institute. Rivers, W. H. R. 1911 address to the Portsmouth meeting of the British Association. Robins, R. H. 1967 A Short History of Linguistics, London, Longman. ––– 1971 ‘Malinowski, Firth, and the “Context of Situation”’, in Ardener, E. (ed.) 1971a. Rosaldo, M. Z. and Lamphere, M. (eds) 1974 Woman, Culture and Society, Stanford, Stanford University Press. Ross, A. 1958 On Law and Justice, London, Stevens. Ross, A. S. C. 1958 Etymology with Special Reference to English, London, Deutsch. Ruel, M. 1969 Leopards and Leaders: Constitutional Politics among a Cross River People, London, Tavistock. Sapir, E. 1916 Time Perspective in Aboriginal American Culture: a Study in Method, Ottawa, Canada Geological Survey. ––– 1921 Language: an Introduction to the Study of Speech, New York, Harcourt. Saporta, S. (ed.) 1961 Psycholinguistics, New York, Holt, Rinehart and Winston. Saussure, F. de 1878 Mémoire sur le Système Primitif des Voyelles dans les Langues Indo-Européennes, Geneva, now translated in Lehmann (ed.) 1967. ––– 1916 (2nd edition 1922) Cours de Linguistique Générale, Paris/Geneva, Payot. ––– 1964 Course in General Linguistics, translated from Saussure 1916 by W. Baskin, New York, Philosophical Library, 1959; London, Peter Owen, 1960 (reprinted 1964). Savitsky, N. and Tarachow, S. 1941 ‘Lilliputian Hallucinations during Convalescence of Scarlet Fever’, Journal of Nervous and Mental Disease, 93, pp. 310–12. Schwimmer, E. (ed.) 1978 Yearbook of Symbolic Anthropology, London, Hurst. Sebeok, T. A., Hayes, A. S. and Bateson, M. C. 1964 Approaches to Semiotics, The Hague, Mouton. Sechehaye, A. 1933 ‘La pensée et la langue ou: comment concevoir le rapport organique de l’individuel et du social dans le langage?’, Journal de Psychologie, (numéro exceptionnel) 30 (1), pp. 57–81. ––– 1940 ‘Les trois linguistiques saussuriennes’, Vox Romanica V, pp. 1–48, Zurich; reprinted in Godel (ed.) 1969. Seligman, C. G. 1910 The Melanesians of British New Guinea, Cambridge, Cambridge University Press. Seligman, C. G. and Seligman, B. Z. 1911 The Vedas, Cambridge, Cambridge University Press. Shannon, C. E. 1948 ‘A Mathematical Theory of Communication’, Bell System Technical Journal, 27, pp. 379–423; also in Shannon and Weaver, 1949. Shannon, C. E. and Weaver, W. 1949 The Mathematical Theory of Communication, Urbana, University of Illinois. Shaw, G. B. 1916 Pygmalion, Harmondsworth, Penguin, 1941. Simonis, Y. 1968 Claude Lévi-Strauss, ou la ‘Passion de l’Inceste’, Paris, Montaigne. Smith, A. G. (ed.) 1966 Communication and Culture, New York, Holt, Rinehart and Winston. Southall, A. (ed.) 1961 Social Change in Modern Africa, Oxford, Oxford University Press. Spencer, P. 1965 The Samburu, London, Routledge and Kegan Paul. ––– 1973 Nomads in Alliance: Symbiosis and Growth among the Rendille and Samburu of Kenya, Oxford, Oxford University Press.

334 References Sperber, D. 1975 Rethinking Symbolism, Cambridge, Cambridge University Press. Spier, L., Hallowell, A. and Newman, S. (eds) 1941 Language, Culture and Personality: Essays in Memory of Edward Sapir, Menasha Wis., Sapir Memorial Publications Fund. Srinivas, M. N. 1977 ‘The Changing Position of Indian Women’, Man, (n.s.) 12 (2), pp. 221–38. Steiner, F. 1956 Taboo, London, Cohen and West (also 1967, Pelican). Steiner, G. 1975 After Babel: Aspects of Language and Translation, Oxford, Oxford University Press. Stone, B. G. 1929 ‘Notes on the Buea District’ (manuscript), Victoria Divisional Office, Cameroon. Sturtevant, E. 1942 The Indo-Hittite Laryngeals, Baltimore, Linguistic Society of America. Sturtevant, W. 1964 ‘Studies in Ethnoscience’, American Anthropologist, 66 (2), pp. 99–131. Swadesh, M. 1950 ‘Salish Internal Relationships’, International Journal of American Linguistics, 16, pp. 157–67. Talbot, P. A. 1912 In the Shadow of the Bush, London, Heinemann. ––– 1926 The Peoples of Southern Nigeria, Oxford, Oxford University Press. Tambiah, S. J. 1968 ‘The Magical Power of Words’, Man, (n.s.) 3 (2), pp. 175–208. Taylor, D. 1958 ‘On Anthropologists’ Use of Linguistics’, American Anthropologist, 60, p. 940. Teitelbaum, M. S. 1972 ‘Factors Associated with the Sex Ratio in Human Populations’, in Harrison, G. and Boyce, A. (eds) 1972. Thomas, K. V. 1971 Religion and the Decline of Magic, London: Weidenfeld and Nicolson. Thomas, N. W. 1910 Anthropological Report on the Edo-speaking Peoples of Nigeria, pt. 1, London, Harrison. ––– 1914 Specimens of Languages from Southern Nigeria, London, Harrison. Thompson, D’Arcy, W. 1966 On Growth and Form, Cambridge, Cambridge University Press. Thurneysen, R. 1899 ‘Der Kalender von Coligny’, Zeitschrift für Celtische Philologie, 2, pp. 523–44. Thurnwald, R. 1912 Forschungen auf den Salomon-Insel and dem Bismarck-Archipel, Berlin. Tolkien, J. R. R. 1954 The Return of the King, The Lord of the Rings, vol. 3 London, Allen and Unwin (Danish edition, 1974, Ringenes Herre, Copenhagen, Gyldendals Bogklub). Tonkin, J. E. A. 1971a ‘The Use of Ethnography’, Journal of the Anthropological Society of Oxford, 2 (3), pp. 134–6. ––– 1971b ‘Some Coastal Pidgins of West Africa’, in Ardener, E. (ed.) 1971a. Trier, J. 1931 Der Deutsche Wortschatz im Sinnbezirk des Verstandes, Heidelberg, Carl Winter. Trubetzkoy, N. 1933 ‘La Phonologic Actuelle’, Journal de Psychologie, (numéro exceptionnel) 30 (1), pp. 227–46. ––– 1968 Introduction to the Principles of Phonological Descriptions, The Hague, Nijhoff. Turlot, F., Durupt, M. J. and Holin, F. 1969 La Population Du Cameroun Occidental, Société d’Etudes pour le Développement Economique et Sociale, Paris. Turnbull, C. M. 1972 ‘Demography of Small-scale Societies’, in Harrison, G. and Boyce, A. (eds) 1972. Turnbull, H. W. 1959 The Correspondence of Isaac Newton, vol. 1, Cambridge, Cambridge University Press. Turner, V. W. 1964 ‘Symbols in Ndembu Ritual’, in Gluckman, M. (ed.) 1964. ––– 1965 ‘Ritual Symbolism, Morality and Social Structure among the Ndembu’, in Fortes, M. and Dieterlen, G. (eds) 1965.

References

335

––– 1966 ‘Colour Classification in Ndembu Ritual’, in Banton, M. (ed.) 1966. ––– 1967 The Forest of Symbols: Aspects of Ndembu Ritual, Ithaca, Cornell University Press. ––– 1969 The Ritual Process, London, Routledge and Kegan Paul. Ullman, S. 1951 The Principles of Semantics, Glasgow, Jackson. ––– 1959 The Principles of Semantics, 2nd edition, with additions, Oxford, Basil Blackwell. ––– 1963 ‘Semantic Universals’, in Greenberg (ed.) 1963a. Van Gennep, A. 1909 Les Rites de Passage, Paris. Vendler, Z. 1967 Linguistics in Philosophy, Ithaca, Cornell University Press. Vendryes, J. 1921 ‘Le Caractère Social du Langage et la Doctrine de F. de Saussure’, Journal de Psychologie, 18, pp. 617–24; also in Vendryes, 1952, pp. 18–25. ––– 1933 ‘Sur les Tâches de la Linguistique Statique’, Journal de Psychologie, 30, pp. 172–84. ––– 1952 Choix d’Etudes Linguistiques et Celtiques, Paris, Klincksieck. Ventris, M. and Chadwick, J. 1956 Documents in Mycenaean Greek, Cambridge, Cambridge University Press. Voegelin, C. F. and Harris, Z. S. 1945 ‘Linguistics in Ethnology’, Southwestern Journal of Anthropology, 1 (4), pp. 455–65. Voegelin, C. F. and Harris, Z. S. 1947 ‘The Scope of Linguistics’, American Anthropologist, 49 (4), pp. 588–600. Waismann, F. 1968 The Principles of Linguistic Philosophy Harré, R. (ed.) London, Macmillan. Wallerstein, E. 1960 ‘Ethnicity and National Integration in West Africa’, Cahiers d’Etudes Africaines, 1 (3), pp. 129–39. Wartburg, W. von 1943 Einführung in Problematik und Methodik der Sprachwissenschaft, Tübingen, Max Niemeyer. ––– 1969 Problems and Methods in Linguistics (translation of Wartburg, W. von 1943), Oxford, Basil Blackwell. Weinreich, U., Labov, W. and Herzog, M.I. 1968 ‘Empirical Foundations for a Theory of Language Change’, in Lehmann and Malkiel (eds) 1968. Wells, J. 1906 The Oxford Degree Ceremony, Oxford, Clarendon Press. Wells, R. S. 1947 ‘De Saussure’s System of Linguistics’, Word, 3 (1), pp. 1–31; reprinted in Joos (ed.) 1957. Whiteley, W. 1966 ‘Social Anthropology, Meaning and Linguistics’, Man, (n.s.) 1 (2), pp. 139–57. ––– 1971a Language Use and Social Change, Oxford, Oxford University Press. ––– 1971b ‘A Note on Multilingualism’, in Ardener, E. (ed.) 1971a. Whorf, B. L. 1952 Collected Papers on Metalinguistics, Washington, Department of State, Foreign Service Institute. ––– 1964 (1956, 1st edition) Language, Thought and Reality, edited by J. Carroll, Cambridge, Mass., MIT Press. Wiener, N. 1948 Cybernetics (2nd edition 1961), Cambridge, Mass., MIT Press. Wilson, B. (ed.) 1970 Rationality, Oxford, Basil Blackwell. Winch, P. 1958 The Idea of a Social Science and its Relation to Philosophy, London, Routledge and Kegan Paul. Withycombe, E. G. 1969 The Oxford Dictionary of English Christian Names, Oxford, Clarendon Press. Wittgenstein, L. 1963 Philosophical Investigations, Oxford, Basil Blackwell (first English edition, 1953).

336 References Woolf, V. 1928 A Room of One’s Own, London, Hogarth. Worsley, P. 1966 ‘The End of Anthropology?’, paper for 6th World Congress of Sociology (mimeographed). Wright, E. M. 1913 Rustic Speech and Folklore, Oxford, Oxford University Press. Y Cymro 1970 March 25. Yeats, W. B. 1950 The Collected Poems of W. B. Yeats (2nd edition), London, Macmillan.

Index

Apology: the following index must be regarded primarily as a citation and author index. As a subject index, it is subsidiary to the titles of the papers themselves, which provide broad-ranging cross-reference that has not been replicated below: for example, ‘social anthropology’ and ‘population’ have not been cross-referenced in relation to chapter 7, nor ‘social anthropology’ and ‘language’ in relation to chapter 1. The entire volume might be regarded as a multi-dimensional cross-reference between social anthropology and linguistics, society and language, and persistent cross-reference of this kind must therefore be assumed. To attempt to capture this indexically would only be to recreate, in a less-accessible form, the volume itself. Many words used by Ardener are both ubiquitous and central to his work, and at the same time transformed by his treatment, which makes them unsuitable index headings – symbol, belief, and structure (for example). Furthermore, many aspects of Ardener’s thought are realized in dualities and oppositions (which are often simultaneously dissolved), in a conceptual space of many dimensions; again, an index can only aspire to provide hints of this. Lastly, Ardener was always anxious to keep his conceptual structures moving, and was quick to abandon a phrase or idea that seemed to risk becoming merely formulaic or conventional. The continuity and development of his own work is, therefore, not necessarily terminologically expressed within itself. His work demands to be read and pondered, not consulted. Adamnan, 309 Life of St Columba, 309 Age-structure, of tribal populations, 114–16 Althusser, L., xxxv, 46, 54, 60–2, 197 Anderson, A.O., 272 Anderson, M.O., 272 Andersson, T.M., 272

Anglo-Saxon, as hyphenated ethnicity, 67 l’Année sociologique, 19 Arbitrariness, 10, 18, 42, 55, 178, 210; and automatism, in language, 178–185; in classification, 164; in ethnic definition, 111; in Mousterian interment, 183; in the social, 157

338 Index Ardener, E., ‘Belief and the Problem of Women’, xix Cameroon Government publications, xvii Coastal Bantu of the Cameroons, xvi ‘A Directory study of Social Anthropologists’, xxiv Divorce and Fertility, viii ‘Documentary and Linguistic evidence for the rise of the trading polities between Rio del Rey and Cameroon, 1500–1650’, xvii ‘Edward Sapir (1884–1939), xxiv ‘Evidences of Creation’, xxiv ‘The “Kamerun” idea’, xvii Kingdom on Mount Cameroon: documents for the history of Buea, 1844–1898, xvii, xxiv ‘The nature of the reunification of Cameroon’, xvii Plantation and Village in the Cameroons, xvii ‘The Political History of Cameroon’, xvii ‘Social Anthropology and the Historicity of Historical Linguistics’, xxiv Social Anthropology and Language, ASA monograph 10 (ed.), 1 ‘Social and demographic problems of the Southern Cameroons Plantation Area’, xvi ‘Witchcraft, Economics and the continuity of Belief ’, xvii, xxiii Ardener, S., xvi, xvii, xix, xx, xxi, xxiv, xxv, xxvii, 78, 84, 117, 118, 119, 122, 127, 128, 130, 152, 183, 212, 221, 283, 284, 285, 299, 305, 307 Armstrong, R. A., 138 Arnold, Matthew, 216 Association of Learned Societies in the Social Sciences, xi Association of Social Anthropologists of Britain and the Commonwealth (ASA), xix, xxiv, xxv, 2, 205; annual conferences: 1968, xxiv; 1969, xxiv, I, 4; 1973, 129; 1982, xxxiv; 1984, xxxiv; 1985, xxxiv; 1987, xx; for monographs, see under individual editors Astounding Magazine, 270

Atlantis, not discovered, 144 Austen, Jane, 183 Automatisms: and arbitrariness, 178–185; of language, 100, 143, 171, 173; in Mousterian interment, 183; of social space, 221 Avars, 113 Azande, 107; witchcraft, empirically unfalsifiable, 53 Bakweri, xvi, 72, 93, 107–8, 127–8, 299–300; agriculture, 78; ‘dying out’, 121–2; ethnicity, 65; as ‘remote’, 215–16, 221; rituals, 79 (see also iengu); women, 77–84, 158; zombie-beliefs, 79, 84, 88–94, 99, 121, 221 Bally, C., 17, 22, 249, 289 Baldwin, Stanley, 120 Bamenda plateau, 215 Bamileke, 66, 70 Banaji, J., 62 Bangwa, 288 Bantu, tribal and linguistic units, 66–9, 114 Banyang, 288 Barltrop, R., 303 Barnes, J. A., 47, 56, 110, 294 Barnes, R., 261 Barnett, C., 120 Barth, F., 55, 56, 110, 295 Barthes, R., 17, 23, 25, 88, 197, 290 Bates, R. H., 111 Bateson, G., 210, 263, 290 Beals, R. L., 287 Beattie, J. H., 5, 10, 36, 210, 253, 284 Beck, B. E., 36 Beeston, F., 286 Behaviour, 105–8, 303; behaviourism: 31, 40, 52, 62, 93, 167; animal, 106, 160–3; as an area of semiotics, 147; primate, as culturally outlandish, 160–3; reversals, 151 Beidelman, T. O., 36 Belief, inadequacy of the category, 52, 75, 279–81 Bellenden, John, 272 Benedict, B., 122, 125 Benedict, R., 263

Index Benveniste, E., 18, 35 Berchenshaw, 287 Bergsland, K., 244 Berlin, B., 13, 163, 164, 169, 172, 288 Berlitz, 7 Bernstein, B., 26, 29, 295 Berry, J., 38 Biafran War, 111, 123, 125 Bini, 68 Blacking, J., 37 Blake, William, 305 Blanc, R., 115 Bloch, M., 164, 183 Bloomfield, L., 18, 30–2, 35, 258, 259, 260, 290, 294, 312, 313 Language, 30 Bloomsburies, the, 199 Boas, F., 3, 6, (Boasians), 198, 256, 257, 260 Body classification, 14, 25, 39, 184, 303–4; English and Ibo, 166–72; French, Gaelic, Rumanian and Welsh, 171; and sub-category boundaries, 169–70 Boece, Hector, 272 Bogatyrev, P., 197, 290 Bohannan, L., 125 Bohannan, P., 287 Bopp, F., 47, 286, 293 Borges, J. L., 162 Ficciones, 162 Boundary features (pollution, taboo, liminality, etc.), 11–12, 39, 76–7, 90–2, 131–3, 167 Bovin, M., 300 Bowen, E. S., 300 Bott, 56 Boyce, A. J., 109, 110, 286 Braithwaite, R. B., 149 Brandes, S., 285 Breton identity, 216–17 Bricolage, 42, 89, 148 Brontë, C., 196 Jane Eyre, 196 Brontë, E, 305 Brown, R. W., 13 Brugmann, K., 3, 42, 238, 242, 243, 246, 248, 251, 252, 259, 293, 312, 313

339

Buchanan, G., 272 Buea, 116; provincial archives, xvi, xvii, xviii Burling, R., 43 Buyssens, E., 22, 23 Byron, Lord, 305 Byzantium, 155 Callan, H., 119, 163, 304 Cameroon, xvi–xxvii, xxxiii, 65–8, 70, 72, 77–8, 81, 115, 118, 121, 123, 128, 173, 288; as ‘remote’, 214–18, 221 Campbell, Lord Archibald, 139 Campbell, J. F., 174 Cannibalism, 116 Carmichael, A., 139, 141, 301 Capell, A., 15, 287 Carnap, R., 22, 290 Carroll, J. B., 15, 164, 165 Carroll, Lewis, 217 Through the Looking Glass, 217 Casagrande, J. B., 35 Cassirer, K, 26 Catastrophe theory, 154 Category, see classification Celtic: identity, definition of, 216; as ‘nature’, 162; other world, in the Antipodes, 213; time systems, 135, 140; twilight, 12, 287–8 Chadwick, H. M., 113, 272, 314 Chagnon, N. A., 110, 111, 116 Chapman, M., xix, xx, xxiv, xxxiii, 216, 217, 285, 286, 309; xxiii–xliv The Gaelic Vision in Scottish Culture, 216 History and Ethnicity, ASA monograph 27 (ed., with Tonkin and Mcdonald), xxiv Chilver, E. M., xvii, 8, 128 Chomsky, N., xxix, 3, 18, 23, 24, 26, 31, 40–2, 50, 54, 194, 238, 248, 249, 252, 260, 263, 291–3, 295, 306; his movement, as a mythical system, 42; compared to Lévi-Strauss, 40–2; 291–3; Hockett’s critique of, 293 Aspects of the Theory of Syntax, 292 Current Issues in Linguistic Theory, 292 Syntactic Structures, 40, 292 Chowning, A., 39

340 Index Churchill, W. S., 198–9 Clarke, J., xix, 284, 291 Classification, xxxii, 9–16, 21, 38–40, 50– 1, 107, 158, 163–72; of body, 165–72; and cross-cultural misunderstanding, 165–8; and density, 223, 275; of events, 87; of food, 123; linguistic and tribal, 66–70; and materiality, 164–5, 178–82, 281; and social space, 212 Coconut, the, glory has departed from,145, 148–9, 154 Cohen, A., 212 Belonging (ed.), 12 Cohen, Abner, 110, 111, 112, 284 Cohen, I. B., 287 Cohen, L. J., 9, 293 Cohen, M., 19 Colby, B., 246 Coleridge, S. T., 144, 302 Ancient Mariner, 144 Kubla Khan, 144 Coligny Calendar, 135–6 Collinder, B., 22, 289 Colonialism, impact of 116; inducing stability, 203; and primitivism, 205; and tribal classification, 67–8 Colour classification, 9–13, 143, 163–4, 287–8 Colson, 295 Communication Theory, 31, 49 Comparative philology, see historical linguistics Comte, A., 59, 61 Conklin, H. C., 9, 11 Cooper, D., 309 Corbusier, 199 Courtenay, B. de, 30, 35 Cowgill, W. 313 Cox, J. C., 117 Crabb, D. W., 67, 243 Crick, M., xxxiv, 159, 163, 170, 207, 303 Crossland, R.A., 312 Crystal, D., 286 Cultural Anthropology (American), 46, 209 Cultural Materialism, 51, 60, 294 Culture/Nature opposition, 76–8, 82–5, 89, 131–3, 162, 182, 300

Cunnison, I., 36 Custer, General Charles, 94, 111, 112, 113, 121 Czech school of folklore analysis, 197 Darwin, C., 155 Deacon, J. J., 147, 302 Dead stretch, 94, 96–8, 143, 147 Deconstruction, xxxv Definition space, 149–54; and catastrophe theory, 154 Delancey, xvii Delbrück, B., 312 Demography, 66, 109–10; and demographic consciousness, 117–18; folk–, 114–17, 121–4; and pressure on food, 118–19; and social adaptation, 124–6 Denison, N., 296, 297, 298, 299 Density, within category; see subcategory phenomena Derrida, J., 194, 197 Descartes, R., 292 Devereux, G., 122 Divine kings, buried alive?, 116 Dominance, problem of, 186–90; theory of, as general anthropological theory, 188 Dominant structures, 127, 130–3, 183, 186–90, 221–3, 300 Dominion over palm and pine, rather hearty, 62 Doroszewski, W., 19 Dorsey, J., 261 Douglas, M., xvii, xxiv, 9, 11, 15, 36, 55, 56, 61, 69, 76, 84, 167, 204, 266, 284, 294, 296, 309, 316; the ‘oysters and champagne’ argument, 118–19, 125–6 Witchcraft Confessions and Accusations, ed., xxiv Dronke, U., 272 Duala, 69, 79 Dualisms, in social anthropology, collapse of, xxix–xxx, xxxv, 279–81 Dube, L., 186, 285 Dumezil, G., 197 Dumont, L., 8, 36, 176

Index Duncan Mitchell, G., 285 Durkheim, E., 9, 18, 19, 22–3, 26, 35, 36, 44, 59–61, 195, 198, 212, 294, 304 Elementary Forms of the Religious Life, 60, 195, 198 Dwelly, E., 136, 137, 138, 139, 141, 142, 301, 304 Edo, 67 Edwards, J., 307 Efik, 66, 69, 288 Einstein, A., 64, 199, 257 Ejagham, 288 Ekoi, 66, 67, 68 Eliot, T. S., 199 Ellen, R. F., 303 Ellis, J., 248 Emic/etic, 31, 35 Empiricism, xxviii, xxx–xxxvii, 40, 60, 93, 108, 144, 279–81; in functionalism, 47; inadequately opposed to idealism, 144; and world–structure, 146; see also world-structure Engels, F., 19, 61, 62 England: ethnicity in, 111; handshake in, 166–72 Engler, R., 17, 289 English, colour terminology, 9–13 Epale, S. J., xvi Epistemological break, 46, 57, 148, 205, 208, 210 Epstein, A. L., 7, 10, 203 Esu, xvi, 118–19, 215 Ethnicity: and demography, 114–26; and history, xx, 110–11; and identity, 211; and population, 65–71, 112–13; and self-definition, 110–17, 121; smallscale, 121, 125 Ethnography, see fieldwork Ethnolinguistics, 40 Ethologism, 119, 304 Ethology: and language, 159–63; animal and human vocalization, 179, 182, 304 Evans-Pritchard, E. E., xviii, 3, 7, 8, 17, 19, 44, 53, 55, 59, 72, 107, 116, 125, 135, 148, 152, 153, 162, 176, 232,

341

237, 245, 260, 265, 291, 294, 296, 307, 317; encouraged interest in language, 35–7 The Nuer, 36, 55, 296 Witchcraft, Oracles and Magic, 55, 296 Event density, 211, 221–3, 275 Events, 71, 86–8, 90–101, 107–8, 134– 148, 159, 178, 221–3, 275-6, 302 Eversley, D. E., 117 Evolutionism, 155, 156, 157, 202, 279 Exchange systems, as communication, 33 Female/Male opposition, 127–33, 174, 187–90, 300; in Ibo, 12; in Bakweri, 72–83, 89 Ferguson, A., 36, 176, 212, 307 An Essay in the History of Civil Society, 307 Ferguson, C. A., 14, 297, 298 Fertility, social constraints on, 122–3; and ethnicity, 124–6 Field, M., 234 Fieldwork, xxxi, 62, 88–9, 94, 275; E. W. Ardener’s, xvi–xviii, xx, xxiii; M. Fortes’, 5–6; ‘at home’, 211–13; and interpreters, 5; and language, 5–7; as modernist technical advance, 201; and supposed stability of primitive societies, 203–4; and women, 72–5, 84–5, 128 Finnegan, R., 36 Firth, J. R., 4, 16, 18, 29, 38, 40, 48, 180, 291, 296 Firth, R., xv, 2, 37, 38, 287 Fishman, J. A., 246 Fitzgerald, 174 Folk-demography, 115–26 Folk-etymology, 39–40 Forde, D., xv Forster, E. M., 203 Fortes, M., 5, 6, 237, 263 Dynamics of Clanship among the Tallensi, 5 Foster, M., 285 Fox, R., 119, 132 The Imperial Animal, 119 Frake, C. O., 9 Frazer, J. G., 198, 210, 283 Totemism and Exogamy, 198

342 Index Freedman, M., 8, 294 Freedom, from structure, 148, 154 Freud, S., 33, 54, 59, 61, 199, 203, 257, 267, 287 Fries, C. C., 290 Fulani, 68, 70 Fulbe, 70 Functionalism, xxviii, xxix, 7–8, 18, 27, 42, 52, 55, 89, 279, 281; ad–hoc terminology of 63; from ‘function to meaning’, 37; high-, 194; as modernism in social anthropology, 193–210; observational model of, 75; and primitivism, 203; relation of, to structuralism, 45, 62; and sociolinguistics, 295–6 Galileo, G., 56 Geertz, C., 209 Gellner, E., 194, 198, 203, 306, 307 Genre and life, 196–7, 205–6 (see also literary criticism; textuality and life) Germanic peoples, 113 Gerontocracy, 123–4 Gibbon, E., 112, 113, 156 Decline and Fall of the Roman Empire, 156 Gibbons, S., 43 Cold Comfort Farm, 43 Glass, D. V., 117 Gleason, H. A., 29, 31, 244 Glottochronology, 3, 8, 244–5 Gluckman, M., 7, 36, 150, 156, 203, 295 Gödel, R., 289 Godelier, M., 195, 207 Golding, W., 26 The Inheritors, 26 Goodenough, E. H., 9, 35 Goody, J., 294 Gosse, E., 248, 264, 269, 314 Goths, 112, 155 Grant, I. F., 136 Grant, J. S., 309 Graunt, J., 117 Green, M. M., 128, 287, 303 Greenberg, J. H., 8, 35, 242–3, 258, 259, 312

Gumperz, J. J., 244 Guthrie, M., 8, 66, 67, 242, 242, 312 Gypsys, as muted group, 133 Haas, M., 313 Halle, M., 31, 35, 38, 49, 312 Hamp, E. P., 20 Handshake, in English and Ibo, 166–72 Hardman, C., xix, 130 Harré, R., 180, 284, 285 Harris, M., 2, 51, 52, 55, 57, 60, 61, 62, 294, 306 Harris, R., 284 Harris, Z. S., 31, 32 Methods in Structural Linguistics, 31 Harrison, G. A., 109, 110 Hart, H. L., 22 Hastrup, K., xv, xxvii, 195, 265, 274–7, 303 Hausa, 70 Hauschild, R., 245, 313 Hawthorn, G., 122 Hayes, E. N. and T., 290 Hazelwood, A., xvii, 283 Hebrides, 218–21, 309 Hegel, F., 60, 61 Henson, H., 3, 4, 172, 245, 286, 303, 304 Hermeneutics, 194 Hertz, R., 17, 36, 43, 132, 174, 187 Herzfeld, M., vii–xiii Herzog, M., 250 Highland Clearances, 216, 308 Highlands and Islands Development Board (HIDB), 219, 309 Highlands of Scotland, 216, 218, 308 Historical linguistics, 3, 31, 33, 143, 286, 291 Historicist anthropology, wiped out by Malinowski, 205 Hjelmslev, L., 4, 9, 11, 22, 23, 30, 36, 49, 163, 249, 287, 290, 296 Hocart, A. M., xxxv, 39 Hockett, C. F., 17, 31, 42, 290, 293, 295, 297, 328 Hodder, Dr, 306 Hoenigswald, H.M., 313

Index Hoijer, H., 9, 312 Holden, P., xix Holinshed, Raphael, 272 Hollingsworth, T. H., 117 Hollis, M., xxxv Hollow category, 69, 70, 71, 136 Holt, H., 309 Homans, G., 52 Hook, S., 9, 293 Householder, F. W., 17, 31, 32, 43 Hudson, D., 305 Human Sciences, at Oxford, xix, 105, 107 Humboldt, W. von, 259, 288 Humphrey, C., 37, 286 Huns, 112; ‘swarming’ of, 113 Hurault, J., 114 Hurst, M., xx Huxley, F. 310 Huxley, T. H., 106 Hymes, D., 4, 8, 243, 286, 291, 292, 295, 296, 299, 312 Ibibio, 67 Ibo, xvi, 67, 68, 70, 89, 107–8, 119, 123, 128, 182, 303; behaviour terminology, 106–7; colour terminology, 9–12, 287; cultural time in, 140, 143–4, 148; definition of ‘tribe’ in, 111; demography, 125; handshake, 166–72; kinship terminology, 169; as ‘remote’, 213–214; women’s uprising, 127, 128 Idealism, xxx, xxxi, 101, 171; fear of 143; see also world–structure Identity, 71, 211–12, 216–17 Igwe, G. E., 303 Ikwerri, 111 Information theory, 3 Institute of Social Anthropology, Oxford, xvii International African Institute, 214 International Phonetic Alphabet, 162 International Phonetic Association, 29 Ipsen, G., 288 –isms, declaration of, 199–200 Isubu, 69, 78 Ittmann, J., 79, 300, 310

343

Jackson, A., xxxiv, 285 Anthropology at Home, ASA monograph 25, xxxiv Jackson, K. H., 14, 21 Jacquard, A., 122 Jakobson, R., 18, 28, 29, 30, 31, 35, 38, 41, 49, 63, 88, 180, 253, 259, 290, 296 Japanese culture, 89 Jarvie, I. C., 194, 198, 203, 210, 230, 237, 283, 291, 306, 315 Jerschina, J., 206 Johnson, Dr Samuel, 308 Jolles, A., 288 Jones, D., 31, 259 Jones, G., and Jones, T. J., 12 Jones, Sir William, 291 Joos, M., 290 Journal of the Anthropological Society of Oxford, xxii, xxxi, xxxii, xxxiii, xxxv Joyce, J., 199, 202 Kaberry, P. M., xv, xvii, 8, 128, 299 Aboriginal Women, 128 Women of the Grassfields, 128 Karcevskij, S., 30 Kardiner, A., 37 Kay, P., 13, 163, 164, 169, 172, 287 Kennedy, President, J. F., 193, 306 Keynes, M., 199 Kilmister, C., 239 King, R., 23, 251, 252, 312 Kingsley, C., 251, 269, 314 Kinship: ‘atom of ...’, 31; fictive, 113, 122, 125; systems, and recruitment, 124 Kinship terminology, 10, 13–14, 21, 124, 143; semantic density within, 169–70 Koelle, S. W., xx, 291 Koerner, K., 255, 256, 257, 260, 263, 314 Koestler, A., 203 Kohn, T., 309 Kole, 69, 78 Krahe, H., 312, 313 Kroeber, A., xxxv, 33, 170, 256, 257, 288 Kruszewski, 30 Kuhn, T., 88 *kuku, 178–9

344 Index Kuper, A., 198, 204, 306 Labov, W., 250 La Fontaine, J., 128, 133, 284, 299–300 Lacan, J., 194, 197 Laine-Kerjean, C., 135 Language: and categories of thought, 5, 7; diatypes in, 297–9; match of, with world-structures, 148, 152, 176–80; problems of definition of, 66, 67, 70, 111; as ‘refuge from materiality’, xxxiii; relation to other semiotics, 147; and the social, xxix, 96–104, 146; stretching of, in novel situations, 145, 153, 208, 213; typologies of 294, 297–9 Language shadow, 91, 94, 95, 100, 103, 104, 148, 215 Langendoen, D. T., 37, 40 Langue/parole, 17, 19, 23, 28, 29, 41, 295, 296 Lawrence, D. H., 199, 203 Le Vine, V., xviii Leach, E. R., xv, 9, 10, 13, 16, 37, 38, 39, 41, 48, 52, 56, 59, 76, 135, 151, 162, 180, 194, 198, 203, 204, 237, 238, 246, 249, 250, 252, 253, 265, 266, 268, 283, 291, 294, 296, 306, 312 Leacock, E., 285 Lean-jawed traveller, the, 6 Lees, R.B., 243 Lefebvre, H., 64 Lehmann, W. P., 250, 251, 291 Leibnitz, G., 292 Leith-Ross, S., 126 Leitner, G. W., 4 Lennenberg, E. H., 13 Leroy, M., 313 Leroy, R., 234, 312 Leroy-Gourhan, Z., 179, 183 Lévi-Strauss, C., xxvii, xxix, xxxv, 3–4, 9, 15–18, 22–36, 38, 40–3, 46, 49–63, 73, 76–7, 79, 84, 88–9, 131–2, 154, 159, 162, 171, 194–7, 204–5, 207, 234, 237, 238, 248, 249, 250, 252, 253, 254, 264, 266, 267, 287, 290–4, 296, 311; as anthropologist of genre,

197; bricolage, 42, 234, 270, 311; compared with Chomsky, 40–2, 248–50, 291–3; correlates language groups and kinship systems, 34; ‘mechanical’ and ‘statistical’ models, 34, 237-8, 252–3; struggles with linguistic insights, 27–35 ‘Language and the analysis of social laws’, 33 ‘Linguistics and anthropology’, 34 Mythologiques, 54, 76, 79 La Pensée sauvage, 34, 35, 41 ‘Postscript’, 35 Les Structures élémentaires de la Parenté, 33, 267 Tristes Tropiques, 196 Lévy-Bruhl, L., 44, 198 Lewis, H., 11, 14 Lewis, I. M., xvii, 90, 152, 283 Lexicostatistics, 8 Liengu, Bakweri rites surrounding, 78–84, 89, 299–300 Lienhardt, R. G., 36, 284 Linguistics: American anthropological, 9, 35, 56, 57, 295; Chomskyan, 241–3; and history, 8, 13–14, 20; London school of, 3, 37, 40; Neogrammarian, 3, 4, 42, 287; revulsion for, in social anthropology, 31, 184; and the social space, 178–80; see also Prague school Linguistic sign, 17–18; and ritual sign, 24–5 Literary criticism, 209; out of phase with social anthropology, 191–2; and poststructuralism, 195; structuralist praxis for, 197 Loch, J., 216, 308 Locke, J., 290, 292 London School of Economics (LSE), xv, xvi Loomis, R. S., 12, 288, 308 Lorimer, F., 122, 125 Lost generation, the, 120–1, 124 Lounsbury, F. G., 9, 10, 39 Lowes, J. L., 144, 145, 302 The Road to Xanadu, 144 Lukes, S., xxxv, 195 Lyons, J., 292, 293, 313 Lysenko, 59

Index McCabe, 195 Mac a’ Ghobhainn, I. (Ian Crichton Smith), 308 Macalpine, N., 137, 138, 142, 304 Macaulay, D., 308, 309 McDandruff, Platypus, 270 McDonald, M., xv, xx, xxiv, xxxii, 216, 217, 279, 285 Macdonald, S., xix Mach, Z., 287 McLuhan, M., 26, 290 McNeill, 292 Macpherson, J. (Ossian), 174, 216 Maguire, M., xix, 130 Magyars, 112 Malinowski, B., xv, xxv, 3–7, 16–19, 26, 28, 35–40, 43, 45–6, 59, 62, 156, 172, 192–4, 198–9, 200–7, 209, 237, 256, 257, 263, 283, 284, 286, 291, 293, 294, 295, 296, 305, 306; announces functionalism, 193–4; and context of situation, 39; and homonyms, 39; and language, 3, 5–7, 36–40; and modernism in social anthropology, 198–204; and Polish literary modernism, 206; compared to Sapir, 256, 263 Argonauts of the Western Pacific, 207 Coral Gardens and their Magic, 37, 40 The Sexual Life of Savages, 202, 203, 306, 310 Malkiel, V., 250 Malmberg, B., 14, 17, 287 Mama and papa, 180–1 Mandelbaum, D., 255, 260, 261, 262 Maquet, J. J., 294 Marriage stability, xxviii, xxxii, 110, 121, 150; parameter collapse, in theory of, 150–1 Martens, F., 145 Voyage in to Spitzbergen and Greenland, 145 Martins, H., 306 Marr, 59 Marx, K., 19, 35, 59, 60, 61, 62, 176, 199, 207 Marxism, xxxiv, xxxv, 60; Althusser’s position in, xxxv, 61; and critique of

345

structuralism, 46, 62, 280; as ‘flatearthism’, 59; ‘like’ structuralism, 58; as limited problematic, xxxvi, 62, 208; as modernism, 205, 207–8; and ‘muted groups’, 129–30, 133; as quasi–positivist system, 57; and rejection of relativism, 164; relationship to ‘new anthropology’, 61; structural, collapse of, 194, 197, 205, 207 Materialism, 101, 146; failure of, in dealing with world structure, 188, 208 Materiality, never free from conceptualization, 208 Mathieu, N.–C., 127, 129, 132, 133, 300 Mauss, M., 9, 27, 33, 35, 36, 44, 55, 61, 287, 303 Maxwell, J. (Maxwell’s Demon), 33, 43, 53, 58, 253 Maybury–Lewis, D., 7, 51 Mboko, 78 Mead, M., 128, 290 Measurement and definition, xxviii, xxxii, xxxiii, 48–50, 117, 136–7, 185, 210, 298; in parameter collapse, 149–50 Mechanical and statistical models, 34, 303–5 Meek, C. K., 8 Meillet, A., 240, 242, 244, 245, 312 Men, articulacy of, 74; and models of society, 74–6, 83–5 Meyer, R., 288 Mickiewicz, A., 206 Miller, D., 306 Milner, G., 4, 63, 286, 287, 312 Mitchell, J. C., 47, 56, 110 Mode of production, and muted groups, 129–30 Mode of registration, 87–8, 95 Mode of specification, 91–2, 93, 95, 300 Models: use of, in social anthropology, 7, 31, 42, 88, 289; ‘male’, in fieldwork, 73 Modernism, 191–210, 305–6; early criticism of, 193, 209; end of, in social anthropology, 197; key figures in, 199; and Marxism, 207–8; and the

346 Index moderne, 199, 204–5; as movement of manifestos, 192; in Poland, 206; and primitivism, 203; and romanticism, 192, 305–6 Mohrmann, C., 290 Mongols, 112 Moore, H., 306 Moore, O. K., 34 Morris, C., 35, 290 Mowatt, D. 272 Müller, M., 3, 139; interest in myth, 286 Munby, A. J., 186, 305 Munro, J., 304 Murdock, G. P., 2, 157, 180, 243 Mussolini, B., 120 Muted groups, 129, 183; and class, 129–30, 133; and prophecy, 148, 152 Nadel, S. F., 252, 263, 291 Natural language users, underestimation of, 169 Ndembu symbolism, 24 Needham, C., 36 Needham, R., xxxv, 2, 9, 10, 17, 36, 39, 43, 47, 50, 52, 57, 63, 75, 132, 174, 234, 250, 288, 289, 294, 296 Rethinking Kinship and Marriage, ASA monograph 11 (ed.), 2 Structure and Sentiment, 47 Neo-anthropology, see post–functionalism New Archaeology, the, 194, 306 Newton, Isaac (Newtonian), 9, 10, 287, 293 Ngie, 115, 116 Ngolo-Batanga, 67 Nicholson, A., 138, 141, 142, 301 Nigeria, xvi, xvii, 66–8, 111, 125, 128, 140, 213–14, 303 Nigeria, xvii Nigerian Institute of Social and Economic Research, see West African Nixon, Richard (Nixonian), 151, 152 Nixonian White House, as isolated worldstructure, 151–2 Njeuma, M., xvi, xvii, 286 Norman, F., 290 Norway, 111 Nuer, 176; prophets, 148, 152–3

Nutini, H., 252, 253 Nye, R., xvii O’Connor, F., 135, 301 Ojo, G., 168, 182 Oli, 79 Olmstead, D. L., 34 Onomatopes, 178–82 O’Rahilly, T., 258 Ornstein, R., 302 Ortega y Gasset, J., 55, 193, 209, 306 Ortner, S. B., 300 Osthoff, H., 42, 251, 312 Overing, J., xxxiv, 2, 285 Reason and Morality, ASA monograph 24 (ed.), xxxiv Ovesen, J., 305 Oxford University Women’s Studies Committee, xix Paine, R., 296 Palmer, L.R., 312 Paluch, A., 286 Panchronic approach, 21, 23–4, 98, 290 Paradigm shift, 148, 151 Paradigm and Syntagm, 46, 55–6, 63–4, 88, 93, 96–9, 295, 296, 300 Parameter collapse, 149–52, 302 Parkin, D., xxv, xxxiv, 2, 207, 285, 286 Semantic Anthropology, ASA monograph 22 (ed.), xxxiv Parman, S., 309 Parry, H. B., 284 Parry, T., 301 Passy, P. 259 Paul, H., 312 Pearse, P., 301 Pedersen, H., 47, 312 Phoneme, 27–32, 43, 290; Bloomfieldian and Prague views, 30–2; first use of, 30 Phonetics, and Henry Sweet, 28–9, 290 Piaget, J., 197, 266 Picasso, P., xxxii, 199, 203 Pidgins and Creoles, 294, 297–8 Pierce, C. S., 290 Pike, K. L., 9, 15, 28, 35 Pippidi, A., 171 Plekhanov, G., 61

Index Pocock, D. F., xxix, 8, 36, 37, 176, 237, 294, 307 Pop, M., 171 Popper, K., 55, 57 Population, and hollow category, 69–71 Population homeostasis, 118, 121, 125–6 Porzig, W., 288 Positivism, xxx, xxxi, 43, 48, 55, 57–60, 62; critique of, 279–81; and demography, 71; pseudo-, 62; quasi-, 57–8; as ‘religion of the masses’, 59; and time, 142 Post-functionalism, 2, 42, 46, 57, 61–2, 104, 204, 294, 306 Post-modernism, 205–6 Post-structuralism, xxxv, 86, 104, 195, 279, 306 Prague school linguistics, 22, 28, 30–5, 41, 59 Preble, E., 37 Predictive statements, 49–53, 58, 135, 146 Prescription and preference, 50–2, 55, 139 Pride, J. B., 286, 296 Primitivism: and colonialism, 205; and modernism, 203, 306 Programme analogy, 48–9, 53, 55, 5960, 63–4, 87, 107; and myth-logic, 53–4 Prophecy, xxvii, xxx, 134–5, 146, 148, 152–4, 222 Propp, V., 197 P-structures, 90–102, 148, 210, 300; and ethnocentricity, 94 Psychoanalysis, 57, 58 Pym, B., 214, 217, 285, 308 Less than Angels, 308 Pym, H., 309 Radcliffe-Brown, A. R., 14, 19, 55, 57, 59, 159, 209, 291 Rationality, debate surrounding, xxxv, 46, 210 Reason, D., 303 Recruitment component, 113, 124, 158 Rees, A., and Rees, B., 135 Reflexivity, xxviii, 163, 182, 207, 212 Relativism, 10, 11, 15, 60, 166, 183, 185, 208, 212–13, 280; as philosophical

347

bogey, 212; useless confusion surrounding, 307; versus universals, 164 Remote area: Cameroon as, 214–16; event-density in, 221; paradoxes of, 218–21; Western Scotland as, 216–17 Renaissance, 156 Renan, E., 216 Rendille, 114 Reyher, R. H., 123 Reynolds, V., 160, 285 Rice-Davies, M., 164 Richards, A. I., xv, 72, 83, 128, 263, 299, 300, 301 Richards, E., 308 Richards, M., 249 Richardson, I., 243 Ricoeur, P., 194 Rites of passage, 83 Roberts, J. M., 13 Robins, R. H., 37, 286 Roedean, girls at, compared to bush-pigs, 85 Roman Empire, 155–7 Ross, A., 22 Ross, A.S.C., 239, 240, 246, 251 Rousseau, J.–J., 132 Royal Anthropological Institute, xxiv, xxv Ruel, M., 311 Russian, translation of personal names into English, 174–5 Samburu, 114, 123 Sapir, E., xxiv, 3, 9, 24, 30, 32, 165, 253–63, 296, 314 Sapir-Whorf hypothesis, 165, 261–3 Sartre, J.–P., 57 Saussure, F. de, xxi, xxii, 3, 9, 16–24, 26–9, 32, 35, 39, 47, 56, 59, 63, 88, 98, 168, 179, 194, 197, 199, 200, 201, 236, 238, 240–2, 246, 248, 249, 250, 252, 256, 259, 260, 280, 286, 287, 288, 289, 290, 296, 297, 299, 306; and chess-analogy, 20–2, 240–1; influence of French sociology on, 22 Cours de Linguistique générale, 17, 23, 194, 256, 260, 288–9

348 Index Savitsky, N., 312 Schleicher, A., 287 Schneider, D., 52 Schraeder, xvii Schuchardt, H., 289 Schwimmer, E., 284 Scott, Sir Walter, 16 Scottish Gaelic, xx; areas, as remote, 216– 21; cultural time in, 135–44, 148, 177, 301–2; identity, 216–18; names, translation of, 176–7, 185, 304 Sebeok, T. A., 290 Sechehaye, A., 17, 19, 26, 289 Secord, P., 180 Seligman, C. G., 198 Melanesians of British New Guinea, 198 Sellar, P., 216, 308 Semantic anthropology, 159, 170, 207 Semantic field, 15, 105, 143, 288 Semantic density, see sub-category phenomena Semantic materialism, 173 Semantics: social, 189; in linguistics, 289–90 Semiology, 18, 22–3, 25, 35, 41, 147, 287, 290 Semiotic, 23–7, 146–7, 182–4, 289–90; frames, 182; non-linguistic, 24–7, 41, 43–4, 55–6, 100, 103, 179, 290 Serbo-Croat, 67 Sex-ratios, 116–17 Shannon, C. E., 32, 49, 292 Shaw, G. B., 29, 290 Pygmalion, 29 Signifier/signified, 17, 24, 168, 178 Simonis, Y., 26, 31, 32, 33, 41 Simultaneity, xxix, xxxii, 55, 93, 95, 104, 158, 171–3, 178, 207–8, 302; of behaviour and category, 168; of classification and action, 184; of definition and action, 184; of discovery and definition, 280; of language, category and object, 169 Singularity, in world-structure, 148, 149, 151–3, 215, 221–3 Sioux, 111, 113 Slays, 113

Social, the: compared to language, 99; as manifold of ideas and action, 108; as manifold of people and reality, 130; as self-defining system, 157; as manifold of thought and behaviour, 86 Social adaptation, 118–20, 124–6, 155–8 Social fitness, see social adaptation Social links with biology, 62, 64; through demography, 65–71; through environment, 182; through infrastructure, 98–100, 146; through language, 181–2; through perception, 172–3; through recruitment criteria, 158 Social Science Research Council, xix Social Space, 134, 142, 146–8, 180, 211, 218; as classification, 172; as definition space, 154, 159; events in, 221–3, 280; neither idealist nor materialist, 184; and prophecy, 152; singularity in, 215; translation of, 178, 184 Socialist Party of Great Britain, 158, 303 Sociolinguistics, 9, 17, 40, 105, 108, 294, 296–8 Sociology, 7; destroyed by statistics, 62; French school of, 19, 22, 23, 27, 36, 59, 289; separation of theoretical and empirical in, 108 Sommerfelt, A., 20, 290 Southall, A., 283 Space: ‘real’, the search for, 143–5; specialism, in West, 213 Spencer, P., 114, 118, 123 Sperber, D., 268 Srinivas, M. N., 186 S-structures, 90–101, 148, 300 Statistics, social anthropological reaction to, 47, 109–110 Steane, C., 299 Steiner, F., 304 Steiner, G., 160, 174, 183, 304 Stendhal (Henri Beyle), 183 Stocking, 206 Stone, B. G., 299 Structural–functionalism, 194, 204, 209–10, 279; narrow intellectual range of, 207

Index Structuralism, xxxv, ch. 1 passim, 56–62, 89, 134, 192–7, 204–8, 279–81, 305–6; and binary oppositions, 24, 132, 153–4, 187–8; collapse of vogue for, 153; cultural materialist critique of, 51, 60 and 294; and empiricism, 159; failure of, 171, 196–7, 205, 207, 209; High-, 142, 159; High-, break with, 185, 205, 208; as last phase of modernism, 207; and Lévi-Strauss, 42, 60; Saussure, as father of, 22; subsumes functionalism, 45, 46 Structures of realization, 89 (see also S-structures) Sturtevant, E., 46, 243, 294 Sub-category phenomena, 39, 167, 169–70, 172–3, 211, 223, 275, 304 Swadesh, M., 8, 243, 258 Sweet, H., xxxii, 28–9, 30, 38, 259, 290, 291 Symbolic behaviour, regrettable collocation, 108, 119 Symbolic interactional(-ism), 108 Synchrony/Diachrony, 19, 21–4, 97, 201–3, 290, 288–92 Tacitus, 155 Talbot, P. A., 8, 67, 68, 310 Tallensi, 5, 6 Tambiah, S. J., 13, 37, 291 Tanga, 79 Tarachow, S., 312 Tarde, 19 Tardits, C., 285 Taxonomic space, 66–8, 71 Taylor, D., 287 Teitelbaum, M. S., 117 Template, xxiv, 64, 79, 89, 99, 189, 210; of dominance, 187; see also p-structures Textuality and life, 171, 185, 209, 275 Thomas, K. V., 90 Thomas, N. W., 8 Thomson, D. (alias Ruaraidh MacThomais), 216, 308 ‘Srath Nabhair’, 308 Thurneysen, R., 135 Tiger, L., 119, 132

349

Time: cultural, 135–43, 301-2; and historical density, 276; ‘real’, search for, 144–5; space, and language, 142–6; structural, 222 Timelessness, supposed, in third world, 202 Tolkien, J. R. R., 177, 178, 185, 235 The Lord of the Rings, 177; fictive philology in, 177–8 Tonkin, E., xx, xxiv, xxxi, 285, 286, 297 Toynbee, A., 237 Transaction (-alism), 63, 194, 294, 295–6; as highest stage of functionalism, 296 Transformation, 22–4, 37, 40–2, 60, 98, 157, 292–4, 296, 306; of population definition, 70; of world-structure, 145, 153 Translation: of culture, 160, 169, 173–6, 182–4; lexical, 169; literary, 160, 174, 183; of personal names, 176–7; total, 173–85 ‘Tribal’ units, 66–8, 109–17, 121–2 Trier, J., 15, 38, 288, 313 Trobriands, 18, 198 Trollope, A., 183 Trubetzkoy, N., 28, 30, 31, 32, 41, 253 Turlot, F., 115 Turnbull, C., 116 Turnbull, H., 287 Turner, V. W., 13, 24, 25, 61, 83, 260 Tweedy, Major, 115 Tylor, E. B., 18, 210 Ullmann, S., 9, 15, 38, 171, 288, 290, 313 Universals, 10–11, 164, 173, 182–5 Urry, J., 206 Vandals, 155 Van Gennep, A., 83 Rites of Passage, 198 Vendryes, J., 19, 246 Vendler, Z., 241, 250 Voegelin, C. F., 32, 255–56, 259 Vogt, H., 244 Waismann, F., 290 Wales, R. J., 293 Wallerstein, E., 111

350 Index Warmington, W. A., xvi, 78, 84, 119, 122, 127, 221, 283, 299, 326 Wartburg, W. von, 13, 14, 15, 38, 39, 171, 289, 329 Watergate enquiry, 101 Weaver, W., 32 Webber, J., 305 Weinreich, U., 250 Weisberger, L., 288 Wells, R. S., 28, 290 Wells, J., 247 Welsh, xii; colour terminology, 9–12, 287 West Africa, ix West African Institute of Social and Economic Research, viii, ix Western Isles of Scotland, 217, 307 Whatmough, J., 290 Whiteley, W., 4, 36, 143, 286, 296, 297 Whorf, B. L., xxi, 9, 15, 101, 102, 164, 165, 169, 172, 183, 261, 262, 263, 296 Wiener, N., 33, 34, 41, 43, 53, 237–8, 253, 254 Cybernetics, 33 Wilson, B., xxxv Winch, P., 210 Winteler, J., 30 Witchcraft, 90–3, 158; Bakweri, 90–3; Esu, 118–19 Wittgenstein, L., 22, 178, 290 Philosophical Investigations, 290

Women: absence from ethnographies, 75–6, 128–9; anthropologists, 128–9; and dominance, 187–90; inarticulacy of, 73–4; 129; models of society, 74, 83–5, 127–8; as ‘muted group’, 129–30; as ‘natural’ in male models, 77–8, 82–3, 131–3; presence in symbolism, 76, 79–82, 127; visibility and invisibility of, 186 Woolf, V., 127 World-structure, xxvii, xxxii, 95, 100–2, 119–20, 123–6, 130–3; as cross– semiotic, 148; as definition space, 134, and language, 148–52; neither empiricist nor idealist, 144, 184, 207, 210, 275; and prophecy, 152; as reality-defining, 146, 148; as selfdefining system, 130; as social-space, 146; and space, 143, 143–5; the term dismantled, 153; and time, 136, 143–5; transformation in, 151–3 World War I, 120 Worsley, P., 203, 294 Wovea Islanders, 78, 79 Wright, E. M., 162 Yanomamö, 110–11 Yasa, 79 Yeats, W. B., 134, 135, 301 ‘The Statues’, 134