The Foetal Condition E-Book

Luc Boltanski

• Herausgeber: John Wiley & Sons
• Kategorie: Geisteswissenschaft
• Sprache: Englisch

Seitenzahl: 868

Veröffentlichungsjahr: 2013

Beschreibung

Abortion is a contentious issue in social life but it has rarely been subjected to careful scrutiny in the social sciences. While the legalization of abortion has brought it into the public domain, it still remains a sensitive topic in many cultures, often hidden from view and rarely spoken about, consigned to a shadowy existence.

Drawing on reports gathered from hospital settings and in-depth interviews with women who have had abortions, Luc Boltanski sets out to explain the ambiguous status of this social practice. Abortion, he argues, has to remain in the shadows, for it reveals a contradiction at the heart of the social contract: the principle of the uniqueness of beings conflicts with the postulate of their replaceable nature, a postulate without which no society would achieve demographic renewal.

This leads Boltanski to explore the way human beings are engendered and to analyze the symbolic constraints that preside over their entry into society. What makes a human being is not the foetus as such, ensconced within the body, but rather the process by which it is taken up symbolically in speech - that is, its symbolic adoption. But this symbolic adoption presupposes the possibility of discriminating among embryos that are indistinguishable. For society, and sometimes for individuals, the arbitrary character of this discrimination is hard to tolerate. The contradiction is made bearable, Boltanski shows, by a grammatical categorization: the “project” foetus - adopted by its parents, who use speech to welcome the new being and give it a name - is juxtaposed to the “tumoral” foetus, an accidental embryo that will not be the object of a life-forming project.

Bringing together grammar, narrations of life experience and an historical perspective, this highly original book sheds fresh light on a social phenomenon that is widely practised but poorly understood.

Leseprobe

The Foetal Condition

The Foetal Condition

A Sociology of Engendering and Abortion

Luc Boltanski

Translated by Catherine Porter

polity

First published in French as La condition fœtale. Une sociologie de l’engendrement et de l’avortement © Editions Gallimard, Paris, 2004

This English edition © Polity Press, 2013

Polity Press 65 Bridge Street Cambridge CB2 1UR, UK

Polity Press 350 Main Street Malden, MA 02148, USA

All rights reserved. Except for the quotation of short passages for the purpose of criticism and review, no part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the publisher.

ISBN: 978-0-7456-8349-2

A catalogue record for this book is available from the British Library.

The publisher has used its best endeavours to ensure that the URLs for external websites referred to in this book are correct and active at the time of going to press. However, the publisher has no responsibility for the websites and can make no guarantee that a site will remain live or that the content is or will remain appropriate.

Every effort has been made to trace all copyright holders, but if any have been inadvertently overlooked the publisher will be pleased to include any necessary credits in any subsequent reprint or edition.

For further information on Polity, visit our website: www.politybooks.com

. . . the evening star is the morning star . . . Gottlob Frege, ‘On Sense and Reference’

Contents

Introduction

1.  The Anthropological Dimensions of Abortion

2.  The Two Constraints on Engendering

3.  Understandings

4.  The Parental Project

5.  Constructing Foetal Categories

6.  The Justification of Abortion

7.  The Experience of Abortion

Conclusion: Forgetting Abortion

Notes

Works Cited

Index

Introduction

The place of abortion in the changes that have affected the politics of life

Among the principal changes that have marked the last third of the twentieth century and the beginning of the twenty-first (including the formation of a new ‘spirit of capitalism’, for example), one can unhesitatingly attribute considerable importance to transformations that have affected the politics of life, most notably changes in the conditions of reproduction, gestation and childbirth. As women’s roles in society, representations of the family, relations between the sexes, modalities of sexuality and affectivity and other major aspects of private life have been transformed, our relation to the possibilities offered by technological developments has ranged from admiring fascination to uneasy reticence. The changes in question have been subject all along to a great deal of analysis and commentary, because they have been viewed, not unjustifiably, as opening the way to inflections in our idea of what it means to be human; they have even led us to reconsider certain aspects of Western anthropology that had previously been taken for granted. Let us note, however, that whether the commentators have looked upon these changes favourably or, as has often been the case, with a critical eye, they have tended to focus on the most spectacular innovations, especially those associated with medically assisted reproduction; in other words, practices that are relatively rare (such as the use of surrogate mothers) or that do not yet exist (such as human cloning) have been the primary focus of attention. Cloning, for example, has given rise to an abundant literature over a short period of time, even though to date the process has not been applied to humans.1

As I could not hope to address this proteiform thematics in all its aspects, I chose to approach it indirectly, by focusing on an event that is limited in scope but that seems to me to have played a particularly important role in the evolution that is still under way. This crystallizing event was the legali-zation of abortion, which occurred in the major Western countries between the mid-1960s and the mid-1970s, precisely at a time when broader changes affecting human life were either beginning to appear or becoming so significant that they could not be ignored. The role played by the legalization of abortion in the transformations associated with the women’s movement and in those that have affected private life in its familial, affective or sexual dimensions can hardly be questioned. But we may also suppose that the development of biotechnologies, and of techniques for medically assisted reproduction in particular, would have run into considerable difficulty if the ban on abortion had not been lifted; its disappearance removed an obstacle to research on intrauterine life and embryos.

A second reason for taking up this topic was its very difficulty. At the centre of disputes that have often been extremely harsh and that seem poised to reignite at any moment, the question of abortion is the very prototype of an inappropriate object for a sociologist, because it seems impossible to approach with the requisite detachment. Attesting to this, in France, is the virtual absence of publications on the subject over a period of nearly two decades, between 1982 (when an excellent special issue of the Revue française de sociologie devoted to abortion came out, edited by François-André Isambert and Paul Ladrière) and the early years of the new century (when several books on abortion appeared). In contrast, publications on abortion remained abundant throughout this period in the United States. While the vast body of literature on the subject includes much work of great integrity and real scientific value, work carried out most notably by anthropologists studying contemporary societies, it also includes a large number of polemical books or articles written in support of positions favourable to abortion (pro-choice); works supporting the opposite positions (pro-life) are much less common, at least in the academic context. These two ways of situating oneself with regard to the question of abortion – either avoiding it or entering into it as if charging into an arena to do battle – are indicative, moreover, of the different ways in which the question has arisen in the United States and in France: as a central conflict that sometimes verges on civil war, in the first case, and as a taboo topic to be avoided, a prohibition that could not be prudently transgressed, in the second.

My intention in this book, then, is precisely to treat abortion as if it were a sociological object like any other, that is, to invoke the celebrated notion of ‘axiological neutrality’. So easy to affirm as a principle and so hard to adopt in practice, neutrality is nevertheless one of the axioms that has allowed sociology to be constituted as an academic discipline. It can be sidestepped without harmful consequences when the object is already solidly established as a research topic, but it remains indispensable for addressing problems that have not yet achieved intellectual existence except in the rhetoric of conflict. To grasp such a problem using the methods and language of sociology, it is thus absolutely necessary to set aside the urgency of practical issues in order to proceed as if it were possible to consider the matter from the outside, and, in a sense, irresponsibly, that is, while refusing to raise, even for oneself, the questions that a ‘man of action’ cannot avoid, according to the division of labour that Max Weber spelled out once and for all in his celebrated lectures on science and politics as vocations (Weber 2004). Distance from one’s topic, which underlies the idea of axiological neutrality, is achieved in the work presented here by the requirements and constraints of model construction, a task that involves taking utterances and various other traces deposited in the social world and attempting to organize them by testing their cohesion and their robustness. This process is somewhat analogous to the way the so-called natural sciences set aside precisely what we commonly call nature – for example, when we are out for a walk in a natural setting – so as to concentrate on analysing samples that have been selected, duly labelled and transported into the equipped space of a laboratory.2 This is to say that at no point in this book will I formulate what readers ordinarily expect from a work on abortion, or for that matter from a discussion of almost any of the questions that are at the centre of still burning conflicts: namely, an opinion, even though opinions on abortion are precisely part of the data whose logic I am seeking to reconstruct. Having lived through the 1970s, when it was impossible to deal with a social topic – social class is a good example – without being challenged to reveal one’s position (‘Where do you stand?’), I am not unaware that such a posture has every chance of encountering suspicion or rejection. Nevertheless, it is a posture I shall maintain throughout.

Two theoretical goals

Adopting a relatively distant position with respect to the directly political components of my topic was made easier by the fact that my decision to study abortion was dictated at least as much by theoretical considerations as by the attention that people legitimately expect sociologists to pay to the contemporary social world. On the level of sociological theory, my research had two primary goals. Both reflect a desire to re-engage with questions that I had deliberately set aside more than twenty years earlier, when I turned away from the problematics that had dominated the social sciences in the 1960s and 1970s. With respect to these problematics, in my new work I wanted to break in particular with several key oppositions: between unconscious reality and self-deceiving consciousness, between what belongs to structure and what stems from phenomena, and especially between the real but hidden motives dominated by interests and the often altruistic but illusory reasons on which actors claimed to base their actions. My intention was to relaunch a research programme in the realm of moral sociology that had been at the centre of Émile Durkheim’s preoccupations, but that the structuralist positivism of the 1960s and 1970s, relying on narrow conceptions of Marxism and psychoanalysis, had rejected. Now, moral sociology does not necessarily require that all moral references on the part of actors be taken at face value, but it does require at the very least that sociologists take such references seriously, in order to study the way actors themselves deal with the gap between normative requirements and reality, whether by critiquing the world as it is or, on the contrary, by justifying themselves in response to critiques.

The wish to develop a programme bearing on critical operations and on justification – in other words, the wish to substitute a sociology of critique for a critical sociology – led me to set aside one question, however, that general sociology cannot ignore: how to deal with the differential between the components of the social world that are exposed to broad daylight at a given moment in time and those that, without being unknown, are nevertheless not well known, as if there were a sort of tacit agreement to close one’s eyes to them. My primary theoretical goal in this book, then, is to take up again on a new basis, without engaging a problematics of the unconscious in the strong sense, a question that has essentially to do with social bad faith, with the separation between what is known officially and what is known in an unofficial or tacit mode. This question has been a familiar one for a long time; it is at the heart of Pierre Bourdieu’s anthropological work (in his courses, Bourdieu liked to recall, with reference to Marcel Mauss, that ‘societies always pay themselves with their own counterfeit coinage’), and I learned to do sociology under Bourdieu’s tutelage. The question is not completely absent from the work on critique and justification that I undertook in collaboration with Laurent Thévenot, where it appeared in the form of the opposition, central to the model of the ordinary meaning of justice that we developed together, between moments when one opens one’s eyes and those when one closes one’s eyes (Boltanski and Thévenot 2006). Still, I find now that I did not pay sufficient attention to this question, either in my teaching or in my subsequent work. In the present book it will become clear how abortion, as a possibility and as a practice, constitutes a privileged terrain for analysing the different ways in which things that matter to society can be known and reported, as requirements or as effects, in the mode of ethical or political generalizations or in the anecdotal mode. (In the latter case, one treats phenomena as if they were isolated and avoids putting them in a series that would associate them with other phenomena of the same type, so as not to have to draw out their consequences.)

This attention to procedures of avoidance led me to place at the centre of the present work a classic question in the social sciences whose importance I had not fully appreciated in my earlier work on the relation between justification and action: the question of contradictions and the social arrangements that seek to attenuate or circumvent them. As the conclusion to this work will make clear, the question of contradiction is linked, for me, with the question of normativity, and I shall try to describe – obviously without exhausting the topic – some formulas for dealing with contradiction. I shall make a particular effort to distinguish between two sorts of solutions: those that consist in distributing various types of normative requirements among temporally different situations and sequences, requirements that are credited with universal validity to an equivalent degree even though they are incompatible with one another (this is to a great extent the path explored in On Justification), and those that consist in establishing a hierarchy among the various consequences of an action, in a logic of the lesser evil (solutions of the latter sort will predominate in the present context).

My second theoretical objective, which is not completely independent of the first, was to try to bring together three distinct approaches associated with intellectual traditions that often have difficulty getting along. The first is an approach that can be qualified as grammatical. It takes hold of facts, selected from a corpus, and seeks to organize them so as to establish a model that will allow them to be arranged in relation to one another according to a logic capable of integrating them intelligibly and without remainder, rather in the way that linguistics goes about establishing distinctive features in phonology, or generative schemas whose organization defines a model of competence in syntax.3 Such an approach, which adopts a position of exteriority with respect to the object or, in a different terminology, which has an objectivist character, does not imply raising questions about phenomena, that is, about the way in which persons experience the world when they encounter the phenomena of which the model offers an organized representation. I shall thus seek to sketch what may be called a grammar of engendering (chapter 2), by specifying certain of the constraints that weigh on the fabrication of new human beings so that they may take their place without too much difficulty among the humans who are already present, and also (at least in a number of societies) among the dead, in so far as the latter remain present in memory. In the first chapter, I shall present the properties of abortion that seem to me the most pertinent and also the most intransigent for sociology. The question of abortion will serve as an operator for working out the components of the grammar of engendering that abortion unveils, in a way, by making these components salient in their contradictory dimensions, which the social arrangements that surround the engendering of human beings (and most notably kinship arrangements) aim precisely to surmount.

The second approach that I have sought to develop in this work consists in starting from the experience of persons in such a way as to describe the manner in which they live in their flesh their encounter with the components and defining features of the act of abortion that have been integrated into the model. But, instead of emphasizing the distance between the lessons offered by the grammatical and the experiential approaches, as one often does in undertakings of a structural type, I shall attempt on the contrary to show how these two approaches can converge, and how it is possible to rediscover through experience – although the languages of description will differ – the components whose relevance has been demonstrated by the grammatical approach. In a radical shift of theoretical registers, I shall then seek support in a conceptual field developed in phenomenology, in an effort to surmount (or at least to bypass) the very lively tensions between the experiential approach, which seeks to describe the intentions immanent to behaviours, and the grammatical approach that I adopted initially, which has often been criticized from the phenomenological standpoint for striving to reduce social phenomena to a universe that can be calculated according to rules.4 However twisted and fraught with pitfalls this path may be, it is perhaps the only one that will make it possible to specify the concept of practice in order to articulate models of competence established from a position of exteriority with regard to narratives that persons offer about their lives, when, ‘emplotting’ these lives, to borrow Paul Ricoeur’s term (Ricoeur 1984), they raise questions about the intentions and motivations that lay behind their own actions. This is how the concept of flesh, put to work in the first part of the book in a strictly structural fashion, since its defining features are established solely in opposition to the concept of speech (so as to establish the distinction between engendering through flesh and engendering through speech), is taken up again and re-elaborated, with a different orientation, in chapter 7, where I seek to account for the experience of flesh during pregnancy, as a dimension of a woman’s relation to her own body.

The third approach, finally, has a historical character. It consists in taking into account the way in which certain constraints that may be understood as possessing an anthropological (and thus in a sense ahistorical) dimension, when they are operative at a specific moment in time, can generate different states of reality. Although they can coexist, at least in part, these states gain intelligibility when they are described in chronological order. I seek to show how the constraints in question (which will be described in chapters 1 and 2) have been manifested differently – and as a result have weighed differently on the actions of persons subjected to them – in different historical contexts, the term ‘historical’ being used in a very broad sense (this is the object of chapters 3–6). I shall then evoke factors that can be viewed as exogenous, in many cases, in the sense that they present themselves (to use the vocabulary of economics) as externalities affecting the relation that persons may have with the grammatical components of the model of engendering presented in the early part of the book, without radically modifying these components.

Unexpectedly, my research has led me across a variety of terrains, into various areas of social science where I am far from an expert. But for me this was one of the most interesting aspects of the project. The generation to which I belong is perhaps the last that will dare to manifest the ‘amateurism’ (or on whose part such a manifestation will be tolerated with a certain indulgence) that nourished a number of works in the social sciences considered ‘classic’ today, an approach that the professionalism of our disciplines – modelled, perhaps mistakenly, on what one imagines to have been the evolution of the so-called hard sciences – threatens to banish forever. The fact remains that, despite the advice generously proffered by eminent colleagues in the disciplines touched upon here, I am aware of the very imperfect character of the enterprise; its impeccable achievement would have required, as they say, a lifetime.

Vocabulary issues

Some clarifications regarding word choices: I have preferred, most often, to use the term ‘abortion’ rather than the French neologism interruption volontaire de grossesse (voluntary interruption of pregnancy) that appeared with the Veil law of 1975; the latter term seemed too marked, historically and socially, to fit the very general phenomenon I sought to study.5 A problem of the same sort arose when it came to qualifying the being that comes to be implanted in the flesh following sexual intercourse. In current practice, several different terms are used in accordance with the state of development of the pregnancy: pre-embryo, embryo, non-viable foetus, viable foetus and so on. But, beyond the fact that the borders between the beings these terms are supposed to designate are far from firmly established (indeed, they are often in dispute), it became clear to me that making terminological judgements was part of my task; having taken on the challenge of describing the logic of the terminology, I could not settle for adopting it naively. Thus I chose to use the term ‘foetus’ exclusively, as a convention, to designate the being in question. Seeking to stress the symbolic dimensions of the events that accompany the entrance of new beings (or their failure to enter) into the world of humans, I largely excluded from my vocabulary terms that had medical, biological or demographic origins or connotations, for example ‘reproduction’, ‘procreation’ or even ‘womb’ (for which I generally substituted the phenomenological term ‘flesh’). Moreover, to designate what happens when a woman finds herself pregnant, I opted for the term ‘engendering’ rather than, for example, ‘having a child’, for – and this fact is precisely at the core of my research – not every being engendered is the occasion for the birth of a child.

Finally, I use the term ‘constructivism’ to designate the method of model construction deployed here, and the term ‘constructionism’ to speak of approaches described as ‘the social construction of reality’.

Credits

It would have been impossible to bring this research to fruition without the collaboration and teamwork of many individuals.

The surveys and observations conducted in hospital settings were carried out by Marie-Noël Godet, an engineer at the Centre National de la Recherche Scientifique; over a period of eighteen months, she went several times a week to one of the principal family planning clinics in the Paris region and also to the gynaecological clinic of a mid-size city in the Nord region. She was able to observe women (or, occasionally, couples) during their pre-abortion counselling interviews or their consultations with a doctor; she collected data on roughly one hundred cases. She was of course not allowed to record the interviews, but she transcribed from memory the essential elements of the conversations she heard each day, and this allowed her to accumulate a very rich corpus. She met separately with a number of people working in both public and private gynaecological services. She was able to sit in on several internal meetings during which doctors and nurses discussed the problems they faced, and on two occasions she was admitted to the operating room. In addition, she conducted a series of fifteen interviews with prominent individuals – doctors, for the most part – who had played an important role in the movement leading to legalized abortion, and she collected documents on several aspects of abortion (legal and medical in particular) as it had evolved over a period of some thirty years. Finally, Godet, whose training in clinical psychology and competence in the realm of psychoanalysis made significant contributions, played a very active role throughout my study, especially by reminding me regularly that, among the factors that come into play in social life, some are not directly accessible via the usual concepts and methods of sociology; in other words, she emphasized the relatively autonomous character of psychic life, especially in its affective dimensions.

In addition to the hundred or so observations made in hospitals or clinics, forty in-depth personal interviews (lasting one to two hours, and recorded on tape) were conducted with women who had had abortions. (With one exception, our interviewees were all women; one of our great regrets was that we had insufficient resources to undertake parallel interviews with men.) I carried out some of these interviews myself, despite the unfortunate handicap of belonging to the other sex. Most of the interviews were conducted by Susana Bleil, a doctoral candidate at the École des hautes études en sciences sociales (EHESS), and by Valérie Pihet, a research assistant at the Centre de sociologie de l’innovation (CSI). Bleil and Pihet were able to establish relations of trust with the persons they met. Although it has been legal in France for nearly thirty years, abortion is still a life event about which it is difficult to speak. So that the interviews could proceed under favourable conditions, that is, so they could provide us with the knowledge we lacked without constituting an ordeal for those who agreed to speak with us, we chose to begin by meeting with people who already belonged to our circle of personal relations, and then asked those individuals to introduce us to others (a ‘snowball’ sample). The flaw in this method is obviously that it restricts the range of the social field in which the study is carried out. Thus we reached mainly young, urban women, either students or workers in the service sector; most of them had no religious affiliations. Nevertheless, when we compared the data collected during the in-depth interviews with the data gathered in the hospital context, where the spread of social classes, geographical origins and religious affiliations was much broader, we did not find major divergences, and this convinced us that the information gleaned from interviews could be generalized.6 Marie-Noël Godet participated in the analysis of the data collected in the hospital context; Susana Bleil did the same for the interviews.

Valérie Pihet also collaborated with me on a search for images, and she collected pictures of foetal life that allowed us to set up an installation in the context of the Iconoclash exhibition curated by Bruno Latour at the Centre for Art and Media in Karlsruhe in May 2002.

I myself took charge of gathering most of the documentation; this led me to make a number of excursions, no doubt often a bit erratic, into various areas of the social sciences. I directed the working group under the aegis of the Groupe de sociologie politique et morale (EHESS-CNRS [Centre national de la recherché scientifique]). But my principal tasks lay in building the analytic framework that allowed me to integrate the data collected and to write the present book. Responsibility for this text is mine alone, and the errors that will undoubtedly be found here are my responsibility as well.

Special thanks

Throughout the preparation and writing of this work, I benefited from the support and collaboration of many people. Almost daily discussions with Élisabeth Claverie played a very important role in the overall design of the project. This book is also in part hers. Jean-Élie Boltanski helped considerably in establishing the grammar of engendering presented in chapter 2, by applying his knowledge in the field of formal linguistics to the incongruous object that I presented to him. I also learned a great deal from exchanges with Christian Boltanski and Hans Ulrich Obrist about the various paths that lead to questions of singularity and its negation from the vantage points of sociology, anthropology and also the plastic arts and poetry.

The pages presented here owe more than I can say to Cyril Lemieux’s comments on a series of drafts that he was kind enough to read, and to the long discussions that I had with him throughout the entire process. I am indebted as well to exchanges with Frédéric Keck (who is responsible for my rediscovery of certain works by Claude Lévi-Strauss that I had read – badly – long ago) and with Sébastien Laoureux, who shared his knowledge in the area of phenomenology. Finally, I benefited on many points from the counsel of friends and colleagues – especially my EHESS colleagues – to whom I told my story and whom I did not hesitate to pepper with questions in my search for details and references. I name them here as a group: Catherine Alès, Jérôme Alexandre, André Burguière, Philippe Descola, Marie-Angèle Hermitte, Claude Imbert, Paul Jobin, Rose-Marie Lagrave, Hervé Le Bras, Nicolas Offenstadt, Joan Stavo-Debauge, Anne Christine Taylor, Isabelle Thireau. I also learned a lot from some of the participants in my seminar at EHESS, especially Roser Cusso, Caroline Ibos, Catherine Rémy, Bénédicte Rousseau, Anne Paillet and Isabelle Baszanger. I profited, too, from attending a seminar on ‘the secret’ led by Cyril Lemieux, Dominique Linhardt and Emmanuel Didier, as well as seminars given by Paul Rabinow at EHESS and at ENS (École normale supérieure).

The pages that follow also owe a good deal to those who were kind enough to read the manuscript ‘straight out of the box’ – Damien de Blic, Sabine Chalvon, Ève Chiapello, Caroline Ibos and Bruno Latour; I thank them for their critiques and comments.

Drafts of parts of this work have been presented in various colloquia and seminars. I am grateful in particular to Mario Perniola, who invited me to speak about my research in the colloquium titled ‘Natura, Coltura, Cultura’ that he organized at the Università degli Studi di Roma (Tor Vergata) in February 2002 (the text of my talk was published in the journal Agalma); Bruno Latour, who invited me to speak at the École des mines in Paris and made room for me in the Iconoclash exhibition mentioned earlier; and Claude Imbert, who allowed me to present my work to the anthropologists, historians, sociologists and philosophers brought together at a colloquium she organized in June 2003 at Trinity College, Cambridge.

This undertaking benefited from the constant support of the Groupe de sociologie politique et morale and from that of Jacques Revel, president of EHESS. It could not have been completed without the friendly attention of Éric Vigne, whose role in the renewal of the social sciences far exceeds what can ordinarily be expected of an editor.

Finally, I want to thank all those who made our research work possible, either by welcoming us in family planning clinics or by agreeing to talk with us. This book would not have been possible without their generosity.

1

The Anthropological Dimensions of Abortion

The comparatist approach of George Devereux

To refer to practices in their most general – that is, anthropological – dimension is to invite disapproval from today’s social sciences, which have probably never before insisted as strongly as they do now on separating disciplines oriented towards culture from those oriented towards nature. In the current view, the latter disciplines have full responsibility for identifying the invariants whose universal character is thought to depend on their biological roots (and especially on the biological underpinnings of the mind) or, put another way, on the effects that constraints determined by the biological characteristics of human beings (who eat, reproduce, die and so on) bring to bear on life in society. The disciplines oriented towards culture, in contrast, have the task of establishing the inventory of what is left over, that is, the differences between human groups that are thought to result chiefly from their adherence to different systems of belief. In the order of nature, everything is understood to be the same everywhere; in the order of culture, everything is understood to be different. It was precisely in reaction against this split, which positivism had made so compelling, that general sociology and social anthropology were constituted over a century ago, with a project defined from the start as comparatist. General sociology and social anthropology thus took their principal task to be cataloguing the ways in which practices that appeared to manifest a kind of family relationship could nevertheless be substantiated differently in different societies (in the case of Émile Durkheim and his followers, for example, these practices would include sacrifice, prayer, exchange, kinship, practices of classification, oaths, crime and so on). The same can be said for psychoanalysis: at least after its encounter with cultural anthropology, and without abandoning its fundamental concepts (the unconscious, repression and so forth), psychoanalysis had undertaken to examine, for example, how different schemas for organizing unconscious drives could correspond to different practices of socialization, or how taking into account the tensions proper to each culture made it possible to trace pathways leading from collective myths to individual dreams and vice versa.

With respect to my topic, the social anthropologist and psychoanalyst George Devereux was the first to undertake a systematic study of the practice of abortion by considering it both in its general dimensions and in the specific forms it has taken in different societies. As Devereux explains in his introduction to A Study of Abortion in Primitive Society, his primary aim was theoretical, or rather ‘methodological’ (Devereux 1955). He sets forth four goals: (a) to provide empirical support for the validity of the ‘axiom that cultural diversity demonstrates the tremendous plasticity and variability of human behavior’; (b) to furnish empirical data in support of ‘the methodological thesis that the intensive analysis of the context and implications of a particular institution in a single tribe … can … yield universally valid conclusions’ (with reference to Durkheim and Freud) and, conversely, to show that ‘the self-same propositions could also be derived from a study in breadth of the variations of the same culture-trait or institution in a large number of societies’, in such a way as to justify ‘simultaneously, and by identical means, both studies in depth and studies in breadth’ (ibid., vii); (c) to demonstrate the compatibility of the anthropological and psychological approaches, owing to the fact that a precise correspondence exists between cultural behaviours and affects1 (Devereux views abortion as a practice that lends itself particularly well to the demonstration he intends to conduct because – and it will become clear why this feature is important for my project – ‘abortion does not occupy anywhere a focal position in culture’ [ibid., viii], so that, not being the object of ‘culturally’ precise and explicit prescriptions, it leaves wide open the possibility of a great diversity of individual behaviours); finally, (d) to present a more or less exhaustive set of materials about abortion in order to facilitate future research.

George Devereux gathered (and methodically published in the annex to his book) a corpus bearing on four hundred ‘pre-industrial societies’. He used Yale University’s Human Relations Area Files as his principal source, under the guidance of Ralph Linton (who joined Yale’s Department of Anthropology late in his life), and especially George Peter Murdock, the anthropologist who had set up the Area Files, starting in 1938, with the goal of developing a comparative and ‘transcultural’ anthropology. Devereux completed his documentation by drawing on his personal archives and on oral and written communications supplied by various colleagues. The Area Files are a huge set of dossiers derived from an exhaustive study of virtually all known anthropological literature (found in books, articles or unpublished manuscripts) and also of what can be called an important pre-anthropological literature (narratives written by travellers, missionaries, colonial administrators and so on) deemed to have sufficiently reliable documentary value. The data were recorded in these files according to a dual classification system: by cultural zones and societies on the one hand, by themes on the other. There is an entry devoted to questions pertaining to pregnancy and abortion and a subentry indexing abortion.2 Since Devereux constituted his corpus, the Area Files have continued to grow. There is a copy in the Laboratory of Social Anthropology in the Collège de France, in three different formats (depending on the age of the files and the format of digital transcription): the material is available on paper, on CD-ROMs and by subscription on the Internet, so that by consulting these files one can complete the information in George Devereux’s work (or verify it, if doubts arise).3

The data contained in the Area Files do not lend themselves well to systematic – let alone statistical – treatment, chiefly because the information is very heterogeneous and of unequal value: it was collected in different periods in widely divergent societies and according to disparate methods by people who were as dissimilar in their ethnographic skills as in their theoretical orientations. As Devereux notes, observations about the same society made by different observers are sometimes in conflict. As a result, one must resign oneself to regarding assertions drawn from these materials as assumptions rather than as factual certainties.

Without necessarily sharing Devereux’s theoretical presuppositions or assenting to all the developments (which sometimes contain remarkable intuitions) in a book that is rich in detail but rather disconcerting in structure, one can nevertheless use some of the observations and remarks included in this survey, along with the results of complementary investigations into the Area Files, as a basis for sketching out a rough framework apt to highlight some of the principal questions that the practice of abortion raises for sociology. For my part, I have chosen to emphasize, at least as a working hypothesis, four properties of abortion that are not explicitly singled out by Devereux, or at least not stressed, but towards which numerous indications in his material – and also, occasionally, in his analyses – nevertheless converge.

A practice universally understood to be possible

A first property, this one clearly affirmed by George Devereux, is the presumably universal character of the practice.4 Devereux notes that information about abortion is available for about 60 per cent of the societies included in the Area Files. This of course does not mean that abortion is absent from the remaining 40 per cent; given the very heterogeneous character of the information in the files, it simply means that ethnographers have not always taken this dimension of existence into account in their monographs, or that their informants did not mention it. What seems universal, moreover, is less the practice of voluntary abortion – which is very unevenly attested, it would seem, varying according to the society and the era (although solid statistical data can almost never be established) – than acknowledgement of the possibility of this practice. There are no examples in the corpus of a situation in which an informant, male or female, when questioned on this point, did not know what the question referred to or who, when an explanation was offered, expressed astonishment that such a thing could exist. The possibility of making a foetus exit the womb before birth for the purpose of destroying it thus seems to belong to the fundamental framework of human existence in society.

The means used to this end are themselves very numerous; they are fairly well known today, not only in societies studied by ethnology but also in ancient societies, especially those of Graeco-Roman antiquity, as well as in medieval and modern Western societies,5 in China and in Japan (La Fleur 1992).6 The most widespread procedures involve the use of abortifacient medications, usually drawn from plants (with emetic, laxative, purgative or astringent effects among others; these are known in practically all societies for which information is available); the use of mechanical means, either internal (introducing a stalk or stick into the vagina) or external (jumping up and down, striking the abdomen or compressing it with a belt, applying hot materials such as water, ashes or stones to the abdominal wall, and so on); or a combination of these procedures (such as introducing medications into the vagina or manipulating the sex organs). These various chemical or mechanical procedures have to be understood in each case in relation to the local theories about reproduction and gestation on which confidence in the effectiveness of a given procedure is based. Magical means are also used (sitting under a certain tree, consuming a certain food or drink, wearing an amulet, and so on): customarily distinguished from mechanical and chemical means, recourse to magic very often requires carrying out a transgressive act (for example, eating a forbidden food). Devereux points out the possible existence, among the Hopi Indians, of a means he calls ‘psychosomatic’, in which the intense desire to abort is viewed as having abortifacient effects in and of itself. In most of the societies about which information is available, the means available for the practice of abortion seem to belong to common knowledge, even if certain persons (who usually act as midwives as well) are considered more knowledgeable or more skilful than others. In fact, many of the means used for abortion are hard to handle and known to be more or less dangerous. They arouse fear. And yet this does not keep them from being called on when the need to abort appears compelling.

The object of general condemnation

A second property of abortion is that it is very often subject to condemnation.7 Only rarely is abortion accepted as a matter of principle, even in societies where it is frequently practised. Reactions go from shocked disapproval to the most violent indignation towards this ‘shameful’ or ‘horrible’ act; moreover, its practice is often attributed to neighbouring peoples or to the inhabitants of bordering villages while presented as unknown ‘among ourselves’. Such indignation does not seem to be merely feigned in order to satisfy the expectations of a foreign observer who is deemed a priori to be opposed to abortion (for example, in cases where the information comes from travel narratives or missionaries’ recollections); it is also noted in reports by highly professional ethnographers. Nor is it an attitude specific to men, for women often manifest the same ‘horror’ when the act is mentioned, although their indignation might be interpreted as a sign that they have internalized masculine values. Abortion is not something one talks about, at least not without embarrassment; when people do discuss it, their intent is most often to make clear that, even though they know that the practice exists, it surely cannot concern their intimate circle – members of their kinship group – or even the collective body to which they belong.

The degree of disapproval expressed ultimately seems to vary not only from society to society but also according to circumstances within a given society, in relation to a casuistics that depends on cultural characteristics: for example, generally speaking, disapproval may be less pronounced when incest or coupling with an animal is suspected (among the Navajo), or when it is presumed that the mother will give birth to an illegitimate child (especially in patrilinear societies), or when a multiplicity of potential fathers makes it impossible to identify the true father and obliges him to marry the pregnant woman (except in societies that recognize multi-paternity8), or when the mother is thought to have been impregnated by a demon and destined to give birth to a monster (among the Jivaro and many other groups9). References to attenuating circumstances based on characteristics of the foetus – features that were unknown and unknowable before the advent of modern imaging techniques – must not be taken too literally, moreover, as would be the case if they were linked to specific controlled tests; they are best viewed rather as sketching the contours of an argumentative register that can be mobilized whenever someone seeks to attenuate the disapproval directed towards abortion. Thus the argument that a woman had an abortion because the child she would have delivered would have been illegitimate (in many traditional societies, this meant that it would have had neither a name nor a kinship group10) always seems ‘self-evident’ in some respects, even though in practice there are always other possibilities, such as finding the pregnant woman a husband who agrees to take on the paternity of the child she is carrying.

Tolerance for abortion

A third important property of abortion can be seen in the fact that condemnation of the practice quite often seems to go hand in hand with considerable tolerance for it on the part of the very persons who express indignation when it is mentioned. Although it is not hard to find examples, in various domains, of gaps between articulated norms – or laws, in societies where a written body of law exists – and the pragmatic expression of their implementation, in the case of abortion the gap between the rule and its application seems particularly striking, and it seems to be found in one form or another in most of the societies for which information is available. Only very rarely are serious efforts made to identify, pursue and punish the persons responsible. And we shall see in chapter 3 that this feature is also characteristic of Western medieval and modern societies dominated by Christian churches whose Fathers had condemned abortion, but in which, before the second half of the nineteenth century, roughly speaking, the authorities could fulminate against that act or call for its prohibition without having much concrete effect: their condemnations neither triggered police investigations nor modified practices.11 The fact that women who had abortions and those who helped them do so were most often not pursued or punished does not mean that the practice went unsanctioned, however. In many societies, the informants mention the existence of sanctions, but these are either immanent to the act itself (such as sterility) or diffuse penalties that affect the kinship group or even the collective body as a whole12 (for example, in the wake of an act of vengeance carried out by the spirit of the aborted foetus), as is often the case when transgressive practices affect the order of the world.

A compilation of ethnographic data makes it possible to identify another intriguing feature that is congruent with the indignation–tolerance pairing. Where abortion is practised, it is usually carried out in secret, or at least in the shadows. But most often it appears as what can be called an open secret. This situation can draw our attention to an opposition that plays an important role with regard to our object, one whose implications I shall try to develop in chapter 3: the opposition, analysed in depth in Pierre Bourdieu’s ethnological work, most notably in the texts devoted to kinship, between what belongs to the official order and is endowed with a ‘public, solemn, collective’ character, and what stems from the unofficial order and is condemned to a ‘shameful’ or even ‘clandestine’ mode of existence.13 This opposition may involve the distribution of different types of action or different forms of power. In Pierre Bourdieu’s study of Kabyl society, it is associated with the opposition between men and women, between masculine society and feminine society. Men hold official power over what is explicitly collective and public, and in particular over representations of kinship (Bourdieu emphasizes that the realm of kinship has an eminently political character in traditional societies); women exercise a power that, while genuine (especially where marriage is concerned, according to Bourdieu), remains hidden and leaves ‘the appearance of power … to men’ (1972, 41).

The distinction between the world of men, the official realm of written or common law, religion, politics and the public square – the exterior world – and the world of women, the unofficial realm of the home, magic and witchcraft – the interior world – has been thematized by many anthropologists who have studied forms of masculine domination,14 and it seems to be made quite generally in human societies. It encompasses first and foremost everything that has to do with gestation and birth, a realm that in most traditional societies is confined to secrecy within the female context, the one situated in the home (inside as opposed to outside, consistent with the private–public opposition); within the home itself there is a space reserved for women,15 one that in many societies (for example, the Achuar, studied by Philippe Descola [1996], or the Baruya, studied by Maurice Godelier [1996]) is off limits to men.16 The space of the home is exempt from the political logic of the polity, that is, from the realm of justice and, more profoundly, from ‘society’ in the modern sense of the word.17

This distinction between the official and the unofficial, it must be noted, is particularly relevant to my topic. Among the set of practices associated with the feminine pole, abortion is probably one of those most forcefully kept out of the public space; it takes place in the shadows, exclusively among women. This explains why information about it is so scarce and so difficult to verify, at least in comparison to the information available about kinship nomenclature, for example; the latter is part of masculine knowledge and can be communicated fairly readily by male informants to anthropologists of the same sex. (Until the feminization of the profession of anthropology over the past several decades, moreover, it is clear that little progress could be made towards developing an anthropology of the practices of engendering.) Expanding on Bourdieu’s distinction between official masculine power and unofficial feminine power, I would suggest that abortion constitutes the very paradigm of properly feminine power (as opposed to power over kinship and its representations), especially in traditional societies where the homologies between political space and domestic space confer great significance on all practices related to procreation. But this power by itself remains illegitimate and hidden, whether it is used without men’s knowledge and in order to do them harm (for example, to avenge an infidelity on the father’s part by eliminating his progeny) or, on the contrary – and both types of examples appear among the ‘motives’ enumerated by George Devereux – with their complicity and in their interest, notably sexual (so that men do not have to comply with prohibitions on relations during pregnancy and lactation).

But the distinction between what is official and public and what is unofficial and tacit is not exclusive to the realm of action. It can also point to different modalities of knowledge, as in Malinowski’s famous example of the young Trobriander man who had violated the rules of exogamy with his maternal cousin, the daughter of his mother’s sister. This fact was known and condemned, but no consequences ensued until the girl’s lover insulted the guilty man in public, accusing him of incest before the entire community. The next morning, in sight of the assembled community, the young man in question climbed up a coconut tree, jumped off and fell to his death.18 What is at stake, then, does not involve a difference in information (the facts are the same, whether they are known unofficially or officially), but has to do with the order of accusation and, consequently, with the imputation of responsibility. While some facts may be known unofficially and remain without consequences as long as no one decides to make the matter public, with the attendant risks to the accuser, the situation looks quite different if a public accusation is introduced, setting in motion a process that has to proceed to a determination of the truth or falsity of the charge and thus must go all the way to the imposition of sanctions, either on the guilty party or on the person who made an unjust accusation. Where abortion, although officially condemned, is said to be unofficially tolerated, a process of this sort is involved. Abortion is tolerated not only because it takes place in a context – the female universe – that cannot be penetrated from the public space, not only because no efforts are made to find out what goes on in the world of women, but also because, even though people in some sense know perfectly well what is going on, they can behave as if what they know is irrelevant; put another way, they can close their eyes and behave as if they do not know.

The dearth of representations

To conclude this rapid inventory of the general features of abortion that seem to me particularly pertinent for setting up a problematics, I shall posit a fourth property whose existence cannot readily be demonstrated; this of course makes my case more difficult, even though the existence of this property strikes me as highly probable, if not certain. My hypothesis is that abortion has been very broadly underrepresented (this may well be what Devereux means when he asserts that ‘abortion does not occupy anywhere a focal position in culture’). To establish the existence of a property such as underrepresentation, one would have to be able both to rely on some sort of inventory of all known representations of abortion and also to provide an operational meaning for the idea that there is a somehow ‘normal’ level of representation for various types of practices. Despite the difficulties, however, it is hard to dismiss the impression that in this case there is indeed a deficit of representation or, more generally, a collective reluctance to transcribe abortion and the aborted foetus in a symbolic register.

Neither abortion nor aborted foetuses appear to have been represented in objects or images with any frequency: not in primitive or traditional societies, not in antiquity or in Western painting. (In Japan, though, one does find representations of the Kappa, a monster resembling a slain newborn and/or a foetus, and there are also hekeshi, figurines representing aborted infants to which offerings are sometimes made [Jolivet 2002].) Abortion may be a practice difficult to depict; we might suppose that it could be more easily narrated. Yet it seems that (at least until fairly recently) it has also been absent in narratives, or has been represented in stories only in veiled terms. Direct references to abortion in myths, folk tales or literary works are hard to find, at least prior to the era of naturalist novels. In the latter, we occasionally come across scenes involving abortion written in a critical tone (abortion is sometimes more or less indistinguishable from infanticide). These texts appeared after the prohibition of abortion (and thus its inscription into nineteenth-century law) had conferred legal and medical visibility on it as a ‘social scourge’ associated with alcoholism and prostitution in the lower classes, according to a thematics inspired by public health reformers (we shall return to this point in chapter 3). Direct representations of abortion in literature and film have become much more common in recent decades. However, even now, the inclusion of abortion in a narrative or in images is most often accompanied by political and moral justifications that inscribe it within a critique of the existing order; it is rarely presented straightforwardly, as if it ‘went without saying’. Although it is a common practice (until the recent development of contraception, there was probably one abortion in France for every two births; today the ratio is probably one abortion for every three or four births, depending on the year and the mode of statistical accounting [see chapter 4]), abortion is never treated as an ordinary occurrence.

It is also noteworthy that, with the exception of texts related to medicine, abortion is rarely mentioned in philosophical works; indeed, it is entirely absent from classical philosophy. The possibility of abortion seems to have had no impact at all on the concepts of the human condition developed in Western philosophy, unlike suicide, for example (which has also been an object of predilection for the field of sociology from the outset). To be sure, there are references to abortion in some prescriptive Western texts from the fields of religion, law and medicine, especially among certain Church Fathers (we shall see examples in chapter 3), but these are relatively rare, often scantily developed and probably reserved for a very limited audience.

Finally, abortion does not seem to be associated anywhere with any form of ritualism or symbolism. Aborted foetuses are crudely buried, burned or drowned, with no specific words or gestures of accompaniment. Nevertheless, in many societies (and perhaps in nearly all), there seems to be a belief that aborted foetuses are transformed into spirits, sometimes particularly malevolent and dangerous ones (notably among the Hopi) against which people would do well to protect themselves by uttering certain prayers.19

The fact that abortion has largely been kept out of the sphere of representation can be linked with two of the properties I have already mentioned: on the one hand, its association with the unofficial world of female practices; on the other hand, the fact that it is generally subject to condemnation and can thus be considered (officially) transgressive. One can find examples of strictly female practices that have been widely represented (as have some homosexual practices); it is even easier to find examples of unquestionably transgressive practices that have been represented in myths, stories, pictures and so on, owing precisely to their transgressive nature. This is the case with incest, intrafamilial murder and also infanticide; according to Muriel Jolivet (2002), the last of these was frequently depicted on the walls of Buddhist monasteries between the Edo and Meiji eras, a period when the practice was very widespread in Japan.

The underrepresentation of abortion doubtless has to be associated more generally with the virtual absence of the foetus, until recent times (we shall return to this point in chapter 5), from the field of social relations, where we find not only living human beings but also, depending on the context, dead people, animals or plants, supernatural beings, futuristic creatures and so on. To be sure, many so-called primitive societies have ideas about conception, gestation and procreation, even if they are not all developed to the same extent (Godelier and Panoff 1998). Similarly, in Western societies, from classical antiquity on we find ideas developed essentially in the field of medicine (with echoes in natural philosophy and theology) about procreation and thus about the foetus.20 But these ideas remain confined to relatively limited realms of knowledge and do not give the foetus any real presence in society. Socially, for ordinary persons and for institutions, the focus is predominantly on pregnant women and on infants. Until recently, foetuses were not recognized as beings endowed with specific identities that had value in and of themselves; this is attested most notably by the paucity of representations of foetuses, the limited character of the legal corpus concerning them, and the virtual absence of rituals associated with their exit from the world of the living and their entry into the realm of the dead, whether their departure can be attributed to spontaneous miscarriage or to deliberately provoked abortion. If we set aside figurines and images intended for the instruction of doctors and midwives, which became particularly abundant starting in the second half of the eighteenth century (Gélis 1988), foetuses are strangely absent both from visual representations (religious images representing Christ in the Virgin’s womb are quite rare21) and from poetry, literature and myth – indeed, from discourse in general. Similarly, despite specifications in Roman law designed to settle thorny inheritance issues (Thomas 1996), foetuses have historically taken up very little space in law or religion; they are virtually absent from the polity, from politics, and more broadly from the symbolic constructions that were superimposed on the social order and that in fact constituted the social order as such. When a foetus emerged prematurely from the womb and did not survive, it was not given a name, nor was it the object of any funerary rites. Now, being the object of funerary rites is a very important index of belonging to human society (scholars of prehistory see rituals of this type as criteria which they credit with a determining role in the process of humanization [Tattersall 1998]). One can hardly even say that the foetus ‘dies’; it is as though one has to be born, and to be born alive, in order to be able to die. This absence is no less remarkable in the history of Western philosophy, where – except for certain Greek texts that deal with nature as phusis22 and later developments of this notion in natural philosophy – the foetal state of humanity has scarcely been taken into account. This is particularly true of classical philosophy, which looked steadily towards the horizon of mortality to construct an ontology, even a political ontology,23 of human beings; with rare exceptions, it completely neglected not only the foetus, but also, more generally, the very fact of birth,24 as Paul Ricoeur notes in his studies of the relationship between memory and history (Ricoeur 2004, 357).

Questions abortion raises for sociology