Subido por bernardo malaver

Johanna Oksala - Foucault on Freedom (Modern European Philosophy) (2009)

Freedom and the subject were guiding themes for Michel Foucault
throughout his philosophical career. In this clear and comprehensive
analysis of his thought, Johanna Oksala identifies the different interpretations of freedom in his philosophy and examines three major divisions
of it: the archaeological, the genealogical, and the ethical. She shows
convincingly that in order to appreciate Foucault’s project fully we must
understand his complex relationship to phenomenology, and she discusses Foucault’s treatment of the body in relation to recent feminist
work on this topic. Her sophisticated but lucid book illuminates the possibilities which Foucault’s philosophy opens up for us in thinking about
j o h a n n a o k s a l a is a Research Fellow in the Department of Philosophy at the University of Helsinki. She has published articles on Foucault,
phenomenology and feminist philosophy.
General Editor
Robert B. Pippin, University of Chicago
Advisory Board
Gary Gutting, University of Notre Dame
Rolf-Peter Horstmann, Humboldt University, Berlin
Mark Sacks, University of Essex
Some recent titles
Daniel W. Conway: Nietzsche’s Dangerous Game
John P. McCormick: Carl Schmitt’s Critique of Liberalism
Frederick A. Olafson: Heidegger and the Ground of Ethics
Günter Zöller: Fichte’s Transcendental Philosophy
Warren Breckman: Marx, the Young Hegelians, and the Origins of Radical
Social Theory
William Blattner: Heidegger’s Temporal Idealism
Charles Griswold: Adam Smith and the Virtues of Enlightenment
Gary Gutting: Pragmatic Liberalism and the Critique of Modernity
Allen Wood: Kant’s Ethical Thought
Karl Ameriks: Kant and the Fate of Autonomy
Alfredo Ferrarin: Hegel and Aristotle
Cristina Lafont: Heidegger, Language and World-Disclosure
Daniel Dahlstrom: Heidegger’s Concept of Truth
Michelle Grier: Kant’s Doctrine of Transcendental Illusion
Henry Allison: Kant’s Theory of Taste
Allen Speight: Hegel, Literature and the Problem of Agency
J. M. Bernstein: Adorno
Will Dudley: Hegel, Nietzsche and Philosophy
Taylor Carman: Heidegger’s Analytic
Douglas Moggach: The Philosophy and Politics of Bruno Bauer
Rüdiger Bubner: The Innovations of Idealism
Jon Stewart: Kierkegaard’s Relations to Hegel Reconsidered
Nicholas Wolterstorff: Thomas Reid and the Story of Epistemology
Michael Quante: Hegel’s Concept of Action
Wolfgang Detel: Foucault and Classical Antiquity
University of Helsinki
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo
Cambridge University Press
The Edinburgh Building, Cambridge cb2 2ru, UK
Published in the United States of America by Cambridge University Press, New York
Information on this title:
C Johanna Oksala 2005
This book is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.
First published 2005
Printed in the United Kingdom at the University Press, Cambridge
A catalogue record for this book is available from the British Library
isbn-13 978-0-521-84779-7 hardback
isbn-10 0-521-84779-6 hardback
Cambridge University Press has no responsibility for the persistence or accuracy of URLs
for external or third-party internet websites referred to in this book, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
page vii
List of abbreviations
part i language
Philosophical laughter
An archaeology of order
The three epistemes
The birth and death of man
The being of language
The Foucaultian failure of phenomenology
The history of science
The analytic of finitude
The anonymity of language
A view from nowhere
The subject of change
The freedom of language
part ii body
A genealogy of the subject
The constitution of the subject
The problem of circularity
Anarchic bodies
The body of power
The discursive body
co n ten t s
The resistance of the body
The anarchic body
Female freedom
The anonymous subjectivity of the body
The historical constitution of the body
Female freedom?
part iii ethics
The silence of ethics
History of ethics
Ethics as practice
The ethical subject
Ethics as aesthetics
Philosophy lived
The freedom of philosophy
The freedom of critical reflection
Freedom as ethos
The different meanings of freedom
The other
Ethical subject and the other
Subjectivity as passivity
The other as precondition of ethics
Conclusion: freedom as an operational concept
I am fortunate to have had some of the leading philosophers in my
field to read parts or versions of this work at different stages of its
development: Rosi Braidotti, Simon Critchley, Thomas Flynn, Gary
Gutting, Sara Heinämaa, Jana Sawicki and Dan Zahavi. I am deeply
grateful to them for their perceptive comments, good advice and constructive criticism.
I want to thank the many networks of colleagues and good friends
who have inspired, supported and discussed my work. I also want to
thank my students, whose critical questions and fresh insights have
contributed to my views on Foucault.
A different version of chapter 5 originally appeared as the article
‘Anarchic Bodies: Foucault and the Feminist Question of Experience’
in Hypatia, A Journal of Feminist Philosophy, vol. 19, no. 4. I am grateful
for the permission to reprint it here.
Last but not least, I want to thank my family for their love, support
and remarkable patience.
Books in English
The Archaeology of Knowledge. Trans. A. M. Sheridan Smith.
New York: Pantheon Books, 1972.
Death and the Labyrinth: The World of Raymond Roussel. Trans.
Charles Ruas. New York: Doubleday, 1986.
Discipline and Punish: The Birth of the Prison. Trans. Alan Sheridan. Harmondsworth: Penguin, 1977/1991.
The History of Sexuality, vol. i, An Introduction. Trans. Robert
Hurley. Harmondsworth: Penguin, 1978.
The Order of Things: An Archaeology of the Human Sciences.
London: Routledge, 1970/1994.
The History of Sexuality, vol. ii, The Use of Pleasure. Trans.
Robert Hurley. Harmondsworth: Penguin, 1985.
Articles, essays and interviews in English
‘The Art of Telling the Truth’, in Michel Foucault: Politics,
Philosophy, Culture, Interviews and Other Writings 1977–1984,
ed. Lawrence Kritzman, trans. Alan Sheridan and others.
New York: Routledge, 1988, 86–95.
‘Body/Power’, in Power/Knowledge: Selected Interviews and
Other Writing 1972–1977, ed. Colin Gordon, trans. Colin
Gordon, Leo Marshall, John Mepham and Kate Soper.
Brighton: Harvester Press, 1980, 55–62.
‘The Confession of the Flesh’, in Power/Knowledge:
Selected Interviews and Other Writing 1972–1977, ed. Colin
list o f abbreviat io ns
Gordon, trans. Colin Gordon, Leo Marshall, John Mepham
and Kate Soper. Brighton: Harvester Press, 1980, 194–228.
‘The Concern for Truth’, in Michel Foucault: Politics, Philosophy, Culture, Interviews and Other Writings 1977–1984, ed.
Lawrence Kritzman, trans. Alan Sheridan and others. New
York: Routledge, 1988, 255–67.
‘Critical Theory/Intellectual History’, in Michel Foucault:
Politics, Philosophy, Culture, Interviews and Other Writings
1977–1984, ed. Lawrence Kritzman, trans. Alan Sheridan
and others. New York: Routledge, 1988, 17–46.
‘The Ethic of Care for the Self as a Practise of Freedom’,
in The Final Foucault, ed. James Bernauer and David Rasmussen. Cambridge, Mass.: MIT Press, 1988, 1–20.
‘On the Genealogy of Ethics: An Overview of Work in
Progress’, in The Foucault Reader, ed. Paul Rabinow, trans.
J. Harari. Harmondsworth: Penguin, 1984, 340–72.
‘The Hermeneutic of the Subject’, in Ethics, Subjectivity
and Truth: Essential Works of Foucault 1954–1984, ed. Paul
Rabinow, trans. Robert Hurley and others. New York: New
Press, 1997, 93–106.
‘Introduction’, in Herculine Barbin, Being the Recently Discovered Memoirs of a Nineteenth-Century French Hermaphrodite,
ed. Michel Foucault, trans. Richard McDougall. Brighton:
Harvester Press, 1980, vii–xvii.
‘Introduction’, in Georges Canguilhem, The Normal and
the Pathological, trans. Carolyn R. Fawcett. New York: Zone
Books, 1991, 7–24.
‘Minimalist Self’, in Michel Foucault: Politics, Philosophy, Culture, Interviews and Other Writings 1977–1984, ed. Lawrence
Kritzman, trans. Alan Sheridan and others. New York: Routledge, 1988, 3–16.
‘Nietzsche, Genealogy, History’, in The Foucault Reader, ed.
Paul Rabinow, trans. J. Harari. Harmondsworth: Penguin,
1984, 76–100.
‘On the Ways of Writing History’, in Aesthetics, Method
and Epistemology: Essential Works of Foucault 1954–1984, ed.
James D. Faubion, series ed. Paul Rabinow, trans. Robert
Hurley and others. New York: New Press, 1998, 279–
list o f a b b r ev iati ons
‘Politics and Ethics: An Interview’, in The Foucault Reader, ed.
Paul Rabinow, trans. J. Harari. Harmondsworth: Penguin,
1984, 373–80.
‘Postscript, An Interview with Michel Foucault by Charles
Ruas’, in Death and the Labyrinth: The World of Raymond Roussel, trans. Charles Ruas. New York: Doubleday, 1986, 169–
‘A Preface to Transgression’, in Aesthetics, Method and Epistemology: Essential Works of Foucault 1954–1984, ed. James D.
Faubion, series ed. Paul Rabinow, trans. Robert Hurley and
others. New York: New Press, 1998, 69–87.
‘Politics and Reason’, in Michel Foucault: Politics, Philosophy, Culture, Interviews and Other Writings 1977–1984, ed.
Lawrence Kritzman, trans. Alan Sheridan and others. New
York: Routledge, 1988, 57–85.
‘Questions on Geography’, in Power/Knowledge: Selected Interviews and Other Writing 1972–1977, ed. Colin Gordon, trans.
Colin Gordon, Leo Marshall, John Mepham and Kate Soper.
Brighton: Harvester Press, 1980, 63–77.
‘The Return of Morality’, in Michel Foucault: Politics, Philosophy, Culture, Interviews and Other Writings 1977–1984, ed.
Lawrence Kritzman, trans. Alan Sheridan and others. New
York: Routledge, 1988, 242–54.
‘Sexual Choice, Sexual Act: Foucault and Homosexuality’,
in Michel Foucault: Politics, Philosophy, Culture, Interviews and
Other Writings 1977–1984, ed. Lawrence Kritzman, trans.
Alan Sheridan and others. New York: Routledge, 1988, 286–
‘Subject and Power’, in Hubert L. Dreyfus and Paul Rabinow, Michel Foucault: Beyond Structuralism and Hermeneutics.
Hemel Hempstead: Harvester Press, 1982, 208–26.
‘Sex, Power, and the Politics of Identity’, in Ethics, Subjectivity
and Truth: Essential Works of Foucault 1954–1984, ed. Paul
Rabinow, trans. Robert Hurley and others. New York: New
Press, 1997, 163–73.
‘Structuralism and Post-Structuralism’, in Aesthetics, Method
and Epistemology: Essential Works of Foucault 1954–1984, ed.
James D. Faubion, series ed. Paul Rabinow, trans. Robert
Hurley and others. New York: New Press, 1998, 433–58.
list o f abbreviat io ns
‘Subjectivity and Truth’, in The Politics of Truth: Michel Foucault, ed. Sylvere Lotringer and Lysa Hochroth. New York:
Semiotext(e), 1997, 171–98.
‘A Swimmer Between Two Words’, in Aesthetics, Method
and Epistemology: Essential Works of Foucault 1954–1984, ed.
James D. Faubion, series ed. Paul Rabinow, trans. Robert
Hurley and others. New York: New Press, 1998, 171–
‘Technologies of the Self’, in Ethics, Subjectivity and Truth:
Essential Works of Foucault 1954–1984, ed. Paul Rabinow,
trans. Robert Hurley and others. New York: New Press, 1997,
‘Truth and Juridical Forms’, in Power: Essential Works of
Foucault 1954–1984, ed. James D. Foubion, series ed. Paul
Rabinow, trans. Robert Hurley and others. New York: New
Press, 1997, 1–89.
‘Two Lectures’, in Power/Knowledge: Selected Interviews and
Other Writing 1972–1977, ed. Colin Gordon, trans. Colin
Gordon, Leo Marshall, John Mepham and Kate Soper.
Brighton: Harvester Press, 1980, 78–108.
‘Truth and Power’, in Power/Knowledge: Selected Interviews
and Other Writing 1972–1977, ed. Colin Gordon, trans.
Colin Gordon, Leo Marshall, John Mepham and Kate Soper.
Brighton: Harvester Press, 1980, 109–33.
‘What is an Author?’, in The Foucault Reader, ed. Paul Rabinow, trans. J. Harari. Harmondsworth: Penguin, 1984, 101–
‘What is Critique?’, in The Politics of Truth: Michel Foucault,
ed. Sylvere Lotringer and Lysa Hochroth. New York: Semiotext(e), 1997, 23–82.
‘What is Enlightenment?’, in The Foucault Reader, ed. Paul
Rabinow, trans. J. Harari. Harmondsworth: Penguin, 1984,
Books in French
L’archéologie du savoir. Paris: Gallimard, 1969/2001.
Les mots et les choses. Une archéologie des sciences humaines. Paris:
Gallimard, 1966/1996.
Raymond Roussel. Paris: Gallimard, 1963.
list o f a b b r ev iati ons
Surveiller et punir. Naissance de la prison. Paris: Gallimard,
Histoire de la sexualité, vol. ii, L’usage des plaisirs. Paris: Gallimard, 1984/1994.
Histoire de la sexualité, vol. i, La volonté de savoir. Paris: Gallimard, 1976.
Articles, essays and interviews in French
‘Cours du 14 janvier 1976’, in Dits et écrits 1954–1988,
vol. iii, 1976–1979, ed. Daniel Defert and François Ewald.
Paris: Gallimard, 1976/1994, 175–89.
‘L’éthique du souci de soi comme pratique de la liberté’,
in Dits et écrits 1954–1988, vol. iv, 1980–1988, ed. Daniel
Defert and François Ewald. Paris: Gallimard, 1984/1994,
‘Introduction par Michel Foucault’, in Dits et écrits 1954–
1988, vol. iii, 1976–1979, ed. Daniel Defert and François
Ewald. Paris: Gallimard, 1978/1994, 429–42.
‘Le jeu de Michel Foucault’, in Dits et écrits 1954–1988, vol.
iii, 1976–1979, ed. Daniel Defert and François Ewald. Paris:
Gallimard, 1978/1994, 298–9.
‘La philosophie analytic de la politique’, in Dits et écrits
1954–1988, vol. iii, 1976–1979, ed. Daniel Defert and
François Ewald. Paris: Gallimard, 1978/1994, 534–51.
‘Pouvoir et corps’, in Dits et écrits 1954–1988, vol. ii, 1970–
1975, ed. Daniel Defert and François Ewald. Paris: Gallimard, 1975/1994, 754–60.
‘Politique et éthique: une interview’, in Dits et écrits 1954–
1988, vol. iv, 1980–1988, ed. Daniel Defert and François
Ewald. Paris: Gallimard, 1984/1994, 584–90.
‘Qu’est-ce qu’un auteur?’, in Dits et écrits 1954–1988,
vol. i, 1954–1969, ed. Daniel Defert and François Ewald.
Paris: Gallimard, 1969/1994, 789–821.
‘Qu’est-ce que les Lumières?’, in Dits et écrits 1954–1988,
vol. iv, 1980–1988, ed. Daniel Defert and François Ewald.
Paris: Gallimard, 1984/1994, 562–78.
‘Qu’est-ce que les Lumières?’, in Dits et écrits 1954–1988,
vol. iv, 1980–1988, ed. Daniel Defert and François Ewald.
Paris: Gallimard, 1984/1994, 679–88.
list o f abbreviat io ns
‘Structuralisme et poststructuralisme’, in Dits et écrits 1954–
1988, vol. iv, 1980–1988, ed. Daniel Defert and François
Ewald. Paris: Gallimard, 1983/1994, 431–57.
‘Sur les façons d’écrire l’histoire’, in Dits et écrits 1954–1988,
vol. i, 1954–1969, ed. Daniel Defert and François Ewald.
Paris: Gallimard, 1967/1994, 585–600.
‘La vie: l’expérience et la science’, in Dits et écrits 1954–
1988, vol. iv, 1980–1988, ed. Daniel Defert and François
Ewald. Paris: Gallimard, 1984/1994, 763–76.
Freedom is a concept that it is repeatedly used today in discourses
ranging from political philosophy and rhetoric to self-help guides, yet
it seems that it has never been less clear what it means. This is not only
due to conceptual confusion or lack of philosophical precision. The
effects of the rapid process of economic and cultural globalization have
made many of our traditional ways of thinking and living redundant,
and have raised critical questions about our ‘freedom’ to command our
lives. On the other hand, neo-liberalism and the extreme individualism
characterizing our culture have made ‘freedom’ itself a contestable
One strand in this present ‘crisis’ of freedom is the critique of
an autonomous subject which characterizes post-structuralist thinking.
Michel Foucault’s thought – and post-structuralist thinking as a whole –
is often read as a rejection of the subject. This ‘rejection’ is interpreted
in varying terms. The subject cannot ground knowledge, meanings or
morality. It is not the agent of social or epistemic changes, but rather
the effect of them. There is no subject in itself prior to the normalizing
cultural coding that turns the human being into a subject. All possible ways to comprehend oneself and to act in a coherent fashion are
conditioned by a historically varying cultural matrix.
The charges against Foucault’s thought in contemporary debates
often focus on the question of the freedom of the subject and the
notions that are understood as intrinsically tied to or dependent on
it: autonomy, authenticity, responsibility, political agency. According to
many of Foucault’s critics, the denial of an autonomous subject leads
to the denial of any meaningful concept of freedom, which again leads
to the impossibility of emancipatory politics. When there is no authentic subjectivity to liberate, and power, as the principle of constitution,
fo u c au lt o n fr ee dom
has no outside, the idea of freedom becomes meaningless. Since we
are always the products of codes and disciplines, the overthrow of constraints will not free us to become natural human beings. Hence, all
that we can do is produce new codes and disciplines.1
I will argue that, rather than dismissing post-structuralist thinking as
politically dangerous and trying to hold on to the autonomous, humanist subject for political or simply conservative reasons, it is more fruitful
to take seriously the major impact post-structuralist thought has had on
our ways of thinking about the subject, and also to try to rethink freedom. The post-structuralist understanding of the subject clearly makes
problematic many of our traditional and accepted ways of conceiving
of freedom. It cannot be understood as an inherent capacity or characteristic of the subject. We cannot say that we are born free. Neither can
freedom be linked to emancipation: it does not lie in finding our true
or authentic nature and liberating it from the constraints of power or
society. For Foucault, freedom is not the freedom of protected rights
that must be safeguarded. Neither does there seem to be much point
in arguing that it is the ability to choose between different courses of
action and to govern oneself autonomously, if our choices themselves
are culturally constituted. Freedom cannot be conceived of negatively
either: it cannot be linked to the ability to think or act despite external
constraints, when the external constraints are understood as the condition of possibility of subjectivity. I will show that Foucault’s thought,
however, opens up alternative ways of thinking about freedom. It provides us with important tools for trying to answer the question, perhaps
more burning than ever: what is freedom?
While it is thus strongly argued by many commentators that there
is no freedom in Foucault’s thought, at the same time, and seemingly
paradoxically, others argue that the main motive and theme in his work
is precisely freedom. Gary Gutting (1989, 1), for example, writes that
Foucault’s thought is a search for ‘truths that will make us free’. John
Rajchman (1985, 50) claims that Foucault is ‘the philosopher of freedom in a post-revolutionary time’. Given the obvious differences in commentators’ understandings of philosophy, and of Foucault’s thought in
particular, it seems plausible to look for the source of the contrasting
interpretations in the different ways of understanding freedom in his
philosophy. My work will explicate the different meanings of freedom
that can be found in Foucault’s works, and inquire into the possibilities
he opens up for us in thinking about freedom today.
1 See e.g. Walzer 1986, 61.
in tro d u c t io n
Before focusing on the topic of freedom, I will explicate the understanding of the subject to which the question of freedom in Foucault’s
thought is essentially tied. When it is argued that there is no freedom
in it, the argument rests on the claim that there is no autonomous
subject. When, on the other hand, it is argued that freedom is what
Foucault’s thought is fundamentally about, it is often claimed that this
is due to the fact that his work reveals constraining forms of subjectivity
as historically contingent.
Foucault himself claimed that the general theme of his research was
the subject (e.g. SP, 208). Even though many commentators argue that
his own interpretations of his work were continuously changing, not
compatible, and were therefore not to be trusted,2 I take this claim to
be significant. I will argue that Foucault’s archaeologies and genealogies not only contain implicit assumptions and presuppositions about
the subject while their actual objects of study, focus and domain are
elsewhere – for example, systems of thought, power, social history – but
that they also contain explicit efforts to rethink the subject. Foucault
characterized his work as a genealogy of the modern subject: a history of
the different modes by which, in our culture, human beings are made
subjects. He further distinguished three modes of objectification that
transform human beings into subjects. These modes correspond with
three relatively distinct periods in his thought (SP, 208.)
The first is the modes of inquiry that give themselves the status of
science. Human beings are turned into subjects in processes of scientific study and classification, for example, into speaking subjects in
linguistics, subjects who labour in economics, subjects of life in biology.
Foucault’s archaeology deals with this first mode in analyzing systems of
knowledge. In The Order of Things he showed how the discourses of life,
labour and language historically developed and structured themselves
as sciences, and how human sciences further constituted man as their
object of study.
The second phase of Foucault’s work, his genealogies, studied what
he himself called ‘dividing practices’ (SP, 208). These are practices of
manipulation and examination that classify, locate and shape bodies
in the social field. His books Discipline and Punish and the first volume of The History of Sexuality are inquiries into this second mode of
objectification. He shows how modern disciplinary technologies constitute the subject as their object of control: human beings are examined, measured and categorized. This process defines them as modern
2 See e.g. Hoy 1986, 2.
fo u c au lt o n fr ee dom
individuals. The disciplinary mechanisms do not shape subjectivity only
by external coercion; they also function through being ‘interiorized’.
In The History of Sexuality, for example, Foucault shows how our belief in
a true sexual nature is a disciplinary mode of knowledge that makes us
objects of control as well as subjects of sexuality. Our self-understanding,
sexuality and even embodiment are constituted by the normative ideas
of what is healthy, true and beautiful.
The third phase of Foucault’s work, represented by volumes ii and iii
of The History of Sexuality, studies the way the human being turns himself
or herself into a subject. It is an analysis of the subject’s relationship
to itself in the domain of sexuality. He asks how human beings recognize and constitute themselves as subjects of sexuality. The subject’s
self-understanding and relationship to the self are important dimensions in the constitution of forms of subjectivity. The subject is studied
now not only as an effect of power/knowledge networks, but also as
capable of moral self-reflexivity – critical reflection on its own constitutive conditions – and therefore also of resistance to normative practices
and ideas. Subjects constitute themselves through different modes of
self-understanding and self-formation.
Foucault’s ‘ethical turn’ does not essentially change his understanding of the subject, however, it is only the perspective that shifts. He still
denies the autonomy of the subject: the subject is always constituted in
the power/knowledge networks of a culture, which provide its conditions of possibility. The modes of self-knowledge and techniques of the
self that subjects utilize in shaping themselves as subjects of sexuality,
for example, are not created or freely chosen. Rather, they are culturally and historically intelligible conceptions and patterns of behaviour
that subjects draw from the surrounding society. Self-understanding is
internally tied to historically varying social and discursive practices –
techniques of governmentality. The governing of oneself is tied to the
governing of others.
My study of Foucault’s understanding of the subject is traversed by
two axes: feminist philosophy and phenomenology. Phenomenology
acts as Foucault’s interlocutor and as a point of comparison. Feminist
philosophy traverses the work in the sense that it motivates the questions
I pose to Foucault. Even though my starting point is Foucault’s thought,
the aim is also at reappropriation, bringing it closer to my own questions
and concerns stemming initially from feminist philosophy. Rethinking subjectivity is essential in feminist philosophy, as several feminist
writers have argued. Rather than arguing that women too are subjects
in tro d u c t io n
when the subject is understood as the independent, autonomous and
rational subject of the Enlightenment, a lot of contemporary feminist
theoreticians consider it important to question traditional notions of
subjectivity. Theorists from diverse philosophical frameworks – such as
Luce Irigaray, Donna Haraway and Rosi Braidotti – have argued that
we should not seek to simply redefine the subject in neutral terms, but
we should rather define a whole new sense of subjectivity: both problematize the subject and embrace a new subjectivity for women. While
the Enlightenment characteristics of the subject – autonomy, independence and rationality – have been firmly associated with masculinity,
Braidotti’s nomadic subjects, Haraway’s cyborgs and Irigaray’s images
drawn from female morphology represent new figurations aiming to
subvert traditional imaginings of female subjectivity.3
Feminist theory does, on the one hand, share with Foucault and poststructuralist thought the aim of rethinking the subject of the Enlightenment. On the other hand, the idea of feminist emancipation is both
historically and theoretically connected to the Enlightenment ideal of
freedom: the autonomy of subjects. This tension between the modernist legacy of feminist theory and its radical challenging of some of
the most fundamental assumptions of the Enlightenment characterizes much of the contemporary feminist debate. This debate has often
been cast in terms of feminism versus postmodernism. Since there
is no consensus over the meaning of either one of these terms, the
debate has taken many different forms, ranging over diverse issues and
positions.4 My aim is not to take part in it, but rather to study critically some of the underlying ideas constitutive of the tension between
emancipatory politics and post-structuralist understanding of the
The feminist task of rethinking female subjectivity is often understood as one of finding an in-between position: we must manage to
argue for the culturally constituted status of female subjectivity without
losing agency, singularity or the Enlightenment values of freedom and
3 See e.g. Irigaray 1977/1985; 1984/1993, Haraway 1991, Braidotti 1994.
4 The debate involves large epistemological questions about how postmodern feminist critiques of objectivity can avoid falling into relativism; debates about whether the concept
of gender functions as a false generalization transcending boundaries of culture, class
and race, or as a unifying and empowering notion; aesthetic issues contesting the borders
between high and mass culture; analyses of the material changes involved in postmodernism, for example, in the structure of the family, and in work and class distinctions.
See e.g. Butler 1990, Haraway 1991, Hekman 1990, Nicholson 1990, Braidotti 1991 and
fo u c au lt o n fr ee dom
equality motivating the feminist movement. Susan Hekman (1990, 81),
among others, has argued that Foucault’s understanding of the subject
is a fruitful approach for feminist theory for the very reason that it manages to question the dichotomy between a constituting subject that is
autonomous and active versus a constituted subject totally determined
by external circumstances. Hekman claims that Foucault’s conception
of the subject avoids the eclecticism of many feminist approaches by
describing a subject that is capable of resistance and political action
without any reference to elements of a disembodied and autonomous
Cartesian subjectivity.5 I agree with Hekman that Foucault’s understanding of the subject may well provide a fruitful point of departure
in feminist efforts to rethink subjectivity, but my stance is more critical. I will argue that Foucault managed to retain the subject’s capacity
for resistance, self-reflection and criticism, but only by leaving open
important questions. My work will explicate these questions and discuss
the problems involved in answering them. I will also argue that when
Foucault’s thinking about the subject is applied to feminist theory, the
question of female emancipation has to be rethought.
Another axis in my study of Foucault, in addition to feminist philosophy, is phenomenology. Foucault is normally presented as being
in opposition to phenomenology, both to its fundaments in Husserl’s
thought and to existentialist reinterpretations.6 The common claim is
that he rejected Husserl’s transcendentalism and focused on concrete
historical facts. He did not align himself with Husserl and his philosophy of transcendental (inter)subjectivity, but rather followed Nietzsche
and the ‘postmodern’ thinkers celebrating the death of the subject,
meta-narratives and reason.
My study questions this common understanding of Foucault’s relationship to phenomenology. I will argue that the simple opposition is
based on a narrow reading of phenomenology, and on a simplification of Foucault’s thought, and that there are interesting connections
between Foucault and phenomenology which are not adequately understood. I will show that, although Foucault clearly rejected existentialist
readings of phenomenology, he did not deny all links to it. The aim
of his critique of phenomenology was rather to reveal its problems (as
he saw them), and to deal with them through a different approach.
5 On Descartes’ conception of the body and feminist critiques of Cartesian mind–body
dualism, see e.g. Reuter 2000, Judovitz 2001.
6 The few exceptions are e.g. Mohanty 1997, Flynn 1997, Han 1998/2002, Visker 1999.
in tro d u c t io n
I will argue that Foucault’s thought links up with the phenomenological tradition in at least two senses: (1) it is a critical inquiry into the
conditions of possibility of knowledge and the historicity of reason;
and (2) as a philosophical study of the subject, it is an effort to rethink
critically the phenomenological subject.
It may seem difficult to defend a view linking Foucault’s thought
to phenomenology, given the fact that he explicitly distanced himself
from it in various texts and interviews. The Order of Things, for example,
contains explicit criticism, which I discuss in chapter 2. In his introduction to the English translation, he furthermore presents his whole
method specifically as an alternative and antidote to phenomenology.7
His criticism of phenomenology in OT is, however, partly self-criticism.
Foucault’s first published works – a monograph Maladie mental et personalité (1954) and an introduction to the French translation of Ludwig
Binswanger’s Dream and Existence (1954) – were both strongly influenced by existential phenomenology. He argues in the first edition
of Maladie mentale et personalité that to understand mental illness we
have to take into account the lived experience of the patient, we need
‘a phenomenology of mental illness’. The second edition, published
in 1962, was radically rewritten. Keith Hoeller (1993) notes that it
reflects the views of mental illness that Foucault put forth in Madness
and Civilization in 1961: we need a historical study of madness. Hoeller
dates the marked turn in Foucault’s thought from the lived experience to a broader historical and political analysis of its preconditions in
these intervening years. Foucault himself describes his turn away from
I belong to the generation who as students had before their eyes, and
were limited by, a horizon consisting of Marxism, phenomenology, and
existentialism . . . at the time I was working on my book about the history of
madness [Folie et déraison]. I was divided between existential psychology
and phenomenology, and my research was an attempt to discover the
extent that these could be defined in historical terms . . . That’s when I
discovered that the subject would have to be defined in other terms than
Marxism or phenomenology.
(PS, 174)8
7 See OT, xiv.
8 The French original is not available. Wherever possible, I will give the French or German
original of the long English citations in a footnote.
fo u c au lt o n fr ee dom
Hence, while it is uncontestable that Foucault was a critic of phenomenology and not a phenomenologist, phenomenology nevertheless
forms an important background from which he sought to differentiate
and distance his own thought. He started from phenomenology, but
he also significantly returned to it in his late texts by reformulating his
relationship to it: it no longer appears in terms of an opposition, but is
rather presented as a continuum. In a text on the Enlightenment written in late 1970s, he turns to Husserl’s late writings, reading him not
essentially as presenting a philosophy of the subject, but as inquiring
into the legitimacy of reason. Foucault associates the Enlightenment
firmly with critique, a critical attitude that questions not only obstacles
to the use of reason, but also reason itself and its limits. According to
Foucault, this critical attitude took the form of questioning reason in
its connection with power, ‘the relationships between the structures of
rationality which articulate true discourse and the mechanisms of subjugation which are linked to it’ (WC, 45). Foucault saw the critique of
reason as responsible for excesses of power taking different forms in the
history of philosophy from the Hegelian left to the Frankfurt School.
Husserl is also used as an example here, who, according to Foucault,
referred to the crisis of European humanity as something that involved
the changing relationship between knowledge and technique. Foucault
considered Husserl’s thought as importantly questioning rationalization and hence studying reason as a historical phenomenon.
In an introduction to the English translation of Georges Canguilhem’s The Normal and the Pathological, Foucault argues that, in his
late works, Husserl was not asking traditional epistemological questions about the universal nature of knowledge or its timeless conditions of possibility, but he was rather posing a critical question about
our epistemic history as well as about our present reality.9 He thereby
situates Husserl in the tradition of thought that questioned western
rationality about its claims of universality and autonomy, and hence
penetrated the historico-critical dimension of philosophy. Foucault
9 Foucault distinguishes two different modalities according to which French thinkers appropriated Husserl’s thought after his Paris lectures in 1929. One was the existentialist reading of Sartre, which took Husserl in the direction of a philosophy of the subject, and the
other was Cavaillès’ reading, which, according to Foucault, brought it back to its founding
principles in formalism and the theory of science (INP, 8–9). Foucault situates his own
thought in the tradition of Cavaillès, which developed as the history of thought and the
philosophy of science.
in tro d u c t io n
And if phenomenology, after quite a long period when it was kept at the
border, finally penetrated in its turn, it was undoubtedly the day when
Husserl, in the Cartesian Meditations and the Crisis, posed the question of
the relations between the ‘western’ project of a universal development
of reason, the positivity of the sciences and the radicality of philosophy.
(INP, 11)10
According to Foucault, the critique of rationality led Husserl to develop
a new mode of questioning (VES, 767). Husserl did not just study the
universal structures of knowledge, he also proposed an inquiry into the
historical meaning of knowledge, that is, into the meaning that the ideas
of science and philosophy have for us now, at this very moment. Foucault
thus considered his thought to be in line with phenomenology to the
extent that the answers to the question ‘What is philosophy?’ would
be similar: philosophy is understood essentially as a critical practice
responding to our present. It is, however, not only critical towards other
forms of knowledge or practices of living, but it is also significantly selfcritical. It must turn to question its own conditions of possibility, the
legitimacy of reason and its own historicity.
I will argue that understanding Foucault’s background in phenomenology and relating his work to it is important for understanding
his philosophical position. I will show how many of Foucault’s central
philosophical issues and methodological directions are motivated by
the problems arising out of the phenomenological enterprise. By constructing a dialogue in this book between Foucault and three major phenomenological thinkers – Husserl, Merleau-Ponty and Levinas – I aim to
bring to light some common forms of questioning and points of fruitful
exchange as well as of fundamental contrast. By focusing on Foucault’s
relationship to phenomenology, however, I do not want to claim that
it is the only or even the most important influence on his thought.
His work had many different themes and influences: Nietzsche’s
philosophy, structuralism, French historiography and philosophy of science, for example. Any study of Foucault’s thought representing one
choice of many possible perspectives is therefore a distortion of his
multifaceted and original thought.
10 ‘Et si la phénoménologie, après une bien longue période où elle fut tenue en lisière,
a fini par pénétrer à son tour, c’est sans doute du jour où Husserl, dans les Méditations
cartésiennes et dans la Krisis, a posé la question des rapports entre le projet occidental
d’un déploiement universel de la raison, la positivité des sciences et la radicalité de la
philosophie.’ (IMF, 432)
fo u c au lt o n freedom
This book is divided into three parts: Language (chapters 1, 2, and 3),
Body (chapters 4, 5, and 6) and Ethics (chapters, 7, 8, and 9). These three
parts explicate the three constitutive modes of subjectivity in Foucault’s
thought, and they also correspond loosely with the three chronological
periods in it: archaeology, genealogy and his late writings on ethics.
The structure of the book is primarily thematic, however. I do not offer
a chronological reading of the development of Foucault’s thought, or
a philosophical reconstruction of ‘Foucault’s theory of the subject’.
Instead, I ask what freedom means at different points in his work and
study its preconditions as well as its problems. My argument is that
language, the body and ethics are the domains in which the different senses
of freedom can be found. My focus on certain Foucault texts, and the
omission of others, are based on this thematic priority.
The first part of the book, Language, inquires into the idea of freedom present in Foucault’s archaeology. The focus of my reading is
on The Order of Things, which studies the question of language most
explicitly. I explicate Foucault’s philosophical position by contrasting it
to Husserl’s phenomenology, particularly to his late work, The Crisis of
European Sciences and Transcendental Phenomenology. This is illuminative
in terms of understanding the philosophical implications of Foucault’s
treatment of the history of science in OT. I also make a stronger claim
about the importance of reading OT in relation to phenomenology. I
will show how many of the central philosophical issues, as well as the
methodological directions, that are present in OT are motivated by the
problems arising out of the phenomenological enterprise.
In OT Foucault advocates the idea of language as something that
always outruns the subject, who can never completely master it. Language is not simply an instrument of expression, it also generates an
excess of meanings. Foucault gives language a regulative role in the
mode of scientific discourse, but it also demarcates a domain of freedom in the mode of literature, particularly as avant-garde writing. There
is an ontological order of things implicit in the theories of scientific discourse. Language as avant-garde writing is, however, capable of forming alternative, unscientific and irrational ontological realms: different
experiences of order on the basis of which different perceptual and
practical grids become possible, and hence lead to new ways of seeing
and experiencing. While Foucault’s archaeology is generally viewed as
emphasizing the necessary structures of thought and opposing humanist aspirations of looking for the freedom of man, there is an antihumanist understanding of freedom as an opening of new possibilities
in tro d u c t io n
of thought implicitly safeguarded in archaeology. Foucault tried to show
not only how the limits of knowledge, representation and experience
are constituted, but also what distorts them. It is not that freedom characterizes the subject, it characterizes language in the mode of literature,
which is capable of undermining the stability of our ways of seeing,
understanding and acting.
The second part of the book, Body, explicates the genealogical approach and engages with feminist appropriations of Foucault.
Foucault’s understanding of the historical construction of the body
through mechanisms of power has influenced feminist theory profoundly, but the feminist appropriations are based on varying readings
of what exactly Foucault meant. He did not present a theory of the
body anywhere, not even a unified account of it, and thus his view has
to be discerned from his genealogical books and articles, which aim
at bringing the body into the focus of history. I will discuss Foucault’s
understanding of the sexual body in volume i of The History of Sexuality,
and ask what he meant by the idea of bodies and pleasures as a form
of resistance to power. I argue that we must take this idea seriously and
not simply dismiss it as a naive fall back to the prediscursive body. I will
show that if we are to understand Foucault’s idea of bodily resistance,
we must leave behind the conception of the body as a mere material or
bio-scientific object. The question of resistance only opens up if we take
the experiential body – the body as experiencing in everyday practices –
as the starting point. I will argue for a certain kind of irreducibility of
embodied experience which does not mean that experience is treated
as ahistorical or prediscursive, but that it can never be wholly reduced
to discursive meanings. Elisabeth Grosz writes (1995, 222) that a distinction must be drawn between discourse and experience, even given
the understanding that language or systems of representation are the
prior condition for the intelligibility of experience. I argue that this
distinction is crucial for understanding the resistance of the body. The
experiential body is the permanent contestation of discursive definitions, values and normative practices. Even if linguistic intelligibility
structures and constitutes the limits of experience, these limits are never
firmly set and are constantly transgressed. The existence of the abject
and the mute is furthermore necessary for constituting the intelligible
and the normal. Foucault’s genealogy, like his archaeology, thus also
displays a dimension of freedom in the sense of a constitutive outside
to the discursive order: embodied experiences capable of distorting,
multiplying and overflowing definition, classification and articulation.
fo u c au lt o n freedom
The idea of understanding the Foucaultian body as the experiential
body raises the question of its relationship to the lived body as described
by the phenomenologists. Foucault’s dialogue with phenomenology,
which features in part ii of this book, thus takes place with MerleauPonty. Chapter 6 presents a Foucaultian reworking of Merleau-Ponty’s
lived body, and I argue that this could provide feminist theory with
new ways of understanding ‘female freedom’. Emancipation requires
a body that is conceived in ways that are open to reinterpretations and
multiple meanings, rather than one that is pure of cultural inscription.
Part iii, Ethics, discusses the understanding of freedom that is to be
found in Foucault’s late work. His thinking on ethics elaborates his
understanding of freedom by introducing a deliberate dimension to it:
ethics as care for the self is the deliberate part of freedom. Freedom
is not only a non-subjective opening of possibilities, but it can also
be deliberately cultivated and practised by its subjects. Subjects exercise freedom in critically reflecting on themselves and their behaviour,
beliefs and the social field of which they are a part. They materialize and
further stylize the possibilities that are opened around themselves. Care
for the self as a practice of freedom means challenging, contesting and
changing the constitutive conditions of subjectivity as well as its actual
forms. It means exploring possibilities for new forms of subjectivity, new
fields of experiences, pleasures and relationships, and modes of living
and thinking. The quest for freedom in Foucault’s ethics becomes a
question of developing forms of subjectivity that are capable of functioning as resistance to normalizing power.
Foucault’s late essays on the Enlightenment also postulate the idea
of freedom as a historical ethos originating from the Enlightenment:
the championing of political freedom in the modern sense cannot be
found in any pre-Enlightenment tradition of thought. Foucault does
not simply embrace traditional Enlightenment ideas, however, he also
submits them to critical reappropriation. What, for him, characterized
the philosophical ethos originating in the Enlightenment is that it is a
permanent critique of our own philosophical era. By linking his thought
to the Enlightenment, he made the normative move of adopting the
values associated with it – critical reasoning and personal and political
autonomy – as the implicit ground on which his critiques of domination, abusive forms of power and rationality rest. The Enlightenment
provided him with the historical – not transcendental – values on which
to base his implicit critiques. Its ethos represents a commitment to a
specific historical tradition within which we think about human life and
in tro d u c t io n
politics, and freedom as the autonomous use of reason is a precondition
of critical reflection on our present.
The last chapter of this book, focuses on the question of the other in
Foucault’s ethics. I explicate two interconnected problems that I claim
are riddling his understanding of the ethical subject: the autonomy
of the subject and the role of the other in its constitution. I will show,
through a reading of Emmanuel Levinas, how these problems are interconnected. With the help of Levinas, I argue that we must acknowledge
the fundamental importance of the other person for the constitution
of ethical subjectivity.
To discuss the centrality of freedom in Foucault’s thought is to situate
him in a long lineage of thinkers who have held freedom as a central
theme in philosophical reflection. It has been argued that an emphasis
on freedom characterizes French philosophy in particular and also lies
in France’s revolutionary heritage.11 The discussion of freedom in the
context of political philosophy forms only one fraction of its pervasive
importance in the French philosophical tradition, however. Descartes
understood freedom most of all as an epistemological question: freedom is the condition of possibility of knowledge because it enables us
to refrain from believing things that are not completely certain and
hence to avoid error. Sartre saw freedom primarily as an ontological
question: in Being and Nothingness freedom is a structure of the being
of man and thus forms a part of a comprehensive ontology. By tracing
the meanings of freedom in the different phases of Foucault’s thought
I will argue that it is best understood as an epistemological and ontological question also in his thought, rather than as a strictly political
question. Freedom lies in the ontological contingency of the present,
in the unpredictability of our ways of thinking, acting and relating to
other people. Thinking can only take place within a historically defined
framework of rules and practices, but this framework is never a necessity. Philosophy as an ontology of the present is the historical analysis of
the limits that are imposed on us and an experiment with the possibility
of crossing them (WE, 50). Its aim is to free us from ourselves: shatter
our preconceptions concerning the present by showing the historical
contingency of our thought and practice.
11 See e.g. Cohen 1997, Gutting 2001, 382.
The madman jumped into their midst and pierced them with
his eyes. ‘Whither is God?’ he cried; ‘I will tell you. We have
killed him – you and I. All of us are his murderers. But how
did we do this? How could we drink up the sea? Who gave
us the sponge to wipe away the entire horizon? What were we
doing when we unchained this earth from its sun? Whither is it
moving now? Whither are we moving? Away from all suns? Are
we not plunging continually? Backward, sideward, forward,
in all directions? Is there still any up or down? Are we not
straying as through an infinite nothing? Do we not feel the
breath of empty space? Has it not become colder? Is not night
continually closing in on us? Do we not need to light lanterns
in the morning? Do we hear nothing as yet of the noise of the
gravediggers who are burying God? Do we smell nothing as
yet of the divine decomposition? Gods too decompose. God
is dead. God remains dead. And we have killed him.’
(Nietzsche 1887/1974,181)
The instant bestseller that made Foucault famous, The Order of Things
(OT), arose out of laughter. Foucault opens the book by writing that the
book arouse out of a passage in Borges, from a laughter that shattered
all the familiar landmarks of his thought and continued long afterwards
to disturb and threaten with collapse the age-old distinction between
the Same and the Other. This passage quoted ‘a certain Chinese
encyclopedia’ which presented a wholly other system of thought and
therefore broke up ‘all the ordered surfaces and all the planes with
which we are accustomed to tame the wild profusion of existing
things’ (OT, xv). The shattering impossibility to think in certain ways
fo u c au lt o n freedom
motivated Foucault’s philosophical inquiry in OT and led him to ask:
what is it possible to think?
Foucault’s archaeological question about the limits of thought resembles the transcendental question of phenomenology concerning the
conditions of possibility of knowledge: what are the conditions of possibility of our experiences, the constitution of objects and modes of
knowledge? I will argue that, although the mode of questioning is similar, the methodological route followed by Foucault diverts sharply from
phenomenology. Hence, the methodological question that must follow
the transcendental one – how can we study the conditions of possibility
of our thought since, by definition, we cannot set ourselves outside of
it – is given a very different answer by Foucault.
I will also show how, in Foucault’s thought, the question of the limits
of thought, by implication, must be followed by the question of what
falls outside the discursive limits of being. I will argue that, by explicitly
questioning the conditions of possibility of knowledge, Foucault also
implicitly inquires into an otherness that escapes the existing order
imposed upon being by scientific discourse. Even after being brought
within the order of the same, this otherness is still capable of forcing a
shattering laughter upon us.
Foucault points to his phenomenological background himself in several contexts. In an essay entitled ‘Subjectivity and Truth’ (ST), he
explicitly describes the development of his thought – the genealogy
of the modern subject – as a reaction to phenomenology.1 He claims
that as the intellectual climate in France became more critical of phenomenology in the 1960s, two hitherto hidden theoretical paradoxes
implicit in phenomenology could no longer be avoided. The first one
was that phenomenology had failed to found a philosophy of scientific
knowledge, and the second that it had failed to take into account the
formative mechanisms of signification and the structure of systems of
meaning. Different approaches developed in France as efforts to deal
with these paradoxes: logical positivism, semiology, Marxism, psychoanalysis and structuralism (ST, 174–5).
The close relationship between Foucault’s archaeology and phenomenology has also been noted by some of Foucault’s commentators.2 Beatrice Han argues in her book Foucault’s Critical Project that
a profound reading of Foucault’s relationship with phenomenology
1 See also e.g. SPS, 442.
2 See e.g. Dreyfus and Rabinow 1982, Gutting 1989, Lebrun 1989/1992.
ph ilo so ph ic al lau g h t e r
shows the continuity of the two projects as well as their roots in a common Kantian origin. According to Han, for Foucault, phenomenology
is both an interlocutor and a favoured target, because it also attempts
to overcome the obstacle of pure transcendentalism. Despite appearances, archaeology is profoundly connected to phenomenology in that
it attempts to find a solution to the same problem and adopts a method
that is similar in aspects such as its descriptive rather than explicative
outlook (Han 1998/2002, 5).
This first chapter presents a concise explication of the main philosophical arguments in OT. The explication prepares the way for the
reading put forward in the second chapter, which contrasts OT to
Husserl’s thought, particularly to his late work, The Crisis of European
Sciences and Transcendental Phenomenology. Following the presentation
of the negative or critical side of archaeology as an effort to overcome
the problems inherent in phenomenology, the third chapter evaluates
the positive contribution of Foucault’s archaeology to philosophical
questioning. It focuses on the topic that is the overriding theme of
the present book: the question of freedom in Foucault’s thought. I will
argue that OT contains an idea of otherness – a realm outside the discursive order of things – which represents freedom for Foucault. This
possibility of an otherness outside of what can be scientifically ordered
and brought within the realm of knowledge does not lie in some pure
and prelinguistic sphere of originary experience, however, but in language itself.
An archaeology of order
Foucault argues in OT that there is a level of order, ‘a positive unconscious of knowledge’, that eludes the consciousness of the scientist and
yet is formative of scientific discourse. This level gives us the organizing principles of knowledge, the unconscious structures that order scientific discourses. Even though individual scientists never formulated
these principles, nor were they even aware of them at the time, this
level nevertheless defines the objects proper to their study. It constitutes the condition of possibility for forming concepts and building
theories (OT, xi). Hence, beyond the level of scientific discoveries, discussions, theories and philosophical views exists an archaeological level
formative of them. This level consists of the ontological order of things
assumed to exist, and also of the principles that organize the relationships between things and words in terms of what exists and how it
fo u c au lt o n freedom
can be known and represented. It is the historical a priori of scientific
This a priori is what, in a given period, delimits in the totality of experience
a field of knowledge, defines the mode of being of the objects that appear
in that field, provides man’s everyday perception with theoretical powers,
and defines the conditions in which he can sustain a discourse about
things that is recognized to be true.
(OT, 158)3
Foucault’s objective in OT is to study this historical a priori, the formative
level of scientific discourse. His method is a historical inquiry into the
ordering codes of our culture since the sixteenth century: how the space
of knowledge has been arranged, how discursive practices have been
structured. By studying how scientific discourses have been ordered, he
seeks to reveal or unveil the ontological order of things assumed to exist
prior to them. This order behind ordering fundamentally but ‘unconsciously’ forms and structures scientific discourse. Foucault describes
his analysis:
Quite obviously such an analysis does not belong to the history of ideas
or of science: it is rather an inquiry whose aim is to rediscover on what
basis knowledge and theory became possible; within what space of order
knowledge was constituted; on the basis of what historical a priori and in
the element of what positivity, ideas could appear, sciences be established,
experience be reflected in philosophies, rationalities be formed, only,
perhaps, to dissolve and vanish soon afterwards.
(OT, xxii)4
The question that guides Foucault’s archaeology is thus a transcendental question in the sense that it concerns the condition of possibility
of knowledge: what determines different forms of scientific knowledge
and makes possible certain discussions and problems? The condition
3 ‘Cet a priori, c’est ce qui, à une époque donnée, découpe dans l’expérience un champ
de savoir possible, définit le mode d’être des objets qui y apparaissent, arme le regard
quotidien de pouvoirs théoriques, et définit les conditions dans lesquelles on peut tenir
sur les choses un discours reconnu pour vrai.’ (MC, 171)
4 ‘Une telle analyse, on le voit, ne relève pas de l’histoire des idées ou des sciences: c’est
plutôt une étude qui s’efforce de retrouver à partir de quoi connaissances et théories
ont été possibles; selon quel espace d’ordre s’est constitué le savoir; sur fond de quel a
priori historique et dans l’élément de quelle positivité des idées ont pu apparaı̂tre, des
sciences se constituer, des expériences se réfléchir dans des philosophies, des rationalités
se former, pour, peut-être, se dénouer et s’évanouir bientôt.’ (MC, 13)
ph ilo so ph ic al lau g h t e r
of possibility must, however, in connection with Foucault’s archaeology, be understood in a specific way. It is not universal a priori, but
conditions that are importantly historical: they are formed in history
and also change in it. Neither can they be found through an analysis
of the subject’s faculties or experiences. They condition, significantly,
also the subject’s experiences. By revealing the conditions of possibility
of the thought of a particular period, Foucault seeks to reveal the nonsubjective conditions that make subjective experiences of order and
knowledge possible.5 His aim is thus to write a history of the transcendental:
a historical description of the varying conditions of possibility of knowledge in different periods.
Foucault introduces the concept of episteme (épistémè) in OT. There
has been much discussion and confusion about what exactly he means
by this, and he himself has offered different definitions.6 One common
but mistaken reading is to understand it to simply refer to the existing field of different forms of knowledge in a specific period. Claire
O’Farrell, for example, describes episteme as the differing configurations of knowledge at different periods.7 Understood in this way,
the main function of the concept would be to enable Foucault to
locate the points of discontinuity in the western history of thought.8
Foucault’s archaeological inquiry in OT distinguishes three epistemes,
and analyzes the epistemic systems underlying three historical epochs:
the Renaissance, the classical age and modernity.
The notion of the episteme is not only a tool for writing the history of
thought and describing ruptures in it. I argue that it is, most of all, a tool
for understanding the historical conditions of possibility of knowledge
5 Beatrice Han (1998/2002) shows that Foucault’s inquiry into the historical a priori of
knowledge is a search for a principle of determination, which is non-subjective. It has
‘the function of introducing into the field of knowledge a principle of nonsubjective
determination, which defines for a given period and geographical area the historical
form taken by the constitution of various forms of knowledge’ (45).
6 The notion of episteme presented in OT was widely criticized for being a totality as well
as a static notion that excludes change. In the book that followed OT, The Archaeology of
Knowledge (AK), Foucault answers this criticism and modifies his stance. He denies that
there is one episteme for the science of a particular period: the relations that he describes
are valid only in order to define a particular configuration (AK, 159). He also writes that
episteme is not an immobile figure but rather ‘an infinitely mobile group of scansions,
shifts, and coincidences which establish and dismantle themselves’ (AK, 250).
7 O’Farrell (1989, 54–5) argues that when the notion of episteme is introduced for the
first time in OT, Foucault uses it in two different ways: firstly to denote the entirety of
western knowledge, and secondly to describe different configurations of knowledge at
different periods.
8 See e.g. Machado 1989/1992, 14.
fo u c au lt o n freedom
in a particular period. Hence, episteme refers to the historical a priori
of an epoch. Rather than referring to the different configurations of
knowledge, it refers to their historical conditions of possibility. Foucault
writes, for example: ‘In any given culture and at any given moment,
there is only one episteme that defines the conditions of possibility of
all knowledge, whether expressed in a theory or silently invested in
a practice’ (OT, 168). He also describes his aim in OT in terms of
transcendental analysis.
What I am attempting to bring to light is the epistemological field, the
episteme in which knowledge, envisaged apart from all criteria having reference to its rational value or to its objective forms, grounds its positivity
and therefore manifests a history which is not that of its growing perfection, but rather that of its conditions of possibility.
(OT, xxii)9
The first break or discontinuity that Foucault’s archaeological inquiry
reveals positions the dawning of the classical age roughly halfway
through the seventeenth century. The second epistemic break took
place at the beginning of the nineteenth century and, according to
Foucault, marked the beginning of the modern age (OT, xxii). By documenting breaks or discontinuities between the epistemes, Foucault
argues against the continuous development of European science and
rationality. The history of western science cannot be understood as a
steady progression from unscientific forms of knowledge to ever more
accurate and rational methods and theories. Foucault’s aim in OT is
to show that, on the contrary, the whole order of existing things on
the basis of which we think today is radically different from that of
the classical thinkers, just as their mode of being of words and objects
was completely different from that of the Renaissance thinkers. The
epistemic changes fundamentally affect what kinds of things can be
the objects of knowledge. The fundamental changes in the mode of
being of what could be known resulted in a change in modes of knowledge to the point that some of the theories of the previous episteme
were no longer even recognized as belonging to the field of scientific
knowledge. Neither were scientific discoveries simply improvements of
9 ‘ce qu’on voudrait mettre au jour, c’est le champ épistémologique, l’épistémè où les connaissances, envisagées hors de tout critère se référant à leur valeur rationnelle ou à leur
formes objectives, enfoncent leur positivité et manifestent ainsi une histoire qui n’est pas
celle de leur perfection croissante, mais plutôt celle de leurs conditions de possibilité’
(MC, 13).
ph ilo so ph ic al lau g h t e r
previous theories. Foucault writes about the birth of natural history at
the beginning of the classical age, for example, that it ‘did not become
possible because men looked harder and more closely’, but because
the mode of being of the natural order had changed (OT, 131–2).
To better understand how modes of knowledge in Foucault’s thought
are determined by non-subjective, discursively formed historical a prioris, I will next give a short description of how his concept of episteme
functions in OT. This vaguely defined and abstract notion is fleshed
out by detailed historical studies of the birth of the different sciences.
This emphasis on history exemplifies Foucault’s method of discarding abstract conceptual definitions and instead elaborating the vague,
often frustratingly ambiguous, philosophical ideas through concrete
historical description. Unfortunately, this method also, at times, makes
it difficult to reconstruct Foucault’s philosophical position coherently.
The three epistemes
The Renaissance episteme. Foucault starts his explication of the three
epistemes with a relatively short description of Renaissance episteme.
The chapter has a telling title, ‘The Prose of the World’. Foucault’s aim
is to show that what characterized the episteme of the Renaissance was
the implicit assumption that words and things formed a unified texture
and were linked through resemblance. Adjacent things were similar or
they emulated and reflected each other from afar. Some similitudes
were not visible, but were rather subtle resemblances of relations: the
relation of the stars to the sky could be found between man’s skin moles
and the body of which they were the secret marks. Things that resembled each other were further drawn together. Through sympathy similar
things assimilated and transformed, through antipathy they maintained
their isolation (OT, 19–24). The shortness of a certain line on a man’s
hand, for example, reflected, through analogy, the image of a short life.
The analogy between body and destiny, again, indicated sympathy that
created communication between them.
Knowledge meant uncovering the resemblances that linked things
to one another in an infinite chain of similitudes. To know meant to
interpret: to find a way from the visible marks to what they resembled.
For these chains of similitudes to become knowledge they had to be
linked to words, but the way words were linked to things was, again,
through resemblance. Signs could signify only insofar as they resembled what they indicated. They signified by forming another order of
fo u c au lt o n freedom
resemblances: they were signatures of resemblances that made it possible to recognize them. According to Foucault, ‘The sixteenth century
superimposed hermeneutics and semiology in the form of similitude’
(OT, 29). To search for a meaning meant bringing to light a resemblance. To search for a law governing signs was to discover things that
were alike. Language was thus one of the things in the world to be interpreted. There was a common order for words and things, ‘the relation
of languages to the world is one of analogy rather than signification’
(OT, 37).
Because words were signs woven into the overall network of similitudes that had to be interpreted, observation, accepted authority and
magical divination were on the same level of knowledge. According to
Foucault, for Renaissance knowledge ‘there is no difference between
the visible marks that God has stamped upon the surface of the earth, so
that we may know its inner secrets, and the legible words that the Scriptures, or the sages of Antiquity, have set down in the books preserved
for us by tradition’ (OT, 33). He argues that, although it is easy for us to
interpret that rational beliefs and esoteric knowledge were in conflict
during the Renaissance, and that esotericism lost the battle as western science and rationality developed, if we study the archaeological
level of knowledge – the level that makes different forms of knowledge
possible – this does not hold. The Renaissance episteme not only made
possible the coexistence of what, for us, are rational arguments, erudition and magic, it also made it impossible to distinguish them as
different forms of knowledge, since they all relied on the same principles of organization. The prose of the world was formed as one infinite
text binding together words and things through resemblance.
The classical episteme. With the advent of the classical age there occurred
a change. At the beginning of the seventeenth century similitude ceased
to be the organizing principle of words and things. It was, rather, understood as the occasion for error, something that tempted man to draw
conclusions from the deceiving senses (OT, 51). The hierarchy of analogies was substituted by the analysis of identity and difference. Rationalism was to replace old superstitious beliefs. It was firmly believed in the
classical age that being had a universal order that could be analyzed
by a universal method and that could be represented by signs that mirrored perfectly this order of being. Knowledge was organized in a table,
it could be displayed as a perfect system. The method was to find the
simple nature of beings and proceed from them to more complex ones
ph ilo so ph ic al lau g h t e r
by comparison and ordering. If the analysis was carried out correctly, it
would yield perfect certainty. To know no longer meant to draw things
together by finding a shared nature in them, it meant to discriminate
(OT, 54–5, 75).
The relationship of words to things thus changed fundamentally.
The common order dissolved into two separate orders that were nevertheless inseparably linked in a transparent relation of representation.
There was an external order of objects and an ideal order of signs.
Signs, including language, were representations that directly represented their objects. The order of language was therefore the order
of the world and could mirror it perfectly. Knowledge became representation. Foucault writes polemically that ‘language in the classical
era does not exist’ (OT, 79). Language as an object of knowledge was
not possible. It only had a function, it represented, but it was not an
object of knowledge in itself. It was the transparent medium of representation, conveying the immutable and perfect order of the visible
[L]anguage is no longer one of the figurations of the world, or a signature
stamped upon things since the beginning of time. The manifestation and
sign of truth are to be found in evident and distinct perception. It is the
task of words to translate that truth if they can; but they no longer have
the right to be considered the mark of it. Language has withdrawn from
the midst of things themselves and has entered a period of transparency
and neutrality.
(OT, 56)10
Foucault’s point in his long discussion of the three classical forms of
knowledge – general grammar, natural history and analyses of wealth –
is to show how the same order underlies them, how they were all authorized by the same organizing principles. The form of these modes of
knowledge is that of an ordered table of representations in which different beings could be named and could find their place in a language.
General grammar was not the classical predecessor of modern linguistics because its object of study was not language as we understand it.
It was a mode of knowledge limited and structured by the classical
10 ‘le langage n’est plus une des figures du monde, ni la signature imposée aux choses
depuis le fond des temps. La vérité trouve sa manifestation et son signe dans la perception évidente et distincte. Il appartient aux mots de la traduire s’ils le peuvent; ils n’ont
plus droit à en être la marque. Le langage se retire du milieu des êtres pour entrer dans
son âge de transparence et de neutralité.’ (MC, 70)
fo u c au lt o n freedom
episteme. In the ontological order of things language did not exist as a
historical, variable object to be known.
Similarly, just as there was no language to be known, neither could
life be an object of knowledge. In terms of natural history, life did not
create a fundamental distinction in the order of things, it was not an
obvious threshold dividing certain forms of knowledge from others. It
was only one category in the classification of beings (OT, 161). According to Foucault, the idea of biology was not possible in the classical
episteme: ‘Natural history is nothing more than the nomination of the
visible’ (OT, 132). The task of natural history was to represent the visible
order of nature by ascribing a name to living beings, and in that name
to name the character that situated them within the taxonomic system
of identities and differences. The goal was to lay out a unified grid that
covered the entire vegetable and animal kingdom (OT, 141): ‘The naturalist is the man concerned with the structure of the visible world and
its denominations according to characters. Not with life’ (OT, 161).
The analysis of wealth obeys the same configuration as natural history and general grammar. In the Renaissance episteme, just as words
had the same reality as what they said, and living beings were marked
by visible signs, the signs that indicated and measured wealth carried
the marks of wealth in themselves. Money, gold and silver coins were
themselves precious and desirable. In the classical episteme, however,
money no longer had intrinsic value, but instead became the pure representation of wealth. The sign that coins bore was merely the exact and
transparent mark of the measure they constituted (OT, 170). Money
had the power to represent all possible wealth, it was the instrument
through which wealth, as well as the objects of need and desire, could be
turned into a universal system of identities and differences. Foucault’s
general claim is thus that the classical knowledge ascribed to each thing
represented a name and laid out a linguistic grid across the whole of
representation. The form of knowledge that named the being of all
representation in general was philosophy: the theory of knowledge and
the analysis of ideas (OT, 120).
The modern episteme. There occurred another a major change at the end
of the eighteenth century. Foucault claims that the visible order of
things was torn apart: a space opened up behind it that required explanation. Knowledge no longer meant perceiving and representing the
visible order in language. It now meant disclosing the invisible basis of
living beings in life, of different discourses in the being of language,
of the systems of wealth in labour. Language, life and labour become
ph ilo so ph ic al lau g h t e r
questions in their own right. Knowledge meant reconstituting the
hidden unities that underlay the dispersion of visible differences. The
consequence was the collapse of the classical episteme (OT, 268–9).
While the classical thinkers saw knowledge in the form of an ordered
table, in the modern age, according to Foucault, it was understood in
terms of series and development (OT, 262). The realm of knowledge
was no longer one of identities and differences forming one ordered
table, but an area made up of organic structures, of internal relations
between elements whose totality performs a function (OT, 218). Objects
of knowledge had become susceptible to external influences and subject to time. The historicity of empirical beings had become definitive
of knowledge in the modern age. Things were defined not by their
place in a timeless system of classification, but by their place in history.
This meant that there had to be a whole new conception of order. The
form of knowledge, as well as the mode of being of its objects, were
fundamentally altered: the modern manner of empiricism had begun
(OT, 250).
Thus, European culture is inventing for itself a depth in which what
matters is no longer identities, distinctive characters, permanent tables
with all their possible paths and routes, but great hidden forces developed
on the basis of their primitive and inaccessible nucleus, origin, causality
and history.
(OT, 251–2)11
The fundamental change from one episteme to another is again recognizable in the specific changes in the areas of knowledge that Foucault
examines. Language evolved with its own density, and therefore also
philology. Words no longer simply represented things, but for a word
to be able to convey meaning it had to belong to a grammatical totality
which, in relation to the word, was primary, fundamental and determining (OT, 281). In biology, living beings were ordered and conceived of
on the basis of functional homogeneity, not visible identities and differences. Life was a non-perceptible, purely functional aspect, it had
‘left the tabulated space of order and become wild again’ (OT, 277).
In economics, wealth was no longer considered to be distributed over a
table as a system of equivalencies, but was organized and accumulated
in a temporal sequence.
11 ‘Ainsi, la culture européenne s’invente une profondeur où il sera question non plus des
identités, des caractères distinctifs, des tables permanentes avec tous leurs chemins et
parcours possibles, mais des grandes forces cacheés développées à partir de leur noyau
primitif et inaccessible, mais de l’origine, de la causalité et de l’histoire.’ (MC, 263)
fo u c au lt o n freedom
Foucault’s point, from the perspective of the history of science, is
thus to show how modern forms of knowledge originated from a fundamental break in the history of thought and were not simply more
advanced developments of previous modes of knowledge. This enables
him to show how certain controversies and oppositions, which are traditionally regarded as fundamental by historians, are in fact part of the
same epistemic order. Although the birth of biology, economics and
philology were part of the same archaeological upheaval, and although
the epistemic forms of these disciplines are similar, what distinguishes
the epistemic structure from the classical one is that there is no longer
a homogenous or uniform basis of knowledge. It is now possible for
different empirical sciences to have different knowledge bases. While
man as both a subject and an object of knowledge provides the basis for
human sciences, natural sciences are grounded in the unproblematized
being of their object.
According to Foucault, the dispersal of the basis of knowledge in the
modern age was tied to the failure of representation. Representation
had become problematic, it had became opaque. It could no longer
define the common order of things and knowledge. The parallel orders
of words and things were diverted and torn apart: there were, on the
one hand, ‘things, with their own organic structures, their hidden veins,
the space that articulates them, the time that produces them’, and on
the other, representation, ‘a purely temporal succession, in which those
things address themselves (always partially) to a subjectivity’ (OT, 239–
40). The representing relation had become problematic. Epistemology
had taken the central place in philosophy that was previously held by
ontology. The questions that haunted philosophy in the modern age
were questions about the legitimacy and possibility of representation:
what linked things to words and propositions and provided the basis of
true knowledge?
According to Foucault, Kant was the first to question representation
on the basis of its rightful limits, and therefore Kant’s thought marks
the threshold of the modern episteme (OT, 242). Kant’s transcendental
question about the condition of possibility of knowledge marked the
end of the classical episteme and the age of representation. Questioning
the rightful limits of representation meant looking beyond the space
of representation for its condition of possibility. In Descartes’ thought,
ordered knowledge was possible because clear and distinct ideas were
in perfect correspondence with things, and this correspondence was
created by God. God was the ultimate ground of knowledge, not man.
ph ilo so ph ic al lau g h t e r
Man was simply another being in God’s well-ordered table. His place
as the organizer of the representations was never problematised in the
classical episteme. According to Foucault, man as an ordering subject
thus did not appear in western thought until Kant. In Kant’s thought,
man has taken the place of God as the organizer of the world, but he
has done this by virtue of being tied to finitude, by being limited. It is
the limits of his knowledge that make knowledge of the world possible.
Foucault describes the philosophy following Kant up to the present
as different efforts to deal with the dissolution of the homogenous field
of orderable representations. He distinguishes between two correlative
new forms of thought. On the one hand, there is the reformulation
of transcendental philosophy in the form of phenomenology: the questioning of the conditions of possibility of representation from the point
of view of the experiencing subject. On the other hand, there are the
forms of thought in which knowledge is grounded in the mode of being
of the object. Foucault writes that in the latter modes of thought, ‘the
conditions of possibility of experience are being sought in the conditions of possibility of the object and its existence whereas in transcendental reflection the conditions of possibility of the objects of experience are identified with the conditions of possibility of experience itself’
(OT, 244). Foucault refers here, on the one hand, to positivism: ‘There
are philosophies that set themselves no other task than the observation of precisely that which is given to positive knowledge’ (OT, 244).
The other possible counterpart to transcendental philosophy of the
subject are the various ‘metaphysics of the object’. For these forms of
thought there are ‘transcendentals’ – ‘Will’ for Schopenhauer, ‘Life’ for
Nietzsche – which form the conditions of possibility of the always partial
knowledge of the subject. These ‘metaphysics of the object’, according
to Foucault, posit an objective foundation behind experience, with its
own rationality that the subject is never able to bring to light completely.
The modern episteme thus has become severely fractured. There is no
one unified archaeological basis of knowledge.12
12 Foucault argues that there are, in fact, two forms of fracture. Firstly, there is the split
between forms of knowledge finding their basis in transcendental subjectivity and those
finding a basis in the mode of being of the object. Secondly, because of the emergence of
empirical fields of which mere internal analysis of representation can no longer provide
an account, there is also a split between the field of a priori sciences, pure, formal
sciences, and the domain of a posteri sciences, empirical sciences. These two forms of
fracture give rise to the modern problem concerning the relations between the formal
field and the transcendental field, the domain of empiricity and the transcendental
foundation of knowledge (OT, 248).
fo u c au lt o n freedom
Foucault’s caricatured picture of modern philosophy as following
Kant’s thought in different variations is clearly simplistic and problematic. Nevertheless, even if Foucault’s characterization of modern
philosophy is not accurate as far as other thinkers are concerned,
it is revealing in terms of his own philosophical position. Foucault
clearly sees the question of the problematic status of representation –
particularly in the form of language – as the main question of modern
philosophy, and hence as the question that his thought for its part must
address. Although Foucault rejects the solution offered by phenomenology, he moves uneasily between the other two philosophical alternatives that he sketches for us, namely positivism and ‘metaphysics of the
Foucault’s attack against phenomenology in OT gains its full force in
chapter 9, ‘Man and his Doubles’, but it begins already to take shape in
the paragraphs in which he lays out in bold strokes the predicament of
all post-Kantian philosophy. The epistemological conjuncture in which
phenomenology is born, in a sense, already seals its fate. Foucault’s
critical attack advances on multiple fronts. He seeks to situate phenomenology historically, and thus to show how the self-understanding
implied in its fundamental project is only one necessary moment in
modern thought. The questions it considers as fundamental to philosophy are those that were made possible and necessary by the modern
episteme. This episteme not only has given birth to the phenomenological enterprise as a whole, but has also made possible the modes of
experience of the phenomenologist, which serve as the starting point
of his inquiries. Phenomenology becomes Foucault’s key example of
a problematic form of thought made possible by the modern episteme and the birth of the empirical sciences of man. I will return to
Foucault’s criticism of phenomenology in detail in the next chapter.
To pave the way for this, I will now explicate Foucault’s controversial
and often misunderstood concept of ‘man’. For Foucault, man marks a
form of thought that the modern episteme has given birth to, and the
death of man is the mark of the beginning of the next one.
The birth and death of man
For Foucault, man refers to a human being, but a human being only
insofar as he is understood in a certain way, in a way that was not possible in the classical age, for example. Man is a being who is the source of
ph ilo so ph ic al lau g h t e r
knowledge of the world, and at the same a being in the world that can be
known. Foucault calls man an empirico-transcendental doublet. Man,
finite, yet a basis of all knowledge, is an invention of modern thought.
In his double aspect, man is at the same time a fact among other facts
to be studied empirically, and he is the transcendental ground of all
knowledge. He is formed by a complex network of background practices that he can never fully understand, and yet he is the possibility
of their elucidation. He is a product of a history whose beginning he
cannot reach and at the same time he is the writer of that history.
Foucault claims that the new conceptual space in which the human
sciences were formed in the modern age took shape in the figure of
In classical thought, the personage for whom the representation exists,
and who represents himself within it, recognizes himself therein as an
image or reflection, he who ties together all the interlacing threads of
the ‘representation in the form of the table’ – he is never to be found
in that table himself. Before the end of the eighteenth century, man did
not exist – anymore than the potency of life, the fecundity of labour, or
the historical density of language. He is quite a recent creature, which
the demiurge of knowledge fabricated with his own hands less than two
hundred years ago: but he has grown old so quickly that it has been only
too easy to imagine that he had been waiting for thousands of years in
the darkness for that moment of illumination in which he could finally
be known.
(OT, 308)13
Foucault thus claims that our understanding of the human being as
an ‘empirico-transcendental doublet called man’ is by no means necessary or unproblematic. Although taken for granted by us to the extent
that it is difficult for us to conceive of any other ways of thinking about
the relationships between the subject, knowledge and history, Foucault
diagnoses man as the problem of the modern episteme. According
13 ‘Dans la pensée classique, celui pour qui la représentation existe, et qui se représente
lui-même en elle, s’y reconnaissant pour image ou reflet, celui qui noue tous les fils
entrecroisés de la “représentation en tableau”, – celui-là ne s’y trouve jamais présent
lui-même. Avant la fin du XVIIIe siècle, l’homme n’existait pas. Non plus que la puissance
de la vie, la fécondité du travail, ou l’épaisseur historique du langage. C’est une toute
récente créature que la démiurgie du savoir a fabriquée de ses mains, il y a moins de
deux cents ans: mais il a si vite vieilli, qu’on a imaginé facilement qu’il avait attendu dans
l’ombre pendant des millénaires le moment d’illumination où il serait enfin connu.’
(MC, 319)
fo u c au lt o n freedom
to Foucault, modern thought has been a series of attempts to overcome the paradox inherent in the figure of man. It has been searching
a discourse whose tension would keep separate the empirical and the transcendental, while being directed at both; a discourse that would make it
possible to analyze man as a subject, that is, as a locus of knowledge which
has been empirically acquired but referred back as closely as possible to
what makes it possible, and as a pure form immediately present to those
(OT, 320–1)14
According to Foucault, the first attempt by post-Kantian thinkers to overcome the paradox of man was through reductionism: if they reduced
man to his empirical side they could not account for the possibility
of knowledge, and if they emphasized the transcendental side, they
could not claim scientific objectivity or account for the contingency
of man’s empirical nature. Foucault associates this stage with the positivism of Comte and the thought of Hegel and Marx. Foucault places
phenomenology within the next stage in which the problem was stabilized, that is, in the coexistence of empiricism and transcendentalism
in an ambiguous balance. His discussion of the three doubles of man –
empirical/transcendental, the cogito/the unthought, the retreat/the
return of origin – is an explicit criticism of phenomenology. He engages
with Merleau-Ponty, Sartre, Husserl and Heidegger respectively. I will
focus in the following chapter on Foucault’s criticism of Husserl,
although Foucault’s argument follows the same structure in connection with all of the three doubles: a mode of thought that centres on
man – a human being in the order of positivity and in the order of a
foundation, as both the source of meaning as well as the outcome of the
natural world, human culture and history – necessarily remains ambiguous and circular. It superimposes the transcendental and the empirical
dimensions of man.15 Phenomenology can only show how ‘what is given
14 ‘un discours dont la tension maintiendrait séparés l’empirique et le transcendantal,
en permettant pourtant de viser l’un et l’autre en même temps; un discours qui permettrait d’analyser l’homme comme sujet, c’est-à-dire comme lieu de connaissances
empiriques mais ramenées au plus près de ce qui les rend possibles, et comme forme
pure immédiatement présente à ces contenus’ (MC, 331).
15 Beatrice Han (1998/2002) argues that Foucault’s critique of phenomenology and of
all post-Kantian theory, which is denounced as being imprisoned by the ‘analytic of finitude’ in OT, applies to Kant himself. According to Han, Foucault’s presentation of Kant’s
position in OT is ambivalent. On the one hand, Foucault credits Kant with being the
ph ilo so ph ic al lau g h t e r
in experience and what renders experience possible correspond to one
another in an endless oscillation’ (OT, 336).
Foucault’s bold aim is thus to show how, after a series of convoluted attempts to deal with the paradoxical predicament of the modern
episteme, modern thought has arrived at a dead end. ‘And so we find
philosophy falling asleep once more . . . this time not the sleep of Dogmatism but that of Anthropology’ (OT, 341). Foucault’s judgement of
philosophy in general – but particularly of phenomenology – is cutting:
phenomenology is ‘a sleep – so deep that thought experiences it paradoxically as vigilance, so wholly does it confuse the circularity of a dogmatism folded over upon itself in order to find a basis for itself with the
agility of a radical philosophical thought’ (OT, 341). Hence, the grand
aim of Foucault in OT is to show, first, how all previous philosophy has
erred and then to give it new direction and impetus. Just as Nietzsche
heralded the death of God as promising philosophical thought a new
beginning, Foucault clearly and consciously imitates him in heralding
the death of man as an event important enough to inaugurate a new
episteme: ‘The void left by man’s disappearance’ is ‘the unfolding of a
space in which it is once more possible to think’ (OT, 342).
initiator of the discursive space in which Modernity deploys itself by critically questioning
the conditions of possibility of representation, and consequently of all possible knowledge. On the other hand, for Foucault, Kant was the insurmountable limit of Modern
thought because his thought initiated the monopolization of the field of possible knowledge by ‘man’ and his doubles. The critical question of the conditions of possibility of
true knowledge was, for Kant, intrinsically connected to the introduction of the distinction between the empirical and the transcendental. Han reads Foucault’s unpublished
complementary doctoral thesis, which was a commentary to his translation of Kant’s
Anthropology from a Pragmatic Point of View, and argues that in this text Foucault’s aim
was to show how the strict separation between the transcendental and the empirical put
forward by Kant in The Critique of Pure Reason undergoes an inflection in the Anthropology
from a Pragmatic Point of View symbolized by the recentring of Kant’s triple interrogation
on to the question ‘What is man?’ This displacement is problematic for Foucault, as it
makes the contents of empirical experience work as their own condition of possibility;
moreover it seeks within human finitude the elements of a transcendental determination henceforth made impossible in principle by the anthropological confusion between
the empirical and the a priori (Han 1998/2002, 3). Han writes: ‘Commentary permits us
to establish that it is not the “Kantian critique” in its totality that marks “the threshold of
our Modernity”, but that the line of division passes within Kant’s work itself, separating
the original formulations of the transcendental theme from its later versions’ (35). This
means that Foucault’s work is not only a critical reaction to phenomenology, but also
and more profoundly to Kant. According to Han, it is the identification of the Kantian
aporia that provided Foucault with a guide from which he was able to build a contrario an
original method and conceptual apparatus, and that allowed him to reopen the critical
question of the conditions of possibility of knowledge while attempting to throw off the
last anthropological constraints (Han 1998/2002, 4–5).
fo u c au lt o n freedom
The being of language
The new possibilities for thinking opened up by the death of man are
connected with a new understanding of language. Foucault argues that
in Western culture the being of man and the being of language have
never, at any time, been able to coexist together as a basis of knowledge, and that their incompatibility has been one of the fundamental
features of our thought. This forms the root of ‘the most important
philosophical choice of our period’ (OT, 339). Foucault suggests that
it is in the analysis of language as something more fundamental than
man that the new possibilities for thought lie. The birth of man was possible because of the collapse of the classical discourse, but now we are
again on the brink of a new episteme. Man’s being is weighed down by
‘the reappearance of language in the enigma of its unity and its being
as by a threat’ (OT, 338).
Although the modern episteme detached language from representation, which meant that it could appear as a philosophical problem in
its own right, Foucault suggests that philosophical reflection held itself
aloof for a long time because of the dispersion of language, because of
its multiple modes of being (OT, 304).16 According to Foucault, language did not appear in the field of thought directly and in its own right
until the end of the nineteenth century with Nietzsche. Now, however,
Foucault claims that it is the most important question confronting us.
‘The whole curiosity of our thought now resides in the question: what is
language, how can we find a way round it in order to make it appear in
itself, in all its plenitude?’ (OT, 306). This question heralds a new episteme (OT, 307). Foucault suggests a solution to the paradoxes of the
modern episteme in an inquiry based on the functioning of language.
16 While language was posited and reflected upon only in the form of an analysis of representation in the classical age, in the modern age it appeared as a question in its own
right. Language itself had become an object of science, instead of having the sole right
to represent the natural order of things. According to Foucault, this ‘demotion of language’ was compensated for in three ways. Firstly, because language was not only an
object of science but also the necessary medium for any scientific knowledge, there
arose the aim of neutralizing scientific language so that it could mirror the world as
exactly as possible. Foucault calls this the positivist dream of a purified, logical language
(OT, 296–7). Secondly, because language had lost its transparency and resumed instead
the enigmatic density it possessed at the time of the Renaissance, this resulted in the
revival of the techniques of exegesis (OT, 298). And thirdly, there was the birth of literature in the modern sense. According to Foucault, modernist writing is not about
creating beauty or pleasing our senses. What it lays before us is the enigmatic being of
language, the bare existence of words with their indefinite meanings (OT, 300).
ph ilo so ph ic al lau g h t e r
Foucault’s position is thus nominalist rather than phenomenological
in the sense that the historical a priori ordering scientific practices is discursively constituted.17 By studying how discourses have been ordered
at different historical times, he seeks to reveal or unveil the ontological order of things assumed to exist independently of them. Scientific
discourses always presuppose certain historically varying things as existing in the world, and this implicit, presupposed ontology grounds the
explicit order of things. However, not only do scientific discourses rely
on certain implicit ontological assumptions, they also explicitly create
them, at least in the realm of human sciences. This reading of the ontological order not as the transcendental ground of scientific discourses
but as their implicit effect – presupposed but by that same movement
also created by them – is explicit in certain passages in OT, such as:
If one wishes to undertake an archaeological analysis of knowledge
itself . . . One must reconstitute the general system of thought, whose
network, in its positivity, renders an interplay of simultaneous and apparently contradictory opinions possible. It is this network that defines the
conditions that make a controversy or a problem possible, and that bears
the historicity of knowledge.
(OT, 75)18
Hence, one can say that Foucault considers the whole network of knowledge, in its positivity, also to constitute its own ‘unconscious’, the historical a priori that makes certain problems and ways of questioning
possible. Thus, he does not claim that there exists two separate ontological levels, the network of scientific discourses, on the one hand, and
the conditions that form it, on the other. There is only the network of
discourses, which can, however, be retrospectively analyzed on different levels. Archaeological analysis reveals the structures of this network
that have remained unconscious to its practitioners while conditioning
their thought and experience.
The nominalist idea that discourse systematically forms the objects
of which it speaks, as well as the ontological order on the basis of which
they become possible, is made more explicit in the book that followed
17 For more on Foucault’s nominalism, see e.g. Flynn 1994, Rouse 1994.
18 ‘Si on veut entreprendre une analyse archéologique du savoir lui-même . . . Il faut
reconstituer le système général de pensée dont le réseau, en sa positivité, rend possible
un jeu d’opinions simultanées et apparemment contradictoires. C’est ce réseau qui
définit les conditions de possibilité d’un débat ou d’un problème, c’est lui qui est porteur
de l’historicité du savoir.’ (MC, 89)
fo u c au lt o n freedom
OT, The Archaeology of Knowledge.19 Foucault sets out to present a method
for studying discourse; a rule-governed practice of making scientific
statements (AK, 138).20 Discursive practices constitute their objects
of study through rule-governed transformations, they do not simply
seek to articulate the already existing and ordered things themselves.
Discourse is not reducible to language and speech, but must rather be
understood as a formative practice that constitutes its objects and also
circumscribes its subjects in creating all feasible positions from which
it is possible to make scientific statements. Foucault now writes about
the task of archaeology:
A task that consists of not – of no longer – treating discourses as groups
of signs (signifying elements referring to contents or representations)
but as practices that systematically form the objects of which they speak.
Of course, discourses are composed of signs; but what they do is more
than use these signs to designate things. It is this more that renders them
irreducible to the language and to speech. It is this ‘more’ that we must
reveal and describe.
(AK, 49)21
19 Beatrice Han (1998, 87–93) argues that Foucault’s nominalist position is not consistent
in OT, and that there is a major shift between OT and AK. The historical a priori is
defined in OT as an implicit relationship between words and things, or between being
and language: the conditions of possibility of forms of knowledge consist of different
relationships between the being of signs and being in general. Foucault considered it his
task to chart the different forms of this relationship between words and things through
history, from the Renaissance to modernity. According to Han, his form of analysis
already makes it clear that he must still have presupposed an ontology of ‘words’ and
‘things’ in OT – an ontology of two separate and autonomous modes of existence –
to be able to analyze the changing forms of their relationship. Hence, although he retrospectively claimed that his book dealt with neither words nor things, but rather with
‘objects’, which are discursively constituted, his position becomes consistently nominalist only after OT.
20 Foucault acknowledged in AK that he used the notion of discourse in at least three
different ways: (1) as a general domain of all statements, (2) as an individualizable
group of statements, (3) as a regulated practice that accounts for a certain number of
statements (AK, 80). I will restrict myself to the last meaning, which, I argue, is central
to understanding Foucault’s philosophical position. For more on Foucault’s concept of
discourse see e.g. Dreyfus and Rabinow 1982, Frank 1989/1992.
21 ‘Tâche qui consiste à ne pas – à ne plus – traiter les discours comme des ensembles de
signes (d’éléments signifiants renvoyant à des contenus ou à des représentations) mais
comme des practiques qui forment systématiquement les objets dont ils parlent. Certes,
les discours sont faits de signes; mais ce qu’ils font, c’est plus que d’utiliser ces signes
pour désigner des choses. C’est ce plus, qui les rend irréductibles à la langue et à la
parole. C’est ce “plus” qu’il faut faire apparaı̂tre et qu’il faut décrire.’ (AS, 66–7)
ph ilo so ph ic al lau g h t e r
In AK, Foucault also reformulates his mode of questioning in order to
emphasize the idea that his aim is not to reveal a set of transcendental
rules or conditions that are ontologically separate from the empirical
practices. Rather, the analysis of discourse aims to isolate ‘the conditions
of existence’ of statements as opposed to the conditions of possibility
(AK, 28). These conditions of existence determine what actually exists
and thus, in principle, imply a finite group of statements. Foucault
thus argues that his analysis of scientific discourses moves on a solely
descriptive plane. He does not concern himself with possible statements
or with the conditions of possibility for making statements, but rather
focuses on actual statements.
Foucault refers to disciplinary unities such as natural history by using
a broader term discursive formation. He claims that the unity of a discursive formation is not defined by a unity of any of its elements, for
example, its objects or themes, but rather by the rules that govern the
formation of its statements and objects. His major premise in AK is
thus that discursive formations are identifiable as rule-governed systems. The aim of an archaeology of scientific discourse is to identify
these rules of formation in their historical existence. Such rules are the
historical a priori in AK: they are regulative and formative but historical. They provide ‘an a priori not of truths that might never be said,
or really given to experience, but an a priori of a history that is given,
since it is of things actually said’ (AK, 127). The rules governing discursive formations are not abstract laws or formal principles, but refer
to the ways in which the statements are actually related. ‘The regularity
of statements is defined by the discursive formation itself. The fact of
its belonging to a discursive formation and the laws that govern it are
one and the same thing’ (AK, 116). It is possible to abstract the rules
from the practices only retrospectively, as a result of an archaeological
Hence, the rules are not transcendental in the sense that they form
the condition of all possible statements. For Foucault, there is nothing
more than the particular and limited amount of statements and the
relations that exist between them. The rules of discursive formation
are simply the description of these existing relations (AK, 116–17).
They determine what, in a particular historical situation, in a particular
field, did count as scientific objects, what sort of things were said about
them, who said them, and what concepts were used in the saying (AK,
71). While, in OT, Foucault presented archaeology as a modification of
fo u c au lt o n freedom
transcendental philosophy in the sense that his aim was to isolate the
conditions of possibility of scientific knowledge for an epoch, in AK he
restricts his analysis more strictly to particular discursive fields and their
conditions of existence.
I will show in the following chapter how the central role Foucault
gives to language, as well as the methodological paths that he explores
in studying it, can be read as reactions to the problems arising out of phenomenology. I take the main motivation in Foucault’s criticism of phenomenology in OT to be to reveal the problems involved in grounding
knowledge on the being of man and to press upon us the need to find
a new direction for philosophical inquiry in the study of language. By
comparing Foucault’s archaeology to Husserl’s phenomenology, especially to the formulations Husserl gives his project in his late work The
Crisis of European Sciences and Transcendental Phenomenology, I hope to
bring to light both the common forms of questioning as well as the
profound methodological differences.
The philosophical laughter that inaugurates OT reappears later on
in the book as laughter directed at phenomenology as a prime example
of a thought centred on man. Philosophical laughter is what Foucault
thinks is needed in order to shatter ‘the familiar landmarks’, the domineering forms of thought, and to inaugurate a new mode of thinking.
However, he must have been aware of the vulnerability of his own philosophical position, since his philosophical laughter could only ever be
silent laughter. He ends his criticism of phenomenology in chapter 9
by writing:
Anthropology constitutes perhaps the fundamental arrangement that
has governed and controlled the path of philosophical thought from
Kant until our own day. This arrangement is essential, since it forms
part of our history; but it is disintegrating before our eyes, since we are
beginning to recognize and to denounce in it, in a critical mode, both a
forgetfulness of the opening that made it possible and a stubborn obstacle
standing obstinately in the way of an imminent new form of thought. To
all those who still wish to talk about man, about his reign or liberation,
to all those who still ask themselves questions about what man is in his
essence, to all those who wish to take him as their starting point in their
attempts to reach the truth, to all those who, on the other hand, refer all
knowledge back to the truths of man himself, to all those who refuse to
formalize without anthropologizing, who refuse to mythologize without
demystifying, who refuse to think without immediately thinking that it is
ph ilo so ph ic al lau g h t e r
man who is thinking, to all these warped and twisted forms of reflection
we can answer only with a philosophical laugh – which means, to a certain
extent, a silent one.
(OT, 342–3)22
22 ‘L’Anthropologie constitue peut-être la disposition fondamentale qui a commandé et
conduit la pensée philosophique depuis Kant jusqu’à nous. Cette disposition, elle est
essentielle puisqu’elle fait partie de notre histoire; mais elle est en train de se dissocier
sous nos yeux puisque nous commençons à y reconnaı̂tre, à y dénoncer sur un mode
critique, à la fois l’oubli de l’ouverture qui l’a rendue possible, et l’obstacle têtu qui
s’oppose obstinément à une pensée prochaine. A tous ceux qui veulent encore parler
de l’homme, de son règne ou de sa libération, à tous ceux qui posent encore des questions sur ce qu’est l’homme en son essence, à tous ceux qui veulent partir de lui pour
avoir accès à la vérité, à tous ceux en revanche qui reconduisent toute connaissance aux
vérité de l’homme lui-même, à tous ceux qui ne veulent pas formaliser sans anthropologiser, qui ne veulent pas mythologiser sans démystifier, qui ne veulent pas penser sans
penser aussitôt que c’est l’homme qui pense, à toutes ces formes de réflexion gauches et
gauchies, on ne peut qu’opposer un rire philosophique – c’est-à-dire, pour une certaine
part, silencieux.’ (MC, 353–4)
Gérard Lebrun (1989/1992) suggests in an article on Foucault’s
archaeology that it is worth rereading The Order of Things, at least once,
as an anti-Krisis. To present such a reading will be my aim in this chapter.
Lebrun points out that The Order of Things and Husserl’s Crisis of European Sciences bear many rhetorical, thematic and structural affinities.
Both Foucault and Husserl engage in a constructive rereading of the
history of philosophy, both prepare their main arguments about modern and contemporary thought with careful analyses of seventeenthcentury thought, and both set as their task the elucidation of the
fundamental conditions of possibility of scientific knowledge. Some
of Foucault’s key concepts in OT, such as ‘historical a priori’ originate
from Husserl (Lebrun 1989/1992, 32).1 Yet despite the similarity of
their tasks, themes and vocabulary, the two thinkers end up with fundamentally different understandings of the historicity of reason, the
conditions of possibility of knowledge, and the development of the sciences. OT sets out to ultimately reveal the failure of phenomenology
and to chart a way out the impasse.
Foucault’s criticism of phenomenology in OT can be divided roughly
into two different types. I will present his discussion here according to
this twofold schema, although the issues in question necessarily overlap.
Firstly, there is implicit criticism of the phenomenological understanding of the philosophy of science as presented in Husserl’s Crisis. By laying out a very different understanding of the development of science in
1 Husserl explicates the notion of historical a priori in his essay ‘The Origin of Geometry’.
Although Foucault’s historical a priori originates from there, it gets a very different meaning in Foucault’s thought. I will return to the difference between Husserl’s and Foucault’s
notions at the end of this chapter.
t h e fouc au lt ian failu re o f phe nom e nol ogy
OT, Foucault implicitly challenges Husserl’s method and central ideas
about the history of thought. In his introduction to the English translation, he presents the method of the book specifically as an alternative
and antidote to phenomenology.
If there is one approach that I reject, however, it is that (one might call it
broadly speaking, the phenomenological approach) which gives absolute
priority to the observing subject, which attributes a constituent role to
an act, which places its own point of view at the origin of all historicity –
which in short, leads, to a transcendental consciousness. It seems to me
that the historical analysis of scientific discourse should, in the last resort,
be subject, not to the theory of the knowing subject, but rather to a theory
of discursive practice.
(OT, xiv)
Secondly, there is explicit criticism of phenomenology in the text of OT,
in chapter 9, which focuses on the paradox of man. Because the chapter
is notoriously cryptic, to be able to follow Foucault’s arguments one
cannot simply lay them out for evaluation: they must first be rebuilt from
fairly sparse and ambiguous references. I will therefore explicate what
I take to be the main points in Foucault’s criticism of phenomenology
in relation to Husserl’s thought.
The history of science
Foucault’s engagement with Husserl in OT takes issue mainly with
Husserl’s late work, particularly The Crisis of European Sciences. By the
time of he had written The Crisis, Husserl had been forced to acknowledge that the ‘Cartesian way’ into phenomenology was misleading. One
reason for this was that there were fundamental structures of the world
which the Cartesian way overlooked, or presupposed unproblematically, in its explication of the ego. In The Crisis, therefore, Husserl
develops the important concept of the life-world. This refers to a horizon of meaning that is always ‘already there’ for the experiencing
subject before any efforts are made to articulate or understand it.
The life-world constitutes the universal field of all actual and possible
The life-world, for us who wakingly live in it, is always already there,
existing in advance for us, the ‘ground’ of all praxis whether theoretical or
extratheoretical. The world is pregiven to us, the waking, always somehow
fo u c au lt o n freedom
practically interested subjects, not occasionally but always and necessarily
as the universal field of all actual and possible praxis, as horizon. To live
is always to live-in-certainty-of-the-world.
(Husserl 1954/1970a, 142)2
Husserl argues that the life-world forms the basis of all human praxis,
including all scientific praxis. All theories of science – not just empirical
sciences but also the a priori sciences, such as mathematics and logic –
derive their meaning and grounds for validity from the life-world. They
presuppose the existence of a directly experienceable world, and set
themselves the task of making this ‘prescientific and subjective knowledge’ into perfect and objective knowledge. In the process, however,
they lose sight of this ground on which their theoretical formulations
rest and can only rest. The ‘objective-true world’ that results from the
theoretical activity of the sciences and of the scientists is a construction
of something that, in principle, is not perceivable or experienceable in
its own being. Husserl writes:
The contrast between the subjectivity of the life-world and the ‘objective’,
the ‘true’ world, lies in the fact that the latter is a theoretical substruction, the substruction of something that is in principle not perceivable,
in principle not experienceable in its own proper being, whereas the
subjective, in the life-world, is distinguished in all respects precisely by its
being actually experienceable.
(Husserl 1954/1970a, 127)3
The life-world is, by its being, experienceable and is therefore the realm
of original self-evidences. It consists of what is intuitable in principle
(Husserl 1954/1970a, 127). It is thus both the reference point of meaning for the theories of objective science as well the concrete unity that
2 ‘Die Lebenswelt ist . . . für uns, die in ihr wach Lebenden, immer schon da, im voraus für
uns seiend, “Boden” für alle, ob theoretische oder außertheoretische Praxis. Die Welt ist
uns, den wachen, den immerzu irgendwie praktisch interessierten Subjekten, nicht gelegentlich einmal, sondern immer und notwendig als Universalfeld aller wirklichen und
möglichen Praxis, als Horizont vorgegeben. Leben ist ständig In-Weltgewißheit-leben.’
(Husserl 1954/1962, 145)
3 ‘Der Kontrast zwischen dem Subjektiven der Lebenswelt und der ‘objektiven’, der
‘wahren’ Welt liegt nun darin, daß die letztere eine theoretisch-logische Substruktion
ist, die eines prinzipiell nicht Wahrnehmbaren, prinzipiell in seinem eigenen Selbstsein
nicht Erfarhrbaren, während das lebensweltlich Subjektive in allem und jedem eben
durch seine wirkliche Erfahrbarkeit ausgezeichnet ist’ (Husserl 1954/1976, 130).
t h e fouc au lt ian failu re o f phe nom e nol ogy
encompasses them. Objective science is one kind of human praxis and
is therefore itself also part of the life-world.
The task that Husserl sets for his ‘new science’ in The Crisis is to break
‘the scholastic dominance of objective-scientific ways of thinking’ and
to inquire into the life-world as the only possible ground for the sciences and their claims to objective validity (Husserl 1954/1970a, 129).
The life-world must become a subject of investigation in its own right.
This cannot mean a simple description of our surroundings, however.
Husserl clearly sees the problems involved in a project aimed at elucidating something that is both the basis of objective validity or the
‘scientifically true’ world, as well as what encompasses it and all theoretical pursuits, including its own. He asks: ‘How are we to do justice
systematically – that is, with appropriate scientific discipline – to the
all encompassing, so paradoxically demanding, manner of being of the
life-world?’ (131).
Husserl claims to achieve this goal by performing a series of inquiries,
starting from the pregiven life-world and moving back to its constitution in transcendental subjectivity, and ultimately in transcendental
intersubjectivity. The first step is the epoche of objective science resulting in the disclosure of the life-world. This means that all our objectivescientific opinions and cognitions are put out of play and ‘one can place
oneself completely upon the ground of this straightforwardly intuited
world’ (123). After suspending all our scientific beliefs about the world,
we come to realize that the life-world is not merely subjective and relative, but that it has a general structure. Prescientifically, the world is
already a spatio-temporal unity containing bodies with causal relations
between them. The structures of the life-world, however, are now intuited directly and are not theoretical abstractions.
To elucidate these invariant structures of the life-world would be the
task of life-world ontology. However, the phenomenologist must not
be content with staying at this level, which still remains the level of the
natural attitude, but should proceed to ask how the life-world is pregiven
to us: what is its manner of being and what makes it a universal basis
for any sort of knowledge (146).
We notice thereby that the first step which seemed to help at the beginning, that epoche through which we freed ourselves from all the objective
sciences as grounds of validity, by no means suffices. In carrying out this
epoche, we obviously continue to stand on the ground of the world; it
fo u c au lt o n freedom
is now reduced to the life-world which is valid for us prescientifically; it
is just that we may use no sort of knowledge arising from the sciences
as premises, and we may take the sciences into consideration only as
historical facts, taking no position of our own on their truth.
(Husserl 1954/1970a, 147)4
It then becomes necessary to accomplish a second epoche the ‘transcendental epoche’, which reveals the transcendental correlation between
the world and world consciousness.5 This is accomplished by the ‘withholding of natural, naive validities and in general of validities already in
effect’ (135). By placing himself above his practical and natural interests, and even his own natural being, the philosopher arrives at an attitude that is ‘above the pregivenness of the validity of the world’ (150) and
is therefore capable of discovering its mode of giveness in the constituting consciousness. After all natural interests have been put out of play,
the world appears ‘purely as a correlate of the subjectivity which gives it
ontic meaning, through whose validities the world is at all’ (152). The
life-world is studied as multiplicities of manners of appearing, which
refer back to the ego-pole as constitutive of their unity. The general
structure of all experience, as well as the manner of being of the world,
are comprehended under three headings: the ego-pole, the subjective
as appearance tied together synthetically, and the object-poles (171).
Foucault’s critique of phenomenology in OT questions the possibility of accomplishing either one of the steps or epoches that Husserl
presents in The Crisis. While Foucault explicitly attacks the second in
his discussion of the double of man, ‘the cogito and the unthought’,
4 ‘Wir bemerken dabei, daß jener nächste Schritt, der anfangs zu helfen schien, jene
Epoché, in der wir uns aller objektiven Wissenschaften als Geltungsbodens entheben
mußten, keineswegs schon genügt. Im Vollzug dieser Epoché stehen wir offenbar noch
weiter auf dem Boden der Welt; sie ist nun reduziert auf die vorwissenschaftlich uns geltende Lebenswelt, nur daß wir keinerlei Wissen, das aus den Wissenschaften herstammt,
als Prämisse verwenden und die Wissenschaften nur in der Weise historischer Tatsachen,
ohne eigene Stellungnahme zu ihrer Wahrheit, in Rechnung ziehen dürfen.’ (Husserl
1954/1962, 150)
5 Recent studies in phenomenology argue that reduction is not strictly speaking something
that is accomplished. It is not a planned procedure or an expected result, but rather,
unanticipated experience. Juha Himanka (1999) emphasizes that reduction allows us to
move from the habitual to the unforeseen and the unexpected: after reduction we are
able to look as if for the first time. On this kind of interpretation of reduction, see e.g.
Waldenfels 1993, Himanka 1999 and 2001, Heinämaa 2002. See also already Fink 1933,
Merleau-Ponty 1945/1994.
t h e fouc au lt ian failu re o f phe nom e nol ogy
I will show here that the method and overall understanding of the history of science present in OT also makes it impossible to accomplish
the first epoche.
Husserl’s aim in The Crisis, in concise terms, is to show how the lifeworld is the starting point for all scientific abstractions and theories. All
scientific truths, even the detailed calculations of observed phenomena given by complicated measuring devices, must derive their ground
of validity from the life-world, which alone is directly experienceable.
Foucault, on the other hand, understands scientific development as
irreversibly removed from direct intuition and experience.
In order to explicate the difference between Foucault’s and Husserl’s
understanding of the philosophy of science, I will argue here that many
of the central ideas that Foucault inherits from Gaston Bachelard’s
and George Canguilhem’s thought are illuminative.6 A third important influence would be Jean Cavaillès, whose posthumously published
essays contain an explicit critique of Husserlian phenomenology and a
call to abandon the subject and to develop in its place a philosophy of
the concept.7
6 Gary Gutting (1989) explicates the relationship between Foucault’s archaeology and Canguilhem’s and Bachelard’s philosophies of science in his book Michel Foucault’s Archaeology
of Scientific Reason. Gutting argues that Foucault’s method in OT is primarily an application and an extension of Canguilhem’s history of concepts, as well as a refinement of
Bachelard’s understanding of the historicity of scientific conceptions. Gutting (1989, 54,
218–20) also points out, however, that Foucault’s archaeology is not only a continuation
of Bachelard’s and Canguilhem’s work. His archaeology is an original method for the
study of history of science, with its own distinctive topics and concerns. Even in areas
where Bachelard’s and Canguilhem’s influence is particularly strong, Foucault extends,
adapts and transforms their ideas and methods. As a historian of biology, Canguilhem
focused on concepts that were in fact deployed by the biologists whose work he was analyzing. Foucault, however, deals not only with first-order biological concepts, but also
with concepts that define the conditions of possibility for formulating such concepts.
According to Gutting, Foucault’s extension of the history of concepts thus undermines
the privileged role of disciplines in the history of thought and introduces a level of conceptual history that is more fundamental than that of the first-order concepts of scientific
disciplines. While Foucault’s work follows Bachelard in emphasizing the epistemological factors working below the level of first-order concepts as well as the consciousness
of the scientists, unlike Bachelard, Foucault does not see this level as entirely negative.
For Bachelard, such factors are residues of outdated modes of thought that obstruct the
path of scientific development. For Foucault, this deep epistemological level has positive
significance: it embodies the conditions that make possible the formation of new concepts. Since my principal topic is Foucault’s relationship to phenomenology and not to
French philosophy of science, my presentation of Foucault’s relationship to Bachelard’s
and Canguilhem’s thought relies considerably on Gutting’s work.
7 See Cavaillès 1947/1994, 473–560.
fo u c au lt o n freedom
Bachelard’s philosophy of science as a critique of phenomenology. Bachelard’s
work reiterates the central idea in Husserl’s thought, that scientific
entities and theories are essentially historical, and that the development
of science is best understood by reflecting on its concrete historical
employment. However, while The Crisis strongly argues that sciences
have a derivative status in relation to philosophy understood as rigorous
science, Bachelard claims that the achievements of science in fact create
and must create philosophical understanding (Bachelard 1934/1949,
3–5; Gutting 1989, 13–14).8
In his book Le nouvel esprit scientifique (1934/1949), Bachelard sought
to show how modern physics radically altered our metaphysical understanding of the nature of reality: of matter, movement, space and time.
He further argued that it implied a non-Cartesian epistemology, which
condemns the doctrine that knowledge must proceed from simple
natures that can be known directly to more complex ones. Modern
physics reveals that there are no simple ideas or essences which could be
understood in themselves, and which would serve as a basis for further
deductions. All simple ideas or essences are in fact always derivative and
can initially only be understood as parts of complex systems of thought
and experience (Bachelard 1934/1949, 142–8). Modern science thus
forces us to rethink our ideas about direct intuition by questioning its
primacy as well as its reliability. The intuitive clarity of scientific claims
is achieved in a discursive manner, by progressive clarification resulting from varying examples, and by putting theoretical notions to work
(Bachelard 1934/1949, 145).9
Bachelard’s notion of the epistemic break characterizes the way
in which scientific knowledge splits off from, and even contradicts,
common-sense experiences and beliefs. It places objects of experience
into new categories that reveal properties and relations not available to
ordinary sense perception. Bachelard rejects any view that makes the
8 Gary Gutting (1989, 23) argues that it is important to note that Bachelard’s subordination
of philosophy to science is not an instance of positivistic scientism. Bachelard did not
consider philosophy itself to be a part of science. Philosophy is a reflection of the sciences,
and its methods and results do not share the empirical character of scientific disciplines.
See also Canguilhem 1994a, 33, 43.
9 In his book La philosophie de non (1940/1949) Bachelard also argued against the idea
of a unitary epistemological base underlying modern science. Reason, rather, must obey
science: it must not privilege immediate experience, but rather balance it with scientific
experience, which is more richly structured (144). Paul Rabinow (1994, 13) elaborates
on this by writing that Bachelard’s aim was not to attack science, but rather to show it in
action in its specificity and plurality.
t h e fouc au lt ian failu re o f phe nom e nol ogy
contents of ordinary, subjective experience more real than scientific
objects (Gutting 1989, 14–19). According to him, ‘Objective research
pursued in a laboratory commits us to progressive objectification that
gives reality at the same time to both a new experience and a new form
of thought’ (Bachelard 1934/1949, 172).
Foucault’s aim of searching for the conditions of possibility for scientific modes of knowledge in the structuring of the discourses themselves
follows Bachelard’s views in this respect. Foucault’s fundamental idea
of the episteme – the development of the sciences in accordance with
a historical a priori that unites, regulates and structures them – is in
direct opposition to Husserl’s argument that the lived experience is the
indispensable starting point and condition of possibility of scientific
discourse. By studying scientific discourses themselves, Foucault claims
to be able to reveal the ordering codes that condition, regulate and
structure even the ordinary, non-scientific experiences of an age. Even
though his explicit criticism in OT is mainly targeted at the effectiveness
of Husserl’s second, transcendental epoche, Foucault’s method and the
philosophical implications of the book refute Husserl’s first epoche, the
reduction to the life-world.
According to Husserl, the first epoche reveals to us a universal a priori
belonging purely to the life-world. From this it is possible to proceed
to life-world ontology, which explicates the universal structures of the
life-world, as well as to transcendental inquiry into the constituting function of subjectivity. What we need first is ‘a separation in principle of the
latter (universal life-world a priori) from the objective a priori, which is
always immediately substituted for it. It is this separation that is effected
by the first epoche of all objective sciences’ (Husserl 1954/1970a,
Husserl thus holds that science always involves some kind of a rupture
with prescientific experience, while at the same time it can only find
its basis in the life-world. The view of scientific discourse that Foucault
advocates in OT, on the other hand, questions the possibility of separating what belongs purely to the life-world and what is a second-order
idealization of scientific discourse. Science constructs ordinary experiences and blurs the line separating them from its own theoretically
informed experiences. Scientific idealizations become part of the ordinary life-world. Foucault’s notion of the episteme covers, in Husserl’s
terms, both the life-world and the objective world of science. A change
in episteme means a change in the perceptual, practical and theoretical
grids: the mode of being of words, as well as of objects, changes. Husserl
fo u c au lt o n freedom
acknowledges the historical dimension of the sciences and considers
the relation between the prescientific life-world and the scientific world
as dynamic.10 Nevertheless, even if their relation changes historically,
reduction to the life-world presupposes that this distinction is in place,
that we know what belongs to the scientific world and what belongs
to the life-world. For Foucault, on the other hand, even everyday perceptions are structured according to the changing historical a prioris of
scientific discourse, and hence the distinction becomes impossible in
principle. He would thus agree with Husserl that there is a more fundamental level underneath the level of particular scientific theories,
but he would claim that this level is not that of life-world experiences.
The fundamental, archaeological level structures not only the order of
discourses but also the subjective experiences of this order. Foucault
also searches for the conditions of possibility of scientific knowledge as
well as of life-world experiences, but rather than turning to the constituting transcendental subjectivity, he argues that these conditions are
essentially non-subjective and anonymous, and that they can be made
explicit only in regard to the past by the historical analysis of scientific
Moreover, contrary to Husserl’s aim in The Crisis, Foucault aims to
show in OT that the history of science is not a continuous and unified
development of rationality. This idea is also central to Bachelard’s work.
Bachelard argues that, strictly speaking, there is no such thing as the
history of science, only various histories of different regions of scientific
work. Philosophy thus cannot uncover one single, unified conception
of rationality from reflections on the history of science, it can only
find various regions of rationality (Gutting 1989, 14).11 Bachelard also
emphasizes sharp breaks in the history of science and in the conception of reason, and claims that for a phenomenologist there can be no
sharp breaks, because scientific discoveries are always additions of truth
about the objects given to ordinary experience. For a phenomenologist, science rests on the indispensable and fundamental perceptual
framework and proceeds from it (Gutting 1989, 27).
Foucault’s presentation of the history of western thought in distinct
epistemes in which the epistemic breaks make the central questions
10 See, in particular, the readings of Husserl that emphasize the importance of intersubjectivity and his phenomenology of the social world, most notably Steinbock 1995 and
Zahavi 1996/2001.
11 See also Machado 1989/1992, 4.
t h e fouc au lt ian failu re o f phe nom e nol ogy
of one episteme incompatible with, sometimes even totally incomprehensible in the next, is thus an implicit criticism of Husserl’s method
of reading the history of philosophy and science in The Crisis. Husserl
begins his search for the origins of the contemporary crisis in philosophy and science with a critical study of Galileo’s mathematization of
nature. Husserl shows how the mathematical idea of nature was given
priority in philosophy, and how this led to the fatal separation of philosophy and the natural sciences and to the forgetting of the life-world
as ‘the meaning-fundament of natural science’ (Husserl 1954/1970a,
48). After arriving at the discovery of transcendental subjectivity, first
Descartes, and later Kant, made the mistake of objectifying it and therefore postponed the forging of a true basis of science.
In the light of Foucault’s history of thought, it would be pointless to blame Galileo for leading western thought in the wrong direction, or Descartes and Kant for failing to discover transcendental subjectivity. Not only do epistemic breaks make the questions of one
episteme impossible in another, but Foucault’s method of writing the
history of thought in terms of concepts rather than individual discoveries also argues against Husserl’s approach. Canguilhem’s influence on
Foucault’s archaeology is apparent here.
Canguilhem’s history of science as a critique of phenomenology. Canguilhem’s
aim as a historian of science was to write histories of concepts, particularly the concepts of life sciences such as reflex, cell and the immune
system. According to Canguilhem, concepts regulate the production
of forms of knowledge: ‘The history of science should be a history
of the formation, deformation and rectification of scientific concepts’
(Canguilhem 1994a, 110). He discusses the origins of cell theory to
argue that theories never proceed from simple facts. Cell theory was
not formulated because cells could be observed with the new aid of a
microscope; the first premise of cell theory was not that living things are
composed of cells, but that all living things consist of nothing but cells.
Such an assertion could not be justified by observations made through
a microscope (Canguilhem 1994a, 161). Canguilhem shows how ideas
and concepts borrowed from political theory in fact dominated debates
in early cell theory; whether cells are elementary, independent organisms akin to the individuals of liberalism, or simply parts of a whole
unable to exist independently. While facts act as a stimulus to theory,
they neither engender the concepts that provide the theories with their
fo u c au lt o n freedom
internal coherence, nor initiate the intellectual ambitions that theories
pursue (177).
Another example that Canguilhem gives to illustrate the fundamental role of concepts in the development of science is Claude Bernard’s
discovery in 1855 of the concept of ‘internal secretion’, a concept that
‘only a few years earlier would have been taken as a contradiction in
terms, an impossibility as unthinkable as a square circle’ (Canguilhem
1944a, 265–6). The concept implied that animals were influenced and
regulated not only by an external environment, but also by an internal
environment. This meant, on the one hand, that the old distinction
between the plant kingdom and the animal kingdom, according to
which plants could and animals could not synthesize simple organic
compounds, was blown away. Life was seen in a new way, without the
distinction between plant and animal. On the other hand, it also meant
the birth of physiology: without the idea of an internal environment,
there could be no autonomous science of physiology (266–9).
Scientific concepts are thus not simply parts of the theories used to
interpret phenomena. They often provide us with the cognitive and perceptual grids for understanding a phenomenon that make it possible
to formulate scientific theories explaining it. Foucault’s aim in OT of
showing how concepts such as life and labour fundamentally structured
the theories of a discipline, follows this idea. As Gary Gutting (1989,
34) notes, the fundamental, regulative status of concepts makes it possible to write historical accounts of their formation and transformation
that operate at a different, more fundamental level than accounts of a
succession of explanatory theories.
According to Canguilhem, this view of the development of science
also makes it pointless to search for early precursors of major scientific
discoveries: ‘Strictly speaking, if precursors existed, the history of science would loose all meaning, since science itself would merely appear
to have a historical dimension’ (Canguilhem 1994a, 49). A precursor is
generally understood as a thinker or a researcher who proceeded some
distance along a path that was later explored all the way to the end
by someone else. He is thus someone who belongs to more than one
age: he is a man of his own time, but also simultaneously a contemporary of those later investigators credited with completing his unfinished
project. He is therefore a thinker whom the historian believes can be
extracted from his cultural milieu and inserted into others. Such adaptability is only obtained at the cost of neglecting the ‘historicity’ of the
object under study. Canguilhem thus concludes:
t h e fouc au lt ian failu re o f phe nom e nol ogy
So long as texts and other works yoked together by the heuristic compression of time have not been subject to critical analysis for the purpose
of explicitly demonstrating that two researchers sought to answer identical questions for identical reasons, using identical guiding concepts,
defined by identical systems, then insofar as an authentic history of science is concerned, it is completely artificial, arbitrary and unsatisfactory
to say that one man finished what the other started or anticipated what
the other achieved. By substituting the logical time of truth relations for
the historical time of these relations’ inventions, one treats the history of
science as though it were a copy of science and its object a copy of the
object of science. The result is the creation of an artifact, a counterfeit
historical object – the precursor.
(Canguilhem 1994a, 51)12
The ‘discovery’ of a precursor is thus usually based on the failure to recognize fundamental conceptual differences that underlie superficially
similar formulations. It also relies on an individualist model of scientific
discovery. In Canguilhem’s view, as soon as the methods and problems
of modern science become adjusted to each other, and instruments
become so highly specialized that their very use implies the acceptance
of common working hypotheses, then it will be true to say that science
shapes scientists just as much as scientists shape science (107).
Foucault reiterates this idea in OT by showing, for example, that the
so-called precursors of evolutionary theory in the classical age, despite
superficial similarities, were in fact relying on a wholly different understanding of nature. For them, nature was only conceivable as a unified, ahistorical table in which changes were shifts of the whole towards
a higher state of perfection. It was only the emergence of the modern concept of life understood as a historical, dynamic phenomenon
that made possible the idea of the evolutionary development of
a species.
Contrary to Husserl’s presentation of Galileo’s fundamental discovery of mathematized nature, and to Descartes’ and Kant’s mistakes
of objectifying transcendental subjectivity, Foucault would argue that
these individual discoveries or ‘errors’ were, in fact, made possible,
even necessary, by certain epistemic conditions and concepts. Galileo’s
idea of calculable nature was only one part of the classical idea of
an orderable nature as mathesis. The mathematization of nature that
Husserl attributed to Galileo, was, for Foucault, simply a typical mode of
12 The original French is not available.
fo u c au lt o n freedom
knowledge in the classical episteme, structured in terms of representation.13 Similarly, Foucault would object to the idea of viewing the
philosophies of Descartes and Kant in terms of mistakes. As Gérard
Lebrun (1989/1992, 24) writes, for Foucault ‘There is no point in
regretting that Descartes missed the idea of the transcendental ego,
when he was far from being able to foresee it’. Or, to see in Kantianism
a way of thinking that allowed the continued predominance of ‘the evidence of objectivity’ was not only futile, but also missed the essential
point. Kant’s thought inaugurated an epistemic break and with him the
classical mathesis was swept away forever (Lebrun 1989/1992, 27).
This same criticism also, implicitly, includes phenomenology itself.
From Foucault’s point of view, phenomenology is not sensitive enough
to the historicity of its own project – its historical conditions of possibility in Kant’s thought and in the modern episteme. Rather than
phenomenology being able to reveal the true ground of the sciences,
its own project, in fact, depends on their discoveries. The coming into
being of new positive elements – life, work, language – which to a large
extent determined what it was to be a human being, posed the problem of how a human being was to understand himself as the subjective
ground of science, as well a being dependent on a life-world he could
not subjectively create. Phenomenology is, thus, a child of its time, an
offspring of the modern episteme, which discovered life, labour, language and therefore also man, a double figure, dependent on the positivities it discovered. ‘This is why phenomenology . . . has never been
able to exorcise its insidious kinship, its simultaneously promising and
threatening proximity, to the empirical sciences of man’ (OT, 326).
In ‘The Vienna lecture’, Husserl presented phenomenology as the
culmination of the European teleology of the infinite goal of reason.
Its aim was to realize the ancient ideal of philosophy as an infinite task,
and to inaugurate ‘the total reorientation of the task of knowledge’
(Husserl 1954/1970b, 298).14 Foucault’s historical contextualization
of the phenomenological project as a necessary moment in the modern episteme is thus an implicit criticism of the aims or pretensions
of phenomenology. He does not consider phenomenology to be the
culmination of the history of philosophy, but, rather, gives it a far more
13 Cf. Lebrun 1989/1992, 23.
14 According to Husserl, phenomenology alone has been capable of attaining ‘the higher
stage of reflexivity which is decisive for the new form of philosophy and of European
humanity’ by realizing the aim of philosophy as a rigorous science on which all objective
sciences are based (Husserl 1954/1970b, 292).
t h e fouc au lt ian failu re o f phe nom e nol ogy
modest role. It is a sensitive acknowledgement of the paradox of man,
a particular configuration in modern thought, but a failed effort to
solve this paradox. Moreover, its ultimate failure means that all further
reformulations of its project are equally futile. The true solution to
the problem would require abandoning the starting point of the whole
phenomenological project, questioning what has in fact rendered the
whole of our contemporary thought historically possible. According to
Foucault, the question we must ask is: how will we think after the death
of man?
The analytic of finitude
It is often argued that Foucault’s criticism of phenomenology in OT
is inaccurate and that it is based on sweeping generalizations that are
not supported by textual references. Gary Gutting (1989, 223), for
example, notes that because Foucault’s discussions of the three ‘doubles’ contain some of the most convoluted and obscure passages in
OT, readers can easily be fooled into thinking that there is a level of
profound criticism that they have failed to penetrate. Gutting argues
that whatever profundity there is in these analyses only concerns
Foucault’s way of understanding the major projects of recent continental philosophy and relating them to one another. The contortions of his
interpretative analysis serve only to hide the weakness of his criticism
of phenomenology.
Because Foucault does not provide any textual references, it is a
matter of interpretation who exactly is the target of his criticism of
phenomenology in chapter 9, ‘Man and his Doubles’. There are clear
echoes of the three doubles in Merleau-Ponty’s, Husserl’s, Sartre’s
and Heidegger’s thought. It could be argued that the discussion of
the ‘actual experience’ in connection with the first double refers to
Merleau-Ponty, while the criticism of the second mainly characterizes
Husserl’s thought and the third double points to Heidegger.15 My aim
here, however, is not to decodify Foucault’s cryptic and missing references. Instead, I will ‘translate’ the three doubles into a criticism of
Husserl’s work, although it could be argued that this would apply in
15 See Dreyfus and Rabinow 1982, 33–41, Gutting 1989, 204. Didier Eribon (1991, 157)
claims that OT was a challenge to Sartrean hegemony in France at the time. He points
out that the book contained numerous attacks on Sartre that Foucault omitted from
the final version.
fo u c au lt o n freedom
differing degrees to all phenomenological projects. My aim in reconstructing Foucault’s three doubles as an explicit criticism of Husserl’s
thought is not to argue for the philosophical superiority of Foucault’s
thought in relation to Husserl. What is clearly contestable is whether
the problems at issue are insurmountable within the phenomenological enterprise itself, and require us to reject it, as Foucault claims. My
point in reconstructing Foucault’s criticism of phenomenology is firstly
to argue that, despite its undeniable vagueness, it points to profound
philosophical questions about the relationships between history and
philosophy, the transcendental and the empirical, language and the
subject. Secondly, I will show that the problems that Foucault points to
are not external but rather internal to phenomenology: they were evident to Husserl himself to a certain extent, but certainly to his followers,
both in France and Germany. Hence, my argument is that Foucault’s
work should be seen not as a total break with phenomenology but
rather as part of a continuum, even if the relationship is fundamentally
The cornerstone of Foucault’s criticism of phenomenology is his
claim that Kant’s distinction between the empirical field of knowledge
and its a priori conditions becomes blurred in the modes of thought
that Foucault calls ‘the analytic of finitude’.16 This is a configuration of
knowledge, or a mode of thinking, that is characterized by the fact that
knowledge is grounded on the human being in his finitude. This leads
to the paradox of man. Foucault discusses three forms of this paradox –
the three doubles of man, as he calls them – empirical/transcendental,
the cogito/the unthought, the retreat/the return of origin.
The empirical and the transcendental. The first paradox, the empirical
and the transcendental, refers to the paradoxical role a human being
has as an empirically limited being and a transcendentally determining subject: ‘such a being that knowledge will be attained in him of
what renders all knowledge possible’ (OT, 318). Man is both part
of the world and, as such, an object of empirical sciences, and at
the same time the transcendental ground of all knowledge, including these very same empirical sciences. Foucault argues that this analytic of finitude transforms questions about the empirical limits of
the knowing subject into questions about the conditions of possibility of knowledge. The ‘anthropological’ becomes the transcendental
16 Cf. Han 1998/2002, 17–20.
t h e fouc au lt ian failu re o f phe nom e nol ogy
when the empirical determinations of the knowing subject are understood as the condition of possibility of the subject’s knowledge.
Foucault credits Husserl with an acute diagnosis of the paradox of
man. He presents phenomenology as ‘the sensitive and precisely formulated acknowledgement of the great hiatus that occured in the modern
episteme at the turn of the eighteenth and nineteenth centuries’ (OT,
325). He argues that Husserl acknowledged the historicity of the knowing subject, its dependence on positivities it did not create, nor can ever
fully elucidate. At the same time, Husserl strongly argued that empirical sciences cannot provide the condition of possibility of knowledge.
Foucault clearly echoes Husserl by arguing that naturalism and historicism have common roots; two modes of thought that appeared out of a
discourse centred on man. His criticism of the claims of the autonomy of
naturalism and historicism can furthermore be seen as a representation
of Husserl’s anti-naturalism. The problem, according to both Husserl
and Foucault, is the foundation of knowledge, which naturalism and
historicism are unable to study. Husserl’s criticism of naturalism and
historicism, especially in his Philosophy as Rigorous Science (1911/1965),
seeks to show that naturalism and historicism are unfounded modes of
thought. Their presuppositions concerning the possibilities of knowledge are never made explicit, and the rules governing their concepts,
theories and values of research are never revealed or grounded. These
theories claim objectivity, but the claims are, in fact, empty rhetoric.
The status of what is true and stable scientific discourse remains
Husserl’s line of argument is reiterated by Foucault, when he argues
that the first efforts to solve the paradox of man meant trying reductionism. However, attempts to reduce man to his empirical side, as
positivism did, could not account for the possibility of knowledge, and
when the emphasis turned to the transcendental side, in the manner
that, for example, Hegel’s phenomenology did, it was not possible to
claim scientific objectivity, or to account for the contingency of man’s
empirical nature. Foucault writes:
For the threshold of our modernity is situated not by the attempt to apply
objective methods to the study of man, but rather by the constitution of
an empirico-transcendental doublet which was called man. Two kinds
of analysis then came into being: There are those that operate within
the space of the body . . . ; these led to the discovery that knowledge
has anatomo-physiological conditions, that it is formed gradually within
fo u c au lt o n freedom
the structures of the body, that it may have a privileged place within it,
but that its forms cannot be dissociated from its peculiar functioning;
in short, that there is a nature of human knowledge that determines its
forms and that can at the same time be made manifest to it in its own
empirical contents. There were also analyses that . . . functioned as a sort
of transcendental dialectic; by this means it was shown that knowledge
had historical, social or economic conditions, that it was formed within
the relations that are woven between men, and that it was not independent of the particular form they might take here or there; in short that
there was a history of human knowledge which could both be given to
empirical knowledge and which prescribe its forms . . . they claim to be
able to rest entirely on themselves, since it is the contents themselves that
function as transcendental reflection. But in fact the search for a nature
or a history of knowledge, in the movement by which the dimension
proper to the critique is fitted over the contents of empirical knowledge,
already presupposes the use of a certain critique – a critique that is not
the exercise of pure reflection, but the result of a series of more or less
obscure divisions.
(OT, 319)17
Both Foucault and Husserl thus claim that naturalism and historicism
stand on the same epistemological ground. Naturalism, which presupposes that empirical sciences can answer philosophical questions
about the possibility of knowledge, and historicism, according to which
17 ‘Car le seuil de notre modernité n’est pas situé au moment où on a voulu appliquer
à l’étude de l’homme des méthodes objectives, mais bien le jour où s’est constitué un
doublet empirico-transcendantal qu’on a appelé l’homme. On a vu naı̂tre alors deux
sortes d’analyses: celles qui se sont logées dans l’espace du corps . . . on y découvrait que
la connaissance avait des conditions anatomo-physiologiques, qu’elle se formait peu à
peu dans la nervure du corps, qu’elle y avait peut-être un siège privilégié, que ses formes
en tout cas ne pouvaient pas être dissociées des singularités de son fonctionnement; bref,
qu’il y avait une nature de la connaissance humaine qui en déterminait les formes et
qui pouvait en même temps lui être manifestée dans ses propres contenus empiriques.
Il y a eu aussi les analyses qui . . . ont fonctionné comme une sorte de dialectique
transcendantale; on montrait ainsi que la connaissance avait des conditions historiques,
sociales, ou économiques, qu’elle se formait à l’intérieur des rapports qui se tissent entre
les hommes et qu’elle n’était pas indépendante de la figure particulière qu’ils pouvaient
prendre ici où là, bref qu’il y avait une histoire de la connaissance humaine, qui pouvait
à la fois être donnée au savoir empirique et lui prescrire ses formes . . . elles prétendent
pouvoir ne reposer que sur elles-mêmes, puisque ce sont les contenus eux-mêmes qui
fonctionnent comme réflexion transcendantale. Mais, en fait, la recherche d’une nature
ou d’une histoire de la connaissance, dans le mouvement où elle rabat la dimension
propre de la critique sur les contenus d’une connaissance empirique, suppose l’usage
d’une certaine critique. Critique qui n’est pas l’exercise d’une réflexion pure, mais le
résultat d’une série de partages plus ou moins obscurs.’ (MC, 329–30)
t h e fouc au lt ian failu re o f phe nom e nol ogy
factual history is capable of providing these answers, have common
roots in how man is understood as an object of their study while at the
same time, although this is not admitted, he is the foundation of all
knowledge. The paradoxical role man has as the source of all knowledge and as an object of it is ignored. Discourse attempting to be both
empirical and critical cannot but be both positivist and eschatological
at the same time. ‘Man appears within it as a truth both reduced and
promised. Precritical naivete holds undivided rule’ (OT, 320).
Foucault presents phenomenology as ‘a radical contestation of positivism and eschatology’ and argues that it ‘has tried to restore the
forgotten dimension of the transcendental’ and to ‘exorcise the naive
discourse of truth reduced wholly to the empirical’ (OT, 321). However,
phenomenology fails because it remains caught in the mode of thought
that is structured by the paradoxical figure of man as an empirical/
transcendental double.
In his late work, Husserl explored the idea that the life-world
becomes the determiner of the subject’s experiences in the sense that
he or she is an empirical object limited by his or her environment. At the
same time, Husserl also became convinced of the transcendental importance of intersubjectivity, and consequently of the transcendental significance of the life-world. The life-world becomes the horizon of meaning
that makes individual meaning-giving acts possible. Therefore if we are
to understand the experiences that are ultimately constitutive of the
world, we need to elucidate their implicit background, to make this
background an object of phenomenological analysis. Husserl wanted
to show how, by means of the two epoches, the initially unthinkable
background could be elucidated. The first epoche brings the life-word
into the view of the phenomenologist, making it possible for him to
proceed to analyze its structures. The second epoche places the phenomenologist above this necessary horizon of his own thought, and
therefore enables him to study its constitution in transcendental intersubjectivity.
According to Foucault, the second epoche leads to the paradox of
man as an empirical/transcendental double. This is because the knowing subject is essentially understood as fundamentally dependent on
positivities it did not create.
For can I, in fact, say that I am this language I speak, into which my
thought insinuates itself to the point of finding in it the system of all its
own possibilities, yet which exists only in the weight of sedimentations
fo u c au lt o n freedom
my thought will never be capable of actualizing altogether? Can I say that
I am this labour I perform with my hands, yet which eludes me not only
when I have finished it, but even before I have begun it? Can I say that I
am this life I sense deep within me, but which envelops me both in the
irresistible time that grows side by side with it and poses me for a moment
on its crest, and in the imminent time that prescribes my death?
(OT, 324–5)18
According to Foucault, Husserl was trying to make explicit ‘the horizon that provides experience with its background of immediate and disarmed proof’ (OT, 327). But he could only accomplish this through the
already existing figure of man, and was therefore trapped in inevitable
circularity. The life-world makes all thought and action of the subject
possible, it is the horizon within which all beings must be situated in
order to be. Yet Husserl’s aim was to show how the world in its pregivenness can be understood through the constitutive acts of the phenomenologizing subject. How can a being that can only have validity
within the pregiven world constitute that world? Foucault concludes:
‘The phenomenological project continually resolves itself, before our
eyes, into a description – empirical despite of itself – of actual experience, and into an ontology of the unthought that automatically shortcircuits the primacy of the “I think”’ (OT, 326).
Husserl himself, however, was well aware of the apparent circularity
of his project. He himself formulates Foucault’s ‘paradox of man’ in
The Crisis:
How can a component part of the world, its human subjectivity, constitute
the whole world, namely constitute it as its intentional formation, one
which has always already become what it is and continues to develop,
formed by the universal interconnection of intentionally accomplishing
subjectivity, while the latter, the subjects accomplishing in cooperation,
are themselves only a partial formation within the total accomplishment?
The subjective part of the world swallows up, so to speak, the whole
world and thus itself too. What an absurdity! Or is this a paradox which
18 ‘puis-je dire, en effet, que je suis ce langage que je parle et où ma pensée se glisse au point
de trouver en lui le système de toutes ses possibilités propres, mais qui n’existe pourtant
que dans la lourdeur de sédimentations qu’elle ne sera jamais capable d’actualiser
entièrement? Puis-je dire que je suis ce travail que je fais de mes mains, mais qui
m’échappe non seulement lorsque je l’ai fini, mais avant même que je l’aie entamé?
Puis-je dire que je suis cette vie que je sens au fond de moi, mais qui m’enveloppe à la
fois par le temps formidable qu’elle pousse avec soi et qui me juche un instant sur sa
crête, mais aussi par le temps imminent qui me prescrit ma mort?’ (MC, 335)
t h e fouc au lt ian failu re o f phe nom e nol ogy
can be sensibly resolved, even a necessary one, arising necessarily out
of the constant tension between the power of what is taken for granted
in the natural objective attitude . . . and the opposed attitude of the
‘disinterested spectator’?
(Husserl 1954/1970a, 179–80)19
Husserl continues to argue that the paradox vanishes once the epoche
has been radically and universally carried out. At this point, we do
not have human beings either as subjects constituting the world or as
objects dependent on it, because we have achieved the ‘attitude above
the subject–object correlation which belongs to the world and thus the
attitude of focus upon the transcendental subject–object correlation’
(Husserl 1954/1970a, 181). We are led to recognize the constitutive
function, not of human beings, but of transcendental subjectivity constitutive even of the ‘phenomena’ of human beings (180–3). After the
reduction, the reflecting subject is annulled as man, and transcendental subjectivity, previously concealed, reflectively turns to inquire about
itself. Thus, for Husserl, Foucault’s paradox of man is only a paradox if
the reduction has not been accomplished. Man as both a subject and
an object of knowledge is a figure of the natural attitude and its reality
is suspended in the reduction. Underlying both sides of the double is
transcendental subjectivity, above or beyond the subject/object distinction and constitutive of all worldly objectivities.
What still remains a problem in Husserl’s account is how mundane
subjectivity and transcendental subjectivity are related to each other, if
they cannot be conflated. Transcendental subjectivity is not a human
being in either one of its two aspects or sides. Yet, in order to reveal
transcendental subjectivity, the starting point of the investigation must
be mundane subjectivity, subjectivity as it is in everyday life. The phenomenologist is thus both part of the universal field of the life-world,
as well as curiously being able to stand outside of it. Husserl’s assistant
19 ‘Wie soll ein Teilbestand der Welt, ihre menschliche Subjektivität, die ganze Welt konstituieren, nämlich konstituieren als ihr intentionales Gebilde? – Welt, ein immer schon
gewordenes und fortwerdendes Gebilde des universalen Konnexes der intentional leistenden Subjektivität – wobei sie, die im Miteinander leistenden Subjekte, selbst nur
Teilgebilde der totalen Leistung sein sollen? Der Subjektbestand der Welt verschlingt
sozusagen die gesamte Welt und damit sich selbst. Welch ein Widersinn. Oder ist es
doch eine sinnvoll auflösbare, sogar eine notwendige Paradoxie, notwendig entspringend aus der beständigen Spannung zwischen der Macht der Selbstverständlichkeit der
natürlichen objektiven Einstellung . . . und der sich ihr gegenübersetzenden Einstellung
der “uninteressierten Betrachters”?’ (Husserl 1954/1962, 183)
fo u c au lt o n freedom
Eugen Fink formulates this problem clearly in The Sixth Cartesian Meditation.20 He takes issue precisely with the relationship between the transcendental ego and the human ego, with their necessary difference on
the one hand, and their necessary identity on the other.21 He raises
two questions in particular: (1) how ‘pre/non-existent’ transcendental
subjectivity relates to mundane/human subjectivity, forming a unity in
difference; and (2) how that same transcendental agency as an absolute
constitutive source relates to the world that is its constitutive end product, forming with it a unity in bipolar differentation (Bruzina 1995,
lvi). Fink indicates and anticipates solutions by arguing that the ‘fullsided subject’ of phenomenology is neither the transcendental subject
taken purely in its transcendentality, nor the human subject taken as
uninvolved with the transcendental, but it is rather transcendental subjectivity appearing in the world. He writes:
In the universal epoche, in the disconnection of all belief-positings,
the phenomenological onlooker produces himself. The transcendental
tendency that awakens in man and drives him to inhibit all acceptednesses nullifies man himself; man unhumanizes himself in performing
the epoche, that is, he lays bare the transcendental onlooker in himself,
he passes into him. This onlooker, however, does not first come to be by
the epoche, but is only freed of the shrouding cover of human being.
(Fink 1932/1995, 39–40)22
20 There are different interpretations as to the extent to which Fink’s presentation of
phenomenology is attributable to Husserl. Ronald Bruzina (1995, xxviii, xxxii), for
example, writes in his translator’s introduction to Fink’s Sixth Cartesian Meditation that
Husserl’s phenomenology, at least as it reached its maturity in his last years, was not
just Husserl’s – it was Husserl’s and Fink’s. The differences from Husserl that emerged
in Fink’s thinking were genuine problems for and within transcendental phenomenology, which developed intrinsically within it rather than antagonistically confronting or
undercutting it from the outside. According to Bruzina, Husserl agreed in principle
with what Fink was writing, even though he might not have grasped the depth of the
implications of Fink’s thought or the radicality with which the fundamental ideas of
phenomenology were being challenged, thus implying the need for critical rethinking.
Dan Zahavi (1994), on the other hand, argues that Fink’s position is ultimately incompatible with and fundamentally foreign to Husserl’s approach, because Fink did not
acknowledge the constitutive importance of transcendental intersubjectivity, or consequently the radical transformation that Husserl’s thinking underwent in the last period
of his life, due to his preoccupation with intersubjectivity.
21 See in particular chapter 5, ‘Phenomenologizing as the Action of Reduction’, and
chapter 8 ‘Phenomenologizing as a Theoretical Experience’.
22 ‘In der universalen Epoché, in der Ausschaltung aller Glaubenssetzungen, produziert
sich der phänomenologische Zuschauer selbst. Die transzendentale Tendenz, die im
Menschen erwacht, die ihn dazu treibt, einmal alle Geltungen zu inhibieren, hebt den
t h e fouc au lt ian failu re o f phe nom e nol ogy
Fink thus sought to avoid the paradox by positing a difference in a unity.
He recognized that the phenomenologizing onlooker must always operate with already given habituated abilities which, however, must go
through a transformation after the reduction. To show exactly how
this peculiar transformation takes place ‘would be a major particular
and far-reaching task of the transcendental theory of method’ (70).
I will not go more deeply into Fink’s suggested solution here.23 My
point is simply to show that Foucault’s paradox of man as an empiricotranscendental double reiterates the phenomenological question of
the ‘full-sided subject’ – how mundane subjectivity and transcendental
subjectivity are related to each other.
The cogito and the unthought. I will now move on to the second of Foucault’s pradoxes – cogito and unthought – which surfaces as a consequence of the phenomenological method: self-reflection as a way of
investigating subjectivity. The substance of this paradox is that the phenomenologizing subject is fundamentally constituted by what eludes
his reflection – the unthought – yet reflection is the method for illuminating it.
Foucault claims that, while modern thought has been forced to focus
on the thinking subject as the precondition of knowledge, the modern
cogito – for example Husserl’s transcendental ego – is not transparent
to itself. It cannot reduce the whole being of things to thought in the
way Descartes’ cogito could ‘without ramifying the being of thought
right down to the inert network of what does not think’ (OT, 324).
Self-reflection can no longer lead to the affirmation of being, or, as
Foucault writes, ‘I think’ does not anymore lead to the evident truth
of ‘I am’. The cogito cannot provide epistemic immediacy and selfcertainty because there are prereflective conditions of knowledge that
obfuscate the evident truths of reflection. Instead, modern subjectivity is
permeated by an unthought that eludes reflection, but that nevertheless
must determine the ways of questioning it.
Menschen selbst auf, der Mensch entmenscht sich im Vollzug der Epoché, d.h. er legt
den transzendentalen Zuschauer in sich frei, er vergeht in ihn. Dieser ist aber nicht
erst durch die Epoché geworden, sondern ist nur frei geworden von der verhüllenden
Verkleidung des Menschseins.’ (Fink 1932/1988, 43–4)
23 Of the contemporary phenomenologists, J. N. Mohanty and David Carr have extensively
studied the problem of the relation between the transcendental subject and the empirical subject. See e.g. Mohanty 1989, 1997, and Carr 1999. On the relationship between
Foucault’s critique of the paradox of subjectivity and contemporary phenomenological
understandings of it, see Oksala 2003.
fo u c au lt o n freedom
Husserl claims that the life-world can be elucidated as a realm
of subjective phenomena that has remained anonymous (Husserl
1954/1970a, 111). Although it is essentially an obscure horizon, it
can still be opened up through methodological regressive inquiry, taking the phenomenologist back to its constitution in transcendental
intersubjectivity. The phenomenological attempt to describe the pregivenness of the life-world in terms of its transcendental constitution
is, however, itself a theoretical activity. Reflection is always already a
theoretical attitude that involves an objectification of that which is the
object of it. This inevitably means that the life-world, when elucidated
through phenomenological description, will lose its pregivenness. A
second problem is that not only is the pregiven experience inaccessible
to reflection, but it is also inexpressible. All expression, any attempt to
put something into words, objectifies. It transforms lived experience
into conceptual entities, such as the life-world, for example.
Foucault describes phenomenology as ‘the endeavor to raise the
ground of experience, the sense of being, the lived horizon of all our
knowledge to the level of discourse’ (OT, 299). The phenomenologist
thus argues both that the life-world is pretheoretical, even prelinguistic, and that its essential structures can be described and articulated
through phenomenological analysis; life-world ontology and transcendental phenomenological inquiry. Foucault writes about phenomenology:
What it touches it immediately causes to move: it cannot discover the
unthought, or at least move towards it, without immediately bringing it
nearer to itself – or, even, perhaps, without pushing it further away, and
in any case without causing man’s own being to undergo a change by that
very fact, since it deploys the distance between them.
(OT, 327)24
The act of describing the life-world necessarily gives it a theoretical form, which futhermore becomes constitutive of man’s own selfunderstanding, since he is dependent on the life-world as a part of it.
The pure description of the pregiven life-world becomes, in fact, an
interpretative act, constitutive of both the world around us as well as of
ourselves as parts of it.
24 ‘Elle fait aussitôt bouger ce qu’elle touche: elle ne peut découvrir l’impensé, ou du
moins aller dans sa direction, sans l’approcher aussitôt de soi, – ou peut-être encore
sans l’éloigner, sans que l’être de l’homme, en tout cas, puisqu’il se déploie dans cette
distance, ne se trouve du fait même altéré.’ (MC, 338)
t h e fouc au lt ian failu re o f phe nom e nol ogy
Furthermore, the unthought in Husserl’s phenomenology does not
only refer to the prereflective horizon of all thought and praxis. Husserl
also concedes that the intentional activity of the subject is founded
upon and conditioned by an obscure and blind passivity, by drives and
association. He argues in Analysen zur passiven Synthesis, for example,
that there are constitutive processes of an anonymous and involuntary
nature taking place in the underground or depth dimension of subjectivity, which can only be uncovered through an elaborate ‘archaeological effort’ (Zahavi 2001, 2). For Husserl, this ‘archaeological effort’
means a reflective investigation of consciousness. In the context of
Foucault’s criticism, this means that a paradox surfaces again, albeit
in a different form. The task is to make explicit through reflection
something that, by definition, eludes reflection. Through reflective
inquiry the depth-dimension of subjectivity becomes thematic and loses
its prereflective elusiveness. As Dan Zahavi notes, reflection does not
merely repeat or copy the original experience, it changes the givenness of the experience reflected upon – otherwise there would be no
need for reflection (Zahavi 2001, 6). Thus, the paradox inherent in
reflection means that we are obviously confronted with a fundamental
limit: when I reflect, I encounter myself as a thematized ego, whereas
functioning subjectivity always eludes my thematization and remains
anonymous (8).
Unlike Foucault, however, Zahavi argues that this does not constitute a major sceptical challenge for the phenomenological enterprise,
but only creates a harmless and unavoidable impasse. Firstly, although
reflection cannot apprehend the anonymous life in its very functioning, neither is it supposed to. The aim is to lift the naivete of prereflective experience, not to reproduce it. Secondly, although it must
be acknowledged that there are depth-dimensions in the constitutive
process that do not lie open to the view of reflection, this does not necessarily imply that they remain forever completely ineffable, beyond
phenomenological investigation. They can be disclosed, not through
direct thematization, but through an indirect operation of dismantling
and deconstruction (Zahavi 2001, 8–9). I do not take a stand on this
question. My point, again, is simply to show that Foucault’s criticisms
of the double figure of man as a structural problem inherent in phenomenology – the phenomenologizing subject as both the object and
subject of reflection – is accurate in the sense that it points to a genuine
impasse in the phenomenological project, even if it is understood as a
harmless one.
fo u c au lt o n freedom
The problem of explicating the unthought linguistically can be
extended to include the explication of any of the results of the phenomenological inquiry after the reduction. If the epoche is carried
out radically, then all wordly habitualities and abilities, such as logic,
conceptuality and language, must also fall subject to bracketing. The
phenomenologist must first bracket the existence and validity of language, but he is then forced to bring it back through the back door so
to say, to be able to explicate and communicate the findings of his phenomenological inquiries. All notions like ‘transcendental ego’ and ‘flow
of consciousness’ revealed by the epoche are worldly concepts already
tied to shared and existing meanings and linguistic expressions.25
Husserl only touches on the problem of language in his analyses
of the life-world.26 In The Crisis he notes that, after the transcendental epoche has been performed and the life-world has become a mere
‘component’ within transcendental subjectivity, words taken from the
sphere of the natural attitude become dangerous and ‘the necessary
25 Derrida presents one of the most influential forms of this criticism in his introduction
to Husserl’s essay ‘The Origin of Geometry’. In this essay, too, Husserl inquires into
the production of the ideal objects of science, and shows how they originate in the
life-world, from its pre-scientific, sensibly evident truths. Derrida shows (1962/1989,
63) that in order to study how the subjective egological evidence of the senses derived
from the life-world can give rise to the ideal objects of science, Husserl must ask how
this sense-evidence becomes objective and intersubjective. Derrida credits Husserl with
acknowledging that the only possibility is through language. According to Derrida,
however, Husserl ultimately fails to question how his own project must also depend on
language. He argues that Husserl distinguishes ideal objectivities from the concepts
of language. Ideal objects must be free from all factual subjectivity, from any de facto
language and also from the fact of language in general. Nevertheless, Husserl claims that
ideality comes to objectivity by means of language. According to Derrida, the paradox
that emerges is, thus, that although objective ideality must be independent of language,
nevertheless ‘without the apparent fall back into language and thereby into history, a
fall which would alienate the ideal purity of sense, sense would remain an empirical
formation imprisoned as fact in a psychological subjectivity – in the inventor’s head.
Historical incarnation sets free the transcendental, instead of binding it’ (77). Hence,
while being aware that only language can make possible ideal objectivity, Husserl does
not follow up the consequences that this idea has for the phenomenological enterprise.
26 Earlier, Husserl dealt with language extensively in Logical Investigations. The First and
Forth Investigations put forward a theory of expression and signification. This was before
Husserl had introduced the idea of reduction in Ideas I, however, and the question of
language in connection with the phenomenological reduction could thus not be initially asked. The second edition of Logical Investigations, which came out in 1913, was
extensively rewritten by Husserl in the light of his new understanding of phenomenology. In the introduction to the second volume he takes up the problem of language,
but solves it rather quickly by referring to a necessary transformation of sense. (Husserl
1913/2001, 171–2.) For more on language in Husserl, see e.g. Mohanty 1976, Edie
1976, Hutcheson 1981.
t h e fouc au lt ian failu re o f phe nom e nol ogy
transformation of their sense must therefore be noticed’ (Husserl
1954/1970a, 174). He does not, however, explain how this transformation takes place or what exactly it means.
Fink took up the problem of language in more detail and set out to
solve it in The Sixth Cartesian Meditation. He claimed at the outset that
the problem would disappear ‘if there could be a proper transcendental
language’ (Fink 1932/1995, 84), but that phenomenological inquiry
does not lead to the construction of a new language, nor could it ever
do so. ‘Language is indeed retained as habituality right through the
epoche . . . the phenomenological onlooker must make use of it, if he
at all wants to give predicative expression to his cognitions’ (86).
Like Husserl, Fink saw a solution emerging from the necessary tranformation of sense. He argues that, in taking over language, the phenomenologizing onlooker transforms its natural sense. ‘If this kind of
transformation did not occur, then the phenomenologist would slip out
of the transcendental attitude every time he spoke’ (Fink 1932/1995,
86). He describes the transformation of sense as an uneasy tension, a
rebellion inside the words of the natural attitude:
On the one hand, the natural meaning of the word and sentence points
analogously to a corresponding transcendental sense, while, on the other
hand, the intended transcendental meaning protests, as it were, against
its expressional formulation; the sense to be expresssed does not rest
quietly in the expressional form, it is in constant rebellion against the
constraint imposed upon it by the formulation in natural attitude.
(Fink 1932/1995, 88–9)27
The necessary transformation of sense taking place in language has
important implications for the phenomenological method. Although
phenomenology becomes communicable through its necessary articulation in language, it must still retain a radically personal character.
There is thus no phenomenological understanding that comes simply by
reading reports of phenomenological research; these can only be ‘read’
at all by performing the investigations themselves. Whoever fails to do
that just does not read phenomenological sentences; he reads queer
27 ‘Einerseits weist die natürliche Wort-und Satzbedeutung analogisierend auf einen
entsprechenden transzendentalen Sinn hin, andererseits aber protestiert gleichsam die intendierte transzendentale Bedeutung gegen ihre Ausdrucksfassung; der
auszudrückende Sinn kommt in der Ausdrucksform nicht zur Ruhe, er ist in ständiger
Rebellion gegen den ihm durch die Fassung in natürlichen Worten und Sätzen angetanen Zwang.’ (Fink 1932/1988, 97–8)
fo u c au lt o n freedom
sentences in natural language, taking a mere appearance for the thing
itself to his own self-deception . . . The transformation that natural
language, as expressive of that which is existent, undergoes in being
claimed by the phenomenologizing I must always be kept in mind as
transformation of ontic-naive meanings into ‘analogically’ indicated,
transcendental-ontic meanings. It signifies a lapse into ‘dogmatism’ (that
of the natural attitude) if explicit knowledge of this necessary transformation dies away, and the phenomenologist thereby in his explications
falsifies the object of his theoretical experiences.
(Fink 1932/1995, 92–3)28
Although the necessary transformation of sense taking place after the
reduction seems to alleviate the problematic role of language in the
phenomenological method, it still remains a difficult step to understand. An obvious objection springs to mind: how can we know that
the personal transformation that I, as a phenomenologist, undergo in
performing the epoche is the same as the transformation of other phenomenologists, and thus results in an identical transformation of the
sense of language? The transformation of sense must enable communication in the same language. Even though the transformation of sense
resulting from the epoche must be absolutely singular, it must also be
generic enough not to result in multiple ‘queerings’ of language.
To properly deal with these questions, one would have to go into
Husserl’s complex account of semantics in Logical Investigations and
Formal and Transcendental Logic. I will not do this, nor will I take a
stance on the extent to which Fink manages or even attempts to solve
the question of phenomenological language. Nevertheless, he should
be credited with a clear formulation of the problem: he inquires into the
necessary condition of possibility of the phenomenological project in
language, and asks how its findings can be expressed in a language that
is essentially a worldly phenomenon. My point is to show that against this
28 ‘Es gibt hier demnach kein phänomenologisches Verstehen durch das blosse Lesen
phänomenologischer Forschungsberichte, sondern solche können überhaupt erst ‘gelesen’ werden im Nachvollzug der Forschungen selbst. Wer das unterlässt, liest gar
nicht phänomenologische Sätze, sondern liest absonderliche Sätze der natürlichen
Sprache, nimmt die blosse Erscheinung für die Sache selbst und betrügt sich . . . Die
Verwandlung, die die natürliche Sprache, als Aussprechen von Seiendem, durch die
Inanspruchnahme durch das phänomenologisierende Ich erfährt, muss immer als
Verwandlung der ontisch-naiven Bedeutungen in die sich “analogisch” anzeigenden
transzendental-ontischen Bedeutungen bewusst bleiben. Es bedeutet ein Verfallen in
den “Dogmatismus” (der natürlichen Einstellung), wenn das ausdrückliche Wissen um
die notwendige Verwandlung erlischt und damit der Phänomenologe den Gegenstand
seiner theoretischen Erfahrungen auslegend verfälscht.’ (Fink 1932/1988, 101–2)
t h e fouc au lt ian failu re o f phe nom e nol ogy
background, Foucault’s attempt to find the conditions of possibility of
knowledge in the workings of discourse – the being of language instead
of the being of man – is not only a break with phenomenology, but also
an effort to solve one of its central problems.
The retreat and the return of the origin. The third paradox – the return
and the retreat of the origin – refers to man as both dependent on a
history whose beginning will always elude him, and as the condition of
possibility of writing that history. Man is essentially a historical being,
he is burdened by a history that is not of his own making and ‘it is always
against a background of the already begun that man is able to reflect
on what may serve for him as origin’ (OT, 330). Yet the world becomes
a historical reality through man: ‘It is in him that things (those same
things that hang over him) find their beginning’ (OT, 332). According
to the phenomenologists, the structures of human consciousness are
understood as the condition of possibility of all factual history.
Husserl introduces the notion of historical a priori in his essay ‘The
Origin of Geometry’. He emphasizes the historicity of ideal entities and
theories by arguing that they always have a historical origin, and that
this fact is essential to them and to their constitutive power. Sciences
can only stand ‘within the historical horizon in which everything is historical’ (Husserl 1939/1989, 172). Husserl goes on to argue that this
historicity cannot be revealed by factual history, in which the conclusions are always drawn naively and straightforwardly from the facts. It
‘never makes thematic the general ground of meaning upon which all
such conclusions rest, has never investigated the immense structural
a priori which is proper to it’ (174).
Husserl thus again argues against the naivety of historicism in regard
to the grounding of the sciences as well as the resulting relativism. He
proposes a methodological inquiry that can reveal the essential structure of the necessary historical horizon of all sciences, ‘the apriori structure contained in this historicity’ (Husserl 1939/1989, 172). This is
the historical a priori for Husserl. The proposed method for revealing
it is free variation: ‘In running through the conceivable possibilities
for the life-world, there arises, with apodictic self-evidence, an essentially general set of elements going through all the variants, and of
this we can convince ourselves with truly apodictic certainty’ (177).
Even if we know almost nothing of the historical surrounding world
of the first geometers, we do know that it had an invariant, essential
structure, which could be revealed to us through the method of free
fo u c au lt o n freedom
variation. Thus, the original meaning of geometry can be rediscovered
by us.
It follows that a study of the historical a priori is essential for Husserl,
because all factual history presupposes it. Without it, historical inquiry
would be a meaningless enterprise, since it alone can provide ‘the
truly apodictic self-evidence extending beyond all historical facticities’
(Husserl 1939/1989, 175). An analysis of the historical a priori provides
the ground on which it is possible to identify or recognize different cultural codes and historical events. ‘We need not first enter into some
kind of critical discussion of the facts set out by historicism; it is enough
that even the claim of their factualness presupposes the historical
a priori if this claim is to have a meaning’ (176).
Although geometrical idealities are thus essentially historical for
Husserl in the sense that they can only originate in concrete historical events, they only become understandable as idealities on the basis
of a universal and apodictic horizon of meaning. The historical a priori
signifies this universal structure of meaning. As Beatrice Han writes,
Husserl’s a priori is not ahistorical like the Kantian apriori, but it is
‘suprahistorical’: it exists essentially to guarantee the possibility of recovering, beyond the sedimentations of history and tradition, the primary
evidences originally thematized by the ‘proto-founder’ of geometry
(Han 1998/2002, 4). The historical a priori thus conveys the historicity
of scientific idealities, but this historicity means only that they originate in history – in contrast to all formal a prioris – but they do not
depend on factual history in the sense that they would change in it. The
original meaning of geometry can be reactivated because, for Husserl,
‘the human surrounding world is the same today and always’ (Husserl
1939/1989, 180).
The way Foucault appropriates Husserl’s notion of historical a priori
can be seen as being symptomatic of the more general way in which he
both criticizes and reappropriates phenomenology. By developing his
own version of the historical a priori, he admits the necessity of finding
a historical version of the transcendental, and thus acknowledges the
importance of the Husserlian effort. Foucault’s reformulation of the
historical a priori is, however, a critical reaction to the problems he
identifies in it in its Husserlian form.
By appropriating Husserl’s historical a priori, Foucault acknowledges
the claim that factual history cannot reveal the basis of knowledge,
which is also its own condition of possibility. This is why he clearly distinguished his archaeology from the history of ideas, representations and
t h e fouc au lt ian failu re o f phe nom e nol ogy
modes of behaviour. He claims that there is a more fundamental level
that structures and unites observations, discussions and concepts by
ordering and determining what can appear as an object of knowledge
and how it can be known. This transcendental level is referred to as
the historical a priori by Foucault. For him, however, the historical
a priori is not a universal structure of meaning, but, rather, is revealed
by factual history. It changes in history while also forming the conditions of possibility of knowledge of a period. It is thus not a principle
of the constitution of objects in general, but always limited to local and
particular forms of knowledge.29
Foucault’s method of studying the historical a priori is different, too.
One of the central tenets of transcendental philosophy is that, while
the aim is to study the conditions of possibility of thought and experience, we cannot, by definition, step outside of them and adopt a view
from nowhere. This is why the phenomenological method is essentially
characterized by reduction as the only way to achieve clarity about the
conditions of possibility constituting our thought while starting from
within our own experience. Foucault rejects the possibility that historical a priori could be revealed by starting from the meaning-giving activity
of a subject. For him, the only means of studying the transcendental is
through history: while it is impossible to study the conditions of possibility of our own thought, it is possible to reveal the fundamental level
determining the order of knowledge of a different age.
Hence, Foucault rejects the transcendental subject and postulates
instead a transcendental without a subject. In this way he seeks to solve
the paradox of man in all its forms. He also historized the transcendental more radically by introducing his own version of the historical
a priori that changes in history. Through his criticism of phenomenology in OT, he sought to show the inadequacy of the phenomenological
understanding of language as well as of the history of science, and he
supported the radical historicity of the conditions of possibility of knowledge. By explicating the paradox of man that the phenomenological
method leads to, he stressed the necessity of finding a new method. I
will now take a closer look at Foucault’s proposed method, archaeology,
of which the ‘aim is to free history from the grip of phenomenology’
(AK, 203). Is there philosophy after the death of man?
29 Cf. Han 1998/2002, 64–5.
The Order of Things has been severely criticized by both historians and
philosophers.1 Since its publication, the philosophical criticism has centred around two themes. Firstly, a common charge is that Foucault does
not problematize his own position, but assumes it to be situated outside
of the epistemic orders he studies. This means that he ends up reiterating the problem of empirical/transcendental circularity of which
phenomenology stands accused. Secondly, it has been claimed that
Foucault’s alternative to the subject-centred approach of phenomenology leads to serious difficulties in conceiving change and consequently
also freedom. Archaeology is therefore a step backwards rather than a
step forwards from phenomenology: it does not manage to solve the
problems with which phenomenology is riddled, but rather adds to
them by creating a host of new ones.
Both strands of the criticism are connected to the question of the
subject. On the one hand it raises questions about the subject as the
1 Gary Gutting (1989, 175–9, 221) notes that, of all Foucault’s books, OT has been the
most severely criticized by historians. In order to assess the impact of these criticisms,
he distinguishes several different historical levels on which OT operates. The first is that
of specific history: the interpretation of particular texts in their own terms. Second, the
level of constructive history, which builds general interpretative frameworks connecting
a range of texts. Third is the level of critical history: the use of the outcomes of specific and constructive history to question the self-understanding of various contemporary
disciplines. Following this schema, Gutting argues that Foucault’s interpretations of particular authors have drawn some criticism, but the primary objection has been to the lack
of detailed evidence for the sweeping claims of his constructive history. Nevertheless, he
defends Foucault by pointing out that the value of his work is most of all as a source
of fruitful suggestions rather than as accurate generalizations. Foucault’s account of the
modern episteme is important primarily as the basis of his critical history of the human
sciences. My focus here is on the philosophical criticism of Foucault’s archaeology. For
more on Foucault as a historian, see e.g. Veyne 1971/1997, Flynn 1994 and Goldstein
th e a n o n ymity o f la n g ua ge
writer of its own history. Who is the writer of archaeology? To what extent
was Foucault himself determined by the discursive structures under
study? The other strand concerns the subject as the agent of change.
How can we understand change if we do not study the intentions and
motives of the subject? Does freedom not become an impossible idea?
My aim in this chapter is to explicate the question of the subject in
connection with the criticism of OT. I will not attempt to clear up all the
ambiguities in Foucault’s archaeology, however. It contains several contradictions, problems and errors that have been thoroughly discussed
by other commentators.2 My principle goal is to show how the central,
philosophical role Foucault assigns to language has important consequences in terms of how we conceive of our subjectivity as well as of our
freedom. I will show that, despite its firm refusal to accept the subject
as the basis of explanation, Foucault’s archaeology does not eradicate
it as a question. Neither does it eradicate freedom, but charts instead
new dimensions of it, which are not tied to the subject’s expressions
and initiatives, but rather make them possible.
A view from nowhere
A central theme in the criticism of OT has been the question of the
starting point of the description. What is Foucault’s own position with
regard to the epistemes that condition all scientific practices? Is his own
thought not also inevitably determined by the epistemic order underlying his archaeology? How can he analyze, using language, the grid
that orders language? How can he escape the order that our discourse
constitutes, and understand an order of a different kind, a different
Dreyfus and Rabinow suggest in their influential book Michel Foucault, Beyond Structuralism and Hermeneutics, that the archaeologist is situated outside of the epistemic orders he studies and therefore does
not speak from within any horizon of intelligibility. They compare
2 See, in particular, Han 1998/2002. Roberto Machado (1989/1992, 17), on the other
hand, argues that one of the essential characteristics of archaeology is exactly the multiple
ways in which it can be defined, and another is its fluidity as a mode of research. The successive shifts are not marks of inadequacy or a lack of rigour: they illustrate the deliberate
and well-considered provisional nature of the analysis. The tensions in archaeology are
more prominent after the book that followed OT, The Archaeology of Knowledge, in which
Foucault claims to present one method, archaeology, that characterizes all of his work.
In reality, however, the methods, theoretical frameworks and aims in OT and in AK are
not identical. Cf. Machado 1989/1992, 12, 17; Han 1998/2002, 52–4, 67–8.
fo u c au lt o n freedom
archaeology to phenomenology in that the archaeologist, like the phenomenologist, claims to be able to accomplish a bracketing, which
enables him to situate himself outside of his own thought and thus
to study its underlying structures.3
[T]he archaeologist, like Husserl’s transcendental phenomenologist,
must perform an ‘ego split’ in order to look on as a detached spectator at the very phenomena in which, as an empirical interested ego . . .
one can’t help being involved. Foucault the archaeologist looks on, as a
detached metaphenomenologist, at the historical Foucault who can’t, if
he thinks about human beings in a serious way, help thinking in terms of
the meaning and truth claims governed by the latest discursive formation.
(Dreyfus and Rabinow 1982, 87)
Despite Foucault’s criticism of the phenomenological method and the
procedure of reduction, Dreyfus and Rabinow claim that he ends up
making a similar move himself in trying to elucidate the epistemic conditions of scientific practices. This leads him to exactly the same type
of circularity or ‘double trouble’ that he criticized phenomenology for.
If discursive practices take the place of the transcendental in the sense
that they condition what is said and known, and discourse is understood as a historically existing network of forms of knowledge and practice, then the empirical content becomes the transcendental condition.
Foucault attempted to pass from the analysis of positivities to the historical a priori foundations that provide the ground of its own possibility –
the very thing for which he had criticized phenomenology (Dreyfus
and Rabinow 1982, 93).
Dreyfus and Rabinow thus understand archaeology to be an ahistorical, ‘quasi-structuralist theory’ conducted from a position of phenomenological detachment. They criticize it for falling into the trap of
3 Dreyfus and Rabinow (1982, 85–90) argue that Foucault’s archaeology is a radicalization
of Husserl’s reductions. Their argument is that the phenomenologist only brackets the
truth of the statement he is studying, that is, its validity, but an archaeologist even brackets
its meaning. Foucault’s approach is thus more radical in the sense that the ‘bracketing’
includes more entities: not just the truth, but also the meaning. This is a problematic
characterization of both Foucault’s and Husserl’s methods, however. Heinämaa points
out that Dreyfus and Rabinow’s comparison is based on a controversial interpretation
of Husserl’s concept of noema as being similar to Frege’s sense. They presuppose that
truth and meaning are two separate entities that can be bracketed one after the other.
This is possible only if the noema is understood to be like the Fregean sense, between
the act and the reference and distinct from both. Such an understanding is by no means
self-evident or undisputed (see Heinämaa and Oksala 2000). For alternative readings of
the noema, see e.g. Drummond 1990 and Haaparanta 1994.
th e a n o n ymity o f la n g ua ge
the empirical/transcendental doublet, and hail genealogy as Foucault’s
breakthrough and corrective method, which avoids this trap by aligning itself with hermeneutical insights in certain respects. While Foucault
the archaeologist aims at viewing the discursive practices with external
neutrality, Foucault the genealogist realized that he could only describe
them from the inside. Dreyfus and Rabinow go as far as to argue that
the insoluble contradictions in archaeology ultimately led Foucault to
abandon it as his method, and to take up genealogy a few years later
(Dreyfus and Rabinow 1982, 99–100).
Their criticism of archaeology relies on some problematic assumptions, however. They understand it as a form of structuralism, a characterization that Foucault himself vehemently denied.4 They also read
Foucault’s thought in terms of a radical break: archaeology failed as
a method and was replaced by genealogy. This interpretation was also
denied by Foucault on several occasions. He claimed that archaeology
and genealogy were complementary methods of investigation representing different axes of analysis, not alternatives.5
In an interview in 1967 in which Foucault responds to the controversy
surrounding the publication of OT, he explains his own position as
the author of the book. While he argued that his position should be
understood as anonymous, he did not claim to be situated outside of
the discursive order characterizing his time. On the contrary, he was
anonymous exactly because he was situated inside his own episteme and
was a product of a specific historical development.
[I]t should be possible to define the theoretical model to which not only
my book belongs but also those which belong to the same configuration
of knowledge [savoir]. It is doubtless the one that now allows us to treat
history as a set of actually articulated statements, and language as an
4 Foucault himself explicitly refused this label in the preface to the English translation of
OT, for example, and the significance of this refusal has been further discussed by commentators. Gary Gutting (1989, 228) argues that, although there is a close link between
Foucault’s work and structuralism, we must understand why Foucault insists on several
occasions that he is not a structuralist. According to Gutting, the link derives from the
fact that, like structuralist work on language, culture and the unconscious, archaeology
displaces man from his privileged position. At the same time, it is a historical method of
inquiry, concerned not with structural possibilities but with actual occurrences and their
effects. Beatrice Han (1998/2002, 45) also points out that Foucauldian understanding
is distinguished from the structuralist model in that the historical a priori is understood
neither as universal nor as invariant; rather, it undergoes the historical transformations
that archaeology is to identify. On Foucault’s relationship with structuralism, see also
e.g. Caws 1988, Dosse 1991, 1992, Frank 1989/1992.
5 See e.g. SP, UP.
fo u c au lt o n freedom
object of description and an ensemble of relations linked to discourse, to
the statements that are the objects of interpretation. It is our age and it
alone that makes possible the appearance of that ensemble of texts which
treat grammar, natural history, or political economy as so many objects.
So, in that respect and only in that respect, the author is constitutive of
the thing he is talking about . . . So the subject is, in fact, present in the
whole book, but it is the anonymous ‘one’ who speaks today in everything
that is said.
(OWH, 286)6
Foucault self-consciously places his analysis in the general anonymity of
all the investigations that were at the time revolving around language
(OWH, 290). The new epistemic configuration signalled by the death
of man made possible his questions about language. He also explained
his method at this time by noting that ‘archaeology owes more to
Nietszchean genealogy than to structuralism properly so called’ (294).
It is my contention that the hermeneutical insight that the thinker
is always situated inside a horizon of intelligibility that conditions his
thought does not suddenly appear with genealogy, but is already central in OT. Although archaeology is strictly opposed to hermeneutics
in the sense that it does not attempt to interpret the deep meaning of
sentences, it is nevertheless a historically situated form of analysis and
this is a definitive characteristic of it.
Several commentators have remarked how OT relies heavily on
Heidegger’s anti-humanism.7 In particular, there are striking similarities between OT and Heidegger’s essay ‘The Age of the World Picture’.
Heidegger’s aim in this text, like Foucault’s in OT, is to reveal the ontological order upon which the sciences as well as the everyday practices
6 ‘on devrait pouvoir définir le modèle théorique auquel appartient non seulement mon
livre mais aussi ceux qui appartiennent à la même configuration de savoir. Sans doute estce celle qui nous permet aujourd’hui de traiter de l’histoire comme ensemble d’énoncés
effectivement articulés, de la langue comme objet de description et ensemble de relations
par rapport au discours, aux énoncés qui font l’objet de l’interprétation. C’est notre
époque et elle seule qui rend possible l’apparition de cet ensemble de textes qui traitent de
la grammaire, de l’histoire naturelle ou de l’économie politique comme autant d’objets.
Si bien que l’auteur, en cela, et en cela seulement, est constitutif de ce dont il parle . . . Si
bien que le sujet est en effet présent dans la totalité du livre, mais il est le “on” anonyme
qui parle aujourd’hui dans tout ce qui se dit.’ (SFH, 591)
7 See e.g. Han 1998/2002, Elden 2001. Foucault hardly ever referred directly to Heidegger.
In his last interview he made a surprisingly strong statement about him, however. ‘For me
Heidegger has always been the essential philosopher . . . I have never written anything
on Heidegger and I wrote only a small article on Nietzsche; these are nevertheless the
two authors I have read the most.’ (RM, 250).
th e a n o n ymity o f la n g ua ge
characteristic of an age are grounded. What becomes a question of scientific study and the method through which it is approached is determined by the ontological order through which the world is interpreted
and presents itself. Heidegger writes: ‘Metaphysics grounds an age, in
that through a specific interpretation of what is and through a specific comprehension of truth it gives to that age a basis upon which it
is essentially formed’ (Heidegger 1952/1977, 115). The metaphysical
ground plan sketches out in advance the manner in which something
can appear as an object of scientific investigation. ‘Only within the perspective of this ground plan does an event in nature become visible
as such an event’ (119). The world picture, like Foucault’s concept of
episteme, thus refers to the overall schema, the implicit order of things,
on the basis of which reality is comprehended.
This normally hidden ontological order can only be made visible by
comparing it with the different modes of thinking of the past. Heidegger
writes, for example, that the metaphysics underlying the modern age
can be characterized by throwing it into relief over against the medieval
and ancient world pictures (Heidegger 1952/1977, 128). Similarly,
Foucault emphasizes that it is only against the background of what is
different in history that the epistemic structures show up.
I can, in fact, define the classical age in its particular configuration by the
twofold difference that contrasts it with the sixteenth century, on the one
hand, and with the nineteenth century, on the other. But I can define the
modern age in its singularity only by contrasting it with the seventeenth
century, on the one hand, and with us, on the other hand; so, in order to
effect this transition, it is necessary to bring out in all our statements the
difference that separates us from it. It is a matter of pulling oneself free
of that modern age which begins around 1790 to 1810 and goes up to
about 1950, whereas for the classical age it’s only a matter of describing
(OWH, 293)8
Revealing the ontological structures of the past becomes possible only
from the vantage point of the present. It is possible for us to describe
8 ‘Je peux, en effet, définir l’âge classique dans sa configuration propre par la double
différence qui l’oppose au XVIe siècle, d’une part, au XIXe , de l’autre. En revanche,
je ne peux définir l’âge moderne dans sa singularité qu’en l’opposant au XVIIe siècle,
d’une part, et à nous, d’autre part; il faut donc, pour pouvoir opérer sans cesse le partage,
faire surgir sous chacune de nos phrases la différence qui nous en sépare. De cet âge
moderne qui commence vers 1790–1810 et va jusque vers 1950, il s’agit de se déprendre
alors qu’il ne s’agit, pour l’ âge classique, que de le décrire.’ (SFH, 598–9)
fo u c au lt o n freedom
the modern episteme only when we manage to pull ourselves free from
it and analyze it as a past that is in some important ways different from
our present. The ability to recognize the differences signals the break
that separates us from it.9
In the book that followed OT, The Archaeology of Knowledge, Foucault
avoids the notion of episteme and introduces that of archive, which he
defines as ‘the general system of the formation and transformation of
statements’ (AK, 130). In this connection, he states explicitly that it is
not possible for us to describe our own archive (AK, 130). We cannot
free ourselves from the rules ordering our own discursive practices and
submit them for archaeological analysis. Foucault thus argues against
the possibility of studying the conditions of possibility of our own knowledge at all, whether or not these conditions are understood as historical
or not. All that he presents is a historical description, a study of the conditions of existence, which is conducted from the vantage point of the
present in regard to the past.
Unlike the Husserlian phenomenologist, Foucault thus does not
claim to be able to study the constitutive conditions of our own thought
and experience. It is my contention that this impossibility is exactly why
he had to engage in historical study in his quest to understand the
semantic relationships between words and things. It is only from the
vantage point of the present ontological order that the semantic relationships of another epoch can be described. This description cannot
reveal any ultimate foundations. Not only are the ontological orders
historical, they can only appear as such from our own interpretative perspective. This perspectivism is the positive condition that allows such
orders to appear at all. Foucault’s question of the historical limits of
language, for example, is a question that can only self-consciously be
asked after the linguistic turn in philosophy. Reading history through
our questions, concepts and ways of thinking makes it possible to reveal
what is different.
Hence, Foucault does not hold that the other epistemic orders are
completely cut off from us. But neither is his archaeology anachronistic in the sense that past forms of thinking were treated as directly
accessible and understandable. Because they are based on a different
9 Thomas Flynn (1997, 251–5) argues that Foucault’s aim was to distance himself from the
modern episteme by questioning its basic presuppositions and by viewing it from without.
He refers to the end of OT, where Foucault charts the privileged positions of ethnology
and psychoanalysis in our present-day knowledge, and argues that he was undertaking an
ethnology of his own culture.
th e a n o n ymity o f la n g ua ge
ontological order, they cannot be directly accessible to us, but revealing
them requires archaeological work. George Canguilhem (1994b, 78)
explicates Foucault’s archaeology by arguing that, while the episteme
of a given era cannot be fully grasped via the intellectual history of
that era, which is subtended by the episteme of a different era, the two
are not entirely foreign to one another. What remains is precisely the
archaeological project, ‘the fact that painstakingly, slowly, laboriously,
indirectly, we can dive deep down from our own epistemic shores and
reach a submerged episteme’ (Canguilhem 1994b, 78).
Dreyfus and Rabinow (1982, 95) summarize their devastating criticism of archaeology by suggesting that if it is to avoid self-elimination it
must either only study the past, or else, like therapy and phenomenology, it must see to it that its task is interminable. It is my contention that
archaeology studies only the past. To the extent that it is intended as a
method of writing history, this should be fairly uncontroversial.10 It is
also possible to argue, however, that Foucault’s philosophical method is
history understood as archaeology. Perhaps the most fruitful way to read
OT is to see it as questioning the independence of history and philosophy. Foucault’s studies show how historical discourses and practices
are both empirical facts and constitute the only possible philosophical basis of knowledge. They can be analyzed on an empirical level to
explain historical events, and on a philosophical, ‘transcendental’ level
to reveal structural regularities and conditions of possibility. Transcendental questions are, for Foucault, of necessity questions of history: the
historical limits of knowledge and experience.11 As he writes in the
10 Gary Gutting (1989, 227–8, 244, 269) argues that archaeology must be read primarily
as a distinctive approach to the history of thought. It is a historical counterpart of the
structuralist counter-sciences and a move away from the modes of thought centred on
the concept of man. It is only open to the devastating philosophical criticism that Dreyfus
and Rabinow develop in their study of Foucault when it is understood as a general philosophical theory. Gutting argues that it should not be construed along these theoretical
lines. Foucault’s aim was not to develop a general theory of discursive regularities at all.
What might appear to be foundational philosophical theories of language, for example,
are better construed as no more than attempts to show that the archaeological approach
can be coherently formulated without relying on the modern philosophical category of
the subject.
11 Claire O’Farrell writes (1989, 33) that history reveals the limits of the formation of ideas
and objects which are both part of historical order and beyond that order. History is
thus in a unique position to attach philosophy to empirical realities while studying their
conditions of possibility at the same time. Shortly before his death Foucault himself
described his books as ‘historical studies’ but noted that they were not the work of a
historian, as they were embarked on as a ‘philosophical exercise’ (UP, 15).
fo u c au lt o n freedom
preface to OT, his aim is to bring to light the history of the conditions
of possibility of knowledge (OT, xxii).
The subject of change
In an interview, Foucault also further explained his aim of analyzing
change in terms of a structure. The interviewer mentioned that the
articles that attacked OT contained the words ‘to freeze history’, which
recurred like a leitmotif and seemed to formulate the strongest accusation against it. Foucault replied that, in the context of the history of
ideas, description of change follow two strategies: (1) one uses concepts such as influence, crisis, sudden realization or the interest taken
in the problem; or (2) one moves to a level of explanation that is exterior to the level of analysis of the statements themselves, such as an
explanation based on social conditions, mentality or the worldview. He
explained that he wanted to avoid giving himself these two expedients
and to describe statements by bringing out the relations of implication,
opposition and exclusion that might connect them. Rather than inventing radical breaks between epistemes, he was in fact trying to chart the
rapid changes in a precise way by establishing a set of transformations
between two sets of scientific discourse. This set of transformations preserves a certain number of theoretical elements and displaces certain
others (OWH, 282–3). Rather than denying change, Foucault was thus
describing it, but only on the level of discourse, the domain of statements obeying rules. He was not attempting to provide causal explanations for the changes in history, but was, in a ‘positivist’ fashion,
describing certain processes in it.12 The historical analysis of discursive practices will reveal numerous types of relationships and modes
of connection that should be analyzed in different ways and along different axes. The archaeological method represents only the horizontal
axis, which describes the theoretical coherence of discourses among
themselves in a given period (OWH, 285).
Foucault thus does not argue that discursive structures do not
change or that they contain some autonomous mechanism of change.
The subjects who engage in different types of practice still instigate the
changes, but developments in the sciences cannot be explained by the
12 Arnold Davidson (1997b, 10–13) argues that the influence of structuralism on Foucault’s
thought shows here: only a synchronic analysis allows one to define the field within which
a causal explanation can then operate. An analysis of the logical structures of knowledge
has to precede any effort to provide causal explanations.
th e a n o n ymity o f la n g ua ge
subject’s structures of consciousness. Here the fundamental influence
that Jean Cavaillès’ thought had on Foucault’s understanding of the
history of thought is apparent. Cavaillès’ philosophy was an effort to
combine necessity and radical innovation in logic and mathematics,
areas of science that are generally regarded as particularly rule-bound.
Although Cavaillès initially followed Husserl’s thought closely, in his
late work he became increasingly critical of it, arguing that mathematical objects, concepts, rules and procedures cannot be grounded
in the structures of consciousness. This means that mathematics is an
autonomous becoming; its historical development cannot be accounted
for or grounded in any discipline other than itself. Its development
is genuinely unpredictable. Analysis cannot find the new ideas within
those already in use (Cavaillès 1947/1994, 504). David Webb (2003)
summarizes Cavaillès’ position by writing that, for him, each act of thinking takes place within a defined context of rules that it did not itself
frame, but this does not represent the end of its freedom. At stake is
not the radical spontaneity of thinking, but the conviction on Cavaillès’
part that no set of formal conditions of thinking can be so primary or
fundamental that they necessarily encompass all the changes in conceptual objects and their relations that the future may bring. Thinking
may therefore engage in the necessary unfolding of ideas, while at the
same time preserving its power to participate in the transformation of
any given set of conceptual objects and relations (Webb 2003, 69).
In connection with Foucault’s very similar understanding of the history of science, we must thus interpret that the idea of an episteme
underlying the scientific thinking of an age does not imply epistemic
determinacy. Rather, the implication would be that the development of
science is always genuinely unpredictable.
In the foreword to the English edition of OT Foucault lists three
problems that were left open in the book: change, causality and the subject. While he explicitly states that he did not even attempt to address
the first two, the question of the subject was not simply left aside. It
was posed – albeit from a perspective that was different from the phenomenological one – even though he was not able to give any definitive
answer at that point.
I do not wish to deny the validity of intellectual biographies, or the possibility of a history of theories, concepts or themes. It is simply that I
wonder whether such descriptions are themselves enough, whether they
do justice to the immense density of scientific discourse, whether there
fo u c au lt o n freedom
do not exist, outside their customary boundaries, systems of regularities
that have a decisive role in the history of the sciences. I should like to
know whether the subjects responsible for scientific discourse are not
determined in their situation, their function, their perceptive capacity,
and their practical possibilities by conditions that dominate and even
overwhelm them.
(OT, xiii–xiv)
One of Foucault’s questions in OT was thus exactly to what extent subjects are determined by rules and structures unknown to them. His aim
was to chart the conditions of possibility of their initiatives. Rather than
eradicating the subject or taking it as the unproblematic starting point
of his analyses, he wanted to question it. He also explicitly denied in
AK that he had excluded the question of the subject. He writes that his
aim instead is to ‘define the positions and functions that the subject
could occupy in the diversity of language’ (AK, 200). The archaeological approach does not mean the denial of the subject’s initiatives:
‘These positivities are not so much limitations imposed on the initiative
of subjects as the field in which that initiative is articulated . . . I have
not denied – far from it – the possibility of changing discourse: I have
deprived the sovereignty of the subject of the exclusive and instantaneous right to it’ (AK, 209).13 Although the subjects of scientific discourse are regulated and even partly constituted by the rules immanent
to the discourse itself, they are not completely determined by them: the
subject is not without the initiative or the capacity to effect changes in
the discourse. Archaeology is simply not about the subject’s abilities to
cause changes, but rather focuses on the more fundamental discursive
structures that make different initiatives possible or impossible.
Foucault was thus trying to develop a method that allows us to study
discourse as a relatively autonomous field of regularities and transformations without positing the subject as the cause and principle of these
unities, regularities and transformations. Archaeology is a method for
analyzing and accounting for the constitution of meanings that are not
dependent on individual speakers. It is an alternative to subject-centred
approaches to the history of thought. Although statements are uttered
by individual speakers, in making a statement the speaker takes up a
position that has already been defined – quite apart from his mental
activity – by the rules of the relevant discursive formation. Foucault thus
13 ‘Il s’agit moins des bornes posées à l’initiative des sujets que du champ où elle
s’articule . . . Je n’ai pas nié, loin de là, la possibilité de changer le discours: j’en ai
retiré le droit exclusif et instantané à la souveraineté du sujet.’ (AS, 272)
th e a n o n ymity o f la n g ua ge
claims that every statement has a subject, but that this subject is not a
‘speaking consciousness’ but rather ‘a position that may be filled in certain conditions by various individuals’ (AK, 115). The subject position
is established by the rules of the discursive formation.
By focusing on the systems of the actual statements that define the
space in which speaking subjects operate, archaeology seeks to question the fundamental role of the human subject in the constitution of
knowledge. Statements are studied historically in their own right, not
as means of understanding the thoughts of the dead. The source of
scientific discourse is an anonymous field of discursive practices, not
the meaning-giving subject. Foucault writes ironically in AK that it is
unpleasant ‘to reveal the limitations and necessities of a practice where
one is used to seeing, in all its pure transparency, the expression of
genius and freedom’ (AK, 210).
Even though archaeology describes the discursive conditions that
limit and make different subject positions possible, it is not until Foucault’s genealogy introduces the notion of productive power that his
idea of a constituted subject gets its full force. Archaeology describes
the possibility and availability of various subject positions. It also concerns the ways in which scientific discourses produce subjects as their
object of study, for example, the speaking subject in general grammar,
philology and linguistics, and the labouring subject in the analysis of
wealth or economics. Nevertheless, these studies only bring out partial
analyses of subjectivity. As Foucault himself admitted later, his archaeology presented only one of the axes of the constitution of the subject
(SP, 208).
The freedom of language
By establishing a new perspective on the question of the subject, archaeology also maps out new ideas of freedom that are not tied to the idea
of a founding subject, its nature, initiatives or abilities. The intention is
to draw a line marking the discursive limits of thought and experience
of an age, and implicitly therefore also to question what falls outside of
these limits.
Foucault not only gives language a regulative role in the mode of scientific discourse, but also allows that it demarcates a domain of freedom
in the mode of literature. While scientific discourses form an ontological order of things that is implicit in their theories and practices, language in the form of literature is capable of forming alternative, unscientific and irrational ontological realms: different experiences of order
fo u c au lt o n freedom
on the basis of which different perceptual and practical grids become
possible, and hence new ways of seeing and experiencing emerge.
Foucault generally emphasizes the necessary structures of knowledge
and opposes the humanist aspirations of looking for the freedom of
man, but there is an anti-humanist understanding of freedom as an
opening of new possibilities of thought implicitly safeguarded in archaeology. He attempts to show not only how the discursive limits of scientific knowledge, representation and experience are constituted, but also
what escapes them.
According to John Rajchman (1985, 11–12), Foucault’s early
thought conformed to the spirit of the influential movement in France
in the 1960s that saw a revolution emerging from avant-garde writing.14 Foucault later came to realize that the central questions of our
era are not, after all, about commentary, language and avant-garde art,
but rather about the politics of documentation, secrecy and individuality, which have made subjectivity our basic problem. Rajchman clearly
points to an important shift of emphasis in Foucault’s thought, which is
often identified as his break from archaeology and move towards more
politically motivated genealogy. I will argue, however, that the question
of the subject was already prominent in his early thought, and, furthermore, that it was interwoven with the question of language.
Foucault’s book Death and the Labyrinth: The World of Raymond Roussel
is among his works that have been least commented on.15 In writing
a book about Raymond Roussel – an eccentric writer admired by the
Surrealists – he raises questions about the writing subject as the origin
of meaning. In writing some of his books, Roussel used a method he
described in his posthumous work Comment j’ai ecrit certain de mes livres.
He would take a phrase containing two words, each of which had a
double meaning, and use the least likely meaning as the basis of a story.
He would transform a common phrase, a book title, or a line of poetry
into a series of words with similar sounds. By selecting a sentence at
random, he would draw images from it by distorting it. John Ashbery
(1986, xxiii–xxv) describes the method by referring to Roussel’s work
in which the hazards of language result in strange but beautiful rhyming
events, and banal mechanisms create convincing juxtapositions.
14 Rajchman (1985, 111) identifies Julia Kristeva, Roland Barthes, Jacques Lacan and
Jacques Derrida as central figures in this movement.
15 Foucault said (PS, 185) in an interview that he was happy that no one had paid much
attention to this book, because it had remained his ‘secret affair’. He also claimed that
it did not have a place in the sequence of his books.
th e a n o n ymity o f la n g ua ge
Foucault was clearly fascinated by Roussel’s experiments with
language. He said in an interview that Roussel’s work was extremely
interesting because it was not only a linguistic experiment, but also an
experiment with the nature of language (PS, 175). Roussel’s experiments were mechanical processes that blindly followed certain rules
and principles, and yet they were capable of creating new and beautiful meanings. It is clear in relation to Foucault’s own interests that the
machine-like production of surreal beauty illustrated for him the idea
that language produces meanings independently of the initiatives of the
subject. While the Surrealists’ interest in Roussel was connected to their
idea of automatic writing and the utilization of the unconscious mind,
Foucault’s motivation was somewhat different: he was interested in the
‘unconscious’ of language, not of the writing subject. He saw Roussel’s
work as celebrating the role of chance in the interstices of language.
Roussel’s process of writing purified discourse of ‘all the false coincidences of inspiration’, in order to confront ‘the unbearable evidence
that language comes to us from the depth of a perfectly clear night and
is impossible to master’ (DL, 39).
It seems that chance triumphs on the surface of the narrative in those
forms which rise naturally out of the depths of the impossible; in the
singing mites, the truncated man who is a one-man band, the rooster
that writes his name by spitting blood, Fogar’s jellyfish, the gluttonous
parasols. But these monstrosities without family or species are necessary
associations; they obey mathematically the laws governing homonyms
and the most exacting principles of order; they are inevitable . . . At
the start no instrument or stratagem can predict their outcome. Then
the marvelous mechanism takes over and transforms them, doubles their
improbability by the game of homonyms, traces a ‘natural’ link between
them, and delivers them at last with meticulous care. The reader thinks
he recognizes the wayward wanderings of the imagination where in fact
there is only random language, methodologically treated.
(DL, 38)16
16 ‘En apparence le hasard triomphe à la surface du récit, dans ces figures qui surgissent naturellement du fond de leur impossibilité – dans les cirons chanteurs, dans
l’homme-tronc qui est un homme-orchestre, dans le coq qui écrit son nom en crachant
du sang, dans les méduses de Fogar, ombrelles gloutonnes. Mais ces monstruosités sans
espèces ni familles sont des rencontres obligées, elles obéissent, mathématiquement, à
la loi des synonymes et au principe de la plus juste économie; elles sont inévitables . . .
Au départ, il y a ces lots, dont aucun instrument, aucune ruse ne prévoit la sortie; puis le
merveilleux mécanisme s’en empare, les transforme, double leur improbabilité par le jeu
des synonymes, trace entre eux un chemin “naturel”, et les livre enfin dans une nécessité
méticuleuse. Le lecteur pense reconnaı̂tre les errements sans chemins de l’imagination
là où il n’y a que les hasards de langage traités méthodiquement.’ (RR, 52–3)
fo u c au lt o n freedom
What is important in connection with a different understanding of freedom is that Foucault read Roussel’s work as an effort to capture what
lies outside the discursive order of things by means of language. As he
noted, Roussel ‘doesn’t want to duplicate the reality of another world,
but, in the spontaneous duality of language, he wants to discover an
unexpected space, and to cover it with things never said before’ (DL,
16). The task of avant-garde literature was thus not so much to create an alternative ontological order – another world – but to show the
instability of the order of things that we take for granted.
This idea of literary writing as constitutive of alternative forms of reality also underlies Foucault’s interest in Surrealism. In an interview he
gave in 1966, he discusses the importance of André Breton to contemporary thought. He was asked what Breton and Surrealism represented
to a philosopher of 1966 who concerned himself with language and
knowledge (STW, 171). Foucault replied that, for Breton, writing had
the power to change the world. He reiterated the idea present in OT
that language and writing used to be understood as transparent instruments in which the world was reflected. However, Breton was one of the
figures with whom the status of writing changed. As Foucault suggested,
‘Perhaps there is a writing so radical and so sovereign that it manages
to face up to the world, to counterbalance it, to offset it, even to utterly
destroy it and scintillate outside it’ (STW, 173). He argued that Breton
had contributed to the changing status of writing in two ways. Firstly,
he had remoralized it by demoralizing it completely: the ethic of writing no longer came from what one had to say, from the ideas that one
expressed, but it emerged from the very act of writing. ‘In that raw and
naked act, the writer’s freedom is fully committed at the same time as
the counter-universe of words takes form’ (STW, 173).
While freedom as an attribute of the writing subject loses strength,
there is a freedom in language itself, in the creation of unexpected
worlds. This is the second characteristic of literary language: writing
solidifies and asserts itself apart from everything that might be said
through it. Writing and art constitute reality, they create objects. Underlying all the activities of the Surrealists, whether writing, painting or wandering around the city, is the aim to constitute new, previously unimaginable objects, to see things differently and expand the domain of what
can be thought and imagined.
For Foucault, literature thus forms ‘a sort of counter-discourse’ (OT,
300) freed from the principles of order regulating scientific as well
as everyday discourses. Its aim is precisely to transgress the limits of
th e a n o n ymity o f la n g ua ge
discourse, and thus to make them visible and contestable, to discover
a ‘madness’ in language as ‘that formless, mute, unsignifying region
where language can find its freedom’ (OT, 383). Literary writing can
express this freedom of language and also shatter our familiar order
of thought. This also means, significantly, that it can make visible the
limits of the discursive order. John Rajchman (1985, 24) takes this idea
as far as to claim that the picture of an avant-garde writer as the hero
of our age is built into the very structure of OT, and that this determines its plot. Literature, represented by figures such as Cervantes and
Sade, is never placed directly within an episteme, but is rather in a
position from which it can articulate its limits. Don Quixote signals
the end of the Renaissance episteme, and Sade, by taking classification
to its extreme limits, shows the impossibility of the classical modes of
knowledge. According to Rajchman, arts as described in OT are thus
meta-epistemic, allegories of the deep arrangements that make knowledge possible. Furthermore, Foucault himself adopted an avant-garde
or vanguard position in relation to our modernity, and claimed to have
announced a whole new form of thought.
By being able to demonstrate the limits of a discursive order, literary
writing is also able to reveal important limits of subjectivity. Because
the discursive order is constitutive of the limits of subjectivity, counterdiscourse in the form of avant-garde writing, for example, can question
these limits. Foucault aimed to show how modes of subjectivity are constituted in scientific discourse, and also how these limits are transcended
in avant-garde writing and art.17
Foucault takes up the question of the relationship between the subject and language in an essay entitled What is an Author? His starting
point is the claim that writing has freed itself from the dimension of
expression and refers only to itself. He does not simply accept this claim,
but seeks to problematize it further. It is not enough to repeat the empty
affirmation that the author has disappeared. Instead we must locate
the space left empty by this disappearance: follow the distribution of
gaps and breaches, and watch for the opening that the disappearance
17 I will argue in chapter 7 that Foucault’s understanding of art as being capable of revealing
and transgressing the limits of subjectivity was explicitly brought out in his late work on
ethics. He saw ‘the arts of existence’ as a possibility for challenging normalizing power.
Language and writing become tools for recreating the self – ways of opening new forms of
experience, modes of thinking and living. According to Foucault, one writes to become
someone other than who one is: language is a possibility for modifying one’s way of being
through the act of writing. See e.g. UP, 8.
fo u c au lt o n freedom
uncovers (WA, 105). Foucault asks what happens in the openings of
the discourse into which the writing subject has disappeared.
Foucault suggests that analyses of discourse should proceed through
the notion of an author function. The author’s name can no longer be
thought of as simply referring to the actual person writing the books
on the cover of which it appears, but it has other characteristics that
can be analyzed and explicated. It performs certain unique roles with
regard to narrative fiction. Analyses proceeding through the concept
of author function can provide new methods for literary study, a new
approach to a typology of discourse as well as a historical analysis of it.
Foucault also attaches a third, distinctly philosophical importance to
this type of study: the re-examining of the subject.
I realize that in undertaking the internal and architectonic analysis of a
work (be it a literary text, a philosophical system, or scientific work), in
setting aside biographical and psychological references, one has already
called back into question the absolute character and founding role of the
subject. Still perhaps one must return to this question, not in order to
re-establish the theme of an originating subject, but to grasp the subject’s
points of insertion, modes of functioning and systems of dependencies.
(WA, 117–18)18
Foucault’s approach, far from eliminating the question of the subject,
in fact poses this question explicitly, but from a new angle. It is no longer
a case of analyzing the subject as the originator of discourse, but rather
one of analyzing it as a variable and complex function. The author
becomes one example of a discursively constructed subject position
that has come into being with the development of a certain kind of
narrative discourse, and which holds a specific function and position
in relation to it. Therefore the questions that Foucault suggests we ask
about the subject concern how, under what conditions, and in what
forms can something like a subject appear in the order of discourse.
What place can it occupy in each type of discourse, what function can
it assume, and by obeying what rules (WA, 118)?
18 ‘Je sais bien qu’en entreprenant l’analyse interne et architectonique d’une œuvre (qu’il
s’agisse d’un texte littéraire, d’un système philosophique, ou d’une œuvre scientique),
en mettant entre parenthèses les références biographiques ou psychologiques, on a déjà
remis en question le caractère absolu, et le rôle fondateur du sujet. Mais il faudrait peutêtre revenir sur ce suspens, non point pour restaurer le thème d’un sujet originaire,
mais pour saisir les points d’insertion, les modes de fonctionnement et les dépendances
du sujet.’ (QA, 810.)
th e a n o n ymity o f la n g ua ge
Foucault does not claim that there are no subjects writing books
with their pens or on their computers. What he claims is that we cannot
understand what an author is by only studying the individual writing a
book. We need a new method. Apart from presenting a new method,
however, he also seems to suggest that his approach has an additional
advantage. He ends his essay with a curious, Utopian twist that brings us
back to the question of freedom. The author as a functional principle
not only organizes the work in a certain way, but also limits, excludes and
chooses. It is the means by which the free circulation, manipulation,
composition, decomposition and recomposition of fiction is impended
(WA, 119). The author for Foucault is thus both a contingent and a constraining figure.19 Although there will never be a completely free circulation of texts, the modes of constraint are historically changing, and
it is therefore possible that one day we might live in a culture in which
we are not limited by the figure of the author, but rather surrounded
by an anonymous murmur, an endless proliferation of meanings (WA,
Hence, by a curious twist, the methodological ‘disappearance’ of the
subject in Foucault’s thought does not signal the disappearance of freedom. It is, rather, the case that freedom is understood as the endless
proliferation of meanings, which undermines the stability of the historical a priori determining possible ways of seeing, understanding and
acting. Rather than thinking of the subject in terms of individuals, and
of freedom as something they have or do not have, he suggests that we
attempt to think of the subject as a discursive effect and freedom as
a non-subjective opening up of possibilities for multiple creative practices. This does not mean that he denies the possibility of understanding
subjects as individuals or agents, and freedom as a capacity that is tied
to their initiatives. His analysis simply does not operate on this level. It
charts new dimensions of freedom. He is trying to find freedom on a
level that orders and regulates subjective expressions and initiatives.
In the realm of scientific discourse, Foucault emphasized the rules
and formal conditions of thinking, and questioned the possibility of
saying something completely new. In the realm of literature this possibility is emphasized, however, because the ontological order of things,
the historical a priori, can be suspended, even thrown out. Language
solidifies the identity of things by repetition, it creates an ontological
order taken as unquestioned reality, but it can also act as ‘a thin blade
19 Cf. DL, 156.
fo u c au lt o n freedom
that slits the identity of things, showing them as hopelessly double and
self-divided even as they are repeated’ (RR, 23). There is a dimension of
language capable of undermining reality instead of only materializing
and solidifying it according to pre-existing rules.
Hence, as it is often argued that Foucault wanted to construct a
history and politics without human nature,20 it could equally well be
argued that he wanted to rethink freedom without human nature. In
the same way as archaeology questions the privileged role of the subject
in the constitution of knowledge without eradicating it as a question, it
also seeks to understand freedom in non-subjective terms without eradicating the notion. Freedom characterizes language rather than the
subject. The limits of freedom are the limits of the discursive order, and
they must not be conflated with the limits of the social order or of acceptance. Although Foucault showed interest in marginal subjectivities
throughout his work, he did not romanticize marginal lifestyles, nor did
he see them as exemplifying freedom. The limits that he attempted to
identify demarcated discursively constructed ontological realms. What
lies outside of them is not socially unacceptable, it is unintelligible in
existing modes of order.
The history of madness would be the history of the Other – of that
which, for a given culture, is at once interior and foreign, therefore to be
excluded . . . whereas the history of the order imposed on things would
be the history of the Same – of that which, for a given culture, is both
dispersed and related, therefore to be distinguished by kinds and to be
collected together into identities.
(OT, xxiv)21
Clare O’Farrell (1989, vii) points to this passage, and argues that Foucault’s whole work can be read as a history of limits, of that edge between
the systems societies impose upon order (the Same), and that which
is outside or beyond that order (the Other). His work of the 1960s
presents, in this sense, a consistent ontological view of a changing
boundary between the Same and the Other, apparent in the events
of history (O’Farrell 1989, 40, 90). In Madness and Civilization, the
confrontation between the Same and the Other was between reason
20 See Davidson 1997b, 15.
21 ‘L’histoire de la folie serait l’histoire de l’Autre – de ce qui, pour une culture, est à
la fois intérieur et étranger, donc à exclure . . . l’histoire de l’ordre des choses serait
l’histoire du Même, – de ce qui pour une culture est à la fois dispersé et apparenté, donc
à distinguer par des marques et à recueillir dans des identités.’ (MC, 15)
th e a n o n ymity o f la n g ua ge
and madness, while in The Birth of the Clinic the Other was represented
by death, and in The Order of Things by the being of language. O’Farrell
nevertheless claims that, during the 1970s, Foucault gradually constructed a vision of society and history in which the Same and the
Other were totally coextensive and indeed interchangeable, inextricably bound together in their movement. Notions of power and politics
came to occupy an important place in his thought at the expense of
the Other. According to O’Farrell, the problem of the limit did not
reappear until the 1980s, when the notion of the subject took centre
stage and the Same and the Other become distinct and free terms again
(41, 91, 115).
I will argue, against O’Farrell, that the limit, and freedom as its
transgression, never disappears from Foucault’s thought. I now turn
my attention to Foucault’s thought in the 1970s, and my focus is on
the question of the body. Although the realm of freedom opened up by
literary writing – or in O’Farrell’s conceptual terminology, the Other –
is harder to locate in the tightly knit networks of power and knowledge,
it is nevertheless present. The body will come to represent resistance
to power, and it will open up new ways of understanding freedom. In
the works following The Order of Things and The Archaeology of Knowledge,
Foucault further developed the idea that subjects are formed in discursive practices, but he also turned to study other axes of the constitution
of the subject in non-discursive practices as well as in the technologies of
the self. This means that the realms of freedom are also expanded. Freedom emerges not only through the practice of writing as constitutive of
an ontological otherness to the discursive order, but also in other kinds
of practice. At the same time, writing becomes entangled with subjectivity in even more integral ways: it becomes inseparable from sexuality
and life: ‘The private life of an individual, his sexual preferences, and
his work are interrelated not because his work translates his sexual life,
but because the work includes the whole life as well as the text. The
work is more than the work: the subject who is writing is part of the
work’ (PS, 184).
So far, all that has given color to existence still lacks a history.
Where could you find a history of love, of avarice, of envy, of
conscience, of pious respect for tradition, or of cruelty? Even
a comparative history of law or at least of punishment is so
far lacking completely. Has anyone made a study of different
ways of dividing up the day or of the consequences of a regular
schedule of work, festivals and rest?
(Nietzsche 1887/1974, 81–2)
In an interview conducted in 1983, Foucault situates his thought, as
well as the trends of thought influenced by structuralism, linguistic
theory and psychoanalysis in the 1960s as explicit efforts to rethink
the phenomenological subject. The phenomenological theory of the
subject was his point of departure, and his thought has been an effort
to make distance from it (SPS, 442).
I would say that everything that took place in the sixties arose from a dissatisfaction with the phenomenological theory of the subject, and involved
different escapades, subterfuges, breakthroughs, according to whether
we use a negative or positive term, in the direction of linguistics, psychoanalysis or Nietzsche.
(SPS, 438)1
The linguistic turn of structuralism and post-structuralism in French
thought was thus, according to Foucault, a reaction to phenomenology.
1 ‘Donc, je dirais que tout ce qui sest passé autour des années soixante venait bien de cette
insatisfaction devant la théorie phénoménologique du sujet, avec différentes échappées,
différentes échappatoires, différentes percées, selon qu’on prend un terme négatif ou
positif, vers la linguistique, vers la psychoanalyse, vers Nietzsche’ (SEPS, 437).
fo u c au lt o n freedom
The importance of the structuralist method and also of Nietzsche’s
thought to Foucault lay in the fact that they provided ways of rethinking
the phenomenological subject and the historicity of reason as forms
of rationality. In his lecture ‘Subjectivity and Truth’, he also describes
his choice of direction in terms of rethinking the phenomenological
I have tried to get out from the philosophy of the subject through a
genealogy of this subject, by studying the constitution of the subject across
history which has led us up to the modern concept of the self. This has
not always been an easy task, since most historians prefer a history of
social processes (where society plays the role of the subject) and most
philosophers prefer a subject without history.
(ST, 176)2
Reading Foucault’s genealogy as an effort to rethink the phenomenological subject opens up a new angle on his relationship with phenomenology. I have argued in the previous section that Foucault’s
archaeology and phenomenology shared a common mode of questioning in inquiring into the conditions of possibility of scientific knowledge. Here I will cast their relationship in a slightly different mould
through the question of the subject: my aim in this chapter is to explicate
Foucault’s genealogical conception of the subject as a critical reaction
to the phenomenological subject. I will ask what Foucault meant by his
controversial claims that power makes individuals subjects and that the
subject is an effect of power. I will argue that we must not understand
the constitution of the subject as a causal process.3 Power relations are
immanent to the social reality and have empirical causal effects, but
they are also paradoxically ‘transcendental’, in the sense that they are a
condition of possibility for the constitution of the subject. Pheng Cheah
(1996, 126), for example, argues that power is quasi-transcendental for
Foucault, because it is both the immanent causal origin of empiricality
and physicality and a condition of possibility for grasping social reality, a
grid of its intelligibility, which cannot itself be accessible to cognitive or
practical-intentional mastery and control. While Foucault’s genealogy
of the subject is thus a critical reaction to the phenomenological subject,
I argue that it nonetheless presents a modification of transcendental
analysis: the network of power/knowledge is understood primarily to
provide the condition of possibility for the subject, not the material
2 Originally delivered in English.
3 Cf. Flynn 1989, 189.
a gen ealo g y o f t he sub j e c t
cause. Historical and transcendental constitution are again inseparable in the sense that the conditions of possibility of the subject are to
be found in the historical practices and discourses structured by power
By claiming that power relations are productive of forms of the subject, Foucault does thus not simply suggest that individuals are produced
as subjects just as cars are produced from various materials in a factory.
Rather, we must understand the subject to be intrinsically entangled
with power and knowledge. Power/knowledge network constitutes the
subject in the sense of forming the grid of intelligibility for its actions,
intentions, desires and motivations.
My discussion ends by considering the problems involved in this
‘transcendental’ reading. I will question the apparent circularity of
Foucault’s understanding of the constitution of the subject, and will
suggest ways of combating this problem.
The constitution of the subject
Foucault’s genealogy is often characterized by the fact that it introduces into his thought non-discursive practices and the relations of
power.4 This kind of characterization is, however, misleading in many
ways. Archaeology already treated discourse as a set of practices, making
the opposition of discursive and non-discursive difficult to uphold. It
also explicitly postulated a close tie between the discursive and the nondiscursive. Archaeology provided a distinctive approach to the relations
between discourses and non-discursive domains such as ‘institutions,
political events, economic practices and processes’ (AK, 162). It aimed
to determine how the rules that govern a discursive formation ‘may be
linked to non-discursive systems’ (AK, 162). Foucault explicitly stated in
The Archaeology of Knowledge that discourse poses the question of power,
because it is by nature an object of political struggle (AK, 120). He
4 Foucault introduced his conception of genealogy in an article Nietzsche, Genealogy, History (1971/1984), in which he presents an interpretation of Nietzsche’s genealogy. The
tone of the article is polemical, and Foucault presents his own ideas alongside those of
Nietzsche without clear distinctions. Although his thought was strongly influenced by
Nietzsche, there were also other major influences on it. His genealogy is by no means a
faithful reapplication of Nietzsche’s method. Foucault does not introduce his own conception of genealogy as a coherent method in any of his other writings either. His idea of
genealogy has therefore to be built up from different books, articles and interviews, which
means that it is hard to give a uniform presentation of it. For a systematic explication of
‘Foucaultian’ genealogy, see Kusch 1991.
fo u c au lt o n freedom
did not, however, elaborate on how the close tie between the discursive
and the non-discursive should be understood, except that he clearly
denied both that they symbolically reflect each other, and that nondiscursive changes should be studied in terms of a causality communicated through the consciousness of the speaking subject (e.g., AK, 163).
By suspending causal analysis of discursive changes, he denied that he
was trying to give discourse ‘the status of pure ideality and total historical
independence’, but was rather doing it ‘in order to discover the domain
of existence and functioning of a discursive practice’ (AK, 164–5).
Foucault’s focus on discourse was thus a methodological choice, not the
ontological choice of discursive idealism. He did not hold that discursive
formations were completely autonomous, nor did he later give up this
position – despite its problems – in favour of a completely new position
emphasizing non-discursive practices. Instead, he further developed
his central philosophical claim that scientific objects are constituted in
history through discursive practices. Foucault’s genealogy in the 1970s
also looked more comprehensively at the tie between discursive and
non-discursive practices through the notion of power/knowledge.
Beatrice Han (1998/2002) also argues that a shift of attention from
discursive to non-discursive practices is not sufficient to mark the difference between archaeology and genealogy.5 According to her, what
is important in the introduction of genealogy is the claim that it is
impossible to understand the conditions of possibility of scientific discourse without taking into account the development of new forms of
power. With genealogy Foucault is able to analyze the way non-epistemic
demands not only control the effective predication of scientific truths,
but also shape the overall conditions of possibility of scientific discourse
(Han 1998/2002, 104, 109–10). The conditions of possibility under
investigation are thus no longer a set of purely epistemic rules, but
a power/knowledge network consisting of all kinds of practices: institutional, architectonic, juridical and medical. The episteme, or what
Foucault now refers to as the ‘regime of discourses’, only constitutes
the specifically discursive element of a more general regime, the dispositif or apparatus (Han 1998/2002, 137–8). Foucault clarifies their
relationship by writing that ‘the episteme is a specifically discursive
apparatus, whereas the apparatus in its general form is both discursive
5 Gary Gutting (1989, 271) also argues that genealogy does not replace or even seriously
revise Foucault’s archaeological method, but rather combines it with the complementary
technique of causal analysis, which establishes an essential symbiotic relation between
power and knowledge.
a gen ealo g y o f t he sub j e c t
and non-discursive, its elements being much more heterogeneous’ (CF,
It is not a question of power relations presenting a new level or a simple addition to previous analyses of discursive practices, but rather that
the idea of the fundamental entanglement of power and knowledge,
power/knowledge, becomes central: knowledge and power are intrinsically tied together, they condition each other and cannot be understood independently of each other. Foucault defines the new notion of
apparatus by writing:
What I am trying to pick out with this term is, firstly, a wholly heterogeneous ensemble consisting of discourses, institutions, architectural
forms, regulatory decisions, laws, administrative measures, scientific statements, philosophical, moral and philanthropic propositions – in short,
the said as much as the unsaid. Such are the elements of the apparatus. The apparatus itself is the system of relations that can be established
between these elements.
(CF, 194)6
The apparatus now becomes the ‘transcendental’ in Foucault’s thought,
in the sense of providing the conditions of possibility for the emergence
of scientific objects. It also, crucially, becomes the condition of possibility for the subject. While archaeology facilitated partial analysis of
the subject by approaching it through the discursive subject positions
made possible by the episteme and produced through a set of formative rules, the introduction of genealogy as Foucault’s method in the
1970s provided a way to present a more comprehensive account of the
constitution of the subject. Genealogy analyzes the constitutive effects
of non-discursive practices as well as of the scientific truths dependent
on them.
The subjection of the body. Central to Foucault’s genealogy is the idea
of ‘subjection’ (assujettissement), which refers to the process of constituting subjects. The body is an important instrument in this process. In his first major genealogical book, Discipline and Punish Foucault
6 ‘Ce que j’essaie de repérer sous ce nom, c’est, premièrement, un emsemble résolument
hétérogène, comportant des discours, des institutions, des aménagements architecturaux,
des décisions réglementaires, des lois, des mesures administratives, des énoncés scientifiques, des propositions philosophiques, morales, philanthropiques, bref: du dit, aussi
bien que du non-dit, voilà les éléments du dispositif. Le dispositif lui-même, cest le réseau
qu’on peut établir entre ces éléments.’ (JMF, 299)
fo u c au lt o n freedom
analyzes the ways disciplinary technologies subject prisoners by manipulating and materially inscribing their bodies. Their bodies are separated
from others in practices of classification and examination, but also concretely and spatially. They are manipulated through exercise regimes,
diet and strict time schedules. These processes of subjection are essentially objectifying: through processes of classification and examination
the individual is given a social and a personal identity: he/she is objectivized as mad, criminal or sick, for example.
Disciplinary power thus constitutes criminal subjects through concrete bodily manipulation and discursive objectification. These two
dimensions strengthen each other. On the one hand, material subjection made theoretical objectification possible, resulting in the birth of
human sciences such as criminology, criminal psychiatry and pedagogy.
The development of the corresponding sciences, on the other hand,
helped the development and rationalization of disciplinary technologies. The two dimensions furthermore link together effectively through
normalization. Scientific discourses produce truths that function as the
norm. Norms further the subjection by reducing individuality to a common measure. They also make possible the subjection of individuals
through the internalization of norms. In Discipline and Punish, Foucault
analyzes the disciplinary strategy that utilizes the idea of an inner core or
essence in the subjection of criminals. Where prisoners are concerned,
disciplinary power does not aim at repressing their interests or desires,
but rather at constructing them as normal. This is done on and through
the bodies of criminals who subject themselves to power to the extent
that its aims become their own inner meaning of normal. In a later text,
in a famous passage, Foucault writes: ‘There are two meanings of the
word subject: subject to someone else by control and dependence, and
subject tied to his own identity by a conscience or self-knowledge. Both
meanings suggest a form of power which subjugates and makes subject
to’ (SP, 212).
Hence, a novel aspect of modern disciplinary power is that it is not
external to the bodies that it subjects. Although the body has also in
the past been intimately tied to power and social order, Foucault claims
that disciplinary power is essentially a modern phenomenon. It differs from earlier forms of bodily manipulation, which were violent and
often performative – public tortures, slavery and hanging. Disciplinary
power does not subject the body to external violence, it is not external
or spectacular. It focuses on details, on single movements, on their timing and rapidity. It organizes bodies in space and schedules their every
a gen ealo g y o f t he sub j e c t
action for maximum effect. This is done in factories, schools, hospitals
and prisons through fixed and minutely detailed rules, constant surveillance and frequent examinations and check-ups. Bodies are classified
according to their best possible performance, their size, age and sex.
Unlike older forms of bodily coercion, disciplinary power thus does
not only causally mutilate and shape the criminal body. The criminal
literally incorporates the objectives of power, which become the norm
for his own aims and behaviour. Foucault formulates this poetically by
writing that the ‘soul’ is an effect of the material reality of the subjection of the body (DP, 30). It is in this context that he presents his most
extreme formulations of subjects as effects of power.
The man described for us, whom we are invited to free, is already in
himself the effect of a subjection much more profound than himself. ‘A
soul’ inhabits him and brings him to existence, which is itself a factor in
the mastery that power exercises over the body. The soul is the effect and
instrument of a political anatomy; the soul is the prison of the body.
(DP, 30)7
In his ‘Two Lectures’, which dates from around the same period as
Discipline and Punish, Foucault discusses the relationship between body,
power and the subject. These lectures represent one of his early efforts
to conceptualize power in new terms.8 He wanted to find an alternative to theorizing power in terms of sovereignty as well as in terms of
rights, repression or economy. He turned to the ‘Nietzschean’ alternative, according to which the basis of the relationships of power lie
in the ‘hostile engagement of forces’ (TL, 91). Foucault also wanted
to construct a model that would better account for the new form of
power, disciplinary power. This type of power functions through material operators; it presupposes a ‘tightly knit grid of material coercions
rather than the physical existence of a sovereign’ (TL, 104). Foucault
suggests that we should not look for the centre of power, but rather
study it at its extremities, the points at which it ‘becomes capillary’
7 ‘L’homme dont on nous parle et qu’on invite à libérer est déjà en lui-même l’effet d’un
assujettissement bien plus profond que lui. Une “âme” l’habite et le porte à l’existence,
qui est elle-même une pièce dans la maı̂trise que le pouvoir exerce sur le corps. L’âme,
effet et instrument d’une anatomie politique; l’âme, prison du corps.’ (SEP, 38)
8 Foucault presents his account of power as a series of propositions in HS and elucidates
it further in numerous interviews and essays. Foucault’s account of power should not be
understood as a universally applicable theory of power. His goal in HS is rather to find a
method that will help us to understand ‘a certain form of knowledge regarding sex’ (HS,
92). See also e.g. Cousins and Hussain 1984, 2 and Gutting 1994, 19–20.
fo u c au lt o n freedom
(TL, 96). Neither should we look for the individuals who dominate, or
question their motives, but rather study ‘the myriad of bodies which are
constituted as peripheral subjects as results of the effects of power’ (TL,
98). Rather than studying how subjects exercise power, Foucault turns
the question around and asks how the subject emerges as an effect of
power. Now he clearly formulates his project in terms of constitutional
[R]ather than ask ourselves how the sovereign appears to us in his lofty
isolation, we should try to discover how it is that subjects are gradually,
progressively, really and materially constituted through a multiplicity of
organisms, forces, energies, materials, desires, thoughts, etc. We should
try to grasp subjection in its material instance as a constitution of subjects.
(TL, 97)10
The emphasis on material bodies as the locus of a struggle between
power relations is also strong in Foucault’s introduction of the concept
of biopower at the end of The History of Sexuality, volume i. Biopower is
a particularly modern form of power, which links together power, sexuality and the body. Its function is to ‘invest life through and through’
by the administration of bodies and the calculated management of life
(HS, 139). This ‘power over life’ evolved in two basic forms, which were
linked together. The first was the disciplinary power that centred on
the individual body as a machine. The second focused on the species
body, the population. It effected the regulatory control of propagation,
births and mortality, the level of health and life expectancy, for example
(HS, 139–40). Foucault’s aim again is to isolate the functions that the
body has as a central component in the power relations of society.
The forms of the subject. If we understand subjection to mean, on the
one hand, that material bodies are causally formed into subjects in
processes of subordination with different manipulative technologies
of power, and on the other hand that individual bodies internalize
the objectives of power to the extent that they become the meaning
of their subjectivity, Foucault’s understanding of the subject seems to
9 See also e.g. EPF, TP.
10 ‘plutôt que de se demander comment le souverain apparaı̂t en haut, chercher à savoir
comment sont progressivement, réellement, matériellement constitués les sujets à partir
de la multiplicité des corps, des forces, des énergies, des matières, des désirs, des pensées;
saisir l’instance matérielle de l’assujettissement en tant que constitution des sujets’
(CJ, 179).
a gen ealo g y o f t he sub j e c t
be some kind of combination of behaviourist and social constructivist
accounts of the subject. In fact, these kinds of labels have invariably
been stuck on him. It has been argued that the core idea of Foucault’s
genealogy is a reduction of subjects to bodies: he denies the existence
of subjectivity, now understood as individual consciousness and transformative agency, and redefines it as disciplined, corporeal materiality
causally formed and shaped. Power causally shapes material bodies into
subjects through subjection, and subjectivity is an illusory effect of the
subordination of bodies. This kind of conception often underlies the
feminist criticisms of Foucault, according to which his understanding of
the subject in terms of passive bodies is problematic for feminist theory
because it cannot account for women’s experiences or agency.11 Power
is not seen as the constitutional moment and precondition of any kind
of subject – autonomous or determined – but is rather understood as
the constraining grip that squeezes bodies into subjects.
In the context of analyzing the functioning of disciplinary power,
Foucault conceives of subjects as bodies. Furthermore, subjection always
has as its correlate the process of objectification, the body of the criminal
as an object-body, the body as useful and docile. This does not, however, mean that he adopts a mechanical view of the body and simply
equates subjects with object-bodies.12 In his later work he insisted that
the subject is not a substance: ‘It is a form and this form is not above all
or always identical to itself’ (EPF, 10). The subject must thus be understood as the intelligible form of a human being, not as an object-body.
Human beings become socially recognized as well as recognize themselves through a culturally and historically constituted form, and there
is not just one universal form, but rather forms of this form, forms of
the subject. ‘It is precisely the historical constitution of these different
forms of the subject relating to the games of truth that interest me’
(EPF, 10).
It is thus not merely that power causally turns bodies into subjects
according to its requirements: the idea of subjection is more complex
and sophisticated. The disciplinary manipulation of bodies does not
11 See e.g. McNay 1991, Soper 1993, Deveaux 1996.
12 According to Foucault, the body is not understood as a passive mechanism. In the next
chapter I will argue that it is exactly in the body that a margin of excess and the possibility
for subverting the normalizing aims of power take place. Foucault suggests ‘bodies and
pleasures’ as a possible form of resistance. Although power constitutes the mastery and
awareness of one’s body, it is the body that produces the resistance, ‘the counter-attacks’
against the same power (B/P, 56).
fo u c au lt o n freedom
produce subjects per se, but is only one part of a complex network of
power/knowledge, or ‘apparatus’, which forms the constitutive conditions of subjectivity. Bodily subjection partakes in the formation of a
discursive order, through the birth of human sciences, for example.
This discursive order then feeds back to the non-discursive practices
by creating material effects. Human sciences and their truths create
objects not only of science, but also of reality: desires, forms of experience, certain kinds of bodies. Scientific discourse and practice constitute not only conceptual objects and identities, but also the subjects
who make them materialize. Hence, the manipulation of bodies forms
only one dimension of a complex power/knowledge network, which
further constitutes subjects through the material effects generated by
scientific truths.
Ian Hacking (1984, 115, 122) argues that Foucault restricts his analysis to the human sciences exactly for the reason that it is only in the
human sciences that scientific truths have constitutive effects on the
subjects under study. In the natural sciences our invention of new identities and categories does not ‘really’ change the way the world works.
Even though we may create new phenomena that did not exist before
our scientific endeavours, what happens in our experiments is constrained by the world: if we do certain things, certain phenomena will
always appear. But in the social sciences we may generate kinds of people and kinds of action as we devise new classifications and categories.
Categories of people come into existence at the same time as kinds of
people come into being to fit those categories, and there is a two-way
interaction between these processes.13
Beatrice Han (1998/2002, 125) also explains this process in her
discussion of a course given by Foucault in 1974 at the College de
France, in which the effects of truth specific to medical discourse
on hysteric patients were analyzed. Medical discourse elaborated a
13 One of Hacking’s examples is the notion of psychic trauma. He discusses it along with the
three axes that Foucault distinguishes in his analyses of the constitution of the subject:
knowledge, power and ethics. First, there is the person as known about, as having a kind
of behaviour and sense of self that is produced by psychic trauma. There is a vast body
of knowledge in the growing field of tramatology. Second, in addition to the power of
courts and legislatures, there is the anonymous power that this concept has in people’s
lives. It organizes their ideas and emotions by creating a new sense of self. At the third,
moral level, the new sense of self as a victim of childhood trauma, for example, also
creates a new moral being. An understanding of who one is and why one is as one is has
implications for our understanding of a person’s responsibilities and duties (Hacking
2002, 18–20).
a gen ealo g y o f t he sub j e c t
theoretical object, following a process made possible by the hospital
structure and therefore by the techniques of subjection practised on
the patient. But by the same token, this discourse generated a real
object corresponding to its knowledge. The conceptual objectification
of the illness hysteria was therefore doubled by a second material form
of objectification, in which the hysterical woman reproduced in her
very person the phenomena. The objectification process was thus transposed from the theoretical level to that of reality, where in turn it produced concrete effects, since real forms of illness ended up corresponding to the newly constituted concept of the patient’s sickness.
Hence, Foucault’s idea of subjection does not simply mean that
power externally and causally forms human beings into subjects by disciplining them in their materiality. Bodily manipulation produces or
constitutes modern forms of the subject by being an integral component of biopower, which not only controls subjects but also constitutes
them through the normalizing effects of scientific truths. By urging
us to dispense with the subject, Foucault does not deny that there are
actual subjects exercising influence on their environments and on the
course of historical events. What his genealogical account of the subject
denies is the foundational status of the subject. This is done on several
different levels. On an ontological level, Foucault denies all metaphysical claims that posit it as a substance or endow it with an essence. The
subject is the intelligible form of a human being, not a static substance.
In contrast to phenomenology, Foucault further seeks to historisize this
form more radically. He does not only historically situate pre-existing
subjects, but also puts forward a stronger version of historical constitution. Historically variable practices not only condition what is possible to
know about the subject, but they also engender its experiences. Human
sciences and the disciplinary practices tied to them constitute not only
conceptual objects or identities, but also the subjects who materialize
This ontological claim about the subject is motivated and argued
for on both epistemological as well as ethical/political levels. Above
14 J. N. Mohanty (1997, 79) explains Foucault’s position concisely by suggesting that
Foucault reworks the problem of constitution by rejecting transcendental subjectivity
as constitutive of all objectivities, and by offering instead a theory of how the subject
itself is historically constituted. The principle of this constitution is not history in the
traditional sense, however, for the idea of one continuous historical process unfolding
itself at one time is itself a construction, behind which stands the transcendental subject.
Mohanty argues that the principle of constitution, ‘the transcendental’, is for Foucault,
power (83–4).
fo u c au lt o n freedom
I discussed how Foucault’s criticism of phenomenology in The Order
of Things aimed to show the instability of all efforts to ground knowledge on the subject. What is constituted in historical practices – namely
the subject and its experiences – cannot provide the basis of explanation for these practices. The subject cannot have a foundational role
in the constitution of meaning, because it is dependent on the nonsubjective mechanisms of signification and on historically changing
discursive structures. The ethical or political motivation arises from
the claim that the modern philosophies of the subject are intrinsically
entangled with the modern forms of subjection. I will return to this
idea in more detail below, when I focus on Foucault’s late work on
According to my constitutional reading of Foucault, genealogy thus
shares with phenomenology the transcendental mode of questioning as
opposed to a purely empirical study of the subject, but it does not share
the methodological starting point in the subject. Foucault rejects the
phenomenological subject that is the ground and source of knowledge,
meaning and value, and asks how the subject itself and its experiences
are historically constituted through discursive games of truth, practices
of power and technologies of the self. He does not deny subjectivity, but
neither does he construct a general theory of it. He studies the historical
constitution of different fields of experience – sexuality, delinquency,
madness. His thought is in a sense left circulating around the place left
empty by the transcendental subject.
The problem of circularity
Foucault’s modification of the transcendental mode of questioning
seems to open up a host of problems. We have to ask: how should
we understand the power/knowledge network as the transcendental?
Several commentators have pointed out that Foucault’s totalizing formulations at times give the impression that power/knowledge is a metaphysical, monolithic entity or a structural invariant. Charles Taylor
(1986, 88) compares Foucault’s conception of power to ‘a strange
kind of Schopenhauerian will, ungrounded in human action’. Beatrice
Han (1998/2002, 143) argues that some of Foucault’s formulations
reactivate the type of Hegelian schema so disliked by him, in which
power/knowledge take different historical forms. Han points out that
the idea that power/knowledge could be an essence definable in itself
would return to exactly the sort of metaphysics that genealogy sought
a gen ealo g y o f t he sub j e c t
to combat by giving primacy to perspective and interpretation, against
any essentialist ontology.
It should be emphasized, however, that Foucault repeatedly used
the notion of game in describing his understanding of the power/
knowledge network. He wrote, for example, that ‘relations of power,
they are played; it is these games of power (jeux de pouvoir) that one
must study in terms of tactics and strategy’ (PA, 542, trans. Davidson
1997a, 4). He explicitly related this study to the Anglo-American analysis of everyday language games, which does not give itself ‘the task of
considering the being of language or the deep structures of language;
it considers the everyday use that one makes of language in different
types of discourse’ (PA, 541, trans. Davidson 1997a, 3). Similarly, the
task of his genealogy would be to analyze what happens in everyday relations of power, rather than attempting to define the essence of power
or to affect it with a pejorative or laudatory qualification.15
Foucault also described his genealogy as a study of the ‘games
through which one sees certain forms of subjectivity, certain domains,
certain types of knowledge come into being’ (TJF, 4). It was ‘a radical
critique of the human subject by history’, which was accomplished by
showing ‘the historical construction of a subject through a discourse
understood as consisting of a set of strategies which are part of social
practices’ (TJF, 3–4). Thomas Flynn (1994, 30) offers an illuminating
explanation of Foucault’s conception of practice using Wittgenstein’s
concept of game: practices are shaped by a preconceptual, anonymous,
socially sanctioned body of rules that govern one’s manner of perceiving, judging, imagining and acting. This game analogy can be pushed
further to illuminate the way socially recognized subjects are constituted through practices that are specific to particular social and historical contexts. All games contain meanings that are not reducible to the
intentions or acts of individual players, but which make them possible.
In a similar way, the power/knowledge network contains meanings and
even intentions, but this intentionality is not in the intentions or acts of
individual subjects, but rather constitutes the individual as a subject.16
15 On Foucault’s relationship to Wittgenstein and Anglo-American analytic philosophy, see
Davidson 1997b.
16 This idea also makes understandable Foucault’s claim that power relations are nonsubjective but intentional. In the first volume of The History of Sexuality, in which he
presents his account of power as a series of propositions, he writes that power relations
are ‘imbued, through and through, with calculation: there is no power that is exercised
without a series of aims and objectives. But this does not mean that it results from
fo u c au lt o n freedom
The transcendental reading of Foucault’s account of the subject
must thus be qualified in the important respect that there are no a
priori rules or categories of intelligibility that make individual meaninggiving acts possible. Rather, the historically variable rules only exist
to the extent that the ‘players’ unconsciously follow them. The intentionality of power relations only arises out of the repeated actions of
the intentional subjects. Foucault insisted that power relations do not
pre-exist the individuals who are to be inserted in them as inert or consenting targets. Power only exists when it is exercised. Individuals are
the vehicles of power, not its points of application (TL, 98).
Even if we therefore pass off the extreme essentialism as a result
of inadequately contextualized formulations, and accept that power/
knowledge always refers to particular practices and not to metaphysical essences, we still seem to encounter the problem of circularity.
Foucault’s genealogy does not deny subjectivity. It is not a behaviourist
account, but neither is it a social constructivist account. While some proponents of social constructivism would also claim that the idea of power
being productive of subjectivity entails that the subject ‘internalizes’ its
aims and objectives, the important idea in Foucault’s account is that
the subject and the constitutive matrix are not understood as external
to each other, but are rather regarded as importantly continuous and
entangled in complex ways through the idea of a constitutive apparatus.
Power/knowledge, understood as a network of practices, is not ontologically distinct from the subject. On the contrary, the emergence of
any kind of subject is dependent on and ontologically tied to the existence of the power/knowledge network: it constitutes the subject by
providing the historically variable conditions for its emergence.
Foucault’s genealogies are thus not descriptions of how prepersonal
beings are turned into subjects through causal processes of social conditioning, but rather analyses or genealogical mappings of the conditions of possibility of certain practices and forms of the subject.
The circularity of Foucault’s understanding of the subject thus does
not take the simple form of presupposing some kind of unconditioned subject to give an account of how a socially conditioned subject
the choice or decision of an individual subject’ (HS, 95). This claim has often been
interpreted as bestowing power with some ‘occult’ qualities, or as simply nonsensical.
(See e.g. Waltzer 1986, 63; Taylor 1986, 85–6.) The claim is that power should be studied
as a non-subjective form of intentionality: it cannot be theorized as an object-like entity,
structure or state of affairs. It is an intentional, productive practice: a set of actions upon
other actions (SP, 220).
a gen ealo g y o f t he sub j e c t
emerges.17 The circularity is more complex. The constitutive matrix
of the subject, or its transcendental condition of possibility, is a
power/knowledge network consisting of discursive and non-discursive
practices constantly feeding back on each other. These practices, however, are only operational to the extent that the subjects are their agents.
The subject as an effect of power relations is also the subject exercising
power. If power/knowledge is identical with the apparatus, the overall
network of practices, then the constituted – subjects as agents of these
practices – and the constituting – the practices – become confused.
What is supposed to explain the subject in fact already presupposes its
existence insofar as there can be no practices without subjects.
One way to defend Foucault’s account against this criticism of circularity is to insist that his aim was not to produce a general theory of
the subject. What thus sets his account apart from behaviourist or constructivist theories, for example, is not just the transcendental mode
of questioning, but also the strictly partial and historically contingent
character. Even if psychiatric practices and the corresponding scientific
truths constitute mentally ill subjects, for example, this is only one of
myriad forms of the subject. There are different sets of practices, partly
overlapping but distinct nevertheless, which constitute the subject. This
constitution is thus not one unified and circular process, but consists
of regions of practices in which different subjects act in different ways
at different times.
Rather than constructing a unified theory of the subject, Foucault
focused on historically specific forms of it. The analysis must always be
partial; there can be no unified theory or unique way of representing the
subject. It is always multiple and dispersed. Foucault’s genealogies show
how certain practices in specific domains – for example criminology
in nineteenth-century France – function as historical conditions of a
particular form of the subject. The transcendental conditions are thus
again importantly historical: they are contingent, historically located
and variable. Foucault’s focus is not on the subject as such, but on the
conditions which make possible certain experiences and actions: the
forms of rationality and the relationships between our thought and our
17 Nick Crossley (1994, 191), for example, has argued that what Foucault needs is an
account of ‘a prepersonal being, a being which is less than a fully fledged subject but
more than an object’. This being would then be turned into a subject through disciplinary technologies. This kind of account would, however, be impossible in Foucault’s
genealogical framework, since Foucault very strongly rejects all a priori theories of a
fo u c au lt o n freedom
practices in western society. The descriptions of how human beings are
turned into subjects are always partial and limited to specific historical
Another direction to take Foucault’s understanding of the subject
in order to lead it out of the impasse of circularity is to make a
sharp distinction between power/knowledge as an overall network and
the power the subject exercises in acting upon the other’s actions.
Foucault kept modifying or ‘redefining’ his account of power. In one of
his last interviews he distinguished three different levels in his analyses
of power: relationships of power as strategic games between individuals;
states of domination; and governmental technologies (EPF, 19). We
can interpret this as meaning that relationships between individuals,
whether they are states of domination or strategic games, form a different level of analysis from governmental technologies understood as the
non-subjective techniques of power/knowledge. Hence, although the
power/knowledge network must consist of practices in which subjects
act and exercise power, there is an important shift between the levels of
the overall network and singular actions. The overall network, on the
one hand, holds constitutive power by virtue of its stable, shared and
sedimented meanings, while the singular actors, on the other hand,
retain the possibility of giving creative and singular meanings to their
actions, and hence, also offer the possibility of resistance.
Judith Butler elaborates this possibility in her book The Psychic Life of
Power. She takes up the question of circularity in Foucault’s account, but
writes, ‘Luckily, the story survives this impasse’ (Butler 1997, 12). She
argues that subjection has a double aspect: one is subjected to regulatory
power, but this subordination is also the condition of possibility of the
subject’s existence. Butler admits that this double aspect of subjection
appears to lead to a vicious circle: the agency of the subject appears to
be an effect of its subordination. This conception of the subject leads,
furthermore, to pessimistic political views according to which forms of
capital or symbolic domination are such that our acts are always already
‘domesticated’ in advance (Butler 1997, 17). Butler argues, however,
that power as a condition of possibility of the subject is not the same
as power considered as the subject’s agency – the power the subject
wields by virtue of being a subject in the social matrix. A significant
and potentially enabling reversal occurs when power shifts from its
status as a condition of agency to the subject’s ‘own’ agency (12). She
suggests that to understand this shift in subjection – how the subject is
formed in subordination while becoming the guarantor of resistance
a gen ealo g y o f t he sub j e c t
and opposition at the same time – requires thinking the Foucaultian
theory of power together with the psychoanalytic theory of the psyche
(3). We must ask how the formation of the subject by the regulatory
and productive effects of power involves the regulatory formation of
the psyche. The resulting conception of the subject would work as a
notion of political agency in post-liberatory times (18).
I will not go more deeply into Butler’s solution here, but I will return
to it in the next two chapters in connection with the question of resistance. I will show that the enabling shift in subjection that Butler locates
on the level of the psyche could also be thought to take place on the level
of the body. Hence, rather than complementing Foucault’s account of
the constitution of the subject with a theory of the psyche, I will suggest that we return to his own formulations about the resistance of the
body. Although Foucault’s genealogy combats some of the problems
connected with the phenomenological subject, I will show in the next
chapter that he never managed to completely leave it behind. There are
gaps or silences in his conception of the subject, particularly in connection with sexuality and embodiment, which could be read as implicitly
pointing in the direction of phenomenology.
Be a philosopher, be a mummy, represent monotono-theism
by a gravedigger-mimicry – And away, above all, with the body,
that pitiable idée fixe of the senses! Infected with every error of
logic there is, refuted, impossible even, notwithstanding it is
impudent enough to behave as if it actually existed!
(Nietzsche 1889/1990, 45)
Foucault’s understanding of the historical constitution of the body
through the network of power/knowledge has influenced feminist
theory profoundly: it has provided a way to theorize the body in its materiality while avoiding all essentialist formulations, and has given tools for
understanding the disciplinary production of the female body.1 While a
shared focus on bodies has opened up important connections between
feminist theory and Foucault’s thought, Foucault’s apparent denial of
the body’s capacity for resistance has also been critically pointed out by
several feminist writers. It has been argued that Foucault understands
the body as too culturally malleable, and that his conception of it is thus
one-sided and limited for feminist purposes.2
1 Susan Bordo (1989, 1993) has appropriated Foucault’s ideas about power and the body in
order to study the different ways that women shape their bodies – from cosmetic surgery
to dieting and eating disorders – and has analyzed these ‘micro-practices’ of everyday
life as disciplinary technologies in the service of normalizing power. According to Bordo,
these normative feminine practices train the female body in docility and obedience to
cultural demands, while at the same time they are paradoxically experienced in terms
of ‘power’ and ‘control’ by the women themselves. For other feminist appropriations
of Foucault, see e.g. Diamond and Quinby 1988, Butler 1990, Hekman 1990, Braidotti
1991, Sawicki 1991 and McNay 1992.
2 See e.g. Bigwood 1991, McNay 1991, Soper 1993.
an a r c hic bo d ies
In order to evaluate Foucault’s account of the body from the perspective of feminist theory, we must first ask what exactly is meant by
the Foucaultian body. The feminist appropriations are based on varying
readings of it. Foucault did not present a theory of the body anywhere,
or even a unified account of it, and thus his conception of it has to be
discerned from his genealogical books and articles, which aim at bringing the body into the focus of history. I will therefore start by taking a
brief look at the central texts in which he discusses the body. My main
focus will be on The History of Sexuality, vol. i, which, I argue, presents
the most fruitful account of the body from the point of view of feminist
philosophy. I will then study three different ways of understanding the
Foucaultian body on the basis of these texts. I will finish by arguing for
a fourth reading, which, unlike most feminist appropriations, gives a
central role to Foucault’s claims about the body as a locus of resistance
to normalizing power.
The body of power
In terms of genealogy, the body is a central theme in Foucault’s thought.
He writes in an article entitled Nietzsche, Genealogy, History that the task
of genealogy is to focus on the body. Nietzsche had attacked philosophy
for its denial of the materiality and vitality of the body, for its pretentious metaphysics that deals only with abstractions such as values, reason
and the soul. Traditional history has been a ‘handmaiden for philosophy’, recounting the necessary birth of these philosophical abstractions
(NGH, 90). Genealogy must, however, be ‘a curative science’, charting
the long and winding history of metaphysical concepts in the materiality of bodies (NGH, 90). Rather than contemplating what is understood
as high and noble, it will focus on the things nearest to it: the body, the
nervous system, nutrition, digestion and energies (NGH, 89). Foucault
writes polemically that the philosopher needs the genealogy of the body
to ‘exorcise the shadow of his soul’ (NGH, 80).
This focus on the body, however, does not mean that it is a genealogical constant, an anchorage point for the unfolding events of history.
According to Foucault, what characterizes genealogy is exactly that it is
without constants: it will reveal in history no recognitions or ‘rediscoverings of ourselves’ (NGH, 88). It makes no assumptions of foundations,
constants or ‘natural’ origins. It is in this context that Foucault outlines
the conception of the body as an unstable entity inscribed and shaped
by history. According to him, we believe that the body obeys only the
fo u c au lt o n freedom
universal laws of physiology, and that history has no influence on it.
In reality, the body is ‘totally imprinted by history’, it can be used and
experienced in many different ways, and its characteristics vary according to cultural practices. It is moulded by the rhythms of work, rest and
holidays; shaped by eating habits and values. Foucault goes so far as to
write that ‘nothing in man – not even his body – is sufficiently stable to
serve as the basis for self-recognition or for understanding other men’
(NGH, 87). The body is ‘the inscribed surface of events’, ‘the locus of
a dissociated self’ (NGH, 83).
Foucault’s aim in this presentation of the body as a historical construct is not to develop some kind of extreme social constructivist theory
of the body. He does not consider the body here as an object of a theory,
but rather as essential to his genealogy in two different ways. The first
point is political or ethical: Foucault wants to use genealogy to study
the history of the very things we believe do not have one: for example, values, essences, identities and the body. Gary Gutting (1994, 10)
argues that, whereas much traditional history attempts to show that our
present situation is inevitable, given the causes revealed by it, Foucault’s
histories aim to show the contingency – and hence surpassability – of
what history has given us.3 Foucault’s point is thus not to argue for an
extreme view of the body as a cultural construction, but to place under
suspicion all claims of its immutable being: essences, foundations and
The second is a methodological point: Foucault wants to bring the
body into the focus of history and to study history through it. He does
not want to write histories of mentalities or ideas, but to focus on the
material practices of the subjection of bodies in order to rethink the
methods of history. Genealogies are thus methodologically distinct in
that they criticize the idea of power operating by the ideological manipulation of minds, as well as all theories in which the subject is understood
as an autonomous free spirit. Foucault’s aim is to show the inadequacy
of these conceptions of the subject by revealing the material instances
that produce the subject.
Foucault’s first major genealogical work, Discipline and Punish,
demonstrates these central points of genealogy. It questions the
inevitability of the prison system by tracking down the historical developments that brought it about. It also examines the functioning of this
3 Foucault himself also explicitly characterized genealogies as histories that have political
meaning, utility and effectiveness. See e.g. QG, 64; TL, 83.
an a r c hic bo d ies
system through the bodies of criminals, and thus brings under scrutiny
the connection between power and the body by analyzing the ways in
which the body is consciously manipulated by disciplinary power. As I
pointed out in the previous chapter, discipline is a form of power that
operates through the body. It consists of various techniques, which aim
at making the body docile and useful. It has both a practical dimension – institutions such as prison, school, hospital – and an abstract
dimension, represented by the human and social sciences, which developed in tandem with them, such as criminology, psychology and pedagogy. Disciplinary power thus demonstrates Foucault’s central idea of
the intertwining of power and knowledge: disciplinary techniques are
important instances of the power/knowledge network.
Foucault’s thought continued to focus on the body throughout the
1970s. His next major work, The History of Sexuality, vol. i, thematizes the
body through the question of sexuality and studies the development of
sexuality as a discursive construct during the last two centuries. Foucault
argues against ‘the repressive hypothesis’ which claims that sexuality in
the Victorian era was repressed and discourse on it silenced. It was
rather that sexuality became the object of a new kind of discourse –
medical, juridical and psychological – and that discourse on it actually multiplied. It also became importantly linked to truth: these new
discourses were able to tell us the truth about ourselves through our
sexuality. Sexuality became essential in determining not only a person’s
moral worth, but his or her health, desire and identity.
The society that emerged in the nineteenth century – bourgeois, capitalist
or industrial society, call it what you will – did not confront sex with a
fundamental refusal of recognition. On the contrary, it put into operation
an entire machinery for producing true discourses concerning it. Not
only did it speak of sex and compel everyone to do so; it also set out to
formulate the uniform truth of sex.
(HS, 69)4
Despite the emphasis on sexuality as a discursive construct, this study is
also a genealogical investigation of the body. Foucault refutes the possible accusations against his analysis of sexuality as a discursive construct
4 ‘La société qui se développe au XVIIIe siècle – qu’on appelera comme on voudra bourgeoise, capitaliste ou industrielle – n’a pas opposé au sexe un refus fondamental de le
reconnaı̂tre. Elle a au contraire mis en œuvre tout un appareil pour produire sur lui
des discours vrais. Non seulement, elle a beaucoup parlé de lui et contraint chacun à en
parler; mais elle a entrepris d’en formuler la vérité réglée’ (VS, 92).
fo u c au lt o n freedom
that completely evades the materiality of the body and the biologically
established existence of sexual functions (HS, 150–1). His response is
that the purpose of the study is, in fact, to show
how deployments of power are directly connected to the body – to bodies,
functions, physiological processes, sensations and pleasures . . . what is
needed is to make it visible through an analysis in which the biological and
the historical are not consecutive to one another, as in the evolutionism of
the first sociologists, but are bound together in an increasingly complex
fashion in accordance with the development of the modern technologies
of power that take life as their objective. Hence, I do not envisage a
‘history of mentalities’ that would take account of bodies only through
the manner in which they have been perceived and given meaning and
value; but a ‘history of the bodies’ and the manner in which what is most
material and most vital in them has been invested.
(HS, 151–2)5
Foucault’s aim is to study the body outside of the dichotomies of
biology–culture, materiality–ideology, prediscursive–discursive. Discursive representations and historical practices are intertwined with material bodies. By claiming that he is studying bodies in their ‘materiality’,
he does not mean that he is approaching them through bio-scientific
theories. These theories are themselves an outcome, as well as being
dependent on certain scientific discourses and historical practices, as
Foucault’s archaeological studies have demonstrated.
Foucault takes up the question of sex (sexe) explicitly at the end of
HS.6 He invents an imaginary opponent who claims that Foucault’s history of sexuality only manages to argue for the discursive construction
of sexuality because he evades ‘the biologically established existence of
sexual functions for the benefit of phenomena that are variable, perhaps, but secondary, and ultimately superficial’ (HS, 150–1). The imaginary critic thus raises the question that many feminist critics of Foucault
have also raised: even if the manifestations of sexuality are culturally
5 ‘comment des dispositifs de pouvoir s’articulent directement sur le corps – sur des corps,
des fonctions, des processus physiologiques, des sensations, des plaisirs; loin que le corps
ait à être gommé, il s’agit de le faire apparaı̂tre dans une analyse où le biologique et
l’historique ne se feraient pas suite, comme dans l’évolutionnisme des anciens sociologues, mais se lieraient selon une complexité croissant à mesure que se développent les
technologies modernes de pouvoir qui prennent la vie pour cible. Non pas donc “histoire
des mentalités” qui ne tiendrait compte des corps que par la manière dont on les a perçus
ou dont on leur a donné sens et valeur; mais “histoire des corps” et de la manière dont
on a investi ce qu’il y a de plus matériel, de plus vivant en eux.’ (VS, 200)
6 HS, 150–9.
an a r c hic bo d ies
constructed and variable, there is nevertheless a material foundation
in the body, a prediscursive, embodied giveness, which cannot be bent
at will.7
Foucault responds to his opponent by firstly denying that his analysis
of sexuality implies ‘the elision of the body, anatomy, the biological, the
functional’ (HS, 151). On the contrary, what is needed is an analysis that
will overcome the biology–culture distinction (HS, 152). Secondly, he
refutes the idea that sex is a material foundation, the ‘other’ with respect
to power: ‘It is precisely this idea of sex in itself that we cannot accept
without examination’ (HS, 152). He then proceeds to write about sex
in quotation marks. He brackets the accepted meaning of the term in
order to be able to study its constitution: how the idea of ‘sex’ took form
in the different strategies of power, and what role it played in them. In
a much quoted passage he writes:
We must not make the mistake of thinking that sex is an autonomous
agency which secondarily produces manifold effects of sexuality over
the entire length of its surface of contact with power. On the contrary,
sex is the most speculative, most ideal, and most internal element in a
deployment of sexuality organized by power in its grip on bodies and
their materiality, their forces, energies, sensations and pleasures.
(HS, 155)8
This passage, among others, has contributed to the feminist interpretations that read Foucault’s claims about sex as interventions in debates
on the sex–gender distinction. In the 1980s the distinction between
biological sex and social gender, which upheld feminist theory in the
1970s, was subjected to fundamental criticism.9 One strand in this criticism was the questioning of the idea of sex as a natural foundation.
Feminist theorists – Judith Butler in particular – referred to Foucault
7 See e.g. Bigwood 1991, 60.
8 ‘Il ne faut pas imaginer une instance autonome du sexe qui produirait secondairement
les effets multiples de la sexualité tout au long de sa surface de contact avec le pouvoir.
Le sexe est au contraire l’élément le plus spéculatif, le plus idéal, le plus intérieur aussi
dans un dispositif de sexualité que le pouvoir organise dans ses prises sur les corps, leur
matérialité, leur forces, leur énergies, leur sensations, leur plaisirs.’ (VS, 205)
9 The conceptual distinction between biological sex and social gender formed the framework that was used to formulate the questions and the possible answers in regard to female
subjectivity. This distinction has, however, been subjected to fundamental criticism in feminist research. It has been argued that it suffers from manifold theoretical ambiguities,
it reproduces the mind–body as well as biological–cultural distinctions that feminist theorists themselves have criticized, and it is politically amorphous and unfocused. See e.g.
Gatens 1983/1991, Butler 1990, Heinämaa 1996.
fo u c au lt o n freedom
and argued that sex, like gender, was a social product or a cultural
Foucault’s discussion of sex at the end of HS is not, however, a comment on feminist discussions of sex or gender, and therefore to read it
through this distinction can be misleading. The original French word
sexe can refer to the categories of male and female in the sense of sex
organs – anatomy and biology that differentiates males from females –
but Foucault’s stress is clearly on the sense of the natural function, an
embodied foundation or principle that belongs in common to both
men and women. Sex is understood as a hidden cause behind observable characteristics and behaviour. Foucault argues that ‘sex’ is a complex idea that was formed inside the deployment of sexuality. It is an
idea of an inner truth, an idea that ‘there exists something other than
bodies, organs, somatic localizations, functions, anatomo-physiological
systems, sensations, and pleasures; something else and something more,
with intrinsic properties and laws of its own: “sex”’ (HS, 152–3). He elaborates the idea further by writing that sex is a ‘form of secret causality’,
an interplay of ‘the visible and hidden’ (HS, 152).
Foucault thus did not problematize sex in the sense that feminist
theory has done. He did not question how the categories of male
and female are constructed or what consequences they have for the
behaviour or empowerment of women. By claiming that sex is imaginary (HS, 156) and the most ideal element in strategies of power
(HS, 155), he was not arguing that femaleness is imaginary, ideal or arbitrary. Rather, he was trying to problematize a certain kind of explanatory
framework of sexuality: the idea of a foundation or an invisible cause
that supports the visible effects.
Foucault takes up the question of sex as a principle of explanation
for the classification of bodies into females and males in his introduction to the book Herculine Barbin: Being the Recently Discovered Memoirs of
a Nineteenth-Century French Hermaphrodite. Foucault now poses the question whether we really need the idea of a true sex. Using the example of
a hermaphrodite, he attempts to make visible how deep in our thinking lies the idea that everybody has a definite and naturally given sex
that is the truth and cause of our behaviour as well as of the observable
sexual characteristics. This true sex determines the individual’s gender
identity, behaviour and desire for the opposite sex. What the body of
a hermaphrodite can show is that there is no true sex to be found in
10 See Butler 1990.
an a r c hic bo d ies
our body, but that this idea is rather a product of the development of
scientific discourse and juridical procedures (IHB, x–xi).
Foucault refers to the Middle Ages, when it was common practice to
think that a hermaphrodite was a person that combined both masculine
and feminine characteristics. When the individual had legally reached
adulthood, he or she could choose which sex to keep. This conception was superceded by scientific theories about sex, which developed
around the same time as juridical concepts and practices relating to
the idea of a true sex. Everybody had only one true sex, which could
be settled conclusively by experts. All the characteristics of the opposite
sex in one’s body and soul were deemed arbitrary, imaginary or superficial. The true sex further determined the individual’s gender role,
and his or her moral responsibility was to behave according to this true
sex. The doctor, as the expert in recognizing this true sex, had to ‘strip
the body of its anatomical deceptions and discover the one true sex
behind organs that might have put on the forms of the opposite sex’
(IHB, viii–ix).
Here, Foucault is using the notion of sex in more or less the way it
is understood in the sex/gender discourse. He discusses the idea of a
true sex from which gender, understood as social roles and culturally
acquired characteristics, follows. This idea also underlies the feminist
distinction between sex and gender, which feminists have used to argue
that natural sex does not determine social gender. Foucault does, however, also critically appraise the idea of natural, scientifically true sex by
revealing its historical construction. His aim, again, is to question the
whole explanatory framework of natural foundations and secondary
effects. He does not claim here, either, that sex as the categories of
maleness and femaleness was invented in a particular historical period
and that we could give them up when we wanted to. He rather analyzes the ways in which these categories were scientifically founded and
explained in discourses of truth, and how this ‘pure’ explanation in fact
constituted these categories so that they were understood as ‘natural’.
The discursive body
Foucault’s conception of the body is often summarized by saying that
he understands the body to be discursively constructed, but this claim
is interpreted in very different ways. The first source of confusion is
the fact that the notion of discourse is understood in different ways.
Sometimes ‘discursive’ is understood in a strictly linguistic sense, as
fo u c au lt o n freedom
something that is linguistic or linguistically structured. Sometimes
discourse is understood in a more general sense, as a cultural practice
and discursive means, more or less the same as ‘cultural’ or ‘culturally
constructed’. Furthermore, there are at least three different interpretations of discursive – linguistic or cultural – construction.11
First, we can understand discursive in the strict sense of linguistic,
and argue that by denying the prediscursive body Foucault is claiming
that the way we identify and understand the body is linguistically constructed. The prediscursive body is, by necessity, impossible to identify
and theorize, because as soon as we name it and start to talk about it
we have already brought it into the realm of discourse. William Turner
(2000, 112), for example, explicates Foucault’s and Butler’s thinking
by writing: ‘Our conceptions of our bodies, whether as material, or
important, or neither, come to us through language; the belief in a
preculturally material body as the ultimate ground of identity itself
depends on the circulation of meanings in a culture.’
I call this first reading the weak version of the idea of a discursive
body, because by denying the prediscursive body it in fact says nothing
about the body itself. Even if the linguistic representations are inevitably
constructed in networks of power, the body itself, in its materiality, could
escape this linguistic construction. However, Foucault himself refutes
this reading:
Hence I do not envisage a ‘history of mentalities’ that would take account
of bodies only through the manner in which they have been perceived
and given meaning and value; but ‘a history of bodies’ and the manner
in which what is most material and most vital in them has been invested.
(HS, 152)12
The second interpretation, what I call the intermediate reading, claims
that the cultural construction covers the body itself in its materiality and
11 The possibilities of understanding the discursive body are further increased if discursive
is understood in the psychoanalytic sense of symbolic. Although Foucault’s relationship
with psychoanalysis is explicitly critical in HS, one could argue that behind this explicit
relationship lies an unacknowledged debt. Charles Shepherdson (2000, 182), for example, argues that the canonical reception that opposes Foucault and Lacan does not do
justice to the complexity of their relation. On Foucault’s relationship to psychoanalysis,
see also e.g. Miller 1989/1992.
12 ‘Non pas donc “histoire des mentalités” qui ne tiendrait compte des corps que par la
manière dont on les a perçus ou dont on leur a donné sens et valeur; mais “histoire des
corps” et de la manière dont on a investi ce qu’il y a de plus matériel, de plus vivant en
eux’ (VS, 200).
an a r c hic bo d ies
not just in its cultural and linguistic representations. However, there
remains a stable core of the body, imposing a limit that cultural manipulation cannot cross. David Couzens Hoy (1999, 5), for example, takes
up Foucault’s claim in ‘Nietzsche, Genealogy, History’ that genealogy
exposes ‘a body totally imprinted by history and the process of history’s
destruction of the body’ (NGH, 83). Hoy argues that this sentence
implies that there must be something natural before history, which is
then destroyed by it. This intermediate reading thus accepts that the
Foucaultian body is culturally constructed, even in its materiality, to a
certain extent, but that we must posit some kind of universal invariance.
The border between nature and culture in this kind of reading is variably drawn. Dreyfus and Rabinow (1982, 111) argue that the stable core
in Foucault’s account is drawn from Merleau-Ponty’s phenomenology
of the lived body and consists of ahistorical structures of the perceptual field, such as size constancy, brightness constancy and up–down
Nevertheless, this second reading seems incompatible with Foucault’s explicit effort to dismantle the nature–culture dichotomy on
which it heavily relies. In Foucault’s thought, we cannot make any claims
about what in our bodies is natural and what is culturally variable. This
distinction between nature and culture should itself be understood as
an effect of a certain discourse that produces the idea of a natural body.
This kind of interpretation, while relying on certain passages of NGH,
must also either ignore or attribute to Nietzsche the strong formulations
of a culturally constructed body that also occur in it. Foucault writes in
NGH, for example, that ‘nothing in man – not even his body – is sufficiently stable to serve as a basis for self-recognition or for understanding
other men’ (NGH, 87).
The third or strong reading denies any dimensions or variations of
embodiment that are not historically and culturally constructed. We
can only understand as well as experience our bodies through culturally mediated representations, but bodies themselves are also shaped in
their very materiality by the rhythms of culture, diets, habits and norms.
The body is, furthermore, not only constructed by cultural signifying
practices, but its constitution is understood as inscription that marks it
so that it becomes a signifying practice itself. Elisabeth Grosz (1994),
for example, reads into Foucault’s account of the body a description
of the corporeal inscription of the body-as-surface. She argues that in
our culture there is a form of body writing and various techniques
of inscription that bind all subjects both violently and in more subtle
fo u c au lt o n freedom
forms. In the first case, violence is demonstrable in social institutions
of correction and training, such as prisons, juvenile homes and psychiatric institutions. Less openly violent are the inscriptions of cultural and
personal values, norms and commitments according to categorizations
of the body into socially significant groups such as male and female.
The body is involuntarily marked, but it is also incised through ‘voluntary’ procedures, life-style and habits. Grosz claims that there is nothing
natural or ahistorical about these modes of corporeal inscription that
make bodies into particular types (Grosz 1994, 141–2).
Judith Butler appropriates Foucault’s thought when she argues in
Gender Trouble that the body gains meaning within discourse only in the
context of power relations. According to her, the body ‘is not a being,
but a variable boundary, a surface whose permeability is politically regulated, a signifying practice within a cultural field of gender hierarchy
and compulsory heterosexuality’ (Butler 1990,139). Butler criticizes all
feminist efforts to liberate the female body from the determinations of
patriarchal power. The culturally constructed body cannot be liberated
to its ‘natural’ past, nor to its original pleasures, but only to an open
future of cultural possibilities (Butler 1990, 93). The feminist task that
Butler sets at the end of her book is the radical proliferation of gender,
the displacement of gender norms by their parodic repetition.
Both Butler and Grosz, while appropriating Foucault’s conception
of the body, also criticize it for its hidden commitment to foundational,
primordial materiality. Unlike some of the proponents of the second
reading, they see this as a contradiction within Foucault’s account which
denies the possibility of a body outside power, while at the same time
implicitly presupposing it. According to Butler, Foucault seems to argue
for the cultural construction of bodies, but his theory in fact contains
hidden assumptions which reveal that he also understood them to be
outside the reach of power (Butler 1990, 94–5). He saw the body as
a medium, a blank page on which cultural values are inscribed. Thus
he must have envisaged a body prior to this cultural inscription: the
body as pure materiality prior to signification and form. According to
Butler, this idea is not compatible with Foucault’s genealogical account
of sex and sexuality. She demands a radicalization of Foucault’s theory: a
genealogy of the body in its discreteness that addresses the issue of how
bodily meanings as well as margins are invested with power (132–3).13
13 Sara Heinämaa (1996, 299) argues that Butler herself occasionally uses the terminology
of raw material and production in Gender Trouble. The metaphor of production brings
an a r c hic bo d ies
While the strong interpretation of the Foucaultian body has resulted
in influential feminist appropriations – for example, Butler’s and
Grosz’s – it also contains serious problems from a feminist point of
view. The wide feminist criticism that Butler’s understanding of the
body in Gender Trouble received testifies that there are at least some difficulties involved in trying to encapsulate female embodiment through
strong cultural constructivist accounts of the body. I will take up three
sets of questions that I see as problematic in Butler’s account of the
body as presented in Gender Trouble,14 and therefore also in Foucault’s
account when it is understood according to the strong reading.15 I will
only briefly outline the first two. The third cluster of questions will be
the focus of my discussion.
The resistance of the body
The first problem can be called the question of identification. Butler,
appropriating Foucault, argues in Gender Trouble that the unity of
with it the idea of raw material, that is, a natural substance that is prior to and independent of the process of production. In the traditional sex/gender thinking that Butler
sets out to criticize, females and males are treated as the raw material of gender production. Butler’s new claim is that even females and males are products, and thus there must
be something that precedes these constructions and which passes as their raw material:
a body that is free from the sexed categories of culture. Even though the line between
naturally given and culturally produced is drawn in a new way, it is still there. Heinämaa
argues that this is why, in order to get rid of the nature–culture distinction, Butler has
to elaborate on her claim by adding that the production of sex is a process which even
generates its own raw material, and that, in effect, becoming sexed must be conceived
of as a process of repetitive and citational action.
14 Butler’s books that followed Gender Trouble, particularly Bodies that Matter and The Psychic
Life of Power, partly answer these questions. She argues in Bodies that Matter that to defend
a culturally constructed body does not mean that one understands cultural construction
as a single, deterministic act or as a causal process initiated by the subject and culminating
in a set of fixed effects. In place of these conceptions of construction, she suggests a
return to the notion of matter as ‘a process of materialization that stabilizes over time to
produce the effect of boundary, fixity, and surface we call matter’ (Butler 1993, 9). Butler
emphasizes the gaps and fissures that are opened up in this process of materialization:
the constitutive instabilities become the deconstitutive possibilities. She also turns to
psychoanalysis in order to understand these disruptions as imaginary contestations that
effect a failure in the workings of the law, but also importantly as occasions for a radical
rearticulation of the symbolic domain. The political dimension of her work is thus again
safeguarded: even if the female body cannot be liberated, the meaning of what counts
as a valued and valuable body can be altered.
15 Despite Butler’s criticism of Foucault’s understanding of the body in Gender Trouble, her
conception is often conflated with Foucault’s in feminist literature and referred to as
the post-structuralist body. See e.g. Bigwood 1991.
fo u c au lt o n freedom
gender is an effect of a regulatory practice that seeks to render gender
identity uniform through compulsory heterosexuality. Cultural representations of the body and its sex as a natural and necessary ground
for gender identity have a normative function in the power/knowledge
strategy that forces individuals into two opposed gender categories.
Gender identity is discursively constructed as a normative ideal and
then performatively produced by those acts that are understood to
be its effects. This normative unity is never fully installed, however.
The dichotomies of male–female, masculine–feminine are constantly
undermined by gender discontinuities in the sexual communities in
which gender does not necessarily correspond with sex. Butler’s theory, while appropriating Foucault’s thought, also reveals the limitations
of the Foucaultian framework when applied to the question of gender:
it leads to an oversimplified notion of gender identity as an imposed
effect. We cannot understand the constitution of gender identity only
through the normative ideals and practices that prevail in our culture,
nor through the individual’s conscious choices. There are experiences,
sensations and lives that do not properly fit within the limits of the normal. People identify with stigmatized subject positions, or even socially
abject positions, and often this identification is strongly tied to their
bodies. As Stuart Hall (1996, 10–11) formulates the problem, Foucault’s thought ‘reveals little about why it is that certain individuals
occupy some subject positions rather than others’ or ‘what might in
any way interrupt, prevent or disturb the smooth insertion of individuals into the subject positions’. Corporeal feminist theory has emphasized that embodiment is definitive of subjectivity. Feminist theory must
therefore develop some understanding of why or how subjects identify
with the sexual subject positions, for example, to which, according to
Foucault’s analysis, they are summoned.16
The second and related problem could be called the problem of
singularity. The feminist question concerns not only how different subject positions are constituted, but also how subjects fashion, stylize and
produce these positions, and why they never do so completely. If female
embodiment as well as subjectivity is an effect of a constitutive power,
16 In interviews about sexuality or the politics of sexuality, Foucault stressed the dangers
of legal control imposed on sexual practices. In connection with homosexuality, he
showed concern about the strategic role played by sexual preference within a legal and
social framework and said very little about the meaning of homosexual behaviour as
such. He strongly refused to offer any comment as regards to the distinction between
innate predisposition to homosexual behaviour and social conditioning (SC, 288). All
he would grant is that there is ‘a certain style of existence . . . or art of living, which
might be called “gay”’(SC, 292).
an a r c hic bo d ies
how is it possible that there are several different variations of it? Even
if we identify with the subject positions to which we are summoned –
in the case of gender, even if my gender identity appears almost selfevident both to myself and to others – I still have the singular style
of living my female embodiment. Despite providing an explanation
of the normative construction of the female body, feminist theory has
not yet accounted for or explained in any way the variations of female
The third, and from my perspective the most interesting, set of questions concerns the possibility of resistance to normative power. The
only possibilities for resistance against subjection that a strong interpretation of the discursive body seems to allow open up through the
gaps in the struggle with competing regimes. The subjection of bodies
is never complete because the deployments of power are always partial and contradictory. Foucault insists that ‘where there is power there
is resistance’ (HS, 95). The points of resistance are distributed in an
irregular fashion throughout the power network. They are the ‘odd
term in relations of power’ (HS, 96), its blind spot or evading limit.
Power is thus not deterministic machinery, but a dynamic and complex strategical situation. In her book The Psychic Life of Power, Butler
analyzes and concisely explicates this idea of resistance in Foucault’s
thought. In Foucault, resistance appears (a) in the course of a subjectivation that exceeds the normalizing aims by which it is mobilized, or
(b) through convergence with other discursive regimes, whereby inadvertently produced discursive complexity undermines the teleological
aims of normalization. Butler concludes: ‘Thus resistance appears as the
effect of power, as part of power, its self-subversion’ (Butler 1997, 93).
Hence, she puts forward the view that resistance constitutes ‘a hazard’ in
normalizing power as the only viable account of resistance in Foucault.
The idea that bodies themselves could generate any resistance, she thus
sees as either a naive mistake by Foucault or as simply impossible within
his framework.
From a feminist point of view, this means that, while a focus on bodies seems to open up important connections with Foucault’s thought,
the apparent denial of the body’s capacity for resistance seems to refute
all feminist political goals. Lois McNay (1992), for example, argues in
her book Foucault and Feminism that Foucault’s historical studies give
the impression that the body presents no resistance to the operations
of power. Although Foucault insists that power is always accompanied
by resistance, he does not elaborate on how this resistance manifests itself through the body. McNay (1992, 12) argues that this is
fo u c au lt o n freedom
particularly problematic for feminist theory, given that a significant aim
of the feminist project is the rediscovery and revaluation of the experiences of women. Foucault cannot account for women’s strategies of
resistance: for the fact that women did not simply slip passively into
socially prescribed feminine roles (McNay 1992, 41).
I will next argue for a fourth reading of Foucault’s understanding of the body, which does not operate with the dichotomies of
nature–culture, prediscursive–discursive. I will take a closer look at what
Foucault meant by the idea of bodies and pleasures as a form of resistance to power, and argue that this idea implies an understanding of
the body as always extending the discursive frameworks that attempt to
control and contain it. I argue that this fourth reading could provide
a fruitful starting point for the feminist accounts that seek to understand not only the cultural construction of the female body, but also its
propensity to resistance.
The anarchic body
In an important passage of The History of Sexuality, vol. i, Foucault writes:
We must not think that by saying yes to sex, one says no to power; on the
contrary, one tracks along the course laid out by the general deployment
of sexuality. It is the agency of sex that we must break away from, if we
aim – through a tactical reversal of the various mechanisms of sexuality –
to counter the grip of power with the claims of bodies, pleasures and
knowledges, in their multiplicity and their possibility of resistance. The
rallying point for the counterattack against the deployment of sexuality
ought not to be sex desire, but bodies and pleasures.
(HS, 157)17
Here Foucault suggests that it is in the body that the seeds for subverting
the normalizing aims of power are sown. The body becomes a locus of
resistance. Foucault elaborates further on this in an interview:
Mastery and awareness of one’s own body can be acquired only through
the effect of an investment of power in the body . . . But once power
17 ‘Ne pas croire qu’en disant oui au sexe, on dit non au pouvoir; on suit au contraire le
fil du dispositif général de sexualité. C’est de l’instance du sexe qu’il faut s’affranchir si,
par un retournement tactique des divers mécanismes de la sexualité, on veut faire valoir
contre les prises du pouvoir, les corps, les plaisirs, les savoirs, dans leur multiplicité et
leur possibilité de résistance. Contre le dispositif de sexualité, le point d’appui de la
contre-attaque ne doit pas être le sexe désir, mais les corps et les plaisirs.’ (VS, 207–8)
an a r c hic bo d ies
produces this effect, there inevitably emerge the responding claims and
affirmations, those of one’s own body against power, of health against
the economic system, of pleasure against the moral norms of sexuality,
marriage, decency. Suddenly, what had made power strong becomes used
to attack it. Power, after investing itself in the body, finds itself exposed
to a counterattack in the same body.
(B/P, 56)18
Passages like these have been read by Foucault’s critics as contradictory
claims about the body’s return to ‘a non-normalizable wildness’ (Butler
1997, 92). In Gender Trouble Judith Butler criticizes Foucault’s understanding of the body along these lines.19 According to her, when we
consider those ‘textual occasions on which Foucault criticizes the categories of sex and the power regime of sexuality, it is clear that his own
theory maintains an unacknowledged emancipatory ideal that proves
increasingly difficult to maintain, even within the strictures of his own
critical apparatus’ (Butler 1997, 94). While Foucault advocates the critical deconstruction of sexuality and sex in HS, he does not extend it
to the sexed body, but naively presents bodies and pleasures as the
site of resistance against power. Butler argues that Foucault’s mistake
is reiterated in his short but significant introduction to the journals of
Herculine Barbin. He romanticizes Herculine’s sexuality as a world of
pleasures and a happy limbo of non-identity by mistakenly presuming
the hermaphrodite body and sexuality to be free from the categories
of sex and of identity. According to Butler, Foucault fails to recognize the concrete relations of power that both construct and condemn
Herculine’s sexuality (94). He thus reads Herculine’s memoirs wrongly
by suggesting that they can reveal something about the undefined
field of pleasures, which is not marked by the categories of sex. The
18 ‘La maı̂trise, la conscience de son corps n’ont pu être acquises que par l’effet de
l’investissement du corps par le pouvoir . . . Mais, dès lors que le pouvoir a produit
cet effet, dans la ligne même de ses conquêtes, émerge inévitablement la revendication
de son corps contre le pouvoir, la santé contre l’économie, le plaisir contre les normes
morales de la sexualité, du mariage, de la pudeur. Et, du coup, ce par quoi le pouvoir
était fort devient ce par quoi il est attaqué’ (PC, 754–5).
19 Elizabeth Grosz reiterates this criticism in her book Volatile Bodies. See Grosz 1994, 155.
In her next book, Space, Time and Perversion, however, she presents what she calls ‘the
most generous reading’ of what Foucault means by bodies and pleasures. She argues
that Foucault is suggesting that the body may lend itself to economies and modes of
production that are other than the ones that produce ‘sexuality’. A different economy
of bodies and pleasures may find the organization of sexuality, the implantation of our
sex as the secret of our being, curious and intriguing instead of self-evident (Grosz 1995,
fo u c au lt o n freedom
overthrow of a univocal sex cannot result in sexual multiplicity or undefined pleasures. Herculine’s body and pleasures are still produced by
power, only her body is produced as a sign of fatal ambivalence in the
practices and discourses based on the idea of a univocal sex (95–6).
Are Foucault’s references to the body as the locus of resistance
merely naive slippages into the idea of a prediscursive body? Do we
have to accept Butler’s reading of the Foucaultian body according to
which these passages are implicit contradictions within his thought?
Are bodies and pleasures within the same discursive order as sex and
sexuality? What can bodies and pleasures as an alternative to sex desire
mean? I will argue that we must take seriously Foucault’s idea of bodies
and pleasures as forms of resistance, and not treat it as a naive fallback to the prediscursive body, or as a return to the non-normalizable
Gilles Deleuze wrote in private notebooks originally given to
The last time we saw each other, Michel told me, with much kindness and
affection, something like, I cannot bear the word desire; even if you use
it differently, I cannot keep myself from thinking or living that desire =
lack, or that desire is repressed. Michel added, whereas myself, what I call
pleasure is perhaps what you call desire; but in any case I need another
word than desire.
(Deleuze 1994/1997, 189)20
In his notebooks Deleuze poses the same question to Foucault as have
many other commentators: if in Foucault’s thought the network of
power/knowledge is constitutive, what is the status of the phenomena
of resistance against it? Deleuze finds in Foucault three possible directions in which one could go for an answer: the relations of forces, truth
and pleasure (Deleuze 1994/1997, 188). He takes up the possibility of
pleasure and emphasizes the importance Foucault assigned to life as a
way of giving a possible status to forces of resistance.21 Deleuze ends by
20 ‘La dernière fois que nous nous sommes vus, Michel me dit, avec beaucoup de gentillesse
et affection, à peu près: je ne peux pas supporter le mot désir; même si vous l’employez
autrement, je ne peux pas m’empêcher de penser ou de vivre que désir = manque,
ou que désir se dit réprimé. Michel ajoute: alors moi, ce que j’appelle “plaisir”, c’est
peut-être ce que vous appelez ‘désir’; mais de toute façon j’ai besoin d’un autre mot
que désir.’ (Deleuze 1994, 63)
21 In his book Foucault (1986/1988), Deleuze attributes to Foucault the idea of life as
resistance: ‘that power does not take life as its objective without revealing or giving rise
to a life that resists power’ (94). See also HS, 144–5.
an a r c hic bo d ies
suggesting that Foucault’s ‘pleasures body’ is the correlative of his own
idea of ‘body without organs’, the body as a site of the production of
positive forces and creative differences.22
It is also my contention that if we wish to consider Foucault’s idea of
bodily resistance, we must leave behind the conception of the body as
a mere material object, the body as an object of natural sciences and
disciplinary technologies. If we conceive of the body as a passive object,
it is possible to discipline it, but equally impossible to theorize about
its resistance to normalizing power. The question of resistance arises if
we take the experiential body – the body as experiencing in everyday
practices of living – as the starting point. While Foucault conceived of
the body strictly in terms of an object of disciplinary manipulation in
Discipline and Punish, I argue that such a conception does not underlie his account of the body in The History of Sexuality, vol. i, in which
he presupposes a more dynamic understanding of the body through
sexuality. He does not explicitly mention experience in this work, but
his claim about bodies and pleasures presupposes an understanding
of the experiential body insofar as pleasure can only be understood as
an experience of pleasure, not solely as a concept or a practice. While
feminist theory has widely appropriated HS in connection with issues
of sexuality and sex, it is normally interpreted as identical to DP in
its account of the body as an object of disciplinary manipulation. Disciplinary power in connection with sexuality is simply complemented
with biopower and deployments of sexuality. Such a reading, however,
overlooks what I think are some of the most interesting aspects of the
account of the body in HS, and which point to the potential of bodily
22 Deleuze understands bodies as specific modes of events and intensities, series of processes, as active forces. To think of a ‘body without organs’ is to refuse any single organization or interpretation of the body and to view it as an active and positive multiplicity.
The body is productive because it connects with other organs, flows and intensities. See
Deleuze and Guattari 1980/1987. A study of the similarities and differences between
Deleuze’s and Foucault’s conceptions of the body is beyond the scope of this work. On
feminist interpretations of Deleuze’s conception of the body, see e.g. Braidotti 1994
and Grosz 1994.
23 Elizabeth Grosz (1994, 146) argues that for Foucault the body is the target of power
and a stake in the struggle for power’s control over a materiality that is dangerous
to it, precisely because it is unpredictable and able to be used in potentially infinite
ways, according to infinitely variable cultural dictates. According to Grosz, Foucault
derives his understanding of the body mainly from Nietzsche, who understands the
body’s capacity for becoming as something that can never be known in advance or be
charted. The body’s limits cannot be definitively listed because it is always in a position
of self-overcoming, of expanding its capacities (Grosz 1994, 124).
fo u c au lt o n freedom
A much earlier text than Discipline and Punish illuminates Foucault’s
idea of bodies and pleasure as put forward in The History of Sexuality,
vol. i. ‘A Preface to Transgression’, written over ten years before HS
and dedicated to Bataille, takes up the question of sexual experience.
Foucault questions the limits of experience and the acts of transgression, and notes that ‘transgression is an action that involves the limit’,
it demands it for its existence (PT, 73). The limit and the transgression depend on each other: a limit could not exist if it were absolutely
uncrossable and reciprocally, transgression would be ‘pointless if it
merely crossed a limit composed of illusions and shadows’ (PT, 73).
The excess of experience, the transgression, not only presupposes the
limit, it constitutes it in overcoming it and momentarily opens it up to
the limitless. It forces ‘the limit to face the fact of its imminent disappearance, to find itself in what it excludes’ (PT, 73). Foucault argues
that transgression is thus not a victory over limits, it is not a negative
or a revolutionary act. Rather, it reaffirms limited being, while also
momentarily opening up a zone of limitlessness to existence (PT, 74).
However, this affirmation contains nothing positive either, in the sense
that no content can bind this experience, which, by definition, has
no limits. Foucault suggests that perhaps it is simply an affirmation of
a division, a testing of the limit, a contestation. Transgression is not
related to the limit ‘as black to white, the prohibited to the lawful, the
outside to the inside, or as the open space of a building to its enclosed
spaces’ (PT, 73–4). The relationship between limit and transgression
is rather like ‘a spiral that no simple infraction can exhaust’ (PT, 74).
Limit and transgression are thus irrevocably tied to each other, they
constitute each other and constantly reaffirm and contest each other.
Transgression creates a limit that exists only in the movement which
crosses it.
If we read the idea of sexual experience as an overcoming of limits
into Foucault’s understanding of the body, we can interpret this in two
ways. Firstly, the experiential body can transgress the limit between the
normal and the abnormal. There are transgressive experiences that fall
outside the limits of the normal, but which are necessary for constituting
its limits. In an interview Foucault opposed the term desire because it
functions as a calibration in terms of normality: ‘I am advancing this
term (pleasure), because it seems to me that it escapes the medical and
naturalistic connotations inherent in the notion of desire . . . There is no
“pathology” of pleasure, no “abnormal” pleasure’ (quoted in Halperin
1995, 93–4).
an a r c hic bo d ies
While modern techniques of power use desire to attach to us a
true self defined in part by our sexuality, the notion of pleasure
cannot be used in the same way as a tool for social and scientific
regulation.24 The sexual body understood as a body experiencing pleasure and not as a determinable object of the sciences and of institutional practices becomes a possibility for subverting normalization.
Foucault’s view would thus be that bodies and pleasures can gain meaning in discourse only, but that this discourse would be different than
the psychologico-medical discourse that produces our conception of
normal sexuality.
I argue that Foucault also makes a more radical claim by taking up
limit experiences. He argues that the limit between discursive intelligibility and unintelligibility can also be crossed in experience. Like
Bataille, he is interested in experiences at the limits of language. This is
a more radical interpretation, because even the abnormal experiences
may still be within discursive intelligibility. Foucault advanced pleasure
and opposed desire as a grid of intelligibility:
That notion has been used as a tool, as a grid of intelligibility, a calibration
in terms of normality . . . Desire is not an event but a permanent feature
of the subject: it provides a basis on to which that psychologico-medical
armature can attach itself. The term pleasure, on the other hand, is virgin
territory, unused, almost devoid of meaning . . . It is an event ‘outside
the subject’, or at the limit of the subject, taking place in that something
which is neither of the body or the soul, which is neither inside nor
outside – in short, a notion neither assigned nor assignable.
(quoted in Halperin 1995, 93–4)25
24 David Halperin (1995, 93–4) takes up Foucault’s distinction between desire and pleasure as his way of distancing himself from the idea of desire associated with Deleuze’s
philosophy. According to Halperin, Foucault’s famous and rather cryptic remarks at the
end of volume i of The History of Sexuality, about the political importance of attacking
sexuality and promoting pleasures at the expense of sex, make more sense when they
are set in the context of his insistent distinction between pleasure and desire. Halperin
emphasizes the idea that pleasure is an event ‘at the limit of the subject’: intense pleasure
is desubjectivating, impersonal.
25 The interview, conducted on 10 July 1978, was published in Dutch in 1982 (‘Vijftien
vragen van homosexuele zijde san Michel Foucault’, Interviews met Michel Foucault, ed.
M. Duyves and T. Massen, Utrecht: De Woelrat) and in French in 1988 (‘Le Gai savoir’,
Mec Magazine). There is a transcription of the original interview in the Centre Michel
Foucault in Paris, but the text has been omitted from the four-volume collection of
Foucault’s texts, Dits et écrits 1954–1988. The translation here is by David Halperin,
who also provides the original French text in a footnote (181, p. 217). ‘Cette notion
a été utilisée comme un outil, une mise en intelligibilité, un étanlonnage en terme de
fo u c au lt o n freedom
According to Foucault, the power/knowledge network constitutes the
subject and also all forms of experience, but this does not mean that
they are discursively constituted. As I showed in chapter 4 above, in
Foucault’s genealogy the regime of discourse, the episteme, only constitutes the specifically discursive element of a more general regime, the
dispositif or apparatus, which is both discursive and non-discursive.26
Hence, we can interpret that not all experiences are discursively constituted, even though their (linguistic) intelligibility is. There are experiences that fall outside of what is constituted by discourse in the sense
that these abject or transgressive experiences are rendered mute and
unintelligible in our culture. They might, for example, be experiences
induced by drugs or experiences that we try to make intelligible by classifying them as forms of madness. However, what Foucault also seems
to suggest is that sexual experiences of pleasure can never be wholly
reduced to discursive meanings either. The sexual body is always discursive in the sense that it is an object of scientific discourses and disciplinary technologies. Nevertheless, the sexual body as experiencing
is capable of multiplying, distorting and overflowing its discursive definitions, classifications and coordinates. The experiential body can take
normal language to the point where it fails, where it loses its power of
definition, even of expression. This does not mean a return to a prediscursive body, however. Rather, it is that the body as a contestation exists
on the limits of discourse, in those moments ‘when language, arriving
at its confines, overleaps itself, explodes and radically challenges itself
in laughter, tears, the eyes rolled back in ecstasy’ (PT, 83). The experience of the limit can be realized in language but only at the moment
‘where it says what cannot be said’ (PT, 86).27
Sexual experiences transgress and also constitute the limit between
the norm and what falls outside of it. The experiential body is not
normalité . . . Le désir n’est pas un événement, mais une permanence du sujet, sur
laquelle se greffe toute cette armature psychologico-médicale. Le terme de plaisir de
son côté est vierge d’utilisation, quasiment vide de sens. Il n’y a pas de “pathologie” du
plaisir, de plaisir “anormal”. C’est un événement “hors sujet”, ou à la limite du sujet,
dans ce quelque chose qui n’est ni du corps ni de l’âme, qui n’est ni à l’intérieur, ni à
l’exterieur, bref une notion non assignée et non assignable.’
26 See e.g. CF, 197.
27 Cf. Shepherdson 2000, 5. According to Shepherdson, Lacan understands transgression
and law very similarly. The rule of law does not repress or prohibit, but produces its own
exception. The symbolic order functions only on the basis of this exception or excess.
The excess is not a natural phenomenon that disrupts the machinery of culture; it is
rather a peculiar feature of culture itself, an effect of language, which includes its own
malfunction, a remainder that marks its limits (Shepherdson 2000, 175–80).
an a r c hic bo d ies
outside the norms, but neither is it fully within them. It cannot be
reduced to either one of these alternatives. The very process of normalization sets the limits for normal experiences, but these limits open
up possibilities of transgression that affirm the potential limitlessness
of the body. The Foucaultian body is capable of generating resistance,
of presenting not malleability but excess and transgression as pleasure.
This resistance is not a return to a wild and natural body, however,
but it is resistance made possible by the normalizing power. Foucault
writes about sexuality in a passage that some readers would perhaps
find surprising and unlike him:
[Sexuality] is a part of our world freedom. Sexuality is something that
we ourselves create – it is our own creation, and much more than the
discovery of a secret side of our desire. We have to understand that with
our desires, through our desires, go new forms of relationships, new forms
of love, new forms of creation. Sex is not a fatality: it’s a possibility for a
creative life.
(SPPI, 163)28
In Bodies that Matter, Butler herself posits an outside to the culturally
constructed body. She writes that there is an ‘outside’ to what is constructed by discourse, even though this is ‘not an absolute “outside”, an
ontological thereness that exceeds or counters the boundaries of the
discourse’ (Butler 1993, 8). Rather, it is a constitutive ‘outside’ ‘which
can only be thought – when it can – in relation to that discourse, at and
as its most tenuous borders’ (8). Butler argues that the bodies that fail
to materialize the norm provide the necessary ‘outside’ for the bodies
that qualify as bodies that matter (16). Hence, Butler seems to hold
that when Foucault assumes an outside to the discursively constructed
body, he effects a naive slippage to an outside as ‘ontological thereness’,
whereas her own notion of an ‘outside’ is always in quotation marks
because it is not an ontological outside, but only becomes possible in
relation to discourse. I argue, however, that Foucault’s understanding of the sexual body, like Butler’s account, is an effort to dismantle
the dichotomy of the culturally constructed versus the natural, and to
inquire into the discursive limits of the body and experience.
Hence, when we discuss the different interpretations of a discursive
body we must make one more distinction. Rather than referring only
to the different senses or degrees of the cultural construction of the
28 This interview was conducted in Toronto in 1982, and appeared originally in English.
fo u c au lt o n freedom
body as a material and bio-scientific object, we must also distinguish
when we talk about the cultural construction of experiences. The experiential body is a locus of resistance in the sense that it forms the spiral
of limits and transgressions. Power/knowledge inscribes the limits of
normal experiences, but it is exactly the existence of these limits that
makes their transgression possible. The experiential body is constituted
by power/knowledge networks, but the limits of its experiences can
never be firmly set because they cannot ever be fully determined and
articulated. The experiential body can multiply, distort and overflow
the meanings, definitions and classifications that are attached to experiences, and in this sense it is capable of discursively undefined and
unintelligible pleasures, for example. It is the permanent contestation
of discursive definitions, values and normative practices. Transgressive
experiences are necessary ‘outsiders’ because they constitute the limits
of the normal and the intelligible. In Foucault’s genealogy, as in his
archaeology, there is thus a dimension of freedom in the sense of a
constitutive outside to the discursive order, even though there is no
outside to the apparatus or cultural network of practices as a whole.
If language can never be totally mastered and brought within the discursive order, neither can experience be ever wholly defined. It always
remains contestable and resistant to articulation.
Reading the Foucaultian body as the experiential body raises the
question of its relationship to Merleau-Ponty’s lived body. As we saw
in connection with the intermediate reading of the Foucaultian body,
Dreyfus and Rabinow see Foucault’s background in phenomenology
as formative of his account of the body.29 They suggest that he appropriates Merleau-Ponty’s idea of the lived body but seeks to situate it
more radically in history and particular historical contexts (Dreyfus
and Rabinow 1982, 111–12). They also suggest that an elaborated idea
of a lived body would provide Foucault with a locus of resistance against
normalizing biopower because of the nascent logos of the body. The
29 Merleau-Ponty’s and Foucault’s conceptions of the body are more often contrasted
rather than seen as similar. David Levin (1991), for example, argues that, while Foucault
describes a passive body, moulded, even totally rebuilt by regimes of power, MerleauPonty’s account emphasizes the activity of the body in shaping our cultural environment.
According to Levin, the lived body thus shapes society while the Foucaultian body is
shaped by it. As I have shown, however, Foucault’s conception of the body as an object
of disciplinary manipulation put forward in DP is complemented by an understanding
of the body as experiencing sexual pleasure in HS. The contrast between the passive
Foucaultian body and the active, lived body can therefore not be upheld if Foucault’s
thought as a whole is taken into account.
an a r c hic bo d ies
body would function as a stable reference point for truth, and hence
provide a position from which to criticize the deployments of power
(167). Dreyfus and Rabinow write:
If the lived body is more than the result of the disciplinary technologies that have been brought to bear upon it, it would perhaps provide a
position from which to criticize these practices, and maybe even a way
to account for the tendency towards rationalization and the tendency of
this tendency to hide itself.
(Dreyfus and Rabinow 1982, 167)
While I share the starting point of Dreyfus and Rabinow’s argument of
relating Merleau-Ponty’s and Foucault’s accounts of the body, my argument in the next chapter proceeds in an almost diametrically opposite
direction. Rather than grounding the possibility of resistance on a foundationalist reading of Merleau-Ponty’s body-subject, which assumes the
body to have a nascent logos, I will construct a non-foundational reading
that is compatible with Foucault’s understanding of the body as historically constituted.30 By foundationalism I refer here to a philosophical
position that seeks in an analysis of the body’s structures a universal
and stable foundation for subjectivity and a reference point for truth.
I will show that the historicity of the body-subject in Merleau-Ponty’s
phenomenology of the body means that the body-subject is fundamentally constituted in history. Hence, the important idea in Merleau-Ponty
from a Foucaultian perspective is not the alleged foundationalism of
Merleau-Ponty’s lived body; the fact that the body has invariant and
universal structures. The important idea is that the body is generative
of an individual sexual style. Even when Foucault acknowledges the
importance of experiences, particularly in the realm of sexuality, his
understanding of experience remains curiously non-subjective and disembodied. He cannot elaborate on it through passive understanding
of the body-object, which featured in his previous writings, and perhaps
30 David Couzens Hoy (1999, 6–11) also argues against Dreyfus’ and Rabinow’s effort to
base resistance on the body’s universal and invariant structures. Hoy points out that
firstly, even if there are bodily invariants, they may be too thin to serve as the basis of
criticism and resistance. Secondly, Foucault’s genealogy is an effort to show that the way
of thinking that there is a normal, natural or universal way to exist would itself be a variant
of normalization. Instead of adopting the assumption of invariance, genealogy seeks to
show how the body has been lived differently historically, and how it can therefore
become ‘more’ than it is now. According to Hoy, critical resistance thus flows from the
realization that the present’s self-interpretation is only one among several others that
have been viable, and that it should keep itself open to alternative interpretations.
fo u c au lt o n freedom
therefore the later understanding of the experiential body that he introduces in connection with sexuality is not further elaborated other than
with a few remarks. I argue that a radically historized phenomenology of the body could therefore fill in some of the gaps in Foucault’s
understanding of bodies and pleasures as a form of resistance.
Butler writes in The Psychic Life of Power about Foucault’s understanding of the body: ‘Perhaps the body has come to substitute for the psyche
in Foucault – that is, as that which exceeds and confounds the injunctions of normalization’ (Butler 1997, 94). According to Butler, this
means that ‘Foucault has invested the body with psychic meanings that
he cannot elaborate within the terms that he uses’ (95). While Butler’s
conclusion seems to hit the point as far as what is lacking from Foucault’s conception of the body, I will argue that the only way to account
for these gaps is not to turn to ‘psychic meanings’ but instead to study
the meanings generated by the lived body.
To conclude, I have two aims in my effort to create an uneasy alliance
between Foucault and Merleau-Ponty in the next chapter. On the one
hand, I argue that phenomenological insights concerning the lived
body could enrich Foucault’s idea of the body as a locus of resistance.
On the other hand, I also want to argue that the feminist appropriations of Merleau-Ponty’s lived body can benefit from a ‘Foucaultian’
interpretation of it, because it can provide feminist theory with new ways
of understanding ‘female freedom’. Emancipation does not require a
body with nascent logos and pure of cultural inscription, but one that
is inscribed in ways that are open to reinterpretations and multiple
Iris Marion Young’s book Throwing like a Girl and Other Essays in Feminist
Philosophy and Social Theory represented one of the most notable efforts
to apply Merleau-Ponty’s phenomenology of the body to explicitly feminist issues.1 The essay ‘Throwing like a Girl’ traces some of the basic
modalities of feminine comportment, manner of moving and relation
in space. With the help of these modalities Young seeks to make understandable the ways in which women in our society typically comport
themselves and move differently from the ways in which men do. She
argues, with the help of Merleau-Ponty’s phenomenology of the body,
that the modalities of feminine comportment, motility and spatiality are
restricted modes of embodiment. According to Young, Merleau-Ponty
describes the lived body as a transcendence that moves out from the
body in its immanence in an open and unbroken directness upon the
world in action. The lived body as transcendence is pure fluid action,
the continuous calling forth of capacities that are applied to the world.
In the case of feminine movement the most primordial intentional act –
the motion of the body orienting itself with respect to and moving within
its surroundings – is inhibited (Young 1990,148). A woman ‘lives her
body as a thing, she remains rooted in immanence, is inhibited, and
retains a distance from her body as transcending movement and from
engagement in the world’s possibilities’ (150).
Jean Grimshaw criticizes Young’s analysis of female embodiment of
the problematic opposition of the repressed female body and the ‘free’
or unrepressed male body. In her view, Young idealizes masculine movement by assuming it unproblematically as a norm. Merleau-Ponty could
1 It can be argued that Simone de Beauvoir had already put forward a phenomenological
description of female embodiment in The Second Sex. See e.g. Kruks 1990, Vintges 1992,
Heinämaa 2003.
fo u c au lt o n freedom
be accused of giving ontological priority to certain kinds of immediate
bodily movements and actions. His analysis of embodiment fails adequately to recognize the ways in which the meaning of all actions, however immediate, is always mediated by culture (Grimshaw 1999, 103).
Consequently, rather than there being, on the one hand, normal, uninhibited or natural motility of men, and on the other a pathological,
culturally repressed motility of women, we should see both male and
female motility as culturally mediated modalities of movement, dependent on the cultural coding of the activities in question. According to
Grimshaw, the female body cannot be simply ‘freed’ from repressive
or inhibitive sexist oppression, nor can it be adequately understood
simply in terms of capitulation to ideological pressures to conform to
a particular norm of the feminine body (115).2 Grimshaw concludes
that what Young’s analysis suggests is that Merleau-Ponty’s conception
of the ‘normal’ needs problematizing: within the category of ‘normal’
there may be different ‘modalities’ that cannot be identified simply as
This point is also forcibly brought home by Judith Butler in her devastating criticism of Merleau-Ponty’s account of sexuality (Butler 1989).3
While Grimshaw’s and Young’s focus is on the motility of the body, Butler’s focus is on its sexuality. She claims that, although Merleau-Ponty’s
account of sexuality seems to offer significant arguments against all
naturalistic accounts by presenting sexuality as a mode of dramatizing
and investigating a concrete historical situation, it in fact contains tacit
normative assumptions about the heterosexual and male character of
sexuality. Not only does Merleau-Ponty fail to acknowledge the extent
to which sexuality is culturally constructed, but his descriptions of its
universal features reproduce certain cultural constructions of sexual
normalcy (Butler 1989, 92). Despite the alleged openness and cultural
malleability of sexuality advocated by Merleau-Ponty, Butler claims that
certain bodily structures emerge as existential and metaphysical necessities, and that this ultimately destroys Merleau-Ponty’s non-normative
pretensions (89).
I argue in what follows that underlying both Young’s adaptation
of Merleau-Ponty’s phenomenology of the body, and Grimshaw’s and
Butler’s criticism of it, is a foundationalist reading of Merleau-Ponty’s
2 See also Grosz 1994, 144.
3 For more recent criticism accusing Merleau-Ponty of bodily fundamentalism, see Sullivan
1997. For an alternative reading, see Silvia Stoller’s reply to Sullivan, Stoller 2000. For
criticisms of Butler’s reading of Merleau-Ponty, see Heinämaa 1997 and Waldenfels 1998.
fema le fr eed o m
body-subject. This criticism can also be conceptualized in terms of a
distinction between historical situatedness and historical constitution,
or what Judith Butler calls ‘a stronger version of historical situatedness’
(Butler 1989, 90–1).
Butler argues that the historicity of the subject in Merleau-Ponty’s
account of sexuality is historical situatedness, not historical constitution. This means that the body forms a universal foundation for subjectivity that only assumes different guises in different historical situations.
Rather than history constituting all bodily experiences and modalities,
there is a normal, foundational body that can assume different modalities depending on the varying historical situations. While Young thus
argues that the modalities of feminine bodily comportment and movement ‘have their source in the particular situation of women as conditioned by their sexist oppression in contemporary society’ (Young 1990,
153), Butler criticizes Merleau-Ponty’s account of sexuality by noting
that sexuality cannot be understood as an expression of a historically
situated subject’s existence.
Yet, to say that the subject is historically situated in a loose sense is to
say only that the decisions a subject makes are limited – not exclusively
constituted – by a given set of historical possibilities. A stronger version of
historical situatedness would locate the history as the very condition for
the constitution of the subject, not only as a set of external possibilities
for choice. If this stronger version were accepted . . . sexuality is itself
formed through the sedimentation of the history of sexuality, and the
embodied subject, rather than an existential constant, is itself partially
constituted by the legacy of sexual relations which constitute its situation.
(Butler 1989, 90–1)
I will argue, against foundationalist readings, that the historicity of the
body-subject in Merleau-Ponty’s phenomenology is not historical situatedness but historical constitution. Merleau-Ponty does not understand the body-subject as ‘an existential constant’. The structures of
the body do not form a universal foundation for all forms of subjectivity. Instead, subjectivity, even on the level of the anonymous body, is
always historically constituted. I will defend my argument by constructing two possible readings of Merleau-Ponty’s body-subject. The first puts
it forward as foundational and historically situated, and can be seen
to underlie Young’s, Grimshaw’s and Butler’s accounts. I will argue
against this by presenting a second reading which emphasizes transcendental intersubjectivity – language, tradition and community – as
fo u c au lt o n freedom
the reality-constituting principle. In this second reading, MerleauPonty’s understanding of the body-subject is seen as compatible rather
than opposed to post-structuralist approaches, for example Foucault:
the body-subject is fundamentally structured as intersubjectivity, and is
always historically constituted.
I will finish by showing how the two different readings of MerleauPonty’s body-subject result in two different views of the freedom of the
body. While the first one is open to criticism for naively assuming a
realm of ‘female freedom’ outside of sexist oppression, I will argue that
the second reading suggests new and interesting ways of thinking about
the freedom of the female body.
The anonymous subjectivity of the body
Merleau-Ponty’s Phenomenology of Perception presents a detailed description of the experience of the living body. It introduces the concept
of the lived body (corp propre, corps phénomenal) and seeks to study
the body of our lived experience, not of bio-scientific descriptions.
One of his central claims is that the human body is not an object, a
bio-mechanical system described by scientific materialism, but a precondition of all objects. It is our basic framework of meaning and
truth, both the source and the sedimentation of significations and
The term lived body is also used in an attempt to describe and understand the fundamental interrelatedness of consciousness and embodiment, and thus to bypass the dualisms of mind–body, interior–exterior
and consciousness–nature. In the phenomenological description the
lived body must be understood as a totality of external and internal perceptions, intelligence, affectivity, motility and sexuality. Merleau-Ponty
aims to show, on the one hand, how the mind is always based on corporeality while not reducible to computational or neurological processes,
and on the other hand, how the body’s physiological processes, such as
perceptions, are never purely mechanical but always incorporate values
and meanings. Using various case studies of pathological embodiment,
he brings to light the bodily abilities that a normal subject has and takes
for granted. We know, for example, how our body is positioned in relation to its environment without having to complete a cognitive process.
We can imitate movements without having first to translate those movements into verbal representations that the body follows or carries out.
In short, we have a ‘bodily knowledge’ of the world.
fema le fr eed o m
Merleau-Ponty uses the concepts of body schema and intentional arc
to further describe the nature of this bodily knowledge; the mode
and structure through which our body is intertwined with the world.
The notion of body schema is often mistakenly equated with imaginary
body.4 It is not, however, an image in the sense of being a mental representation of the body. It is not imaginary, and cannot be understood as
some kind of a ghost of the material body. It is the material body itself,
a structured capacity for actions and intentions. It describes the bond,
the intentional arc, that connects us to the world.
[T]he life of consciousness . . . is subtended by an ‘intentional arc’ which
projects around us our past, our future, our human setting, our physical,
ideological and moral situation, or rather which results in our being
situated in all these respects. It is this intentional arc which brings about
the unity of the senses, of intelligence, of sensibility and motility.
(Merleau-Ponty 1945/1994, 136)5
According to Merleau-Ponty, the mode or structure through which our
body is intertwined with the world is possible through a special kind of
intentionality, a prereflective, bodily intentionality. He adopts Husserl’s
term operative intentionality (fungierende Intentionalität) to describe intentionality that is not limited to thetic acts – acts of consciousness that
have the structure of act-object – but describes our basic bodily bond
to the surrounding world. This basic bond or relationship is essentially
non-thetic: it cannot be fully analyzed into acts and their objects, but is
more like an intertwining of our body and the world.6 Merleau-Ponty
writes: ‘We found beneath the intentionality of acts, or thetic intentionality, another kind which is the condition of the former’s possibility:
namely an operative intentionality already at work before any positing
or any judgement’ (Merleau-Ponty 1945/1994, 429).
Through operative intentionality the body is directed to the world, its
acts intend it and anticipate it. The body is ‘what opens me out upon the
world and places me in a situation there’ (Merleau-Ponty 1945/1994,
4 See e.g. Gatens 1996, 12, 69–70. Merleau-Ponty’s notion of body schema was strongly
influenced by psychoanalytical accounts of the body image, however, as well as by Paul
Schilder’s research in neurophysiology.
5 ‘la vie de la conscience . . . est sous-tendue par un “arc intentionnel” qui projette autour
de nous notre passé, notre avenir, notre milieu humain, notre situation physique, notre
situation idéologique, notre situation morale, ou plutôt qui fait que nous soyons situés
sous tous ces rapports. C’est cet arc intentionnel qui fait l’unité des sens, celle des sens et
de intelligence, celle de la sensibilité et de la motricité.’ (Merleau-Ponty 1945/1972, 158)
6 See Heinämaa 2002.
fo u c au lt o n freedom
165). Although Merleau-Ponty takes fundamental intentionality to be
intentionality of the body rather than of consciousness, this does not
mean that it is unconscious or mechanical. Bodily intentionality does
not have the structure of an act and object because this separation
of subject and object is accomplished on the level of individual consciousness. It is thus primary in the sense of being a precondition for
intentional consciousness. According to Merleau-Ponty, our relationship to the world is, in the first place ‘not a matter of “I think” but of
“I can”’ (Merleau-Ponty 1945/1994, 137).
Phenomenological account of sexuality. Basic bodily intentionality is also
what underlies Merleau-Ponty’s account of sexuality in Phenomenology of
Perception. Sexuality is an intending of the world and of the other body
in a way that precedes intellectual signification and the separation of
subject and object. It is an intertwining more basic than all conceptual
and theoretical formulations of it.
When I move my eye, I take account of the movement, without being
expressly conscious of the fact, and am thereby aware that the upheaval
caused in my field of vision is only apparent. Similarly sexuality, without being the object of any intended act of consciousness, can underlie
and guide specified forms of my experience. Taken in this way, as an
ambiguous atmosphere, sexuality is co-extensive with life.
(Merleau-Ponty 1945/1994, 169)7
Through a detailed discussion of Goldstein and Gelb’s case study of
a brain-damaged patient called Schneider, Merleau-Ponty argues that
sexuality is not a separable function or aspect of behaviour. Schneider
was a veteran of the First World War and was wounded in the occipital
region of the brain, which is generally believed to process visual data.
What is peculiar about his case is that there was no specific function that
was deficient. It was rather that his problems affected his whole way of
being: his way of moving, perceiving and his relation to the world. As far
as his sexuality was concerned, the breakdown of his sexual activity and
interests had internal links with the whole of his active and cognitive
7 ‘Quand je bouge les yeux, je tiens compte de leur mouvement, sans en prendre conscience expresse, et je comprends par lui que le bouleversement du champ visuel n’est
qu’apparent. De même la sexualité, sans être l’objet d’un acte de conscience exprès,
peut motiver les formes privilégiées de mon expérience. Prise ainsi, c’est-à-dire comme
atmosphère ambiquë, la sexualité est coextensive à la vie.’ (Merleau-Ponty 1945/1972,
fema le fr eed o m
being. According to Merleau-Ponty, the very structure of perception and
erotic experience underwent a change in Schneider (156). Sexuality
lost its meaning for him. In a normal subject, sexuality permeates the
whole living being and its situation. It is present in all acts and qualities,
‘like an odour or like a sound’ (168).
Although Merleau-Ponty’s account of sexuality is allied with psychoanalysis in emphasizing the fact that sexuality does not express
or present itself to us in determinate conscious acts, the opposite
pole of conscious acts is not put forward as the unconscious, but as
bodily anonymity, which also characterizes sexuality. Throughout Phenomenology of Perception, Merleau-Ponty presents us with examples of the
anonymity of perceptions. He argues that every perception takes place
in an atmosphere of generality and is presented to us anonymously.
I cannot say that I see the blue of the sky in the sense that I can say
that I decide to devote my life to mathematics. I can see blue because I
am sensitive to colours, whereas I am a mathematician because I have
decided to be one (Merleau-Ponty 1945/1994, 215). In connection
with sexuality, Merleau-Ponty takes the example of falling asleep. It is
not an act of positing consciousness where the body simply expresses
a conscious intention. It is a bodily act or an act realized only through
the body.
I lie down in bed, on my left side, with my knees drawn up: I close my
eyes and breathe slowly, putting my plans out of mind. But the power of
my will or consciousness stops there . . . There is a moment when sleep
‘comes’, settling on this imitation of itself which I have been offering to
it, and I succeed in becoming what I was trying to be.
(Merleau-Ponty 1945/1994, 163–4)8
Tacit cogito. In the chapter entitled ‘The Cogito’ in Phenomenology of
Perception, Merleau-Ponty presents his account of subjectivity – conveyed by his notion of body-subject – through a critique of the Cartesian cogito. There can be no subjectivity separable from our bodies
because all consciousness is founded on perceptual consciousness; even
the pure ideas of intellect derive their meaning from the world of
8 ‘je m’étends dans mon lit, sur le côté gauche, les genoux repliés, je ferme les yeux, je
respire lentement, j’éloigne de moi mes projets. Mais le pouvoir de ma volonté ou de
ma conscience s’arrête là . . . Il y a un moment où le sommeil “vient”, il se pose sur cette
imitation de lui-même que je lui proposais, je réussis à devenir ce que je feignais d’être’
(Merleau-Ponty 1945/1972, 191).
fo u c au lt o n freedom
perception. Similarly, self-consciousness arises through direct contact
with the world perceived not by consciousness observing itself perceiving, or by inference from any idea of itself. Consciousness is not closed
in on itself, but is always a relation to the world, a form of conduct.
All knowledge of the world is, at the same time, knowledge of the self;
‘it is through my relation to “things” that I know myself’ (MerleauPonty 1945/1994, 383). At the same time, Merleau-Ponty writes that
‘there is an element of final truth in the Cartesian return of things or
ideas to the self’ (369). The perceiving subject is a precondition of the
experience of a transcendent world. The world only becomes meaningful through the perceptual acts of a transcendent subject. However,
the cogito that Descartes should have discovered at the root of our
experiences of the world is not the cogito of explicit experiences. In
the place of the Cartesian cogito, Merleau-Ponty postulates a tacit cogito that is anonymous or prepersonal. This tacit cogito underlies or
precedes the emergence of explicit self-consciousness, which comes
into being through language. Martin Dillon characterizes the tacit cogito by writing that its corporeal reflexivity is latent and unexpressed,
whereas the Cartesian cogito is personal and individual and its explicit
reflexivity is thematized in language and thought. The tacit cogito is
thus subjectivity that is yet curiously prereflective and unaware of itself
(Dillon 1988, 105–8). It is bodily consciousness, an intertwining with
the world. The structures of the world are structures of the body, not
of the consciousness.
[W]hen I reflect on the essence of subjectivity, I find it bound up with
that of the body and that of the world, this is because my existence as
subjectivity is merely one with my existence as body and with the existence
of the world, and because the subject that I am, when taken concretely,
is inseparable from this body and this world.
(Merleau-Ponty 1945/1994, 408)9
Hence, Merleau-Ponty clearly claims that the tacit cogito, or the anonymous body-subject, is foundational in relation to individual, personal
subjectivity. It forms the latter’s condition of possibility. What is not so
clear in his account of subjectivity, however, is the relationship between
9 ‘Si, réfléchissant sur l’essence de la subjectivité, je la trouve liée à celle du corps et à
celle du monde, c’est que mon existence comme subjectivité ne fait qu’un avec mon
existence comme corps et avec l’existence du monde et que finalement le sujet que je
suis, concrètement pris, est inséparable de ce corps-ci et de ce monde-ci.’ (Merleau-Ponty
1945/72, 467)
fema le fr eed o m
the anonymous body and intersubjectivity. Although he considers
subjectivity as fundamentally anonymous and general, it is tied to a
singular body: ‘It is essential for me not only to have a body, but to
have this body’ (431). My existence is tied to my singular experience
of myself as a body as well as to my experience of the world. This is
important because the prereflectivity or anonymity of perceptions does
not mean coincidence with the perceived. As Merleau-Ponty writes, ‘We
do not mean that the primordial I completely overlooks itself. If it did,
it would indeed be a thing, and nothing could cause it subsequently to
become consciousness’ (404). Self-awareness, even though tacit, thus
becomes a precondition for the experience of a transcendent world.
There has to be a space between the here of perception and the there
of the phenomenon, and there has to be some kind of awareness of the
here for the there to appear as such (Dillon 1988, 103).10
Although the tacit cogito or body-subject does not refer to personal
subjectivity, but should rather be understood as an anonymous and preconscious layer of subjectivity, it must remain centralized or singularized
in such a way that the acts and their object do not coincide. The perceiving body becomes constituted through its perceptions as a subject
endowed with tacit reflexivity. This reflexivity or tacit self-consciousness
becomes, on the one hand, a precondition for perception, while perceptions, on the other hand, constitute subjectivity as incarnate in the
body (Dillon 1988, 105.) The anonymity of perception, rather than
breaking or decentralizing subjectivity, in fact seems to presuppose not
undifferentiated anonymous life but rather a subjectivity that is singularized in such a way that it is equated with an individual perceiving
Foundational account of the body-subject. How, then, is the body-subject
related to intersubjectivity: is it foundational in the sense of forming
the latter’s condition of possibility? I will argue that Merleau-Ponty’s
understanding of subjectivity in Phenomenology of Perception allows for two
possible readings. The first is that we distinguish three separate ‘layers’
of subjectivity: (1) anonymous, prepersonal subjectivity; (2) personal,
individual subjectivity; and (3) interpersonal intersubjectivity. Thus,
anonymous, prepersonal subjectivity is understood as foundational for
10 Dan Zahavi also argues that, for an experience to be anonymous means that it lacks
explicit self-awareness; it does not mean that it lacks self-awareness, differentiation and
individuation altogether. According to Zahavi, Merleau-Ponty occasionally flirts with this
radical interpretation, but on closer scrutiny, this must be rejected. See Zahavi 2001.
fo u c au lt o n freedom
the other two. It forms the condition of possibility, not only for individual consciousness but also for intersubjectivity understood as comprising the linguistic community, culture and history. In this reading,
Merleau-Ponty presents a foundational account of subjectivity. There
is a rudimentary level, the perceptual flow of the singular subject, on
which all forms of subjectivity are founded.
Even though, according to this reading, the anonymous existence
of the body forms a universal foundation, the subject is still not an
ahistorical constant. This is because the body-subject is, for MerleauPonty, always and by necessity historically situated and circumscribed.
The phenomenological account of the lived body shows that it is always
situated within or intertwined with its environment. It actively takes
up its situation in the world and transforms it through its bodily acts,
attitudes or styles. This activity of the body is, moreover, normatively
generative: the body has optimal ways of acting in the world. Normality
for the lived body can, thus, according to this reading, be understood
as what is optimal for it. Optima are instituted within experience by the
very fact that the body takes a perspective on things and is embedded in
the surrounding world. As bodies we can be more open to the givenness
of objects or more closed to them.11 Following Husserl, Merleau-Ponty
refers to normal as optimal when he writes that ‘for each object, as for
each painting in a gallery, there is an optimal distance from which it
demands to be viewed, an orientation through which it gives the most
of itself . . . the distance from me to the object is not a size which
decreases or increases, but a tension that oscillates around a norm’
(Merleau-Ponty 1945/1994, 302).
In this reading, phenomenology of the lived body can be criticized
for approaching the notions of normal and abnormal with respect to the
lived body and its immediate surroundings, not with respect to an interpersonal community. What characterizes the living body is its ability to
instigate norms, and norms are founded on the experience of the lived
body. This understanding of the foundational role of the structures of
the body in establishing the normal seems to be in stark contrast to
the feminist theorists who claim that normal and abnormal are always
defined in a social context and attached to the polarity of positive–
negative. Norms offer possibilities for reference and judgement; instituting and identifying norms are always acts of power.
11 See Steinbock 1995, 138–43. Steinbock argues that, when Husserl refers to normal as
optimal, the optimal as a norm is instituted and generated from within experience.
fema le fr eed o m
Hence, according to my first reading, while emphasizing that the subject is always historically situated, Merleau-Ponty does not problematize
how the ‘normal subject’ itself is fundamentally constituted in history.
As Elmar Holenstein (1985/1999, 87) argues, he neglects to question
both the possible historical and sociological dependence of the structuring and orientation of the world as well as of the body. Merleau-Ponty’s
account of the body-subject would reject the strong version of historical constitution, according to which intersubjectivity – understood as
language, tradition and community – provides the very condition of
possibility for individual subjectivity as well as for objective reality. The
historicity of the subject in Merleau-Ponty’s phenomenology of the body
would, as Judith Butler claims, be historical situatedness, not historical
constitution. History as constitutive of the body-subject, even on the
level of rudimentary perceptual flow, would be redundant.
The historical constitution of the body
I will leave aside here the methodological questions connected with
studying intersubjectivity – language, history, cultural normativity – that
arise out of Merleau-Ponty’s phenomenology of the body.12 I will argue
that, despite possible methodological problems, his understanding of
the body-subject is open to a second reading, which is more in line with
Foucault’s insights about the historical constitution of the body. According to this second reading, transcendental intersubjectivity – language,
tradition and community – is understood as the reality-constituting principle providing the conditions of possibility for all forms of subjectivity
12 As I have made clear in my discussion of phenomenology and Foucault in chapters 1
and 2, an obvious difference between Merleau-Ponty and Foucault concerns their
methodologies. Although Merleau-Ponty emphasizes the fact that the constitution of
subjectivity is fundamentally dependent on history, language and society, he applies the
phenomenological method. Through phenomenological reduction one can reveal not
only the eidetic structures of a singular subjectivity, but also intersubjectivity, understood
as tradition and society. For a phenomenologist, intersubjectivity is not an objectively
existing structure in the world which could be studied from a third-person perspective.
Rather, it can only be disclosed through a description of the living body’s structures of
experience. Intersubjectivity can thus unfold itself only in the relation between singular
subjects. See e.g. Zahavi 1996, 237. Unlike Merleau-Ponty, Foucault does not analyze the
lived meanings of a singular body as constitutive of the meaningfulness of the world. He
analyzes sense constitution only by studying its conditions of possibility in the intersubjective or socially shared meaning structures, such as language. A concrete description of
the historical development of forms of rationality, technologies of subjection and techniques of the self can reveal the different modes through which forms of subjectivity are
fo u c au lt o n freedom
as well as for objective reality. Instead of the tacit cogito being a foundational layer of subjectivity on which the personal and intersubjective
depend, I will argue that it is a dimension of intersubjective sense constitution. It is not a foundation, but a constitutive condition.
Merleau-Ponty clearly emphasizes the reciprocity of all constitutive
processes. For instance, subjectivity and the world can never be understood in isolation from each other. He also explicitly states that transcendental subjectivity is transcendental intersubjectivity (Merleau-Ponty
1945/1994, 361–2). How transcendental intersubjectivity is understood is, however, decisive for the way we understand its relationship to
the body-subject.
Dan Zahavi, among others, effectively argues for an intersubjective
transformation of Husserl’s phenomenology in his late, posthumously
published writings.13 These texts by Husserl had the greatest influence
on Merleau-Ponty, who saw the main thrust of Husserl’s work to be
contained in the manuscripts.14 Zahavi shows that, from the winter
of 1910/11 up until his death in 1938, Husserl’s aim was to develop
a transcendental theory of intersubjectivity. According to Zahavi, what
has made Husserl’s account difficult to explicate and understand is that
he operated with several different kinds of intersubjectivity. He did not
only understand it to refer to the subject’s cultural context, to the fact
that we are constantly confronted with intersubjective meanings such as
social institutions and cultural products. Neither does intersubjectivity
refer exclusively to other people’s actual presence in the subject’s field
of experience. The core in Husserl’s reflections on intersubjectivity lies
in its fundamental reality-constitutive function.
In terms of understanding Merleau-Ponty’s body-subject as intersubjectivity, Husserl’s major claim is that the experience of objective validity is made possible by the experience of the transcendence of foreign
subjectivity. Objects cannot be reduced to being merely my intentional
correlates if they can be experienced by others. Our primal experience
of others permanently transforms our categories of experience. The
objective validity of my experiences does not, after the initial encounter,
require the other’s actual presence. The precondition for objective reality is, however, that it can only be constituted by a subject that has
13 See e.g. Zahavi 1996, 1996/2001.
14 Zahavi refers to, for example, Ideen II, Erste Philosophie II, Erfahrung und Urteil, Analyzen zur
passive Synthesis, Zur Phänomenologie der Intersubjektivität I–III. See also e.g. Merleau-Ponty
1945/1994, 92–3; 1960/1964.
fema le fr eed o m
experienced other subjects (Zahavi 1996, 233).15 Just as for Husserl,
transcendental intersubjectivity forms the condition of possibility for
our experiences of objective reality, similarly for Merleau-Ponty, other
people function as a precondition for the objectivity of perceptions. In
Phenomenology of Perception, Merleau-Ponty discusses in detail the phenomenon of hallucinations. The fact that I classify the voices and visions
of my interlocutor as hallucinations means that I find nothing similar
in my visual field. The world that I perceive as objective reality is not
only a correlate of my consciousness. An object appears to me only
from one possible angle, but the other possible angles are implied in
my perception of it. For something to be an object it must thus carry
with it the potentiality of being perceived simultaneously from multiple angles. Intersubjectivity in Merleau-Ponty’s thought refers not only
to the subject’s shared cultural context, but also to the fundamental
condition of possibility of objective reality. Intersubjectivity forms the
condition of possibility for perception.
Even if the role of intersubjectivity in the constitution of objective
reality is understood as an intertwining of the perceiving body with
other people’s bodies and the natural world, the body-subject can still
be understood as foundational. The constitutive process takes place in
a threefold structure – subjectivity–intersubjectivity–world – and this
structure is essential and universal. Intersubjectivity can, however, also
be understood as comprising the linguistic community and historical
tradition. Zahavi argues that this sense is present in Husserl’s late thinking. Intersubjectivity is now understood as the linguistic community that
forms the fundamental condition of possibility for singular subjectivity.
Zahavi argues that the question of normality and the concept of the
home-world become constitutional core concepts for Husserl. Homeworld refers to the normatively significant life-world, with its unique
language and tradition. It is the familiar life-world that is normatively
15 Zahavi further argues that it would be a mistake to assume that Husserl understood
intersubjectivity as something which is exclusively attached to concrete bodily mediated
interaction. He held a more fundamental view: the being of the subject as experiencing
and constituting implies a reference to other subjects already prior to its concrete experience of them, that is, a priori. Husserl argues that an analysis of perceptual intentionality
leads to a disclosure of the apodictic intersubjective structure of the transcendental ego.
Our perceptions of objects are characterized by the horizontal appearance of the object,
where a certain aspect is present and others are absent. All possible aspects can thus
never be actualized by a single subject and therefore can only be accounted for through
reference to a plurality of possible subjects. The being of the perceiving subject is thus
referred to others already by the fact of the ontological structure of the object (Zahavi
1996, 233–4).
fo u c au lt o n freedom
relevant to us. It is the world that our body intends and spins around us.
The subject’s embeddedness in this living tradition and its anonymous
normality forms a third type of intersubjectivity. Normality is understood as conventionality, which in its being transcends the individual.
Our horizon of anticipations is structured in accordance with the intersubjectively handed-down forms of apperception (Zahavi 1996, 239–
42). Social normativity cannot therefore be regarded only as secondary
or derivative of individual, lived normativity.16
When intersubjectivity is understood as social normativity, the bodysubject cannot be understood as historically situated, but rather as historically constituted. According to this view, there can be no universal
or inherent normativity of the living body. The anonymous body is
not foundational for social normativity, but the relationship between
the living body and the surrounding culture is complex and chiasmic.
The structures of the body are structures of the world, but not only of the
natural world. The shared normativity of the living tradition also constitutes and structures the intentionality of perceptions, sexuality and
Merleau-Ponty does not defend the view that posits the body as
immune to the influence of history. According to him, ‘Man is a historical idea, not a natural species’ (Merleau-Ponty 1945/1994, 170).
This claim is often interpreted to mean that the fundamental structures of the anonymous body – for example temporality, spatiality and
sexuality – form a foundation that simply assumes different guises in
different historical situations. Merleau-Ponty’s emphasis on history can,
however, be interpreted through Husserl’s theory of transcendental
intersubjectivity as defending a stronger version of historical constitution. The structures of the anonymous body come into being only
as intersubjectively generated. Merleau-Ponty’s conditions of possibility for perception must not be understood as ahistorical or universal
forms, but rather as dynamic and developing structures derived from
our cultural environment, constantly in a state of change. The anonymous body is not a natural foundation on which intersubjectivity, understood as tradition and community, forms a secondary layer. Nor is it the
16 Anthony Steinbock (1995, 267) argues that, when Husserl turned to generative phenomena, he no longer addressed the problem of normality and abnormality in terms of
normal as optimal for the living being and abnormal as one-sidedly dependent on the
normal, but rather treated them intersubjectively in terms of ‘homeworld’ and ‘alienworld’. By doing this, Husserl implicitly reinterpreted the concepts of normality and
fema le fr eed o m
case that, in emphasizing transcendental intersubjectivity as constitutive of the body-subject, culture is privileged over nature. It is rather
that transcendental intersubjectivity represents the very effort to dismantle the nature–culture dichotomy and to rethink nature as well as
culture. Merleau-Ponty’s aim throughout Phenomenology of Perception is
to argue against all dichotomous and causal modes of thinking, which
reduce lived phenomena to primary causes and secondary effects.
The phenomenological conception of intersubjectivity comprises
areas that are traditionally posited not only on the side of culture
but also on the side of nature: geography, physiology, materiality, the
body. The fundamental bodily nature of subjectivity that Merleau-Ponty
emphasizes with his notion of the body-subject is not part of nature
or culture, but it problematizes the distinction between them. The
anonymity of perception must be understood as a layer or dimension of
intersubjectivity that we cannot localize or isolate.17 Our relationship
to the world is fundamentally ambiguous and therefore resists conceptualization. This does not mean that there exists a sphere of ‘pure’
bodily experiences independent of intersubjectively constituted structures, meanings and linguistic representations of the body.
In his essay The Child’s Relations with Others, Merleau-Ponty studies
intersubjectivity in very concrete terms (Merleau-Ponty 1964, 96–155).
He discusses the development of a child’s corporal schema and shows,
through a detailed study of child psychology, how intersubjectivity forms
the condition of possibility for singular subjectivity. Merleau-Ponty
gives a detailed account of how a child’s bodily awareness, as well as
its perceptual consciousness, develop as a consequence of being in
an intersubjective situation, and how these therefore correspond to
cultural variations in the child’s environment. The development of a
child’s corporal schema is tied internally to the process that leads to
the distinction between itself and others. It develops only in a concrete
social, historical and cultural situation. The corporal schema is not an
a priori form that the child receives intact at birth, but is historically
constituted and structured as intersubjectivity. The structures of the
anonymous body come into being only as historically sedimented structures derived from our cultural environment. Subjectivity, even on the
level of the anonymous body, is always dynamic. It is constituted and
structured by language, community and culture. Our perceptions of the
17 Cf. Visker 1995, 120.
fo u c au lt o n freedom
world form an organic whole, not because of a universal foundation in
the body, but because of opening on to the same world.
Female freedom?
When Merleau-Ponty’s body-subject is not understood as universal or
foundational, but as essentially dynamic and historically constituted,
the implication for feminist theory is that there are no normal or foundational modes of female embodiment, motility or sexuality. There is no
inhibited female corporeality and free and normal male corporeality
in societies of sexist oppression, but rather two differently gendered
and historically constituted experiences and modalities of embodiment. What is called normal depends on the values of the society in
This view comes close to Foucault’s idea that power/knowledge networks constitute normalcy. According to Foucault, modernity is characterized by life becoming an object of scientific discourse intrinsically
tied to political aims and technologies. Biopower targets individual bodies and the population’s health as a whole. An important consequence
of its development is the growing importance assumed by the action
of the norm, at the expense of the juridical system of the law. Unlike
laws which function according to the binary logic of the forbidden
and the permitted, norms are individualizing: they make it possible
to demarcate distributions, measure differences, construct scales and
classify in various categories. Biopower is dependent on this individualizing knowledge about particular bodies, and about the population as a
whole. It is power for normalizing judgement, power to identify scientific
criteria for what is normal.
According to Foucault, norms are thus an important part of the
power/knowledge network, and as such constitutive of the subject. Scientific discourse creates norms that are utilized by political discourse
and institutional practices and vice versa: political problems are taken
up by scientific discourse and its experts, on whose authority the normal is identified. Structures of power/knowledge create not only new
objects of science, but also new kinds of subjects.
Foucault’s studies of ‘dividing practices’ show how the ‘normal’ subject is constituted by a distinction and physical separation between normal and abnormal subjects. Scientific normativity and its third-person
accounts contribute to the constitution of our lived bodily sense of the
normal. They also shape the liminal encounters of the home-world and
fema le fr eed o m
the alien-world, the normal and the abnormal. As Steinbock argues
(1995, 180), for Foucault it was not a matter of inquiring into how we
circumscribe a prefabricated alienness, but of understanding how alienness itself is constituted. His analyses show, for example, how the scientific status of psychiatric knowledge disqualifies the patient’s knowledge
and thereby makes his life-world an alien-world, while constituting the
doctor’s world as the home-world. Foucault’s thought thus effectively
politicizes the level of intersubjective sense constitution by bringing in
the idea of power being constitutive of forms of the subject.
Merleau-Ponty’s phenomenological descriptions of the lived body,
on the other hand, show how the body becomes an instigator of normative as well as transgressive reiteration. The body is not a surface or a site
on which psychic meanings are played out. Neither is it a mute container
of subjectivity. The body-subject is generative in the sense of being constitutive of different meanings: subjectivity means the embodied capacity to creatively respond to existing norms. As Rudi Visker notes, style
for Merleau-Ponty is the moment of singularity (Visker 1995, 119).18
Although perception is fundamentally intersubjective and anonymous,
it is also an expression; it creates a singular style. It actualizes a sexual
style that the subject lives through all its engagements with the surrounding world. Instead of viewing perception as a process whereby
the hitherto meaningless takes on meaning through the foundational
structures of the body, we must understand it as an essentially open
and ambiguous process. The sensory meanings that are ordered and
constituted intersubjectively through our coexistence with the world
take several forms. Perception therefore always remains indeterminate
and incomplete.19 The lived body is characterized by a fundamental
indeterminacy because its relationship to the world is essentially open
and dynamic.
The constant responding of bodily acts to existing cultural meanings
is what interlocks the lived normativity of a singular body and social
normativity. The relationship between subjective, bodily normativity
and the intersubjective horizon of meanings is dynamically interlocked
18 In his article relating Merleau-Ponty’s and Foucault’s thought on the question of
experience/discourse, Rudi Visker also argues that we do not have to choose between
them. ‘What is at stake here is not the attempt to reduce discourse to existence, or
existence to discourse, but to find in their mutual intrication some indication of what
it could mean for us to be those subjects who take up positions we did not ourselves
generate’ (see Visker 1995, 126).
19 According to Merleau-Ponty, sensation can be anonymous only because it is incomplete
(Merleau-Ponty 1954/1994, 216).
fo u c au lt o n freedom
in constant oscillation: shifting, resisting and adapting. While individual bodily normativity responds to established sets of norms – female
embodiment is constituted in a certain way in a patriarchal culture –
it never mechanically reiterates the existing norms. The body is constantly materializing different social norms, it reiterates them but always
through its individual style. It is not a replica or a carbon copy of preestablished normativity, but rather materializes an individual style of
being. The constitution of meaning, even in the singular living body,
is always intersubjective, but never mechanical. The body-subject is initiatory and capable of resistance, and at the same time constituted by
intersubjective normativity. According to Merleau-Ponty, ‘The question
is always how I can be open to phenomena which transcend me, and
which nevertheless exist only to the extent that I take them up and live
them’ (Merleau-Ponty 1945/1994, 363).
Hence, my argument is that a non-foundationalist reading of
Merleau-Ponty’s body-subject can provide feminist theory with an
account of the female body that acknowledges its generative status
instead of viewing it only as a passive product of cultural crafting. At
the same time, Merleau-Ponty’s thought refutes the possibility of feminist theory returning to a fixed or pure female embodiment or essential femininity.20 Because the body-subject is always historical as well
as generative, the emergence of new ‘sexual styles’, new sets of bodily normativities, constantly shifts the meaning of sexual difference.
The intersubjective horizon of meaning is transformed because what
the lived body generates is unpredictable. As Elizabeth Grosz writes
in Volatile Bodies, what fascinates her is the ability of bodies to always
extend the frameworks that attempt to contain them and seep beyond
their domains of control. ‘Bodies are not inert; they function interactively and productively. They act and react. They generate what is new,
surprising, unpredictable’ (Grosz 1994, xi).
The lived body cannot be emancipated from sexist oppression to
free modes of motility or sexuality. Nevertheless, Merleau-Ponty’s phenomenology of the body potentially furthers understanding not only of
the cultural constitution of the body, but also of its resistance against this
constitution. There is ‘freedom’ in the unpredictability of our embodied experiences that establishes the always incomplete character of the
body’s cultural constitution. This freedom is not to be understood as an
inherent capacity or an attribute of the body as such, but is more like
20 On the possibility of ‘feminist phenomenology’, see also Oksala 2004.
fema le fr eed o m
a ‘Foucaultian’ understanding of freedom as the freeing or opening of
new possibilities for living our bodies, sexualities and lives.
This freedom of the body is thus not political freedom. The fact that
the body will offer resistance to sexist forms of power does not mean
that we can give up political struggles for feminist issues connected
with the body – such as abortion or rape – and simply let the body do
the job for us. Even if the cultural constitution of the body is never
complete or uniform, the resistance it shows is never enough because
it cannot rearticulate the terms of the body’s cultural constitution.21
The rearticulation of the intersubjective horizon of meanings cannot
be accomplished by creative bodies alone, but only by creative politics.
Nevertheless, even if the female body cannot be emancipated through
Merleau-Ponty’s phenomenology in this sense, our ways of inscribing
and reading the body have important political consequences, particularly in the realm of sexual politics. Perhaps it is necessary for us to
rethink emancipation: it requires not a body pure of cultural constitution, but one that is constituted in ways that are open to reinterpretation and multiple meanings. The undefined freedom of the lived body
opens up a space in which political freedom can be sought.
21 Cf. Butler 1997, 89.
My whole tendency and I believe the tendency of all men who
ever tried to write or talk on Ethics or Religion was to run
against the boundaries of language. This running against the
walls of our cage is perfectly, absolutely hopeless. Ethics so
far as it springs from the desire to say something about the
meaning of life, the absolute good, the absolute valuable, can
have no science. What it says does not add to our knowledge
in any sense. But it is a document of a tendency in the human
mind which I personally cannot help respecting deeply and
would not for my life ridicule it.
(Wittgenstein 1965, 11–12)
Foucault never developed a theory of ethics, yet his two last books, The
History of Sexuality, volumes ii and iii, could be characterized as being
concerned primarily with ethics. They deal with the sexual morality
of ancient Greece and the Roman Empire. The question that guides
Foucault’s inquiry is: ‘How, why and in what form was sexuality constituted as a moral domain?’ (UP, 10). The focus of the inquiry is thus
on the manner in which sexual activity was problematized, mainly by
philosophers and doctors in texts written as guides for others. In the
second volume of The History of Sexuality, The Use of Pleasures, the period
under study is the classical Greek culture of the fourth century bc. The
third volume, The Care of the Self, deals with the same problematization
in the Roman Empire of the first two centuries ad.
History of ethics
What emerges out of Foucault’s historical studies of sexual morality is a
particular conception of ethics that he traces to antiquity. He begins by
fo u c au lt o n freedom
making a distinction between morality as a moral code and the morality
of behaviours. The former refers to ‘a set of values and rules of action
that are recommended to individuals through the intermediary of various prescriptive agencies such as the family, educational institutions,
churches, etc’ (UP, 25). By morality of behaviours, he refers to the
effective behaviour of people in relation to the code; how their actual
behaviour matches the rules and values that are recommended to them.
These components of sexual morality are studied through the history
of morals and the social history of sexual practices, respectively. They
are not the objects of Foucault’s ‘history of desiring man’ (UP, 6).
Apart from these components of morality there is, according to
Foucault, still one important component, which he calls ethics. He
argues that for an action to be moral it is not enough to reduce it
to an act that conforms to a moral rule. Ethics refers to the manner in
which one forms oneself as a subject of a morality acting in reference
to its prescriptive elements; the modes by which subjects problematize their activity and use of pleasures. It is necessary to undertake a
study of ethics to be able to make visible the difference between the
morality of antiquity and that of Christianity. Foucault argues that, contrary to what is often believed, on the level moral ideals, structures and
codes of behaviour, there are striking similarities between antiquity and
Christianity. Both express concern, even fear, about the effect of sexual
expenditure on an individual’s health, both value conjugal fidelity and
abstention, and both attach a negative image to homosexual relations.
What constitutes a strong contrast between these two cultures, however,
is the way in which they integrate moral ideals or demands in relation to
the self and thus the forms in which sexual behaviour is problematized.
In ancient Greece the themes of sexual austerity were not an expression
of deep or essential prohibition, but ‘the elaboration and stylization of
an activity in the exercise of its power and the practice of its liberty’
(UP, 23).
Hence, the important discovery behind Foucault’s ‘history of ethics’
is that ethics understood as a certain kind of relation to oneself differs
from one morality to the other, just as do systems of values, rules and
interdictions.1 Only by keeping this component of morality in view can
the difference between Christian morality versus ancient morality be
1 Arnold Davidson (1986, 230–1) argues that Foucault’s understanding of ethics is an
original contribution to the study of morality because it shows how a study of ethics can
be fruitful even when there is little or no change in the moral codes examined. By isolating
the relation to oneself as a separate component of morality, Foucault opens up a domain
t he silen c e o f eth i c s
made clear. The main emphasis in Christian morality is on the code,
its systemacity, its richness, its capacity to adjust to every possible case
and to embrace every area of behaviour. The morality of antiquity, on
the other hand, represents a morality in which the code and rules of
behaviour are rudimentary. The emphasis is on the relationship with
the self and the practices of the self rather than on conformity to a
law. More important than the actual rules or contents of the law is the
relationship that one has with the self the choice about the style of
existence made by the individual (UP, 29–30).
Foucault further distinguishes between four different aspects of
ethics as the relation to oneself. The first he calls the ethical substance:
the part of the subject constituted as the material that is going to be
worked over. According to Foucault, for us the ethical substance would
generally be feelings, or in the sexual realm, the desire, whereas for
Kant, for example, it was intentions. Secondly, there is the mode of
subjection; the way people are invited to recognize their moral obligations. One could, for example, practise conjugal fidelity because one
acknowledges oneself to be part of a group that practises it, or because
one wants to offer one’s self as an example for others to follow. Thirdly,
there is what Foucault calls the ethical work, the self-forming activity –
the actual means by which one attempts to change oneself in order to
become a moral subject. This may take the form of a sudden renunciation of pleasures, self-interrogation, meditation, listening to others
or the memorization of scriptures, for example. Finally there is the telos:
the mode of being characteristic of the ethical subject one is aspiring
to become. One’s ultimate aim may be immortality, self-mastery, purity,
happiness, freedom (UP, 26–8).
In The Use of Pleasure Foucault analyzes the four aspects of the relation
to oneself in the sexual morality of ancient Greece through three austerity themes of the code: health, wives or women, and boys. The aim or
telos of this morality was the beauty of life through self-control: superiority over appetites and pleasures and mastery in relation to one’s body,
one’s household and society. The third part of The History of Sexuality,
The Care for the Self, takes up the same themes and problematizations in
the Roman Empire, and analyzes the gradual changes that had taken
place in the realm of ethics.
of analysis that can be profitably investigated both when moral codes are relatively static
and when they undergo great upheaval. His ethics provides us with a way of writing a
history of ethics that will not collapse into a history of moral codes.
fo u c au lt o n freedom
Ethics as practice
The History of Sexuality, volumes ii and iii, thus presents us with a historical study of the forms of an ethical problematization of a remote past.
Ethics refers to a specific component of morality and provides a useful
analytical tool for studying its history. Foucault’s work on ethics should
not be read solely as a new methodological approach to historical studies of sexual morality, however. His notion of ethics refers not only to
a component of morality that deals with the ways individuals constitute
themselves as moral subjects, but also to a certain way of understanding
morality. I argue that when his last books are combined with his late
interviews and other texts, an idea of ethics in the prescriptive sense of
the word emerges too; a conception of ethics as an individual ethos, an
attitude or a way of life. Foucault’s late work on ethics represents a continuation of his on-going concern with forms of subjection, and makes a
contribution to the task of rethinking ethics in the ‘postmodern’ world.
Foucault explicitly admitted that he wrote the last two volumes of
The History of Sexuality in terms of a contemporary situation (CT, 263).
He denied, however, that he was suggesting that we adopt the ethics
of ancient Greece. He condemns outright the ancient Greek ethics of
pleasure in many ways as something quite disgusting, and refers to how
it was linked to the ideas of a virile society, to dissymmetry, exclusion of
the other and an obsession with penetration, for example (GE, 346).
Yet, he suggests there is something we can learn from it.
My idea is that it’s not necessary to relate ethical problems to scientific
knowledge. Among the cultural inventions of mankind there is a treasury
of devices, techniques, ideas, procedures, and so on, that cannot exactly
be reactivated but at least constitute, or help to constitute, a certain point
of view which can be useful as a tool or analyzing what is going on now –
and to change it.
(GE, 349–350)2
In ancient Greece morality was not related to religion or religious preoccupations, nor was it related to social, legal or institutional systems.
Its domain was the relationship one had towards the self, namely, the
aesthetics of existence. What Foucault found striking was the similarity
of the ethical problems with the problems of our society:
2 The interview is the result of a series of working sessions with Michel Foucault conducted
by Paul Rabinow and Hubert Dreyfus at Berkley in 1983, and was originally published in
t he silen c e o f eth i c s
I wonder if our problem is not, in a way, similar to this one, since most of us
no longer believe that ethics is founded in religion, nor do we want a legal
system to intervene in our moral, personal, private life. Recent liberation
movements suffer from the fact that they cannot find any principle on
which to base the elaboration of a new ethics. They need an ethics, but
they cannot find any other ethics than an ethics founded on so-called
scientific knowledge.
(GE, 343)
Ancient Greek morality was not a system of rules or codes of conduct,
but an ethos; it was the subject’s mode of being and a certain manner of acting visible to others. Moral actions were further indissociable
from forms of self-activity, a set of practices relating to the principle
of epimeleia heautou, of taking care of oneself. Foucault writes that the
expression epimeleia heautou was a very powerful one in Greek: ‘It does
not mean simply being interested in oneself, nor does it mean having
a certain tendency to self-attachment or self-fascination . . . it describes
a sort of work, an activity; it implies attention, knowledge, technique’
(GE, 359–60). This care of the self was considered as both a duty and
a technique, a basic obligation and a set of carefully worked out procedures (HES, 95).
Foucault thus clearly points to the potential of ethics as a care for
the self in the secular, postmodernized world. He argues that we have
inherited the tradition of Christian morality with its values of selfrenunciation and self-sacrifice, as well as the secular tradition that sees
in external law the basis for morality. Against these traditions the care
for the self appears as immorality, egoism or a means of escape from
rules and responsibilities towards others (TES, 228). The care for the
self that Foucault advocates should be understood as stemming from a
wholly different conception of ethics, however, ethics as practice, creative activity, the permanent training of the self by oneself. To be an
ethical subject means to engage in ‘a process in which the individual delimits that part of himself that will form the object of his moral
practice, defines his position relative to the precept he will follow, and
decides on a certain mode of being that will serve as his moral goal’
(UP, 28).
The ethical subject
Foucault refers to the practices of the self in more general contexts as
technologies of the self, which are an important point of focus both in
fo u c au lt o n freedom
his later understanding of the subject as well as in his effort to rethink
the possible forms of morality for us today.
Technologies of the self . . . permit individuals to effect by their own
means, or with the help of others, a certain number of operations on
their own bodies and souls, thoughts, conduct, and a way of being, so as
to transform themselves in order to attain a certain state of happiness,
purity, wisdom, perfection or immortality.
(TES, 225)3
According to Foucault, technologies of the self have acquired different
forms in different moral systems over the centuries, through antiquity
and after the transition to Christian morality. In a seminar text, ‘Technologies of the Self’, written at around the same time as the last volumes
of The History of Sexuality, he formulates his study of pagan and early
Christian sexual morality in terms of ‘a hermeneutics of the self’. His
aim is to study the link between the obligation to know oneself in telling
the truth about oneself, and the obligation to submit to certain prohibitions against sexuality. He asks: ‘How had the subject been compelled
to decipher himself in regard to what was forbidden? . . . What must
one know about oneself in order to be willing to renounce everything?’
(TES, 224).
Foucault claims that the precept ‘to be concerned with oneself’ was
the form that this hermeneutics of the self takes in ancient Greece.
The other ancient maxim, ‘Know thyself’, was always subordinated
to it. The concern for oneself underlies Socratic dialogues. Socrates
presents himself in Plato’s Apology before his judges as a master of
epimeleia heautou. He tells his judges that they should concern themselves with themselves, that is, with ‘wisdom, truth and the perfection
of the soul’ (TES, 226). According to Foucault, there occurred a profound transformation in the moral principles of western culture with
the development of the Christian tradition, as well as with the growing
importance of the secular tradition. In Christianity, technologies of the
self adopted the paradoxical position of being the means of deciphering the truth about the self while at the same time the ultimate aim was
Foucault takes Gregory of Nyssa’s treatise On Virginity as an example. For Gregory, taking care of oneself meant the movement of
3 The text derives from a seminar Foucault gave at the University of Vermont in 1982, and
was originally published in English.
t he silen c e o f eth i c s
Gregory exhorts one to light the lamp and turn the house over and
search, until gleaming in the shadow one sees the drachma within. In
order to recover the efficacy which God has printed on one’s soul and
which the body has tarnished, one must take care of oneself and search
every corner of the soul.
(TES, 227)
Foucault thus argues that there has been an inversion between the
hierarchy of the two principles of antiquity, ‘Take care of yourself’ and
‘Know thyself’. Knowledge of the self was understood in Greco-Roman
culture as a consequence of taking care of the self and therefore was
subordinated to it. In the modern world, it constitutes the fundamental
principle. Foucault was not only referring to our religious tradition. He
also claims that the principle of knowing the self underlies all those
‘philosophies of the subject’ in which knowledge of the thinking subject
constitutes the first step in the theory of knowledge (TES, 228).
Foucault’s studies of the history of ethics can thus be seen as a continuation of his attempt to rethink the subject, this time the forms of the
self: the forms of understanding which the subject creates about himself or herself and the practices by which he or she transforms his or
her mode of being. Rather than understanding the ethical relationship
one has to oneself as a relationship of knowledge, Foucault advocates
an understanding of it as a ‘care’ or ‘concern for oneself’. With his
explication of ancient Greek ethics, he clearly wanted to further argue
the point that there is no true self that can be deciphered and emancipated, but that the self is something that has been – and must be –
created. There is a whole new axis of analysis present in his late studies
of the subject, however.
The last two volumes of The History of Sexuality appeared in a very
different form from the one that Foucault had originally planned and
proposed.4 He indicates in the introduction to volume ii that there was
an analytical axis missing from his previous work. To be able to study
the history of ‘the experience of sexuality’, he also needed, besides
the methodological tools with which his archaeologies and genealogies
4 The back cover of the first volume of The History of Sexuality announced the five forthcoming volumes: The second volume was to be called The Flesh and the Body and it was going
to deal with the problematization of sex in early Christianity; volume iii, The Children’s
Crusade, with the sexuality of children; volume iv, Woman, Mother, Hysteric, with the ways in
which sexuality had been invested in the female body; volume v, Perverts; with the person
of the pervert; and volume vi, Population and Races, with bio-politics. See Davidson 1994,
fo u c au lt o n freedom
had provided him, to ‘study the modes according to which individuals are given to recognize themselves as sexual subjects’ (UP, 5). He
then turned to studying the historical constitution of the self: the
forms of understanding subjects create about themselves and the ways
they form themselves as subjects of a morality, for example. While
his earlier genealogical studies investigated the ways in which the
power/knowledge network constitutes the subject, in his late work the
emphasis is on the subject’s own role in implementing or refusing
forms of subjectivity. His late work thus brings into focus a new component of the constitution of the subject – modes of relation to oneself –
and thus presents a more elaborated understanding of the subject than
is found in his earlier writings.
Many commentators refer to a third phase in Foucault’s thinking,
and note a marked change in his concerns. How this change is interpreted varies, however. Commentators such as Peter Dews (1989) see
it as an ‘abrupt theoretical shift’ and a ‘return of the subject’, while
others, such as Paul Rabinow (1984) understand it, in my view more
correctly, simply as a shift of emphasis.5 Foucault himself describes this
change in his thinking in various contexts in terms of a recasting of his
Perhaps I have insisted too much on the technology of domination and
power. I am more and more interested in the interaction between oneself
and others, and in the technologies of individual domination, in the
mode of action that an individual exercises upon himself by means of
the technologies of the self.
(TES, 225)
Technologies of the self are not separate from technologies of domination, which had been the focus of Foucault’s earlier studies. He points
out the necessary link between them. He argues that if one wishes to
analyze the genealogy of the subject in western civilization, one must
take into account the interaction between techniques of domination
and techniques of the self. This means analyzing the points at which
the technologies of domination of individuals over one another overlap processes by which the individual acts upon himself. Conversely, the
analyses must also take account of the points at which the techniques
of the self are integrated into structures of coercion or domination
5 See also e.g. Davidson 1986, 230.
6 See also e.g. UP, 11.
t he silen c e o f eth i c s
(ST, 181). These contact points are what Foucault calls governmentality
(TES, 225).
Hence, technologies of the self do not introduce a totally autonomous subject to Foucault’s late thinking. As he commented, even
if he was interested in the way in which the subject constituted himself or herself in an active fashion, by the practices of the self, ‘these
practices are nevertheless not something that the individual invents by
himself. They are patterns that he finds in his culture and which are
proposed, suggested and imposed on him by his culture, his society
and his social group’ (EPF, 11). Neither are technologies of the self
simple extensions of techniques of domination disguised as voluntary,
however. Foucault must presuppose a subject with some relative independence with regard to the constitutive power/knowledge network in
order to describe a subject capable of critical self-reflection and ethical
work on the self. As Gilles Deleuze (1986/1988, 101) argues, Foucault’s
fundamental idea is that of a dimension of subjectivity derived from the
power/knowledge network without being dependent on it. The subject
constituted by the power/knowledge network is now capable of turning
back upon itself: of critically studying the processes of its own constitution, but also of subverting them and effecting changes in them.
This understanding of the subject as being, on the one hand, constituted by the power/knowledge network, while on the other hand
retaining a relative independence from it, is, in my view, one of the
most problematic aspects of Foucault’s late thinking on ethics. I will
explicate my criticism in detail in chapter 9, but I will first defend Foucault against a number of other criticisms that have been levelled against
his ethics. I will argue that it is important to understand correctly his
idea of an ‘aesthetics of existence’, as well as his aim in inquiring into
the possibility of contemporary ethics. Foucault’s ethics must be read
as a continuation of his genealogy of the subject and of his on-going
concern with oppressive forms of subjection.
Ethics as aesthetics
The new focus on the government of the self by one’s self is crucial
in Foucault’s elaboration of resistance. Ethics is the domain in which
he situates it. Ethics becomes an important mediator in the triangle
of relationships between the subject, knowledge and power. In his late
thinking Foucault returns to the idea, found in his early work, of the
subversive role of art. The ethical practices of the self are closely linked,
fo u c au lt o n freedom
or even fused with aesthetics. When asked what kind of ethics it was
possible to build in our society, he replied:
[I]n our society, art has become something that is related only to objects
and not to individuals or to life. That art is something which is specialized
or done by experts who are artists. But couldn’t everyone’s life become a
work of art? Why should the lamp or the house be an art object but not
our life?
(GE, 350)
The process by which subjects care for themselves, form themselves
as ethical subjects, resembles the creation of a work of art. Foucault
characterized the ethical practices of the self also in terms of arts of
What I mean by the phrase [arts of existence] are those intentional and
voluntary actions by which men not only set themselves rules of conduct
but also seek to transform themselves, to change themselves in their singular being, and to make life into an œuvre that carries certain aesthetic
values and meets certain stylistic criteria.
(UP, 10–11)7
This idea of creating oneself as a work of art has fuelled a lot of heated
criticism against Foucault.8 He has been accused of retreating into
amoral aesthetics, privileging an elitist notion of self-centred stylization, and undermining all possibilities of emancipatory politics. The
7 ‘Par là il faut entendre des pratiques réfléchies et volontaires par lesquelles les hommes,
non seulement se fixent des règles de conduite, mais cherchent à se transformer euxmêmes, à se modifier dans leur être singulier, et à faire de leur vie une œuvre qui porte
certaines valeurs esthétiques et réponde à certains critères de style.’ (UPL, 16–17).
8 See e.g. Wolin 1986, Fraser 1989. Foucault’s practices of the self incorporate certain
intellectual attitudes – an attitude of permanent criticism – yet they are principally bodily
techniques that focus on everyday life and on the choices one makes in one’s way of life,
diet and habits. They also incorporate one’s sexuality and aspects of one’s gender. To
turn one’s life into a form of art involves one’s body, its experiences and pleasures, not
the renunciation of them. This connection between ethics, sexuality and embodiment
seems to open up interesting connections between Foucault and feminist ethics. Yet perhaps surprisingly, feminist theorists have commented very little on Foucault’s late work,
and generally their critical stance is derivative of an established ethical and theoretical
framework. In Foucault and Feminism, Lois McNay (1992), for example, follows Habermas
and argues that there is a problematic lack of normative grounding to Foucault’s implicit
criticism of modern society, and that his thinking therefore slips into a politically and ethically disabling relativism. Some feminist writers, such as Jean Grimshaw (1993), simply
dismiss Foucault’s studies of Greek morality as elitist and male-dominated, that therefore
sidestep questions that are crucial for feminist theory.
t he silen c e o f eth i c s
idea has also provoked a number of sympathetic responses, but the
ways in which it is interpreted still vary. Bernauer and Mahon (1994,
155) suggest that Foucault’s aesthetics of existence resists a ‘science of
life’. They claim that to think of human existence in terms of aesthetic
categories releases it from the realms of scientific knowledge and its
psychological and biological norms. Lois McNay (1992, 161) places
Foucault in the tradition of the romantic/modernist quest to retrieve
a more intense or worthwhile form of experience, which escapes the
deadening effects of the instrumental rationality pervading contemporary culture. She suggests that Foucault’s ethics explores the ways
in which individuals may redefine their existence in an experientially
impoverished world by adopting an ethical/aesthetical attitude in their
actions towards the surrounding world.
I will argue that Foucault’s ethics-as-aesthetics should be understood
primarily as a continuation of his permanent questioning of the limits
of subjectivity and the possibilities of crossing them. His ethics represents an attempt to seek ways of living and thinking that are transgressive in the extent to which, like a work of art, they are not simply the
products of normalizing power. The target of these practices is not primarily the aesthetically impoverished forms of experience, but rather
modes of normalization: the forms of power that produce forms of
David Boothroyd (1996) points out that in order to properly understand the link between ethics and aesthetics in Foucault’s thought, it is
important to understand the transgressive role Foucault assigns to art in
his thinking. In many of his texts he repeatedly and significantly returns
to the figure of the artist and the work of art: Van Gogh, Goya, Bataille,
de Sade, Artaud, Roussel, Magritte, Las Meninas, Don Quixote. The work
of art is always presented as exhibiting a certain resistance to the system,
of being able to work at the borders of a system of thought without being
totally incorporated into it. Art is thus able to mark the border between
the sayable and the unsayable, and constitutes a relation between the
inside and the outside of thought. According to Boothroyd, it is in
this context that individual life as a work of art provides the basis for
thinking practical forms of transgression.
Resistance against forms of subjection cannot be situated outside
the networks of power in Foucault’s thought, since subjectivity is only
possible within them. This means that resistance also becomes possible
only within them, through the subject’s lived practices, which help to
constitute forms of subjectivity; through the refusal and the adoption
of forms of subjectivity. Foucault’s later work on ethics is an inquiry into
fo u c au lt o n freedom
resistance: it represents the possibility of contesting determinations, of
refusing what we are.9
The problem with modern state power is that it is normalizing power:
it is individualizing and yet totalizing. It ignores individuality, difference
and becoming. At the same time, it splits up community life, forces the
individual back on himself and ties him to his own identity in a constraining way. The modern state is a sophisticated structure, in which
individuals can be integrated, but under one condition: that individuality will be shaped in a new form, and submitted to a very specific
pattern (SP, 212).
The way to contest this normalizing power is by shaping one’s self
and one’s lifestyle creatively: by exploring possibilities for new forms of
subjectivity, new fields of experiences, pleasures, relationships, modes
of living and thinking. It consists of creative activity as well as critical interrogation of our present and the contemporary field of possible experience.10 The quest for freedom in Foucault’s late thought, in
short, becomes a question of developing forms of subjectivity that are
capable of functioning as resistance to the normalizing power. Ethics is
a practice that stretches the limits of the subject.
It is in the context of normalizing power that we can also better
understand the importance of the ancient practices of the self for
Foucault’s thinking. As he points out, one cannot find any normalization in Stoic morality, there was no attempt to normalize the population. One reason for this was that the principle aim or target of this
kind of morality was an aesthetical one. It was a morality involving personal choice, the choice to live a beautiful life and to leave to others
memories of a beautiful existence (GE, 341).11
9 See SP, 216.
10 Foucault was not, however, suggesting that these practices of freedom are sufficient
in themselves to resist abusive forms of power or states of domination. ‘We need to
create the rules of law, the techniques of management, as well as the ethics, the ethos,
the practice of self, which will allow relations of power to circulate with a minimum of
domination’ (EPF, 18). I will return to this issue in the following chapter.
11 Timothy O’Leary (2002) discusses the extent to which Foucault overemphasizes the aesthetical aims of the ancient ethical practices by reading a completely different, modern
notion of aesthetics into their thinking. By reading Stoicism through the lens of modern
aesthetics, for example, he overemphasizes the free, creative, form-giving aspects of it
and neglects the normative role of reason and nature. O’Leary points out, however,
that the peculiarities of Foucault’s reading must be understood as part of his concern
to develop a contemporary, post-Christian ethics of self-transformation. His historical
studies are critical interventions in contemporary debates about ethical subjectivity and
therefore they must balance the concern for truth and historical accuracy with a concern
for the present.
t he silen c e o f eth i c s
Foucault’s aesthetics of existence should thus not be understood as
a narcissistic enterprise, nor as purely aesthetic in the narrow sense of
visuality, of looking stylish.12 As Timothy O’Leary (2002, 138) writes,
Foucault’s aesthetics of existence is an aesthetics not because it calls on
us to make ourselves beautiful, but because it calls on us to relate to ourselves and our lives as to a material that can be formed and transformed.
Foucault was very critical of the self-absorption and introspection characterizing our culture, and pointed out that the ancient practices of the
self were almost diametrically opposed to the present ‘culture of the self’
(GE, 362). His aim was not self-stylization conducive to narcissism, but
a personal transformation, a becoming, finding ways of thinking, living
and relating to other people that were currently unimaginable.
Philosophy lived
When Foucault’s ethics is understood as personal practice, it means that
ethical acts are primary in the sense that they will not find their justification in any general theory or principle. Foucault therefore invites
readers who are more accustomed to normative ethics to ask the obvious question: are all creative and transgressive acts ethical, and if not,
which ones are? Rape and murder could be seen as creative and transgressive, since for a lot of us it would certainly constitute a new field of
experience. Since there are no normative guidelines or rational justifications, there seems to be no way to make distinctions between different
acts and no way to determine which ones are ethical. As Foucault’s critics argue, what is wrong with his aesthetics of existence is that it can
never provide the critical framework necessary for being able to condemn certain actions, such as rape or murder, as being simply wrong.13
Therefore it fails to create the normative space of judgement which
these critics assume ethics should provide.
For Foucault, however, it was an impossibility to provide people with
normative grounding, guidelines, rules or criteria for passing moral
judgements. The task of an intellectual is not to tell others what to
do, people have to build their own ethics (MS, 16). What Foucault is
12 Paul Veyne (1986/1997, 231) also argues that style in Foucault’s thought does not mean
a distinction; the word should be understood in the sense in which the Greeks used it,
for whom an artist was first of all an artisan, and a work of art first of all a work.
13 Jürgen Habermas is perhaps the best-known critic of Foucault, and has accused him of
a lack of normative grounding in his analyses. See e.g. Habermas 1985/1987. See also
e.g. Walzer 1986, Taylor 1986, Fraser 1989. The planned discussion between Foucault
and Habermas never took place. For a reconstruction of the Foucaultian portion of this
exchange, see Flynn 1989.
fo u c au lt o n freedom
advocating is the rejection of universal foundations of ethics. His aim
was to offer an alternative to naturalistic ethics as well as to all other
universal forms of morality. He openly stated, ‘The search for a form of
morality acceptable to everybody strikes me as catastrophic’ (RM, 253–
4). His stance towards ethics also denies the possibility of normative
ethics: ethics understood as an abstract normative code, a collection of
rules or principles that would guide and justify our actions. The care
for the self that he advocates cannot be a universally applicable theory,
but only a practice, a transformative activity aiming to create a space
for the ethical. Foucault’s is not an ethics of normative judgement, but
one of critical practice and creativity.
Does this mean, therefore, that since normative questions are unanswerable in Foucault’s thought, ultimately his ethics fails to tell us anything about what is ethical? Does his ethics turn out to be, after all, not
ethics but aesthetics? It is my contention that his position seems to leave
room for at least three possible readings. (1) He would hold a relativist
position claiming that a philosopher should not go around patronizing,
but should leave it up to the people to articulate their own moral values
and the justification for them. The ethical aims of the practices of the
self would thus be absolutely individual and relative to personal values.
I will argue against this relativist position, but I still find two possible
directions to take Foucault’s ethics. (2) He would hold that there is
always an implicit level of values behind the practices of the self, even
though it might be unsystematically articulated and varies in different
historical and cultural contexts. For the ancient Greeks, this level consisted of the moral values and ideals dominant in their society, such as
mastery of the self, nobility, brilliance and beauty. For us, this normative background consists of the moral values we have inherited from
the Enlightenment, such as freedom, autonomy and equality. Commitment to the Enlightenment tradition and its values forms the shared
moral framework within which our personal ethical practices are situated. Foucault’s writing, in fact all writing, lies on normative notions of
some kind. What is significant, however, is the status of these notions
and the ways they are argued for. I will discuss this interpretation in the
next chapter.
(3) The reading which I will explicate in this chapter holds that
ethics cannot be grounded on any articulated moral framework at all.
This reading is thus based on a radically different conception of ethics.
The underlying values and normative criteria are unarticulated, not
because the philosopher should not provide them for us, but because
he or she cannot provide them. Ethical situations are characterized by
t he silen c e o f eth i c s
an experience of a fundamental limit to what can be brought to the
realm of language and knowledge.
For philosophy, this reading would seem to mean, however, that we
have to pass over in silence what we cannot talk about.14 If we want to
interpret that Foucault shares this Wittgensteinian insight that ethics
as a subject matter of philosophical inquiry is impossible, what, then,
is the point of him writing about it? How could philosophy say what
cannot be said? I will argue that, although Foucault’s ethics can never
be completely salvaged from the point of view of normative ethics, nor
should this be attempted, it is possible to show that as a style, a manner
of writing, it does express ethical values. These values are not communicated as linguistic propositions providing a normative ground or
framework; they must be understood as part of Foucault’s philosophy
as lived. His understanding of ethics as personal practice – care for
the self – therefore also challenges the role of philosophy. Philosophy
cannot provide privileged access to moral truths, it cannot ground the
ethical work one performs on oneself because it can only be this work:
philosophy lived as an ethos, an ethical practice.
To support my reading I will briefly turn to Derrida’s critique of
Foucault’s book Madness and Civilization. In making this detour to
Foucault’s early work I will suggest that Derrida shows us one possible way out of Foucault’s impasse concerning the silence of ethics and
the role of philosophy. In his essay ‘Cogito and the History of Madness’,
Derrida takes up the question of the limits of language in Foucault’s
thought. He criticizes Foucault’s attempt in Madness and Civilization to
write an archaeology of silence that lets madness speak for itself before
it is interned by the language of reason. To give a voice to madness itself
within philosophy is what Derrida claims is the impossibility inherent
in the very terms of Foucault’s project. Any effort to restore to madness its voice and its right to speak is already a form of repression, a
form imposed upon it and therefore again an act of imprisonment and
Derrida thus argues that Foucault did not seem to suspect the innocence of his own language or the order of rationality that it, by necessity,
14 Probably the single most often cited statement from the philosophical literature of
the twentieth century is the concluding line from Wittgenstein’s Tractatus: ‘Whereof
one cannot speak, thereof one must be silent’ (6.54). Read in combination with other
statements – such as the line that ‘there can be no ethical propositions’ (6.522), a radical
understanding of ethics, or rather the impossibility of it, in the usual sense of the word,
emerges. Ethics is not a term for a subject matter alongside other subjects. It comes
from our having a human world and a capacity to decipher ethical meanings in it. It
frames or gives form to our propositions about facts, but cannot be any one of them.
fo u c au lt o n freedom
conveyed. While being aware of the impossible task of giving an authentic voice to madness, Foucault nevertheless wrote a book with an impossible aim. As Derrida formulates this, Foucault’s determination to avoid
the trap that would have led him to write a history of untamed madness
with the restraining language of reason is the maddest aspect of his
project (Derrida 1967/1978, 34). As with ethics, so in connection with
madness, it seems, Foucault was trying to say something that could not
be said in philosophy.
Nevertheless, Derrida ends up suggesting that, although the silence
of madness cannot be expressed in the logos of the book, it can be
found in its pathos. The resolution of the difficulty is practised rather
than formulated. This again echoes Wittgenstein and his distinction
between saying and showing.15 What cannot be said in the book can be
shown by it. Foucault’s book on madness becomes an act that renders
madness present. Madness escapes philosophy only to appear in its
intensive style and its power to arouse emotions. Derrida formulates
this by writing: ‘What I mean is that the silence of madness is not said,
cannot be said in the logos of this book, but is indirectly made present,
metaphorically, if I may say so, in the pathos – I take the word in its best
sense – of this book’ (Derrida 1967/1978, 37).
Similarly, we can think that, for Foucault, the ideal of freedom is
something that can be shown with philosophy even if it cannot be said
in its (propositional) language. Even though values are not and cannot
be communicated as an explicit normative ground or framework, they
can be understood as part of Foucault’s philosophy as lived. His books
speak directly of the lack of freedom and hence indirectly of freedom:
the oppressive treatment of madness, the internment of criminals, sexual normalization and marginalization. Typically, they begin with his
perception that something is terribly wrong in the present and through
a study of history aim to show the contingency of our present. This opens
up the possibility for seeing to what extent it could be different. Philosophy as an ethical practice does thus not necessarily mean that we
write lots of academic books on ethics, but that our philosophical life is
ethical. As Foucault wrote: ‘The key to the personal poetic attitude of
a philosopher is not to be sought in his ideas, as if it could be deduced
15 The famous distinction between what can be said and what can only be shown
obtains a decisive philosophical significance in Tractatus. According to Wittgenstein,
what can be shown is, in the end, what matters, as the privileged object of philosophical insight. If all epistemic worries were suddenly, one beautiful morning,
resolved by scientific inquiry, ‘the problems of life have still not been touched at all’
t he silen c e o f eth i c s
from them, but rather in his philosophy-as-life, in his philosophical life,
his ethos’ (PE, 374).
Thus, philosophy cannot simply pass over in silence what it cannot
speak about. If we depart from the Tractatus view that only propositions
make sense, we can and we have to write about ethics. The task of interrogating the limits of our present is not something that is done once
so that the philosopher could quickly announce that behind our limits
nothing exists. It is rather that philosophy as a critical practice is a movement of always having to turn back: to question time and again what is
taken for granted, to start thinking anew. Philosophy must be practised
as an attempt to transgress what is given; as an interrogation of the limits of our thought, language and world.16 Perhaps at times this critical
practice of philosophy does succeed in penetrating the dimension of
the ethical.
To sum up, Foucault’s thought rejects formalist ethics and challenges
the idea that philosophy can provide privileged access to moral truths.
What is at stake are not rational arguments for a morally good code
of conduct, but a way of life that involves the whole of one’s being:
philosophy lived. As philosophy was a way of life, a spiritual exercise for
the ancient Greeks, so it was for Foucault.17 Philosophy is something
16 Paul Veyne writes (1986/1997, 231) that, during the last eight months of Foucault’s
life, the writing of his two last books played the role for him that philosophical writing
and the personal journal played in ancient philosophy: that of a work of the self on
the self, a self-stylization. Thomas Flynn (1989, 195–6) argues that Foucault’s trenchant
nominalism, the inquiring scepticism, the distrust of power relations and his penchant
for an aesthetics of existence suggest that he empathized with the Cynics’ concept of
philosophy as ethos or life. His involvement in various political struggles of his time
are marks of the parrhesiast, the truth-teller. Without being the applications of a complete political theory, such actions by their exemplary nature hold before us our own
endangered freedom as in a critical mirror.
17 Arnold I. Davidson (1997a, 195–6) explicates the profound importance that Pierre
Hadot’s writings had on the last works of Foucault. According to Davidson, Hadot’s
focus on the notion of spiritual exercises is a way of primarily emphasizing that in
ancient schools of thought philosophy was a way of life. It was an invitation to transform
oneself and one’s way of life as well as a quest for wisdom. Philosophy understood as a
form of life required spiritual exercises that aimed at realizing a transformation of one’s
vision of the world and a metamorphosis of one’s personality. These aims are visible
in Socratic dialogues, they are spiritual exercises practised in common. What is most
important is not the solution to a particular problem, but rather the path traversed
in arriving at this solution. Davidson argues that Foucault’s aim is to link the practices
of the self exhibited in the domain of sexual behaviour to the spiritual training and
exercise that govern the whole of one’s existence. Parallel to Hadot’s argument that
spiritual exercises gradually became almost eclipsed by the conception of philosophy
as an abstract, theoretical activity, Foucault argues that, in the realm of ethics, morality
understood as codes of behaviour gradually came to be emphasized at the expense of
forms of subjectivation (Davidson 1997a, 199–201). See also e.g. Hadot 1983/1997.
fo u c au lt o n freedom
that enables one to learn to think differently, and through thinking to
change one’s mode of being, to transform oneself: ‘Transformation of
one’s self by one’s own knowledge is . . . something rather close to the
aesthetic experience’ (MS, 14).
Despite their dispute on reason and madness, Foucault and Derrida
seem to agree that philosophy cannot and should not attempt to be
a totalizing gesture, the master discourse. It should not tower above
all other forms of knowledge, ruling us by passing moral judgements.
If what is required from philosophy is precisely awareness of its limits
and pretensions and an ability for self-criticism and even irony, then
Foucault’s disconcerting thought can be seen to do just that. But it
also shows us a Utopian moment: a deep commitment and belief in
something that I take to resemble freedom. If Foucault’s archaeology
of silence had an impossible aim, possibly his ethics of silence did too.
Even so, it is one more thrust against the walls of our cage.
Enlightenment is man’s release from his self-incurred tutelage. Tutelage is man’s inability to make use of his understanding without direction from another. Self-incurred is this
tutelage when its cause lies not in lack of reason but in lack
of resolution and courage to use it without direction from
another. Sapere aude! ‘Have the courage to use your reason!’ –
that is the motto of enlightenment.
(Kant 1784/1997, 7)
Apart from the focus on the self, Foucault’s ethics has another
dimension: critical responding to one’s time. Critical work encompasses
the subject’s personal work on him/herself, and also a critique of society, power relations and structures. While I have argued that to criticize Foucault’s late work for a lack of normative guidelines misses
the point of his effort to rethink ethics, the need for a normative
grounding becomes more pressing in connection with politics. The
subject’s ethical work on him/herself may be based on unthematized
values and experiences of liberation, but a shared conception of freedom seems necessary in emancipatory politics. Concepts empower, they
incite discussions, arguments, dialogues. Normative ideals such as freedom, equality and justice articulate Utopian possibilities and give imaginations a concrete form that can be communicated and shared as a
common political ideal and goal.
The connection between philosophy and politics in Foucault’s
thought is, to say the least, ambiguous. David Couzens Hoy (1998,
18–20), for example, argues that, although Foucault’s writings seem
to be politically engaged, exactly how they generate this effect is not
clear. Hoy finds evidence in Foucault’s writings that he asserted both
fo u c au lt o n freedom
that philosophy and politics are profoundly linked, and that they are
not linked.1 Wendy Brown (1998, 33) argues similarly that Foucault’s
responses to expressly political questions in interviews are frequently
vague, oblique, deflective or simply bland. While Hoy seeks to redeem
Foucault’s thought for progressive politics by pointing to his involvement in various political struggles, Brown argues for the opposite. She
claims that Foucault’s thinking opposes all traditional understandings
of politics, and instead reformulates the political as opposition to politicization on the one hand, and to policy on the other (Brown 1998, 42).
She challenges the idea that Foucault’s particular political positions and
enthusiasms were an outcome of his genealogical studies: a genealogical politics has no necessary political entailments.
I will argue in this chapter that Foucault’s understanding of the connection between philosophy and emancipatory politics turns on his
stance on the Enlightenment. In order to understand his late thinking on ethics we have to read it in connection with his other writings, his genealogies of subject and power, but particularly with his
writings on the Enlightenment. In what follows, I will briefly present
the common form of criticism against the political implications of
Foucault’s thought, and then present two possible readings of his position in regard to emancipatory politics, arguing for the latter. I will
conclude by distinguishing four different meanings of freedom that I
find in his work.
The freedom of critical reflection
The criticism that claims that Foucault’s thought refutes emancipatory
politics usually revolves around two themes. Firstly, because Foucault
defends the view that social relations are inevitably power relations,
and that there can be no Utopian position outside or beyond power, it
is claimed that the idea of any kind of liberation becomes impossible.
Since all relations are power relations, there is no possibility of progress
in the sense that social relations will become less oppressive. Nancy
Fraser, for example, has argued for this point: ‘Because Foucault has
1 Hoy (1998, 20–1) ends up arguing for the former stance: the idea of a profound link
between philosophy and politics represents Foucault’s more mature view. Hoy claims that,
for Foucault, philosophy is imbedded in an ethos critically involved in minimizing domination. In the ethos the political is personal and the personal is political. Thomas Flynn
(1989, 188) also argues that if politics is the art/science of governance, if governance is
the directing of power relationships, and if power, for Foucault, is all-pervasive, then so
too is the ‘political’: every facet of human life carries a political dimension and stands
subject to ‘political’ analysis.
t he fr eed o m o f philo s op hy
no basis for distinguishing, for example, forms of power that involve
domination from those that do not, he appears to endorse a one-sided,
wholesale rejection of modernity as such’ (Fraser 1989, 32–3).
As I have argued in my discussion of Foucault’s genealogy, this kind of
conclusion is based on a misunderstanding of his conception of power.
A careful reading of his late texts on power, in particular, makes it
possible to distinguish between different meanings and levels of power:
(1) individual power relations; (2) domination; and (3) power as a
strategic situation, an overall network. While it is impossible to step out
of the social field structured by power relations, it is possible to effect
changes in it: for example to free subjects from states of domination –
situations in which the subject is unable to overturn or reverse the power
relation – to a situation in which power relations are interchangeable,
variable and allow for strategies for altering them. Foucault goes as far
as to set this as an explicit task.
I don’t believe there can be a society without relations of power, if you
understand them as means by which individuals try to conduct, to determine the behaviour of others. The problem is not of trying to dissolve
them in the Utopia of a perfectly transparent communication, but to give
one’s self the rules of law, the techniques of management, and also the
ethics, the ethos, the practice of self, which would allow these games of
power to be played with the minimum of domination.
(EPF, 18)2
Hence, although there can be no overall liberation from power, there
can and will be ‘particular’ emancipations from different systems of
domination: from oppressive relations of power and the effects of the
employment of certain normalizing techniques. Foucault again states
[W]e know from experience that the claim to escape from the systems of
contemporary reality so as to produce the overall programs of another
society, of another way of thinking, another culture, another vision of
the world, has led only to the return of the most dangerous traditions. I
prefer the very specific transformations that have proved to be possible
in the last twenty years in a certain number of areas that concern our
2 ‘je crois qu’il ne peut pas y avoir de société sans relations de pouvoir, si on les entend
comme stratégies par lesquelles les individus essaient de conduire, de déterminer la
conduite des autres. Le problème n’est donc pas d’essayer de les dissoudre dans l’utopie
d’une communication parfaitement transparente, mais de se donner les règles de droit,
les techniques de gestion et aussi la morale, l’êthos, la pratique de soi, qui permettront,
dans ces jeux de pouvoir, de jouer avec le minimum possible de domination.’ (EPL, 727)
fo u c au lt o n freedom
ways of being and thinking, relations to authority, relations between the
sexes, the way we perceive insanity or illness.
(WE, 46–7)3
This view of liberation as resistance to particular states of domination
involves a negative view of freedom. Foucault claims that power relations
arise when there is action upon the actions of others. Power only functions on free action, it is an action on action (SP, 221). It is thus always
exercised on ‘free’ subjects, free, however, here meaning no more than
being able to act in a variety of ways. Subjects free of domination are
capable of instigating shifts in power relations by acting in different
ways to influence each other’s behaviour. Why or how they should do
this is not, however, discussed in this understanding of freedom.
While the first objection to Foucault’s thought in connection with
emancipatory politics results from a clear misunderstanding of his conception of power, the second objection is more serious. His critics claim
that his analysis of power provides us with no normative grounds for
condemning domination and arguing for liberation. If we do not share
some kind of idea that freedom is desirable, then we cannot see why disciplinary power represents domination and an abuse of power. In short,
how can we oppose domination or criticize oppression without any ideal
of freedom: the desirability of liberation and human autonomy?
Foucault explicitly claims that the role of philosophy is to be critical
of power: it is to keep watch over the excessive powers of political rationality (SP, 210). When asked in one of his last interviews if philosophy
had the duty of sounding a warning on the danger of power, he replied:
That duty has always been an important function of philosophy. On the
critical side – I mean critical in a very broad sense – philosophy is precisely the challenging of all phenomena of domination at whatever level
or under whatever form they present themselves – political, economic,
sexual, institutional.
(EPF, 20)4
3 ‘on sait par expérience que la prétention à échapper au système de l’actualité pour
donner des programmes d’ensemble d’une autre société, d’un autre mode de penser,
d’une autre culture, d’une autre vision du monde n’ont mené en fait qu’à reconduire les
plus dangereuses traditions. Je préfère les transformations très précises qui ont pu avoir
lieu depuis vingt ans dans un certain nombre de domaines qui concernent nos modes
d’être et de penser, les relations d’autorité, les rapports de sexes, la façon dont nous
percevons la folie ou la maladie.’ (QL1, 575)
4 ‘Cette tâche a toujours été une grande fonction de la philosophie. Dans son versant
critique – j’entends critique au sens large – la philosophie est justement ce qui remet en
question tous les phénomènes de domination à quelque niveau et sous quelque forme
qu’ils se présentent – politique, économique, sexuelle, institutionelle.’ (EPL, 729)
t he fr eed o m o f philo s op hy
Foucault thus presents philosophy as emancipatory insofar as it is a
critique of power: a critical analysis of domination, abusive forms of
power and the constraining forms of subjectivity that power produces.
How exactly philosophy manages this critical task without a normative grounding, is, however, the question that Foucault’s critics repeatedly take up. I will present this critique, in a nutshell, through Sheila
Benhabib’s work.
Seyla Benhabib (1987) argues, in connection with feminist theory,
that there are two necessary moments in social critique.5 Firstly, there
is the explanatory-diagnostic analysis in which the social crisis is examined. Secondly, there is the anticipatory-Utopian moment, a which articulates the normative groundings of the critique. This second aspect
is primarily normative and philosophical, ‘it involves the clarification
of moral and political principles, both at the metaethical level, with
respect to their logic of justification, and at the substantive, normative
level, with reference to their concrete content’ (Benhabib 1987, 81).
What Foucault’s thought is thus lacking, according to his critics, is the
normative and philosophical level necessary for any form of critique to
be a critique and not just a description. These critics thus claim that
Foucault’s thought might be interesting, insightful and even true, but
it is not critical.
The first line of defence of Foucault’s position is to simply agree that
if critique is understood in the way Benhabib understands it, then what
Foucault offers us is not a critique but pure description or diagnosis.
Foucault’s aim was to take up the Socratic or Nietzschean task of exposing the familiar as an illness. He writes of genealogy that its task is to
become a curative science (NGH, 80). Foucault’s description can, however, be turned into an effective critique by its readers. Hoy (1998, 26–7)
argues for this kind of reading of Foucault’s genealogy. He writes that
genealogy observes and interprets contingent social formations and
phenomena from the inside, without positing a transcendental perspective or transcendentally necessary universal standards. The genealogist
tries to see as strange what the culture takes to be familiar. ‘Insiders’
reading a genealogical investigation come to see for the first time subliminal aspects of their behaviour. According to Hoy, simply becoming
conscious of social practices that were previously unconscious often
leads to difficulty in continuing to engage in those practices.6 Hence,
5 See also Benhabib 1986.
6 Foucault himself also seemed to suggest something like this when he wrote, ‘The history
of various forms of rationality is sometimes more effective in unsettling our certitudes
and dogmatism than is abstract criticism’ (PR, 83).
fo u c au lt o n freedom
becoming aware of types of social behaviour will lead to readers becoming critical of that behaviour. However, as Brown argues, genealogical
diagnosis itself would have no necessary political entailments. Foucault
sought to diagnose our present, our political rationality, the forms of our
subjectivity and the kind of deployments of power that have produced
them. Brown writes (1998, 34–7) that genealogy opens up a political
space that harbours no explicit political aims but which is replete with
challenging exposures and destabilizations. It reveals the necessary as
contingent, and can thus be deployed to incite possible futures, but not
to prescribe any political positions or specific futures.
It can, furthermore, be argued that it is not a question of Foucault’s
diagnostic position being uncritical and politically defective, but that
the novelty of his thought lies exactly in the fact that he manages to
challenge our traditional ideas of what critique is. His work severs critique from prescription and effectively questions the claim that the
clarification of a normative grounding is a prerequisite for criticism.7
Hence, we may disagree with Foucault, we may disagree with his diagnosis, the aims that he set himself and his implicit moral engagements.
An explicit normative grounding, however, will not bring about general consensus. Explicitly stating that a universal principle of freedom
should underline any social practice will not solve the problem of the
different interpretations and value judgements connected with modern forms of power. As Benhabib herself clearly points out, the second
level of critique is normative and philosophical, it deals with morality
or ethics, and therefore no rational consensus on matters of fact can
ultimately solve it.
Foucault was clearly not advocating the impossibility of critical judgement and action; his engagement in various political and ethical confrontations during his own time are a demonstration of this. According
to my first readings, however, on the basis of his philosophy, he could
only pass these ethical and political judgements as a person, and they
contained no ideals for everybody to follow. Philosophers cannot be
politicians: the task of philosophy is to call into question our understanding of politics.
I have especially wanted to question politics, and to bring to light in
the political field, as in the field of historical and political interrogation,
some problems that had not been recognized there before. I mean that
the questions I am trying to ask are not determined by a pre-established
7 See e.g. Flynn 1989.
t he fr eed o m o f philo s op hy
political outlook and do not tend towards the realization of some political
project. This is doubtless what people mean when they reproach me for
not presenting an overall theory. But I believe precisely that the forms
of totalization offered by politics are always, in fact, very limited. I am
attempting, to the contrary, apart from any totalization – which would be
at once abstract and limiting – to open up problems that are as concrete as
possible, problems that approach politics from behind and cut societies
on the diagonal, problems that are at once constituents of our history
and constituted by that history.
(PE, 375–6)8
In one sense, Foucault’s work is emancipatory – promoting freedom –
in that it might engage its readers in critical practices.9 It is also emancipatory in another sense, in that the studies themselves as inquiries into
ways of understanding our selves and our society represent the freedom
the subject exercises in turning to question its own constitutive conditions.10 One of the most important features of Foucault’s late thinking
is that he recognized the significant role played by the freedom of critical thinking: the freedom to reflect on one’s self and potentially change
aspects of the self, but also through reflection to change society at
The freedom that opposes domination and abusive power in this view
is the freedom embedded in critical inquiries and practices. Freedom
is not a moral or political principle underlying specific practices, it is
a practice. Neither is it a state of being of the subject or a legal or
institutional structure. The subject exercises freedom in withdrawing
from itself and problematizing its behaviour, beliefs and the social field
of which it is part. The practices of freedom are the practices capable
of changing the constitutive conditions of our subjectivity as well as its
8 ‘En fait j’ai surtout voulu poser des question à la politique et faire apparaı̂tre dans
le champ de la politique comme de l’interrogation historique et philosophique, des
problèmes qui n’y avaient pas droit de cité. Les questions que j’essaie de poser ne sont
pas déterminées par une conception politique préalable et ne tendent pas à la réalisation
d’un projet politique défini. C’est sans doute cela que les gens veulent dire lorsqu’ils
me reprochent de ne pas présenter de théorie d’ensemble. Mais je crois justement que
les formes de totalisation offertes par la politique sont toujours, en fait, très limitées.
J’essaie, au contraire, en dehors de toute totalisation, à la fois abstraite et limitative, d’ouvrir
des problèmes aussi concrets et généraux que possible – des problèmes qui prennent la
politique à revers, traversent les sociétés en diagonal, et sont tout à la fois constituants
de notre histoire et constitués par elle.’ (PEI, 586–7)
9 See e.g. McWhorter 1999.
10 Thomas Flynn (1991, 115), for example, notes that the genealogical charting of the
advent of the modern subject is itself a form of liberation.
fo u c au lt o n freedom
actual forms. John Rajchman formulates well this sense of freedom in
Foucault’s thought:
Our real freedom is found in dissolving or changing the polities that
embody our nature, and as such it is asocial or anarchical. No society
or polity could be based on it, since it lies precisely in the possibility of
constant change. Our real freedom is thus political, though it is never
finalizable, legislatable, or rooted in our nature.
(Rajchman 1985, 123)
Freedom as ethos
According to my first reading, Foucault’s work is critical in the sense
that, although purely descriptive, it nevertheless represents the critical practice of freedom in posing questions about the constitutive
conditions of subjectivity. I will, however, argue for a second reading, according to which it is not purely descriptive, but incorporates
a normative dimension. This normative dimension is, I claim, what
gives his thought its political character.11 Foucault’s analyses are undertaken with the explicit aim of changing social reality in the direction of
The ideal of freedom as emancipation from the effects of power is
an important part of the Enlightenment thinking and the subsequent
understanding of emancipatory politics. Foucault, however, is notorious for his clear objection to the universalistic discourse of Enlightenment emancipation: there is no inherent human nature justifying
the demands for human freedom or guaranteeing the possibility of
progress. Foucault warned us that the Enlightenment ideal of individual autonomy was one effect of normalizing power, power that is
totalizing and individualizing at the same time (SP, 213). According to
Foucault, Enlightenment humanisms have furthermore either masked
forms of disciplinary power that operate to produce forms of modern
individuality, or have participated in extending domination (Sawicki
1996, 169).
Consequently, when, shortly before his death, Foucault wrote a reading of an article by Kant entitled ‘What is Enlightenment’ (WE), in
which he located himself squarely within the Enlightenment tradition
of philosophy, many of his readers were surprised and confused.12 I
11 Cf. Hindes 1998.
12 For more on Foucault’s writings on the Enlightenment, see also ATT, WC.
t he fr eed o m o f philo s op hy
take this move to be both theoretically as well as morally significant,
however. I argue here that Foucault’s critical reappropriation of the
Enlightenment was motivated by the urgency to elaborate his understanding of freedom and to build on it by introducing into it a deliberate
dimension.13 By rooting his thought in the inheritance of the Enlightenment, he implicitly professed his faith in its values: the increase of
autonomy among individuals and the importance of philosophy, that is,
philosophy understood as critical thought. His writings on the Enlightenment can be read as a clear gesture of distancing himself from the
ultra-relativist, neoconservative and postmodern labels that had been
stuck on him.14
Foucault was interested in Kant from very early in his life and wrote
his complementary doctoral thesis on him. This interest culminated in
The Order of Things and Foucault did not refer directly to Kant again
until the end of the 1970s. While strongly refusing all other political and philosophical labels, he explicitly asked to be called ‘simply
Nietzschean’ (RM, 251). By accepting this label, what he was endorsing,
however, was not the nihilism and pessimism that many of his critics see
as the baneful influence of Nietzsche, but the idea that a philosopher is
not someone who attempts to totalize his own time by building a system,
but one who establishes a diagnosis of the present.15 When he turned
to write about Kant again, it was precisely this idea of philosophy as a
way of questioning our own present that he took up.
Foucault’s text ‘What is Enlightenment?’ can be read as an interesting interpretation of the main impact and implications of the Enlightenment. What it meant for him was a set of political, economic, social,
institutional and cultural events on which we still depend for the most
part, and which therefore constitutes a privileged domain for analysis.
It is also an enterprise for linking the progress of truth and the history
of liberty, and in doing so, according to Foucault, it formulated a philosophical question that remains for us to consider. What he found to be
its most significant aspect, however, was that it initiated a new type of
philosophical investigation. What characterizes the philosophical ethos
13 Cf. Sawicki 1996, 170. See also Sawicki 1994.
14 See also e.g. O’Leary 2002.
15 Nietzsche distanced himself from the optimism of the Enlightenment and is commonly
criticized for his individualism and lack of concern for political community. According
to Walter Kaufman, for example, Nietzsche gave up hope for his own people and for
mankind, and addressed himself only to single human beings (Kaufman 1950/1974,
421). For Nietzsche, ethics could only be a task for the ‘single one’. While he was in this
sense an anti-political thinker, I argue that this does not hold true of Foucault.
fo u c au lt o n freedom
originating in the Enlightenment is that it is a permanent critique of our
own philosophical era. The philosopher should not occupy his mind
with eternal or timeless truths, but must position himself at his own
particular and historical moment and find the meaning of philosophy
The task is not just to bring to philosophy the distinctive concerns
of the present time, such as questions stemming from on-going political struggles. The aim is not just to offer a new topical content for
philosophical inquiry, but more fundamentally, to question its meaning. What does it or can it mean to engage in philosophy today?
What is happening today? What is happening now? And what is this
‘now’ within which all of us find ourselves; and who defines the moment
at which I am writing? . . . the question Kant is answering . . . is not simply:
what is it in the present situation that can determine this or that decision
of a philosophical order? The question bears on what this present actually is, it bears firstly on the determination of a certain element of the
present that is to be recognized, to be distinguished, to be deciphered
among all the others. What is it in the present that produces meaning
now for philosophical reflection?
(ATT, 87)16
Through his reading of Kant, Foucault explicitly presents his own work –
from his early archaeological writings to his genealogies of the modern
subject as well as the diagnosis of modern forms of power – essentially as
an Enlightenment project: a series of historico-critical analyses studying
‘the events that have led us to constitute ourselves and to recognize ourselves as subjects of what we are doing, thinking and saying’ (WE, 46).
This critical ontology of ourselves is a philosophical, ethical and political task all at the same time: it is ‘an attitude, an ethos, a philosophical
life in which the critique of what we are is at one and the same time the
historical analysis of the limits that are imposed on us and an experiment with the possibility of going beyond them’ (WE, 50). Foucault
calls the philosophical ethos characterizing his work a limit-attitude. He
16 ‘qu’est-ce qui se passe aujourd’hui? Qu’est-ce qui se passe maintenant? Et qu’est-ce
que c’est que ce “maintenant” à l’intérieur duquel nous sommes les uns et les autres;
et qui définit le moment où j’écris? . . . la question à laquelle Kant répond . . . n’est
pas simplement: qu’est-ce qui, dans la situation actuelle, peut déterminer telle ou telle
décision d’ordre philosophique? La question porte sur ce que c’est que ce présent,
elle porte d’abord sur la détermination d’un certain élément du présent qu’il s’agit de
reconnaı̂tre, de distinguer, de déchiffrer parmi tous les autres. Qu’est-ce qui, dans le
présent, fait sens actuellement pour une réflexion philosophique?’ (QL2, 679–80)
t he fr eed o m o f philo s op hy
wants to turn the Kantian question around: rather than asking what
limits knowledge has to renounce transgressing, he is asking, ‘In what
is given to us as universal, necessary, obligatory . . . what place is occupied by what ever is singular, contingent and the product of arbitrary
constraints?’ (WE, 45). He was thus not interested in showing what are
the necessary conditions determining the limits of reason, but in revealing the extent to which the limits presenting themselves as necessary
are actually contingent.
Hence, Foucault does not simply embrace traditional Enlightenment
ideals, but submits them to critical reappropriation. Through reservation, he denied that his work was simply for or against the Enlightenment. He refused the ‘blackmail of the Enlightenment’, the idea that
‘one has to be “for” or “against” the Enlightenment’ (WE, 43). For
him, reappropriating the critical ethos of the Enlightenment ‘means
precisely that one has to refuse everything that might present itself in
the form of a simplistic and authoritarian alternative’ (WE, 43). Foucault also clearly distances himself from humanism, warning us that ‘We
must escape from the historical and moral confusionism that mixes the
theme of humanism with the question of the Enlightenment’ (WE, 45).
He thus saw humanism not as a critical questioning of the present, but as
a diverse and inconsistent set of themes designed to justify and promote
particular values. It necessarily leans on conceptions of what it means
to be human borrowed from religion, science or politics, and thus functions as a form of justification, not as a form of critique. Enlightenment
and humanism are therefore ‘in a state of tension rather than identity’
(WE, 44).
The limit-attitude characterizing the philosophical ethos of the
Enlightenment has to be translated into specific inquiries. It is in this
context that Foucault presents his analyses of the three axes of the
constitution of the subject – knowledge, power, ethics – as the concrete forms into which the limit-attitude translates. The ontology of
ourselves poses the questions: ‘How are we constituted as subjects of our
knowledge? How are we constituted as subjects who exercise or submit
to power relations? How are we constituted as moral subjects of our
own actions?’ (WE, 49). Archaeology and genealogy are methods in
this inquiry into the constitution of the subject conducted as a study of
practices or ‘practical systems’ (WE, 48). Nevertheless, these historicocritical reflections must also be put to the test of contemporary reality,
‘both to grasp the points where change is possible and desirable, and
to determine the precise form this change should take’ (WE, 46).
fo u c au lt o n freedom
Even though, for Foucault, philosophy cannot ground knowledge
in essential and universal truths, in analyzing our present it can break
contingent and historical limits and constraints. The critical work on
our limits is what, in Foucault’s thought, links ancient ethics and
his understanding of the philosophical ethos of the Enlightenment.
What is common to both is the necessity for a critical attitude towards
the self that shows both an awareness of the contingency of one’s
traits and a willingness to rework them (Moss 1998, 4). Furthermore,
both ethical work on one’s self and philosophy as a historico-critical
practice have ‘freedom’ as their implicit goal: the aim is to investigate limits and constraints, abusive forms of power, and in revealing
them as contingent, to open up the possibilities for experimenting
to transgress them. In the practices of freedom the subject exercises
its relative autonomy to gain more autonomy: it creates a space for
freedom. While this goal can be unarticulated in the realm of ethics,
in the realm of politics it becomes a historical ideal, an ethos of our
Foucault associates the Enlightenment firmly with critique, a critical
attitude that questions not only obstacles to the use of reason, but also
reason itself and its limits. In an earlier lecture on the Enlightenment
given in 1978 and published as, ‘What is Critique’, Foucault claimed
that this critical attitude has, in modern philosophy, taken the form
of questioning reason in its connection with power, ‘the relationships
between the structures of rationality which articulate true discourse and
the mechanisms of subjugation which are linked to it’ (WC, 45). This
critique of reason as responsible for excesses of power has taken different forms in the history of philosophy. In Germany, for example, from
the Hegelian left to the Frankfurt School, there has been a criticism
of the relationships between ‘the fundamental project of science and
techniques whose objective was to show the connections between science’s naive presumptions, on the one hand, and the forms of domination characteristic of contemporary society, on the other’ (WC, 38–9).
In France, this critical work was taken up, on the one hand, by phenomenology through the question of what constitutes meaning and,
on the other hand, by the history of the sciences.
Where are we with this rationalization which can be said to characterize
not only western thought and science since the sixteenth century, but also
social relationships, state organizations, economic practices and perhaps
even individual behaviors? . . . This problem, for which in France we must
t he fr eed o m o f philo s op hy
now shoulder responsibility, is this problem of what is the Aufklärung? We
can approach it in different ways. And the way in which I would like to
approach this – you should trust me about it – is absolutely not evoked
here to be critical or polemical.
(WC, 43–4)17
I argue that, by linking his thought to the Enlightenment, Foucault
makes the normative move of adopting the ideals associated with it –
critical reason and personal autonomy – as the implicit ground on which
his critiques of domination, abusive forms of power and reason rest. The
Enlightenment provides him with the historical – not transcendental –
values on which to base his critiques. Unlike Kant, he endorses freedom
not as an abstract and universal ideal, but as an outcome of a certain historical development: historical and sociological facts. The ideal of freedom behind the philosophical critique of domination or our political
rationale emerges from historically concrete and specific practices, and
can only emerge out of them. The championing of political freedom in
the modern sense cannot be found as such in any pre-Enlightenment
tradition, but is rather a product of a specific historical tradition of
thought – the Enlightenment – which we are part of in any case. The
ideal of freedom is a commitment to a tradition according to which
we think about human life and politics. As Thomas Flynn (1989, 197)
points out, this means that Foucault neither offers nor seeks foundations beyond the presumed commitment of his audience to freedom
autonomy. To those who greet him with a ‘So what?’, there can be no
Freedom is thus not human autonomy as the transcendental condition of moral action, as it is for Kant, but rather it is the contingent
historical condition of critical reflection on our present. Neither is freedom a Kantian regulative idea for Foucault. It does not provide us with
a criterion for action – we must act as if we were free – but only gives a
source of motivation, a commitment to a historical value. By revealing
constraining forms of subjectivity as historically contingent, Foucault’s
analyses can be read as actively advocating social change in the direction of ‘freedom’. Although such change must be understood in terms
of specific and partial transformations rather than as general political programmes, Foucault’s thought is far from political nihilism. The
analyses of our limits are analyses of freedom.
17 The French original is not available.
fo u c au lt o n freedom
The different meanings of freedom
I have attempted to show so far in this, and the previous chapters, that
when freedom in Foucault’s thought is brought into the realm of ethics
and politics, it becomes a concrete practice as well as a historical ideal,
an ethos. Foucault writes, rather cryptically, ‘Freedom is the ontological
condition of ethics. But ethics is the considered form that freedom takes
when it is informed by reflection’ (EPF, 4, trans. modified). When the
different conceptions of freedom present in the different phases of
Foucault’s thought are elucidated, we can understand what Foucault
means by this claim.
Paul Patton argues that by ontological condition, Foucault refers to
the concrete capacities of the subject: ‘Just as for Foucault political
power exists only in the concrete forms of government of conduct, so
freedom exists only in the concrete capacities to act of particular agents’
(Patton 1998, 45). I argue, however, that freedom as an ontological
condition does not, for Foucault, refer to any capacity or potentiality of
the subject. While the understanding of the subject in Foucault’s ethics
implies that it is capable of engaging in reflexive practices of freedom –
forms of critical self-reflection and care for the self – this capacity is
not what Foucault meant by positing freedom as an ontological condition of ethics. Freedom is not an ontological characteristic of the
subject. We are not born free, nor have we an inherent capacity to realize our freedom. There is no inherent human freedom in either sense;
what Foucault aimed to do is, on the contrary, to separate the idea of
our nature from freedom. John Rajchman points out that Foucault’s
genealogy is a continuation of Nietzsche’s philosophy in this respect
(Rajchman 1985, 121).
I argue that, by presenting freedom as the ontological condition
of ethics, Foucault, in fact, refers to the ontological contigency of the
present: freedom is the opening up of possibilities of an age.18 This
is the sense of freedom that I have explicated in the previous parts
in connection with Foucault’s understanding of language on the one
hand, and bodies and pleasures on the other. Freedom in this sense
is the condition of possibility of Foucault’s ethics: it is the moment
of the unexpected as opposed to the normalized, the unforeseen as
opposed to the determined. My point has been to show, however, that
ethics and politics – understood in Foucault’s thought as concrete
18 See Rajchman 1985, 46.
t he fr eed o m o f philo s op hy
practices – delineate a realm where freedom understood as an ontological condition can be given ‘a considered form’. When freedom is
brought into the realm of ethics and politics – informed by reflection –
it becomes a concrete practice of freedom as well as a historical ideal,
an ethos of our culture. I am now in a position to sum up the different
meanings of freedom operative in Foucault’s thought.
Freedom as ontological contingency. I have shown in connection with
Foucault’s archaeology how freedom refers to the indeterminacy of
discursive structures: language can never be fully mastered or tamed,
but results in unexpected orders and unimaginable conjunctions of
meanings. Foucault not only gives language a regulative role in the
mode of scientific discourse, he also demarcates a domain of freedom
in the mode of literature, particularly as avant-garde writing. While
scientific discourses form an ontological order of things that is implicit
in theories and practices, language as avant-garde writing is capable
of forming alternative, unscientific and irrational ontological realms:
different experiences of order on the basis of which different perceptual
and practical grids become possible, and hence new ways of seeing
and experiencing emerge. While Foucault’s archaeology is generally
viewed as emphasizing the necessary structures of scientific discourse
and opposing humanist aspirations of looking for the freedom of man,
it contains an anti-humanist understanding of freedom as an opening
of new possibilities of thought and experience. Foucault’s aim was not
only to show how the limits of knowledge were constituted, but also to
study what distorted them.
In connection with Foucault’s genealogy, I have argued that this
indeterminacy or ontological contingency characterizes not only discursive structures, but also embodiment and experience. Subjection
sets the limits for normal experiences, but these limits make possible transgressions that affirm the limitlessness of bodies and pleasures.
The Foucaultian body is capable of generating resistance, of presenting excess and transgression, not just malleability. This resistance is
not, however, a return to a wild and natural body, but rather is made
possible by the normalizing power. The body is a construction of scientific discourses and disciplinary technologies, but is also capable of
multiplying, distorting and overflowing its discursive definitions, classifications and coordinations. Even if linguistic intelligibility structures
and partly constitutes experience, there are nevertheless experiences
that fall outside of discourse in the sense that these abject or transgressive experiences are rendered mute and unintelligible in our culture.
fo u c au lt o n freedom
At the same time, they are necessary ‘outsiders’ because they constitute
the limits of the normal and the intelligible. In Foucault’s genealogy,
like in his archaeology, there is a dimension of freedom in the sense of
a constitutive outside to the discursive order, even if there is no outside
to the apparatus or network of practices as a whole.
Practices of freedom. Foucault’s late thinking identifies ethics as the
deliberate dimension of freedom. Ethics is a practice of freedom.
Hence, while freedom in the previous sense is an ontological condition
of ethics, ethics as a practice is the deliberate form it assumes. Foucault’s
thinking on ethics thus develops a fuller understanding of freedom,
elaborates it by introducing a deliberate dimension to it. Freedom is
not only a non-subjective opening of possibilities, it can be deliberately
cultivated and practised by subjects. The subject exercises freedom in
critically reflecting on itself and its behaviour, on beliefs and the social
field of which it is part. It materializes the possibilities that are opened
around it. The practices of freedom may challenge, contest and even
change the constitutive conditions of our subjectivity as well as its actual
forms. Ethics as practices of freedom means exploring possibilities for
new forms of the subject, new fields of experiences, pleasures, relationships, modes of living and thinking. It consists of creative activity as well
as the critical interrogation of our present, and of the contemporary
field of possible experience. The quest for freedom in Foucault’s ethics
is a question of developing forms of the subject that are capable of
functioning as resistance to the normalizing power.
Freedom as the ethos of the Enlightenment. Foucault’s essays on the
Enlightenment put forward the idea of freedom as a historical ideal
originating from it. He does not simply embrace traditional Enlightenment ideals, but submits them for critical reappropriation. What, for
him, characterizes the philosophical ethos originating in the Enlightenment is that it is a permanent critique of our own era. By linking
his thought to the Enlightenment, he makes the normative move of
adopting the ideals associated with it – critical reason and personal
autonomy – as the implicit ground on which his critiques of domination, abusive forms of power and reason rest. The Enlightenment
provides him with the historical – not transcendental – values on which
to base his critiques. The ideal of freedom is a commitment to a specific
historical tradition within which we think about human life and politics.
Freedom is the contingent historical ethos and precondition of critical
reflection on our present.
t he fr eed o m o f philo s op hy
Negative freedom. In his afterword to Dreyfus and Rabinow’s book
Michel Foucault, Beyond Structuralism and Hermeneutics, Foucault defines
power and freedom as very similar to the standard Anglo-American
view. A relationship of power is a set of actions upon other actions.
This means that the one over whom power is exercised is ‘thoroughly
recognized and maintained to the very end as a person who acts; and
that, faced with a relationship of power, a whole field of responses,
reactions, and possible inventions may open up’ (SP, 220). Power presupposes freedom in the sense that to be free means that one has a
field of possibilities in which several ways of behaving can be realized.
‘Power is exercised only over free subjects, and only in so far as they are
free’ (SP, 221). Freedom is not the opposite of power, but is rather its
precondition and permanent provocation.
While the three previous definitions all opposed the humanist idea of
freedom as a characteristic of a human being, and attached to it instead
practices, experiences or language, this fourth definition posits freedom in the standard vocabulary of political philosophy as an attribute
of the subject. This characterization is not typical of Foucault, however.
Tuija Pulkkinen (1996/2000, 85) argues that Foucault’s afterword to
Dreyfus and Rabinow’s book gives a problematic view of his thinking
because it was written for an American audience already engaged in a
debate on what kind of ‘definition of power’ Foucault’s work involves.
Pulkkinen sees in the background of Foucault’s text the urge to meet
the demands of his new audience and this explains why he formulates
the questions as well as the definitions of power in a way that is familiar
to his Anglo-American readers. We might assume that his treatment of
freedom in this text follows a similar motivation.
I argue that, in general, we can conclude that freedom in Foucault’s
philosophy is divorced from the subject understood as a stable entity
and reference point, and is attached instead to practices, forms of experience or the being of language. The understanding of the subject
underlying Foucault’s thinking of ethics and politics is consistent in
this sense with his earlier explicit rejections of the humanist subject.
Foucault sought to develop a way to think of ethics and politics that
does not rely on any ahistorical, ontological assumptions about the
subject. The subject is neither the starting point nor the foundation of
morality, any more than it was of epistemology or history. Throughout
his work Foucault warned us against fixed meanings of what a human
being is. To be consistent, his ethics cannot be built on any foundational
fo u c au lt o n freedom
understanding of the ethical subject, but on the contrary, must aim to
break essences, constants and human natures. Ethics becomes possible exactly through the movement of revealing forms of subjectivity as
contingent and questioning constraining essences.
Foucault’s late understanding of the subject, however, implicitly presupposes a curious autonomy: the subject is capable of engaging in practices of self-reflection, considered self-mastery and creation. It turns
back upon itself, recognizes itself as a subject of a certain morality,
problematizes normalized subjectivity and seeks to create itself a new.
At the same time, and seemingly paradoxically, the subject does not
invent itself, but only deploys modes of behaviour and forms of thinking of its cultural context. It internalizes ‘the exterior’, and by an act of
appropriation turns it into a singular ethical style. This is, furthermore,
a style of problematization, of contesting the normalizing practices formative of the subject. Foucault thus refuses to develop any general and
invariant understanding of the subject of ethics and politics, while at
the same time he locates ethics in the reflexive practices of the self.
The questions that follow are questions about the ‘freedom’ of the
subject. How can we understand the capacity of the subject for critical
self-reflection? How is the constituted subject capable of engaging in
truly critical practices?
In the next chapter I will focus more explicitly on the question of the
subject in Foucault’s ethics. I will explicate two interconnected problems that I claim are riddling Foucault’s understanding of the ethical
subject: the autonomy of the subject and the role of the other in its
constitution. I will show, through a reading of Emmanuel Levinas, how
these problems are interconnected. As Foucault recognized, important
preconditions for the morality of antiquity were that the subjects of that
morality were free and active masters of themselves: slaves and women
had no morality. In other words, moral subjects were always free men.
My criticism will focus on this implicit understanding of the ethical
subject underlying Foucault’s thinking. I will argue that his attempt to
rethink ethics, starting from the ethics of antiquity and its practices of
the self, leaves the meaning as well as the possibility of ethical subjectivity unproblematized. I will pose the question whether we can conceive
of a form of contemporary ethics based on the model of ancient ethics
without problematizing the understanding of ethical subjectivity underlying it. Is the ethical subject fundamentally an active master of the
For men of courage physical sufferings (and privations) are
often a test of endurance and of strength of soul. But there is
a better use to be made of them. For me then, may they not be
that. May they rather be a testimony, lived and felt, of human
misery. May I endure them in a completely passive manner.
Whatever happens, how could I ever think an affliction too
great, since the wound of an affliction and the abasement to
which those whom it strikes are condemned opens to them the
knowledge of human misery, knowledge which is the door to
all wisdom?
(Weil 1952/1997, 31)
Unlike the claim of some of his critics, Foucault’s ethics is not a solitary
pursuit, nor does he prioritize isolated individuality. Ethical subjectivity
is given a form in the practices of the self, but these practices always
take place and derive their meaning from an interpersonal situation.
Care for the self, according to Foucault, implies complex relationships
with others: relationships and duties towards one’s family members,
society at large, one’s spiritual master or guide. The ethos of freedom
and self-mastery can only take concrete shape and become a style of life
in a particular interpersonal situation in which the ethical acts become
ways of dealing with the surrounding community. The self that is cared
for is never isolated, but always linked to larger societal structures.1
Moreover, the ethical relationship always exists between free individuals. When Foucault was asked whether care for the self released
1 Cf. Gros 2001.
fo u c au lt o n freedom
from the care of others ran the risk of absolutizing itself and therefore
becoming an exercise of power on others, he replied:
[T]he risk of dominating others and exercising over them a tyrannical
power only comes from the fact that one did not care for one’s self and
that one has become a slave for his desires. But if you care for yourself
correctly, that is to say if you know ontologically what you are . . . you can
not abuse your power over others.
(EPF, 8)2
Foucault thus considered the ethical practice of caring for oneself as
one way of controlling and limiting abuses of power. ‘It is the power over
self which will regulate the power over others’ (EPF, 8). For the Greeks,
the ethos of freedom was also a way of caring for others: when one cares
for oneself properly this means that one also cares for the other. Ethical
practices provide a way of resisting domination, and ethical conduct will
be a conduct of non-domination.
Thus, Foucault does not advocate blatant egoism, nor does he neglect
the care for others. He does, however, relocate the ethical from interpersonal relationships to the relationship one has with one’s self, and
this way rethinks the realm of ethics. In Foucault’s thought, care for the
self is a precondition of care for others: ‘One must not have the care
for others precede the care for self. The care for self takes moral precedence in the measure that the relationship to self takes ontological
precedence’ (EPF, 7). Foucault understood the relationship to one’s
self to be ontologically primary and thus a precondition for ethical relationships to others. When one cares for one’s self properly this means
that one also cares for the other. However, care for the self is not ethical because it necessarily implies care for others. The meaning of the
ethical does not originate from the other, but from a certain kind of
relationship to one’s self.
This ontological priority given to the relationship to one’s self in
the ethical realm poses serious problems for efforts to think forms of
contemporary subjectivity and ethics. I will argue in this chapter that
we must acknowledge the fundamental importance of the other for
the constitution of ethical subjectivity. Despite the problems and even
2 ‘le risque de dominer les autres et d’exercer sur eux un pouvoir tyrannique ne vient
précisément que du fait qu’on ne s’est pas soucié de soi et qu’on est devenu l’esclave
de ses désirs. Mais si vous vous souciez de vous comme il faut, c’est-à-dire si vous savez
ontologiquement ce que vous êtes . . . vous ne pouvez pas à ce moment-là abuser de votre
pouvoir sur les autres.’ (EPL, 716)
t he o th er
paradoxes entailed by the idea, we must seek to understand the way in
which other people form the precondition for ethical subjectivity. This
means turning around Foucault’s ontological order of primacy: the
most fundamental ethical question does not concern my relationship
to myself, but to the other. This means also that the question of how
to live a good life must be subordinated to the more primary question
which alone can give life an ethical meaning: how do I respond to the
other person?
Ethical subject and the other
It is my contention that the problems with Foucault’s ethics stem from
the conception of ethical subjectivity he adopts from ancient Greece.
The ethical subject was a free man whose main moral concerns were
related to the question of how to live a good and honourable life in
the polis, not of how to respond ethically to the irreducible humanity
of the other. In a culture built on institutional slavery and sexual practices of domination, the moral dilemmas clearly did not focus on the
ethical demand of the other. Rather, the ethical problems posed by the
existence of others were related to questions of how to rightfully govern them. As a moral subject, the free man had to know how to care
properly for his wife, children and slaves.
Timothy O’Leary (2002, 43) argues that Foucault’s reading of classical Greek ethics plays down its political motivations and aims. Classical
ethics was an ethics whose primary concern was with establishing and
maintaining relations of domination with one’s inferiors. The mastery
over oneself was not the goal of this ethics only because it was a way
of giving life more intensity and beauty, it was, most of all, the moral
condition of possibility of one’s mastery over others. As Foucault himself also at times recognized, it was an ethics that was resolutely ‘virile’
in character; if one was a ‘man’ in relation to oneself, one could be
a ‘man’ in relation to others. Self-mastery (enkrateia), both as a theme
and as a social practice, was inseparable from the mastery of others.
This isomorphism between self-mastery and mastery of others was particularly striking in the field of sexuality. This entire field was governed
by the primary oppositions active–passive, penetrator–penetrated and
dominator–dominated (O’Leary 2002, 62).
The whole ethical framework and conception of ethical subjectivity
is thus, in significant respects, foreign to our modern understanding
of ethics. While this is clearly what Foucault found appealing about it
fo u c au lt o n freedom
in certain respects – the complete lack of normalizing and moralizing
technologies of the self, for example – there are, nevertheless, other
fundamental problems connected with it that he was not able to solve.
He himself asked, in connection with Greek sexual ethics: ‘How can we
have an ethics of acts and their pleasures which would be able to take
into account the pleasure of the other?’ (GE, 346). This question is, I
argue, perhaps asked, but no attempt is made to answer it in Foucault’s
thought. Although the subject is always constituted in an intersubjective and relational matrix, the constitutive community of others is an
impersonal power/knowledge network. The personal other of the ethical relation – the sexual or ethical partner – is not given a decisive role,
but instead rendered almost invisible and more or less irrelevant.
In Foucault’s thought, an ethical relationship becomes possible
when the subject has already been constituted as an ethical subject by
him/herself through practices of the self. An ethical relationship to the
other follows from ethical subjectivity rather than from being constitutive of it, and the other becomes contingent and exchangeable. As long
as I have given my subjectivity a beautiful, courageous or honourable
form and my ethical subjectivity is intact, I can relate to anyone in any
circumstance in an ethical way. However, can a relationship be ethical if
the personal other involved is contingent and not fundamentally constitutive of it? What kind of ethical relationship is not constituted by the
singular, personal other who makes the relationship what it is? How or
why should ethics mean anything at all without a personal other? How
or why does anybody give an ethical meaning to the task of caring for
The obvious place to look for an answer to these questions seems to
be in Levinas, who has studied the question of the ethical relationship
to the other to an extent that is perhaps incomparable in western philosophy. However, to stage a dialogue between Foucault and Levinas also
seems an impossible task. The incommensurability of these two thinkers
derives from the originality of their projects and their distinctive views
of the role and scope of philosophy. Engaging in a dialogue with
Levinas also seems impossible simply because of the notorious obscurity of his language, and would, by definition, require that one speaks
his language. Levinas’ aim of breaking with the tradition of western
philosophy and its language of ontology means that his own language
is pushed to the limits of incomprehensibility.
Ethics, furthermore, means very different things in Levinas’ and
Foucault’s thought. For Levinas, it is first philosophy. It is the condition
t he o th er
of possibility of a morality and frames the questions of how to live, think,
act and treat others. It is not the study of values, virtues, rights or duties.
It is not a distinct area of philosophy, not even an object of cognitive
contemplation at all, but more like a passive opening to alterity. It precedes me: my choices, will and reason. I cannot choose it or practise it,
it chooses and possesses me. The fundamental experience of the other
constitutes ethics and precedes my subjectivity. For Foucault, on the
other hand, ethics refers to a component of morality that consists of
one’s relationship to one’s self. A study of ethics means a study of the
different modes by which subjects problematize their relationship to
different moral codes in different historical situations.
Apart from the obvious incommensurability of the linguistic styles
and their understanding of philosophy, the ethical frameworks of these
two thinkers seem to be almost diametrically opposed. While Levinas’
thought also builds upon the philosophical tradition of antiquity, it is,
importantly, influenced and structured by the Judeo-Christian tradition. Religious images and symbols intertwine with rational concepts
and categories.3 While Foucault’s model for the ethical subject is built
on the figure of a free man of antiquity aspiring to activity, virility, mastery and honour, Levinas’ text carries the ideals present in the Bible of a
people driven into slavery: infinite responsibility, humbleness and subordination to the other. Levinas explicitly opposes the conception of
subjectivity of the philosophical tradition originating in ancient Greece.
He writes: ‘The rational subjectivity bequeathed to us by Greek philosophy . . . does not feature that passivity which . . . I have identified with
the responsibility for the other’ (Levinas 1989, 206).4
The responsibility that, for Levinas, characterizes or even constitutes
ethical subjectivity has to be understood in a specific sense, however.
While I will suggest here that the question of ethical subjectivity is
an important area of critical exchange between Foucault and Levinas,
the question of responsibility of the subject does not constitute a simple contrast between these two thinkers. Barry Smart (1997) takes up
the question of responsibility in Foucault’s and Levinas’ thought and
argues that, despite Foucault’s later emphasis on the subject’s ethical
self-constitution, there remains a significant absence in his discussion
of the subject of any consideration of the question of moral responsibility (Smart 1997, 83). Preoccupation with the self leaves open the
3 On Levinas’ religious writings, see e.g. Levinas 1963, 1968.
4 The central idea of responsibility in Levinas’ thought is, according to him, also a central
feature of Judaism (see e.g. Levinas 1989, 264).
fo u c au lt o n freedom
question about caring or taking responsibility for others. Smart turns
to Levinas’ philosophy and advocates an understanding of ethics based
on moral responsibility for others. He argues that it is ‘from the initial
moral bearing of being, taking or assuming responsibility for the other
that a particular ethical practice of caring for the self follows’ (87).
Smart demands an ethically responsible agent for Foucault’s practices
of the self in the sense of an individual who takes seriously his/her
responsibility for the other. He assumes the subject of morality to be
the modern subject characterized by autonomy and rationality, who
then adopts responsibility as a subsequent act or attitude. Smart ends
his article on Foucault’s ethics by accusing Foucault of irresponsible
egoism. He asks if ‘the subject of responsibility for others has become
a subject of indifference in our time?’ (91).
Levinas’ notion of responsibility – like most of his concepts – does
not, however, translate directly into the standard language of moral
philosophy. Responsibility is not an ethical attitude for Levinas, as it is
for Smart, but it is, rather, a fundamental structure of subjectivity. For
Levinas, ethics is responsibility that is anterior to any volition. Responsibility is irreducible, something one cannot deny or shed, but neither
can one take or assume it. I do not take responsibility, rather in the
encounter with the other, responsibility falls upon me. I am passive in
its regard and my humanity consists of this passivity – of not being able
to shed fundamental responsibility. For Levinas, responsibility is thus
not an ethical attitude that we could choose or promote in others. By
emphasizing the subject’s absolute responsibility for the other, Levinas
does not ‘place the emphasis firmly and deliberately on care for others,
rather than care for the self’ (Smart 1997, 89). Despite the absolute
and fundamental responsibility for the other, people kill and torture
each other, not to mention polluting the environment and letting their
fellow men go hungry. It is always possible to relate to the other as to an
object, to oppress, exploit and even kill him or her. Even then, however,
according to Levinas, I am not free from my responsibility for him, the
relationship of responsibility can never be suppressed. Even in his or
her death, my responsibility for the other is not resolved.
Levinas’ contribution to rethinking ethics and ethical subjectivity in
particular is thus not in the simple advocation of responsibility and
condemnation of irresponsible egoism, nor is this where his thought
contrasts with that of Foucault. The domain of interesting exchange is,
in my view, in their understanding of the constitution of ethical subjectivity and its relationship to radical alterity. In my argumentation
t he o th er
for this I will next briefly explicate Levinas’ conception of ethical
Levinas’ account of subjectivity is modified throughout his writings,
and like most of his notions is stretched in his late texts to the limits
of comprehension, to a point where it almost vanishes. Indeed, it can
be argued that there is no subject in Levinas.5 Ethics for him is beyond
the realm of constitutional subjectivity, and therefore puts the whole
conception of subjectivity into question. The subject of ethics is not
the source of meaning and action. It is being that does not express
itself, being that cannot be known or understood but only assigned as
responsible for the other. Any explication of a single idea or concept
in Levinas’ work, furthermore, always contorts the movement of his
thought. My aim here is not to provide a comprehensive explication of
his conception of subjectivity, however, but simply to take up some central features of it that enable me to articulate the problems in Foucault’s
understanding of the ethical subject.6
Subjectivity as passivity
In his late work Otherwise than Being, or Beyond Essence, Levinas sets out
to do the impossible: to articulate, using the language of philosophy,
the inevitable silence of philosophy. He seeks to describe in language
what by definition is unthematizable, a sphere that cannot be an object
of knowledge, understanding or any other intentional act. To write
about ethics means to describe what is beyond being and non-being,
the ‘otherwise than being’. Ontology traditionally refers to an area of
philosophy that studies being in all its forms and modes. It seeks to
comprehend being and thus to bring it into the realm of knowledge.
For Levinas, however, ethics is beyond being and therefore beyond
ontological inquiry. It is not a relation of knowledge, but it is in direct
opposition to ontology.7 However, before philosophy can describe what
5 See e.g. Bailhache 1994.
6 My presentation of Levinas’ understanding of subjectivity is based mainly on his late
thought, particularly Otherwise than Being or Beyond Essence (1974/1981). There are significant changes as well as shifts of emphasis between this book and Totality and Infinity
(1961/1969), for example, on the question of the ethical significance of sensibility, on
the role of language and on the emphasis on justice and the third party. Discussion of
the changes is, however, beyond the scope of this study. See e.g. Critchley 1992, Peperzak
7 Ontology is Levinas’ general term for any relation to otherness that is reducible to
comprehension or understanding (Critchley 2002, 11).
fo u c au lt o n freedom
lies beyond being – what is otherwise than being – it must give up the
language of ontology without sinking into incomprehensibility.
An important distinction in Levinas’ thought is therefore the one
he makes in Otherwise than Being between saying (le dire) and the said
(le dit).8 The language of philosophy has traditionally consisted only of
the said: of sentences and arguments that have a truth-value. Everything
that can be named, discussed and debated can only be expressed in the
said. Philosophy has therefore neglected the other aspect of language,
the saying. By saying, Levinas refers to speech as aimed at the other.
Speaking to the other – addressing and responding – is the condition
of possibility of all language and philosophy. There can be no language
without the other, because words are always for the other. Saying is thus
similar to Levinas’ idea of responsibility: it is a relationship of openness
and vulnerability to the other which cannot be thematized or brought
to the said.
The only way to approach the saying in philosophy is through the
said. However, the effort to thematize saying in the said is always, by
necessity, an act of its destruction: to succeed in thematizing what
cannot be thematized, paradoxically, means to fail. Philosophy is thus
doomed to this recurring failure, and therefore has to start anew time
and time again. Hence, it is necessary to do the impossible in order
to write about ethics. The ethical relation to the other will always be
beyond the language of philosophy, rationality, totality, order – and
being. ‘Ethics is not a moment of being; it is otherwise than being, the
very possibility of the beyond’ (Levinas 1989, 179).
This fundamental paradox in philosophical language underlies the
description of subjectivity in Otherwise than Being. Ethical subjectivity
is described in the language of philosophy, which is, nevertheless, constantly undoing itself: subjectivity is equated with passivity, vulnerability,
sensibility, maternity, materiality, responsibility and substitution. These
terms may each express something about it, but it cannot be reduced
to any one of them. Levinas writes, for example, that ethical subjectivity is essentially passivity, but this is not passivity in the sense in which
we normally understand it, as the opposite of activity. Subjectivity as
ultimate passivity does not belong to the order in which the alternative
of active/passive retains its meaning: ‘Our western passivity’ refers to
8 Otherwise than Being is often read as Levinas’ attempt to address the questions Derrida
posed in ‘Violence et metaphysique’ (1967/1987). On Derrida’s relationship to Levinas’
thought, see e.g. Bernasconi 1991, Critchley 1991, 1992.
t he o th er
a subject who is passive when he does not give himself the contents
of perceptual and cognitive acts. This passivity is not passive enough
because the subject is still receptive. Sensations are produced in a subject who grasps himself through these sensations and conceives them.
Levinas seeks a new degree of passivity, ‘more passive than passivity’, that
does not take charge of itself, but that breaks the unity of subjectivity.
The passivity of subjectivity ultimately means its fission (Levinas 1986/
1998, 89).
The radical understanding of subjectivity as responsibility prior to
commitment is also developed further. Levinas said in an interview:
‘In this book [Otherwise than Being] I speak of responsibility as the
essential, primary and fundamental structure of subjectivity’ (Levinas
1982/1985, 95). Responsibility is the absolute principle of becoming a human subject, and as such is the primordial structure of subjectivity. Through substitution, literally putting myself in the place of
the other, I become responsible even for the other’s responsibility, I
become a hostage for the other. The possibility of putting oneself in
the place of the other is a condition of possibility for solidarity and
ethical behaviour. ‘It is through the condition of being a hostage that
there can be in the world pity, compassion, pardon and proximity –
even the little there is, even the simple After you, sir’ (Levinas 1974/
1981, 117).
Responsibility for the other is not, however, only a demand, a weight
on the shoulders. It is also the freedom that constitutes the subject’s
uniqueness. Levinas had already repudiated the modern humanist idea
of the essential similarity and equality of subjects in his book Time and
the Other (1947/1987). My responsibility constitutes my singularity and
uniqueness because no one can carry my responsibility for me. ‘No one
can substitute himself for me, who substitutes myself for all’ (Levinas
1989, 115). My selfhood comes into being through my responsibility
for the other. The inescapable responsibility makes me an individual I:
to be myself can only mean being for the other.
The identity of the subject is thus determined by the uniqueness of
his or her responsibility for everybody and everything. Responsibility is
fundamental and yet impossible. I cannot shed its demand/command
because it is a fundamental structure of my being, but neither can I
ever fulfil it: the more just I am, the greater is my responsibility. I am
responsible for that which has preceded me and that which will outlive
me. No one else can take my place and carry my responsibility, which
is always more than anyone else’s. ‘Responsibility for the neighbour is
fo u c au lt o n freedom
precisely what goes beyond the legal and obliges beyond contracts; it
comes to me from what is prior to my freedom, from a non-present, an
immemorial’ (Levinas 1989, 180).
Like the phenomenological subject, the subject for Levinas is always
the singular I. Husserl’s subject is singular for methodological reasons, but for Levinas the singularity is integral for his understanding
of ethics.9 The ethical demand can only be placed upon me. Ethical
subjectivity cannot be objectified or studied as generic. Responsibility,
as the assignment of the other, always arises in a singular and particular
situation and concerns only me.
Unlike the phenomenological subject, the subject that Levinas
aims at describing does not constitute the world in perceptions or
any meaning-giving acts. An ethical relation is not like a relation of
knowledge; it cannot be reduced to knowledge about the other. The
other cannot be posited by any constituting, intentional act of the subject, because he or she cannot be an intentional object, but always overflows the limits of perception and comprehension. This constitutes the
paradox of the presence of alterity in a finite act of a self-possessed
subject. To encounter something truly other is, by definition, impossible, because it would be incomprehensible and unexperienceable.
For something to be able to preserve its alterity means that it must
exceed my categories of experience and understanding, and therefore
be a non-experience, unexperienceable. This core problem in Levinas’
thought is also already articulated in the early work Time and the Other:
‘How can an event that cannot be grasped still happen to me? What
can the other’s relationship with a being, an existent, be? . . . How can a
being enter into a relationship with the other without allowing its very
self to be crushed by the other?’ (Levinas 1947/1987, 77). It is this
paradox that founds subjectivity, which is not reducible to consciousness or to any kind of intentionality. Subjectivity as passivity does not
constitute the other through meaning-giving acts, but the other breaks
up the unity of transcendental constitution.
The paradoxical understanding of subjectivity becomes even more
pronounced in Otherwise than Being. The reason the subject cannot constitute the other is not only because the other as radical alterity always
overflows the limits of experience, but also because the subject itself
9 Although Levinas’ conception of the subject is often read as a critique of Husserl’s
transcendental Ego, it can also be understood as its radicalization. (See e.g. Levinas
1987/1994, 151–8.) For an illuminative analysis of the differences between Husserl’s
and Levinas’ thought, see e.g. Bernet 1998.
t he o th er
does not exist as an ethical subject prior to the encounter. As Simon
Critchley notes (2002, 12), Levinas does not posit, a priori, a conception
of ethics that then instantiates itself in certain concrete experiences. It
is, rather, that the ethical is an adjective that describes, a posteriori as it
were, a certain event of being in a relation to the other irreducible to
comprehension. The ethical relationship to the other is described in
terms of a relationship of proximity, extreme closeness to the point of
obsession. It is a relationship without the mediation of language, any
principle, any ideality: ‘To thematize this relationship is already to lose
it’ (Levinas 1989, 110). It is this proximity to the other that constitutes
ethical subjectivity.
Alphonso Lingis notes (1981, xvi) that embodiment becomes central in the descriptions of the ethical relationship in Otherwise than Being.
The ethical relationship acquires, if not an erotic, then a sensuous character. Though realized in language, it is described as sensuous contact
and closeness. Because the alterity of the other cannot be grasped by the
subject in any intentional act, contact can only be through sensibility,
understood as susceptibility and sensuality but also as pain. In extreme
pain the subject loses the mastery of its world: rational consciousness,
dignity, even perceptual reality. Pain reveals a level of subjectivity at
which intentionality has broken down, and the subject of rational consciousness has been reduced to destitution and subjected to what is
beyond it. The extreme passivity of ethical subjectivity is thus only possible because of embodiment. The subject of ethics is essentially an
embodied, carnal subject, a subject capable of being hurt, wounded,
shamed and humiliated. Only out of this bodily vulnerability can ethics
arise as responsibility for the other, passivity in the face of overwhelming
pain and pleasure.
Hence, Levinas’ aim is to describe subjectivity prior to its ontological constitution as an entity among others. Ethical subjectivity cannot
be defined by consciousness, rationality or intentionality, but precedes
them. Since the ethical subject is not an entity or a being thematizable
in language, we cannot even call it a subject. It has no stable form, but
is constituted time and again in encounters with the other, in openness
to the other, in responsibility for the other. The unthematizable relationship to alterity is fundamentally constitutive of ethical subjectivity.
It is what is anterior or beyond the constituting subject and what makes
individual constitution possible. ‘Ethics is when not only do I not thematize another; it is when another obsesses me or puts me in question’
(Levinas 1986/1998, 99).
fo u c au lt o n freedom
The other as precondition of ethics
It is my contention that from the point of view of Foucault’s ethics, Levinas’ understanding of ethical subjectivity puts forward important ideas
that are overlooked by Foucault. I will summarize them here in simple
terms. Ethical situations are always absolutely singular, and ethical subjectivity is constituted anew in each encounter with the singular other.
The ethical relationship is thus constitutive of its terms, rather than
presupposing that they are already constituted before entering into it.
In the encounter with the other, my ethical subjectivity as responsibility
arises through my ability to put myself in the place of the other, or in
Levinas’ terms, through becoming a hostage for the other. From this
fundamental structure of my subjectivity arises my singular responsibility for the other and also the possibility for ethical behaviour towards
him or her. The other is thus fundamentally constitutive of ethical subjectivity simply by virtue of his or her existence.
Moreover, the other is always radically incomprehensible to me.
Despite being able to put myself in his or her place, I also know that he
or she has a unique way of experiencing the world that I can never completely grasp. Unlike everything else in the world, the other presents
a radical limit to what I can experience. This basic phenomenological
insight can be expressed simply by noting that I can never experience
the world exactly the way the other experiences it. This idea has radical
ethical consequences in Levinas’ thought, however. The other introduces radical alterity and plurality into my constituted world and thus
deprives me of my sovereignty of it. As an ethical subject I am essentially passive and vulnerable, because the other puts my experiences
into question.
My aim in briefly turning to Levinas is thus to show that his thought
studies the fundamental conditions of possibility of ethical subjectivity
in the other that are overlooked by Foucault. The ethical subject cannot
be determined by an exterior power/knowledge network, but neither
can it be ‘set free’ by complementing it with an interior, a capacity for
self-reflection. Only the other can give ethical meaning to the practices
of the self. Alone on a desert island I do not have ethical problems
and I cannot constitute myself as an ethical subject even though I may
continue to engage in practices of the self. To problematize one’s place
in the world and to engage in any kind of practice will only be ethically
meaningful because of the fundamental encounter with the other. In
short, for there to be ethics there has to be a more fundamental level
t he o th er
of subjectivity than the one that Foucault describes in his analyses of
the third axis of its constitution. The fundamental encounter with the
other makes ethical questions possible.
If we turn to Levinas in seeking to explicate the preconditions of
ethical subjectivity, this does not mean that we have to compromise
Foucault’s aim of breaking human essences, however. For Levinas, subjectivity as passivity or responsibility does not mean a universal essence
or a fixed core of a human being. It is not something that can be posited
as an entity, foundation or a principle for a certain kind of morality.
Ethical subjectivity as responsibility is fundamental and primordial, but
it is never generic. It is constituted anew in every singular encounter
with the other. Like Foucault, Levinas repudiates the humanist understanding of the subject, but for different reasons. In Otherwise than Being
Levinas writes:
Modern antihumanism, which denies the primacy the human person,
free and for itself, would have for the signification of being, is right over
and beyond the reasons it gives itself. It clears the place for subjectivity
positing itself in abnegation, in sacrifice, in a substitution which precedes
the will. Its inspired intuition is to have abandoned the idea of a person,
goal and origin itself, in which the ego is still a thing because it is still a
being. Strictly speaking, the other is the end; I am a hostage, a responsibility and a substitution supporting the world in the passivity of assignation,
even in an accusing persecution, which is undeclinable. Humanism has
to be denounced only because it is not sufficiently human.
(Levinas 1974/1981, 127–8)10
From a Levinasian perspective it would thus seem that while Foucault
managed to ‘clear the place’ of problematic humanist conceptions of
the subject, he was not able to find an alternative understanding of
ethical subjectivity that would still make ethics meaningful. A reflexive
and critical relationship to one’s self can be constitutive of an aesthetical
style of living, but only a relationship to the other can give it an ethical
10 ‘L’antihumanisme moderne, niant le primat qui, pour la signification de l’être,
reviendrait à la personne humaine, libre but d’elle-même, est vrai par-delà les raisons
qu’il se donne. Il fait place nette à la subjectivité se posant dans l’abnégation, dans le
sacrifice, dans la substitution précédant la volonté. Son intuition géniale consiste à avoir
abandonné l’idée de personne, but et origine d’elle-même, où le moi est encore chose
parce qu’il est encore un être. A la rigueur autrui est “fin”, moi je suis otage, responsabilité et substitution supportant le monde dans la passivité de l’assignation allant
jusqu’à la persécution accusatrice, indéclinable. L’humanisme ne doit être dénoncé
que parce qu’il n’est pas suffisamment humain.’ (Levinas 1974, 203)
fo u c au lt o n freedom
Recognizing the fundamental role of the other in the constitution of
ethical subjectivity can also untangle another set of problems riddling
Foucault’s late understanding of the subject, namely the question of
encountering limits. For Foucault, ethics becomes possible because of
the subject’s act of turning back upon him/herself: the subject maintains a critical distance to him/herself and participates in the constitution of moral, sexual and other forms of subjectivity. This critical
distance in the act of self-reflection is what constitutes the subject’s
relative autonomy, and critical thinking becomes one of the concrete
forms that ethics takes. It is through considered practices of freedom
that normalized subjectivity is contested and a space for the ethical is
created. This late understanding of reflexive subjectivity raises a host
of questions. The subject that engages in ethical practices is a subject
constituted and embedded in power/knowledge networks. How can
the constituted subject engage in truly critical thought questioning its
own constitutive processes?
Even if we grant that the subject constituted in power/knowledge
networks is not totally determined by them but is capable of critical
self-reflection, we must still ask in connection with ethics, how it can, by
turning back upon itself, find anything but a constituted and normalized interior? The form that ethical subjectivity assumes in Foucault’s
thought is that of a subject who engages in practices of creative selfformation. Ethics implies freedom in the sense of conscious practices
contesting the borders of the habitual. But how is a subject embedded
in the power/knowledge network going to encounter these borders?
How can the subject engage in anything truly different, anything that
would break or exceed the normalized self? How is the subject able
to encounter something radically other through self-reflection: find
different ways of being a subject?
David Halperin argues that even though Foucault himself did not
explicitly thematize the impersonal character of the self, his late writings
nevertheless imply an impersonal understanding of it. It was the late
antique philosophers, whom Foucault studied, who identified the self
with the soul, which was not a principle of individuation but rather an
errant particle of the Divine. The self, in the ancient philosophical view
of it, was not the locus of a unique and private psychological depth in
the model of humanism, but the site of radical alterity: the space within
each human being where she or he encounters the not self, the beyond
(Halperin 1995, 74–6). Halperin claims that the dimension of the self
that makes it a site of irreducible alterity nowadays is no longer the
t he o th er
divine spark that dwells within it, as it was for the ancient philosophers,
but rather the subject’s determination by history. ‘It is no longer divinity
but history that guarantees us an experience of the Other at the core
of our own subjectivity and brings it about that any direct encounter
with the self must also be a confrontation with the not-self’ (Halperin
1995, 104). The study of history becomes a spiritual exercise when it is
conducted as an inquiry into our own alterity (Halperin 1995, 105).
If we accept the Levinasian idea that ethical subjectivity is constituted
by the other, understood as the other person, I suggest that we do not
necessarily need to study history to encounter our own alterity. We
can also think that an encounter with other people will bring about a
confrontation with the not-self, but also an experience of the other as
passivity, responsibility and vulnerability at the core our subjectivity.
The other as radical alterity importantly opens the constituted subject to what it is not, to what it cannot grasp, possess or know. The arts of
existence aiming to transgress normalized individuality would succeed
in opening up an ethical sphere exceeding totality and determination
because the other is capable of introducing alterity to the constituted
subject. The other makes ethical subjectivity possible, but also breaks
the totality of constituted experience by introducing a plurality in being
that resists all efforts of totalization and normalization. Only the other
ultimately reveals the limits of subjectivity and gives the attempts to
transgress them an ethical meaning.
For Foucault, freedom refers to the indeterminateness of the constitutive matrix and to the contingency of all structures. It is the virtual
fractures that appear in the invisible walls of our world, the opening up
of possibilities for seeing how that which is might no longer be what it
is. Freedom does not mean that everything is possible, but neither is
the present a necessity. Foucault writes:
I would like to say something about the function of any diagnosis concerning the nature of the present. It does not consist in a simple characterization of what we are but, instead – by following lines of fragility in
the present – in managing to grasp why and how that-which-is might no
longer be that-which-is.
(CT/IH, 36)1
Even though freedom for Foucault is thus not an attribute of the subject, this does not commit us to political apathy and cynicism. Freedom
is anarchic in the sense that it disturbs and even breaks every totality, but it must nevertheless not be understood as some absolute and
mystical outside. The virtual fractures for thinking and being otherwise will not just appear in the invisible walls of our world, they can
only emerge from our practices. We must try to open up possibilities
1 ‘Ce que je voudrais aussi dire à propos de cette fonction du diagnostic sur ce qu’est
aujourd’hui, c’est qu’elle ne consiste pas à caractériser simplement ce que nous sommes,
mais, en suivant les lignes de fragilité d’aujourd’hui, à parvenir à saisir par où ce qui est et
comment ce qui est pourrait ne plus être ce qui est. Et c’est en ce sens que la description
doit être toujours faite selon cette espèce de fracture virtuelle, qui ouvre un espace de liberté, entendu comme espace de liberté concrète, c’est-à-dire de transformation possible.’
(SEPS, 448–9)
c onc l us i o n : fr eed o m as an o perat i ona l c onc e p t 209
for seeing to what extent that which is might no longer be what it is.
Foucault continues: ‘In this sense, any description must always be made
in accordance with these kinds of virtual fracture which open up the
space of freedom understood as a space of concrete freedom, i.e., of
possible transformation’ (CT/IH, 36).
I will conclude by putting forward one more definition of freedom
that does not directly emerge from Foucault’s thinking, but which nevertheless, in my view, captures something essential about it: freedom
as an operational concept. According to dictionaries of philosophy, an
operational definition is the characterization of a concept through the
operations performed to check it, such as the characterization of weight
as that which scales measure and intelligence as that which IQ tests measure. Freedom as an operational concept would thus mean that freedom
is defined and gains a meaning only through the concrete operations
through which its existence is tested. It emerges through the particular,
political and/or personal struggles that try and test its limits, possibilities or extent. Foucault writes about philosophy that it is important in
that it should be put to the test of contemporary reality:
But if we are not to settle for the affirmation or the empty dream of
freedom, it seems to me that this historico-critical attitude must also be
an experimental one. I mean that this work done at the limits of ourselves
must, on the one hand, open up a realm of historical inquiry and, on the
other, put itself to the test of reality, both to grasp the points where change
is possible and desirable, and to determine the precise form this change
should take.
(WE, 46)2
Freedom can only gain meaning through our practices of resistance and
the fleeting experiences of liberation resulting from them, both collective and personal. It is always dangerous and precarious. Sometimes its
testing turns into riots and violence, and what emerges is not freedom
but anger and resentment. Sometimes it results in nothing but dry pages
filled with exercises of common sense. Freedom is a fragile moment, a
2 ‘Mais pour qu’il ne s’agisse pas simplement de l’affirmation ou du rêve vide de la liberté, il
me semble que cette attitude historico-critique doit être aussi une attitude expérimentale.
Je veux dire que ce travail fait aux limites de nous-mêmes doit d’un côté ouvrir un domaine
d’enquêtes historiques et de l’autre se mettre à l’épreuve de la réalité et de l’actualité, à la
fois pour saisir ler points où le changement est possible et souhaitable et pour déterminer
la forme précise à donner à ce changement.’ (QL1, 574)
fo u c au lt o n freedom
virtual fracture, which seldom endures the test of contemporary reality,
but which nevertheless only emerges from it. It is a meaningless, empty
word without the permanent testing of its possible forms, without ‘the
patient labour giving form to our impatience for freedom’ (WE, 50,
trans. modified).
Ashbery, John (1986), ‘Introduction: On Raymond Roussel’, in Michel Foucault,
Death and the Labyrinth: The World of Raymond Roussel. New York: Doubleday,
Bachelard, Gaston (1934/1949), Le nouvel esprit scientifique. Paris: Presses
Universitaire de France.
(1940/1949), La philosophie de non, essai d’une philosophie du nouvel esprit scientifique. Paris: Presses Universitaire de France.
Bailhache, Gérard (1994), Le sujet chez Emmanuel Levinas. Fragilité et subjectivité.
Paris: Presses Universitaires de France.
Beauvoir, Simone de (1949/1988), The Second Sex. London: Pan Books.
Benhabib, Seyla (1986), Critique, Norm and Utopia. A Study of the Foundations of
Critical Theory. New York: Columbia University Press.
(1987), ‘The Generalized and the Concrete Other: The Kohlberg–Gilligan
Controversy and Feminist Theory’, in Seyla Benhabib and Drucilla Cornell
(eds.), Feminism as Critique. Essays on the Politics of Gender in Late Capitalist
Societies. Cambridge: Polity Press, 77–95.
Bernasconi, Robert (1991), ‘Scepticism in the Face of Philosophy’, in Robert
Bernasconi and Simon Critchley (eds.), Re-reading Levinas. London: Athlone
Press, 149–61.
Bernauer, James and Mahon, Michel (1994), ‘The Ethics of Michel Foucault’,
in Gary Gutting (ed.), The Cambridge Companion to Foucault. Cambridge:
Cambridge University Press, 141–58.
Bernet, Rudolf (1998), ‘Encounter with the Stranger: Two Interpretations of the
Vulnerability of the Skin’, in Ernest Wolfgang Orth and Chain-Fai Cheung
(eds.), Phenomenology of Interculturality and Life-World. Freiburg: Verlag Karl
Alber, 89–111.
Bigwood, Carol (1991), ‘Renaturalizing the Body’, Hypatia 6.3, 54–72.
Bordo, Susan (1989), ‘The Body and the Reproduction of Femininity: A Feminist
Appropriation of Foucault’, in Allison Jaggar and Susan Bordo (eds.), Gender/
Body/Knowledge. New Brunswick: Rutgers University Press, 13–33.
(1993), ‘Feminism, Foucault and the Politics of the Body’, in Caroline
Ramazanoglu (ed.), Up Against Foucault. London and New York: Routledge,
referen c es
Boothroyd, David (1996), ‘Foucault’s Alimentary Philosophy: Care of the Self
and Responsibility for the Other’, Man and World 29, 361–86.
Braidotti, Rosi (1991), Patterns of Dissonance: A Study of Women in Contemporary
Philosophy. Cambridge: Polity Press.
(1994), Nomadic Subjects, Embodiment and Sexual Difference in Contemporary Feminist Thought. New York: Columbia University Press.
Brown, Wendy (1998), ‘Genealogical Politics’, in Jeremy Moss (ed.), The Later
Foucault: Politics and Philosophy. London: Sage, 33–49.
Bruzina, Ronald (1995), ‘Translator’s Introduction’, in Eugene Fink, Sixth
Cartesian Meditation, The Idea of a Transcendental Theory of Method. Bloomington: Indiana University Press, vii–lxviii.
Butler, Judith (1989), ‘Sexual Ideology and Phenomenological Description’, in
Jeffner Allen and Iris Marion Young (eds.), The Thinking Muse. Feminism
and Modern French Philosophy. Bloomington: Indiana University Press,
(1990), Gender Trouble: Feminism and the Subversion of Identity. London and
New York: Routledge.
(1993), Bodies that Matter: On the Discursive Limits of Sex. London and New York:
(1997), The Psychic Life of Power: Theories in Subjection. Stanford: Stanford University Press.
Canguilhem, Georges (1994a), A Vital Rationalist: Selected Writings from Georges
Canguilhem, ed. François Delaporte. New York: Zone Books.
(1994b), ‘The Death of Man, or Exhaustion of the Cogito?’, in Gary Gutting
(ed.), The Cambridge Companion to Foucault. Cambridge: Cambridge University
Press, 71–91.
Carr, David (1999), The Paradox of Subjectivity. Oxford: Oxford University Press.
Cavaillès, Jean (1947/1994), Sur la logique et la théorie de la science, in CEuvres
complètes de philosophie des sciences. Paris: Hermann, 473–560.
Caws, Peter (1988), Structuralism. The Art of the Intelligible. New Jersey: Humanities
Cohen, Paul M. (1997), Freedom’s Moment: An Essay on the French Idea of Liberty from
Rousseau to Foucault. Chicago: University of Chicago Press.
Cousins, Mark and Hussain, Athar (1984), Michel Foucault. London: Macmillan.
Critchley, Simon (1991), ‘Bois – Derrida’s Final Word on Levinas’, in Robert
Bernasconi and Simon Critchley (eds.), Re-reading Levinas. London: Athlone
Press, 162–89.
(1992), The Ethics of Deconstruction. Levinas and Derrida. Oxford: Blackwell.
(2002), ‘Introduction’, in Simon Critchley and Robert Bernasconi (eds.),
The Cambridge Companion to Levinas. Cambridge: Cambridge University Press,
Crossley, Nick (1994), The Politics of Subjectivity: Between Foucault and Merleau-Ponty.
Aldershot: Avebury Press.
Davidson, Arnold (1986), ‘Archaelogy, Genealogy, Ethics’, in David Couzens Hoy
(ed.), Foucault: A Critical Reader . Oxford: Blackwell, 221–33.
(1994), ‘Ethics as Aesthetics: Foucault, the History of Ethics, and Ancient
Thought’, in Gary Gutting (ed.), The Cambridge Companion to Foucault.
Cambridge: Cambridge University Press, 115–40.
referen c es
(1997a), ‘Introductory Remarks to Pierre Hadot’, in Arnold I. Davidson (ed.),
Foucault and his Interlocutors. Chicago: University of Chicago Press, 195–
(1997b), ‘Structures and Strategies of Discourse: Remarks Towards a History
of Foucault’s Philosophy of Language’, in Arnold I. Davidson (ed.), Foucault
and his Interlocutors. Chicago: University of Chicago Press, 1–17.
Deleuze, Gilles (1986/1988), Foucault. London: Athlone Press.
(1994), ‘Désir et plaisir’, Magazine Littéraire 325, 58–65.
(1994/1997), ‘Desire and Pleasure’, in Arnold I. Davidson (ed.), Foucault and
his Interlocutors. Chicago: University of Chicago Press, 183–92.
Deleuze, Gilles and Felix Guattari (1980/1987), A Thousand Plateaus: Capitalism
and Schizophrenia. Minneapolis: University of Minnesota Press.
Derrida, Jacques (1962/1989), ‘Introduction to The Origin of Geometry’, in
Edmund Husserl, The Origin of Geometry. Lincoln: University of Nebraska
Press, 25–153.
(1967/1978), Writing and Difference. London: Routledge & Kegan Paul.
Deveaux, Monique (1996), ‘Feminism and Empowerment: A Critical Reading of Foucault’, in Susan J. Hekman (ed.), Feminist Interpretations of
Michel Foucault. University Park: Pennsylvania State University Press, 211–
Dews, Peter (1989), ‘The Return of the Subject in Late Foucault’, Radical Philosophy 51, 37–41.
Diamond, I. and Quinby, L. (eds.) (1988), Feminism and Foucault: Reflections on
Resistance. Evanston, Il.: Northeastern University Press.
Dillon, Martin (1988), Merleau-Ponty’s Ontology. Bloomington: Indiana University
Dosse, François (1991), Histoire du structuralisme, vol. i, Le champ du signe, 1945–
1966. Paris: Éditions le Découverte.
(1992), Histoire du structuralisme, vol. ii, Le chant dy cygne, 1966 à nos jours. Paris:
Éditions le Découverte.
Dreyfus, Hubert L. and Rabinow, Paul (1982), Michel Foucault, Beyond Structuralism
and Hermeneutics. Hemel Hempstead: Harvester Wheatsheaf.
Drummond, John J. (1990), Husserlian Intentionality and Non-Foundational Realism:
Noema and Object. Dordrecht: Kluwer.
Edie, James M. (1976), Speaking and Meaning. Bloomington: Indiana University
Elden, Stuart (2001), Mapping the Present. Heidegger, Foucault and the Project of a
Spatial History. London: Continuum.
Eribon, Didier (1989/1991), Michel Foucault. Cambridge, Mass.: Harvard
University Press.
Fink, Eugen (1932/1988), VI Cartesianische Meditation. Die Idee Einer Transzendentalen Methodenlehre, in Husserliana documente, Dordrecht: Kluwer.
(1932/1995), Sixth Cartesian Meditation. The Idea of a Transcendental Theory of
Method. Bloomington: Indiana University Press.
(1933), ‘Die phänomenologische Philsosophie Edmund Husserls in der
gegenwärtigen Kritik’, Kant-Studien 1.1, 229–383.
Flynn, Thomas (1989), ‘Foucault and the Politics of Postmodernity’, Noûs 23,
referen c es
(1991), ‘Foucault a Parrhesiast: His Last Course at the Collège de France’, in
James Bernauer and David Rasmussen (ed.), The Final Foucault. Cambridge,
Mass.: MIT Press, 102–18.
(1994), ‘Foucault’s Mapping of History’, in Gary Gutting (ed.), The Cambridge
Companion to Foucault. Cambridge: Cambridge University Press, 28–46.
(1997), Sartre, Foucault, and Historical Reason, vol. i, Toward an Existentialist Theory
of History. Chicago: University of Chicago Press.
Frank, Manfred (1989/1992), ‘On Foucault’s Concept of Discourse’, in Michel
Foucault, Philosopher: Essays Translated from the French and German by Timothy J.
Armstrong . Hemel Hempstead: Harvester Wheatsheaf, 99–115.
Fraser, Nancy (1989), Unruly Practices: Power, Discourse and Gender in Contemporary
Social Theory. Cambridge: Polity Press.
Gatens, Moira (1983/1991), ‘A Critique of the Sex/Gender Distinction’, in
Senja Gunew (ed.), A Reader in Feminist Knowledge. London and New York:
Routledge, 139–57.
(1996), Imaginary Bodies. London and New York: Routledge.
Goldstein, Jan (ed.) (1994), Foucault and the Writing of History. Oxford: Blackwell.
Grimshaw, Jean (1993), ‘Practices of Freedom’, in Caroline Ramazanoglu (ed.),
Up Against Foucault. London and New York: Routledge, 51–72.
(1999), ‘Working out with Merleau-Ponty’, in Jane Arthurs and Jean Grimshaw,
(eds.), Women’s Bodies. Discipline and Transgression. London: Cassell, 91–116.
Gros, Frédéric (2001), ‘Situation du cours’, in Michel Foucault, L’herméneutique
du sujet. Cours au Collège de France (1981–1982), ed. Frédéric Gros. Paris:
Seuil/Gallimard, 497–526.
Grosz, Elizabeth (1994), Volatile Bodies: Toward a Corporeal Feminism. Indianapolis:
Indiana University Press.
(1995), Space, Time, and Perversion: Essays on the Politics of Bodies. London and
New York: Routledge.
Gutting, Gary (1989), Michel Foucault’s Archaeology of Scientific Reason. Cambridge:
Cambridge University Press.
(1994), ‘Introduction. Michel Foucault: A User’s Manual’, in Gary Gutting
(ed.), The Cambridge Companion to Foucault. Cambridge: Cambridge University
Press, 1–27.
(2001), French Philosophy in the Twentieth Century. Cambridge: Cambridge University Press.
Haaparanta, Leila (1994), ‘Intentionality, Intuition and the Computational
Theory of Mind’, in Leila Haaparanta (ed.), Mind, Meaning and Mathematics.
Dordrecht: Kluwer, 211–33.
Habermas, Jürgen (1985/1987), The Philosophical Discourse of Modernity.
Cambridge: Polity Press.
Hacking, Ian (1984), ‘Five Parables’, in Richard Rorty, Julius Schneewind and
Quentin Skinner (eds.), Philosophy in History: Essays on the Historiography of
Philosophy. Cambridge: Cambridge University Press, 103–24.
(2002), Historical Ontology. Cambridge, Mass.: Harvard University Press.
Hadot, Pierre (1983/1997), ‘Forms of Life and Forms of Discourse in Ancient
Philosophy’, in Arnold I. Davidson (ed.), Foucault and his Interlocutors.
Chicago: University of Chicago Press, 203–24.
referen c es
Hall, Stuart (1996), ‘Introduction: Who Needs ‘Identity’?’, in Stuart Hall and
Paul du Gay (eds.), Questions of Cultural Identity. London: Sage, 1–17.
Halperin, David M. (1995), Saint Foucault: Towards a Gay Hagiography. Oxford:
Oxford University Press.
Han, Beatrice (1998/2002), Foucault’s Critical Project. Between the Transcendental
and the Historical. Stanford: Stanford University Press.
Haraway, Donna (1991), Simians, Cyborgs, and Women: The Reinvention of Nature.
London and New York: Routledge.
Heidegger, Martin (1952/1977), ‘The Age of the World Picture’, in The Question
Concerning Technology and Other Essays. New York: Harper & Row, 115–54.
Heinämaa, Sara (1996), ‘Woman – Nature, Product, Style? Rethinking the
Foundations of Feminist Philosophy of Science’, in Lynn Hankinson Nelson
and Jack Nelson (eds.), A Dialog Concerning Feminism, Science and Philosophy of
Science. Dordrecht: Kluwer, 289–308.
(1997), ‘What is a Woman? Butler and Beauvoir on the Foundations of Sexual
Difference’, Hypatia 12.1, 20–39.
(2002), ‘From Decisions to Passions: Merleau-Ponty’s Interpretation of
Husserl’s Reduction’, in Ted Toadvine and Lester Embree (eds.), MerleauPonty’s Reading of Husserl. Dordrecht: Kluwer.
(2003), Toward a Phenomenology of Sexual Difference. Husserl, Merleau-Ponty, Beauvoir. Lanham, Md.: Rowman & Littlefield.
Heinämaa, Sara and Oksala, Johanna (2000), ‘The Philosophical Practice of
Thinking: Between Husserl and Foucault’. Unpublished manuscript.
Hekman, Susan (1990), Gender and Knowledge: Elements of Postmodern Feminism.
Cambridge: Polity Press.
Himanka, Juha (1999), ‘Reduction in concreto. Two Readings of the Idea of
Phenomenology’, Recherches Husserliennes 11, 51–78.
(2001), ‘Before and After the Reduction’, Journal of the British Society for Phenomenology 32.2, 188–204.
Hindes, Barry (1998), ‘Politics and Liberation’, in Jeremy Moss (ed.), The Later
Foucault. Politics and Philosophy. London: Sage, 50–63.
Hoeller, Keith (1993), ‘Editor’s Forword’, in Michel Foucault and Ludwig
Binswanger, Dream and Existence. New Jersey: Humanities Press, 7–17.
Holenstein, Elmar (1985/1999), ‘The Zero-Point of Orientation: The Placement
of the I in Perceived Space’, in Don Welton (ed.), The Body. Classic and
Contemporary Readings. Oxford: Blackwell, 57–94.
Hoy, David Couzens (1986), ‘Introduction’, in David Couzens Hoy (ed.), Foucault:
A Critical Reader . Oxford: Basil Blackwell, 1–25.
(1998), ‘Foucault and Critical Theory’, in Jeremy Moss (ed.), The Later Foucault.
Politics and Philosophy. London: Sage, 18–32.
(1999), ‘Critical Resistance: Foucault and Bourdieu’, in Gail Weiss and Honi
Fern Haber (eds.), Perspectives on Embodiment: The Intersections of Nature and
Culture. London and New York: Routledge, 3–22.
Husserl, Edmund (1911/1965), ‘Philosophy as Rigorous Science’, in Phenomenology and the Crisis of Philosophy. New York: Harper, 79–122.
(1913/2001), Logical Investigations, vols. i and ii. London and New York:
referen c es
(1939/1989), The Origin of Geometry, Lincoln: University of Nebraska Press.
(1954/1962), Die Krisis der europäischen Wissenshaften und die transzendentale
Phänomenologie, Husserliana, vol. vi, The Hague: Martinus Nijhoff.
(1954/1970a), The Crisis of European Sciences and Transcendental Phenomenology.
Evanston, Il.: Northwestern University Press.
(1954/1970b), ‘The Vienna Lecture’, in The Crisis of European Sciences and
Transcendental Phenomenology. Evanston, Il.: Northwestern University Press,
Hutcheson, Peter (1981), ‘Husserl and Private Language’, Philosophy and
Phenomenological Research 42, 118–28.
Irigaray, Luce (1977/1985), This Sex Which Is Not One. Ithaca: Cornell University
(1984/1993), An Ethics of Sexual Difference. London: Athlone Press.
Judovitz, Dalia (2001), The Culture of the Body. Genealogies of Modernity. Michigan:
University of Michigan Press.
Kant, Immanuel (1784/1997), ‘Was ist Aufklärung?’, in Sylvere Lotringer and
Lysa Hochroth (eds.), The Politics of Truth: Michel Foucault. New York:
Semiotext(e), 7–20.
Kaufman, Walter (1950/1974), Nietzsche. Philosopher, Psychologist, Antichrist.
Princeton: Princeton University Press.
Kruks, Sonia (1990), Situation and Human Existence: Freedom, Subjectivity and Society.
London and New York: Routledge.
Kusch, Martin (1991), Foucault’s Strata and Fields: An Investigation into Archaelogical
and Genealogical Science Studies. Dordrecht: Kluwer.
Lebrun, Gérard (1989/1992), ‘Notes on Phenomenology in Les Mots et les Choses’,
in Michel Foucault, Philosopher: Essays Translated from the French and German by
Timothy J. Armstrong . Hemel Hempstead: Harvester Wheatsheaf, 20–37.
Levin, David Michel (1991), ‘Visions of Narcissism: Intersubjectivity and the
Reversals of Reflection’, in M. C. Dillon (ed.), Merleau-Ponty Vivant. Albany:
State University of New York Press, 47–90.
Levinas, Emmanuel (1947/1987), Time and the Other. Pittsburgh: Duquesne
University Press.
(1961/1969), Totality and Infinity. An Essay on Exteriority. Pittsburgh: Duquesne
University Press.
(1963), Difficile liberté. Essais sur le judaisme. Paris: Albin Michel.
(1968), Quatre lectures talmudique. Paris: Éditions de Minuit.
(1974), Autrement qu’ être ou au-delà de l’essence. The Hague: Martinus Nijhoff.
(1974/1981), Otherwise than Being, or Beyond Essence. The Hague: Martinus
(1982/1985), Ethics and Infinity. Pittsburgh: Duquesne University Press.
(1986/1998), Of God Who Comes To Mind. Stanford: Stanford University Press.
(1987/1994), Outside the Subject. Stanford: Stanford University Press.
(1989), The Levinas Reader. Ed. Sean Hand. Oxford: Basil Blackwell.
Lingis, Alphonso (1981), ‘Translator’s Introduction’, in Emmanuel Levinas,
Otherwise than Being or Beyond Essence. The Hague: Martinus Nijhoff,
referen c es
Machado, Roberto (1989/1992), ‘Archaeology and Epistemology’, in Michel
Foucault, Philosopher: Essays Translated from the German and the French by Timothy
J. Armstrong . Hemel Hempstead: Harvester Wheatsheaf, 3–19.
McNay, Lois (1991), ‘The Foucauldian Body and the Exclusion of Experience’,
Hypatia 6.3, 125–40.
(1992), Foucault and Feminism. Cambridge: Polity Press.
McWhorter, Ladelle (1999), Bodies and Pleasures: Foucault and the Politics of Sexual
Normalization. Bloomington: Indiana University Press.
Merleau-Ponty, Maurice (1945/1972), Phénoménologie de la perception. Paris:
(1945/1994), Phenomenology of Perception. London and New York: Routledge.
(1960/1964), Signs. Evanston, Il.: Northwestern University Press.
(1964), The Primacy of Perception and Other Essays on Phenomenological Psychology,
the Philosophy of Art, History and Politics. Ed. James M. Edie. Evanston, Il.:
Northwestern University Press.
Miller, Jacques-Alain (1989/1992), ‘Michel Foucault and Psychoanalysis’, in
Michel Foucault, Philosopher: Essays Translated from the French and German by
Timothy J. Armstrong . Hemel Hempstead: Harvester Wheatsheaf, 58–64.
Mohanty, J. N. (1976), Edmund Husserl’s Theory of Meaning. The Hague: Martinus
(1989), Transcendental Phenomenology, An Analytic Account. Oxford: Basil
(1997), Phenomenology: Between Essentialism and Transcendental Philosophy.
Evanston, Il.: Northwestern University Press.
Moss, Jeremy (1998), ‘Introduction: The Later Foucault’, in Jeremy Moss (ed.),
The Later Foucault, Politics and Philosophy. London: Sage, 1–17.
Nicholson, Linda (ed.) (1990), Feminism/Postmodernism. London and New York:
Nietzsche, Friedrich (1887/1974), The Gay Science. New York: Vintage.
(1889/1990), Twilight of the Idols/The Anti-Christ. Harmondsworth: Penguin.
O’Farrell, Clare (1989), Foucault, Historian or Philosopher? London: Macmillan.
O’Leary, Timothy (2002), Foucault and the Art of Ethics. London: Continuum.
Oksala, Johanna (2003), ‘The Birth of Man’, in Dan Zahavi, Sara Heinämaa
and Hans Ruin (eds.), Metaphysics, Facticity, Interpretation. Dordrecht: Kluwer,
(2004), ‘What is Feminist Phenomenology? Thinking Birth Philosophically’,
Radical Philosophy 126, 16–22.
Patton, Paul (1998), ‘Foucault’s Subject of Power’, in Jeremy Moss (ed.), The Later
Foucault, Politics and Philosophy. London: Sage, 64–77.
Peperzak, Adrian (1993), To the Other. An Introduction to the Philosophy of Emmanuel
Levinas. Indiana: Purdue University Press.
Pheng, Cheah (1996), ‘Mattering’, Diacritics 26.1, 108–39.
Pulkkinen, Tuija (1996/2000), The Postmodern and Political Agency. Jyväskylä:
SoPhi, University of Jyväskylä.
Rabinow, Paul (1984), ‘Introduction’, in Paul Rabinow (ed.), The Foucault Reader .
Harmondsworth: Penguin, 3–29.
referen c es
(1994), ‘Introduction: A Vital Rationalist’, in Georges Canguilhem, A Vital
Rationalist, Selected Writings from Georges Canguilhem, ed. François Delaporte.
New York: Zone Books, 11–22.
Rajchman, John (1985), Michel Foucault: The Freedom of Philosophy. New York:
Columbia University Press.
Reuter, Martina (2000), Questions of the Body, Sexual Difference and Equality in Cartesian Philosophy. Helsinki: Yliopistopaino.
Rouse, Joseph (1994), ‘Power/knowledge’, in Gary Gutting (ed.), The Cambridge
Companion to Foucault. Cambridge: Cambridge University Press, 92–114.
Sawicki, Jana (1991), Disciplining Foucault: Feminism, Power and the Body. London
and New York: Routledge.
(1994), ‘Foucault, Feminism, and Questions of Identity’, in Gary Gutting (ed.),
The Cambridge Companion to Foucault. Cambridge: Cambridge University Press,
(1996), ‘Feminism, Foucault and “Subjects” of Power and Freedom’, in Susan
J. Hekman (ed.), Feminist Interpretations of Michel Foucault. University Park:
Pennsylvania State University Press, 159–78.
Shepherdson, Charles (2000), Vital Signs, Nature, Culture, Psychoanalysis. London
and New York: Routledge.
Smart, Barry (1997), ‘Foucault, Levinas and the Subject of Responsibility’, in
Jeremy Moss (ed.), The Later Foucault: Politics and Philosophy. London: Sage,
Soper, Kate (1993), ‘Productive Contradictions’, in Caroline Ramazanoglu (ed.),
Up Against Foucault. London and New York: Routledge, 29–51.
Steinbock, Anthony J. (1995), Home and Beyond: Generative Phenomenology after
Husserl. Evanston, Il.: Northwestern University Press.
Stoller, Silvia (2000), ‘Reflections on Feminist Merleau-Ponty Skepticism’,
Hypatia 15.1, 173–82.
Sullivan, Shannon (1997), ‘Domination and Dialogue in Merleau-Ponty’s
Phenomenology of Perception’, Hypatia 12.1, 1–19.
Taylor, Charles (1986), ‘Foucault on Freedom and Truth’, in David Couzens Hoy
(ed.), Foucault: A Critical Reader . Oxford: Basil Blackwell, 69–102.
Turner, William (2000), A Genealogy of Queer Theory. Philadelphia: Temple
University Press.
Veyne, Paul (1971/1997), ‘Foucault Revolutionizes History’, in Arnold I.
Davidson (ed.), Foucault and his Interlocutors. Chicago: University of Chicago
Press, 146–82.
(1986/1997), ‘The Final Foucault and his Ethics’, in Arnold I.
Davidson (ed.), Foucault and his Interlocutors. Chicago: University of Chicago
Press, 225–33.
Vintges, Karen (1992), Philosophy as Passion: The Thinking of Simone de Beauvoir.
Indianapolis: Indiana University Press.
Visker, Rudi (1995), ‘Raw Being and Violent Discourse: Foucault, MerleauPonty and the (Dis-)Order of Things’, in Patrick Burke and Jan Van Der
Veken (eds.), Merleau-Ponty in Contemporary Perspectives. Dordrecht: Kluwer,
referen c es
(1999), Truth and Singularity: Taking Foucault into Phenomenology. Dordrecht:
Waldenfels, Bernhard (1993), ‘Interrogative Thinking: Reflections on MerleauPonty’s later Philosophy’, in Patrick Burke and Jan van der Veken (eds.),
Merleau-Ponty in Contemporary Perpectives. Dordrecht: Kluwer, 3–12.
(1998), Grenzen der Normalisierung. Frankfurt-on-Main: Suhrkamp.
Waltzer, Michel (1986), ‘The Politics of Michel Foucault’, in David Couzens Hoy
(ed.), Foucault: A Critical Reader . Oxford: Basil Blackwell, 51–68.
Webb, David (2003), ‘Cavaillès, Husserl and the Historicity of Science’, Angelaki
8.3, 59–71.
Weil, Simon (1952/1997), Gravity and Grace. Lincoln: University of Nebraska
Wittgenstein, Ludwig (1922/1961), Tractatus Logico-Philosophicus. London and
New York: Routledge.
(1965), ‘A Lecture on Ethics’, Philosophical Review 74, 3–12.
Wolin, Richard (1986), ‘Foucault’s Aesthetic Decisionism’, Telos 67, 71–86.
Young, Iris Marion (1990), Throwing Like a Girl and Other Essays in Feminist Philosophy and Social Theory. Indianapolis: Indiana University Press.
Zahavi, Dan (1994), ‘The Self-Pluralisation of the Primal Life: A Problem in Fink’s
Husserl-Interpretation’, Recherches Husserliennes 2, 3–18.
(1996), ‘Husserl’s Intersubjective Transformation of Transcendental Philosophy’, Journal of the British Society for Phenomenology 27.3, 228–45.
(1996/2001), Husserl and Transcendental Intersubjectivity, A Response to the
Linguistic-Pragmatic Critique. Athens, Oh.: Ohio University Press.
(2001), ‘Anonymity and First-Personal Giveness. An Attempt at Reconciliation’, in David Carr and Christian Lotz (eds.), Subjektivität – Verantwortung –
Wahrheit. Neue Aspekte der Husserlschen Phänomenologie. Frankfurt-on-Main:
Peter Lang, 61–72.
aesthetics of existence, 159, 160, 165,
166–7, 168–9, 171, 174, 207
analytic of finitude, 54
apparatus (dispositif), 96–7, 102, 106,
130, 132
a priori, historical, 20–3, 35, 36n19, 37,
40, 47, 48, 67–9, 72, 87
archaeology, 3, 10–11, 18, 19, 20–2, 35,
36, 37–8, 68, 69, 70–1, 74, 76–8,
80–2, 88, 94, 95, 96, 97, 132, 163,
170, 171, 184, 185, 186, 189, 190
Archaeology of Knowledge, The, 21n6, 36–8,
76, 80, 81, 89, 95
archive, 76
Ashbery, John, 82
lived, 12, 119, 132–3, 134, 135,
138–40, 144, 148, 151, 152
phenomenology of the body,
see phenomenology
Boothroyd, David, 167
Bordo, Susan, 110n1
Braidotti, Rosi, 5
Brown, Wendy, 176, 180
Bruzina, Ronald, 60n20
Butler, Judith, 108–9, 115, 120–2, 123,
125–6, 131, 134, 136–7
Canguilhem, Georges, 8, 45, 45n6,
49–51, 77
Carr, David, 61n23
Cavaillés, Jean, 8n9, 45, 79
contingency, historical, 13, 107, 170,
172, 187, 188
Critchley, Simon, 203
critique, 8–9, 12–13, 173, 175, 178,
179–80, 182, 186–7, 188, 190
Crossley, Nick, 107n17
Bachelard, Gaston, 45, 45n6, 46n9, 47,
Barthes, Roland, 82n14
Bataille, Georges, 128, 129
Beauvoir, Simone de, 135n1
Benhabib, Seyla, 179, 180
Bernauer, James, 167
Binswanger, Ludwig, 7
Birth of the Clinique, The, 89
and pleasures, 11, 124, 126, 127–8,
129, 134
and power, 11–12, 97–102, 113, 120,
and resistance, 11, 89, 109, 110, 111,
122, 123–6, 127, 131, 134, 153,
experiential, 11–12, 127, 128–9,
130–1, 132, 134
female, 110, 110n1, 120, 122–3, 124,
135–8, 150, 152, 153
Foucault’s understanding of, 11–12,
110–14, 117–1, 124, 128, 131,
132–4, 145, 150
Davidson, Arnold, 78n12, 158n1,
Death and the Labyrinth: The World of
Raymond Roussel, 82
Deleuze, Gilles, 126–7, 165
Derrida, Jacques, 64n25, 82n14,
169–72, 174, 200n8
Descartes, René, 13, 28, 49, 51–2,
Dews, Peter, 164
Dillon, Martin, 142
Discipline and Punish, 3, 97, 98, 112,
discourse, 11, 32, 35, 36, 36n20, 38, 55,
67, 72, 74, 77, 78, 80–1, 84–5, 86,
95–6, 102–3, 113, 117, 130, 131,
in d ex
counter-, 84–5
scientific, 19–20, 35–7, 47–8, 55, 78,
79–80, 81–2, 87, 96, 98, 102, 117,
130, 150–1, 189
discursive formation, 37, 81
Dreyfus, Hubert, 71–3, 77, 119, 132–3,
Enlightenment, 5, 8, 12–13, 170, 172,
176, 182–4, 185–7, 190
episteme, 21–3, 27–30, 33–4, 47, 48–9,
75, 76, 77, 78, 79, 96–7, 130
classical, 21, 22, 24–7, 28, 29, 31, 52,
modern, 21, 22, 27–30, 31, 33, 34, 52,
Renaissance, 21, 23–4, 26, 85
epoche, 43–5, 47, 57, 59, 60, 64–5, 66
feminist, 166n8
Foucault on, 4, 12, 85n17, 157–61,
162, 163, 165–7, 169–71, 172–3,
175, 186, 191–2, 193–4, 195–6,
Levinas on, 196–205
experience, 127–31, 132, 133, 163, 174,
189–90, 191
criticism of Foucault, 101, 114
theory, 4–6, 11, 101, 110–11, 115–16,
120, 121, 122–3, 124, 127, 134,
135–8, 144, 150, 152, 153, 179
Fink, Eugene, 60n20, 61, 65–6
Flynn, Thomas, 76n9, 105, 173n16,
176n1, 181n10, 187
Frankfurt School, 8, 186
Fraser, Nancy, 176
freedom, 12–13, 132, 152–3, 168, 170,
172, 175, 178, 180, 181, 182–3, 186,
187, 188–92, 193, 208–10
and language, see language
female, 12, 134, 150–3
practice of, 181–2, 186, 188, 189,
190, 206
gender, 115–17, 120, 121–2, 123,
genealogy, 3, 11, 73, 74, 81, 82, 94,
95n4, 96, 97, 101, 104–5, 106, 107,
111, 112, 112n3, 113–14, 119, 120,
130, 132, 163–4, 175, 177, 179–80,
185, 188, 189–90
of the subject, 3, 18, 164, 184
governmentality, 4, 164
Gregory of Nyssa, 162–3
Grimshaw, Jean, 135–7, 166n8
Grosz, Elisabeth, 11, 119–1, 125n19,
127n23, 152
Gutting, Gary, 2, 45n6, 46n8, 50, 53,
70n1, 73n4, 77n10, 96n5, 112
Habermas, Jürgen, 169n13
Hacking, Ian, 102
Hadot, Pierre, 173n17
Hall, Stuart
Halperin, David, 129n24, 129n25,
Han, Beatrice, 18–19, 21n5, 32–3n15,
36n19, 68, 73n4, 96, 102, 104–5
Haraway, Donna, 5
Heidegger, Martin, 32, 53, 74–5
Heinämaa, Sara, 72n3, 120–1n13
Hekman, Susan, 6
Herculine Barbin, 116
hermeneutics, 73, 74
of the self, 162
Himanka, Juha, 44n5
historicism, 55, 56–7, 67–8
historicity, 9, 55, 67–9, 133, 137, 148
history, 21, 49–50, 67, 68–9, 77–9, 112,
114, 118, 119, 132
History of Sexuality, vol. i, 3–4, 11, 100,
105n16, 111, 113, 124, 125, 127–8,
History of Sexuality, vol. ii, 4, 157,
159–60, 162, 163
History of Sexuality, vol. iii, 4, 157,
159–60, 162, 163
Hoeller, Keith, 7
Holenstein, Elmar, 145
Hoy, David Couzens, 119, 133n30,
175–6, 179
Husserl, Edmund, 6, 8–9, 10, 32, 38,
40–6, 47–8, 49, 51–2, 53–4, 55–60,
60n20, 62–3, 64n25, 64n26, 68,
69, 72, 79, 139, 144, 146–7, 148,
intentionality, 139–40, 203
Irigaray, Luce, 5
Kant, Immanuel, 28–30, 32–3n15, 49,
51–2, 54, 68, 159, 182, 183, 184–5,
186, 187
Kaufman, Walter, 183n15
and power, see power
conditions of possibility of, 18, 20–3,
29, 38, 55, 61, 67, 68–9, 73, 76, 78,
empirical, 28, 54–5, 56
scientific, 18, 22, 48
Kristeva, Julia, 82n14
in d ex
Lacan, Jacques, 82n14, 118n11,
language, 10–11, 25–6, 27, 34, 34n6, 36,
74, 76, 87–8, 89, 118, 129, 130, 132,
170, 171, 172–3, 191
and freedom, 10–11, 19, 81–2, 84–5,
88, 89, 189
and phenomenology, see phenomenology
Levinas on, 199–201
Lebrun, Gérard, 40, 52
Levin, David, 132n29
Levinas, Emmanuel, 9, 13, 192,
life-world, 41–5, 47–8, 49, 57, 58, 59, 62,
64, 67
Lingis, Alphonso, 203
literature, 10, 34n16, 81, 84–7
avant-garde, 10, 34n16, 82–5, 189
Machado, Robert, 71n2
Madness and Civilization, 7, 88, 169, 171
Mahon, Michel, 167
Maladie mental et personalité, 7
as empirico-transcendental doublet,
31–3, 57, 73
doubles of, 32–3, 53–9
paradox of, 53, 54–3, 67–9
McNay, Lois, 123–4, 166n8, 167
Merleau-Ponty, Maurice, 9, 12, 32, 53,
119, 132–3, 134, 135–47, 148–50,
Mohanty, J. N., 61n23, 103n14
naturalism, 55, 57
Nietzsche, Friedrich, 6, 9, 29, 33,
34, 94, 95n4, 99, 111, 119, 183,
nominalism, 35
O’Farrell, Claire, 21, 77n11, 88–9
O’Leary, Timothy, 168n11, 169, 195
order, 19–20, 81, 89
Order of Things, The, 3, 7, 10, 17–18,
19–23, 33, 35, 36, 37, 38, 40–1,
44–5, 47, 48, 50, 51, 53, 69, 70, 71,
74–5, 76, 78, 79–80, 84, 85, 89,
Patton, Paul, 188
Pheng Cheah, 94, 95
phenomenology, 4, 6–9, 29, 41–5,
and archaeology, 72–3, 77, 94
and language, 64–6, 142, 145
and science, 42–3, 45, 47–9, 52, 55
and the subject, 41, 44, 47, 58, 59–61,
93–4, 109, 137–8, 141–50, 151,
of the body, (see also bodies, lived),
135, 136, 137–41, 142–6, 147,
148–50, 151–3
Foucault and, 6–9, 30, 32–3, 35, 38,
40–1, 44–5, 47, 48–9, 52, 53–8,
67–9, 70, 76, 94–5, 103–4, 132, 134,
Plato, 162
pleasure, 126–7, 129, 132, 158, 159
and body, see bodies
and resistance, see bodies and resistance
positivism, 18, 29, 30, 32, 55, 57
post-structuralism, 1, 2, 5, 93
power, 12, 94–5, 97, 98–102, 106, 108–9,
113, 118, 123, 126, 129, 167, 176–9,
184, 186, 190, 191, 194
and body, see bodies
and games, 105, 108
and knowledge, 4, 96, 97, 102,
104–5, 107, 108, 110, 113, 122, 126,
130, 132, 150, 165, 196, 204,
and sexuality, 115, 116
bio-, 100, 103, 127, 132, 150
normalizing, 12, 111, 127, 131, 132,
167, 168, 177, 182
practices, 11, 12, 36, 77, 104, 105–8,
119, 129, 132, 170, 171, 185, 191,
discursive, 20, 72–3, 78, 89, 96–7
dividing, 3, 150, 151–3
of the self, 159, 161, 165, 166n8,
168–70, 172, 193, 196, 204
scientific, 35, 71, 72, 102
critique of the, 9, 12–13, 175, 184–5,
186, 187, 190
ontology of the, 13, 75–6, 184
psychoanalysis, 18, 93, 118n11, 139n4,
Pulkkinen, Tuija, 191
Rabinow, Paul, 46n9, 71–3, 77, 119,
132–3, 164, 191
Rajchman, John, 2, 82, 82n14, 85, 178,
relativism, 5n4, 67, 170, 172
resistance, 11, 108, 109, 123, 165,
167–8, 194, 209
and the body, see body
responsibility, 197–8, 201–2, 203,
Roussel, Raymond, 82–5
in d ex
Sartre, Jean-Paul, 8n9, 13, 32, 53
Schilder, Paul, 139n4
science, 3, 74–5, 129, 185
and discourse, see discourse
and knowledge, see knowledge
and phenomenology, see phenomenology
history of, 22, 28, 48, 49–51, 79, 80
human sciences, 31, 35, 98, 102, 103,
self, 4, 163–5, 171, 174, 175, 181, 206
care of the, 11, 12, 161, 162–3, 170,
171, 174, 188, 193–4
practices of the, see practices
relationship to the, 4, 158–9, 163,
technologies of the, 4, 89, 161–5, 196
sex (sexe), 114–17, 122, 124–6
sexuality, 4, 89, 104, 109, 113–15, 116,
120, 124–6, 127, 129, 133–4, 157–8,
162, 163
Merleau-Ponty on, 136–7, 138, 140–1,
Shepherdson, Charles, 118n11, 130n27
Smart, Barry, 197–8
Steinbock, Anthony, 144n11, 148n16,
structuralism, 9, 18, 73, 74, 93
subject, 3, 12, 70–1, 88, 108–9, 199,
201, 202–3
conditions of possibility for, 94–5, 97,
100, 106–8, 182
ethical, 13, 158–9, 160, 161, 163–4,
166, 167, 185, 192, 195–6, 197,
198–9, 200–1, 202, 203–5, 206,
Foucault’s understanding of, 6, 7,
79–82, 85–7, 89, 94–5, 98, 100–2,
103–4, 105, 106–8, 130, 150–1, 162,
163, 164, 165, 167–8, 179, 180, 185,
187, 188, 190, 191–2, 194–5, 196,
in phenomenology, see phenomenology
subjection, 97–8, 99, 100, 101–2, 103,
108–9, 123, 159, 160, 165, 167, 189
Surrealism, 82–4
Taylor, Charles, 104
disciplinary, 3, 98, 103, 110n1, 130,
governmental, 108, 164–5
of the self, see self
conditions, 21, 37, 69, 72, 107
intersubjectivity, 6, 43, 57, 60n20, 62,
137, 143–4, 145–50, 151–2, 153
subjectivity, 43, 44, 47, 48, 49, 59–61,
64–5, 69, 146
Veyne, Paul, 169n12, 173n16
Viskers, Rudi, 151
Webb, David, 79
Wittgenstein, Ludwig, 105, 170, 171,
171n14, 172
Young, Iris Marion, 135–7
Zahavi, Dan, 60n20, 63, 143n10, 146,
146–7n14, 147n15