poststructuralist discourse analysis ||

Poststructuralist Discourse AnalysisSubjectivity in Enunciative PragmaticsJohannes AngermullerISBN: 9781137442475DOI: 10.1057/9781137442475Palgrave Macmillan

Please respect intellectual property rights

This material is copyright and its use is restricted by our standard site license terms and conditions (see palgraveconnect.com/pc/connect/info/terms_conditions.html). If you plan to copy, distribute or share in any format, including, for the avoidanceof doubt, posting on websites, you need the express prior permission of PalgraveMacmillan. To request permission please contact [email protected].

Poststructuralist Discourse Analysis

10.1057/9781137442475 - Poststructuralist Discourse Analysis, Johannes Angermuller

Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Postdisciplinary Studies in DiscourseSeries edited by Johannes Angermuller, University of Warwick, UK, and Judith Baxter, Aston University, UK

Postdisciplinary Studies in Discourse engages in the exchange between discourse theory and analysis while putting emphasis on the intellectual challenges in dis-course research. Moving beyond disciplinary divisions in today’s social sciences, the contributions deal with critical issues at the intersections between language and society.

Titles include

Johannes AngermullerPOSTSTRUCTURALIST DISCOURSE ANALYSISSubjectivity in Enunciative Pragmatics

Postdisciplinary Studies in DiscourseSeries Standing Order ISBN 978–1–137–50038–0 Hardback(outside North America only)

You can receive future titles in this series as they are published by placing a standing order. Please contact your bookseller or, in case of difficulty, write to us at the address below with your name and address, the title of the series and the ISBN quoted above.

Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire RG21 6XS, England


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Also by Johannes Angermuller

THE MOMENT OF THEORY

Rise and Decline of Structuralism in France

THE DISCOURSE STUDIES READER

Main Currents in Theory and Analysis (co-editor)

POSTDISCIPLINARY STUDIES IN DISCOURSE (co-editor)


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

This page intentionally left blank


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Poststructuralist Discourse AnalysisSubjectivity in Enunciative Pragmatics

Johannes AngermullerUniversity of Warwick, UK


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

© Johannes Angermuller 2014

All rights reserved. No reproduction, copy or transmission of thispublication may be made without written permission.

No portion of this publication may be reproduced, copied or transmittedsave with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6–10 Kirby Street, London EC1N 8TS.

Any person who does any unauthorized act in relation to this publicationmay be liable to criminal prosecution and civil claims for damages.

The author has asserted his right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988.

First published 2014 byPALGRAVE MACMILLAN

Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited,registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS.

Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010.

Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world.

Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries.

ISBN 978–1–137–44246–8

This book is printed on paper suitable for recycling and made from fullymanaged and sustained forest sources. Logging, pulping and manufacturingprocesses are expected to conform to the environmental regulations of thecountry of origin.

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

Library of Congress Cataloging-in-Publication DataAngermuller, Johannes, 1973– author.Poststructuralist discourse analysis : subjectivity in enunciative pragmatics / Johannes Angermuller, University of Warwick, UK.pages cmSummary: “French thinkers, such as Lacan, Althusser, Foucault and Derrida, have been widely perceived as theorists of the linguistic turn. Yet, the linguistic and semiotic traditions which informed the theoretical imagination of these theorists so decisively have hardly been accounted for outside French linguistics. This book presents past and present developments in French discourse analysis, while also paying special attention to the development of enunciative pragmatics, which hinges on the discursive construction of subjectivity. Five textual fragments by these theorists, all written around 1966 when the controversy over structuralism was at its height, are analysed in detail in relation to the question of how theoretical texts are used in discourse where one constantly needs to define one’s position vis-a-vis others. The book will be valuable to students, researchers and practitioners within discourse analysis, pragmatics, linguistics and semiotics, as well as all those interested in the analysis of the social production of

meaning”—Provided by publisher.Includes bibliographical references.ISBN 978–1–137–44246–8 (hardback)1. Discourse analysis—History. 2. Pragmatics—History. 3. Linguistic analysis—History. 4. Poststructuralism—France. 5. Language and languages—Pronunciation.6. Sociolinguists. I. Title. P302.A534 2014401'.41—dc23 2014025133

Typeset by MPS Limited, Chennai, India.

The cover picture represents the interior of the old National Library of France (BNF), where a great deal of poststructuralist discourse was produced until the 1980s.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

vii

Contents

List of Figures viii

Acknowledgements ix

1 Introduction: Poststructuralism and Enunciative Pragmatics 1

2 A History of Discourse Analysis in France 7 From discursive formation to enunciative heterogeneity 7 Discourse as utterance and enunciation: the field

of enunciative pragmatics 24 Elements of enunciative discourse analysis: indexicality,

polyphony, preconstruct 38

3 A Methodology of Discourse Analysis 54 From understanding to analyzing discourse 54 A discourse analytical research design 58 Polyphony and scenography: the activity of the reader 63

4 Analyzing Intellectual Discourse: Variations on the Critique of Humanism 72

Five protagonists of theoretical discourse 72 Jacques Lacan: the return to (the subject of) Freud 78 Louis Althusser: Marxism as anti-humanism 93 Michel Foucault: the end of the age of ‘Man’ 109 Jacques Derrida: the metaphysics of the text 117 Tel Quel: narrating the revolution 129

5 Conclusion: The Subject of Discourse 140

Notes 146

References 154

Index 163


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

viii

List of Figures

2.1 The self-referential character of the utterance 23

2.2 The system of adjectives 42

2.3 The images of locutor, allocutor and the third party 44

4.1 The narrative structure of the vanguard/rearguard discourse 137

5.1 A three-floor model of discursive subjectivity 142


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

ix

Acknowledgements

This book, which initially came out in French (Analyse du discours poststructuraliste. Les voix du sujet dans le langage chez Lacan, Althusser, Foucault, Derrida et Sollers. Limoges: Lambert Lucas 2013), greatly ben-efited from the discussions with my friends at CEDITEC (Paris, Créteil) and with Dominique Maingueneau in particular, whose support was crucial for this book project. I would like to thank the members of DiscourseNet (especially Felicitas Macgilchrist, Yannik Porsché, Jaspal Singh and Jan Zienkowski) as well as my wonderful colleagues from the Centre for Applied Linguistics at Warwick for their helpful com-ments with the English version of this book. I also want to specially thank Daniel Marwecki and Clare Simmons, who helped me with some parts of the translation, as well as Gerard Hearne for proofreading the manuscript. The work leading to this publication has received fund-ing from the European Research Council under the European Union’s Seventh Framework Programme (FP7/2007-2013)/ERC grant agreement no. 313172 (DISCONEX).


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

This page intentionally left blank


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

1

The controversy over structuralism reached its peak around 1966–7, when a new generation of French theorists, including Jacques Lacan, Louis Althusser, Michel Foucault and Jacques Derrida, came onto the intellectual scene. Inspired by Marxism and psychoanalysis, these intel-lectuals are today known for their critical epistemologies that point to the symbolic constitution of the subject and insist on the constitutive role of language in society. Yet while these theorists have been greeted as representatives of the linguistic turn in the social sciences and humanities, the linguistic and semiotic traditions themselves, which have so decisively stimulated the imagination of the interdisciplinary theoretical debate, are hardly known outside a rather restricted circle of specialists. Not surprisingly, these thinkers have often been perceived as sweeping theorists of language in society, but of rather limited help when it comes to analyzing linguistic and semiotic texts.

By making key canonical texts from French Theory the object of rigorous linguistic scrutiny, Poststructuralist Discourse Analysis attempts to bridge this gap and to present discourse analysis as it has developed in France since the late 1960s, notably enunciative pragmatics, also known as the linguistics of enunciation (énonciation). For some lin-guists, ‘enunciation’ refers to fundamental operations of language as a grammatical system (for example in Culioli), whereas for others it des-ignates the pragmatic dimension of language use. Generally speaking, enunciative pragmatics asks how utterances (énoncés) mobilize sources and voices, speakers and points of view, locutors and enunciators at the moment of enunciation. Following Benveniste’s famous defini-tion of enunciation as the ‘enactment of language through an act of individual usage’ (Benveniste 1974: 80),1 enunciative pragmatics asks how linguistic expressions, markers, traces and shifters (marqueurs,

1Introduction: Poststructuralism and Enunciative Pragmatics


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

2 Poststructuralist Discourse Analysis

repères, indices énonciatifs …) reflect those who speak in a the context of enunciation. In line with the pragmatic turn in French linguistics since the late 1970s, the growing interest in the question of enuncia-tion has contributed to a new enunciative strand in pragmatics which has not only helped to direct attention to questions of subjectivity, context and communication but also to the sociohistorical dimensions of discourse more generally, as can be seen in Foucault’s project for an ‘enunciative discourse analysis’ (1969: 143[123])2 and Maingueneau’s enunciative-pragmatic work on genre and scenography (1993). One might also think of the often overlooked theorizations of enunciation in Deleuze/Guattari (1980), Lyotard (1988) and some of Lacan’s semi-nars of the 1960s and 1970s (for example 1973).

As there is no direct English equivalent of énonciation, translations have often failed to render the rich tradition of pragmatics that has characterized French linguistics in the post-war period. While in English ‘enunciation’ usually means the articulation of speech, especially in a clear and distinct way, the linguistic term énonciation designates lin-guistic activity more generally. The correlate of énonciation is énoncé, that is a specific semiotic realization of a communicative act, which often has the form of a phrase. In English, énonciation and énoncé are not easily distinguished as both are sometimes interchangeably trans-lated by ‘utterance.’ In the following, I will use ‘utterance’ in the sense of énoncé and take ‘enunciation’ to be the equivalent of énonciation. Much more could be said about the surprisingly difficult task of translating linguistic terminology from French into English, and vice versa. Suffice it to say that with all these terms—that is énonciation: the act or process of using language; énoncé: the utterance as a product of this process; énoncer: utter, say, voice, speak; énonciatif: communicative, pragmatic, discursive, indexical, subjective; énonciateur: speaker, voice, source, perspective—all deriving from the common root énonc-, a new and distinctive tradition has formed, that is enunciative pragmatics, which accounts for the construction of subjectivity in the many voices of discourse.

Having emerged from structuralist linguistics and semiotics, enunciative pragmatics connects to various disciplinary fields and traditions at the crossroads of language and society (Angermuller et al. 2014). One can cite the philosophy of language as seen in the later work of Wittgenstein (1997), who criticizes the idea of pure language and points out that we can-not use language without engaging in some sort of creative action. With an interest in the question of how, in what modalities and under what circumstances utterances are produced, enunciative pragmatics follows


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Introduction 3

Austin’s idea (1962) that utterances, whether oral or written, reflect speech acts produced by somebody with a specific illocutionary force. Moreover, for qualitative social scientists, enunciative pragmatics adheres to the idea that language is always tied up with practices in which social identities, relationships and subjectivities are constituted. Thus, one can ask how individuals are constructed by means of ‘membership categorization devices’ and defined as social beings in turn-taking sequences (Sacks 1986) or how polyphonic instances such as animator, principal and author are orchestrated in the interactive situation (Goffman 1981). Yet, unlike actor-centered strands in social research, which place emphasis on observable social practices in their setting, enunciative pragmatics does not claim to have immediate access to the practice of using language. Enunciative prag-matics deals with written or spoken utterances circulating in a discursive community rather than with meaning-producing subjects and situated practices.

Enunciative pragmatics belongs to the universe of linguistic prag-matics. As a subfield of linguistics, pragmatics catalogues linguistic expressions that reflect the use of utterances by somebody ‘here’ and ‘now’. A range of linguistic phenomena, such as deixis, presupposi-tion, argumentation, implicatures and negation, testify to how this activity is linguistically encoded. Here, language serves to construct relationships, as has been argued by politeness theorists who point to the desire in any communicative action to save or improve one’s face (Brown and Levinson 1987), like the research on expressing stance, style and identity. In a systematic way, the pragmatic idea of language as social action was theorized in the Systemic-Functional Linguistics of Halliday (1978) which, crucially, has contributed to social semiotics (Hodge and Kress 1988; Leeuwen 2005), appraisal theory (Martin and White 2005) and Critical Discourse Analysis (Fairclough 1992). From this perspective, the individual needs to make certain linguistic choices, given the communicative functions to be fulfilled in a social situation (see also the functional pragmatics of Ehlich 2007).

The major features of enunciative pragmatics vis-à-vis these other strands in linguistic pragmatics may be summarized as follows: (1) an emphasis on the opaque materiality of (mostly written) texts whose meaning cannot be read from the surface; (2) the break with mean-ingful experience and subjective interpretation through an analytical practice which highlights the formal linguistic markers of enunciation (for example I, but, not …); (3) a preference for the non-subjectivist study of discursive subjectivity over more semantic, hermeneutic or content-related approaches.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Enunciative pragmatics prolongs the anti-humanist intellectual heritage of structuralism and poststructuralism and breaks with subjec-tivist conceptions of meaning-making. The subject is not a source of meaning; rather, as Bakhtin (1963) argues, it is a product of the many voices staged by the utterances of a discourse. At the same time, enun-ciative pragmatics offers a wealth of analytical tools to account for the positions subjects occupy in discourse. Therefore, the methodology of enunciative pragmatics allows us to analyze how, in the act of reading and writing, utterances are contextualized with respect to who speaks, when and where. Inspired by the critique of the sovereign subject in Foucault (1969), it shows how subjectivity is constructed in a multitude of voices, sources and speakers and tied to the linguistic forms and formal markers which organize the enunciation.

It should now be clear that the title of this book—Poststructuralist Discourse Analysis—refers not only to a theoretical discourse which is sometimes labeled ‘poststructuralist’ (even though this label is hardly known in France, see my sociohistorical account of French intellectuals in France, Angermuller 2015). Following the critical constructivist spirit that has come to pervade sociolinguistics, conversation analysis and Critical Discourse Analysis (Baxter’s Feminist Poststructuralist Discourse Analysis can be cited as an example, 2003), it also outlines enunciative pragmatics as a poststructralist framework which breaks with the static, homogenizing and abstract approaches to language one commonly associates with structuralism. Yet, while both French Theory and enun-ciative pragmatics have emerged as a reaction to structuralism, there has been little exchange between them, the first having its base in the literary and cultural field and the interpretive social sciences, the latter in French linguistics.

In responding to the demand for this long overdue encounter, this book delineates the contours of a poststructuralist discourse analysis. In this view, discourse presents itself as an open and dynamic terrain of protean perspectives and nested voices in which the discourse participants are confronted with the difficult practical task of defining their place in discourse. Discourse is considered to be a linguistically encoded practice of positioning oneself and others and creating dis-cursive relationships with others within a play of polyphonic voices. As opposed to a structuralist vision of a grand discourse from above, we will zoom in to the level of small textual passages and discover the complex polyphonic play of voices staged by the utterances of discourse. The objective is to account for the traces the subject leaves in its utterances—a subject which must not be confounded with a


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Introduction 5

constituted origin or source of meaning (that is an ‘author’ or ‘actor’) but which should rather be seen as a set of shifting and unstable places and positions which the discourse participants process as they enter dis-course. Yet even though this monograph deals with texts, one must not forget that intellectual discourse as a positioning practice is not only linguistically but also socially constrained. If it catalogues the linguistic resources through which the discourse participants negotiate their positions in intellectual discourse, the social, institutional and political resources mobilized in the positioning practices of their field have been accounted for elsewhere (Angermuller 2015).

With the discourse analysis techniques of enunciative pragmatics, it aims to reveal the gaps and fissures, the bugs and glitches, the conflicts and antagonisms in discourse. Thus, Poststructuralist Discourse Analysis can be said to be directed against three theoretical adversaries: the humanist, who believes in autonomous subjects as the source and ori-gin of social and linguistic activity; the realist, who believes in objective realities that exist independently of discourse; and the hermeneuticist, who believes in a world of transparent and homogeneous meaning. It is critical of silencing the voice of the Other, of policing resistant practices and controlling disobedient knowledge, of homogenizing the social through representations of ‘the’ society, ‘the’ culture or ‘the’ discourse.

This book consists of four parts. To map the evolution of discourse analysis in France, Chapter 2 will conduct a detailed discussion of the enunciative-pragmatic turn in French linguistics. Then, in Chapter 3, I will sketch out a poststructuralist methodology in discourse research which investigates the ways in which a written text refers to its con-text. In Chapter 4, I will apply the discourse analytical instruments to a sample of key theoretical texts from around 1966 (Lacan, Althusser, Foucault, Derrida, Sollers), which will reveal the subtle play of voices and references via which these theorists negotiate their positions in discourse. In the conclusion, I will plead for ongoing critical reflection on the subject by taking into consideration our own symbolic practices when we read, speak and write.

In bridging pragmatics and poststructuralism, this monograph addresses all those who are interested in reconciling discourse theory with discourse analysis. By radicalizing the critical constructivist ten-dencies in sociolinguistics and Critical Discourse Analysis, it responds to a need of rigorous analytical instruments for analyzing language in its social dimensions. Even though the book is firmly rooted in linguistics, one does not have to be a linguist to adopt its frame of analysis. Philosophers of language may be interested to see how to account for


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


discourse by departing from utterances as the smallest units of analysis. Literary critics can find inspiration in an approach that accounts for the question of auctoriality and intertextuality. Cultural analysts will observe the creative appropriation of subject positions in a polyphonic play of voices. And social scientists will discover how social order and agency are constructed and represented through the textual markers of polyphony. Yet, just as with any other text, the meaning of this text, too, needs to be constructed by readers coming from a discursive com-munity whose background is more or less out of reach, at least for the originator of this text.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

7

From discursive formation to enunciative heterogeneity

The field of discourse analysis comprises a variety of traditions and approaches. As their smallest common denominator, we can consider the assumption that signs, sentences, and texts do not have any inherent meaning, but that meaning presupposes a context and that linguistic activity is not limited to the level of signs and sentences. In the Anglo-American debate, influenced by pragmatism and analytical philosophy, discourse is in many cases understood as organized turn taking, for example as conversations and interactions. By contrast, the tradition that has emerged since the 1960s in France tends to associ-ate discours with written texts circulating in larger social communities (Maingueneau 1987; Détrie et al. 2001; Charaudeau and Maingueneau 2002; Mazière 2005; Münchow 2010).

French discourse analysis begins to develop under the auspices of two formalist tendencies in linguistics: distributionalism, inspired by the North American linguist Zellig Harris, and structuralism, which goes back to Ferdinand de Saussure. Discourse analysis receives its name from Harris, close to the behaviorist language theory of Leonard Bloomfield and still known as teacher of Noam Chomsky’s.

In his Discourse analysis (Harris 1963), first published in 1952, Harris examines the organization of linguistic phenomena beyond the level of individual sentences (in French: énoncé) and thus defines discourse as the transphrastic organization of texts. Like Bloomfield, Harris insists on the role of linguistic forms, whose regularities are determined by procedures of comparison and exchange. Harris traces the syntactic and transformative rules which allow the formation of a discourse from indi-vidual sentences. His question is how formal elements extend beyond the

2A History of Discourse Analysis in France


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


sentence and form a syntactic order. The meanings that are marshaled by discourse, however, play no role for him. This meaning-free discourse analysis, which is to give important impulses for Chomsky’s transforma-tive grammar, was propagated in the 1960s by Sumpf and Dubois (1969). It will also influence Michel Pêcheux’s Automatic Discourse Analysis (Pêcheux 1969) and inspire certain lexicometric instruments from the 1970s on, such as the text-statistical software Alceste by Max Reinert. Indeed, the quantifying instruments of corpus linguistics traditionally play an important role in French discourse analysis (cf. Lebart and Salem 1994). Thus, lexicometric methods have been developed since the 1960s, for example in the research laboratory of Saint Cloud, where the various contexts and uses of ‘key words’ (mots-pivots) were inventoried until the 1980s (Tournier et al. 1975; Bonnafous and Tournier 1995; Lebart et al. 1998). In the empirical applications of Harris’ method, the syntactic and transformative relations between the graphic forms of a corpus are often of lesser importance than the lexicometric determination of its distribu-tions and co-occurrences (cf. Robin 1973).

While Harris’ distributionist method focuses on the syntactic linkages between the sentences of a corpus, the semiotic approaches deriving from Saussure’s theory aim at the semantic values of signs. Saussurian linguistics, which describes language (langue) as a system of difference without a positive term, is a second impetus for the emergence of discourse analysis in France (Saussure 1962). Saussure shows how the variety of linguistic phenomena can be traced back to a limited number of distinctive elements and grammatical rules. In the transition from the linguistic system (langue), that is the entirety of grammatically pos-sible sentences, to the actualized parole, the speaker needs to select and combine the smallest constitutive elements of a language in order to form sentences. Their meaning depends on the difference of these ele-ments vis-à-vis all possible elements of a language. Meaning is thus not inherent in the individual sign, but results from its position in a system, which consists of nothing but differences. In contrast to Harris, Saussure is interested in the meaning of signs. If signs exhibit a dual structure: signified (signifié) and the signifier (signifiant), Saussure insists on the arbitrary nature of their association. Saussure knows neither the notion of discourse nor the term discourse analysis. Yet he underscores the more general relevance of the formal-linguistic model for the social sciences and humanities by pointing out the role linguistics plays in a more general science of social life—semiology. Among his most impor-tant successors in France, we can cite Émile Benveniste. Benveniste was to compare a great number of languages with the method of difference


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

A History of Discourse Analysis in France 9

and exchange. Following the Danish Saussurian Louis Hjelsmlev, A.J. Greimas developed a general semiotics which conceived the meaning of signs as a product of semiotic operations. Furthermore, a number of cul-tural semioticians of the 1960s used Saussure’s differentialist model to decode the cultural logic of modern and postmodern societies (Barthes 1967; Baudrillard 1968; Bourdieu 1972).

The field of discourse analysis in France was established at the end of the 1960s. In the light of intellectual paradigms of the time such as structuralism, psychoanalysis and Marxism, two philosophers discov-ered linguistic and social theory (Pêcheux 1969; Foucault 1969): Michel Pêcheux, taking up Althusser’s and Lacan’s theoretical investigations of the symbolic, the social and the subject, became the theoretical head of a school which influenced numerous linguists of the 1970s. Michel Foucault, the discourse theorist with great international renown, also turned to the problem of discourse in the 1960s (Foucault 1966). However, even though Foucault reaches out to a broader intellectual public, his influence on the French field of discourse analysis remains rather diffuse (cf. Maingueneau and Angermuller 2007).

If Foucault did not found a school, this was not only due to his aversion to long-term disciplinary and theoretical commitments, but also to the fact that he no longer published any works on discourse analysis after the Order of Discourse (1971). While Pêcheux and Foucault offer only few methodological instruments for empirical research, their accomplishments are more theoretical, for they situate discourse analy-sis in the broader intellectual debate on subjectivity and ideology. In the spirit of Saussure’s vision of an overall semiology of social life, they search for the rules organizing the production and linkage of the utter-ances of discourse. From Althusser and Lacan, they take the idea that discourse has an institutional place (Foucault) and that the subject has a symbolic position in discourse (Pêcheux). Against this theoretical back-ground, the field of discourse in France develops and receives its major impulses from linguists interested in social and political questions (Maingueneau 1976; Henry 1977; Marandin 1979; Courtine 1981; Charaudeau 1997), linguistically informed historians (Robin 1973; Guilhaumou et al. 1994) and sociologists (Leimdorfer 1992; Achard 1993; Jenny 1997; Chateauraynaud 2003).

In the course of the 1970s, the field of discourse analysis consolidated itself and turned its attention to empirical research. Especially from political discourse analysis, several larger studies emerge, drawing on a growing repertoire of text-analytical and corpus-linguistic instruments (Robin 1973; Seriot 1985). There were further developments on the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


theoretical front, as well. We can mention, for instance, an increasing problematization of discourse as closed structure and the internal analysis in the manner of Harris and Saussure. According to Pêcheux’s scheme of periodization, discourse analysis passes through three major phases. For DA 1 (discourse analysis around 1970) discourse was a closed machine, operating in a completely autonomous manner. DA 2 (around 1975) discovered the dimensions of the overarching before and elsewhere of discourse. DA 3 (around 1980) questioned container-models of discourse while abrogating the inside/outside distinction. Now discourse analysis has entered its ‘deconstructive’ phase, as its inner or absent borders receive theoretical attention (Pêcheux 1990). Calling into question the container-model of discourse does not remain without consequences for the analytical procedure. DA 1 was concerned with the question concerning how elements of discourse are assembled to form a complete whole. Its objects of analysis were autonomous texts, whose transphrastic organization must be examined (cf. for example Harris or Saussure). DA 2 directed its sights to what happened elsewhere and strayed into the text from outside (see the problem of discursive formation, interdiscourse and preconstruct in Pêcheux). The assumption of a discourse with stable boundaries was to finally become problematic by DA 3. Now it is no longer possible to compile ‘natural’ text corpora or to delimit the precise units of discourse. It becomes nec-essary to reflect on the heterogeneity of the utterances of a discourse whose external borders must be constantly negotiated anew. And only now must one differentiate systematically between texts and discourse: whereas DA 1 had understood discourses as the transphrastic organiza-tion of autonomous texts, DA 2 had discovered the institutional places where texts were equipped with specific meanings and preconstructed knowledge mobilized. Finally, DA 3 conceived of texts as material surfaces in which discursive practices were inscribed.

While the gradual de-limiting of discourse grows apace with the crisis of the structural model, the pragmatic dimension of discursive activity becomes increasingly important. The pragmatic turn that linguistics has seen since the 1970s is supported by the reception of Anglo-American currents, such as Wittgenstein’s philosophy of language or Austin’s speech-act theory. Récanati (1979a) and Berrendonner (1981), for exam-ple, enquire into the rules underlying the production of linguistic events and how their performativity, reflexivity and opacity can be accounted for. This pragmatic turn involves a fundamental change of perspective—from the analysis of formal structures (‘texts’), to the analysis of language use in the specific contexts of enunciation (‘discourse’). The distinction


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


between discourses and texts is fundamental for the tradition of discourse analysis that has been developing in France since the 1970s. Texts now turn out to be merely abstract entities; they need a context of enuncia-tion to be ‘complete.’ This is why most discourse analysts, in one way or another, see themselves as discourse pragmaticians. The object of analysis is not a system of signs (langue), but the rules of usage by which symbolic forms mobilize specific contexts. The French discussion about discourse pragmatics continues, for the most part, to be marked by the anti-subjectivist stance that had characterized its structuralist predeces-sors. Thus, linguistic activity is not conceived of as symbolic action which presupposes a speaking subject as a unified source of meaning produc-tion. Rather, it is understood as the production of events which are not necessarily under the sway of intentions (cf. Moeschler and Reboul 1994).

Yet the first symptoms of a pragmatic turn do not come from linguis-tics but from the field of discourse analysis itself. It is Michel Foucault, who in his discourse-analytical period (1966–1971) broke with the struc-tural model by integrating notions from pragmatics such as speech act (as is argued in detail by García 2009 and Larrauri 1999). Therefore, from the very beginning, the question of discourse pragmatics is a constitutive problem for French discourse analysis. This evolution can be illustrated by juxtaposing Foucault’s historical study The Order of Things (1966) with the more explicitly theoretical Archeology of Knowledge (1969).

No work of Foucault’s realizes the structuralist program in a more encompassing and systematic way than The Order of Things. This work is based on a theory of the epistemai. Episteme (from Greek επιστημη ‘knowledge’) denotes the specific historical mode which systematically and encompassingly structures what can be thought and said in a society at a given point in time. Like Saussure’s langue, this term denotes the generative and classificatory principles of a system. With the help of this term, Foucault investigates historical systems of knowledge which comprehensively organize scientific and proto-scientific discourses. According to Foucault’s historical episteme, the age of analogy (until the end of the sixteenth century) was followed by the classical age of representation (seventeenth and eighteenth centuries), which was in turn followed, by the age of Man (nineteenth century). With the theory of episteme, Foucault tried to ‘explain’ the knowledge of an epoch with a specific code or cultural grammar. The advantage of this method is obvious: an episteme reduces the horizon of the thinkable and sayable to a common ‘formula’: ‘analogy,’ ‘representation,’ ‘Man.’ Like Saussure’s langue, Foucault’s episteme denotes a grammatical principle, a generative code which organizes the knowledge of a socio-historical formation


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


in a systematic manner. In consequence, the entirety of what can be thought and said is conceived as a product of an all-encompassing episteme, whose rules and laws need to be decoded.

For many observers, the Archeology is the theoretical elaboration of Foucault’s historical analyses undertaken in The Order of Things. However, the Archeology offers one fundamental contrast to the work of 1966. In contrast to the Order of Things, Foucault then emphasized the difference between a structural-semiotic and a discourse-analytical approach:

The question posed by language [langue] analysis of some discur-sive fact or other is always: according to what rules has a particular utterance been made [construit], and consequently according to what rules could other similar utterances be made? The description of the events of discourse poses a quite different question: how is it that one particular utterance appeared rather than another? […] we must grasp the utterance in the specificity and singularity of its event, determine its conditions of existence, fix its limits as precisely as possible, estab-lish its correlations with other utterances that may be connected with it, and show what other forms of statement it excludes.

(Foucault 1969: 39f.[30f.])1

The ‘occurrence of the utterance/event’ (l’instance de l’événement énon-ciatif) now replaces the grammatical understanding of the episteme (Foucault 1969: 41[31]).2 The realized sentence (phrase) of structural linguistics is replaced by a specifically enunciated utterance (énoncé), which he conceives of as an ‘elementary unit of discourse’ (Foucault 1969: 107[90]).3 The shift from a structural to an enunciative and prag-matic notion of discourse has far-reaching consequences for the theory and method of discourse analysis. The rules described by Foucault in the Archeology no longer designate a comprehensive episteme generating abstract knowledge through a given set of rules (episteme). The rules Foucault deals with in the Archeology organize the manifestation of actual, singular and specific utterances:

An utterance exists outside any possibility of reappearing; and the relation that it possesses with what it enunciates is not identical with a group of rules of use. It is a singular relation: and if in these conditions an identical formulation reappears, with the same words, substantially the same names—in fact, exactly the same sentence—it is not necessarily the same utterance.

(Foucault 1969: 118[100f.])4


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


The emergence of an utterance does not leave the rule untouched on the basis of which it emerges. As Wittgenstein (1997) has remarked, the rule is constituted as a rule only in its application. Its existence can therefore not be separated from the specific utterance whose enunciation the rule organizes. What is of interest here is the fact that utterances are enunciated. Utterances are no longer either ‘correct’ (when real-izing the episteme) or ‘false’ (and therefore outside of the domain of discourse). We therefore have to ask what is ‘their mode of existence’ (Foucault 1969: 143[123]).5 Yet in light of the numerous discourse-pragmatic developments adumbrated since the 1970s (see below), the Foucauldian approach seems, in retrospect, peculiarly hybrid. Thus, Foucault conceives of énonciation as a ‘singularity’ (1969: 133[114]), whereas the énoncé is characterized by its ‘repeatable materiality’ (1969: 133f.[114]).6 Courtine is right to point out that Foucault’s definition of the utterance differs from terminologies prevalent today. Accordingly, Foucault’s notion of enunciation (énonciation) is ‘close to discourse-analysis’s current understanding of the term […]; the utterance (énoncé) is associated with repetition [in semiotics and structuralism].’ (Courtine 1981: 45).7 Oswald Ducrot (1984: 95) has provided the well-known definition of the utterance (énonciation) by contrasting the sentence as an ‘abstract linguistic being’ (phrase: ‘an invention of that strange science of grammar’) with ‘what is observable by the linguist: the utter-ance [énoncé], which is regarded as a specific manifestation, as a hic et nunc occurrence’ (1984: 174).8 Seen from this angle, Foucault mingles a structural- semiotic approach to discourse with speech- act-theoretical considerations (Foucault 1969: 110ff.[93ff.]):

Too repeatable to be entirely identifiable with the spatio-temporal coordinates of its birth […], too bound up with what surrounds it and subtends it to be as free as a form […], it [the utterance] is endowed with a certain modifiable heaviness, a weight relative to the field in which it is placed, a constancy that allows for various uses, a temporal permanence that does not have the inertia of a mere trace or mark. (1969: 138[118])9

Regardless of this rather peculiar hybridization of sign and speech act, the enunciative discourse analysis outlined by Foucault can serve as a model for the investigations that follow below. For the enunciative approach to discourse, the smallest units of discourse are the utterances (énoncés) produced by enunciations or utterance acts (énonciations). Utterances constitute facts of discourse produced in discursive


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


acts—facts to which further acts can refer. Enunciative discourse analysis does not deal with abstract, grammatically correct sentences but with utterances that must have actually been enunciated in order to exist. In fact, one can grasp the contexts in which utterances attain their specific meanings only by recognizing the singularity of actually produced enunciations. In this way, we can account for the organiza-tion of discourse beyond the interactive situation. Whereas the Order of Things renders the production of knowledge as the actualization of semiotic meaning by a given set of rules (episteme, langue, code …) independent of its specific context, the Archeology of Knowledge under-scores the role of specific contexts, fields and formations without which the meaning of an utterance cannot be determined. Is it only a case of terminological negligence that discourse is not mentioned anywhere in the work of 1966? In fact, discourse could become an object of Foucault’s investigations only when he gave up the structural-semiotic framework to which he subscribed in 1966. From 1969 onward, Foucault discovered that discourse is about more than isolated abstract signs. If discursive activity is now conceived of as the creation of networks or discursive formations of utterances, it is the theoretical shift towards the problem of the enunciation that has made the analysis of discourse possible. Therefore, in one form or another, discourse analysis has to engage in an ‘enunciative analysis’ (1969: 143[122]). In contrast to the Order of Things, analytical objective of which was to discover a code of what can be said and thought historically, the attention is now directed to the specific organization of certain utterances. Therefore, the dis-course analyst no longer has to take on a bird’s-eye view from which perspective the entirety of the symbolic manifestations of a society can be traced back to an overriding code; rather, the analysis assumes its starting point ‘from below’—from the complex materiality of variously constituted, specific utterances:

The enunciative analysis, then, is an historical analysis, but one that avoids all interpretation: it does not question things said as to what they are hiding, what they were ‘really’ saying, in spite of themselves, the unspoken element that they contain, the proliferation of thoughts, images, or fantasies that inhabit them; but, on the contrary, it questions them as to their mode of existence, what it means to them to have become manifest, to have left traces, and perhaps to remain there, awaiting the moment when they might be reused once more; what it means to them to have appeared when and where they did—they and no others. (Foucault 1969: 143[123])10


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


The enunciative analysis aims neither at the deciphering of an underlying set of rules nor at the interpretive reconstruction of lived meaning. It claims to observe the utterances of discourse in their ‘positivity’, as ‘things said,’ behind the opaque surface of which there is no stable meaning waiting to be understood. Therefore it is logically consistent when, in his inaugural speech at the Collège de France, Foucault qualifies his discourse-analytical project as a ‘felicitous positivism’ (positivisme heureux) (1971: 72[234]).

But while Foucault underlines the distance to interpretive-hermeneutic approaches to knowledge, he gives few hints as to how an enunciative analysis of utterances might look. The empirical analyses in The Order of Things neither operate with ‘utterances’ nor with ‘discourse’—a term which was to be introduced only in his Archeology. In this context, we can note Foucault’s rather allusive remarks on how to imagine the connection of utterances to ‘discursive formations.’ Unlike the knowledge organized by an episteme, utterances do not constitute any transparent spaces of knowledge and representation. These three features mark the difference between the systems of knowledge or epistemai of 1966 and the discursive formations of 1969:

• First, Foucault emphasizes the dynamic character of a discursive for-mation. Unlike English, the French formation discursive can also have a processual connotation, namely ‘becoming,’ ‘creating,’ ‘forming.’ Thus, a formation is constituted in the enunciation of utterances in the double sense of an (unfinished) process and an (open) system of elements. As a forming formedness, a formation is situated in a permanent historical flux which erodes any overriding principles of order.

• Second, a discursive formation is composed of heterogeneous elements, that is utterances which exist in different enunciative modalities. A discursive formation is no unitary structure whose elements can claim equal value and are enunciated on the same discursive level. Commands and evaluations coexist with descriptions; direct speech is nested within indirect speech; irony alternates with literal speech. In short, in utterances a swarming plurality of different, even contra-dictory voices and points of view can be heard, which testifies to the characteristic impurity of the discursive formation.

• Third, we must note the incompleteness of the formation. Its ele-ments do not add up to an integrated structure. In the formation, scattered utterances link to form a bundle whose elements remain in many cases under- or over-specified and which do not display clear


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


inner/outer boundaries. And since the formation neither emanates from a definite origin nor strives towards an end, an arbitrary moment inheres in every representation of discourse. No part can represent the whole since it cannot be clear where discourse starts and where it comes to an end. It is the constitutive openness of the formation which counters any attempt to proclaim a summary, a tableau, in which all elements of the discourse occupy their functional place.

Although Foucault lacks the instruments for a methodical application of the enunciative approach, the question of the discursive formation is one of the enduring legacies of Foucault’s discourse analysis. Thus, in the early 1970s, Michel Pêcheux, too, takes up this term. After analyzing, in Automatic Discourse Analysis (1969), the transphrastic organization of texts (DA 1), he turns to the problem of the discursive anterior and exterior (DA 2). Against the background of Louis Althusser’s Marxist philosophy, Pêcheux mobilizes a number of analytical instruments in order to deal with the question of the discursive formation. Thus, Althusser turns against deterministic (‘causal-expressive’) class theories which take the economy as the last instance of the social. According to Althusser, institutions (‘state apparatuses’) have a logic of their own that cannot be reduced to any class antagonism. In his famous essay on ‘ideo-logical state apparatuses,’ Althusser distinguishes between repressive institutions (government, army, police and so on) and ideological state apparatuses (culture, school, religion, politics and so on) (Althusser 1995: 106f.[16f.]). While the former, by means of (the mere threat of) brute force, ensure that individuals occupy the places assigned to them in the social structure, the latter operate by means of ideology. Althusser views ideology as a symbolically ordered matrix which ‘represents the imagi-nary relationship of individuals to their real conditions of existence’ (1995: 216[36]).11 By stressing the materiality of ideology (in contrast to ‘the ideal […] and spiritual existence of ‘ideas’ ’ 1995: 108[39]),12 he turns against a notion of ideology as false consciousness. Ideology does not deceive the individuals living within it by simulating a ‘wrong’ picture of ‘real’ conditions. Rather, ideology is what appears to the indi-viduals as given beyond any doubt, as spontaneously self-evident and as something to be followed voluntarily. Ideology belongs to the essential properties of the individual, who becomes a subject only at the moment when she or he is called—interpellated—by it:

I shall suggest that ideology ‘acts’ or ‘functions’ in such a way that it ‘recruits’ subjects among the individuals (it recruits them all), or


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


‘transforms’ the individuals into subjects (it transforms them all) by that very precise operation that I have called interpellation or hailing, and which can be imagined along the lines of the most commonplace everyday police (or other) hailing: ‘Hey, you there!’ (Althusser 1995: 226[48])13

Althusser’s concept of ideology stands for the transition from DA 1 to DA 2, since it defines the relations of discourse (or ‘ideology’) to the institutions (or ‘repressive and ideological state apparatuses’) and to the socio-structural position that the individuals (as ‘subjects’) occupy. In discourse, the relations between the individuals are symbolically defined—a process, which need not be reflected or controlled by the interpellated individuals. Thus the entrance into ideology—the subjec-tivation of the individual—is no process of the self-unfolding of ideas. It takes place on the level of material practices and institutional relations.

Althusser’s theory of ideology and subjects constitutes the socio-theoretical background for Pêcheux’s investigations of the problem of discursive formation. Following Althusser’s concept of ideology, Pêcheux defines a discursive formation as an ensemble of utterances which emanate from a certain institutional setting. According to the famous definition of Haroche, Henry and Pêcheux, ideological formations, while relating to certain class positions, consist of one or more intrinsically related discursive formations that determine what can or must be said (articulated in the form of a Philippic, a sermon, a pamphlet, a presenta-tion, a program), from a given position in a given economic situation: the main point here is that we do not deal with the nature of words but also (and above all) with the constructions in which these words are united, insofar as the constructions determine the meaning which the words take on: as we have noticed in the beginning, the words alter their meaning according to the positions adopted by those who make use of them. One can say in a more explicit way: the words ‘change their meaning’ by switching to another discursive formation. (Haroche et al. 1971: 102f.)14

For Pêcheux et al., the discursive formation allows for the fact that from certain institutional positions, certain things can and must be said. Discursive formations mark an institutionally organized terrain which is transversal to the ideological and socio-cultural lines of conflict.

In Foucault’s Archeology, utterances are not produced in a space devoid of rules, either. Yet there are striking differences to Pêcheux’s concept of formation: first, Foucault operates with a concept of rule which presup-poses positive utterances only. Foucault’s question is how an utterance


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


is enunciated and how it exists in discourse. Thus, ‘censored’ discourse, too, needs to be enunciated in a certain way, for example in the mode of a prohibition or a taboo. By contrast, Pêcheux et al. suggest a regime in which certain utterances can and have to be expressed while others are left unexpressed.

Second, while for Foucault discourse unites utterances with different enunciative modalities, Pêcheux et al. exhibit a more homogeneous com-prehension of formations shaped by certain textual genres (‘Philippic,’ ‘sermon’…). Third, Pêcheux et al. do not replicate Foucault’s enunciative and pragmatic turn. The discursive formation allows Pêcheux et al. to delimit socio-historical and institutionally homogenized regions of dis-course within which the structural method can be applied.

In a word, Foucault introduces the concept of formation in order to overcome the structural model, while Pêcheux utilizes the concept of formation in order to save the structural model. Pêcheux links the concept of discursive formation to the discourse-analytical project modeled after Saussure. Despite the problems resulting from the structural model, which Pêcheux becomes increasingly aware of in the course of the 1970s, there are two important achievements which underscore the sociological relevance of his concept.

First, the discursive formations organize the subjectivity of the indi-viduals entering into discourse. Though Foucault had already defined discourse as a ‘field of regularity for various positions of subjectivity.’ (1969: 74[60]),15 he did not indicate how subjectivity and discourse go together. In Les Vérités de La Palice [Language, Semantics, Ideology. Stating the Obvious], Pêcheux devotes his attention to this problem in extenso, which, like Althusser, he tries to solve in the sense of a ‘non-subjectivist theory of the subject’ (Pêcheux 1975: 122[91f.]).16 Pêcheux conceives the subjectivation of the individual—of the Diskursträger, the discourse bearer entering into discourse—as a ‘Münchhausen effect.’ Like the German baron who claims to have dragged himself out of the marsh, the effect of self-evident meaning precipitated by and through the entry into discourse is experienced as the sine qua non of the subject’s own origin:

In the case of ‘the subject,’ what we are dealing with here is an (imaginary) process inside the non-subject, which is the network of signifiers, in Lacan’s sense: the subject is ‘taken in’ by this network—‘common nouns’ and ‘proper names’, ‘shifting’ effects, syntactic constructions and so on—such that he results as ‘cause of himself’, in Spinoza’s sense. (Pêcheux 1975: 141[108])17


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Pêcheux follows Althusser’s interpretation of Lacan insofar as he understands subjectivity as an illusion of inner unity induced by the use of language. The individual’s subjectivation is understood as the appropriation of a subject position which had already been defined anterior to the individual’s entry into discourse. Discourse recruits its subjects insofar as it provides empty symbolic locations (such as names), by which the individuals are addressed. Hence, the individuals become discursively visible subjects in a ‘process of interpellation-identification that produces the subject in the place left empty’ (Pêcheux 1975: 143[110]).18 According to Pêcheux, subjectivity is organized via the socio-symbolic order into which the individual enters. By way of his reflections on subjectivity, Pêcheux opens his discourse-analytical project to the classical questions of social theory. Thus, the symbolic site which the individuals occupy in their discourse is always a social site as well. It constitutes the address of the individual’s position in the social arrangement, to which other subjects of discourse can refer.

Second, besides their excursions into social theory and the question of the subject, Pêcheux and his fellow advocates contribute some analytical instruments to the discussion which intend to reveal how the institutional interior and exterior impinges upon the discursive forma-tion. They pose the question as to how utterances mobilize a knowledge which is characterized as uncontestable. As an example, we may refer to their analysis of the phenomena of the preconstruct (préconstruit) and the presupposition (présupposition). The preconstruct denotes what the utterance takes from elsewhere and displays as non-negotiable knowl-edge. Thus, the utterance (1) As one of its theoretical pioneers, Michel Pêcheux contributed significantly to the establishment of discourse analysis in France in the 1970s, implies the fact F1 that French discourse analysis indeed was established in the 1970s. Preconstructs often operate with nominal constructions (such as ‘establishment of X’), which refer to an utterance enunciated elsewhere (2): ‘X is established by Y.’ Now, for (1) the utterance (2) is cited as a preconstruct precisely because the questions of where, how and by whom (2) was expressed are no longer relevant in (1). In (2) the conditions of production are erased, and it is this erasure of speaker and context which characterizes the precon-struct as the discourse of an anonymous institution or a commonly shared knowledge. Unlike reconstructive approaches, enunciative discourse analysis considers the contents of preconstructed products (what?) to be less important than their rules of production (how?). The enunciative approach aims at an analysis of the way preconstructed


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


knowledge and asserted meaning are organized in discourse. There is no analysis of whether the preconstruct is in fact accepted or refused by the individuals. To put it another way: the preconstructed knowledge of (2) is not encyclopedic knowledge, existing independently of (1) in the heads of the members of society; on the contrary, it is inseparably linked with (1).

While the preconstruct refers to the question of how the ‘institutional outside’ of discourse can become an object of analysis, the concept of presupposition allows us to focus on what is taken for granted in the utterance. Consequently, utterances not only operate with semantic contents but also with sets of implicit assumptions and immediate inferences depending on the context. The utterance (1), for example, presupposes that Michel Pêcheux is a scholar who produced important theoretical texts about discourse analysis during the 1970s in France. The utterance does not explicitly state that Pêcheux wrote texts with far-reaching consequences in the 1970s, but this can be inferred from what is said, given that one knows that discourse analysis is a field of study and the knowledge produced at universities usually is written knowledge.

Presuppositional knowledge is often closely connected to the context of the utterance. For example, it is important to know in which sequence (1) appears. In the conversation situation, an utterance fol-lows certain other utterances. Thus, the reader of a written text can connect the utterances with preceding and following utterances. Thus, the implications are different if (1) is followed by (1') Michel Pêcheux’s discourse theory follows Althusser’s ideology theory or by (2') After studying philosophy, Michel Pêcheux turned to linguistic theory. This is to say that (1') implies: ‘Althusser plays an important role in discourse analysis in France’ while (2') implies ‘Originally, Pêcheux was not a linguist.’ For argumentation theory the discovery of the presuppositional structure of utterances is especially important, because the argumentative con-nection of two utterances not only implies what is said but also what is not said.

In the context of Pêcheux’s school, it is especially Paul Henry who, following Gottlob Frege, points out the presuppositional rules of dis-course. Henry underscores the autonomy of these rules vis-à-vis the ‘subject,’ which he understands in the terms of Lacanian psychoanalysis not as the ‘center, source, unit of inwardness’ but rather as a ‘language effect’ (Henry 1977: 21).19 For Henry, the attempt to account for presup-positional meanings in the manner of speech-act theory runs the risk of reinstating the speaking subject.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


By linking presupposition and speech-act, one again reinstates the subject as the source of this act. This is true even if one acts on the assumption that this subject internalizes a universal subject regulating its activities or if one turns society into an agency which juridically rules the meanings of this subject. (Henry 1977: 82)20

Demarcating himself from Ducrot, whom he sees as a disciple of John Austin, as well as from Noam Chomsky, Henry advocates a non- subjectivist foundation for discourse theory: ‘For Foucault, a subject does not really exist on the enunciative level. Instead, there is a subject position which can be taken by different individuals’ (Henry 1977: 84).21

However, the group around Pêcheux does not offer a methodical account of the subject-positions in a discursive formation. Moreover, with the transition from DA 2 to DA 3 the concept of discursive formation loses its theoretical pliancy. Now the problems related to the heterogeneity of discourse step into the foreground—for example, its inconclusiveness and its layered density (its épaisseur and feuilletage). Now the insight gains momentum that discourse cannot easily be split up into distinct domains. The notion of a constituted formation of elements which can be represented by a closed corpus is replaced by the notion of an ‘absent border’ (frontière absente) which renders the closure of discourse impossible. In this sense, Courtine argues for a discursive formation ‘which is heterogeneous toward itself.’ In accord-ance with Lacanian psychoanalysis he regards ‘the closure of a discur-sive formation [as] fundamentally unstable; it does not consist of a border constituted once and for all, distinguishing inside from outside, but is inscribed as a border between different discursive formations which moves according to the stakes involved in the ideological bat-tle’ (Courtine and Marandin 1981: 24).22 If the container model of discourse is abandoned for the sake of discourse as an open terrain constituted around its interdiscursive borders, internal rifts and dis-locations, a discursive formation can no longer be seen as a homoge-neous, closed totality. For if discourse shows no stable inside/outside borders, it is impossible to represent the formation by a sample. Nor is it possible to compile ‘natural’ corpora. The compilation of a corpus becomes a construction of the analytical process—a construction which predetermines the results to be achieved to a considerable extent. Does this not mean that procedures from corpus linguistics, as Courtine sug-gests, ‘smooth the discursive ruggedness, suture the rifts arising in each discourse, planing down on one end, plugging and filling out on the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


other; making a full body and a flat surface’ (Courtine and Marandin 1981: 23).23 Rather than determining the meaning of an utterance, Pêcheux now turns to its complex, ‘vertical’ stratification. An utterance can now hide and encapsulate other utterances and form enunciative sequences (Pêcheux 1981: 143f.).

The focus of Pêcheux’s DA 3 on the heterogeneity of discourse recalls the crisis of Saussure’s and Harris’s structuralist models. Other authors also increasingly underscore the heterogeneity of a discourse which can no longer be accounted for within a structuralist frame.

• Heterogeneity in the sense of different overlapping sources of meaning. It was Benveniste who strengthened the argument that utterances obtain their meanings from a variety of sources. Following Benveniste, the ‘semiotic’ meaning, ‘the self-referential and, to a certain extent, self contained meaning’ (1974: 21), is complemented by a ‘semantic’ meaning that ‘emerges from the linked signs which mutually adopt and conform to one another according to the circumstances’ (1974: 21).24 Therefore, signs can have a stable and abstract as well as a local and dynamic sense. This idea is extended by Ducrot’s dis-tinction between meaning (the signification of sentences actualized by the langue) and sense (the sens of concretely uttered utterances) (Ducrot 1984: 95).

• Heterogeneity in the sense of a connection between text and context. The discourse-analytical trends of the 1950s and 1960s define as the context of a sentence mostly the neighboring sentences whose combination forms a ‘discourse’ (that is text). Since the 1970s, ‘extra-linguistic’ aspects (that is context) have also entered the analysis. Thus, Benveniste captures the spatio-temporal and personal coordinates of the utterance’s context with the ‘formal apparatus of the enunciation’ (I, here, now). Ducrot accounts for the context by conceiving the meaning as the product of the connection between a linguistic component and the context of the utterance (Ducrot 1984). Maingueneau (1993), on the other hand, dissolves the separation between text and context with the concept of scenography. Thus the text organizes its own context, which is presupposed and validated by the utter-ance at the same time. This turn to the question of the context results from insight into the non-saturated nature of signs and symbols: they need a context to which they refer in a certain way in order to be ‘complete.’


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


• Heterogeneity as self-reflexivity. As a product of the enunciation, the utterance is the trace of an activity which the utterance reflects. According to Récanati an utterance relates a representative (représentant) x to a represented (représenté) y, such that ‘the represented reflects itself in the very moment in which it represents something’ (Récanati 1979a: 21): the self-referential [sui-référentiel] character of the utterance, which discourse pragmatics illustrates as grounds the opacity of the utterance. The reflexivity of the utterance becomes manifest when it points to the act of the enunciating as a fact (Figure 2.1).

x y

Figure 2.1 The self-referential character of the utterance

• Heterogeneity in the sense of the non-unity of language. When does the heterogeneity of the utterance manifest itself? Récanati points out pragmatic paradoxes (for example I am dead), ‘cases involving a contradiction between what the utterance says and the enunciation shows’ (1979a: 206).25 In contrast, Authier-Revuz examines cases in which the ‘non-coincidence’ of language—that is ‘everything that marks the ‘not being one’ of communication: lack of understand-ing, uneasiness, lack, misunderstanding, ambivalence’—is outfitted with ‘meta-enunciative comments’ or ‘enunciative loops’ (boucles énonciatives) (for example in the form ‘p is called q,’ ‘p not to say q’…) (Authier-Revuz 1995: iif.).26 Ambiguity and misunderstanding are no accidents of language; language does not distinguish itself by transparency and communicative success. Following Culioli, ‘understanding [is] a special case of misunderstanding.’ (Culioli 2002: 28)27

• Heterogeneity in the sense of the co-existence of different enunciative modalities. As Foucault indicates in the Archeology (Foucault 1969), formations derive from utterances existing in different enunciative modalities—an idea that Austin anticipates with his differentia-tion of various forces by which sentences are expressed (cf. footing in Goffman 1981: 124). A classic example of the manifestation of heterogeneity is indirect speech. As Rosier points out, citing and cited speech can interweave in different ways, which is marked in the text typographically, attributively, enunciatively (1999: 139).


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Therefore, irony and negation, too, can be understood as an orchestrated ensemble of voices which are kept at a certain variable distance to the locutor (locuteur) (see below on polyphony in Ducrot).

In view of these numerous objections coming from different camps against a homogenizing understanding of discourse, it is hardly sur-prising that the theory of the discursive formation associated with Pêcheux remains problematic. For even as it conveys the visionary charm of the theoretical pioneers of discourse analysis, it has hardly been applied in empirical research. After Pêcheux’s death (1982), the discursive formation seems to withdraw from the discourse-analytical debate, as a period of consolidation and specialization begins. While some of Pêcheux’s close colleagues turn back to their disciplinary roots and introduce certain discourse-analytical perspectives into lin-guistics (Fuchs 1981b), history (Guilhaumou et al. 1994) or sociology (Achard 1995), linguistics in France remains an important source of ideas for discourse-analytical research. But Pêcheux’s death also marks a momentous change in the theoretical base of discourse-analytical research. Discourse analysis begins to demarcate itself from the inter-nal analysis in the manner of Harris and Saussure and clears the way for more recent ‘post-structural’ attempts, particularly in the area of the linguistics of enunciation. While the interest in structural models of discourse dwindles, the focus is more and more on the enunciative dimension of discourse, on the different enunciative levels, sources and dimensions which make up the heterogeneity of the utterance. While discourse itself oversteps all boundaries, its utterances thicken and grow ever more opaque.

Discourse as utterance and enunciation: the field of enunciative pragmatics

As the structural model fades at the end of the 1970s, the issue of enun-ciation increasingly takes center stage in linguistic theory and practice (Maingueneau 1981; Fuchs 1981a; Cervoni 1987). A pragmatic turn takes place in discourse analysis and linguistics. Pragmatics is the label for a number of approaches which reflect in one way or another upon the event-character of language and thus challenge the strict separation between text (as a linguistic object) and context (as an extra-linguistic object). Thus, in a pragmatic perspective the analysis no longer features abstract texts isolated from their enunciative context, but texts which are enunciated, used, and contextualized in discourse. If texts receive a


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


certain meaning only in their contexts, discourse analysis is bound to understand its task as a form of discourse pragmatics. Pragmatics in this broad sense is proclaimed by various linguists and discourse analysts from different theoretical camps (Ducrot 1980a; Henry 1984; Parret 1987; Adam 1989; Maingueneau 1997). However, a narrower under-standing of pragmatics prevails as well—the ‘Anglo-American’ variant which, following J.L. Austin and H. Paul Grice, focuses on speech acts and the use of language. The linguists in the tradition of speech-act theory also consider the notion of enunciation (énonciation) as their basic analytical problem.

Outside France, the debate on énonciation and énoncé is hardly known (but see Williams 1999). This lack of reception may be due not least of all to the problems of terminological translation. As a rule, the terms énonciation and énoncé can be translated by ‘enunciation’ (or ‘utterance act’) and ‘utterance’ (or more rarely ‘statement’) in English. The prob-lem with this solution is that it not only conceals the reciprocally active contexts of the procedural and resultant aspects of linguistic activity (in the sense of discourse as production and product). It also risks insinuating an intentional model of meaning and discourse (subjectivity as an original interior and language as an instrumental/expressive exterior). It is a hallmark of the French debate on ‘enunciation’ and ‘utterance’ that it generally does not rely on theories of the speaking and acting subject. This version of pragmatics is about the systematic correlation between linguistic forms and discursive practices. It includes not only analytical strands from speech-act theory but also structural and logical-semantic theories of enunciation.

The French theory of enunciation indeed receives its first impulses from structuralism. Its pioneer is Émile Benveniste who defines the enunciation as ‘the enactment of language through an act of indi-vidual usage’ (Benveniste 1974: 80).28 The analysis has no direct access to the enunciation; rather, it deals with the formal marks or traces (marques) by means of which the text indicates the context of enunciation. Benveniste—and this identifies him as a disciple of Saussure’s— situates the problematics of enunciation in the lan-guage system (langue), whose formal inventory includes the ‘formal apparatus of the enunciation’ (appareil formel de l’énonciation). The for-mal apparatus of enunciation comprises all the pronouns and adverbs which, like the personal pronouns I and you (not, however, he, she or it) or the adverbs here and now, reflect the utterance by making the spatio-temporal personal coordinates of their enunciation contexts deictically visible (from the Greek δεικνυμι, I show) (Cervoni 1987).


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Comprising the personal pronouns of the first and second persons (I/we, you/you) and their inflections (us, mine), these deictic marks or traces of the enunciation point to their object rather than saying it. Thus, deictics like I, here and now—Jakobson’s shifters (1995: 388)—refer to a context of enunciation which they ‘engage’ (embrayer in the mechanical sense of clutch, as a gear engages a motor)—a context which delivers the required contextual information about the locutor and its spatial and temporal position. According to Benveniste, the texts that operate with deictic marks of enunciation belong to the register of ‘discourse’ (discours), and texts without deictic marks belong to ‘history’ or ‘report’ (histoire) (cf. the distinction between ‘enunciative’ versus ‘utterative’ texts in Greimas and Courtès 1993: 124[102]). Kerbrat-Orecchioni (1980) sum-marizes Benveniste’s ideas on enunciation as the search for linguistic procedures (shifters, modalizers, evaluative terms and so on) through which the speaker leaves his or her imprint in the utterance, inscribes himself (implicitly or explicitly) into the message and positions himself in relation to it (the problem of ‘enunciative distance’). Thus, it is necessary to prove and describe the units (of whatever nature and level they may be) as the traces of inscription that the subject of enunciation leaves in the utterance (Kerbrat-Orecchioni 1980: 26).29

As an anthropological constant, subjectivity is embedded in the linguistic structure of the langue through enunciative marks or markers. If it is language that ‘speaks’ individuals, their subjectivity is organized through the formal apparatus of the enunciation. Accordingly, subjectivity exists independently of the individual acts of appropria-tion of langue by the individuals, who become ‘speaking subjects’ upon entering the symbolic order.

However, Benveniste is neither the first one nor the only language theorist interested in enunciation. Besides historical precursors such as Charles Bally (1965), who succeeded Saussure in Geneva, Georges Guillaume (1974) or Karl Bühler (1965, whose Origo-Theory serves as the theoretical base for the ‘formal apparatus of the enunciation’), a number of alternative attempts have been sketched out since the end of the 1970s. Sometimes the field of enunciation linguistics has been divided into a ‘neo-structural’ camp headed by Benveniste, and a logi-cal-pragmatic trend of Anglo-American provenance (Fuchs 1981a: 42). While all these strands conceive language in its complex indexicality, they are themselves marked by considerable heterogeneity. This is why I suggest dividing the strands of enunciation theories into the following three theoretical main poles (cf. Grunig’s distinction between indexi-cal, psychological, speech-act-theoretical and situative-interactionist


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


pragmatics 1981): the structural approach represented by Benveniste, logical semantics in the vein of Oswald Ducrot, and radical pragmatics of Anglo-American origin.

Oswald Ducrot, a paradigmatic representative of the logical-semantic pole in the field of enunciation theory, investigates how utterances connect with each other in the enunciation. Drawing from logic (cf. Brøndal 1943; Blanché 1968; Grize 1982; Danon-Boileau 1987) and argumentation theory (cf. Toulmin 1952; Perelman 1963; Anscombre and Ducrot 1983; Plantin 1990), he attempts to determine the rules according to which sentences follow each other and thus form argumentative units. Ducrot defines the utterance (énoncé) as an analyti-cal object which he distinguishes from the sentence—the ‘invention of that peculiar science called grammar’ (1984: 174).30 While he under-stands the sentence (phrase) as ‘an abstract linguistic being, self-identical with itself across its different occurrences’, he sees the utterance (énoncé) as ‘a specific appearance, the realization hic et nunc of the sentence’ (1984: 95).31 With the transition from the sentence (that is grammar) to the utterance (of discourse), Ducrot continues down the path blazed by Benveniste with the discovery of ‘semantic’ (as opposed to ‘semi-otic’) meaning. Now the utterance is investigated in its connection with other utterances. This allows a contextually specified sense (sens) to originate, in contrast to the meaning (signification) of the abstract sentence. Ducrot does not want ‘to trim language so that its elements obey the rules of the logician’ (Ducrot 1989: 76).32 What Ducrot learns from logic is how utterances operate with presuppositional knowledge (cf. Gottlob Frege). Thus, the utterance (1) Does the fridge work again? presupposes a number of circumstances that are presented as a non-negotiable background, namely (P1a) that there is a fridge or (P1b) that the fridge broke down at some time. Ducrot makes clear that these presuppositions are anchored in the utterance itself and are organized in a certain way through what comes before and after the utterance. Thus, different argumentative orientations are evoked if (1) is followed by either (2a) I have brought ham or (2b) The fuse has blown. Depending on the sequence, we may infer either (P2a) The ham must be cooled or (P2b) A technical problem has occurred, where (2a) refers because of (P2a) to (P1a) and (2b) because of (P2b) to (P1b). A discourse is formed if the utterances interlink with such argumentative ensembles and evoke presuppositional knowledge. This ‘discursivization’ of the utterance refers neither to a certain state of the world nor to what the speaking subject knows, wants or intends: ‘The concatenation of utterances has an internal origin; it is based on the very nature of the utterance, or if


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


you prefer, on its sense [sens], […] which contains an allusion as to how it should continue’ (1980b: 11).33

In this context, connectors like but or because, which define an argu-mentative relation between two utterances, play an important role. For example (3) I have bought ham, but the fridge is broken. While but marks both utterances (3a) and (3b) as ‘true,’ the first utterance is kept at a distance in a certain way by the locutor, whereas the second utterance is marked as a direct product of the source of enunciation: It is true that (3a) but (3b) as a result of which a new light is shed on (3a) and it is concluded: (3c) If the fridge does not work again soon, then (3a) was for nothing. In order to designate the implied points of view, Ducrot calls the discursive being ‘that is responsible for the utterance’ (1984: 193)34 the locutor (locuteur), while the enunciator (énonciateur) is a figure or voice held at a certain distance by the speaker. In this line, Rabatel (2010: 370) insists on the fundamental distinction between locutor and enunciator:

The locutor is the first instance which produces the utterances materially. […] The enunciator is the instance which positions itself vis-à-vis the objects of discourse to which it refers and, by so doing, takes them in charge. The notion of enunciator corresponds to an (enunciative) position which the locutor adopts in its discourse to envisage the facts, the notions, under this or that angle, taking them on its account or on the account of others. The disjunction locutor/enunciator can explain the ways in which the locutor as enunciator turns around the objects of discourse, envisages the facts, words and discourses, the notions, situations, events, phenomena of this or that point of view, in the present, in the past or in the future, in relation to itself or to others.35

Thus, in (3), the locutor designates the being directly responsible for the second utterance, yet for the first utterance only in order to call it into question by the second one. Utterances mobilize different voices (voix) and sources of enunciation (sources énonciatives), which found the characteristic polyphony of the utterance and send the utterance searching, as it were, for connections to other utterances. Applying Mikhail Bakhtin’s theory of dialogue and polyphony to individual utter-ances (cf. Genette 1972) Ducrot envisions the utterance as a heteroge-neous bundle of overlapping voices (Bres 1999; Nølke et al. 2004; cf. Goffman’s distinction of animator, principal, author 1981: 197ff.; and the debate on intertextuality Kristeva 1973; Moirand 2001; Fairclough 2003: 39ff.).


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Another field of application for Ducrot’s theory of polyphony is irony. Irony emerges when an utterance (for example, (4): George W. Bush is America’s greatest intellectual uttered with a certain intonation) is expressed without the locutor taking responsibility for what is said. In this case, the speaker lets an enunciator (énonciateur) speak, only to reject it straight away (L: ‘It is not true that (4)’). The locutor is only indirectly present since it cannot accept the spoken utterance in the wake of exag-geration, intonation or some obvious discrepancy. Negation functions in this way, too. Rather than simply turning a positive value (p) into a negative value (as in a mathematical operation), negation is a complex operation in which two figures cooperate, namely an enunciator (énon-ciateur) (E: (p)), and a locutor who holds the enunciator at a distance (L: ‘It is not true that E: (p)’): ‘Ironic speech means that the locutor L presents the enunciation as expressing the position of an enunciator E, which one knows that the locutor L is not responsible for and even believes to be absurd’ (1984: 211).36 By conceiving of the utterance as a theater where different voices cross and mingle, Ducrot underlines the impossibility of tracing what is said in the utterance back to a unified origin of meaning.

What are the consequences for the analysis of texts if we divide the origin of enunciation into different speech agents, into locutor and enunciator?

• First, the utterance is borne not by a single speaker but by different speakers. There is exactly one locutor in the utterance (the deictic center of the enunciation) and at least one enunciator (accepted or rejected by the locutor). Therefore, a single utterance may bear the traces of an exchange, dialogue or conflict between the discourse participants.

• Second, the variety of enunciators who speak through the utterance tends to be continued, resolved or reoriented through other utterances. The utterance does not exist as a self-sufficient unit of meaning. Since it generally exists as a response to questions (which do not necessarily have to be asked), it has a tendency to orient a sequence of utterances: ‘A suggestion to act or an obligation to respond are given as effects of the enunciation.’ (Ducrot 1984: 174)37

• And third, the locutor and the enunciator are discursive beings who belong to the utterance and cannot be anchored in a speaking subject or in one individual out there in the world. In this sense, Ducrot subscribes to the structural critique of the speaking subject: ‘There is nothing psychological about the concept of the enunciation


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


[…]; it does not imply that the utterance is produced by a speaking subject’ (Ducrot 1989: 76).38 Like Benveniste, Ducrot assumes that ‘a very large number of enunciative phenomena can be integrated into the linguistic structure itself. It is a matter of expanding structuralism while retaining the fundamental ideas of structuralism’ (Ducrot 1992: 64ff.).39

• Ducrot’s utterance theory has turned out to be fruitful for a large number of linguistic studies. Thus, in his analyses of the textual construction of the point of view [point de vue], Alain Rabatel extends the analysis of single utterances to narrative texts in which, as a result of complex enunciative operations, certain points of view are dominant while others are erased (Rabatel 1998). With the introduc-tion of over- and under-locutor and over- and under-enunciator and the notion of ‘enunciative effacement,’ Rabatel refines Ducrot’s terminology and points to the various ways in which utterances can be accepted or rejected—directly or indirectly. Laurence Rosier dif-ferentiates locutor and enunciator in order to analyze the quotation of utterances in reported speech: ‘Reported speech is the placing in relation of discourses [speech, discours], where one discourse creates a particular enunciative space while the other is held at a distance and is connected to another source [of enunciation] in an equivocal or unequivocal way’ (Rosier 1999: 125, in italics in the original).40 The crossing from citing to cited discourse can be signaled in various ways. Rosier gives special prominence to quarrel markers (discordan-ciels), which include connectors (but), modal transitions (for example from the indicative to the imperative), assertive or negating particles (no, yes), interjections (Shoot!), nominalized sentences or lexical breaks (1999: 153). In reported speech, different discourses with different sources of enunciation are interlaced. He who speaks does not understand only (by) himself.

Finally, radical-pragmatic theories of enunciation usually draw from theories of speech acts or conversational maxims. As opposed to struc-tural or logical-semantic approaches, they generally focus directly on the enunciation and its linguistic contexts. Unlike the afore-mentioned strands, where the discursive events correlate with certain linguistic forms, these theories of Anglo-American provenance postulate the pos-sibility of a direct access to event, action and situation (as in symbolic interactionism, conversation analysis or Goffman). Not only these pragmatistic tendencies (cf. for instance Kerbrat-Orecchioni 1990) but also discourse pragmatics (Austin, Grice) have crucially contributed to


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


the French theory of enunciation. Austin’s speech-act theory highlights the different enunciative modalities or ‘illocutionary forces’ according to which discursive acts exist (Austin 1962), whereas Gricean discourse pragmatics (1989) looks for ‘discursive laws’ which are inevitably assumed in every communicative exchange.

Jean-François Lyotard is often overlooked when we consider the intel-lectual pioneers of the discourse-pragmatic tradition in France. Better known is Michel Foucault, who insists on the enunciative modalities of existence of utterances in the discursive formation (cf. above on his Archeology of Knowledge 1969). In the Archeology, Foucault theorizes discursive practice as a sudden ‘irruption of the event’ (Foucault 1969: 37[28]). Lyotard (1979, 1988), too, underscores the contingency which distinguishes the concatenation of sentences to more complex aggrega-tions of knowledge. By contingency, Lyotard does not mean ‘arbitrari-ness’ or ‘non-necessity.’ Contingency characterizes the new aspect that an utterance gains when it is selected from all possible utterances and chained to other utterances. Discourse is not about the reproduction of what is already known. It organizes the production of unique events which introduce something that was not yet there in the terrain of symbolic relations.

Lyotard and Foucault certainly formulate a momentous critique of the structural paradigm. However, the linguistic discussion about the enunciation is led by language theorists like François Récanati or Alain Berrendonner, who crucially contribute to the establishment of speech-act theory in France. While for Récanati linguistics ‘examines language as such, that is as a system constituted by rules and norms, pragmatics considers it, so to speak, from an external perspective—not language itself is the focus, but the use made of language’ (Récanati 1979b: 8).41 Thus, Récanati insists upon the difference between the ‘occurrence,’ that is the concrete spatio-temporal appearance of a sign, and the ‘type,’ that is the sign of which the occurrence represents an appearance. Récanati examines the rules which organize the specific occurrence of a type, ‘but pragmatics does not examine what is particu-lar or individual in language usage’ (Récanati 1981: 20[10]).42 While the abstract sentences of semiotic theory, which signify the same (no matter where, when and how they are expressed) and realize a pre-given, over-riding code (langue), the utterances of speech-act theory are enunciated in a specific way.

The use of signs represents not only the world or an object, but also the use itself, as it were. It is this double reference to the world and to itself that makes the discursive process heterogeneous and reflexive.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


The reflexivity of discourse can be illustrated by ‘type words’ like I, whose ‘occurrence reflects the fact of its own enunciation’ (Récanati 1979a: 9).43 And the heterogeneity of discourse is shown by pragmatic paradoxes—‘cases involving contradictions between what an occur-rence says and what its enunciation shows’ (Récanati 1979a: 206).44 Discourse takes place simultaneously on different levels which may contradict each other. As a materiality referring to itself, discourse mediates between thought and the world. Text and context are no separate worlds. Pleading against the ‘Y model,’ which comprehends meaning as the result of the confluence of a linguistic component and an enunciative environment, theorists like Berrendonner aim to weaken the boundary between sign and context, and to grasp the production of meaning as an autonomous act that can be reduced neither to abstract semiotic meaning nor to a given set of contextual variables (Berrendonner 1981: 22).

Pragmatics in the vein of speech-act theory points out the specific enunciative modalities in which utterances exist in discourse. For H. Paul Grice, pragmatics is about the description of the specific meaning effects produced by discursive acts. Grice is especially interested in the ‘laws’ that allow partners in a communicative situation to recognize what is meant in the situation. As a basic discursive law, Grice points out the principle of cooperation that can be differentiated into various maxims of conversation (quality: Say nothing that you think could be false! quantity: Be informative! or relevance: Be relevant!). These maxims work in all communicative situations, no matter whether they are obeyed by the speakers or not. To say something irrelevant or to say too much in a conversation produces communicative effects—implicatures—which can be recognized by the listeners as conveying an implied intention or meaning. For example in the following dialogue: ‘Who won the 2000 presidential elections in the U.S.?’—‘The evangelical suburbs.’ Given the maxim Be informative!, the answer alludes less to the competing candidates than to a conflict between social groups. As a political state-ment the answer refers to the role of social and religious questions in Northern American politics. In a Gricean perspective, discourse is sym-bolic action which the communication partners scan for the locutor’s implied intentions against the background of the conversation maxims. These discursive laws are valid regardless of whether the discourse is grammatically correct or not. Thus, Grice’s maxim theory implies a powerful critique of meaning models based on codes and grammars that privilege discursive practice complying with the rule while classifying all deviations and violations as noise devoid of meaning.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


In the French debate on discourse, Grice’s criticism of language- as-code is represented by Anne Reboul and Jacques Moeschler, who are critical of approaches which see language as self-sufficient and which hold that ‘the interpretation of a sentence consists of decoding, that is using the code constituted by the language system (langue) in which the sentence is expressed in order to restore the message’ (Reboul and Moeschler 1998a: 17ff.).45 Reboul and Moeschler also distance themselves from an approach to discourse modeled after the phoneme and the morpheme. For Reboul and Moeschler, discourse should not be seen as a ‘linguistic unity beyond the sentence’ (Reboul and Moeschler 1998b: 8).46 Consequently, the analysis should not conceive its object as a structure sui generis beyond sign and sentence (analyse de discours). Instead, they plead for a discourse analysis (analyse du discours) that starts from utterances enunciated in a specific way without reducing them to larger discursive structures.

In contrast to structural and logical-semantic theories of enunciation, this radical-pragmatic current focuses on the speech actions that trigger certain cognitive processes among the discourse participants. The turn to the problems of cognition is supported in particular by Dan Sperber and Deirdree Wilson (1989), who merge Grice’s conversation maxims into one single principle of ‘relevance’ (pertinence). In this view, inten-tions can be signaled linguistically (that is with the help of symbols and signs) or non-linguistically (by gestures or even non-action). In the communicative process, we continually search the environment of the cognitive system for traces signaling intention. From this perspective, the symbolic organization of the message is secondary to the question of how listeners recognize it as a sign of intentionality. Communication and discourse become synonymous with mental representations on the part of listeners, which emerge as a result of linguistic action based on discursive laws. Since in the perspective of radical pragmatics, the context of enunciation merges into a cognitive context, texts are not privileged objects of analysis. Every action or non-action in the cogni-tive system’s environment has the potential for becoming an object of communicative search processes. Accordingly, radical pragmatists plead for a turn from semiotic to cognitive models in discourse analysis.

To sum up, the structural, the logical-semantic and the radical- pragmatic approaches to enunciation can be distinguished as follows. The structural theory of enunciation sticks to the language system (langue) as an object of linguistic analysis which contains a set of markers reflecting the enunciation. In this view, sentences can refer to the enun-ciation and engage the context (Benveniste’s discours) or not (histoire).


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


While the analysis has no access to the actions, events, and contexts of enunciation, discourse is produced when individuals appropriate for themselves the sentences of the language system and actualize the code of what can be said. The logical-semantic theory of enunciation, too, recognizes the existence of the language system (langue). However, it sees the sentences of the language system as merely abstract phe-nomena, which need to be combined with their enunciative contexts in order to be analyzed as specific utterances of discourse. Therefore, the analysis deals with utterances and not with sentences. Encased in a set of contextual parameters, utterances organize the possible con-texts of enunciation. The individual cannot completely appropriate the enunciation to him/herself because the origin of enunciation is split into different voices and points of view. By putting negotiable and non-negotiable knowledge into relation, the utterance contains instruc-tions about the way it is to be concatenated with other utterances. Finally, radical-pragmatic theories of enunciation break with the notion of language-as-code (langue), deeming access to the specific discursive events and their actual contexts of enunciation fundamentally possible. Discourse operates with specific events that exist in certain enunciative modalities (Austin); they can be recognized by the cognitive system as signals for intentionality (Grice). Radical-pragmatic approaches presuppose a cognitive system, an individual who searches his or her environment for traces of intended meaning.

Many of the enunciative approaches that have been developing since the 1970s do not clearly represent one of these three poles (structural, logical-semantic, radical-pragmatic theories of enunciation). This is the case for the discourse-analytical perspective which Dominique Maingueneau has brought into the discussion with the concept of sce-nography. By scenography, Maingueneau understands a world which is both presupposed and represented in the enunciation of the text: ‘Discourse coincides with the way in which it regulates its own origin; it represents a world of which its enunciation is a part’ (Maingueneau 1995: 40).47 He is especially interested in ‘constituting’ discourses (dis-cours constituants like philosophy or religion), which have to solve the paradoxical task of creating the discursive conditions of what can be said. This approach starts from the hypothesis that

only a discourse which constitutes itself by thematizing its own constitution can play a constituting role with respect to other discourses. Its constitution can be investigated with recourse to three dimensions:


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


• constitution as […] a process by which the discourse is set in motion and constructs its own emergence in interdiscourse;

• the way discursive cohesion is organized, constitution in the sense of an arrangement of elements that form a textual totality;

• constitution in a juridical and political sense; the establishment of a discourse which, as a norm and guarantor, regulates the behav-ior of a collectivity. (Maingueneau and Cossutta 1995: 113)48

Maingueneau draws from Benveniste’s theory of deictic ‘engagement’ (embrayage) with the context of enunciation. However, he radicalizes the pragmatic turn by breaking with the idea of a universal langue. In the enunciation, text and context reciprocally tie in with each other; the border between ‘linguistic’ forms and ‘external-linguistic’ contexts collapses (cf. Maingueneau 1993).

Another variation of enunciation theory is proposed by Jacqueline Authier-Revuz, who sees the enunciation as an operation within the language system (langue) even though the latter must not be conceived of as a closed code. Instead, language is constantly confronted with its own ‘non-unity.’ As a Saussurian with a psychoanalytic background, she emphasizes the constitutive heterogeneity of language, which has at its command certain markers which reflect its heterogeneity again and again (Authier-Revuz 1982). Authier-Revuz examines the ways in which language reflectively comments upon itself in ‘enunciative loops’ (boucles énonciatives) by means of which language attempts to become one with itself. Enunciative loops are introduced in various ways, for example through formulae such as that is …, not to mention x, or in terms of x, by quotation marks designating reported speech, or by retro-spective attempts to clear up things said elsewhere:

Through these disjunctions—through which everything that marks the disunity of communication: non-comprehension, uneasiness, lack, misunderstanding, ambivalence—‘language recalls itself’, in its reality, to its speaker; and simultaneously […] ‘the speaker recalls language’; his enunciation appears hollowed out in these points by a distance between the observer and the observed object, which establishes the speaker—who is no longer ‘one’ with his words—in a state of non-coincidence with the words he enunciates. (Authier-Revuz 1995: iiff.)49

In speech, language trips over itself and provokes hesitations that confront the speaker with the real of langue, with its opaque materiality,


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


with the internal Other. Thus, a discourse about language emerges which exceeds the control of the speaking subject. As in psychoanalysis, so, too, for Authier-Revuz the notion of the split subject prevails, which tries to achieve an illusion of inner unity by using language. As soon as the subject starts to speak, his or her words become opaque and elude him or her. It is this elusive clouding of discourse that calls for meta-enunciative comments about the enunciation. While Benveniste tries to extend the language system (langue) by the dimension of the enunciation, Authier-Revuz grasps the enunciation as a constitutive component of the langue, whose rifts and lacunae demand constant reflexive suturing in enuncia-tive loops. The point is not to add the enunciation to the langue, but rather to detect the internal instability of the langue in the enunciation.

It is Antoine Culioli who, since the late 1960s, has influenced the theoretical discussion of the enunciation like no one else. In contrast to Authier-Revuz, who embeds the enunciation into the langue, Antoine Culioli sees the operations of the enunciation as a general mechanism of meaning production which shatters the framework of the langue. In contrast to his teacher Benveniste, as well as to Ducrot (who grafts the enunciation as it were onto the language system), for Culioli the operations of the enunciation continue to be at work even on the level of forms and concepts. Culioli’s analytical object is the ‘enunciative marker’ (marqueur)—a formal trace of the utterance which stands for a ‘kind of condensation of procedures which trigger and activate rep-resentations’ (Culioli 2002: 172).50 The enunciative marker is the end product of a complex process of construction in which formal-linguistic and cognitive operations converge. Therefore, language is ‘an open system’ (1999: 48);51 it is in dynamic change, continuously in search of formal solutions for the production of meaning.

A text is not a representation of a reality cut to measure in advance and stable for all speakers. If one deals with an utterance or a textual sequence, one deals with an arrangement [agencement] of markers. […] Markers are the representations of ideas. (1985: 16)52

Linguistic activity takes place on three levels: (1) the level of cognitive operations, ‘which are not directly accessible,’ (2) of utterances, that is the ‘materiality of the text, which is directly accessible’ and (3) the meta-linguistic ‘practice of the linguist who will simulate the level of operations.’ (2002: 185; see also 1990: 22ff.)53

By means of enunciative markers, ‘a constituting object ‘dives’ into a system of reference with inter-subjective and space-time coordinates’


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


(2002: 36).54 Thus, the utterance is surrounded by a space of references in which the linguistic object takes on a certain value: ‘To enunciate means to construct a space, to orient, to determine, to establish a network of referential values, in short: a system of references [système de repères]. Every utterance is positioned with respect to a situation of enunciation that is defined by an enunciating subject S0 (or, more exactly, a first enunciating subject) at a time of enunciation T0, to just name these two referential variables [repères]’ (1999: 49).55 In order to produce or recog-nize an utterance, the individual needs to determine a point of reference which allows us to distinguish between ‘I certainly think that’ [Je crois bien que…] and ‘I think that…’ [Je crois que…], between the meanings of ‘a little’, ‘a lot’ or ‘too much’, or, more simply, to understand the difference between ‘in front’ and ‘behind.’ Culioli is not interested in the factual situations of enunciation. He wants to describe the construction of linguistic forms, which includes a complex of enunciative operations.

The definition of the enunciative system of references raises the question of how individuals engaging in linguistic exchange reach a mutual agree-ment over what is said. Thus, ‘with the discovery of the utterance, there arose the problem of ‘inter-subject’ relations and the fundamental prob-lem of the non-symmetry between production and recognition’ (1999: 11).56 While Culioli sees linguistic activity as a process of constructing meaning ‘which triggers a representation in the other’ (2002: 32),57 the emerging representations on the two sides of the communicative process are not congruent: ‘There is no symmetry between two subjects. We may have the illusion of making the ‘communication’ symmetrical, transpar-ent, because sometimes we have the illusion that we have been understood perfectly’ (2002: 28).58 When we speak with each other, we have to ‘live with permanent backfires and approximations, and still we somehow get out of it’ (2002: 221)59 because ‘we succeed in conquering clarity’ (1985: 2).60 Nevertheless, even if we consider ‘understanding as a special case of misunderstanding’ (2002: 28)61 or as ‘a kind of optimization of flops [of communication]’ (2002: 189),62 the individuals still manage to tune the production and recognition of the utterance. By way of the utterance, sub-jects are arranged ‘inter-subjectively’; they are positioned, put in relation: ‘to the degree that we speak, language serves to regulate us, to regulate with respect to the other and with respect to ourselves’ (2002: 196).63 Subjects are arranged ‘inter-subjectively’ by defining a source of enunciation (the enunciators) and an end of enunciation (the co-enunciators), both of which are part and parcel of the utterance. It is the play between the forms and meanings of the reflexive relations (which language constantly takes upon itself) that prevents the establishment of an exchange ‘without


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


loss’ between the partners of communication as well as the competition between the two different referential systems of the enunciator and the co-enunciator. If communication ‘is based on this more or less successful, more or less desired adjustment between the referential systems of two enunciators’ (Culioli 1999: 48),64 then the problem of the ‘inter-subjective’ relation exhibits itself ‘in’ the utterance, namely before ‘real’ individuals engage in a ‘real’ communicative exchange.

From Benveniste, Culioli adopts the concept of the enunciative marker, which he develops for the purposes of a general constitution theory of utterances (cf. Vogüe 1992: 85). With the logical-semantic theory of enunciation Culioli shares the assumption that utterances operate with more than one speaker or enunciator and that they instruct the subjects about the ‘inter-subjective’ arrangement. Even though for Culioli there is no access to the cognitive operations and representations without passing through the material forms, he also forges links with the cognitive approaches of radical-pragmatic origin. By bringing together different theoretical currents in the field of French enunciation theory, Culioli firms up the theoretical base of enunciative linguistics, which distinguishes itself both from grammatical and objectivist (symbolic practice as the actualization of a code) as well as situative and subjectiv-ist (symbolic practice as situated acting) approaches to discourse.

For readers who are not familiar with enunciative linguistics or discourse pragmatics, the following section gives an overview of the numerous formal markers of enunciation as well as the contextualization rules that they involve. We can then search texts for these markers of enunciation by means of which the enunciation is inscribed in them in at least three ways: by referring to the enunciation context (indexicality or discursive deixis), by orchestrating the sources of enunciation (polyphony), and by mobiliz-ing what was enunciated before and elsewhere (preconstruct).

Elements of enunciative discourse analysis: indexicality, polyphony, preconstruct

The enunciative approach which will be applied in the following analyses draws from structural, logical-semantic and radical-pragmatic theories of enunciation. Like radical-pragmatic approaches, enunciative analysis does not consider discourse as the product of an overarching grammar or code which the utterances reproduce in some teleological manner (cf. Reboul and Moeschler 1996). Yet unlike actor-based trends such as interactionism and conversation analysis it does not depart from specific situations and actions but insists on the opaque materiality


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


of the symbolic forms. A form is a (repeatable) sign which triggers cognitive operations and representations as soon as it is recognized by the reader who fills the gaps of the text with specific (unrepeatable) contexts. When predicated by other forms, a form becomes a sentence. Existing in certain enunciative modalities, sentences become utterances by ‘plunging’ into a system of references as well as by orientating their ‘before’ and ‘after.’ When utterances are combined with other utterances, discursive formations emerge. Following Foucault, we can understand a formation discursive as referring both to a constituting process (‘forming’) as well as to a constituted structure (‘formedness’). As the smallest constitutive ‘unit’ of discourse, an utterance can be examined by enunciative analysis in three ways: with respect to the deictic reference to the context of enunciation, with respect to the poly-phonic organization of the sources of enunciation and with respect to the mobilization of preconstructed knowledge.

Unlike qualitative research, enunciative analysis does not aim at reconstructing the meaningful whole, the social context, the subjective world or objective setting in which the utterance is produced. The anal-ysis begins by scanning the symbolic material for forms that in some way reflect the enunciation. If this procedure does not answer how texts are ‘really’ understood, it emphasizes that texts cannot be understood in whatever way. Thus, the forms with which texts operate instruct the reader about the relevant co- and contexts. These contextually under-specified forms organize the discourse by having the readers search for the contexts in which they are enunciated. Like Foucault’s Archeology, enunciative analysis (1969: 143[122]) is conceived as a ‘felicitous positivism’ (1971: 72[234]),65 which starts from the forms that orientate the reader staking out a framework for possible interpretations.

In the last 30 years, the research on enunciation has produced exten-sive inventories systematically listing the linguistic forms that organize the enunciation. Based on Benveniste’s structural theory of enunciation, a first compilation was produced by Catherine Kerbrat-Orecchioni (1980). Kerbrat-Orecchioni gathers the elements that make up the formal appa-ratus of enunciation. While the deictics I, here and now, which refer to the person, the space and the time of the enunciation, can be considered as the ‘classic’ elements, Kerbrat-Orecchioni points out the numerous derivatives that show the discursive object in enunciation, rather than saying or representing it (Barbéris et al. 1998).

As Benveniste (1966: 251ff.[217]) suggests, there are two types of personal pronouns: pronouns of the ‘person’ and of the ‘non-person.’ While the subjective or first person I and the ‘non-subjective’ or


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


second person you belong to the former, she/he/it and they belong to the latter. This differentiation between ‘person’ and ‘non-person’ is justified by two different types of reference: on the one hand the deictic or exophoric reference of the ‘person,’ which juts out, as it were, into the context of enunciation, and on the other hand the co-textual or endophoric reference of the ‘non-person,’ which remains submerged in the textual surroundings. Proper names referring to a given person (for example the author’s name on the cover of this book) testify to a third, ‘absolute’ reference which indicates its object in a conventional way (1980: 36).

The reference structure in plural pronouns is complex. We combines either a deictic, subjective component (I ) with a deictic, non-subjective component (you) or a deictic, subjective (I ) with a co-textually defined component (such as the non-person she). These pronouns produce a number of derivatives, for example from I can be derived me, my, mine, myself … (1980: 41). Even expressions for family members can some-times be considered an example of deictic reference to the person (such as dad and mom, which may refer to the parents of the speaking person). Non-European languages often have at their disposal an even more differentiated deictic system of family relations (1980: 54).

Another group are demonstrative pronouns like that, this, the referring to a temporal or spatial position (1980: 44). There are considerable differences between these pronouns in different languages (celui-ci/celui là, der/dieser/jener, while Slavic languages like Russian often do without any article). Sometimes these pronouns can act both as deictic and as co-textual references (like the German der/die/das). In order to signal deictic reference, the use of these pronouns is often accompanied by a gesture such as pointing. Moreover, a number of co-textual references are available (such as in English first or latter), which may also include relative pronouns.

Moreover, languages have at their service a large number of forms that define the spatial and temporal position in which discourse takes place. In order to systematize the different ways of temporal positioning, Kerbrat-Orecchioni (1980: 45) makes a distinction between T0, the time of the enunciation, and T1, the moment defined in the verbal context or co-text. Thus, we can mention adverbs of time that operate with deixis (that is reference to T0) like just now, yesterday, a few hours ago, recently or tomorrow, next year, soon, tonight, this morning, immediately. Expressions such as at that time, the day before, the week after, shortly before; the day after, a short time later; some other time usually define their point of reference time through co-text (that is with T1) (1980: 45). In French


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


the different systems of reference (which Benveniste places under two registers: discours and histoire) are connected with two different tenses, passé composé and passé simple. In addition, temporal prepositions such as since (depuis) (reference point: T0) can be distinguished from, for instance, from … until (à partir de) (reference point: T1). Both types of reference can be triggered by temporal adjectives such as: modern, old, future, recent… (1980: 48). Finally we can point out the deictic reference of the present tense. Similarly to the French passé compose, the English present perfect is aligned to T0, as opposed to the co-textual reference of the past tense or passé simple aligned with T1 (in German the reference of past and future tenses can be both deictic and co-textual).

Spatial orientation can also be organized with the aid of deixis. This holds true not only in the case of demonstrative pronouns, but also for other ways of indicating a spatial position: near versus far, before versus behind, right versus left, come versus go. Whereas movement verbs can operate with absolute reference (Peter goes down the stairs), co-textual reference (Peter approaches London) or deictic reference (in French the verbs aller/venir which imply a spatial position of the speaker) (1980: 49ff.).

Through deictic markers, subjectivity is inscribed into language. By using these forms, the speakers leave their subjective trace in language. According to Benveniste, ‘languages [le langage] are so organized that they permit each speaker to appropriate to himself an entire language [langue] by designating himself as I’ (1966: 262[226], italics by E.B.).66 Kerbrat-Orecchioni goes a step further by distinguishing between two ‘discourses’ (see Benveniste’s discours and histoire). The first is a ‘subjective discourse,’ ‘where the speaker explicitly (I think it is ugly!) or implicitly (This is ugly!) presents himself as a criticizing source of the utterance’ (1980: 71).67 The second ‘discourse,’ by contrast, is an ‘objec-tive discourse’ which ‘tries to erase all evidence of the existence of an individual speaker’ (1980: 71).68 As Kerbrat-Orecchioni points out, the transition between objective and subjective discourses occurs gradually and is not dichotomic (ranging from objective adjectives like single or yellow by way of ‘mixed’ adjectives such as small or thick to subjective adjectives like good or hideous). Semantic and evaluative components often mix in subjectively colored words (‘subjectivemes’). For example, Paul is a fool is not only an utterance about Paul’s intellectual capacities, but also an evaluation of his intelligence.

Kerbat-Orecchioni attempts to list adjectives and verbs which signal the subjectivity of the speaker. She contrasts affective and evaluative adjectives, the latter being further divided into axiological and non- axiological adjectives. Affective adjectives ‘imply an affective


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


engagement of the speaker’ (1980: 81)69 while evaluative adjectives mark the proximity or distance of the speaker. Evaluative adjectives are non-axiological if they are used ‘according to a normative standard that the speaker has for a given category of objects’ (1980: 86).70 In contrast, axiological adjectives prove to be ‘doubly subjective,’ because ‘firstly, their use varies […] with the special nature of the subject of enunciation whose ideological competence they reflect, […] secondly, they show a positive or negative attitude on the part of L [the speaker, J.A.] toward the object at hand’ (1980: 91) (Figure 2.2).71

Verbs are also classified by their subjective components. Kerbrat-Orecchioni makes a distinction between verbs of judgment (prefer, hate), of locution (pretend, confess) and of belief (fancy, think, know) (1980: 109ff.). Verbs are usually nominalized in the same way as adjec-tives (love, lovely → love; confess → confession, proud → pride). Most nouns deriving from these verbs keep the enunciative features of the original expressions. The same applies to adverbs, which may set a subjective tone to the whole utterance (for example Fortunately, you have already come). In these cases the mixing between objective and subjective components becomes even clearer. Yet it is doubtful whether the enunciative functions of these words can be determined abstractly, that is without taking into account the specific contexts of enunciation in which utterances are produced or received. This is where Kerbrat-Orecchioni, too, sees the limits of structural approaches, which departs from an abstract notion of the language system (langue) according to which every linguistic form corresponds to a given grammatical rule which cannot be deviated from without violating it. Thus, the structural theory of enunciation especially has difficulties in coming to terms with the phenomenon of reported or indirect speech, with displaced speech

objective

single/married,

male/female,

colors…

subjective

affective

funny, intensive

evaluating

non-axiological

big, far

axiological

good, pretty

Figure 2.2 The system of adjectives


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


where the speaker is not present in the space of enunciation, or with enallage, the non-conventional use of words (1980: 57ff.).

In her empirical investigations, Kerbrat-Orecchioni makes a distinc-tion between the ‘subjectivities’ of texts. She distinguishes ‘subjective,’ ‘interpretive’ and ‘axiological’ subjectivities, each operating with its own repertoire of formal markers. Yet it would be problematic to see subjectivity as a property of the text, as in the good old days of the structuralist critique of the subject, when the symbolic order was held to be a system of subject-positions for which individuals were ‘recruited.’ Texts do not define homogeneous subject-positions by means of which the symbolic order recruits its subjects. Therefore, Kerbrat-Orecchioni’s structural grid appears rather rigid for the problems of empirical dis-course. Nevertheless, what can be salvaged from Kerbrat-Orecchioni is the reference to the manifold ways in which texts refer to their contexts by means of deictic markers of enunciation.

The second strand of enunciative analysis leads back to the theory of polyphony as developed by Oswald Ducrot (1984) following Mikhail Bakhtin (1929). According to Ducrot the utterance orchestrates different ‘voices’ which its locutor holds at a certain distance. The utterance is associated with other utterances depending on its voices that are orchestrated in a certain way. However, Ducrot is not interested in the bodily human speaker of the utterance but in the speakers or, rather, enunciators belonging to the utterance. Thus, we can say that the speaker of the utterance on a delivery van, Rent me, does not refer to the human who wrote the words on the van but to the van itself. The question of who is speaking can never be answered in immediate evidence. In order to answer the question how the utterance defines its sources of enunciation or enunciators, it is necessary to dissect the different enunciative layers that make up the utterance.

Ducrot’s theoretical impulses have been elaborated by Henning Nølke, Kjersti Fløttum and Coco Norén in the ‘Scandinavian theory of linguistic polyphony’ or ScaPoLine (Théorie scandinave de la polyphonie linguistique), which allows a systematic description of the polyphonic configuration of the utterance (2004). According to Nølke et al. an utter-ance is composed of at least four components: the locutor, the points of view (Ducrot’s ‘voices’), the discursive beings (Ducrot’s ‘enunciators’ or ‘speakers’) and enunciative connections. The locutor L (locuteur- en-tant-que-constructeur) is the ‘puppet master,’ the responsible agent of the utterance who orchestrates the different points of view ‘pov.’ Points of view are ‘semantic unities whose source supports the point of view.’ The discursive beings (the ‘speakers’ or ‘discursive figures’) ‘saturate the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


sources’ or fill the enunciative positions of the enunciation, whereas enunciative connections ‘tie the discursive beings to the points of view’ (2004: 30).72 Although never appearing himself, the locutor marks his presence in the utterance by constructing ‘images’ of himself vis-à-vis the allocutor A and the third party who exists individually as a ‘non-person’ or collectively (as ONE/WE (ON) or as LAW (LOI)). There are two ‘images’ of the locutor L: the ‘responsible’ speaker l0 of the utterance as well as the textual locutor which is formally realized as pronouns, names and so on. Accordingly, the two images from the allocutor are a0 and its textual realizations. While l0 and a0 refer to the locutor and his allocutor both speaking in the enunciation of the utterance, the textual realizations refer, in contrast, to the formally marked locutor and his counterpart. Nølke et al. do not divide the third persons (tiers) in order to avoid complicating the model, but they distinguish between two col-lective thirds: a heterogeneous ONE/WE (as a background of opinions, postures and knowledge in their manifold existence) and a homogene-ous LAW (whose normative presence calls for recognition) (Figure 2.3).

With these terminological distinctions in mind we can now system-atically determine the speakers who have their say in the utterance. ScaPoLine makes a distinction between simple and complex points of view.

the responsible speaker of the utterance: l0

L:

the textual speaker

the allocutor of the utterance: a0

A:

the textual allocutor

individual third

third party: heterogeneous: ONE/WE

collective third

homogenous: LAW

Figure 2.3 The images of locutor, allocutor and the third party


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


The simple point of view takes on the form of a predication, for example Pierre goes for a walk or The weather is nice. It is constituted by a semantic content as well as by a judgment about the content, made by default in the state of lacking further information: ‘It is true that …’ The source of the point of view remains undetermined. Each utterance contains at least one simple point of view whose semantic content is posed. (2004: 33)73

From an analytical point of view, the more interesting case is when the utterance is not produced in the default-mode ‘It is true that….’ In complex utterances, different points of views are superimposed and encapsulated both hierarchically and relationally. An example for a hierarchically encapsulated point of view is the utterance (1) Perhaps Peter has not eaten. Nølke et al. (2004: 41) describe the polyphonic configuration of this utterance by isolating three points of views povx. These three points of view intermesh with each other in a certain way. As ‘perhaps (q),’ the utterance contains a semantic content q (‘Peter has not eaten’) which has its own polyphonic structure, namely ‘not p’ (p: ‘Peter has eaten’):

pov1: [x] (TRUE (p))

pov2: [l2] (NO (pov1))

pov3: [l0] (PERHAPS (pov2))

In the formal representation it becomes clear that in the utterance Perhaps Peter has not eaten, there exist three nested points of view pov1, pov2, and pov3 (pov3 and pov2 refer to pov2 and pov1 respectively). Thus, each point of view (the enunciators being inside the square brackets) is tied to another point of view by an enunciative connection ‘TRUE, NO, PERHAPS….’ Out of the three points of view that the utterance offers, the locutor l0 finally opts for pov3 since the negation is qualified by the adverb perhaps. The enunciator x stands for a source of enuncia-tion which is lost as it were in the process of composing the utterance from its different points of view. However, depending on the context of enunciation, this ‘wild card’ enunciator can turn into an image of the allocutor or of a third party. Thus, in the utterance, different points of view come together that are each supported by a different source of enunciation. Meanwhile the locutor operates with more than one point of view, which he rejects one after the other until only one remains. In this polyphonic jumble, the locutor continues to romp in the company


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


of the rejected points of view, we never speak alone; we let others speak along with us in the enunciation, either those who are pitted against the locutor (the allocutor A) or those who murmur in the background (a third party). These voices may refer to the passive presence of the ONE/WE or the prescriptive existence of the LAW. For this reason, as a unit through which the Other speaks as well, the utterance stages a kind of ‘drama’:

The author of the drama is L. It constructs the polyphonic play, without taking part in it itself. The actors of the drama are the discursive beings [in the following ‘enunciator’ or ‘speaker,’ J.A.]. L creates their roles, it can create roles for his own images. Accordingly, it can create roles for other figures (in particular for the allocu-tor), who are present in the world of which the theater is a part. (Nølke et al. 2004: 55)74

Just as in Benveniste’s theory of deixis, the theory of polyphony assumes that the discursive dynamism of texts (their connection with the contexts of enunciation) is a process triggered by certain forms that orientate and instruct the interpretive process. There are formal con-straints on interpretation, but signs do not interpret themselves. They require a discursively competent individual who, in recognizing the forms, completes the text with contextual information according to the rules and instructions contained in the forms.

As such, the utterance contains information on the protagonists, the situation of enunciation, etc. This information is available to the interpreter who applies the interpretive strategies, of which the most important principle will be: ‘Try to saturate as far as possible all variables which are transmitted in the signification’ (2004: 24).75

Analytically speaking, the most interesting forms include those that connect points of view in a complex way, that is either in an encap-sulated hierarchy or in a relational sequence. Standard examples of hierarchically nested points of view are negation and irony. Negation disposes of numerous formal operators, such as not, nobody, never, un-, without, instead of… which present a semantic content p whose enuncia-tor is rejected by the locutor. For the polyphonic theory of negation, the positive content must always be ‘spoken aloud’ in the negating phrase. Accordingly, This wall is not white manifests indeed a point of view pov1 for which the wall is white. Through the operator not, pov1 is rejected by the point of view pov2 which is supported by the image of the locutor l0.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


pov1: [a1] (TRUE (p: ‘The wall is white’))


Another case for complexly interlocking points of view is irony. Irony, as a rule, does not work with formal operators, but derives from otherwise triggered cognitive disturbances of the ‘default’ interpretation. Thus, the irony of Under W.’s Presidency, U.S. foreign policy finally discovered the humanitarian cause is signaled by the tension between different styles: ‘W.’ versus ‘humanitarian cause.’ Cliché-ridden knowledge plays a central role here. It may create a ‘too obvious’ contrast between what is said and the ideological position of the speaking subject. However, ironic utter-ances are always in danger of being interpreted as factual utterances (that is according to its default-value). For this reason, other, non-symbolic hints are usually needed (such as intonation, eye-rolling and so on).

pov1: [a1] (TRUE (p: ‘Under W.’s Presidency, U.S. foreign policy finally discovered the humanitarian cause’)), where p is ABSURD


Moreover, relationally linked points of view can be named. Argumentative connectors play a central role in this context, which Ducrot had already analyzed in the case of but. This argumentator brings two utterances in contact, whereby both the foregoing utterance p and the successive utterance q are accepted by the locutor (‘It is true that p and q’). But (‘p but q’) creates an effect whereby, in the light of q, the preceding utterance p suddenly appears in a new perspective, which in turn calls into question certain presupposed points of view. Thus, the above-mentioned utterance I brought ham, but the refrigerator is broken can be formulated as following:

pov1: [l1] (TRUE (p: ‘I brought ham.’))

pov2: [l2] (TRUE (q: ‘The refrigerator is broken.’))

pov3: [a3] (TRUE (P from (pov1) BUT (pov2)))


pov5: [a5] (QUESTION (Q from (pov4))

Pov3 is the argumentative result of the utterance. It ties the second part to the first by assigning its presupposition (P: ‘Ham needs cooling’) to


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


the allocutor a0 (or to a third party) and by having the allocutor raise a question Q (for example ‘What do I do with the ham?’). In the for-mal representation of the polyphonic configuration the sense of the utterance extends to five superimposed enunciative levels. However, presuppositional knowledge is hard to formalize (see pov5). While Ducrot points out that the utterance operates with both explicit and presupposed contents, relational points of view and the problem of presuppositions are handled only superficially by Nølke and co. To give an example, the utterance Peter has stopped smoking presupposes that Peter had smoked at an earlier time. As a rule, presupposed contents belong to non-negotiable knowledge, which is evoked by stopped doing or but in a certain way. Being more or less attached to the utterance, presuppositions are points of view which cannot be questioned without threatening the communicative pact between the writing and reading individuals. Thus, presupposed knowledge is often used in exclusive discourses with clear demarcation lines between inside and outside. For instance, in ideologically charged utterances (such as The integration of immigrants is one of our most important political tasks), everyone who does not subscribe to the implied assumption that foreigners are not integrated is forced into an allocutor or third party position. By defining the sequential organization, the utterance marks zones of controversial and uncontroversial knowledge.

From a polyphonic perspective, the utterance is teeming with differ-ent points of view. Yet it is not always necessary to dissect the utterance up to the last possible enunciative layer. There can be no complete list of all the possible discursive beings inhabiting the utterance since it does without a fixed source of enunciation. Let us say the utterance p (‘The weather is nice’) operates with a source of enunciation x: x(p) (for example, I say: ‘The weather is nice’) which can be traced back to yet another source y: y (x(p)) (that is He says: ‘I say that the weather is nice’). Theoretically, this process of incorporating ever more points of view can go on indefinitely: z (y(x(p))) (on this point see Kerbrat-Orecchioni 1980: 246). The Scandinavian theory of polyphony argues that the reader will stop this process at some point, when a default-value can be reached that saturates the sources of enunciation (Nølke et al. 2004: 24).

In order to simplify the representation I suggest a ‘test’ which verifies the polyphonic configuration without requiring too much formalism. This test ‘conjugates’ the polyphonic constitution of an utterance, for example: (0) I am not a racist, but the number of foreigners is too high. In the Scandinavian polyphony theory at least five layered points of view can be made out:


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


pov1: [a1] (TRUE (p: ‘X is racist’))


pov3: [a3] (TRUE (q: ‘The number of foreigners is OK’))

pov4: [l4] (TOO MUCH (pov3))

pov5: [l0] (TRUE (P from (pov2) BUT (pov4))), where from Pq it follows r: ‘The borders need to be closed.’

In the polyphonic test, the points of view that constitute the utterance are transformed into utterances of their own. Thus, the ‘polyphonic test’ reveals the Other that the locutor evokes in the enunciation. In order to test the polyphonic configuration of the utterance, it is necessary to transform its points of view into a dialogue of different individuals. This is indeed what we unconsciously do in everyday situations when we try to understand ambiguous or ambivalent utterances. We interpret the utterance by distinguishing its points of view and linking them with certain real or imaginary interlocutors. Thus, the negation in the first part of the utterance contains an implied exchange where somebody says that ‘X is a racist,’ which moves somebody else to answer: ‘No, x is not a racist.’ In this case, x is made deictically visible by the pronoun I. The second part of the utterance is also organized by some dialoguical mechanism that defines what is explicitly said through the demarca-tion from what is presupposed (cf. in more detail Ducrot 1980b). Thus, the utterance (0) can be split into different utterances: (1): X is a racist, (2): X is not a racist, (3): The number of foreigners is exactly right, (4): the number of foreigners is too high, which bears witness to the polyphonic character of both parts of the utterance. These utterances may animate a dialogue between two individuals (here: between a pro-immigrant and an anti-immigrant person). But they can also be merged again to form one utterance through hierarchical composition. A negated utterance (5) can be produced, for example, from (3) and (4): No racist finds the number of foreigners just right. Or by means of argumentative operators like but the utterances can be arranged relationally, which may generate the ironic utterance (6): Indeed, I am racist, but there are too few foreigners. As a result of these polyphonic rearrangements, the ideological values of the produced utterances can change. Indeed, (5) and (6) operate with the same points of view as (0); yet they are orchestrated differently by the locutor. Is it not by ‘restaging’ the polyphonic configuration from (0) to (5) and (6) that the latter appear as more or less efficient ideologi-cal counterarguments to (0)? If (5) and (6) belong to the same discourse


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


as (0), their argumentative effects are different because their points of view are composed in different ways. Does the polyphonic test not assume that the Other is always present in discourse and a monophonic discourse is not possible? Mustn’t ‘the question ‘who is responsible?’’ (Nølke et al. 2004: 56)76 always be asked anew? And isn’t the question—who speaks in the utterance?—ultimately undecidable?

A third analytical perspective can be gained from Pêcheux’s concept of the preconstruct. With this concept, Pêcheux tries to account for what was said before and elsewhere (as opposed to that what is ‘con-strued’ by the utterance itself). By way of the preconstruct, something outside protrudes into discourse and presents itself as a self-explanatory knowledge, needing no justification. A central component in Pêcheux’s theory of discursive formation, the preconstruct refers to the social and institutional conditions of production in which a discourse emerges. Additionally, it underscores the interdiscursive nature of discourse. Thus, discourse is not the expression of a single source; there are always other people speaking along; this is what the preconstruct’s sudden appearance testifies to. The preconstruct manifests itself in the fissures and rifts of interdiscourse, by which discourse signals that it is not one with itself.

Pêcheux et al. conceive of the preconstruction as an alternative to Ducrot’s theory of the presupposition. While Ducrot takes the presup-position as knowledge resulting from the way in which specific utter-ances are connected in a sequence, Pêcheux considers the preconstruct as a knowledge whose origin lies elsewhere (cf. Henry 1977). Unlike the presupposition (a ‘discursive’ knowledge produced in a sequence of utterances), the preconstruct is an ‘ideological’ knowledge that requires recognition by the subject. The preconstruct usually reflects a society’s commonplace ideas which impose themselves as self-evident truths. Pêcheux tries to grapple empirically with the preconstruct by relying on a discursive semantics which extends Saussure’s structural theory of language by a discursive dimension (Haroche et al. 1971). If, according to Pêcheux et al., the constitution of meaning is incumbent upon langue, it is necessary to account for the place where symbolic practice takes places: ‘We find out that the words can change their meaning accord-ing to the position of those who use them’ (Haroche et al. 1971: 97).77 This theoretical program is put into practice in their studies of explica-tive (non-restrictive) and determinative (restrictive) relative clauses. For example, the non-restrictive relative clause, which in English is separated by commas, signals preconstructed knowledge that is in no need of further justification. In view of the porous boundary between


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


restrictive and non-restrictive relative clauses, Henry emphasized the general importance of the problem of the preconstruct, which informs a large number of linguistic and discursive phenomena (1975). Ducrot’s presupposition is rejected for theoretical reasons as well. Members of the Pêcheux School reproach Ducrot for his stance on presupposition and enunciation, for rehabilitating the speaking subject and doing away with the structural critique of ideology in discourse analysis.

Around 1980, with the structural model falling into crisis, the preconstruct began to lose its appeal. In turning to the heterogeneous tendencies of DA 3, Pêcheux seems to have given up both concepts (the preconstruct and the discursive formation). Yet even though today the preconstruct no longer plays a central role in the theory or in the practice of discourse analysis, it may gain new relevance within an enunciative framework. In contrast to presuppositions, which emerge directly from utterances as they combine to form discursive formations, the preconstruct needs to be considered a ‘foreign’ element in the discursive process—a heteronomous component that draws its legitimacy from non-linguistic sources (such as socio-institutional lines of conflict or class rule). The preconstruct is a knowledge which is—as it were—cut off from its source of enunciation. Rendered anonymous, this knowledge works as long as it ‘goes without saying,’ needing no explanation.

What are the formal markers of preconstructs? In his study on Soviet discourse, Seriot (1985: 246ff.), following the Culioli circle, points out the role of transformations such as the nominalization of utter-ances, whereby an utterance loses not only the textual images of the locutor (I, mine …), but also its source of enunciation l0: ‘the nominal-ized utterance is preconstructed, that is its origin is not in the speaking subject; it is like an object of the world which is ‘already there,’ an object that exists prior to discourse […]: its conditions of production are wiped out’ (1985: 248).78 Using speeches made at congresses of the Communist Party of the USSR, Seriot shows how the frequent nominal constructions have the effect of purging the spoken material of its sources of enunciation. By means of nominalization, utterances are transformed into individual words: I want, you want, he wants, we want… → will; I am proud that…, you are proud of… → pride. In this way, nominalizations render what was said before and elsewhere by some-body else, and the subject of discourse has no choice but to appear in the name of an anonymous institutional power which has already said and decided everything: ‘In these [preconstructed] utterances without subject and reason, the enunciator is nothing but a witness: he ‘sees’ objects and processes in which he does not participate as an actor.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


His saying retreats behind his seeing’ (1985: 259).79 In the absence of individuals assuming the position of an I, discourse becomes ‘wooden.’ Certain verbs, like testify of, contribute to, require, characterized by, is accompanied by (for example The pride of x testifies to the will to y …), are especially suitable for this type of discourse. These verbs ‘merely bring two preconstructs into relation, merely state one relation between ‘objects of the world’ which in reality are pre-stated objects in a nether-world of the discourse’ (1985: 254).80 Thus, certain endings can give an ‘ideological’ touch to preconstructs, for example, the ending -ism which forms preconstructed movements or collective trends out of adjectives (for example liberal) or names/nouns (Marx). Therefore, liberalism or Marxism usually implies the existence of a group of speakers saying something like I am liberal or We believe in Karl Marx’s theoretical insights. Along with unnecessary relative clauses, nominalizations and endings, other forms and mechanisms can be expected. They turn the utterances expressed by a speaker into preconstructs, which means that the dis-course manifests itself as an interdiscourse, without, however, allowing this other speaker to become visible.

In conclusion, enunciative discourse analysis deals with the question how texts evoke by means of the formal markers of enunciation (marqueurs) the contexts, speakers, knowledge the reader needs to understand their meaning. The utterance is the smallest analytical unit of discourse (that is a sentence existing in a specific enunciative modal-ity). In order to analyze the symbolic material, it is necessary to identify in a first step the forms of the text that instruct the reader about what cannot be found in the text: the enunciation and its contexts. The sec-ond step is the question how discursive order emerges. By orchestrating their enunciative levels, sources, and contexts, utterances combine with other utterances and form discursive formations. Therefore discourse is not just a ‘horizontally’ organized ensemble of elements. Its utterances are also ‘vertically’ layered units whose enunciative levels do not necessarily constitute a harmonious whole. Given the vertical stratification and heterogeneity of the utterances, discourse does not constitute a completely structured space where each utterance occu-pies a defined place and transmits a precise meaning. Texts and contexts always combine anew in discourse; the space in which the utterance is enunciated is in constant motion as well as constitutively open.

Methodologically, enunciative discourse analysis goes beyond structuralism and hermeneutics. Unlike structuralism and semiotics, it does not consider discursive order as the product of a code or grammar which produces differential systems of smallest distinctive unities.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


And in contrast to hermeneutical approaches, its objective is neither to reconstruct the meaning lived by the subjects nor to reveal the inter- subjectively shared knowledge expressed by the symbolic material. Enunciative discourse analysis has no immediate access to the meaning and knowledge originating in the discursive process; what is accessible to empirical research are the forms by means of which the texts have the individuals search for a meaning that everyone must discover in her or his own fashion. Let us leave the business of understanding and interpretation to individual readers. As discourse analysts we focus our gaze on the forms, messages and instructions scattered among the symbolic material, sending the reader on quests for what is expressed, meant and intended.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

54

From understanding to analyzing discourse

In specialized as well as in everyday discourse, readers usually use texts to understand their meaning. Understanding is indeed the spontane-ous mode of dealing with intellectual discourse since texts are written to convey a message to be understood by the readers. Accordingly, in the social sciences and humanities, many interpretive methodologies have emerged that have recourse to understanding for example in ethnographic and historical research. Yet even though understanding is crucial for any strategy that aims to develop a certain expertise of a discourse, it can pose problems when discourse is to be made an object of empirical investigation.

• The black box problem. While interpretive research sometimes treats the analysis of the textual material as an art, hermeneutic and phe-nomenological approaches claim to reconstruct an original meaning of texts that can reveal itself to the competent reader through some kind of direct transmission. However, rather than exploring what a discourse means, should we not concentrate on how the discourse participants interpret the textual material?

• The bird’s-eye view problem. To the degree that meaning is considered to be just a quality of the world ‘out there,’ some of the more histori-cal strands presuppose an omniscient observer. Discourse turns into a meaningful totality which the analyst is encouraged to grasp as such. Yet do we not have to account for the interpretive freedoms that we enjoy when we engage in the interpretation of texts?

• The problem of interpretive authority. Interpretive methodologies tend to claim a special authority over what are good and bad interpretations.

3A Methodology of Discourse Analysis


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

A Methodology of Discourse Analysis 55

Yet while they can make sense in the case of everyday discourse, the highly specialized discourse of intellectuals and academics inevitably raises the question of interpretive authority. Thus, if we had to base our analyses on the ‘correct’ interpretation of the texts of intellectual discourse, we would have to perform exactly what the symbolic producers do themselves: to argue about what others meant to say in order to mark their difference from other produc-ers. Would we not have to challenge the discursive expertise of the intellectual, who in his or her very existence is bound up in struggles over meaning?

With these points I do not deny that we usually indeed understand (or believe that we understand) what is meant when we have a conversation at the bus stop or in the classroom, read a newspaper or a dissertation on the sociology of intellectuals, write an email or a comprehensive history of Western philosophy. As an alternative to such an interpretive orienta-tion, I suggest a formal-qualitative methodology of discourse research that brackets the question how the texts circulating among intellectuals are ‘really’ understood by the discourse participants and how they can be interpreted and reconstructed ‘correctly’ by the discourse analyst.

In accordance with ‘quasi-qualitative’ research (Paillé 1996; Angermuller 2005), formal-qualitative tendencies of semiotics, argumentation, stylis-tics, narratology, logic, rhetoric and sometimes also of conversation and frame analysis, commonly deal with objects whose complexity can hardly be accounted for by the positivist paradigm with its focus on measurement, statistical correlation and causal explanation. To be sure, formal-qualitative research usually does not follow quantitative or positivist research designs. But unlike many interpretive strands in qualitative social research, it does not conceive its object following a reconstructive-interpretive methodology.

By pointing out the formal-qualitative methodology of French discourse analysis, I want to highlight the following features that have characterized its theory and practice in the last 30 years:

• Materiality of form. The analysis privileges the opaque materiality of symbolic forms. Such forms are not just the secondary expressive container of a primary meaning’s content. They constitute a material surface devoid of any concealed ‘beyond’ (meaning, intention, know-ledge, interpretation, interest …). The signs and practices of discourse are ‘material’ insofar as they possess no inherent meaning. Meaning is an effect of the interplay of the symbolic-material elements with the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


context in the interpretive process. Like hermeneutics and interpretive social research, formal-qualitative discourse analysis underscores the meaningfulness of its object. But from a formal-qualitative perspec-tive, discourse operates with too much meaning, as it were, to allow recourse to ‘its’ meaning. Because ‘the’ meaning is neither subjec-tively nor intersubjectively verifiable and constantly exceeds what it ‘means’ to say, formal- qualitative methodology emphasizes the opaque materiality of its material. The meaning of the material does not impose itself in the guise of immediate self-evidence. Nor can it be reconstructed as an act of subjective or intersubjective agreement. For this reason, the formal-qualitative approach to discourse analysis does not work with hermeneutic understanding, interpretation and coding. Taking its point of departure in the graphic forms of the text, it prefers to postpone the moment of interpretation to the end of the research process.

• Break between object and theory. Formal-qualitative approaches accentuate the epistemological break between object and theory.1,2 The production of analytical knowledge is a process of active construction, extracting elements from the object-discourse and transferring them to the theory-discourse without leveling the distance between them. In this process, the categories of the object-discourse are refracted by the categories of the theory-discourse. In this discontinuous translation from object- to theory-discourse, new knowledge is generated which is not already present in the object. The question, then, is not whether the theoretical knowledge produced in this process gives an ‘objective’ account of the object, but how the theory-discourse intervenes in the object-discourse by re-arranging and transforming it.

• Reduction of complexity. Another characteristic of formal-qualitative research is its analytical procedures. By breaking the object up into its smallest constitutive elements and structuring mechanisms, it aims to ‘explain’ complex objects. Thus, the variety of empirical phenomena should, at least ideally, be reduced to the fundamental rules of production. Instead of reconstructing a socially shared stock of knowledge (‘what?’), discourse analysis focuses on the rules which organize the construction of discourse and its formations (‘how?’). These rules are transversal to individual or collective experience. Therefore, they cannot necessarily be deduced from the participants’ perspectives. In order to reduce the complexity of its object, discourse analysis does not seize its object as a singular meaningful


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


entity. Instead, it understands its object as a quarry out of which general analytic models are to be extracted.

The main difference between interpretive and formal- qualitative approaches lies in what is understood by ‘construction’ or ‘constructiv-ism.’ Interpretive approaches often rely on hermeneutic- phenomenological or interactionist notions of construction. For them, social reality is what the actors assume to be real. It is the actors who construct their world by communicating about it and interpreting it. For formal-qualitative approaches or ‘radical constructionists,’ by contrast, the world and its actors are constructed. In this view, actors’ greater or lesser control over what they do, say and understand is a (necessary) fiction. Responsibility and intentionality are apportioned in a complex, symbolically mediated process of construction which exceeds the control of individuals.

If I have classified the dominant currents of discourse analysis in France as ‘formal-qualitative’ tendencies, this is to emphasize their proximity to as well as their distance from standard qualitative social research. If one can observe a certain overlapping between the two, especially when looking at their questions and objects, one also has to highlight their methodological premises. While qualitative social research aims at reconstructing intersubjective structures of mean-ing resulting from actors defining a situation, formal-qualitative research understands both ‘meaning’ and ‘actor’ as effects of symbolic practice.

Last but not least, the plea for a formal-qualitative methodology is to be seen in the light of the written discourse that will be dealt with in this study. Everyday interactive situations, privileged by qualitative approaches, play a rather subordinated role in the intellectual discourse I wish to study. Intellectual discourse predominantly deals with written texts, read, cited and recited in different manners and in different con-texts, by readers with different discursive competencies.3 These texts are written without their producers knowing who will read them and in which situations they will be read. And often readers know the name of an author, but know little or nothing about the specific context in which the text emerged. With recourse to the analytical resources of French discourse analysis, viz. the linguistic theory of ‘enunciation,’ the following three sections will outline an enunciative-pragmatic approach to discourse. This approach will account for the multitude of possible contexts with which a given text can be associated (see Angermuller 2011, 2012). Like a message in a bottle washing up on shore, about


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


which no one knows who sent it on its way or who will find it, the text is on a never-ending search for its contexts of enunciation.

A discourse analytical research design

In order to analyze theoretical texts according to the enunciative dimension, a number of problems have to be resolved: constituting the corpus, carrying out the analysis and presenting the results.

• Constituting the corpus. Discourses usually operate with symbolic mate-rial that neither individuals nor groups can survey in its entirety. In order to investigate a particular discourse, a corpus needs to be constituted whose texts can be analyzed in more detail. It is generally not the nature of the discourse that justifies the constitution of a corpus; it must be justified theoretically. With respect to intellectual discourses, the constitution of a corpus is a particularly difficult task since these discourses criss-cross different regions of the intellectual field (for example academic, political and aesthetic regions) without their being organized around a given institutional place. For this reason, I will analyze intellectual discourse as an interdiscourse where various topics, regions and contents intersect. Even the smallest units of discourse—individual utterances—find their place on these interdis-cursive borders. The corpus is no homogeneous whole, but is composed of snippets and fragments which can be connected and extended in any number of directions. As a constitutively open-ended interdis-course, intellectual discourse must do without clearly defined external borders along which a natural corpus might be constituted. Against this background, it is both practically and theoretically impossible to reproduce a discursive totality by means of a representative, selected sampling of texts. The following analyses therefore do not pretend to offer a cross-section of some overarching field of knowledge; neither do they register the production of meaning of a social group; nor do they extract any representative sample from a discursive whole. A corpus consists of interdiscursive fragments—ensembles of utter-ances which reach out into different regions of the discursive space. In contrast to discursive approaches which try to order knowledge in a surveyable tableau, I will insist on the layered density of individual utterances that make up discourse.

• Carrying out the analysis. Once the corpus and its individual utter-ances are identified, the question arises how to analyze the symbolic material. The smaller the corpus, the less promising it is to use


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


quantitative methods. However, even in the following corpus, which consists of five discursive fragments, a complete and systematic examination of graphical forms must be undertaken. The forms can then be differentiated according to the contextualizing instructions that they contain. From an enunciative perspective, the most inter-esting forms are the marqueurs, the formal markers of enunciation, non-conceptual expressions such as the deictics (I, here, now…), the logical-argumentative operators of polyphony (not, but, if…), certain forms such as nominalizations and suffixes as well as typographical indications (like quotation marks and italics). These markers of enun-ciation, which signal the heterogeneity of discourse, trigger series of cognitive operations that each individual will carry out differently yet not arbitrarily. The individuals need not build up some stock of shared knowledge. What they share are the contextualization proce-dures condensed in the enunciative markers, with the help of which the individuals embark on the quest for a meaning that each reader must discover on his or her own. Thus, the interpretive process is subject to rules which are necessarily attached to certain forms. From this perspective, there are no rules without forms. The forms are the rules. And whoever attempts to describe the rules of discourse independently of its formal organization, risks missing the object of discourse analysis.

• Presenting the results. Finally, the results have to be presented. Rather than at the manifold contents, the analysis aims at the generalizable and reproducible rules by which texts refer to their contexts. What kind of contexts these ‘really’ are and what the individuals ‘really’ understand—this is what the enunciative discourse analysis has no access to. The enunciative approach departs from the symbolic material and its material forms, which send the individuals on the quest for a meaning each person will have to master individually. By formalizing the results, we emphasize that texts and contexts are not associated spontaneously in discourse. The interpretive process is subject to certain constraints which have left their formal traces in the text in one way or another. By breaking the utterance down into its various points of view, the formal description distances itself from the spontaneous understanding of the actors. This break prevents the discourse analyst from competing with readers who usually strive to grasp the content-based ‘what’ of the text. Unlike readers who search until they have found their meaning, discourse analysts enquire into the formal instructions the readers have to fol-low in order to gain knowledge about its context and to discover the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


meaning of the text. Unlike readers whose goal is to understand the text, discourse analysts bracket the problem of meaning in order to concentrate on the procedures that readers carry out to understand. Unlike the readers, who may solve the interpretive problems of the text in a thousand different ways—though not in just any way— discourse analysts aim at discovering the rules, mechanisms and laws of discourse, condensed in symbolic forms which no individual with a minimum of discursive competence can evade.

In order to analyze the selected texts, I suggest differentiating the forms of the symbolic material into two different types which I broadly designate as concept-words and enunciative markers (Culioli’s marqueurs). With this differentiation we can capture two different types of cognitive processes triggered once the reader recognizes a linguistic form. Concept-words generally refer to the content-related ‘what’ of intellectual discourse, that is to a knowledge which is attached rather loosely to these signs in the discursive process. These signs trigger series of associations and representations which can vary significantly depending on the individual reader. The degree of freedom at the indi-viduals’ disposal while interpreting these signs is therefore rather high. Enunciative markers, on the other hand, are forms which signal the ‘how’ of discourse, the mode of existence of these contents. Markers are not interpreted; they are spontaneously recognized. Through markers, language places high linguistic constraints on the interpretive process. This is because the individuals cannot recognize these forms without immediately and automatically applying the rules which they contain. If there are signs which constitute an ‘intersubjective bond’ of the symbolic, it is precisely those signs through which the enunciation is inscribed into the text.

Thus, the text operates with symbolic forms which allow for more or less interpretive freedom. In contrast to concept-words, which involve varying series of associations and representations according to the individual readers, enunciative markers trigger cognitive operations car-ried out by all readers in a more or less identical manner (even though they may lead to different results). The markers of enunciation include the markers of deixis, polyphony and of preconstructs, that is the elements of the formal apparatus of the enunciation (I, here, now…), logical- argumentative operators (not, but, if…), but also certain syntactic transformations (nominalizations, -isms), as well as typographic traces (such as inverted commas and italics). With enunciative markers, the constraints on interpretation are high. These markers are generally


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


not an object of reflexive comprehension because the reader cannot recognize them without effectuating their inbuilt rules, as in the case of I: Look for the person with whom the locutor is associated!; of not: Assign what is being said to the allocutor!; or of preconstructed nominalizations: Consider the utterance as a universal truth managing without a specific source of enunciation!

While concept-words refer to the specific knowledge of the producers and co-producers in the field, enunciative markers presuppose more basic discursive competencies, that is the command over linguistic rules in a narrower sense, which all participants of discourse equally share. ‘Informed’ readers associate a specific knowledge with concept-words, which is put into relation and perspective with the help of enunciative markers. ‘Uninformed’ readers, by contrast, only recognize the markers without having developed a differentiated knowledge of the contents and contexts of intellectual discourse. Enunciative discourse analysis simu-lates the ‘uninformed’ reader who knows how the contents of the discourse are orchestrated through the formal markers of enunciation, but can only shrug his shoulders when confronted with the question of ‘what’ the discourse is about. In so doing, discourse analysis demarcates itself from the history of ideas, generally not interested in the formal ‘how’ but in the ‘what’ of the contents of discourse.

Nouns and verbs tend to be concept-words; the signs reflecting the enunciation, by contrast, are to be counted among the markers of enun-ciation. The latter are all those signs by means of which the enunciation leaves its imprints in the symbolic material, for example Benveniste’s formal apparatus of enunciation or the logical-argumentative operators of Ducrot and ScaPoLine. The enunciation is also reflected in the case of nominalizations (‘now’ as an adverb—‘the now’ as a noun), through which their source of enunciation is concealed. There are many ‘mixed’ cases (especially adverbs, adjectives and verbs) in which certain (‘objec-tive’) conceptual components are colored subjectively. Thus, it is neces-sary to thoroughly examine the symbolic material in order to answer the question of how the enunciation is formally reflected.

As long as the concept-words and the markers of enunciation are regarded as signs with inherent meanings, abstracted from their con-texts, we are not yet doing discourse analysis. As isolated forms, these words are nothing more than ‘dead’ semiotic material. They begin to take a discursive life of their own only when concept-words and enunciative markers come together in the symbolic material to form certain syntactical-grammatical entities, that is sentences. Thus, the contents of discourse (‘what’) evoked by concept-words are bound to


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


exist in a certain way (‘how’), organized by the markers of enunciation. Discourse analysis does not deal with the question of how sentences are constructed grammatically, a task which can be left to grammar-ians, semioticians and traditional linguists. The objective of discourse analysis is to understand how these sentences are informed by directives of the enunciation and how they exist as specific utterances in discourse.

Before the symbolic material can be analyzed, we need to formally identify the sentences of the corpus. In a second step, these sentences can be analyzed as specifically produced utterances of a discourse. Only now does one begin to deal with the discursive processes triggered by these utterances, which make certain specific contexts of enunciation deictically visible, allow different sources of discourse to have a voice, and may presuppose and preconstruct certain stocks of knowledge. In a third step, one has to enquire how the interpretive processes triggered by utterances configure the social relations between the individuals entering into the discourse. In following this methodological proce-dure, we will account both for the (linguistic) world of symbolic forms as well as the (social) world of the individuals, producers and groups. If the abstract signs, the grammatical sentences, the repeatable texts are, as it were, located ‘beneath’ the discursive, then the non-repeatable contexts, the interactive situations and the social relations between the individuals are located ‘above.’ The question of discourse analysis is: how are texts and contexts linked through the enunciation?

In order to enquire into how texts instruct readers about the contexts in which they are uttered, the analyses will treat in turn three problem areas:

• The problem of indexicality or discursive deixis: how do utterances point to their contexts in the enunciation? What is marked as a relevant context?

• The problem of polyphony and dialogism: how do utterances orches-trate their voices and speakers? Who is responsible for what is said? Who’s speaking?

• The problem of the preconstruct: what are the traces left by what was said before and elsewhere? What are the signals for self-evident knowledge?

This analytical procedure directs its focus on the heterogeneity of a complex, convoluted and knotted discursive object. Instead of an ideal immaculateness which draws its meaning from itself, the following


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


discourse analyses examine texts which are connected with an amalgam of the most different (cognitive, situative, institutional…) contexts. Instead of a transparent medium through which a timeless truth shim-mers, they deal with texts which do not easily impart the illusion of lived meaning even among readers with a great discursive competence, who have differentiated knowledge of the contexts of production and approach reading with great patience. We therefore must not under-stand the discourse as a closed totality in which all elements mutually define one another. Rather, if we want to account for how these utter-ances are uttered in specific contexts, we should reflect the blanks and gaps, what is not said and what cannot be ultimately decided. Instead of retrieving a core of meanings, we will ask how utterances can be cited, made use of and reinscribed into different contexts. Instead of a lonely monologue, the analyses will register a chorus of voices which are entangled and superimposed upon each other, which are brusquely rejected or respectfully accepted, which face each other in antagonistic opposition or in harmonious accord. If the reader wants to know how the other individuals—those who have written the text and those who will read them—set themselves apart from each other, refer to each other and position themselves in the intellectual field, the reader must some-how orient her- or himself in this maze of allusions, in these murmuring voices, in this parcours strewn with pitfalls. How much easier it would be if the actors could just assume social positions directly, without the detour through symbolic forms. The problem is, they just can’t.

Polyphony and scenography: the activity of the reader

From a discourse analytical point of view, texts, even the most conceptual ones, are not closed-meaning containers. Discourse emerges whenever a text is linked with its contexts in acts of reading and writing (in their ‘enunciation’). In this sense, the enquiry will be concerned with how a theoretical text demands from the reader an answer to the following question: ‘By whom, for whom, where, when, under which circumstances and with which purposes and so on am I uttered?’ This is a question which the utterance cannot answer itself. It needs a cognitive agent to solve its interpretive problems: the reader. Rather than resuscitating the meaning originally intended, the reader has to determine the meaning of the text’s utterances in each act of reading anew, namely with the help of the formal markers, traces and instructions scattered throughout the text. When the reader recognizes the symbolic forms with which the text operates, he or she will begin to


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


look for the enunciative context, without which the text would remain devoid of meaning.

As an opaque symbolic materiality, the text is posed between the individual and the world. At the same time, the individual’s knowledge about the world is always symbolically mediated. His or her knowl-edge needs to be inscribed into the symbolic material. In order to gain access to the social world, to the producers of the text and to other individuals in the field, the individual thus has to enter the discourse, use language and produce texts. By having the individual search for an answer to its question, the text enables the reader to construct knowl-edge about the social world in which other individuals act and position themselves in relation to each other. Therefore, the task of discourse analysis is to analyze ‘the enunciative dispositif which connects a textual structure and a social location’ (Maingueneau 1996: 6).4

Texts are not repositories for pure ideas, content or messages to be read off directly from the symbolic material. They need a reader who completes them by adding missing contexts and thus associating the many anonymous sources and voices of discourse with definite individuals occupying social positions. Thus, to understand the text’s social relevance, the reader has to look for the guarantors, references and authorities that the locutor quotes, implicitly or explicitly, in order to legitimate the content for which he or she does not claim responsibil-ity. The reader also has to enquire into those individuals and producers whose discourse is rejected by the locutor. The analyses underscore the many ways in which the text confronts the reader with the question: who is speaking? The social efficacy of texts seems to lie precisely in the fact that they allow the reader a certain degree of freedom in terms of determining the sources of enunciation and associating them with actors in the social world.

Meaning is not produced in a lonely monologue or soliloquy; a variety of points of view, perspectives and all kinds of fictive and subpersonal speakers are mobilized in the production of meaning—this is the fun-damental insight that is commonly associated with Mikhail Bakhtin. While Bakhtin used a host of terms to designate this phenomenon, there has been controversy over how to define the notions of polyphony and dialogism. As polyphony is only used in his work on Dostoevsky’s novel, some have suggested that it be limited to a particular literary genre (cf. intertextuality in Kristeva). Nowakowska proposes dialogism as a more general term to cover the relation with the Other in all linguistic and social praxis (2005: 25; Moirand 2010). However, as dialogism usually implies a pattern of question and response between two partners (Bres 2005: 58),


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


I prefer polyphony as a more general term for the characteristic presence of many voices in any discourse (cf. Perrin 2006: 7ff.). From this perspec-tive, the Ducrot/Nølke theory of linguistic polyphony can account for only a certain aspect of polyphonic discourse: the (hierarchical) relations between the voices within an utterance, notably those which are linguis-tically encoded and reflected as formal markers in the linguistic material (Nølke 2010), which has been criticized regarding the abstract hierarchies implied in language being understood as a static system (Nowakowska 2005: 28). With a reader model of polyphonic discourse, I place the theory of linguistic markers of polyphony in the context of a pragmatic model of meaning produced by the active reader cooperating with text in context. In this view, we have recourse to markers such as I, not or but to orchestrate the many voices which we need to deal with in discourse. And this is a task which calls for our practical interpretive competence as members of a socio-historical community.

If we consider discourse as a decentered space of voices, texts draw from a plurality of sources which are not always easy to identify. Theoretical texts are no exception—even those that seem to correspond perfectly to the norms of academic writing (that is texts with citations, bibliographies, see Fløttum 2005). However, as texts are always contex-tually underspecified, they demand an active reader who will search for a context in which they ‘make sense.’ In order to manage the many anonymous voices, open gaps and undetermined sources that need to be processed in the act of reading, the reader tends to have recourse to inter-pretive hypotheses and to mobilize certain schemes of interpretation which presuppose a scenography. While Maingueneau (1993) links the problem of the scenography to the genre of a discourse, I want to point out the role of the reader. It is the reader who constructs the scenography by associating discursive speakers with certain individuals holding this or that position in discourse.

In order to come to terms with the interpretive problems of a text, the reader has recourse to interpretive schemes. ‘Structuralism’ is such an interpretive scheme that allows the reader to frame theoretical texts in a certain way. Thus, against a ‘structuralist’ background, the producers of the following textual fragments will be seen to relate to each other and to form a movement at a certain time and place. The label ‘structural-ism’ presupposes a group of producers, a movement or a school (that is ‘structuralists’ such as Lacan, Foucault, Althusser, Derrida and Sollers) that can be distinguished from other groups with differing theoretical orientations (for example ‘humanists’). Such a frame can help the reader to solve the interpretive tasks posed by the text. By framing the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


text as a product of the structuralist movement or conjuncture, the reader can reduce the multitude of discursive voices they encounter when reading to a limited set of relevant subject positions (for example a conflict between ‘structuralists’ and ‘humanists’). Texts from different producers can then be seen as dealing with a common topic, theme or problem (that is ‘the structural model in the human sciences’) and thus as contributing to a discourse.

Let us assume that readers of the following textual excerpts are famil-iar with the ‘structuralism’ hypothesis and will therefore frame what they read according to a certain antagonism between ‘structuralists’ and ‘non-structuralists.’ In that case, they will try to situate the producers, figures and voices of intellectual discourse on this or that side of the dividing line separating ‘structuralists’ from ‘non-structuralists.’ As long as the reader manages to recognize the many sources appropriately, the label ‘structuralism’ will have its symbolic efficiency. Yet, in their search for who says what, with and against whom and so on, readers will not restrict themselves to what they find in the text. Every piece of informa-tion may turn out to be relevant as long as it helps them to determine the positions of the producers of the discourse. In order to map the relations of proximity and distance between the various producers in the field, they will draw from their own non-conceptual knowledge about the dis-course, for instance from what they know about the careers, institutional affiliations, networks or political stances of the protagonists.

Grouping the voices of intellectual discourse together and giving them a name is an essential operation for the reader who is trying to understand what a text is supposed to mean. However, in a given context of reception, not just any label will do. For the reader familiar with the history and insti-tutional organization of the French field, for instance, it is rather difficult to group together permanently, under the same heading (such as ‘post-structuralism’), philosophers (Derrida, Althusser) and ex- philosophers (Foucault), academics (Foucault, Derrida, Althusser) and autodidacts (Lacan, Sollers), structuralists (all of them) with non- structuralist anti-humanists (for example Deleuze, Lyotard … see Angermuller 2015). That is why the reader is likely to prefer labels other than ‘poststructuralism,’ such as the ‘moderns’ (Aron 1984), the ‘Freudian-Marxist-Structuralist’ movement (Boudon 1980) or ‘the thinking of the 1960s and 1970s’ (Ferry and Renaut 1988), which lay more emphasis on the spirit of a period than on the unity of a movement.

The different paradigmatic labels which are attached to these texts on their journeys through different contexts of reception remind us that theoretical texts, too, are read in different contexts. And theoretical


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


texts also have to be complemented by the contexts in which they are enunciated. Theoretical texts, then, need a reader to solve the interpre-tive problems they pose to him or her. Hence, the reader’s task is not to realize a given structure or transcendental grammar of what can be said or thought. They must understand the text ‘here’ and ‘now’; only they can connect symbolic material, which can be repeated, with the contexts of enunciation, which cannot be repeated. And indeed, the whole of their interpretive talent is needed to understand a text. What meaning is eventually constructed depends on the individual reader, for every reader has to walk through the labyrinth of references on his or her own and follow the instructions scattered in the text. And he or she has to develop knowledge of the enunciative context that the text activates. We all try to fill in the text’s gaps by calling upon our contextual knowledge. We all are more or less well-versed experts in the interpretive problems that the text asks us to solve. We are all reader-philosophers who try to understand a text with the ‘philosophy’ we have of the text to hand.

Thus, by following the instructions in the text, the reader seeks to discover its meaning, which depends on the discursive competence he or she has acquired. Readers with low competence will generally limit their attention to the immediate contexts of enunciation. They will scan those contexts, especially the co-text and paratext, for information miss-ing in the text. Readers with a more developed discursive competence can, by contrast, rely on cognitive contexts. They have a memory that provides them with more or less confirmed hypotheses concerning the context of enunciation. These hypotheses are sustained as long as they do not encounter resistance or cause disturbance in their cognitive environment. Therefore, even one and the same reader can approach a text in different ways. A text can be ‘worked through’ or ‘skimmed over,’ meaning that texts can be dealt with more or less systematically. Therefore any meaning is, not least, a function of the time and energy available to the reader, since sooner or later there must come an end to any process of interpretation.

Of course a text can even be read without differentiated prior knowl-edge about the context of enunciation. This is what the discourse analyses above do. They focus on the markers of enunciation in order to simulate the ‘uninformed’ reader. Equipped with a basic grammatical competence, he or she thoroughly tries to solve the text’s interpretive problems while trying systematically to process the contextualizing instructions of the formal markers of enunciation. For such a reader, however—in contrast to the reactions of the ‘informed’ reader—the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


cited names and concepts, such as ‘name-of-the-father,’ ‘self-presence’ or ‘intertextuality,’ will provoke little more than an embarrassed shrug of the shoulders.

The following analyses deal with a discourse revolving around a recurrent antagonism between what can be called a humanist and an anti-humanist ‘scenography’ (cf. Maingueneau 1993). According to Maingueneau, texts come along, as it were, with a surrounding world which they both refer to and presuppose. Thus, the hegemonic force of the texts we will examine in the following chapters results from the ways in which they allow the reader to switch from a humanist to an anti-humanist scenography during the boom-period of the ‘modern’ disciplines known as the human sciences. We can perhaps describe the scenography of the traditional humanities as a world in which the role of the scholar is to preserve and transmit an exclusive stock of knowl-edge. The producer is perceived as an erudite scholar who preserves the heritage bequeathed by his predecessors. While he (for he is of course masculine) enters an axiological world of hierarchical values, styles and taste, his world is a world of strong subjectivities. He lives in a centered world in which every element is defined by its functional role vis-à-vis the humanist mission of education (shaping personality and subjective taste, transmitting cultural values, developing Western civilization…). In the humanist universe, enunciation refers to the transcendental origin from which meaning flows. Everything goes back to this original center where the personal, spatial and temporal coordinates of human-ist discourse come together. The authority of the humanist scholar derives from this center, which is also the center of the tradition and the institution. Reading and writing relate to this center, which allows the humanist producer to enter a discursive order where every element has its place. By helping intellectuals to define their position in academia, the humanist scenography ‘makes sense’ of those regions of the field where the producers respond to the needs of institutional pedagogy. Given the ideal of consensualist social relations and continuist histori-cal time, the humanist scenography seems particularly appropriate in a universe where the producers act as teachers and disciples, where disci-plinary boundaries and canonical authorities contribute to the cohesion of the institution.

While humanism draws from a meaning-endowing center (‘subject,’ ‘culture,’ ‘tradition,’ ‘institution’…), the anti-humanist world of struc-tural theorists, such as Lacan, Althusser, Foucault, Derrida, Deleuze, Barthes and so on, follows the modernist logic of the theory vanguard. Thus, the humanist world of ‘old school’ scholarship is challenged by the anti-humanists who perceive the humanist I as a privileged place


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


for the inscription of ideology and challenge the authority of the author. The discursive space of anti-humanists is one in eternal conflict, with the margins in constant revolt against the center. The anti-humanist temporality is a modernist one, in that it attempts to subvert tradition and reinstate a new order. The coordinates of the person, space and time are not anchored in any transcendental center; meaning is not guaran-teed by the subject of humanism.

By means of the instruments of enunciative analysis, we will analyze the selected theoretical texts as antagonism between two voices: of one speaker a1, who represents the cause of ‘man,’ the ‘subject,’ the ‘author,’ ‘humanism,’ ‘living presence’…, and one speaker l0, who rejects a1. In its formal guise, the theoretical figure of anti-humanism implies at least these two basic points of view:

pov1: [a1] (TRUE (p: ‘Man is alive’))


This is the polyphonic structure of anti-humanist discourse which confronts the reader with a degree of antagonism between locutor l0 and allocutor a1, who represents the cause of ‘man.’ By defining an intellectual ‘now’ T0, which distinguishes ‘new’ from ‘old’ tendencies, the speakers become historical vectors of a discourse in which both producers and products are exposed to ongoing processes of obsoles-cence. The text then gives the reader an interpretive instruction: Scan the social context for possible protagonists for the speakers l0 and a1 and determine which age group and generation they belong to! If the reader is confronted repeatedly with interdiscursive antagonism between l0 and a1 and an intellectual ‘now,’ T0, their knowledge about the field can solidify into a topos (‘the anti-humanism of the 1960s and 1970s’), which certain slogans (‘death of man/subject’) can easily reactivate. By scanning other texts and the more general intellectual sphere, the reader may then ask how other producers and groups of producers posi-tion themselves in the interdiscourse of anti-humanism. Perhaps he or she can imagine the implicit enemies of the ‘anti-humanist’ intellectu-als: say, previous avant-garde philosophers ranging from Bergson to Sartre who emblazoned ‘life’ and ‘humanism’ on their banners. Or he or she may think, on the other hand, of the canonical disciplines of the time led by academic philosophy and the humanités. And other produc-ers, feeling they are being addressed, can react—one might remember Sartre’s ‘answer’ to Foucault, Lacan and Althusser (1966), the argument between Barthes (1960) and Picard (1965) over the nouvelle critique,


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Dufrenne’s criticism of anti-humanism (1968), Crémant’s polemics against structuralism (1969) and Fougeyrollas’s denunciation of ‘obscu-rantism’ (1976). The same holds true for the international debate on poststructuralism. A famous reply was delivered by Habermas who, we might remember, felt interpellated by what he perceived to be the ‘young conservative theoretical movement’ in France (1993). It is these reactions, interpretations and speculations which link different texts together so as to produce a discursive formation with common sources of enunciation.

While it is up to the reader to construct more or less stable subject positions from the brouhaha of intellectual discourse, it is a charac-teristic feature of these producers that they and co-producers in the field are generally not cited explicitly. How then can the reader gain knowledge about how the producers relate to each other in the field? To this end, she or he can draw upon a range of strategies. The first strategy might consist of furnishing the speakers of utterances a1 and l0 with the proper names of the co-text representing certain theoreti-cal positions in the intellectual debate. A second strategy is masking the locutor (see the Foucault analysis) or concealment of responsibility (see Althusser). A third strategy consists of leaving the naming of the anonymous speakers of discourse to the reproducer (such as Sollers) or to the reader. It is then up to the reader to search the discursive space and associate the utterances’ speakers with appropriate individuals and social positions. What the text cannot furnish, at least not by itself, is the extent to which the relations between the producers which it makes visible are hierarchical and unequal. Who is dominant and who is dominated cannot become clear without recourse to the context in which the reader unfolds the text’s interpretive potential.

Against this background, the reader constantly attempts to contextu-alize a text against a background of hegemonic lines of conflict, such as between humanism and anti-humanism as suggested in previous readings. The reader will want to situate the many speakers and voices with which texts operate in intellectual discourse: Which side is X1 on? Where should we situate the idea X2? Does the fact of X3 reveal a certain proximity between X4 and X5? What does the feature of X6 mean in terms of X7? By looking for answers to these questions, the reader enters a discursive space in which they learn to navigate between ‘here’ and ‘there,’ ‘us’ and ‘them,’ ‘now’ and ‘then’.

Yet, in the course of the following analyses of small textual passages from Lacan, Althusser, Foucault, Derrida and Sollers, we will see that in theoretical discourse the various elements are not always neatly


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


distributed on this or that side of the antagonistic divide. While the reception of theories usually takes great pains to assign a clear and unambiguous place to every idea and position, the production of theories typically blurs the binary organization of discourse. Indeed, by looking closely at the selected passages, we will realize that they may contribute to undoing the hegemonic lines of conflict as effectively as they reinforce them. Accordingly, the reader may find that any hypoth-esis of a ‘structuralist’ or ‘anti-humanist’ hegemony is either weakened or confirmed. For this reason, the reader’s knowledge about what it is that the text is ‘really’ about is never fixed once and for all; and the only way to grasp more firmly what a text means, to dissipate doubts and uncertainties concerning the ‘correct’ interpretation, is to process more textual material, to ask what additional contextual knowledge is needed in order for intentions, meanings and messages to be understood. Thus, the reader constructs the meaning of the text in a never-ending process by working through textual material—utterance by utterance by utterance….


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

72

Five protagonists of theoretical discourse

The ‘corpus’ of the following analyses consists of five textual snippets deriving from leading representatives of the intellectual generation of structuralism: Jacques Lacan, Louis Althusser, Michel Foucault, Jacques Derrida and Philippe Sollers, who occupy points on a broad spectrum of differing positions in the intellectual discourse. They differ

— first of all, in terms of their political orientation (on the left: Althusser; periodically on the left: Foucault, Sollers; moderately leftist: Derrida; ‘non-leftist’: Lacan);

— second, in terms of the social networks associated with them (found-ers of schools: Lacan, Althusser; academic maverick: Foucault, Derrida; essayist in the public sphere: Sollers);

— third, in terms of their contrasting disciplinary fields (philosophy: Althusser, Derrida; sciences humaines: Foucault; transdisciplinary: Lacan);

— with respect to their different educational pathways (the elite philoso-phers from École normale: Foucault, Althusser, Derrida; autodidacts: Lacan, Sollers);

— and, finally, in terms of their various institutional anchorings (profes-sorship: Foucault; lower-level academic positions: Althusser, Derrida; free scholars: Lacan, Sollers).

The order of presentation of the analyzed texts roughly reflects the ‘hierarchy’ among these producers. The ‘senior members,’ Lacan and Althusser, as leaders of the two great theoretical schools of the era, head the analyses, which close with an examination of a textual passage from

4Analyzing Intellectual Discourse: Variations on the Critique of Humanism


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Analyzing Intellectual Discourse 73

Tel Quel penned by Sollers, Lacan’s junior by 35 years, who functions here as a recipient and commentator.

Lacan, Althusser and Foucault can be regarded as the intellectual troika of the time, who stand in for important theoretical trends of the time (psychoanalysis, Marxism, sciences humaines theories…) as well as for different institutional places where the theoretical conjuncture of structuralism took place (‘free-lance’ theorist, caïman at the ENS and professor at the Collège de France). While Lacan (born in 1901) was the conceptual wizard of his generation, Althusser (born in 1918) played the role of the faithful soldier of both the structuralist scientific project and the Communist Party. Finally, Foucault (born in 1926) was the bril-liant theoretical chameleon with intellectual all-round qualities.

Next to this intellectual trio, who challenged Sartre’s hegemony and emerged as the spokespeople of a new intellectual generation, several other intellectuals who also served as leading representatives of this generation have to be mentioned, such as Lévi-Strauss, who belongs to the same age group as Lacan and who coined the label of structuralism. Roland Barthes, whose work reflects the diversity of intellectual tenden-cies at the time, enjoyed a considerable intellectual presence. Derrida, Deleuze and Lyotard must not be omitted either. The three philosophers took up positions as avant-garde philosophers on the margins of their disciplines while crossing the border to neighboring fields, Derrida with regard to literature and Deleuze and Lyotard with regard to politics and psychoanalysis. In contrast to the ‘troika’ of intellectuals Lacan, Althusser and Foucault, who emerged as undisputed leaders in their respective areas by crossing and short-circuiting the various regions of the intellectual field (theory, politics and also to some degree aesthetics) directly or indirectly, the intellectual standing of these ‘second-rank’ intellectuals does not compare with those of the troika. If these aca-demic theoreticians devise their theoretical projects as broad intellec-tual visions, they address themselves to rather specialized audiences (for example, philosophy in the case of Derrida, Deleuze and Lyotard, and anthropology in the case of Lévi-Strauss) or they act as commentators and propagators of others’ ideas (see Barthes). The latter applies as well to the journal Tel Quel, directed by Sollers, which offers a platform to several of the above mentioned theoreticians and constantly created a stir with its unexpected theoretical and political turnarounds.

By analyzing the intellectual discourse of these ‘heroic’ figures of theoretical discourse, I do not want to imply that their texts reproduce the ‘great’ ideas of the time nor that they reflect the social conditions of production in the field. These texts are no representative samples of


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


a larger whole. Rather, their meaning derives from the intersections of interdiscourse where a multitude of perspectives, positions and topics come together. The question, therefore, is how these texts bring into relation the various elements and regions of a widely ramified discourse. Consequently, texts need to be considered as interdiscursive nodes where the rifts, antagonisms and contradictions of discourse crystallize and from which different regions of the intellectual discourse can be reached. Thus, it is the number of potential connections and conjunc-tions that establish the symbolic efficacy of texts in intellectual dis-course. The texts are not containers of an already constituted meaning, knowledge or space. Rather, they allow us to enter a discursive space yet to be constituted.

What justifies the selection of a ‘corpus’ that consists of not even three pages or fewer than 50 utterances? All the excerpts originated around the year 1966, at the apex of the structuralist conjuncture. They appeared in prominent places in the works of well-known theorists of the structuralist generation, which comprised the leaders of important theoretical schools in psychoanalysis (Lacan) and in Marxism (Althusser), the major figurehead of the sciences humaines movement (Foucault), an academic philosopher (Derrida), and a liter-ary critic (Sollers). These excerpts, which reflect the diverse problems and questions of this theoretical generation, will be analyzed in the following pages as fragments of a discourse which cannot be surveyed, at least not definitively, either by the producers (the authors) or by the co-producers (the readers in the field). Neither can it be surveyed by second-degree observers (discourse analysts, intellectual sociologists or other professional observers of this discourse) in pursuit of the formal traces left by discursive activity in the symbolic material. The question that the following intellectual discourse analyses will attempt to answer is, therefore: how do these discourse fragments help the producers and co-producers to enter into a discursive space ramified and differentiated in a multitude of ways?

The first text is taken from Seminar XI, called The Four Fundamental Principles of Psychoanalysis, which Jacques Lacan held in 1964 at the École Normale Supérieure. It is in his public seminars, held for the first time in 1953 at Hôpital Sainte-Anne, that Jacques Lacan (born in 1901) had started to develop his psychoanalytical theories in the wake of Sigmund Freud’s discovery of the unconscious. Initially, his seminars comprised a limited number of psychoanalysts and philosophers—‘a kind of research laboratory’ (Roudinesco 1993: 343[260]).1 In the course of the 1960s, however, the audience grew to include hundreds.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Like no other circle, Lacan’s seminar soon turned into a cross-section of Parisian intellectual life. Among the participants were Koyré, Kojève, Lévi-Strauss, Hyppolite, Ricœur, Certeau, Guattari, Irigaray, some of Althusser’s disciples (including Jacques-Alain Miller) as well some of the later so-called nouveaux philosophes. From 1964 until 1969, Lacan held his seminar at the École Normale Supérieure, where more and more listeners attended from the expanding universities, among them numerous students of psychology and medicine. At that time, the seminar began to resemble a secular mass. In 1963, Lacan was excluded from the International Psychoanalytical Association (IPA) because of his practice of variable length sessions, and founded his own school, the École Freudienne de Paris (EFP) in 1964. It grew from initially 134 to about 200 members by 1967 before reaching almost 1,000 in the course of the 1970s (Roudinesco 1993: 404ff.[317ff.]). It is in this situation that Lacan tried to consolidate the rich conceptual output of the former years in a seminar which has generally been acclaimed as one of the more accessible entries to his thought. Lacan transformed the seminar into a platform for the EFP, where he trained a large number of psycho-analysts in France in the following years. Initially, Lacan’s seminars were not meant for publication, even though some parts soon appeared in the Écrits from the publishing house Seuil (1966). In 1973, Seminar XI was the first to be published by Lacan’s son-in-law Jacques-Alain Miller, who transformed the seminar notes into a book published by Seuil (Lacan 1973). Up to today there have been numerous typescripts of this seminar as well as others, and Miller’s edition has evoked controversial reactions. As, for the time being, no critical edition of Seminar XI exists, I will draw on the circulating notes known as ‘version J.L’ which can be obtained from http://www.ecole-lacanienne.net.

The second text derives from Althusser’s 1967 foreword produced for the second edition of Pour Marx [For Marx], whose first edition had been published in 1965. While Lacan introduced Freud to a broader intellectual audience, Althusser opened Marxism—till then only scantly regarded in the academy—to psychoanalytical and semiotic enquiry. Between 1948 and 1980, when his career tragically ended, he occupied the position of a caïman at École Normale Supérieure (ENS), giving supple-mentary courses to students preparing for the agrégation (among them Foucault and Derrida, who became friends of Althusser’s). Althusser established a school of Marxist theory in this institutional surrounding, rather shielded from the political and social upheavals of the time. Thus, he published Lire le Capital [Reading Capital] (1965) in 1965 with his students Étienne Balibar, Roger Establet, Pierre Macherey and Jacques


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Rancière. The texts of the Althusser circle were controversially received within and beyond the Communist Party, of which Althusser was a card-carrying member until his death. Many observers have interpreted Althusser’s structuralist Marxism as an indirect political statement, especially concerning Stalinism and Maoism (Matonti 2005: 79). After the publication of the works from 1965, a number of ‘self-criticisms’ (autocritiques) were to follow. In the course of the 1970s, the intellec-tual enthusiasm which had characterized Pour Marx, receded under the impact of various political disillusionments. The later editions of Pour Marx were to feature additional fore- and afterwords, including a lengthy appraisal by Étienne Balibar in the appendix of the edition published by Maspero (1996).

The third text concludes Michel Foucault’s rich historical study Les Mots et les choses [The Order of Things] (1966: 398[421ff.]). After returning from abroad and obtaining his habilitation in 1961 (1961), Foucault, born in 1926, became a professor of philosophy in Clermont-Ferrand. With Les Mots et les choses, Foucault achieved a breakthrough and gained the attention of a broader intellectual public. Published by Gallimard and reaching six-digit sales figures, the book was received as a milestone for the structuralist debate as well as for the many new research tendencies developing in the sciences humaines. Foucault emerged as a politically engaged intellectual a little later, after his election to the Collège de France at the end of 1969, which crowned a brilliant normalien career. Given his wide-reaching and ever-shifting research interests, Foucault remained an intellectual maverick throughout his career. Soon after the publication of Les Mots et les choses, Foucault would emphasize his distance from structuralism.

The fourth text is taken from Jacques Derrida’s monograph La Voix et le phénomène [Speech and Phenomena] (1967c), which introduced Saussure’s differentialist model into philosophy. Along with Grammatologie [Grammatology] (1967a), the essay collections L’Écriture et la différence [Writing and Difference] (1967b) and Marges [Margins of Philosophy] (1972), this monograph on Husserl’s theory of the sign laid the founda-tion for the deconstructivist project of the normalien-philosopher. Born in Algeria in 1930 and, at a very young age, enduring the anti-Jewish pogroms during the war years, Derrida went on to work as an assistant at the Sorbonne between 1960 and 1964, then as a maître-assistant at the ENS, before he became directeur d’études in 1983 at the École des Hautes Études en Sciences Sociales (EHESS). A circle of avant-garde philosophers and writers (like Sarah Kofman, Jean-Joseph Goux, Hélène Cixous) gathered around this renowned connoisseur of contemporary German


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


philosophy, whose main works were published by the small publishing house Minuit. Derrida’s circle never reached the size of Lacan’s or Althusser’s following. Politically less conspicuous than Althusser, Derrida was situated on the margins of a philosophical discipline having fallen into crisis in the wake of the boom in the sciences humaines of the 1960s. As one of the participants of the 1966 congress at the John Hopkins University, Derrida has received considerable international attention from the late 1970s on. Against the background of his rising fortunes in the U.S., the relationship with Foucault cooled in the course of the 1970s. The relationship with Lacan was also characterized by critical dis-tance. Derrida served as a main philosophical reference for Tel Quel until he was replaced in this role by Lacan in the early 1970s.

Philippe Sollers is the author of our last textual excerpt, which appeared in the journal Tel Quel (Seuil) in 1967 and was published again in the collection Théorie d’ensemble [Set Theory] (Tel Quel 1968). As early as the 1950s, Sollers (born in 1936) became known to a broader public through his novels (Sollers 1957). At the beginning of the 1960s, Seuil entrusted him with the foundation and direction of the journal Tel Quel. Under the star of the Nouveau Roman this journal initially dis-cussed current tendencies in literature and aesthetics. Sollers had a firm grip on the editorial line of Tel Quel. In trial-like tribunals, ‘deviating’ staff members were excommunicated. From the middle of the 1960s onwards, the journal began to follow the many alternating theoreti-cal and political conjunctures, accompanied spectacular reversals and about-faces. After the apolitical aestheticism of its first years, the journal subscribed to deconstructivist text and écriture-theories (1965–1969) before turning to philosophies of désir inspired by Lacan. The short alli-ance with the Communist Party (1967–1972) was followed by a short period of enthusiasm for Mao’s China (see the special issue on China in 1972). The dogmatic ethos which elevated every new trend to the ultimate avant-garde was as characteristic of the journal’s style as was the way in which the young, largely unknown editorial staff members celebrated intellectually renowned contributors to the journal (such as Derrida, Foucault and Barthes, who led Théorie d’ensemble) as—in Kauppi’s phrase—‘totemic’ figures (1990: 109[140]). In 1977, a special issue on ‘America’ announced a liberal turn which marked the end of a period of remarkable intellectual effervescence, and in 1982 Sollers discontinued publication of the journal. With nothing more than a B.A. from a minor business college (ESSEC), Sollers has been working as a freelance author. Sollers married Julia Kristeva after her arrival in France and befriended Barthes. The selected text originated in the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


deconstructivist period, under the impact of the student revolt, when the journal moved closer to the French Communist Party.

In one way or another, all of these texts decline the topos of ‘Man.’ I will now attempt to demonstrate how they stage a polyphonic specta-cle between the antagonistic figures of the (structuralist) ‘avant-garde’ and the (humanist) ‘tradition.’

Jacques Lacan: the return to (the subject of) Freud

(1) Le symptôme c’est d’abord le mutisme dans le sujet supposé parlant. (2) S’il parle, il est guéri de son mutisme évidemment. (3) Mais cela ne nous dit pas du tout pourquoi il a commencé à parler, pourquoi il a guéri de son mut-isme. (4) Cela nous désigne seule-ment un trait différentiel qui est celui, comme il fallait s’y attendre, dans le cas de la fille muette, celui d’hystérique.(5) Or, ce trait différentiel est celui-ci que c’est dans ce mouvement même de parler que l’hystérique constitue son désir (6) de sorte qu’il n’est pas étonnant que ce soit par cette porte que Freud soit entré dans ce qui était, en réalité, les rapports du désir au langage, à l’intérieur duquel, dans ce champ, il a découvert les mécanismes de l’inconscient.(7) Que ce rapport du désir au langage comme tel ne lui soit pas resté voilé est justement là un trait de son génie (8) mais ce n’est pas encore dire qu’il n’ait été pleine-ment élucidé même et surtout pas par la notion massive de transfert.

(1) The symptom is first of all the silence in the supposed speaking subject. (2) If he/she speaks, he/she is cured of his/her silence evi-dently. (3) But this does not tell us anything about why he/she began to speak. (4) It merely designates for us a differential feature which, in the case of the silent girl, as was only to be expected, is that of the hysteric.(5) Now, the differential feature of the hysteric is precisely this—it is in the very movement of speaking that the hysteric con-stitutes her desire. (6) Thus it is hardly surprising that it should be through this door that Freud entered what was, in reality, the relations of desire to language, and discovered the mechanisms of the unconscious.(7) That this relation of desire to language as such did not remain concealed from him is a fea-ture of his genius, (8) but this is not to say that the relation was fully elucidated—far from it—by the massive notion of the transference.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


(9) Que pour guérir l’hystérique de tous ses symptômes, la meilleure façon soit de satisfaire à son désir d’hystérique, qui est, pour nous, à nos regards, elle, l’hystérique de poser son désir comme désir insat-isfait, laisse entièrement hors du champ, la question spécifique de ce pourquoi elle ne peut soutenir son désir que comme désir insat-isfait, (10) de sorte que l’hystérie, dirais-je, nous met sur la trace d’un certain péché originel de l’analyse. (11) Il faut bien qu’il y en ait un. (12) Le vrai n’est peut-être qu’une seule chose, c’est le désir de Freud lui-même, à savoir le fait que quelque chose, dans Freud, n’a jamais été analysé.(13) C’est exactement là que j’en étais au moment où, par une sin-gulière coïncidence, j’ai été mis en position de devoir me démettre de mon séminaire, (14) car ce que j’avais à dire sur les noms du Père ne visait à rien d’autre qu’à met-tre en question l’origine, à savoir par quel privilège le désir de Freud avait pu trouver, dans le champ de l’expérience qu’il désigne comme l’inconscient, la porte d’entrée.(15) Remonter à cette origine est tout à fait essentiel si nous voulons mettre l’analyse sur les pieds dont il ne manque pas un d’entre eux.

(9) The fact that, in order to cure the hysteric of all her symptoms, the best way is to satisfy her hys-teric’s desire—which is for her to posit her desire in relation to us as an unsatisfied desire—leaves entirely to one side the specific question of why she can sustain her desire only as an unsatisfied desire. (10) Thus hysteria places us, I would say, on the track of some kind of original sin in analysis. (11) There has to be one. (12) The truth is perhaps simply one thing, namely, the desire of Freud himself, the fact that something, in Freud, was never analyzed.(13) I had reached precisely this point when, by a strange coinci-dence, I was put into the position of having to give up my semi-nar. (14) Because what I had to say on the Names-of-the-Father had no other purpose, in fact, than to call into question the origin, to discover by what privi-lege Freud’s desire was able to find the entrance into the field of experience he designates as the unconscious.(15) It is absolutely essential that we should go back to this origin if we wish to put the analysis on its feet.

Left column source: Jacques Lacan (1973[1964]): Le Séminaire. Livre XI. Les quatre concepts fondamentaux de la psychanalyse. Paris: Seuil, S. 16; right column source: Jacques Lacan (1986[1964]): The Four Fundamental Concepts of Psychoanalysis, Penguin Books Ltd, Harmondsworth, Middlesex, England, pp. 11–12, translation adapted.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


This is the closing passage of Lacan’s lecture held on January 15, 1964, the first of his 11th seminar. It was to be published almost ten years later by his son-in-law, Jacques-Alain Miller, under the title of Les quatre concepts fondamentaux de la psychanalyse (1973). The following analysis is based on the ‘version J.L.,’ which was initially written in shorthand, then typewritten by Lacan’s secretary and possibly reviewed by Lacan himself. The manuscript exhibits all the trademarks of improvised speech, especially the linguistic bumps and oddities (9: qui est, pour nos, à nos regards, elle, l’hystérique …; la question spécifique de ce pourquoi …; 15: dont il ne manque pas un d’entre eux) and Lacan’s characteristic transi-tions with de sorte que [so] (6, 10). Still, it is already adapted to certain requirements of written speech, such as punctuation and division into paragraphs. By dividing the symbolic material into 15 utterances, which do not always follow the punctuation proposed by the drafter of the protocol, I have defined the smallest analytical units.

Like all texts of Lacan’s, this text also puts the interpretive competence of its readers to a hard test. A first reaction of the reader in order to come to terms with the interpretive problems provoked by this passage may consist in searching the co-text for additional hints. Thus, we may discover that the text immediately preceding this excerpt is about the role of speech in the psychoanalytical situation. However, it remains unclear whether Lacan, by talking about the ‘mute girl,’ refers to a specific case (of Freud’s or his own practice), or whether he poses a very general problem of analytical practice. Since no source is cited for this case, the second hypothesis appears more probable. The editorial variants are contradic-tory. While the manuscript ends after (15) in the ‘version J.L,’ in Miller’s edition Lacan goes on to make some organizational remarks about the seminar (15). It is possible that Miller draws on his own notes in the sub-sequent passage. Yet in both cases, the conceptual work of the seminar ends with (15).

Lacan’s high conceptual productivity is reflected in the numerous psychoanalytical terms: sujet [subject] (1), symptom [symptom] (1, 9), hystérique/hystérie [hysteric/hysteria] (4, 5, 9, 10), désir [desire] (5, 6, 7, 9, 12, 14), langage [language] (6, 7), inconscient [unconscious] (6), transfert [transference] (8), analyse/analyser [analysis/analyze] (12, 15), nom du Père [Name-of-the-Father] (14). Some of the terms originate from other theoreticians (especially from Sigmund Freud: unconscious, transference, hysteric…); others are Lacan’s creations (Name-of-the-Father). Yet none of these signs is marked as a citation clearly referring to a foreign source of enunciation (with the possible exception of transfert [transference] (8)). Lacan’s idiosyncratic conceptual style is associated, as it were, with


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


all those terms. Even those terms extracted from other contexts are conceptually enriched by Lacan (for example desire, initially originating from Freud’s Wunsch [wish] and playing a particular role in Lacanian psychoanalysis). Terms like nom du Père [Name-of-the-Father] are also characteristic for Lacan’s concepts, namely in their undecidable oscilla-tion between pairs of homophones (nom [name]/non [no]).

The boundary between literal-semiotic and metaphorical-conceptual meaning is often hard to draw. This fact is underlined by the follow-ing signs, which are conceptually charged by Lacan without becoming psychoanalytical terms in their own right: parler/parlant [speak/speaking] (1, 2, 3, 5), mutisme/muet [silence/mute] (1, 2, 3, 4), guérir [heal] (2, 3, 9), péché originel [original sin] (10, <11>). Additionally, more nouns can be identified: trait (différentiel) [(differential) feature] (4, 5, 7), mouvement [movement] (5), porte/porte d’entrée [door] (6, 14), rapports [relations] (6, 7), mécanisme [mechanism] (6), question [question] (9), trace [trace] (10), priv-ilège [privilege] (14), origine [origin] (14, 15), pieds [feet] (15). There are also verbs which usually do not carry specific conceptual contents and may be labeled as organization words, subordinated to pure concept words, as they help organize the conceptual contents without carrying a specific conceptual knowledge of their own.

The focus of the producers and co-producers of intellectual discourse commonly centers on the concept words, which activate their concep-tual knowledge. In contrast to the ‘intellectualist’ reading of theoretical texts, typical for the co-producers but also in the history of ideas as well as in intellectual sociology, discourse-analytical readings concentrate on the symbolic ‘residues’—on the formal markers of enunciation strewn among the conceptual terms. The co-producers usually pass over this symbolic residue since it is immediately recognized and therefore not subjected to reflexive efforts of interpretation. In focusing on these markers, the discourse-analytical reading of the text demarcates itself from intellectualist readings. By bracketing the conceptual contents evoked by the text, the discourse-analytical reading concentrates on the question of how the text reflects the circumstances of its enunciation.

In a first step, the discourse-analytical reading examines the text for those forms which deictically refer to the enunciation and its contexts, that is for Benveniste’s markers of enunciation which inscribe subjectivity into language. Thus, the pronouns of the first person pronoun, je [I] (10, 13, 14) and its derivatives me/mon [me/mine] (13, 14) attest to the person speaking in the context of enunciation. As co- and contexts can offer many clues to authorship, readers usually do not encounter problems relating this I to the person Jacques Lacan. Thus, by using


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


I, the text not only changes from the mode of the (objective) report (Benveniste’s histoire), typical for conceptual academic work, to the (subjective) discourse (discours), typical for the narration of personal experiences; it also makes visible the situation of the person designated as I, that is Lacan, who had to terminate the preceding seminar due to a unique coincidence [singulière coincidence] (13), which carries a subjective coloration as well.

This deictic reference to the speaking person is noteworthy for two reasons: first, this I instructs the reader—especially in the utterances (13, 14)—to envision the specific speech situation in which the semi-nar of the preceding year had ended. This leads the reader out of the conceptual reflections of the preceding utterances and evokes a specific context (‘my seminar’) in which certain individuals actually spoke and acted. In this way, the I of (13, 14) confronts the reader with a number of complex tasks which he or she may solve by having recourse to his or her knowledge or simply to imagination: to find out that the leader-ship of the École Normale Supérieure had put an inglorious end to Lacan’s seminar because of the cloud of smoke hanging in the seminar (official reason); to find out that the motley crew congregating around this illus-trious conjurer did not properly fit into the respectable and exclusive environment of an academic elite school (unofficial reason); to find out why Althusser, as a caïman of the ENS, some of whose students were attending the seminar, could not or did not want to prevent Lacan’s expulsion. Of course, the readers will ask these questions only insofar as they are familiar with the biographical, institutional and social char-acteristics of the person denoted by I.

In this passage, the I of (13) is not only important because it represents a window as it were through which the position of Lacan and his school in the intellectual field comes into view (of course only if the listeners or readers know how to open this window with their contextual knowl-edge). But it is also relevant since knowledge about the specific context of his seminar, which enters into the discourse through the pronoun I in (13), is marked as a knowledge presupposed by the conceptual knowledge of (14) (nom du Père). As will become clear further on, in terms of the discourse’s polyphonic structure the role of the argumenta-tive operator car [because] must be highlighted, too, for it emphasizes a causal relation between the two utterances. However, even in the absence of such an argumentative operator, both utterances would be coordinated by a specific argumentative relation. Thus, an argumenta-tive linkage has to be established between what is presupposed by the utterance (13) (‘expulsion from the ENS’) as well as by the utterance (14)


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


(Name-of-the-Father). Now, many readers may have difficulties following, in the absence of additional information, the argumentative transition from (13) to (14) (that is not only those readers appreciating Lacan as a productive thinker but also those in whose view Lacan is no more than an intellectual charlatan). Yet whatever the readers really under-stand, the heterogeneous character of a discourse must be underscored in which utterances drawing from different sources are tied together. Thus, the knowledge of conceptual terms or proper names, designated in absolute reference (see for example Freud in (14)), enmeshes with the deictically mobilized knowledge of I. It becomes apparent that, in intellectual discourse formations, it is not only conceptual entities that connect with each other, entry points are also defined thanks to which individuals are put into position as persons in discourse. Thus, while the social and institutional setting of the speaker in (13) and (14) becomes visible through what the utterance evokes via the deictic pronoun I, the speaker also positions him- or herself theoretically and conceptually vis-à-vis what is said in the utterance.

The first person also appears in little reflexive insertions which comment on the discursive process, such as in dirais-je [I would say] (10). However, this I is to be understood more as a reader’s guide than as a real person, helping the reader throughout the reading and signaling a certain distance of the speaker to what is said. The reflexive qualifications évidemment [obviously] (2), comme il fallait s’y attendre [as was only to be expected] (4), en réalité [in reality] (6) also underscore the heterogeneity of the material. What is being said becomes the object of repeated metadiscursive comments and qualifications. By means of these formulas, the speaker attempts to catch up with its discourse, to adjust it with reflexive remarks such as à savoir… [namely…] (12, 14). Propelled by these flops and backfires which constantly disrupt the discursive process, Lacan constantly corrects and patches what is said. In these reflexive loops, he tumbles over his discourse so to speak, in which, ultimately, he cannot shelter.

The pronoun of the first person presents itself also in the plural form: nous/nos [we/us] (3, 4, 9, 15). Other than the je [I] of (13, 14), which discontinues the conceptual work when the person of the enunciative context appears, these we/us seem to be embedded harmonically in the emerging conceptual system. Who are the speakers signified by we/us? It could be the locutor himself, a dissimulated I, a ‘we of modesty’ (which exists in French). Or else, this we could represent a universal audience including the locutor himself, as it sometimes is the case in pedagogi-cal texts. Against both hypotheses one can argue that for both wes the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


situation may not be academic enough, since Lacan is improvising orally. It is thus perfectly possible that we refers to the real people of the speech situation and is either made up of a deictic, subjective compo-nent (I) and a deictic, non-subjective component (you) or of a deictic, subjective component (I) and a ‘non-person’ (he/she/it) denoted in the co-text. However, this we cannot contain any ‘non-persons,’ for all non-persons appearing in the co-text are excluded from the group denoted by we. Who are those non-persons? On the one hand there is le sujet [the subject] (1), il [it] (2), la fille muette [the silent girl] (4), l’hystérique [the hysteric] (4, 5, 9), elle [she] (9). On the other hand, there is Freud (6, 12, 14), il [he] (6, 14), lui [him] (7), son [her] (9). Against the background of these two non-persons, that is she, the hysteric, and he, Freud, the we of this text seems to unify the subjective person Jacques Lacan with the non-subjective persons of the speech situation, the participants of the seminar. Yet how are both non-persons related to we? While she denotes an arbitrary figure of discourse, taking up the role of the subject to be analyzed in the analytical situation, he is the figure of discourse who founds the analytical situation and is therefore a guarantor of what is being said. This seems to be suggested by the configuration between the persons and non-persons, which is established by (6). Seen from (6), the we speaks about the hysteric in the name of Freud. However, maybe a certain distance to the Freudian discourse is already signaled in (6), namely through the insertion of en réalité [in reality], which announces a divergent position on the part of the locutor. The gap between what is attributed to ‘Freud’ and that for which the director of the utterances takes responsibility widens in (7, 8) before becoming the explicit topic of (11). The problems for the reader to understand what is meant there-fore originate less from what the text is saying than from the changing modes of attributing the conceptual contents to the figures of discourse. Thus, until (6) Freud serves as a guarantor for what is said about she/her. However, this hypothesis becomes unstable by (12), where ‘Freud’ is the-matized as an object of psychoanalysis and thus, following her, takes his place on the couch of ‘us’ psychoanalysts. In so doing, the non-person ‘Freud’ fulfills a rather peculiar double function. As guarantor, ‘Freud’ ini-tially legitimates what the ‘we’ is saying about the subject (‘she’). Then, however, ‘Freud’ is presented as a subject which ‘we’ psychoanalysts have to analyze. In (14, 15) this ambiguous position taken by ‘Freud’ in Lacanian discourse as both guarantor and subject of analysis is provided with specific conceptual terms of its own: désir de Freud [desiring Freud/Freud’s desire], inconscient [unconscious]. The ambivalent return to Freud is thus established as a central problem in Lacan’s doctrine.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Analyzing Freud in the name of Freud: this is the task which the text confronts its readers with on two levels at once, which overlap and dynamically displace each other in complex ways. Thus, the text sends its listeners and readers on a quest for knowledge which on the one hand evoked in an encyclopedic way by concept words and, on the other, indicated in a deictic way by enunciative markers. Presumably, it is this demanding task posed on different levels that gives the text its symbolic efficacy in the intellectual discourse and seems to surround the producer of the text, Lacan, with a special aura. For if the discovery of the unconscious can be traced back to the desire of the discoverer of the unconscious, who else but Lacan could demonstrate Freud’s desire as ‘unsatisfied’ to us (9)? By simply following the performative mode in which Lacan’s theory is enunciated, many participants of his seminars and even more numerous readers of his books came to believe Lacan to possess a secret knowledge. Or to put it differently: doesn’t the theory, namely the criticism of the unitary subject, work precisely because it is presented with the prophetic gesture of a knowing subject?

Yet the text passage not only makes certain figures of discourse vis-ible by deictic reference. It also operates with a number of presupposed speakers, which will be examined later on. It is the many speakers that the text has parade onto the discursive stage that accounts both for its efficacy in intellectual discourse and for its difficulties. So numerous and various are the speakers that the reader has problems identifying who precisely is speaking. Attributing what is said to the various sources of enunciation turns out to be a problem which is not solved without further ado by the reader, even if he or she has the time and energy to read (or listen to) Lacan carefully. The interpretive difficulties of the reader are aggravated, of course, by the fact that the ‘version J.L.’ has to do without Lacan’s characteristic intonation, with which he organ-ized his speech like a virtuoso and thus helped create his rhetorical charisma. Audio and video recordings of his seminar have only been systematically available since the beginning of the 1970s. However, no matter how he presented his discourse on January 15, 1964, it is this manuscript (or rather its different versions) which the large majority of participants of intellectual discourse draw upon when referring to Seminar XI. That the speakers who file past in Lacan’s discourse are not accompanied by the accents of his sonorous oral speech makes this text certainly more difficult to read. Yet this absence changes nothing in the polyphonic organization of the text, which was every bit as characteris-tic of the oral delivery (to the degree that all of the enunciative markers are correctly recorded).


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


How can the polyphonic configuration of this excerpt be described? As will be demonstrated, the text performs a polyphonic spectacle in which numerous figures of discourse make an appearance, in addition to the already mentioned persons and non-persons. Thus, a character-istic case of polyphony is attested by (7, 8), where mais [but] calls into question what is said in (7) against the background of (8) and where the negation signals a conflict between two of the speakers. According to ScaPoLine (Nølke et al. 2004), six points of view can be identified which are superimposed upon each other in (7, 8). In the formaliza-tion of the polyphonic configuration, the speakers representing specific propositional contents (p, q…) can be easily identified: on the one hand as images of L (l2, l3, l0), on the other as the images of A (a1, a4, a5). Thus, according to ScaPoLine/Ducrot, the polyphonic configuration of (7, 8) can be formally described as follows:

pov1(7): [a1] (TRUE (p: ‘This relationship remained concealed from him’)), with which pov1 signifies a first point of view (7), which is not accepted by L and is therefore attributed to an image of A (a1).

pov2(7): [l2] (NO (pov1(7))), where l2 is one of L’s images.

pov3(7): [l0] (VALUE+ (pov2(7))), where l0 is the image of L which finally represents the ‘director’ of the utterance and where VALUE+ (‘genius’!) displays a positive evaluation by L.

pov4(8): [a4] (TRUE (P of (pov3(7)) BUT (pov6(8))), where P displays a presupposition of pov3(7), which is attributed to a second image of A (a4).

pov5(8): [a5] (TRUE (q: ‘The relationship was fully elucidated…’))

pov6(8): [l0] (NO (pov5(8))), through which the director l0 of pov3(7) is retrospectively cast into doubt and turns into one of the images of L, that is pov3(7): [l0→ l3].

The technical character of this analysis demands some patience on the reader’s part. Let us, therefore, give an alternative way of describing the polyphonic discourse of these two utterances. The polyphonic configu-ration of these utterances can be tested by dividing the points of view into autonomous utterances of a ‘real’ dialogue. At least two dialogues can thus be generated from the six points of view mentioned above which are contained in the two utterances (7, 8).

First, a dialogue 1 between A1 and L2:


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


A1: ‘The relation of desire to language remains concealed from Freud.’

L2: ‘No, it did not remain concealed from him,’ whereuponL3 assents: ‘What L2 has just said, is an indication of Freud’s genius.’

And secondly, a dialogue 2 between A5 and L6:A5: ‘The relationship has been fully elucidated by the massive

term of transference,’ whereuponL6 counters: ‘No, in the light of dialogue 1, this cannot be said.’

This test declines the polyphonic configuration of both utterances by allo-cating their discursive speakers to ‘real’ individuals. Such a test reveals the different points of view superimposed upon each other in the utterances in a conflictual or a harmonious manner. Against this background, the con-tents evoked by the utterances cannot be traced back to a unitary instance of meaning production (a ‘speaking subject’); rather, the utterances configure the interrelations between the different figures of discourse. L keeps all of them at a distance except for one (l0). In order to position themselves in the field, producers and co-producers must resort to such a multiplicity of discursive speakers, with the help of which they (as images of Lx) involve the others (Ax) in a fictitious (or real) dialogue. Which real individuals are then identified with which speakers—this is a question the reader has to answer by drawing from his contextual knowledge. From a discourse- analytical perspective, we can point out that social relationships are already implied and configured within the utterances of discourse. These utterances bring various speakers of the discourse into relation even before the latter are associated with real individuals in the social world.

However, the allocation of a speaker to an individual usually presumes that the individual has a permanent institutional address, a name to which specific contents can be assigned once and for all. It is precisely the task of the reader to name the speakers of the utterance and thus to ‘fix’ the meaning of discourse. On the basis of the instructions scattered in the text, the reader searches for suitable candidates in the enunciative context. In the case at hand, for instance, she will likely come across a name—‘Freud’ (7, 9)—which serves as a guarantor for what is said. As the utterance’s ultimate instance of responsibility, the locutor who does not wish to assume full responsibility for what is said and may instead attribute xyz to the guarantor. The reader can then interpret (7) as an arena for a contest between different figures of discourse. They appear either against ‘Freud,’ or in the name of ‘Freud,’ they appear as ‘opponents’ or ‘followers’ of Freud. Accordingly, a1 would figure among


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


the opponents of L which reproach Freud for not having examined the relationship between desire and language. The speaker l2, by contrast, turns out to be a follower of Freud by rejecting this accusation; and l3 concurs with l2. Since the reader will recognize l3 as the ‘director’ of the utterance (that is as L’s ultimate voice l0), the reader should be able to tell what is the relationship of the theoretical guarantor, Freud, to the locutor-author, Lacan (who is the leader of the Seminar XI and who is mentioned as the author of the manuscript ‘J.L.’). However, mais (but) in (8) may call into question this hypothesis. The utterance (8) gives a voice to two figures which are rejected by the director-speaker, viz. to a4, whose view is presupposed by l2, and to a5, who explicitly carries some of the conceptual contents. With the rejection of a4 and a5 in (8) by L, some of what is said in (7) is corrected again. While in (7) the reader assumes the person of Lacan to be a follower of the followers of the guarantor ‘Freud,’ this assumption is called into question in light of (8), for in (8) the locutor Lacan turns out to be a critic of ‘Freud.’ Yet the speaker l0 of (8) does not coincide with a1 of (7), that is an opponent of Freud’s. The ‘director-speaker’ l0, who is finally responsible for the utter-ance, distances himself from this ‘sweeping’ critic of Freud. Thus, what has been said in pov3 is not assigned to A, but remains connected to one of the images of L, namely l3. By means of L’s images, Lacan dances around ‘Freud’ as it were—sometimes dearly embracing him, sometimes keeping him at a certain distance. In doing so, Lacan acts as a ‘construc-tive’ critic of Freud, radicalizing Freud’s psychoanalytical project by making the blind spot of Freud an object of analysis.

Therefore, the complex polyphonic spectacle enacted by (7, 8) has three main characters: the opponent of Freud, the follower of Freud and the constructive critic of Freud. Equipped with the author’s signature, Lacan is at first attributed to the side of the ‘followers,’ then to the side of the ‘(constructive) critics.’ Which discursive roles do the readers assign to the producers of the intellectual field not mentioned in the text (but present in their cognitive contexts)? This question requires that the readers search the text and associate its discursive figures, speakers and voices to ‘real’ producers of the field. Louis Althusser (cf. 1996[1965], 1993) would be a candidate who is commonly situated as an intellectual ‘ally’ of Lacan’s (that is he would be associated with the images of L). In contrast, Paul Ricœur, who wrote on Freud, too (1961), is likely to be located in the oppositional camp (that is with A). It is the specific polyphonic configuration of the speakers of the utterance that explains the symbolic efficacy of our textual excerpt. Thus, by orchestrating the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


reader’s knowledge about the producers of the intellectual field in a cer-tain way, the markers of polyphony but and not instruct him or her about who says what.

The utterances at hand allow the readers to attribute their knowledge to at least three sources of enunciation: opponents, followers, and followers of the followers, and it can be assumed that such a tripolar set of speak-ers gives rise to specific representations among the readers regarding the relations between the producers of the field. Reading along, the picture of a great battle will perhaps emerge, with two armies facing each other: the forces of the past on the one side—Freud’s ‘opponents,’ representing the traditional humanities (humanités) of the universities—and the ‘follow-ers’ on the other side, standing in for the forces of intellectual modernity reliant on interdisciplinary theoretical references such as ‘Freud.’

What scene might the reader envision while reading the text? Amid this anonymous mass of rank-and-file protagonists of the discursive battle, the clatter of swords and the smoke of guns, a third figure stands out. Bearing a sonorous name (‘Lacan’), this knightly figure stands on the edge of the battlefield; again and again he incites the ‘followers’ to defend the heritage of Freud, and even against Freud! As a ‘follower of the followers,’ Lacan distinguishes himself as a widely renowned war strategist in a battle whose soldiers abide (mostly) in eternal anonym-ity. Thus, the text enacts a polyphonic spectacle before the reader’s inner eye, enjoining a constant search for the ‘real’ individuals the text alludes to. In asking which personae dramatis correspond to the pro-ducers and co-producers of the field familiar to him or her, the reader attends a polyphonic spectacle which she can embellish before her inner eye by drawing from her contextual knowledge. While the person of Lacan, as a ‘follower of the followers,’ plays the role of a visionary strategist in the theoretical argument of the time, the reader presumably faces much greater difficulties in finding suitable protagonists for the roles of ‘opponents’ and ‘followers’ in the field. He or she cannot even be completely assured that further protagonists existed at all—for there is not one name of a single infantryman who could bear witness to the actions of battle.

Surely, the reading of the text is complicated by the fact that many of the utterances’ speakers are not equipped with a fixed address or name. However, the appearance of nameless infantrymen is neither specific of Lacanian discourse nor can it be considered a deficit. Indeed, it is safe to assume that discourses which address large audiences cannot but resort to these supernumeraries, who may bear various conceptual contents of


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


the discourse without the readers needing to know their names. If the competent reader can in many cases relate the contents of the discourse to the individuals (or groups) of the social field or space, it surely is not always necessary to make explicit the sources, especially in social milieus as concentrated as the intellectual circles of Paris. Yet it is not the namelessness of such infantrymen, but their constant desertion that easily turns the reading of Lacan’s texts into an interpretive odyssey, no matter what their readers may already know about the contexts in which they originated. As stray voices, these speakers traverse the dis-course. Lacan’s text marshals speakers about whom no one can be sure to which camp or which side they belong at a given point.

In order to account for this particular difficulty in the Lacanian dis-course, let us consider the utterances (2, 3), which at first sight appear to exhibit the same polyphonic configuration as (7, 8). As is also the case in (7, 8), it is the function of mais [but] to shed new light retroactively on what has already been said in (2). This can be formalized as follows:

pov1(2): [x1] (TRUE (p: ‘He speaks’)), where it is undetermined if the speaker is accepted or refused by L.

pov2(2): [l0] (TRUE (q: ‘He is cured,’ IF x1 = l1), where pov2 designates a second speaker who is finally accepted by the locutor as l0. The final judgment of l0, however, depends on x, that is l1 will follow x’s opinion.

pov3(3): [l3] (TRUE (pov2(2))), that is what is said before mais [but].

pov4(3): [a4] (TRUE (r: ‘This tells us about why…’))

pov5(3): [l5] (NO (pov4(3)))

pov6(3): [a6] (TRUE (P of (pov3(3)) BUT (pov4(3))), where P is a presup-position of pov3(3) which is attributed to another speaker a6.

pov7(3): [l0] (NO (pov6(3))), that is the locutor cannot but oppose what is presupposed by pov4(3) and has to retrospectively denounce responsibility for what is said in pov2(2), that is pov2(2): [l0→ x2?].

In this formal description, the seven points of view, complexly interlac-ing in (2, 3), are identified and distinguished. We can see that in (3) only pov4, pov5 and pov7 are explicitly attested. Pov3, by contrast, picks up what has been said before and pov6 refers to what is presupposed by pov3. With the formal dissection of the enunciative levels of (2, 3), the different figures of discourse can be identified which carry the different


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


points of view. These are first of all L’s images (thus l3, l5 and the speaker through which L finally decides to speak: l0), and, second, those speak-ers which appear in order to be immediately rejected by L: the images of A, namely a4 and a6.

The reading of the Lacanian text (as well as the formal depiction of its polyphonic structure) turns out, therefore, to be a challenge for the reader. Not only must she or he come to terms with a multitude of points of view, but must also identify their sources of enunciation. As (2, 3) do not give any clues regarding possible real-world candidates for the images of L and A, the reader is invited to develop his or her own hypotheses about the protagonists of this discourse and to search the co-text for solutions to this problem. For instance, just before our passage, Lacan mentions a certain Fenichel, who is blamed for carrying out psy-choanalysis ‘on the level of platitude’ [‘au niveau de la platitude’]. If a2 and a4 were attributed to a figure such as ‘Fenichel’, then the utterances (1, 2) could be traced back to such a speaker who, in L’s perspective, displays a ‘naïve’ understanding of psychoanalysis. It must be noted, however, that before and after (3) the name of Freud is mentioned, too. It therefore appears more likely that the psychoanalytical situation, as alluded to by a2 and a4, must be understood as the classical one, that is the one defined by Freud. Even if the polyphonic configuration of (2, 3) seems to correspond to the one of (7, 8), the question remains as to how the speaker of pov2 is accepted or rejected by the locutor L. Yet as long as this first speaker can be attributed neither to L nor to A, the speak-ers carrying the other points of view will remain ambiguous as to their role in the conflict between L and A. We can take as an example the speaker of the point of view pov4, who, with cela [this], takes up what had been said in (2) but also the speaker of pov5, directly dependent on pov4 (whereby it remains unclear whether cela [this] refers to pov1 or rather to pov2). The situation is complicated by the fact that the speaker of pov2 retrospectively loses his function as the director-speaker due to the attribution of l0 to pov7. Therefore, x2, too, begins to swing back and forth between L and A. As it is unclear on which side the speaker of pov2 stands, it becomes difficult to position pov3 and pov6, emerging as they do with the word mais [but] in reaction to the speaker of pov2. Thus, if ultimately the enunciative sources of pov1, pov2, pov3, pov4, pov5 and pov6 cannot be determined once and for all, where does the speaker of pov7 finally stand, that is the speaker who claims to act as the director l0 of the utterance but whose speakers do as they want?

It is the unclear status of the first speaker x1 that undermines posi-tions of all the speakers in (2, 3). It is x1 who causes the interpretive


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


problems of this passage. As a double agent, x1 is now in the service of L, now in the service of A. Since all other speakers of (2, 3) are directly or indirectly dependent on this speaker, they too turn into wobbly privates who indecisively reel between the fronts. Even a single double agent can therefore suffice to profoundly undo the battle formation. And precisely in Lacan’s texts there are numerous double agents who undo the clear-cut division into frontlines. What had begun as an orderly conflict between two armies, in which each soldier takes up his place in a hierarchical formation, changes into a guerrilla struggle. Now, it is as if everybody were running around aimlessly, wildly shooting, without knowing who is friend or foe.

It is the double agents that turn Lacan’s texts into an interpretive chal-lenge for its readers. Even those who carefully read page by page may not necessarily have the impression of having taken hold of much meaning: no thoroughly conducted arguments, no clear sentences, no concise definitions. There is nothing but the persistent rustling of words and letters, out of which Lacan’s conceptual formulas (sujet supposé parlant (1), nom du père (14)) protrude, without the reader being able to attribute substantive contents to either L or A. Is this why the reader may have the impression that every author that Lacan discusses prolongs his own conceptual work in some way or another? No matter whether what is said in (2) is attributed to the founder of psychoanalysis (‘Freud’) or to a platitude-doctor (‘Fenichel’), Lacan himself always seems to be speaking along with the sources he cites. Nowhere is Lacan’s own position easy to identify; but nowhere does he seem to be completely absent either. The hypothesis that psychoanalysis is about the speech and the silence of the subject, as suggested in (2), can thus appear as the opinion shared by more or less everybody: by Freud, by Lacan and even by Fenichel. If the reader does not consent to this idea, that is to the hypothesis of the primacy of the signifier, he or she has to exert considerable interpretive resistance. Not only does she have to refuse a definition whose author-ship she doesn’t know, she also may have to face opponents who oper-ate ‘undercover.’ It thus turns out that, although Lacan never explicitly takes full responsibility for what is said, there is no way to question the definition that is given of psychoanalysis (namely the linguistic practice of analyzed and analyzing subject). For in order to cure the ana-lyzed subject, the analyzing subject has to bring the former to say the unsayable. This is what is marked as non-negotiable knowledge. This is precisely the doctrine which cannot be criticized without ending the conversation between the teacher and his students and without being excluded from the psychoanalytical community.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


If the reader does not want to take this risk, she has, in one way or another, no other choice than to follow the formal instructions of the text about the allocation of voices. At first, she will try to situate the emerging double agents in one camp or the other. This will continue until, betrayed by them, she will have to start the interpretive process all over again. In order to come to terms with the ambiguities and undecidabilities of the text, the reader will then begin to devise proactive strategies of interpretation which she can develop and practice in one of the many Lacan reading circles in Paris. What is indeed necessary is an interpretive guerrilla who knows how to deal with the double agents, the stray speakers and wobbly privates of this discourse. When the reader, well versed in the interpretive struggle, begins to develop an ear for what is murmuring on the margins of the discourse, and begins to speak about what cannot be said without further ado, only then can she be sure of having successfully traversed Lacanian analysis.

Louis Althusser: Marxism as anti-humanism

(1) C’est dans cette conjoncture qu’ont été conçus et publiés les textes qu’on va lire. (2) Il faut les rapporter à cette conjoncture pour apprécier leur nature et leur fonction: (3) ce sont des essais philosophiques, ayant pour objet des recherches théoriques, et pour objectif d’intervenir dans la con-joncture théorico-idéologique existante, pour réagir contre des tendances dangereuses. (4) Très schématiquement, je dirais que ces textes théoriques contiennent une double « intervention », (5) ou, si l’on préfère, ils « interviennent » sur deux fronts, pour tracer, selon l’excellente expression de Lénine, une « ligne de démarcation » entre la théorie marxiste, d’une part, et des tendances idéologiques étrangères au marxisme, d’autre part.

(1) It was in this conjuncture that the texts you are about to read were conceived and published. (2) They must be related to this con-juncture to appreciate fully their nature and function: (3) they are philosophical essays, with theoreti-cal investigations as their objects, and as their aim an intervention in the present theoretical ideo-logical conjuncture in reaction to dangerous tendencies. (4) Very schematically, I should say that these theoretical texts contain a double ‘intervention,’ (5) or, if you prefer, they ‘intervene’ on two fronts, to trace, in Lenin’s excel-lent expression, a ‘line of demar-cation’ between Marxist theory on the one hand, and ideological tendencies foreign to Marxism on the other.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


(6) La première intervention a pour objet de « tracer une ligne de démar-cation » entre la théorie marxiste et les formes de subjectivisme philosophique (et politique) dans lesquelles elle a été compromise ou qui la menacent: avant tout l’empirisme et ses variantes, clas-siques ou modernes, pragmatisme, volontarisme, historicisme, etc. (7) Les moments essentiels de cette première intervention sont: recon-naissance de l’importance de la théo-rie marxiste pour la lutte de classe révolutionnaire, distinction des dif-férentes pratiques, mise en évidence de la spécificité de la « pratique théorique », première recherche sur la spécificité révolutionnaire de la théorie marxiste (distinction nette entre la dialectique idéaliste et la dialectique matérialiste), etc.(8) Cette première intervention se situe essentiellement sur le terrain de la confrontation entre Marx et Hegel. (9) La seconde intervention a pour objet de « tracer une ligne de démarcation » entre les fondements théoriques véritables de la science marxiste de l’histoire et de la phi-losophie marxiste, d’une part, et les notions idéalistes pré-marxistes sur lesquelles reposent les inter-prétations actuelles du marxisme comme « philosophie de l’homme » ou comme « humanisme », d’autre part. (10) Les moments essentiels de cette seconde intervention sont: mise en évidence d’une « coupure épistémologique » dans l’histoire de la pensée de Marx, différence fonda-mentale entre la « problématique »

(6) The object of the first intervention is to ‘draw a line of demarcation’ between Marxist theory and the forms of philo-sophical (and political) subjectiv-ism which have compromised it or threaten it: above all, empir-icism and its variants, classical and modern— pragmatism, volun-tarism, historicism, etc. (7) The essential moments of this first intervention are: recognition of the importance of Marxist theory in the revolutionary class struggle, a distinction of the different practices, a demonstration of the specificity of ‘theoretical prac-tice,’ a first investigation into the revolutionary specificity of Marxist theory (a total distinction between the idealist dialectic and the materialist dialectic), etc.(8) This first intervention is situated essentially in the terrain of the confrontation between Marx and Hegel.(9) The object of the second intervention is to ‘draw a line of demarcation’ between the true theoretical bases of the Marxist science of history and Marxist philosophy on the one hand, and, on the other, the pre-Marxist idealist notions on which depend contemporary interpretations of Marxism as a ‘ philosophy of man’ or a ‘Humanism.’ (10) The essential moments of this second interven-tion are: the demonstration of an ‘epistemological break’ in the history of Marx’s thought, a basic difference between the ideological


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Left column source: Louis Althusser (1996[1965]): Pour Marx. Paris: Maspero, p. 262; right column source: Louis Althusser (1969): For Marx, New York: Allen Lane, The Penguin Press, pp. 12, 13; translation adapted.

This excerpt is found at the beginning of the second part (‘II.’) of the preface from the October 1967 edition of the essay collection Pour Marx, first published in 1965. After the polyphonic anarchy of Lacan’s seminar, Althusser’s preface is characterized by the lucid organization of its contents. If Althusser’s text is more readable, this is perhaps not only due to the intellectual background of its author: Althusser was a trained academic teacher as opposed to the autodidactic impresario Lacan. Moreover, the preface is not a place where, as in a seminar, one can engage in free speculation. In the preface, the author comments on the following text, perhaps he or she prepares the reader for what to expect, but he or she does not yet delve into the actual conceptual work or venture into bold theoretical experiments. Probably this implicit pact between the reader and the author allows for the business-like tone of this text, whose contents seem to be imparted without the enthusiasm which is characteristic of many other texts of Althusser’s.

Like the inventory list of a department store or a warehouse, this text lists the available stock of contents that the sales manager (the locutor) has labeled with small stickers. These labels signal clearly to the customer which products are under guarantee and which are not (through subjectively colored adjectives, like dangereux [dangerous] (3), and étranger [foreign] (5)). All the articles are arranged on their shelves in an orderly fashion, so that the goods under guarantee (‘Marxism’) are displayed on one side of the shop and those without a guarantee (‘subjectivism,’ ‘historism’) on the other. It is up to the customer to decide which aisles he will push his cart down. On the side of the shop with guaranteed products, he can be sure what awaits him, and if something happens to go wrong with the acquired good he can count on the friendly help and advice of the manager. On the other side of

idéologique des œuvres de jeunesse et la « problématique » scientifique du Capital; premières investigations sur les spécificités de la découverte théorique de Marx, etc. (11) Cette seconde intervention se situe essentiellement sur le terrain de la confrontation entre les œuvres de jeunesse de Marx et Le Capital.

‘problematic’ of the Early Works and the scientific ‘problematic’ of Capital; first investigations into the specificity of Marx’s theoretical discovery, etc. (11) This second intervention is situated essentially in the terrain of the confrontation between Marx’s Early Works and Capital.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


the shop, he must rely on himself should the goods not live up to their expectations. The department store thus divides the assorted goods for two different classes of readership: one half of the products are for those who trust the judgment, taste and experience of the manager; the other half is for those who do not trust him. As there is no ‘Mixed Wares Department’ (that is articles with and without guarantee), and since the customer does not want to lose time milling about, he will tend to stay in the part of the shop he originally chose.

Just as the division of the goods for sale forces the customer to choose one side of the shop or the other, the text confronts the reader with the choice between those contents accepted by the locutor and those rejected by him. In this case, there are no gray areas where the locu-tor’s speakers cavort with those of the allocutor. Unlike in the previous example, these utterances have as a rule only one speaker lx at their disposal, who necessarily coincides with the responsible speaker 10. All terms, theories and contents can be clearly traced to the locutor, who makes it quite clear what he accepts and what he rejects. However, the presence of the locutor does not mean that he is in all cases the source of the discourse or that he reveals all his sources. On the contrary, sometimes the function of the locutor is restricted to commenting on preconstructed contents that have been enunciated before and elsewhere. Thus, the locutor relies to an extent on preconstructs in (7) and (10), where nominal constructs like reconnaissance de l’importance de x [recognition of the importance of x] (7), distinction des x [distinction of x] (7), or mise en évidence de x [demonstration of x] (7, 10), or premières investigations sur les spécificités de la découverte théorique de x [first inves-tigations into the specificity of the theoretical discovery of x] (10) mobilize contents presented as being so natural and obvious that their sources of enunciation can be kept anonymous. Rendering knowledge anonymous is a strategy that can be observed especially in highly institutionalized discourse. Anonymous knowledge is indicated by the nominalization of utterances whereby they are separated from their sources of enuncia-tion. Thus, recognition of the importance of x (7) can be seen as the prod-uct of an utterance (−2), enunciated by a (fictitious) speaker l−2 saying x is important, an utterance that, in the transition to another utterance (−1) I recognize the importance of x, enunciated by speaker l−1, loses as it were its speaker l−2. Through the transformation into the nominal construct, recognition of the importance of x, enunciated by l0, the locutor l−1 gets lost as well. Therefore, only the tip (l0) of an iceberg is visible, the greater part of which (that is l−2 and l−1) remains under water. The negative speakers represent anonymous voices which are mobilized by


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


l0. The ‘oblivion’ of these speakers presents no problem as long as the preconstructed knowledge cited by l0 does not meet with interpretive resistance on the part of the reader. Thus, by marking ‘x’—Marx and his theoretical achievements—as uncontroversial knowledge, the utterance defines an audience which includes readers who accept the discourse of l−2 and l−1 with no protest and who do not wish a controversial discus-sion about x. However, readers who are not in agreement with l−2 and l−1 must engage in interpretive resistance by launching a debate which l0 does not have in mind. For l0 the question is only about how the reader reacts (positively or negatively) to the recognition of the importance of x. What the reader thinks about the importance of x is not at stake (that was l−1’s question), and even less so whether or not x is really that important (for which the reader must go back to l−2). Should the recalcitrant reader not want to (or not be able to) expend the effort, there is only one choice left: simply closing the book and putting it aside.

As we shall see below, a similar analysis can be made of the other nominal constructs and ‘–isms.’ These preconstructs serve to exclude specific bits of knowledge from the controversy and thus to distinguish the targeted readership as an exclusive circle of initiates. Thus, the locutor never speaks alone—in the background he allows a group of speakers to murmur along, supporting the locutor’s position without necessarily revealing their own identity. Not only the presence of pre-constructs underscores the heterogeneity of this discursive fragment, which relies on different enunciative sources. The ostensible clarity and precision of the text notwithstanding, the presence of the Other is indi-cated by numerous other traces. Yet, in contrast to the Lacanian text, the Other is not so much conjured up by logical and argumentative operators (not, but). On one hand, the Other manifests himself in the dialogic configuration of the utterances (1, 2, 3); on the other, he is reflected by the typographical marks of the text, through the quotation marks and the italics.

Even though in (1, 2, 3), nothing is said that does not fall under the ultimate responsibility of the locutor, it is still clear that the locutor finds himself ‘in company.’ The situation is comparable to a telephone conversation that the locutor is having with an unknown other. Even if the reader, who may be eavesdropping from the next room, catches only the words of the locutor, he can guess the questions of the Other, at least partially. Accordingly, the available utterances are revealed as discursive fragments of a dialogue of which only a part is manifest, namely the locutor’s discourse, from which the reader must surmise the discourse of the Other. Thus, the utterance (1) can be understood


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


as the answer from the locutor to this question asked by the Other or allocutor: (1′) In which context did the texts of this essay collection originate? In the same way, (2) and (3) react to questions such as, for example, (2′) How can these texts be appreciated or valued? and (3′) Are these really political texts? It will not be difficult for the informed reader to identify the interlocutors sitting across from Althusser, interpellating him: his party comrades, who interrogate Althusser about what he meant to say in this text. By complementing the locutor’s answers with the ques-tions posed by the Other, the reader can gain an understanding of the conflict-ridden situation in which Althusser has landed by publishing this text. Whether or not Althusser was ever really interrogated by his comrades at some point cannot, of course, be definitively answered. It is perfectly possible, for example, that the answers that l0 gives in (1, 2, 3), are responses to questions (1′, 2′, 3′) which have never actually been posed. Even though the Other a0′ may have never existed physically, he still puts pressure on Althusser to justify himself. Yet it should be kept in mind that the questions (1′, 2′, 3′) deduced by the reader as posed by the speakers a0′ are not simply products of the reader’s interpretive fancy. To complete the dialogue from which the utterances (1, 2, 3) have been taken, the reader is subject to certain argumentative rules, that is discursive laws that define the position of any utterance as prior to or posterior to other utterances in a sequence. An utterance can be connected with many other utterances and may form most different sequences. However, there are rules determining how the utterances are linked with each other in the discourse.

In these utterances, which can be read as fragments of an imaginary dialogue, various discourses meet, enmesh and intersect. Discursive heterogeneity is also indicated by typographical marks of the text, such as italics and quotation marks, which signal the beginning of another discourse and reinforce or dilute the responsibility of the locutor for what is said. Italics such as in philosophiques [philosophical] (2), intervenir [intervene] (2) and la première/seconde intervention [the first/second interven-tion] (6, 9) reflect the locutor. They mark especially important content and optically underscore the construction of the argument. By making the respective word stand out from the others (such as in the case of empirisme [empiricism] (6) or théorie [theory] (7)), italics are a common device used to comment on the discourse and thus to steer the course of reading. Other such devices are adverbs, like très schématiquement [very schematically] (4), which qualify the entire following sentence (and not only an adjective or verb), or si l’on préfère [if you prefer] (4). With these reflexive commentaries the locutor adjusts his position in discourse and


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


intervenes in the reading process as a guiding agency. This role is attested in (4) by the use of I, which does not actually show any living individual of the context of enunciation. Quotation marks have a similar function, as can be seen in the case of « intervention »/« intervenir » [‘intervention/intervene’] (4, 5), « philosophie de l’homme » [‘philosophy of man’] (9), and « humanisme » [‘humanism’] (9). Obviously the locutor does not want to mark the cited conceptual contents as unreservedly his own, holding them as if with clothes-pegs at a safe distance. In the first case, he may draw from the ambiguity of the French intervention/intervenir that has different meanings according to the context, for example in the politi-cal field (‘military intervention’), in medicine (‘surgical operation’), and in the intellectual debate (‘statement’). By using quotation marks, Althusser signals that he is aware of the word’s polysemy, and, by play-ing with its various meanings, emphasizes his claim to mediate between the fields of politics and philosophy. In the case of œuvres de jeunesse [Early Works] (10, 11) and Capital (10, 11), by contrast, the italics signal the intellectual property of another author (that is Marx) whose work is under copyright. The locutor cannot use copyright-protected works under his own name without coming into conflict with the norms of academic institutions and the publishing industry.

Quotation marks make the heterogeneity of discourse visible. In the case of coupure épistémologique [epistemological break] (10), pratique théorique [theoretical practice] (7) and problématique [problematic] (10), they signal the transition from a citing to a cited discourse. While the locutor keeps the cited discourse at a certain distance, it is not uncom-mon for intellectual discourse of the time not to explicitly state the sources of the cited discourse. Evidently, an informed Parisian reader-ship is presupposed that is able to attribute the ‘epistemological break’ to Gaston Bachelard (whose intellectual prestige Althusser tries to partake of in this way). With ‘theoretical practice’ and ‘problematic,’ Althusser quotes himself, thereby alluding to the preface’s co-text and Reading Capital, which also appeared in 1965. The self-quotation emphasizes the property claims that Althusser makes in regard to his own published work. With respect to ligne de démarcation [line of demar-cation] (5, 6, 9), by contrast, the locutor reveals his source, even osten-tatiously so with the parenthetical phrase selon l’excellente expression de Lénine [in Lenin’s excellent expression] (5)—which explicitness may well lead us to presume that ‘Lenin’ (as opposed to ‘Bachelard,’ who is not mentioned) is not an uncontroversial reference in the rather academic contexts of the philosophical debate. As the politically informed reader and surely also Althusser himself knows, ‘Lenin’ is a reference which


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


introduces the official resolutions of Soviet party conventions, but it is hardly a reference in philosophical tracts from the non-socialist world. If Althusser nevertheless invokes ‘Lenin,’ there can be two explanations. Either Althusser is invoking Lenin as a philosopher, in which case Althusser is implicitly asking the (co-)producers of the philo-sophical field to recognize Lenin as a canonic author for the history of philosophy. Or Althusser is referring to Lenin as a political authority. In this case, he makes clear to his readers that he accepts the precedence of politics over theory, and thus the interpretive authority of the Communist Party in questions of philosophy. In either case, the citation refers the reader to an ongoing intertextual struggle in the intellectual field, in which various groups manifest their claims to dominance by invoking certain reference figures.

The difficult tightrope walk that Althusser is trying to achieve between the political and philosophical field becomes apparent in other ways as well. The reader has to access the political and historical knowledge evoked in the text differently from the theoretical and philosophical concepts that are, as we saw, arranged and displayed in an orderly fash-ion by the sales manager/locutor. It is only after (6) that the texts resem-ble a kind of department store in which each conceptual/theoretical item is orderly situated next to another, all in their places, all furnished with a label of origin. In the first utterances, by contrast, conceptual terms are less prominent than a host of deictic and anaphoric references, for example the demonstrative pronoun cette (conjoncture) [this (particular conjuncture)] (1, 2, 3), the first- person singular (4, 5) and plural (1), as well as the generic on [one/we] (1,5). With the demonstrative pronoun cette [this], repeated three times, and with the deictic adjective existant [(currently) existing], the text evokes a knowledge in the interpretation of which the reader has a great deal of freedom. Since the demonstrative pronoun refers the reader to the preceding co-text in a rather general way, the reader must indeed have a look at the whole first part (‘I.’) of the preface, where Althusser deals with the historical background of the book, the (non-)reception of Marx in France, the political situation of the Communist Party since Stalin as well as theoretical questions. Therefore, the context made visible by this remains rather diffuse, which invites the reader to scan not only the co-text of the preface, but also to search his cognitive context, his general knowledge about the historical context in which this text was produced. Evidently, this hardly covers everything that the reader finds paratactically linked in the available co-text and in contextual information. What this emphasizes is the necessity of accessing another type of knowledge—not the conceptual


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


knowledge of philosophical terms, but the political and historical knowledge that this anaphorically refers to. Thus, the pronoun this leads the reader out of the department store of theoretical problems and con-cepts (6ff.) into an open-air marketplace, where he must search for the political and historical contents without the help of an informed sales manager. After the department store of concepts and theories he finds himself at a flea market of historical and political circumstances, where he is hardly guided by anybody. Therefore, while letting his associations play, the reader may have recourse to the 1965 preface (which gives further information on the context), to the posthumous epilogue by Étienne Balibar from 1996, to treatises (such as this one) or simply to his own historical memory.

The deictic particles and ‘subjectivemes’ from (1, 2, 3, 4, 5) signal different spheres of knowledge existing elsewhere to which the reader can switch with the help of a shuttle like this. These forms join two separate regions of knowledge: on the one hand, the philosophical and theoretical problems that the text attempts to deal with; on the other, the elements of the political and historical context, which the reader must put together himself. Accordingly, the producer (Althusser, the individual who has signed Pour Marx) can be recognized in two guises: on one hand as a subject who is time and again evoked by deictic reference—this is the activist Althusser, who takes part in the activities of his communist cell; on the other hand, as a subject orchestrating the technical terms and concepts after (6)—this is the philosopher Althusser, who describes himself as a relentless critic of ‘humanism’ and as a representative of a theoretical avant-garde. In this excerpt, the phi-losopher reminds us that he is also an activist, and therefore emphasizes the fact that he belongs to two different universes: the political and the academic field. By appearing in two different guises—as a deictically indicated party supporter and as an administrator of theory offering a variegated assortment of concepts—the locutor brings two different regions of the intellectual field together. By short-circuiting these oth-erwise separate universes and performing an acrobatic balancing act on the borderlines between philosophy and politics, Althusser underscores his intellectual ambitions. With this strategy, the activist Althusser gets (potential) access to a realm that normally remains closed to pure phi-losophers: the numerous intellectual publication outlets of the French Communist Party: La Pensée, Cahiers de l’analyse, La Nouvelle critique, Les Lettres françaises, Clarté, Europe. Althusser can even count on the attention of large daily newspapers like L’Humanité. At the same time, Althusser can underscore his distance within the academic field to the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


discipline of philosophy, which had plunged into a crisis following the boom in the human sciences and the turmoil on university campuses. As a Marxist philosopher, he can draw upon a theoretical capital which is widely recognized in the social and cultural debate of his time. By combining various political and theoretical factors, Althusser contrib-utes (just as Sartre and Merleau-Ponty did before him) to an intellectual conjuncture in which a new generation of symbolic producers emerges.

Like Lacan’s text, this text contains a range of conceptual words that stand for pure theoretical problems, and it operates with enunciative markers that orchestrate the ways in which conceptual knowledge will exist in discourse. Unlike Lacan’s text, however, this text leaves few open questions concerning the attitude the locutor takes vis-à-vis the contents. With pedagogical precision the conceptual terms are inventoried and sorted. Numerous cues exist in the text to guide the reader: première intervention [the first intervention] (6, 7, 8), la seconde intervention [the second intervention] (9, 11), d’une part [on the one hand] (9), d’autre part [on the other hand] (10). In this way, the itemized terms are ordered in an argumentative relationship, whereby the text not only summarizes to some degree the preface from 1965 and the philosophi-cal contents of For Marx, but also fixes as it were the meaning of the discourse in short paraphrasing expressions, namely in (6, 7) through (8) and in (9, 10) through (11). By constantly producing paraphrases of itself, the text testifies to the attempt to reduce the unintended back-ground noises, errors and false starts of discourse and thus to keep at a minimum the plurality of possible interpretations. Permanently refor-mulated, adjusted, and repaired, discourse turns out to be a continuing process of reflexive self-correction, in which deviations and ambiguities are headed off, the discourse is brought ‘in line’ and potential traitors are made harmless. With the binary formula entre … et … [between … and …] (6, 7, 8, 9, 10, 11), the conceptual terms take their functional place in a discursive space organized by a double antagonism: first by the opposition between Marx vs. Hegel (8), and secondly in (11) the œuvres de jeunesse [Early Works] vs. Capital (10, 11). The theoretical matter referred to by a wide range of terms is now arranged around this double antagonistic line. The elements associated with marxiste [Marxist] emerge on the locutor’s side of the borderline: théorie [theory] (6, 7), science (9), philosophie [philosophy] (9), and on the other side of the line: subjectivisme, empirisme, pragmatisme, volontarisme, historicisme (6), humanisme (9), philosophie de l’homme [philosophy of man] (9). The locutor’s position on this side of the line is made clear by subjective verbs such as compromettre [compromise] (6) and menacer [threaten] (6).


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Moreover, the reader familiar with Althusser’s work will recognize the subjective marks of a wide range of adjectives as well, through which he signals that he accepts certain points (scientifique [scientific] (9, 10), théorie/théorique [theory/theoretical] (6, 7, 9, 10)), while rejecting other points (idéaliste [idealistic] (7, 9), idéologique [ideological] (10)). By means of these ‘subjectivemes,’ conceptual terms are located on this or that side of the antagonistic borderline, see also dialectique idéaliste … dialectique matéraliste [idealistic … materialistic dialectic] (7) and problé-matique idéologique … problématique scientifique [ideological … scientific problematic] (10). Thus, two antagonistic chains emerge organizing the conceptual contents xL + yL + zL … and xA + yA + zA …, which are attrib-uted respectively to the locutor L and to the allocutor A. All conceptual terms bear the subjectivity of the locutor who assigns the cited contents to a clear position, either on this or the other side of the antagonism.

In Althusser’s case, the conceptual contents of theoretical discourse are equipped with explicit labels of origin (‘L’ or ‘A’), by means of which the author is clearly positioned in the hegemonic conflict. Moreover, the voices which the discourse orchestrates in the enun-ciation refer to institutional collectivities rather than to individual producers. This discourse cites an array of collective movements, which is especially underscored by the many ‘-isms’: subjectivisme, empirisme, pragmatisme, volontarisme, historicisme (6), humanisme (9). Thus, the nouns with the suffix ‘-ism’ can be seen as resulting from the transformation of an utterance whose speaker sides with a specific propositional content p.

pov1: [l0](p), where p represents a demand for x (‘subject,’ ‘empirical research,’ ‘action,’ ‘will,’ ‘history,’ or ‘man’).

The speaker l0 of this (fictitious) utterance refers to a discursive figure who, by producing the utterance, institutionalizes the contents (p) as the doctrine of a group or movement. The reader will perhaps associate this figure with particular names from the history of philosophy, whether English (‘Locke,’ ‘Hume’ for ‘empiricism’), North American (‘Dewey,’ ‘James’ for ‘action’ and ‘will’), or German (‘Hegel’ for ‘history’). The association of l0 with specific names depends in large part on the ability of the reader to connect l0 with producers of the field known to him or with canonic figures. However, with the transformation of the utterance into a ‘p-ism,’ the necessity of naming l0 disappears. What remains from this transformation is the (preconstructed) fact that a school in fact has been founded (‘empiricism,’ ‘pragmatism’…) regardless of when, where


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


and by whom. Thus, in its transformation into a ‘p-ism,’ the utterance is separated from its context of enunciation. The p-ism can now appear as a conceptual term for a school founded somewhere, at some time, by someone, demanding x, in relation to which the locutor then takes a position. Thus, L does not necessarily reject the preconstructed knowledge designated by the suffix ‘-ism.’ The noun Marxism (and its derivatives Marxist …), for example, are always accepted by L. Unlike the case of the ‘-isms’ rejected by L, the label Marxism derives from the founder of the school (l0, that is ‘Marx’) and not from the demand p, thanks to which we should speak of an ‘l-ism.’ In the transformation of the utterance into an ‘l-ism,’ the utterance ‘loses’ the demand p. Therefore, the discourse processes only the fact that l0 founded a school where- and whenever, which demands whatever.

With the help of such ‘-isms,’ the locutor can bring into play a preconstructed knowledge originating elsewhere. In (6), we see that his task consists in placing these institutionalized facts of discourse in a relation of opposition (that is the l-ism Marxism vs. the p-isms like subjectivism …), whereby he rejects that which is signified by subjectivism and accepts that which is signified by Marxism. The separate speech posi-tions involved in (6) can be formally broken down in the following way:

pov1(6): [l1](x), where l1 is the founder of the theoretical school (l1 = ‘Marx’) and x is the (now lost) demand made by l1 (l-ism)

pov2(6): [y](p), where y is the (unknown) enunciative source of the collective doctrine and p = ‘we demand x’ (x = ’subject’) (p-ism)

pov3(6): [a4](TRUE pov2(6)), as the verb compromise presupposes that the locutor would reject what was said in the subject position.

pov4(6): [l0](TRUE pov1(6)).

In (6), only the point of view pov4 is ‘constructed’ by the locutor of the utterance, whereas in pov1(6) and pov2(6) the locutor has recourse to knowledge constructed elsewhere. While these preconstructs have lost their sources, the role of the locutor is limited to qualifying them as either ‘wrong’ (in the case of subjectivism, cf. pov2(6)) or ‘right’ (in the case of Marxism, cf. pov1(6)). In this way, as a referee who has different teams compete against each other, the locutor of the utterance declares one the champion in the end. Who exactly these teams are, however, is not clear to the referee. Sometimes he does not even know where they come from or what they stand for. However he can differentiate the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


teams easily by their jerseys: on one side is the team who wear the name of their trainer (‘Marx’), on the other side the opponents with their vari-ous colors (‘subjectivism,’ ‘empiricism’ and so on), who drop out of the game one after the other. Despite the alliance which it has to confront, Marx’s team wins in the end.

The referee does not seem to have much trouble deciding the winner and the losers. The balance of power is so clear-cut that he could assign each participant to the one camp or the other—if only, in the last round of the game, there didn’t appear someone who cheers first one team, then the other. This individual is not unambiguously committed to one side or the other, which means trouble for the team of ‘Marx’ just when they thought they were bound for glory. Some team members now find themselves strangely infected by the figure’s gleeful cheering for the other camp, and threaten to break away. This individual seriously distracts the team of ‘Marx,’ because he wears the familiar white beard and has the same name as that emblazoned on their jerseys: ‘Marx.’ It is none other than this ‘Marx’ who seems to be depriving the ‘Marxists’ of their victory at the very last moment. And this is happening pre-cisely at (9)—or to be more exact—from the very moment in which Marx’s Early Works are connected with an ‘ideological’ (that is not with a ‘scientific’) problematic (10). Looking back from the utterances (10, 11), the reader realizes that the interprétations actuelles du marxisme … [contemporary interpretations of Marxism] can already be traced back to this ‘false’ Marx. But since the referee declares the author of the Early Works a loser, ‘Marx’ cannot be a member of the very team or school he had founded. As is well known, Marx is not necessarily a Marxist. If the person to whom his pupils look up doesn’t actually belong to the very school named after him, don’t the ‘Marxists’ run the risk of follow-ing a person who in fact hasn’t the slightest desire to be their teacher? Thus, the question that this text poses for its readers is: how can those who enter into the intellectual debate in the name of Marx deal with the problem that Marx, of all people, turns out to be a border-crosser, a wobbly candidate, a double agent to whom the discursive positions on this and the other side of our antagonistic divide can both refer?

With the theme of an epistemological break between the ‘early’ and the ‘late’ Marx, Althusser engages not only in Marxological exegesis, for by venturing this hypothesis Althusser touches even more on the power relations in a world that constantly confronts Communist intellectuals with the question: ‘Are you in or are you out?’ This is a particularly explosive question for Althusser, acting as he does in two worlds: as a Marxist philosopher he produces in an academic field that demands


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


autonomy; as an active member of the French Communist Party he is part of an organization whose leadership claims the ultimate authority in theoretical matters. Particularly Communist intellectuals like Althusser, who are active within the party but draw on experience and expertise acquired elsewhere, are confronted with contradictory expectations. On one hand these intellectuals are greeted as welcome supporters of the Communist cause, since their prestige raises the esteem of the party; on the other hand, it is difficult for them to speak in the name of that spe-cific expertise acquired in another field without appearing to criticize (perhaps ‘between the lines’) the Communist Party leadership, which may react with expulsion. For these party intellectuals, the borderline between ‘voice’ and ‘exit’ is particularly tenuous.

With the thesis of an ‘epistemological break’ in the thought of Marx, Althusser balances precariously on this borderline. In doing so he inter-rupts, gathers together and multiplies the attention of his peers in the intellectual field, who would otherwise produce in separate regions. The rejection of the ‘humanist’ Marx may be understood as the attempt to straddle two separate fields with one project: in the academic field for the structuralist theorists of the human sciences who have inscribed on their banners the death of author/subject/man, and against the traditional humanities and the philosophers of the preceding generation (Sartre); in the political microcosm of the Communist Party against competitors like Roger Garaudy, who represented a Christian-influenced, Hegelian interpretation of Marx, and who, as a member of the Politburo and leader of the party’s Centre d’études et de recherche marxiste (CERM), had considerable theoretical authority, and whose humanism won the acceptance of the leadership at the 1961 party convention at Choisy-le-Roi, where the members of the Politburo gathered with the party’s philosophers (Kauppi 1990: 117[152]). The reader well versed in the history of ideas will perhaps also connect Althusser’s ‘epistemological break’ with other debates of the period, such as for example the personality cult in the Soviet Union and Mao’s China, which Althusser alludes to a few pages before in his preface (cf. Bowd 1999: 45ff.; Matonti 2005: 72ff.).

However, as a closer look at the example can show, the thesis of an ‘epistemological break’ can unfold its symbolic efficacy for intellectual discourse even if the reader is barely or not at all informed about Althusser’s possible friends and enemies in the field at that time. The text activates more than just the historical knowledge of the reader; it deals with the question of who speaks in whose name, whose authority is being called upon to make what claims under which premises. Thus,


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


the text confronts the reader with the embarrassing problem that the discourse figure himself, on whose authority the discourse of the ‘Marxists’ is based, is not clearly assigned to either camp. How can we, who call upon Marx, take a clear position in a conflict completely dominated by two antagonistic blocks, when it is ultimately unclear on which side our guarantor, Marx, stands? This question must of course destabilize the position of all those who speak in the name of Marx, including the referee himself (that is Althusser). By taking the side of the ‘Marxists,’ although displaying certain reservations in respect to their guarantor ‘Marx,’ the locutor effectively signals to the reader, ‘It’s not enough to simply act as the representative of a school (‘Marxism’). In order to partake of the authority of your school, you must have a clear attitude toward its founder (‘Marx’).’ But what type of reader is the text addressed to, and which authority is in doubt? That is the ‘prob-lematic’ of the text, calling for an answer. While the break between the early and late Marx is constantly revisited and formulated in For Marx as well as in other texts, the text never says whose authority is being called into question. It is as if the text repeatedly asked the question but avoided providing an answer under all circumstances. It is what is not said that the discourse revolves around.

So it is up to the reader to choose and nominate fitting candidates for the tournament staged by the text, and to think up a scenario to fill in the blanks of the text, for example who the competing teams are and how the final result is affected by the circumstance that Marx roots for the wrong team in the end. In a first scenario, the emergence of ‘Marx’ provokes a crisis of legitimacy for the winning team, which would cause the following twist of fate: after declaring the ‘Marx’ team the winner, the referee, who also recognizes the authority of ‘Marx,’ questions the legitimacy of the victory, as the support of ‘Marx’ seems doubtful. In this scenario the reader might read the text as the critique of a dissident who finds it debatable that the players of the victorious team (for exam-ple the leadership of the Communist Party) are truly fighting in the name of an indisputable authority (‘Marx’). In this scenario, the reader might see the person responsible for the discourse (‘Althusser’) as a critic who measures the party and its representatives against the authority in whose name they act.

However, this scenario stands or falls depending on the position that the referee takes in relation to ‘Marx.’ We can be sure that the ‘Marxist’ team is playing on behalf of ‘Marx.’ It is also a given that the referee will declare the ‘Marxist’ team the winner. However, it is not clear whether the referee comprehends ‘Marx’ as the ultimate source of


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


authority. The text offers no clear answer to this. If we imagine that the referee’s decision is not dependent on the authority ‘Marx,’ but rather on the effective strength of the team, then another scenario is possible. In this situation the ‘Marxist’ team would still climb the rostrum after the game. Still the emergence of ‘Marx’ for the victorious team would raise uncomfortable questions, but the reactions would be different. ‘Marx’ would now be a bewildered agitator, a volatile troublemaker, whose contradictory actions raise eyebrows everywhere, and for whom the referee has no sympathy. Instead of giving long thought to his deci-sion, the referee will try to call this mischief-maker to account and to quieten down the players on the victorious team, who want to stand up for their old trainer. The reader would then understand this text as a call to order by a party cadre who denounces all those who deviate from the party line as traitors, in the name of the legitimate victor (the Party)—even if that dissenter is named ‘Marx.’ In this case, the reader would recognize Althusser as a loyal and incorruptible soldier of the party, whereas a primal, tragic betrayal of the party can be traced back to the discourse figure ‘Marx,’ all of which emphasizes the importance of constant vigilance and control.

Which scenario is the right one for the reader to envision? Which sce-nario is truer to the intentions of the author? Unfortunately, we can least expect an answer from the producer himself, Althusser, who in this and all following texts seems to use all his talent as a writer in order not to give an answer. The decision of how the text should be understood is left up to the reader, who must glean an answer by reading ‘between the lines.’ The reader can thus opt for one of the scenarios at one time and for the other scenario at another, or even for both at the same time, as is typical for dis-course in total institutions in which the border between critical distance and dogmatic loyalty blurs and everyone is both traitor and watchdog. Althusser seems again and again to circumscribe the empty gaps in the text—the attitude of the locutor toward ‘Marx’—thus maintaining the bal-ancing act between the philosophers and the activists. To the co-producers the text can be presented both as a critique of an authoritarian style of leadership (the dissident scenario), or as a disclosure of an early betrayal of Marxism and a threatening division in the party (the Stalinist scenario). Althusser uses these unsutured gaps to cross two incompatible worlds. It is these gaps that establish the symbolic efficacy of his text in the intellectual field. It is these gaps that emphasize the impossible position Althusser occupies as a philosopher and as an activist, as an advocate of democrati-zation and as an opponent of factionalism within the party (cf. Althusser’s unpublished manuscripts in Valentin-Mc Lean 2005). This position is one


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


that underscores the singular aura of its owner and allows him to become a leading actor in the intellectual discourse of the post-war period. But it is also a position that places him under enormous internal pressure, as we can see from the numerous ‘self-critiques’ which will become bleaker and bleaker in the course of the 1970s and which the two prefaces of For Marx perhaps foreshadow. Althusser does not disperse this tension, as so many in his generation did, by exiting the party. As we know, the final act of the drama staged by Althusser was to provide an ‘exit’ undreamt of by even the most theatrical among us. His murder of his wife Hélène—a résistance fighter at whose instigation he had joined the party in 1948—turned his exit from the intellectual stage into a dramatic spectacle whereby Althusser the ‘dissident’ could once again pillory the party and its misguided devel-opment—yet at the same time remain true to Althusser the ‘Stalinist,’ who commits no betrayal of his party.

Michel Foucault: the end of the age of ‘Man’

(1) Une chose en tout cas est certaine: (2) c’est que l’homme n’est pas le plus vieux problème ni le plus constant qui se soit posé au savoir humain. (3) En prenant une chronologie rela-tivement courte et un décou-page géographique restreint – la culture européenne depuis le XVIe siècle – on peut être sûr que l’homme y est une invention récente. (4) Ce n’est pas autour de lui et de ses secrets que, longtemps, obscurément, le savoir a rôdé. (5) En fait, parmi toutes les mutations qui ont affecté le savoir des choses et de leur ordre, le savoir des iden-tités, des différences, des caractères, des équivalences, des mots, – bref au milieu de tous les épisodes de cette profonde histoire du Même – un seul, celui qui a commencé il y a un siècle et demi et qui peut-être est en train de se clore, a laissé

(1) One thing in any case is certain: (2) man is neither the oldest nor the most constant problem that has been posed for human knowledge. (3) Taking a relatively short chron-ological sample within a restricted geographical area—European cul-ture since the sixteenth century—one can be certain that man is a recent invention within it. (4) It is not around him and his secrets that knowledge prowled for so long in the darkness. (5) In fact, among all the mutations that have affected the knowledge of things and their order, the knowledge of identities, differences, characters, equivalences, words—in short, in the midst of all the episodes of that profound history of the Same—only one, that which began a century and half ago and is now perhaps drawing to a close, has made it possible for the figure


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


apparaître la figure de l’homme. (6) Et ce n’était point la libération d’une vielle inquiétude, passage à la conscience lumineuse d’un souci millénaire, accès à l’objectivité de ce qui longtemps était resté pris dans des croyances ou dans des philosophies: (7) c’était l’effet d’un changement dans les dispo-sitions fondamentales du savoir. (8) L’homme est une invention dont l’archéologie de notre pensée montre aisément la date récente. Et peut-être la fin prochaine.(9) Si ces dispositions venaient à disparaître comme elles sont appa-rues, si par quelque événement dont nous pouvons tout au plus pressentir la possibilité, mais dont nous ne connaissons pour l’instant encore ni la forme ni la promesse, elles basculaient, comme le fit au tournant du XVIIIe siècle le sol de la pensée classique, – alors on peut bien parier que l’homme s’effacerait, comme à la limite de la mer un visage de sable.

of man to appear. (6) And that appearance was not the liberation of an old anxiety, the transition into luminous consciousness of an age-old concern, the entry into objectivity of something that had long remained trapped within beliefs and philosophies: (7) it was the effect of a change in the fun-damental arrangements of knowl-edge. (8) As the archaeology of our thought easily shows, man is an invention of recent date. And one perhaps nearing its end.(9) If those arrangements were to disappear as they appeared, if some event of which we can at the moment do no more than sense the possibility—without knowing either what its form will be or what it promises—were to cause them to crumble, as the ground of Classical thought did, at the end of the eighteenth century, then one can certainly wager that man would be erased, like a face down in sand in the edge of the sea.

Left column source: Michel Foucault (1966): Les Mots et les choses. Une archéologie des sciences humaines. Paris: Gallimard, p. 398; right column source: Michel Foucault (2004[1970]): The Order of Things: An Archaeology of the Human Sciences, London and New York: Routledge, pp. 421–422.; translation adapted.

This passage closes Michel Foucault’s Les Mots et les choses [The Order of Things], the historical survey of early modern knowledge in Europe with which the author achieved his breakthrough in the French pub-lic sphere. The public success of this book is surprising, given that its 400 pages are based on mostly forgotten texts from over four centuries which the author compiled from dusty library archives. The book’s main concern is a presentation of historical material that utterance (3) refers to. The short conclusion of (3) serves to theoretically position Foucault, who discusses the relationship of historical knowledge to the problem


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


of l’homme [man] (2, 3, 4, 5, 8, 9). The concept word man serves a double function: on the one hand, it refers to conceptual contents which sum up what had been said in the previous sections about the knowledge of the nineteenth century. On the other hand, it moves the problem of ‘man’ into the middle of a polemic argument in which the locutor posi-tions himself in a specific way. The enunciative context in which the person Foucault speaks in the present time T0 becomes deictically visible in this context twice. First, in the passé composé of (5) (qui a commencé [that which began], a laissé apparaître [has made it possible … to appear]) and second with the possessive pronoun of the first person notre [our] (8). How is the deictically highlighted context integrated in the theoretical argument between the protagonists of the intellectual discourse?

Just like the English present perfect (have + past participle), the French passé composé defines a point of time in the past T0′ which is dependent on the subjective or present point of time T0, when the enunciation takes place. In contrast to (9), where the narrative past tense (the French passé simple, which does not exist in English: comme le fit au tournant … [as it did …]), identifies the transition to the era of man around 1800 as an objective point in the past (that is T−1), the use of the passé composé implies the presence of the speaking person in the context of enun-ciation. This person also shows itself through the pronoun notre [our] (8), by means of which the author identifies himself as a member of the community of contemporary readers, and links the ‘thought’ (8) with the presence of T0 of the enunciative context. In so doing, the historical knowledge that is presented by the author becomes attached to the concept word ‘man,’ which turns into an important stake in the intellectual debate of the time wherein the locutor positions himself in a certain way. In projecting the problems of the present onto the his-torical material, Foucault articulates a representation of the present in which a now of the intellectual debate (T0) is defined. The definition of a now T0 is anything but an innocent maneuver, as it forces the reader to locate her- or himself and the other producers of the field with reference to T0. If the era of man, beginning in T0′ or T−1 (that is around 1800), ends at T0, then this is not the case for the producers of the field, whom the reader recognizes as representatives of a ‘philosophy of man,’ easily turn into figures of the past, becoming T−1 objects of the intellectual nar-rative but no longer appearing to themselves as T0 narrators ‘here and now.’ Certainly, mapping the field historically by anchoring T0 in the intellectual discourse is a central strategy whereby the producers distin-guish themselves from their competitors and attempt to dominate the present debate in the intellectual field. It is texts such as these that bear


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


witness to the aging of intellectual products (such as those of Sartre) and that characterize others (the disciplinary philosophy, the canonical humanities …) as ‘traditional’ or ‘obsolete.’

Between T0′/T−1 and T0 the text situates the (anonymous) protagonists of humanism whose names are made explicit in a simultaneously pub-lished article of Foucault’s (June 1966). Thus, the anonymous humanist others alluded to in the utterances (2, 4, 6) may include intellectuals such as Saint-Exupéry, Camus, Teilhard de Chardin and—with reserva-tions—Sartre (Foucault 1994b: 541[698]). The polyphonic operator does not trigger a confrontation between two types of speakers who were already apparent in the discourses of Lacan and Althusser: the oppo-nents and the supporter of ‘man.’ These utterances present a simple polyphonic structure:

pov1(x): [a1] (TRUE (px))

pov2(x): [l2] (NO (pov1(x))

While x stands in for the utterance’s number, px represents the respec-tive contents: p2: ‘Man is the oldest and the most constant problem that has been posed for human knowledge’ (2); p4: ‘It is around man and his secrets that knowledge prowled for so long in the darkness’ (4); p6: ‘And that appearance was the liberation of an old anxiety, the transition into luminous consciousness of an age-old concern, the entry into objectivity of something that had long remained trapped within beliefs and philosophies’ (6). Against this background, the polyphonic organization of the utterance’s contents can be ‘tested’ by dividing both points of view into two persons engaged in a (fictitious) dialogue. Those individuals would be a questioner, A, and a responder, L, who always says ‘no’ to A’s questions:

A: ‘p2?’L: ‘No.’A: ‘p4?’L: ‘No.’A: ‘p6?’L: ‘No.’

This dialogue between a humanist-A and an anti-humanist-L, which can be taken up by real individuals—let’s think of Sartre’s ‘answer’ to Foucault (1966)—can now be supplemented with the propositional


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


contents of (3, 5, 7) which give argumentative support and additional content to L’s monotonous rejection of A’s position:

A: ‘p2?’L: ‘No, because p3.’A: ‘p4?’L: ‘No, because p5.’A: ‘p6?’L: ‘No, because p7.’

This question-answer game ends in (8), where L concludes what has been said by programmatically recapitulating the results. This polyphonic way of organizing conceptual contents is typical for Foucault’s sumptu-ously Baroque style, as it allows the manifold historical knowledge of the producer to join the conceptual fray with his intellectual opponents.

Now, the question is what position the locutor of (8) takes regarding this dialogue between the humanist (A) and his anti-humanist counterpart (L). Does he simply continue to give L further argumentative support? Or does he act as an autonomous judge over A and L, one who ends the testimony of A and L by reaching a verdict at p8? At this point, the locutor of (8) seems to be clearly taking sides. If in (8) L is to appear as a neutral judge, then the reader has every reason to doubt his impartiality, for the locutor of (8) is too close to the locutors of (2, 3, 4, 5, 6, 7). In the argument over ‘man,’ the locutor’s verdict clearly favors the anti-humanist camp. Or does the locutor-Foucault perhaps not leave a small back door open, to provide an escape route for the verdict to be examined in a later appeal? At first glance, the concluding utterance of (9) appears to simply confirm what is said in (8). But if the outcome of the trial is already fixed, why is the sentence so bloated with conditions, reservations and supplementary remarks that the reader can hardly glean the locutor’s position at a glance?

If (9) is meant to repeat only what is said in (8), it would seem rather odd that (9) relies so heavily on the markers of polyphony. Thus, the argumentative operators si [if], ne … ni … ni [neither … nor] and mais [but] conjure up a number of speakers who assist as it were the judge-locutor in formulating his verdict. It is as if the locutor cannot render his verdict without drawing upon a battalion of witnesses, assistants, and counselors. Thus, the many points of view of this utterance can be articulated in the following formal description:

pov1(9): [x1] (TRUE (p: ‘The arrangements were to disappear …’), where it is unknown if the speaker is accepted or rejected by L.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


pov2(9): [x2] (TRUE (q: ‘Some [event] were to cause these arrangements to crumble’)), where the parenthesized element in q relates to the pronoun dont [of which] which connects to a number of syntactically subordinate points of view:

pov1(pov2(9)): [l1] (TRUE (r: ‘We can at the moment sense the pos-sibility of [event]’)

pov2(pov2(9)): [a2] (TRUE (s: ‘We know the form and the promise [of the event]’)

pov3(pov2(9)): [l3] (NO (pov2(pov2(9)))

pov4(pov2(9)): [a4] (TRUE (P from (pov1(pov2(9)) BUT (pov3(pov2(9))), where P indicates what is presupposed by pov1(pov2(9)) (P would possibly be: ‘The possibility of predicting an event implies a knowl-edge about what it means.’), allocated to the other speaker a4.

pov3(9): [x3] (TRUE (t: ‘Man is erased like …’))

pov4(9): [x4] (PERF (pov3(9)), where PERF designates a performative speech act (‘wager’): ‘One wagers on …’

pov5(9): [l5] (MOD (pov4(9)), where MOD represents the modal verb ‘can.’2

pov6(9): [l0] (TRUE (pov5(9)), IF x1 = l1 and x2 = l2), where IF represents the transition to the conditional mode pov1(9) and pov2(9) thanks to si.

The formal representation of the points of view demands some con-centration on the part of the reader. Yet we can immediately identify the position of the locutor vis-à-vis the various speakers carrying the contents of the utterances. We see the locutor’s four adjutants (x1, x2, x3, x4), whereby x1 and x2 are his ‘assessors,’ on whose verdicts he will finally make his own independent decision. Assessor x2 even has at his disposal a ‘team’ of speakers with an individual opinion each. Of these, x2 finally accepts l1 and l3 while rejecting a2 and a4. In contrast, x3 might play the role of a judge who has already ruled on a precedent case and whose verdict can now be adopted by L. The role of a juridical observer, however, would devolve upon x4. The speaker x4 bets on the judge-locutor to refer to the verdict of the other judge. What the locutor then says is that he deems ‘possible’ the result expected from the mouth of the legal observer x4, although only if both assessors are not mistaken in their judgments.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


While in the largely monophonic utterance (8), the locutor-judge l0 declares his verdict p (‘Man is an invention of recent date’) without relying on other speakers, he deploys an entire apparatus of speakers in (9). These speakers prepare, write and submit the verdict to the judge for his final approval. Yet what does the judge do in the end? Instead of declaring the expected victory of the anti-humanist party, he still with-holds his signature from the verdict prepared by his adjutants. Finally, he announces only the possibility of his signing soon. The case thus remains in limbo.

Even though it is not wholeheartedly accepted by the locutor in the end, the verdict p (‘Man is an invention of recent date’) is perfectly available right from the beginning: (2), (3), and (5) already contain p. However, it appears as though the locutor-judge always looks for an excuse in order to postpone his definitive signature under p. Thus, the locutor of (2) does not qualify p as true, but rather as what is called ‘certain’ in (1). Instead of reaching a final verdict and simply saying p (as could be the case in a monophonic utterance like: pov1(2): [l0] (TRUE (p))), the locutor l0 prefers to let another speaker (pov1(2)) express what he has asserted to:

pov1(2): [l1] (TRUE (p))

pov2(2): [l0] (TRUE (pov1(2)), where TRUE means that (2) is enunci-ated in the mode of ‘Il est certain que …’ (‘It is certain that’).

Just as in (9): on peut bien parier que p [one can wager that p], the locutor leaves it to his images (l1) to say p. In this way, l0 supports l1’s claim to truth in the form of ‘It is certain that p.’ If both images of L qualify p to be ‘true,’ this does not mean that p is therefore doubly true. Quite the contrary, saying that p is true reduces p to a mere claim. For if the truth of p did not cause any problem for the locutor, why would he then not simply say: ‘Man is an invention’? An analogous observation applies to (3): on peut être sûr que p [one can be certain that p]. Yet here, the locutor’s assent is also limited by the modal verb can and relativized by another ‘adjutant’ (en-gerund!). Finally, we must note the adverb peut-être [maybe] (5) which qualifies the entire utterance as well as a part of (8). Since the locutor is ultimately responsible only for maybe and not for p itself (which would read in Polyphonese: pov1(2): [l1] (TRUE (p)); pov2(2): [l0] (MAYBE (pov1(2))), he once again avoids committing himself to p.

Yet I want to emphasize that if, in his written texts (which differ from his oral statements), Foucault seldom takes sides for a certain position


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


as clearly as he does in this fragment; this text, too, follows suit with the strategy, typical of Foucault, of masking or erasing the locutor. Foucault’s locutor never decides in the final analysis. Each of the trials staged by an utterance can be appealed again in the next utterance. Thanks to this strategy Foucault (who shortly after the publication of Les Mots et les choses will distance himself from structuralism) enjoys a great deal of flexibility when positioning himself. In April, 1967, he still half-ironically describes himself in an interview as a ‘choir boy of structuralism’ (Foucault 1994c: 581).3 And between September 1967 and 1969 (Foucault 1994d: 603[187]), he begins to distance himself from this intellectual movement before marking, in the Archéologie du savoir (1969), his staunch opposition to the movement as the leader of which he had been widely seen since Les Mots et les choses. Foucault’s dexterity lies in his ability—despite all the theoretical reshufflings and realignments of a field in perpetual ferment—to take such changes into account in his own writings and yet to avoid jeopardizing his own cred-ibility through open contradictions; to ensure the basic coherence of his work and thus to secure the ongoing quotability of his name. A position repudiated once and for all (such as the rejection of ‘humanism’ in the texts of around 1966) can hardly be revised later without incurring a serious loss of intellectual credibility. For this reason, at the end of the 1970s he will turn back to the problem of the subject by framing it as an elaboration and radicalization of earlier positions.

If Foucault partakes in the intellectual polemics of the time yet avoids at all costs siding too much with one or the other party, this strategy bears witness to his caution vis-à-vis all forms of appropriation in an intellectual field where the singularity of the producer and his project must be constantly defended. In fact, each final commitment (see Althusser) runs the risk of becoming vulnerable to the aging process of a discourse in which constant breaks with the past are among the few biding certainties of the time. Yet Foucault is not always successful in his balancing act between masking and profiling himself. In retrospect his academic career certainly appears to be a brilliant success; but in the academic field, with its more pedagogical relations geared toward reproduction, the strategy of continuously switching positions and always leaving a back door open sometimes encounters obstacles. The dilemmas of this strategy, often running contrary to the need for plan-ning, predictability and longevity on the part of academic institutions, is attested to by the fact that Foucault needed two attempts to make it into the ENS; that especially in the more conceptual disciplines (such as philosophy) he couldn’t always and immediately convince his teachers


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


(Pestaña 2006: 49ff.); that he didn’t found an intellectual school in France; that his books entered only gradually into the university curriculum. Surely, one can recognize in his work an ever critical and challenging spirit following an éthique de la déprise, an ethics of ‘leaving behind,’ in which every new book must cast doubt on the one pre-ceding it. However, along with Pestaña (2006: 56), one can also see Foucault’s work as the product of the creative tension of a spirit whose multiple interests do not always unite in a coherent and systematic whole. Yet however one may evaluate Foucault’s intellectual versatility, the fact remains that, in a discourse which constantly offers new ideas and remains open to new ideas, such a strategy will grant the producer attention well beyond specialized academic fields and well beyond the period marked with the structuralist label.

Jacques Derrida: the metaphysics of the text

Left column source: Jacques Derrida (1967c): La Voix et le phénomène. Paris: Quadrige/PUF, p. 66; right column source: Jacques Derrida (2003): Speech and Phenomena. And Other Essays on Husserl’s Theory of Signs. Evanston: Northwestern University Press, 1973, p. 59f.

(1) La présence à soi du vécu doit se produire dans le présent comme maintenant. (2) Et c’est bien ce que dit Husserl: (3) si les « actes psychiques » ne s’annoncent pas eux-mêmes par l’intermédiaire d’une « Kungabe » [sic], s’ils n’ont pas à être informés sur eux-mêmes par l’intermédiaire d’indices, c’est qu’ils sont « vécus par nous dans le même instant » (im selben Augenblick). (4) Le présent de la présence à soi serait aussi indivis-ible qu’un clin d’œil.

(1) The self-presence of experience must be produced in the present taken as a now. (2) And this is just what Husserl says: (3) if ‘mental acts’ are not announced to them-selves through the intermediary of a ‘Kundgabe,’ if they do not have to be informed about them-selves through the intermediary of indications, it is because they are ‘lived by us in the same instant’ (im selben Augenblick). (4) The present of self-presence would be as indivisible as the blink of an eye.

This section concludes the fourth chapter (from a total of six) of La Voix et le phénomène [Speech and Phenomena]—Derrida’s monograph on Husserl’s sign theory, as it is outlined in the beginning of the first part of the second edition of Husserl’s Logical Investigations. In contrast to the previous examples, which partake in contemporary polemics in one form or another, this text seems to deal merely with philosophical ques-tions without intervening in the intellectual debates of the time. Aside


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


from Husserl’s, hardly any other names appear in Derrida’s text. Besides remarks on the philosophical canon (Plato, Hegel, Heidegger …), one finds only two references to the translators of Husserl (such as ‘P. Ricœur’ (62[111]) or ‘S. Bachelard’ (82)) but to no other contemporary producers of the philosophical field. As a philosophical text of a philosopher for philosophers, this text adheres to the specific rules of the philosophical field, as do many of Derrida’s texts. Thus, in order to gain recognition as a philosopher by philosophers, the producer generally follows two rules: first, only refer to timeless reference figures and second, base your-self on their authority. In this way, philosophical discourse is vaulted by a history of ‘immortal’ producers with whom the producers of the field communicate with the goal of becoming ‘immortal’ themselves. Nevertheless, philosophical texts also address ‘mortal’ co-producers in the contemporary field, who read and quote certain texts. As we will see in the following, however conceptually pure and self-referentially closed they attempt to be, Derrida’s texts also refer to their contexts in a certain way.

For more than 30 years, the objective of a (mostly Anglo-American) industry in Derrida exegesis has been to focus on the following questions: how does Derrida position himself toward Husserl? What is Derrida’s criticism of Husserl? More generally speaking, what does Derrida (or his text, as die-hard deconstructivists may say) ‘mean’? I do not intend to compete with the many Derrida specialists and commentators by reconstructing what he ‘really’ meant to say. However, in order to situate the section above in the wider deconstructive argument, we can sum up Derrida’s reasoning as follows: starting out from the Husserlian distinction between ‘expression’ (Ausdruck, expression) and ‘indication’ (Anzeichen, indice), Derrida investigates the relation between meaning and sign, that is between ‘the represented and representative in general, the signified and the signifier, simple presence and its representation, presentation as a Vorstellung and re-presentation as a Vergegenwärtigung’ (Derrida 1967c: 58[52]).4 Thus, on the one hand, Husserl postulates that a sign remains a mere indication (a ‘Kundgabe’) as long as it is not expe-rienced in the living present as a meaningful expression. On the other hand, Husserl stresses that there can be no sign which ‘would take place but ‘once’ ’ (Derrida 1967c: 55[50]).5 While Derrida underlines the cen-tral function of this distinction in Husserl’s phenomenology, he bases his own theoretical project on the aporetic consequences he perceives in Husserl’s attempt ‘to save presence and to reduce or derive the sign, and with it all powers of repetition’ (1967c: 57[51]).6 This is what establishes him as the author of philosophical deconstructivism.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Derrida’s texts mobilize complex stocks of conceptual knowledge; they are part of longer theoretical arguments in which individual utter-ances have a precise argumentative value as is testified by utterance (1). This utterance is loaded with Derrida’s characteristic ‘jargon’ and with the terms that Derrida has coined: présence à soi [self-presence], vécu [experience], présent [present], maintenant [now]. In this usage, présent and maintenant do not have a deictic function. Instead of referring to the enunciative context of the utterance, they are concept words in an elaborated argumentation that is unfolded in the co-text.

Do these terms evoke pure, theoretical contents which can exist out-side the enunciative context? Can the outside, as the deconstructive doctrine postulates, be completely kept out of the text? Indeed, a ‘time-less’ text such as this one does not immediately deictically refer to its context of enunciation. Let us examine the discursive instances through which the locutor engages in a conversation with his philosophical interlocutors. Who speaks in utterance (1)? In (1) there are no markers of enunciation which would, like the nots or buts of the previous examples, divide what is said into conflicting speakers or sources of enunciation. Do we therefore have to believe that this utterance is ‘monophonic’? In fact, this utterance contains only one speaker lx, who necessarily concurs with the director l0: pov1(1): [l0] (TRUE (p)), where X and Y represent the concept words and p for ‘X doit se produire dans Y’ [‘X must be produced in Y’]. Now, some sort of polyphonic process is triggered by the modal verb devoir [must], which allows a ‘constative’ utterance of one speaker to be cited by another, ‘normative’ speaker. In the following formalist representation, the image of L (in particular l1) accounts for the ‘consta-tive’ utterance p, and the other image of L (l0) represents the ‘normative’ demands concerning p:

pov1(1): [l1] (TRUE (p: ‘X must be produced in Y’))

pov2(1): [l0] (MOD (pov1(1))), where MOD designates the modal verb devoir [must].

However one describes the polyphonic function of modal verbs, both speakers pov1 and pov2 turn out to be images of L. Must everything that is said therefore be traced back to L (that is to the person Derrida), who as the author whose name is indicated on the cover of the book, holds the full copyright over its contents? Why can the reader, despite the ostensi-ble ‘monophony’ of the utterance, not get rid of all doubts and possibly allocate what is said to a third party, namely ‘Husserl’? In fact, the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


‘monophony’ of (1) is deceptive since in (2) the person who is responsible for the spoken is identified, for while l0 denotes the person of ‘Derrida,’ this ‘Derrida’-l0 reveals to the readers the name of the person responsible for what is said in (1), namely ‘Husserl.’ Not only does the utterance (2) give us the source of (1); in ‘et c’est bien’ [‘this is just what …’] we can also recognize the full authorial acumen of the deconstructive philoso-pher, for the locutor reminds us that—retrospectively as it were—we can associate this source of enunciation equally with ‘Husserl.’

As opposed to a reference such as ‘(1), says Husserl’ or to the academi-cally even more correct reference ‘(1) (Husserl 1901)’, it is this formula ‘et c’est bien’ which signals to the reader: ‘Yes, I (the ‘Derrida’-locutor) am the originator of what is said in (1) but, as you know, what I said in (1) is in reality based on ‘Husserl’.’’ In English, it is precisely the little word just: ‘And this is ‘just’ what …’ that betrays the polyphonic finesse with which Derrida makes the reader switch between a quoting and a quoted discourse.

The absence of enunciative markers does not necessarily turn an utterance into a ‘monophonic’ utterance. In fact, ‘monophony’ may be an unfortunate expression to the degree that it presupposes a limited number of voices in the utterance. In fact, in a ‘monophonic’ utterance of the p0-type (where p0 consists of a subject and a predicate), the reader will identify one single point of view with one speaker (that is pov1(1): [l0] (TRUE (p0))) only as long as he or she does not have to assume the presence of any other speakers. In fact, a ‘monophonic’ utterance p0, uttered by an individual L0 (described as L0(p0)), can be quoted by another individual L1 who does not explicitly identify himself. The latter would be made explicit only if the utterance were transformed to L1(p1: ‘L0 says that p0’). It is equally imaginable that L1 is quoted by a third individual L2 saying p2 (‘L1 says that p1: ‘L0 says that p0’ ’). In theory, there are no limits for p0 to be quoted by ever more individuals, and the only reason for the reader not to incessantly produce these nested utterances is a pragmatic one: nobody can afford the effort of produc-ing all these quotations in quotations in quotations… The reader will therefore stop searching for ever more quoting individuals L1 (or L2, L3 … Lx) as soon as the circumstances allow him to do so. Only when noth-ing speaks against terminating the interpretive process, that is when no evidence exists for p0 not to refer to L0 (or L1, L2 … Lx−1), will the reader venture the hypothesis that p0 is in fact uttered by L0 (or L1, L2 … Lx−1). The ‘monophony’ of p0 is therefore the reader’s wager on the default-speaker L0, a wager which the reader can revise at any time in the light of new information.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


In our example, in the transition from (1) to (2), the reader’s initial bet on L0 is called into question, for (2) suggests the possibility that (1) may be uttered not only by a ‘Derrida’-L0 but also by a ‘Husserl’-L0. In this case, (1) could be transformed into the utterance Husserl says that X has to be produced in Y. In this case, the reader would of course not take a risk if he placed a bet on the possibility that the speaker of this utterance was ‘Derrida.’ Therefore, polyphony does not always have to be signaled via operators such as not or but. As Ducrot points out in the case of irony, the association the reader makes between L and some individual person can be questioned retrospectively by informa-tion creating obvious contradictions, by exaggerations or by gestures and intonation. Thus, the utterance Why is Derrida’s writing so simple and comprehensible? may produce irony if its speaker turns out to be a pseudo-image of L. In the same way, the speaker of (1) is not necessar-ily the speaker which he initially pretends to be. In contrast to classical irony, however, the speaker of the utterance is not completely rejected, as the speaker of the quoting discourse L (‘Derrida’) continues to draw upon the authority of the speaker of the quoted discourse Q (‘Husserl’). We are dealing with a case of quasi-irony at most, which is supported by the fact that the question whether the locutor himself (‘Derrida’) speaks or another speaker (‘Husserl’) is quoted by the locutor (‘Derrida’) remains unanswered. Thus, the reader can read (1) as playing through two different scenarios, namely:

pov1(1): [l1] (TRUE (p))

pov2(1): [l0] (MOD (pov1(1)))

or

pov1(1): [q1] (TRUE (p))

pov2(1): [q2] (MOD (pov1(1)))

pov3(1): [l0] (QUOTE (pov2(1)))

The (controlled) undecidability of allocating the utterance’s voices is a method which Derrida applies with great virtuosity in his texts. Deconstruction’s argumentative force derives from precisely these utterances which the reader can perceive to be signed by Derrida as well as by speakers of the quoted discourse. This is what constitutes the deconstructive gesture of Derrida’s theoretical project: reading a text deconstructively means quoting utterances of a text and confronting


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


them with utterances from another discourse, for example (1), which can be traced back to the sources of the quoted as well as of the quoting discourse. It is these undecidable quotations with the help of which Derrida marks his ‘subversive’ distance toward the canonic texts of his discipline while drawing on the authority of their producers. In this way, Derrida offers a method of reading which enables the readers of theoretical texts to produce texts about texts. Uttered in the name of the speakers of the quoted texts, these texts keep the quoted texts at a distance without L rejecting them. In so doing, the deconstructive producer can set him- or herself apart from the disciplinary mainstream with its established hierarchies and connections and maintain the avant-gardist claims of his or her project by means of controlled deviations, distinctions and innovations. Thus, while addressing non- philosophers with his philosophical texts about philosophical texts, Derrida succeeds in the unlikely feat of partaking in a theoretical conjuncture whose decisive impulses come from outside the philosophical field.

Now, (3, 4) utilize the ‘classic’ operators of polyphony, whereby the quoted discourse evokes a range of speakers to which the locutor, in turn, takes a certain stance. In order to formalize the polyphonic configura-tion of this example in which a quoted and a quoting discourse intersect, I will define ax as images of the allocutor A, which are rejected by cx, the images of the speaker of the quoted discourse C. The dependence of the speakers who are responsible for what is said after when (that is lx) is designated by the placeholder speakers xx, who will take on what is said before when. The images of L, that is lx, coincide with the speakers of the quoting discourse:

pov1(3): [a1] (TRUE (p: ‘The mental acts are announced to themselves through the intermediary of a ‘Kundgabe’)), whereby the transition from quoting to quoted discourse indicated by the quotation marks (‘mental acts’) is not taken into account in order not to make the analysis even more complicated (see the similar transition from pov5(3) to pov6(3) and from pov9(4) pov10(4))

pov2(3): [x2] (NO (pov1(3)))

pov3(3): [a3] (TRUE (q: ‘They have to be informed about themselves through the intermediary of indications.’))

pov4(3): [x4] (NO (pov3(3)))

pov5(3): [c5] (TRUE (t: ‘They are lived by us in the same instant.’), IF x2 = c2 and x4 = c4)


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


pov6(3): [l6] (QUOTE (pov5(3))), where the quoting discourse QUOTE is signaled by the quotation marks and by the italics of the original wording.

pov7(4): [a7] (TRUE (r: ‘The present of self-presence is divided.’))

pov8(4): [a8] (MOD (pov4(9))), where MOD represents the modaliza-tion by the suffix ‘-ible’ from ‘indivisible’.

pov9(4): [z9] (NO (pov8(4), IF x2 = z2 and x4 = z4)), prefix ‘in-’ from ‘indivisible’!

pov10(4): [l0] (QUOTE (pov9(4))), where the quoting discourse QUOTE is signaled by the conditional of the verb and the italics.

When trying to associate these ten speakers of both utterances with the two discourse figures (‘Derrida,’ ‘Husserl’) attested in the co-text, the reader will not have great difficulty relating what is said by cx to ‘Husserl’ and what is said by lx to ‘Derrida.’ But ‘Derrida’ not only appears as lx, as the individual who guarantees that what is said in the quotes and presented in italics really originates from C. The images of A (ax), which are rejected by C (‘Husserl’), also seem to reflect the opinion of L, too. The ‘aporia’ that the deconstructive philosopher now unfolds before the eyes of the reader is founded on precisely this uncertain relation between L and C. While L still identifies himself with C in the light of (2), in (3) L identifies with A, the ‘anti-C.’ The ‘magic’ of the deconstructive process thus results from the fact that in (1, 2) L still appears to represent for the reader the cause of C, whereas in (3) C rejects A, who now says exactly what L had expressed in (1, 2) with the support of C. As a consequence of these complex polyphonic opera-tions, C loses some credibility, yet without L becoming ‘disloyal’ toward C. By affirmatively repeating the discourse of C, L steps back behind C, who, so to speak, appears to dissolve on his own.

By letting the position of C dissolve in an act of self-demolition without L overtly intervening, Derrida could venture to take a stance in intellec-tual debates outside philosophy while remaining a philosopher perfectly faithful to the rules of the philosophical game. If Husserl’s phenomenol-ogy, along with Sartre’s existentialism, Merleau-Ponty’s psychology and Tran Duc Thao’s Marxism had been at the center of theoretical discus-sion in the years following the Second World War, then by the 1960s not only phenomenology, but philosophy as a whole was losing prestige and being driven into increasingly defensive positions. It becomes increasingly difficult for specialists in phenomenology to appear as


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


theoretical spearheads of a wider intellectual movement. This change did not escape Derrida’s notice either. While his colleagues at ENS, such as Michael Foucault and Pierre Bourdieu, still worked in phenomenological philosophy in the early 1950s, they soon turned to the expanding field of the sciences humaines. Additionally, with Husserl’s work translated by the early 1960s (the cited volume of Logical Investigations was translated in 1961), the novelty value of Husserl’s philosophy diminished for young, ambitious philosophers such as Derrida.

By offering a philosophical reading of philosophical texts, Derrida produced texts which enabled his philosophical readers in the field to read philosophical texts philosophically—that is, in full recognition of their canonical authority—while giving them protection from the looming devaluation of phenomenology on the symbolic market of ideas. Derrida accomplished this balancing act by producing exquisitely composed verbal scores which present philosophical themes that are distributed among the voices of intellectual discourse and masterfully conducted while still following all the rules of philosophical composi-tion. The echo did not fail to materialize. He achieved a resounding success especially with those who strived to adjust their philosophical capital to the new theoretical tendencies of the sciences humaines and to follow up with the new reference figures of anti-humanism such as Saussure. Derrida’s intellectual and theoretical jolt not only gave rise to effects resembling conversions and illuminations, in which the readers quickly abandoned old theoretical positions in order to embrace new ones, more in tune with the reigning intellectual atmosphere. His project also aroused the moral indignation of those who perceived his deconstructive readings as no more than the sleights-of-hand of a sophist rhetorician whose aim was to banish questions about meaning, truth, and reason from Western philosophy.

Yet how can this text, which comes in the most classical guise of aca-demic philosophy, be read as a document of theoretical anti-humanism or structuralism? The text’s hegemonic function in the intellectual debate of the time derives from the way in which it helped readers change their discourse, namely from a ‘metaphysical’ discourse to one critical of metaphysics. Thus, Derrida’s text can be read as an instruc-tion manual for the reading of canonical texts with which the reader can move through diverse regions of discourse; the trick lies in how Derrida’s texts instruct the reader in the citation of texts whose contents the quoting speaker rejects in the name of the quoted speaker. In this way, the quoted text turns out to be the material repository for two different discourses: for a first, quoted discourse, whose ‘metaphysics’


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


is problematized by the second, quoting discourse, yet in which the source of the first discourse is nonetheless recognized as an uncontested philosophical authority. In the aforementioned case, the transition from quoting to quoted discourse is signaled by typographical means such as quotation marks and italics which reflect the heterogeneity of the discourse fragment. The attentive reader comes upon the source of these contents—of the ‘mental acts’, ‘Kungabe [sic],’ ‘vécus par nous dans le même instant (im selben Augenblick)’ (3) as well as the ‘clin d’œil’ (4) in italics, which the German-speaking reader will understand as an allusion to the ‘instant’ in (3)—if he skips back to the beginning of the chapter, where he finds a longer quotation of Husserl’s contain-ing the stated phrases, minus the orthographically botched ‘Kungabe.’ In the following passage, too, there are typographically marked inserts which refer back to the exact wording of the original text. The reader will find the original text quoted by Derrida on page 43 at § 8, ‘The expressions in the solitary life’ of the second volume of the Logical Investigations. The following diagram shows Husserl’s original citation on the left, the English translation of Husserl as it appears in the English

‘[…] In der monolo-gischen Rede können uns die Worte doch nicht in der Funktion von Anzeichen für das Dasein psychis-cher Akte dienen, da solche Anzeige hier ganz zwecklos wäre. Die fraglichen Akte sind ja im selben Augenblick von uns erlebt […].’

‘[…] (1″) In a mon-ologue words can perform no function of indicating the existence (Dasein) of mental acts, since such indica-tion would there be quite purposeless. (2″) For the acts in question are them-selves experienced by us at that very moment […]’

‘[…] (1′) Dans le mon-ologue, les mots ne peuvent toutefois nous servir dans la fonction d’indices de l’existence (Dasein) d’actes psy-chiques, car une telle indication n’aurait ici aucune finalité (ganz zwecklos wäre). (2′) Les actes en question sont en effet vécus par nous-mêmes dans le même instant (im selben Augenblick). […]’

Left column source: Edmund Husserl (1984[1901]): Logische Untersuchungen. Zweiter Band, Erster Teil: Untersuchungen zur Phänomenologie und Theorie der Erkenntnis. The Hague, Boston, Lancaster: Martinus Nijhoff, p. 43 (§8 ‘Die Ausdrücke im einsamen Seelenleben’); middle column source: Edmund Husserl (1970[1901]): Logical Investigations. Volume II. London: Kegan Paul, pp. 279–280 (§8 ‘Expressions in solitary life’), also cited in Jacques Derrida (1967): Speech and Phenomena. Paris: Quadrige/PUF, p. 54 (from Chapter 4: ‘Meaning and Representation’); right column source: cited after Jacques Derrida (1967) dans La voix et le phénomène. Paris: Quadrige/PUF, p. 54 (chapitre iv « Le vouloir-dire et la représentation »).


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


translation of Derrida’s Speech and Phenomena in the middle (which corresponds to the English translation of Husserl’s Logical Investigations of 1970); and finally Derrida’s French translation of the Husserl pas-sage on the right (which differs from the official French translation of Husserl’s Logical Investigation of 1961).

In (1′, 2′), the French philosopher’s text negotiates a first transition, namely from a source text (1″, 2″) to a translated text (1′, 2′). If this passage deals with the more technical problem of the translation from one language to another, we also need to register the numerous marks the locutor has left in (1′, 2′) in the form of German words, by which he indicates the limited responsibility for his translation while informing the reader: ‘I try here to reproduce the original text for you to the best of my knowledge. But in French I say something which Husserl does not want to say in German.’ Thus the task of the chapter in which this cita-tion appears is to recover what is said unintentionally and to paraphrase what gets lost in the transition from one text to the other. Therefore, in ever new paraphrases, the text attempts to capture what is lacking in the utterances of the French translation (1′, 2′). Paraphrasing this loss implies another transition, namely from (1′, 2′), the French quote of Husserl, to various paraphrases which surround the immediate co-text of (1′, 2′). This is what the utterances (1, 2, 3, 4) do, which are uttered in the mode of the quoting discourse. Typographically distinct as direct quotes in (3), they mix with fragments of the quoted discourse of (1′, 2′). Accordingly, in (1, 2, 3, 4), the quoting discourse introduces terms absent in the quoted discourse of (1′, 2′) which are needed to correct (1′, 2′) and render (1″, 2″). These conceptual ‘patches’ are présence à soi du vécu [self-presence of experience] (1), présent comme maintenant [present taken as a now] (1), présent de la présence à soi [present of self-presence] (4), which make up Derrida’s characteristic jargon.

How does the term présence relate to other terms with which it is associated: soi [self], maintenant [now], vécu/vie [experience/life], clin d’œil [instant], aucune finalité [quite purposeless]? Présence [presence] not only fills the lack of (1′, 2′) but also gets charged with a large number of theoretical questions and philosophical problems. These are reflected elsewhere in lists of concept words which exist in relations of equiva-lence, such as: ‘We have experienced the systematic interdependence of the concepts of meaning, ideality, objectivity, truth, intuition, perception, and expression. Their common matrix is being as presence’ (1967c: 111[99]).7 If présence plays a central role in the argumentation of the text at hand, this may be due to its nature as a pseudo-concept word. As a derivation of the deictic adjective présent, présence is a hybrid sign. Therefore, it not only evokes the encyclopedic knowledge


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


of the philosopher but also operates with deictically shown contexts of enunciation which are not really shown in the discourse. It is characteristic of the adjective présent that it may indicate both the time (maintenant [now]) as well as the place of the enunciation (ici [here]). Thus, the person who expresses the utterance (0) Je suis présent [I am present] signals both his temporal and spatial presence. In the transformation into présence, the utterance is cut off as it were from the person (H) of the enunciation context. What remains of (0) is a kind of preconstructed présence existing independently of the specific act of enunciation. Thus, if the suffix -isme were a morphologically possible ending for présenc-, the text would probably show a p-ism (*présentisme) rather than présence. In the case of présence, therefore, information is given about the propositional contents of the ‘original’ utterance (0) but not about the ‘original’ locutor H.

Now, in (1, 2, 3, 4) the deictic preconstruct présence, which L needs to introduce in order to pass from (1″, 2″) to (1′, 2′), becomes the object of a conflict between the speaker of the quoting discourse L and the speaker of the quoted discourse C. For what does L reproach C concern-ing présence? According to L, by arguing in the name of présence, C draws upon a questionable authority, namely upon the authority of the ‘Ur’-locutor H. Insofar as C’s concept word (‘expression’, ‘indication’, ‘expe-rience’…) can be paraphrased with présence and its derivatives, C relies on an authority that is so obviously and so generally accepted that it does not become an object of discursive controversy, that is about H, the locutor, who assumes responsibility for the utterance (0) I am present. It is self-evident who H is, who is ‘present’ not only ‘here’ and ‘now’ but ‘always’ and ‘everywhere.’ For L, however, it does not go without saying that C can speak for this nameless H. If C’s concept words presuppose the discourse of H, C relies on H to formulate his doctrine. From the point of view of C, H is indeed so well-known a source that he needs to be neither mentioned nor named. This H is any individual, the homo sapiens or zoon logon, the human being as such, which is cited as the ultimate authoritative source of all conceptual contents in C’s discourse. Against C, L (Derrida) asks the question: ‘Can the philosopher speak on behalf of H, the human being? Is the human being as such a legiti-mate philosophical authority?’ This is the question raised by the text, raised as the concept words of the quoted discourse are paraphrased by the speaker of the quoting discourse by means of presence and similar concept words. Vis-à-vis this question, the locutor positions himself by providing présence with the attributes of ‘Western metaphysics’: does Husserl’s phenomenology, or rather the entire philosophical tradition, not find itself within a ‘metaphysics of presence’ (115[102])?


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


As Derrida tells us, even the deconstructive philosopher is finally unable to step out of the history of metaphysics. The reader familiar with the theoretical debates of the time can read from this critique of metaphysics Derrida’s more or less dissimulated stance against human-ism, whose code word is présence. By paraphrasing Husserl’s text with the help of présence, the text is initially embedded in the cited discourse operating with preconstructed deixis. Yet in the quoting discourse, the quoted discourse, which turns on this universal zero point of enuncia-tion, becomes qualified as ‘metaphysical.’ As the theoretically informed reader recognizes, the quoting discourse draws from the Saussurian critique of essence and origin, according to which the value and contents of any concept word can be determined only by their difference to all the other signs.8 Hence, if we want to understand Derrida’s text, we will have to read the text he quotes twice—once framed by the quoted discourse, in which deixis is allowed, and the second time framed by the quoting discourse, in which deixis is forbidden and where there is only a play of difference devoid of origin and time.

It is the prohibition of preconstructed deixis in the quoting discourse which effects an ‘enunciative reprogramming’ of philosophical read-ing. The reader will indeed experience the enunciative reprogramming of canonical philosophical texts as a change of discourse, namely as a change from ‘humanism,’ whose concept words refer back to a meaning-endowing center, to ‘anti-humanism,’ in which all terms define them-selves through their position in a system of differences, namely in the decentered world of text, writing and différance. If in this way Derrida manages to intervene in discussions outside philosophy, he achieves this unlikely feat by reframing the very historical dispute about structuralism of around 1967 as a timeless problem of philosophical history. Once again he demonstrates the acumen of a producer who attempts to defend a discipline on the decline.

By veiling its polemical purpose and its addressees in the contemporary debate, this text does not weaken its symbolic efficacy in the intellectual field. On the contrary, insofar as he provided philosophical ammunition to the anti-humanist cause, Derrida found short cuts to cross through regions of the intellectual field hitherto separated from each other and occupied a singular position in the field. As the founder of a renowned philosophical project, he became a leading representative of the intel-lectual generation of structuralism around 1970. But even a producer like Derrida, writing with surgical precision, could not overcome the conflicting constraints he had to negotiate in the field. Soon after 1970, as the spectacular growth rates of the educational system slowed and the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


intellectual field entered a period of consolidation, Derrida, who lacked secure roots in an academic position to consolidate his success, found things growing quieter around him. If he, unlike many other representa-tives of his intellectual generation, remained loyal to a classic discipline such as philosophy, he could not but arouse the suspicion of some non-philosophical co-producers in the field, despite all subversive efforts on his part to maintain his independence. His famous fellow students of ENS, for example, accused him of a ‘scholastic’ attitude serving the reproduction of an obsolete discipline (Bourdieu 1979: 578ff.[494ff.]) or of being a ‘school-masterly critic’ engaged in explication de texte (cf. the second edition of Foucault 1961). At the same time, without an institu-tional position from which he could exert institutional influence on the discipline, he remained rather marginal in his own discipline. Nor did his texts find their way into the curricula of universities and schools. If for many non-philosophers, his project was too much indebted to the exegetic reading of canonical texts in classical disciplines, for many philosophers they were too much connected to the glitter of an ephem-eral intellectual fad. He was unable to satisfy all sides at the same time. Derrida would probably have remained no more than a minor intellec-tual star disappearing as quickly as he had appeared on the intellectual firmament, had it not been for his success on other intellectual stages outside France. In the late 1970s, Derrida started a second career in the United States, where he redefined his intellectual role. After his career as one of the philosophers of ‘structuralism’ in France, he now became a major protagonist of Theory international style, of ‘poststructuralism.’ Only the international circulation of his texts made Derrida a veritable star of the intellectual scene, and only at this point did he gain entry to the hall of fame in France, too, as a luminary of philosophical discourse.

Tel Quel: narrating the revolution

Les concepts de texte, d’inter-textualité, d’écriture sont explic-itement à la base d’une mutation dans notre civilisation, et les noms que nous citons de façon répétée: Lautréamont, Mallarmé, Marx, Freud, en sont les symptômes mas-sifs et, à notre avis, encore à venir.

The terms text, intertextuality, writing explicitly build the base of a change in our civilization, for which the names that we keep quoting: Lautréamont, Mallarmé, Marx, Freud, are its massive symptoms and, in our opinion, still to come.

Source: Philippe Sollers (1968b): « Le réflexe de réduction » In: Tel Quel (ed.), Théorie d’ensemble. Paris: Seuil, p. 394, translation by J.A.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


This quote is found in a contribution from Phillip Sollers to Théorie d’ensemble [Set Theory] (Tel Quel 1968). This volume contains a selection of articles that appeared in the journal Tel Quel, edited by Sollers. Under the title ‘Le réflexe de réduction’ [‘The reduction reflex’], Sollers sums up the literary and theoretical program of the journal in which a number of writers (among others Georges Bataille, Alain Robbe-Grillet, Francis Ponge, Pierre Klossowski, William Burroughs) as well as theorists (such as Jacques Derrida, Roland Barthes, Michel Foucault, Tzvetan Todorov, Jean Starobinski) had published since its foundation in 1960. The edito-rial board consists of staff writers who contribute to the special issues headed by external authors. The symbolic hierarchy between the mostly younger and unknown staff and the external authors is reflected in the structure of Théorie d’ensemble, which is led by three widely known theorists (Derrida, Foucault, Barthes), to whom the staff writers, above all Sollers, continuously show their reverence.

After the producers of theory, we now come to the mediators of the-ory. The quoted excerpt raises the question how producers of intellec-tual discourse draw on ideas which they have not produced themselves. Vis-à-vis the theory producers quoted again and again in Tel Quel, such as Derrida, Althusser, Lacan, Barthes, Foucault and Kristeva, Sollers posi-tions himself as a reproducer who draws from these celebrated produc-ers in order to define the theoretical profile of his still barely established journal and to present himself as the spokesman of an intellectual vanguard active on all fronts. Undoubtedly, Sollers’ strength lies in his ability to quickly absorb theoretical and literary tendencies and to make them accessible to a wider non-specialist public. As a theory reproducer, he stakes a claim on the theoretical legitimacy of theory producers, with whom he forms a symbiotic relationship. For the producer of a long-term theoretical venture, Tel Quel offers an audience keen on following the most recent developments in the intellectual debate. Not bound by any particular long-term theoretical commitments, the reproducer chooses from the various projects offered on the symbolic market and defines the ‘cutting edge’ and ‘avant-garde’ projects over and against those which he or she dismisses as ‘traditional’ or even ‘reactionary.’ Therefore, the reproducer is in an asymmetrical relation with the producer, who, as a prestigious contributor to the journal, can retract his symbolic support at any moment. Yet by qualifying the producer as belonging to an intellectual ‘avant-garde,’ the reproducer renders him or her a valuable service which makes the producer at least to some degree dependent on the reproducer. In this mutually defined relationship both sides can get their money’s worth. For this reason,


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


the symbolic hierarchy between the producer and the reproducer not only reveals the differing volumes of their cultural-theoretical capital, but also testifies to the differing demands and modes of labor in the intellectual field. The reproducer’s function is not limited to relaying, repeating, or summarizing the producer’s theoretical project. His task is to reframe it in a certain way, which requires a type of discursive competence in its own right. My goal will be to point out the distinc-tive features of a ‘discourse of reproduction’ in the selected text, absent in the ‘discourse of production’ as it has been analyzed in the previous sections: first, the rendering visible of one’s own persona and its inte-gration into a theoretical argumentation; second, the listing of certain producers of the field as authoritative references; third, the framing of a ‘revolutionary’ narrative.

In French academic texts, the first person plural nous/notre is some-times used as an ‘I of modesty’ (as opposed to the ‘royal we’ in English), which instructs the reader about the argumentative course of the text. However, at least the first occurrence of the pronoun does not represent an ‘I of modesty,’ since ‘notre civilization’ clearly points beyond the text to the context in which the speaker produces his or her discourse. For texts of producers who take a position in relation to socio-historical questions, such a localization of writer and reader in the context of enunciation is not unusual, as the examples of Althusser and Foucault prove. However, the occurrences of the following two pronouns are rather untypical, as they denote a group of people among whom the reader is not included. Here, Sollers speaks together with the staff writ-ers of the journal. They turn to the reader as a group, reminding him of their existence, pointing out to him that they (and no one else) quote the names and that the lines reflect their opinion (and not those of others). No matter whether the author is aware of whom exactly he alludes to with we/our in the act of writing, the fact remains that the speaking group of the enunciation context is referred to, which frames the political-theoretical-literary contents in a certain way. Instead of simply notifying the reader of the contents (‘p’), the speaker does not forget to append to these contents the message that the group wants to say certain contents (‘we say that p’). The reader is thereby forced to complement what has been read with contextual information about the group. Perhaps the reader knows that the journal, in its eight years of existence, has published many articles about aesthetic and theoretical questions and themes, that the average age of a Tel Quel writer is only the mid-1920s, that these barely known writers rarely possess higher academic degrees or even come from an academic background. And if


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


the reader has been sensitized by Bourdieu’s sociology, he or she will take cognizance of their middle-class Parisian background. Therefore, whatever the reader knows about this group is activated by we/our and attaches itself to the foregrounded conceptual-theoretical contents. If the reader knows the group as ‘provocative bohemian Parisians,’ this information is activated by the deictic pronoun we/our (without which it would remain merely a knowledge based on free-association, which the reader may or may not mobilize). In so doing, the reader’s knowl-edge about the deictically reflected group is brought into direct contact with the foregrounded contents. The question of who is speaking is no longer ‘irrelevant’ or even ‘illegitimate,’ as is often the case in academic discourse as well as in the foregoing examples of intellectual discourse. The ‘who’ of the group obtains the same argumentative weight as the ‘what’ of the theoretical positions and concepts negotiated in the text. In this way, the relations of authority are inverted. The argument becomes apodictic. One does not argue that ‘I think x is correct, because it is supported by y’ but rather: ‘Because I, who, as you know, represent y (that is the ‘avant-garde,’ the ‘only victorious party’ and so on), x must be correct.’

Even though the person speaking is quite often shown in the discourse of production, the mingling of concepts and biographies, of works and lives is a risky business, as the text becomes vulnerable to the aging of intellectual discourse. The deictically indicated person can occupy new positions later on, and the context in which the text was created can change, which may have an immediate effect on the credibility of the argument. Therefore producers tend to leave it to the reproducers to speak of their person and to designate them as members of certain groups of producers. In the discourse of reproduction, the pronouns we/our possess a double function: on the one hand, the deictically shown context of enunciation gains thereby in argumentative value; on the other hand, these pronouns signal the distance separating the repro-ducers from the producers as figures of uncontested authority. But how the text alludes to those contemporary producers who are nowhere mentioned, this is a question which will now lead over to the second characteristic of the reproducer’s discourse.

The reader who has attentively read the previous issues of the journal or followed the evolutions of the intellectual book market will recognize the three terms texte [text], intertextualité [intertextuality], écriture [writing] as abbreviations for certain producers in the intellectual spotlight of the times, namely ‘Derrida’ (texte and écriture), ‘Barthes’ (texte and écriture) and ‘Kristeva’ (intertextualité). The same holds true for the proper names


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


‘Marx’ and ‘Freud,’ which evoke the theoretical projects of ‘Althusser’ and ‘Lacan.’ The informed reader can possibly even connect ‘Lautréamont’ with the name of a contemporary producer: Maurice Blanchot, who had published a literary essay with Minuit not long before titled Lautréamont et Mallarmé (1963). As representatives of a literature which had broken free of (in the words of Bourdieu) extra-literary instances and the public marketplace, ‘Lautréamont’ and ‘Mallarmé’ stand for the aestheticist trends which include contemporary writers grouped around Georges Bataille, such as Pierre Klossowski and Maurice Blanchot, as well as the Nouveau Roman (Alain Robbe-Grillet, Michel Butor, Claude Ollier, Claude Simon). Like quotations which are so well-known that naming their authors is superfluous, these concept words and names bear the signatures of contemporary producers in the field, figures whose author-ity can be marshalled in arguments and who simultaneously come equipped with the weight of references from the nineteenth century. The structures of authority in intellectual discourse are made effective precisely because the text does not make its sources explicit, because who is ‘actually meant’ is alluded to only through certain historical predeces-sors. Thus if the reader can guess which producers are ‘actually meant’ in the text, he or she will tacitly consent to the author’s suggestion that these producers are continuing a tradition that began in the nineteenth century. Therefore the search for the unnamed sources is provoked by concepts de [terms] and noms que nous citons de façon répétée [names we keep quoting], which signal the transition from a quoting to a quoted dis-course. The qualification of the quoted terms as ‘terms’ and the proper names as ‘names’ is only possible in the discourse of reproduction. Thus, this peculiar doubling underscores the distance of the locutor to the conceptual and theoretical contents he quotes. The concepts encoun-tered by the reader here do not represent self-constructed knowledge, but knowledge which carries the signature of other producers in the field. In contrast to those theory producers who rely on certain producers who long ago dropped out of the intellectual game of the field (for example Lacan’s reliance on Freud, Althusser’s on Marx, Derrida’s on Husserl …), Sollers offers an armada of references including active producers of the field to whom he sometimes even maintains close personal ties—such as to his friend Roland Barthes and his wife Julia Kristeva.

In the discourse of reproduction, the producers who continue an intellectual tradition going back to their historical reference figures may appear as protagonists of a contemporary movement with com-mon goals and opponents. For the cited producer, such paradigmatic effects offer advantages and disadvantages. On the one hand, he is now


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


indeed able to appear as the spokesman of a group of producers, thereby multiplying the weight of his project in intellectual discourse. However, he also becomes dependent upon the interventions and careers of other producers, whose prospective fate he will share. It thus comes as no surprise that producers do not necessarily accept the ‘movement effect’ emerging in the discourse of reproduction. While Althusser, for example, enthusiastically and permanently sides with the structuralists, Foucault’s identification with structuralism is of brief duration, and Lacan and Derrida keep the forces of appropriation at bay by producing a self-referential theoretical language. Yet even if the producers voice only little enthusiasm for labels such as ‘structuralism,’ this does not mean that they reject the theoretical projects of their peers (as we know, Foucault was a friend of Althusser’s and also made clear his closeness to Lacan, see for example 1994a). And the fact that producers take up these labels mostly in quite marginal places, such as footnotes or inter-views, does not necessarily express any fundamental reservations with respect to the groups and movements of which they are supposed to be a part. Instead, such ‘marginal’ recognition emphasizes the division of labor in which the producer and the reproducer act according to certain well-defined roles. By rejecting the label, the producer marks his independence and singularity vis-à-vis the other producers in the field. However, he or she will avoid rejecting the label so unambiguously that the reproducer can no longer adorn himself with his or her name. The producer may be perfectly at ease with the label; but he or she is not allowed to say so.

Both sides can profit from this game as long as neither of them ques-tions their tacit agreement, whereby the producer leads his fellow pro-ducers in the achievement of symbolic distinctions, and the reproducer does not too prominently display any names of the producer’s rivals. Especially in an era as eventful as the years around 1970, the collabora-tion between producer and reproducer is subject to cancellation, as their strategies of symbolic production follow different logics. Thus, Derrida is superseded by Lacan in his function as Tel Quel’s ‘house philosopher’ in the about-face of the early 1970s, which led Tel Quel first to the PCF and then to Maoism (Kauppi 1990: 111[142ff.]). With this spectacular turnaround, Sollers turned to more literary questions under the auspices of the philosophies of desire. No matter who took the initiative to end their symbiotic relationship, even at the end of their game both Sollers and Derrida respect their roles: Derrida as a producer whose symbolic capital is based on a long-term theoretical project; Sollers as a reproducer is constantly trying to reinvent himself by embodying at every moment the prevailing vanguard model in the field.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


Sollers’ mode of distinguishing himself in the field with an avant-garde logic of leap-frogging his own (conceptual) shadow is reflected textually in the way in which all conceptual contents accepted by the locutor are placed at the end of an implied time bar, while rejected contents are placed at the beginning. The difference to the conceptual work in the discourse of production is conspicuous. The locutor is no longer a craftsman producing new concepts and forging new theories from the inchoate tumult of intellectual discourse. He is a trader hawking the fin-ished products he has acquired from the craftsman. The craftsman’s and the trader’s activities have little in common. If the texts of intellectual discourse are produced by the first, they are sold by the latter (of course the two can unite in the same person and thus not seldom create the highest possible impact in the field). How does the trader get his goods to his customers? By telling them a story—one which requires a specific analytical approach to do it justice.

To wind up this analysis, I would like to draw from narrative theory to illustrate how specific narrative structures allow intellectual discourse to be made repeatable and thus. After breaking down the manifold points of view and sources of enunciation, I will now ask how the anarchic tumult of voices of the producer’s discourse is seized and ‘packaged’ in the discourse of reproduction. While turning to the question how intel-lectual discourse is received and popularized, we leave the framework of enunciation theory and ask how the protean and ramified knowledge mobilized by the producer is channeled and fixed. From the surplus of possible interpretations and perspectives, there emerges a narra-tively measured terrain of more or less stable socio-symbolic relations. Individuals who may have lost their way in the polyphonic whirlwind of intellectual discourse suddenly feel as if their feet had alighted on firm ground again. On islets left in the midst of the calming waves, they begin to settle down in what they take to be durable living conditions. As ‘actors’ they can find their way around this symbolically consoli-dated terrain and found a ‘society.’

A systematic introduction to narrative theory as developed by Frederic Jameson (1981) following A.J. Greimas (1966), cannot be given here (cf. Angermuller 2003, 2006). Like other structural sign theories, the narrative approach considers meaning—to put it briefly—as the product of semiotic operations. The value of an element is defined by its position in relation to all other elements in the system. Greimas’ achievement lies in formulating the smallest possible system generating meaning: the semiotic square, whose four constitutive elements are generated by the three fundamental operations of contrariety, contradiction, and impli-cation. Accordingly, Jameson uses the semiotic square to identify the


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


vital elements of a narrative and to decode the historical temporality they represent.

Such a semiotic approach undoubtedly marks a break with the enunciative-pragmatic background of the previous analyses—a break which can be justified by positing that we are now dealing with a ‘trader text’ and no longer with a ‘craftsman text.’ As a trader, the locutor takes the text as abstract material which he inventories, orders, and classifies before offering it on the market of symbolic goods. The trader auctions off concepts, terms and names whose contents he does not necessarily know. To attain the highest possible return, he or she must create from the source material a compact parcel which can be easily dispatched and transported. His or her narrative turns the discourse of production into a package in which all elements are well fixed and ready for sale.

Sollers’ narrative could hardly be more simple, as it knows only one critical event, only one turning point deciding on who is counted among the traditional rearguard and who among the revolutionary van-guard. The selected text example contains the following elements and protagonists of the vanguard, namely the terms or names text, writing, intertextuality, Lautréamont, Mallarmé, Marx, Freud and the proper names of the corresponding producers: ‘Derrida,’ ‘Barthes,’ ‘Kristeva,’ ‘Blanchot,’ ‘Althusser,’ ‘Lacan.’ These elements are qualified as vanguard through their integration into a wider narrative context in which they face a

Source: Philippe Sollers (1968a): ‘Écriture et révolution (entretien avec Jacques Henric).’ In: Tel Quel (ed.), Théorie d’ensemble. Paris: Seuil, p. 69, originally in Lettres françaises, April 1967; translation J.A.

D’emblée, en mettant l’accent sur le texte, sur ses déterminations historiques et son mode de valori-sation métaphysique des concepts « d’œuvre » et « d’auteur »; en met-tant en cause l’expressivité subjec-tive ou soi-disant objective, nous avons touché les centres nerveux de l’inconscient social dans lequel nous vivons et, en somme, la distri-bution de la propriété symbolique. Par rapport à la « littérature », ce que nous proposons veut être aussi subversif que la critique faite par Marx de l’économie classique.

By focusing on the text, its histori-cal conditions and its metaphysical mode of valorizing the terms ‘work’ and ‘author,’ by questioning the subjective or reputedly objective expressivity, we have touched the nerve center of the social uncon-scious in which we live and, in summary, the distribution of sym-bolic property. In regard to ‘litera-ture’ our suggestion wants to be as subversive as Marx’s criticism of classical economy.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


rearguard. Some of the rearguard’s elements—work, author, subjective expressivity—can be found in the following excerpt of the co-text. With these elements, the narrative structure of Sollers’ discourse, of which the first excerpt above revealed only one half, can be completed.

Vanguard and rearguard are easy to distinguish: the former consists of the discursive elements accepted by the locutor, the latter contains the rejected elements. Each vanguard element has its rearguard counterpart. This way the discourse generates a never ending flow of binary oppositions: text vs. work, intertextuality vs. author, écriture vs. subjective expressivity or parole … By utilizing the theory of the semiotic square, the historical temporality of this discourse can now be described as an effect of the semiotic operations which produce the discourse of vanguard versus rearguard. This discourse operates with (at least) two contrary forces, namely with a ‘theoretical’ contrariety building upon Sollers’ distinction between ‘signifier’ and ‘signified,’ and the well-known ‘political’ contrariety between ‘repression’ and ‘libera-tion.’ Since these four constitutive components also relate in terms of implication (arrows) and contradiction (no arrows), they form a square from whose elements one can obtain a revolutionary narrative in which rearguard and vanguard occupy their precise positions on the right and on the left (Figure 4.1).

‘work’ signifier signified

‘author’

‘parole’

‘literature’ repression liberation

‘idealism’ ‘Marx’ (Althusser)

‘humanism’ ‘Freud’ (Lacan)

… …

‘text’ (Derrida)

‘intertextuality’ (Kristeva)

‘Lautréamont...’ (Blanchot)

‘writing’ (Barthes)

Figure 4.1 The narrative structure of the vanguard/rearguard discourse

The rear- or vanguard chains are created by the contraction between the two left and right corners respectively of the square, which constitute the start and end points of the (dotted) historical time bar. Thus, the ele-ments attested in the textual extract (see box) can be formally expanded so as to occupy the four positions of the square which are played out


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


throughout the text. While more vanguard and rearguard elements can be generated, the plurality of elements can be reduced to a simple, stable structure. Moreover, certain elements can be exchanged over time with-out putting at risk the underlying logic. In 1967, the Communist Party appeared in the right chain before being integrated into the left chain in 1971. And the principle of leap-frogging one’s own shadow remained intact since, at the end of the 1960s, the Nouveau Roman and, at the beginning of the 1970s, Derrida disappeared from the right chain. Only at the end of the 1970s, with the turn to the post-Maoist nouveaux philos-ophes, was the avant-garde strategy exhausted—a strategy which drew its symbolic energy from the permanent crossing of academic and aesthetic products, of theory and literature. The advent of audiovisual mass media redefined the rules of the intellectual game. It was a clear sign of this development when Sollers discontinued publication of the journal in the late 1970s in order to devote himself exclusively to literature again.

Yet this change reflected not only the new political conjuncture of the late 1970s. In order to assess what it means for Sollers to engage in a vanguard narrative production, his biographical situation around 1970 must also be taken into consideration. Still too young and without the necessary academic credentials to appear on equal footing with other representatives of the intellectual generation, Sollers makes a virtue out of necessity by telling a story in which he equips the producers he cites with certain roles. In his story, all narrative elements and actants are defined precisely in respect to past and present. The narrative structure of Sollers’ discourse is closed. Its closure makes the text easy to con-sume, reduces the interpretive effort necessary for its comprehension and fixes the dynamic flow of the intellectual discourse unfolding in the texts of the theory producers. Thus, Sollers’ display window reunites the creative products of the ‘structuralist’ generation and makes them readable. Sollers’ accomplishment is to exhibit them in an attractively decorated collection and thus to convince the customers of the latest trend the Parisian fashion scene has to offer. Through eloquent and competent guidance, he makes his customers acquainted with the vari-ous philosophies of the represented couturiers. Indeed, the customer has little difficulty in distinguishing between the exhibited creations which are each presented with a clear and profiled program. Through a clever use of lighting, the differences come advantageously to the fore and relevant details are accentuated. Impressed by the consumer-friendly environment, the customer finally decides on his costume without entertaining any doubts that it follows a sophisticated, systematic and perfectly composed pattern.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


If only he or she were to glance into the workshop where his costume was produced. There he would not only come across conditions con-trasting starkly with the sparkling polished world of Sollers’ display window; he would also become aware of the work the producers had to invest into their costume-texts. Instead of an aseptically ordered high-tech laboratory, he would see a sweatshop where everything flies around in wild chaos, with the producer in the center, crafting his product out of whatever he gets hold of. The final product cannot hide its characteristic impurity. It is not a self-contained text, not a pure semiotic structure, but an unstable discursive patchwork in which textual fragments are sewn together with extracts from the contexts and numerous references to other producers. Furthermore, instead of a workroom complying with all security regulations, the visitor finds a producer who has to deal with the perfidies of his production material. The producer is continuously getting entangled in the various threads that are supposed to hold the costume together. More often than not, he sews himself up and has to re-adjust his work in the end. And when-ever he pricks his finger, he involuntarily winces and loses the thread. In the end, the visitor sees the more or less unfinished results leaving the workplace. Instead of the shiny polished products ready-made for display in a chic boutique, there are torn pockets here, loose hems there. Most parts still bunch and sag, and despite all the effort the tailor has put into quality control, it seems that with every hole he darns, a new one opens up. Only when he gets home, unwraps the narrative packag-ing and examines the material more closely, does the customer realize that he has acquired an impure, unfinished product, created under rather shady conditions. Does he then notice the work invested in his product? He will, but only by examining the traces of the constraints, tension and labor scattered throughout the material, which testify to a troublesome origin and leave their unique imprint on the text.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

140

By placing the problem of the subject in the center of their epistemological investigations, Lacan, Althusser, Foucault, Derrida and Sollers made themselves the representatives of an anti-humanist moment in French intellectual life. Thus, in his ‘return to Freud,’ Lacan elaborates upon a split subject entangled in the pitfalls of the symbolic. For Althusser, taking up arms with Marx against humanism, individuals are hailed and recruited as the subjects of ideology. Foucault shows the historical limits of the figure of ‘Man,’ which he perceives as the product of a system of what can be said and thought. And by uncovering the aporias of living presence, Derrida formulates a critique of the idealist tradition, from Plato to Husserl.

In analyzing a few textual excerpts from these theorists, I have made recourse to enunciative-pragmatic tools of analysis which reflect on the position of the subject in language. As a ‘poststructuralist’ approach to the discourse of these structuralist theorists, enunciative pragmatics does not renounce critique of the free and autonomous subject envisaged as center and origin of the production of meaning. Yet at the same time, it critically interrogates structuralist models which perceive meaning as the simple application of a set of grammatical rules or the result of a combination and selection of elements. From an enunciative-pragmatic viewpoint, there are no given subject positions in the symbolic; the subject is not the property of a semiotic code; nor is it an effect of a textual play of differences. Rather, the discursive positions of subjects are creatively and dynamically constructed by readers cooperating with texts in context. As texts are contextually underspecified, they need a practical instance—the reader—whose interpretive capacities and con-textual knowledge need to be mobilized in the production of meaning (see Angermuller 2013).

5Conclusion: The Subject of Discourse


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Conclusion 141

While the reader can read a text in many different ways, the text is not contextualized in pell-mell fashion. In order to comprehend its meaning, the reader is effectively subject to certain linguistic con-straints. In written discourse, there can be no immediate access to the enunciation itself, to the act of speaking, writing or reading. What is accessible to the analyst, as well as to any other reader, are the formal traces of enunciation, the enunciative markers instructing the reader about how the discourse was uttered. For it is thanks to these formal markers of enunciation—for instance I, not, quotation marks—that the text defines the interpretive limits within which the reader may engage in the search for whom it is that is speaking in the discourse.

Indeed, this question requires that individuals participating in the intellectual debate develop some knowledge about the subjects of intel-lectual discourse. And this is precisely what she or he does by following the formal instructions of the text. His or her mission is to deal with a discourse whose heterogeneity must be taken into account, a discourse which mobilizes a host of voices and figures, a discourse in which nei-ther discourse nor its subjects constitute a transparent and harmonious whole.

Such a discourse can be described as having several ‘floors’ which the reader must traverse, from bottom to top, if he or she wants to know who says what and to learn who takes what position in the intellectual field. On the first floor, the reader encounters a great number of voices, speakers or ‘enunciators’ who populate the utterances of discourse. Orchestrated by the text’s formal markers of enunciation, every voice says something. Yet, as there is always an excess of voices, the reader will need to reduce the text’s polyphonic complexity. Therefore, on the second floor, the reader will discover that behind the multitude of voices on the floor below there is a limited number of discursive subject positions, such as ‘author’ (locutor), other (allocutor) and other instances, to which the voices can be attributed. Here, the reader real-izes that the hive of voices on the floor below is not without a certain logic. These voices have been orchestrated by the locutor in his or her role as the stage director of the discursive spectacle. It is by means of enunciative markers that the locutor signals which speakers he or she accepts and which ones he or she rejects in the discourse. But if, still on the second floor, the reader discovers that the numerous speakers of discourse are staged in certain ways by the locutor, the reader’s mission has not yet ended. He or she must ascend further, to the third floor, where the task is to associate the locutor and allocutor with various social actors, to attribute them names and give them an institutional


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


address in the discursive space, which thus finally allows him or her to match what is being said to historical beings with interests, motivations and intentions that occupy a certain place in the social (Figure 5.1).

Figure 5.1 A three-floor model of discursive subjectivity

Third floor: naming the subject positions in the social

L = for example ‘John Miller’, Professor of X at Y and so on; A = ?

Second floor: attributing the voices to the locutor and other instances

L = l2, l3, l5; A = a1, a4

First floor: a multitude of subpersonal voices

x1, x2, x3, x4, x5

↑

↑

Such is the floor plan of the analyses carried out in the preceding chapters. In order to account for the social space in which intellectuals position themselves, the reader must ascend as it were the floors of a het-erogeneous and polyphonic discourse—from the many voices of subject positions to named actors in the social, which demands a considerable effort of interpretation and contextualization (cf. Angermüller 2012). In contrast to approaches which consider only the third floor—that is to say: the actors, their ideas, objectives and institutional places—my analyses have concentrated on the ‘street level,’ where the reader deals with texts in their refractory opacity. Unlike theories of the actor as a transparent and unitary source of meaning, discourse turns out to be a heterogeneous phenomenon occupying several levels, revealing in this way the divisions of a subject inscribed in the symbolic and enmeshed in the snares of discourse.

Considerable interpretive labor needs to be invested if the reader is to come to terms with the texts’ interpretive problems. Depending on the circumstances and their background, different readers will solve them in different ways. Academic ‘insiders’ are likely to have less difficulty climbing up the three floors of intellectual discourse than non-academic ‘outsiders.’ As experts trained in dealing with interpretive complexity, the academic reader can rely on a developed discursive competence in how to contextualize a text. In schools and universities, they have learned how to produce legitimate representations of who speaks in discourse. And as the members of an academic community, they can


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Conclusion 143

draw from large stocks of specialized knowledge about the social and historical contexts in which a text originates. These academic experts are easily thought of as dealing with purely conceptual knowledge. And yet they cannot understand a concept without mobilizing their non-conceptual knowledge about who speaks. Indeed, in order to enter the abstract play of academic and intellectual concepts they have to have a certain knowledge about the social and historical background, without which a theoretical text cannot make sense. According to the three-floor model of polyphonic discourse, even in academic communities where the question of the author and his or her context is considered illegitimate, specialized academic readers need to prove their practical interpretive capacity to construct positions of the subject in the social. Therefore, the reader will not achieve the most basic understanding of the conceptual ‘content’ without accounting for the dialogic organization of theoretical discourse. Even the most abstract and ‘objective’ text would remain nothing but white noise if the reader was unable to distinguish between what is said by one person and what is said by another.

It is in this sense that interpretation must be seen as a profoundly social activity. In the act of reading and writing, the reader links a text with its context of enunciation. Unlike models that hold that context is ‘added’ to the text in discursive practice, enunciative-pragmatic approaches to discourse conceive of context as a dynamic and creative construction by the reader as she or he reads along. There is no text with a surface of immanent meaning on the one side to be placed in a context as a constituted socio-historical totality on the other. The text needs the reader to search at all costs for whatever context is necessary to develop an understanding of what it means. Therefore, context is neither stable nor a given; it points to the dynamic knowledge that the reader builds up about the wider social and historical context in which the text originated. The question, therefore, is not what the context of the text is, the question is how the reader is instructed by the text’s formal markers about the context(s) in which it can make sense.

As a consequence, just as the subjects need to be constructed through a reduction of discourse as a polyphonic multitude, the social context must be seen as the result of the interpretive labor of the discourse participants. Thus, for a discursive theory of the social, society is not a horizon that is given once and for all before the discourse participants engage in their interpretive practices. Rather, society is constituted by transforming the social, as the open terrain of the polyphonic multitude, into a structured territory of social actors. In the act of reading or writing, the discourse participants do not deal with society


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26


as a constituted structure or order; they deal with the social as an unstructured multiplicity. Meaning, in other words, is not something out there, a reality to be read from the textual surface, to be grasped by everybody; it needs to be revealed by discursive practices. The social order is constructed by a diversity of practices of which only the ‘textual’ ones have been investigated here, while many others need to be addressed elsewhere. Therefore, the limitations resulting from the analysis of written material notwithstanding, we need to ask how the social order is instituted in discourse. Involving ‘textual’ as well as ‘non-textual’ practices, it is discourse that allows the social to be represented as an ordered, structured and constituted society.

While this book has dissected the complex polyphonic organization of the discourse of Theory, it has also aimed to show that even discourse as abstract, universal and conceptual as theoretical texts cannot but put into practice its social dimension, since intellectuals, theorists and researchers, too, need to speak, to engage in a dialogue and to commu-nicate with others. Thus, understanding a text demands that the reader provide an answer to the question: ‘Who is speaking?’ Sometimes, the answer to this question can be given as if by itself, the source and intention of the discourse seeming to be self-evident. But sometimes the answer to ‘Who says what, in the name of whom and against whom, in what circumstances, toward what goal?’ is not so easily found. In the intellectual game, the interpretation of what texts are saying often demands considerable skill and competence on the part of the reader. There are numerous formal traces that the reader can take into account; numerous, too, are the possible contextualizations; and even more numerous are the ramifications of the knowledge being mobilized. Indeed the interpretive possibilities open to a reader are so numerous that identifying which positions certain intellectual producers occupy in relation to other producers turns out to be anything but a banal ques-tion. This is why those who consider what intellectuals say merely as a given, omit perhaps the most interesting question of all: how do we know who wishes to say what?

If the entry into discourse involves interminable work, beset by failures and accidents, this also holds true for the reading individual, before whose eyes pass cascades of letters. This holds even truer for the writing individual who sets letters to paper. Again and again, the individual says more than they mean to say. He or she is always caught up by what he or she does not say. And constantly they become entan-gled in snares of the written word. The writing individual is a tragic figure who does not come to terms with him or herself and remains


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Conclusion 145

incomprehensible to his or herself; for whenever they act symbolically, a split needs to be sutured between their symbolic and non-symbolic being: between speakers, locutors and voices of discourse on the one hand, and the physical individual, the reading and writing agent, with their cognitive and interpretive capacity, on the other. The voices and speakers are beings who populate the utterances of discourse, whereas the reading individual is confronted with the practical task of under-standing and interpreting what is said. The individual is an active, creative and imaginative being who searches for the context in which symbolic material is enunciated. But in order to act symbolically, they depend on the speakers of the discourse embedded in the symbolic material. Like a puppeteer who must let his puppets say everything he wants to express, the living (but mute) individual becomes entangled in the strings with which he tries to control its dead (but speaking) figures. To the critics of humanism who proclaim the ‘death of man,’ one can respond thus: Ah yes, individuals do exist, and as reader-philosophers they are alive and kicking, experts in the construction of meaning. However, they have to make do (and do their best) with the speakers and figures of a dead symbolic order without time or origin.

If even individuals belonging, like the intellectuals, to the class of professional, full-time producers of texts, time and again stumble over their own speakers; if their words, once they have been put down on paper, confront them as opaque material; if their control over what is said repeatedly runs away with them—then how can these experts of the written and spoken word come to terms with the dilemma of their symbolic existence if not by continuing to speak, to write further texts, to produce more discourse? Yet the prospect that one day everything will have been said, and saying itself becomes superfluous—this is a day no intellectual must dread, neither those who write nor those who read.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

146

1 Introduction: Poststructuralism and Enunciative Pragmatics

1. « cette mise en fonctionnement de la langue par un acte individuel d’utilisation » (Benveniste 1974: 80).

2. « L’analyse énonciative est donc une analyse historique » (Foucault 1969: 143).

2 A History of Discourse Analysis in France

1. « La question que pose l’analyse de la langue, à propos d’un fait de discours quelconque, est toujours: selon quelles règles tel énoncé a-t-il été construit, et par conséquent selon quelles règles d’autres énoncés semblables pourraient-ils être construits ? La description des événements du discours pose une tout autre question: comment se fait-il que tel énoncé soit apparu et nul autre à sa place ? […] il s’agit de saisir l’énoncé dans l’étroitesse et la singularité de son événement; de déterminer les conditions de son existence, d’en fixer au plus juste les limites, d’établir ses corrélations aux autres énoncés qui peuvent lui être liés, de montrer quelles autres formes d’énonciation il exclut. » (Foucault 1969: 39ff.)

2. « l’instance de l’événement énonciatif » (Foucault 1969: 41)3. « l’énoncé est bien l’unité élémentaire du discours. » (Foucault 1969: 107)4. « Un énoncé existe en dehors de toute possibilité de réapparaître; et le rapport

qu’il entretient avec ce qu’il énonce n’est pas identique à un ensemble de règles d’utilisation. Il s’agit d’un rapport singulier: et si dans ces conditions une formulation identique réapparaît – ce sont bien substantiellement les mêmes noms, c’est au total la même phrase, mais ce n’est pas forcément le même énoncé. » (Foucault 1969: 118)

5. « sur quel mode elles existent, ce que c’est pour elles d’avoir été manifestées, d’avoir laissé des traces et peut-être de demeurer là, pour une réutilisation éventuelle; ce que c’est pour elles d’être apparues – et nulle autre à leur place. » (Foucault 1969: 143)

6. « singularité », « matérialité répétable » (Foucault 1969: 133, 134)7. « Le couple énoncé/énonciation fonctionne différemment dans l’Archéologie

et dans la tradition linguistique que reprend l’AD: si la notion d’énonciation utilisée par Foucault est proche de celle dont se sert l’AD (activité de produc-tion d’un discours par un sujet énonciateur dans une situation d’énonciation), l’énoncé se trouve par contre lié à la notion de répétition. » (Courtine 1981: 45)

8. « un être linguistique abstrait » (Ducrot 1984: 95); « une invention de cette science particulière qu’est la grammaire. […] Ce que le linguiste peut prendre pour observable, c’est l’énoncé, considéré comme la manifestation particulière, comme l’occurrence hic et nunc d’une phrase. » (Ducrot 1984: 95)

Notes


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Notes 147

9. « Trop répétable pour être entièrement solidaire des coordonnées spatio- temporelles de sa naissance […], trop lié à ce qui l’entoure et le sup-porte pour être aussi libre qu’une pure forme […], il est doté d’une certaine lourdeur modifiable, d’un poids relatif au champ dans lequel il est placé, d’une constance qui permet des utilisations diverses, d’une permanence temporelle qui n’a pas l’inertie d’une simple trace. » (Foucault 1969: 138)

10. « L’analyse énonciative est donc une analyse historique, mais qui se tient hors de toute interprétation: aux choses dites, elle ne demande pas ce qu’elles cachent, ce qui s’était dit en elles et malgré elles le non-dit qu’elles recouvrent, le foisonnement de pensées, d’images ou de fantasmes qui les habitent; mais au contraire sur quel mode elles existent, ce que c’est pour elles d’avoir été manifestées, d’avoir laissé des traces et peut-être de demeurer là, pour une réutilisation éventuelle; ce que c’est pour elles d’être apparues – et nulle autre à leur place. » (Foucault 1969: 143)

11. « représente le rapport imaginaire des individus à leurs conditions réelles d’existence » (Althusser 1995: 216)

12. « l’existence “d’idées” sans support réel et matériel » (Althusser 1995: 108)13. « Nous suggérons alors que l’idéologie “agit” ou “fonctionne” de telle sorte

qu’elle “recrute” des sujets parmi les individus (elle les recrute tous), ou “transforme” les individus en sujets (elle les transforme tous) par cette opéra-tion très précise que nous appelons l’interpellation, qu’on peut se représenter sur le type même de la plus banale interpellation policière (ou non) de tous les jours: “hé, vous là-bas.” » (Althusser 1995: 226)

14. « une ou plusieurs formations discursives interreliées, qui déterminent ce qui peut et doit être dit (articulé sous la forme d’une harangue, d’un sermon, d’un pamphlet, d’un exposé, d’un programme etc.) à partir d’une position don-née dans une conjoncture donnée: le point essentiel ici est qu’il ne s’agit pas seulement de la nature des mots employés, mais aussi (et surtout) des constructions dans lesquelles ces mots se combinent, dans la mesure où elles déterminent la signification que prennent ces mots: comme nous l’indiquions en com-mençant, les mots changent de sens selon les positions tenues par ceux qui les emploient; on peut préciser maintenant: les mots “changent de sens” en passant d’une formation discursive à une autre. » (Haroche et al. 1971: 201ff.)

15. « un champ de régularité pour diverses positions de subjectivité » (Foucault 1969: 74)

16. « théorie non subjectiviste du sujet » (Pêcheux 1975: 122)17. « il est question du sujet comme procès (de représentation) intérieur au

non-sujet que constitue le réseau des signifiants, au sens que lui donne J. Lacan: le sujet est “pris” dans ce réseau – “noms communs” et “noms propres”, effets de “shifting”, constructions syntaxiques, etc. – de sorte qu’il en résulte comme “cause de soi”, au sens spinoziste de l’expression. » (Pêcheux 1975: 141)

18. « il y a le procès de l’interpellation-identification qui produit le sujet à la place laissée vide. » (Pêcheux 1975: 143)

19. « La psychanalyse traite donc le sujet comme un effet. Plus précisément, le sujet dont elle fait sa matière première est effet du langage. C’est en fin de compte cette mise en place du sujet par rapport au langage qui met la psychanalyse en position de rompre avec l’idéologie de la transparence. En


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

148 Notes

outre poser le sujet comme effet exclut de le tenir pour centre, source, unité d’une intériorité, etc… » (Henry 1977: 21)

20. « Or justement, en liant présupposition et acte de langage, on remet en place un sujet, en tant que source de cet acte, même si l’on considère que ce sujet intériorise un sujet universel qui en règle l’activité ou si l’on fait de la société une instance qui régit juridiquement pour ce sujet les significations. » (Henry 1977: 82)

21. « Chez Foucault, au niveau énonciatif, il n’y a en effet pas de sujet mais une position de sujet susceptible d’être occupée par des individus divers. » (Henry 1977: 84)

22. « Nous considérons ainsi une FD [formation discursive] comme hétérogène à elle-même: la clôture d’une FD est fondamentalement instable, elle ne consiste pas en une limite tracée une fois pour toutes séparant un intérieur et un exté-rieur, mais s’inscrit entre diverses FD comme une frontière qui se déplace en fonction des enjeux de la lutte idéologique. » (Courtine and Marandin 1981: 24)

23. « gommer les aspérités discursives, à suturer les failles qui béent dans tout discours, à raboter d’un côté, à combler et colmater de l’autre; à faire de tout discours un corps plein et une surface plane. » (Courtine and Marandin 1981: 23)

24. « La sémantique, c’est le “sens” résultant de l’enchaînement, de l’appropriation à la circonstance et de l’adaptation des différents signes entre eux. Ça c’est absolument imprévisible. C’est l’ouverture vers le monde. Tandis que la sémiotique, c’est le sens refermé sur lui-même et contenu en quelque sorte en lui-même. » (Benveniste 1974: 21)

25. « Les paradoxes pragmatiques sont des cas de contradiction entre ce que dit un énoncé et ce que montre son énonciation » (Récanati 1979a: 206)

26. « c’est à travers ses non-coïncidences – tout ce qui marque de non-un de la communication: incompréhension, inquiétude, manque, malentendu, ambiguïté, – que “le langage se rappelle” » (Authier-Revuz 1995: iiff.)

27. « la compréhension comme cas particulier du malentendu » (Culioli 2002: 28)28. « cette mise en fonctionnement de la langue par un acte individuel

d’utilisation » (Benveniste 1974: 80)29. « la recherche des procédés linguistiques (shifters, modalisateurs, termes éval-

uatifs, etc.) par lesquels le locuteur imprime sa marque à l’énoncé, s’inscrit dans le message (implicitement ou explicitement) et se situe par rapport à lui (problème de la “distance énonciative”). C’est une tentative de repérage et de description des unités, de quelque nature et de quelque niveau qu’elles soient, qui fonctionnent comme indices de l’inscription dans l’énoncé du sujet d’énonciation. »

30. « une invention de cette science particulière qu’est la grammaire » (Ducrot 1984: 174)

31. « un être linguistique abstrait, identique à lui-même à travers ses diverses occurrences […] occurrence particulière, la réalisation hic et nunc de la phrase » (Ducrot 1984: 95)

32. « [b]ouleverser le langage pour que ses éléments obéissent aux règles aux-quelles ils sont soumis dans les formules du logicien » (Ducrot 1989: 76)

33. « Là l’enchaînement des énoncés a une origine interne, il est fondé sur la nature même de l’énoncé, ou, si l’on préfère, sur son sens [qui] contient une allusion à son éventuelle continuation. » (Ducrot 1980b: 11)


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Notes 149

34. « à qui l’on doit imputer la responsabilité de cet énoncé » (Ducrot 1984: 193)35. « Le locuteur est l’instance première qui produit matériellement les énon-

cés. […] Elle est proférée (ou écrite) par un locuteur (ou scripteur), dotée d’une matérialité, subordonnée à l’expérience sensorielle. L’énonciateur est l’instance qui se positionne par rapport aux objets du discours auxquels il réfère, et, ce faisant, qui les prend en charge. La notion d’énonciateur corre-spond à une position (énonciative) qu’adopte le locuteur, dans son discours, pour envisager les faits, les notions, sous tel ou tel PDV pour son compte ou pour le compte des autres. La discjonction locuteur/énonciateur rend compte des possibilités que le locuteur se donne, en tant qu’énonciateur, pour tourner autour des objets du discours, pour envisager les faits, les mots et les discours, les notions, les situations, les événements, les phénomènes de tel ou tel PDV, dans le présent, le passé ou le futur, par rapport à soi ou par rapport aux PDV d’autrui. » (Rabatel 2010: 370)

36. « Parler de façon ironique, cela revient, pour un locuteur L, à présenter l’énonciation comme exprimant la position d’un énonciateur E, position dont on sait par ailleurs que le locuteur L n’en prend pas la responsabilité et, bien plus, qu’il la tient pour absurde. » (Ducrot 1984: 211)

37. « c’est que cette incitation à agir ou cette obligation de répondre sont don-nées comme des effets de l’énonciation. » (Ducrot 1984: 174)

38. « Le concept d’énonciation dont je vais me servir n’a rien de psychologique, il n’implique pas que l’énoncé est produit par un sujet parlant. » (Ducrot 1989: 76)

39. « je pense que l’on peut intégrer un très grand nombre de phénomènes énonciatifs à la structure linguistique elle-même. Il s’agit d’un élargissement du structuralisme, qui respecte l’idée fondamentale de ce structuralisme. » (Ducrot 1992: 64ff.)

40. « Le discours rapporté est la mise en rapport de discours dont l’un crée un espace énonciatif particulier tandis que l’autre est mis à distance et attribué à une autre source, de manière univoque ou non. » (Rosier 1999: 125)

41. « la syntaxe et la sémantique étudient le langage en tant que tel, c’est-à-dire en tant que constitué par un système de règles ou de conventions, alors que la pragmatique l’étudie d’un point de vue en quelque sorte extérieur: elle étudie non le langage lui-même, mais l’usage qui en est fait. » (Récanati 1979b: 8)

42. « La pragmatique étudie la parole ou, mieux, le discours, mais elle n’étudie pas ce qu’il y a de particulier et d’individuel dans l’utilisation du langage. » (Récanati 1981: 20)

43. « chaque occurrence du mot “je” réfléchit le fait de sa propre énonciation » (Récanati 1979a: 9)

44. « des cas de contradiction entre ce que dit un énoncé et ce que montre son énonciation » (Récanati 1979a: 206)

45. « Il se suffit à lui-même et l’interprétation d’une phrase consiste à la décoder, c’est-à-dire à utiliser le code constitué par la langue dans laquelle elle est exprimée, pour restituer le message. » (Reboul and Moeschler 1998a: 17ff.)

46. « une unité linguistique supérieure à la phrase » (Reboul and Moeschler 1998b: 8).

47. « Le discours ne fait qu’un avec la manière dont il gère sa propre émergence, l’événement de parole qu’il institue; il représente un monde dont son énon-ciation est partie prenante. » (Maingueneau 1995: 40)


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

150 Notes

48. « seul un discours qui se constitue en thématisant sa propre constitution peut jouer un rôle constituant à l’égard d’autres discours. On peut étudier cette constitution selon trois dimensions:

– La constitution comme action d’établir légalement, comme processus par lequel le discours s’instaure en construisant sa propre émergence dans l’interdiscours;

– Les modes d’organisation, de cohésion discursive, la constitution au sens d’un agencement d’éléments formant une totalité textuelle;

– La constitution au sens juridico-politique, l’établissement d’un discours qui serve de norme et de garant aux comportements d’une collectivité. » (Maingueneau and Cossutta 1995: 113)

49. « c’est à travers ses non-coïncidences – tout ce qui marque le non-un de la communication: incompréhension, inquiétude, manque, malentendu, ambiguïté, – que “le langage se rappelle”, dans sa réalité, à l’énonciateur; et dès lors que, comme dans les énoncés évoqués, l’énonciateur “rappelle le langage”, son énonciation apparaît, en ces points, comme creusée par l’écart d’une distance – celle de l’observateur à l’objet observé – établissant l’énonciateur, qui ne fait plus “un” avec ses mots, dans une non-coïncidence aux mots qu’il énonce. » (Authier-Revuz 1995: iif.)

50. « Un marqueur, si vous voulez, c’est une espèce de résumé, de concentré de procédures qui déclenchent et activent des représentations. » (Culioli 2002: 172)

51. « Le langage est un système, mais un système ouvert. » (Culioli 1999: 48)52. « Un texte n’est pas un représentant stable d’une réalité prédécoupée, stable

pour tous les locuteurs – Quand on a un énoncé ou une suite textuelle, on a affaire à un agencement de marqueurs. […] Les marqueurs sont des représentants de représentations. » (Culioli 1985: 16)

53. « l’activité de langage, le niveau des opérations, qui n’est pas directement accessible; les énoncés, c’est-à-dire la matérialité du texte, qui est directe-ment accessible; l’activité métalinguistique, le travail du linguiste qui va simuler le niveau des opérations. » (Culioli 2002: 185)

54. « on “plonge” donc, un objet en voie de constitution dans un système de références, un système de repérage avec des coordonnées espace-temps et intersubjectives. » (Culioli 2002: 36)

55. « énoncer, c’est construire un espace, orienter, déterminer, établir un réseau de valeurs référentielles, bref un système de repérage. Tout énoncé est repéré par rapport à une situation d’énonciation, qui est définie par rapport à un sujet énonciateur S0 (ou pour être exact, un premier sujet énonciateur), à un temps d’énonciation T0 pour ne considérer que ces deux repères. » (Culioli 1999: 49)

56. « Avec la découverte de l’énoncé, apparaissait le problème des relations inter-sujets et le problème fondamental de la non-symétrie entre production et reconnaissance. » (Culioli 1999: 11)

57. « Le sens, c’est d’abord de déclencher chez autrui une représentation. » (Culioli 2002: 32)

58. « ça n’est pas symétrique entre deux sujets, ça va pouvoir nous donner l’illusion de la symétrisation, puisque nous avons parfois l’illusion d’avoir été parfaitement compris, l’illusion donc de la transparence, de la “commu-nication”. » (Culioli 2002: 28)


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Notes 151

59. « vivre avec des ratés permanents, des approximations, et, malgré tout, à nous tirer plus ou moins d’affaire. » (Culioli 2002: 221)

60. « on arrive à conquérir la clarté. » (Culioli 1985: 2)61. « la compréhension comme cas particulier du malentendu » (Culioli 2002: 28)62. « une espèce d’optimisation des ratages » (Culioli 2002: 189)63. « au fur et à mesure que nous parlons, le langage sert à nous réguler, à

nous réguler par rapport à autrui et par rapport à nous-mêmes. » (Culioli 2002: 196)

64. « fonde sur cet ajustement plus ou moins réussi, plus ou moins souhaité, des systèmes de repérages des deux énonciateurs » (Culioli 1999: 48)

65. « positivisme heureux » (Foucault 1971: 72)66. « Le langage est ainsi organisé qu’il permet à chaque locuteur de s’approprier

la langue entière en se désignant comme je. » (Benveniste 1966: 262)67. « dans lequel l’énonciateur s’avoue explicitement (“je trouve ça moche”)

ou se pose implicitement (“c’est moche”) comme la source évaluative de l’assertation. » (1980: 71)

68. « qui s’efforce de gommer toute trace de l’existence d’un énonciateur indivi-duel » (1980: 71)

69. « impliquent un engagement affectif de l’énonciateur » (1980: 81)70. « l’usage d’un adjectif évaluatif est relatif à l’idée que le locuteur se fait de la

norme d’évaluation pour une catégorie d’objets donnée » (1980: 86)71. « leur usage varie […] avec la nature particulière du sujet d’énonciation dont

ils reflètent la compétence idéologique, […2)] ils manifestent de la part de L une prise de position en faveur, ou à l’encontre, de l’objet donné. » (1980: 91)

72. « sont des entités sémantiques porteuses d’une source qui est dite avoir le pdv », « sont des entités sémantiques susceptibles de saturer les sources, les liens énonciatifs relient les ê-d aux pdv » (Nølke et al. 2004: 30)

73. « Le pdv simple prend la forme d’une prédication comme par exemple Pierre se promène ou il fait beau. Il est constitué d’un contenu sémantique et d’un jugement porté sur ce contenu, par défaut ‘il est vrai que’. La source du pdv reste indéterminée. Chaque énoncé contient au moins un pdv simple dont le contenu sémantique est posé. » (Nølke et al. 2004: 33)

74. « chaque énoncé constitue un drame. L’auteur du drame, c’est LOC. C’est lui qui construit le jeu polyphonique, mais il n’y participe pas (directement) lui-même. Les acteurs du drame sont les ê-d. LOC crée leurs rôles et il peut créer des rôles pour des images de lui-même; tout comme il peut créer des rôles pour d’autres personnages – notamment l’allocutaire – qui sont présents dans le monde dont fait partie le théâtre. » (Nølke et al. 2004: 55)

75. « En tant que telle, l’énoncé renferme donc des indications concernant les pro-tagonistes, la situation énonciative, etc. Ces informations sont à la disposition de l’interprète lorsqu’il applique les stratégies interprétatives, dont le principe le plus important sera “Chercher à saturer, dans la mesure du possible, toutes les variables véhiculées par la signification!” » (Nølke et al. 2004: 24)

76. « il faut poser la question: “qui en est responsable?” » (Nølke et al. 2004: 56)77. « on constate que les mots peuvent changer de sens selon les positions

tenues par ceux qui les emploient. » (Haroche et al. 1971: 97)78. « Ce qui est important ici est que l’énoncé nominalisé est préconstruit, c’est-

à-dire qu’il n’est pas pris en charge par le sujet énonciateur, mais se trouve


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

152 Notes

comme un objet du monde “déjà là”, préexistant au discours […]: ses conditions de production ont été effacées. » (Seriot 1985: 248)

79. « Dans ces énoncés sans sujet ni cause, l’énonciateur n’est qu’un témoin: il “voit” des objets, des processus dont il n’est pas l’Agent, son dire s’efface derrière son voir. » (Seriot 1985: 259)

80. « il [l’énoncé] ne fait que mettre en relation deux préconstruits, que constater une relation entre des “objets du monde” qui sont en réalité des objets préas-sertés dans un ailleurs du discours. » (Seriot 1985: 254)

3 A Methodology of Discourse Analysis

1. Compare Bachelard’s plea for an epistemology of the break (also Cavaillès 1976; Canguilhem 1977). Bachelard’s opponents are the ‘gradualists of cul-ture’ (1971: 185), who emphasize the gradual emergence of culture from a shared historical meaning (sens commun) and its various ‘influences’ (1971: 187ff.). Bachelard insists on the ‘pitfalls of language’ (1971: 192), which make the possibility of comprehensively understanding the historical artefact (‘from the within’) seem limited. Thus, the position of the scientific analyst is defined by a number of breaks ‘between the problems then and the problems of the present,’ between the ‘everyday and the scientific language’ as well as in the language of science itself, which he sees in a state of ‘permanent semantic revolution’ (1971: 192).

2. « continuistes de la culture » (1971: 185); « les pièges du langage »; « Entre les difficultés de jadis et les difficultés du présent, il y a une totale discontinuité »; « Le langage de la science est en état de révolution sémantique permanente »; « Une constante transposition du langage rompt alors la continuité de la pen-sée commune et de la pensée scientifique » (1971: 192)

3. The field of academic discourse is still a recent one (for instance Allori 1998; Hyland 2006).

4. « le dispositif d’énonciation qui lie une organisation textuelle et un lieu social déterminés » (Maingueneau 1996: 6)

4 Analyzing Intellectual Discourse: Variations on the Critique of Humanism

1. « une véritable laboratoire de recherche. » (Roudinesco 1993: 343)2. ScaPoLine does not deal with modal verbs. Yet if certain adverbs (like peut-être

[maybe]) can carry a point of view, the same should be true for modal verbs like can.

3. « Je suis tout au plus l’enfant de chœur du structuralisme. » (Foucault 1994c: 581)

4. « entre le représenté et le représentant en général, le signifié et le signifiant, la présence simple et sa reproduction, la présentation comme Vorstellung et la re-présentation comme Vergegenwärtigung. » (Derrida 1967c: 58)

5. « Un signe qui n’aurait lieu “qu’une fois” ne serait pas un signe. » (Derrida 1967c: 55)

6. « désir obstiné de sauver la présence et de réduire ou de dériver le signe » (Derrida 1967c: 57)


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

Notes 153

7. « Nous avons éprouvé la solidarité systématique des concepts de sens, d’idéalité, d’objectivité, de vérité, d’intuition, de perception, d’expression. Leur matrice commune est l’être comme présence. » (Derrida 1967c: 111)

8. Interestingly enough, in Speech and Phenomena Derrida does homage to one non-philosopher several times: Benveniste, to whose theory of the for-mal apparatus of the enunciation Derrida refers more than once. Yet with Beveniste (and against Derrida), we can argue that deconstrunction does nothing but clean the philosophical discourse from deictic markers and thus change the mode of the discours into the mode of histoire.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

154

Achard, Pierre (1993): La sociologie du langage. Paris: Presses Universitaires de France.

——— (1995): “Formation discursive, dialogisme et sociologie.” Langages 117: 82–95.

Adam, Jean-Michel (1989): “Pour une pragmatique linguistique et textuelle.” In: Claude Reichler (ed.), L’Interprétation des textes. Paris: Minuit, pp. 183–222.

Allori, Paola Evangelista (ed.) (1998): Academic Discourse in Europe. Thought Processes and Linguistics Realisations. Roma: Bulzoni.

Althusser, Louis (1993): Écrits sur la psychanalyse. Freud et Lacan. Paris: Stock/Imec [transl. Writings on Psychoanalysis: Freud and Lacan. New York: Columbia University Press, 1996].

——— (1995[1969]): Sur la reproduction. Paris: Presses Universitaires de France [transl. Essays on Ideology. London: Verso, 1984; contains “Ideology and Ideological State Apparatuses”, which is the short version of Sur la reproduction].

——— (1996[1965]): Pour Marx. Paris: Maspero [transl. For Marx. The Penguin Press, 1969].

Althusser, Louis, Etienne Balibar, Roger Establet, Pierre Macherey and Jacques Rancière (1965): Lire le Capital. Paris: Quadrige/Presses Universitaires de France [transl. Althusser, Louis and Etienne Balibar (1997): Reading Capital. London: Verso].

Angermüller, Johannes (2003): “Transformation und Narration: Zur Methodologie einer formal-operationalen Textanalyse am Beispiel eines biographischen Interviews mit einer Armenierin in St. Petersburg.” In: Raj Kollmorgen and Heiko Schrader (eds.), Postsozialistische Transformationen: Gesellschaft, Wirtschaft, Kultur. Theoretische Perspektiven und empirische Befunde. Würzburg: Ergon, pp. 199–220.

——— (2005): “ ‘Qualitative’ Methods of Social Research in France: Reconstructing the Actor, Deconstructing the Subject.” Forum Qualitative Research 6(3): Art. 19, http://www.qualitative-research.net/fqs-texte/3-05/05-3-19-e.htm.

——— (2006): “Die marxistische Hermeneutik Fredric Jamesons.” In: Stephan Moebius and Dirk Quadflieg (eds.), Kultur. Theorien der Gegenwart. Konstanz: UVK, pp. 297–308.

——— (2011): “From the Many Voices to the Subject Positions in Anti-globalization discourse. Enunciative Pragmatics and the Polyphonic Organization of Subjectivity.” Journal of Pragmatics 43: 2992–3000.

——— (2012): “Fixing Meaning. The Many Voices of the Post-liberal Hegemony in Russia.” Journal of Language and Politics 11(2): 115–134.

Angermuller, Johannes (2013): “How to Become an Academic Philosopher. Academic Discourse as a Multileveled Positioning Practice.” Sociología histórica 3: 263–289.

——— (2015): The Moment of Theory. Rise and Fall of Structuralism in France. London: Continuum.

References


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

References 155

Angermuller, Johannes, Dominique Maingueneau and Ruth Wodak (eds.) (2014): The Discourse Studies Reader. Main Currents in Theory and Analysis. Amsterdam, Philadelphia: John Benjamins.

Anscombre, Jean-Claude and Oswald Ducrot (1983): L’Argumentation dans la langue. Bruxelles: Mardaga [transl. Argumentation in Language. University of Chicago Press, 1990].

Aron, Jean-Paul (1984): Les Modernes. Paris: Folio. Austin, John L. (1962[1955]): How to Do Things with Words. The William James

Lectures delivered at Harvard University in 1955. Oxford, New York: Oxford University Press.

Authier-Revuz, Jacqueline (1982): “Hétérogénéité montrée et hétérogénéité con-stitutive. Éléments pour une approche de l’autre dans le discours.” DRLAV 26: 91–151.

——— (1995): Ces Mots qui ne vont pas de soi. Boucles réflexives et non-coïncidences du dire. Paris: Larousse.

Bachelard, Gaston (1971): Épistémologie. Paris: Presses Universitaires de France. Bakhtin, Mikhail (1929): Problemy tvorchestva Dostoevskogo. Leningrad: Priboj. ——— (1963): Problemy poetiki Dostoevskogo. Moskva: Sovetskij Pisatel’ [transl.

Problems of Dostoevsky’s Poetics. Minneapolis: University of Minnesota Press, 1984].

Bally, Charles (1965[1932]): Linguistique générale et linguistique française. Bern: Francke.

Barbéris, Jeanne-Marie, Jacques Bres and Paul Siblot (eds.) (1998): De l’actualisation. Paris: CNRS Editions.

Barthes, Roland (1960): Sur Racine. Paris [transl. On Racine. New York: Hill and Wang, 1964].

——— (1967): Système de la mode. Paris: Le Seuil [transl. The Fashion System. New York: Hill and Wang, 1983].

Baudrillard, Jean (1968): Le Système des objets. Paris: Gallimard [transl. The System of Objects. London: Verso, 1996].

Baxter, Judith (2003): Positioning Gender in Discourse: A Feminist Methodology. Houndmills, Basingstoke: Palgrave Macmillan.

Benveniste, Émile (1966): Problèmes de linguistique générale, 1. Paris: Gallimard [transl. Problems in General Linguistics. Coral Gables, Florida: University of Miami Press, 1971].

——— (1974): Problèmes de linguistique générale, 2. Paris: Gallimard. Berrendonner, Alain (1981): Eléments de pragmatique linguistique. Paris: Minuit. Blanché, Robert (1968): Introduction à la logique contemporaine. Paris: Armand

Colin. Blanchot, Maurice (1963): Lautréamont et Sade. Paris: Minuit. Bonnafous, Simone and Maurice Tournier (1995): “Analyse du discours, lexico-

métrie communication et politique.” Langages 119 (Septembre): 67–81. Boudon, Raymond (1980): “The Freudian-Marxian-Structuralist (FMS) move-

ment in France: Variations on a Theme by Sherry Turkle.” The Tocqueville Review 2(1): 5–24.

Bourdieu, Pierre (1972): Esquisse d’une théorie de la pratique: précédé de trois études d’ethnologie kabyle. Genève: Droz [transl. Outline of a Theory of Practice. Cambridge, New York: Cambridge University Press, 1977].


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

156 References

——— (1979): La Distinction. Critique sociale du jugement. Paris: Minuit [transl. Distinction. A Social Critique of the Judgement of Taste. New York and London: Routledge, 1984].

Bowd, Gavin (1999): L’Interminable enterrement. Le communisme et les intellectuels français depuis 1956. Paris: Digraphe.

Bres, Jacques (1999): “Vous les entendez? Analyse du discours et dialogisme.” Modèles linguistiques XX(2): 71–86.

——— (2005): “Savoir de quoi on parle: Dialogue, dialogal, dialogique; dia-logisme, polyphonie” In: Jacques Bres, Patrick Pierre Haillet, Sylvie Mellet, Henning Nølke and Laurence Rosier (eds.), Dialogisme et polyphonie. Bruxelles: De Boeck & Duculot, pp. 47–61.

Brøndal, Viggo (1943): Essais de linguistique générale. Copenhague: Einar Munksgaard.

Brown, Penelope and Stephen Levinson (1987): Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press.

Bühler, Karl (1965[1934]): Sprachtheorie. Die Darstellungsfunktion der Sprache. Stuttgart: Gustav Fischer [transl. Theory of Language. The Representational Function of Language. Amsterdam, Philadelphia: J. Benjamins, 1990].

Canguilhem, Georges (1977): Idéologie et rationalité. Paris: Vrin [transl. Ideology and Rationality in the History of the Life Sciences. Cambridge, MA: MIT Press, 1988].

Cavaillès, Jean (1976[1947]): Sur la logique et la théorie de la science. Paris: Vrin. Cervoni, Jean (1987): L’Énonciation. Paris: Presses Universitaires de France. Charaudeau, Patrick (1997): Le Discours d’information médiatique. La construction

du miroir social. Paris: Nathan. Charaudeau, Patrick and Dominique Maingueneau (2002): Dictionnaire d’analyse

du discours. Paris: Seuil. Chateauraynaud, Francis (2003): Prospéro. Une technologie littéraire pour les sciences

humaines. Paris: CNRS Editions. Courtine, Jean-Jacques (1981): “Quelques problèmes théoriques et méthodo-

logiques en analyse du discours, à propos du discours communiste adressé aux chrétiens.” Langages 62: 9–128.

Courtine, Jean-Jacques and Jean-Marie Marandin (1981): “Quel objet pour l’analyse du discours?” In: Bernard Conein, Jean-Jacques Courtine, Françoise Gadet, Jean-Marie Marandin and Michel Pêcheux (eds.), Matérialités discursives. Lille: Presses Universitaires de Lille, pp. 21–33.

Crémant, Roger (1969): Les Matinées structuralistes. Paris: Robert Laffont. Culioli, Antoine (1985): Notes du séminaire de D.E.A. Poitiers: Département de

Recherches Linguistiques [transl. Cognition and Representation in Linguistic Theory. Texts selected, edited and introduced by Michael Liddle. Amsterdam, Philadelphia: John Benjamins B.V., 1995 (selections)].

——— (1990): Pour une linguistique de l’énonciation. Opérations et représentations. Tome 1. Paris: Ophrys.

——— (1999): Pour une linguistique de l’énonciation. Formalisation et opérations de repérage. Tome 2. Paris: Ophrys.

——— (2002): Variations sur la linguistique. Entretiens avec Frédéric Fau. Paris: Klincksieck.

Danon-Boileau, Laurent (1987): Le Sujet de l’énonciation. Psychanalyse et linguistique. Paris: Ophrys.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

References 157

Deleuze, Gilles and Félix Guattari (1980): Mille plateaux. Capitalisme et schizophrénie. Paris: Minuit [transl. A Thousand Plateaus. Capitalism and Schizophrenia. London: The Athlone Press, 1988].

Derrida, Jacques (1967a): De la grammatologie. Paris: Minuit [transl. Of Grammatology. Baltimore: Johns Hopkins University Press, 1976].

——— (1967b): L’Écriture et la différence. Paris: Le Seuil [transl. Of Grammatology. Baltimore: Johns Hopkins University Press, 1976].

——— (1967c): La Voix et le phénomène. Paris: Quadrige/Presses Universitaires de France [transl. Speech and Phenomena, and Other Essays on Husserl’s Theory of signs. Evanston: Northwestern University Press, 1973].

——— (1972): Marges de la philosophie. Paris: Minuit [transl. Margins of Philosophy. Chicago: University of Chicago Press, 1982].

Détrie, Catherine, Paul Siblot and Bertrand Verine (2001): Termes et concepts pour l’analyse du discours. Une approche praxématique. Paris: Honoré Champion.

Ducrot, Oswald (1980a): “Analyses pragmatiques.” Communications 32: 11–60. ——— (1980b): Les Échelles argumentatives. Paris: Minuit. ——— (1984): Le Dire et le dit. Paris: Minuit. ——— (1989): Logique, structure, énonciation. Lectures sur le langage. Paris: Minuit. ——— (1992): “Entretien avec Oswald Ducrot.” In: C. Lopez Alonso and A. Sere

de Olmos (eds.), Où en est la linguistique? Entretiens avec des linguistes. Paris: Didier, pp. 59–75.

Dufrenne, Mikel (1968): Pour l’homme. Paris: Le Seuil. Ehlich, Konrad (2007): Sprache und sprachliches Handeln. Berlin, New York: Walter

de Gruyter. Fairclough, Norman (1992): Discourse and Social Change. Cambridge, Oxford:

Polity Press. ——— (2003): Analysing Discourse. Textual Analysis for Social Research. London,

New York: Routledge. Ferry, Luc and Alain Renaut (1988[1985]): La Pensée 68. Essai sur l’anti-humanisme

contemporain. Paris: Gallimard [transl. French Philosophy of the Sixties. An Essay on Antihumanism. Amherst: University of Massachusetts Press, 1990].

Fløttum, Kjersti (2005): “Moi et autrui dans le discours scientifique: l’exemple de la négation ne…pas ” In: Jacques Bres, Patrick Pierre Haillet, Sylvie Mellet, Henning Nølke and Laurence Rosier (eds.), Dialogisme et polyphonie. Bruxelles: De Boeck & Duculot, pp. 323–337.

Foucault, Michel (1961): Folie et déraison. Histoire de la folie à l’âge classique. Paris: Plon [transl. Madness and Civilization. A History of Insanity in the Age of Reason. London: Routledge, 2001].

——— (1966): Les Mots et les choses. Une archéologie des sciences humaines. Paris: Gallimard [transl. The Order of Things. An Archeology of the Human Sciences. London: Routledge, 2002].

——— (1969): L’Archéologie du savoir. Paris: Gallimard [transl. The Archaeology of Knowledge & The Discourse on Language. London: Routledge, 1989].

——— (1971): L’Ordre du discours. Paris: Gallimard [transl. The Archaeology of Knowledge & the Discourse on Language. New York: Pantheon, 1972, pp. 215–237].

——— (1994a): “Des espaces autres.” In: Dits et écrits, tome 4. 1980–1988. Paris: Gallimard, pp. 752–762.

——— (1994b[1968]): “L’homme est-il mort?” In: Dits et Ecrits, tome 1. 1954–1969. Paris: Gallimard, pp. 540–544.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

158 References

——— (1994c): “La philosophie structuraliste permet de diagnostiquer ce qu’est aujourd’hui.” In: Dits et Ecrits, tome 1. 1954–1969. Paris: Gallimard, pp. 580–585.

——— (1994d[1967/1969]): “Qui êtes-vous, professeur Foucault?” In: Dits et Ecrits, tome 1. 1954–1969. Paris: Gallimard, pp. 601–620.

Fougeyrollas, Pierre (1976): Contre Lévi-Strauss, Lacan et Althusser. Trois essais sur l’obscurantisme contemporain. Paris: Spag-Papyrus.

Fuchs, Catherine (1981a): “Les problématiques énonciatives: Esquisse d’une présentation historique et critique.” DRLAV 25: 35–60.

Fuchs, Cathérine (1981b): La Paraphrase. Paris: Presses Universitaires de France. García, Francisco Vázquez (2009): “Foucault y la filosofía analítica.” Ambiente

Jurídico 11: 109–143. Genette, Gérard (1972): Figures III. Paris: Seuil. Goffman, Erving (1981): Forms of Talk. Philadelphia: University of Pennsylvania

Press. Greimas, Algirdas J. (1966): Sémantique structurale. Paris: Larousse [transl.

Structural Semantics. An Attempt at a Method. Lincoln: University of Nebraska Press, 1983].

Greimas, Algirdas Julien and Joseph Courtès (1993[1979]): Sémiotique. Dictionnaire raisonné de la théorie du langage. Paris: Hachette [transl. Semiotics and Language. An Analytical Dictionary. Bloomington: Indiana University Press, 1982].

Grice, H. Paul (1989): Studies in the Way of Words. Cambridge: Harvard University Press.

Grize, Jean-Blaise (1982): De la logique à l’argumentation. Genève et Paris: Droz. Grunig, Blanche-Noëlle (1981): “Plusieurs pragmatiques.” DRLAV 25: 101–118. Guilhaumou, Jacques, Denise Maldidier and Régine Robin (1994): Discours et

archive. Expérimentations en analyse du discours. Liège: Mardaga. Guillaume, Georges (1974[1933–1958]): Langage et science du langage. Paris: Nizet. Habermas, Jürgen (1993[1985]): Der philosophische Diskurs der Moderne. Zwölf

Vorlesungen. Frankfurt am Main: Suhrkamp [transl. The Philosophical Discourse of Modernity. Twelve Lectures. Cambridge, MA: MIT Press, 1987].

Halliday, M.A.K. (1978): Language As Social Semiotic. London: Edward Arnold. Haroche, Claudine, Paul Henry and Michel Pêcheux (1971): “La sémantique et la

coupure saussurienne: langue, langage, discours.” Langages 24: 93–106. Harris, Zellig S. (1963): Discourse Analysis Reprints. The Hague: Mouton & Co. Henry, Paul (1975): “Constructions relatives et articulations discursives.”

Langages 37: 81–98. ——— (1977): Le Mauvais Outil. Langue, sujet, discours. Paris: Klinksieck. ——— (1984): “Les faits sont têtus mais ne parlent point. Quelques remarques

sur l’énonciation et la pragmatique en linguistique.” DRLAV 31: 107–119. Hodge, Robert and Gunther Kress (1988): Social Semiotics. Cambridge: Polity. Husserl, Edmund (1984[1901]): Logische Untersuchungen. Zweiter Band, Erster

Teil: Untersuchungen zur Phänomenologie und Theorie der Erkenntnis. The Hague, Boston, Lancaster: Martinus Nijhoff [transl. Logical Investigations. London, New York: Routledge, 2001].

Hyland, Ken (2006): “Disciplinary Differences: Language Variation in Academic Discourse.” In: Ken Hyland and Marina Bondi (eds.), Academic Discourse Across Disciplines. Bern: Peter Lang, pp. 17–45.

Jakobson, Roman (1995[1990]): On Language. Cambridge, MA, London: Harvard University Press.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

References 159

Jameson, Fredric (1981): The Political Unconscious. Narrative As a Socially Symbolic Act. Ithaca, NY: Cornell University Press.

Jenny, Jacques (1997): “Méthodes et pratiques formalisées d’analyse de contenu et de discours dans la recherche sociologique française contemporaine. Etat des lieux et essai de classification.” Bulletin de Méthodologie Sociologique 54: 64–112.

Kauppi, Niilo (1990): Tel Quel. La Constitution sociale d’une avant-garde. Helsinki: Finnish Society of Sciences and Letters [transl. The Making of an Avant-Garde: Tel Quel. Berlin: Mouton de Gruyter, 1994].

Kerbrat-Orecchioni, Catherine (1980): L’Énonciation. De la subjectivité dans le langage. Paris: Armand Colin.

——— (1990): Les Interactions verbales. T I-III. Paris: Armand Colin. Kristeva, Julia (1973): “The Ruin of a Poetics.” In: Stephen Bann and John E.

Bowlt (eds.), Russian Formalism: A Collection of Articles and Texts in Translation. Edinburgh: Scottish University Press, pp. 102–119.

Lacan, Jacques (1966): Écrits. Paris: Le Seuil [transl. Écrits. A selection. New York: Norton, 1977].

——— (1973): Le Séminaire. Livre XI. Les quatre concepts fondamentaux de la psych-analyse. Paris: Le Seuil [transl. The Four Fundamental Concepts of Psychoanalysis. New York, London: W.W. Norton, 1973].

Larrauri, Maite (1999): Anarqueología. Teoría de la verdad en Michel Foucault. Valencia: Episteme.

Lebart, Louis and André Salem (1994): Statistique textuelle. Paris: Dunod. Lebart, Ludovic, André Salem and Lisette Berry (1998): Exploring Textual Data.

Dordrecht, Boston, London: Kluwer. Leeuwen, Theo van (2005): Introducing Social Semiotics. London, New York:

Routledge. Leimdorfer, François (1992): Discours académique et colonisation. Paris: Publisud. Lyotard, Jean-François (1979): La Condition postmoderne. Paris: Minuit [transl.

The Postmodern Condition. A Report on Knowledge. Minneapolis: University of Minnesota Press, 1984].

——— (1988): Le postmodernisme expliqué aux enfants. Paris: Galilée [transl. The Postmodern Explained. Correspondence, 1982–1985. Minneapolis: University of Minnesota Press, 1993].

Maingueneau, Dominique (1976): Initiation aux méthodes de l’analyse de discours. Problèmes et perspectives. Paris: Hachette.

——— (1981): Approches de l’énonciation en linguistique française. Paris: Hachette. ——— (1987): Nouvelles Tendances en analyses du discours. Paris: Hachette. ——— (1993): Le Contexte de l’œuvre littéraire. Énonciation, écrivain, société. Paris:

Dunod. ——— (1995): “L’énonciation philosophique comme institution discursive.”

Langages 119 (Septembre): 40–62. ——— (1996): “L’analyse du discours en France aujourd’hui.” In: Sophie Moirand

(ed.), Le Discours: Enjeux et perspectives (numéro spécial Le Français dans le monde, juillet 1996). Paris: Hachette, pp. 8–15.

——— (1997): Pragmatique pour le discours littéraire. Paris: Dunod. Maingueneau, Dominique and Johannes Angermüller (2007): “Discourse

analysis in France. Dominique Maingueneau in conversation with Johannes Angermüller.” Forum Qualitative Research 8(2), http://www.qualitative-research.net/fqs-texte/2-07/07-2-21-e.htm.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

160 References

Maingueneau, Dominique and Frédéric Cossutta (1995): “L’analyse des discours constituants.” Langages 117: 112–125.

Marandin, Jean-Marie (1979): “Problèmes d’analyse du discours. Essai de descrip-tion du discours français sur la Chine.” Langages 55: 17–88.

Martin, James Robert and Peter R.R. White (2005): The Language of Evaluation. Appraisal in English. Basingstoke: Palgrave/MacMillan.

Matonti, Frédérique (2005): Intellectuels communistes. Essai sur l’obéissance politique. La Nouvelle Critique (1967–1980). Paris: Découverte.

Mazière, Francine (2005): L’Analyse du discours. Paris: Presses Universitaires de France.

Moeschler, Jacques and Anne Reboul (eds.) (1994): Dictionnaire encyclopédique de pragmatique. Paris: Seuil.

Moirand, Sophie (2001): “Du traitement différent de l’intertexte selon les genres convoqués dans les événements scientifiques à caractère politique.” Semen 13: 97–118.

——— (2010): “Retour sur une approche dialogique en analyse du discours.” In: Marion Colas-Blaise, Mohamed Kara, Laurent Perrin and André Pettitjean (eds.), La question polyphonique ou dialogique en sciences du langage (Recherches linguistiques 31). Metz: Université Paul Verlaine, pp. 375–398.

Münchow, Patricia von (2010): “Langue, discours, culture: quelle articulation?” Signes, discours et sociétés 4, http://www.revue-signes.info/document.php?id=1439.

Nølke, Henning (2010): “L’ancrage linguistique de la polyphonie.” In: Marion Colas-Blaise, Mohamed Kara, Laurent Perrin and André Pettitjean (eds.), La question polyphonique ou dialogique en sciences du langage (Recherches linguistiques 31). Metz: Université Paul Verlaine, pp. 19–38.

Nølke, Henning, Kjersti Fløttum and Coco Norén (2004): ScaPoLine. La théorie scandinave de la polyphonie linguistique. Paris: Kimé.

Nowakowska, Aleksandra (2005): “Dialogisme, polyphonie. Des textes russes de M. Bakhtine à la linguistique contemporaine.” In: Jacques Bres, Patrick Pierre Haillet, Sylvie Mellet, Henning Nølke and Laurence Rosier (eds.), Dialogisme et polyphonie. Bruxelles: De Boeck & Duculot, pp. 19–32.

Paillé, Pierre (1996): “Qualitative (analyse).” In: Alex Muchielli (ed.), Dictionnaire des méthodes qualitatives en sciences humaines et sociales. Paris: Armand Colin, pp. 180–184.

Parret, Herman (1987): Prolégomènes à la théorie de l’énonciaton. De Husserl à la pragmatique. Bern: Peter Lang.

Pêcheux, Michel (1969): Analyse automatique du discours. Paris: Dunod [transl. Automatic Discourse Analysis. Amsterdam; Atlanta, GA Rodopi, 1995].

——— (1975): Les Vérités de La Palice. Paris: Maspero [transl. Language, semantics and ideology. Stating the obvious. London: Macmillan, 1982].

——— (1981): “L’énoncé: enchâssement, articulation et dé-liaison.” In: Bernard Conein, Jean-Jacques Courtine, Françoise Gadet, Jean-Marie Marandin and Michel Pêcheux (eds.), Matérialités discursives. Lille: Presses Universitaires de Lille, pp. 143–148.

——— (1990): L’inquiétude du discours. Paris: Edition des Cendres. Perelman, Chaïm (1963): The Idea of Justice and the Problem of Argument. London:

Routledge. Perrin, Laurent (2006): “Introduction.” In: Laurent Perrin (ed.), Le sens et ses voix

(Recherches linguistiques 28). Metz: Université Paul Veraline, pp. 5–17.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

References 161

Pestaña, José Luis (2006): En devenant Foucault. Sociogenèse d’un grand philosophe. Broissieux: Editions du Croquant.

Picard, Raymond (1965): Nouvelle critique ou nouvelle imposture. Paris: Pauvert. Plantin, Christian (1990): Essais sur l’argumentation. Introduction à l’étude linguis-

tique de la parole argumentative. Paris: Kimé. Rabatel, Alain (1998): La construction textuelle du point de vue. Lausanne, Paris:

Delachaux et Niestlé. ——— (2010): “Retour sur les relations entre locuteurs et énonciateurs.” In:

Marion Colas-Blaise, Mohamed Kara, Laurent Perrin and André Pettitjean (eds.), La question polyphonique ou dialogique en sciences du langage (Recherches linguistiques 31). Metz: Université Paul Verlaine, pp. 355–373.

Reboul, Anne and Jacques Moeschler (1996): “Faut-il continuer à faire de l’analyse de discours?” Hermès 16: 61–92.

——— (1998a): La pragmatique aujourd’hui. Une nouvelle science de la communica-tion. Paris: Le Seuil.

——— (1998b): Pragmatique du discours. De l’interprétation de l’énoncé à l’interprétation du discours. Paris: Armand Colin.

Récanati, François (1979a): La transparence et l’énonciation. Pour introduire à la pragmatique. Paris: Le Seuil.

——— (1979b): “Le développement de la pragmatique.” Langue française 42: 6–20.

——— (1981): Les Énoncés performatifs. Paris: Minuit [transl. Meaning and Force: The Pragmatics of Performative Utterances. Cambridge: Cambridge University Press. 1987].

Ricœur, Paul (1961): De l’interprétation. Essai sur Freud. Paris: Seuil [transl. Freud and Philosophy. An Essay on Interpretation. New Haven: Yale University Press, 1970].

Robin, Régine (1973): Histoire et linguistique. Paris: Colin. Rosier, Laurence (1999): Le discours rapporté: histoire, théories, pratiques. Bruxelles:

Duculot. Roudinesco, Elisabeth (1993): Jacques Lacan. Esquisse d’une vie, histoire d’un

système de pensée. Paris: Fayard [transl. Jacques Lacan. An Outline of a Life and a History of a System of Thought. Oxford: Polity Press, 2005].

Sacks, Harvey (1986[1972]): “On the Analyzability of Stories by Children.” In: John Gumperz and Dell Hymes (eds.), Directions in Sociolinguistics. The Ethnography of Communication. Oxford: Blackwell, pp. 325–345.

Sartre, Jean-Paul (1966): “Jean-Paul Sartre répond.” L’Arc 30: 87–96. Saussure, Ferdinand de (1962[1916]): Cours de linguistique générale. Paris: Payot

[transl. Course in General Linguistics. London: Fontana, 1974]. Seriot, Patrick (1985): Analyse du discours politique soviétique. Paris: Institut

d’Études slaves. Sollers, Philippe (1957): Le défi. Paris: Seuil. ——— (1968a): “Écriture et révolution (entretien avec Jacques Henric).” In: Tel

Quel (ed.), Théorie d’ensemble. Paris: Le Seuil, pp. 67–79. ——— (1968b): “Le réflexe de réduction.” In: Tel Quel (ed.), Théorie d’ensemble.

Paris: Le Seuil, pp. 391–398. Sperber, Dan and Deirdree Wilson (1989): La Pertinence. Communication et

cognition. Paris: Minuit [transl. Relevance: Communication and Cognition. Oxford, Cambridge: Blackwell, 1993].


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

162 References

Sumpf, J. and J. Dubois (1969): “Problèmes de l’analyse du discours.” Langages 13: 3–7.

Tel Quel (1968): Théorie d’ensemble. Paris: Le Seuil [transl. in ffrench, Patrick; Lack, Roland-François (ed.) (1998): The Tel Quel Reader. London, New York: Routledge; contains selected contributions only].

Toulmin, Stephen (1952): The Uses of Argument. Cambridge: Cambridge University Press.

Tournier, Maurice et al. (1975): Des tracts en mai 1968. Paris: Armand Colin. Valentin-Mc Lean, Frédérique (2005): Dissidents du parti communiste français. La

révolte des intellectuels communistes dans les années 1970. Paris: L’Harmattan. Vogüe, Sarah de (1992): “Culioli après Benveniste. Énonciation, langage, intégra-

tion.” Linx 26: 77–108. Williams, Glyn (1999): French Discourse Analysis. London: Routledge. Wittgenstein, Ludwig (1997[1953]): Philosophische Untersuchungen/Philosophical

Investigations. Second Edition. Oxford: Blackwell.


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

163

allocutor → locutorargumentation 20, 27–28, 47–60,

59–61, 82–83, 131–133

context 10–15, 22–35, 39–47, 56–71, 82–83, 87–90, 98–101, 111, 118–119, 127–131, 140–145

conversation maxim, 30–33, 46

deixis → indexicalitydistributionalism 7–8

formation, discursive 7, 10–11, 14–26

enunciation (énonciation) 1–4, 11–15, 22–38,

formal apparatus of enunciation 22–26, 39, 60–61

marker, enunciative 1–6, 26, 33–38, 43, 51, 59–67

utterance (énoncé) 1–4, 10–17, 22–38

Foucault 9–23, 31, 109–118

heterogeneity 10, 21–28, 35, 83, 98–99

humanism 4–5, 66–78, 145

ideology 16–21, 48–52indexicality 25–26, 38–43, 62,

81–85, 100–101, 111, 126–127, 132

preconstructed deixis 118, 128interdiscourse 21, 35, 50, 58, 74interpretation 3–4, 14–15, 19, 39,

46–49, 54–71, 93

knowledge 10–20, 31, 34, 48–53, 55–73

locutor 24–30, 43–49, 61allocutor 44–48

marker, enunciative → enunciationmateriality 3, 10, 13–17, 35–38,

52–65, 71 meaning 3–10, 22, 32, 52–63nominalization 51–52, 59–62, 96Marxism 9, 16

participant 4–6, 20, 33, 39, 52–71, 143–144

Pêcheux 9–10, 16–24, 50–51point of view 28, 30, 43–46polyphony 6, 24, 28–29, 43–50,

59–65, 69, 85–90poststructuralism 65–71, 129, 140, pragmatics 10–12, 143

enunciative pragmatics 1–5radical pragmatics 30–38

preconstruct 9–20, 38–39, 50–52, 60–62, 96–97, 103–104, 127–128

presupposition 19–21, 27, 48–51psychoanalysis 20–21, 35–36, 62–83

reader → participant

Saussure 7–11scenography 22–34, 68–71sentence 7–8, 12, 22, 27, 31, 33–34 speaker 2, 28–29, 35, 38, 41–43, 52,

102–117speech act 3, 10–13, 20–32square, semiotic 135–138structuralism 10, 65–71subjectivity 2–4, 17–21, 26–38, 48–57

text 1–6, 10–11, 23, 32–33, 36–38

utterance → enunciation

Index


Co

pyr

igh

t m

ater

ial f

rom

ww

w.p

alg

rave

con

nec

t.co

m -

lice

nse

d t

o M

on

ash

Un

iver

sity

- P

alg

rave

Co

nn

ect

- 20

14-1

1-26

poststructuralist discourse analysis ||

Documents