Mary Douglas 9781800739802

Citation preview

MARY DOUGLAS

Anthropology’s Ancestors Edited by Aleksandar Bošković, University of Belgrade; Institute of Archaeology, Belgrade; Max Planck Institute for Social Anthropology, Halle/Saale As anthropology developed across geographical, historical, and social boundaries, it was always influenced by works of exceptional scholars who pushed research topics in new and original directions and who can be regarded as important ancestors of the discipline. The aim of this series is to offer introductions to these major figures, whose works constitute landmarks and are essential reading for students of anthropology, but who are also of interest for scholars in the humanities and social sciences more generally. In doing so, it offers important insights into some of the basic questions facing humanity. Volume 4 Mary Douglas Paul Richards and Perri 6 Volume 3 Françoise Héritier Gérald Gaillard Volume 2 William Robertson Smith Aleksandar Bošković Volume 1 Margaret Mead Paul Shankman

MARY

DOUGLAS ••• Paul Richards and Perri 6

berghahn NEW YORK • OXFORD www.berghahnbooks.com

First published in 2023 by Berghahn Books www.berghahnbooks.com © 2023 Paul Richards and Perri 6

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 Names: Richards, Paul, 1945 May 14- author. | 6, Perri, 1960- author. Title: Mary Douglas / Paul Richards and Perri 6. Description: New York : Berghahn Books, 2023. | Series: Anthropology’s ancestors; Volume 4 | Includes bibliographical references and index. Identifiers: LCCN 2023007752 (print) | LCCN 2023007753 (ebook) | ISBN 9781800739796 (hardback) | ISBN 9781800739819 (paperback) | ISBN 9781800739802 (ebook) Subjects: LCSH: Douglas, Mary, 1921-2007. | Ethnosociology. | Social structure. | Social conflict. | Anthropologists—Great Britain—Biography. Classification: LCC GN21.D68 R53 2023 (print) | LCC GN21.D68 (ebook) | DDC 301.092 [B]—dc23/eng/20230227 LC record available at https://lccn.loc.gov/2023007752 LC ebook record available at https://lccn.loc.gov/2023007753

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

ISBN 978-1-80073-979-6 hardback ISBN 978-1-80073-981-9 paperback ISBN 978-1-80539-367-2 epub ISBN 978-1-80073-980-2 web pdf https://doi.org/10.3167/9781800739796

CONTENTS

• • • Preface Acknowledgements Introduction

vi viii 1

Chapter 1. A Biographical Sketch

14

Chapter 2. Purity and Danger

29

Chapter 3. Four Sides to Every Question

40

Chapter 4. Risks and Solidarities

57

Chapter 5. Institutions and Thought Styles

67

Chapter 6. Ritual and Categories

82

Chapter 7. Gifts, Goods and Economic Development

98

Chapter 8. Strife

110

Conclusion. Institutional Renewal

123

Appendix. Selected Works by Mary Douglas Not Discussed in the Text

133

Works by Mary Douglas Discussed in the Text

139

References

142

Index

151

PREFACE

• • • The influence of Professor Dame Mary Douglas (1921–2007) upon each of the social sciences and many of the disciplines in the humanities is vast. The list of her works is also vast, and this presents a problem of choice for the many readers who want to get a general idea of what she wrote and its significance but who are somewhat baffled about where to begin. Our book offers a short overview and suggests why her key writings remain significant today, in a world of social tensions threatening creative and cooperative solutions to a sustainable planetary future. Her aim was to understand such tensions and to propose ways of living with or working around them. Our previous, longer book on Mary Douglas’s thought (6 and Richards 2017) was directed to debates across the social sciences. That work is a much fuller analysis of her output than we can attempt here, and it also examined contributions by other scholars who have drawn upon Douglas’s legacy. The present book – as readers will expect in a series entitled ‘Anthropology’s Ancestors’ – is specifically concerned with exploring the importance of Douglas’s intellectual legacy for anthropology’s enduring concerns, and for a renewed anthropological contribution to wider public debates about deep and perplexing social conflicts (such as so-called ‘culture wars’). A fresh look by anthropologists at Douglas’s work is timely. Her contribution was to explain the range of plurality in human thought and the dynamics of intransigent conflict, and nothing could be more topical in a world of authoritarian and populist politics and internet social media storms. In this book, we also give some recognition to Douglas’s sustained engagement with philosophical debates. In the second half of the twentieth cen-

PREFACE

vii

tury, most leading anthropologists found themselves unable to avoid arguments with philosophers of social science, of language and of meaning, for the philosophers were keen to address anthropological problems. Douglas herself positively relished these exchanges. A fresh look at Douglas’s work allows us to claim there was something enriching for the discipline in both drawing upon and being challenged by philosophers. Perhaps most importantly, we are aware of how far Douglas’s influence has travelled – into worlds she can hardly have envisaged at the outset of her career, involving areas of creativity and analysis as diverse as art, music, fashion, technology, cultural studies, management and conflict studies. Douglas was a profound thinker, and readers approaching her from a diverse range of directions deserve a succinct account of her work that also seeks to convey some of the depth of her thinking.

ACKNOWLEDGEMENTS

••• We are most grateful to Marion Berghahn for inviting us to write this second book on Douglas for Berghahn Books, to the series editor Aleksandar Bošković for his patience with us as we worked on the manuscript, and for his sound editorial advice, to Tom Bonnington and Caroline Kuhtz for their meticulous support in preparing the text and to Sarah Sibley for most careful and efficient copyediting. We also wish to thank Jim Douglas and Janet Farnsworth for providing the cover photograph of Mary Douglas. Both authors owe apologies to other research collaborators who have tolerated our dilatoriness on other projects while we have worked on this book. Paul Richards wishes to thank his daughter, Marjorie Kumba-Komeh Rozemarijn Richards, a student at the Amsterdam Fashion Institute (University of Applied Sciences Amsterdam) for opening his eyes to the wider world of twenty-first century sustainable multiracial creativity into which much of the thinking of Mary Douglas so clearly fits. We both wish to extend profound thanks to Professor Richard Fardon, who read the final manuscript with great promptness and fine attention to detail, offering us valuable criticism and advice. We are most grateful to Danny Cohen for suggestions on many points in the text. Any deficiencies in the book are, of course, entirely of our own making.

INTRODUCTION

••• T

he anthropologist Mary Douglas built her career on the insight that humans are classifying animals. We like to know where our things are. We tidy up, and pack accordingly, with drawers and cupboards for this and that. But on what basis do we classify the items we tidy? Some people may shove all their clothes into a cupboard and slam the door. Others take time to sort out underwear from overwear and neatly pair all their socks. Much of this depends on social expectation. We may sort socks because we do not want to be late for work next morning. In this case, our strategies of classification depend on conventions (about when work starts) and social expectations of those to whom we are accountable (the boss’s requirement we turn up on time, not wearing odd socks). When we classify something as having passed a threshold – the spread of a virus constituting a pandemic, or the scale of ethnic slaughter in a certain place reaching a level that amounts to genocide – then duties are activated for organisations and for states. Equally, when classifications stigmatise people, they become not just the subject of conflicts about justice but the focus of active organisation and counter-organisation (witness for example the Black Lives Matter social movement). In this way, we code our world and match ourselves to the expectations of others. A simple – as we shall see, probably too simple – way to describe an institution is to state how its implicit or explicit rules apply to a case when the case has been classified in a particular way: as a sock for a day at the office rather than a hiking sock, as an outbreak but not yet a pandemic, as shocking abuse but not a war crime. All institutions classify even though they do much

2

MARY DOUGLAS

more besides. Considered as rules, institutions use classification to regulate social life. Yet all the categories in our rules generate anomalies. Some socks can be used for multiple purposes; to which drawer do they belong? Some cases prove very difficult to classify unambiguously. Is a shipwreck the property of the descendants of the original owners of the ship, or of the contemporary salvage company, and who owns the cargo that was stolen from another group of people but was classified as state property five hundred years ago? Contrary to some social science theories, institutions do not typically come into being because they are good or efficient. We can waste endless amounts of time arguing whether (for example) driving on the left or the right is better. An agreement to do it one way (a convention) is needed, but arguments proliferate about which one to accept, and why. For example, prioritising ‘equity’ (or fairness) over contract law (or the reverse) are both practices that can be justified by appeal to some (but always contested) conception of what is good, right or efficient. Some institutions arise by coercion, as did many of those associated with imperialism; today, land tenure rights may be granted by authorities who first became ‘authorised’ for this task by colonial conquest. Others – such as rules through which children inherit land or property – may only acquire much-contested justifications of morality or efficiency long after they have become established. Moreover, institutions, Mary Douglas pointed out, never stand alone. They interlock in a web of often equally arbitrarily chosen conventions through which we attempt to impose some order and predictability on the otherwise teeming chaos of everyday social life. It is hard to pick out and change an institution when it is interlocked with so many others. Institutions that govern who can marry whom, who inherits what, what obligations guest and hosts have to each other and to the wider community may arise initially as solutions to practical problems. Yet what is counted as a problem and what is counted as practical is intimately linked to large numbers of other institutional assumptions about how we live our lives. For that reason, we defend our institutions as bundles. We may come to see some of them as being somehow nat-

INTRODUCTION

3

ural or God-given – the only possible way of doing things – not because we are biased by inexplicable preferences (for left over right, for example) but because we sense how much else might have to change if we were to change one of our rules. Faced with disorienting fear, some people will retreat into an institutional ‘silo’ and slam the door (Tett 2015). Faced with conflicts of laws, courts must determine which silo takes priority. Now we have arrived at the point of deep conflict – the hard confrontations between different cultures and different sets of rules to which we apply terms such as nationalism, racism, ethnicity and religious intolerance. Such clashes, Douglas thought, are rooted in the processes through which institutions become tightly bundled as a way of life. We may not dare concede the arbitrariness of our assumptions (least of all when we sense they are to our advantage – as with racial privilege, for instance) because to question one of our institutions is to question them all. We may double down on our prejudices for fear that anarchy might prevail. Douglas came to understand that addressing deep conflict was her central task. For this reason, she set about trying to understand how classification works and how institutions are ordered and sustained, pondering how better to cope with the problem of institutional intransigence. Intolerance is often fatal. We need to find ways to cooperate over planetary survival, for fear that clashes of institutions will kill us all. The problem, however, is that often there is no referee. The authority of arbitration and courts is limited. Parties may have all kinds of ways to continue contention whether despite or through procedures designed to end such conflicts. In many settings, there is no one to say which of two sets of conflicted rules or cultural understandings should prevail. Of course, plenty have tried. Ignoring Acemoglu and Robinson’s own caution that ‘you can’t engineer prosperity’, too many development consultants have concluded from their book, Why Nations Fail (2012: 446–50), that the answer is to pick the best institutions and spread them across the world in an all-inclusive way. The hapless president of a struggling African country is asked by international development advisers ‘Why can’t you be

4

MARY DOUGLAS

more like Denmark?’ (see Fukuyama 2011). But there is no being ‘more like’ – the only option is to become Denmark because institutions come in mutually supporting bundles. Even if a ‘bad’ institution could be identified, excising it would be more complex and dangerous than brain surgery; the patient might die. Faced with the impracticality of everywhere becoming Denmark, Douglas’s approach is different. First, she tries to identify and map the social processes through which we generate institutions and then wonders how to influence that process. Institutions are explained functionally. Ritual action cultivates organisational arrangements (X), which in turn shape styles of thought (Y) to sustain and thus protect (Z) the form of social organisation in a group (Douglas 1986: 33; Chapter 5 below). The process achieves its effect without the group being aware of how it works. But it is not ‘functionalist’. Nothing about the process is necessarily efficient or ‘meeting society’s needs’, and as the dynamic proceeds among different groups and looser clusters of individuals, it generates conflict. This is one of Douglas’s most important but most controversial claims. Most people like to think they control their own thoughts. We imagine that it is only other people who are fooled by the company they keep. Douglas’s retort is that all ideas are filtered by social processes, and she uses ethnography to prove it. Even the very idea that our thoughts are our own is a product of a society organised to encourage and reward individual action.1 If this explains a central process in how institutions arise, we also need to know how institutions are legitimated and stabilised. Often institution builders start with a rough analogy, along the lines of (for example) the ruler is up and the people are down, but group work on the analogy invests it with a greater clarity of outline and gathers around it a sense of collective commitment. In terminology adopted from the French social theorist Émile Durkheim (1995 [1912]), the result is a group totem – something symbolising who we are, socially speaking. Waving a flag or attending a parade or a swearing-in is part of a grand public ceremony celebrating the nation. But Douglas had also taken seriously the American sociologist Erving Goff-

INTRODUCTION

5

man’s insight that much ritual activity is also carried out in minor daily exchanges (Goffman 1967). We are not always thinking at the level of tribe or nation. Often, we are simply trying to get a project to work or to win a game, and not every game is a cup final. When a sports team prepares to attempt a difficult passage of play, members will often mimic the task, knock hands together or perhaps huddle in a group. They are gearing up to engage in a collective task with an uncertain outcome through ritual. Ritual action speaks about organisational challenges ahead. Before a match, the coach gives a ‘team tactics talk’. The sentences of this talk would probably be useless if written down. But the social interaction in which the speech is delivered and listened to are consequential. The coach’s words have efficacy and become inspirational to the team only because the speech builds upon weeks of everyday ritual activity in which collective action is rehearsed. Teams are built ritually in members’ routine conversations during training and refreshment breaks, not by trainers’ rhetoric or by ceremonies to present trophies. Now Douglas can tackle the outcomes (the Y variables). Her model is configured with a causal arrow passing from social life, through ritual activity, to how people think (6 2014a: 93). Deciding and acting based on this ritually cultivated manner of thinking will, over time, lead to institutionalised outcomes. In the sports example, this might be a game won. With the mobilisation of nationalism or other group sentiment, it might be economic success or victory in a war. Others might limit their ambitions to organising a birthday party or a day out among friends. On whatever scale, the result has been achieved through collective effort stabilised by an institution or a set of interacting institutions. Institutional success is not easily achieved, because it requires a kind of trick. People come to believe they will succeed through skill, natural superiority or because God is on their side. It is too easy to see through rational appeals to institutional mobilisation. Someone else will soon be along urging the exact opposite, and on rational grounds also. But ritualisation blinds us to our self-fulfilling prophecies. Success is achieved because we know in our hearts this is the right thing to do. Conversely, the activi-

6

MARY DOUGLAS

ties supported by an institution are in trouble when it no longer seems the natural way to respond. Prior to 2020, it used to be viewed as ‘natural’ to go to a workplace run by an employing organisation. The great hiatus of the pandemic of Covid-19 caused much white-collar work to be done from home. Now it is hard to get these workers back to offices, with significant consequences for (for example) urban transport and city centre coffee bars. After the end of lockdown, the British minister in charge of government efficiency developed a passive-aggressive tactic of dropping in on civil servants’ empty offices and leaving a card on which was written ‘Sorry you were out when I visited. I look forward to seeing you in the office very soon.’2 This signalled that their absence had been noted and work done elsewhere was discounted. The tactic is unlikely to succeed in inducing staff to come back to the office, because it attempts to confront a Y variable with another Y variable – in a battle of competing ideas. Douglas’s argument at this point is that ideas never directly change ideas, at least where institutions are involved, because this short-circuits the institution and its many subterranean interconnections with other institutions.3 What is needed is to look again at the practices shaping our institutional ideas. How and why are the practices of work changing? During pandemic lockdowns, people discovered how to work effectively from home, so institutional understandings changed as well. Rituals of work once reached out to include dress-down Fridays. Given the Covid-19 pandemic, sometimes this now extends to stay-athome Mondays as well. (Both authors come from Lancashire, so we cannot resist adding that Lancastrians at the heart of the British industrial revolution early discovered the cult of what overworked mill operatives referred to as ‘Saint Monday’, an unfailing source of life-restoring spiritual help.) Douglas’s explanation to the employer would be that a deep institutional change in where work ought to take place is underway. These dynamics might of course be reversed in time. But if so, any reversion is likely to arise from dynamics in social relations and practices, not from appeals to beliefs and ideals. By 2021, working from home had already become institutionalised in some settings. After the pandemic

INTRODUCTION

7

of Covid-19, a five-day working week in the office cannot be restored by decree or by appeals to ideas of employee duty alone. The British minister for government efficiency may be a modernday King Knut athwart an institutional tide. Those who work from home have discovered how to organise work from home, and this organisational knowledge cannot now be undiscovered. A theory of how institutions emerge and regulate social life and social change was laid out in Douglas’s most important book, How Institutions Think (1986). It is short but by no means an easy read, and one of the purposes of the present book is to prepare those interested to give it a go. It proposes something missing in many other accounts of institutions – a clear, conceptual separation between independent and dependent variables. The independent variable is always social action, which encompasses social relations and the organisation or ordering of those relations, including the stylisation of social interaction that takes place through ritual activity. The dependent variables are, in the first instance, styles of thinking. Action based on those ways of thinking then leads to outcomes. These can include the sustaining of institutions, resulting in, for example, collective achievements and constitutional settlements, but also the undermining of institutions, resulting in intractable conflicts and socially effervescent occurrences such as riots and massacres. Douglas consistently rejects the notion that institutions are the product of ideas – that a nation (for example) is the product of wise elders sitting down to draft a sacred text (a constitution). First there must be a set of social relations and interactions among some people intent on becoming a nation; the constitution emerges as its expression. ‘Originalists’ insist that the American constitution must be interpreted by reference only to the intentions of the framers (that is, as a sacred text), but in practice it has had to be reinterpreted afresh and through practices and rituals to address every generation’s problems, to sustain its institutionalisation (Ackerman 2000). The Italian statesman d’Azeglio commented in his memoir that after unification ‘We . . . made Italy. Now we must make Italians.’ The declaration of unification, independence and statehood itself did not create Italy or Ital-

8

MARY DOUGLAS

ians. Citizens’ identification with the newly united nation took at least a generation to cultivate via participation in common life and governance. Through everyday interactions (many of them with ritual connotations – consider the stylised iconography of a stamp bought to send a letter on a national postal delivery service or the small but decorated piece of card authorising journeys on a national rail network), it now more and more made sense to relate to others as fellow citizens of a national community. Citizens are made through the everyday practices of citizenship. Douglas argues that these existing fields of social action and interaction are the only drivers of institutionalisation. Douglas began her academic training with a degree in philosophy, politics and economics (PPE) but later switched to anthropology, at a time when the discipline was pondering the implications of policy changes driven by global cooperation during the Second World War. The Roosevelt and Truman administrations made it clear European colonial empires had to be demolished as part of the price for US cooperation in winning the war against Hitler. British anthropologists became engaged in debates about what this would mean in the African colonies. African self-government was on the way, but the colonial territories were predominantly rural and agrarian and had only partially recovered from social damage sustained in the last and most intense periods of the slave trades to the Americas and Middle East. Colonialism had stepped into this breach to impose by force of arms a kind of sullen and resentful ceasefire – ‘they made a desert and called it peace’.4 In what condition were these once turbulent and battle-scarred but now conquered and quiescent African countries to rule themselves? This question forced anthropologists, many of whom were fundamentally anticolonial in their sympathies, to direct their attention to local institutions and how they worked, and this was the training ground for Douglas’s interest in social classification as a key to institutional order. Douglas went to the Belgian Congo to carry out fieldwork among the Lele people of the Kasai province (Chapter 1, below). The biggest social concerns of the Lele were how to keep up their numbers. Reproduction and marriage were major issues, as in all

INTRODUCTION

9

the Congo, where colonial forced labour had spread venereal diseases and depressed birth rates. Because both intergenerational and interfamilial tensions ran high, another major challenge was to find ways to resolve disputes, given that the Lele lacked any tradition of strong rulers. Douglas homed in on how the Lele used classification to locate themselves within wider natural and social worlds and withstand their abundant hazards (Douglas 1977 [1963]). She realised that foodstuff consumption rules – who could eat what, when – played a major part in marking and sustaining social boundaries. The cult of the pangolin (the scaly anteater) was a highly significant institution, since it brought attention to those who were the most successful in the struggle to reproduce (the begetters of female and male children). Classification reinforced institutional boundaries. Anomalies confirmed rules by testing those boundaries. In her first widely successful book (Douglas 1966), Purity and Danger (Chapter 2, below), Douglas was mainly concerned with examining similarities and differences in the way various societies used classification and built institutions. In her later work, however, she became highly interested in the various conditions under which institutional choices are made. Forms of institutions are constrained by degrees of what Durkheim (1951 [1897]) called social regulation and social integration – that is to say, the degree to which social life is ordered, on the one hand, by rulebased constraint (or else by discretion) and, on the other, by groups with internal bonds and external boundaries around their members (or in some cases around loose clusters of transactionally linked individuals). Defining the two dimensions in terms of social interaction and operational practices also allowed Douglas to keep causes clearly separate from symbols or ideas. Douglas’s term for ‘social regulation’ was ‘grid’, and correspondingly she labelled social integration as ‘group’. Cross-tabulating these dimensions and attending to the cells provided her with a fourfold scheme. This typology became an important tool for challenging institutional economists in their belief that there were, basically, only two kinds of institutions – markets and bureaucracies (Williamson 1975)5 – and that the task of the economist was to secure

10

MARY DOUGLAS

an optimal division of labour between them, based on some universally applicable principles of institutional design. Some readers may ask whether Douglas’s early ethnographic work, which was developed within the milieu of decolonisation, is still worth reading today. The struggle for colonial independence was a passing phase. When Douglas completed Purity and Danger (1966), formal sovereignty over much of Africa by imperial powers had ended. But arguments about decolonisation did not end there. This is because these arguments became institutionally embedded in, for example, contested viewpoints on nationhood, immigration, asylum and trade in the postcolonial world. Today, decolonization remains a prominent focus in the humanities and social sciences, promoted or opposed with ardent ferocity by those who see themselves as occupying different mental and political spaces. This is an institutional perspective in which ruling elites and dominant cultures must be challenged forcefully, since colonial mentalities still imprison minds and destroy life chances. From an opposing institutional perspective, ‘woke’ arguments sustain a vindictive, victim culture, thereby undermining technological answers to urgent material challenges; complain not about the injustices of colonialism but look at the opportunities it provided. This is exactly the kind of deadlocked situation where Douglas felt she had something distinctive to contribute. Conflicted ideas were always, Douglas argued, a symptom of dynamics driven at an institutional level. A first step in breaking the deadlock was to revisit the classification of people in order better to map the ways in which social agency shapes patterns of exclusion and marginalisation. Change cannot readily come from outside, however. This is because where institutions have made boundaries between an ‘inside’ and an ‘outside’, ritual processes will seek to protect the ‘inside’ from interference. An engineered institution, Douglas concluded, will struggle to take root, since its artificiality is clear for all to see, undermining its functionality. A prior requirement, therefore, is to provide useful tools to assist people to make changes of institutionalisation from within. In her later work, she enters an important reservation. Some groups

INTRODUCTION

11

lacked the freedom to experiment with institutional change. Blaming poor ‘culture’ or local institutions often implies blaming the poor for their own predicament. Although local efforts often fail, importing better institutions is an even more complicated game with less clear prognosis of success. She picks up an important suggestion of Amartya Sen (1999), that what the poor need to overcome poverty is freedom, including freedom to experiment with institutional changes, although Douglas envisaged that freedom not only as individual choice, as Sen had done, but as being a social choice as well. Douglas et al. (1998) recognised that Sen’s conception of ‘positive’ freedom was a good deal more social and institutional than the conception of choice entertained by many economists. The hope is that an easing of constraints changing local practices will cultivate institutions better adapted to local circumstances. Douglas also recognised that in the most deadlocked cases positive solutions may be highly elusive and mutual avoidance may be the only practical option. Doubtless, however, she would have shaken her head in despair at current attempts to foment culture wars, often for no better purpose than the protection of authoritarian regimes. These kinds of campaigns cannot work over the longer term because they are wrong about culture. Culture is the product of social agency, not its cause. The drivers of culture are rooted in social life, and specifically in the ways in which social classification stabilises emergent institutions. Peace is the product of institutional accommodation and reform. A world convulsed by identity-based feuding requires a method of analysis that does not take identities at face value but seeks to explain their emergence, conflict and decay not merely as manipulation but as the work of institutionalisation and deinstitutionalisation. After the present introduction, our short book is divided into eight main chapters, followed by a brief concluding discussion. Chapter 1 provides an account of Douglas’s career and identifies five major strands in the development of her intellectual concerns. Chapter 2 offers an account of the main themes addressed in Purity and Danger, in which she explains how social classification grounded in natural analogy serves both to bind groups

12

MARY DOUGLAS

and to separate and stigmatise individuals. Chapter 3 offers an account of her fourfold scheme for mapping different strategies of institutionalisation. Douglas’s work on risk is examined in Chapter 4. Chapter 5 considers the ways in which social life is regulated by institutionalisation and offers a summary of her model of institutional dynamics as laid out in her most important book, How Institutions Think. Chapter 6 offers an account of the role of exemplars of categories in the making of distinct social worlds, largely drawing on one of the most important of her later articles, ‘Rightness of Categories’ (Douglas 1999d [1992c]). In Chapter 7, we consider Douglas’s work on consumption and goods and her later interest in debates about institutions in economics in the 1980s. Chapter 8 explains how her ‘fieldwork’ on the Hebrew Bible illuminated arguments about deep conflicts and provided a path to understanding terrorism, communal violence and civil wars, and the mitigation of such conflicts. The conclusion argues that grasping the point of Douglas’s insistence on social dynamics as drivers of ideas, and not ideas as drivers of social dynamics, provides a basis for comprehending a modern world convulsed and confused by populist politics, autocracy, social media bubbles and internet miscommunication.

NOTES 1. Douglas termed the outcome of this kind of social filtering of thought as ‘cosmology’ but later preferred (for reasons we explain in Chapters 3–5) ‘thought style’. Her key focus is on any collective activity through which a sense of institutional patterning is imposed on the profusion of social life. 2. See, e.g. 29.04.2022, Financial Times, https://www.ft.com/content/ c8f678b3-6c54-44d2-9de8-059112c590e2. Office use increased again in late 2022 for a mix of social reasons and high costs of winter home heating, even its most ardent advocates acknowledge that flexibility is likely to persist at least for some categories of workers (https://www .theguardian.com/commentisfree/2022/dec/04/new-love-affair-withoffice-step-towards-better-philosophy-of-work). This is likely to be partly because employees want to do some work at home and employ-

INTRODUCTION

13

ers want to retain and attract skills, and also partly because employers have seen opportunities to reduce their commercial real estate costs. For both workers and employers, change in working practice is driven by a combination of social pressures of bonding and accountability and also financial constraints, not by ideas or exhortation. 3. We need to be clear what Douglas means by ideas. If we ask a friend how they hope to meet us, they might say ‘my idea is to come by bus’. That refers to an arrangement, a plan or a promise and is included within social action (Austin 1975 [1962]). Douglas is referring to abstractions – democracy, fairness, the nation and so forth. Max Weber (1947: 87–118) called such abstractions ‘ideal types’, having in mind a model against which social reality is compared. These abstractions are descriptive summaries of situations, not causes. 4. Tacitus (2009 [1948]) credited this remark to a Caledonian warrior describing the Roman conquest of the Scots. 5. Some economic institutional theories also recognise clans (Ouchi 1980) or ‘networks’ (Thompson et al. 1991).

CHAPTER 1

A BIOGRAPHICAL SKETCH

••• I

t is commonly supposed that ideas cause actions. Douglas stands this on its head, arguing that actions cause ideas. Where does this come from? The answer is from her life, or more specifically from reading the eminent French sociologist Émile Durkheim (1858–1917) against the background of a rather unusual childhood culminating in secondary school education at the convent of the Sacred Heart in Roehampton, south-west London. Douglas was herself the product of a colonial milieu. Part of her ancestry was Irish. Ireland was one of England’s first colonies. Her maternal great grandfather, Daniel Twomey, was from County Cork. He made sufficient money buying farm animals and selling meat to ships about to cross the Atlantic that he was able to educate all his children, who then departed impoverished rural Ireland in various professional capacities (Fardon 1999: 4). Daniel Twomey’s son, Mary Douglas’s grandfather, Sir Daniel Harold Ryan Twomey (1864–1935) became a colonial judge in Burma. Her father, Gilbert Tew (1884–1951), who was English, also served in Burma, as a District Commissioner. In early childhood, she lived with her maternal grandparents in Devon while her father was serving overseas. Douglas was thus constantly aware in her childhood of the competing polarities and multiple identities characteristic of colonial social worlds. Migrants from rural Ireland were prone to be despised by sections of English society, both for rural backwardness and for their Roman Catholic religion. Douglas saw neither Catholicism nor alleged backwardness as an obstacle to an anthropological

A BIOGRAPHICAL SKETCH

15

career. In fact, they became a stimulus to thinking in profoundly new ways about social identity, social change and the role of religious agency in both reinforcing and changing social norms. Thus, there was no conflict between being religious and exploring social organisation in an open and sceptical manner. No conflict, that is, if religion was understood as a set of mental states cultivated by social organisation through ritual tools. This argument she encountered (influenced by her Oxford training in anthropology) in the work of Durkheim (especially Durkheim 1995 [1912]). Academics trade in ideas and frequently overrate their value to the point where entire schools of thought insist ideas cause actions. Durkheim argued for the reverse. The priority for the social scientist is to understand action – including not only practical drivers of human social life such as farming or trade but also the kinds of ritual actions Durkheim categorised as elements of religion (cult activities performed to bring the members of the group into emotional alignment). It is worth paying attention, therefore, to what her intellectual biographer, Richard Fardon (1999), writes about Douglas’s schooldays and their significance in shaping her view of the social world. Although the Sacred Heart was a closed institution, the teaching nuns were well educated and something of an intellectual elite. They saw themselves as female cousins to the Jesuits and were permitted to read books prohibited to ordinary Catholics. The school imposed elaborate internal discipline, dictated by a strict ritual calendar. No one moved through the school except in the right order and along a specified path. Punishments and rewards were small but prompt and less liable to foster lingering resentment. In short, to a receptive (and orphaned) student like Douglas, school ritual inculcated a lifelong positive feeling for social order based on hierarchy.1 Having been taught French at school, with some time in Paris to further polish her knowledge of the language, Douglas was able to read Durkheim in the original, including books and papers never translated into English. English-speaking students seeking to understand Durkheim have long been at the mercy of poor translations. Recent scholarship has corrected some of the

16

MARY DOUGLAS

misunderstandings that arose (Stedman-Jones 2001; Warfield Rawls 2005) – notably, that Durkheim argued for a kind of group mind – but Douglas was spared these errors at source. Durkheim convinced her that what people believe (and the culture they adopt) results from the ways in which they act and were organised. Action and organisation, not belief, constitute the independent variable in social life. Education at the Sacred Heart was focused on the humanities. There was no social science, but papal encyclicals were studied, which introduced the young Douglas to the church’s social teachings, in which laissez faire capitalism and individualism were deplored and social justice stressed. Douglas expressed an interest in studying social sciences at the London School of Economics, but this was discouraged, since it was seen as a politically radical institution. Philosophy, politics and economics (PPE) at Oxford was thought to be more suitable, especially because there was a small, subsidised lodging house at Oxford for Catholic women of limited means (now St Anne’s College). For postgraduate studies, she switched to social anthropology. It is sometimes argued that, as an emergent field of study within the first half of the twentieth century, social anthropology was complicit in colonial rule, and that colonial institutions framed the fieldwork of so many of its foundational practitioners. In fact, much British academic social anthropology at this period was distinctly anti-colonial. It openly challenged many of the assumptions of metropolitan and racial superiority on which British colonial rule rested. Several key practitioners of this emergent social science were from the colonial periphery and themselves members of racial and ethnic minorities. One of these was Meyer Fortes, a Jewish Fabian socialist from South Africa who helped inspire Douglas’s ambitions to undertake immersive ethnographic fieldwork in rural African communities. Fortes’s meticulous studies among the Tallensi of the Northern Territories of the Gold Coast (Ghana) (Fortes 1945, 1949) set a standard against which Douglas determined to measure herself, even if she felt a lifelong sense of inadequacy in terms of her own fieldwork.2 Raided for centuries from north and south for both the Atlantic and Saharan slave trades, the Tallensi had taken to the hills

A BIOGRAPHICAL SKETCH

17

and devised a way of coordinating their community through shrines but without mobilisation under paramount chiefs or Islamic emirs or residence in walled towns (unlike many of their regional neighbours). The society as described by Fortes was layered, comprising autochthones and a group of later arrivals – a horse-borne cadre known as the Namoo – who seemingly traced their origins as a breakaway of the Songhai state 400 years previously. Fortes compiled his account of Tallensi life based on documentation of intricate social connections and interactions. This microscopic complexity is the point. It is like the dense weave of a piece of cloth. Where colonial administrations treated groups such as the Tallensi as being backward and ‘primitive’, anthropologists of a Fortesian bent sought to demonstrate that people without political centralisation were successfully managing their own lives according to their own rules. Douglas had some doubts. In bringing out the functionality of Tallensi social life, had Fortes placed too much emphasis on its densely integrated character? Might there not have been more rents in the social cloth than his cohesive account allows? The threat of social breakdown seems to have lurked beneath the surface, as apparent in a remark by one Tallensi elder, who reflected that under British colonial rule he remained alive in his old age where he might otherwise have long been dead from a poisoned dart (Lynn, M. Lynn and S. Lynn 2013: 14). This serves to suggest that levels of internal dissent were high and potentially lethal. Fortes could only carry out his painstaking work in documenting the social networks binding Tallensi society because the British Governor hovered in the background with the option of military intervention should need arise. The Fortesian doctrine of structural-functionalism cannot be allowed to obscure the fact that life in autonomous peripheral African rural societies was liable to be tough, violent and short. Douglas, herself, was alert to signs of fundamental social disagreement and the need to take full account of it in any anthropological theorisation. It was in her background. Part Irish among the English, with a Protestant father but raised a Roman Catholic and educated by exiled French nuns at a time (in the 1930s) when significant parts of the British population expressed sympathy

18

MARY DOUGLAS

for the racial policies of Nazi Germany, Douglas was searching for an anthropology that provided scope to address deep social disagreement. She found the approach she was looking for in the African ethnographic fieldwork of Edward Evan Evans-Pritchard, holder of the Oxford chair of social anthropology. Evans-Pritchard had made his name from a study of the Azande (Evans-Pritchard 1937), a group living across the colonially imposed border between Anglo-Egyptian Sudan and the Belgian Congo. Like the Tallensi, the Azande were a hierarchically layered society, divided into aristocratic and commoner elements. Daily life was beset by misfortunes, some of which could only be explained by the presumed action of witches, but no commoner could accuse an aristocrat of being a witch. As in many African societies without a clear and agreed bureaucratic system of leadership, the Azande had once relied on the power of the poison ordeal to decide intractable social disputes. This poison – a decoction made from a common African tree (Erythrophleum suaveolens) – contained two active principles, one that induced vomiting and the other a heart attack. An accused person volunteering to prove their innocence would drink the preparation. Truth-tellers vomited; liars died. Colonial regimes promptly banned the ordeal. Thereafter, a backlog of unsolved cases blocked local dispute resolution systems, resulting in various adaptations. In some cases, human diviners were used but were not a success; they could all too easily be bribed (Richards 2020). Ingeniously, the Azande kept the poison but adapted it so that the poison was administered to chickens. This system – now an oracle, not an ordeal – revealed witches but no longer carried out punishment, so further adaptation was needed. Integral within the Azande scheme of belief was a whole series of remedies whereby people accused of witchcraft could be purged of their malice. A popular system of justice was created without it fatally undermining the position of the dominant elites. Douglas’s own family background and upbringing had prepared her to appreciate the social dualisms and issues of class identity addressed by Evans-Pritchard’s subtle analysis of Azande dispute management by occult means. Evans-Pritchard had real-

A BIOGRAPHICAL SKETCH

19

ised that far from being a manifestation of ‘primitive’ irrationality, Azande oracles served as a coherent scheme for managing misfortune without causing a larger system of political settlements to be challenged. It might not be ideal, but it was founded on a set of institutional compromises that seemed to work. Evans-Pritchard referred to this notion of pragmatic functionality when noting that while living among the Azande he regulated his own domestic affairs by these same practices. When in Rome . . . . In the Azande social context, witchcraft beliefs made practical sense. Take away the social context and no one need be convinced. Evans-Pritchard’s subsequent work on the Nuer (e.g. 1940, 1951, 1956) was also influential upon Douglas’s thought. She came to appreciate Evans-Pritchard’s analysis of how countervailing forces and feedback dynamics sustained delicate settlements between competing imperatives and contrasting notions of accountability – for example, the rival demands of lineage (family descent) and local community, or the tensions between collective duty and status-based pride. Evans-Pritchard taught her that on many occasions metaphysical arguments really turn on practical tensions between different kinds of social claims (Douglas 1980). From his approach, she derived a method of analysing her own African ethnographic data on the intricate tensions and competing demands for social accommodation among the Lele people. Surprisingly to some, Evans-Pritchard, the son of a protestant clergyman, also became a Catholic convert. He had managed to step over a gap that might have seemed unbridgeable to some within English society at the time. Douglas’s institutionalised religious upbringing did not have to be abandoned to become a successful anthropologist. She became Evans-Pritchard’s doctoral student.3 Under Evans-Pritchard’s guidance, Douglas then left for Africa to carry out her own fieldwork. In the late 1940s, the Lele were a small, decentralised matrilineal group speaking a Bantu language and largely dependent on hunting, fishing and cassava cultivation on rolling uplands adjacent to the Kasai River, in the Belgian Congo. Prior to Douglas’s arrival, the region had attracted the interests of the British soap manufacturer Lever

20

MARY DOUGLAS

Brothers. Adjacent to the Brabanta oil palm plantation on the Kasai River, there was a Belgian administrative post and a Catholic mission. The circumstances under which she encountered the Lele were only revealed in a piecemeal fashion (and then never completely). Her writings addressed the impact of Belgian rule and the mission but remained largely silent about the plantation. European industrialisation demanded an ever-increasing supply of African vegetable oils. Aggressive commercialisation of oil palm production was a factor in a major uprising in colonial Sierra Leone in 1898, which threatened the security of a key asset of British imperialists – the deep water harbour at Freetown, the maritime base for further colonial acquisitions in western Africa and beyond (Richards 2020). Further development of oil palm plantations in Sierra Leone, the Gold Coast or Nigeria was felt to be a security risk to British rule. Applications for land for plantations in British colonial West Africa were rejected. The British rulers hesitated to give away property that was so clearly owned by their colonial subjects. Henceforth, commercial tree crop agriculture was a field reserved for peasant small holders. The Belgian regime in the Congo had no such scruples; any land not currently occupied by villagers belonged to the state and could be reallocated as concessions to external investors. Lever Brothers established five major oil palm plantations along the Congo-Kasai River system. The furthest upstream, Brabanta, was located at Basongo on the Kasai, close to its head of navigation at Port Francqui (later Ilebo). In Britain, Lord Leverhulme was a model employer, building high-quality accommodation for his employees at Port Sunlight on Merseyside; in Africa, matters were somewhat differently arranged. Marchal (2008) describes the system on the Leverhulme oil palm plantations in the Congo between 1911 and 1945 as ‘forced labour’; men who refused recruitment to the plantation to harvest palms were subject to severe punishments by colonial state authorities, including imprisonment and beatings. On how this affected the Lele, living at the back door of the plantation, Douglas does not tell us. She had indicated to her Oxford supervisors that she would like to work among a matrilineal group. These are groups in which de-

A BIOGRAPHICAL SKETCH

21

scent or inheritance is determined through the mother’s brother, not the father. A reason for her choice may have been to examine how such a group would manage the tension between male elders wishing to attract younger kin to live under their authority and a rule that children were expected to live with their mother’s people (Fardon 1999: 58–59). Once it became clear to Douglas (1977 [1963]: 85–112) that clans were not corporate bodies and that there was much flexibility in decisions about where to live, she lost interest; instead, the matrilineal ‘puzzle’ (Richards 1950) became for Douglas merely an element in the wider structural tensions and anomalies generated by intergenerational relations between younger and older men, between aristocrats and pawns, and by the politics of marriage and reproduction more generally.4 Proximity to the Brabanta plantation and the neighbouring administrative and mission post at Basongo may have been an equally significant (if more practical) consideration in her choice of field site. The suggestion that Douglas should work among the Lele came from Georges Brausch, a Belgian colonial official she met in Brussels at an ethnological congress (Douglas 1977 [1963]: xv; Fardon 1999: 48). She would travel safely if somewhat lengthily on one of the river boats supplying the plantation, and the Brabanta-Basongo complex offered possibilities of medical help should she become ill. Douglas was rather tight-lipped about Brabanta (its dubious employment practices were perhaps closer to her field site than she liked to think).5 Nevertheless, the European wives of managers and technicians at the plantation welcomed her, and even visited her in the field, which suggests it was close enough for the occasional casual unscheduled visit. In her monograph on the Lele (Douglas 1977 [1963]), the impact of colonisation is addressed, and clearly it was profound. The Lele had only come under Belgian administrative control as recently as 1935, and the plantation and the Catholic mission would assuredly have been destabilising presences. Her main fieldwork was undertaken in 1949–50, with a shorter return visit in 1953, after she had defended her doctoral thesis. She regretted not learning the Lele language as thoroughly as she would have wished. How far this hampered her fieldwork is unclear. She

22

MARY DOUGLAS

acknowledges the help of a young Lele man, Makum, recruited originally by the nuns as her cook but who became a friend, confidant and research assistant (Wachtel 2013 [2003] [reprinted in Fardon 2013b]). She eventually settled in Makum’s own village, where she was regarded as his ‘sort of ’ aunt. While her work remains largely silent about the plantation (but see Douglas 1977 [1963]: 261–63), the impact of the mission and of the Catholic church more generally is tackled more explicitly and critically in papers she published based on a return visit to the Lele in 1986 (Douglas 1999b; see also Douglas 1988 [reprinted in Fardon 2013b]). The nuns at Basongo (adjacent to Brabanta) looked after her practical requirements when she arrived and visited her in the field from time to time (Douglas 1977 [1963]: xv; Wachtel 2013 [2003] [reprinted in Fardon 2013b]). The Flemish missionary fathers will also have been interested in what the Lele believed, and why. They had made their own enquiries but doubtless were keen to hear the account of a trained anthropologist, as significantly the plantation and the mission were major accelerators of social and economic change. This gave Douglas an important research question: how were the Lele, a group with their own modes of economic production and social organisation, reacting to momentous colonial changes, such as roads, religious conversion and the banning of the poison ordeal? How, more generally, does any human group reshape its agreements concerning the rules of collective social life? Here, then, is the central focus on institutions and institutional dynamics that became her life’s work. Back from Congo, Douglas took a lectureship in a new department of anthropology at University College London, headed by Daryll Forde. Forde, a prehistoric archaeologist by PhD training but known for his research on kinship and descent in southeastern Nigeria, became an important mentor. The department remained her main affiliation for much of her professional life, including in her lengthy retirement. In the 1960s, she concentrated on the comparative analysis of societies studied in the anthropological literature, in order to develop, in Purity and Danger, a theory of institutional cultivation of styles of classification and

A BIOGRAPHICAL SKETCH

23

management of anomalies, and then, in Natural Symbols, an account of the variety of elementary forms of institutional ordering. In 1975, the first edition of Implicit Meanings gathered together the most important of her essays on themes important in her work up to this date.6 Douglas’s husband headed the research department of the British Conservative Party in the 1970s, when Edward Heath was the party leader and prime minister. An economist, James Douglas was a staunch One Nation Tory, committed to moderate and negotiated solutions to political problems. He was warned that he would not survive the political advent of Mrs Thatcher. In 1977, the Douglases decided to leave Britain ahead of this gathering political storm. Both found academic positions in the United States. Mary worked first at the Russell Sage Foundation in New York and collaborated with the distinguished American political scientist Aaron Wildavsky. The Douglases knew Wildavsky well; James Douglas had assisted Wildavsky in securing interviews with leading British civil servants for Wildavsky’s major work applying anthropological insights to public budgeting in what he came to call ‘the Whitehall village’ (Heclo and Wildavsky 1981 [1974]). With Wildavsky, Mary Douglas wrote Risk and Culture (1982), using her theory of social cultivation of perception of anomalies to engage in debates about risk assessment and technological hazards. Later she took a professorial appointment at Northwestern University, on the outskirts of Chicago, from where she began to work on religion in America. In the late 1980s, she was also able to make her only rather brief return visit to the Lele. What she saw of violent anti-sorcery fury caused her deep disquiet and triggered a focus on understanding institutional dynamics for conflict containment and reconciliation (Douglas 1988). It was during her time in the US that she revisited her earlier work on the Hebrew Bible and planned further excursions in Bible Studies. She started to learn Biblical Hebrew and absorbed a large amount of high-level mentorship from leading Biblical scholars. The Douglases returned to their home in Highgate in London in 1988, whereupon Mary Douglas resumed her old association with the department of Anthropology at UCL, first as

24

MARY DOUGLAS

professor emerita and later as a Fellow of the College. She gathered around her a group working on the risk agenda and began also to process the shock of her return to the Lele. It is convenient to see her body of work as exploring five major intertwined themes. The first concerns Africa, and African ethnography. In the 1950s, Douglas began to publish papers based on her Lele material, several of them in Africa, the journal edited by Daryll Forde (Douglas 1999d [1975]). Forde was a polymath with a broad multidisciplinary overview of anthropology as the comparative study of human cultural variation. Her own comparative interests thrived under Forde’s broad-minded mentorship. Douglas also encountered in Forde’s department a Belgian doctoral student, Jan Vansina, who had trained in medieval history but was now planning anthropological work in the Belgian Congo under Forde’s supervision. His fieldwork was to be among the Bushong people, neighbours of the Lele to the east. This prompted Douglas to look at her own material through a comparative ethnographic lens. The Lele and the Bushong lived in comparable environmental settings but organised their production systems differently. These organisational differences suggested a basis for explaining variations in thought patterns and beliefs, drawing on carefully controlled ethnographic comparison (Douglas 1962). The Lele people were of concern to her thinking to the last, and she remained convinced that all theory in social anthropology had to pass the test of adequate ethnography. The second of her focal themes emerges in Purity and Danger (1966), still her most widely read book.7 Here, she attempted a general statement on ways in which organisational commitments are buttressed by social classification and how ritual and symbolism both advance and protect schemes of social classification by allowing for the management of anomalies. She needed a wider range of materials than available from her Lele work alone, and the book draws heavily on the work of fellow anthropologists. It also uses the Hebrew scriptures as an ethnographic source on ancient society. Her aim was to examine similarity and difference in forms of social accountability both among contemporary peoples and between ancient and modern groups.

A BIOGRAPHICAL SKETCH

25

She followed Purity and Danger with further work on the mechanisms by which social life produced a classification of things, resulting in the book Natural Symbols (1973 [1970]), inspired by work on competing codes of linguistic expression by the educational sociologist Basil Bernstein (1971), and containing the first outing for her typology based on ‘grid’ and ‘group’. In that book, she reverted to a suggestion by Durkheim (1951 [1897]) that all social life, at root, was organised around two polarities – a recognition that on the one hand some social rules must be enforced, and on the other that much of what we do as a group is based on accountability to other people. Douglas realised that societies could be mapped and compared by locating them in relation to these two axes of ‘grid’ and ‘group’, combinations of which she thought might then have predictive power. This was a definite advance on Purity and Danger, in which she had written ‘beliefs in control of the environment vary according to the prevailing tendencies in the political system’ but said little about how to measure such tendencies. Now she had a way of explaining Bernstein’s codes of linguistic communication, though probably not in a way Bernstein might have envisaged. Drawing from her own Roman Catholic upbringing and sympathetic observation of the Irish immigrant community at worship in London, she was able to explain different styles of ritualisation in terms of variations in grid and group. Douglas continued to refine the approach, but it took a decade or more before the explanation was fully elaborated (Cultural Bias [Douglas 1978] is a major statement of her typology and the theory that underpinned it). Having been introduced to philosophy, politics and economics at Oxford, and being married to an economist, Douglas was also keen to find ways to engage with arguments from economic theory. Her interests in economics formed a third strand in her thinking. Her 1979 book, The World of Goods, written with the economist Baron Isherwood, showed how the economics of consumption and what economists call ‘preferences’ needed to be substantially rethought to address the implications of a large body of evidence that commitments in our use of goods and services are shaped by ritual activities. Consumption is overwhelm-

26

MARY DOUGLAS

ingly about how we display ourselves to others or pool or share tools and services in the mobilisation of group activities, and nowhere is this more obvious than in the daily rituals associated with preparing and sharing food (Douglas 1972b). It is through an understanding of ritual that we link consumption preferences to social organisation. If social organisation and ritual activity shape belief, maybe a wider audience needed to hear her argument? This began a fourth thread, increasingly taken up during the 1980s, in which Douglas sought the company of engineers, political scientists and sociologists of technology. Why do we fear to live next to a nuclear power station but cross the road in a busy city without a thought? Douglas’s answer was characteristically Durkheimian; practical concerns and organisational factors select the risks upon which we focus. It is also necessary to ask who will get blamed if risk management goes wrong. This resulted in her 1985 book, Risk Acceptability According to the Social Sciences, and the approach attracted both wide interest and substantial controversy. She turned to both environmental activism and epidemiology as spheres (among others) within which this perspective could usefully be applied. This fourth strand of her thinking was consolidated in a set of public lectures that eventually became the book How Institutions Think (1986). This was a key piece of conceptual and theoretical clarification, and Douglas wished she had written it much earlier in her career. Here, she aligned herself not only with Durkheim but also with an important but neglected contributor to the history and philosophy of science, Ludwik Fleck. Durkheim and Fleck both argued that ‘elementary cognitive process[es] depend on social institutions’ (p. 45). The way the laboratory is organised influences the judgments that scientists make. Her student, Steve Rayner, went on to demonstrate this ethnographically by documenting different responses to safety procedures among laboratory teams organised differently (Rayner 1986). The science was the same, but organisation and thought style varied. In How Institutions Think, she felt that at last she had been clear about dependent and independent variables. The X variable

A BIOGRAPHICAL SKETCH

27

comprised social action; the Y variable was the cultural result. Of course, as Durkheim himself had argued, the result sometimes fed back on activities. Waving a flag on the battlefield may rally the troops. Singing a hymn may calm grief. But that is only because the flag and the hymn have already been invested with significance on the multiple ritualised occasions on which they have been waved or sung, serving to mobilise the emotions in support of an institution. In retirement, supported by an able group of collaborators, including former PhD students with noted careers of their own, Douglas built on the gains of How Institutions Think, applying her theoretical insights to an increasingly broad field of public concerns, including climate change, epidemiology, poverty and violent conflict (notably, African civil wars). Work on this fifth and final strand ran from the early 1990s and was grounded (empirically) in what she referred to as ‘fieldwork’ in the Bible. With some command of ancient Hebrew, she drew upon several books of the Bible (Numbers, Deuteronomy, Leviticus, Ezra and Nehemiah) as data sources to develop an argument about the institutional dynamics of conflict containment and reconciliation, balancing earlier work on how conflicts deepen and amplify. This fifth phase was also bound up with some final thoughts on economics, in which she called development agencies to account. An essay in a volume on the role of culture in public policy (Douglas 2004a [reprinted in Fardon 2013a]) offered a powerful critique of the influential theories of Douglass North (1990, 1992) concerning the evolutionary role of institutions in economic development. This was a direct outcome of her Durkheimian commitment to the idea that practical factors bearing on organisation (for example, lack of access to markets or land) are causal (X variables) and cultural factors – for example, strategies for coping with poverty, or even manifestations of fatalism and apathy – are only ever results (Y variables). Her objection is to the notion of enrolling culture as a cause of poverty instead of seeing it as an effect, thus scandalously making the poor the authors of their own precariousness. Mary Douglas died in 2007, a few days after being made a Dame of the British Empire, a very senior public award for

28

MARY DOUGLAS

women of distinction. An editorial in a leading French newspaper wondered why the British had taken so long to acknowledge her world-class contribution as a public intellectual, but then (as her Bible had predicted) a prophet is not without honour except in her own country.

NOTES 1. Her mother, Phyllis Tew (née Twomey), died of cancer in 1933, before Mary Tew and her sister started secondary school. Concerning her Catholic schooling, she wrote that ‘the sense of pattern was reassuring, given the basic insecurity of being separated from our parents’ (Douglas 2005 [reprinted in Fardon 2013b: 18]). 2. Late in life, Douglas continued to regret her (relatively) poor grasp of Lele language (Wachtel 2013 [2003] [reprinted in Fardon 2013b]). Fortes was noted for his extraordinary field linguistic competence. 3. A lifelong admirer, Douglas published an insightful study of EvansPritchard’s theoretical contributions (Douglas 1980) a few years after his death. 4. Two later papers address the topic (Douglas 1964, 1969). The second is the more significant. It appears as part of a volume marking Daryll Forde’s retirement and tries to explain matriliny in agroecological terms, an argument Forde will have found interesting. 5. Comparing a map of the Brabanta holding (West Kasai Province, DR Congo), https://mapforenvironment.org/layer/info/555/Brabanta-co ncession-boundaries-Socfin#10.11/-4.4731/20.4995) and the map at p. 15 of Douglas (1977 [1963]) suggests that plantation land comes quite far inside Secteur Lumbundje-Loange, the area where the Lele constituted the largest proportion of the local population, and where we presume Douglas worked. 6. The second edition (Douglas 1999d) drops some essays and adds important new ones (notably, ‘Rightness of Categories’) and includes a second preface where she revises her understanding of Durkheim’s notion of ‘sacred contagion’. 7. Judged by number of citations – currently running at over 33,000, according to Google Scholar.

CHAPTER 2

PURITY AND DANGER

••• D

ouglas’s own style of thought is dialectical. Some readers find this problematic. Typically, she both proposes and undermines antinomies or oppositions, such as purity and danger, order and chaos, or ‘primitive’ and ‘modern’ modes of thought. In Purity and Danger, Douglas plays with these oppositions with the intention of revealing a grand design, which the author herself at times senses only partially. She takes the reader down one path towards rejecting the idea that there is a difference between primitive and modern and then changes tack to suggest that there might be significant differences after all, but not how we first imagined them. At times a line of thought peters out: she admits there is a flaw in her thinking, she lacks evidence, or more work is needed. In places she tiptoes towards an argument with many qualifications and at other times jumps right in, making bold assertions that are then modified as the argument is elaborated. Some reviewers of the book thought this made for exciting reading; for them, it was a refreshingly honest insight into a fine mind at work. Others were exasperated; in a strongly critical review, Spiro (1968) concluded Purity and Danger was full of unsupported assertions. Given her dialectical approach, honest though it may be, it was not always easy for the reader to pin down exactly what she had in mind. Grasping the book’s importance, generations of teachers of anthropology have encouraged their students to open it and dip in. A grab-and-go approach has been supported by reprinting individual chapters in anthologies. This happened to Chapter 3 (on the abominations of Leviticus)

30

MARY DOUGLAS

on multiple occasions. Douglas later tried to stop this bite-sized sampling, since her overall intention was lost. She thought that Purity and Danger needed to be taken whole, or it could not be grasped at all. To attempt a reading of such a complex and ambitious book, some guideposts are required. Fardon (1999: 84) offers a helpful five-part scheme, adopted here, though we modify some of the labels. – Introduction and chapter 1: differences are unfounded – Chapters 2 to 4: similarity is reinstated – Chapter 5: difference is reinstated – Chapters 6 to 9: social dimensions of difference in primitive societies are more fully explored – Chapter 10: similarity is revisited, in an apotheosis, linking life and death. Below, we will comment on three key points of inflexion in the argument (Chapters 2, 5 and 10) to clarify Douglas’s larger aims and then conclude with a comment on the legacy generated by the book for readers today. Chapter 2 offers the reader a good view of her working methods. She starts with medical materialism – the idea that rituals of purification or separation are hygienic in intent. Whatever people thought they were doing, rituals of purification (so medical materialists assert) prevented poisoning or encountering infectious agents. Douglas has no doubt that rituals of purification might have incidental health benefits but comments (dipping into the Hebrew Bible) that it misses the point of what Moses was doing to treat him as a public health official and not a spiritual leader. Her argument in this chapter hinges on reinstating similarity between primitive and modern views of the world. Dirt, she reminds us, is matter out of place.1 Moderns are as obsessed as the ancients with keeping the place tidy (or grumbling about those of us who do not). Why? Why do we organise our storage to keep books and breakfast cereals in separate places? Why is the dog allowed in the house but the pet rabbit kept in

PURITY AND DANGER

31

the garden? Her answer is that moderns are as obsessed as ancients with creating ordered systems. We choose the pattern. Anything that breaks the pattern risks being seen as out of place, an anomaly, or an ambiguity. Humans seem not to be able to do without classifying, but all classification creates anomaly. There will always be objects, persons, circumstances and occurrences that do not fit. There are several ways of not fitting; some cases may fall under two different categories when they should only fall under one, as in the case of Lele young people under the thumb of a male elder when they ought to be under the authority of the elders of their mother’s clan, and some cases that one would expect to be clearly classified appear to fall under no category at all, such as novel risks. Whereas different classificatory preferences develop in different times and places (some of us are relaxed about shoes indoors, others are not), all societies must address anomaly and ambiguity because no classification is all-embracing; no scheme can be applied without encountering exceptions. Understanding how human groups deal with these otherwise disturbing loose ends is core business for anthropology. The chapter reaches its goal with Douglas expounding five ways of responding to anomaly. First, the anomaly can be reinterpreted. The Nuer, she remarks, reclassify a monstrous birth as a hippopotamus born by mistake and gently lay it in the river, where it belongs. Second, the aberration may be physically controlled. Among the rural Yoruba of western Nigeria, a cock that crows in the middle of the night risks having its neck wrung so that it does not live to contradict its identity as a bird crowing at dawn (much as an alarm clock that goes off at random in the night might expect to be binned). Third, anomalous things can be shunned. Leviticus abhors ‘creeping things’; abhorrence emphasises ‘the negative side of the pattern of things approved’. Fourth, anomalous events can be labelled as dangerous, from which protection may be sought. ‘Attributing danger is one way of putting a subject beyond dispute’ (p. 49). Fifth, ambiguous symbols can be used in ritual in the same way they appear in poetry or myth, to enrich meaning or call attention to other levels of existence.

32

MARY DOUGLAS

Some anomalies may be celebrated, as in some African communities, where people with intersex conditions may be given special status. Finally, faced with anomalies beyond our powers to influence, we may simply shrug and cope. Here she anticipates the book’s conclusion, where ritual ambiguity is seen as a way of unifying the mysteries of life and death. Thus prepared, the reader is led into the book’s most famous chapter (3), on the dietary laws of the Book of Leviticus. The Hebrew cosmos was divided into separate spheres – water, land, air. Daily interactions brought these ancient pastoralists, transitioning to farming, into regular contact with a large variety of nonhuman living creatures. Each creature was clearly associated with the domain in which it lived. Birds are two-legged creatures with feathers that fly. Fish are scaly creatures that swim. An animal ‘not equipped for the right kind of locomotion within its element’ (p. 69) is contrary to holiness. Some, however, did not so clearly fit the class to which they seemed to belong. Consider, for example, a water snake (swims, but no scales) or a pig (has cloven hooves but does not chew the cud). Douglas’s basic interpretation is that the forbidden creatures are all in some way anomalous in Hebraic classification. These anomalies had to be treated with care, since they appeared to challenge the entire framework of classification around which human dominion of nature was organised. Such creatures should not be used for food. Taking note of animals that were clean and unclean at every turn ‘inspired meditation on the oneness, purity and completeness of God’ (p. 71). Food taboos that at first sight seemed puzzling made sense when the social and organisational context was considered. The idea that appreciation of nature puts us in touch with an order and unity outside ourselves is not one we necessarily find strange. The gap between the ancients and ourselves is less than we imagine. This interpretation ‘rang a bell’ with many of Douglas’s readers but did not meet with the approval of Melford Spiro, an anthropologist with rabbinical training. He objected that Douglas had insufficient scholarly knowledge to provide a convincing contextualised reading of the source material (Spiro 1968). In a return

PURITY AND DANGER

33

to the Hebrew Bible in retirement, she also returns to Spiro’s reproach and addresses it decisively (Chapter 8, below). In Chapter 5, difference is reinstated. Anthropology provides ethnographic evidence of radically different ideas about how the world works. The Azande know the poison oracle is a bodily response to a biologically derived substance; they make it from boiling up a plant. But they address it as a person, with the capability of providing unerring answers to questions concerning witchcraft as a cause of misfortune. Faced with poor returns, hunters in the Ituri forest consider the forest to be in a bad mood and sing to it to cheer it up. You can hardly imagine a modern car mechanic approaching a repair in the same way, she remarks.2 She then embarks on a defence of the category ‘primitive’. Admittedly, it can be used in a pejorative sense, but this is not what she intends. Cosmologies of causality, as exemplified by Azande diviners or Ituri hunters, emerge from the way the activity is organised and the challenges presented by such organisation, including the division of labour and the division of produce resulting from such labour. This reflects the influence of Durkheim (1984 [1893]), who differentiated societies in terms of mechanical and organic solidarity. Mechanical social solidarity, Durkheim argued, was the outcome of low levels of division of labour. Where everybody shared the same basic mode of livelihood, there was little scope to recognise complementarity of function or tolerate differences based on specialised activity. So primitive means undifferentiated – there is no shame in that (she protests).3 On the other hand, there seems no reason why Douglas should not simply proceed on this Durkheimian basis and drop ‘primitive’ in favour of the terms ‘more differentiated’ and ‘less differentiated’ societies. Even better, perhaps, if she had arranged her examples along a continuum of differentiation, with some plausible criteria for measuring complexity of organisation, because her point throughout is to stress that it is human activity – and perhaps primarily human economic activity – that is responsible for the varied sets of cosmologies and institutions that then emerge. Her insistence on rehabilitating a terminology redolent of colonial racism seems perverse. It could easily be dropped

34

MARY DOUGLAS

without damage to her larger argument that belief is a product of social organisation. So why does she doggedly maintain a distinction seemingly only guaranteed to cause offence, not least to her African professional colleagues? The answer appears to reflect a desire to position herself as a loyal student of Evans-Pritchard. She commends the French philosopher of anthropology Lucien Lévy-Bruhl for having taken further the link between organisation and belief, as proposed by Durkheim, while promptly criticising him for having assumed a distinction between ‘primitive’ and ‘advanced’ world views a priori. Evans-Pritchard, she asserts, set Lévy-Bruhl back on the right track by demonstrating how Azande witch-finding practices allowed practical response to tensions and contradictions in a world of peasant subsistence production.4 She might as easily have argued that practices – whether based on analogy or analysis – sustain beliefs everywhere, and this is what she does in later work. If the Azande were primitives then we are all primitives. In Purity and Danger, however, she persists in dialectical mode. This sent a signal to her colleagues, who would also be her reviewers. This was the first time she was placing her ethnographic work on the Lele in a larger comparative setting. In the decade in which the book was in the making, the bulk of ethnographic writing continued to concern globally ‘remote’ rural communities, and many anthropologists saw time ‘in the bush’ as a rite of professional passage. To gain a sympathetic reading, she needed to show these paragons of arduous fieldwork that she was fully taking their own hard-won material into account. ‘Primitive’ made clear to her professional peers it was their sort of book – proper anthropology. Chapters 6–9 cover the gamut of these kinds of emblematic studies. Many kinds of ritual prohibitions, avoidances and cleansings are shown to be grounded in local tensions and contradictions concerning organisational arrangements over marriage, sex, labour, gender and age. But equally, the material suggests that social classification only works up to a point. There are always anomalies – cases that do not fit. Anomalies must be either avoided or embraced. Ritual creativity is pressed into service to

PURITY AND DANGER

35

address anomalies and prevent entire communities stumbling over exceptions. With her peers now firmly woven into her argument, she can end the book by pulling the threads together in a synthesis touching on ultimate mysteries (Chapter 10). She proposes to talk about death as a corrupting factor. Here the Lele cult of the pangolin takes centre stage. Douglas admits she does not know much about the mysteries of this cult, other than that membership is restricted to males who have fathered both a boy and girl by the same mother, and that all the family belong to clans associated with the founding of the settlement. At the time of her investigations, the Lele were also in the process of becoming Catholic converts. The reader senses some creative synthesis between the ethnographer and her informants in which the gentle pangolin, curving into a scaly ball when hunted rather than seeking to escape, becomes somehow equated with the sorrowful mysteries of the crucifixion. Far-fetched though this might seem, it gives Douglas a trump card to play in relation to social classification. The pangolin is an anomaly. Indeed, as a mammal covered in scales like a fish and found living in trees, it is doubly out of place. It is an anomaly of anomalies and can be eaten only by cult members descended from founders of the community. In her 1972a article ‘Selfevidence’, Douglas argued that the pangolin symbolised the position of sons-in-law, who had to find resolutions between the demands of patriarchs and their mother’s families; eating the most anomalous animal was, she argued, a symbolic demonstration by young men who had overcome the position they found themselves as youths by achieving the status of sons-in-law who return to live in a village founded by their own their matriclan, having fathered a boy and a girl, a much desired outcome. Having the right, denied to others, to eat the pangolin demonstrated a capacity by members of the cult to release themselves from the structural anomalies of Lele social relations more generally. In the 1966 argument in Purity and Danger, Douglas then makes a startling (and ethnographically unsupported) jump. The anomaly of anomalies is a way of latching shut the circle of life. Life ends

36

MARY DOUGLAS

in dirt, and dirt is matter out of place – another name for corruption and decay. The book ends face-to-face with the anomaly of death itself. This poetic ending is perhaps best understood as a by-product of her (Hegelian) strategy for writing the book, in which she first stresses similarity then assesses difference and finally attempts a grand concluding synthesis. Douglas sensed, however, that her conclusion to Purity and Danger was metaphorical rather than analytical, and only sketchily supported by ethnographic evidence. A later work on classification (‘Rightness of Categories’, Douglas 1999d [1992c]) advances an altogether tighter argument (Chapter 6, below). One of the benefits of Purity and Danger was that it helped Douglas to decide to ditch primitive culture once and for all. In finding a way out of this cul-de-sac, she hung tightly on to the coat tails of the Scottish nineteenth-century theologian William Robertson Smith (Bošković 2021), the scholar she credits with having first drawn ‘attention away from belief . . . to the practices associated with them’ (p. 111). Douglas explicitly added that this focus on practice is ‘not a peculiarity of primitive culture’. Rather more problematically, she then claimed that ‘the primitive world view . . . is rarely itself an object of contemplation and speculation . . . and to this extent must be taken to be unaware of itself, unconscious of its own conditions’ (p. 113). Are we to conclude that the primitive worldview is unaware of its own practices? This seems to be a contradiction, and she knew she needed to resolve it. Perhaps the only way was to abolish the difference? This might be done in one of two ways. One is to offer ethnographic evidence that ‘primitive culture’ was indeed fully ‘aware of itself ’ (an approach sometimes termed ethnoscience; see, for example, Atran [1993]). The other option is to argue, perhaps more radically (the path she took), that a lack of reflective self-awareness was true of all cultures, including cultures of science, seen by many as the epitome of rational self-enlightenment. In later work, Douglas argued not only that this lack of selfawareness was universal but that it has a specific function for institutions of obscuring the process through which the institution

PURITY AND DANGER

37

was formed. It does this by naturalising the analogy on which the institution is based. All institutionalisation seeks protection through naturalisation. In justifying an institution, we first set up a loose analogy and then try to ground it in nature – the lion (the top predator) is the king of the animals (if we are organised in a hierarchy), or everything comes round in its due season (if we are organised in an enclave), and so forth. This process of naturalisation of institutional analogies is as protective of scientific organisation as any other field of human endeavour. David Hull, with whom she collaborated editorially on a set of essays dedicated to the philosopher Nelson Goodman (Douglas and Hull 1993), argued that evolutionary theory could be used to explain progress in biological science, perhaps an example of taking the naturalisation argument to its limit! The highly organised practitioners of science often find themselves outraged that their carefully framed hypotheses about (for example) epidemics or climate change were (in Douglas’s scheme) contending for position on a level playing field with all kinds of dubious entities – processes of blame, specious accusation, social exclusion, obsession with purity and hygiene, and visitations by witches. They are hard pressed, however, to show why they should be an exception to Douglas’s general model of institutional development, when evidence is assembled to show that the institutions of science behave no differently from other kinds of institutions regarding the exercise of power and patronage, the distribution of rewards, the presence of gender or racial bias, or the governance of safety procedures. From Purity and Danger onwards it became clear to Douglas that we are all in the same boat, institutionally speaking. So let us now briefly try to sum up what Purity and Danger achieves. The book demonstrates that humans are classificatory animals and that classes do not inhere in objects or people themselves but in our desire for a tidy world. We find it necessary to know where things and people are but also seek assurance that everything is in its place. These schemes do not derive from experience but from a shared (collective) desire for order. Order is forged by ritual interaction, not inspection. It is in this sense that

38

MARY DOUGLAS

humans are ritual (as well as classificatory) animals. No ritually generated scheme for ordering the world, however passionately sung, or danced, is ever perfect. There are always things or persons that do not fit. Exceptions occur; new experiences arise. Anomalies are inevitable. They may be adjusted for, debarred, exploited, embraced or endured but cannot be ignored. Dirt is matter out of place and can be swept up, but death turns us all to dust, an anomaly we are biologically obliged to embrace. In responding to anomalies, we drive processes of change and innovation. The dialectic of primitive and modern (an obstacle in appreciating the book to some readers today) was a device for presentation. Like many such devices, it became an encumbrance. Douglas herself knew it risked tripping her up, and in her later work she abandoned it. Henceforth, her approach was based on the claim that although institutions vary, and so styles of thinking also vary, underlying all variation there is a single common process – social organisation. Sociality is the driver for all explanatory schemes for mapping the world.

NOTES 1. This much-quoted remark is often attributed to Douglas. Douglas herself cited the eighteenth-century aristocrat Lord Chesterfield as its source, but it has not yet been traced in his writings. Fardon’s (2013c) research concludes that the nineteenth-century statesman Lord Palmerston used something close to Douglas’s phrase in an 1852 speech, in which Palmerston himself claimed to have heard it from others. Fardon finds that the phrase was later reworked by others, including William James, to become closer to Douglas’s formulation. 2. In fact, we can – the ethnography of car mechanics throws up a few surprises (Dant 2010). 3. Measuring the degree of differentiation in the division of labour presents major problems, which neither Durkheim (1984 [1893]) nor Douglas (1966) recognise. The role-intricacy of Lele or Tallensi life shows that even if by some measure they used a smaller number of skills than other peoples, their social systems were certainly neither less complicated nor less complex as a result.

PURITY AND DANGER

39

4. Douglas returned to the theme, and to Lévy-Bruhl, with a more sophisticated treatment. In a lecture given at the Collège de France in April 2001 (Douglas 2008), Douglas argues that while Lévy-Bruhl was wrong to suppose it ‘primitive’, he rightly diagnosed a thought style of analogical thinking that contrasts clearly with analytical styles. She finds analogical thinking widespread in the ethnographic record, in the Torah, in Homeric and many other ancient literatures, as well as in the everyday speech codes of many people living in developed economies. Its cultivation, Douglas argues, is to be explained by forms of social organisation.

CHAPTER 3

FOUR SIDES TO EVERY QUESTION

••• I

n the late 1960s, London was experiencing major social change. An economic boom drove the importation of labour from other regions. Immigrants from the New Commonwealth (former colonial dependencies, especially the Caribbean) were staffing public transport and medical care. Immigrants from western Ireland – a long-established British labour ‘reserve’ – sought daily paid labouring work in the building sector of an expanding city. Notices appeared in many cheap lodging houses and rented properties, stating, ‘No blacks, no Irish’, starkly manifesting a classification of difference that anthropology could not ignore but that cut across a racial boundary. What was needed to explain a form of stigmatisation that linked immigrants from Jamaica and Donegal? Goffman’s book on stigma had described the process of classificatory stigmatisation and pointed to ‘group alignment’ dynamics (Goffman 1963), and his later studies on ritual interaction (Goffman 1967) had indicated that everyday ritual was central to the process of social organisation quite generally, but he had not successfully explicated the causal mechanism. Moreover, ‘ritualism’ had by the 1960s become in itself a stigmatising term (Douglas 1982a). Purity and Danger had made Douglas aware that she needed better explanatory mechanisms. Carving the world into ‘primitive’ and ‘modern’ explained little about social classification as a mechanism. Her faith in Durkheim’s insight that what we think

FOUR SIDES TO EVERY QUESTION

41

is driven by who we are, how we act and how we are organised, and linked to belief and understanding by ritual, remained undiminished. The problem was to develop a better explanation of how ritual worked. Natural Symbols was published in 1970, and in a revised edition in 1973. Fardon (1999: 104) characterises it as ‘a defence, both passionate and reasoned, of the importance of ritual to social life’, adding that ‘four long books might have begun to explain clearly what she was trying to do; a single short one could only put down markers against arguments to be developed later’. As someone of part Irish descent, and a Roman Catholic, she also had personal stakes in the issue. She needed ways to defend Irish labouring families (and herself ) from the charge of mindless ritualisation. Disdain for ritual (she thought) was a social prejudice, and perhaps as firmly rooted as racism, where prejudices are based on observable physical differences such as skin colour. It reflected the outworking of a particular type of exclusionary social classification. Perhaps sensing herself to be on the wrong end of the notice ‘No blacks, No Irish’, she positively embraces derogatory social terminology. Chapter 3 is bluntly titled ‘The Bog Irish’ and offers a sympathetic account of why working-class Irish families in London continued to embrace the Friday abstinences mandated by the Catholic church. This in-your-face approach to ritualisation disoriented some of the book’s reviewers. In Cambridge, Edmund Leach (missing the point of the game she was playing) had decided that ‘the object of the exercise is to adapt her learning to the service of Roman Catholic propaganda’ (Fardon 1999: 103). The Edinburgh anthropologist James Littlejohn thought that the book’s passion was akin to desperation and demonstrated that ‘verily, God is dead’ (Fardon 1999: 124). But her objective lay elsewhere – namely, to model the multiple ways in which social organisation triggers ritualisation, in turn shaping how belief is framed and used. This meant seeing ritualisation where others might only have seen reason. Also, it meant exploring its opposite – the antiritualism evidently gripping several of her more hostile reviewers. The opening sentences of Chapter 2 nail the issues: ‘Those on the New Left who are in revolt against empty rituals do not see

42

MARY DOUGLAS

themselves walking in the footsteps of [earlier] ardent Protestant reformers . . . alienation from the current social values usually takes a set form: a denunciation not only of irrelevant rituals, but of ritualism as such’ (Douglas 1973 [1970]: 21). Her colleagues – at that time developing a progressive Marxist anthropology in the Department of Anthropology at University College – were the New Puritans, smashing statues. But smashing statues is itself a type of ritual. She needed a single explanatory frame that embraced both ritual and anti-ritual. Developing this unified framework occupies a major part of the book. Fardon (1999: Ch. 5) suggests that Natural Symbols is perhaps Douglas’s most important book. He is referring to the scope of the issues it raises, not the clarity of exposition or power of its conclusion. She herself regarded it as the workshop that gave birth to all her later work. Undoubtedly important, the book has too many moving parts to be a success in its own terms. It resembles a workshop yard full of items marked for later use rather than a well-organised manufactory. The reader might also wish to set it aside for later attention, but perhaps not before taking note of Chapter 4, ‘Grid and Group’, which is her first stab at outlining a mechanism for linking social organisation and belief, via ritual action of all kinds, including anti-ritual. Douglas, and others, later used the phrase ‘grid-group analysis’ to characterise what was especially distinctive about her typology of social classification and its consequences. But the typology was not itself the theory, as is made apparent in her later book How Institutions Think, where she manages without explicitly mentioning grid and group. Douglas also has a distinctive Durkheimian theory of causality that rather than explaining little ideas by big ideas as so many traditions of social science do instead argues that through ritual interaction social organisation sustains styles of thought. This theory was set out only sketchily in Natural Symbols but was worked out much more fully in How Institutions Think. The problem with Douglas’s typology, in its first form, is that many readers found it unclear what she meant by the dimension of ‘grid’, other than something constraining. The meaning appears to shift even within the confines of the book. There are later

FOUR SIDES TO EVERY QUESTION

43

and much more coherent accounts, to which we will turn below. But first it is helpful to get a sense of what she was trying to do by paying attention to her source of inspiration. This is explained in Chapter 2, ‘From Inner Experience’. Here she outlines her debt to the educational sociologist Basil Bernstein. Bernstein (1971, 1975) was concerned with the apparent limited educational attainments of children from a working-class background. It was not that they were less bright than children from the homes of middle-class professionals but that they seemed less well-adapted to or motivated by certain kinds of learning tasks. Specifically, they were poor in discussion. Did they lack ambition, or was the curriculum at fault in not engaging them adequately? What part did home background play? This last question implied to Bernstein the need to address home-based differences in types of discussion between parents and children. Might there be homes in which children were not encouraged to interrogate their parents? Perhaps they brought this code of restricted speech into the classroom? Bernstein was interested in establishing a scale representing positioning between poles he termed restricted and elaborated speech. These forms of speech, he concluded, are ‘contingent on a form of social structure’ (p. 26). Douglas, building on Purity and Danger, realises that this ‘form’ both of speech and social structure reflects a kind of classification. ‘Gridded’ at home by a scheme of classification, children were controlled by pressure from other people. In Chapter 4, Douglas tried to capture her generalisation from Bernstein’s idea of a ‘restricted code’ in a simple diagram (Douglas 1973 [1970]: 64), labelled ‘grid and group’. This diagram is hard to interpret. It appeared to represent the intersection of two dimensions. Where they meet at the middle was identified as a zero point. Each of the four vertices was marked with an arrowhead, implying intensification outwards (as confirmed by four ‘+’ signs). The horizontal axis, labelled ‘group’, was marked (to the right) as ‘ego, increasingly controlled by other people’s pressures’ and (to the left) ‘ego, increasingly independent of ’ such group pressures. Problems set in with the vertical dimension, which was labelled at the top of the diagram ‘grid’, with the

44

MARY DOUGLAS

arrow pointing upwards labelled as ‘system of shared classification’, and at the bottom of the diagram, with an arrow pointing downwards, as ‘private system of classification’. Earlier, she had remarked that ‘this private system of classification would point away from communication with others, eventually to madness’ (p. 63). Nothing was said about how such idiosyncratic schemes might arise, nor is there any instance given of what they might comprise. As a theory of mental ill-health, this was simply a stab in the dark. To illuminate what she had in mind by ‘high grid’, she returned to Fortes on the ethnography of the Tallensi. Douglas viewed the Tallensi as a ‘high grid’ society: ‘here, the public system of rights and duties equips each man with a full identity, prescribing for him what and when he eats, how he grooms his hair, how he is buried or born’ (p. 66). She then turned to the peoples living around Lake Nyasa in central Africa, ‘where long wars . . . with slave-raiders had already broken up local social structures before the colonial freeze came down’ (p. 67). The peoples of this region were ‘much lower on the line of coherent classification than the Tallensi . . . their culture promises them contradictory rewards and holds out impossible goals’. The hypothesis seemed to be that some African groups were better than others in protecting local schemes of social classification in the face of potentially destabilising external factors, such as slave-raiding. British colonial protection had then locked these schemes of ‘grid’ positioning in place, as reflected in the structural-functionalism of the great anthropological monographs on rural Africa from the 1930s through to the 1960s by Fortes, Evans-Pritchard and others. Her basic point was to suggest that ‘any bureaucratic system which is sufficiently secure and insulated from criticism will tend to think the same way’ (p. 66). But what about social systems not so protected? For that we need to understand the interaction of ‘grid’ and ‘group’. This is where Natural Symbols provided only partial answers, in part because she is busy addressing her most immediate concerns with the ‘no blacks, no Irish’ notices. Skimming over a range of important theoretical issues more fully explored in

FOUR SIDES TO EVERY QUESTION

45

later work, she then draws a parallel in the prejudices displayed by London landlords and the revolutionary intolerance sweeping through the academic social sciences. Her argument is that anti-ritualism (or iconoclasm) was itself a type of ritual engagement and certainly no barrier to entry into academic debate for persons like herself ‘gridded’ by a Catholic heritage. For those who were there, her choice of destination evokes that turbulent era, but for a reader today perhaps it seems puzzlingly obsessive. The core problem in trying to define ‘grid’ as a particular type or practice of classification was that it did not help her theory, which was supposed to explain classification by reference to social ordering: social ordering itself therefore needed to be defined independently of classification, to avoid the risk of circularity. Her later accounts of ‘grid’ and ‘group’ became clearer when detached from the context of social science debates associated with the revolutionary late 1960s. Douglas’s next major statement of her typology appeared in Cultural Bias. There she defined ‘group’ as the degree of social incorporation within a bounded group (Douglas 1982a [1978]: 190), which is very close to Durkheim’s concept of social integration. Although her brief opening definition of ‘grid’ rather confusingly stated that it was the degree of ‘individuation’, the paper’s subsequent discussion was much clearer. Grid was ‘the crosshatch of rules’, which at one extreme permits transactions to be entered into freely and at the other regulates interaction, restricts options and controls behaviour; for example, by limiting permissions by rank. Douglas was then able to use her two dimensions to code both ‘social context’ – meaning the social ordering – and ‘cosmology’ – meaning the way people think – to show that weak or strong social regulation and weak or strong integration in social organisation cultivates weak or strong cognitive regulation and integration in people’s understanding of their world. Most of the paper, though, was devoted to measures of ‘cosmology’, or how people represent the world or connect (or do not connect) their representations, rather than to measures of social ordering. In Cultural Bias, strong grid with weak group was described as ‘insulated’ social organisation and later

46

MARY DOUGLAS

‘isolation’ (p. 235), which interestingly prefigures Douglas’s later preferred term, which was ‘isolate’ ordering (e.g. Douglas 1996). Weak grid and weak group made ‘individualist’ organisation. At this stage, there were no separate labels for the two forms made by strong group together with either weak or strong grid, but her examples were drawn respectively from bureaucratic contexts, and from voluntary groups and sects. The paper also considered the case of hermit life, which Douglas took to lie outside the four categories, but perhaps might be a very extreme form of individualism in which even transaction is eschewed. Unfortunately, few of her examples of hermit life were ones in which social ties of any kind had genuinely been severed. Cultural Bias was an advance on the account in Natural Symbols, not least because the grounding in Durkheim’s two dimensions from Suicide made them much clearer, and because in that paper she was able precisely to distinguish ritualised social relations from the thought that they cultivated. Yet empirical evidence with which to support the central explanatory claims, or even to validate the typology at the descriptive level, was still to be collected. For much of the rest of her life, she worked with former students and colleagues to encourage a range of empirical studies for precisely this purpose. She began this collaborative work with the edited collection Essays in the Sociology of Perception (Douglas 1982b). In that collection, Ostrander’s (1982) opening piece provided a detailed comparison between Douglas’s two dimensions and others from other traditions of anthropological and social theory but finally (p. 26) reduced the definition of ‘grid’ to ‘the degree of restriction on how individuals may behave’ and ‘group’ (p. 25) to ‘the individual’s degree of identity with a specific group’. In doing this, Ostrander explicitly drew the connection with Durkheim’s (1951 [1897]) measures of social regulation and social integration. Each of the theoretical and empirical studies collected in the volume then developed indicators for these dimensions appropriate to the fields of ‘social context’ and ‘cosmology’ that they examined. Since that collection was published, studies using Douglas’s typology, often based on fresh

FOUR SIDES TO EVERY QUESTION

47

specific indicators for particular empirical purposes, have proliferated across every field of the social sciences and humanities, but the grid-group analysis and ‘cultural theory’ traditions are especially well represented in political science, public policy and public management, science and technology studies, management studies, and health behaviour and health services research. The one discipline in which Douglas’s method has probably had least impact has been anthropology itself. Towards the end of the 1990s, Douglas took the opportunity to summarise the approach for a new generation of scholars working across social science disciplines in what is perhaps the best single, short account of her typology. The article appeared in 1999 and was titled ‘Four Cultures – The Evolution of a Parsimonious Model’, published in a special issue of Geojournal following a workshop at the University of Amsterdam (Douglas 1999c [reprinted in Fardon 2013a – page references are to the version in Fardon’s collection]). The paper reports on work by Douglas and colleagues for which she uses the label ‘cultural theory’. Although – as the original publication dates of the articles collected in her (1992a) Risk and Blame collection make clear – Douglas had begun to use this term around 1987–88, it was perhaps an unfortunate choice because it caused confusion with the much broader field of cultural studies, which was becoming a recognised field at about the same time. Worse, the label focused on the ‘cosmology’ or ‘thought style’, which is what Douglas was trying to explain; yet using ‘culture’ in the name of the theory risked leaving readers with the wrong impression that culture was doing the explaining when variation and dynamics in social organisation are the explanatory forces. As presented in ‘Four Cultures’, cultural theory was underpinned by a ‘parsimonious model’, a revised and clarified version of ‘grid and group’, as first explained in Natural Symbols. The article was a succinct restatement (‘a summary of the early history of the theory’) of what Douglas had tried to achieve using grid and group. It began with a bold and clear claim: ‘Cultural theory is good at explaining irreconcilable differences’ (p. 53). It was not a theory of conflict resolution but a means of examining

48

MARY DOUGLAS

why differences in viewpoints become so intractable. It sought to explain the ‘culture wars’ forming around rival claims to social, cultural or national identity. Douglas diagnosed the condition in the following terms: ‘when disagreement is so deeply entrenched the protagonists to a serious debate seem to be acting irrationally . . . there are reasons for their intransigence, reasons which go deep into their loyalties and moral principles’ (p. 53). Her first move was to label this, referring back to her 1978 paper, as ‘cultural bias’ and to define it as ‘steady preference for one or another set of institutional forms and consequently a commitment to the kinds of knowledge that go with it’. We are here in the terrain of what one of President Trump’s supporters once called (to widespread derision) ‘alternative facts’. Such mockery is misplaced, since it ignores the possibility that all parties to a dispute will marshal evidence according to their biases. We all suffer from ‘cultural biases’. We live in information bubbles, but Douglas explains that these are bubbles of institutional preference. Unless we see the institutional framing and the different viewpoints these sustain, we will never even begin to understand why we disagree. The beginning of wisdom was to recognise ‘a four-fold set of biases, each defined against the others and each sustained by the activity of aggressive self-definition’ (p. 53). Why any more than one? Rational choice theory, so dominant in many of the social sciences, largely works with a single form of rationality, in the form of the maximisation of subjective expected utility under constraints. Douglas had argued (Douglas and Ney 1998) that this approach drastically under-performed in explaining variation in styles of reasoning. If institutional economics had introduced an allowance for variation in ways of organising, it had not recognised the depth to which that variety cultivated distinct styles of rationality. But why only four styles? In part, this is because any explanation needs to be parsimonious, otherwise description and dissection of social difference would never end. Douglas argued frequently against the common presumption in history and anthropology that every case is unique because this did not only

FOUR SIDES TO EVERY QUESTION

49

rule out comparative analysis but also made explanation impossible, for any explanation implicitly refers beyond the unique case. Methodologically, four forms are enough to suggest range without getting lost in detail: ‘four . . . types of social environment are enough to generate . . . four . . . types of cosmologies, which stabilise . . . four . . . kinds of organisation’ (p. 54). These, though, are pragmatic reasons for limiting the extent of plurality. There were also deeper theoretical reasons that the four forms were jointly exhaustive, at least at the level of generality that is appropriate for truly elementary forms. In part, her argument is that at the elementary level variation is limited by the basic common acts of human experience and physicality. The range of variation in social integration and social regulation, or group and grid, measures the limits of what is possible for a species that must not only eat, find shelter, sleep and reproduce but also sustain the meaning of relations with others for survival. Douglas argued that these two dimensions are indeed as basic as Durkheim (1951 [1897]) had suggested. Deductively, their cross-tabulation, and a focus on the cells rather than the apices of the diagram, must yield only four forms. To put the point more forcefully, Douglas argued for a kind of ‘realism’ about the status of the dimensions of social regulation and social integration. The two dimensions and the four forms are not just a heuristic device. They are not just Weberian ideal types. Variation in strength and weakness on each dimension is limited by physical and social requirements for human thriving, but the four basic combinations of weak and strong conditions on the two dimensions are, she argued, the standard poles of social organisation around which human social life is always attracted,1 and the tensions and fragile, provisional settlements among them shape our conflicts and drive social change. Choosing labels for the forms in those four cells continued to be a problem, since words have connotations, and some may appear (to some) to be pejorative. Although in the essays collected in Risk and Blame Douglas had referred to strong grid and strong group as hierarchical, she recognised that ‘hierarchy’ is a bad word to some, ‘market’ a bad word to others. Therefore, for the

50

MARY DOUGLAS

sake of a level playing field, in ‘Four Cultures’ she preferred labelling the four forms A, B, C and D. She then produced Figure 4.1: ‘a cultural map’ (p. 55). The derivation of this map is not fully explained. It would perhaps have been clearer if she had identified the vertical line as ‘grid’ and the horizontal line as ‘group’ and explained that by ‘grid’ she meant degrees of social regulation and by ‘group’ she meant degrees of social integration. Something similar had been attempted in Durkheim’s (1951 [1897]) foundational book on suicide. This identified three distinct types of suicide (four if one includes a lengthy footnote on ‘excessive regulation’), where in effect those taking their own lives are described as exiting life from the apices of the dimensions, rather than from the cells (for a mapping of Durkheim’s on to Douglas’s scheme, see 6 2007: 133). These different forms of suicide were seen by Durkheim as results of distinct kinds of institutional pressure. The bankrupt trader can see no more purpose to a life in which wheeling and dealing is no more possible. The failed military leader takes the loaded pistol for the honour of regiment. Communes are prone to collective suicide (e.g. the Jonestown cult). In the long footnote, Durkheim speculates whether slaves’ willingness to embrace death in the face of extreme regulation was a fourth type (p. 239). Earlier in the article, Douglas writes that ‘there are only four stable organisational forms’, with the remaining possible mixtures ‘assumed by the theory to be transitional’ (p. 54). In ‘Four Cultures’, Douglas prefers to allow her cultural map to stand unadorned. Perhaps she thought her readers retained some sense of what she was arguing from her earlier work. She ploughs on regardless; in part because she is more anxious to explain the origins of cultural theory as the product of collaboration with colleagues. In her other work from the same period, these four stable forms were recognised through a variety of labels, but one set comprised (from bottom left, reading clockwise) ‘individualism’, ‘fatalism’, ‘hierarchy’ and ‘sect’ (or ‘egalitarianism’). Later, ‘fatalism’ was replaced by ‘isolate ordering’ because the latter term captures social organisation rather than a mindset or belief. Douglas also replaced ‘sect’ with ‘enclave’ to focus on the

51

FOUR SIDES TO EVERY QUESTION

elementary form rather than on a specific empirical case arising in doctrine-centred religions and doctrine-defined political movements. Figure 3.1. shows Douglas’s account of her typology in its final form. In it, we use Durkheim’s original terms, ‘social regulation’ and ‘social integration’, for the two dimensions to emphasise the theoretical roots of the typology. The figure also uses the labels for the forms on which Douglas settled by the 1990s and early 2000s. It also shows in the extreme corners the conditions of disorganisation to which each of the forms can lead after the ritual   
   
 


  
   


  


   


 
  


 !


 
%

"



"! 
!


 


! 
!""


 %

!


  


  

 !


 " 
!



#



"


 %

!


 
#





! 

!
%

 

 






!
 

 
%%
#





  
%
 




#

 








"! 


"






→

  
   
 
  


 
! %
  "
&!
 
 " 

%

#
! 
!

"%
#
"!


 ! 

 


  

  
   
 
  


 
 
$ 


 


 " 
%
$! 





 

 "
 

Figure 3.1. Positive and negative feedback dynamics of the four elementary forms. Source: 6 and Richards 2017.

52

MARY DOUGLAS

process of reiteration and amplification pushes each of the ways of organising to its most drastic extreme. As we shall see in later chapters, these kinds of positive feedback processes greatly concerned Douglas in her later work, especially in the case of enclaved ordering, and in her final writings she sought to develop an account of how they might be contained. These processes of positive feedback are shown in the solid arrows. Much of Douglas’s work was concerned with the conflicts which arise between people cultivated by their institutional ordering to see their world differently, and who therefore react against other ways of ordering. Negative feedback of this type among the forms is shown by the hollow arrows. There is another slightly submerged stepping stone in the river that Douglas is trying to cross in ‘Four Cultures’. She still lacks a fully specified mechanism. Too much emphasis on the cultural map risked arid typology – classifying this and that group or community only as A, B, C or D and treating forms as static rather than using the theory to explain their dynamism, rise and fall, clashes and settlements. For by 1999, it had become clear from the emerging body of empirical work by her colleagues (e.g. Coyle and Ellis 1994) that, even when studied cross-sectionally, cases are typically made up of mixes, settlements or tensions reflecting relationships between multiple forms. Attention now had to shift to how the dynamics of the theory made possible these mixes, settlements, accommodations and plural rationalities (Verweij and Thompson 2006) and how it sustained persistent tensions in which none of the forms was eliminated (6 2003). Douglas was more interested in solving the problem of how people create or adopt a culture, defined as a stable configuration of social practices, cosmologies and organisation. She rejected the standard criticism of Durkheim and his followers that in some way Durkheimian sociology was built on the idea of a group mind. ‘Personal autonomy’ was a social fact, but how then did such autonomy result in ‘creating or adopting a culture’? Answers had begun to appear in work she did with colleagues during the 1970s and 1980s on household consumption (Douglas and Isherwood 1979). ‘People run their homes in all sorts of ways. No one tells

FOUR SIDES TO EVERY QUESTION

53

them to have meals punctually, or to have meals at all. Yet if they adopt a time pattern or refuse to adopt one at all the choice rules out other things they can do’ (p. 56). Pressures from employers’ and schools’ timetables often do constrain when people must eat or at least give their children breakfast; social regulation limits choices, imposes costs and cultivates commitments. Recognising that organisational choices create and sustain institutional values brought into her thinking some provisional clarity about cause (X) and effect (Y), and she goes on to illustrate how these X and Y variables could be arrayed in a quantitively analysable space, drawing on work by her colleagues Karl Dake and Michael Thompson (1993, 1999) and on unpublished studies by Gerald and Valerie Mars on household cultures and self-reported behaviour (Douglas 1999c: 57). Again, however, there is a degree of under-explanation. The scattergram borrowed from Dake and Thompson (1993, 1999) has vertical and horizontal axes simply labelled Dimension 1 and Dimension 2, which are not grid and group; rather, they are dimensions generated statistically but inductively from the data using a procedure called discriminant function analysis. Dake and Thompson are able to mark the locations on the scattergram that correspond to ‘centres of gravity’ for each of Douglas’s four forms, based on the anthropological work on the 77 households that underpins the scattergram. Douglas concludes this stage of her presentation with the remark, ‘it would be feasible to devise measures of how well a given community achieves its own ideals for all its members and to compare those ideals with those of the researcher’. This is vague, to say the least. The feasibility is aspirational, not demonstrated. Nor is it clear what such a comparison would yield. Her starting point appears to have been that ‘alternative facts’ are incommensurable. What seems to be needed is a clearer demonstration of the ways in which different patterns of social activity generate different cosmologies, in turn stabilising different forms of social organisation. Douglas describes her approach as an example of a ‘strong programme’. This is an allusion to the Edinburgh school of sociology

54

MARY DOUGLAS

of science (Bloor 1991 [1976]), in which a key point was to treat organisational and institutional factors in science as independent (X) variables and scientific results (whether later replicated) as dependent (Y) variables. ‘Weak’ programmes used social variables only to explain falsified explanations in science (such as phrenology). This made no sense, since science proceeds according to specific organisational forms that are independent of the later support from or correction by the outputs of science (whether a scientific explanation is factually correct or not). The same institutional mechanism is at work whether it helps someone to reach a true conclusion or allows them to come to a false one. Each of the four types of bias enables us to appreciate truths selectively as well as leading us to accept falsehoods. Douglas appears to have had this point in mind in applying the label ‘strong programme’ to her grid-group scheme; social organisation is prior to and stands in causal relationship with cosmology. Social action is a causal (X) variable, and the beliefs that people hold about the nature of their society is a dependent (Y) variable, and never the other way round. The last part of the paper (pp. 58–61) is then concerned with exemplifying each of these four forms and arguing that they are dynamic rather than static types. In Douglas’s perspective, there can be no means of reconciliation merely via argument calling to argument. We have a world divided into four siloes, each armed with its own distinctive sets of ‘keep out’ messages. How do these siloes emerge? The theory – as Michael Thompson (1996, 2008 and Thompson, Ellis and Wildavsky 1990) had extended it from Douglas’s work – proposes that each silo arises in countervailing response to disappointment or disadvantage experienced from life under any of the others. When people seek to circumvent, rebel against, negotiate with or control institutions that disappoint or disadvantage them, they inevitably reach for one or more of the forms that give them stronger or weaker social regulation or social integration than the institutions that have offended them (Thompson 1996, 2008; Thompson, Ellis and Wildavsky 1990). Equally, the ritual reinforcement of institutions can exaggerate and amplify them to the point of selfdisorganisation. In either dynamic, there will be surprises.

FOUR SIDES TO EVERY QUESTION

55

Dynamic modelling of these feedback processes can only take us so far. How were settlements and accommodations possible, if not the work of ideas? What was needed was ethnographic evidence and, in particular, observation of the ritual processes through which social activity and social organisation came to be articulated. How do conflicted parties approach each other? If you had put this question to the women’s movement for peace in Sierra Leone at the height of a brutal civil war that threatened to destroy that small African country in the mid-1990s, they would have answered ‘we will cook for the rebels’ (Richards 1996). When we talk about ritual, people often tend to think in terms of grand occasions, such as the services recently held to mark the platinum jubilee of the British queen. The greater part of ritual is minor quotidian social exchange – a handshake or a shared drink, polite talk or the rules for a meeting (Goffman 1967; Schwartzman 1989). In Sierra Leone, ‘cooking for the rebels’ is an everyday but profound expression of social concern. It signifies a basic level of social inclusion. To take another example, the game of cricket in the English county of Yorkshire currently has a problem. Club cricket is the core of the game. Playing remains a popular leisure time activity, boosted by a large South Asian immigrant population originally attracted by work in the Yorkshire cloth industry. But the various leagues are differentiating themselves into those made up of ‘white’ and ‘Asian’ teams only. According to some accounts, the differentiating factor is not the game itself but the quotidian rituals associated with socialising after the game – specifically, the part in this socialisation played by alcohol and sausage rolls, proscribed for those obedient to Islam. In Sierra Leone, where conversion to Islam or Christianity was common in the colonial period, and where families marry across faith lines, the issue of accommodating ritual food and drink avoidances within mixed faith families and among neighbours is a well-practised art. Interfaith socialisation includes the preparation of gifts among neighbours of appropriate food at major Muslim and Christian holidays and is hardly an issue to excite comment. So well-adjusted is this interfaith accommodation that

56

MARY DOUGLAS

even the protagonists of the rebel war were unable to find any purchase in manipulating differences of faith or ethnicity. The membership of the rebel Revolutionary United Front was found at the end of the war to reflect the same patterns of faith-based and ethnic differentiation found in the country as a whole. Douglas was right to place so much emphasis on quotidian ritual as a key to understanding the ways in which social life is channelled. Everyday ritual interaction orders (Collins 2004) are hugely important in stabilising vast amounts of social activity into recurrent patterns of organisation. Her account of the mechanism, as this chapter has suggested, may have been muddled and incomplete in Natural Symbols but gradually became clearer. The ceaseless surge of social life is stabilised and rendered meaningful by quotidian ritual, and indeed by anti-ritualism, which produces its own distinctive kinds of rituals. Rituals impose patterns by repetition; anti-ritualistic rituals do so by iconoclasm. From these things, organisation and predictability can be developed. But more work was needed. How, specifically, can we explain the emergence and functioning of institutions, and what part is played by social classification? This required a fuller account of both the grounding of institutions in analogy (Chapter 5, below) and how chains of social exemplars and systems of classificatory denotation are formed (Chapter 6, below).

NOTE 1. The concept of a set of ‘attractors’, meaning a small number of forms or positions in which elements in a system provisionally settle but when they move they can only move between these few forms or positions, has a central place in complexity theory. Thompson (2008: 67–81) developed Douglas’s argument and her answer to the question ‘why four forms?’ within a framework of complexity theory, treating the forms as attractors.

CHAPTER 4

RISKS AND SOLIDARITIES

••• A

fter Natural Symbols, Douglas’s grid-group scheme took on a life of its own. It became a device through which a number of her collaborators sought to contribute to debates about environmental and technological hazards, and this resulted in divergences from her own line of development, resulting in some misunderstanding among her readers as to where exactly she was heading. By the late 1990s, Douglas felt a need to clarify. In her summary of the history of the grid-group scheme (Douglas 1999c), she noted two interpretations. The first was based on the continuous variation of grid and group, allowing an almost endless variety of outputs. Steve Rayner pioneered this approach to modelling the two dimensions as making up a continuously differentiable space (e.g. Rayner 1984). The second was the ‘strong programme’, resulting in only four stable cosmological forms (maybe three, if isolates are left out [see Rayner 1999], or maybe five, if hermit forms are admitted [Thompson 1996]). In a typical example of ‘strong programme’ thinking, Michael Thompson produced a vivid graphic suggesting the four basic ways in which environmental dynamics could be viewed from the four main social perspectives captured by the grid-group scheme (Thompson 1996, 2008). Thompson was careful to keep the X and Y values separate, ‘deriving preferences from ways of life’ (Thompson, Ellis and Wildavsky 1990). Four different forms of social interaction sustained four different cosmological perspectives. Another collaborator, Aaron Wildavsky, although sometimes recognising the causal

58

MARY DOUGLAS

force of social context (Wildavsky 1994a) took another view, seeing cosmologies as drivers of political processes (Wildavsky 1994b). Douglas wrote her (1982) book Risk and Culture jointly with Wildavsky. The authors add a note of explanation, stating the book could not have been written other than jointly. But examination of the references suggest which authors were mainly responsible for which chapters. Some chapters are rich in reference to Wildavsky’s field of political science and others are rich in reference to Douglas’s area of African ethnography. The book addresses the issue of risk. Life is a hazardous journey, and we would never even begin it if we worried about everything that might go wrong. Douglas argued that under at least three of her four basic types of institutional ordering risks were understood as anomalies, with a low-risk world expected as a kind of default. In eighteenth-century Europe, risk had a neutral connotation (Bernstein 1996; Hacking 1975). Those at the gambling tables expected gain. Losses then quickly came to be seen as synonymous with ‘danger’, and so did risk itself (Douglas 1992a), classified as a condition of exception – in other words, an anomaly. Perhaps only under isolate ordering, in which chances of adversity and affliction are an ever-present condition of life, are risks not anomalous. But this alone did not explain why we are so selective in which risks we worry about. Some risks haunt us (nuclear energy, perhaps) while others raise no qualms (the risks associated with crossing the street). So why do we worry about some risks more than others? Douglas’s answer is that social organisation sets the agenda (the X variable). Whether we are right or wrong to weight our risk agenda in the way we do (the Y variable) only time will tell. But specifically, the risk weightings can be measured in terms of grid and group. The chancers in the bottom left-hand corner are exhilarated by danger – the prizes go to risk-takers. Those whose daily life unfolds under hierarchical ordering are convinced that someone is in charge and everything will be well. Those who are ordered by the internally egalitarian rules of a commune see the outside

RISKS AND SOLIDARITIES

59

world as unremittingly hostile and rush to reinforce the defences at the first sign of an incursion. The isolates shrug their shoulders, convinced there is no alternative than to take what life throws at them. Wildavsky seems to have been deeply concerned by one group in particular – the sectarians. In her own later work, Douglas (1996) would call the world of strong group and weak grid that of the ‘enclave’. But it was probably Wildavsky, seeing the issue in terms of ideas rather than of social organisation, who persuaded Douglas in Risk and Culture to accept the term ‘egalitarian’ for this cell in her matrix. (Those who are strongly integrated but weakly regulated tend to mark their boundaries very heavily, so reinforcing inequality between members and nonmembers, even if they treat all members as equal.) In Chapter 8 of their book, some of Wildavsky’s own political concerns become explicit. America is a border country, settled by sectarian minorities escaping religious persecution in Europe. It was in this sense God’s own country and had resisted the development of a strong centralised state. Industrialisation created new wealth and social mobility but generated insufficient jobs for the highly educated children of sectarian religion, who then took their communitarian cosmology and applied it to environmental causes, using state subsidies and mail order membership recruitment methods to attack not only the military part of the military-industrial machine but its industrial half as well, in the name of wilderness protection. Wildavsky’s own preferences appeared to lie with those whose values had been shaped by low grid and low group – the rugged American individualists (for Wildavsky’s own views, see his 1991 and 1995 books). Understandably, Risk and Culture outraged many environmentalists and cast Wildavsky’s stranded co-author in the camp of the reactionaries.1 Forty years later, this book does not read well – that is, now that the evidence is much clearer that fossil-fuel industrialism is a major factor in climate change and potential ecosystem breakdown. The underlying argument also violated the principle that Douglas had laid down for her own theoretical programme – that ritual performance acting on social life generates a framework of

60

MARY DOUGLAS

beliefs and representations used to stabilise forms of organisation. In the Douglas-Wildavsky version of the argument presented in Risk and Culture, pre-formed ideological dispositions – for example, a sectarian ‘cosmology’ – appeared to be doing the organising. This was precisely the general-ideas-cause-particular-decisions (Y causes Y) formulation Douglas had been at pains to reject. Too much of Chapter 8 appeared to suggest that the ecowarriors draw on a pre-formed enclaved ideology to organise their memberships to attack the state. Maybe those attracted to campaigning careers working for agencies such as Greenpeace had found congenial employment suited to ideals inculcated in advanced education (as the book claims), but little or no evidence is offered that they had been socially ordered in enclaved social worlds. Obscuring the distinction between organisation and cognition, and identifying culture with ideology, led to shallow explanation and reinforced suspicions that grid and group were little more than a neat party trick, not a way of framing a serious causal explanation. The authors were also short on ethnographic observation. Douglas and Wildavsky gave some hint that they were aware of this deficit, when they acknowledged special help provided by three research assistants in documenting environmental groups (see especially Chapter 7, where some of the organisational differences between Friends of the Earth and the Sierra Club are described), but too often the X variable was specified only in rather general terms. The book assimilated individualistic and hierarchical ordering into a pooled ‘centre’ ground, cast enclaved ordering to the ‘periphery’ of American politics, and gave minimal recognition to isolate ordering. This was far from being the level playing field for examining competing notions of social or political order towards which Douglas’s earlier work had pointed. Risk and Culture was a setback to her own programme, and it took her some time to recover. Douglas needed a better account of technological risks. She also needed to restate her own understanding of her theory, free of the uses to which it had been put in Risk and Culture. The required remaking took place particularly in three subsequent books. The first two – a monograph, Risk Ac-

RISKS AND SOLIDARITIES

61

ceptability According to the Social Sciences (1985), and a collection of essays, Risk and Blame (1992) – set straight the account on risk. A third book, How Institutions Think (1986), then carefully laid out once more the basis of her entire theory. We will examine the first two titles here but defer discussion of her arguments in How Institutions Think to a separate chapter. Only a few years after Risk and Culture, Douglas published arguments that served to differentiate her own further thoughts on risk from those she had signed up to in the book with Wildavsky. Risk Acceptability According to the Social Sciences (Douglas 1985) warns readers that it was not a book about risk assessment or about risk management but a development of arguments earlier advanced in Purity and Danger that humans pay attention to disasters as omens or punishments, and that there is always mutual adaptation of views about natural dangers and views about how society works. Douglas regretted that reviewers found Risk and Culture either novel or outrageous, and always difficult. Without disrespecting her distinguished collaborator, she offers this further account as a way of easing some of the difficulties readers had encountered. She laments the lack of ‘sustained theorising about the social influences which select particular risks for public attention’, thus aligning this new book with Risk and Culture, which had a similar aim. Douglas then notes that risk assessment involves moral choice and that current theories of choice in social science are weak in interpreting ‘a strong moral response on the part of the public’ (p. 4). As she puts it, risks occupy different places in each of the four kinds of ‘moralised cosmos’ that her typological work had examined. The significance of this point is high. Risk Acceptability made clear – as Risk and Culture had not, because the argument was obscured by that earlier book’s focus on ideas and ideologies – that risk mattered crucially for her entire programme, not merely because risks are chances of misfortunes but because they are always conceived as anomalies in people’s schemes of classification. Risk is therefore the barometer for measuring the pressure under which a social ordering is operating. Douglas turns to the political philosopher John Rawls, who argued that the basis

62

MARY DOUGLAS

of justice is fairness (Rawls 1970, 1993), but makes her own characteristic argument that fairness means different things to different people according to the organisations within which they are embedded. She cites (p. 6) work by Selznick (1969) showing that to unskilled manual workers fairness meant equal treatment for all but that professional and management cadres thought that fairness involved recognition of individual ability. Differentiation from the political perspectives in Risk and Culture now became explicit. ‘Since the present distribution of risk reflects only the present distribution of power and status fundamental political questions are raised by the justice issue.’ Race enters into it: ‘In the US, nonwhites have higher expectations of being injured at work than whites’ (p. 10). Moral and ethical dilemmas raised by risk assessment are inescapable. She cites as an example a community that ‘once heroically accepted the risks of coal mining’ but ‘now suffer[s] the misery of unemployment’. ‘Who is to tell them not to have a liquid natural gas site in their neighbourhood if they perceive it as a source of income and pensions?’ (p. 11). The dilemma remains central to global environmental politics today. Why should poor people absorb the greatest costs, whether we are talking about climate change itself or the measures needed to mitigate it? Douglas’s concern with social justice (Chapter 7) remained central to her project and climaxed in a fierce denunciation of the World Bank and its adoption of new institutionalism as an approach to explaining poverty, tantamount (she thought) to blaming the poor for their own poverty. Typically, it resulted from a wrong understanding that culture was somehow a causal variable. Cultures of poverty, Douglas argued, were the result not the cause of material deprivation. They were attempts to make life work, after a fashion, despite deep and fundamental underlying inequities in wealth. For cultures of poverty to change, the underlying structures of wealth distribution would have to change first. On this she was as certain as any of the Marxist anthropologists with whom she debated in late-1960s London. It meant that many of the critics of Risk and Culture had almost entirely misunderstood her own project, and Risk Acceptability was the first step in righting the record.

RISKS AND SOLIDARITIES

63

In Risk Acceptability, Douglas presented her sustained critique for the first time, using 1980s’ ‘new institutionalist’ arguments that lurking behind the psychology of risk perception were methodological individualist assumptions.2 In the final stage of the argument, she extended that critique to the conception of bounded rationality as a single universal phenomenon. Risk and Blame (1992a) continued the process of clarifying her aims by bringing an analytical focus to the topic of blame. Blame is a missing term in risk analysis, perhaps for fear this is where the lawyers become involved. To attribute blame is, in effect, to accuse other people of both causal and moral responsibility for harms or for risks of future harms, and at least implicitly if not overtly to suggest the need for sanctions and reparations or corrections. Having become a synonym for danger, therefore, risk became inextricably linked with blame because anomalous misfortunes are tied by institutions to moralities of responsibility. As Douglas put it (1992a), it is a ‘forensic’ concept; having identified the chance of a harm, our inquiry first into causation and second into roles for amelioration take us directly into attributions of responsibility and therefore of blame. Thus, there can be no adequate social theory of risk unless processes of social apportionment of blame are understood. Moreover, blaming is performative. To level blame is to activate a ritual process of examination and involves accusation, demonstration of evidence, defence, evaluation, adjudication and perhaps even redemption. These range from the Lele poison ordeal to actions before the Supreme Court, not to mention trials in the court of public opinion. Every type of social ordering has its distinct repertoire of highly ritualised practices for the enactment of each of these phases of an extended process of blame, restitution and social restoration. Risk is part of the Durkheimian mechanism of social life in which accidents are invested with enduring meaning through being reinforced with ritual activity, as apparent from the sudden appearance of flowers and other tributes at fatal accident sites. One especially dramatic example is the London wall of remembrance for victims of Covid-19, a spontaneous manifestation of concern both for victims and with the political mismanagement

64

MARY DOUGLAS

of the Covid-19 pandemic. The British Prime Minister, not incapable of understanding its significance, felt unable to visit it except by night and alone. Here, surely, is an inescapable instance of Douglas’s argument about social action seeking outlets through ritualisation with the aim of institutionalising public concern. The promised public enquiry into British handling of the pandemic continued to be mired in prevarication and delay, but its day in court cannot be prevented for ever.3 Risk and Blame (Douglas 1992a) gathers several of Douglas’s essays relating not only to environmental threats and technological hazards but also to infectious disease (leprosy and HIV and AIDS), which then serve to reintegrate some of the themes brought into consideration in Purity and Danger, such as the social roots of ideas about contamination. Among the most impressive items is an essay on ‘Witchcraft and Leprosy: Two Strategies for Rejection’ (Douglas 1992b). This marks an important step forward in her thinking, because it now explicitly links diagnosis of witches and diagnosis of disease. In both cases, the damage is done by stealth. Appearances are deceptive. ‘The witch looks like anyone else.’ Likewise, in its early stages, leprosy is invisible, and the infection spreads unobserved. Noting that biological causes were first established by the Norwegian scientist Gerhard Hansen, Douglas goes on to point out that leprosy also represents a social challenge over how to treat those who have been marked out by physical manifestations of skin disease. Are these people to be quarantined, and if so, how are their basic needs to be provided for; are they forbidden to mix, or to work? This turns leprosy into a social problem. It is beside the point to ask whether or which manifestations of ancient disease were really Hansen’s Disease or some other kind of infection. The social challenges of what to do – how to decide who or what to blame, and whether to exclude or integrate – run independently of biological causes. This is something she demonstrates by pointing to the way the arrow of causality for leprosy shifts around according to shifts in organisational challenges and moral ideas about the probity of rulers and the changing nature of work. Citing research by the medi-

RISKS AND SOLIDARITIES

65

eval historian Mark Pegg, she concludes that the disease varied in the extent to which it was seen as a moral disqualification from society (Pegg 1990). In medieval Europe, leprosy was seen as a mark of divine disapproval of those at the top of the feudal tree. It was a disease of failing hierarchs and struck down incompetent kings and queens or corrupt senior clergy. In the early modern period, and because of economic change, with the emergence of labour markets, the diagnosis shifted and became associated with footloose workers and the unemployed. The degraded and diseased outcasts of labour market failure were housed in leprosaria (specialist accommodation for those diagnosed with the condition). In other words, application of disqualification rules shifts as feudal hierarchy is replaced by market economy. The problem of insidious harm is exposed as a social phenomenon. The leprosy of the crusader King of Jerusalem, Baldwin IV, who was not disqualified from rule by his physical condition, is then offered as a limiting case (different social organisation among distant rulers made for different rules). Infection risk cannot be assessed from biological data alone. It is necessary to attend not only to the behaviour of the disease but also to the social life of the people experiencing it. The essays in Risk and Blame round out Douglas’s theory of conflict amplification. The theory, however, to be symmetrical would also have to be able to explain conflict limitation or containment and even include an explanation of reconciliation. In How Institutions Think, Douglas worked out some of the basic cognitive elements of her approach to this, the other side of her theory of conflict, but a full examination of the ritual underpinnings had to wait until Douglas returned to the Hebrew Bible as a source of data from which to develop a more satisfactory account of how conflicts might be dampened through management of ritual processes.

NOTES 1. Richard Fardon has examined the manuscript drafts of the book and noted a degree of internal tension in the writing process (personal communication, 12 August 2022).

66

MARY DOUGLAS

2. Methodological individualism is the notion that the analyst should assume, as a starting point, that decisions are to be explained by reference to factors defined at the level of the individual natural person: these may be beliefs and desires or– as in Freud’s early theory – drives. Social relations are assumed only to enter explanations by way of their impact on individual level forces. This reduces all bonds and accountabilities to others – whether parents or acquaintances or communities or nations or identities – merely to constraints on particular decisions, and – as Durkheim argued – fails to recognise their enduring and structuring force and – as the next chapter discusses – their shaping of thought, preferences, cognition and classification which make up those particular decisions. But it is the underlying dogma from which much economics (and social science more generally) proceeds. It is ‘methodological’ to distinguish it from ‘substantive individualism’, which is the thesis that social life is predominantly organised around transactional, instrumental relations between people pursuing self-regarding goals and that appearances of, for example, hierarchy or enclave ordering are misleading, and also from ‘normative individualism’, which is the notion that we ought to stand on our own two feet and impose the fewest costs and burdens on anyone else. 3. It finally began, several months late, on 28 June 2022.

CHAPTER 5

INSTITUTIONS AND THOUGHT STYLES

••• B

y the early 1980s, Douglas’s model of social dynamics was now becoming more clearly outlined. The tumult of everyday social life, with its manifold ups and downs, its accidents and triumphs, is tamed by a giant accounting machine running on ritual. Douglas thought the key to its understanding had been provided by Marcel Mauss, in his hugely influential essay, The Gift (1990 [1950]). Durkheim had attempted a general theory of how immaterial social facts generate real-world consequences, but it was a theory without reliable method. How are social facts to be pinned down? According to Douglas, The Gift turned the Durkheimian project into a true social science. ‘The Gift was like an injunction to record the entire credit structure of a community’ (Douglas 1990: xiii). Mauss, she wrote, had ‘rendered an extraordinary service to Durkheim’s central project by producing a theory that could be validated by observation’ (Douglas 1990: xvi). To map cultural exchanges – ‘all dues, gifts, fines, inheritances and successions, tributes, fees, and payments’ – is to know all who benefit and ‘who gets left at the end of the day without honour and citizenship’. Some might want to label it an honourand-blame machine. A huge range of performative rituals normally assigned to the humanistic end of the anthropological spectrum could now be brought within the range of a science of society. An approach to mapping this terrain is explicit. ‘With such a chart in hand’, she

68

MARY DOUGLAS

writes, ‘the interpreter might be capable of sensing the meanings of ballads, calypsos, dirges and litanies; without it, one guess will do as any other’ (Douglas 1990: xvi). The honour-and-blame machine ground out social classifications through which a provisional social order might be maintained. But the ritual orderings were not all the same. Dissimilar orderings were cultivated in different times and places. How are these differences to be explained? The fourfold typology provided ideal types, and Thompson had supplied it with some theoretical modelling of feedback dynamics. But what was required was a clear account of the causal process by which social organisation, in any of the forms, shaped thinking and the use of classification and the categories. This was the task she set for herself in How Institutions Think. Strikingly, the book found no need to deploy the grid-group typology. There is a clear sense of a reset. Indeed, Douglas wondered if this was the book with which she ought to have begun her career. There is nothing to stop us taking it that way. It is short and, as always with Douglas, written with limpid clarity. But it requires steely concentration to absorb the implications of its far-reaching arguments (not to mention to accommodate its rather disconcerting tendency to dart off after side issues in some places). If today’s student reads only one book by Douglas, it might be this. The core questions addressed in How Institutions Think are these: what exactly is the work of institutions, and how is it done? Durkheim’s (1995 [1912]) approach was based on his understanding that the basic categories of all explanation – time, space and causality – ‘have a social origin’ (Douglas 1986: 12). His major examples included how aboriginal Australians used cult activity and animal symbolism to organise scattered and mobile communities of gatherers and hunters into groups of people that they could and could not marry. It is widely appreciated that Durkheim offered a powerful account of how cult ritual invested animal totems with organisational significance. More profoundly, though, Durkheim presented an account of how ritual life sustained the most elementary implicit categories by which people understood the ordering of their world and

INSTITUTIONS AND THOUGHT STYLES

69

their capacity to have an impact upon it and upon each other (Warfield Rawls 2005). She then turns to Ludwik Fleck, a Polish bacteriologist and historian of science who had written a major work in the 1930s entitled Genesis and Development of a Scientific Fact (1979 [1935]) to develop a theory in the sociology of knowledge to apply to understanding scientific discovery. In Genesis, Fleck analysed ways in which social pressure prioritised the search for a cure for syphilis over a cure for tuberculosis, even though the latter disease was the more deadly in its impact on society. Distinguishing ‘esoteric’ from ‘exoteric’ circles, Genesis argued that predominant styles of reasoning in science were driven by the social organisation of the ‘collective’ in which they were sustained. The impact of his work was lost in the chaos of the Second World War, but an English translation in the late 1970s gave the book a new lease of life because it was read as a precursor to arguments in those years by Kuhn (1970 [1962]), Feyerabend (1975) and especially Lakatos (1970, 1976) about styles of scientific inference in response to anomalies. Fleck built on the Durkheimian approach, but an advantage for Douglas was that Fleck was talking about science and not aboriginal Australian classification. She saw a chance to bridge the troublesome gaps between modern and primitive thought bedevilling Purity and Danger. The Polish bacteriologist also provided some helpful new concepts. Borrowing from Fleck, Douglas’s (1986) How Institutions Think replaced the problematic politicised implications of ‘ideology’ or the slightly obscure term ‘cosmology’ with the term ‘thought styles’. Although at one point Douglas equates thought styles with Durkheim’s ‘collective representations’, in fact the shift from Risk and Culture is more profound. For if ideologies and representations have to do with the content of what people believe then thought style captures important features of the ways in which people characterise their thoughts and modes of reasoning. This might include (for example) how far back or forward in time they benchmark their decision-making and which parts of the past are forgotten, integrated with or delinked from schemes of lay classification, what use people make of analogies, how rigidly or flexibly they deploy catego-

70

MARY DOUGLAS

ries, and above all how they reason and make inferences when confronted with anomalies (6 2011). The concept of thought style allowed Douglas to recognise ‘group mind’ as only one special case, for there are distinct thought styles that cultivate less commonality and also greater flexibility and openness, or even casual and opportunistic ways of reasoning about cases that also need to be explained. Thus, How Institutions Think shifted the focus on the ‘Y variable’, which her theory was designed to explain, from content of beliefs to manner of reasoning. So, what causes styles of thought to vary? Durkheim had attracted criticism for writing of society as if it were a single mind writ large. Better, Douglas insists, to put it the other way round – to think of the individual mind being ‘furnished as society writ small’ (p. 45). But because the concept of ‘society’ is problematic – there can be large disagreement about how it is to be bounded and the ways in which it could have agency – it is better to focus on institutions, which are the expression of social organisation or what in her (1980) book on Evans-Pritchard she had called social ‘accountabilities’. Douglas offers a characterisation of what she does and does not mean by the word ‘institution’ (Douglas 1986: 46). It is ‘a legitimised social grouping’ and can include, for instance, the family, a game or a ceremony. To describe an institution as a social grouping is perhaps unfortunate because it might be taken to exclude the weakly social integrated forms of isolate and individualistic ordering. It might be more consistent with the way that Douglas uses the term ‘institution’ to substitute for ‘social grouping’ the word ‘cluster’, or, more exactly, a specification of the forms of social ordering in any social context that govern how people are accountable in that context. Legitimising authorities might include persons such as a parent or a judge, or may be diffuse, such as assent to some general founding principle (born of this village, I have rights in its land). She excludes any ‘purely instrumental or provisional practical arrangement’. A decade after publishing How Institutions Think, Douglas offered what is perhaps a better definition of an institution for her purposes. ‘To institute is to establish order . . . An institution is

INSTITUTIONS AND THOUGHT STYLES

71

specifically an ordering of social relationships into regular patterns’ (Douglas 1995a). ‘Ordering’ encompasses loosely integrated clusters, tightly integrated groups in closely regulated and lightly regulated settings, and the ‘regularities’ of the ‘patterns’ can be measured by the strength or weakness of social regulation and social integration. Douglas argues that institutions are often explicit rules for behaviour, violation of which might lead to sanctions or costs; institutions of that narrow kind may affect what we decide to do. But permissions and prohibitions are only the most obvious kinds of institutions. There are also institutions of a less overt kind expressed in terms of (for example) kinship and collaboration over work. Anthropologists were especially familiar with the former, and historians of science had examined the latter, drawing upon the sociology of knowledge to understand how scientists respond to anomalies in relationships between theories and data. These deeper institutions encode the ordering of our social relations, at home or work. They implicitly and subtly shape the way in which we are able to think, well before we come to make particular decisions about what to do. Those who are institutionalised are not conscious of their institutionalisation. ‘Half of our task is to demonstrate this cognitive process [of naturalisation] at the foundation of the social order. The other half . . . is to demonstrate that the individual’s most elementary cognitive process depends on social institutions’ (Douglas 1986: 45). She starts her task in How Institutions Think with ‘the DurkheimFleck idea of a social group that generates its own view of the world, developing a thought style that sustains the pattern of interaction’ (p. 32). How precisely do institutions cultivate styles of thought? Douglas’s answer is that institutions work through education and induction into a group. For example, the individual scientist has absorbed the communicative conventions of her trade and can produce new and independent statements and findings couched in common code. Learning such schemes is part of a ritual ‘game’ through which the group is constituted. Deployment of a thought style is the mechanism through which the institutions cultivating that style are then reinforced.

72

MARY DOUGLAS

Why do we then hold on so stubbornly to what we think we know? How Institutions Think ‘attribute[s] the inability to be converted by reasoned argument to the hold that institutions have on our processes of classifying and recognising’ (p. 3). These processes in turn evoke public utterances that generate ‘bitter disagreement [and] selective deafness in which neither of the parties to a debate can hear what the other is saying’ (p. 3). The feedback loop is now complete: institutions cultivate the thought style of certainty (Douglas 2001 [1993]). We act on our certainties; in doing so, we reinforce our institutions and undermine institutions that conflict with our own. This is the first stage of a positive feedback dynamic. The book’s subsequent tasks include to understand how institutions come to be established, how they work, and what they allow us to do and prevent us from doing. The argument begins slowly. The first chapters engage in quite a lot of positional play – regarding Durkheim, Fleck, the philosophy of science, scale issues, and Mancur Olson’s approach to coordination problems and public goods. In particular, Olson’s work was important to her argument because he addresses the issue of how collective action can be geared towards the production of public goods (Olson 1965). Public goods are available, like fresh air, to everyone. The creation of public goods tends to involve large-scale projects for their production. Olson is concerned with the coordination of collective action when free riding is built into the activity. He exempts small-scale groups from these problems because smallness of social scale is supposed to engender public trust. From her work on Lele villages, riven by fear of betrayal and witchcraft activity, Douglas knows that this is not so. Olson’s arguments must be applied to African villages as well as to concerns with public goods in large-scale industrial societies. This has the unanticipated advantage of placing Durkheim and the Durkheim-Fleck programme back into the social science mainstream. Durkheim had been cast to the margins because too many readers supposed that he dealt in the small-scale and archaic societies only of interest to anthropologists.

INSTITUTIONS AND THOUGHT STYLES

73

Towards the end of Chapter 2, she arrives at a definition of her main task; ‘In Durkheim’s work the whole system of knowledge is seen to be a collective good that the community is jointly constructing. This is the process we need particularly to focus upon in the subsequent chapters’ (p. 29). Chapter 3 (‘How Latent Groups Survive’) and Chapter 4 (‘Institutions are Founded on Analogy’) then dive deeply into institutional mechanisms. Chapter 3 provides Douglas’s account of institutional development. The argument begins by grappling with the problem of functional explanation. Functional explanation was often associated in social sciences such as social anthropology with structural functionalism, which was less a coherent theory than an article of belief asserting that anything with any sort of durability in social life must serve a purpose. However, much structural functionalist theorising came to be associated with the inference that if something served a purpose that purpose would meet collective ‘needs’, would be socially stabilising, and would reduce conflict – that is, it is assumed the purpose must be a benign, conflict-mitigating one. In critiquing the design of functional explanations in structural functionalism, the philosopher Jon Elster (1983) had offered some useful rules for a coherent functional explanation in social science. The standard that Elster proposed was that X can be explained by its function Y, for a group Z, if and only if Y is an effect of X, Y is beneficial to Z, Y is unintended by actions producing X, the causal relation between X and Y is unrecognised by actors in group Z, and if Y maintains X by a causal feedback loop passing through Z (Douglas 1986: 33). ‘Unintended’ and ‘unrecognisable’ are key terms. Without these two qualifications, the group is making an arrangement or a contract. Everyone sees how it works and can sign up or not. But this is not the case with institutions. Many institutions emerge without being planned; even planned ones soon evolved. Y may be recognised as beneficial, but that was an unintended consequence. We become wedded to our institutions because they have the capacity to fool us. Douglas gives as an example a fully formed functional argument at work defending institutional judgment drawn from a list

74

MARY DOUGLAS

of examples assembled by Robert Merton. Merton (1968 [1949]) describes the case of a community that ‘holds the purse strings of educational funds and believes in the mental inferiority of [African Americans].1 Their belief justifies them in withholding schooling from these families, and they are naively delighted when the scholarships won by their own children confirm their belief, justify their allocations, and maintain their control’ (p. 42). It is a self-fulfilling prophecy, and a clear example of ‘how institutions think’. The self-fulfilling prophecy is a major obstacle to the rational examination of the relative merits of institutions. Other institutions are self-disorganising (e.g. Hedström 1998), and Douglas would later explore self-undermining processes in highly enclaved sectarian settings (Douglas and Mars 2003). Indeed, her argument leads logically to the conclusion that each of the four ways of organising can be become disorganising when ritual action reinforces them too far. Next, Douglas applies Elster’s template to show how an institution is built up. Without saying so, she is giving the example of an enclaved institution (bottom right in the grid-group chart). Enclaves were on her mind from the attention she and Wildavsky had paid to them in Risk and Culture, and now she was trying to add what had been missing in that book – namely, a mechanism. She starts with a group drawn from her Lele ethnography. Consider a village dependent on collective effort in farming, where any individual can threaten to leave (X). The result (Y) is weak (noncoercive) leadership, which is useful for the group (Z) because thereby individuals minimise unwelcome demands on their time and resources. Weak leadership (Y) is an unintended consequence of X (the threat to leave); indeed, villagers may regularly deplore that the chief is weak. But there is a hidden causal loop – people stay in the village because weak leadership means less coercion; they are relatively free. This is a fully formed functional explanation according to Elster’s rules. But it is not functionalist, because Douglas uses it to explain contention, conflict and even disorganisation. Let us suppose that the villagers practise (X) equality and participation in use of labour and access to harvests and come to

INSTITUTIONS AND THOUGHT STYLES

75

expect each other to follow this way of conducting social relations. This established expectation will mean that people need to know who is in and who is out of the community, thus creating (unintentionally) a boundary (Y) around the group (Z). This boundary consolidates membership and prevents free riders, so is an advantage to the group, but the causal loop is unperceived and unintentional, since individuals see only the need to prevent themselves from being suckered by others. It has the ‘self-policing effect of a convention’ (p. 39). Douglas then considers the effects of cheating and betrayal (X) resulting in a shared belief in an evil conspiracy (Y). This belief is protective of the group (Z), is unintended, and depends on a causal connection not perceived by the members (that the real danger is wavering commitment not the supposed conspiracy). Action taken in pursuit of that belief will reinforce the boundary, thus completing the causal loop. We now have three kinds of processes of institution formation that fit together into a bundle – weak leadership, strong boundaries and conspiracy mindedness. ‘This particular type of social group thinks along certain grooves . . . In choosing to join this idealistic band . . . no one opts for the whole package of behaviour and beliefs. But they go together’ (p. 40). They are combined as follows: Y (belief in conspiracy) is an effect of X (weak leadership and strong boundary). Y is beneficial in keeping the community Z, together, but is unintended by the group, so it cannot be charged with duplicity. Causal links are not perceived. Y maintains X by splitting the community, or expelling when treachery is suspected, thus producing a history to make every would-be leader nervous. (p. 41) Note, though, that the process begins with practices and ways of conducting social relations. It does not begin with any ideology or doctrine or abstract justification for equal participation in harvesting, or any grand abstract scheme of political thought about equality; such things, if they arise at all, will arise later, as a result of causal loops, and typically only when people are pressed

76

MARY DOUGLAS

by other accountabilities to produce justifications of this kind for their ways of life. Moreover, Douglas’s functional explanation, which meets Elster’s standards, is not functionalist. For example, fears of conspiracies and weak leadership and strongly defended boundaries might well lead to major problems for the group and over the longer term could threaten its continued existence. Far from meeting their ‘needs’, over time their institutions and thought styles could prove self-undermining. The strongly defended boundary could also spark tensions with other people, and frustrations among would-be leaders and among those accused of conspiracies could give rise to other kinds of conflicts. Rather than being a stabilising process, the repeated reiteration of the functional loop over time can be used to explain conflict and disorganisation. As we shall see, this is exactly what Douglas found herself having to do with it when she returned to the Lele in the 1980s and observed the effects of social conflict and breakdown. Douglas has now described the process through which an enclaved institution with its own distinctive thought style has been formed. The reader is left to figure out how to apply the argument to Douglas’s other forms of ordering. Her discussion of Olson directs us to think about kinds of institutions ordered by competitive individualism, and familiar from work by economists. Collective action problems are characterised by market functionality. Individuals can maximise their advantage by bargaining (X), and prices regulate supply and demand as if by a hidden hand (in Adam Smith’s phrase). Regulating supply and demand is an unintended consequence (Y), and unrecognised by actors (Z), who are intent only on meeting their own needs. The argument in relation to hierarchy might run on the following lines. Hierarchies form around up-and-down movement of information and control. This is an X. There is a shared belief in the greater importance attached to the higher echelons (Y). An example is that of a person recently interviewed by a journalist about the prospect of yet another British Prime Minister coming from a single private school, Eton College, whose reply was ‘I have served in the army, and I do not expect officers to come

INSTITUTIONS AND THOUGHT STYLES

77

from state schools’. Seemingly, this kind of attitude (let us label the Y value ‘elitism’ in this case) reflects respect for a socially distant officer cadre but protects the functionality of a group organised around command-and-control (Z) while also being causally unintentional and unperceived. Douglas wrote with great insight on the ways in which this kind of institutional functional ordering was cultivated in her Catholic boarding school (Douglas 2005). Isolate ordering presents a different challenge. Here the X comprise sets of practices intended to foster as complete autonomy as possible because aggregation in groups threatens danger in extremely hostile environments – such as among maroons on the edge of societies ordered by slavery. Y values will then probably be expressed in terms of intense suspicion and hostility. Unexpectedly, this extreme magnetic repulsion creates a kind of society, notwithstanding the intention to be alone. For example, regions of maroon occupation are sometimes known for initiation practices fostering extreme self-reliance, endurance and secrecy (Z), thus shaping an unintended and unperceived skeletal social formation populated with initiates with similar feelings and instincts (Richards 2022). There are even cases of isolate ordering among people in high office, where isolate despots make decisions alone; this cultivates a distinct thought style, which can lead to unintended consequences (Coyle 1994; 6 2011, 2014a,b). With the process of formation of institutional mechanisms described, in outline at least, for any social unit regardless of scale, Douglas then moves on to describe the process through which these tentative bundles of institutions, each with its distinctive thought style, can gain greater stability.2 Stability often arises from legitimacy. To acquire legitimacy – Douglas suggests – every institution needs a formula that claims to found its rightness in reason and nature. This involves the naturalisation of social classification. ‘There needs to be an analogy by which the formal structure of a crucial set of social relations is found in the physical world, or in the supernatural world, or in eternity, anywhere, so long as it is not seen as a socially contrived arrangement’ (p. 48). To be seen as socially contrived would blow the game. We would be contin-

78

MARY DOUGLAS

uously chafed by our own ingenuity; our organisational arrangements would no longer seem natural and effortlessly conducive.3 As Douglas’s argument in How Institutions Think advances, some ethnographic examples follow – the way, for example, that left- or right-handedness can map on to a gender-based division of labour, conveying a sense of unequal complementarity between wife and husband as between left and right hand (85 per cent of humans are naturally right-handed, so the analogy of biased handedness is readily available). ‘So, the equation “female is to male as left hand is to right” reinforces the social principle with a physical analogy’ (p. 49), and a hierarchical ranking is classified. From such means are potentially wayward humans locked into an institutional structure through analogy with the body. Materials for forging analogy abound. In Africa, a continent of long-distance historical folk wanderings and abundant bush, current rural populations take note of their origins (a proverb often cited is ‘if you do not know where you are going you had better know where you are coming from’) and build institutions by seeking the blessing of the ancestors on land they cleared and settled. It is the ancestors, as Douglas notes, who validate access to land, in communities where formal titling of property is rare or absent. The ancestors are on hand to prevent claim-jumping. Quite elaborate institutions can be built and naturalised through various processes of claiming kinship with ancestors. To be born into one of these networks is often enough to validate a strong claim to local citizenship. But Douglas is good enough an African ethnographer to know that claims to kinship only become effective in some cases. The ancestral metaphor is widely available but not used in the same way in all places (the Lele belonged to a widely scattered minority of African matrilineal societies, where they do things differently; land comes from the uncle not the father, for example). In some cases, the available kinship model is not used at all. Social attachment is more through client status resembling feudal vassalage (parts of northern Nigeria, for example). So why the differences? Are they simply accidents of history? The potential analytical pitfall, Douglas remarks, is ‘to isolate the ancestral cult from the entire social structure’. When the system is in good working order ‘the configuration of ancestors projects

INSTITUTIONS AND THOUGHT STYLES

79

the social structure’. But ‘[t]o say that it is a good metaphor of society does not explain why some metaphors work catalytically to promote collective action, and others do not’ (p. 50). The ancestral model only becomes an effective institution when ‘articulation of ancestors’ actions articulates the social process’ (p. 50). An instance is when the ancestors are appealed to in adjudicating land claims. The ancestors then become ‘the guardians of property rights’. But the system can crash. What happens when everyone has ancestral backing for their claims? These claims then cancel out. Something further is needed. Maybe what I then recognise is mutuality – that my ancestor protects my space, while I respect that your ancestor protects yours. This buttresses the institutional system of ordering as a natural system grounded in the world established by the ancestors. Fights can be contained, out of respect to those who came before us. ‘Many more such analogies that confer natural status on social relations abound in the anthropological literature’ (p. 52). Douglas then makes a bold leap (something we have already seen she is rather good at). She cites a study by the psychologist De Soto (1960) in which individuals are reported to recognise in their social situation formal (mathematical) relations of complementarity, exclusion and inclusion, and transitivity. ‘By using formal analogies that entrench an abstract structure of social conventions in an abstract structure imposed upon nature, institutions grow past the initial difficulties of collective action’ (p. 53). The significance of De Soto’s finding for Douglas’s argument is that it suggests that is in the formal, stylistic features of thought, more than in normative ideological beliefs, that social relations imprint themselves upon cognition. Here, then, the form of the mechanism for ordering the wild profusion of social life is at last fully revealed. Institutions of social organisation cultivate capacities for fixing patterns using analogical reasoning for legitimation, which in turn depends on a capacity for recognising degrees of similarity. And what is the source of this capacity for recognising similarity? ‘Sameness is conferred on the mixed bundle of items that count as members of a category; their sameness is . . . fixed by institutions’. It is the institution, not nature, doing the shaping.

80

MARY DOUGLAS

Recasting Durkheim and Mauss’s (1963 [1902–3]: 11) central argument, Douglas shows that we reproduce features of our social organisation not so much in the content of our beliefs but in the formal and stylistic features of our thought; our thought styles are as strongly or weakly cognitively integrated and regulated as our social ordering is strongly or weakly socially integrated (by group) and regulated (by grid). Those two dimensions are explicitly absent from How Institutions Think because they are not necessary for persuading social scientists of the central causal mechanism. But they are implicit in Douglas’s use of an enclaved example to set out that mechanism, and equally in her account of strongly and weakly integrated and regulated styles of linking (or not linking) concepts and analogies, and in using categories rigidly or flexibly, etc. In short, the grid-group dimensions have a ghostly centrality to the book’s argument, which integrates it tightly into her oeuvre as a whole. The remainder of the book focuses on other aspects of thought style, beyond analogy and how it is used. Chapter 6 explores how institutions remember and forget, as a key example of styles of thought with reference to long and foreshortened past in memory and future in planning. Other chapters examine in greater detail how institutions cultivate different styles of deployment of categories in classification. There is more to be said about classification in our next chapter. Douglas concludes How Institutions Think with a benediction. ‘The Durkheim-Fleck program points to a way of return. For better or for worse, individuals really do share their thoughts and they do to some extent harmonise their preferences, and they have no other way to make the big decisions except within the scope of the institutions they build’ (p. 128).

NOTES 1. We have modernised Merton’s terminology. 2. This is after a digression on institutions and information. Douglas wishes to acknowledge the economist Andrew Schotter (1981), who had reworked in information-theoretic terms Oliver Williamson’s ar-

INSTITUTIONS AND THOUGHT STYLES

81

guments about institutions and transaction costs. She found this compatible with her own views in some respects, though she doubts that institutions as readily achieve equilibrium as economists often suppose. Although Schotter restricted his model to information about what people think rather than styles of how people think, his conception of information encoded in institutions comes close to Douglas’s arguments concerning the ways in which institutions cultivate cognition. 3. Douglas’s requirement here is probably stronger than it ought to be for the purposes of her theory, the point of which is to handle the full range of variation in forms of social ordering. In highly individualistically ordered settings of trading, people are often aware of how contrived pricing arrangements are, even though this background awareness fades into the background when they are in the thick of work and decision-making guided by those institutions; Dinler’s (2019) anthropological study of waste pickers and merchants’ awareness of how different prices emerge and are managed in recycling in Turkey provides an example.

CHAPTER 6

RITUAL AND CATEGORIES

••• D

ouglas was a gifted writer, and that gift is often particularly well displayed in her essays. If asked to name a single essay that comes closest to capturing the essence of her thought, a good candidate might be ‘Rightness of Categories’ (Douglas 1999d [1992c]). This appeared as a chapter in a book she edited with David Hull, a philosopher of science, who worked on biological taxonomy.1 Published in 1992, How Classification Works was offered as a tribute to the philosopher Nelson Goodman, whose (1978) ideas on ‘world-making’ had attracted Douglas’s interest and admiration. Hull and Douglas were for a time colleagues at Northwestern University in suburban Chicago. Her essay in honour of Goodman was also later added to the second edition of her book of essays Implicit Meanings (Douglas 1999d). ‘Rightness of Categories’ is a succinct account of Douglas’s approach to culture. Culture was always the product of the ways humans come together in groups and order their affairs. But we tend to put the matter the other way round and explain actions and organisation in terms of cultural causes. Douglas does not agree. Culture is a dependent variable, even if she clearly recognises feedback loops, so that once institutions have cultivated a style of thought people decide and act in ways influenced by that style of thought, and so reinforce their institutional ordering (by positive feedback) or conversely undermine other forms of social ordering (by negative feedback). In the essays in Risk and Blame, such as the one on witchcraft and leprosy discussed in Chapter 4 above, she examined these dynamics carefully. A

RITUAL AND CATEGORIES

83

study with Gerald Mars on self-radicalisation dynamics within enclaved groups (Douglas and Mars 2003) then shows that positive feedback dynamics could lead to disorganisation, and even self-disorganisation. But ‘culture’ – or thought style – has no existence independent of processes of social ordering that would allow it to be credited with being an independent causal variable. We appeal to culture as the reason we do things, as if a law from God. But it is our social ordering talking. A cultural norm is not independent of the ways in which we accept to be organised; it is an explanation of who we are, or an attempt at self-justification. Fardon (1999) suggests that Douglas saw cultural arguments as working in two ways – as classification and as contention. We use culture-as-classification to sort ourselves out into groups – whether this person is a citizen or a foreigner, whether these people are refugees or economic migrants, and so forth. Cultureas-contention becomes apparent when, for example, we demand social accountability or apportion blame. People may insist that ‘we expect citizens to do their duty’ or confidently assert (plentiful contrary evidence notwithstanding) that ‘economic migrants cause crime to increase’. What then is of interest is how these classifications and contentious statements arise. Douglas’s central point is that there is no contention without classifying what we disagree about, and no classifying anything without provoking people who are ordered differently into challenging that classification; both aspects of the process generate anomalies and lead us to respond to these anomalies accordingly. The philosopher Immanuel Kant (1933 [1787]) had suggested that to make sense to each other humans require shared assumptions about the world. Kant labels these as a priori intuitions. But where do they come from? They cannot be simply given in our cognitive hard-wiring, or else we would all agree about the way the world is. Manifestly we do not. There must be some way in which groups are able to decide about rightness of categories. Durkheim argued that basic general categories – space, time and relatedness – are sustained through social processes. The present tense is important in this statement. It is insufficient to answer that a sense of society is somehow historically prior to other

84

MARY DOUGLAS

categories (that in some way the world has been ‘socially constructed’). This is because relatedness is a category as well, and we are doing no more than asserting relatedness as a category to trump all others. A better answer, Douglas realised, is that through the elaboration of daily social life humans constantly make and remake space, time and relatedness through the ways in which we invoke the past and future, distance and proximity, connection, and so forth, both in our interactions with each other and in reacting against others. This thesis that our social life stylises how we think about (for example) time has been supported empirically (Rayner 1982; Peck and 6 2006: 50–77). The tools of this process are classification and contention. What we make and remake daily then varies according to the scale and density of human interaction. Time may be made to appear endless, short or even circular. Space can be entirely locally constrained or made to appear limitlessly vast, even though we actually live (as Kant – an enthusiast for geography – was at pains to point out) on the surface of a sphere and our supposedly limitless travels only take us round in circles (Richards 1974). What is of equal importance is to understand that our tools of world-making have some inherent constraints. No scheme of classification is perfect. There are always exceptions that test the scope or the interpretation of a rule. How anomalies are addressed feeds cultures of contention. It is where real-time adjustments have to be made to cope with classificatory imperfection. Options include careful rule-based adjustment to define new subcategories as special cases, debarring and rejecting anomalies, exploiting them as opportunities for arbitrage, or simply accepting them and coping as best we can. Here we arrive at Douglas’s central concerns with the everlastingly messy realities of everyday life, our desire to tidy up without our institutional coping routines being overwhelmed, and our constant need for both material intervention and legal and ritual creativity to transcend the limitations and contradictions of inherited schemes of classification. The first half of ‘Rightness of Categories’ becomes an exposition of Goodman’s (1978) work, since she wishes to show us how

RITUAL AND CATEGORIES

85

his argument aligns with her own. How do we classify? Classification is a kind of abstraction from experience. But what kinds of archetypes do we draw upon to structure experience? The archetypes are found everywhere – up-down, front-back, linear ordering, core-periphery, part and whole – but a problem is that these are not used in the same ways in all places. If they are given in intuition (whatever Kant meant by that) then intuition works differently in different places. Douglas’s answer to this conundrum draws on Durkheim and Mauss’s (1963 [1902–3]) essay on classification – that the classification of things follows the classification of people. Social relations and performance ground the categories. Rodney Needham (1963) in his introduction to Durkheim and Mauss (1963 [1902–3]) had ticked off Durkheim and Mauss for arbitrarily prioritising ‘society’. In ‘Rightness of Categories’ Douglas works out an answer to Needham’s criticism. By convention, for example, on British railway lines leading to London, the up train is the one heading for the capital. That would be confusing for people brought up in Sierra Leone. A railway line pulled the country together in the colonial period, and even though the railway is long gone people heading out of the capital (Freetown) towards the provinces will invariably describe themselves as ‘traveling upline’. Comparing travellers in Britain and Sierra Leone, we have a single archetype (up-down) but two opposed meanings. There is nothing either in railways or the lie of the land to determine whether ‘up’ is ‘down’ or vice versa. So how does each categorisation acquire its locally specific meaning? The answer is through repeated usage. In both cases, generations of passengers learnt the local convention in order to catch the right train. Incorrigibles ended up in the wrong place. Still aligning herself with Goodman, Douglas digs further, looking at how thinking is shaped by samples, and exemplification more generally. A sample is not a member of a class. It is merely an illustration that can serve to exemplify rules or options. A sample of a piece of cloth simply tells you what that cloth is like, not what kinds and colours of cloth might be available. Her advice is to consider these rules not in terms of their formal character but on the practices that shape them. ‘Isomorphic im-

86

MARY DOUGLAS

ages are ten a penny’, she writes, but ‘what significant properties they share depends on everyday practical organization and the assumptions and prejudices that go with it’ (Douglas 1999d [1992c]: 287).2 Remaining with Goodman, she then begins to explore the ways through which ‘everyday practical organization’ is unfolded. ‘To exemplify’, she notes, ‘is to display’. Display is public enactment. It is through public endorsement of such display that social norms are confirmed. Humans have a gift to sort out elements of display. The gym instructor orders the class to ‘copy what I do’. She bends her knees, and the class does likewise, but the class does not shout out ‘copy what I do’. This implies a grasp of what Michael Polanyi called ‘tacit knowledge’. Human groups are good at tacit understanding – picking up unspoken clues about what is going on. Douglas comments that ‘anthropologists are perforce very interested in tacit understandings, foregrounding of some meanings, backgrounding of others’ (Douglas 1999d [1992c]: 290). This talent for the improvised and collective editing of a social performance is the essence of what Goodman meant by ‘world-making’. This may seem a simple enough point. But there is a great deal at stake in Douglas’s use of it. Central to Durkheim’s (1995 [1912]) programme was to argue that rituals do more than help us to bond (although many sociologists still stress this function over all others: e.g. Sennett 2013). In ritual, Durkheim argued, we fix categories. Showing how this works is a key task for Douglas. In arguing in ‘Rightness of Categories’ that using exemplars is a ritual performance, she implicitly – for nowhere in the article does Douglas mention his name – develops an answer to the philosopher Willard Quine’s (1960) ‘indeterminacy of reference’ thesis. Quine had famously argued that one cannot learn the meaning of a word through ostension – that is, by pointing to an example. When the informant tries to teach the anthropologist the word ‘gavagai’ by pointing to a fleeing rabbit,3 Quine asks, how is the anthropologist to know whether what is meant is ‘rabbit’, ‘running’, ‘grey’, ‘white-tailed’ or ‘lunch’? Asking whether future

RITUAL AND CATEGORIES

87

cases of uses of the word are the same as this one only goes so far, and understanding the answers depends on having a prior translation of grammar and a theory of the people’s categories for relevant ‘sameness’. One cannot learn a language piecemeal from material connections, Quine concluded; one had to learn it wholesale from a grand scheme of hypotheses about the whole system of words (‘confirmation holism’) or not learn it at all. Douglas thought that this was quite wrong – about how children learn, about how anthropologists learn, and about how people use social encounters to convey ideas. Her point is that in a specific institutional context such as being an informant with an anthropologist, or a teacher with a child, there are clear roles and implied social contexts constraining what is done; ostension is a ritual act and disambiguates by being performed as a part of a system of ritual interactions. Douglas notes that in these kinds of contexts rich in ritual roles, ostension can be used to disambiguate using exemplars, but the disambiguation rests on the ritual process, just as it does in the exercise class in the gym. The consequences of this argument are very large indeed for Douglas’s programme. It provides her with a social and causal mechanism by which Goodman’s world-making can get done. Goodman (1978: x) described his argument as ‘a radical relativism under rigorous restraints’. In ‘Rightness of Categories’, Douglas finally supplied a key constraint herself (the limited plurality of elementary forms of social organisation in her fourfold typology), which the philosopher focused only on language and ideas could not. This limited plurality constrains what can be done ritually, in an institutionally ordered social context, to disambiguate and therefore to fix categories. This answers Quine (1960) by providing the social, organisational and material4 basis for the rightness of categories and for the way in which rightness is – or is not – protected against anomalies in different thought styles. The argument that Douglas makes here is designed to fill out the causal mechanism for Durkheim’s central programmatic claim in his (1995 [1912]) Elementary Forms of Religious Life,

88

MARY DOUGLAS

that categories are fixed through ritual action. Goodman then causes Douglas to reflect further on examples. An exemplar refers directly to itself and is not necessarily expressed verbally (a model exemplifies fashion without uttering a word). A sequence of exemplars (a parade of fashion models on the catwalk, perhaps) is a preliminary classificatory process, serving to flag up the distinctiveness of a brand. Recurrent social displays are this way stamped with a recognisable patterning. Without this step no similarity can be recognised. This leads to the important question of how anything gets ‘selected and shaped to serve as an exemplar of a class’. This involves a process of denoting (drawing attention to people, things or activities) through which correspondences between bits of the world and signs that refer to them are established. This process, she claims, is ‘artificial, but not arbitrary’. In other words, it is configured around the type of activity being attempted, or around the organisational challenge to be resolved; again, Douglas argues, the limited plurality of ways of organising social contexts and the bodily limits of performing in those contexts provide the ‘rigorous restraints’ on diversity in world-making. Branding, in the world of fashion, can continue to serve as our example. No one would imagine that brands exist out there ‘in nature’. They are crafted for the purpose of mobilising desire and gaining adherents. A brand is a product of a classificatory process through which fashion producers seek to organise markets. But denoting a brand is essentially the same process as involved in ‘world-making’ more generally. The raw material of the process of denotation is communication, but the ethnographer can step back from the noisy forum in which ‘bits of the world and the signs that refer to them’ are brought together, to assess the more general process of world-making, from which panoptic point of view the importance of repetition becomes clear. ‘Members of a class often have to be enumerated, the list reviewed, the criteria for membership re-assessed . . . this [recurrent work of indication] has to happen for similarities to be established’ (Douglas 1999d [1992c]: 291). Branding is another name for ‘the recurrent work of indication’.

RITUAL AND CATEGORIES

89

It is at this point that notation becomes an issue. The teacher’s class register is a simple example. Are all the children present, and correct? Has one absconded or wandered in from the street? In which case, steps need to be taken. In art and science, notational schemes serve a similar purpose of keeping tabs on complex processes with many moving parts. Musical scores and mathematics are mentioned as examples. The purpose of these notation schemes is to generate frameworks for comparison – to answer with greater confidence whether any specific item belongs to a particular class (for example, whether a wrong note has been played). Fine art and music are further compared. An exact replica of a Rembrandt painting would be treated as a valueless fake, whereas an accurate performance of a musical score brings the music alive.5 Tracing provenance to eliminate the possibility of fakery allows the uniqueness of a great work of art to speak to the individuality of the owner. Typically, the item is enjoyed in contemplation. Realising a musical score may require huge amounts of skill and coordination, and success is registered in the moment, through dancing in the aisles, or the enthusiastic applause of a delighted audience. In short, beneath differences in schemes of denotation, we should expect to find differences of social projects. In science, Goodman finds two kinds of distinct social process at work – autographic and allographic (Douglas 1999d [1992c]: 296). The first involves justification ‘with reference to a verified history of production’. Success is crowned when the name of the original author is attached to the method or process. The second (allographic) process ‘corresponds to work done in expectation of repeated anonymous performances of teams of teachers and researchers’ (Douglas 1999d [1992c]: 296). These two streams are consolidated by adoption of different schemes of exemplification and denotation. Douglas is attracted by this insight because it can be applied across a range of fields in both the arts and the sciences – ‘a notable advance’ – but also because it clearly speaks to her concern to develop a general account of the ways in which social agency imprints itself on ideas about nature, without re-

90

MARY DOUGLAS

verting to some of the awkward ‘primitive-modern’ dialectics of Purity and Danger. The village healer, who paid dearly to be initiated into knowledge of an obscure cure and wishes the village to know her patient is healed, is every bit as autographic as Nobel prize committees in science seeking to draw attention to a breakthrough discovery with a verified history. Scientists mobilising huge teams to tackle urgent common problems (Covid-19, for example) and making this knowledge available as ‘open source’ are no less tapping into allographic social processes than the village elders advocating collective sacrifices as a way to protect against Ebola virus disease (Kamara, Mokuwa and Richards 2022). Above all, Douglas now has a theory of change. She finds herself ‘at the centre of a whole flurry of reciprocal monitoring’ (Douglas 1999d [1992c]: 298). ‘The results of repeated checking may require the classification to be adjusted.’ ‘Authentication has a touchstone – the world that is being made.’ Adjustment, though, is only one strategy for managing anomalies in world-making. As Douglas had shown in Natural Symbols, adjustment is the approach cultivated in hierarchical ordering. Where hierarchy adjusts for anomalies, enclaves debar them, in individualistic ordering people exploit them, and in isolate life people may accept or cope with them; mixes or hybrids are also possible. Douglas’s point is that each of these strategies produces change, either through reinforcement or pushing back against results people achieve using the other strategies. The structural functionalism of her mentors and teachers in anthropology has been left far behind. In offering a thorough exposition and consideration of Goodman’s arguments in the first part of ‘Rightness of Categories’, Douglas stresses their alignment with her own approach, and thereby provides her Durkheimian theory with a more precisely specified mechanism: an explanation of how public processes of exemplification and denotation support schemes of classification generated within changing communities. In the second part of the paper, she undertakes a reconsideration of her Lele ethnography, using these insights. The circumstances need to be mentioned. A grant from the Spencer Foundation allowed her to make a short

RITUAL AND CATEGORIES

91

return visit to the Lele in 1986. Now she had to take account of thirty-five years of far-reaching and at times violent social change since her first ethnographic work, and this challenged her ideas about what the anthropologist needed to explain. The local system of cults based on food prohibitions was undermined when the missionaries told Lele converts they could eat anything, but had she been right about these cults in the first place? She decided major revision was needed. In her earlier accounts including Purity and Danger, she had explained food taboos in what she thought of as metaphorical terms. But faced with trenchant criticism from anthropologists influenced by the structuralism of Levi-Strauss and others, together with her now renewed field interaction with the Lele, she decided her earlier metaphorical account of food prohibitions was too loose-limbed. What she needed was a comprehensive account of the entire system of Lele food taboos. Goodman had taught her that notation systems prepare exemplars, reflecting social concerns, but how is the system learnt? Here, she turned to another philosopher of science, Thomas Kuhn.6 Engaging with debates in the history and philosophy of science was neither a digression nor a self-indulgence for Douglas. These debates had made the concept of anomalies central to all sociology and anthropology of knowledge. When Kuhn, Lakatos and others rejected Popper’s initial (1959 [1935]) view that a single anomaly would suffice to sink a theory, they began to explore the range of ways in which scientists handle anomalies between theory and data.7 Indeed, in his essay in Douglas’s 1982 edited collection, Essays in the Sociology of Perception, Bloor (1982) had shown that Lakatos’s (1976) account of anomaly management exactly mapped onto Douglas’s typology of strategies. Goodman’s work was only indirectly and in part a response to these debates, but it offered her a way to develop an argument that might show that her approach could stand up in the history of sciences debates in exact parallel with the debates about classification being conducted in anthropology. Kuhn (1974) was concerned with the organisation of science and how as a system of classification science is learnt. The teach-

92

MARY DOUGLAS

ing of science is a public process organised around exemplars. Students are tested on how well they understand sets of standard examples, and since an exemplar will contain ‘richer and more varied material than whatever it comes conventionally to exemplify’ (Douglas 1999d [1992c]: 298), it will also provide the mind with resources to reflect on new problems and potentials for discovery. Students are also tested on how well they can handle notational systems through which examples are conveyed, by undertaking, for example, compulsory courses in mathematics and statistics. This process is two-way. The scientific community moderates the process of teaching. Teachers are expected to offer good questions and are open to sanction if these do not meet collective standards. The ‘rightness of questions’ is subject to scrutiny and debate among the teachers – are the range and level sufficient in the light of the latest developments of the field? This process involves moderation by external examiners, especially at the highest levels (for example, in the evaluation of PhDs). Extending her argument in answer to Quine, Douglas argued that the process of using exemplars to teach and test learning from teaching is Durkheimian. Durkheim taught that social norms are reinforced by ritual. Rituals involve a focusing of the emotions via which social norms are invested with sacred force. A university examination room or the ‘aula’ for a public defence of a PhD is among the settings in which science engenders such emotion. ‘Classrooms and exams are kinds of public performances in which correct repetition is required . . . exemplars are learnt during mutual testing . . . and the emotional tension is why they are so well learnt’ (p. 299). The conclusion is that teaching and examining exemplars and denotative systems illuminate how science is handed on, in terms of content and capacity, while at the same time constituting scientists as members of a community. Successful communication or teaching in handling exemplars, including a demonstrated capacity to see around and beyond what the sample exemplifies, may lead directly to access to resources for science – laboratory time, technical support, research grants and access to publication outlets. But success from the use of exemplars in one institutional form is not guaranteed if

RITUAL AND CATEGORIES

93

countervailing institutional pressures are too great, as the fate of the Lele in the 1980s showed all too clearly. When applied to scientific explanation, this line is sometimes labelled a social constructivist approach and treated with scepticism by scientific practitioners themselves. Therefore, it is important to note that Kuhn and Douglas are arguing only for the mutual constitution of science and scientific community because of a system of exemplification, and not suggesting that social factors determine scientific facts. Douglas now shifts to her main purpose in ‘Rightness of Categories’ by refocusing her entire argument on a revised understanding of her Lele ethnography. Living in a hazardous world with many unknowns, Lele villagers approach the task of grappling with that world from an explanatory perspective comparable with processes in the sciences, as analysed by Goodman and Kuhn. The Lele, too, proceed by building explanations from exemplars and denotation. The system then must be grasped in its entirety, and not piecemeal. The daily working environment of the rural Lele in the 1940s and 1950s was not at all comparable with the working conditions of scientists – Lele villagers lived by hunting and hoe agriculture on the edge of the rainforest – but the mutual constitution of community and knowledge is directly comparable to processes of world-making via exemplification, denotation and ritual reinforcement mapped out by Durkheim, Goodman and Kuhn and applied across the human range of religion, arts and science. Similarity and difference are now abandoned. When it comes to the ways in which communities think about and teach their understanding of the natural world, we are all essentially the same. This now extends to a recognition that local systems of classification have some capacity for change; the rule that the potential of an exemplar is not exhausted by enumerating what it exemplifies applies to Lele classification and scientific taxonomy alike. There is a proviso; changing social problems and consequent social response are the most likely drivers for any such exploration. She is not implying that Lele intellectuals engage in speculation for its own sake; but we doubt she was claiming that for science

94

MARY DOUGLAS

either. That is probably something for the self-appointed thinkers of social media. What she is claiming is that that there is only one explanation for ‘the bizarre ideas of the Lele about their world, and the bizarre ideas of physicists about theirs’ (p. 299). She then tracks back through the Kuhnian account of scientific teaching and applies a version of it to her Lele material. ‘We have to ask how the Lele choose their equivalent of exam questions and what are the puzzle-solving techniques they have to learn in solving them’ (p. 299). Like contemporary professional risk analysis, Lele religious practice was strongly oriented towards averting misfortune. Public debates are likened to meetings of exam boards in which hapless professors are sanctioned for not asking their students sufficiently probing questions. Public puzzle-solving performances (recall the poison ordeal) use exemplars intended ‘to make and contest claims’. Regular rehearsal ‘entrenches . . . medical categories through standard question-and-answer’. ‘Curing confirms and death disconfirms the classification.’ The punchline appears: ‘In the long run the system to which [the] medical categories conform is the system of social relations’ (p. 299). The pangolin, and the question of who is allowed to eat it, now has an explanation. The entire animal kingdom is graded in terms of relative success in human reproduction. Lele identity is defined by comparison with neighbouring groups in ways that are rooted in hunting. Therefore, domestic animals (goats and pigs), which are not hunted, are forbidden to adults. Children (who have yet to reproduce) can eat anything. Women should be careful about chickens. Only those men who had been most successful in the reproductive game – the begetters of a female and male child by the same mother – can eat the pangolin and share in its cult. Anomaly celebration was in fact anomaly reintegration for formerly anomalous young men, now sons-in-law. ‘Before I realised the weakness of the metaphorical argument’, Douglas writes, ‘I was content to say that the Lele projected the universe upon the animal world’ (p. 299), but now she understands how the mapping was done – how (in the subtitle of the section) ‘exemplars constitute communities’. When the Catholic church

RITUAL AND CATEGORIES

95

undermined the system of categories by insisting that anyone could eat anything, exemplification broke down and community breakdown followed. Her new understanding is that ‘the underlying principles . . . deployed for the animal categories – territory, overlordship and enmity’ were ‘prime examples of [Lele] political relationships’ (p. 301). Rules about who can eat what serve to remind people where their loyalties lie. When sickness then intervened, there was a public examination of whether the exemplars had been given due weight and whether the implications of the publicly attested system of denotations had been followed. This reinforced the system of food rules and in turn buttressed Lele political relationships. However, her new understanding had to face up to social change. The Catholic mission had lobbed a grenade into the Lele social system by telling converts they could eat anything. The cults were challenged and old beliefs exposed. And yet grievances, many now resulting from the economic involution of a postcolonial rural society, were piling up without check or resolution. This posed a problem she was compelled to address when she returned to writing about the economy (Chapter 8, below). ‘Rightness of Categories’ first appeared in a book about classification in science and had to deliver on an obligation to address debates about cultures of science. For this reason, she concluded with a section on interests. However, a comment citing her coeditor, David Hull, hints at the grid-group cast of her thinking. ‘In the history of science, we find all kinds of cultural variation, entrepreneurial individualists, hierarchical control points and small enclaves united in protest against some established mainstream’ (p. 305), after which she ends with a quotation from Richard Rorty: ‘Human communities can only justify their existence by comparisons with other actual and possible communities’ (p. 307). Her mind was moving back towards ways in which her scheme for mapping human organisational differences and thought styles might be used for resolving collective shocks. What happens when events render local schemes of social classification completely dysfunctional? How could groups like the Lele cope with

96

MARY DOUGLAS

postcolonial corruption in the Congo or the threat of New War, and what part was being played by external agents of change such as the Catholic mission or World Bank?

NOTES 1. In standard theory of classification, not all classifications are taxonomies. A taxonomy is a scheme of classification that is (a) strictly hierarchically ordered (b) made up of a closed set of higher-level categories (c) based on a small number of explicit and consistent principles or rules. Thus, the periodic table or Linnaeus’s system of species are taxonomies. Most likely Lele classification of animals was not strictly taxonomic in this sense; in the same way, the standard classification of risks discussed by Douglas in her 1980s work is not taxonomic. Here, we refer to taxonomy only with respect to category systems used in biology, and otherwise to classification. 2. Page references are to the reprint of the essay in the second edition of Implicit Meanings (Douglas 1999d). 3. The Philosophy Dictionary of Arguments (https://philosophy-scien ce-humanities-controversies.com/listview-list.php?concept=Gavagai) considers this to be an imaginary word from an unknown language necessary to Quine’s thought experiment on the indeterminacy of translation. 4. Bodily acts of pointing or showing are foundational. Here, Douglas was extending an argument made by Mauss (2006 [1935]) in his famous essay on ‘techniques of the body’ that gait, deportment, manner of walking or swimming, etc. do not simply enact a particular form of social organisation; they also implicitly teach it and enable others to learn its specific categories and its peculiar relations with other people. 5. Goodman 1976 [1968] doubtless overstates his case that only note-perfect performances count. Plenty of performances with some mistakes successfully interpret a score, while note-perfect but lifeless performances miss the musical point. Douglas’s argument is that ritual effectiveness in exemplification can tolerate a range of deviation while still fixing categories, and therefore avoid the problems that Goodman’s own account of musical world-making ran into. For a critique of Goodman’s account of the relationship between notation and performance, see Ridley (2004: 106–9). 6. Thomas Kuhn (1922–1996), by training a physicist, is best known for his theory of paradigm shift in scientific thinking (Kuhn 1970 [1962])

RITUAL AND CATEGORIES

97

based on the notion that science changes by revolutions. His later work (e.g. Kuhn 1974) focused much more on the use of exemplars in teaching and on the role of shifts in classification as the basis for assessing incremental change. 7. But for his later, more nuanced view see Popper (1976).

CHAPTER 7

GIFTS, GOODS AND ECONOMIC DEVELOPMENT

••• A

s noted in Chapter 1, Douglas had some initial training at Oxford in philosophy, politics and economics. Her husband was an economist, and she retained a lifelong lively interest in the subject (even taking a sabbatical to refresh her knowledge, Wachtel 2013 [2003] [reprinted in Fardon 2013b]) and made a distinctive contribution to economic anthropology through studies of consumption. Her key work on institutions (Douglas 1986) coincided with a revival of interest in institutions in economics, as apparent in work especially by Williamson (1981, 1985), Schotter (1981) and North (1990, 1992). We will discuss her work on consumption and then turn to her critical commentary on the new institutionalism in economics. Her work on consumption was targeted widely because it was intended to pose a challenge to some of the most basic assumptions of the economists in setting up their discipline. Her aim was to use the study of consumption to develop a set of theoretical arguments about goals and restraints in economic life, to supply reasons and justifications for economic behaviour, and to bring out the importance to economics of ritual acts and forms of gift giving – aspects of social life that tended to be neglected in mainstream economics. She saw the preoccupation of economists with market forces as a limitation, mainly because one must first explain preferences; who does and who does not want goods or services, for what and why?

GIFTS, GOODS AND ECONOMIC DEVELOPMENT

99

Her contributions to economic anthropology applying arguments about rituals of consumption are contained mainly in the book she wrote with an econometrician, Baron Isherwood, The World of Goods (1979), and in a collection of papers on money and food republished in excerpted form in the edited collection In the Active Voice (1982a). A clear pointer to the route she intended to follow is to be found in an earlier essay, ‘Deciphering a Meal’ (Douglas 1972b), exploring the ritual interaction order of eating together or alone, which she later added to the second edition of the essay collection Implicit Meanings (Douglas 1999d). The ritual processes of different kinds of collective and solo eating shape our wants both for foods and for the foods we avoid; the social organisation of eating must be examined at the daily level of ritual at the dining table or bus queue, or wherever we eat, in which duties to and claims of others upon our eating are implicitly managed. The collaboration with Isherwood left the statistical analyses of consumption and class to the second part of their co-authored book, which is presumably where Isherwood took the lead, since Douglas freely confessed that she did not relish the mathematical aspects of economics. Some of the arguments of The World of Goods were then reiterated and given further examples in a set of papers collected in Thought Styles. The 1998 book Missing Persons (with Ney) and an article on ‘Human Needs and Wants’ deepened Douglas’s critique of standard economic perspectives and their lack of recognition of heterogeneity in wants and needs for goods. A key paper in her late output (Douglas 2004a) then tackled the ‘new institutionalism’ and approaches to poverty alleviation in development economics. In The World of Goods, Douglas’s central argument was that consumption – whether of a routine kind or in the form of special events such feasts or fasts – was a type of ritual, or rather four types of ritual. Although this argument drew on Veblen’s early (1991 [1899]) work on ‘conspicuous consumption’, which itself drew on an earlier discussion by Marshall of preferences for honour, in Douglas’s hands the notion is vast in scope and goes far beyond status-display. It may go so far as to include the possi-

100

MARY DOUGLAS

bility (though this is not an implication Douglas develops) that eating disorders might reflect ritual ordering gone wrong (there is evidence that social trauma plays a significant part in conditions such as anorexia and bulimia: Seubert and Virdi 2019). Her focus throughout her body of work on consumption is that through feasting and fasting, and other means of encouraging or ascetically limiting consumption, people engage in learning about their social world while also acquiring the typical blinkers and partisanships that come with becoming members of specific social groups or looser clusters. People use consumption, Douglas argues, as a way of constructing an intelligible social universe. Experimentation, or refusal to experiment, with clothes, décor, equipment and associating forms of consumption with special occasions, such as religious holidays, and even the possession of certain ritual objects, not excluding books or certain kinds of music, were ritual activities signalling group membership or nonmembership on one organisational dimension while signifying discretion or authority on another. In short, consumption submits to analysis by her method of grid and group. In effect she is elaborating upon Marcel Mauss’s (1990 [1950]) insight that the economy is socially embedded and can be more fully revealed by following interactions not normally detected or analysed by economists, such as patterns of gift exchange (Chapter 6, above). In a preface she supplied for a new English edition of Mauss’s celebrated (1990 [1950]) essay on The Gift, she elaborates on the claim (Chapter 5, above) that Mauss provided the Durkheimian approach with a viable analytical methodology, based on mapping patterns of exchanges, including goods but also immaterial rewards. Mauss claimed every gift requires a return. Douglas makes clear that the path of return was often indirect, if it existed at all. If a complete inventory were to be made, it might then be discovered that transactions accumulated upwards, spread outwards or by-passed some individuals and groups altogether. It was her hope that through tracking such patterns of consumption the institutional biasing effects of the four elementary forms would be made plain, thus providing her

GIFTS, GOODS AND ECONOMIC DEVELOPMENT

101

theory with empirical confirmation. The pattern, Douglas notes, is not conveyed in the ideas and rationalisations of informants but would become apparent only when the interactional tally of ritual exchanges is totalled. Then we would see, she suggests, who is honoured and who is debased in what she calls a ‘drama [of gift exchanges] directly cued to public esteem’ (Douglas 1990: xiv). Douglas was involved in several large-scale surveys of household consumption, along with Gerald and Valerie Mars, and Karl Dake and Michael Thompson, but it is unclear whether she, or anyone else, ever attempted the total mapping of gift transactions she infers as being called for by Mauss’s essay on gifting. Esther Mokuwa (personal communication) tells us that inspired by reading Douglas’s comments on Mauss she once attempted such an exercise among villages bordering the Gola Forest in eastern Sierra Leone, using a team of highly trained field surveyors. The first pass was straightforward, and seventeen categories of gift with distinct names in the Mende language were identified and their frequency recorded in the daily interactions of a large sample of informants. But following up on the ramifications of these transactions – for example trying to find out how many gifts were returned or redirected and when, how and to whom – became a daunting task and had to be shelved for lack of time and resources. Economists have constructed input-output models of entire economies but so far as we know never using the cultural categories that would have satisfied the demands of Douglas’s vision of what Marcel Mauss had in mind. There are some points of comparison between what Douglas was attempting to conceptualise in The World of Goods and Pierre Bourdieu’s work on consumption and class. Evidently, she knew Bourdieu’s work La distinction: critique sociale du jugement (Bourdieu 1984 [1979]) in the original French because Douglas and Isherwood (1979) cite it, and she wrote a (1981) review for the Times Literary Supplement. She argued that far from being stable, consumption preferences are often periodic or episodic in ways that are best explained by the ritual function of consumption. Ritual calendars often play a crucial part in the regulation

102

MARY DOUGLAS

of this vast network of exchanges. What is appropriate to eat at Christmas, at Diwali, at the end of Ramadan, on a twenty-first or sixty-fifth birthday, on holiday or at a weekly team working lunch shape what we want. People never just want nutrition, but rather they are conditioned by society to want culturally formed eating practices or react against conformity and therefore must reach for or find themselves in some other form of social ordering with a distinct calendar of eating periodicities. In effect, Douglas is placing the arguments about the periodicities of social time made by Durkheim’s student Henri Hubert (1999 [1905]) as central to a social theory of consumption. Where Bourdieu (1984 [1979]) suggests that discourses on consumption might be used to explain how things are consumed, Douglas will have none of this. As always in her later work, ideas are something to be explained (Y variables), not causal factors (X variables). Instead, it is the ritual process of organising and disorganising people that shapes ideas about consuming, as indeed Bourdieu (1970) had recognised in his own earlier work on the Berber people of Kabylia (in Algeria). In How Institutions Think, Douglas had briefly engaged with the work of the American economist Oliver Williamson, awarded the Nobel Prize in Economics for his work on transaction costs. Williamson (1975, 1981, 1985) had broadened the scope of his discipline (after some decades in which it had been mainly preoccupied with market forces) by revitalising the theory of the firm. Why might traders sometimes seek protection by working in firms? Some traders, in some fields, found the costs of acquiring sufficient information to trade independently were too high. It made sense for them to cede some autonomy and move under the protective umbrella of a firm. This reduced the costs of doing business. Williamson developed a general theory of institutional forms as the result of ways of controlling transaction costs. Douglas recognised that what he had proposed was a twoculture model of economic activity – in which markets and hierarchy (a mode of organisation typically associated with the firm) existed side by side. She asked why one would stop there. According to her theory of elementary forms of social organisation (grid

GIFTS, GOODS AND ECONOMIC DEVELOPMENT

103

and group), Williamson had paid attention to only two of four possible organisational options. Might it now be possible for economists to work with a model of all four elementary forms of culture? The extent to which Douglas had followed the debates of economists about the ‘new institutionalism’ became apparent in a rather unexpected place – an encyclopaedia article originally published in Italian (Douglas 1995a [2008] [reprinted in Fardon 2013a]), which was not published in Douglas’s original English version until after her death. The essay in question, ‘Institutions: Problems of Theory’, deftly summarises the history of its topic in both economics and sociology, including a comment on the contribution of an otherwise obscure group of nineteenth-century French Catholic thinkers on the topic of institutions and natural law. Spencer and Durkheim find their place, as do later theorists such as Talcott Parsons and Ronald Coase. She then addresses several new institutionalisms in economics. The achievements of both Oliver Williamson and Douglass North are discussed, followed by a section on ‘Perverse and Benign Institutions’, making clear that transaction cost challenges of coordination, information and trust are as severe in small-scale societies such as the Lele as in supposedly ‘advanced’ industrial economies. The item is then rounded out with a profound reflection on ‘The Effect of Institutions on the Individual Mind’. This is where the thrust of her article is focused. ‘A model of the mind that constructs only one kind of rational agent does not take into account the responsiveness of persons to the minds of other persons’. In short, economics needs to find social relations. Better models are required because ‘the very emptiness of the concept of the self-interested rational person [as supposed in so much of institutional economics] makes it practically impossible even to formulate the problems of commitment and moral judgment’ attracting attention in other social sciences. This is an invitation for economists to join with these other social sciences in pursuing ‘the questions which a full theory of institutions would require’. In 1998, Douglas published an extended analytical examination of the economic literature on preferences in a co-authored

104

MARY DOUGLAS

piece on ‘Human Needs and Wants’ (Douglas et al. 1998) in a broadly inclusive four-volume collection of social science contributions to the study of climate change, edited by Steve Rayner, her former PhD student, who had notably expanded her work on grid and group. ‘Human Needs and Wants’ extended the critique made in Douglas and Ney’s (1998) book Missing Person: A Critique of Personhood in the Social Sciences and elsewhere of methodological individualist approaches in economics, individual psychology, development studies and rational choice sociology. The article criticised both the material ‘basic needs’ and fuller ‘human needs’ approaches to the development of indices for measuring human needs on a universal basis. Both works argued that her typology could provide a richer set of conceptions of where wants and needs come from and of how people rank different kinds of wants, reflecting the social organisation by which they live and through which they link with and conflict with others. In 2004 she published the paper ‘Traditional Culture: Let’s Hear No More About It’ (Douglas 2004a), which proved to be a strongly critical engagement with new institutionalism, addressing the work of a second Nobel economics laureate, Douglass North. ‘Traditional Culture . . . ’ stands as a major summation of her arguments on institutions and grid and group. She begins with an outcry. Extreme poverty in a world of plenty is a moral outrage. As with the slave trade, future generations will wonder why we ever allowed global extreme poverty to happen. But we fail to address it because we cannot see how to stop it. The causes are diffuse. It is unclear who to blame. Douglas then identifies a pernicious source of dithering – the inadequate state of the theory of institutions in development economics. New institutionalism proposed a theory in which institutions – as rules shaping economic life – begin in a tentative and ramshackle form but then become more refined and effective with use. Successful institutions drive out ineffective institutions. Initially, North (1992) focused on the distinction between informal and formal institutions, though in later work (e.g. North, Wallis and Weingast 2009) he recognised that every formal institution depends on a penumbra of informal ones to sustain and

GIFTS, GOODS AND ECONOMIC DEVELOPMENT

105

interpret it. In Douglass North’s terms, at least as set out in his last major statement published shortly after Douglas’s death, the institutional refinement dynamic is an evolutionary transition from closed to open access orders (ibid.), which may not be inevitable but where it is completed produces what he and his co-authors regarded as unambiguously superior outcomes. Economic development, therefore, is the result of introducing openness to institutions of law and governance that were previously controlled and captured by elites. North’s two-step-with-transitional-stages model of institutional change has been widely influential. For example, Acemoglu and Robinson (2012) argue that countries achieving successful transitions from poverty create or adopt ‘inclusive institutions’ triggering virtuous circles of innovation, economic expansion and wealth distribution.1 Douglas has two major problems with this (in her view) limited reading of new institutionalist theories. Institutions cannot be carted around from place to place, like a generator for emergency electricity, guaranteed to provide light wherever required. They grow out of a specific social context. People trust institutions where they see that the institutions are part of their own ritual and social ordering. As she had shown in How Institutions Think, breathing life into institutions is a complex sociological process, based on identification of suitable analogies then embedding these in community life by repeated ritual action. Secondly, the process of institution building is not unidimensional but shaped by a double set of criteria for regulation and integration, recognising a greater range of institutional options than North considered. Under the so-called Washington Consensus in the 1980s, global poverty alleviation stressed the role of market integration. But what about all those forms of institutional solidarity not based on (low grid, low group) competitive individualism? Developed countries do not abandon hierarchy in the armed forces, for example, in favour of competitive individualism. Even in the most democratic countries many institutions remain much more closely locked against non-elites than may be officially admitted. So who decides what is a ‘good’ and a ‘bad’ institution and according to which criteria?

106

MARY DOUGLAS

Many institutions do not start off ‘good’ or ‘bad’. Some simpler ones start off simply because a rule is needed. Whether traffic moves better by the left or right is not the issue but whether a decision has been imposed. Napoleon required those parts of Europe he conquered to ride on the right. Remembering to switch over remains a problem for motorists using the tunnel between England and France. Recently arrived British motorists determinedly keeping to the right of the road as they enter the motorway system of northern France are then somewhat amazed to see the French trains they glimpse from their car windows pass on the left. This is because the French railways were built by British engineers, and Britain was never conquered by Napoleon (British trains also pass on the left). Once the signalling and other crucial track side equipment was installed, it was too much effort ever to change, even in a country ‘rationalised’ by the Code Napoleon. These simple cases of conventions are not the typical cases of institutions, however, because most institutions emerge from fragile and provisional settlements of conflicts rather than being simple solutions to ‘coordination games’, to use the term that game theorists apply to problems of the ‘rules of the road’ type. Institutions that emerge from conflicts are not discrete in the way that road rules can be. They are deeply enmeshed in every other set of institution through which daily life is conducted. Douglas reminds us, therefore, that institutions do not operate alone – they belong to a complex web of interacting institutions, and institutional changes have system-wide effects. Bearing this in mind, there must be a preference for locally driven processes of adjustment. Quite what a ‘virtuous circle’ would look like in such circumstances is not altogether clear, but perhaps it would comprise many interlocking circles. The census of the Book of Numbers, with its double rotating circles of Hebrew marching bands (Chapter 8, below), would be only the start of it. Douglas then outlines, for the benefit of new institutionalist readers (the book in which the essay appeared was edited by academics with connections to the World Bank), what a more fully elaborated set of achievable institutions might look like, drawing on the fourfold ordering of the grid-group scheme.

GIFTS, GOODS AND ECONOMIC DEVELOPMENT

107

She explains that people have developed their institutions with materials at hand and with reference to local constraints of grid and group. This means that sometimes they make choices that appear strange to outsiders. Maybe they limit consumption, even when they are malnourished. This fasting is not stupidity but evidence of interfamily cooperation underpinning labour cooperation and social security, as shaped by a shared ritual calendar. Douglas is particularly concerned with the charge sometimes made that the poor are apathetic. Allegedly, they lack the drive to solve their own problems. But is this really a lifestyle choice or an adaptation to highly limiting conditions? To illuminate the second possibility, she directs attention to the top left quadrant of her grid-group scheme. Even in joint writing with her own colleagues she had made little use of this part of the diagram, sometimes referring to it, pejoratively, as a zone of fatalism (Douglas, Thompson and Verweij 2003). In her sole authored work, though, Douglas dismisses this (e.g. Douglas 1996, passim). Her theory does not deal in psychological dispositions. ‘Isolate’ forms of social ordering by strong social regulation and weak social integration will suffice. But why are the denizens of this quadrant isolates in the first place? This is a place of high regulation – either other groups or the environment restrict options for livelihoods and social agency. At the same time, there may be good reasons to minimise contacts with the neighbours, especially if they might betray you to others. Isolate ordering is not necessarily passive nor is it a residual category. Examples of isolate institutional ordering are to be found among enslaved people escaping bondage, or in war zones, where groups struggle to survive without attracting the attention of any of the warring factions. She cites a well-known study by Banfield (1958) of a war-ravaged peasant community in southern Italy, but other, and perhaps better, examples have since been documented (Richards 2022). The point she wishes to make is that this is not ‘apathy’ in the sense of a lack of drive or interest. The issue is lack of freedom. If the World Bank was seriously interested in ending poverty, it would pay attention to lack of freedom, as (she notes with approval) Amartya Sen had long advised.

108

MARY DOUGLAS

The overall point of her critique now comes into view. According to the theory of Douglass North, the poor were poor because they remained locked in closed order institutions, when rich and successful countries had honed open access ones. She finds this a scandalous argument because it appears either to make the poor to blame for their own poverty or to leave them merely passive victims of plots by local elites. Furthermore, the argument is based on a fundamental error – of thinking that culture causes anything. Culture is the product of social life. It is part of the Y variable. Poverty alleviation must begin on the left-hand side of the causal equation, with the X variable. This includes the processes through which groups form institutions in often very difficult and limiting circumstances. If the cultural production of institutions was less circumscribed, better institutions might result. The answer to global poverty is more freedom of action (and material support for such freedom) not boat loads of crated-up and pre-formed Washington Consensus. Thus, her final and pointed advice to the economists, among whom she had begun her intellectual journey, was ‘traditional culture – let’s hear no more about it’. How, then, shall we sum up Douglas’s work on economic issues? Her work on consumption enlarges the field. Consumption is more than price-driven rationality but includes a wider set of cultural considerations, notably the scope humans find in having, or not having, goods, for ritualised control of what would otherwise be the tumultuous chaos of everyday life. The ritualisation of consumption is an everyday medium through which institutions are stabilised. She brings Mauss back into the economic picture, reminding us of the methodological significance of gifting as a way of understanding how social systems are built up from transactions that mainstream economics tends to miss, and likewise Henri Hubert, emphasising the way in which the periodicities of time in our ritual order cultivate and shape and distribute our preferences as consumers, workers or citizens under a specific form of political economy. Perhaps her most important contribution is to enlarge the scope of new institutionalism. Here we need to enter a warning.

GIFTS, GOODS AND ECONOMIC DEVELOPMENT

109

Her grid-group scheme has an important purpose in indicating the range of social selection pressures serving to fix institutions, but we need to keep in mind that this is simply a means to an end and that her ultimate objective is to explain deeply rooted sources of social intransigence. Institutions, she argues, are always bundled in sets and can never be treated independently (Douglas 1995a). Legacies matter, and many institutional bundles are marked (sometimes scarred) with local selection histories. Histories of slavery or class exploitation (institution formation on restricted grid and group) matter a great deal, therefore, in evaluating differences in institutionalised responses to current challenges, such as pandemics or global climate change. Maybe human groups think in silos (Tett 2015), but silos exist for a reason, connected with the history of institutions. This history comprises many tangled thickets. It is thus no use to advocate nostrums – for example, to ‘follow the science’ in relation to a global pandemic – more especially when a group has a history of being neglected by science (consider the history of marginalisation from modern medicine experienced by many African communities, as most recently expressed by contortions in rich countries over Covid-19 vaccine access). Groups will process new challenges but only through reassessing their own institutional commitments because institutions are the guarantee of an orderly and meaningful social life. Ultimately, she recognised and concurred with the merit of Amartya Sen’s argument that greater local freedom was a key missing component in improving the quality of adaptive institutionalised responses to existential crises.

NOTE 1. Both North Wallis and Weingast (2009) and Acemoglu and Robinson (2012) recognise that success in reducing poverty is by no means as simple as importing the ‘right’ (open access or inclusive) institutions, but the net impact in development practice has been to advocate ‘policy transfer’ of institutional templates from other countries.

CHAPTER 8

STRIFE

••• C

onfident about cause and effect, in relation to institutions, Douglas picked up the mantle of the Dutch Enlightenment philosopher Baruch Spinoza (1632–1677) in her later work, using the books of the Hebrew Bible as a data source for examining strife and the possibilities for calming it. Spinoza has a good claim to be a pioneer anthropologist of the Bible. Raised and educated in Amsterdam’s Portuguese Jewish community, he knew Biblical Hebrew well. Expelled as an apostate, he continued his intellectual journey in company with the Collegiants, who were Dutch radical Christians of mainly a Mennonite background (Fix 1991). Radical Christians in the seventeenth-century Dutch republic frequently combed the Old Testament of their Bibles, hoping to find the voice of divine inspiration to buttress their socially radical views on how community life should be lived. Spinoza was their go-to person to help them understand what the ancient texts were saying, but he had disappointing news (Spinoza 2002 [1670]). The Pentateuch was the law book of an ancient polity that no longer existed. There were only two laws of relevance in seventeenth-century Amsterdam – to respect God-given natural law, and to show comradely regard for neighbours. The rules of purity in Leviticus were the institutional codification of a political community long since disappeared. A new world of trade and scientific empiricism needed altogether different styles of thought, and Spinoza was busy giving some of these a requisite philosophical grounding. But he makes clear that Jewish scriptures might have made good institutional sense in their ancient political context.

STRIFE

111

Douglas’s Biblical project sought to resurrect that approach. She focused on community life in ancient Israel c. 500 BCE, homing in on context, and then tried to show how rules and rituals protected key institutional values. The Babylonian captivity had divided Jewish society, and with the return home of the exiles, major institutional accommodation was required. The editing of the Hebrew Bible became a test case for her theory of how social life might become institutionalised through ritual (in this case, liturgical) agency. A key event was the moment when the priest Ezra attained leadership of a faction of returnees in Jerusalem concerned with the restoration of the temple and its rituals. The returnees had encountered a mixed population within which they needed to integrate. Ezra denounced those remaining in occupation of the land who had married non-Jewish wives. He ordered them to undergo divorce. But why? What set of practical or organisational concerns would force such a harsh and drastic step? Douglas immediately thinks in terms of what she had learned about peasant communities in Kasai province faced with Lord Leverhulme and his army of colonial oil palm planters. The only recourse the Lele had to keep them from destitution was an agreed set of local rules about who could farm the land, fish the waters and hunt the forest, and now the resources themselves were being depleted by a colonial land grab. Lele religion was a ritual and symbolic depiction of the practical governance of these disappearing resources. It protected Lele basic interests, even at the expense of many internal contradictions, requiring much patient attention to local dispute resolution. But as land resources dipped further, Lele institutions fell into crisis. Ezra, Douglas concluded, was facing the same kinds of challenge and decided to solve the problem at a stroke by instituting a land grab of his own, enrolling the voice of divine authority to that end. The divorcees would be obliged to remarry, to protect the old religion, and the daughters of the exile returnees could now regain access to land through their new husbands. Former wives would be dumped and driven out as refugees. This suggested the need to look closely at how ritual ordering served to institutionalise Ezra’s enclavist scheme. Douglas then

112

MARY DOUGLAS

sensed that parts of the Bible were being edited by priests hostile to Ezra’s narrowly nationalistic reforms. These priestly editors were engaged in a search for ways of shaping institutions to embrace a wider range of factions within a hierarchical framework. Douglas begins In the Wilderness (2001 [1993]) with a focus on the politics of the enclave, a form of social ordering that had been treated too high-handedly in Risk and Culture. The foreign wives were being expelled because the enclave needs strong boundaries to hold its members together. The dilemmas of Ezra’s leadership are also explored. Enclave leaders lead by charisma but lack strong executive power. How are hierarchs to respond to the threat of factions within? Actual ancient Israel was a mix of Hebrew people with a range of origin stories. Somehow, these would have to be brought together into an analogical synthesis if the nation was to escape factional fragmentation from which only surrounding enemies could profit. The Book of Numbers is a kind of census, Douglas concluded, but written out in performative notation (Douglas 2001 [1993]). It denotes the Hebrew tribes in circular ordering and then denotes them again (from the middle of the book) in reverse order. The picture is that of two counter-rotating concentric rings, and it appears to serve, Douglas argued, to demonstrate a degree of interdependence while accommodating a degree of difference. Maybe it was even conceived as a choreographic scenario? Might ring dancing have been used to counter the enclavist dream of national purity?1 There is a lot of speculation here because the contextual data concerning daily social life in ancient Israel, more than two millennia ago, are thin, to say the least. But the attentive reader of How Institutions Think is in no doubt that this is the kind of device that Douglas had in mind when she refers to fragile social institutions founded on loose analogies and needing to find buttressing in ritual or liturgical practices. Her last book, Thinking in Circles (2007), is a virtuoso exercise in analysing and decoding ring forms of liturgical or poetic composition. This makes it clear that this was once a widespread literary form, though dropping out of use to stabilise institutional change in today’s highly interwoven social worlds convergent on a singu-

STRIFE

113

lar planetary destiny. She returns to the Book of Numbers in Chapter 4, which is titled, with Douglas’s sly wit, ‘Alternating Bands’ (as well as amulets, think marching bands going round in circles but starting from different parts of the town square). Rings are not the only option to give strength to metaphors of institutional ordering, however. There are many other mathematical forms or geometric models that can be pressed into service to give shape to the rough analogies through which institutional forms can be grasped. In Leviticus as Literature (1999a), Douglas found a different kind of buttressing. Here the book traced out a cosmogram, or a pocket version of the tabernacle in the wilderness. Readers journey in their mind’s eye into the tabernacle and finally to the Holy of Holies, where they find nothing but the majesty of God and the certainty of his promises. Douglas understood that this enrolled aniconism (the belief that God should not be represented in any shape or form) in service of repairing the institutions of a fractured community. The institutional model got rid of the factional special pleading associated with ancestors, spirits and magic at a stroke. There was but one God of all the Hebrews, invisible and beyond representation. The cosmogram of Leviticus serves to illustrate, once more, what she had been arguing in How Institutions Think – namely, that fragile institutions need the protection of powerful imaginative props to help with the process of naturalising the analogy on which the institution is based. The empty tabernacle offers a space beyond history or faction in which changeless divinity dwells. Michael, Lord Young of Dartington2 thought we ought to talk to each other about bereavement, but sometimes (she argued) a wall of silence might be a better memorial to the dead (Douglas 1995b). It allows ancestral blessing to flow beyond the boundaries of rancorous division. While admiring the ingenuity of her overall arguments concerning the Hebrew Bible, one of Douglas’s critics among Biblical scholars, Lester Grabbe, suggested that her books on this topic could only be considered a ‘case not proven’ (Grabbe, in Heald et al. 2004: 157–61). No one knew whether the editors of the Pentateuch were building liturgical models to stabilise institutions emergent in existing social worlds or whether they were

114

MARY DOUGLAS

dreaming up model constitutions. Only the former would count as true examples of a Durkheimian mechanism. The latter would be instances of the approach she rejected of trying to explain ideas with other ideas. This suggests she had reached the limits of her fieldwork in the Bible. But perhaps this did not matter. Douglas was an anthropologist concerned with social change, focused on the problem of fundamental social conflict. Around the time of completing How Institutions Think, she had received the grant from the Spencer Foundation that allowed her to make a return visit to the Lele, and this refocused her concerns on the social turbulence of postcolonial Africa. Years earlier, she had sensed that the structural-functionalist account of the Tallensi offered by her esteemed mentor, Meyer Fortes, might have been a product of a relatively benign form of colonial protection. She had first met the Lele in a less favourable world, and their institutions, even then, were in trouble. Their dispute settlement system was clogged with unresolved cases, and there were evident tensions between the generations. Her return to the Kasai in the 1980s showed that things were now much worse. A witch-finding cult was sweeping the land, led by a Lele Catholic priest, in which elders had been targeted and killed. This was a condition that had earlier been recognised by Durkheim as ‘sacred contagion’. In the first preface to her collection Implicit Meanings (first edition 1973, second edition 1999d), she had been critical of this concept, thinking the French sociologist was making some loose analogy based on the idea of infection. But in their joint work on grid and group, Michael Thompson had helped her see that her ideas about ritual stabilisation of fragile institutions might have a cybernetic interpretation. Perhaps Durkheim had been talking about positive feedback and its outof-control tendency towards self-destruction? If so, this was the consequence of a process in which ideas drive ideas without any direct contact with daily social life. It was what goes wrong in the echo chambers of social media, when a pointlessly destructive Twitter storm (or similar) blows up. In the case of the Lele witch-finders, she traced the source of the trouble back to pastoral guidance issued by the Catholic

STRIFE

115

church. Cardinal Ratzinger (the future Pope Benedict) had advised the Theology Faculty of the University of Kinshasa to focus on the threat posed by American Pentecostal missions – they were to meet ideas with ideas. This had perverse consequences for the Lele, as an ideological battle boomed back and forth, generating a tornado of sacred contagion. What the priests should have been doing, she thought, was to start with the everyday realities of Lele social life. More specifically, they should have addressed the causes of the huge immiseration stemming from a disordered postcolonial economy in which international interests exploited the Congo’s wealth in cahoots with corrupt political classes. Perhaps a ritualised version of liberation theology would then help stabilise fragile and disrupted social institutions well enough to allow local remediation to take root? But the process would have to follow the X-causes-Y ordering proposed in How Institutions Think. Analogical strengthening of institutions would need to be grounded in daily social life. The uncontrolled fury of the Lele witch-finders was part of a pattern of postcolonial violence in Africa at the end of the twentieth century. Civil wars and ethnic riots were breaking out in regions as far apart as Senegal, Somalia and Mozambique during the 1990s. Douglas sensed she had the theoretical tools to explain such conflicts, apparently beyond the comprehension of policymakers located outside the continent, who labelled these rebellions without cause. But by now her health was beginning to fail, and perhaps she lacked the strength for a full book on the topic of so-called ‘new war’ (Kaldor 2012). What she did instead was to gather some examples of how her theory worked, mainly from her Biblical fieldwork, and to point out (mainly by means of footnotes) where the lessons applied to this rash of intercommunal African conflicts. The book was called Jacob’s Tears (Douglas 2004b). The explanation of the title comes in an incident described early in the book (pp. 28–31). The Hebrew patriarch, Jacob, led a pastoral clan seeking accommodation within a patchwork of apparently rather fragmented agrarian communities. Jacob had been trying to mend fences locally by reviving an old regional shrine,

116

MARY DOUGLAS

thereby securing concessions for his group and their animals. Then a prince of Shechem sexually assaulted Jacob’s daughter. Rather than rushing to vengeance, Jacob negotiated a settlement in which his people would agree to keep the peace if the men of Shechem were willing to undergo circumcision as a ritual marker of their shared societal bonds. Simeon and Levi, two of Jacob’s sons, overrode their father’s attempts to consolidate peace, taking revenge on a number of local men, and thus turning a group whose good will Jacob courted as allies into implacable enemies. The patriarch, furious that all his patient diplomacy had come to nothing, disinherited his two headstrong sons. This is an explanation of violent conflict in terms of failed institutionalisation. Jacob’s violent sons overrode the carefully cultivated ritual scheme through which it was intended that the two groups of herders and farmers would naturalise their shared interests. With a potentially escalating blood feud looming, there was little hope the parties could stabilise shared institutional arrangements. A better grasp of failed institutionalisation allowed a useful expansion of Douglas’s explanatory apparatus, which she then applied to peace processes. The Australian diplomat John Wear Burton (1915–2010) was a major figure in the field of conflict resolution. Burton was brought to Douglas’s own institution, University College London, by a group within the college (some of them members of the Society of Friends) interested in establishing a School of Peace Studies. Douglas and Burton overlapped in London for a period in the 1960s until he moved to the United States. Burton’s approach to conflict resolution was based on a bargaining framework (Burton 1990). His approach was to find out what the conflicted parties most needed and build up around that knowledge a ‘win-win’ scenario. Douglas rejected this approach not least because it presumed a good deal of prior institutional work before any bargaining could begin to enable people to present their ‘needs’ in reconcilable ways (see, for example, the essay ‘Human Needs and Wants’, Douglas et al. 1998 [reprinted in Fardon 2013a]). Burton’s bargaining design was rooted in only one of the quadrants of the grid-group map – the bottom left (low grid, low group), where deal-making thrives. Groups

STRIFE

117

located in other spaces on the map had collective commitments that cannot so easily be bargained. What happens when, as in Jerusalem, different religious communities all possess holy places sacred beyond price? To talk about a negotiated solution would have little effect and might even escalate into wars of words, and worse. This would be a scenario favouring the eruption of Durkheim’s sacred contagion and might prove overwhelming. How Institutions Think pointed to a possible solution. Ritual stabilisation had to arise from somehow conjoining conflicted sets of social processes. In Leviticus as Literature, Douglas (1999a) argued that the ritual process might have proceeded in distinct stages. A simplified version of her account of the situation in ancient Israel, in the period during which she considered the book of Leviticus might have been edited, would foreground several sects, organised as enclaved groups, practising purification rituals to counter defilement that might arise from sins or from improper contact with nonmembers. Hierarchically organised priests might then be wary of the risks of sacred contagion and concerned that a sectarian style of practising rituals of purification would be divisive. The challenge was to use the capacity of hierarchical institutions ritually to transform opposites into dualities coexisting as complements.3 Just as in her 1966 book, purity was the danger, and the danger in this case was civil war. She wonders how to address it ritually, and specifically whether a different kind of ritual performance could de-fang sectarian purification rituals that might otherwise lead to sacred contagion. Douglas argues that what Leviticus describes is a reworking of the purification ritual to become something more effectively reintegrative, in which everyone – member of a sect or not – stands in need of purification, and therefore purification must be reformed into a collective act rather a boundary-marking one. Rather than trying to replace purification with some other ritual, which might only have fuelled the fire of conflict further, Douglas suggested that the priesthood sought to ‘edit’ the existing ritual. The second stage of this process was the editing of the text of the book of Leviticus itself into a reintegrative cosmogram. This was not exactly a ring (as might suit an enclaved group) but an ordered, balanced

118

MARY DOUGLAS

and integrated compositional structure intended to display and – if the book of Leviticus was designed to be read aloud – perhaps ritually enact the reintegration of the community centred on an aniconic Ark of the Covenant shared by all. Instead of the pure and the impure being opposites, in which the enclaved pure shun the impure, rather we all stand in need of regular purification together, so that the states of purity and impurity (or indeed purity and danger) become routine phases of our lives in a kind of perpetual ritual canon,4 each phase dependent on the other in a collective process of daily profanation and reintegration. Ritual can calm conflict but only by starting with the rituals of the belligerents and gently editing and recomposing them into performances through which conflict can be dampened and sacred contagion danced or sung into peaceful interaction. A more recent example might help further explain what Douglas had in mind. The civil war in Sierra Leone (1991–2002) was triggered when a small group of Pan Africanist students launched a rebel movement, the Revolutionary United Front (RUF). Following a model proposed by guerrilla strategists working for the Libyan leader Colonel Gaddafi, seeking to establish a network of sub-Saharan client states, the rebels infiltrated a remote, forested region of the country, appealing to a politically disgruntled local population for support. The government army pushed back against the rebels but was not strong enough to follow them to ground in their forested lairs. A stalemate ensued, and the RUF became a strongly enclaved social formation (Richards 1996). The opposing sides developed strongly differentiated institutional viewpoints, and there were few points of contact for a negotiated settlement (Peters 2011). The attempt by government forces to eliminate the rebels with help from mercenaries provoked a backlash in which rural civilians caught between two militarised factions were horribly brutalised. Eventually, attempts were made to map out a path to peace (Richards and Vincent 2008). At first this process was run on conventional ‘needs-based’ negotiating lines. But the rebels had become extremely isolated within their own forest bases and had little or no grasp of what might be possible through negotiation.

STRIFE

119

Bizarre dreams were entertained about impossible political outcomes. The tactic of ideas being used to regulate ideas had only perverse outcomes when negotiations broke down and positive feedback drove violence to new and ever more horrifying levels (Richards 2013). A new start was made when negotiations were taken over by the Nigerian president Olusegun Obasanjo. In his previous career, President Obasanjo had been a no-nonsense military engineer. He took the Biafran surrender at the end of the Nigerian civil war and averted a bloodbath by camping his over-excited troops on every available hill crest, tasking them with the job of building their own barracks, while he reappointed officers in the Biafran police who had previously been in pre-war Nigerian service to run the surrendered territories on the ground (Richards 2011).5 The civilian population took the cue and began to rebuild. Normal social and economic life resumed, and very shaky postwar institutions became protected by a myriad of daily quotidian rituals. He now applied the same lesson to the predicament of the RUF. Out of the window went any further political negotiations based on the movement’s Afro-Marxist beliefs. In its place, the ex-combatant cadres were asked what might help them reintegrate with the national society, and they all pointed to the need for meaningful employment. A massive scheme of jobs training was introduced. It was full of incompetence and corruption, and most combatants recognised that six months’ training would not make them brick layers or carpenters. But the reintegration package provided them food, healthcare and a modicum of income, supplemented by selling their tools. Meanwhile, they had spotted other opportunities from years spent in the bush. Villages had abundant produce to sell but lacked roads and transport. Rural road building lay in the future, but cheap Indian and Chinese motorbikes allowed combatants to set up as bike taxi riders in towns, a new mode of transport in postwar Sierra Leone. Some then went into the rural areas from which they originally hailed and became involved in converting footpaths into bike tracks, sometimes adding simple wooden bridges. If their relatives rejected them, they turned to village women trad-

120

MARY DOUGLAS

ers and went into partnership, taking the women to market and sometimes advancing credit from their daily takings ( Jenkins et al. 2021). Daily economic activity required quotidian ritual consolidation (the common courtesies associated with trade, from handshakes and greetings, moving upwards towards rituals sustaining new friendships and marriage), thus stabilising very tentative postwar commercial institutions into a solidified thought style of petty trade in which combatant origins were forgotten. Villagers report on this by remarking that rebels were once alien killers but as bike taxi riders have come back to the village, settled down, had children and are now like the rest of us. Thus, the civil war in Sierra Leone was ended as Douglas imagined it might – through ritualisation working on daily life to embed new institutions of social ordering and commercial cooperation. Douglas partly explained her approach to conflict by writing about it in terms of positive feedback within enclaved groups. Douglas and Mars (2003) show how the violence of terrorist cells in Northern Ireland in the 1980s and 1990s was cultivated by organisational processes through which groups were cut off from a wider social world, living in self-sustaining bubbles of intensifying extremism. The result is lack of regulation by daily life linking potential enemies in common tasks; sacred contagion took over, and unrestrained violence erupted. In effect, this is the negative part of the argument. It is a pity she never wrote the other (positive) half – explaining the ways in which the careful management of ritual or liturgical activity consolidates joint institutional arrangements within peace projects. The obvious example from Northern Ireland is the part played by the Parades Commission in ensuring that Republican and Unionist celebrations of community identities take place at different times and places. This is a modest objective, but it controls flaunting in a setting where embattled communities have decided that there is no sustainable alternative other than to try and live in peace. Silence and avoidance are underestimated tools for accommodation. Representations (in Northern Ireland the abundant public murals expressive of the period of ‘The Troubles’) pose problems; they are graphi-

STRIFE

121

cally evocative, stirring support and rejection in equal measure. Sometimes only an aniconic deity will do. Douglas’s last public words are telling. When she was asked by The Spectator magazine a week or two before she died (in 2007) what her theory had to say about Islamic terrorism; she replied ‘you need to listen to your enemies’ (6 and Richards 2017: 213). This was interpreted (on the peace talks model) as a call for negotiation. Judged against the total picture of her work on intransigent conflicts, what she probably meant was something more like ‘for God’s sake, kindly shut up and listen’. Create or sustain, by all possible means, some practical shared activities motivating a need for institutional accommodation. Talk, however, is not the only or best medium for cultivating mutually beneficial institutional growth. Indeed, institutions think, but at times they benefit from having the silence in which to reflect.

NOTES 1. Her suggestion is that ‘contrapuntal organization of complementary authorities’ (Douglas 2001 [1993]: 67) was a solution proposed by hierarchically ordered priests to the problems of effervescent enclaves. 2. Young was a sociologist and social reformer who chaired the committee drafting the Labour Party’s 1945 election manifesto and who created many influential organisations over the course of a long career. Her essay was written as a piece celebrating his eightieth birthday. 3. It should be added that this interest in dualistic but complementary styles of ordering in hierarchy owed something to the influence of her friend, the French anthropologist Louis Dumont, who had argued that hierarchy is more than simply a relation of command or domination but integrates as well as regulates. Douglas had experienced such forms of organisation as her surrogate parent in her schooling and retained a lifelong feeling for hierarchy (Douglas 2005). Durable hierarchies often have a dualistic structure for classifying loyalties and claims – e.g. spiritual and secular power in the Holy Roman Empire, or department and profession in the British civil service. It was Dumont who first suggested that hierarchy works with opposed categories and finds ways at least to balance them in coexistence and perhaps to design structures

122

MARY DOUGLAS

within which these categories can be complementary. Douglas deepened and extended this insight in chapter 3 of In the Wilderness, a book based on her Gifford Lectures in Edinburgh (Douglas 2001 [1993]: 63–82). There she argued that in hierarchical ordering opposites can be performed contrapuntally, thus deftly turning Dumont’s ideological notion into one of composition ready for ritual performance, using a musical analogy. Perhaps surprisingly, her point of reference was not Dumont’s book Homo Hierarchicus (Dumont 1980 [1966]) but his essays on individualism (Dumont 1986 [1983]). 4. A canon consists of two melodies begun at different points that harmonise when superimposed. A perpetual canon is a canon so arranged that each voice, having arrived at the end of its melody, can begin again and so continue indefinitely. 5. Observations by Paul Richards in the vicinity of Uli airstrip, eastern Nigeria, February 1970.

CONCLUSION INSTITUTIONAL RENEWAL

••• W

hy read Mary Douglas today? Because she offers one of the most powerful explanations available in the social sciences of what drives people to think in the way they do. Many theories take ideas as given and then seek to explain what people do by reference to their prior beliefs. Douglas argued differently. Our thinking, she claims, is driven by social activity governed by institutions. This involves a distinctive approach to the part played by institutions in social, economic and political life – how they emerge and how they change. Institutions are explained functionally. Ritual action cultivates organisational arrangements (X), which in turn shape our styles of thought (Y) to sustain and thus protect the form of social organisation in our group (Z), whether that form be isolate ordered, enclaved, individualistic or hierarchical. Typically, this process occurs without explicit awareness of the feedback processes involved. In other words, people use ideas in complex and strategic ways for group support, and only when we come to understand the forces shaping these complex and strategic ways of thinking can we explain decisions and action. A distinctive aspect of Douglas’s approach is her focus on ritual. ‘As a social animal’, Douglas (1966: 63) wrote, the human being is ‘a ritual animal’. Ritual at the level of everyday conversations and encounters is a means by which we sustain social organisation through densely linked sets of institutions. Ritual does not merely reinforce group identity. As Durkheim (1995 [1912])

124

MARY DOUGLAS

stressed, it does something much more powerful. It fixes categories and sustains distinct ways of managing anomalies arising from our use of categories. The density of the linkages among our institutions then tightly ties our ways of classifying things, with the classifications of actions and responsibilities reflected in our moralities. To classify is to take the first step towards identifying the moral status of anomalies under a particular thought style, and therefore towards understanding how we come to blame others or how we will respond to being blamed by others. Institutions are legitimated and stabilised by analogical thinking and grounded in appeals to nature. This naturalisation may rest on something as simple as a geometrical figure (we belong to a social circle) or an animal metaphor (our boss is leader of the pack). Shared rituals build emotional alignments to strengthen shaky metaphors. Any artifice in setting up the group is backgrounded through this process of naturalisation and emotional alignment. The institution is now ready to act as a framework for regulation of activities – for example, sustaining livelihoods, running a firm, carrying out scientific research, fighting a war or organising a club. The group now has ways to conduct its argument about its work. Tasks and the rules through which they will be executed seem self-evident, until, that is, rival groups with different histories of institutionalisation are encountered and then conflicts can arise. Where coordination problems or outright conflict become troublesome, these problems will not be solved by a free and frank exchange of views. Our ideas on topics such as virtue, liberty, fairness, loyalty, justice and need – or indeed any of the moralities – are underpinned by institutions. Douglas concluded that moral order would vary in relation to constraints on institutional development. Without reference to these constraints through which different institutions have been formed and are sustained, we risk argument boiling over in the empty fury of ‘sacred contagion’, a process of positive feedback where there is no underlying set of mutually important social considerations to hold the process of disagreement in check. It is this final step in Douglas’s explanatory argument that leads us to a second important reason why anthropology, social science

CONCLUSION

125

and citizens at large should take account of Douglas’s legacy. For having explained the institutional amplification of conflict and dysfunction through the ritual cultivation of positive feedback in a functional causal loop, the next challenge is a normative one. How can we address any deep and bitter clash in which people invoke values they treat as non-negotiable? It is this willingness in Douglas’s work to address processes that make for intransigence that provides resources and inspiration for all who are interested in how the most severe conflicts might be channelled and contained, and how the self-radicalising collective effervescence of sacred contagion might be dampened. Debate via social media will not save the world or protect our lives. We risk drowning in a tsunami from a media earthquake of incompatible opinions. There is now nothing for it but to go back to basics, coordinate activities afresh and seek opportunities for re-institutionalisation. Fresh analogies and new or adapted rituals of an everyday and informal kind will almost certainly be needed much more than new grand formal public ceremonials. This performative work requires creative flair for ritual processes applicable to conversations, meetings, gatherings, casual encounters and online, at the level examined by Goffman (Collins 2004, and for West African ethnographic examples, see Fardon 1988). Many social scientists have focused on everyday ritual interaction as a process of sustaining emotions and collective energy for bonding of groups (Collins 2008), but Douglas appreciated very clearly that this could also be a process leading to sacred contagion of a destructive kind. Therefore, she argued that we need to focus more on how rituals classify and less on how they bond. This requires us to understand ritual as a type of communicative dynamics capable of dampening and attenuating as well as amplifying conflict. Douglas first developed her theory of institutions in a world going through decolonisation. Debates about decolonisation are not yet laid to rest. This might seem strange until it is realised how differently institutions have developed in the different parts of the globe. History matters. Not all institutions are raised equal. They are marked and sometimes scarred with the legacies of long

126

MARY DOUGLAS

histories of chequered social interaction, including exploitative trading relations, brutal wars and forced migrations. Appreciation of, and tolerance for, institutional differences require attention to this chequered history of institutional development. Consider the African case (which motivated Douglas throughout her long academic life). Colonial institutionalisation strongly shaped understandings that life was arranged to benefit the colonial power. Typically, transport networks ‘drained’ countries towards a coastal capital, generally also the largest port; getting around internally was often harder than travelling internationally because of the way connectivity was arranged. Sanitation rules similarly projected a message that the colony was a servant of the metropolis; white rulers needed protection from the diseases of the common people. One of the first results of Ronald Ross’s work in Sierra Leone on the mosquito transmission cycle for malaria was that the European colonial officers were removed to a segregated cantonment (Hill Station) with a special suburban railway built for their commute to work (Spitzer 1974: 39–69). This isolated them from the source of the infection – by which the colonial officers meant the Africans they administered. Drained of resources, African colonial territories were then restored to independence in the 1950s and 1960s. Impoverished but independent regimes were offered aid to address public health problems. But postcolonial assistance was never enough to benefit all. International support for public health, for example, would typically be focused on the under-fives. This causes the adult citizens of these postcolonial countries today to wonder whether, somehow, their own health is of lesser importance. Institutionalisation under cramped conditions often takes on an enclaved, self-reliant character; in that case, as Douglas’s theory predicts, wild fears of conspiracies take root. During the pandemic of Covid-19, fieldwork in Sierra Leone encountered two responses among those expressing vaccine hesitation (Mokuwa 2022). The first is the notion that colonial powers (and now international agencies) were only ever worried about African demographic success, fearing this would reshape global political power. This

CONCLUSION

127

feeds the suspicion that vaccination is enforced population control. A second adverse response starts with the observation that vaccination is only something for children; so why now is it necessary for adults? Something dark must be afoot. Both responses reinforce tendencies towards enclaved institutionalisation. To the medically trained, these responses are seen as ‘not following the science’. They are, to put it bluntly, viewed as products of ignorance. Douglas cautions us to take another view. Maybe these local responses make perfect sense as the product of thinking shaped by strong historical restrictions of particular strengths and weakness of grid and group? The requisite response is not to put on a set of lectures on vaccine science but to see what can be done to loosen the conditions and constraints under which local institutional formation takes place. That will require a much more generous approach to improving the livelihoods of poor people globally. Failure to have addressed the question, Douglas tells us in no uncertain terms, is a global scandal but one that is perpetuated because the wrong issues are being addressed. Too much attention is paid to finding and imposing the ‘right’ set of institutions; not enough attention is paid to fostering the social and personal freedoms under which effective institutional accommodation takes place. Freedom for institutional development is something that Douglas’s scheme of grid and group seeks to map and measure, with a view to charting a way round the storms of ‘sacred contagion’ and arriving at calmer zones in which productive institutional accommodation can be fostered and thrive. But Douglas does not aspire to give an engineer’s guide to institutional improvement. This is because successful modifications must be naturalised; the process is fraught or impossible if not backed by local processes of institutionalisation. Thus, her approach is more that of a drama critic or music analyst, observing performances and pointing out which patterns, devices and gestures work and which do not. Douglas has several pieces of general advice. The first is not to engage in a dialogue of the deaf. You will never convince the base, because the base – especially where it is deeply enclaved and al-

128

MARY DOUGLAS

ready in positive feedback in that form of ordering – has naturalised its institutional moorings. Its own point of view seems just obvious, and to insist otherwise is a threat. To challenge ideas with ideas is a waste of time, institutionally speaking, since ideas are the product of social life, not its cause. For in doing so, one is not only seeking to defend an opinion but indirectly asking for the dismantling of the social life supporting the idea one disputes. The process is also dangerous. Asking people to change their ideas causes them to double down, risking the triggering of sacred contagion. Both sides have fortified their institutional choices. There is no scope here for debate. Argument will only get louder. A fight, a riot or a massacre might ensue. Should we insist on new rules for more civil debate? Keeping to rules for debate might sound sensible until you start. Who will enforce them, and how? Regulating the internet is a fool’s errand because it is so limitlessly vast. Stubborn holdouts simply move off to set up new spaces in which their bubbles endlessly proliferate. Douglas thought it might make more sense to focus on mapping the spaces in which different groups have room to manoeuvre, and then seeking to loosen the constraints on those who were, institutionally speaking, most hemmed in. In other words, truly global convergence among institutions is inhibited by some groups suffering over-regulation or lack of integration, and by other groups stigmatising them for the institutional bonds they have developed. It helps no one to be accused of apathy or fatalism, for example. Greater freedom to institutionalise makes more sense than trying to impose competitive or meritocratic solutions on problems for which these institutional orderings are not well suited. Who in their right mind would think that competing claims over the Holy Places in Jerusalem could be resolved by offering a management contract to the highest bidder? The hardest problem is to develop options for cohabitation with people with whom we profoundly disagree. Too often development consultants have read North (1990, 1992), and Acemoglu and Robinson (2012) and concluded that the right prescription would be simply to copy the best institutions and open these to

CONCLUSION

129

all. This is unlikely to work, Douglas suggests, because people living under different institutions will disagree on what constitutes the best institutional models to import, and these institutional innovations need to settle within a complex natural landscape of other institutions. Predicting interaction effects is beyond calculation. Missing also is any consideration of how an institution becomes accommodated to specific social forms of life. This is equivalent to the problem encountered in biology when too much attention is paid to a life form and not enough to the context in which it lives. A lesson she advocates in these circumstances is that low-key ritual interventions are sometimes the best. Cut down on the flag waving. A religion based on personal cleanliness rather than one prostrating before representations of deities may turn out to be more usefully inclusive. Purification rituals that underscore the boundaries of the sect must be ritually edited over time into collective acts that reinforce a common life and shared responsibility, rather than blame those classified as impure when the problem is that we are all impure. Institutions that cultivate silence or allow for avoidance may be what we need to manage the adverse consequences of a social media storm (or any other deep-rooted stand-off among groups with differing views). In short, her advice is that when all else fails pipe down and give quietness a chance. Douglas’s most important intellectual contribution to debates about institutional dynamics is her ‘Y is a function of X, serving to strengthen Z’ model of institutional functionality. She was an anthropologist and so worked hard to bring into the global picture marginal or neglected communities, but she also thought that anthropology belonged to the sciences – it produced insights of general importance. It would make a fitting end to our short book to give a brief worked example relevant to current global concerns. What might she have said about the institutional dimensions of our current global predicament – the pandemic of a novel coronavirus, Sars Cov-2, and the sickness it causes, Covid-19? Of course, at this point we can only offer a vignette; a full Douglasian analysis of the pandemic would take a book (or more). But

130

MARY DOUGLAS

our picture is more than a speculative sketch, since Douglas had already outlined the shape of the argument with which she might have met the pandemic in How Institutions Think (Douglas 1986: 99–101). She asks whether the modern generation is the first to escape ‘old … institutional controls’, the first ‘to come face to face with one another as real individuals’, and thus ‘the first to achieve full self-consciousness’. To test this idea, she proposed a thought experiment. ‘To know how to resist the classifying pressure of our institutions we would like to start a classificatory exercise.’ She shows how hard this is to do from scratch. ‘Unfortunately, all the classifications that we have for thinking with are provided ready-made, along with our social life.’ In each case, we revert to established schemes such as ‘the division of labour in the home’ or categories ‘used by our administrators for collecting taxes, making population censuses, or estimating the need for schools or prisons’. ‘Our minds’, she remarked, ‘are running on the old treadmill already’ (Douglas 1986: 99). At the same time as institutions produce labels, ‘there is feedback of Robert Merton’s self-fulfilling kind’, in which labels both ‘stabilise the flux of life and create to some extent the realities to which they apply’. Acknowledging the philosophers Michel Foucault (1970) and Ian Hacking (2002 [1986]), she notes that labelling is a dynamic process. In particular, she endorses Hacking’s argument that ‘people are not merely labelled and newly made prominent, still behaving as they would whether so labelled or not. The new people behave differently than they ever did before’ (Douglas 1986.: 100). They live up to (or are crushed by) the label they bear (as a knight of the realm, a person placed on the autism spectrum, or whatever). She goes further than Hacking (2002 [1986]), however. Hacking draws a difference between people and things, arguing that ‘what camels, mountains and microbes are doing does not depend on words’. Douglas agrees that ‘a course of injections can kill microbes’; the possibilities are delimited by nature not words. But she then adds two further indisputable points. A course of the wrong injections could also kill people, ‘nor are the microbes

CONCLUSION

131

less responsive to words than humans’ (Douglas 1986: 101). If there is any doubt on this second point, consider the labelling of the Sars Cov-2 virus, in which variant names have become a commonplace of our everyday conversation: ‘This game is cancelled due to risks of coronavirus.’ ‘Work from home because of the spread of the delta variant.’ ‘Have you had omicron yet?’ Douglas’s next move seems eerily prescient of our present predicament. The plants and animals and microbes respond even more vehemently [to their labelling] than humans. The individual bacillus may die, true enough, but in a very short space of time new breeds have emerged, not to conform to the labels but to defy them, millions of new bacilli appear, never imagined before, but immune to the attacks mounted against them under the old labels. (p. 101) It is almost as if the Sars Cov-2 virus was determined to prove us wrong by evolving. Her point is not that microbes have social agency but that ‘the labelling process in both cases is part of a larger constraining action’ (p. 101). That larger constraining action is how we shape and maintain our institutions in a world in which change is endemic. The world changes of its own accord, but we supply the labels. The labels reflect our working arrangements, and this is as true of science as of any other institutions based on social cooperation. So, we need to become very aware of how we use labels. At times, we fall behind the game. One of us had the privilege to attend a brilliant lecture by the eminent German virologist Professor Christian Drosten in the first days of the pandemic. Drosten’s lab had created the first test for the novel virus in a matter of days. The lecture was well attended. The pandemic was big news, and the German press corps had turned up in numbers. Drosten began with a warning to the journalists present to be clear about the topic. Virologists had strict rules for naming new viruses, and the virus, properly, was Sars Cov-2. The World Health Organization (WHO), however, had come up with the label ‘Covid-19’, naming not the virus but

132

MARY DOUGLAS

the disease it causes. Drosten and his colleagues are scientists; the WHO is run by medical doctors. Different working arrangements resulted in different labels.  Faced with two labels, and two distinct sets of institutional working practices, neither of which meant much to ordinary folk, many of the public later opted for ‘coronavirus’ or just ‘corona’ (despite the fact that there are several coronaviruses, including those which cause common colds). Those who were determined to resist (as they imagine) all enforced institutionalisation preferred to deny the existence of both the virus and disease altogether, to their undoubted cost when infected. Labels matter – becoming confused about categories, or ignoring them, can be deadly – but what they tell us about is working practices, not the nature of things in themselves. So, Douglas’s main point still stands. The categories through which we try to apprehend the world are produced by our institutions. We can be caught out by change in the larger world to which we have been blinded by the ways in which our institutional categories govern our thoughts. Given the capacity of our institutions to fool us, it pays to be on our guard to detect institutional biases. Douglas’s grid-group scheme gives us a basis to start. It suggests that over-confidence in vaccines alone as the solution to the pandemic might stem from the institutional biases of interlocking global hierarchies of business, science and medicine. The capacity of the Sars Cov-2 virus to escape the labels within which we attempted to imprison it reflects a complacent underestimation (by politicians, in particular) of the scope of viral evolutionary processes, and as a result, vaccine development must run much harder than once anticipated. Enclave and isolate institutionalisation in regions and among groups needing vaccine protection fosters conspiracy theories, but escape does not require the services of a witchfinder so much as scope to address the conditions of material deprivation channelling these adverse institutional categorisations. These are the main lines along which Douglas thought we might release ourselves from institutional prisons of our own devising.

APPENDIX SELECTED WORKS BY MARY DOUGLAS NOT DISCUSSED IN THE TEXT

••• M

ary Douglas’s oeuvre is vast and her productivity unrivalled. The limits of this short book prevent us from examining more than some key writings. Here we list a selection of some of the most important of her other writings we were unable to make space for in the text, together with some notes on their interest and significance. A full listing of her writings up to the year 1999 can be found at the end of Fardon (1999).1

MONOGRAPHS, EDITED VOLUMES, ARTICLES AND CHAPTERS Douglas, Mary. (ed.). 1970. Witchcraft Confessions and Accusations. London: Tavistock. – Douglas worked with anthropologists and historians to examine accusation and blame, in the wake of Evans-Pritchard’s work on the Azande. Her introduction focuses on institutions that shape classification in blame and, in some cases, reintegration. Douglas, Mary. (ed.). 1973. Rules and Meanings: The Anthropology of Everyday Knowledge. Harmondsworth: Penguin. – A pivotal collection of excerpts from leading twentieth-century philosophers, sociologists, anthropologists, composers and educational sociologists, all of which shaped her own thought and teaching. The collection shows the ways in which institutional rules of

134

APPENDIX

social organisation shape and define meaning in language, classification and what Douglas would later (How Institutions Think) call styles of thought. Douglas used it with her advanced students; it must have been a fascinating if demanding course. Douglas, Mary. 1980. Evans-Pritchard: His Life, Work, Writings and Ideas. Brighton: Harvester Wheatsheaf with Fontana. – An overview of Evans-Pritchard’s thought. Of particular importance for the development of her own thought is her chapter on time, memory and forgetting, and the two chapters comparing social ‘accountability’ systems (institutions) through EvansPritchard’s analysis of Azande and Nuer life. Douglas, Mary. 1984. ‘The Social Preconditions of Radical Scepticism’, Sociological Review 32(1): supplement, 68–87. – One of a series of articles written in response to the rise of postmodernism, in which Douglas considers historical periods in which scepticism, doubt, uncertainty and rejection of substantive claims and even aspirations for objectivity have arisen. Her work seeks to explain these trends by the changing forms of social organisation that cultivate this style of thought. Douglas, Mary. 1985. ‘Pascal’s Great Wager’, L’Homme 93, 25, 1: 13–30. Reprinted under the title, ‘Credibility’, in Douglas, Mary. (ed.). 1992. Risk and Blame. London: Routledge, pp. 235–54. – Another paper on scepticism and uncertainty, this time focusing on the intermediate condition between doubt and certainty, which is the effort to wager on probabilities, and again seeking to explain this thought style by the institutional ordering under which it emerged. Like the 1984 ‘Radical Scepticism’ paper, it examines the institutional situation of intellectual groups without political power and the thought style cultivated under their anomalous institutional position. Douglas, Mary. (ed.). 1987. Constructive Drinking. London: Routledge. – An extension to the context of drinking alcohol, tea, coffee etc., of the argument in Douglas and Isherwood (1979) that consumption is a ritual process driven by social relations and the distinct periodicities of different forms of social organisation.

APPENDIX

135

Douglas, Mary. 1989. ‘The Background of the Grid Dimension: A Comment’. Sociological Analysis 50(2): 171–76. – A response to a critique of her accounts of the dimension by James Spickard in the same journal. Douglas, Mary. 1990. ‘The Body of the World’, International Social Science Journal 42(3): 395–99. – Seeks to explain why some social orders do and others do not cultivate the use of microcosms as part of a larger argument about the social process that stylises human concepts of space. Rejecting explanations of the decline of microcosms that point to urbanisation, Douglas argues that their decline is a response to greater individualistic ordering of social relations. Douglas, Mary. 1991. ‘The Idea of a Home: A Kind of Space’, Social Research 58(1): 287–307. – Analysis of institutional ordering that transforms material housing into a space for home living and that rests on a series of prohibitions that Douglas describes as ‘tyrannical’ (!). Introduces a concept of proto-hierarchy, reflecting the self-organising character of the institutional system of a home. Douglas, Mary. 1993. ‘Emotion and Culture in Theories of Justice’, Economy and Society 22 (4): 501–55 [reprinted in Fardon 2013a: 20–35]. – A critique of philosophical attempts such as those of Rawls (1970, 1993) to understand social justice by reference to emotions and preferences taken as given and universal, without attending to the range of variation in the forms of social ordering that shape those emotions and preferences. Douglas, Mary. 1995. ‘The Language of the Emotions in the Social Sciences’, Greek Economic Review 17(2): 167–76 [reprinted in Fardon 2013a: 11–19]. – A critique of social scientists and psychologists who use emotions to explain action or decision-making. Douglas points out that emotions are things we are accused of feeling, things we claim to feel in order to justify our claims, and also things that we learn to feel because of our location in (often anomalous) social orderings. It is the emotions and our use of them that needs explaining first.

136

APPENDIX

Douglas, Mary. 1995. ‘The Cloud God and the Shadow Self ’, Social Anthropology 3(2): 83–94 [reprinted in Fardon 2013b: 120–34]. – An extended treatment of the concept of personhood, using more ethnographic evidence than appears in Douglas and Ney (1998) and also drawing on Douglas’s 1990s work on ancient Israel but using Elinor Ostrom’s account of institutions to explain variety in forms of personhood. Douglas, Mary. 2000. ‘The Risks of the Risk Officer’. Revised version of Douglas, Mary. 1999. ‘Les risques du fonctionnaire du risque: la diversité des institutions et la répartition des risques.’ Alliage 40: 61–74 [reprinted in Fardon 2013a: 201–15]. – One of her final interventions in risk debates, using a reformulation of her typology by the political scientist and regulation scholar Christopher Hood, to argue that the anomalies, dilemmas and limitations of practical risk management are only explicable when we recognise the clashing institutional dynamics at stake.

LATE OVERVIEWS OF DOUGLAS’S TYPOLOGICAL THEORY: PRESENTATIONS RATHER THAN NEW RESEARCH Douglas, Mary. 2006. ‘A History of Grid and Group Cultural Theory’. Available at https://dokument.pub/a-history-of-grid-and-group-culturaltheory-flipbook-pdf.html and also at https://dokumen.tips/docume nts/mary-douglas-history-of-grid-and-group-cultural-theory.html? page=1. – Includes some important rethinking of her account of enclaved ordering, drawing on work by Emmanuel Sivan and also by Dipak Gyawali, scholars in very different fields who drew upon her theory. Douglas, Mary. 2006. ‘Seeing Everything in Black and White’. Available at http://communitas.co.za/wp-content/uploads/2011/08/dougl as2.pdf. – Similar to ‘History of G&G CT’ above but with more application to rival certainties. Douglas, Mary. 2005. ‘Grid and Group, New Developments’. Workshop on complexity and cultural theory in honour of Michael Thomp-

APPENDIX

137

son, London School of Economics, 27 June 2005. Available at http:// emk-complexity.s3.amazonaws.com/events/2005/MaryDouglas.pdf. – Returns to Bernstein as the point of origin for the theory. Also discusses revised understanding of enclaving following Sivan’s and Gyawali’s work. Douglas, Mary. Bias and Solidarity, German Sociological Association, Congress on ‘Nature and Society’. 2006. – Reflections on the universality of bias, in particular about ‘nature’ whether in respect of the climate or sexuality and its roots in forms of solidarity, none of which should be romanticised. Argues for analysis of ‘repleteness’ (Goodman 1978) or perfect fit (see Douglas 1992c, ‘Rightness of Categories’) as socially driven but also suspect and unstable.

INTERVIEWS WITH MARY DOUGLAS Gosden, Chris. 2004. ‘Grid and Group: An Interview with Mary Douglas’, Journal of Social Archaeology 4(3): 275–87. – Widely ranging interview, from an archaeological perspective. Concludes with discussion about Douglas’s late interest in music and the ritual basis for musical rhythm. Interview with Mary Douglas by Alan Macfarlane, 2006. Video at https://www.youtube.com/watch?v=xl3oMdIRFDs. – An hour-long interview, very widely ranging. Some discussion about religion. See also interviews in Fardon (2013b).

NOTE 1. Initially, we had assumed that Richard Fardon’s listing was based on Douglas’s curriculum vitae. He has explained to us that Mary Douglas had better things to do with her time than to update lists (ironic for a scholar focused on the social meaning of classification!) and that the CV was full of gaps and errors. So, in an era before electronic bibliography and online search engines, he had to undertake the task of going

138

APPENDIX

through journals by hand, to search for her papers, then collate references in her own work and the work of others. His list up to 1999 (when his own intellectual biography of Douglas was published) is more or less complete, though it needs extension, since she was productive to the end of her life (in 2007). This extension is partly catered for by Fardon’s two volumes of previously unpublished and otherwise hard to access items (Fardon 2013a, 2013b), including a good number of her final papers. Richard Fardon’s efforts in bringing her large and scattered output into good order is a significant scholarly service to all those interested in her work and deserves the widest public acknowledgement. Our task in our own books on Douglas’s work would have been nearly impossible without it.

WORKS BY MARY DOUGLAS DISCUSSED IN THE TEXT

••• Douglas, Mary. 1962. ‘Lele Economy Compared with the Bushong: A Study of Economic Backwardness’, in Paul Bohannan and George Dalton (eds), Markets in Africa. Evanston, IL: Northwestern University Press, pp. 211–33. ———. 1964. ‘Matriliny and Pawnship in Central Africa’, Africa 34(4): 3–113. ———. 1966. Purity and Danger: An Analysis of Concepts of Pollution and Taboo. London: Routledge. ———. 1969. ‘Is Matriliny Doomed in Africa?’ in Mary Douglas and Phyllis Kaberry (eds), Man in Africa. London: Tavistock, pp. 121–37. ———. 1972a. ‘Self-evidence’, in Mary Douglas. 1999 [1975]. Implicit Meanings: Selected Essays in Anthropology, 2nd edn. London: Routledge, pp. 252–83. ———. 1972b. ‘Deciphering a Meal’, Daedalus 101(1): 61–82 [reprinted in Douglas 1999d: 231–51]. ———. 1973 [1970]. Natural Symbols: Explorations in Cosmology. London: Barrie and Rockcliff. ———. 1977 [1963]. The Lele of the Kasai. London: Oxford University Press for the International African Institute. ———. 1978. Cultural Bias. London: Royal Anthropological Institute [reprinted in Douglas 1982a: 183–254]. ———. 1980. Evans-Pritchard: His Life, Work, Writings and Ideas. Brighton: Harvester Wheatsheaf with Fontana. ———. 1981. ‘Good Taste: Review of Pierre Bourdieu, La distinction’, Times Literary Supplement, 13 February 1981, 163–69 [reprinted in Douglas 1982a]. ———. 1982a. In the Active Voice. London: Routledge and Kegan Paul.

140

WORKS BY MARY DOUGLAS DISCUSSED IN THE TEXT

———. (ed.). 1982b. Essays in the Sociology of Perception. London: Routledge and Kegan Paul. ———. 1985. Risk Acceptability According to the Social Sciences. London: Routledge. ———. 1986. How Institutions Think. Syracuse, NY: University of Syracuse Press. ———. 1988. ‘The Problem of Evil Among the Lele: Sorcery, Witchhunt and Christian Teaching in Africa’ [reprinted in Fardon 2013b: 79–94]. ———. 1990. ‘Foreword: No Free Gifts’, in Marcel Mauss, The Gift: The Form and Reason for Exchange in Archaic Societies, trans. W.D. Halls. New York: W.W. Norton and Co., pp. vii–xviii. ———. 1992a. Risk and Blame: Essays in Cultural Theory. London: Routledge. ———. 1992b. ‘Witchcraft and Leprosy: Two Strategies for Rejection’, in Risk and Blame: Essays in Cultural Theory. London: Routledge, pp. 83–101. ———. 1992c. ‘Rightness of Categories’, in M. Douglas and D. Hull (eds), How Classification Works: Nelson Goodman Among the Social Sciences. Edinburgh: Edinburgh University Press, 239–71 [reprinted in Douglas 1999d: 284–309]. ———. 1995a. ‘Institutions: Problems of Theory’. Original in English, first publication in Italian, first published in English in Perri 6 and Gerald Mars (eds). 2008. The Institutional Dynamics of Culture, vol. 1. Aldershot: Ashgate [reprinted in Fardon 2013a: 36–52]. ———. 1995b. ‘To Honour the Dead’, in Geoff Dench, Tony Flower and Kate Gavron (eds), Young at Eighty: The Prolific Public Life of Michael Young. Manchester: Carcanet Press, pp. 209–15 [reprinted in Fardon 2013b: 221–27]. ———. 1996. Thought Styles: Critical Essays on Good Taste. London: Sage. ———. 1999a. Leviticus as Literature. Oxford: Oxford University Press. ———. 1999b. ‘Sorcery Accusations Unleashed: The Lele Revisited, 1987’, Africa 69(2): 177–93. ———. 1999c. ‘Four Cultures: The Evolution of a Parsimonious Model’, GeoJournal 47(3): 411–15 [reprinted in Fardon 2013a: 53–62]. ———. 1999d. Implicit Meanings: Selected Essays in Anthropology, 2nd edn. London: Routledge. ———. 2001 [1993]. In the Wilderness: The Doctrine of Defilement in the Book of Numbers. Oxford: Oxford University Press.

WORKS BY MARY DOUGLAS DISCUSSED IN THE TEXT

141

———. 2001. ‘Dealing with Uncertainty’, Mulatuli Lecture 2001, Ethical Perspectives 8(3): 145–55. ———. 2004a. ‘Traditional Culture: Let’s Hear No More About It’, in V. Rao and M. Walton (eds), Culture and Public Action. Stanford, CA: Stanford University Press, pp. 85–108 [reprinted in Fardon 2013a: 284–308]. ———. 2004b. Jacob’s Tears: The Priestly Work of Reconciliation. Oxford: Oxford University Press. ———. 2005. ‘A Feeling for Hierarchy’, in James Heft (ed.), Believing Scholars: Ten Catholic Intellectuals. New York: Fordham University [reprinted in Fardon 2013b: 15–36]. ———. 2007. Thinking in Circles: An Essay on Ring Composition. New Haven: Yale University Press. ———. 2008. ‘Raisonnements circulaires: retour nostalgique à LévyBruhl’, Sociological Research Online 12: 6–12 [orig. manuscript, Mary Douglas. ‘Nostalgia for Lévy-Bruhl: Reasoning in Circles’, lecture at the Collège de France, 4th April 2001]. Douglas, Mary, and Baron Isherwood. 1979. The World of Goods: Towards an Anthropology of Consumption. London: Routledge. Douglas, Mary, and Aaron Wildavsky. 1982. Risk and Culture: An Essay on the Selection of Technological and Environmental Dangers. Berkeley and Los Angeles: University of California Press. Douglas, Mary, and David Hull (eds). 1993. How Classification Works: Nelson Goodman among the Social Sciences. Edinburgh: Edinburgh University Press. Douglas, Mary, and Steven Ney. 1998. Missing Persons: A Critique of Personhood in the Social Sciences. Berkeley: University of California Press. Douglas, Mary, Des Gasper, Steven Ney and Michael Thompson. 1998. ‘Human Needs and Wants’, in Steve Rayner and Elizabeth L. Malone (eds), Human Choice and Climate Change. Vol 1: The Societal Framework. Columbus, Ohio: Battelle Institute, pp. 195–254 [reprinted in Fardon 2013a: 65–135]. Douglas, Mary, and Gerald Mars. 2003. ‘Terrorism: A Positive Feedback Game’, Human Relations 56(7): 763–86. Douglas, Mary, Michael Thompson and Marco Verweij. 2003. ‘Is Time Running Out? The Case of Global Warming’, Daedalus 132(2): 98– 107 [reprinted in Fardon 2013a: 134–45].

REFERENCES

••• 6, Perri. 2003. ‘Institutional Viability: A Neo-Durkheimian Theory’, Innovation: The European Journal of Social Science Research 16(4): 395–416. ———. 2007. ‘Sense and Solidarities: Politics and Human Well-being’, in John Haworth and Graham Hart (eds), Well-being: Individual, Community and Social Perspectives. Basingstoke: Palgrave Macmillan, pp. 127–48. ———. 2011. Explaining Political Judgement. Cambridge: Cambridge University Press. ———. 2014a. ‘Explaining Decision-Making in Government: The NeoDurkheimian Institutional Framework’, Public Administration 92(1): 87–103. ———. 2014b. ‘Explaining Unintended and Unexpected Consequences of Policy Decisions: Comparing Three British Governments, 1959– 74’, Public Administration 92(3): 673–91. 6, Perri, and Paul Richards. 2017. Mary Douglas: Understanding Social Thought and Conflict. Oxford and New York: Berghahn. Acemoglu, Daron, and James A. Robinson. 2012. Why Nations Fail: The Origins of Power, Prosperity and Poverty. London: Profile. Ackerman, Bruce. 2000. We the People, Vol II: Transformations. Cambridge, MA: Belknap Press of the Harvard University Press. Atran, Scott. 1993. Cognitive Foundations of Natural History: Towards an Anthropology of Science. Cambridge: Cambridge University Press. Austin, John Langshaw. 1975 [1962]. How To Do Things With Words, 2nd edn. Marina Sbisà and James Opie Urmson (eds). Oxford: Oxford University Press. Banfield, Edward. 1958. The Moral Basis of a Backward Society. New York: Free Press. Bernstein, Basil. 1971. Class, Codes and Control. Vol 1: Theoretical Studies Towards a Sociology of Language. London: Routledge and Kegan Paul.

REFERENCES

143

———. 1975. Class, Codes and Control. Vol 3: Towards a Theory of Educational Transmissions. London: Routledge and Kegan Paul. Bernstein, Peter. 1996. Against the Gods: The Remarkable Story of Risk. New York: Wiley. Bloor, David. 1982. ‘Polyhedra and the Abominations of Leviticus: Cognitive Styles in Mathematics’, in Mary Douglas (ed.), Essays in the Sociology of Perception. London: Routledge, pp. 191–218. ———. 1991 [1976]. Knowledge and Social Imagery, 2nd edn. Chicago, IL: University of Chicago Press. Bošković, Aleksandar. 2021. William Robertson Smith. Berghahn: New York and Oxford. Bourdieu, Pierre. 1970. ‘The Berber House or the World Reversed’, Social Science Information 9(1): 151–70. ———. 1984 [1979]. Distinction: A Social Critique of the Judgment of Taste, trans. by R. Nice. Cambridge, MA: Harvard University Press. Burton, John (ed.). 1990. Conflict: Human Needs Theory. London: Macmillan. Collins, Randall. 2004. Interaction Ritual Chains. Princeton, NJ: Princeton University Press. ———. 2008. Violence: A Microsociological Theory. Princeton, NJ: Princeton University Press. Coyle, Dennis J. 1994. ‘The Theory That Would Be King’, in Dennis J. Coyle and Richard J. Ellis (eds), Politics, Policy and Culture. Boulder, CO: Westview Press, pp. 217–39. Coyle, Dennis J., and Richard J. Ellis (eds). 1994. Politics, Policy and Culture. Boulder, CO: Westview Press. Dake, Karl, and Michael Thompson. 1993. ‘The Meanings of Sustainable Development: Household Strategies for Managing Needs and Resources’, in Scott D. Wright, T. Dietz, R. Borden, G. Young and G. Guagnano (eds), Human Ecology: Crossing Boundaries. Fort Collins, CO: Society for Human Ecology, pp. 421–36. ———. 1999. ‘Making Ends Meet, in the Household and on the Planet’, GeoJournal 47(4): 417–24. Dant, Tim. 2010. ‘The Work of Repair: Gesture, Emotion and Sensual Knowledge’, Sociological Research Online 15(3): 97–118. De Soto, Clinton. 1960. ‘Learning a Social Structure’, The Journal of Abnormal and Social Psychology 60(3): 417–21. Dinler, Demet. 2019. ‘Market, Morality and ( Just) Price: The Case of the Recycling Economy in Turkey’, Research in Economic Anthropology 39(1): 27–47.

144

REFERENCES

Dumont, Louis. 1980 [1966]. Homo Hierarchicus: The Caste System and its Implications. Chicago, IL: University of Chicago Press. ———. 1986 [1983]. Essays on Individualism: Modern Ideology in Anthropological Perspective. Chicago, IL: University of Chicago Press. Durkheim, Émile. 1951 [1897]. Suicide: A Sociological Study, trans. John Spaulding and George Simpson. New York: Free Press. ———. 1984 [1893]. The Division of Labour in Society, trans. by W.D. Halls. London: Macmillan. ———. 1995 [1912]. The Elementary Forms of Religious Life, trans. Karen E. Fields. New York: Free Press. Durkheim, Émile, and Marcel Mauss. 1963 [1902–3]. Primitive Classification, trans. R. Needham. Chicago, IL: University of Chicago Press [original publication: ‘De quelques forms primitives de classification: contribution à l’étude des representations collectives’, Année sociologique, VI (1901–2), 1903, 1–72]. Elster, Jon. 1983. Explaining Technical Change: A Case Study in the Philosophy of Science. Cambridge: Cambridge University Press. Evans-Pritchard, Edward. 1937. Witchcraft, Oracles, and Magic Among the Azande. Oxford: Clarendon Press. ———. 1940. The Nuer: A Description of the Mode of Livelihood and Political Institutions of a Nilotic People. Oxford: Clarendon Press. ———. 1951. Kinship and Marriage among the Nuer. Oxford: Clarendon Press. ———. 1956. Nuer Religion. Oxford: Clarendon Press. Fardon, Richard. 1988. Raiders and Refugees: Trends in Chamba Political Development, 1750 to 1950. Washington: Smithsonian Institution Press. ———. 1999. Mary Douglas: An Intellectual Biography. London and New York: Routledge. ———. 2013a. Mary Douglas: Culture and Crises: Understanding Risk and Resolution. London: Sage. ———. 2013b. Mary Douglas: A Very Personal Method: Anthropological Writings Drawn from Life. London: Sage. ———. 2013c. ‘Citations out of Place’, Anthropology Today 29(1): 25–27. Feyerabend, Paul. 1975. Against Method, 3rd edn. London: Verso. Fix, Andrew Cooper. 1991. Prophecy and Reason: The Dutch Collegiants in the Early Enlightenment. Princeton, NJ: Princeton University Press. Fleck, Ludwik. 1979 [1935]. Genesis and Development of a Scientific Fact, trans. Fred Bradley and Thaddeus J. Trenn. Chicago: Chicago University Press.

REFERENCES

145

Fortes, Meyer. 1945. The Dynamics of Clanship among the Tallensi: Being the First Part of an Analysis of the Social Structure of a Trans-Volta Tribe. London: Oxford University Press for the International African Institute. ———. 1949. The Web of Kinship among the Tallensi: The Second Part of an Analysis of the Social Structure of a Trans-Volta Tribe. London: Oxford University Press for the International African Institute. Foucault, Michel. 1970. The Order of Things: An Archaeology of the Human Sciences. New York: Pantheon. Fukuyama, Francis. 2011. The Origins of Political Order: From Prehuman Times to the French Revolution. London: Profile. Goffman, Erving. 1963. Stigma: Notes on the Management of Spoiled Identity. Harmondsworth: Penguin. ———. 1967. Interaction Ritual: Essays on Face-to-face Behaviour. New York: Pantheon. Goodman, Nelson. 1976 [1968]. Languages of Art: An Approach to a Theory of Symbols. New York: Hackett. ———. 1978. Ways of World Making. Indianapolis, IN: Hackett. Hacking, Ian. 1975. The Emergence of Probability: A Philosophical Study of Early Ideas About Probability, Induction and Statistical Inference. Cambridge: Cambridge University Press. ———. 2002. ‘Making Up People’, in Ian Hacking, Historical Ontology. Cambridge, MA: Harvard University Press, pp. 99–114 [Originally in Thomas C. Heller, Morton Sosna and David E. Wellbery (eds). 1986. Reconstructing Individualism: Autonomy, Individuality, and the Self in Western Thought. Stanford, CA: Stanford University Press, pp. 161–71]. Heald, Suzette, et al. 2004. ‘Book Review Forum on Mary Douglas’s Leviticus as Literature’, Journal of Ritual Studies 18(2): 152–91. Heclo, Hugh, and Aaron Wildavsky. 1981 [1974]. The Private Government of Public Money, 2nd edn. London: Macmillan. Hedström, Peter. 1998. ‘Rational Imitation’, in Peter Hedström and Richard Swedberg (eds), Social Mechanisms: An Analytical Approach to Social Theory. Cambridge: Cambridge University Press, pp. 306–28. Hubert, Henri. 1999 [1905]. An Essay on Time: A Brief Study of the Representation of Time in Religion and Magic. Oxford: Durkheim Press. Jenkins, Jack, Esther Yei Mokuwa, Krijn Peters and Paul Richards. 2021. ‘Rural-Urban Connectivity Strengthens Agrarian Peace: Evidence from a Study of Gender and Motorcycle Taxis in Sierra Leone’, Jour-

146

REFERENCES

nal of Agrarian Change. Retrieved from https://doi.org/10.1111/ joac.12423. Kaldor, Mary. 2012. New and Old Wars: Organized Violence in a Global Era. Cambridge: Polity. Kamara, Foday, Gelejimah Alfred Mokuwa and Paul Richards. 2022. ‘Keeping Ebola at Bay: Public Authority and Ceremonial Competence in Rural Sierra Leone’, Journal of Modern African Studies 60(1): 1–20. Kant, Immanuel. 1933 [1787]. Critique of Pure Reason, 2nd edn, trans. Norman Kemp Smith. London and Basingstoke: Macmillan. Kuhn, Thomas S. 1970 [1962]. The Structure of Scientific Revolutions, 2nd edn. Chicago, IL: University of Chicago Press. ———. 1974. ‘Second Thoughts on Paradigms’, in Frederick Suppes (ed.), The Structure of Scientific Theories. Urbana, IL: University of Illinois Press, pp. 459–82. Lakatos, Imre. 1970. ‘Falsification and the Methodology of Scientific Research Programmes’, in Imre Lakatos and Alan Musgrave (eds), Criticism and the Growth of Knowledge. Cambridge: Cambridge University Press, 91–196. ———. 1976. Proofs and Refutations: The Logic of Mathematical Discovery. Cambridge: Cambridge University Press. Lynn, Charles, Marjorie Lynn and Sylvia Lynn. 2013. The Long Garden Master: Life and Times of a Colonial Agricultural Officer in the Gold Coast, 1929–1947. Authors OnLine Book. Marchal, Jules. 2008 [2001]. Lord Leverhulme’s Ghosts: Colonial Exploitation in the Congo, trans. Martin Thom. London: Verso. Mauss, Marcel. 1990 [1950]. The Gift: The Form and Reason for Exchange in Archaic Societies, trans. W.D. Halls, with a Foreword by Mary Douglas. London: Routledge. ———. 2006 [1935]. ‘Techniques of the Body’[reprinted in Nathan Schlanger (ed.), Marcel Mauss: Techniques, Technology and Civilisation. New York: Berghahn, pp. 77–95]. Merton, Robert. 1968 [1949]. ‘The Self-fulfilling Prophecy’, in Robert Merton, Social Theory and Social Structure. New York: Free Press, pp. 475–90. Mokuwa, Esther Yei. 2022. ‘Vaccines are for Children Only: Understanding Popular Scepticism about Vaccines for Covid-19 in Sierra Leone’ [Pandemic Preparedness Project, Institute of Development Studies at University of Sussex]. Needham, Rodney. 1963. ‘Introduction’, in Émile Durkheim and Marcel Mauss, Primitive Classification, trans. R. Needham. Chicago, IL: University of Chicago Press, vii–xliii.

REFERENCES

147

North, Douglass. 1990. Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. ———. 1992. Transaction Costs, Institutions, and Economic Performance. San Francisco: ICS Press. North, Douglass, John Joseph Wallis and Barry R. Weingast. 2009. Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. Cambridge: Cambridge University Press. Olson, Mancur. 1965. The Logic of Collective Action: Public Goods and the Theory of the Group. Cambridge, MA: Harvard University Press. Ostrander, David. 1982. ‘One and Two-dimensional Models of the Distribution of Beliefs’, in Mary Douglas (ed.), Essays in the Sociology of Perception. London: Routledge and Kegan Paul, pp. 14–30. Ouchi, William G. 1980. ‘Markets, Bureaucracies, and Clans’, Administrative Sciences Quarterly 25(1): 129–41. Peck, Edward, and Perri 6. 2006. Beyond Delivery: Policy Implementation as Sense-Making and Settlement. Basingstoke: Macmillan. Pegg, Mark G. 1990. ‘Le corps et l’autorité: la lepre de Badouin IV’, Annales ESC, Mars-Avril 2, pp. 265–87. Peters, Krijn. 2011. War and the Crisis of Youth in Sierra Leone. Cambridge: Cambridge University Press. Popper, Karl. 1959 [1935]. The Logic of Scientific Discovery. London: Routledge. ———. 1976. Unended Question: An Intellectual Autobiography, 2nd edn. London: Fontana. Quine, Willard Van Orman. 1960. Word and Object. Cambridge, MA: Massachusetts Institute of Technology Press. Rawls, John. 1970. A Theory of Justice. Oxford: Oxford University Press. ———. 1993. Political Liberalism. New York: Columbia University Press. Rayner, Steve. 1982. ‘The Perception of Time and Space in Egalitarian Sects: A Millenarian Cosmology’, in Mary Douglas (ed.), Essays in the Sociology of Perception. London: Routledge and Kegan Paul, pp. 247–74. ———. 1984. ‘The Politics of Schism: Routinisation and Social Control in the International Socialists/Socialist Workers’ Party’, Sociological Review 32, S1: 46–67. ———. 1986. ‘Management of Radiation Hazards in Hospitals: Plural Rationalities in a Single Institution’, Social Studies of Science 16(4): 573–91. ———. 1999. ‘Mapping Institutional Diversity for Implementing the Lisbon Principles’, Ecological Economics 31(2): 259–74.

148

REFERENCES

Richards, Audrey. 1950. ‘Some Types of Family Structure among the Central Bantu’, in A.R. Radcliffe-Brown and Daryll Forde (eds), African Systems of Kinship and Marriage. London: Oxford University Press for International African Institute. Richards, Paul. 1974. ‘Kant’s Geography and Mental Maps’, Transactions of the Institute of British Geographers 61: 1–16. ———. 1986. Coping with Hunger: Hazard and Experiment in a West African Rice-Farming System. London: Allen & Unwin. ———. 1996. Fighting for the Rain Forest: War, Youth and Resources in Sierra Leone. Oxford: Currey. ———. 2011. ‘Sacred Contagion: Cults of War in Nigeria and Sierra Leone Compared’, in E. Jul-Larson, P.-J. Laurent, P.-Y. Le Meur and E. Leonard (eds), Une anthropologie entre pouvoirs et histoire: conversations autour de l’oeuvre de Jean-Pierre Chauveau. Paris: APAD-IRDKarthala, pp. 535–51. ———. 2013. ‘Elementary Forms of War: Performative Aspects of Youth Militia in Sierra Leone’, in Sandra L. Hausner (ed.), Durkheim in Dialogue: A Centenary Celebration of ‘The Elementary Forms of Religious Life’. New York and Oxford: Berghahn Books, pp. 67–85. ———. 2020. ‘Public Authority and its Demons: The Sherbro Leopard Murders in Sierra Leone’, Africa 91(2): 226–48. doi: 10.1017/ S0001972021000048. ———. 2022. ‘The Moral Economy of Isolates: Escape Agriculture and Warzone Public Authority’, Journal of Peasant Studies. doi: 10.1080/03066150.2022.2052279. Richards, Paul, and James Vincent. 2008. ‘Sierra Leone: Marginalisation of the RUF’, in Jeroen de Zeeuw (ed.), From Soldiers to Politicians: Transforming Rebel Movements After Civil War. Boulder, CO: Lynne Rienner. Ridley, Aaron. 2004. The Philosophy of Music: Theme and Variations. Edinburgh: Edinburgh University Press. Schotter, Andrew. 1981. The Economic Theory of Social Institutions. Cambridge: Cambridge University Press. Schwartzman, Helen B. 1989. The Meeting: Gatherings in Organisations and Communities. New York: Plenum. Selznick, Philip. 1969. ‘Employees’ Perspectives on Industrial Justice’, in Philip Selznick with Philippe Nonet and Howard M. Vollmer, Law, Society and Industrial Justice. New York: Russell Sage Foundation, pp. 183–211. Sen, Amartya. 1999. Development as Freedom. Oxford: Oxford University Press.

REFERENCES

149

Sennett, Richard. 2013. Together: The Rituals, Pleasures and Politics of Cooperation. Harmondsworth: Penguin. Seubert, Andrew J., and Pam Virdi (eds). 2019. Trauma-Informed Approaches to Eating Disorders. New York: Springer. Spinoza, Baruch. 2002 [1670]. ‘The Theological-Political Treatise’, in Michael L. Morgan (ed.), Spinoza: Complete Works, trans. Samuel Shirley. Indianapolis: Hackett. Spiro, Melford. 1968. ‘Religion and Myth: Review of M. Douglas 1966, Purity and Danger: An Analysis of Concepts of Pollution and Taboo’, American Anthropologist 70(2): 391–93. Spitzer, Leo. 1974. The Creoles of Sierra Leone: Their Responses to Colonialism 1870–1945. Madison: University of Wisconsin Press. Stedman-Jones, Susan. 2001. Durkheim Reconsidered. Cambridge: Polity. Tacitus, Publius Cornelius. 2009 [1948; c. AD 98]. ‘Agricola’, in Publius Cornelius Tacitus, Agricola and Germania, trans. Harold Mattingly. Harmondsworth: Penguin, pp. 1–32. Tett, Gillian. 2015. The Silo Effect: Why Putting Everything in its Place isn’t Such a Bright Idea. London: Little, Brown. Thompson, Graham, Jennifer Frances, Rosalind Levačič and Jeremy Mitchell (eds). 1991. Markets, Hierarchies and Networks: The Coordination of Social Life. London: Sage. Thompson, Michael. 1996. Inherent Relationality: An Anti-dualist Approach to Institutions. Bergen: LOS Senteret, University of Bergen. ———. 2008. Organising and Disorganising: A Dynamic and Non-linear Theory of Institutional Emergence and its Implications. Axminster: Triarchy. Thompson, Michael, Richard J. Ellis and Aaron Wildavsky. 1990. Cultural Theory. Boulder: Westview Press. Veblen, Thorstein. 1991 [1899]. The Theory of the Leisure Class. London: Routledge. Verweij, Marco, and Michael Thompson (eds). 2006. Clumsy Solutions for a Complex World: Governance, Politics and Plural Perceptions. Basingstoke: Palgrave Macmillan. Wachtel, Eleanor. 2013 [2003]. ‘Original Minds: Mary Douglas in Conversation with Eleanor Wachtel’, in Richard Fardon (ed.). 2013b. Mary Douglas: A Very Personal Method: Anthropological Writings Drawn from Life. London: Sage, pp. 271–97. Warfield Rawls, Anne. 2005. Epistemology and Practice: Durkheim’s The Elementary Forms of Religious Life. New York and Cambridge: Cambridge University Press.

150

REFERENCES

Weber, Max. 1947. The Theory of Social and Economic Organisation. Talcott Parsons (ed.). New York: Free Press. Wildavsky, Aaron. 1991. The Rise of Radical Egalitarianism. Washington, DC: American University Press. ———. 1994a. ‘Why Self-interest Means Less Outside of a Social Context: Cultural Contributions to a Theory of Rational Choices’, Journal of Theoretical Politics 6(2): 131–59 [reprinted in Sun-Ki Chai and Brendon Swedlow (eds). 1998. Culture and Social Theory. New Brunswick, NJ: Transaction, pp. 231–58]. ———. 1994b. ‘Cultural Pluralism Can Both Strengthen and Weaken Democracy’, Research on Democracy and Society 2: 153–72 [reprinted in Sun-Ki Chai and Brendon Swedlow (eds). 1998. Culture and Social Theory. New Brunswick, NJ: Transaction, pp. 195–216]. ———. 1995. But is it True? A Citizen’s Guide to Environmental Health and Safety Issues. Cambridge, MA: Harvard University Press. Williamson, Oliver E. 1975. Markets and Hierarchies: Analysis and Antitrust Implications. New York: Macmillan. ———. 1981. ‘The Economics of Organisation: The Transaction Cost Approach’, American Journal of Sociology 87(3): 548–77. ———. 1985. The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting. New York: Free Press.

INDEX

••• 6, Perri, 50, 52, 69, 77, 85, 121 Acemoglu, Daron and Robinson, James, 3, 105, 109n1, 128 Africa (continent), 3, 8, 10, 18, 19, 20, 24, 27, 32, 34, 44, 55, 58, 72, 78, 109, 114, 115, 118, 125, 126 Africa (journal), 24 American constitution, 7 Amsterdam, 110 ancestor, 78–79, 113 Aniconic, 113, 118, 121 anomaly, 31, 35–36, 38, 58, 91, 94 Atran, Scott, 36 Australians (aboriginal), 68 Azande (Sudan and Congo people), 18, 19, 33, 34, 133 Baldwin (Crusader king of Jerusalem), 65 Banfield, Edward, 107 Berber (Algerian/Moroccan people), 102 Bernstein, Basil, 25, 43, 58, 137 bias (cultural), 3, 48, 54, 78, 100, 132, 137 Bible (Hebrew), 12, 27, 28, 30, 33, 65, 110, 111, 112, 113, 114 Black Lives Matter (social movement), 1 blame, 26, 37, 63– 64, 67–8, 83, 104, 108, 124, 129, 133

Bloor, David, 54, 91 bonds. See also social integration, 9, 86, 116, 125, 128 Bošković, Aleksandar, 36 Bourdieu, Pierre, 101, 102 Brabanta (plantation), 2021, 22, 28n5 Brands, branding, 88 Brausch, Georges (colonial official), 21 Burton, John Wear, 116 Catholic (church), 14, 15, 16, 17, 19, 29, 21, 22, 25, 28n1, 35, 41, 45, 77, 94, 85, 96, 103, 114 circles, rings, 106, 112–113, 117, 124 Civil war, 27, 55, 115, 117, 118, 119 Civil war (Nigeria), 119, 122n5 Civil war (Sierra Leone), 120 Civil war (Northern Ireland), 55, 118, 120 classification. See also anomaly, 1–3, 8–9, 10–11, 22, 24–25, 31–38, 40–45, 52, 56, 58, 61, 68–69, 72, 77–78, 80, 82–96, 121n3, 124, 125, 129–130, 133, 134, 137 Coase, Ronald, 103 Code Napoleon, 106 Collegiants, 110 Collins, Randall, 56, 125 Colonialism, 2, 8–10, 14, 16–22, 33, 40, 44, 55, 85, 95–96, 111, 114, 115, 125, 126

152

INDEX

composition (poetic, musical, choreographic), 112, 118, 122n4, 122n5, 133 conflict (amplification), 1, 3, 4, 7, 10, 11, 12, 15, 27, 47–48, 52, 65, 72, 74, 76, 104, 114, 115, 116–117, 120, 124–125 conflict (attenuation), 23, 27, 49, 55, 65, 73, 106, 116–118, 121, 125 Congo (Belgian Congo, Democratic Republic of Congo), 8–9, 18–20, 22, 24, 28n5, 96, 115 Conservative Party (UK), 23 constraints. See also social regulation, grid, 9, 11, 42, 48, 53, 84, 87, 107, 124, 127, 128, 131 consumption, 9, 12, 25–26, 52–53, 72, 98–109 passim, 134 cosmology. See thought style, 12n1, 32, 33, 45, 46, 47, 49, 52, 53, 54, 57–61, 69 cosmogram, 113, 117 Covid-19, 63–64, 90, 129, 131–132 Covid-19 vaccination, 109, 126–127 Covid-19 workplace working, home working, 6–7 Coyle, Dennis, 77 Coyle, Dennis and Ellis, Richard, 52 cult. See ritual ‘Cultural Bias’ (essay), 25, 45, 46, 48 cultural theory, 47, 50 d’Azeglio, Massimo (Italian statesman), 7 Dake, Karl and Thompson, Michael, 53, 101 sance, 38, 89, 112, 118 Dant, Tim, 38n2 ‘Deciphering a Meal’ (essay), 99 decolonisation. See colonialism dependent variable, 5, 6, 7, 26–27, 53–54, 58, 70, 82, 102, 108 De Soto, Clinton, 79 Drosten, Christian, 131, 132

Durkheim, Émile, 4, 9, 13, 14, 15, 16, 25, 26, 27, 28n6, 32, 33, 34, 38n3, 40, 45, 46, 49, 50, 51, 52, 62, 67, 68, 69, 70, 71, 80, 86, 90, 92, 94, 100, 102, 103, 11, 117 Durkheim, Émile and Mauss, Marcel, 80, 85 dynamic(s). See feedback enclave, 37, 50, 52, 59, 60, 66, 74, 76, 80, 83, 90, 95, 112, 117, 118, 120, 121n1, 123, 126, 127, 132, 136 Elster, Jon, 73, 74, 76 Essays in the Sociology of Perception, 46, 91 Evans Pritchard, Edward, 18, 19, 28n3, 34, 44, 70, 133, 134 Ezra (priest), 111, 112 Fardon, Richard, 14, 15, 21, 22, 28n1, 28n2, 30, 38n1, 41, 42, 65n1, 83, 98, 125, 133, 135, 136, 137, 138 fatalism, fatalist. See isolate ordering feedback, 4, 6, 10, 12, 19, 22, 23, 27, 40, 47, 52, 54, 55, 57, 67, 68, 72, 73, 82, 83, 105, 114, 119, 120, 123, 124, 125, 128, 129, 130, 136 Feyerabend, Paul, 69 Fleck, Ludwick, 26, 69, 71, 72, 80 Forde, Daryll, 22, 24, 28n4 Fortes, Meyer, 16, 17, 28n2, 44, 114 Foucault, Michel, 130 ‘Four Cultures’ (essay), 47, 50, 52 Friends of the Earth, 60 functional explanation, 4, 10, 19, 36, 56, 73–77, 86, 95, 101, 123, 125, 129 functionalism, 4, 17, 44, 73, 90, 114 Genesis of a Scientific Fact (Ludwick Fleck), 69 Ghana. See Gold Coast Goffman, Erving, 5, 30, 55, 125

INDEX

Gola Forest (Sierra Leone), 101 Gold Coast, 16, 20 Goodman, Nelson, 37, 82, 84, 85, 86, 87, 88, 89, 90, 91, 93, 96n6, 137 goods. See consumption Grabbe, Lester, 113 Greenpeace, 60 grid. See social regulation group. See social integration Hacking, Ian, 58, 130 Hansen’s Disease (leprosy), 64 Hedström, Peter, 74 hierarchy, 15, 18, 37, 49, 50, 58, 60, 65, 66, 76, 78, 90, 95, 102, 105, 112, 117, 121–122n4, 123, 132, 135 Hill Station (Sierra Leone), 126 How Institutions Think, 7, 11, 26, 27, 42, 61, 68, 69, 70, 71, 72, 74, 78, 80, 102, 105, 112. 114, 115, 117, 139, 134 Hubert, Henri, 102, 108 Hull, David, 37, 82, 95 ‘Human Needs and Wants’ (essay), 99, 104 ideas, 4, 6, 7, 9, 10, 12, 13n3, 14, 15, 32, 33, 42, 55, 59, 60, 61, 64, 87, 89, 94, 101, 102, 114, 115, 119, 123, 124, 128 imperialism. See colonialism Implicit Meanings, 23, 82, 96n2, 99, 114 In the Active Voice, 99 In the Wilderness, 112 independent variable, 7, 16, 26–27, 53–54, 58, 60, 62, 83, 102, 108 individualism (form of social organisation), 16, 46, 50, 59, 60, 70, 76, 81n3, 90, 95, 105, 122n4, 123, 135 individualism (methodological), 63, 66n2, 104

153

institutional dynamics. See dynamic(s) institutionalisation, 7, 8, 10, 11, 12, 37, 71, 116, 124, 125, 126, 127, 132 institution(s), 1–4, 5, 6, 7, 8, 9, 10, 11–12, 16, 22, 27, 33, 36–38, 54, 56, 63, 65, 67–81 passim, 82, 98, 103–109 passim, 111–115, 116, 119–121, 123–125, 128–130, 131–132, 133, 134, 136 intolerance, 3, 45 intransigence, 3, 48, 109, 121, 125 Isherwood, Baron, 25, 52, 99, 101, 134 isolate ordering, isolates, isolate life, 27, 46, 50, 57, 58, 59, 60, 70, 77, 90, 107, 123, 128, 132 Italy, Italians, 7 Jacob’s Tears, 115 Kabylia (Algeria), 102 Kaldor, Mary, 115 Kasai (river and province in DR Congo), 8, 19, 20, 28n5, 111, 114 Kuhn, Thomas, 69, 91, 93, 94, 96–97n6 Lakatos, Imre, 69, 91 Leach, Edmund, 41 Lele (Congolese people living near Kasai river), 8, 9, 19–24, 28 n3, 28n5, 31, 34, 35, 38 n3, 63, 72, 74, 76, 78, 90, 91, 93–96, 96 n1, 103, 111, 114, 115 leprosy, 64, 65, 82 Levi-Strauss, Claude, 91 Leviticus (Hebrew Bible), 27, 29, 31, 32, 110, 113, 117, 118 Leviticus as Literature, 113 Levy-Bruhl, Lucien, 34, 39n4 Littlejohn, James, 41 Lynn, Charles, 17

154

INDEX

Makum (research assistant), 22 Marchal, Jules, 20 Mars, Gerald, 53, 74, 83, 120 Mars, Valerie, 53 Marshall, Alfred, 99 matriliny, 19, 20, 21, 28n4, 78 Marxist anthropology, 42 Mauss, Marcel, 67, 96n4, 100, 101, 108 Merton, Robert, 74, 80n1, 130 microcosm, 135 Missing Persons, 99, 104, 136 Mokuwa, Esther, 101, 126 Natural Symbols, 23, 25, 41, 42, 44, 46, 47, 56, 57, 90 negative feedback dynamic. See dynamic(s) New Institutionalism, 108 ‘new war’, 115 Ney, Steven, 48, 99, 104, 136 North, Douglass, 27, 98, 103, 104, 105, 1108, 109n1, 128 North, Douglass, Wallis, John and Weingast, Barry, 104, 105, 109n1 Northwestern University, 23, 82 Nuer (Nilotic people), 31, 134 Numbers (Hebrew Bible), 27, 106, 112, 113 Olson, Mancur, 72, 76 ostension, 86–87 Oxford (university), 16, 18, 20, 25, 98 pangolin (scaly anteater), 9, 35, 94 Parsons, Talcott, 103 Pegg, Mark, 65 performance (of music), 89, 96n5, 127 performance, performative (of rite, ritual, cult). See also ritual, 15, 59, 63, 67, 85, 86–87, 88, 92, 94, 112, 117, 118, 122n4, 125, 127 poison oracle / poison ordeal, 18, 22, 33, 63, 94

Popper, Karl, 91, 97n7 positive feedback dynamic. See dynamic(s) postcolonialism. See colonialism poverty, poor, impoverished, 11, 14, 27, 62, 99, 104–109 passim, 126–127 purification rite, purification ritual, 30, 117–118, 129 purity, 29, 32, 37, 110, 112, 117, 118 Purity and Danger, 9, 10, 11, 22, 24, 25, 29, 30, 34, 36, 37, 40, 43, 61, 64, 69, 90, 91, 118 racism, race prejudice, racial bias, racial privilege, racial boundary, 3, 16, 18, 33, 37, 40, 41 Rawls, John, 62 Rayner, Steve, 26, 57, 84, 104 Revolutionary United Front (Sierra Leone), 56, 118 Richards, Paul, 55, 77, 84, 107, 118, 119, 121, 122n5 ‘Rightness of Categories’ (essay), 12, 23, 28n6, 35, 82, 83, 84, 85, 86, 87, 90, 93, 95, 137 ring. See circle risk, 12, 23, 24, 26, 31, 58–65 passim, 94, 96 Risk Acceptability According to the Social Sciences, 26, 61, 62, 63 Risk and Blame, 47, 49, 61, 63, 64, 65, 82, 134 Risk and Culture, 23, 58, 59, 60, 61, 52, 69, 74, 112 ritual, rite, cult, ritualisation, ritualism, anti-ritualism, 4, 5, 6, 7, 8, 9, 10, 15, 24, 25, 26, 27, 30, 31, 32, 34, 35, 37–38, 40–42, 45, 46, 51, 54, 55, 56, 59, 63, 64, 65, 67, 68, 71, 74, 78, 84, 86–88, 92, 93, 96n4, 98–102 passim, 105, 107, 108, 111, 112, 114, 115, 116,

INDEX

117–118, 119–121, 123, 124, 125, 129, 134, 137 Robertson Smith, William, 35 Ross, Ronald (malaria transmission), 126 rules, 2–3, 9, 17, 21, 22, 25, 45, 55, 58, 65, 71, 84, 85, 93, 95, 96n1, 104, 106, 110, 111, 124, 126, 128 Russell Sage Foundation, 23 sacred contagion, 124 Sacred Heart (school), 14, 15, 16 Schotter, Andrew, 80– 81n2, 98 sect, sectarian. See enclave ‘Self Evidence’ (essay), 35 Selznick, Philip, 62 Sen, Amartya, 11, 107, 109 Seubert, Andrew and Virdi, Pam, 100 Sierra Club, 60 Sierra Leone, 20, 55, 85, 101, 118–120, 126 silence, 113, 120–121, 129 Smith, Adam, 76 social integration (group), 4, 5, 9, 10, 11, 15, 16, 17, 18, 25, 26, 42, 43, 45, 46, 47, 49, 50, 51, 52, 53, 54, 57, 58, 59, 60, 64, 68, 69, 70, 71, 73, 74, 75, 76, 77, 80, 82, 83, 94, 105, 95, 100, 103, 104, 105, 106, 107, 109, 111, 114, 116, 117, 118, 119, 120, 121n3, 123, 124, 125, 127, 128, 132, 133 social media, 12, 94, 114, 125, 129 social organisation, 4, 15, 22, 26, 34, 38, 39, 40, 41, 42, 45, 47, 49, 50, 53, 54, 55, 58, 59, 65, 68, 69, 70, 79, 80, 87, 99, 102, 104, 123, 134 social regulation (grid), 9, 25, 42–47 passim, 49, 50, 51, 53, 54, 57, 58, 59, 60, 68, 71, 74, 80, 95, 100, 101, 102, 104, 105, 106, 107, 109, 114, 116, 120, 121n3, 124, 127, 128, 132

155

Spinoza, Baruch, 110 Spiro, Melford, 29, 32, 33 Stedman-Jones, Susan, 16 stigma, stigmatization, 1, 12, 40, 128 structural functionalism. See functionalism Tacitus, Publius Cornelius, 13n4 Tallensi (people of northern Ghana), 16, 17, 18, 38n3, 44, 114 taxonomy, 82, 93, 96 Tett, Gillian, 3, 109 Tew, Gilbert, 14 Tew, Phyllis, 28n1 Thatcher, Margaret, 23 The Gift (Marcel Mauss), 67, 100 The World of Goods, 25, 99, 101 Thinking in Circles, 112 Thompson, Graham, 13n5 Thompson, Michael, 52, 54, 56n1, 57, 68, 107, 114 thought style, 4, 7, 12n1, 22, 25, 26, 29, 38, 39n4, 42, 47, 48, 67–84 passim, 87, 95, 110, 117, 120, 121n3, 123, 124, 134, 135 Thought Styles (book), 99 ‘Traditional Culture – Let’s Hear No More about It’ (essay), 104, 108 Twomey, Daniel, 14 University College London, 22, 23, 42, 116 Veblen, Thorstein, 99 Verweij, Marco, 52, 107 Wachtel, Eleanor, 22, 28n2, 98 Warfield Rawls, Anne, 16, 69 Washington Consensus, 105, 108 Weber, Max, 12n3, 49 Wildavsky, Aaron, 23, 57, 58, 59, 60, 61 Williamson, Oliver, 9, 80n2, 98, 102, 103

156

INDEX

‘Witchcraft and Leprosy: Two Strategies for Rejection’ (essay), 64 witch-finding, witches, 18, 19, 33, 34, 37, 64, 72, 82, 114–115, 132 World Bank, 62, 96, 106, 107 World Health Organization (WHO), 131, 132

X variable. See independent variable Y variable. See dependent variable Yoruba (people of western Nigeria), 31 Young, Michael (Lord Young of Dartington), 113, 121n2