848

Experience without Subject: Rule-Governed Practices and the Possibility of Critical
Historiography in Foucault

Abstract
Experience without Subject: Rule-Governed Practices and the Possibility of Critical
Historiography in Foucault

In this study I propose to articulate a concept of experience according to which it is
already limit-experience, where “limit” and “experience” are understood in terms of rulegoverned
and spatio-temporally indexed practices criterial for both cognitive and
practical interactions with ourselves and with the world. Moreover, I propose to locate
this concept in the historiography of Foucault and construct its dimensions using
Foucauldian conceptual tools. Finally, I discuss Foucault’s division of his
“methodologies” over the course of his trajectory, avoiding their strict separation—as
archaeology, genealogy, and problematization—in favor of their articulation in terms of
discursive and nondiscursive practices, where “articulation” stands for neither a purely
linguistic, nor logical, nor even causal relation, but traces the contours of an ensemble of
historically constitutive and therefore criterial practices. I want to call that “genealogy”. I
then situate this Foucauldian conception with respect to a number of powerful objections
I reconstruct on the basis of texts by Habermas, Derrida, Merleau-Ponty and Adorno. I
claim that the objections gain their motivation and strength from the constitutive divide
between the transcendental and the empirical. Against this position, and at times against
the grain of some of Foucault’s own formulations, a more faithful characterization of
Foucault’s trajectory is not so much the conversion into the domain of contingency and
particularity of what would otherwise be necessary and universal conditions—
historicizing the transcendental—as it is giving up the transcendental in the forms it has
taken since Kant, and pressing the consequences of this abandonment for a reflection on
history, and by extension subjectivity. The result is a problematic and problematizing
notion of critique: Foucault’s giving up of the transcendental standpoint is not a
repudiation of reflection; it is rather motivated by the conviction that the moment of selfrelation
entailed by reflection cannot be anchored in any unreflected given. It therefore
has more critical force than an argument which would transcendentally secure the
legitimacy of cognitive and normative exclusion by “demonstrating” the irrationality of
those who are so excluded, without accounting for the constitutive role our social
sanctions play in their very creation.
Experience without Subject: Rule-Governed Practices and the Possibility of Critical
Historiography in Foucault
Acknowledgement
A dissertation is probably too small an achievement to merit acknowledgement. But the effort
sustaining the present study has been long and solitary; and so I would like to give thanks to
those who have accompanied me: Tom Flynn, for the unwavering support, infinite credit and
care he has given me; Geoffrey Bennington, Michael Sullivan, Lynne Huffer, and Mark Risjord,
for the interest they have taken in my project and for the generosity with which they offered their
time, even on short notice. And even our disagreements have shaped my thinking by the
exemplary models they have set up for me, as I tried to find the proper voice for its expression.
Stephen Farrelly, Ericka Tucker and Peter Milne, for their friendship on days when I needed it
most; Camila Aschner, for never failing to surprise me and for the “beautiful monsters” to which
she has introduced me; Mehmet, Aysel and Övünç, for being something like my constellation
through rain or shine.
Table of Contents
Introduction .................................................................................................................................. 1
1. From the Theater of Ideas to the Black Box Emitting Sentences ....................................... 20
1.1 Justification: Transcendental or Historical? ...................................................................... 24
1.11 Dependence of the Subject.................................................................................................... 26
1.12 Independence of the Subject ................................................................................................. 30
1.2 Phenomenology, or How Do I Simply Look On? ................................................................ 49
1.21 We Think, Therefore I Am Not ............................................................................................ 53
1.22 Seeing Things (Disappear) .................................................................................................... 63
2. Giving Up the Transcendental: Critical and Empirical Discourse? .................................. 76
2.1 Analytic of Finitude: Kant and the Order of Things ......................................................... 80
2.11 What is an empirico-transcendental double? ........................................................................ 82
2.12 Rights vs. Facts of Knowledge ............................................................................................. 96
2.2 Conditions of Possibility vs. Conditions of Existence ...................................................... 109
2.21 Logos in Archaeology. ........................................................................................................ 111
2.22 Criteria, Rules, and Tribes .................................................................................................. 117
3. Reasons, Causes, Madness: The Articulation of Discursive and Nondiscursive
Practices. .................................................................................................................................... 133
3.1 Madness in the First Person ............................................................................................... 137
3.11 Mental Illness: Explanation or Description?....................................................................... 138
3.12 How Fundamental Is Experience? ...................................................................................... 147
3.2 Madness in the Third Person ............................................................................................. 156
3.21 Habermas and the Transcendental Site Where We Meet .................................................... 157
3.22 Derrida and What We Mean When We Say ―Lock Them Up!‖ ......................................... 172
4. Experience of Freedom: Speaking for Oneself. .................................................................. 185
4.1 Talking Sex .......................................................................................................................... 187
4.11 The Body and its Sexuality ................................................................................................. 188
4.12 A Critical Ontology? .......................................................................................................... 193
4.2 Am I Free Not to Take Up a Standpoint? ......................................................................... 214
4.21 Subjects without Experience ............................................................................................... 216
4.22 Subjects, Objects, and Mere Plants ..................................................................................... 234
Limit-Experience: Hegel (without the Absolute) Again? ...................................................... 254
Bibliography .............................................................................................................................. 285
List of Abbreviations:
AK Archaeology of Knowledge
BC The Birth of the Clinic
DE Dits et écrits: 1954-1988, vols. 1-4
DP Discipline and Punish
EW Essential Works of Foucault, 1954-
1984, vols. 1-3
FF The Final Foucault: Interviews,
1966-1984
FL Foucault Live
FR Foucault Reader
HM History of Madness
HS History of Sexuality, vol. 1
HDS Herméneutique du sujet: Cours au
Collège de France (1981-1982)
IKA Introduction to Kant‟s Anthropology
MF Michel Foucault: Beyond
Structuralism and Hermeneutics,
Hubert L. Dreyfus and Paul
Rabinow.
OD The Order of Discourse
OT The Order of Things
PK Power/Knowledge: Selected
Interviews and Other Writings
PPC Politics Philosophy Culture:
Interviews and Other Writings,
1977-84
UP The Use of Pleasure, vol. 2 of The
History of Sexuality
Note on translations:
I use the English translations of Foucault‘s works when available and indicate the cases when I
modify the translation. I have retained some references to the French originals of History of
Madness, Discipline and Punish, and The Use of Pleasure, which are indicated as HMfr, DPfr,
and UPfr, respectively. In the case of L‟ordre du discours (Gallimard, 1971), I use The Order of
Discourse as its title, but I have retained the page numbers of the English translation by A. M.
Sheridan Smith, appended to the Archaeology of Knowledge. I use his translation, except when I
indicate otherwise.
1
Introduction
I cannot resist mentioning a scene first narrated to me some years ago. A program for the BBC
brings together Searl and Vattimo and, after some discussion, it is clear that they are not seeing
the same things, hearing the same problems, or even speaking the same language (in more senses
than one). Searle, in order to establish some common starting point, proposes that they talk about
inner experience and says: "I have an experience of pain on my back." To which Vattimo, in his
Italian accent, responds: "That's your inner experience? I pity you!" which, as one might expect,
leaves Searle even more mystified.
This scene always makes me laugh, and the following pages contain in part the effort to
understand that laughter, that pity, and that mystery (or the absence of mystery). The experience
of pain on one's back, as Wittgenstein certainly felt, may be as strange, and as ordinary and
mundane, as the experience of madness. These are the starting and ending points of the following
study. What then lies in between?
Experience is the word I use when my justifications come to an end and my reasons give
out, as Wittgenstein might say. But it is also the word on which I fall when every other word
fails me, as if it possesses something indubitable, which moreover would be immediately
accessible to me—the ultimate ground. It is no wonder, then, that it has been taken up in a
number of not entirely consistent ways, as when one refers to: this patch of white, this paper, this
fire, my body; or when Nancy beautifully writes:
[A]ccording to the origin of the word ―experience‖ in peirā and in ex-periri, experience is an
attempt executed without reserve, given over to the peril of its own lack of foundation and
security in this ―object‖ of which it is not the subject but instead the passion, exposed like the
2
pirate (peirātēs) who freely tries his luck on the high seas. In a sense…freedom, to the extent that
it is the thing itself of thinking, cannot be appropriated, but only ―pirated‖: its ―seizure‖ will
always be illegitimate.1
I propose to articulate a concept of experience according to which it is already limitexperience,
where ―limit‖ and ―experience‖ are understood in terms of rule-governed and spatiotemporally
indexed practices criterial for both cognitive and practical interactions with ourselves
and with the world. Moreover, I propose to locate this concept in the historiography of Foucault
and construct its dimensions using the conceptual tools which can be found therein. Finally, I
discuss Foucault‘s division of his ―methodologies‖ over the course of his trajectory, avoiding
their strict separation—as archaeology, genealogy, and problematization—in favor of their
articulation in terms of discursive and nondiscursive practices, where ―articulation‖ stands for
neither a purely linguistic, nor logical, nor even causal relation, but traces the contours of an
ensemble of historically constitutive and therefore criterial practices. I want to call that
―genealogy‖.
It may appear, not without reason, that this is an unlikely site to excavate such a concept
of experience. But in the introduction to one of the last texts that Foucault authorized for
publication, he proposed the following explication of his project: ―What I planned, therefore, was
a history of the experience of sexuality, where experience is understood as the correlation
between fields of knowledge, types of normativity, and forms of subjectivity in a particular
culture‖ (UP, 4). According to this definition, analyzing sexuality as a historically singular
experience entails an analysis of the three axes that constitute it. First, there is the formation of
the sciences (savoirs) that refer to it as object; second, there are systems of power that regulate
1 Jean-Luc Nancy, The Experience of Freedom (Stanford: Stanford University Press, 1993) p. 20
3
its practice; and finally, there are the forms within which individuals recognize themselves as
subjects of this sexuality.
A little further in the text, Foucault specifies how such an analysis involves ―the games of
truth (jeux de verité) in their interplay with one another…their interaction with power
relations…[and] in the relationship of self with self and the forming of oneself as a subject…‖
(UP, 6). If games of truth are the ―rules according to which what a subject can say about certain
things involves the question of truth or falsehood,‖ (DE, 4:632) then it is possible to infer that
sexuality as a singular experience is constituted by historically analyzable practices that involve
three sets of rules: those which operate at the level of formation of discursive knowledge, those
which operate at the level of the exercise of power, and those which determine the forms in and
through which the subject constitutes itself.
Experience, then, is the correlation between knowledge, power, and what Foucault will
call subjectivation. Even if we leave aside the nature of their interrelations, a problem emerges:
is experience constituted or constitutive? The text warrants both readings to the extent to which
Foucault says, in this particular context where the object of investigation is sexuality, ―an
‗experience‘ came to be constituted in modern Western societies, an experience that caused
individuals to recognize themselves as subjects of a‗sexuality‘‖ (UP, 4; my italics). The inclusion
of a reflexive moment (―recognize themselves as‖) amplifies this problematic status of
experience, which Foucault is explicitly introducing as a concept that retrospectively
characterizes the trajectory he has followed from the History of Madness on.2
2 This intention is clearly expressed by the set of questions Foucault provides on p. 7: ―…when he perceives himself
to be mad; when he considers himself to be ill; when he conceives of himself as living, speaking, laboring being;
when he judges and punishes himself as a criminal?‖
4
The argumentative strategy I develop is motivated by a ―counter-problem‖: Why do we
take that to be a problem? In other words, what is the source of the necessity which compels us
to seek a theoretical foundation, and distribute ―experience‖ between what would be the
transcendentally constitutive conditions and the empirical ―conditioneds‖? Therefore, I resist the
temptation to regard Foucault‘s work as a cluster of empirical insights in search of their
philosopher to organize and justify their historical content.
However, the form that resistance takes and the intelligibility it assumes are far from
obvious and entail a number of risks. Specifically, it is not possible to articulate with full
conceptual determination the standpoint which sustains that resistance, because it is, in some
sense that will become clear only gradually throughout the present study, a disavowal of the very
points which have a categorial standing in modernity. More specifically still, because it entails a
complication of the separation into spheres of validity—cognitive truth, normative right, and
aesthetic beauty—which are then distributed between the transcendental and the empirical.
If one yields to the temptation, the multiplicity of concepts Foucault marshals across his
trajectory—episteme, historical a priori, discursive formation, archive, archaeology, games of
truth, genealogy, regimes of truth, techniques of the self, problematization—are so many
attempts to provide coherence to what remains one and the same passion: historicizing the
transcendental.
Against this position, and at times against the grain of some of Foucault‘s own
formulations, a more faithful characterization of Foucault‘s trajectory is not so much the
conversion into the domain of contingency and particularity what would otherwise be necessary
5
and universal conditions—historicizing the transcendental—as if all one had to do were the
addition of historical variability to transcendental frameworks or schemes; as it is giving up the
transcendental in the forms it has taken since Kant, and pressing the consequences of this
abandonment for a reflection on history, and by extension subjectivity. The following claim
Foucault makes crystallizes both the necessity and the difficulty staked by this thesis: ―The
research that I am undertaking here therefore involves a project that is deliberately both
historical and critical, in that it is concerned … with determining the conditions of possibility of
medical experience in modern times…. Here as elsewhere, it is a study that sets out to uncover,
from within the density of discourse, the conditions of its history‖ (BC, xix; modified, my italics).
Two tentative conclusions follow: first, the demonstration of the position I am
formulating, namely, that Foucault‘s philosophical/historical studies are motivated by and carry
out the renunciation of the transcendental, demands that a sense be given to the ‗conditions of
possibility‘ talk that permeates his works and structures his thought; and second, the
understanding so provided should account for both the historical and the critical dimensions
without presupposing the confusion of the transcendental and the empirical Foucault excoriates
in The Order of Things. The question is: Does Foucault need a version of the transcendental to
ground what would then become his empirical/historical studies, or does he successfully manage
to recast the transcendental theme through his non-subjective account of experience?
My argument proceeds in four stages.
First, because I take the apparent necessity of this question seriously—it is not simply a
philosophical mistake we can shrug off at will—I trace two lines, running from Kant to Hegel
and to Husserl, in order to understand the formation, the stakes and the different embodiments of
6
the transcendental theme. My interest is restricted to the juridical conception of ―deduction‖
which Kant proposes in order to displace philosophical enquiry from knowledge of substance—
the question of fact—to the self-knowledge of the rights by which a subject validates its claims
to knowledge—question of right. I propose an interpretation of the transposition of this theme
into the domain of history in Hegel‘s phenomenology; and I contrast that with Husserl‘s
conception of evidence and transcendental subjectivity.
Second, I offer a reading of the ―analytic of finitude,‖ as a critical argument in The
Order of Things, as offering not yet another transcendental or quasi-transcendental argument to
the effect that the target epistemic moves are impossible in principle, but that we have good
reasons to give up trying because they are no longer attractive; and that‘s because the endless
oscillations they generate at best show us that they are not fruitful research programs, and at
worst perpetuate a misrecognition of the practices they claim to ground. Therefore, I interpret
what the extension of the ―analytic‖ so described might be as not so much specific doctrines as
determinate epistemic strategies. So, for instance, the reason why there are few if any logical
positivists today is due not so much to its definitive refutation in its global goal and particular
details, since there are programs that are sympathetic to and continue aspects of positivism; as it
is to its repeated failure to provide the principle (of verification) which its method nevertheless
requires.
Foucault‘s description, however, at times operates at too high an altitude for some of his
criticisms to have bite—notwithstanding the brilliance of his ―synthesis‖. Drawing on Paul
Franks‘s discussion of German Idealism, the stakes of which he formulates in terms of the
7
Agrippan trilemma3, I ―schematize‖ the analytic of finitude. This enables me to provide a clearer
account of why the figures Foucault describes must fail (and in what sense), and to further my
account of why the transcendental theme has such a grip on us. According to the Agrippan
skeptic, only one of three possible answers are available in response to a why-question: either it
will be a brute assertion without a justification, in which case it is itself ungrounded, or it will
raise another why-question, in which case it generates an infinite regress, or it will presuppose
just what is at issue, in which case it is circular. Hence, the justificatory process triggered by a
demand for reasons will either terminate arbitrarily or give rise to a viciously infinite regress or
be circular. Unless an answer is available which circumvents this trilemma, our cognitive
relation to the world will remain unsatisfied.
In terms of the constitutive divide between the transcendental and the empirical, it is
possible to see why the modern attempts to find a satisfactory resolution of this predicament end
up confusing the two or lapsing into either circularity or infinite regress. Within this field what
counts as an adequate response must be capable of absolute grounding, i.e. it must terminate in
an a priori principle from which the conditions for the possibility of all our claims will be
derived but which will itself be unconditioned4.
3 Paul W. Franks, All or Nothing: Systematicity, Transcendental Arguments and Skepticism in German Idealism,
(Cambridge, Mass.: Harvard University Press, 2005), passim. See in particular Chapter 1.
4 Although discussions of the Agrippan trilemma tend to focus on the epistemological nature of the claims at issue
(that is, reasons for belief), it extends to practical as well as ontic reasons (reasons for why I should do x and reasons
for why something is what it is). Franks, op. cit., p. 19 The use of the term ―ground‖ reflects that polyvalence.
Moreover, Franks traces the demand that physical explanation be kept separate from metaphysical explanation to
Leibniz and the success of modern scientific explanations in terms of physical causes alone. This leads to what he
calls the ―explanatory closure of the physical,‖ the thesis that no nonphysical factors be allowed into explanation of
natural phenomena. One consequence of this is a tension between the demands for absolute grounding and for the
explanatory self-sufficiency of physical laws, since the latter cannot provide anything absolute. What is significant
in this account for my purposes is the emergence of the constraints that a) the transcendental and the empirical must
be rigorously kept apart, b) there must be absolute grounding escaping the trilemma, c) nothing empirical can count
8
The analytic of finitude may then be reformulated as tracing out the epistemic field
governed by the structure of this demand. There must be a single absolute first principle to stop
the regress of conditions generated by reflection on what makes experience in general possible.
However, this principle cannot be homogenous with what it conditions, since it would then be
subject to the same conditions as the series it conditions, which is to say that it would no longer
be unconditioned. Hence the empirical cannot be grounded on itself.
If, on the other hand, the absolute principle is heterogeneous with what it conditions, then
it either is immanent in the series (of conditions) or transcends it. If it is the former, then it is a
part of the whole, and hence subject to its law, while somehow standing outside it. Therefore, the
only principle that will satisfy the demand for absolute grounding, thereby skirting the Agrippan
skeptic, is one which is heterogeneous with and transcending the empirical. Since, however, the
principle thus invoked cannot be transcendent in the manner of pre-critical metaphysics, it can
only be transcendental.
What this ―schematic‖ rendering of the analytic of finitude entails is that the
contradictory demand to keep the transcendental and the empirical from reciprocal
contamination, while grounding the latter on the former, condemns the desire for epistemic—but
also practical—satisfaction in advance to disappointment. It is in the light of this that the
different reinscriptions of Kantian transcendental apperception ought to be perceived
I then draw on Wittgensteinian ―grammar‖ in order to interpret the historical a priori as
one possible response to this predicament. Wittgenstein‘s discussion of criteria, rules, and
standards enables me not so much to justify as clarify Foucault‘s claim that ―[historical a priori]
as absolute. The consequence is that the self-sufficiency of the empirical is both asserted and withdrawn, in which
may be found a prefiguration of the confusions Foucault describes in the analytic of finitude.
9
is defined as the group of rules that characterize a discursive practice: but these rules are not
imposed from the outside on the elements that they relate together; they are caught up in the very
things that they connect; and if they are not modified with the least of them, they modify them,
and are transformed with them into certain decisive thresholds‖ (AK, 127, my italics). Two
puzzles are sometimes mutually illuminating when juxtaposed. Foucauldian rules, unlike
transcendental conditions of possibility, are intimately bound with what they determine and they
are capable of historical transformation; and unlike what is merely empirically given, they have
the power to determine or modify what they condition.
The rules constituting the historical a priori are necessary, to the extent that they
determine what counts as a reason, or a true-or-false statement; in short, what count as the limits
of the world at a given time and for a particular group of people. But this a priori is at the same
time utterly contingent, since the rules ―are caught up in the very things that they connect,‖ and
are therefore spatially and temporally situated. I then formulate a response to a long standing
objection to Foucault‘s account of rules of formation of discursive practice in terms of criteria
and forms of life.
The classic development of this criticism is found in Dreyfus and Rabinow: ―If rules that
people sometimes follow account for what gets said, are these rules meant to be descriptive, so
that we should say merely that people act according to them, or are they meant to be efficacious,
so that we can say that speakers actually follow them;‖ but since Foucault‘s account claims to be
both a pure description of discursive events (AK, 27) and prescriptive determination of
10
discourse, he locates ―the productive power revealed by discursive practices in the regularity of
these same practices. The result is the strange notion of regularities which regulate themselves.‖5
It is indeed strange that regularities should regulate themselves, but why? The objection
relies on four interrelated claims: 1) One can provide either a description of regularities among
statements, in which case the rules are invoked only to systematize and give coherence to
phenomena; or one can give conditions of possibility governing (discursive) phenomena, in
which case one must have some account of this operation on phenomena in terms of either
objective causal laws or subjective norms. 2) One must be able to locate these rules either in the
consciousness of the speakers (which they reflectively obey and follow) or in the objective world
(which causes discursive behavior). 3) The only alternative to intentional and causal explanation
is the structuralist one, which provides the formal rules of combination of a system of differential
elements. 4) Foucault‘s account is committed to an essential separation of discursive from
nondiscursive practices.
Against (1), (2) and (3), it should be observed that the alternatives listed are not
exhaustive. Consider the case of the grammar of the language which a group of speakers uses.
Here we have a set of rules that are descriptive and prescriptive: I can understand what an
English speaker is saying only if I obey the rules of English. And that condition can be
satisfied—is in fact satisfied—without my reflective awareness of the rules which I follow. And
these rules do not causally determine what is actually said, at least not in a way captured by
mechanical causality. Moreover, the rules governing the formation of statements in a given
period, where what is at issue is not the virtual possibilities of combination of linguistic
5 Dreyfus and Rabinow, Michel Foucault, Beyond Structuralism and Hermeneutics, (University of Chicago Press,
1983) pp. 81 and 84.
11
elements, but the determination of what constitutes something‘s falling under a concept—of
what counts as something‘s being this or that kind of object—cannot be analyzed by means of
the application of a formal system of rules. Hence it is not a question of structural analysis.
If, however, the charge is that Foucault‘s conflation of causally or normatively productive
rules with the description of mere regularities hinges on his commitment to the explanatory
autonomy of the discursive, then two claims follow: Either Foucault‘s insistence on the selfsufficiency
and self-determination of the discursive in The Archaeology of Knowledge was an
overstatement for which he had no justification, or his commitment to such autonomy was
already qualified, which qualification he then developed explicitly in later works. I am
indifferent between these alternatives in this study, since, in either case, my claim is that the
account of rules operative in the definition of experience does not presuppose an irreducible
separation in kind between the discursive and the nondiscursive, but brings to light their
articulation on one another through the concept of practice.
Third, I offer a reading of the History of Madness in order to pursue the ―analysis‖ of
experience through rule-governed practices. This may seem like an odd choice: First, its
language, and even methodology, are riddled with those of phenomenology and existentialism
such as ―sensibility,‖ ―perception,‖ and ―consciousness‖. Second, even to those who are
sympathetic to a project inspired by Foucault, the work is reminiscent of romantic yearnings with
its evocation of the primitive purity of madness that would finally speak for itself.
This assessment seems universal among readers of the book as different as Habermas and
Derrida, even though it may be based on different reasons. It would then be a sign of progress
when Foucault appears to distance himself from the book, which ―accorded far too great a place,
12
and a very enigmatic one too, to what I called an ‗experience,‘ thus showing to what extent one
was still close to admitting an anonymous and general subject of history‖ (AK, 16, modified).6
But perhaps this exorcism of phenomenology and existentialism misses the mark, to the
extent to which the ―enigmatic‖ concept of experience structuring the History of Madness from
one end to the other is not that of phenomenology (and probably never was). For here too
Foucault uses the vocabulary of conditions of possibility: ―In the reconstitution of this
experience of madness, a history of the conditions of possibility of psychology wrote itself as
though of its own accord‖ (HM, xxxiv). Moreover, there are three appeals to a ―concrete a
priori‖ at key points in the text where Foucault is describing the object of his investigation, one
of which is:
[A] madman is not recognized as such because an illness has pushed him to the margins of
normality, but because our culture situates him at the meeting point between the social decree of
confinement and the juridical knowledge that evaluates the responsibility of the individuals before
the law. The ‗positive‘ science of mental illness and the humanitarian sentiments that brought the
mad back into the realm of the human were only possible once that synthesis had been solidly
established. They could be said to form the concrete a priori of any psychopathology with
scientific pretensions. (HM, 130, my italics)
What is at issue, then, is the articulation of a structure of experience (of madness) which
is the condition for the possibility of the myriad ways in which a group of people at a given time
and place, a ―culture,‖ objectifies madness. And this experience is understood in terms of
practices that are both discursive and nondiscursive, as the summary invocation of institutions,
6 See also his well-known identification (and repudiation) of ―an explicit theme of History of Madness‖ as ―what
madness itself might be, in the form in which it presented itself to some primitive, fundamental, deaf, scarcely
articulated experience‖ (AK, 47).
13
knowledge, and practice of confinement in this quotation attests. Furthermore, this structure of
experience ―is history through and through‖ (HM, xxxii).
The argumentative strategy I follow at this stage involves two steps, each corresponding
to one of these questions. First, I focus on two moments of the history of madness Foucault
writes, ―the ship of fools‖ and ―confinement,‖ in order to delineate how different types of
practices are ―articulated‖ and what such spatial metaphors imply in terms of explanatory power.
Drawing on texts that are commonly labeled genealogical in relation to the History of Madness
clarifies what ―power‖ is supposed to be or do in our understanding of phenomena. Second, I
situate this account of rules in relation to what has been called the ―space of reasons‖.
A number of significant objections leveled at Foucault provide the frame in which I
propose to carry out the development of this aspect of rule-governed practices and defend their
status. The defense, following the logic of the analytic of finitude and the Agrippan trilemma,
involves two strategies: a) since it is not possible to place oneself explicitly outside the space of
reasons without thereby (and implicitly) situating oneself inside it, I show how the concept of
experience need not be construed in terms of the inside-outside logic, b) I claim that insistence
on this dynamic as a transcendental requirement occludes dimensions of practices that become
visible through a Foucauldian account.
Two exemplary challenges provide the occasion for (a) and (b). I claim that there is a
symmetry between the objections which Derrida and Habermas present to Foucault, and by
extension, to the concept of experience I formulate here. It results from their insistence that the
transcendental standpoint must have primacy over the empirical one by right. Hence the
necessarily oblique nature of the responses I advance: I cannot place myself outside the space of
14
reasons without thereby impugning my rights, since every right presupposes that one is always
already placed inside reason‘s space. From this perspective, the space of confinement Foucault
describes can only appear as a self-defeating criticism of reason in the name of a better reason.
And since the figure of that ―better reason,‖ in Foucault‘s story, appears as madness itself,
beyond the institutionalizing work of reason, Foucault‘s refusal to offer explicit justification for
the standpoint from which that history can be written, or his refusal to refer the structure of
exclusion to a quasi-transcendental (non-)principle of iteration, necessarily appears as either
naivety or willful rebellion.
That necessity is what motivates the insistence of Habermas and Derrida. Against this
perspective, I insist that Foucault‘s refusal is also motivated: the history of madness is
simultaneously the history of reason. Therefore, if the genesis of the space of reasons is
inextricably bound up with the spaces of confinement, the history of that genesis cannot be
written from the transcendental standpoint alone. The genitive in ―history of madness‖ should be
understood both subjectively and objectively. But the subjective sense refers not to madness
itself which would finally speak its primitive purity, but rather to the flipside of the limits
criterial for our own space of reasons.
If the description of that entanglement necessarily appears as both subjective and
objective, that necessity is inscribed not in eidetic or transcendental structures but is the result of
historically situated antagonistic relations. Therefore, Foucault‘s articulation of experience as the
matrix of context-bound, but no less constitutive, rules of formation is more attuned to the
historical singularity of events and the critical interrogation of their conditions of emergence.
15
However, and this is the fourth stage, even if I am correct in claiming that the aim of
Foucault‘s investigations is to bring to light the set of historically specific practices criterial for
what counts as knowledge of objects and the proper treatment of subjects, there still remains the
question of whether (and why) that should (or could) be called ―experience‖. The question
becomes more urgent since the works in which Foucault offers his most explicit definitions of
experience are volumes two and three of The History of Sexuality, where what is at issue is ―a
history of the experience of sexuality,‖ (UP, 4) but where almost everything Foucault says
concerns what people have said about sexuality.
Compared with, for instance, Merleau-Ponty‘s acute and fine-grained descriptions of the
lived experience of sexuality, Foucault is particularly vulnerable to criticism if it turns out that he
is privileging discursive practices and, as a result, never saying anything except repeating what
other people said about everything. Therefore, his account of subjectivity, and how it fits into his
accounts of discursive and nondiscursive practices, needs further specification.
A more direct engagement with subjectivity is also required in order to explicate the
critical import of Foucault‘s discourse to the extent to which, in modernity, criticism is motivated
by, carried out in the name and grounded on the capacities of the subject, the fundamental one
being its spontaneity. The question, bluntly, is: Am I free to give up the kind of reflection
constitutive of modernity without thereby rendering freedom itself meaningless and impossible?
I propose to formulate my answers to these two questions through direct engagement
with two thinkers, Merleau-Ponty and Adorno. A critical encounter with Merleau-Ponty is
motivated by two main reasons: First, in the Phenomenology of Perception he thematizes the
experience of sexuality in a section titled ―The Body in its Sexual Being,‖ and the preface to the
16
same work offers a succinct discussion of his appropriation of phenomenology. Second, his
influence on the early Foucault is unmistakable, both in terminology and methodology.
On the other hand, Adorno, probably more than any other philosopher in the twentieth
century—with the possible exception of Derrida—has been sensitive to the recoil of critique on
the critic and lodged his thinking in the very space of that complicity between who criticizes and
what is criticized. A critical encounter with a significant moment of his work, namely, the
reciprocal mediation of subject and object, enables a sharper definition of whether Foucault can
claim any sort of legitimacy and efficacy for his practice of history.
I argue that insistence on the fact of our embodiment which, paradoxically, cannot be
reduced to any other fact, and which would provide a universal and necessary datum of any
kind—ontological or epistemological—fails to grasp Foucault‘s appeal to ―experience‖ at its
proper register—or in any event, at the register where I situate it here. That insistence on the fact
of our body as the site of the universal and necessary experiences of pain and pleasure, of desire
and its frustration, and of hunger, thirst and death is motivated by, I believe, the view that the
alternative is either a formalism about the virtues which makes ethics more rational than its
actuality can support, or a relativism without recourse. And I want to claim that this insistence is
mistaken, from a Foucauldian perspective, not because there is no such fact, but because it never
performs the role this view would have it do.
If Foucault is right that, for instance, Greek antiquity made a very different problem of
the appetites and their management, and consequently, did not take the gender of the partner to
be a very significant element in their reflection on sexual ethics or hygiene, then it does not
explain much of anything to refer to the ―experience of homosexuality‖ in ancient Greece—for
17
there was none in a sense that is salient for modernity; nor does it justify what appropriate
attitude we should have with respect to homosexuality.
In other words, the appeal to the universality and necessity of ground experiences, or of
structures of embodied comportment, or again of the ontological framework subtending my
corporeal existence, does not perform any epistemological role, and it does not provide any
normative purchase. Therefore, Foucault‘s reconfiguration of experience in terms of discursive
and nondiscursive practices cannot (should not) be taken as a prima facie reason for denying the
right to call that experience. Nothing of the softness of a caress, the abrasion of a desire, or the
intensity of hunger, is lost by referring experience to the conditions mediating the subject‘s selfrelation.
Therefore, my claim is that an understanding of experience as the articulation of
discursive and nondiscursive practices, where their articulation is neither a logical, nor linguistic,
nor even causal determination, provides a more honest description of human action precisely
because it refuses to read more rationality into it than is to be found. This articulation concerns
rather ―that less apparent syntax which causes words and things (next to and also opposite one
another) to ‗hold together‘‖ (OT, xviii). I have tried to relate this ―syntax‖ to Wittgenstein‘s
―grammar‖. However, even though I argue against a transcendental reading of Wittgenstein, the
opposite risk of that engagement is that critique may then appear only a matter of ―this is how we
go on‖; and critical work would be restricted to the description of the moves ―we‖ make in
different language games, and if ―they‖ do not move like us, then they are not one of us.
There are times when that may be all we can or even should say and do. At such times the
Foucauldian rejoinder against strategies of exclusion is not to ―cry normalization,‖ but to insist
18
that we realize that that is what we are doing, and accept responsibility for the consequences. It
therefore has more critical force than an argument which would transcendentally secure the
legitimacy of that exclusion by ―demonstrating‖ the irrationality of those who are so excluded,
without accounting for the constitutive role our social sanctions play in their very creation. That
insistence takes two forms: epistemic, which I discuss in terms of conditions of acceptability and
conditions of predication; and normative, which I discuss in terms of power relations and their
strategies. Demonstration, then, is complicit with the formation and exclusion of monsters, which
does not entail that we would thereby be excused from making distinctions in order to collapse
truth and falsehood, right and wrong, into a new night in which all cows are black.
Foucault‘s giving up of the transcendental standpoint is not a repudiation of reflection; it
is rather motivated by the conviction that the moment of self-relation entailed by reflection
cannot be anchored in any unreflected given. And that includes appeals to immediate intuition, or
will, or power, where these are taken as merely given to thought from the outside. There is no
unconditioned standpoint, transcendental or empirical, which would provide a refuge for thought
outside of the practices constitutive of its forms. And that also implies the inseparability of the
forms of thought from what is thought. Therefore, the abandonment of transcendental reflection
is at the same time a radicalization of self-reflection.
Perpetual negativity without grounds? I want to say, rather, ―back to the rough ground‖7.
In what follows, I try to understand why it is that we resist friction and to motivate the thought
that we can only walk if there is friction. And so I provide not a philosophical foundation but a
philosophical motivation for the ―changing forms revealed by a kaleidoscope,‖ which Borges
7 Wittgenstein, Philosophical Investigations, fourth edition (Blackwell Publishing Ltd, 2009) § 107.
19
requires for his imaginary monsters, and which could just as well be ―the slightly dusty archives
of pain‖ that Foucault‘s histories are.8
8 Jorge Luis Borges, The Book of Imaginary Beings (Penguin Classics, 2006) p. xv; Foucault, HM, p. xxxvi
20
Inasmuchas the new true object issues from it,
this dialectical movement which consciousness
exercises on itself and which affects both its
knowledge and its object, is precisely what is
called experience. (Hegel, Phenomenology of
Spirit.)
1. From the Theater of Ideas to the Black Box Emitting Sentences
In the context of his discussion of how epistemology came to be established as a foundational
discipline, attempting to answer the question ―how is our knowledge possible?‖, Richard Rorty
claims that the term ―experience‖ has become the epistemologists‘ name for their subject matter,
―a name for the ensemble of Cartesian cogitationes, Lockean ideas.‖9This conception of
experience, which is said to accompany philosophy-as-epistemology well into the twentieth
century despite some modifications, is a philosophical term of art quite distinct from its sense in
an expression like ―experience on the job‖. Rorty‘s overall aim is to render such a conception of
experience redundant by denying the coherence of the very problem for which it was intended as
a solution.
According to the narrative Rorty constructs, the official self-understanding of philosophy
as a discipline distinct from the sciences—as well as from other forms of cultural expression—
with a privileged role in accounting for the structure of human knowledge and experience, arises
from a number of ―confusions‖: for instance, Locke is charged with having confused explanation
with justification, whereas Kant is guilty of confusing predication with synthesis. However, there
is an even more overarching confusion, or rather a misguided project, which makes the particular
9 Richard Rorty, Philosophy and the Mirror of Nature, (Princeton: Princeton University Press, 1979) p.150
21
ones possible, viz., the project to model all knowledge on sense perception, and more
specifically, on ocular imagery: in short, knowing on seeing.
The fundamental Kantian distinction between concepts and intuitions as two irreducible
kinds of representations in ―inner space,‖ Rorty claims, attests to the fact that Kant does not
manage to break free of the Cartesian frame that spawns the conception of experience mentioned
above. According to this conception, we are asked simply to look at this ensemble (of
representations, ideas, impressions, etc.) and see that they fall into two basic kinds, those that are
immediately given/presented to the mind and those that represent the activity of thought.
However, if we follow Rorty‘s interpretation, we do not stumble upon such a distinction
inevitably when we reflect on our experience, but rather we would not know what counted as
―experience‖ unless we mastered that distinction in advance.
The upshot of the claim is that concepts/intuitions, just as ideas of reflection/ideas of
sensation in Locke, and ideas/impressions in Hume, are terms that admit only contextual
definitions; and that context in this case is a notion of experience as an ―inner space‖ furnished
with items whose causal relations appear problematic. However, this conception of experience in
turn is constructed in response to a mistaken view about knowledge which confuses ―knowing
that‖ with ―knowing of‖. And Rorty makes the following speculation about what the history of
philosophy would have looked like, were it not for this confusion:
If Kant had gone straight from the insight that ―the singular proposition‖ is not to be identified
with ―the singularity of a presentation to sense‖…to a view of knowledge as a relation between
persons and propositions, he would not have needed the notion of ―synthesis.‖ He might have
viewed a person as a black box emitting sentences, the justification for these emissions being
found in his relation to his environment….The question ―How is knowledge possible?‖ would
then have resembled the question ―How are telephones possible?‖ meaning something like ―How
can one build something which does that?‖ Physiological psychology, rather than
22
―epistemology,‖ would then have seemed the only legitimate follow-up to the De Anima and the
Essay Concerning Human Understanding.10
There are three interrelated points in this passage which constitute a challenge to any
enquiry into the concept of experience other than through accepted scientific procedures. First,
the statement ―this table is black,‖ say, is not to be understood as being presented with a black
table, or having the idea or representation of a black table ―in the mind‖; that is, the judgment
whereby I articulate my experience of a black table is not a matter of my mind simply having an
idea of a black table—and this is the insight Rorty credits Kant. Second, this insight should lead
us to think of knowledge not as a relation between a subject and an object, but rather as one
between a subject and propositions. The former view remains captive to the misguided project of
modeling knowledge on perception, and it misconstrues the paradigmatic case of a knowledge
claim as ―knowledge of‖; whereas the latter shifts the ―subject matter‖ of investigation to
propositions and justificatory practices, that is, to ―knowledge that‖. Third, as far as
philosophical investigation is concerned, the subject might as well be a ―black box emitting
sentences,‖ and the questions we might come to ask of this subject are not going to be of a
fundamentally different order than the questions we might feel inclined to ask about telephones.
What is at stake in this shift from subjectivity as a theater of ideas to subjectivity as a
black box, and how does it come about? What happens to ―experience‖ when it is no longer
conceived in terms of representation, but rather it finds its significance solely in the seamless
web of sentences? Is it possible to say anything significant about what experience is like and how
it is possible in a way that ―both goes beyond common sense and yet avoids any need to mess
about with neurons, or rats, or questionnaires‖?11
10 Ibid., p. 152 (My italics.)
11 Ibid., p. 151
23
What follows in this chapter is an attempt to provide some answers to these questions
through a discussion of what might tentatively be called the theories of experience in Kant,
Hegel, and Husserl. It is clear that the writings of any one of these figures would require a
separate study on its own in order to do justice to the complicated way in which they articulate
what they take to be ―experience‖ in relation to a host of other notions; and in order not to fall
back on ready-made labels such as empiricism, rationalism, idealism, materialism. To avoid such
pigeonholing as much as possible, and to keep the discussion focused, I will take the subjectobject
relation, and the role of experience in the imbrication of questions about subjectivity and
objectivity, as a guiding thread. This discussion will in turn set the main parameters of the
conceptual framework for its recasting through Foucauldian tools in the chapters that follow.
Since it could be said that part of the strategy I follow in relation to Foucault‘s concept of
experience parallels that of Rorty, or of some versions of pragmatism, their differences will stand
out only against the background of what they criticize.
At this stage, however, one might wonder why Kant, Hegel, and Husserl should be taken
to play this role, as opposed to many other alternatives. Even though such choices can be
justified only up to a certain point, it is possible to offer some reasons by way of justification.
First, since the subject-object relation, and its modalities, is how I propose to approach the
concept of experience, the figures referred to all have significant and detailed discussions of this
theme. Second, the way in which they approach this theme—for instance, the problem of
whether it is possible to make an absolute distinction between what is ―given‖ in experience as
opposed to what is ―taken,‖ or ―made‖ by the subject—is such that their answers render
problematic the very labels of empiricist, idealist, and the like, as appropriate terms of
appropriation. Third, it is possible to make a case to the effect that there is a line that could be
24
traced from Kant to Hegel, on the one hand, and to Husserl, on the other, in order to formulate
two significantly different ways in which experience can be ―thought‖. Finally, it is possible to
formulate a productive ―dialogue‖ between their conceptions and what Foucault has to say about
experience.12
1.1 Justification: Transcendental or Historical?
The phrase ‗possible experience‘ occurs frequently throughout the Critique of Pure Reason13 and
in some of the key passages where Kant appears to situate the entire critical project in relation to
it. Consider, for example, the following:
In transcendental cognition…this guideline is possible experience… [Transcendental proof] shows that
experience itself, hence the object of experience, would be impossible without such a connection [that is,
a synthetic a priori connection].
The conditions of the possibility of experience in general are at the same time conditions of the
possibility of the objects of experience, and on this account they have objective validity….
[W]hat is left for us? The possibility of experience, as a cognition in which in the end all objects must be
able to be given to us if their representation is to have objective reality for us.
[Pure reason] certainly erects secure principles, but not directly from concepts, but rather always only
indirectly through the relation of these concepts to something entirely contingent, namely possible
experience.14
So much emphasis does Kant seem to place on the notion of possible experience that we
may reasonably set aside some of his other formulations of what the critical project consists in,
12 These preliminary claims may suggest a view of the history of philosophy in terms that come too close to the
―timeless conversation between great thinkers‖ approach; more specifically, it might be taken to be operating on a
conception of history that would seem not only inappropriate in the context of Foucault, but also beg the question by
deciding at the outset that there is something definite which can be identified as experience as such. This is a
problem that will be discussed thematically later in this study, and so my claims should be taken as entailing only a
minimum of either ontological or epistemological commitment at this stage.
13 Immanuel Kant, Critique Of Pure Reason, trans. Paul Guyer and Allen W. Wood, (Cambridge University Press,
1998). All references are to this edition and will be indicated by CPR, followed by the standard edition pagination.
14 CPR, A782-3/B810-11; B197; A217/B264; A736-7/B764-5 (Emphases in the original.)
25
and instead limit ourselves to this one: how is experience possible? Since it will be impossible in
the context of this study to examine the entire conceptual scaffolding Kant builds in order to
answer this question, let alone the many problems of interpretation generated by it, I will attempt
to delineate Kant‘s concept of experience through a reading of the Transcendental Analytic, and
in particular, certain aspects of the Deduction.15
In the Cambridge edition of the Critique of Pure Reason, a footnote is added by the
editors to the section heading ―Transcendental Analytic, First Book, The Analytic of Concepts,‖
which says that: ―The following notes appear at this point in Kant‘s copy of the first
edition.‖16Among these notes is the following:
We remarked above that experience consists of synthetic propositions, and how synthetic a
posteriori propositions are possible is not to be regarded as a question requiring a solution, since
it is a fact. Now it is to be asked how this fact is possible. Experience consists of judgments, but it
is to be asked whether these empirical judgments do not in the end presuppose a priori (pure)
judgments….The problem is: How is experience possible? 1. What does the understanding do in
judgments in general? 2. What do the senses do in empirical judgments? 3. In empirical
cognition, what does the understanding, applied to the representations of the senses, do in order
to bring forth a cognition of objects?…The test for whether something is also experience, i.e., a
fact, is as it were experimentation with the universal propositions under which the particular
empirical judgment belongs. If the latter cannot stand under a universal rule for judging, if no
15 This account will require further complication and modification in light of Kant‘s claims in the Critique of
Practical Reason and the Critique of Judgment; but the relevant aspects of Kant‘s practical philosophy and his
discussions of aesthetics and teleology will be introduced in those parts of this study which treat Foucault. The
literature on the Critique of Pure Reason, and the transcendental deduction in particular, is immense. I skirt the
many thorny problems of interpretation and focus on some key aspects of the subject-object relation and the juridical
function of the deduction. Two texts which I found very helpful in grasping some of the Kantian moves I describe
are: Henry E. Allison, Kant‟s Transcendental Idealism, An interpretation and Defense, (New Haven: Yale
University Press, 1983), especially part 3; Sebastian Gardner, Kant and the Critique of Pure Reason (New York:
Routledge, 1999), especially parts 2 and 6.
16 Ibid., p. 202 (My italics.)
26
concept can be made out of that, then it is a vitium subreptionis [vicious fallacy]. Why in
superstition and credulity.17
These notes, which appear at a point of transition between the Aesthetic and the Analytic,
crystallize some of the key problems Kant deals with and the concepts he introduces in these two
sections. Around the fundamental problem of ―How is experience possible?‖ are constructed
such Kantian terms as understanding/sensibility, concepts/intuitions, and judgment.
1.11 Dependence of the Subject
By this point in the Critique of Pure Reason, Kant has established that there is a basic difference
to be made, at the most general level of a subject‘s relation to an object, between an object‘s
being given to consciousness and its being thought about. Kant uses the generic term
representation (Vorstellung) in order to designate the most basic element of cognition, that is,
anything subjective that can play a role in making judgments or knowledge claims about objects.
As such, its extension appears to be as wide as the Cartesian reformulation of ―idea‖. In line with
this definition, intuitions are those representations whereby objects are given to us, and concepts
are those representations whereby we think about objects. The concept/intuition distinction is
both presupposed by and entails a number of others.
Sensibility is the faculty (or cognitive power/function) that enables objects to be given,
and the understanding is the faculty that makes thought about objects possible. Intuitions relate
to objects immediately, whereas concepts relate to them, if and when they do so, mediately:18 It is
important to note that the German word for intuition is Anschauung, which means ―looking
17 The last mysterious question without its mark appears incongruous in this context. I offer an interpretation below
to spell out how what Kant has to say about experience may be related more explicitly to this worry concerning
superstition and credulity.
18 ―…that through which [a cognition] relates to [objects]…is intuition.… [Objects] are thought through the
understanding, and from it arise concepts.…All thought…must ultimately be related to intuitions…since there is no
other way in which objects could be given to us‖ (Ibid., A19/B33).
27
at‖.19Hence something like the object‘s phenomenal presence to a subject remains in the term.
Kant also writes about intuitions as ―singular representations,‖ that is to say, as representations of
one particular, individual thing, a single object: hence the immediacy of the relation of an
intuition to its object. The intuition/concept distinction, then, reproduces the particular/general
distinction, since a concept is, for Kant, by definition general.
A concept can apply to more than one particular, i.e., it may have more than one
instantiation. Sensibility, as the faculty of intuitions, is a capacity for receptivity: the subject
forms intuitions only through being affected, passively. Finite subjects, that is, subjects whose
sensibility is subject to the conditions of space and time in Kant‘s definition, must be affected in
order for objects to be given to them.
Two other terms should be noted in connection with intuitions and sensibility: First, the
―material‖ aspect of intuition that requires the subject‘s being affected is sensation, which is a
posteriori; second, sensation comprises a manifold, that is, a multiplicity. Even though Kant‘s
argument in the Deduction will complicate the series of dichotomies that are introduced here,
Kant seems to retain, throughout his critical interrogation of conditions of possibility, the
traditional empiricist account of sensation: one aspect of sense experience requires the subject‘s
being somehow impinged upon from the ―outside‖. The understanding, in contrast, is claimed to
be spontaneous and represents the active element in cognition. On the basis of these foundational
distinctions, what I want to understand is the specifically critical claim that subjective agency
plays an irreducible role in the constitution of experience.
Since the guiding thread on which I am focusing is the subject-object relation, it is worth
recalling the well-known Kantian analogy with Copernicus:
19 It also has the connotations of view, opinion, experience, as in the expression aus eigener Anschauung, ―from
personal, or one‘s own experience‖.
28
Up to now it has been assumed that all our cognition must conform to the objects; but all attempts
to find out something about them a priori through concepts that would extend our cognition have,
on this presupposition, come to nothing. Hence let us once try whether we do not get farther…by
assuming that the objects must conform to our cognition, which would agree better with the
requested possibility of an a priori cognition of them, which is to establish something about
objects before they are given to us.20
Kant interprets the problem of cognition, i.e. forming a possibly true or false knowledge claim
about objects, in terms of a ―connecting‖ relation of some kind between a subject and an object.
It is legitimate to ask what this relation consists in and why it requires an explanation. It appears
to be a mysterious relation in need of an explanation because the two poles that are to be
―connected‖ seem utterly heterogeneous and independent of each other: on the one hand, we
have objects that have constitutions without any essential reference to subjects; on the other
hand, we have subjects whose only point of access to reality so conceived is their ability to
somehow represent it.
Now, if we try to understand this ―representing‖ simply as a function of reality‘s
impressing itself on, or impinging upon subjects, then this would seem to presuppose what is in
need of explanation: it would imply that subjects have an ability to transform the impression of
reality into a representation simply by being immanently related to reality. But if, on the other
hand, one interprets this fundamental connection as solely the result of the subject‘s activity, then
such a connection would imply the capacity to ―reach out‖ and ―read off‖ the features of reality.
If that were the case, however, the only condition accounting for how the subject knows so to
―locate‖ and ―read‖ reality would be, once again, the mere assertion of its immanent connection
to reality.
20 Ibid., B xvi
29
Therefore, Kant‘s critical interrogation of the conditions of possible experience is
motivated by his argument that the assertion of the object as originally independent from the
subject must be ―dogmatic‖: any attempt to account for the connecting relation between the two
will either presuppose what needs to be explained (and hence be circular), or postulate a further,
more fundamental relation, which will generate an infinite regress. Moreover the attempt to
avoid these alternatives by invoking a third term, which would be independent of the subject and
the object (e.g. God), would require that one have a true representation of this third term; which
assumption presupposes, once again, the very connection in need of explanation.
Hence the upshot of Kant‘s claim is that we cannot simply step outside the subject-object
relation (our representations) in order then to offer an independent description of the two sides of
the relation. Even if there is a fundamental connecting relation between the subject and the object
which allows the latter to become an object for the former, in order for the subject to represent
this relation, it would need to stand outside its capacity for representation; and that it cannot do.
Therefore, it is the very possibility of objects, how something can become an object for a subject,
that requires explanation; and Kant proposes that we adopt the ―hypothesis‖ of ―objects
conforming to our mode of cognition‖.21
The constitution of objects, at the most fundamental level, is going to be a function of
how the subject is. And if objects are determined at the most basic level by the subject, this
means that the subject is active in cognition, i.e. it must be considered as making it the case that
objects conform to its mode of cognition. It can do this, however, only if it in some sense
21 The analogy with Copernicus implies that, just as the apparent movement of the sun is to be explained in terms of
the movement of the observer, our knowledge of independently existing objects is to be explained in terms of our
mode of cognition: the very possibility of knowledge of independent objects is referred to the mode of cognition of
the subject. In both cases, we have a redescription of what was taken to be an independent reality as an appearance,
dependent on the subject. It is this notion of dependence/independence which will become one of the key
components of Hegel‘s reformulation of what it is for a subject to have experience.
30
actively produces the object. Then everything hinges on how we are supposed to understand this
relation of active constitution/production.22I want to see how this theme gets developed by Hegel
before I take up again the concept of constitution in Kant in relation to Husserl‘s version of
phenomenology.
1.12 Independence of the Subject
Toward the end of the Introduction to the Phenomenology of Spirit,23 Hegel writes that the stages
of consciousness which we ―see‖ unfold before us is the origination of ever new objects that
present themselves to consciousness without its understanding how this happens, ―as it were,
behind the back of consciousness‖ (PS, 56). Hence in the movement of consciousness there
occurs a moment which is not present to the consciousness comprehended in the experience
itself. For us, the phenomenological we, simply looking on, this appears as a process of
becoming. And he adds that ―… the way to Science is itself already Science, and hence, in virtue
of its content, is the Science of the experience of consciousness‖ (Ibid).24 What is this
experience, of which phenomenology is the science? And, how is the experience of reading the
Phenomenology, the paradoxical ―simply looking on,‖ related to that of which it is a science?
Hegel‘s term for experience in the Phenomenology of Spirit is Erfahren and Erfahrung25
The root meaning of Erfahren is setting out on a journey to explore or get to know something;
22 Rorty, for instance, dismisses any notion of constitution as metaphysical: ―It may be shocking to call Kant‘s
account ‗causal,‘ but the notion of ‗transcendental constitution‘ is entirely parasitical on the Descartes-Locke notion
of the mechanics of inner space…‖ (op. cit. n. 31, p. 151). And he continues to claim that Kant‘s use of ―ground‖
instead of ―cause‖ only masks this.The following discussion will show that this claim is not entirely correct.
23 G. W. F. Hegel, Phenomenology of Spirit, (New York: Oxford University Press, 1977) Hereafter PS.
24 The original title for the Phenomenology of Spirit, which survived in the first edition, was Science of the
Experience of Consciousness. Heidegger‘s reading in Hegel‟s Concept of Experience (Harper Row, 1970) starts off
from this observation and tries to establish the subjectivism of Hegel‘s philosophy.
25 For the different uses and connotations of the term in German, see Michael Inwood, A Hegel Dictionary, (Malden,
MA : Blackwell Publishers, 1992) ; on its contrast with Erlebnis, Martin Jay, Songs of Experience, Modern
American and European Variations on a Universal Theme, (Los Angeles: University of California Press, 2005) pp.
9-12.
31
and Erfahrung refers either to this process or to its result. Hence something in the ordinary
meaning of Erfahrung resonates directly with, or lends itself to a description of, the process of
the emergence of new objects of consciousness Hegel describes in the Introduction. The
experience undergone by consciousness on its way to science, which phenomenology is, is a
voyage of discovery. This experience is not specifically empirical, since its primary contrast is
not with thought. Rather in contrast with what we, the phenomenological onlookers, know about
it, experience refers to what consciousness undergoes and finds out for itself as it moves from
one of its ―shapes‖ to the next through their inadequacy26: and here ―inadequacy‖ means the
internal incoherence between its object and its conception of that object. What is interesting,
however, is the implication that, for Hegel, the very transformation of one conception into
another, as a result of this realization, itself has an experiential dimension. In other words, it
cannot be grasped simply in its logical function.
Hence it is important to distinguish this broader conception of experience from the sense
of ―empirical‖ as that which is given to consciousness unprocessed by any conceptual or
imaginative activity.27 As his critique of sense-certainty shows, Hegel agrees with Kant‘s
reflective account of experience. On this view, philosophical reflection must explicate what a
subject of experience must be like in order to count as a subject that can ―have experiences,‖
where such experiences are empirical representations of objects and where such representing is a
function of judgment. Kant‘s transcendental account lays out the transcendental conditions of
possibility of such experience. Even though Hegel claims that Kant‘s determinations were
―arbitrary,‖ especially the categorial distinction between the legitimate conceptualization of
26 ―The experience of itself which consciousness goes through can, in accordance with its Notion, comprehend
nothing less than the entire system of consciousness, or the entire realm of the truth of Spirit‖ (PS, 56).
27 For instance, Hume‘s impressions would be just such brute givens, at least on the standard reading of Hume‘s
empiricism.
32
sensory material one finds in the natural sciences and the application of concepts to transcendent
entities (e.g. God), there are three Kantian claims concerning the possibility of experience that he
accepts (and modifies in the process).
Being in a subjective state, however momentary or enduring, does not suffice to have an
experience of that state.28 For this to count as an experience, I must apply a determinate concept
and take myself to be in, that is, judge that I am in such and such a state. This must be something
I do and must know that I am doing. Therefore, the subject of experience must be identical
through time: the experience must belong to the subject who experiences. Moreover, this
belonging together of representations in a subject must be the result of the subject‘s activity. The
relevant unity of representations can only be brought about by the spontaneous activity of the
subject, since this concerns the connection of representations according to rules (concepts).
Finally, the subject of experience can be the same subject throughout experience, only if it could,
in principle, become conscious of itself as actively applying such laws. Every perception is the
taking of something as some determinate thing, and the Kantian twist which Hegel retains is that,
when I perceive an object, I must implicitly take myself as perceiving that (determinate) object.
Self-consciousness is a condition of experience, and therefore experience is implicitly reflexive.
28 My discussion of a continuity between Kant‘s account in the transcendental deduction and the function which
Hegel attributes to his phenomenology in the system follows the thesis advanced by Robert B. Pippin in Hegel‟s
Idealism: The Satisfactions of Self-Consciousness, (New York: Cambridge University Press, 1989) especially
chapter 2. Even though I think that his stress on the epistemological issues connecting Kant and Hegel at times
occludes other dimensions of Hegel‘s thought, I find his development of the ―apperceptive theme‖ through German
Idealism lucid and compelling. However, the distinction he sometimes wants to make between an ―epistemological
Hegel‖ and a ―social-historical Hegel‖ can no longer be maintained on Hegelian terms, since, if my account is
correct, no sense can be given to ―reasons‖ which would abstract from their entanglement with ―social sanctions‖.
33
However, Hegel criticizes Kant‘s conception of experience (and its subject) as arbitrary,
psychologistic and finite, yet not without preserving its moment of truth. Consider the following
passage:29
If I say ―I,‖ this is the abstract self-relation, and what is posited in this unity is infected by it, and
transformed into it. Thus the Ego is, so to speak, the crucible and the fire through which the
indifferent multiplicity is consumed and reduced to unity. This, then, is what Kant calls ―pure
apperception‖….What human beings strive for in general is cognition of the world; we strive to
appropriate it and to conquer it. To this end the reality of the world must be crushed as it were;
i.e. it must be made ideal.
Hegel‘s ―description‖ of consciousness in the Phenomenology of Spirit is intelligible
against the background of his acceptance of the fundamental ―idealizing‖ move which the critical
interrogation of conditions of possibility (of experience) entails. He provides a reconstructed,
ideal ―history‖ of what a subject would come to regard as the conditions of possibility of its
experiences (where experience includes judgment, justification and knowledge), in such a way
that this history is determined by a phenomenology of what it would be like to participate in
various candidate practices constituting experiences. Even though any actual subject may only
have a vague idea of such conditions, such a subject is still self-consciously making a judgment
about a determinate object or event.
In order better to explicate Hegel‘s modification of Kant‘s appeal to the transcendental
unity of apperception, that is, the implicitly self-conscious nature of experience, I will consider
the following example30. Subject S, in accounting for an action, offers a moral justification. What
are the conditions that make this a moral justification? First, there must be the institution of
morality. If there were no such institution, S would merely ―think‖ that he was giving a moral
29 Hegel, The Encyclopedia Logic, Part I of the Encyclopedia of Philosophical Sciences, (Indianapolis: Hackett
Publishing Company, 1991) pp. 84-85
30 Pippin, op. cit. p. 22
34
justification. Second, S cannot merely reproduce the words sanctioned by this institution, or offer
this judgment as a result of behavioral conditioning, without some self-understanding that that is
what the institution demands and that fulfilling such a demand is what he is doing31. Therefore, S
must take himself to be offering such a justification for it to count as one. If we generalize this
example to cover all possible experiences, we could say that unless a subject could become
conscious of the rules it applies in unifying its representations, it would not be following rules or
representing objects; it would merely associatively produce subjective states. Since the latter are
merely associated, they could not count as representations, that is, as having objects or being
experiences.
This example crystallizes how a Kantian reflection on experience is inseparable from the
question of justification, and hence that of its ―deduction‖32. I want to advance the following
claim: Hegel‘s concept of experience is best understood when fleshed out in terms of his
―transposing‖ of Kant‘s transcendental deduction into a retrospective historical reconstruction.
The central question of Kantian critical philosophy, which subsequent German idealism inherits,
is: how can thought successfully determine a priori what is other than thought? And what is at
stake in answering this question is nothing less than the possibility of experience.
I want to adumbrate the key moments of Kant‘s answer.
31 A crucial question is just how much implicit self-understanding one must ascribe to the subject in order for what
he does to count as following a rule. In the next chapter I raise some objections to this requirement, which is
inseparable from the conception according to which the very notion of experience implies a concern with
legitimation.
32 The key Kantian distinction between ―question of fact‖ and ―question of right‖ occurs at B 116-117. He explicitly
identifies his task as that of the deduction of what legal title we have to the employment of the categories, i.e.
whether we are entitled to apply them in experience. Therefore, the question concerning the objective validity of the
categories is recast in terms of their justification: subjective conditions of thought will have objective validity, if it
can be shown that no object of experience in general would be possible in their absence. And as I claimed above,
―possible experience‖ is the concept which Kant intends to demarcate this legitimacy: the use of the categories is
justified within the limits of possible experience. For a discussion of how this theme gets taken up by nineteenth
century sociology, see Gillian Rose, Hegel contra Sociology, (London: The Athlone Press, 1981).
35
First, experience is empirical knowledge. So the conditions for the possibility of
experience are those which are necessary for our capacity to make possibly true or false
(cognitive) judgments about the world and ourselves. Experience of any kind, even that which
consists of the internal flow of the subject‘s mental states, presupposes this cognitive
ability.33Second, this ability in turn requires that the subject unify a manifold, which is a
synthetic activity: I cannot simply inspect my mental states to have an experience, but rather
must actively bring them to a unity. This activity, however, is complex.
It involves, for instance, the recognition of disparate states as belonging together in one
representation and the discrimination between different states in some manifold of intuition.
Third, if synthetic activity is necessary, then it must somehow be constrained. Or, in other
words, there must be normative limitations on the subject‘s activity of representation. Kant‘s
insight is to deny the object the status of such a constraint, since such a move would beg the
question.34 These objectivity-conferring rules are the conditions under which a unified, implicitly
self-conscious subject of experience is possible.35 Finally, these conditions are the pure rules
which, prior to any experience, already determine what counts in general as an object of
experience. Kant calls these pure rules/concepts that are necessary for the possibility of
experience ―categories‖. They are objectively valid precisely because there could be no object of
33 For instance, judging that such states are flowing in this and no other order.
34 This is why Kant motivates the critical turn by arguing for the infinite regress or circularity of any attempt to
ground the subject-object relation on the dogmatic assertion of their independence and immanent connection.
Another way in which this point can be made is the following by John McDowell, in the context of a discussion of
W. Sellars: ―…suppose we want to conceive the course of a subject‘s experience as made up of impressions,
impingements by the world on a possessor of sensory capacities. Surely such talk of impingements by the world is
‗empirical description‘; [but since] to identify something as an impression is to place it in a logical space [that is
other than the normative one]…experience as made up of impressions…cannot serve as a tribunal.‖ Mind and
World, (Cambridge, MA: Harvard University Press, 2002) p. xv.
35 The transcendental unity of apperception is just such a subject as formal condition of possibility of experience.
36
experience unless what is given in a manifold of intuition had already been synthesized in
accordance with the categories.36
Hegel‘s concept of experience preserves what he takes to be the moment of truth in this
brief sketch of the deduction, while cancelling the formal apparatus of forms of judgment on
which Kant relies. If the Kantian deduction, by invoking the transcendental unity of
apperception, is successful in showing that the objects of experience, in their very givenness,
conform to the categories, then it shows, by the same token, that what counts as given in
experience is also determined by conceptual conditions. Therefore, it is determined ultimately by
the spontaneity of the subject. This, however, renders problematic the peremptory distinction
between sensibility and understanding, intuition and concept. The blurring of the conceptintuition
dichotomy which is implied by the principle underpinning the deduction is precisely
what forms the core of Hegel‘s repudiation of the thing-in-itself as a superfluous postulation of
Kant‘s uncritical presuppositions. Hegel thereby moves from thought‘s relation to the pure
manifold of intuition to thought‘s self-determination37.
Hegel argues that thought‘s self-determination is what accounts for the ideal nature of
experience, because Kant‘s unquestioned assumption of two distinct cognitive faculties
(understanding-sensibility) collapses under the weight of the very terms of the deduction.
Throughout the Critique of Pure Reason there is a lingering sense that the universal and
necessary categorial conditions for possible experience actually determine only our experience,
i.e. finite creatures with sensible intuition and discursive understanding. The experience which
36 Hence the passage I quoted at the beginning of this section: ―The conditions of the possibility of experience in
general are at the same time conditions of the possibility of the objects of experience, and on this account they
have objective validity…‖ (CPR, B197).
37 See Pippin, p. 31: Hegel‘s repudiation of the concept-intuition distinction does not eliminate the role of the given
in knowledge but ―radically relativizes to thought the ways in which the given can be taken to be given.‖
37
the categories make possible is thereby restricted to a phenomenal world; but this generates the
corollary need to anchor the phenomena in a noumenal thing-in-itself, as an irreducible ground.
But if we can only ever have objects of experience under categorial conditions, then we cannot
have legitimate knowledge of things in themselves. Therefore, from a Hegelian point of view,
this conception is ―psychologistic‖ and ―finite‖.
Hegel‘s rejoinder then involves two negative claims. First, Kant gratuitously limits
subjectivity to consciousness. Second, Kant interprets consciousness as a private spectator of
itself and of the world. The positive corrective to these deficiencies in the Kantian account is that
subjectivity is Spirit (Geist) and that consciousness is communal, public, and socially interactive.
Since, however, Hegel still accepts the fundamental Kantian gesture that understands experience
as basically reflexive and continues the search for the conditions of its possibility, the result is an
uncanny view of the a priori. If there are fundamental notions, which are necessary for cognitive
experience and which are neither casual products of objects independent of subjectivity, nor
empty forms, the content of which is supplied ―from the outside‖; then, what can count as such
conditions?
The answer involves the Hegelian Notion (der Begriff). Hegel‘s argument is that there is
a level of fundamental concepts presupposed by the possibility of experience, and therefore
unrevisable in the light of experience. However, this does not entail for Hegel that this ―notional‖
level is unrevisable tout court. This is what renders his account of the a priori uncanny, for the
Kantian categories are static forms, whereas Hegel claims that the Notion ―moves‖38: the
attempts to make use of any given, determinate pure concept in discriminating objects fail. This
failure in turn requires the revision of such a Notion. More specifically, the failure leads to a
38 See, for instance, EL, p. 316 and PS, pp. 287, 350.
38
revision of the Notion‘s exclusiveness or independence in a way which necessitates an
understanding of its relation to concepts that may have initially occurred as its other or
contrary.39 Thus the movement of the Notion is dialectical: thought revises itself in an attempt to
overcome its difficulties, a process that is purely internal.
Hegel‘s rejection of the concept-intuition dichotomy is helpful in fleshing out what is
involved in this dialectical development of the Notion which determines experience at a deep and
peculiarly a priori level. Recall that the key Kantian move, which functions as the linchpin of the
deduction of the categories, is that in any case of my taking this to be P, the intuited particular
can play no cognitively significant role except as an already minimally conceptualized
particular.40
So in any judgment I must have already taken this to be this-such (e.g. this thing, here,
now.) Hegel‘s denial of sensibility as a disparate faculty, then, involves the claim that neither the
interaction between the world and the subject, nor some mental state produced by that
interaction, in and of itself, can function independently of the active role of the Notion. Hence a
mere intuition, as defined by Kant despite his blurring of it in the deduction, cannot be a nonconceptual
item to which concepts are applied from the outside; and so it cannot serve as a
grounding reason. Therefore the Hegelian reinterpretation of conceptuality entails that
conceptualization of intuition is the conceptualization of what we take to be intuitively given: the
intuited content is not simply a brute given, but must be already determined as such in some way.
39 A process which Brian O‘Connor calls the horizontal aspect of conceptuality: Adorno‟s Negative Dialectic,
Philosophy and the Possibility of Critical Rationality, (Cambridge, MA: MIT Press, 2005) Part 1, section 2.
40 In other words, the intuited particular is always already minimally conceptualized. McDowell, for instance, sets
up this aspect of Idealism in terms of the myth of the given, but in order to rehabilitate a role for receptivity which
would be irreducible to its conceptual uptake; see Mind and World, p. 8. The difficulty, then, is how exactly to
account for the constraint intuitions in this quasi-independent role would impose on spontaneity, once it is granted
that intuitions count as constraints on thought only once they have been conceptualized.
39
The Notion is precisely such a non-empirical, spontaneously self-moving condition of
experience. It originally determines the possibility and most basic character of experience.
Moreover, once Hegel jettisons the concept-intuition duality, he no longer takes this relativity of
objects of experience to our conceptual structure as a limitation. Since there no longer is a
possible contrast between our conceptual framework and the world, there can be no such
limitation. And, according to Hegel, it is precisely this move which results in the infinity of
Absolute Knowledge over and above the finitude of subjective idealism. Finally, then, the
problem which Kant generates but cannot satisfactorily resolve is recast: if the phenomenal
world is conditioned by our conceptual framework, then a different conceptual framework would
entail a different phenomenal world; this is what produces the thing-in-itself as remainder and
point of anchorage. Hegel‘s account, if correct, means that the noumenal world is a ―world well
lost‖41.
From this perspective, Kant has no way of showing the objectivity of the conditions for
an implicitly self-conscious experience of objects. He can only relativize claims about objects to
claims about mere phenomena, or he can create an infinite, and hence in a sense impossible, task
of their asymptotic convergence. So the reconciliation between subject and object, the need for
which is made possible and denied at the same time by the turn to a reflexive account of
experience, cannot occur. Hegel regards Kant‘s attempt as ultimately succumbing to the tragedy
of human finitude: by insisting on a fundamental difference between the human and the divine
perspectives, we are trapped in the domain of appearances; whereas only a being with
intellectual intuition could have access to things in themselves. So in order to make good on his
speculative promise that subject and object are identical, that ―everything turns on grasping the
41 Pippin, p. 277 and Rorty, ―The World Well Lost,‖ Journal of Philosophy 69 (1972) pp. 649-666.
40
True, not only as substance, but equally as Subject‖ (PS, 10), Hegel offers a new theory of
subjectivity, i.e. of what it is to be a self-conscious subject in a conceptually mediated relation
with objects: an account of Spirit.
In the Introduction to the Science of Logic, Hegel writes the following about what he
takes to be the task and accomplishment of his phenomenology:
In the Phenomenology of Spirit I have exhibited consciousness in its movement onwards from the
first immediate opposition of itself and its object to absolute knowing. The path of this movement
goes through every form of the relation of consciousness to the object and has the Notion of
science for its result. This Notion therefore needs not justification [in the SL] because it has
received it in that work; and it cannot be justified in any other way than by its emergence in
consciousness, all the forms of which are resolved into this Notion as into their truth. … [A]
definition of science … has its proof solely in the already mentioned necessity of its emergence in
consciousness. … Phenomenology of Spirit is nothing other than the deduction of [the Notion of
pure Science].42
In Hegel‘s own words, then, the Phenomenology of Spirit provides us with the elimination of the
possible separation between the object and the subject. Thus the progress of consciousness on its
―path of despair‖ offers a ―deduction‖ in order to overcome the reflexive worry that in asserting
the conditions of possible experience we are providing only subjective conditions, i.e. describing
the subject‘s criteria for self-certainty. Against such doubts, which still plague the Kantian
deduction, Hegel‘s phenomenological deduction is supposed to establish the objectivity of
Spirit‘s experience of itself, which constitutes absolute knowledge.
In the language of the Phenomenology of Spirit, the Absolute is the Notion, and such
notionality is the logic of a self-determining subjectivity. Hegel‘s strategy then is to co-opt the
skeptic, and in the process manifest the absolute objectivity of the Notion (its basic structure and
42 Hegel, Science of Logic, (New York: Humanity Books, 1969) pp. 48-49.
41
how it determines actuality at a deep level). Skeptical objections can only arise within an
experience determined by some Notion. What this claim implies should be clearer given the
discussion of the implicitly reflexive character of experience above. A condition of conscious
experience is an implicit Notional presupposition; hence the possibility of consciousness relating
itself to objects requires that consciousness take itself to be in some kind of relation to objects in
general: there must be a prior normative presupposition of what there is to relate oneself to. And
since concepts of particular kinds of objects (empirical concepts) can only be formed through
interaction with objects, this Notional criterion of what counts as an object cannot be empirically
verified or falsified.
Thus Hegel argues that ―[c]onsciousness provides its own criterion from within itself, so
that the investigation becomes a comparison of consciousness with itself‖ (PS, 53). It is a
question of determining whether consciousness‘s notional assumption about what there is to
know, and how consciousness could access it, is in fact what consciousness takes it to be. In
other words, the problem concerns whether a certain conception of experience, i.e. the conditions
under which a cognitive relation to objects can be established, can in fact account for such a
relation to objects. Thus the Notion, at this level of analysis, is a criterion for determining the
possibility of objects. It is a criterion of (what counts as) an object itself, and as such it is as deep
a condition of cognition as can be.
Moreover, Hegel claims that any such notion can only be understood as a function of
other possible notions, i.e. any such notion will fail unless transformed and expanded
systematically in relation to other notions. Finally, since this movement of the notion determines
the basic structure of experience, including the possibility of the objects of experience, actuality
42
itself must change. Actuality constituted by the dialectically self-determining Notion is
Wirklichkeit.43
There are, then, five fundamental features of the concept of experience which a line
traced from Kant‘s transcendental unity of apperception to Hegel‘s Spirit makes visible: First,
the subject‘s cognitive relation to objects is possible only if there is a ―subjective‖ presupposition
of some notion of objects in general. Second, this presupposition is implicit, but it is nevertheless
a reflection of the subject‘s spontaneous apperception, its taking the objects of its experience to
be such objects. Third, in any experience of objects, if and when such a notional condition is
inadequate for self-conscious experience, this inadequacy can be determined internally, by a
description of what such an experience would be like. Fourth, the upshot of this description is
that a cognitive relation to objects could not be established with such and such a notional
presupposition. And fifth, the subject constituted by (the failed) notion is opposed to or
dissatisfied with itself.
The Phenomenology of Spirit is therefore the idealized history of Spirit‘s self-education
or self-formation (Bildung), tracing the trajectory of the experience of an idealized subject‘s
opposition to itself: the opposition between its self-determining activity and what it is trying to
determine. This self-consciousness depends on the experienced development of the notion, and
not on a table of categories as in Kant.
Hegel refers to this experience as the ―dialectical movement which consciousness
exercises on itself and which affects both its knowledge and its object.‖ (PS, 55, my
43 ―…as Reason, assured of itself, it is at peace with them, and can endure them; for it is certain that it is itself
reality, or that everything actual is none other than itself; its thinking is itself directly actuality…‖ (PS, 139).
43
italics).44Hegel‘s concept of experience, then, is temporal and historical precisely because it
traces out the movement of the Notion necessary for there to be any experience of particular
objects. The question of the legitimacy of any notional determination (shape of consciousness)
can only arise relative to other possible notions. To the extent that any notion, in isolation, will
fail to adequately establish a coherent relation to objects, experience will be a ―path of despair‖:
we can doubt not only the legitimacy of particular claims, but also the implicit notion
presupposed for there to be any determinate claim.45
The overcoming of despair can only be accomplished if our Notions change within a
progressively more adequate articulation of the Absolute. Hence experience so understood attests
to the necessity of prior self-interpretive activity for consciousness to attain the standpoint of the
fully developed Notion. If all basic positions that assume the non-identity of subject and object
are experienced as unsatisfactory, the sublation of each opposition (between shapes of
consciousness) will result in the identity of subject and object. Therefore, the cumulative
development of this process will be teleological, to the extent to which the identity aimed at must
be presupposed as already determining the movement through the figures of non-identity.
Because dissatisfaction so understood drives the movement of Spirit, ―diremption‖ is
central to the speculative concept of experience. The diremption of subject from object, through
the course of the Phenomenology of Spirit, takes on different guises; or rather, it serves as the
44 He writes: ―In as much as the new true object issues from it, this dialectical movement which consciousness
exercises on itself and which affects both its knowledge and its object, is precisely what is called experience
[Erfahrung]‖ (Ibid).
45 As ―despair,‖ such self-doubt goes beyond Cartesian methodical doubt, since it is a question of there being even
merely mental representations, or ideas. Of course, even Descartes, at the beginning of the Second Meditation,
writes as if his resolution to consider as false everything which admits of the slightest doubt leads him to something
like despair: ―It feels as if I have fallen unexpectedly into a deep whirlpool which tumbles me around so that I can
neither stand on the bottom nor swim up to the top.‖ The Philosophical Writings of Descartes, vol. 2 (Cambridge
University Press, 1985) p. 16
44
most general figuration of a series of divisions such as particular/universal and certainty/truth.
For instance, the description of ―unhappy consciousness‖ traces consciousness in its separation
from and identity with its timeless, universal other, which consciousness takes to be the truth
about itself (PS, 119-139). Hegel writes that ―what it does not know itself is that this its object,
the Unchangeable, which it knows essentially in the form of individuality, is its own self, is itself
the individuality of consciousness‖ (PS, 131). Hence the attempt by consciousness to sacrifice its
material particular life in order to affirm the universal is self-defeating: First, because the ascetic
pleasure consciousness derives from the pain of corporal abnegation keeps the body as a focus of
attention; and second, because this renunciation in fact shows sensuous fulfillment as integral to
spiritual transcendence.
The coming to be of this realization of the entanglement of the universal and the
particular is Hegelian experience. The changing relation between subject and object requires that
the subject redefine who it is and what its object is in response to its failed attempts at relating
itself to its other. As the example of unhappy consciousness evinces, experience is at once a
recognition and a misrecognition. The negative thrust of this dialectical movement, which is
experience, accentuates the failure of any isolated moment to arrive at a satisfactory coherent
self-understanding and articulation. However, the teleological development of the
Phenomenology means that speculative discourse, which articulates the positive moment of this
process, comes to full objectivity only when experiences are given an appropriate narrative order
in memory. Narrative recollection, exemplified by the path traversed by consciousness in the
Phenomenology of Spirit, is open to contingency, while retrospectively and at the same time
imposing a closure. It is therefore the sublation of contingency and necessity.
45
The narrative recollection which is the culmination of experience is also a collective
recollection. Hegel moves away from Kant‘s transcendental subjectivity, with its formal
interpretation of categorial conditions, toward the achievement of some fundamental ―likemindedness‖
as the condition of knowledge46. We thereby shift away from a transcendental
deduction toward an account of the genesis whereby such like-mindedness is achieved as a
result. And it is this genetic account that bears the weight of justification, since we can no longer
appeal to transcendental conditions. This move then entails the reconfiguration of the
problematic of self-consciousness.
Recall the example of the agent who makes a moral judgment. It is possible to generalize
it to cover any claim-making and pursue its conditions further. Any claim-making activity can
count as a possible objective judgment only within the context of the practice or institution
governing such activity. There is such an activity only if a community of participants take
themselves to be participating in it within norms that define it as the practice it is and no other
practice.
Hegel‘s claim is that we must be able to reconstruct why such participants would come to
take such a practice as constraining possible judgments about objects, acceptable facts, legitimate
political institutions, etc. The objectivity of the practice at issue will have been established if our
reconstruction displays a rational inevitability in the development of such progressive selfconsciousness.
This reconfiguration transforms the pursuit of cognitive experience in general,
and knowledge in particular, into participation in a social practice/institution that is rulegoverned,
collective, and teleological. In order to evaluate the rationality of such practices one
46 I borrow the expression from Jonathan Lear; I discuss his argument in the next section.
46
must consider such self-consciously held criteria as social norms and possible grounds for
mutual recognition.
In Hegel‘s account, we do in fact get a necessary, progressive negation of the grounds for
recognition that are inadequate; for instance, the mere exercise of power as one such ground is
shown to generate contradictions which can only be resolved by invoking criteria other than
power. Self-consciousness presupposes recognition, and genuine mutual recognition requires a
genuinely universal basis of recognition47. In Hegel‘s language, this transformation of selfconsciousness
points towards an ―I that is a We and a We that is an I,‖ that is, Spirit48. Spirit then
is a community of mutually recognizing individuals. The basic institutions of such a community
will be legitimate, because they will embody universal recognition. Thus the main thrust of
Hegel‘s transformation is in place: cognitive experience is itself a social institution, and
rationality is ultimately explicable in terms of recognition. The subject of experience, then,
determining for itself its own fundamental Notions, is Spirit: a collective, socially selfdetermining,
self-realizing subject.
The concept of recognition, and its role in Hegel, presents many problems of
interpretation. In accordance with the line I have so far traced, and in anticipation of the theme of
activity-passivity in the following chapters, I want to focus only on a few salient moments. The
section of the Phenomenology of Spirit titled ―Independence and Dependence of Self-
Consciousness,‖49 provides a transition from organic life where there is only a ―sentiment‖ of
self to the domain of subjectivity proper: self-consciousness. It is not unreasonable to claim that
47 PS, pp. 104-111.
48 Ibid., pp. 110.
49 The following discussion is based on PS, pp. 104-119, inclusive of the introductory remarks under ―The Truth of
Self-Certainty‖.
47
this section articulates the conditions of emergence of self-consciousness, where condition talk
means the retrospective recognition of the movement of self-interpretation, through a serious of
failed attempts at satisfactory self-identification, as displaying an internal progressive
development. That is why Hegel‘s argument is not vulnerable to a charge of circularity which
would accrue to any putative unconditional emergence of self-consciousness through
recognition.
We have two figures. The first one is consciousness passively immersed in ―its‖ object:
its only possible mode of object-relation is consumption in response to organic needs. Hegel
does not hesitate to call this ―negation,‖ since it involves a rudimentary denial of dependency on
the object. The satisfaction which the consumption of the object generates is the occasion for
self-certainty, however elementary: need implies lack, and lack in turn gives rise to a sense of
self through the satisfaction which it makes possible. But since satisfaction so derived is
exhausted by and dependent on the negation of the object, it is necessarily transient: each
successful creation of a sense of self must be also a failure to the extent to which it destroys the
very ground of its emergence. This figure, then, is doomed to the repetition of a cycle of
satisfaction and disappointment.
The second figure is consciousness which aims at another desiring consciousness. Since
the object cannot sustain the sense of a self, Hegel claims that only another consciousness can be
the source of self-certainty.50 The transitions in the argument at this stage are notoriously hasty,
but I think the claim is motivated by the following: if the object cannot ground the subject,
because its negation entails its destruction, then only another subject can ground the subject,
50 ―Self-consciousness achieves its satisfaction only in another self-consciousness‖ (PS, 110).
48
since it alone can potentially negate itself without destroying itself.51 And recognition or
acknowledgement is the name for this process.
Therefore, self-consciousness presupposes mutual recognition. But since mutual
recognition is not a given and cannot be taken for granted, self-consciousness is an achievement.
Or rather, it can only be the result of a process that depends on division, negation, and its
sublation. We are to imagine what the primordial encounter between two self-consciousnesses
would be like. Since each is aiming at full independence through securing the other‘s
recognition, their interaction is inherently contradictory: the goal of autonomy (as full selfdetermination),
to the extent that it depends on the other‘s recognition, is untenable. Each aims at
independence, which, however, can only pass through and be guaranteed by the self-negation of
the other. Thus independence and dependence of self-consciousness appear to presuppose and
exclude one another.
Hegel discusses a number of strategies in response to this predicament. First, out of fear
of losing the ground for recognition, neither is willing to differentiate itself from its other, and so
recognition does not really achieve the result for the sake of which it is invoked. Second, one of
them breaks the stalemate, and claims universality by negating the particularity of its physical
existence: it risks death in a struggle to secure the other‘s recognition without acknowledging it
in turn. But this strategy too fails, for internal reasons: to the extent to which it succeeds in
wresting recognition from its other without recognizing it in return, self-consciousness loses the
very basis of recognition. Only as freely offered can acknowledgment do the work which mere
51 ―[Self-consciousness] can achieve satisfaction only when the object itself effects the negation within itself; and it
must carry out this negation of itself in itself…‖ (PS, 109).
49
consumption of an external object could not; but a coerced recognition soon devolves into mere
consumption.
Recognition, then, is bound up with the very concept of Spirit, since, as a collective,
socially self-determining, self-realizing subject, it can have no other ground than mutual
recognition. The appeal to full reciprocity is necessitated by the failure of any other possible
grounding relation, so much so that the very possibility of individual self-consciousness
presupposes Spirit. In the absence of individuals‘ recognition of themselves as mutually
recognizing, the only recourse available is coercion, and that does not lead to the kind of
satisfaction which alone can secure the kind of self that is worthy of the name. But the shift from
individuals enmeshed in coercive relations to the ―I-We‖ implies that full autonomy has to be
abandoned as a goal, since only through acknowledging its dependence on its other can selfconsciousness
become a subject proper. Therefore, the practical unity of the social world,
embodied in socially sanctioned practices and institutions, is the only context in which freedom
can become actual.
1.2 Phenomenology, or How Do I Simply Look On?
Phenomenology is central to what philosophical activity is for both Hegel and Husserl. It is not
obvious, however, how this is so and what the function of the concept is in their work. For Hegel
phenomenology is (only) a necessary moment in the unfolding of philosophy as Science,
whereas for Husserl transcendental phenomenology is philosophical activity at its most rigorous.
Ultimately the difference turns on the different conceptions of constitutive subjectivity at work in
their respective articulations of phenomenology. Nevertheless, there are two relatively
uncontroversial points at which the two phenomenologies converge.
50
First, Husserl and Hegel seem to agree that philosophy must become rigorous science and
that it can attain this status only if it is without presuppositions. What counts as
―presuppositionless‖ is a point of divergence between the two thinkers, given their different
conceptions of evidence and what the subject of experience is like; however, in some sense that
remains to be clarified, both Hegel and Husserl take themselves to be engaged in transforming
philosophy into rigorous science adequate to die Sache selbst.52 This is to be possible by
transcending the superficial first-order conceptualizations and naïve commitments of natural
consciousness in order to articulate a deeper level of experience in conceptual (or essential)
form.
For both Husserl and Hegel, phenomenology involves turning against natural
consciousness through modulated uses of Kant‘s transcendental interrogation of conditions of
possibility. The steps of these ―turns‖ entail a move from the level of experience in which the
primacy of the world (object) is implicitly assumed, to a level of experience governed by the
primacy of the subject. In Husserl the epoche—or rather, the series of reductions—clears the way
for transcendental subjectivity by suspending all belief in the world‘s existence and its value in
providing justifications. The ―natural attitude,‖ since it is common to both everyday
consciousness and the empirical sciences, must be put out of play, if philosophy is to become a
rigorous science. So, for instance, Husserl argues that: ―[e]very ordinary appeal to self-evidence,
insofar as it was supposed to cut off further regressive inquiry, is theoretically no better than an
52 ―Things themselves,‖ in Husserl, and ―subject matter itself,‖ in Hegel. (PS pp. 2-3, especially, ―die Erfahrung der
Sache selbst‖.)
51
appeal to an oracle through which a god reveals itself.… Every kind of self-evidence is the title
of a problem, with the sole exception of phenomenological self-evidence….‖53
And when Hegel proposes to present the unfolding of ―natural consciousness‖ in the
Phenomenology of Spirit, he means something more or less identical with Husserl‘s ―natural
attitude‖: common sense, everyday empirical consciousness, the natural and the social sciences,
and even not yet truly ―scientific‖ philosophy. He claims, for instance, that: ―[i]n any case
Science must liberate itself from this semblance [i.e. notions such as ‗subjective,‘ ‗objective,‘
‗cognition,‘ and even ‗science,‘ in their non-reflected immediacy], and it can do so only by
turning against it‖ (PS, 48). Moreover, his description of ―natural consciousness,‖ when
confronted with the standpoint of genuinely scientific philosophy, as undergoing a kind of
despair, and even death, suggests the degree to which the transformation of one into the other is
experienced as violence. 54
Both Hegel and Husserl view phenomenology—and this is the second point of
convergence—as a response to a historical ―crisis‖ of some sort.55 Thus phenomenology, through
which philosophy finally becomes rigorous science, is a response to an historical present as the
turning point of an unstable process which stands in need of resolution.56And they both insist on
the necessity of taking the transcendental turn for philosophy to become a rigorous science.
53 Edmund Husserl, Crisis of European Sciences and Transcendental Phenomenology (Nortwestern University
Press, 1970) pp. 188-189 (my italics)
54 Cf. his claims that, from the perspective of ordinary consciousness, the standpoint of Science appears as walking
on one‘s head; and that ―the familiar, just because it is familiar, is not really understood‖ (PS, 18).
55 This point is discussed by Merold Westphal in ―Hegel and Husserl: Transcendental Phenomenology and the
Revolution Yet Awaited,‖ in Critical and Dialectical Phenomenology, ed. Don Welton and Hugh Silverman (State
University of New York Press, 1987) pp. 103-135
56 PS, p. 4, 6-7; Husserl, in turn, writes the Crisis of European Sciences—but it is possible to see his concern with
psychologism from Logical Investigations on as already prefiguring this sense of crisis—in response to the threat of
relativism which he sees as a direct consequence of the type of objectivity presupposed by and consequent on the
successes of the natural scientific explanation of the world from Galileo on.
52
Hence transcendental reflection, which redirects thought from the known object to the knowing
subject, is a necessary moment of genuine philosophy.
In the Introduction to the Phenomenology of Spirit, Hegel states that his task is ―an
exposition of how knowledge makes its appearance‖ (PS, 49). But this presupposes that one take
as an object the phenomenal appearance of knowledge. And this in turn entails that it is possible
to distinguish between the observed and the observing consciousnesses. Hegel ―conceptualizes‖
this distinction as one between what appears to consciousness and what appears for us. The
latter, then, is the former‘s activity. It is in terms of this distinction that Hegel can claim to be
speaking from the standpoint of the ―phenomenological ‗we‘,‖ which does not participate in the
acts of the consciousness which ―we‖ observe: ―we‖ are the ―simply looking on‖ subject. (PS,
54-56)
Despite these general points of convergence between Husserlian and Hegelian
phenomenology, there are irreducible differences crucial for their theories of experience. The
differences turn on their concepts of evidence and how they position themselves with respect to
the reflective turn in philosophy, especially with respect to Kantian transcendentalism. What is
of most interest for the question ―what is the subject of experience, and how is it related to the
objects of experience?‖ is the way Hegel and Husserl arrive at two different versions of
transcendental subjectivity. In order to delineate these differences, I propose to follow a two-step
procedure: First, I pursue the theme of collective justification through mutual recognition in
relation to Kant and Hegel. Second, I adumbrate the salient features of Husserl‘s
phenomenological investigation of experience in relation to this theme.
53
1.21 We Think, Therefore I Am Not
Jonathan Lear makes the following claim in ―The Disappearing ‗We‘‖57:
If establishing the objective validity of our representations consists in showing that they are all
that there could be to being a representation, then one ought to expect that a certain type of
reflective consciousness will have an evanescent quality. I do not yet know how to describe this
quality without resorting to spatial metaphors. If our representations have objective validity, then
one will not be able to continue looking down upon them: that sort of reflective consciousness
must ultimately evaporate. And with it goes the detached perspective on ―our representations‖. It
is not obvious, however, that the ―We are so minded:‖ must therefore disappear. Our ability to
append the ―We are so minded:‖ represents a permanent [my italics] possibility of reflective
consciousness. Yet the ―We are so minded:‖ is, like the Kantian ―I think,‖ in an important sense
empty: we gain insight into who ―we‖ are by considering the representations to which we are
willing to append a ―We are so minded:‖ or, by considering which bits of the world we are
willing to consider as representations. The ―We are so minded:‖ must thus stand in an analogous
―master-slave‖ relation to our form of life as the Kantian synthetic unity of apperception stands to
the object of judgment.
On the interpretation Lear wants to motivate, just as Kant needs a ―deductive‖
justification for the categories because there is the skeptical worry that thought and experience
might not correspond, Wittgenstein offers a ―transcendental deduction‖ of some sort, on account
of the possible conflict between comprehension (of linguistic meaning) and practical ability (of
correct use of words). Moreover, just as Kant appeals to what is required for the possibility of
(unified) experience, Wittgenstein appeals to the conditions required for the possibility of
57 Jonathan Lear, ―The Disappearing ‗We‘,‖ Proceedings of the Aristotelian Society, supp. vol. 58 (1984) p. 241.
Lear‘s argument makes a convincing case for reading a transcendental import into Wittgenstein‘s appeal to ―forms
of life‖. In chapter two, part two, I propose, in relation to Foucault‘s ―historical a priori‖ and Wittgensteinian
―criteria,‖ a number of reasons why this should not be taken as entailing any grounding intention. My concern at this
stage is to draw on his interpretation of Kant and Wittgenstein in order to spell out further the kind of legitimation
implicated in Hegel‘s appeal to historical experience.
54
communication. Therefore, an analogy, albeit not a perfect one, exists between the ―I think‖
which must accompany all my representations, and the ―We are so minded:‖ which must
accompany all my comprehensions.
What I find interesting is the extent to which this tracks the line I traced between Kant
and Hegel and, if that is correct, what it entails for what justification becomes when transposed
into the field of communal and historically varying practices. When Lear argues that ―[t]he ‗We
are so minded:‘ must thus stand in an analogous ―master-slave‖ relation to our form of life as the
Kantian synthetic unity of apperception stands to the object of judgment,‖ I think it is not
unreasonable to see there the two aspects of Hegelian phenomenology I emphasized above: First,
the Kantian insight that it is impossible to stand outside the subject-object relation constitutive
for our perspective, in order then to anchor either one in the other, implies that both the
subjective and the objective perspectives are somehow necessary. Second, Kant‘s insistence that
this necessity must be nevertheless grounded in formal categories, which presupposes the
categorial distinction between concept and intuition, meets the Hegelian rejoinder according to
which the two are far less distinct than is required to sustain a formal justification.
Therefore, the move to ―We are so minded:‖ tracks the move to mutual recognition as the
only form of legitimation available to—and Hegel would add, needed for—establishing the
legitimacy of our fundamental claim-making practices. But what then becomes of the claim for
necessity and universality, the two features of the Kantian a priori, through which alone the
objective validity of possible experience could be established?
If my description of the stakes of Hegel‘s confrontation with Kant is accurate, then what
is entailed by the claim that one must ―see the true not only as substance, but also as subject‖ is
that, even though there is a fundamental identity between subject and object, this identity cannot
55
be referred to a pre-established harmony, but must be seen as an achievement: in other words, it
is constituted, and that constitution cannot be understood otherwise than through the contribution
―we‖ make to experience (as Erfahrung). Hence if intuitive experience is conceptually
determined, then the meaning, value, and legitimacy of concepts can only be secured through
their relation to other concepts—since there is nothing ―outside‖ against which they can be
measured. But, as the dialectic of dependence and independence evinces, the ―inside‖ cannot be
referred to a single consciousness taken in isolation: thought itself must be seen as embedded in
the practical unity of a social life comprising practices, customs, and institutions. This
displacement of epistemic and practical validity onto the context of sociality implies that
justification itself becomes historical, i.e. it becomes a question of how ―we‖ have come to take
certain kinds of life as justified.
I want to interpret what this move amounts to through an argument Lear formulates in
relation to Wittgenstein.
[A] person is minded in a certain way if he shares the perceptions of salience, routes of interest,
feelings of naturalness in following a rule that constitute being part of a form of life. Then…not
only must I be able to attach an ―I understand‖ to each of my representations, but it must be
possible for the ―We are so minded:‖ to accompany each of our representations. Thus our
representations stand between two distinct claims: the ―I understand‖ and the ―We are so
minded:‖58
On the analogy of Kant‘s ―deductive‖ strategy, what first appears as an analytic
principle—―we are so minded:‖ defines what it is for something to be a representation of ours—
is, from the transcendental perspective, a synthetic principle—―we are so minded:‖ expresses the
practical unity of social life which constitutes our being so minded. Moreover, as the
58 Ibid., p. 229.
56
representation to which I append ―I think‖ is not mentioned but used, the reflective capacity
thereby invoked does not entail the subject‘s stepping outside of all representations to make a
claim about one of them: it is itself the formation of another representation. Therefore, ―I
understand p,‖ is itself an enactment of a form of life, i.e. of being so minded: ―the ‗I understand‘
shows our participation in a form of life, it does not say anything about a representation which is
merely mentioned. Thus when someone says the word ‗cube‘ to me, my experience of
comprehension is not legitimated by any feeling or other inner experience….[It] is legitimated by
my being like-minded with other ‗cube‘-users.‖59
Therefore what I referred to as the practical unity of social life is the ultimate horizon of
legitimation for the particular ways in which a community continues its activities. If experience
(as Erfahrung) is to be possible, then ―we are so minded:‖ must be able to accompany all my
representations, which are thereby our representations. The consequent means that there must be
a regular relation of agreement between the subjective and the objective, inner and outer. Even
though there may be cases when this agreement may be undone, they must be precisely cases of
its being undone.
Hegel‘s claim for the a priori function of the Notion is that, since it provides the criterion
of what counts as (an object of) experience, it cannot itself be derived from experience.
Therefore, there is no empirical justification for or explanation of why ―we are so minded‖.
However, Lear argues that ―we cannot make any sense of the possibility of being ‗other
minded‘,‖60 and I want to claim that the priority of the Notional unity of Spirit in Hegel entails a
similar conclusion (and its dialectical twist). For Hegel the conditions for the possibility of
experience (so construed) are at the same time fully empirical: who ―we‖ are is revealed to us by
59 Ibid., p. 230.
60 Ibid, p. 232.
57
what we (are so minded to) take as who we are: ―There can (for us) be no getting a glimpse of
what it might be like ‗to be other-minded,‘ for as we try to pass beyond the bounds of our
mindedness we lapse into what (for us) must be nonsense: that is, we lapse into nonsense.‖61
However, the Hegelian recourse to the work of memory, the retrospective recollection of
what we have been and how we have come to be is precisely what allows the appropriation of
other-mindedness. To that extent justification through appropriation occurs by means of
expropriation. Therefore, from a Hegelian standpoint it cannot be said that all we are is the
particular ways in which we ―go on,‖ beyond the bounds of which there is only nonsense, in so
far as Hegel seeks to account for the emergence of sense from nonsense.62
Nevertheless the appearance of the ―we‖ is also its disappearance in Hegel too. If the
objective validity of Erfahrung can only be established by referring it to the practical unity of
social life, thereby showing its constitution by the progressive unfolding of Spirit, the ―for us‖
cancels out. In reflectively finding out who we have been and how we have come to be, we
thereby come to see that being our object is all there could be to being an object. Lear calls a
legitimation which does not provide a foundation for what is legitimated ―groundless
legitimation‖.63His application of this type of legitimation invokes transcendental arguments
broadly considered: one establishes the legitimacy of X by showing what it is to be X.64I think
61 Ibid.
62 It is in this sense that, when Lear writes ―Wittgenstein…is able to awaken us to the possibility that our form of life
is partially constituted by our being so minded without making contrasts with ‗other perspectives‘,‖ I would like to
forward two claims: 1) Since he is not talking about Hegel in this essay, if what he says about Wittgenstein is
correct, then it both comes close to and diverges from the Hegelian shift to Spirit, 2) In the next chapter, I want to
claim that there is some reason for why his reading of Wittgenstein may not be correct, since the function of ―tribe‖
in the Philosophical Investigations is to establish such (fictitious) contrasts. I also want to add that there are places
where Lear writes in terms that come close to the Hegelian appropriation of other-mindedness (or alienation) minus
the appeal to history. See p. 233, n. 37.
63 Ibid, p. 233 I discussed the juridical sense of ―deduction‖ in Kant above; what I say here in relation to Lear and
Hegel presupposes that sense.
64 Ibid, p. 222-223: Such a transcendental argument ―will answer the question ‗How is X possible?‘ when that
question is asked with a straight face rather than a sceptical sneer.‖
58
this feature may be captured by providing a schematic formulation: ―X is possible, only if…,‖
where the consequent specifies the relevant conditions. Thus the argument seeks to show not
what must be the case for some X recognizable by (for) us to be possible, but rather what all Xs
must be, in order to count as X.
So, for instance, it is not that anything we could recognize as ―mindedness‖ has to be like
our ―mindedness,‖ but that there is no concept of being other-minded: ―they‖, if they are to count
as minded in any way at all, must be part of ―us‖. For what would it be to consider an otherminded
group or individual? It establishes a relationship between ourselves and the other—the
subject and the object—such that we must already be treating the other as one of us for it to stand
out from among all the features of the social and natural environment as something about which
we can ask ―Is it like-minded or other-minded?‖
However, if consciousness becomes Spirit in Hegel‘s Phenomenology of Spirit, and if the
process of experience is the unfolding of that movement, then the proper understanding of this
necessity, at once subjective and objective, must be placed in history. Therefore, self-relation in
relation to an other, some aspects of which I outlined in section one, is both social and historical:
social, because the criterial role which the Notion plays in the constitution of experience cannot
be abstracted from its function as social sanction; historical, because the forms of what comes to
play that role and have the relevant (authoritative) force changes historically in a determinate
way.
I want to take one moment from the Phenomenology of Spirit in order to illustrate what
this claim involves and adumbrate a theme which will be important in my interpretation of
59
Foucault. The immediately relevant section is ―Individuality Which Takes Itself To Be Real In
and For Itself‖.65 Hegel writes:
It is…equally a deception of oneself and of others if it is pretended that what one is concerned
with is the ―matter in hand” (die Sache selbst) alone. A subject that opens up a subject matter
soon learns that others hurry along like flies to freshly poured-out milk, and want to busy
themselves with it; and they learn about that individual that he, too, is concerned with the subject
matter, not as an object, but as his own affair. On the other hand, if what is supposed to be
essential is merely the doing of it,…the expression of this particular individuality, then equally it
is learned by all parties that they all regard themselves as affected and as invited to participate,
and instead of a mere ―doing‖…peculiar to the individual who opened up the subject-matter,
something has been opened up that is for others as well, or is a subject-matter on its own account.
(PS, 251)
The problem in the face of which ―consciousness‖ will suffer at this stage of its
Erfahrung is ―the antithesis of doing and being‖ (PS, 244), as consciousness provides different
interpretations of its individuality, which the dialectic of independence and dependence showed
to be required for a satisfactory relation to self. For each new self-interpretation of what this
individuality comprises, the phenomenological account shows whether it is consistent with its
own preconceptions: that is, an account of the conditions under which an individual could
identify its own content and what ―belonging‖ could mean here.66
Hegel‘s starting point is a conception of individuality ―in which…its shape has the
significance solely of putting on the shape of individuality; it is the daylight in which
consciousness wants to display itself. Action alters nothing and opposes nothing. It is the pure
form of a transition from a state of not being seen to one of being seen, and the content [so
displayed] is nothing else but what this action is already in itself‖ (PS, 237). Individual
65 The moment on which I focus occurs in subdivision (a) The spiritual animal kingdom and deceit, or „the matter in
hand‟ itself (die Sache selbst). PS, pp. 236-252.
66 ―Let us see whether this Notion is confirmed by experience, and whether its reality corresponds to it‖ (PS, 242).
60
consciousness so conceived takes itself to be real in and for itself, and dispenses with ―all
opposition and every condition affecting its action; it starts afresh from itself, and is occupied not
with an other, but with itself‖ (Ibid). The result is a picture of social interaction, for the
individual, where he can simply be who he is without having to pass through the mediation of a
potentially alienating other (law, community, etc.). Hegel‘s phrase ―the spiritual animal
kingdom‖ signifies this social space where individuals take themselves to be simply what they
are, just as animals (allegedly) are in the animal kingdom with already given specific differences.
Not surprisingly, the spiritual animal is an unsatisfied animal: such (immediate)
reconciliation with one‘s own nature is barred for self-conscious subjects. Since ―[c]onsciousness
provides its own criterion from within itself, so that the investigation becomes a comparison of
consciousness with itself‖ (PS, 53), the dissatisfaction results from the very discrepancy between
what consciousness takes to be its doing and its being. The significance and consequence of
consciousness‘s actions cannot be secured by anchoring them immediately in its intentions
because ―others hurry along like flies to freshly poured-out milk, and want to busy themselves with it‖
(PS, 251): it is the consequences that the action generates, contingently and independently of
conscious intention, which determine the meaning of the action. Thus consciousness, once it
acts, is taken up by the meanings it cannot control, and appeal to what one is immediately cannot
secure the correspondence between what it does and what the action is.
That consciousness is caught up in a supra-individual context of meanings which it alone
cannot determine entails that its action ―vanishes‖ (PS, 244-245). But consciousness comes to
experience this vanishing of its action as itself vanishing: ―If…we consider the content of this
experience in its completeness, it is seen to be the vanishing work. What is preserved is not the
vanishing: the vanishing is itself actual and is bound up with the work and vanishes with it; the
61
negative itself perishes along with the positive whose negative it is‖ (PS, 245). ―This vanishing
of the vanishing‖ (Ibid), then, is the last ditch effort by consciousness to secure the significance
of its individuality by abstracting the meaning of its act both from itself and from its contingent
consequences. What it really means, ―the matter in hand,‖ or again ―the real thing itself,‖ is
untouched by what is made of it contingently.
But, Hegel continues, ―[t]he truth about this integrity…is that it is not as honest as it
seems‖ (PS, 248-249): since consciousness is ineluctably involved in a struggle for recognition,
what it takes to be the real thing, uncontaminated by its contingent effects, and what the others
take it to be cannot be evaluated independently. Consciousness depends on the others‘
interpretations (of its action) in order to attach a determinate meaning to its action; however, the
others‘ interpretations can only count as meaningful for consciousness through its own
interpretation.
The stalemate of individuality which is characterized by this dialectical interrelation is
inherently ambiguous: ―…in this alternation consciousness keeps, in its reflection, one moment
for itself and as essential, while another is only externally present in it, or is for others…thus
enters a play of individualities with one another in which each and all find themselves both
deceiving and deceived‖ (PS, 250). Therefore, individuality alone cannot secure what it means
for its actions to be its own, to belong to it. The self-relation presupposed by any ascription of
what is proper, when conceived as occurring independently of other (subjects), leads to an
impasse, since they too must make something of consciousness‘s action in order to secure their
own sense of individuality. The claim to individuality and the ascription of what is proper is
never consciousness‘s own property.
62
Therefore the characterization of my action as my own, in order to secure my identity as
an individual, cannot itself be guaranteed by individuals. There must be some supra-individual
context, in relating to which I relate to myself, which secures that what I take to be my own is in
fact my own. If I could never satisfy myself that what I take to be my interest, or reason, or
identity, is in fact my own—and not the impersonal product of an external agency—then the selfrelation
presupposed by subjectivity could not hold. And Hegel‘s point is that, I cannot do that
alone. Hence: ―Consciousness learns that no one of these moments is subject, but rather gets
dissolved in the universal „matter in hand‘ (die Sache selbst); the moments of the individuality
which this unthinking consciousness regarded as subject…coalesce into…this particular
individuality, [which] is no less immediately universal‖ (PS, 252). Individuality, which is an
essential moment of the development of subjectivity, presupposes its realization as Spirit, and
without which it disappears.
The transposition of justification to history is required by such ―impasses‖ as the
dependence and independence of consciousness and the spiritual-animal individual. What
underpins the conflicts so described, and propels their movement, is the requirement that
subjectivity be a self-relation that is also a relation to an other: I must confirm my own
independence without denying my dependence (on either biological or social life). But all the
shapes which this requirement assumes contain a discrepancy between my self-comprehension
and my actuality, until one reaches the standpoint of Absolute Knowledge. ―History [as] a
conscious, self-mediating process‖ (PS, 492), is not the absolute other of subjectivity:
―…Absolute Knowing, or Spirit that knows itself as Spirit, has for its path the recollection of the
Spirits as they are in themselves….Their preservation, regarded from the side of their free
existence appearing in the form of contingency, is History; but regarded from the side of
63
their…comprehended organization, it is the Science of Knowing in the sphere of appearance
(phenomenology)‖ (PS, 493).
Since the concepts through which we comprehend the past result from that very past,
retrospective recollection of history is a form of self-discovery. And since each shape of
consciousness articulates a partial truth—revealed through the very gap between its partiality and
its claim to be absolute—the sublation of all these shapes, preserving their moments of truth, will
be the absolute standpoint; no longer outside of time and history, but precisely their very
internalization and remembrance. Moreover, this recollection as justification, if successful, will
have imported no assumptions from the outside, because every shape of consciousness is both
the contradiction between its self-comprehension and its actuality, and the movement of its
sublation. Hegel‘s claim, therefore, is that justification can only perform its work from the inside
and that there is no outside (to Spirit).
1.22 Seeing Things (Disappear)
If the concept that stands out most in Hegel‘s phenomenology is that of satisfaction, one
interpretation of which I traced above, then the central concept for Husserl is that of
fulfillment.67Husserl‘s conception of evidence, from the Logical Investigations through his
transcendental turn to his investigations concerning the historicity of constitution, is modeled on
the presence of the intended object in intuition. In the Logical Investigations, for instance, even
when he is considering ―signitive acts‖—―merely empty acts,‖ as when I use words without
really thinking about what I am saying—are still referred to an ideal of ultimate fulfillment:
67 I should say ―the central concept which will be important for the purposes of my comparison,‖ since any claim for
such centrality is bound to be tentative.
64
―…many elements of fullness count for us—quite apart from anything genetic, for we know full
well that these…have an associative origin—as final presentations of the corresponding objective
elements. They offer themselves as identical with these last, not as their mere representatives:
they are the thing itself in an absolute sense…. express universal, essential structures, that is,
strictly necessary structures of every conceivable stream of consciousness….The discussion of
possible relationships of fulfillment therefore points to a goal in which increase of fulfillment
terminates, in which the complete and entire intention has reached its fulfillment.‖68
Moreover, Husserl‘s conception of truth is bound up with this conception of evidence:
―the full agreement of what is meant with what is given as such. This agreement we experience
in self-evidence, in so far as self-evidence means the actual carrying out of an adequate
identification.‖69 Therefore, paradigmatic for Husserl‘s conception of evidence and truth is the
fulfillment of an intention through an act of identification, which is modeled on the self-evidence
with which something presents itself to consciousness in perception.70
However, Husserl‘s appeal to intuitive-fulfillment should not be confused with senseperception,
for it is a question of finding the essences of phenomena, and therefore intuition has a
broader sense than that which is tied to sensuous experience: ―…extended concept of Perception
permits…a wider interpretation….[E]ven universal states of affairs can be said to be perceived
(―seen,‖ beheld with evidence).‖71Hence the distinction between sensuous and categorial
intuition provides Husserl with the epistemological warrant to describe essences given in
intuition.
If the goal of enquiry is the ―definition‖ of essences of various types of acts such as
perceiving, imagining, and remembering, by investigating the essential structures of
consciousness wherein intentions find fulfillment, then the starting point for such enquiry
68 Edmund Husserl, Logical Investigations, (New York: Humanities Press, 1970) p. 761, 762.
69 Ibid, p. 765
70 ―Any evidence is a grasping of something itself that is, or is thus, a grasping in the mode of ‗it itself‘, with full
certainty of its being, a certainty that accordingly excludes every doubt.‖ (Cartesian Meditations, p. 15)
71 Ibid, p. 786.
65
becomes a problem: we do not see ―essences‖ in ordinary perception. Husserl insists that
philosophy worthy of the name be presuppositionless, and therefore the entire trajectory of his
development of phenomenology is the ever renewed effort to reflect on a point of departure that
would commit the philosopher to no presuppositions. In line with the theme I delineated in the
previous sections, I want to focus on one moment of this movement, namely, Husserl‘s recourse
to transcendental subjectivity in Ideas I.72
Husserl writes: ―We begin our considerations as human beings who are living naturally,
objectivating, judging, feeling, willing ‗in the natural attitude.‘ What that signifies…can best be
[clarified] in the first person.‖73The natural attitude is characterized by existential and validity
commitments: Not only do I believe in the existence of the world independently of
consciousness, but I also make many validity claims. If, however, we want to bring to light what
the different types of claims are as claims, thematize them as objects of investigation, without
any presuppositions that would prejudge the outcome, then we must suspend the existential and
validity commitments; we must perform the phenomenological epoche: ―We do not give up the
positing we effected…[but] we, so to speak, „put it out of action,‟ we „exclude it,‟ we
„parenthesize it‘.‖74
What remains ―if the whole world, including ourselves with all our cogitare, is excluded‖
is a ―new region of being never before delimited in its own peculiarity‖:75 the sphere of
consciousness and what is presented immanently in it. Husserl claims that we are thereby
capable of formulating a new eidetic, universal insight into the essence of any consciousness
whatever. Moreover the epistemological goal of Husserl‘s investigations motivates him to
72 Edmund Husserl, Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy.
73 Ibid, §27, p. 51 [48]
74 Ibid., § 31-32, pp. 58-59 [54]
75 Ibid., § 33, p. 63 [57-58]
66
designate the reductions, whereby this region of pure consciousness is revealed, as
transcendental. And the mental processes which have the character of acts reveal that,
universally, ―every actional cogito [is] consciousness of something.‖ The mark of the mental,
then, in its active sense, is intentionality.76
Therefore phenomenological analysis consists of describing the modes of givenness of
objects revealed in the region of pure consciousness, which becomes visible as the result of a
series of reductions.77The model of evidence as intuitive fulfillment means that one must retrace
intentions back to their founding intuitions. And there things are given to us ―in person‖.
Moreover, since the region of pure consciousness is without any presuppositions, what the result
of investigation reveals can be established with eidetic necessity and universality. However,
Husserl also argues that ―[a mental process] is never perceived completely, that it cannot be
adequately seized upon in its full unity.‖78This is because it is ―in flux,‖ and thus even though
reflection can ―swim along after it starting from the Now-point,‖79 the stretches it has already
covered are no longer available to perception.
Consciousness is given over essentially to a temporal flow which is always already
anterior to reflective acts. Therefore it cannot be made an object of phenomenological ―seeing,‖
since it is the pre-reflective background against which any phenomenon can stand out through
reflection. Therefore what motivates Husserl‘s ever renewed efforts to account for the condition
of possibility of phenomenology itself is bound up with this process in which consciousness
always plays catch-up with itself: that is, the demand to grasp the essential structures of
76 Ibid., §35, 36, pp. 73-74 [64-65]
77 Husserl‘s distinction between phenomenological and transcendental reductions, which I merely note, is complex;
but it is also not relevant for my purposes.
78 Ibid., §44, p. 97 [82]
79 Ibid.
67
consciousness and its temporal genesis without objectifying and thereby falsifying it.80If the flow
of consciousness is anterior to reflection, then reflective categories of subject and object, or
―constituting‖ and ―constituted,‖ are no longer appropriate to it.
For Husserl, however, this kind of incompleteness pertains to the essence of the
perception of a mental process, and as such it is referred back to ―the form of retention‖ or ―the
form of retrospective recollection‖.81This ―essential incompleteness‖ provides the first crack in
the immanence of the constitution of objectivity by transcendental subjectivity. In his later work
Husserl becomes more attentive to the historical dimension of the constitution of objectivity. The
pure description of modes of givenness of objects cannot account for the different layers of
objectivity and their specific temporality. Since access to some fundamental levels cannot be
gained through direct experience, and since, for instance, the essence of a historical event cannot
be intuited in principle, Husserl is forced to provide a genetic account of how traditions of
meaning are constituted over time and get sedimented in different cultural formations. But
because, from the Logical Investigations on, Husserl is convinced that only relativism can result
from psychologism or historicism, his attention to the essential historicity of objectivity takes a
particular direction. I want to look into just one instance.
In The Origin of Geometry, Husserl claims that ―Science, and in particular geometry,
[with the meaning: necessarily embodied in a communal tradition] must have had a historical
beginning; this meaning itself must have an origin in an accomplishment: first as a project and
80 It is perhaps in this sense that Merleau-Ponty‘s claim in the Preface to the Phenomenology of Perception (p. xv)
becomes intelligible: what the reduction reveals is the impossibility of a complete reduction. I discuss Merleau-
Ponty in relation to Foucault in Chapter 4, part 1.
81 Ideas I, p. 97 [82-83]
68
then in successful execution.‖82He proposes to grasp this beginning through a ―regressive
questioning‖ (Rückfrage), where ―geometry‖ is a stand-in for ―all disciplines that deal with
shapes existing mathematically in pure space-time.‖83Therefore the enquiry concerns the
constitution of ideal objectivity which must nonetheless be embodied in and transmitted through
concrete historical traditions. But the conception of evidence which is to support his ―questioning
back‖ remains essentially the same:
The original being-itself-there, in the immediacy [Aktualität] of its first production, i.e. in original
―self-evidence,‖ results in no persisting acquisition…that could have objective existence. Vivid
self-evidence passes—though in such a way that the activity immediately turns into the passivity
of the flowingly fading consciousness of what-has-just-now-been. Finally this ―retention‖
disappears, but the ―disappeared‖ passing and being past has not become nothing for the subject
in question: it can be reawakened. [To this passivity reawakened] there belongs the possible
activity of a recollection in which the past experiencing [Erleben] is lived through….84
Since any tradition can only be constituted over time, the phenomenologist must question
backwards to reveal the historical sedimentations in experience. The phenomenological attempt
to ground the genesis of reason and objectivity in history, however, is motivated from the start
by the goal of ―defining‖ the very historicity of history, i.e. its essential structures. Husserl‘s
argument, therefore, starts by assuming the factual existence of a first geometer, and proceeds to
specify the conditions under which alone what is given to this psychological individual can
acquire intersubjective and objective ideality and validity. The first step involves its expression
in linguistic ideality of meaning; and the second step requires the detachment from any actual
subjectivity by its embodiment in writing. Only then will ideal geometrical objects become fully
82 Edmund Husserl, The Origin of Geometry; my references are to the translation of this text by David Carr,
provided in Jacques Derrida, Edmund Husserl‟s Origin of Geometry: an Introduction (Lincoln: University of
Nebraska Press, 1989) p. 159.
83 Ibid., pp. 171-172, 158 respectively.
84 Ibid., 163.
69
virtual, thereby securing their persistence through time and space. Even though this makes
ideality vulnerable to its corruption by its material embodiment, the permanent possibility of
reawakening the original intentions through self-evident intuitions remains.
It is not that Husserl deduces a priori the empirical specificity of any and every cultural
and historical practice from an essence. But since the aim is to make manifest the essential
structures of any possible historical society, etc., in this account empirical facts can only have the
status of examples that start off the process of imaginary variation. Intuition of essences de jure
precedes every empirical historical investigation. The a priori sense of the historian‘s activity,
starting with the very idea of history, cannot be derived from particular historical formations,
since only from the perspective of the former are the latter recognized as such. If reflection on
history is necessary, it is in order to secure the possibility of re-apprehending the primary
evidences originally given to the founder of geometry (or of any scientific or cultural formation).
Therefore, the phenomenologist‘s concession that there is a historical dimension of objectivity is
to free phenomenology from the corruption of history. As Husserl says, ―in the
phenomenological sphere, there are no contingencies, no mere matter-of-fact connections
(Faktizitäten); all is essentially and definitely motivated.‖85
The distinction, then, between historicity and history (Historie) wards off the threat of
relativism which could accrue to ideal objectivity through its necessary incarnation in and
transmission across history:
Anything that is shown to be a historical fact, either in the present through experience or by a
historian as a fact in the past, necessarily has this inner structure of meaning….All [merely]
factual history remains incomprehensible because, always merely drawing its conclusions
85 Husserl, Ideas p. 356.
70
naively…from 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.86
Husserl therefore insists on the necessity of a ―concrete, historical a priori,‖ which will provide
the ultimate ground on the basis of which any particular historical investigation becomes
possible. His insistence derives from the conviction that ―the problem of genuine historical
explanation comes together…with ‗epistemological‘ grounding or clarification.‖87Accordingly,
the necessary detour through history remains precisely that: a detour which retains the primacy
of his original model: ―…to discover, through recourse to what is essential to history, the
historical original meaning which necessarily was able to give and did give to the whole
becoming of geometry its persisting truth-meaning.‖88Hence the grounding provided by the
―universal historical a priori,‖ or ―internal history,‖ and bound up necessarily with the ―universal
teleology of reason,‖89 leads to a phenomenological justification of history far from a historical
justification of phenomenology.
Even when Husserl refers to the ―life-world as the forgotten meaning-fundament of
natural science,‖90 his reinscription of the concept-intuition duality as that between intention and
its fulfillment, and the ultimate search for a genuinely presuppositionless starting point, limits the
very limitation which the life-world could provide for ultimate grounding through the meaningbestowing
acts of consciousness. The mathematization of the life-world, and its substitution ―for
the only real world, the one that is actually given through perception, that is ever experienced
and experienceable,‖ is now faulted as itself being no longer faithful to the ―sources of truly
86 Husserl, Origin of Geometry, p. 174.
87 Ibid.. p. 175.
88 Ibid., p. 179.
89 Ibid., p. 180.
90 Edmund Husserl, Crisis of European Sciences and Transcendental Phenomenology, (Northwestern University
Press, 1970) § 9, (H).
71
immediate intuition and originally intuitive thinking.‖91The investigation of the life-world, on the
other hand, must be grasped in its pre-scientific and pre-theoretical everdayness: ―…as history
teaches us, there was not always…a civilization that lived habitually with long-established
scientific interests.‖92But since the life-world was always there, its essence must be grasped and
the pre-scientific interests that ―govern it‖ described.
Thus: ―[T]he life-world is a realm of original self-evidences. That which is self-evidently
given is, in perception, experienced as ‗the thing itself,‘ in immediate presence, or, in memory,
remembered as the thing itself; and every other manner of intuition is a presentification of the
thing itself.‖93It is therefore possible in principle to describe the ―general structure‖ of the lifeworld,
despite its relative features: ―This general structure, to which everything that exists
relatively is bound, is not itself relative.‖94Husserl distinguishes between ―the objective-logical a
priori‖ and ―the a priori of the life-world‖. The former is ―put out of play‖ through the
transcendental reduction of scientific idealizations, but the latter contains everything we find in
the former: the world, prior to its theoretical explanation by the sciences, is already spatiotemporal,
causal, etc.; and since the latter is free from the idealizations which can only be built
up through a ―reference back to the corresponding a priori of the life-world,‖ ―[t]his reference
back is one of a founding of validity.‖95Therefore, because ―among the objects of the life-world
we also find human beings,‖ the phenomenological investigation of this ―universe of the
91 Ibid..
92 Ibid., §33, my italics. Here I merely note in passing this appeal to some putative lesson of history. It should be
clear, however, that, according to the conception of phenomenology I describe, history cannot really teach anything;
or the history which Husserl appears to invoke here could only be history in a very attenuated sense. I take up this
issue in the next chapter in comparing Foucault‘s historical a priori with that of Husserl.
93 Ibid., § 34, (D).
94 Ibid., § 36.
95 Ibid.
72
subjective,‖ in opposition to all objective sciences, is to be ―science of the universal how of the
pre-givenness of the world, i.e. of what makes it a universal ground for any sort of objectivity.‖96
Husserl‘s consistent and rigorous thinking through the very presuppositions of
phenomenology leads him to push across the limits of the region of transcendental subjectivity,
the immanence of which was to be the achievement of transcendental phenomenology. However,
his conceptions of evidence, meaning and intention retain a certain hold on the very direction
which his trajectory follows from pure consciousness to the life-world as the field of ―ultimate
grounds‖ [Gründe].97Thus, when he claims that: ―[a]ll truth and all judicative evidence…are
related back to the primitive basis, experience; and, because experience itself functions in and
not beside the original judgments, logic needs a theory of experience—in order to be able to give
scientific information about the legitimating bases, and the legitimate limits, of its a priori,
consequently about its own legitimate sense‖98; the starting and ending points of the
phenomenological conception of experience remains beholden to the primacy of a constitutive
consciousness and its meaning-bestowing acts as the ultimate legitimating acts.
In the light of my discussion, however limited, of the ―uses of experience‖ in Kant,
Hegel, and Husserl, I focused on one strand that I take to be central for philosophical attempts to
say something about experience which would go beyond common sense without, however,
messing about with rats, or questionnaires, to use Rorty‘s phrase. That strand is the theme of
activity and passivity and the distribution of agency between subjectivity and objectivity. From
the theater of ideas through transcendental apperception to Hegel‘s ―I who is also We,‖ one
finds the renewed efforts to account for the reflexivity of experience and follow through its
96 Ibid., § 38.
97 Ibid.
98 Edmund Husserl, Formal and Transcendental Logic, (The Hague: Martinus Nifhoff, 1969) pp. 211-212.
73
consequences. In Husserlian phenomenology, finally, there is the attempt to refer all such
reflective distinctions to that region which, anterior to all reflection, sustains all theoretical and
practical commitments: ―It is that as yet dumb experience…which we are concerned to lead to
the pure expression of its own meaning.‖99Hence the ever renewed trials of beginning which
Husserl formulates. Rorty‘s response, as I sketched briefly at the beginning of this chapter, to
this entanglement of ―experience‖ with the misguided projects of epistemology, is to drop the
concept or refer it to currently accepted scientific practice.
After all, if the subject, for all we care philosophically, is nothing more than a black box
emitting sentences, then it is plausible to ask of it only those questions which will be of the same
order as the ones we might ask of telephones.100Therefore, Rorty‘s strategy of asserting the
superfluity of the subject (and of experience) by denying the very coherence of the problem(s)
for which it was intended as an answer appears to be effective. The strategy I will attribute to
Foucault in the following chapters is not unlike the one I hint at here. However, I want to
indicate briefly what differences one might expect.
Hume had claimed that when he turned back his mind‘s eye to observe the contents of his
own mind, all he could see were various acts of perceiving, remembering, being in this or that
state; but the self, which allegedly remained the enduring substance throughout the flow of
representations, was nowhere to be seen. However, Kant‘s awakening from his dogmatic
slumber did not neglect to raise a peculiarly poignant question to Hume: when I thus turn back
my mind‘s eye, it may be that all I see is the unceasing flow of this or that state, this or that
99 Husserl, Cartesian Meditations: An Introduction to Phenomenology (The Hague: Martinus Nijhoff, 1977) p. 38-
39 (The Cairns translation is: ―…pure—and, so to speak, still dumb—psychological experience, which now must be
made to utter its own sense with no adulteration.‖ I follow the translation of Merleau-Ponty‘s citation in PP, p. xv
Alternatively, ―die stumme Erfahrung‖ can be translated as ―mute experience‖.)
100 One option Rorty does not consider is that even if the subject is no more than a black box, it could also be called,
by the same token, no less than a black box. As I try to show in the case of Foucault, the dimension of self-relation
seems to be more resilient than its wishful abandonment supposes.
74
representation, with no underlying self in sight; but just what is this very I, the ―eyes‖ of which I
turn back, with which I see myself seeing this or that? Therefore the critical project, and the very
possibility of critique, is bound up with the Kantian insight that, even though the empirical self
may be no different from all the other objects which appear to me in experience, the
transcendental subject cannot be so easily dissolved into the stream of consciousness; since it
seems to be presupposed by the act of dissolving anything at all.
In what follows I articulate many criticisms which may be leveled at Kant from a
Foucauldian perspective. Bu at this juncture, a question not unlike the one I just attributed to
Kant may be posed to Rorty: Was it just that Kant, or Hegel, or Husserl did not know any better
than to remain caught up in the Cartesian net of subjectivity? If, rather, the recourse to
subjectivity so construed can only be made intelligible through the historically specific social
practices which made it necessary in some sense; and if it is precisely the rigor with which Kant,
or Hegel, or Husserl pursued the conditions which had made possible the moves they could make
in a determinate field of possibilities; just what does it take for it to become plausible, now, for
us, to even think of the subject as a black box, and for its reality to be exhausted by the
controlled experimentation on rats in laboratories or the statistical analysis of questionnaires?
The answer is, in part, that in modern societies the subject has indeed become a black box. But
that becoming cannot be grasped neutrally as a mere result of our finally knowing better; nor
celebrated unequivocally as the improvement in conditions of liberty.
75
76
A discourse attempting to be both empirical and
critical cannot but be both positivist and
eschatological; man appears within it as a truth
both reduced and promised. Pre-critical naïveté
holds undivided rule. (OT, 320)
2. Giving Up the Transcendental: Critical and Empirical Discourse?
There is a temptation to regard Foucault‘s work as a cluster of empirical insights in search of
their philosopher to organize and justify their historical content. This temptation should be
resisted. However, the form that resistance takes and the intelligibility it assumes are far from
obvious and entail a number of risks. Specifically, it is not possible to articulate with full
conceptual determination the standpoint which sustains that resistance, because it is, in some
sense that will become clear only gradually throughout the present study, a disavowal of the very
points which have a categorial standing in modernity. More specifically still, because it entails a
complication of the separation into spheres of validity—cognitive truth, normative right, and
aesthetic beauty—which are then distributed between the transcendental and the empirical.
So, for instance, Béatrice Han argues that Foucault succumbs to the very confusion
between the transcendental and the empirical, in different modalities, he had charged post-
Kantian philosophies to have committed.101On this account, the thread binding together the
disparate historical studies is Foucault‘s effort to offer a nonanthropological version of the
101 Béatrice Han, Foucault‟s Critical Project: Between the Transcendental and the Historical, trans. Edward Pile
(Stanford: Stanford University Press, 2002), hereafter BTH. Han‘s study is thoroughly researched and forcefully
argued. Although I do not agree with her conclusions, and even less with the unargued assumption that Foucault
presupposes a philosophical foundation that he cannot provide, the argument of this chapter owes much to her
relentless questioning and objections. Another source of motivation for framing some of the elements of my
argument is Jürgen Habermas‘s The Philosophical Discourse of Modernity: Twelve Lectures, trans. Frederick
Lawrence (Cambridge, Massachusetts: MIT Press, 1987), hereafter PDM, especially lectures 9 and 10.
77
transcendental theme he had first analyzed in the introduction he wrote for his complementary
thesis, a translation of Kant‘s Anthropology from a Pragmatic Viewpoint, the argument of which
supplies the contours of the analytic of finitude articulated in The Order of Things. The
multiplicity of concepts Foucault marshals thereafter—episteme, historical a priori, discursive
formation, archive, archaeology, games of truth, genealogy, regimes of truth, techniques of the
self, problematization—are, from this perspective, so many attempts to provide coherence to
what remains one and the same passion: historicizing the transcendental.
Against this position, and at times against the grain of some of Foucault‘s own
formulations, a more faithful characterization of Foucault‘s trajectory is not so much the
conversion into the domain of contingency and particularity what would otherwise be necessary
and universal conditions—historicizing the transcendental—as if all one had to do were the
addition of historical variability to transcendental frameworks or schemes; as it is giving up the
transcendental in the forms it has taken since Kant, and pressing the consequences of this
abandonment for a reflection on history, and by extension subjectivity. The following claim
Foucault makes crystallizes both the necessity and the difficulty staked by this thesis: ―The
research that I am undertaking here therefore involves a project that is deliberately both
historical and critical, in that it is concerned … with determining the conditions of possibility of
medical experience in modern times…. Here as elsewhere, it is a study that sets out to uncover,
from within the density of discourse, the conditions of its history‖ (BC, xix; modified, my italics).
Two tentative conclusions follow: first, the demonstration of the position I am pushing,
namely, that Foucault‘s philosophical/historical studies are motivated by and carry out the
renunciation of the transcendental, demands that a sense be given to the ‗conditions of
possibility‘ talk that permeates his works and structures his thought; and second, the
78
understanding so provided should account for both the historical and the critical dimensions
without presupposing the confusion of the transcendental and the empirical Foucault excoriates
in The Order of Things. That such an account is necessary should be clear from the epigraph of
this chapter. The question is: Does Foucault need a version of the transcendental to ground what
would then become his empirical/historical studies, or does he successfully manage to recast the
transcendental theme through his non-subjective account of experience?
Experience, both the concept Foucault is proposing and the manner in which he is
conceptually articulating it, operates in a problematic relation to the separation into spheres of
validity (objective knowledge, practical/moral rightness, and aesthetic reflection) and the
discursive/subsumptive reasoning this presupposes. It requires a form of reasoning in which
forms are not indifferent to content, and meaning is not abstracted from social practices. This
experience is governed by forms that are local, non-topic-neutral, and context-bound. The
intelligibility of experience so understood is neither transcendentally constituted nor empirically
given, but intimately tied up with the internal connections between its elements.
In short, it involves a binding of conceptual understanding and perceptual experience (if
we take ‗experience‘ in the Kantian sense). It forecloses conceptual closure by localizing
discourse, and it makes possible a critical engagement with procedural, formal rationality lodged
from the space of a local, context-dependent and particular rationality. To demonstrate all this
would be a tall order. What is even more difficult, however, is that, since this experience and its
articulation cannot resist the demand for universality indefinitely, the very attempt to flesh it out
discursively is fraught. Hence the charges to which Foucault is exposed: from performative selfcontradiction
to presentism to crypto-normativity to oscillations between the transcendental and
the empirical.
79
I think there is a good response to each of these objections, and the concept of experience
provides the matrix from which to provide them. Since what is common to most, if not all, of
these criticisms is the claim that Foucault‘s project, and by extension the concept of experience
that is the focus of this study, lacks a theoretically constituted foundation, and given the just
noted aporia consequent upon discursively articulating what is not susceptible to full conceptual
articulation, my argumentative strategy will be twofold.
First, I will attempt to make the demand for theoretical grounding less attractive by
sketching an account of why it appears inescapable and why its claim to be exhaustive is
deceptive. This part of the argument is the negative moment of what assumes positive shape in
Chapter 4 with a more detailed discussion of the concept of recognition and the hermeneutics of
the subject. Second, I will chart the three dimensions of experience following, selectively and not
always chronologically, the texts where they are either displayed or reflectively discussed. The
present chapter focuses on the dimension of knowledge in order to establish a conception of
rules, which cannot be captured in terms of the dichotomy between the transcendental and the
empirical, which is then developed further in the next chapter. So long as we presuppose the
ineluctable validity of the transcendental and the empirical as lying on logically different levels,
the Foucauldian project appears as a series of oscillations between the two. However, his concept
of experience—centered on a notion of practices—as constitutive of subjectivity is an attempt to
think not between or beyond these two, but involves a reconfiguration and rearticulation.
80
2.1 Analytic of Finitude: Kant and the Order of Things
In the introduction to one of the last texts that Foucault authorized for publication, he proposed
the following explication of his project: ―What I planned, therefore, was a history of the
experience of sexuality, where experience is understood as the correlation between fields of
knowledge, types of normativity, and forms of subjectivity in a particular culture‖ (UP, 4).
According to this definition, analyzing sexuality as a historically singular experience entails an
analysis of the three axes that constitute it. First, there is the formation of the sciences (savoirs)
that refer to it as object; second, there are systems of power that regulate its practice; and finally,
there are the forms within which individuals recognize themselves as subjects of this sexuality.
A little further in the text, Foucault specifies how such an analysis involves ―the games of
truth (jeux de verité) in their interplay with one another…their interaction with power
relations…[and] in the relationship of self with self and the forming of oneself as a subject…‖
(UP, 6). If games of truth are the ―rules according to which what a subject can say about certain
things involves the question of truth or falsehood,‖ (DE, 4:632) then it is possible to infer that
sexuality as a singular experience is constituted by historically analyzable practices that involve
three sets of rules: those which operate at the level of formation of discursive knowledge, those
which operate at the level of the exercise of power, and those which determine the forms in and
through which the subject constitutes itself.
Experience, then, is the correlation between knowledge, power, and what Foucault will
call subjectivation. Even if we leave aside the nature of their interrelations, a problem emerges:
81
is experience constituted or constitutive?102 The text warrants both readings to the extent to
which Foucault says, in this particular context where the object of investigation is sexuality, ―an
‗experience‘ came to be constituted in modern Western societies, an experience that caused
individuals to recognize themselves as subjects of a‗sexuality‘‖ (UP, 4; my italics). The inclusion
of a reflexive moment (―recognize themselves as‖) amplifies this problematic status of
experience, which Foucault is explicitly introducing as a concept that retrospectively
characterizes the trajectory he has followed from the History of Madness on.103
A first approximation to a resolution of this tension is to see that there appears to be an
isomorphism of sorts between what Foucault presents as the three axes constitutive of experience
(knowledge, power, and subjectivation) and the Kantian separation into domains of validity
(objective knowledge, practical/moral rightness, and aesthetic reflection). This appearance is
both problematic and not incidental. Problematic, because there are texts where Foucault
unequivocally claims that ―the episteme has nothing to the with Kantian categories‖ (DE, 2:371);
not incidental, because Kant occupies an ambivalent position in The Order of Things (―The
Kantian critique…marks the threshold of our modernity‖ (OT, 242)) and Foucault will later place
his own project in the space opened up by critical philosophy104.
Therefore, the rearticulation of experience Foucault presents at the end of his trajectory
requires reference to and resolution of a key issue he had raised at its beginning: what is the
102 For an interpretation according to which Foucault‘s invocation of experience in The Use of Pleasure reinscribes
the transcendentally constitutive subject back into what would have been the purely empirical analyses of
genealogy, see BTH, Chapter 5, and especially section 1, pp. 152-58. I shall argue that this interpretation
misconceives the way in which Foucault grapples with the transcendental-empirical divide.
103 This intention is clearly expressed by the set of questions Foucault provides on p. 7: ―…when he perceives
himself to be mad; when he considers himself to be ill; when he conceives of himself as living, speaking, laboring
being; when he judges and punishes himself as a criminal?‖
104 See, for instance, ―What is Enlightenment?‖ in PT, pp. 97-119, and the encyclopedia entry on Foucault he
himself wrote under the pseudonym Maurice Florence, in The Cambridge Companion to Foucault, ed. Gary Gutting,
(Cambridge: Cambridge University Press, 1994) pp. 314-319.
82
relationship between the transcendental and the empirical, and the status of man as an empiricotranscendental
double? If experience is somehow both constituted and constitutive, and if this
tension is to avoid sheer contradiction (or confusion), then we must look at the texts where
Foucault investigated the epistemic space of modernity: the analytic of finitude presented in The
Order of Things and his Introduction to Kant‟s Anthropology.
2.11 What is an empirico-transcendental double?
The analytic of finitude is the various efforts through which philosophical reflection and the
human sciences circle around the ambiguous figure of man, which ―did not exist‖ (OT, 308)
before the end of the eighteenth century. These ―warped and twisted forms of reflection‖ (OT,
343) are a response to the status of man as an empirico-transcendental double, which status is
made possible by and is a response to the dissolution of representation as the self-evident ground
of knowledge in the classical age. Foucault‘s well-known analysis of Las Meninas helps us
understand this transformation, and consequently what is involved in the claim that man is
introduced into the epistemic field constitutive of modernity, thereby initiating the analytic of
finitude.
Foucault‘s analysis of Velazquez‘s Las Meninas makes a case for how ―in this picture, as
in all the representations of which it is…the manifest essence, the profound invisibility of what
one sees is inseparable from the invisibility of the person seeing…‖ (OT, 16). The painting
becomes the representation of Classical representation and the epistemic space it opens up.105 At
105 Foucault‘s periodization presents a problem, to the extent to which he seems to reproduce the not untypical
division into Renaisscance, Classical Age, and Modern Age; that is, roughly, the fifteenth to mid-sixteenth to end of
the eighteenth to the 19th century on. For a gloss on this division, see Ian Hacking, Historical Ontology (Cambridge,
MA: Harvard University Press, 2002) pp. 77-78. In the next chapter, I offer some reasons why the emphasis on
83
several levels, the structure of the painting is organized around a point exterior to it. We can
isolate three elements, superimposed on this point external to but presupposed by the painting,
each corresponding to an element of representation: the king and the queen/object represented,
the spectator/subject viewing the representation, and the painter/subject representing. There
corresponds to each element its represented double in the painting: the image reflected in the
mirror and the man standing on the threshold, both situated in the background, and the selfportrait
of the artist in the middle-ground.
None of the elements, however, is represented as playing the role it does in the process of
representation. The models and the spectator are in the background, the former only a reflection
in a mirror; the painter is not in front of the painting that is ―Las Meninas‖. The effort to
represent representation, i.e. to thematize the act of representation itself, fails because it is built
around an ―essential void: the necessary disappearance of that which is its foundation…. [The]
subject has been elided‖ (OT, 16). The necessity of this failure constitutes the way knowledge is
organized in the Classical age and forms the main steps in Foucault‘s description.
Classical thought precludes a theory of signification, for this would ask the question
―how can we know whether a sign designates what it signifies?‖ Any answer to this question
presupposes that there is something in virtue of which a sign signifies the signified—for
instance, ―resemblance‖ plays just such a role in the Renaissance. What is peculiar to Classical
thought, and serves as its organizing principle, is that there is no intermediary between sign and
signified. There is no doubt concerning whether it represents what it seems to represent, and
discontinuities in Foucault‘s historiography overlooks the extent to which non-global continuities pervade his
histories from one end to the other. Ruptures are not incompatible with a historical continuity or traditions; what is
incompatible, however, is the epistemological function for which Idealist philosophy of history, or some versions of
hermeneutics, appeal to historical or traditional continuity.
84
precisely because of such transparency the problem of their linkage arises: whether there is an
external world, the actual existence of the objects represented, etc.
Because conscious thought is necessarily representative, we cannot ask how
consciousness comes to have the capacity to represent: what is taken for granted is that the
mind‘s thoughts have a coherent representational content. Representation as the organizing
principle of Classical thought thus eludes scrutiny within Classical thought, since one cannot step
back from it in order then to ask how it is possible. If representation were one element in an
ordered system, one object of knowledge among others, then it would be possible to place it in a
tabular series in terms of its identity with and difference from other elements. However because
it is that from which Classical knowledge becomes possible, it withdraws from the field which it
opens up for cognition and therefore cannot be an object of cognition.
Now, it is precisely this epistemic configuration that is transformed by the Kantian
critique, which ―questions the conditions of a relation between representations from the point of
view of what in general makes them possible: it thus uncovers a transcendental field in which the
subject, which is never given to experience (since it is not empirical) but which is finite (since
there is no intellectual intuition) determines in its relation to an object =X all the formal
conditions of experience in general‖ (OT, 243). Man consequently appears as both an object
produced by the world and a subject who constitutes the world.
This is doubly inscribed in the Kantian critique: first, to the extent to which in his Logic
we find, in addition to the threefold questioning, ―an ultimate one: the three critical questions
(What can I know? What must I do? What am I permitted to hope for?) [are then] themselves
referred to a fourth, and inscribed, as it were, ‗to its account‘: Was ist der Mench?‖ (OT, 341);
85
second, to the extent to which the problematic relation Kant‘s first Critique bears to his
Anthropology raises the question whether there was already in 1772 ―a certain concrete image of
man which no subsequent philosophical elaboration would substantially answer‖ (IKA, 19).
Reading Foucault‘s Introduction to Kant‟s Anthropology in tandem with The Order of Things,
there emerges the outline of an argument which links the dual status of man as constituting and
constituted with a fault line running through the Kantian corpus.
Kant then both demonstrates the divide between the transcendental and the empirical,
and blurs that distinction; and the wager of Foucault‘s turn to Kant‘s Anthropology is that it
enables him to demonstrate whether and to what extent ―[t]he Critique would…have been
inclining toward the Anthropology from the beginning…‖ (IKA, 19)106 If it is the case that there
is a Kantian prefiguration of the analytic of finitude, i.e. that Kant both introduces a de jure
wedge between the transcendental and the empirical (when he asks about the conditions of
possibility of experience, of moral conduct, of judgment), and collapses that very divide by
transposing the critical questions onto the question of man, then the analytic of finitude already
plagues modern reflection from its inception: the distinction between the constituted and the
constituting, and their superimposition.
But why should this be troubling for modern thought? The short answer is that, from the
moment when Kant challenges the self-evident status Classical knowledge assigns to
representation and questions by what right we may be said to have representations with objective
validity in the first place, a space opens up by virtue of the novel terms in which this questioning
is carried out, i.e. every empirical determination is to be submitted to transcendental reflection
106 Hence the pervasiveness of the question concerning the relationship between Kant‘s first Critique and his
Anthropology which runs through Foucault‘s Introduction, of what he calls anthropologico-critical repetition. See
especially pp. 106-109.
86
that seeks to determine its conditions of possibility and legitimacy; however, this sharp division
between what is de facto and what is de jure structures a field with a certain number of
determinate strategies, all of which pivot around human finitude, but none of which successfully
accounts for it. Thus begin various forms of reflection that either keep apart the transcendental
and the empirical, or collapse one into the other. This answer, however, does not quite express
the anxiety which both sustains and dooms in advance the strategies of modern philosophical
reflection to come to grips with the recentering on man of the transcendental perspective.
Foucault describes these figures under the heading ―man and his doubles,‖ and their relation to
the transcendental theme requires a fuller account.
The human being as an object of the empirical sciences is finite, which means that it is
produced within the networks of empirical determinations of which it is a part. These are the
networks which empirical sciences investigate under concepts like economic production,
linguistic systems, and physical environment. But transcendental reflection, since it insists on the
ineluctability of questioning every empirical determination as to its sources, objective validity
and legitimacy, shows that this human being must also constitute the world of which it is
nonetheless a part. The Kantian solution consists in making the very finitude of man serve as the
condition of possibility of knowledge. Having to experience spatio-temporal and causally
interacting objects, i.e. the very constraints imposed by our form of sensibility and categories,
makes the objects of experience possible in the first place.
Hence philosophical reflection on knowledge demands that we ground the objective
reality of the empirical determinations in the domains of life, labor and language through
87
recourse to human finitude107: ―At the foundation of all the empirical positivities, and of
everything that can indicate itself as a concrete limitation of man‘s existence, we discover a
finitude…the spatiality of the body, the yawning of desire, and the time of language…. [T]he
limitation is expressed not as a determination imposed upon man from the outside, but as a
fundamental finitude which rests on nothing but its own existence as fact, and opens upon the
positivity of all concrete limitation‖ (OT, 315) The point is that it is precisely and only through
the subject‟s desire, body, and expression that the forces of life, labor, and language, which
determine it empirically, are constituted in the first place. As the Kantian critique seeks to
demonstrate, philosophical reflection must show human finitude to be self-grounding by making
the finite subject its own foundation. Foucault‘s terms for this process of attempted selfgrounding
of finitude are the fundamental and the positive. It is not quite clear what the semantic
extension of these terms is, but it is possible to delineate their function within the analytic of
finitude.108
From the moment when man appears as both object and subject (of knowledge), its
identity with and difference from itself becomes a problem: how can such a being ground itself,
since nothing can literally precede and produce itself—even in the idealist sense of constitution.
The difficulty is expressed by Foucault‘s gloss on the subject‘s status in the Anthropology in
relation to the first and second Critiques: the space in which an anthropology could be
formulated is ―a space in which self-observation bears not upon the subject as such, nor upon the
107 The three Ls (life, labor, and language) then figure the broader problem of transcendental constitution. As such
they indicate an original passivity, the paradoxical nature of which I will describe shortly.
108 For useful discussions of their meaning, see Thomas Flynn, Sartre, Foucault, and Historical Reason, vol. 2
(Chicago: The University of Chicago Press, 2005), pp. 180, 254, 191; Gary Gutting, Michel Foucault‟s Archaeology
of Scientific Reason, (Cambridge: Cambridge University Press, 1989), p. 200 (―finitude as founding is fundamental,
finitude as founded is positive‖); BTH, part one, passim. Han draws on Foucault‘s IKA, a text which was not yet
translated into English at the time, in order to map out the points of convergence between OT and the themes of the
originary, fundamental, and a priori.
88
pure ‗I‘ of the synthesis, but upon ‗a I‘ that is object and present solely in its singular
phenomenal truth. But this ‗I-object,‘ given to sense in the form of time, is no stranger to the
determining subject; for it is ultimately nothing more than the subject as it is affected by itself‖
(IKA, 39)109 Foucault gathers the possible ways of articulating this self-grounding finitude under
three rubrics. Each corresponds to different modalities of the fundamental-positive pair: a) the
empirical and the transcendental, b) the cogito and the unthought, and c) the retreat and return of
the origin.
I think that the empirical and the transcendental, which occur as one version of the
analytic of finitude in Foucault‘s scheme, must be granted a certain privilege by the terms of
Foucault‘s own analysis. His own description of what motivates the particular entanglements and
contortions of this pair more generally applies to the analytic at issue: the necessary failure of
attempts to collapse the one into the other is what initiates the modern search for ―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…‖ (OT, 320).
Kant‘s question concerning the objective validity of representations requires that there be
a sharp distinction between the transcendental, as that which would constitute experience (as
empirical knowledge), and the empirical, as all that which would then be seen as the content of
109 I take it that the ―subject as such‖ refers to the noumenal self of the Critique of Practical Reason, whereas the
―pure I of synthesis‖ refers to the ―I think‖ of the Critique of Pure Reason. The passage continues by noting: ―Far
from the space of anthropology being that of the mechanism of nature and extrinsic determinations…it is entirely
taken over by the presence of a deaf, unbound, and often errant freedom which operates in the domain of originary
passivity. In short...a field proper to anthropology is being sketched out, where the concrete unity of the syntheses
and of passivity, of the affected and the constituting, are given as phenomena in the form of time.‖
89
such knowledge. As I pointed out above, there is an ambiguity even in Kant as to whether such a
division can be maintained in its purity. One post-Kantian strategy of tackling this problem is
that of reducing the transcendental to the empirical.110 We may turn towards physiology or
biology to provide a positivist grounding of the transcendental aesthetic: knowledge will then be
grounded on conditions that are merely empirical, and the relation of condition to what is
conditioned will be that of causation. Alternatively, we may turn towards history as the locus of
social and economic conditions in order to provide a concrete account of the transcendental
dialectic.111
Both strategies appear to work at the empirical level alone, but Foucault claims that they
presuppose concepts that refer to the transcendental, thereby repeating the figure of man as
transcendental subject and empirical object. The claim should not be very surprising—although
Foucault‘s synthesis of the diverse strands involved is ingenious—since it draws on well-known
criticisms of what could broadly be called positivism and Marxism, where these are understood
not so much as specific doctrines as determinate epistemic strategies.
To the extent that they are epistemic strategies, they presuppose a distinction between
knowledge of empirical objects and reflection on the validity and status of this first-order
knowledge. By the terms of their own strategies, there must be a distinction between empirical
truth and truth of discourse about empirical truth.112 The proper way to account for the
relationship of one to the other is what causes each strategy to founder. For a successful
110 OT, pp. 318-322
111 Foucault‘s gloss here is instructive for the argument of this chapter, since he comes very close to describing his
own project, thereby flirting with reproducing such oppositions as constituted-constituting and transcendentalempirical:
―…a history of human knowledge which could both be given to empirical knowledge and prescribe its
forms‖ (OT, 319).
112 Foucault says ―truth that is of the order of discourse‖ (OT, 320).
90
reduction, one must be grounded on the other. Positivism reduces the latter to the former, and
accounts of a Marxist type—what Foucault calls here eschatological, probably because the
achievement of such grounding is seen as the culmination of a historical process—reduce the
former to the latter. But as soon as positivist accounts attempt to provide a philosophical account
of their reduction of all account-giving to empirical truth, they appeal to principles that transcend
their internal confinement to empirical truth. Even though Foucault does not spell out exactly
what this would comprise, an instance of such a failure might be, say, that there can be no
satisfactory verificationist grounding of the principle of verification.113
Eschatological strategies of reduction, on the other hand, ground empirical truth on
conditions that are not of empirical order, thereby abandoning their claim to concretion and
operation on the empirical level alone. Both strategies then turn out to be self-defeating: either
they carry through the project of reducing the empirical to the transcendental (or vice versa), in
which case they generate a perpetual oscillation from one to the other; or they leave the
distinction between empirical truth and truth of discourse about empirical knowledge untouched,
in which case they reproduce the very transcendental divide they seek to overcome.
On the epistemological field structured by this configuration, phenomenology appears as
a way to avoid the twin shoals of positivism and eschatology. Foucault‘s engagement with the
strategy phenomenology constitutes is admittedly cursory—he spends little more than a page to
113 If this claim is too strong, a more guarded version is sufficient for my purposes: so far no satisfactory principle of
verification has been offered. In fact, the trajectory of logical positivism seems to capture the thrust of Foucault‘s
argumentative strategy in the analytic of finitude quite well. The analytic of finitude, as a critical argument in The
Order of Things, works not by offering yet another transcendental or quasi-transcendental argument to the effect that
the target epistemic moves are impossible in principle, but that we have good reasons to give up trying because they
are no longer attractive; and that‘s because the endless oscillations they generate at best show us that they are not
fruitful research programs, and at worst perpetuate a misrecognition of the practices they claim to ground. The
reason why there are few if any logical positivists today is due not so much to its definitive refutation in its global
goal and particular details, since there are programs that are sympathetic to and continue aspects of positivism; as it
is to its repeated failure to provide the principle which its method nevertheless requires.
91
dispatch it. Moreover, it is not quite clear which specific doctrines or arguments fall under the
term in this context. But at the archaeological level that is at issue here, it may not be necessary
to know the extension of the term, since it should be understood as yet another strategy to come
to grips with the transcendental theme.114Phenomenology recognizes the difficulty of collapsing
the distinction between the transcendental and the empirical by reducing one to the other.
Instead, it turns to the faithful description of lived experience where the subject is revealed as
both transcendental and empirical. Subtle descriptions of the body and cultural formations as
experienced reveal an irreducible spatiality that makes possible any empirical corporeal
comportment and ―the immediacy of [culture‘s] sedimented significations‖ (OT, 321).
Consequently it appears to fulfill the demand to respect the transcendental division, while
at the same time accounting for what makes it possible. However, Foucault‘s claim is that
phenomenological descriptions ultimately revert to the empirical. Hence phenomenology suffers
the same predicament as reductionist attempts to grapple with man as the empiricotranscendental
double.
My contention has been that the transcendental/empirical version of the analytic of
finitude is privileged because it puts into play the main elements of the transcendental theme
Foucault locates in Kantian critical philosophy and the post-Kantian strategies to find a
114I discuss the crux of Foucault‘s real contention against phenomenology later in this chapter. As a pointer in that
direction, it is important to note one point at this juncture: referring to the third term through the invocation of which
phenomenology would restore the forgotten dimension of the transcendental against various reductionist projects,
Foucault says ―Actual experience is…both the space in which all empirical contents are given to experience and the
original form that makes them possible in general….‖ (OT, 321)The English text has ―actual experience,‖ which is
not necessarily wrong but potentially misleading. The term Foucault uses is ―le vécu,‖ which is typically rendered as
―lived experience‖. In light of the distinction introduced in chapter one between Erfahrung and Erlebnis, this
difference makes a difference: phenomenology, in Foucault‘s description at this point, provides an account of selfgrounding
finitude by appealing to lived experience as both empirical and transcendental, while at the same time
doing justice to their irreducible difference. Since Foucault will shortly claim that this account also fails, his
proximity to and difference from phenomenology is intelligible in part through his repudiation of le vécu.
92
successful account of self-grounding finitude. However, his description of this version focuses
on the attempts to reduce the transcendental to the empirical. The inversion of this move,
namely, the reduction of the empirical to the transcendental constitutes ―the cogito and the
unthought‖. The transcendental theme articulated in the pair fundamental/positive centers on the
modern conception of man‘s double status: transcendental subject and empirical object. ―Cogito
and the unthought‖ traces a series of strategies in response to man‘s double status that take the
form of the human subject as experiencing consciousness, which grasps the world in reflective
awareness, and as the never fully present object of that experience.
Modern cogito, unlike the Cartesian cogito, is the attempt at articulating the transparency
of pure consciousness as inherently related to an unthought that cannot be fully incorporated into
reflective awareness or rendered fully conscious. The Cartesian thinking thing, precisely because
it is pure consciousness, can infer from the immediate certainty of ―I think‖ to ―I am‖: if we were
to transpose the process into the terms of modern episteme, we would say that it superimposes
the empirical and the transcendental, and therefore establishes an unproblematic connection
between man‘s reality as reflective consciousness and as object in the world. Any unthought
otherness which cannot be reduced to conscious awareness is only temporarily so. When
Descartes confronts illusions, dreams, and madness, as dimensions of thought that elude
reflective control, he does so only to recover them in the sovereignty of consciousness: even if I
am dreaming, I think.
Modern cogito acknowledges itself as intrinsically bound to an unthought it cannot so
master. In Kant this is inscribed as the thing-in-itself, whereas post-Kantian formulations
reinscribe the unthought as a dimension of our own reality, an otherness in which we must come
to recognize ourselves. Hence modern strategies strive to preserve the irreducible distance
93
between the cogito and the unthought in man, while searching for an account of self-grounding
finitude. But precisely for this reason ―the cogito will not…be the sudden…discovery that all
thought is thought, but the constantly renewed interrogation as to how thought can reside
elsewhere than here, and yet so very close to itself, how it can be in the forms of non-thinking‖
(OT, 324)
Transcendental phenomenology appears, as before, to succeed: phenomenological
reduction reveals the empirical realm, including the empirical self, as constituted by the
transcendental ego, without thereby making the two ontologically distinct entities. Man as
constituting subject and constituted object seems to find, in phenomenological reflection, a stable
account. However, this account fails because the transcendental ego opened up by the epoche
cannot be a pure consciousness, but turns out to be in the manner of the modern cogito, i.e. as
itself related to an unthought that it cannot ground. The phenomenological reduction, or the
series of reductions, fails to open up and preserve the division between the transcendental and the
empirical. Consequently, the transcendental ego in its many phenomenological guises, as
constitutively related to an opaque unthought that is its reverse side, repeats rather than resolves
the problematic double nature of man: ―The phenomenological project continually resolves itself,
before our eyes, into a description—empirical despite itself—of lived experience, and into an
ontology of the unthought that automatically short-circuits the primacy of the ‗I think‘‖ (OT,
326; modified).
The third modality of the analytic comprises attempts to anchor finitude in its relation to
time (OT, 328-335). These take the form of reflections on man‘s historical reality and its origin.
That man, from the first moment of its existence, is given over to a history not of his own
making articulates the subject‘s status as constituted—the positive. If it were possible to recover
94
the totality of this history as the series of conditions producing the subject, it would be possible
to ground man‘s enigmatic status as the empirico-transcendental double. But unlike a purely
natural species, the origin of which would be something other of the same kind, man‘s origin is
essentially other: man, which in this configuration is a being capable of knowing the world of
which he is a part, cannot be conceived as homogenous with its origin.115The origin, as the limit
of a series (of conditions) to which it does not belong, perpetually retreats in the very attempt to
recover it, since any putative origin as a moment in human history will ipso facto be shown not
to be the true origin.
Either the origin of man is a point that lies on human history, in which case it fails to be
truly originary, or there is an originary point where man cannot be present as such, in which case
it is not man‘s origin. Anchoring finitude by tracing the series of conditions that produced man to
their origin fails because the origin as the limit of human history perpetually retreats.
There is another strategy reflection on man as historical reality employs: if there is
history as such, it is only because human consciousness constitutes the world as a temporal series
of meaningful actions—the fundamental. Since human action is not merely an event in the causal
nexus of nature, philosophical reflection refers to human project as that through which the
succession of events becomes historical. The proper origin of man, then, is the point at which
such self- and other-constitution occurs. Even though this point will retreat perpetually when
pursued in terms of positive finitude—the fundamental meaning of humanity cannot be located
empirically—from the perspective of man‘s constitutive activity, this very retreat of the origin
will have been constituted as part of history. Thus begin the modern efforts to recover the
115See Gutting, op. cit., p. 205, for a useful description of this movement.
95
original project in and through which man constitutes history. This movement is what Foucault
calls the return of the origin (to the present) which would reveal the fundamental meaning of
humanity. As such it is tantamount to the recovery of an authentic meaning that would account
for the problematic but constitutive status of man‘s double nature.
These three modalities of the fundamental/positive pair, namely, the empiricaltranscendental
repetition, cogito and the unthought, and retreat and return of the origin, define
the anthropological quadrilateral (OT, 335) as the epistemological field within which modern
philosophical reflections seek to account for self-grounding finitude, their constitutive
problematic.116The function of the analytic of finitude in The Order of Things, as an
argumentative strategy not so much to refute as to de-motivate and undermine the possible
moves one can make within this field, is visible in the oscillations described above. Following
the fate of what could be called ―epistemological figures‖—positivism, Marxism,
phenomenology, etc.—makes a strong impression which undermines the reflective projects made
possible and necessary by the appearance of man as constituting and constituted in the wake of
the Kantian Critique, and calls into question the search for philosophical foundations for the
possibility of knowledge.
However, the wide canvassing and the hasty pace of Foucault‘s descriptions in these
sections somewhat blunt the sting of his criticisms.117The effectiveness of the criticism becomes
116 For a diagramming of this quadrilateral, see Flynn, op. cit. p. 180. I have been arguing that one ‗corner‘ of this
rectangle has a certain privilege over the others as determining Foucault‘s argumentative strategy here, the sense of
which will be further spelled out shortly.
117 For instance, the return and retreat of the origin runs through a series of figures that culminate with the claim that
the return of the origin takes two forms, plenitude and negation of being, that start off the infinite tasks with which
we have become familiar. The names Foucault mentions are: Hegel, Marx, Spengler, Hölderlin, Nietzsche, and
Heidegger. That is too high an altitude for the force of the criticism to have bite.
96
more acute if one traces the relative primacy of the transcendental and empirical repetition. I
think it is possible to flesh out the transcendental theme by delineating some of the salient
arguments we find in both German Idealism and Husserlian phenomenology.118This will further
clarify two of the contentions of the present chapter: First, it helps us understand why, pace Han,
the conceptual stake of the various types of investigation Foucault undertakes (archaeology,
genealogy, problematization) is not (the impossible task of) a nonanthropological reconfiguration
of the transcendental, but rather a giving up of transcendental reflection.119 Second, it enables
one to reach a vantage point from which may be evaluated the question I introduced at the
beginning of this chapter: does the concept of experience Foucault articulates succumb to the
type of oscillation between the constituted and the constitutive he dismisses in The Order of
Things?
2.12 Rights vs. Facts of Knowledge
From the moment when experience is questioned as to what makes it possible in general,
philosophical thinking in terms of conditions of possibility confronts a constitutive division
around which it circles: the subject as empirically limited and transcendentally determining.
What gives this division its intensity and urgency in philosophical attempts to provide a
118 My argument will follow two classic discussions, while also drawing on the discussion of Kant, Husserl and
Hegel in chapter 1of the present study: Dieter Henrich, ―Fichte‘s Original Insight,‖ in Darrel E. Christiansen, ed.,
Contemporary German Philosophy, vol. 1, 1982 and ―Foundation and System in the Science of Knowledge‖ and
―The Paradoxical Character of the Self-Relatedness of Consciousness,‖ in Between Kant and Hegel: Lectures on
German Idealism (Harvard University Press, 2008); Peter Dews, Logics of Disintegration, Post-Structuralist
Thought and The Claims of Critical Theory (London: Verso, 1987), especially Chapter 1.
119 I qualify this ―impossibility‖ at this point, since it could be plausibly argued, as I shall admit in the next chapter,
that the Habermasian and the Derridean appeals to ―quasi-transcendentals‖ count as just such reconfigurations.
However, I shall also argue there that ultimately the transcendental stategy motivating their conclusions presents
other, but no less intractable, difficulties.
97
theoretically constituted foundation for knowledge claims is the distinction, employed by Kant in
the transcendental deduction, between questions of fact and questions of right (quid facti and
quid juris). This distinction, so central to German Idealism and Husserlian phenomenology, even
as they contest its Kantian formulation, thereafter structures the field of epistemological inquiries
in terms of the sharp wedge between questions of legitimacy and questions of fact:120any
empirical limitation can only appear as such because it has already been determined by reason at
the transcendental level.
In Husserlian phenomenology, for instance, this problematic is inflected yet preserved.
To the extent to which any account of consciousness as externally conditioned must be grounded
on data presented to consciousness, the distinction between what is presented and the fact that it
is presented secures the irreducible division between the empirical and the
transcendental/essential. Causal accounts can never capture the intrinsic meaning of appearances
as given to consciousness. The series of reductions of the empirical world culminating in the
transcendental reduction, since they suspend even the belief that there is an external world
independent of consciousness, will reveal the essential structures of experience. And precisely
because the transcendental reduction will have suspended the naïve beliefs and commitments of
the natural attitude, the description of such essential structures will be untainted by contextual,
psychological, and historical particularity and contingency. The constituting activity of
subjectivity will thereby be shown to satisfy the demand for a theoretical foundation of
knowledge claims: the meaning of objectivity must be prior de jure to any objective
120 In fact, this distinction is by no means restricted to what Habermas calls philosophies of consciousness, but
survives transposition onto the intersubjective register of communicative rationality, as I will show in the next
chapter.
98
investigation. Even when Husserl, in the Origin of Geometry, broaches the historical constitution
of objectivity, it is from the foundational perspective of the primacy of juridical conditions.121
Husserl concedes that levels of objectivity are built up genetically and that there are
levels which cannot be objects of direct description. He proposes to reach such levels through the
retrospective movement of Rückfrage in order to uncover their meaning. Since any tradition can
only be constituted over time, the phenomenologist must question backwards to reveal the
historical sedimentations in experience. But this phenomenological attempt to ground the genesis
of reason and objectivity in history results in the oscillations Foucault describes in the analytic of
finitude: the instability is due to the twin demands of the project to provide descriptions of these
historical sedimentations in experience and transcendentally ground them.122
It is not that Husserl deduces a priori the empirical specificity of any and every cultural
and historical practice from an essence. But since the aim is to make manifest the essential
structures of any possible historical society, etc., in this account empirical facts can only have the
status of examples that start off the process of imaginary variation. Intuition of essences de jure
precedes every empirical historical investigation. The a priori sense of the historian‘s activity,
starting with the very idea of history, cannot be derived from particular historical formations,
since only from the perspective of the former are the latter recognized as such. This is clearly an
inflection of the fundamental movement of Kant‘s Copernican revolution.
Hence even Husserl‘s ambition to go beyond the Kantian transcendental reproduces the
dynamic between the fundamental and the positive. If reflection on history is necessary, it is in
121 For a forceful expression of this intention in Husserl, see Jacques Derrida, Edmund Husserl‟s Origin of
Geometry: An Introduction (Lincoln: University of Nebraska Press, 1989) pp. 112-113.
122 I should say, ―and thereby ground them,‖ were the project successful.
99
order to secure the possibility of reapprehending the primary evidences originally given to the
founder of geometry (or of any scientific or cultural formation). Therefore, the
phenomenologist‘s concession that there is a historical dimension of objectivity is to free
phenomenology from exposure to the contingency of history. As Husserl says, ―in the
phenomenological sphere, there are no contingencies, no mere matter-of-fact connections
(Faktizitäten); all is essentially and definitely motivated.‖123
But the tension between acknowledging that transcendental consciousness must take
account of the contingency of historical and social formations, and insisting that this very
contingency be grounded a priori, strains the project to the point where it becomes self-defeating.
The phenomenological reinscription of the transcendental theme leads to an irreducible distance
between: a) pure validity of logical laws, on the side of which one places eidetic certainty and
necessity, and b) facticity of empirical events, where one locates psychologism, history, and
specific cultural practices.124 The insistence that it must be possible in principle to separate
eidetic unity and transcendental form from the specific (determinate and contingent) experiences
in which they are discovered seeks to reduce a dependence on facticity which it simultaneously
confirms. To the extent that phenomenological description must start from a factual example that
is always historically situated and context-bound, the essences arrived at through imaginary
variation cannot erase the trace of this dependence.125 In the light of this dynamic, Foucault‘s
123 Husserl, Ideas p. 356.
124 To this extent Husserl remains beholden to the ideal stipulated in the Cartesian Meditations, as when he writes:
―Naturally everything depends on strictly preserving the absolute ‗unprejudicedness‘ of the description and thereby
satisfying the principle of pure evidence, which we laid down in advance‖ (p. 36).
125 I think Derrida‘s treatment of this duality in Husserl is instructive in relation to my claims about Foucault here.
For Derrida also argues that the Husserlian attempt to keep separate the ideal unity of meaning and its factual
embodiment through the constitutive activity of transcendental consciousness collapses and fractures the latter‘s
immanence. He shows that writing, as a material medium, is a necessary condition for the constitution of ideal
objectivity, since in its absence the actual linguistic intersubjectivity Husserl insists on could disappear (as when the
founder of geometry and his fellow geometers die). Without the detachment from any actual subjectivity writing
100
claim that phenomenology perpetually flirts with the very psychologism against which it defines
its fundamental project becomes intelligible; and so does the enigmatic remark that
phenomenology only fulfills ―with greater care the hasty demands laid down when the attempt
was made to make the empirical, in man, stand for the transcendental‖ (OT, 321)126.
There is one other salient way to spell out more fully the ―warped and twisted forms of
reflection‖ that comprise the analytic of finitude as empirical limitations are converted into
transcendental conditions of their own possibility. No sooner had Kant articulated the
foundational logic of critical philosophy than attempts were made to go beyond the limitations of
transcendental consciousness, especially by Fichte and Schelling127. The peregrinations of the
reflection theory of subjectivity, which comprise an important moment of that debate, are of
immediate significance for the concept of experience. Kant‘s appeal to the transcendental unity
of apperception as the formal unity of the thinking subject, without which the manifold of
intuitions could not come to possess the unity of experience, and his deduction of the categories
as the forms of this subject‘s synthetic activity, generated two major sources of dissatisfaction
for his immediate successors. The derivation of the categories, which are supposed to be a priori,
hence universal and necessary, from the table of judgments leaves much to be desired to secure
the claims made on their behalf; and the postulation of the thing-in-itself as the intelligible,
makes possible, the ideal objectivity and continuity of knowledge claims could not be secured. That guarantee,
however, is indissociable from the possibility of forgetfulness and destruction (Derrida, op. cit.). What distinguishes
his strategy from that of Foucault is that he sees this failure as the result, not of facticity and historical dispersion
shattering the immanence of consciousness, but of a transcendental structure more fundamental, in principle, than
consciousness.
126 See also p. 248 for a clear expression of how Foucault links phenomenology to this problem of giving empirical
contents transcendental value: ―[Phenomenology] is trying…to anchor the rights and limitations of a formal logic in
a reflection of the transcendental type, and also to link transcendental subjectivity to the implicit horizon of
empirical contents, which it alone contains the possibility of constituting….But perhaps it does not escape the
danger that…threatens every dialectical undertaking and causes it to topple over…into an anthropology,‖ which, for
Foucault, involves a blurring of that very distinction and a confusion between the empirical and the transcendental.
127 Henrich, Betwenn Kant and Hegel, op. cit.: for criticism of Kant by Jacobi, et al., see part II, on Fichte, part III,
and on Schelling in relation to Fichte, p. 192ff.
101
unknown and unknowable ground of our experience seems to reintroduce pre-critical
metaphysics.
Fichte insists that any attempt to improve on Kant‘s deficient solutions must follow and
not fall short of transcendental deduction, properly understood. What is relevant for my purposes
is that his system presents two features that persist throughout modern grappling with the
transcendental theme: first, he claims that the irreducible gap between concept and intuition,
which is central to Kant‘s deduction in the Critique of Pure Reason, cannot accommodate the
structure of experience as the activity of an absolute self, which is how transcendental unity of
apperception must be recast, if it is to be the principle of all experience; second, this absolute
self, which is self-consciousness presupposed by and constituting the form and content of
experience, cannot be understood in the standard terms of a theory of reflection.
According to this theory, selfhood implies self-consciousness, and hence it is essentially
reflexive. But this relation to self is cast as the subject‘s turning back on itself and grasping its
identity with itself. This reflective relation, then, is what produces and accounts for selfhood:
since the object grasped by the reflecting subject is the reflection of that very subject itself and
not something other than the subject, a self is thereby constituted. This account, however, is
circular: the relation of reflection constituting the self at the same time presupposes the self.
Unless the reflecting subject was already somehow the self, it could not be identical with the
object of reflection discovered through that activity. And not unlike the problem encountered in
ostensive definitions, since there is nothing inherent in a reflected image to establish a necessary
connection with that of which it is an image, the subject must already be somehow acquainted
with itself in order to recognize this image as his own. Appeal to a third element to establish the
102
necessary connection (of identity) generates an infinite regress, since we would then need to
know how the subject could recognize that element as sufficient for the purpose.
One way out of this reflective account of selfhood is to argue that the self posits itself. If
the twin dangers of circularity and infinite regress are to be avoided, the self must posit itself in
such a way that the subject does not preexist itself but emerges absolutely, and its emergence
coincides with a relation to itself. Since the relation to self and consciousness of the identity of
the relata emerge at the same time in this activity, it also implies a fundamental passivity. This
interbraiding of activity and passivity is not unlike intellectual intuition, which Kant rules out as
a possible mode of cognition for finite subjects with discursive intellects and spatio-temporal
sensibility. Fichte‘s claim, however, is that my consciousness of myself is just such a mode of
awareness: it has the characteristic immediacy of intuitions, yet it is nonsensory like the pure
forms of the understanding.
Moreover, since this recasting of the transcendental unity of consciousness as absolute
self still performs the transcendental function of grounding all experience, it cannot itself be
demonstrated theoretically. And because this condition cannot itself be theoretically
comprehended, it can only be encountered in a mode of awareness comparable to a special type
of vision128. Self-positing subject—which is precisely what this unique activity is—is an
alternative to the reflection theory, but it is still located within the space of the
transcendental/empirical divide at the heart of Foucault‘s discussion in The Order of Things. The
unconditioned conditions of all (conscious) experience of objects are still implicit in and derived
128 It could be argued that the replacement of discursive rationality with other modes of intelligibility in Husserl and
later in Sartre (and perhaps even in Merleau-Ponty) are motivated by considerations similar to those operative in
Fichte. For an account of the theory of reflection in German Idealism in relation to deconstruction, see Dews, op.
cit., Chapter 2, passim.
103
from within consciousness. True to the logic of self-grounding finitude, the effort to ground
(empirical) knowledge on an absolute foundation circles back to the knowing subject, since only
its double nature (as empirico-transcendental) provides the elements that can live up to the
requirements of such a foundation.
It could be argued against Foucault that he stacks the deck in favor of his thesis in the
analytic of finitude by selecting theories which end up with one version of transcendental
subjectivity or another as the absolute grounding principle. Then one would point out other
attempts to provide a more satisfactory articulation of the transcendental and the empirical
beyond the subject-object dichotomy. Schelling‘s criticism of Fichte, and his own system, would
be exemplary in this regard. However, it is possible to demonstrate that even such formulations
of the absolute beyond and prior to the differentiation between subject and object reproduce the
logic sketched out above and offer only competing candidates for the epistemic positions opened
up by the transcendental turn in critical philosophy.
The demand for a secure grounding of experience in a universal and necessary condition
constituting and legitimating any knowledge claim imposes a number of constraints on what can
count as such a condition. It must be: a) unique, b) unconditioned, c) heterogeneous with all that
is empirical, d) yet actual within experience. Schelling proposes the absolute I as the only
principle that can satisfy these constraints. From this perspective, Fichte continues to remain
within the ambit of reflection, since he arrives at his first principle by critically reflecting on the
conditions of knowledge. He therefore avoids dogmatism at the price of losing the unconditioned
character of his principle, because the subject is conditioned by its opposition to the object.
104
Only a principle which cannot become an object, and hence is never given in
consciousness, can ground the subject and the object, instead of deriving one from the other in
circular and conditioned fashion. Schelling‘s name for this principle is absolute I, which is not
the transcendental subject but a name for absolute identity itself. But since this claim amounts to
the postulation of a nonempirical moment which is unconscious of itself and which enters
consciousness only through its effects, and since discursive thought cannot recuperate this
moment in principle, philosophical thought can only become reflection on this constitutive split.
Even insistence on absolute identity in which all oppositions would be cancelled cannot stop this
movement, since, as unobjectifiable and unrepresentable in consciousness, this principle falls
within the ―Cogito and the Unthought‖ described above.
The account in chapter one stressed the juridical sense of Kant‘s transcendental
deduction. Its import in relation to the interpretation of the analytic of finitude provided above
becomes clear if these arguments from Kant‘s immediate successors are read as intimately
connected to that sense. If one understands Foucault‘s description of the analytic of finitude as
engaging with specific doctrines and theses, then his account appears to run roughshod over a
long period of European philosophy and his criticisms remain tenuous. If, however, we recast
that account as limning a series of definite epistemic positions and strategies made possible and
necessary by the introduction of the subject as an empirico-transcendental double in response to
a new configuration of philosophical reflection‘s knowledge of knowledge, then a few threads
binding the modern episteme emerge.
Kant insists that one must distinguish between the transcendental and the empirical
senses of philosophical concepts and that it is imperative not to conflate the two. Even though
the distinction begins to fray in Kant‘s own development of this foundational dualism, one
105
requirement stays in place, in relation to which all modern responses will be structured: that in
philosophizing one must take up the transcendental standpoint, which is separate from and
irreducible to the standpoints we inhabit in everyday engagements with the world or in scientific
explanation of those engagements. Failure to respect this distinction between the transcendental
and the empirical, then, becomes the source of dogmatism stemming from the (illusory)
transcendent standpoint of pre-critical metaphysics. Therefore, the critical project of ―deducing‖
the a priori conditions of experience is inseparable from the demand to justify the standpoint one
occupies in that very deduction.
Paul Franks, in an important study on German Idealism, formulates its stake in terms of
the Agrippan trilemma129. According to the Agrippan skeptic, only one of three possible answers
are available in response to a why-question: either it will be a brute assertion without a
justification, in which case it is itself ungrounded, or it will raise another why-question, in which
case it generates an infinite regress, or it will presuppose just what is at issue, in which case it is
circular. Hence, the justificatory process triggered by a demand for reasons will either terminate
arbitrarily or give rise to a viciously infinite regress or be circular. Unless an answer is available
which circumvents this trilemma, our cognitive relation to the world will remain unsatisfied.
In terms of the constitutive divide between the transcendental and the empirical
articulated so far, it is possible to see why the modern attempts to find a satisfactory resolution of
this predicament end up confusing the two or lapsing into either circularity or infinite regress.
Within this field what counts as an adequate response must be capable of absolute grounding, i.e.
129 Paul W. Franks, All or Nothing: Systematicity, Transcendental Arguments and Skepticism in German Idealism,
(Cambridge, Mass.: Harvard University Press, 2005), passim. See in particular Chapter 1.
106
it must terminate in an a priori principle from which the conditions for the possibility of all our
claims will be derived but which will itself be unconditioned130.
The analytic of finitude may then be reformulated as tracing out the epistemic field
governed by the structure of this demand. There must be a single absolute first principle to stop
the regress of conditions generated by reflection on what makes experience in general possible.
However, this principle cannot be homogenous with what it conditions, since it would then be
subject to the same conditions as the series it conditions, which is to say that it would no longer
be unconditioned. Hence the empirical cannot be grounded in itself. If, on the other hand, the
absolute principle is heterogeneous with what it conditions, then it either is immanent in the
series (of conditions) or transcends it. If it is the former, then it is a part of the whole, and hence
subject to its law, while somehow standing outside it. Therefore, the only principle that will
satisfy the demand for absolute grounding, thereby skirting the Agrippan skeptic, is one which is
heterogeneous with and transcending the empirical. Since, however, the principle thus invoked
cannot be transcendent in the manner of pre-critical metaphysics, it can only be transcendental.
What this ―schematic‖ rendering of the analytic of finitude entails is that the
contradictory demand to keep the transcendental and the empirical from reciprocal
contamination, while grounding the latter on the former, condemns the desire for epistemic—but
130 Although discussions of the Agrippan trilemma tend to focus on the epistemological nature of the claims at issue
(that is, reasons for belief), it extends to practical as well as ontic reasons (reasons for why I should do x and reasons
for why something is what it is). Franks, op. cit., p. 19 The use of the term ―ground‖ reflects that polyvalence.
Moreover, Franks traces the demand that physical explanation be kept separate from metaphysical explanation to
Leibniz and the success of modern scientific explanations in terms of physical causes alone. This leads to what he
calls the ―explanatory closure of the physical,‖ the thesis that no nonphysical factors be allowed into explanation of
natural phenomena. One consequence of this is a tension between the demands for absolute grounding and for the
explanatory self-sufficiency of physical laws, since the latter cannot provide anything absolute. What is significant
in this account for my purposes is the emergence of the constraints that a) the transcendental and the empirical must
be rigorously kept apart, b) there must be absolute grounding escaping the trilemma, c) nothing empirical can count
as absolute. The consequence is that the self-sufficiency of the empirical is both asserted and withdrawn, in which
may be found a prefiguration of the confusions Foucualt describes in the analytic of finitude.
107
also practical—satisfaction in advance to disappointment. It is in the light of this that the
different reinscriptions of Kantian transcendental apperception ought to be perceived: from
Fichte‘s attempt to avoid the regress of self-consciousness by insisting on a self-positing activity
in which subject and object are identical and of which I am immediately aware in first person
self-consciousness; to Schelling‘s protest that this first principle is not heterogeneous enough to
perform the requisite epistemological function and must itself be derived from absolute identity
beyond the subject-object dichotomy.
The fundamental-positive pair in Foucault‘s description stands for the superposition of
the transcendental and the empirical in the structured set of responses to this predicament.
Consequently it is not a head-on engagement with, much less a refutation of, the figures that may
appear therein. It is rather an attempt—the only one available once you no longer take up the
transcendental standpoint—to de-motivate the central role human finitude in its double nature
assumes in modernity. From this perspective, it matters little whether the transcendental is
transposed onto a linguistic register or is inflected towards intersubjectivity. The fundamental, or
finitude as ground, makes empirical determinations of the subject work as conditions of
possibility of knowledge: it is precisely because the subject is endowed with intuitive sensibility
and discursive intellect that empirical knowledge, which is what Kant means by experience in
the Critique of Pure Reason, is possible and valid.
However, this in turn is possible only by constituting empirical determinations as the
positive, or finitude as grounded. Since empirical determinations make sense only against the
background of their transcendental conditions, the relation between the fundamental and the
positive is inherently unstable: empirical finitude appears as already transcendentally grounded
in a retrospective movement and transcendental finitude only ever encounters itself as
108
empirically determined. The effort to ground this original passivity in the order of knowledge
only defers transcendental finitude as perpetually anterior to itself. The analytic of finitude is a
description of the dissatisfaction of the demand to think the empirical concretely and ground it
transcendentally, while barring the pre-critical invocation of prime mover or God. The
dissatisfaction is due not so much to some failure to comprehend a more adequate position, since
the taking up of a place within this site is not a matter of personal whim or intelligence, as it is a
function of ―the demands laid down when the attempt was made to make the empirical, in man,
stand for the transcendental‖ (OT, 321).
The flip side of this, however, is that satisfaction cannot be achieved simply by deciding
to no longer take up the transcendental standpoint. Giving up the transcendental cannot simply
be an expression of caprice but has to offer some reasons why. But this requirement threatens to
bring the entire panoply of modern philosophy in through the back door, and make Foucault a
captive of the same analytic he disclaims. The above account goes part of the way toward
motivating that renunciation: it is a matter of no longer taking as central the knowing subject as
transcendentally determining and empirically determined, where this decentring is a matter of
denying the epistemological functions it performs in the modern episteme. It is also necessary to
follow the consequences of giving up the transcendental and articulate a response to what is
thereby lost. It is Foucault‘s rearticulation of the concept of experience that can meet this
challenge.
109
2.2 Conditions of Possibility vs. Conditions of Existence
Recall the definition of experience quoted above: ―What I planned, therefore, was a history of
the experience of sexuality, where experience is understood as the correlation between fields of
knowledge, types of normativity, and forms of subjectivity in a particular culture‖ (UP, 4).
Foucault claims that experience so understood implies games of truth, and these in turn are a
matter of the ―rules according to which what a subject can say about certain things involves the
question of truth or falsehood,‖ (DE, 4:632). Experience, then, is constituted by historically
analyzable practices that involve three sets of rules: those which operate at the level of the
formation of discursive knowledge, those which operate at the level of the exercise of power, and
those which determine the forms in and through which the subject constitutes itself.
Given the analytic of finitude, the terms Foucault uses—―constitution,‖ ―rules,‖ etc.—can
be neither transcendental nor empirical, on pain of repeating the confusions implied by the
transposition of the transcendental theme on finitude. Foucault‘s association of this transposition
with the emergence of man and anthropology should no longer be puzzling in the light of what
has been said so far, since ―man‖ and ―anthropology‖ figure the epistemological strategies which
circle around the finite subject as transcendentally determining and empirically determined. Then
the difficulty which the definition of experience poses for Foucault can be stated as whether
there may be nonanthropological conditions of knowledge and, if these conditions are to perform
some sort of epistemic role, whether a sense may be given to the a priori that avoids the terms of
anthropology so understood131.
131 Or, given the Agrippan trilemma in terms of which I schematized the analytic of finitude: whether a sense may be
given to experience that would challenge the assumption that unless one provides an absolute grounding for
110
The three axes, the correlation of which is experience, in a sense yet to be determined,
are: knowledge, normativity, and subjectivity. These ―domains,‖ or ―fields,‖ are then supposed to
correspond to the trajectory of Foucault‘s investigations divided between three distinct
methodological principles: archaeology, genealogy, and problematization. It is not
inappropriate—and in fact it may necessary—to organize these methods as successive
developments of a single historical methodology which Foucault modifies in response to his
critics and his shifting interests. One would then go through the considerable terminological drift
created in part by the dispersion of Foucault‘s theoretical formulations across interviews and
articles, and one would be tempted to separate fact from fiction in Foucault‘s efforts to impose
retrospective unity on his disparate studies on knowledge in the human sciences, criminology,
sexuality, etc.
I propose to follow another line of questioning: is it possible to offer a reading of
Foucault‘s claims in his works, taken as a whole, through the lenses provided by the search for a
nonanthropological articulation of conditions of knowledge, which will then afford an
affirmative answer? From this perspective, the chronological discrepancies and terminological
inconsistencies between Foucault‘s statements will be important only to the extent to which they
facilitate or stymie that search132. Finally, such a reading will in part be framed by the criticisms
epistemic, practical or ontic claims that meets the requirements discussed above, one must abandon any claim to
knowledge as illusory and end up with the well-known Schellingean worry of ―an eternal round of propositions,
each dissolving in its opposite, a chaos in which no element can crystallize.‖ The Unconditional in Human
Knowledge. Four Essays 1794-1796 (Lewisburg: Bucknell University Press) p. 63
132 One consequence of this is that much that can interestingly be said about Foucault‘s relation to other thinkers,
for instance Bachelard and Canguielhem, or ―movements‖ such as structuralism will remain peripheral to the
argument developed here. However, precisely this will enable a no less interesting incorporation of these in the
course of what follows.
111
of Habermas and Han to which I eluded in the introduction to this chapter, since their force
derives precisely from the volatile relation between the transcendental and the empirical133.
2.21 Logos in Archaeology.
The ―field of knowledge,‖ structured by the rules of formation of discursive knowledge, which
occurs as the first element in Foucault‘s rearticulation of experience, refers predominantly to the
type of analysis Foucault offers in The Birth of the Clinic, The Order of Things and The
Archaeology of Knowledge, where the object of enquiry is specified as the historical a priori.134
Foucault‘s explication of this concept is by no means straight forward, and it is permeated with
the language of a priori conditions and rules characteristic of the analytic of finitude:
[O]n what conditions was Classical thought able to reflect relations of similarity or equivalence
between things, relations that would provide a foundation and a justification for their words, their
classifications, their systems of exchange? What historical a priori provided the starting-point
from which it was possible to define the great checkerboard of distinct identities established
against the…faceless [and] indifferent background of differences. (OT, xxiv)
[T]he original distribution of the visible and the invisible insofar as it is linked with the division
between what is stated and what remains unsaid. (BC, xi)
[W]ithin what space of order knowledge [savoir] was constituted; on the basis of what historical a
priori, and…what positivity, ideas could appear, sciences be established…only, perhaps, to
dissolve…. (OT xxi-xxii)
133 A more direct engagement with Habermas is offered in the next chapter.
134 In order to focus the argument on the terms I have so far pursued, I leave to the side a number of candidates that
may equally plausibly be offered as the objects of archaeology such as episteme, archive and order. I think the
historical a priori is the concept best suited to delineate Foucault‘s project in this register, and it avoids some of the
pitfalls to which other Foucauldian concepts succumb as he multiplies their number in The Archaeology of
Knowledge, at times beyond necessity, against his own best nominalist instincts. I will refer to these other concepts
shortly in order to explicate the notion of rule common to all of them.
112
A preliminary step towards unpacking the historical a priori and defining its role within
the ―constitution‖ of experience is the apparently innocuous claim that discursive knowledge
implies identifications by means of discriminating differences. But the mode in which this
activity takes place in everyday engagements with the social and natural world obeys rules
different from those governing its production and organization in more systematic groups
comprising what we tend to call the standard sciences. Foucault uses connaissance to refer to the
latter type of knowledge.
From the perspective of his analyses the term does not entail a commitment to the
validity or truth of the group of statements so characterized, and examples include biology and
astronomy as well as astrology, alchemy, and phrenology. Foucault‘s argument turns on the
claim that the statements (énoncés), which connaissances comprise, are rule-governed and
display a certain kind of regularity. This regularity is what archaeological analysis aims to
manifest: ―in a society, different bodies of learning, philosophical ideas, everyday opinions, but
also institutions, commercial practices and police activities, mores—all refer to a certain implicit
knowledge [savoir] special to this society. This knowledge is profoundly different from the
bodies of learning [connaissance] that one can find in scientific books, philosophical theories,
and religious justifications, but it is what makes possible at a given moment the appearance of a
theory, an opinion, a practice…. It is this knowledge [savoir] that I wanted to investigate, as the
condition of possibility of knowledge [connaissance], of institutions, of practices.‖ (EW 2:261-
62)
Hence the system of rules constituting the savoir of a given period provides the
conditions for the possibility of connaissances during that period. This preliminary specification
poses three questions in relation to the analytic of finitude: a) What is the concept of rule which
113
is central to savoir? b) What is the relationship of ―governance,‖ or ―regulation,‖ implied by talk
of conditions of possibility and asserted to hold between savoir and connaissance? c) What is the
claim to systematicity apparently issued by talking about a system of rules?
Foucault‘s most elaborate answer to these questions is found in The Archaeology of
Knowledge, where he distinguishes between different levels of rules and the domains they
constitute. These levels are articulated by means of five central concepts: discursive
formation/episteme, enunciative function/historical a priori, and archive135. Foucault‘s ambition
to provide what appears to be an exhaustive theoretical formulation at times compounds his
problems, but attention to what role these concepts play enables a more salient understanding of
the historical a priori (as I will reconstruct it).
The discursive formation is the set of rules articulating the regularity which obtains
between statements at the discursive level, where regularity is a function of objects, different
modalities of expression, concepts, or strategies of discourse. However, these types of regularity
require a reversal of the standard practice of historical analysis which takes the objects of
discourse as existing prior to and independently of discourse. The specificity of studying
discursive formations, by attending to what appears as the shared object of a number of
statements, for example, is that it dispenses with ―things‖ and substitutes ―for the enigmatic
treasure of ‗things‘ anterior to discourse, the regular formation of objects that emerge only in
discourse‖ (AK, 47). This formation and emergence of objects in turn are to be understood not in
terms of correspondence to a referent, but ―by relating them to the body of rules…and thus
constitute the conditions of their historical appearance‖ (Ibid., 48, my italics).
135 What I claim here about the central concepts of archaeological analysis follows Foucault‘s definitions and
discussion in AK, Part III ―The Statement and the Archive.‖
114
But then Foucault introduces the enunciative function, which appears to reproduce this
constitution at a higher level: there is a set of rules specifying the conditions a group of signs
must satisfy to count as a statement. At the enunciative level, what matters is neither the referent
nor even the object constituted by discourse, but the principle of distribution of possible objects:
―[A statement]…is linked…to a ‗referential‘ that is made up…of laws of possibility, rules of
existence for the objects that are named…and for the relations that are affirmed or denied within
it‖ (AK, 91, my italics.). So, for instance, to say of ―Colorless green ideas sleep furiously,‖ that it
is meaningless is possible only on the basis of a number of restrictions: it does not describe a
dream, is not a poetic expression, not the message a covert agent passes to her superior, etc. But
supposing that we take this statement as meaningless, it nonetheless has a referential as
statement: a rule, for example, according to which color or sleep can be neither attributed to nor
denied of an idea.
Moreover, the enunciative function provides the rules ―determining what position can and
must be occupied by any individual if he is to be the subject of [a given statement]‖ (AK, 96).
And finally, the enunciative rules specify the associated field of a given statement comprising
other statements with which it enters into relations of coexistence, and the type of materiality
which inserts the statement into institutions. Only a group of discursive elements meeting these
general requirements count as a statement.
Thus the enunciative function specifies the structure common to all statements in a given
period, but this is still not the object of archaeology, since ―everything is never said; in relation
to what might have been stated in a language (langue), in relation to the unlimited combination
of linguistic elements, statements…are always in deficit…there are, in total, relatively few things
that are said [at a given period]‖ (AK, 118-19). Therefore there is a further set of conditions that
115
select from among the number of statements made possible by the enunciative function only
those which count as candidates for truth or falsehood. This further set of restrictions, which
operate as the principle of selection accounting for the rarity of statements, is the historical a
priori. What is actually said, historically recognized as a candidate for truth/falsehood, is
specified by the rules which comprise the historical a priori. The domain of objects so constituted
are unlike the virtual possibilities (of combination) of a formal system of differential elements,
which constitute the object of a structuralist analysis; and it is unlike the potentially infinite
richness of meaning presupposed, and aimed at, by hermeneutic interpretation. Archaeology,
then, seeks to specify the conditions of existence of what is actually said in such a way as to
account for their rarity.
These ―conditions of operation of the enunciative function‖ (AK, 127) constitute a
domain where formal identities, thematic continuities, concepts, and argumentative strategies are
deployed. But then Foucault introduces his final set of conditions, the most succinct definition of
which is: ―I shall call an archive…the series of rules which determine in a culture the appearance
and disappearance of statements…their paradoxical existence as events and things‖ (EW
2:309)136. The archive, then, is the ―law of what can be said,‖ and it governs the specific
regularities of groups of statements in their historical existence.
Therefore Foucault‘s account of discursive knowledge—which constitutes one element of
experience—traces a series of sets of conditions or rules, starting with the discursive formation
and ending with the archive. It appears at first blush as if the movement of this series is that of
136 He continues: ―To analyze…the general element of the archive is to consider [the facts of discourse] not at all as
documents…but as monuments.‖ The former would be to subject discourse to a hermeneutic interpretation to
uncover hidden deep meanings, whereas monuments present a surface that is visible. For a restatement of the
definition, see AK, 128-29.
116
generalization. But the language Foucault uses also suggests that the relation is that between
ground and consequence, each successive term grounding the one preceding it. If that were the
case, we would have the movement from the empirical to the transcendental; and since Foucault
seems to terminate this movement not in a greater degree of ideality but with ―a purely empirical
figure‖ (AK, 128), his account would then repeat the confusions constitutive of the analytic of
finitude: that which is empirically conditioned comes to work as transcendental condition.
The terms of the historical a priori provide some initial traction on this issue. Foucault is
fully aware of the risks he is flirting with and is careful to specify what a priori in this context is
not137. It is not a condition of validity for judgments, and as such it does not perform any
legitimizing role; it is not a question of ideal truth which cannot be an object of experience or
which may never be actually said; it does not constitute an atemporal structure; it is not pure
form imposed from the outside on inert matter/content. It should be clear, against the background
of the analytic of finitude, that Foucault wants an historical a priori which will elude the endless
criss-cross of the transcendental-empirical divide. Hence, it is to be a question of: conditions of
emergence and reality of statements; a history that is given, since it is the a priori of things
actually said; the history of discourse distinct from its meaning or truth; the rules of
transformation of statements which is itself transformable; a historicity of discourse that is not a
history tacked on to a formal a priori.
Therefore, whatever historicizing the transcendental may mean, it is not a matter of a
transcendental scheme or set of rules that is simultaneously empirical. However, it is one thing to
insist on this difference and another to show whether and how such a thing as the historical a
137 AK, 127-28
117
priori, which Foucault admits to be a barbarous term (AK, 127), may be conceived otherwise
than transcendentally and empirically. And since barbarism is never so in itself but only in
relation to something else, it has to be possible to make this concept intelligible in terms not
entirely alien to those of modernity. I propose to offer one such account through a comparison
with a similar problem posed by Wittgenstein‘s later work.
2.22 Criteria, Rules, and Tribes
A key indication of the dual tasks in the service which Foucault introduces the historical a priori
is the following: ―[historical a priori] is defined as the group of rules that characterize a
discursive practice: but these rules are not imposed from the outside on the elements that they
relate together; they are caught up in the very things that they connect; and if they are not
modified with the least of them, they modify them, and are transformed with them into certain
decisive thresholds‖ (AK, 127, my italics). This passage crystallizes the peculiar conception of
rules with which Foucault is grappling. Unlike transcendental conditions of possibility, these
rules are intimately bound with what they determine and they are capable of historical
transformation; and unlike what is merely empirically given, they have the power to determine or
modify what they condition.
One consequence of this is that designation of the historical a priori as a ―pure empirical
figure,‖ cited above, can only be another barbarous term, and hence is in need of explanation.
My hypothesis is that the historical a priori has important structural affinities with and
differences from Wittgensteinian criteria, which provide insight into the problems concerning the
transcendental theme. What Wittgenstein says on the subject is arguably more obscure than
Foucault‘s pronouncements, but, since my object is not to engage the thorny issues of
118
Wittgenstein exegesis, I will follow an interpretation that resonates best with the theme I have so
far developed.
Wittgenstein makes a number of Delphic claims about criteria in relation to what he calls
grammatical investigation. He says that ―grammar tells what kind of object anything is,‖ that
―essence is expressed in grammar,‖ and that ―‗inner‘ process stands in need of outward
criteria.‖138He also describes his mode of investigation as grammatical, which is supposed to
yield the criteria governing our language use. The problems pertaining to how, even whether,
such criteria are ―found‖ or ―produced,‖ whether it is a question of what is implicit being made
explicit, and how such criteria determine actual use track those I have articulated in relation to
the rules constituting the historical a priori.
Not unlike Foucault‘s displacement of enquiry from discursive knowledge as consisting
solely of true statements onto the conditions of statements as what come to be perceived as a
candidate for truth or falsehood, Wittgenstein‘s criteria operate at a level in some sense distinct
from that of the traditional epistemological project of justification and extension of true
judgments: they constitute what governs the human capacity to apply fundamental categories to
the social and natural environment. And like archaeological investigation of statements, it is
neither a matter of logical analysis nor (simply) equivalent to the grammars of natural languages.
The criteria, then, are prior to specific acts of discursive practice such as identifying, naming,
judging, and measuring, where what this priority entails is difficult to capture in terms of the
transcendental theme. Criteria constitute (make possible and necessary) categorial understanding
of the world—and so, prior to criteria, there is as yet no object to identify, judge, measure, etc.—
138 Ludwig Wittgenstein, Philosophical Investigations, trans. G. E. M. Anscombe, revised 4th ed. by P. M. S.
Hacker and Joachim Shulte (Oxford, UK: Wiley-Blackwell, 2009) § 373, 371 and 580, respectively. All references
to Wittgenstein are to this work, unless indicated otherwise.
119
yet they are immanent in the very practices that are the different modes of human interaction
with that world.
What criteria are, and what sense of priority is at issue here, is suggested by Stanley
Cavell‘s contrast between Wittgensteinian grammar and traditional epistemological pursuit, in
terms of truth conditions for propositions, of justified true belief139. A not uncommon reading of
grammatical criteria is that Wittgenstein invokes them against the skeptic in order to secure our
claims to knowledge. Criteria, on this interpretation, are precisely ―the means by which the
existence of something,‖ say another person‘s pain, ―is established with certainty.‖140Criteria are
supposed to perform this solely epistemic function by laying down the truth conditions for
statements. Contrary to this reading, Cavell insists that Wittgensteinian criteria cannot fulfill this
epistemological demand, and that they are not intended to: criteria in fact show the truth of
skepticism. I will use his distinction between criteria and standards in order to frame my
discussion of how criteria elucidate the sense of rules operative in the historical a priori. Criteria
―determine whether something is of the right kind, is a relevant candidate,‖ whereas standards
―discriminate the degree to which the candidate satisfies criteria.‖141
Wittgenstein is interested in discursive practices which seek to settle or assess whether
something has a particular status/value. Legal reasoning offers a preliminary step towards
understanding the conditions of activities where what is at stake is the settling/deciding of
something. A judge is not supposed to modify the criteria whereby she judges an individual case.
139 Stanley Cavell, The Claim of Reason: Wittgenstein, Skepticism, Morality, and Tragedy (New York: Oxford
University Press, 1979) pp. 11-13. I follow Cavell‘s discussion of Wittgenstein (especially Part 1, sections 1 and 2),
which resonates with what I want to claim about the historical a priori, except where the limits of transformation of
criteria are concerned, as I will explain shortly.
140 Ibid., p. 6
141 Ibid., p. 11
120
But it may be controversial whether a given case falls under one or another set of criteria, or
even under any set whatever. When the judge decides the case, she is in fact identifying the case,
since her decision presupposes that she has also decided which criteria, if any, apply to the case.
So, for instance, the question is not only whether Roe v. Wade is to be settled in favor of one
party or the other, but also whether it is to be settled under privacy rights, i.e., whether privacy
rights apply to it. Therefore, in a sense that is neither exactly like legislating new law (say by
Congress), nor exactly like applying a law already given, she makes the law by extending the
established criteria, thereby modifying them, from which she begins her deliberation.
Wittgensteinian criteria operate at a level that is both more fundamental and efficacious,
and less secure than that of the judge‘s activity, where application of standards plays a role.
Criteria determine what counts as something: to have criteria for x is to know whether in an
individual case the criteria apply or do not apply. In relation to nonstandard cases, where there is
doubt about application, it may be informative to say that it is both x and not-x. But in the
absence of criteria, not even doubt about application can arise, since there is as yet no case with
respect to which we may debate available standards. If all knowledge claims presuppose a
criterial level, then truth conditional analysis in terms of evidence or reason for belief is not
sufficient to account for discursive knowledge, since criteria specify what counts as evidence or
reason.
The criterial level, then, not unlike the archaeological level, provides the conditions for
practices such as identifying, classifying, and establishing differences between objects. The
―site,‖ or ―mute order,‖ which Foucault evokes in the preface to The Order of Things to specify
the aim of his enquiry, is where one can locate what is criterial for a given period: ―...on what
‗table,‘ according to what grid of identities, similarities, analogies, have we become accustomed
121
to sort out so many different and similar things?‖ (OT, xix) This ‗table‘ is then ―that from which
forms of knowledge [connaissances] and theories become possible‖ (OT, xx-xxi) Therefore,
criteria, like the historical a priori, determine, in a given period, the ways in which people come
to know this or that, and what they cannot fail to come to know. This suggests that a kind of
necessity is operative in the way this determination takes place.
Wittgenstein‘s name for this level is ―form of life,‖ which provides the background of
necessities and ―agreements‖ which conditions particular knowledge claims. What kind of
necessity a form of life provides is tied up with the Agrippan skepticism to which I referred to
schematize the analytic of finitude. There are a number of relevant claims in the Philosophical
Investigations: ―Explanations come to an end somewhere,‖ ―Once I have exhausted the
justifications…[t]hen I am inclined to say, ‗This is simply what I do‘,‖ ―…how do I know [how
to continue the pattern]? If that means ‗Have I reasons?‘ the answer is: my reasons will soon give
out. And then I shall act, without reasons.‖142Since truth-conditional analysis is not selfsufficient
–because what is at stake here is whether anything, and what, counts as evidence or
reason for a particular something—the ―place‖ where ―explanations come to an end,‖ or where
―my reasons give out,‖ cannot be captured in terms of a demand for empirical verification of
statements.
Hence, even though criteria tell ―what kind of object anything is,‖ they are not
operational definitions. We learn what our concepts are by means of criteria, and therefore they
are necessary prior to particular knowledge or identification of objects. Wittgenstein refers this
necessity and priority to the background of systematic agreements between ―us,‖ a community of
142 § 1, 217, 211 respectively.
122
language users143. It is this background, as form of life, which is presupposed, accepted, and
given.144
The example of the judge‘s activity of decision-making brings into relief how a more
fundamental level must be presupposed as being prior to settling particular cases on the basis of
pre-established standards. But there is a significant disanalogy with Wittgensteinian criteria. In
legal reasoning there is still an authority which could change and issue new criteria in response
to a number of factors: for instance, the old ones may no longer be desirable with respect to the
activity of judging this or that kind of social issue. This implies that legitimate modification of
legal criteria is conceivable. However, Wittgenstein‘s claims about forms of life seem to
preclude the possibility of their transformation in these terms. He says: ―It is not only agreement
in definitions, but also (odd as it may sound) agreement in judgments that is required for
communication by means of language.‖145 Discursive practice, then, presupposes ―agreement in
judgments‖.
But no member of a community, let alone all members, could have agreed to all that is
necessary for using language. And since this level of agreement in judgment is presupposed by
all linguistic acts, it cannot itself have been arrived at on a particular occasion. It refers to what
Cavell calls our ―mutual attunement‖ with respect to language.146This attunement cannot itself be
a claim or a statement, since claiming or stating something implies that it can be challenged. But
such a challenge at the criterial level cannot be a question of stating something false as opposed
143 Here I am interested in tracing the sense of priority and necessity, which I claim characterize the historical a
priori as well as Wittgensteinian criteria; the significant differences between the two introduced by this reference to
a community of language users will be addressed shortly.
144 §345 ―What has to be accepted, the given, is—one might say—forms of life.‖ See also, § 242.
145Wittgenstein, §242.
146 Cavell, op. cit., p. 46
123
to true, or refusing to hear what is so said. Hence appeal to criteria is appeal to community: it is
the appeal to the network of necessities on the basis of which what counts as a reason, or as a
statement (as possibly true or false), or again as a justification, is constituted. Therefore it is the
appeal to what constitutes the community itself: ―what has to be accepted, the given, is…forms
of life.‖147
The necessity of a form of life implies that it cannot be understood in terms of contingent
customs or actual contractual agreements. Wittgenstein hints at ―those very general facts of
nature,‖ which determine what counts as normal for a group of creatures ―we‖ call human.148 At
first blush, this appears as a naïve appeal to an empirical human nature that provides the ultimate
ground of all our practices, which are then understood in behaviorist fashion. In anticipation of
what I want to claim about Foucault, I propose to read Wittgenstein‘s naturalism, if that is what
it is, in terms that render problematic how nature is conceived in standard accounts of naturalism.
On the one hand, then, criteria are efficacious and necessary in a way that goes beyond
mere custom, or explicit agreement, or even specific convention. It is possible to imagine the
practices that are merely customary, or contractual, or conventional in this sense as capable of
alteration without disrupting a group of people‘s cognitive and practical engagement with the
world. They therefore reach deeper than merely contingent rule-governed practices, where this
depth cannot be captured by, or explicated in terms of, individual or collective consciousness.
Criteria ―go without saying,‖ not because they are ineffable or obscure, but rather because they
147 Wittgenstein, §345.
148 ―If concept formation can be explained by facts of nature, shouldn‘t we be interested, not in grammar, but rather
what is its basis in nature? –We are, indeed, interested in the correspondence between concepts and very general
facts of nature.‖ §365
124
are too obvious and intimately bound with the very possibilities of what can and must be said in
a given community at a given time.149
On the other hand, criteria are not logically necessary and sufficient conditions which
will then render determinate the empirical decidability of knowledge claims under their
jurisdiction; nor do they indicate constant correlation. The determination of which they are
capable is therefore neither that of logical implication nor causal connection. They determine
application of concepts in statements because they specify what it is for there to be an object
counting as this or that kind of object, or having this or that value/status. And since, on this
account, what counts as this or that kind of object is ultimately a matter of what ―we‖ call as this
or that kind of object, criteria are subject to historical transformation and geographical variation.
It could be argued that this account of necessity, which is understood in neither logical nor causal
terms, is simply that anthropological sleep decried by Foucault transposed into a linguistic
register.
There certainly are passages where Wittgenstein‘s forms of life appear as ways of going
on of particular groups or cultures endowed with transcendental function by virtue of the putative
universality of human nature; thus necessity here would amount to no more than making what is
empirical, in the human, work as its own transcendental foundation. But I want to claim that this
appearance is due to an assumption according to which what count as criteria must perform the
149 So, for instance, Wittgenstein continues §365, which I just quoted: ―(Such facts as mostly do not strike us
because of their generality.) [But this does not imply natural science or even natural history] since we can also
invent fictitious natural history for our purposes.‖ The qualification I add by ―in a given community, at a given time‖
may seem out of place in the context of Wittgenstein. I want to advance two claims: 1) There is a sense in which
criteria, so defined, differ from the historical a priori, which difference I clarify in what follows, 2) The difference,
however, need not be an essential one: Wittgenstein‘s interest is in those practices which display a greater degree of
continuity across historical communities or across a given tradition, and his examples are carefully crafted to
manifest this; Foucault‘s interest, however, is in those practices which are more discontinuous in what could be
called the Western tradition, where the boundaries are fuzzy—which is compatible with the historical a priori as I
reconstruct it—without, however, being meaningless.
125
epistemological function of securing the certainty of our judgments. It is therefore the
assumption that, in response to the Agrippan skeptic, only foundational explanations or
justifications, which establish the possibility of legitimate knowledge claims, could perform the
required task, which is only understood transcendentally. But the place where explanations come
to an end and reasons give out is not intended, and more significantly, does not need to be a
transcendental ground. It that sense nothing grounds our practices but the practices themselves.
Wittgenstein‘s way of problematizing our practices is to refer them to a ―tribe,‖ which is
to imagine them in unfamiliar contexts where issues of epistemological certainty do not arise and
yet they appear ungrounded.150 So, for instance, it is not necessary that a tribe punish those
members which it takes as criminals by imprisoning them under conditions of constant
surveillance; nor is it impossible that it place them in dungeons no light penetrates. But if the
tribe is to have some kinds of things or activities that will count, for it, as criminals and
punishment, then it must accept some things as criminals and some activities as punishment. The
sense of necessity articulated by the consequent of this conditional is neither logical nor causal,
since it does constitute what ―we‖ must take as punishment and yet ―we‖ cannot determine a
priori what kinds of activities or entities will count as criminal or punishment: what a tribe takes
as necessary is capable of historical transformation.
A long standing objection to Foucault‘s account of rules of formation of discursive
practice should be recast in the light of what I have so far said about criteria and forms of life.
The classic development of this criticism is found in Dreyfus and Rabinow: ―If rules that people
sometimes follow account for what gets said, are these rules meant to be descriptive, so that we
150 See § 6, 200, 282, 385, and 419, where the practices are children ―playing trains,‖ calculating, playing a game of
chess, language learning, and ―having a consciousness‖. The examples I use in what follows anticipate and carry out
the assimilation of criteria to historical a priori.
126
should say merely that people act according to them, or are they meant to be efficacious, so that
we can say that speakers actually follow them;‖ but since Foucault‘s account claims to be both a
pure description of discursive events (AK, 27) and prescriptive determination of discourse, he
locates ―the productive power revealed by discursive practices in the regularity of these same
practices. The result is the strange notion of regularities which regulate themselves.‖151
It is indeed strange that regularities should regulate themselves, but why? The objection
relies on four interrelated claims: 1) One can provide either a description of regularities among
statements, in which case the rules are invoked only to systematize and give coherence to
phenomena; or one can give conditions of possibility governing (discursive) phenomena, in
which case one must have some account of this operation on phenomena in terms of either
objective causal laws or subjective norms. 2) One must be able to locate these rules either in the
consciousness of the speakers (which they reflectively obey and follow) or in the objective world
(which causes discursive behavior). 3) The only alternative to intentional and causal explanation
is the structuralist one, which provides the formal rules of combination of a system of differential
elements. 4) Foucault‘s account is committed to an essential separation of discursive from
nondiscursive practices.
Against (1), (2) and (3), it should be observed that the alternatives listed are not
exhaustive. Consider the case of the grammar of the language which a group of speakers uses.
Here we have a set of rules that are descriptive and prescriptive: I can understand what an
English speaker is saying only if I obey the rules of English. And that condition can be
satisfied—is in fact satisfied—without my reflective awareness of the rules which I follow. And
151 Dreyfus and Rabinow, Michel Foucault, Beyond Structuralism and Hermeneutics, (University of Chicago Press,
1983) pp. 81 and 84.
127
these rules do not causally determine what is actually said, at least not in a way captured by
mechanical causality. Moreover, the rules governing the formation of statements in a given
period, where what is at issue is not the virtual possibilities of combination of linguistic
elements, but the determination of what constitutes something‘s falling under a concept—of
what counts as something‘s being this or that kind of object—cannot be analyzed by means of
the application of a formal system of rules. Hence it is not a question of structural analysis.
If, however, the charge is that Foucault‘s conflation of causally or normatively productive
rules with the description of mere regularities hinges on his commitment to the explanatory
autonomy of the discursive, then two claims can be forwarded: Either Foucault‘s insistence on
the self-sufficiency and self-determination of the discursive in The Archaeology of Knowledge
was an overstatement for which he had no justification, or his commitment to such autonomy
was already qualified, which qualification he then developed explicitly in later works. I am
indifferent between these alternatives in this study, since, in either case, my claim is that the
account of rules operative in the definition of experience does not presuppose an irreducible
separation in kind between the discursive and the nondiscursive, but brings to light their
articulation on one another through the concept of practice.152
The objection secures its thrust through the kinds of responses to the Agrippan skeptic
admitted as the only available ones. On that conception, talk of conditions of possibility of
152 I only forward this claim here, which is developed and justified in the next chapter, in order to focus the
argument of the present chapter on the development of the concept of experience in relation to the transcendental
theme. Since my hypothesis is that the concept of ―rules of formation…,‖ or ―conditions of possibility…,‖ which
permeates Foucault‘s reconfiguration of experience, runs through the dimensions of power and subjectivation, as
well as knowledge, the full extent of what this implies is offered in the next chapter through a reading of the History
of Madness. Focusing on the concept of rules operative in Foucault‘s historical analysis of knowledge here enables a
more productive development there. All that I want to render intelligible here is how the historical a priori need
imply neither the unmediated effectivity of nondiscursive practices nor the seamless mediations of discursive ones.
128
statements implies grounding what is empirical in what is transcendental. Foucault is then doubly
damned, since to insist on an historical a priori precludes even empirical grounding of formal
rules in natural laws of physics, for instance, those governing how the brain works. He is thereby
condemned to describing mere regularities of discursive phenomena. But granting these
descriptive regularities the status of conditions of possibility of anything is to become entangled
in the very confusion of the transcendental and the empirical.
From this perspective, it could be argued that even reference to Wittgensteinian criteria
cannot unravel the historical a priori from what it conditions. For Wittgenstein, at least, the
invocation of ―those very basic facts of nature‖ appears to do the requisite grounding. My
discussion so far should cast some doubt on the putative foundational function attributed to
forms of life: our linguistic practices themselves support and sustain the criteria that make them
possible; therefore there is no formal transcendental level. If, however, one insists that linguistic
practices are grounded in social practices, which are then determined by facts of human nature,
the Foucauldian rejoinder would be: there is indeed a dimension of nondiscursive practices
which are articulated with discursive ones, but this articulation is not that which holds between
ground and consequence. Since our access to what counts as human nature itself presupposes the
application of concepts that are criterial for our time and place, no epistemological purchase can
be gained by appealing to nature, human or otherwise. But this does not commit one to the view
that discursive practices spin in a void, unhinged from everything nondiscursive.
In the light of this, Foucault‘s definition of the historical a priori, which provided the
preliminary step toward the rearticulation of experience, becomes no less strange but nonetheless
intelligible: ―[historical a priori] is defined as the group of rules that characterize a discursive
practice: but these rules are not imposed from the outside on the elements that they relate
129
together; they are caught up in the very things that they connect; and if they are not modified
with the least of them, they modify them, and are transformed with them into certain decisive
thresholds‖ (AK, 127, my italics). The term is barbarous only in relation to the analytic of
finitude according to which what has history cannot be a priori. The rules constituting the
historical a priori are necessary, to the extent that they determine what counts as a reason, or a
true-or-false statement; in short, what count as the limits of the world at a given time and for a
particular group of people. But this a priori is at the same time utterly contingent, since the rules
―are caught up in the very things that they connect,‖ and are therefore spatially and temporally
situated.
Unlike Husserl‘s turn to the historical a priori, to which I referred in relation to the
analytic of finitude and which turns out to be suprahistorical, Foucault‘s analysis of knowledge
does not aim to secure the legitimacy of particular knowledge claims by referring them to an
origin or founding act. To the extent that Husserl claims to demonstrate the essential structure of
historicity itself, he seeks to show that the recovery of the original self-authenticating intuitions
grounding claims to knowledge is possible in principle. Against this, the historical a priori
constituting one dimension of experience avoids both the ―historico-transcendental recourse:
trying to seek, beyond all historical manifestation and point of birth, a project which would be
withdrawn from any event,‖ and the ―empirical or psychological recourse: seeking the founder,
interpreting what he wanted to say, detecting implicit meanings which silently slept within his
discourse.‖ (PPC, 23)
Analogous to the distinction between Wittgensteinian criteria and standards, the historical
a priori constitutes the conditions of acceptability of statements in contradistinction to their
130
conditions of predication153. The former are prior to effective determination of a statement as
true or false and make the latter possible by specifying what could become, at a time and place, a
candidate for effective determination of truth or falsehood. This distinction, however, is not one
of opposition or a matter of transcendentally grounding what is empirical, since conditions of
acceptability are referred to nothing psychologically, logically, or causally determining. They
inhabit the very practices they inform and are grounded on nothing more nor less than the
existence of those practices. It is not, therefore, a question of offering a theoretical resolution of
the terms of the Agrippan skeptic, since there can be none, at least none which avoids the
analytic of finitude. The temptation to resist is precisely the interpretation of historical a priori—
and by extension experience (of which it constitutes one dimension)—as offering just such a
theoretical grounding. Foucault is not offering a theory, at least unless any characterization of
our practices is theory154.
The Agrippan skeptic is not quite wrong to the extent that: ―A proposition must
fulfill…conditions before it can be admitted within a discipline; before it can be pronounced true
or false it must be…‘in the truth‘‖ (OD, 224, modified). A statement may be accepted as a
candidate even though it is false (say from the standpoint of the conditions operative in another
period), or even though no empirical verification at the time is available (say it is a hypothesis
awaiting the development of some new technology). Foucault mentions Mendel, the truth of
whose statements was not perceived by his contemporaries. He could be judged false only on
condition that he first belonged in the truth, but ―Mendel spoke the truth…he was not in the
truth‖ (OD, 224, modified). Schleiden, on the other hand, whose denial of vegetable sexuality is
153 For the occurrence of ―conditions of acceptability,‖ see PT, p. 60; Hacking, Historical Ontology, chapters 4 and
5; BTH, pp. 7 and 200.
154 I take up the issue of what this entails for the status of Foucault‘s own writings in Chapter 4.
131
false from the perspective of current biological accounts, was only committing a disciplined
error. Error, therefore, ―can only emerge and be identified within a well-regulated process‖ (OD,
223)
The historical a priori orders the space of this truth, which is not that of the determination
of true statements and their discrimination from those that are false, but rather of what can be
recognized as capable of being true or false, of what can count as a possible object of effective
predication of truth or falsity. And the space of truth defined by conditions of acceptability is
transformed historically, such that Mendel could say the truth without being in the truth, whereas
Schleiden could be in the truth without saying the truth. Outside this space there are only
monsters: ―Mendel was a true monster, so much so that science could not even properly speak of
him‖ (OD, 224). The ideal reconstruction of the formal conditions for the possibility of
legitimate knowledge claims ignores this monstrosity, which rather demands a ―teratology of
learning‖ (OD, 223).
Teratology implies both that which is marvelous or fantastic, and that which is abnormal,
defective and repulsive. This dual implication crystallizes the hypothesis I introduced above,
namely, that it is not necessary to presuppose an essential difference in kind between Foucault‘s
analysis of discursive practice, which would correspond to archaeology, and his analysis of
nondiscursive practice, which would correspond to genealogy. The concept of rules developed in
this chapter extends to the two other dimensions of experience, those of power and
subjectivation. The Order of Things provides the history of the same, whereas History of
Madness is that of the other (OT, xxiv). The interpretation of the historical a priori in terms of
temporal, context-bound rules determining what counts as necessary and what must appear as
impossible for a particular group in a given period brings to light how a history of the same, far
132
from being categorially and in principle separated from a history of the other, forms with it an
imbrication.
The full extent and justification of this hypothesis requires a further account of practices
and what it means for them to be historical, thereby registering the articulation of Mendel‘s
monstrosity on that of those whose abnormality is constituted by and inserted into mechanisms
of control. It should be clear, however, that the question with which the chapter opened, namely,
whether Foucault needs a version of the transcendental to ground what would then become his
empirical/historical studies, or he successfully manages to recast the transcendental theme
through a non-subjective account of experience, can be intelligibly resolved in favor of the latter:
experience must and can be rethought without reduction to subjectivity by means of an
articulation of the conditions of its possibility, where these are neither transcendental nor
empirical, at least in so far as these terms are defined by the analytic of finitude.
133
It would be a mistake to believe that organic
evolution, psychological history, or the situation of
man in the world may reveal these conditions. It is
in these conditions…that the illness manifests itself,
that its modalities, its forms of expression, its style,
are revealed. But the roots of the pathological
deviation, as such, are to be found elsewhere. (MIP,
60)
3. Reasons, Causes, Madness: The Articulation of Discursive and Nondiscursive Practices.
The History of Madness may appear an unpromising site to excavate a concept of experience
through an analysis of rule-governed practices, for two main reasons: First, its language, and
even methodology, are riddled with those of phenomenology and existentialism such as
―sensibility,‖ ―perception,‖ and ―consciousness‖. Second, even to those who are sympathetic to a
project inspired by Foucault, the work is reminiscent of romantic yearnings with its evocation of
the primitive purity of madness that would finally speak for itself. So much so that Ian Hacking
likens the book to Borges‘s Don Quixote penned by Pierre Menard in the twentieth century,
identical to that written by Cervantes, yet absolutely different from it155. The second one would
be stripped of the Romantic illusion that animated the first. Hacking does not say this explicitly,
but I think he would be in agreement with Borges‘s narrator who finds the second Quixote
infinitely richer than the first Quixote: after all, the same passage offering mere rhetorical praise
155 See his foreword to the English translation of History of Madness, pp. xi-xii.
134
of history in Cervantes becomes an explosive thesis when written by ―Menard, a contemporary
of William James….‖156
This assessment seems universal among readers of the book as different as Habermas and
Derrida, even though it may be based on different reasons. It would then be a sign of progress
when Foucault appears to distance himself from the book, which ―accorded far too great a place,
and a very enigmatic one too, to what I called an ‗experience,‘ thus showing to what extent one
was still close to admitting an anonymous and general subject of history‖ (AK, 16, modified).157
But perhaps this exorcism of phenomenology and existentialism misses the mark, to the
extent to which the ―enigmatic‖ concept of experience structuring the History of Madness from
one end to the other is not that of phenomenology (and probably never was).158 For here too
Foucault uses the vocabulary of conditions of possibility: ―In the reconstitution of this
experience of madness, a history of the conditions of possibility of psychology wrote itself as
though of its own accord‖ (HM, xxxiv). Moreover, there are three appeals to a ―concrete a
priori‖ at key points in the text where Foucault is describing the object of his investigation:159
[A] madman is not recognized as such because an illness has pushed him to the margins of
normality, but because our culture situates him at the meeting point between the social decree of
confinement and the juridical knowledge that evaluates the responsibility of the individuals before
156 Jorge Luis Borges, ―Pierre Menard, Author of the Quixote,‖ in Collected Fictions (Penguin, 1999) p. 94 The
passage Borges ―compares‖ is the following: ―…truth, whose mother is history, the rival of time, depository of
deeds, witness of the past, exemplar and adviser to the present, and the future‘s councelor.‖ And he continues, in
relation to Menard‘s Quixote: ―The final phrases…are brazenly pragmatic.‖
157 See also his well-known identification (and repudiation) of ―an explicit theme of History of Madness‖ as ―what
madness itself might be, in the form in which it presented itself to some primitive, fundamental, deaf, scarcely
articulated experience‖ (AK, 47).
158 As in the previous chapter, I am not so much interested in whether Foucault at the beginning of his trajectory
already was the one at the end as interpreting the methodology and concepts of that book to construct and justify
further the notion of practices which I take to be central to experience. It is not a question of showing what the
History of Madness really was, and I incorporate exegetical problems only insofar as they advance or stymie that
construction.
159 For the third occurrence, see HMfr, p. 472.
135
the law. The ‗positive‘ science of mental illness and the humanitarian sentiments that brought the
mad back into the realm of the human were only possible once that synthesis had been solidly
established. They could be said to form the concrete a priori of any psychopathology with
scientific pretensions. (HM, 130, my italics)
…does not simply signal a reorganization of the nosographical space, but, underlying the medical
concepts, the presence and the workings of a new structure of experience. The institutional form
that Tuke and Pinel designed, the constitution around the mad of a containing asylum space
where they were to admit their guilt and rid themselves of it, allowing the truth of their sickness
to appear and then suppressing it…all this now became an a priori of medical perception. (HM,
528, my italics)
Finally, there is a parallel to the distinction between ―being in the truth‖ and ―speaking
the truth,‖ which I analyzed in terms of conditions of acceptability and conditions of predication
at the end of Chapter 2. In an appendix to the 1972 edition of the book, Foucault writes: ―It will
be said not that we were distant from madness, but that we were in the distance of madness‖
(HM, 543, emphasis in the original).160 This almost symmetrical doubling of the language used à
propos Mendel suggests that here too it is a question of specifying the rules governing practices
constitutive of experience.
What is at issue then is the articulation of a structure of experience (of madness) which is
the condition of possibility of the myriad ways in which a group of people at a given time and
place, a ―culture,‖ objectifies madness. And this experience is understood in terms of practices
that are both discursive and nondiscursive, as the summary invocation of institutions, knowledge,
and practice of confinement in these quotations attests. Furthermore, this structure of experience
―is history through and through‖ (HM, xxxii). This historical a priori, which the experience of
madness articulates, should appear less puzzling in the light of what I have argued concerning
160 ―Madness, the absence of an oeuvre,‖ in HM, pp. 541-550.
136
the analytic of finitude. But it also poses new questions, especially given the presence of
passages where Foucault glosses his aim as trying ―…to recapture, in history, this degree zero of
the history of madness, when it was undifferentiated experience, the still undivided experience of
the division itself‖ (HM, xxvii, my italics).161 This appears to bring it squarely under the analytic
of finitude in the modality of retreat and return of the origin. Consequently, two questions must
be answered: First, how are discursive and nondiscursive practices related? Second, what is the
concept of rule operative in practices that would not repeat the confusions of the analytic of
finitude?
The argumentative strategy I will follow involves two steps, each corresponding to one of
these questions. First, I focus on two moments of the history of madness Foucault writes, ―the
ship of fools‖ and ―confinement,‖ in order to delineate how different types of practices are
―articulated‖ and what such spatial metaphors imply in terms of explanatory power. Drawing on
texts that are commonly labeled genealogical in relation to the History of Madness clarifies what
―power‖ is supposed to be or do in our understanding of phenomena. Second, I situate this
account of rules in relation to what has been called the ―space of reasons‖.
A number of significant objections leveled at Foucault provide the frame in which I
propose to carry out the development of rule-governed practices and defend their status. The
defense, following the logic of the analytic of finitude and the Agrippan trilemma, involves two
strategies: a) since it is not possible to place oneself explicitly outside the space of reasons
without thereby (and implicitly) situating oneself inside it, I show how the concept of experience
161 This occurs in the Preface to the 1961 edition of the book, which was dropped from the 1972 edition. The elision
and what it means have a convoluted history. I will note only a few salient points relevant to my argument when I
consider objections of a Derridean inspiration, not in the terms of Derrida‘s explicit confrontation with Foucault, but
through his criticism of speech act theoretical analyses.
137
need not be construed in terms of the inside-outside logic, b) I claim that insistence on this
dynamic as a transcendental requirement occludes dimensions of practices that become visible
through a Foucauldian account.
In Mental Illness and Psychology Foucault introduces a theme that remains constant
through its different methodological reinscriptions: ―[mental illness] is both a retreat into the
worst of subjectivities and a fall into the worst of objectivities‖ (MIP, 56). If his critics are right
to insist on the necessity of a reflection of the transcendental type on this constitutive divide of
the inside from the outside, it would appear that Foucault‘s fascination with madness anticipates
and recoils on the extreme subjectivism and extreme objectivism of his own ―method‖. I argue
that a reconstruction of the ―persons‖ of madness through the concept of experience shows that
there is no such coincidence of subjectivism and objectivism, and therefore that insistence relies
on problematic assumptions. I hope this will highlight the specificity of the story, so similar to
the ones told by Marx and Weber, written by Foucault.
3.1 Madness in the First Person
A first step towards bringing out the specificity of what the experience of madness Foucault
refers to might be goes through his earlier engagement with the subject in Mental Illness and
Psychology.162This study is clearly indebted to the dominant modes of description and analysis at
162 The term ―experience‖ appears some 94 times in History of Madness, whereas ―le veçu‖ appears some 8 times,
and its very centrality and pervasiveness make it somewhat amorphous, which is part of the reason why it is
criticized in Archaeology of Knowledge. Mental Illness and Psychology itself has a complex history. It is a largely
revised version, republished one year after History of Madness, of Maladie Mentale et Personalité (Paris: Presses
Universitaires de France, 1954), which appeals to a Pavlovian physiology in order to ground the existential thematic
of Part 1 on a Marxist account of social contradictions in Part 2. The revised version, which is the one translated into
English, rewrites large sections of Part 2 in order to historicize the very object of the study (in light of the History of
Madness). For a useful discussion of the differences, see Hubert Dreyfus‘s foreword to the English translation, as
138
the time, phenomenology and Marxism; but what matters in the work for my purposes is not
whether it is an uneasy juxtaposition of existential anthropology and Marxist social history, or a
brilliant attempt at their synthesis, so much as the types of explanatory factors Foucault admits
and insists on as necessary components of an adequate comprehension of madness, and their
transformation. And the key to what those are is visible in the identification of his questions at
the outset: ―Under what conditions can one speak of illness in the psychological domain? What
relations can one define between the facts of mental pathology and those of organic pathology?‖
(MIP, 1)
3.11 Mental Illness: Explanation or Description?
The first part of the book, which remains relatively unchanged from 1954 to 1962, answers these
questions by offering what would be progressively more adequate explanatory schemes: from the
mistaken attempt to formulate a metapsychology, anchoring both organic and mental pathology,
through the correct incorporation of individual history into psychoanalytic explanation to the
disavowal of all causal mechanical inquiry in favor of a phenomenological description of the
specific ontological structures organizing the style of one‘s existence. The 1954 version of Part 2
in turn provides a casual grounding of the structures described in Part 1 on actual social practices
situated within the contradictory relations made possible by capitalism. There is an uneasiness
between the two parts to the extent that Marxist explanation too seems to run afoul of the
authentic comprehension of pathology existentially described in Part 1. However, in 1962 this
well as Pierre Macherey, ―Aux Sources de l‘Histoire de la Folie: Une Rectification et Ses Limites,‖ Critique 42
(August-September 1986): 753-774 and Frédéric Gros, Foucault et la Folie (Paris: Presses Universitaires de France,
1997) pp. 7-28. I will touch on the interpretative issues only insofar as they shed light on the
discursive/nondiscursive theme. (References to the 1954 version are indicated as MMP.)
139
problem seems otiose with the repudiation, as explanatory bedrock, of both Marxism and
existentialism: they become subservient to ―the historical constitution of mental illness‖ (MIP,
64) where ―one must not regard [the] various aspects of mental disease as ontological forms‖
(MIP, 84-85).163
Two series of factors then: on the one hand, we have a necessary moment of description
that is irreducible to any causal account and stands in a problematic relation to causal grounding
in the objective contradictions of the social world; on the other hand, we have both moments
subsumed under a historical constitution of their very object, mental illness. Since the latter is
supposed to summarize the specific shift and insights signaled through ―experience‖ of madness,
it is important to understand what this contrast implies. I think there are four salient points that
motivate the methodological innovation of the History of Madness.
The first point to note is the criticism of the attempts to discover a metapsychology which
look at physiology to find the foundation for both mental and organic illnesses. The argument is
based on the premise that ―mental pathology requires methods of analysis different from those of
organic pathology and that it is only by an artifice of language that the same meaning can be
attributed to ‗illnesses of the body‘ and ‗illnesses of the mind‘‖ (MIP, 10). Paradigmatic for
organic pathology is functional explanation in terms of the relations between isolated and atomic
components. However, the unity of human behavior is more than the sum of its parts, therefore
its meaning cannot be captured through functionalist abstractions: ―…the integration of [human
behavior‘s] segments tends toward a unity that makes each possible, but that is compressed and
gathered together in each: This is what psychologists call [in the language of phenomenology]
163 The passage continues: ―…it is only in history that one can discover the sole concrete a priori from which mental
illness draws, with the empty opening up of its possibility, its necessary figures.‖
140
the significant unity of behavior, which contains in each element…the general appearance, the
style, the whole historical anteriority and possible implications, of an existence‖ (MIP, 10-11).
Consequently no explanation that does not take into account this significant subjective unity can
be correct. And since organic pathology implies a conception of the human as a natural species,
fully exhausted through physiological explanations, it cannot be an adequate comprehension of
mental pathology.
The second point is that Freudian psychoanalysis succeeds to the extent to which it was
Freud‟s genius ―to go beyond the evolutionist horizon defined by the notion of libido and reach
the historical dimension of the human psyche‖ (MIP, 31), where the historical dimension is
provided by attention to individual case histories. However, psychoanalytic explanations
misconstrue this dimension by reducing it to an evolutionary process, according to which there is
a normal development of the human psyche through various stages of biological instincts and
where mental pathology is simply regression to past stages. This type of causal explanation is no
less an abstraction than the previous one, since it reduces an individual‘s existence to the normal
functioning of its components: ―In psychological evolution, it is the past that promotes the
present and makes it possible; in psychological history, it is the present that detaches itself from
the past, conferring meaning upon it, making it intelligible‖ (MIP, 30).164
According to Foucault‘s interpretation of this evolutionary logic, mental illness becomes
a natural fall into a past stage, where present action appears only as causal reaction to its
164 I think Foucault‘s argument is effective against the theory of instinctual development which we do in fact find in
Freud and which is inseparable from his scientific ambitions. But Freud‘s own ambivalence in relation to this should
also be admitted. The hermeneutic aspects of psychoanalysis cannot be so easily assimilated to its scientific
pretensions, and the theory of action and rationality, irreducible to a representational account of the psyche, which
psychoanalysis provides, helps illuminate some strands of Foucault‘s own description of practice (as I argue in the
second part of this chapter).
141
antecedent. What is thereby lost is ―the specificity of the morbid personality; the pathological
structure of the psyche is not a return to origins; it is strictly original‖ (MIP, 25-26) Therefore the
specificity of pathological behavior can be grasped only if it is understood as an intentional
response to the lived world of the individual.165 Someone is morbid only ―insofar as present and
past are not linked together in the form of a progressive integration….In contrast with the history
of the normal individual, the pathological history is marked by [a] circular monotony‖ (MIP, 41)
The third point then is the description of these originary structures of lived experience
through which an individual constitutes a world. Individual history is not successive mechanical
determination, and hence cannot be explained causally. Explanation is abandoned in favor of the
description of ―both the experience that the patient has of his illness (the way in which he
experiences himself as sick or abnormal individual) and the morbid world on which this
consciousness of illness opens….The understanding of the sick consciousness and the
reconstitution of its pathological world, these are the two tasks of a phenomenology of illness‖
(MIP, 46).
The relation between the past and the present, as lived by the individual, is circular,
which implies that the meaning of the past, and any efficacy it may have in the present, is
possible only retrospectively from the perspective of a present interpretation. Pathological
behavior then is a strategic response, the meaning of which can be interpreted only on the basis
of the shape of the morbid world sustained by the individual‘s existence. Unless the unique
existential structures constituting one‘s perception of time, space and possibilities of meaning
and action are described, we can never understand why one person at this time in response to this
165 It should be remarked that the term which appears in this connection is ―le vécu‖ (MIP, 44).
142
situation becomes ill, or why another may perceive possibilities where he sees only obstacles.
These structures define the style of existence for an individual (which Foucault calls ―a sort of a
priori of existence‖). (MIP, 42) At this level, mental illness is not unlike an inauthentic relation
to the world, a refusal to appropriate the originary activity by means of which someone sustains
his possibilities of meaning and acting and mistakes such active structures for fateful alien
conditions.
The fourth point is the insufficiency of even the description of originary existential
structures sustaining the morbid lived experience of an individual, and their grounding in actual
social conditions. The 1954 version of Part 2 attempts this grounding through a social history
inspired by Marx. The individual may be active in giving meaning to her world, but this very
activity must be placed in the concrete social world and its contradictions. Explanation then must
move from attention to individual case history to the collective history of which the individual is
a part. Here Foucault appeals to a number of familiar factors such as the struggle over property
following the French revolution and imperial competition. Mental illness experienced by the
individual may be described in terms of its existential structures, but ―this pathological form is
merely secondary in relation to the real contradiction that causes it‖ (MIP, 83).
Even though the 1962 version drops the more strongly Marxist language, the criticism of
―phenomenology of mental pathology‖ retains the explanatory privilege of such phenomena as
competition, class struggle, group rivalry and exploitation (MIP, 82). The essential claim is that
social alienation effected through economic, legal, and institutional mechanisms is the condition
of possibility of mental alienation. So whether it is a question of explaining mental pathology
through a model of regression, or describing its existential structures, it is social history which
provides the fundamental conditions of possibility. For instance:
143
Neuroses of regression do not reveal the neurotic nature of childhood, but they denounce the
archaizing character of the institutions concerned with childhood. What serves as a background to
these pathological forms is the conflict, within a society, between the forms of education of the
child, in which the society hides its dreams, and the conditions it creates for adults, in which its
real present, with all its miseries, can be read. (MIP, 81)
This conflict cannot be captured through a physiological account, and it is irreducible to a
biological theory of the instincts. On the other hand, the description of the mentally ill person‘s
―style of existence‖ is indispensable to accurately understand its uniqueness, but it must
ultimately be grounded in the conflicts produced by the development of capitalism.
Consequently, the shift effected by the History of Madness is motivated by the
explanatory inadequacy of both causal mechanical explanations and phenomenological
descriptions. The incarnations of the former in physiological functional or psychoanalytic
evolutionary explanations understate the specific reality and efficacy of the subjective elements
in mental illness; whereas the latter exemplified by phenomenological or hermeneutic existential
descriptions and interpretations understate the objective social context which produces mental
pathology. The 1962 version of Part 2 is illuminating in that Foucault is playing off Marx against
Heidegger, and vice versa, in a theoretical field where psychoanalysis trumps physiology,
phenomenology trumps psychoanalysis, and Marxist social history provides the ultimate matrix
of mental pathology.
However, it is precisely the insufficiency of Marxist social historical explanation itself
which motivates a reconfiguration of this field. For the attempt to ground ontological structures
on social contradictions through the mediation of Pavlovian reflex mechanisms invokes causal
principles more mythical than those denounced in the first and second points above. Moreover,
this account implies that mental pathology remains an essence, invariant throughout its diverse
144
cultural and historical manifestations, and hence a socialized replica of physiology‘s ―nature‖.
Organic evolution, psychological history, or styles of existence provide the ―forms of
appearance‖ of illness, but its ―conditions of appearance‖ cannot be located there. (MIP, 60)
That‘s why ―the roots of the pathological deviation, as such, are to be found elsewhere‖ (Ibid.,
my italics). But the grounding implied by talk of ―roots,‖ when pursued causally in terms of
objective social contradictions, leads to nowhere.
Foucault‘s response to this deadlock is to historicize both the object and the method of
analysis. Two demands animate the strategy I described so far: first, the need for a non-reductive
account of mental pathology that will preserve both the subjective and the objective dimensions
of mental illness; second, the need to provide a theoretical foundation for this account. True to
the requirements of the Agrippan trilemma, Foucault‘s grappling with the true elements of
phenomenological descriptions of the lived experience of mental illness and its situation in a
concrete social world produce two series of conditions. He could construe these sets
homogenously, in which case what does the conditioning would be assimilated to what is thereby
conditioned; or he could construe them heterogeneously in order to provide an objective
grounding. But the causal story that must be told to make this route pan out resuscitates all the
bad elements implied by candidate theories.
Therefore it is not unreasonable to suspect that these demands themselves are the source
of the problem. I claim that the methodological innovation of the History of Madness consists in
part of giving up the second demand and reformulating the first one. This reformulation entails
five interrelated claims: 1) What counts as mental illness is itself historically constituted by
discursive and nondiscursive practices, 2) Consequently it has no objectivity apart from such
practices, 3) Both the subjective and the objective dimensions articulated above are to be
145
retained, and therefore the story told will involve relations of causality and meaning but 4) Both
dimensions are referred to a level of ―fundamental experience‖ that is on the order of neither
biology nor biography, 5) What is fundamental about this experience refers neither to a causal
nor logical grounding, but rather to its criterial function.
Attention to the mode of argumentation Foucault uses illustrates the explanatory import
of these claims. If one tries to explain ―religious delusion,‖ one might assert that religion is
delusional by nature or that in religious discourse and practice the individual discloses his
suspect psychological origins; one might also reduce religious ritual to its function within the
field of economic relations and exploitation. It is not unlikely that such explanations would
capture some elements of the object of study. But what is thereby left out of the account is the
crucial point that ―religious delusion is a function of the secularization of society‖ (MIP, 81).
Religious discourse and practice could appear delusional only for a group which no longer
permits a seamless assimilation of religious belief into its experience.
Foucault‘s appeal to ―culture‖ in order to mark this shift should not obscure the fact that
what is at stake here is reference to the conditions of existence criterial for a group‘s modes of
perception and action. Similarly, infantile behavior could be a refuge for an individual and its
reappearance could be perceived as pathological fact only against the background of social
practices that establish limits between the individual‘s past and present: only on the basis of
cultural practices which integrate the past by forcing it to disappear can neurosis as regression
become fact.
Phenomenological descriptions capture the irreducible structures through which
consciousness articulates its world pathologically. But these structures, which show the absence
146
of possibilities and the presence of restrictions, are not sustained by the magical causality of a
consciousness fascinated by its world. One must also show the effective causality of the real
constraints under which individuals find themselves. Therefore the conditions of possibility of
mental illness must reflect two dimensions: ―Our society does not wish to recognize itself in the
ill individual whom it rejects or locks up; as it diagnoses the illness, it excludes the patient. The
analyses of our psychologists and sociologists, which turn the patient into a deviant and which
seek the origin of the morbid in the abnormal, are, therefore, above all a projection of cultural
themes‖ (MIP, 60, my italics).
If one can no longer presuppose the independent and anterior existence of mental illness
as an ahistorical object, the essence of which is to be reconstituted, the central problem can no
longer be an objective causal grounding of subjective structures. Foucault could be sanguine
about such a possibility in 1954 only because he had identified the common thread binding both
dimensions in ―concrete man‖ (MMP, 2). Reflection on concrete man as transcendentally
determining and empirically determined was to provide the ultimate foundation. But the analytic
of finitude shows the interminable oscillations to which it leads. Therefore one reason why
scientific psychology is suspect is that it presupposes a philosophical anthropology that is
inextricably entangled in a confusion of the transcendental and the empirical. But simply
―turning, as is so often done, to the ‗economic and social context‘‖ (FR, 334) leaves the basic
assumption of an already constituted object intact.
Hence the specificity of the methodological shift is to consider ―the very historicity of
forms of experience,‖ which implies a ―‘nominalist‘ reduction of philosophical anthropology‖
(Ibid.). The relations of meaning discovered by phenomenology are integral to a proper
understanding of mental illness as much as the relations of causality discovered by social
147
historical analysis. But these must be referred to a fundamental experience which will specify the
conditions of emergence of the very object of enquiry through an analysis of the practices of
exclusion and knowledge. The very conception of madness as mental illness, or deviation from a
norm, is itself a cultural product and can only be understood by referring it to the network of
practices outside of which it has no existence: ―It has been said, only too often, that, until the
advent of a positivist medicine, the madman was regarded as someone ‗possessed.‘ And all
histories of psychiatry…have set out to show that the madman of the Middle Ages and the
Renaissance was simply an unrecognized mentally ill patient‖ (MIP, 64). But this interpretation
falsely assumes that madmen were in fact regarded as possessed and that those who are
possessed are in fact mentally ill.
3.12 How Fundamental Is Experience?
What then is the fundamental experience of madness which articulates the practices constitutive
of what for a given group at a given time counts as ―madness‖? I will take two moments from its
history and my interest will be in bringing into sharper focus its contrast with the explanatory
schemes outlined in the four points above. The immediate background is traced by the figure of
the ship of fools.166The figure crystallizes two contradictory aspects which constitute the very
coherence of the experience of madness in the Renaissance.
166 The following discussion is based primarily on ―Stultifera Navis‖ and ―The Great Confinement‖ in HM pp. 3-78.
Although the criticism by some historians of the history Foucault writes is beyond the scope of this study, I mention
in passing a few remarks in anticipation of some common objections: 1) There were no ships of fools, 2) There is no
evidence for the relative freedom of the mad in the Renaissance and the medieval period, 3) There was medical
treatment of the mad before modern psychiatry, 4) There was no confinement on the scale Foucault argues,
especially in England, 5) The Lockean conception of the mad as human beings whose associations of ideas are awry
cannot be assimilated to the paradoxical animality described by Foucault in relation to the exhibition of the mad. My
reconstruction indirectly addresses (1), (3), (4), and (5) in what follows. The next chapter is in part an answer to (2).
For a bibliography on the sources of some of these criticisms, see Tom Flynn, op. cit. pp. 3-31 and Gary Gutting,
148
On the one hand, there is the cosmic experience of madness: madness is the object of an
imaginary fear, in the sense that its strange powers are represented through images taken as signs
of another world, which constantly threatens to invade this one. But precisely because madness
figures in imagination, communication with it is possible in principle. The limit it traces is also a
surface of contact between two worlds, and to that extent madness can reveal the truth of the
world on this side of the limit, and the possibility of a dialectical reversal of the one into the
other is admitted as a constant danger. And as danger, its mode of existence still involves
participation and is almost equal with reason.
On the other hand, there is the critical experience of madness: madness appears as a
figure in moral satire. In the discourse of Renaissance humanism, madness appears from the
perspective of the standpoint which constitutes wisdom. Wisdom consists in the knowledge
which perceives the madness of reason and the reason of madness: belief in absolute truth
discerned by human reason alone is folly, since it forgets its dependence on an order not of its
own making; whereas madness is the sign of this dependence and its fragility. And since wisdom
entails grasping that reason can be rational only insofar as it uses madness, critical experience
still recognizes the danger of going to the other side of this mutual implication, and therefore
allows its possibility. The experience is not tragic, however, and it is already distanced in irony.
Confinement then becomes the figure of the experience of madness in the Classical
period. There is no longer an objective imagination of madness precisely because madness is
only the unreal fantasy of dreams. There is no longer reference to the tragic powers of the world.
―Foucault and the History of Madness,‖ The Cambridge Companion to Foucault, (New York: Cambridge University
Press, 1994). Gutting‘s arbitration at times relies on a strategy of downplaying the arguments concerning the
Renaissance experience of madness as nonessential, and hence risk missing a crucial aspect of Foucault's claim:
confinement is a specific practice not to be confused with just any locking up of people; ―mad people‖ were locked
up before but not confined.
149
Madness becomes pure opposition, and madman becomes transgressor of social norms. He is
confined in spaces created for all figures of unreason. However, there is also the attempt to
situate and exhaust madness on the table of species. Therefore, confinement too crystallizes
contradictory aspects which constitute the very coherence of the experience of madness in the
Classical period: it is simultaneously excluded and included. But this global contrast between the
Renaissance and the Classical period touches only the most general characteristics that stand out
in Foucault‘s account. The significance of the shift of domain to ―fundamental experience‖ as the
key interpretative term is not visible at this level, and I think the emphasis on what is glossed as
―ruptures‖ in early interpretations of Foucault covers up the intricate linkages that exist in his
analyses in the History of Madness.167 Only those linkages establish an account better than the
ones criticized above.
The phenomenon, therefore, which demands interpretation, is the confinement, roughly
around 1656, of a large number of people in spaces either constructed specifically for that
purpose or converted from previous use as lazar houses. This event does not happen overnight, it
follows a different pace in different places, but it is a fact to the extent that it represents a distinct
type of treatment of a distinct group of people, neither of which existed in the period commonly
called the Renaissance. The treatment is confinement, and the people are those of ―unreason‖: a
167 It is puzzling how the image of Foucauldian historiography, according to which people would live in one
episteme for about 100 years only to wake up one fine day to find themselves in another, could exert so much
influence for so long. Reading his histories, especially the history of madness, one is rather struck by the subtle
continuities that permeate them from one end to the other. I suspect there are two main reasons for this discrepancy:
First, Foucault‘s language in the 1960s tends to emphasize the discontinuity theme, probably in order to better mark
off his distance from philosophical historiography, which establishes continuities through the stipulation of
teleological or dialectical progress. Second, it is in fact not easy at all to perceive precisely what those continuities
are.
150
classification, no longer intelligible for us, gathering together beggars, vagabonds, libertines, the
mad, etc. It is that very et cetera which demands interpretation.168
Around this fact Foucault constructs a narrative in which one finds the interplay of a
number of factors such as philosophical discourse, medical knowledge, physical exclusion
through spatial segregation, and moral condemnation. This interplay is then supposed to
constitute the experience of madness in the Classical period. Foucault claims: ―The economic as
well as the moral demand for confinement was the result of an experience of work‖ (HM, 71, my
italics). But then he says: ―…linked to the experience of madness that resulted from this general
obligation to work‖ (HM, Ibid. my italics). A question immediately arises: just what results from
what?
To explain the practice of confinement it is necessary to appeal to its economic and social
functions. So, for instance, confinement serves the nascent bourgeoisie to control prices and
salaries by manipulating the supply of work force. But it also represents a compromise between
the religious imperative of charity toward the poor, the institutional representation of which is
the Church, and the economic and political imperative to exercise some control over
unemployment. It is also necessary to refer to the mechanisms of absolute monarchies, which
exercised arbitrary power through such instruments as lettres de cachet. Moreover, in the wake
of Luther and Calvin, poverty ―no longer spoke of glorification of pain, nor of salvation proper to
both Charity and to Poverty, but concerned rather the idea of civic duty, and showed the poor
168 And so, in line with what I claimed at the end of Chapter 2, ―the history of the other‖ is not categorially distinct
from but articulated on ―the history of the same‖: the et cetera, that is, the perception of similarities through
discriminating differences, presupposed by scientific and social classifications, is related to practices of exclusion,
where that relation cannot be reduced to either a logical or a causal one, and where ―exclusion‖ may be epistemic
and/or practical; however, it would be too hasty to equate exclusion with oppression tout court. I develop the latter
claim in the next chapter.
151
and destitute to be both a consequence of disorder and an obstacle to it‖ (HM, 57). Thus poverty
becomes the object of a moral experience which condemns it.
To reduce the multiplicity of these factors, which do contribute to our understanding of
what the confinement of figures of unreason involved, to any one of them at the expense of the
others, in order then to postulate a linear causal relation between them, ends up in mythical
explanations of the type described above. But more importantly, it mistakes the effect for the
cause; or rather, since causal language at this level is not quite appropriate, it must be said that
only against the background of practices which define a set of behaviors as transgressive relative
to social norms that a group is objectified as in need of confinement. The hotchpotch figures of
unreason did present a coherent unified object in the Classical age, and the first important step in
Foucault‘s shift is the acknowledgement of the unique creativity of practices: people are not
confined because they are figures of unreason, but the very figures of unreason are constituted
through a set of practices and have no existence apart from them.169
The second consequential step is Foucault‘s refusal to refer these practices either to a
constitutive consciousness, which would be their ultimate ground of justification and meaning, or
to the brute effectivity of material mechanisms. The choice between idealism and materialism
loses its natural necessity after the reconfiguration of the Agrippan trilemma, and ―experience‖
as ultimate explanatory term signals precisely this reconfiguration. The language we find in the
texts uses the spatial metaphors of vertical and horizontal modes of analysis. The distinction
sometimes refers to that between historical materialism, with its establishment of causal
169 Cf., for instance, ―Sadism is not a name finally given to a practice as old as Eros: it is a massive cultural fact that
appeared precisely at the close of the eighteenth century, constituting one of the great conversions in the Western
imagination—unreason transformed into the delirium of the heart….reappears not as a figure of the old, nor as an
image, but as discourse and desire‖ (HM, 361-62).
152
grounding between infra- and super-structures, and positivism, with its reliance on linear
causality.
But the more significant occurrence of the distinction reflects the difference between an
empirical analysis in terms of causal relations and a critical questioning in terms of conditions of
existence, as I reconstructed them in Chapter 2.170 In order for confinement as physical act to
pick out and enclose just this group of people, they had to be marked out as belonging together in
a uniform perception. And the priority which this perception has over the specific acts of
exclusion is neither logical nor causal, but criterial. But because it is criterial, it involves both
discursive and nondiscursive dimensions. Hence Foucault‘s reference, in tracing the contours of
the experience of madness, to medical and philosophical discourse, on the one hand, and
concrete acts of exclusion and evaluation, on the other.
We then get a division into four types of conditions, which in turn correspond to two
types of practices, discursive and nondiscursive: 1) critical consciousness, 2) practical
consciousness, 3) enunciatory consciousness, 4) analytical consciousness.171 The use of
―consciousness‖ in this classification should not occlude its extension: spatially and temporally
indexed, rule-governed practices criterial for a group of people. That these articulate the
experience of madness is clear from Foucault‘s concluding remarks: ―A singular experience
appears…and each element can interact with the other according to the law that is its
own….This experience is neither practical nor theoretical. It is part of the fundamental
experiences in which a culture risks all its values—allowing them to face contradiction‖ (HM,
174, my italics). And despite the fact that the language of discursive and nondiscursive practices
170 The texts are not very perspicuous on this distinction. For an example of the former occurrence, see AK, p. ; and
for an example of the latter, see HM, p. 28.
171 The Introduction to Part Two, HM, pp. 163-174.
153
is not central in the text itself, that this is the main division articulated by the four types of
consciousness is implied by the central thesis of the chapter on confinement: ―…a madman is not
recognized as such because an illness has pushed him to the margins of normality, but because
our culture situates him at the meeting point between the social decree of confinement and the
juridical knowledge that evaluates the responsibility of individuals before the law‖ (HM, 130).
Different methodological principles and concepts which Foucault proposes should be
understood in the light of this project as tracking these two types of practices. The historical a
priori is defined in The Birth of the Clinic using the vocabulary of an articulation of the
―sayable‖ and the ―visible‖; and there is a bewildering proliferation of conceptual machinery
starting with the Order of Discourse through the works typically labeled genealogical: apparatus
(dispositif), discipline, regime, and power-knowledge, to name a few. I propose, similar to the
strategy I followed with respect to the historical a priori, that these concepts be read as
continuing the articulation of the criterial level through a sharpening of its content and the
dynamic interplay of its elements.172
An interesting symmetry between such reinscriptions is thereby displayed. In his analysis
of the development of anatomical pathology in modern medicine, the object of enquiry is defined
as the ―common structure that delineates and articulates what is seen and what is said‖ (BC, xix).
This structure is then said to constitute the spatialization and verbalization of pathology that are
the conditions of possibility of a discourse on disease. The concrete historical analysis in turn
specifies a medicine of types, in which the sayable has priority over the visible: diseases are
172 I am aware that this is not entirely accurate when measured against all of Foucault‘s intentions and some of his
explicit statements. This does not impair the reading offered here because it is not my intention to faithfully
reconstruct the Foucauldian corpus; and because there is sufficient warrant in what he does say to justify such
reconstruction as is offered.
154
defined a priori and their empirical manifestations in individuals are discounted as secondary and
derivative, since no ideal type as such can be given in an individual body. Diagnosis consists in
recognizing the already-said of disease in the visible marks of the body.
The transformation of this configuration of the visible and the sayable into that defining
clinical medicine at the end of the Classical period can then be described: ―medical perception is
freed from the play of essence and symptom, and from the no less ambiguous play of species and
individuals: the figure disappears by which the visible and the invisible were pivoted in
accordance with the principle that the patient both conceals and reveals the specificity of his
disease. A domain of clear visibility was opened up to the gaze‖ (BC, 105). Anatomical
pathology is intelligible only against the background of this articulation between seeing and
saying—according to which all that is visible is wholly sayable and it is visible because it is
sayable—as a disarticulation: the visible is no longer translatable into what can be expressed
immediately and without residues, but implies the priority of the invisible over the sayable.
Anatomical pathology ―[introduces] language into that penumbra where the gaze is bereft of
words‖ (BC, 169).
Since the historical a priori is precisely this articulation of the visible and the sayable,
which is capable of a number of transformations, one could expect Foucault‘s concrete analysis
to appeal to both discursive and nondiscursive practices. The transformation whereby it becomes
possible to discern medical truth in the opacity of the sick body and the folds of the cadaver goes
through the clinic, as an institution for both the teaching and the practice of medicine. And as
such it is implicated in a set of measures of social protection. The sayable/visible pair tracks the
discursive/nondiscursive pair, which in turn subsumes enunciative-analytical/critical-practical
consciousnesses described in History of Madness.
155
Finally, I think some of Foucault‘s invocations of ―power‖ are demystified when read in
light of this symmetry. The texts from the 1970s introduce a new concept, power, which is then
related to a number of others such as dispositif, regime, and discipline. However, what is crucial
is the function it plays in his analyses: ―We should rather admit…that power and knowledge
directly imply one another; that there is no power relation without the correlative constitution of
a field of knowledge, nor any knowledge that does not presuppose and constitute at the same
time power relations‖ (DP, 27, my italics). Like the articulation of the visible and the sayable,
power-knowledge, far from being a substance or ontological principle, refers to the imbrication
of discursive and nondiscursive practices. And when Foucault attributes some sort of creative
power to it, as when he says that the exercise of power creates objects of knowledge and
knowledge entails effects of power, this should be interpreted in light of the criterial function of
such practices.173Moreover the symmetry I have been tracking finds further confirmation at the
level of historical analysis:
It doesn‘t much matter for my notion of the apparatus to be able to say that this is discursive and
that isn‘t. If you take Gabriel‘s architectural plan for the Military School together with the actual
construction of the school, how is one to say what is discursive and what institutional? That
would only interest me if the building didn‘t conform with the plan. But I don‘t think it is very
important to be able to make that distinction, given that my problem wasn‘t a linguistic one. (PK,
198)174
173 See DE 2:753. Foucault‘s gloss on ―creates‖ is ―makes them emerge,‖ which recalls ―conditions of emergence‖.
174 An equally clear statement of the same insight occurs in Foucault‘s discussion of how at a certain moment in the
medieval period it became possible to build a chimney inside the house. At that moment all sorts of new relations
between individuals became possible. But he claims that it would be wrong to conclude from this that ―the history of
ideas and thoughts is useless…. What is…interesting is that the two are rigorously indivisible. Why did people
struggle to find the way to put a chimney inside the house? ...or put their techniques to this use?...It is certain that
this technique was a formative influence on new human relations, but it is impossible to think that it would have
been developed and adapted had there not been in the play and strategy of human relations something which tended
in that direction. What is interesting is always interconnection, not the primacy of this over that, which never has
any meaning‖ (EW, 3:362).
156
Therefore referring the massive number of heteroclite and contradictory events to the
level of fundamental experience, which is the spatially and temporally indexed, rule-governed
practices criterial for a group of people, provides a better grid of historical intelligibility, where
―better‖ means: capable of discerning the historical singularity of events and critical
interrogation of their conditions of emergence. But the introduction of power as a specific
interpretative dimension poses new questions, which imply two strong objections: 1) What
becomes of the rationality of practices so construed? 2) Is it even intelligible to appeal to rules in
characterizing this criterial level which would be fundamental in relation to what we say and see
and do?
3.2 Madness in the Third Person
The intimate connection between knowledge and power may be as old (at least) as Bacon, but
this connection is thought within their categorial separation in modernity. Value-neutrality of
epistemological analysis presupposes that what makes a statement true and what makes it
effective are irreducibly distinct. Against this background, Foucault‘s attempts to express their
mutual implication and articulation are not only false but also unintelligible:
‘Truth‘ is linked in a circular relation with systems of power which produce and sustain it, and to
effects of power which it induces and which extend it. A‗regime‘of truth. This regime is not
merely ideological or superstructural. (PK, 133, my italics)
Truth is a thing of this world: it is produced only by virtue of multiple forms of constraint. (DP,
27)
Relations of power are not in a position of exteriority with respect to other types of relationships
[e.g. economic, epistemic, sexual] but are immanent in the latter; they are the immediate effects of
the divisions, inequalities, and disequilibriums which occur in the latter, and conversely they are
the internal conditions of these differentiations. (HS, 94, my italics)
157
If what Foucault‘s histories describe are the transformations of criterial practices, both
discursive and nondiscursive, and if no categorial distinction between them is to be admitted—
which is what their ―circular relation,‖ or ―internal connection,‖ or ―mutual implication‖ try to
articulate—then the immanent rationality of epistemic and practical activities is undermined:
cognitive claims to truth and practical claims to rightness would be reduced to the social field
marked by coercive relations of force through and through. This is a long standing objection
formulated by critics from the 1960s on, but I will focus on one formulation, by Habermas,
which I take to be exemplary.175
3.21 Habermas and the Transcendental Site Where We Meet
Habermas‘s critique is exemplary because he does not deny the pertinence of nondiscursive
practices to an analysis of discursive forms but insists on the necessity of inscribing them in a
normative frame, which subsumes strategic relations of force, as their ultimate horizon. From
that standpoint, the methodological reduction of truth and rationality which is presupposed by the
historical a priori or experience, as I reconstructed them, can only generate self-defeating
strategies: refusal to engage in explicit normative justification of one‘s own standpoint recoils on
the concepts deployed from that very standpoint. Therefore, if Foucault‘s history of madness,
say, two moments of which are discussed in the previous section, employs the thesis of an
articulation between coercive practices of control and epistemic practices, in order then to make
visible the complicity of the normative standpoint of modern psychiatry with strategies of
domination, then the very conceptual language Foucault uses is implicated in that domination.
175 My reconstruction of the argument which I take be a challenge is based on: Jürgen Habermas, The Philosophical
Discourse of Modernity, op. cit. (hereafter PDM); Moral Consciousness and Communicative Action, (Cambridge:
MIT Press, 1999) (hereafter MCCA); Knowledge and Human Interests, (Boston: Beacon Press, 1972) (KHI); The
Theory of Communicative Action, vols, 1 & 2, (Boston: Beacon Press, 1987) (TCA).
158
For Habermas, however, the cognitive instrumental relation between the subject and the
object must be placed in the broader horizon of communicative reason.176 And Foucault is
charged precisely with ignoring that horizon in favor of privileging what is only one form of
rationality: namely, that of instrumental and strategic relations. Contrary to this narrowing of
horizon, restoration of intersubjective interactions to their properly communicative dimension
would then permit the expression of normative principles implicit in dialogue. Following the
logic of transcendental argumentation described in Chapter 1, we would then get: ―Mutual
understanding and action coordination are possible only if …‖ where the consequent would
articulate necessary conditions with transcendental status.177
But since that status is not derived from a theory of faculties of the mind, its principles
would not be threatened by the circularity or dogmatism which potentially impugn the Kantian
deduction. Habermas‘s strategy is to derive that status by attending to the necessity implied by
having to raise cognitive and practical validity claims in intersubjective dialogue and debate.
Since I must raise validity claims not only when I make cognitive claims, but also when I express
my intentions or when I make judgments of taste, communicative rationality provides the
universal and necessary framework regulating both moral-practical and aesthetic-practical
interactions.178
Habermas too narrates a history not unfamiliar from Marx and Weber. Its crucial
difference from the one told by Foucault is that, relying on some elements of Hegel‘s
176 For various expressions of this strategy, see PDM, pp. 294-327, MCCA pp. 1-20, 116-194, TCA 1, parts I and III.
The criticism of Foucault from that perspective is pursued at length in PDM parts IX and X.
177 For instance, ―The quasi-transcendental necessity with which subjects involved in communicative interaction
orient themselves to validity claims is reflected only in their being constrained to speak and act under idealized
conditions‖ (MCCA, 203).
178 See TCA 1, p. 305 ff for an analysis of validity claims in relation to communicative action. Habermas claims that
his analysis of ―Please bring me a glass of water‖ ―holds true for all speech acts….‖ p. 307
159
transposition of deductive function on to retrospective historical narration, it is to perform a
legitimation of the standpoint of modernity.179It has two important moments for my purposes.
First, it describes the process whereby practical, cognitive, and aesthetic claims no longer
presuppose a foundation in religious or metaphysical worldviews. Second, they rather constitute
three spheres of value: morality, science, and art. For Habermas, their categorial separation
implies that relations of domination result not from the instrumental nature of rationality as such,
but from the colonization of rationalized life-world by economic and administrative systems
governed solely by functional imperatives.180
Thus part of what Foucault describes through an analysis of the transformation of
―work‖, signaled by the names of Luther and Calvin, namely, morality‘s gradually becoming
something to be administered, is incorporated into Habermas‘s account; so is the Marx-inspired
analysis of class-conflict.181 But they are both referred to a colonization of the life-world, which
only impedes the development of the latter‘s communicative potential. Therefore, power, which
Habermas analyzes under media and money, is external to the life-world, which harbors the
liberating possibilities of self-determination implicit in modernity.182 The process of
secularization which liberates the life-world from the suffocating hold of traditional norms is a
positive accomplishment, which is only partially colonized by the norm-free logic of economic
and administrative relations. The latter are always secondary and derivate.
179 ―Deductive function‖ here refers to the argument developed in the first chapter of the present study, where its
juridical sense in Kant is emphasized.
180 See MCCA, p. 17, TCA 2, pp. 194-6 (―completely differentiated validity spheres‖ p. 196); on the colonization
thesis, the following is a typical claim: ―…we have to show that the theory of communication can [explain] how it is
that in the modern period an economy organized in the form of markets is functionally intermeshed with a state that
has monopoly on power, how it gains autonomy as a piece of norm-free sociality over against the life-world, and
how it opposes its own imperatives based on system maintance to the rational imperatives of the life-world‖ (PDM,
349)
181 The passage quoted in n. 180 continues with an analysis of Marx in this register, and TCA 1 engages Weber
directly, as well as his occasional contrasting with Marx.
182 ―The Uncoupling of System and Lifeworld,‖ in TCA 2, pp. 153-197
160
Consequently, the positive moment of Habermas‘s critique claims to incorporate what is
good in Foucault, while avoiding what is bad. Concrete historical analysis is accepted as
indispensable in order to prevent critique from becoming perpetual negativity, but transcendental
reflection is also admitted in order to justify and ground practices by invoking necessary
communicative norms. And the conception of knowledge as categorially differentiated
underwrites this critical reflection. From this standpoint, Foucault‘s invocation of power can only
appear as an ontological substance, and therefore pre-critical dogma.
Habermas‘s analysis of scientific knowledge is instructive in relation to this charge. He
proposes a sophisticated account of how scientific practices presuppose non- or pre-scientific
interests. Interest in technical control grounds empirical-analytical knowledge expressed in
causal laws, whereas interest in intersubjective understanding grounds hermeneutics.183 The
former is valid over objectified processes, but its categorial separation from the latter guarantees
that the intersubjectivity of action-oriented communication will remain irreducible in principle.
Moreover, the interests in technical control and action coordination are not contingent
and arbitrary. These are fundamental orientations grounded in the (self-)reproduction of the
human species through work and interaction.184 Therefore, even though the account appeals to
the natural history of Homo sapiens, which is contingent, the interests underwriting cognitive
and practical practices have a quasi-transcendental necessity. And so the argument I schematized
183 ―Empirical analysis discloses reality from the view point of possible technical control over objectified processes
of nature, while hermeneutics maintains the intersubjectivity of possible action-orienting mutual understanding,‖
Knowledge and Human Interests, p. 191. He continues this passage with a contrast between horizontal level—
referred to interpreting foreign cultures—and vertical level—referred to appropriating one‘s own traditions.
184 ―I term interests the basic orientations rooted in specific fundamental conditions of the possible reproduction and
self-constitution of the human species, namely work and interaction‖ (Ibid., 196) Further down in the passage he
claims that ―[k]knowledge-constitutive interests can be defined exclusively as a function of the objectively
constituted problems of the preservation of life that have been solved by the cultural form of existence as such.‖
161
above through ―Mutual understanding and action coordination are possible only if …‖ is further
justified by Habermas because the antecedent is not up to any one individual. Mutual
understanding and action coordination are rooted in fundamental interests, which we cannot
choose not to have, on pain of ceasing to be human.
Habermas reinscribes this distinction between two kinds of interest in terms of that
between system and life-world in Theory of Communicative Action, and the justificatory work of
the argument is transposed into a linguistic register.185 But this modification does not matter so
much for what I want to argue. The categorial distinction between the three spheres of value as
constitutive of modernity‘s achievement of rationality remains. What provides both a criticism of
Foucauldian description of practices and the positive account that would escape its pitfalls is the
rational reconstruction of the presuppositions of intersubjective communication: there are claims
to which I am implicitly committed when I raise any claim in any of the spheres of value, and all
claims so raised are oriented toward intersubjective agreement as their ultimate horizon. That is
to say, I am implicitly committed to justifying my claim through reasons. This process of raising
claims and justifying them in the reciprocity of a dialogical situation presupposes the goal of
consensual resolution of conflict.
Moreover, this resolution is to be effected only through the force of the better
argument.186 But universal consensus, reached through argumentation, as implicit commitment
presupposed by every claim, is to be distinguished from factual agreement. Factual agreement
can establish truth or rightness only if speakers implicitly understand the conditions under which
their agreement would determine truth. And since that can never exist in fact, an ideal speech
185 ―Intermediate Reflections: System and Lifeworld,‖ in TCA 2, part VI.
186 TCA 1, pp. 28, 42, 348 (―the unforced quality that comes to a conviction only through good reasons or grounds‖);
PDM, pp. 130 and 305
162
situation is necessarily presupposed as regulative for all communicative interaction in principle.
Finally, the ideal speech situation, which would be fully transparent, and hence could only be
conceived in the absence of any coercion or distortion, has the factual force of the
counterfactual.187 In a genuine tour de force, Habermas argues that I must presuppose the ideal
speech situation as already holding in order to engage in any genuine conversation. Such is the
force of the space of reasons.
There are, then, a genuine consensus and a spurious one, and even though we are only
ever mired in the latter in fact, the former guarantees the legitimacy and rationality of our
interactions in principle. It is possible to criticize Habermas‘s recourse to the necessary and
universal presuppositions of communicative rationality based on the Foucauldian themes
developed above, on two registers: First, it distorts our understanding of past practices (the
register of historical events in their singularity); second, it distorts our understanding of present
practices (the register of its critical function). The two distortions both stem from the insistence
on the categorial distinction between validity claims, all of which are referred to the ―unforced
force of the better argument,‖ and nondiscursive practices, which, unless they are already
discursively justified, can only enter the fray to the extent to which agreement is not achieved
through argumentation.
187 PDM, p. 206 : ―But the contextualist concept of language, laden as it is with Lebensphilosophie, is impervious to
the very real force of the counterfactual, which makes itself felt in the idealizing presuppositions of communicative
action.‖ TCA 1, pp. 30-31 : ―The concept of propositional truth is in fact too narrow to cover everything for which
participants in argument claim validity in the logical sense. [Therefore] a more comprehensive concept of validity
that is not restricted to validity in the sense of truth [is required]. But [this does not imply] that we have to…expunge
every counterfactual moment from the concept of validity and to equate validity with context-dependent
acceptability.‖ Or again, MCCA, p. 19: ―Every agreement, whether produced for the first time or reaffirmed, is
based on (controvertible) grounds or reasons. Grounds have a special property: they force us into yes or no
positions. Thus, built into the structure of action oriented toward [consensus] is an element of unconditionality. And
it is this unconditional element that makes the validity (Gültigheit) that we claim for our views different from the
mere de facto acceptance (Geltung) of habitual practices.‖ In the same passage Habermas is clear that he sees
precisely this element as what transcends the specific spatio-temporal occasion.
163
But what reason is there to suppose that agreement can be underpinned and justified by
formal presuppositions? Agreement does not result merely from the giving and taking of reasons,
but presupposes the recognition of what is so given and taken as reasons. In other words, the
important question is not whether I must acknowledge my implicit commitment to the space of
reasons so much as what constitutes a statement as a candidate for rationality, or the conditions
that determine what counts, in a given time and place, as a reason: in short, the criterial level,
which cannot itself be evaluated in terms of reasons, since it is constitutive of the very force
reasons have contextually.
To be sure, Habermas is sensitive to the fact that it is not possible to simultaneously
evaluate the validity of specific claims and of the very framework of ideal speech situation
which facilitates that evaluation.188 But his strategy, through the ―factual force of the
counterfactual,‖ is to inscribe that impossibility in a distinction between de facto and de jure
consensus. This ensures that a speaker can always call a former distortion, thereby showing that
what passed itself off as speech free from coercion was not in fact so. This would imply that the
actuality of full transparency can never be established with certainty: hence Habermas‘s
fallibilism. But the in principle necessity of presupposing its actuality in any specific situation
where validity is at issue is referred to a fact of reason. What results from this is not only the
reproduction of the transcendental theme, and the interminable oscillations consequent on it; it is
188 ―Only in theoretical, practical, and explicative discourse do the participants have to start from the (often
counterfactual) presupposition that the conditions for an ideal speech situation are satisfied to a sufficient degree of
approximation. I shall speak of ‗discourse‘ only when the meaning of the problematic validity claim conceptually
forces participants to suppose that a rationally motivated agreement could in principle be achieved, whereby the
phrase ‗in principle‘ expresses the idealizing proviso: if only argumentation could be conducted openly and
continued long enough‖ (TCA 1, 42). Habermas provides a sustained discussion of what this claim involves in
MCCA, pp. 76-109, where he is trying to justify the principle of universalization itself, without, however, reverting
to what he regards as the weakness of Apel‘s appeal to an ―ultimate justification‖.
164
also the occlusion of the singularity of the situations in which only some statements come to
count as reasons and only some actions become candidates for certain types of evaluation.
The invocation of singularity should not be taken as a mystification, for it concerns
nothing more nor less than the very intelligibility of historically situated practices. If Buffon
could only see an undifferentiated mixture of myths, fables, and meticulous descriptions of
empirical observations in Aldrovandi‘s writings, this is because, for the latter, there was no
reason to differentiate what was written from what was seen; what was observed empirically
through natural signs were as much inscriptions (legenda) as what one read in fables. Hence the
coherent and seamless juxtaposition in one period, of what appears unprincipled mixture of fact
and fiction from the perspective of another: for Aldrovandi, that an animal has a certain
appearance to the naked eye is just as much knowledge of it as the roles it plays in myths. That
difference cannot be captured by calling him more or less credulous than ourselves: he obeyed
different criteria.189
Similarly, I could discount as nonsense or the baby-steps of modern science, the bizarre
way in which Paracelsus tries to make nature do things which it has no intention of doing; but
my discounting would be worse than anachronism: thereby I guarantee that it will never be
intelligible how that could be taken seriously.190 To suppose that both Paracelsus and I stand in
the same space of reasons which, though it may allow diversity at the level of content, formally
underwrites the necessity that we are all issuing validity claims, is at best a vacuous principle, of
no use in rendering visible the logic governing his statements; and at worst, precisely by
transcendentally establishing that we are basically doing the same thing, subject to the same
189 For Foucault‘s description of this scene, see OT, p. 39.
190 For Foucault‘s attempt to render visible what is criterial for Renaissance knowledge, through the categories of
convenience, analogy, emulation and sympathy, see OT, pp. 17-44.
165
normative values, it guarantees the empirical misrecognition of what he is doing. And if he does
not see and think like I do, this can only prove that he is either ignorant, or that he willfully
places himself outside the space of reasons. He thereby abrogates the rights that belong only to
the native inhabitants of that space by right. Perhaps, when the self-exiled foreigner is
Paracelsus, this may be of little consequence; but we have heard the same argument deployed in
relation to foreigners closer to home, and to justify practices definitely not of the order of that
which is discursive.
The point on which I am insisting is not only that Habermas‘s account is too
formalistic—though it is. As I indicated above, he also claims that historical analysis is an
indispensable element of the full work of justification. His work contains a wealth of historical
material, and his analyses are at times conceptually more fine-grained than those Foucault offers,
especially given the latter‘s reticence on what the concept of power entails.191 But that reticence
is motivated, and Habermas sometimes writes as if Foucault simply refuses to appropriate the
lessons which any good reader of Kant and Hegel and speech act theory should.
At first blush there may good reason for that condemnation. Foucault‘s reduction of truth
and rationality, and his insistence on the articulation of power and knowledge, fall afoul of the
presumption of rationality in terms of which we distinguish between actions and events, reasons
and causes. Since I typically interpret an occurrence as action only by attributing beliefs and
desires to the agent, which then constitute his reasons, failure to discriminate categorially
between power and knowledge may appear as conflating events and actions. On this
interpretation, we can make the requisite distinction only if we assume basic norms of rationality.
191 The Theory of Communicative Action, in particular, presents the development of the formal analysis of the
pragmatic presuppositions of communicative action through an incorporation of a social historical narrative. But this
narrative both presupposes and is supposed to in part vindicate the normative superiority of the modern standpoint.
166
I think it is here that reference to psychoanalytic theory proves helpful.192In his argument
against conceptions of the unconscious as a second mind, Jonathan Lear appeals to what he calls
―motivated irrationality‖ as precisely that aspect of behavior which is left out of accounts
emphasizing the necessary presumption of rationality.193According to one picture of how the
unconscious determines behavior, one must assume that there are beliefs and desires which
ultimately motivate the individual‘s action—since otherwise it would merely be a physical event
causally determined in space and time—but add that the individual is simply not aware of these
motives.
So when faced with inexplicable behavior, I must still attribute a motivational set to the
agent which would be sufficient to render his action rational; but since he is not aware that these
are his motives, I posit that there must be an unconscious space in which I can locate these
reasons. They may be bad reasons relative to what one takes as normal, but they are nonetheless
reasons. For in the absence of beliefs and desires which hang together to form a coherent
motivational set, actions dissolve into events. But once we attribute the necessary minimum
number of motives, we must multiply our attributions given the holistic nature of motivation on
this standard account: the unconscious thereby becomes a second mind on its own, imbued with
the magical power to determine the first.
As a result we get two sets of motivations, that which is conscious and that which is
unconscious. The actions following from each set make sense individually, since we can
192 This may appear questionable given Foucault‘s animosity toward psychoanalysis. But nothing in his critical
remarks on psychoanalysis impugns the elements which I appropriate in this context. Moreover, his practice of
historical criticism converges with some insights of psychoanalytic practice.
193 The specific theory motivating Lear‘s argument is Davidson‘s distinction between events and actions. See,
Jonathan Lear, Freud (New York: Routledge, 2005), pp. 30-43. I will modify Lear‘s account slightly in order to
develop the argument I have pursued against Habermas‘s claims.
167
understand each one as articulating a pattern of desires, beliefs, and intentions. Moreover, each
set constitutes reasons, more or less rational, for each action. Therefore the presumption of
rationality secures the in principle intelligibility of actions by removing irrationality one level up,
in the relation between the conscious and the unconscious sets.
Two cases Lear analyzes, one from Freud and one from his own practice, are instructive
in relation to what I want to say about the articulation of power and knowledge.194I take these to
be two snapshots from the lives of these individuals not unlike the snapshots we find in
Foucault‘s histories.
[R] is walking along a road on which he knows his lady-friend will later be travelling in a
carriage. He removes a stone from the road so that the carriage will not be damaged. A bit later he
feels compelled to go back to replace the stone in the road.
So we have two actions: removing and replacing a stone. R himself is puzzled by their
incongruence. But this puzzlement does not imply that he does not see himself as acting: he
knows what he is doing and when pressed, he can offer justifications. If we follow the standard
interpretation of what motivates actions, we will say that he has two sets of reasons, conscious
and unconscious. He believes that the stone is a danger to the carriage and he wants to prevent
that from happening (conscious); he also believes that his friend does not return his love and he
wants to punish her (unconscious). And since the mental is holistic, his unconscious will soon
contain an ever increasing number of motives: he believes that this is a good kind of punishment,
that failure to return love demands punishment, etc.
Each action is fully explicable in relation to its appropriate motivational set, and what is
irrational is the way in which they do not fit together. But this can be understood as a case of
194 Lear, ibid., pp. 27, 31.
168
akrasia, or going against one‘s better judgment. His unconscious motive to replace the stone is
stronger than his conscious motive to remove it. Therefore his actions can be fully reconstructed
as rational—even if some may consider his unconscious reason a bad one. R‘s reasons smoothly
fit into a propositional pattern, but the only problem is that the unconscious one lacks a name.
Contrast this case with the following:
[An] unhappy couple where each partner has, over the years, built up many reasons to be angry at
the other. But…in order to stay together each has devised a strategy of keeping the reasons for
anger out of conscious awareness. Officially and sincerely, each is not angry with the other. But
every now and then a vengeful act slips out—though the partner who acts is not really aware of
what he or she is doing.
The crucial difference between these two cases, which the presumption of rationality
passes over, is that the couple, unlike R, do have an articulated set of reasons for being angry.
But what they miss is indeed the awareness that they are angry. In their case it makes sense to
attribute to them a motivational set that would exhibit a propositional pattern: they are angry
that, but they lack the name for their reasons, and precisely by virtue of that they can
simultaneously hold onto their anger and remain together as a couple.195
But R‘s actions cannot be assimilated to this conception. He does not understand why he
does what he does because he does not yet have angry reasons, and not because he has
articulated unconscious reasons. His action cannot be understood in terms of a propositional
attitude: he is angry at, but not angry that. But this does not imply that his actions thereby
become events, pure and simple. He still knows what he is doing and he may offer some
rationalization to back up that knowledge. The presumption of rationality may demand that we
195 Lear refers this to the ―pre-conscious,‖ which exhibits the same structure as consciousness but the motives are
either not presently conscious or they are actively kept out of present consciousness. See ibid., pp. 27, 11.
169
reconstruct his reasons for him. And it may be that he will sincerely appropriate these reasons
and come to see his actions as motivated by them.
But we thereby construct a false self for him: he comes to think of himself as having this
or that unconscious desire and belief, and then go on articulating all the other unconscious
reasons presupposed by these. But the false image of self he builds as a result preserves the
image of his rationality only to condemn him to ever more intense repetitions of his irrationality.
Against this picture, we must insist that R has neither conscious nor unconscious reasons and yet
what he does is still an action. He is not consciously or unconsciously angry that his friend does
not return his love, but he is nonetheless angry, and the anxiety consequent on his ambivalence
frames his actions. The presumption of rationality covers up this situation precisely because it
describes the action as more rational than it is. What is thereby occluded is R‘s motivated
irrationality.
I claim that Habermas misrepresents practices just in this way, by reading more
rationality into them than there is to be found. He insists, not unreasonably, that we situate the
problematic, contested interactions in the space of reasons. Then the presupposition that we must
all be committed to discursively justifying the norms governing our validity claims ensures that
rational reconstruction of our reasons is possible in principle. Once that work is done, we may
continue the process by explicitly evaluating what is so reconstructed: there are reasons—
conscious or unconscious—and the cases where there is inexplicable behavior may be resolved
by showing that they are a function of inappropriate or bad reasons. Critical reflection brings
them to light, and hopefully we will all come to agree that they are bad reasons. And when we do
not all see that in fact, the situation in which that could happen is built into our doings and
sayings in principle. Once good reasons are named in the process of argumentation, the anxiety
170
generated by contested claims would disappear; just as R would cease (or should cease) to
unconsciously hate when he comes to see that it is motivated by bad reasons.
I should emphasize that I am not arguing from individual to social psychology by
analogy; or rather, there is an analogy, but it is not based on putative similarities between
individual and collective psyche. In fact, it is precisely the picture of action on the presumption
of rationality which assumes that Foucault‘s invocations of power must imply a metasubject, just
like its empirical counterpart but somehow pulling all the strings, while remaining invisible to
conscious or reflective awareness. What I say above should make it clear that it is only on a
particular interpretation of what motivation and action must look like, which is mistakenly
supposed to be universal, that ―power‖ is conceived as a metasubject. Contrary to that
supposition, the analogy on which I base my argument is that, in the description and
interpretation of both individual and social practices, the distinction between reasons and causes,
and therefore that between actions and events, is neither categorial nor exhaustive.196
I want to insist, against Habermas‘s conception, that the Foucauldian articulation of
criterial practices provides a better matrix of intelligibility for the interpretation of our history
and the critical work on our actuality than the recourse to the normative assumptions of
communicative action. Not unlike psychoanalytic interpretation, Foucault too starts from
puzzling behaviors, contradictory statements, and incongruent actions.197 We lock up criminals
and the prison appears as the most natural place for their treatment; and yet we all know that
196 I develop this further in the next chapter by unpacking what is involved in Foucault‘s characterization of strategic
action as ―intentional but not subjective‖.
197 Some of Foucault‘s claims which appear to characterize the a priori as unconscious, most clearly in the Preface to
OT, are perhaps more intelligible in light of this than by reference to Bachelard‘s psychoanalysis of the natural
sciences.
171
prisons create more criminals than they reform. The mad are confined in Hôpitals Généraux in
the seventeenth century; and yet there is nothing medical about the practice and institution of
confinement. The sick body, which for a long time presents nothing but unintelligible opacity to
the observing eye, first becomes the site of an immediate articulation of what can be seen and
said about disease, and then the hidden source of what is most intelligible about disease; we then
look for the truth about life in the immobility of a corpse.
We could insist, in the face of all this, that ―the life-world is…the transcendental site
where speaker and listener meet,‖198 in order then to refer all of our meaningful interactions to
the forms of intersubjectivity of possible understanding. That is our prerogative. And it is not
unreasonable to highlight the advantages which accrue to grounding the objective and social
world on the process of argumentation in which we raise claims, criticize one another, and seek
agreement. But inscribing all that in a quasi-transcendental framework to underwrite universality
and necessity, though it may propitiate our epistemic and practical anxieties, ultimately
misrecognizes the source of our conflicts and the potential for their resolution. It enables us to
construct a false image of our society, and thereby rationalize our practices, but we continue be
locked in the repeated compulsion of social antagonisms. Referring the latter to the result of
merely external forces through a distinction between what conditions in fact and what regulates
in principle is not a work of shedding light but of occlusion.
Enlightenment of good reasons becomes blackmail if we are forced to choose between
the de jure validity of communication free from distortion and the de facto entanglements in
198 TCA 2, p. 126, which continues : ―…where they can reciprocally raise claims that their utterances fit the world
(objective, social, or subjective), and where they can critize and confirm those validity claims….In a sentence:
participants cannot assume in actu the same distance in relation to language and culture as in relation to the totality
of facts, norms, or experiences concerning which mutual understanding is possible.‖
172
power relations. That is a false dilemma, and the two kinds of practices should be grasped in
their reciprocal implication. When Foucault refers to practice as the place ―where what is said
and what is done, rules imposed and reasons given, the planned and the taken for granted meet
and interconnect,‖ (EW, 3:225) what appears at first blush as a confused amalgamation becomes
intelligible in contrast with the life-world as the transcendental place safeguarding subjects‘
reciprocal recognition. It is in fact not possible to fully articulate conceptually the relation of
mutual implication which holds between power and knowledge. That impossibility, however, is
located not in a transcendental necessity but in the very movement Foucault traces in his
histories. We could study the space of reasons in abstraction from the space of confinement;
nothing in principle prevents it. But we thereby misrecognize our history and actuality.
3.22 Derrida and What We Mean When We Say ―Lock Them Up!‖
The exemplary formulation of what I take to be the second possible challenge to experience as
the spatially and temporally indexed, rule-governed practices criterial for a group of people is
found in Derrida.199 It stands in a curious relation of symmetry to the first one. Here the charge
199 Derrida‘s explicit engagement with Foucault occurs in the forceful criticism, in 1963, of Foucault‘s claim, in the
1961 Preface to the History of Madness to do an archaeology of a silence by letting the mad speak for themselves.
(See ―Cogito and the History of Madness,‖ in Writing and Difference (Chicago: The University of Chicago Press,
1978) pp. 31-63. The stakes of that confrontation between the two have a convoluted history and at times dissolve in
its vitriol. I will steer clear of that particular confrontation, except to note that the animosity of Foucault‘s response,
which is undeniable, seems to have led to the prevalent view according to which there is nothing in Derrida‘s text to
merit such a violent response; the implication of that text, however, is that Foucault the historicist is not capable of
grasping the transcendental dimension of Descartes‘s cogito, thereby reproducing the naïve objections of ordinary
consciousness against the eidetic certainties of the philosopher. My discussion will rather focus on arguments
Derrida deploys against speech act theory, which I take to be more pertinent to the account I have constructed.
Moreover, my representation of speech acts will contain only as much as is required to recast the account of
experience so far developed in that terminology. Those familiar with the texts will realize that there are elements
common to both of those engagements, but in order to focus my argument I will not explicitly deal with them. The
language and style of Derrida‘s argumentation pose interpretive problems which I could bypass in my reconstruction
of Habermas‘s position. I will not further qualify and justify here the strategy I follow in my deliberation: that can
only be done in the course of actually carrying it out. But perhaps that is nothing more nor less than Derrida‘s
173
would be not that Foucault‘s refusal of categorial separation leads to an indeterminate conflation
of different types of practices, but that his reference to rules as such in characterizing the criterial
level results in too much determination. Or, in any case, more determination than can be
maintained. But that impossibility, not unlike the fundamental thrust of Habermas‘s argument, is
referred not to the contingency of historical practices, but to a quasi-transcendental level; but
unlike Habermas‘s account, this level involves conditions of possibility that are also conditions
of impossibility.200
The terms in which I constructed Foucault‘s description of criterial practices may be
recast in the mold of speech-act analysis (with a number of modifications, which will be clear
only after the presentation of Derrida‘s argument).201 The first important innovation of a speechtheoretical
analysis of discursive practices is the acknowledgement that not everything
cognitively and practically interesting about statements can be grasped by an exclusive attention
to their descriptive function. Logical analysis essentializes what is only one function of
utterances among many others: it privileges those that are constative, which claim to represent
reality and are evaluated as true or false, at the expense of those that are performative, which are
not descriptions of reality but the performance of an action. Because explicit performatives,
which fit the pattern of ―in saying ‗…,‘ I thereby…,‖ they are a part of the very reality they
purport to describe. Therefore truth-functional analysis does not capture what is specific to their
resolution of the predicament according to which one ―must break the glass, or better the mirror, the reflection, his
infinite speculation….And start to speak‖ (―Cogito and the History of Madness,‖ op. cit. p. 32).
200 My reconstruction in what follows is based on Jacques Derrida, Limited Inc (Evanston: Northwestern University
Press, 1988) (hereafter LI); Writing and Difference, op. cit. (hereafter WD); Of Grammatology (Baltimore: The
Johns Hopkins University Press, 1974) (hereafter OG). On Derrida‘s use of the term ―quasi-transcendental ,― see
Geoffrey Bennington, Derrida (Editions du Seuil, 2008) pp. 223-236, especially p. 229f against ―historicist‖
readings of this problematic notion.
201 My reformulation of what I am referring to as the criterial level in terms of speech-act analysis draws on J. L.
Austin, How to Do Things with Words (Oxford University Press, 1962) (hereafter HW); John R. Searl, Speech Acts:
An Essay in the Philosophy of Language (Cambridge University Press, 1969) (hereafter SA).
174
function. And because they are in some sense constitutive of what they describe, there can be no
absolute distinction between these utterances and the situations in which they occur.
Hence they are more like events and actions—keeping in mind what I said above in
relation to reasons and causes—than representations. The proper evaluative categories would
then be success or failure; or to use a slightly modified version of Austin‘s wonderful expression,
they may be happy or unhappy. Moreover, since constative utterances typically use expressions
such as ―stating that,‖ and ―arguing that,‖ it is possible to rearticulate them in terms of
argumentative performatives. The distinction between the saying of something and its doing is
thereby subsumed under the category of speech acts. We may then assimilate Foucault‘s claims
about the énoncé to the framework of speech-act analysis.202And the implication most pertinent
in this context is that speech acts so construed are indexed to their context, i.e. the object is
neither the proposition, nor the grammatical sentence, but the utterance in speech-situation.
Finally, this utterance-in-situation may be analyzed along three dimensions: locutionary (its
semantic and referential values), illocutionary (the kind of act effected in saying the utterance),
and perlocutionary (the effects produced by the utterance). And again, the levels of analysis
admitted by Foucault may be more or less unproblematically assimilated to this division.
Now, if we pursue an analysis of speech-acts by referring to the context of utterance, and
in particular, when we evaluate the illocutionary force of utterances, we must invoke the
conditions governing their success/happiness. And since the issuing of utterances is a social
activity, these conditions are articulated in institutional conventions. If, for instance, the
chairman of the jury says ―We find the defendant guilty,‖ this will amount to the issuing of a
202 For the discussion of énoncé in relation to the historical a priori, see Chapter 2 of the present study.
175
successful/happy speech act only if the institutional framework, with its system of rules defining
offices, penalties, acceptable formulations, etc., is already in place: only then does the utterance
thereby make the defendant guilty. If I roam around finding people guilty, this will at best be a
disturbance of the air by sound waves, and at worst a disturbance of public order eliciting a call
to 911. The upshot is that a system of rules structures the activity by defining the conventions on
the basis of which an utterance gets its illocutionary force. To be effective, these rules must be
public and definite. But in such a way that my actual knowledge of them becomes a moot point:
if the defendant claims that he was not aware of the jury‘s power to find him guilty, this will not
cancel the punishment consequent precisely on that find.
Derrida‘s argument targets, among other things, the very possibility of there being rules
or conventions which could have this constitutive role without remainder, where ―without
remainder‖ means exhaustive determination of the meaning, value and force of the practices at
issue. The immediate pretext of the argument is Derrida‘s reading of Austin‘s characterization of
speech acts performed by actors on stage or those found in a poem as hollow and etiolated.203
Speech acts performed in ―fiction‖ are peculiar for Austin because, even though they do not
necessarily fail on account of not fulfilling the conditions defining their success, they do not
quite succeed either. So the conditions of success/happiness may still hold, but only in a different
way.
203 The passage from Austin is: ―a performative utterance will, for example, be in a peculiar way hollow or void if
said by an actor on the stage, or if introduced in a poem, or spoken in a soliloquy. This applies in a similar way to
any and every utterance—a sea-change in special circumstances. Language in such circumstances is in special
ways…used not seriously, but in ways parasitic upon its normal use…‖ (HW, 22). The passage is taken up in LI, p.
16ff. I say ―immediate pretext,‖ for Derrida considers other passages as well, but this one retains a certain key role
in how he structures his argument in ―Signature Event Context‖.
176
Austin ascribes the possibility of hollow speech acts to their intrinsic repeatability:
utterances can be repeated beyond the conditions specifying their normal contexts. This
difference is then determined as categorial by Searle, who appeals to intentions of speakers in
order to provide theoretical grounding for the distinction in terms of serious as opposed to nonserious
speech acts.204 Since it could be said that the speech acts analyzed by Foucault are
serious, or rather, since his talk of systems of rules seem to treat all speech acts as if they were
serious, Derrida‘s argument could undermine the very possibility of describing the historical a
priori criterial for a group.
In Derrida‘s reading, the fundamental thrust of making a categorial distinction between
serious and non-serious speech acts implies that the former are taken as normal and the latter as
parasitic on what is normal.205 Thus all copies, quotations, or citations of original performances
would only mimic what they repeat. Derrida points out that, despite this distinction, the very
determination of speech acts as conventional presupposes that they are repeatable. Speech acts
governed by conventions are like rituals and ceremonies in that for a performance to fulfill its
conditions of success/happiness—and hence to be the specific speech act that it is—it must be
the repetition of an already accepted performance. What is determined exceptional as citation is
the general citationality presupposed by all performatives. This general structure of citationality,
or iterability, is then the necessary condition of the very possibility of speech acts capable of
204 For the ―anchoring‖ through intentions in relation to sincere/insincere promises, see SA, p. 60ff; in relation to
reference in real vs. fictional universes of discourse, p. 78ff; in relation to serious vs. non-serious speech acts with
respect to the commitment of the speaker to their consequences, see pp. 189f and 197. Searle‘s argument is more
complex than simply invoking intentions as justificatory bedrock, but it is also beyond the scope of what I need for
my purposes here.
205 ―Austin thus excludes, along with what he calls a ―sea-change,‖ the ―non-serious,‖ ―parasitism‖…all of which he
nevertheless recognizes as the possibility available to every act of utterance….[I]s this general possibility
necessarily one of a failure or trap into which language may fall…? …For ultimately, isn‘t it true that what Austin
excludes as anomaly, exception, ―non-serious,‖ citation…is the determined modification of a general
citationality…without which there would not even be a ―successful‖ performative?‖ LI, pp. 16-17.
177
identification as such. Therefore, that non-serious speech acts mimic serious ones cannot be the
reason for their vacuity: iterability characterizes all speech acts. And Derrida extends the scope
of the argument by placing any sign or mark in this logic of iteration.206
A number of conclusions follow from this line of reasoning. First, if any sign, in order to
be identified as a sign, must be capable of repetition in principle, then its functioning does not
depend on any particular speaker or hearer. Even if it occurs only ever once, that it must be
iterable attaches to the very possibility of that occurrence as the occurrence of a sign. Therefore,
its value is independent of the intentions of any particular speakers. Second, if a sign is
necessarily iterable, then its identification as the same as another sign simultaneously implies its
internal difference from itself. It can occur in another time and place as the same sign only
because it is not identical with itself in the first place. Even if we were to posit an original sign,
the first one of its kind, it can function as a sign only because it already looks the same as its
possible future repetitions. Finally, because the sign is what it is only on the basis of its
difference from itself, it can never be exhaustively inscribed in and determined by a particular
context.
The force of this last claim should be understood at the proper level of its generality: if
the sign is iterable in principle, it can be removed from any particular context and reinscribed in
206 ―[The] force of rupture is tied to the spacing [espacement] that constitutes the written sign: spacing which
separates it from other elements of the internal contextual chain (the always open possibility of its disengagement
and graft), but also from all forms of present reference (whether past or future in the modified form of the present
that is past or to come), objective or subjective….[Is this force of rupture/a force that breaks with its context]
limited…strictly to ‗written‘ communication in the narrow sense of this word? [Is it] not to be found in all
language…and ultimately in the totality of ‗experience‘ insofar as it is inseparable from this field of the mark…of
units of iterability, which are separable from their internal and external context, and also from themselves, in as
much as the very iterability which constituted their very identity does not permit them to ever be a unity that is
identical to itself?‖ (LI, pp. 9-10).
178
another.207 Therefore, even though we only ever find signs in context, no context could be the
original, and therefore final, determination of the value of a sign. And if no linguistic code or
social situation could exhaustively determine the meaning and value of a sign, then the historical
a priori would be constitutively different from itself, to the extent to which traces of other
contexts attach to its status of being the ultimate context of all contexts. And the limits built into
its very concept as criterial would then burst open, not because something from the outside
would undermine its coherence, but because it is internally different from itself.208
I want to advance two responses to this challenge, both necessarily oblique, but I think no
less effective: a concession and a counter-challenge.
First the concession. At times Foucault writes as if what is criterial at a given time and
place could be totalized into an ordered system. The paradigmatic formulation of this is: ―In any
207 LI, passim., but in particular, p. 57: ―Once again…: what is at stake here is an analysis that can account for
structural possibilities. Once it is possible for X to function under certain conditions (for instance, a mark in the
absence or partial absence of intention), the possibility of a certain non-presence or of a certain non-actuality
pertains to the structure of the functioning under consideration, and pertains to it necessarily….[I]t can happen that a
mark can function without the sender‘s intention being actualized, fulfilled, and present, and which to this extent
must be presumed. Even if this (eventual) possibility only occurred once, and never again, we would still have to
account for that one time and analyze whatever it is in the structural functioning of the mark that renders such an
event possible….That is possibility qua eventuality. It might, however, also be said: in fact, that doesn‘t always
happen like that. But at this point, we must pass to possibility qua necessity…and moreover, we must recognize an
irreducible contamination or parasitism between the two possibilities and say: ‗to one degree or another that always
happens, necessarily, like that‘: by virtue of the iterability which, in every case, forms the structure of the
mark…preventing [intention] from being fully present to itself in the actuality of its aim or of its meaning (i.e. what
it means-to-say [vouloir-dire].‖
208 This is also, not incidentally, one of the elements of Derrida‘s argument throughout ―Cogito and the History of
Madness,‖ op. cit. It is possible to find a modulated version of it in ―Structure, Sign, and Play in the Discourse of the
Human Sciences,‖ WD, pp. 278-293, where the important point to note is the way Derrida avoids privileging either
―interpretation of interpretation,‖ analogous to his insistence on the necessary contamination between two types of
possibility in the previous quotation from LI: ―Totalization can be judged impossible in the classical style: one then
refers to the empirical endeavor of either a subject or a finite richness which it can never master. There is too much,
more than one can say. But nontotalization can also be determined in another way: no longer from the standpoint of
a concept of finitude as relegation to the empirical, but from the standpoint of a concept of play….[B]ecause the
nature of the field—that is, language and a finite language—excludes totalization….[I]nstead of being an
inexhaustible field…there is something missing from it: a center which arrests and grounds the play of
substitutions‖; ―For my part, I do not believe that today there is any question of choosing…‖ (WD, pp. 289 and 293,
respectively).
179
given culture and at any given moment, there is always only one episteme that defines the
conditions of possibility of all knowledge [savoir], whether expressed in a theory or silently
invested in a practice‖ (OT, 168). This is untenable, but it is also not presupposed by the
reconstruction I have offered. The historical a priori is not, or should not be understood on the
model of, either formal codes articulating the totality of differential relations between its
linguistic elements, or the system of explicitly formulated rules publicly available in principle for
our consent in order then to become efficacious. That it is not a linguistic system of differential
elements should be clear from what has been said so far. The latter possibility, however, needs
further clarification.
The discursive and nondiscursive practices criterial for what counts as falling under a
concept need not be interpreted on the model of explicit rules that define the practice of, say,
playing backgammon, because the sense of conventionality involved in the latter is not the same
as that which is effective in the former. The rules of backgammon are constitutive because they
define what it is to play that game. They may be collected in a rule book, and if there is any
doubt concerning a particular move, it is possible in principle to settle disputes by appealing to it.
And the sense of conventionality still retains its sense of that which derives from a coming
together or assembly: playing the game presupposes explicit or implicit recognition of the rules‘
binding force.
So we could say that, to bring the example in line with the reconstruction of Derrida‘s
argument, the performances possible within the game presuppose the existence of the institution
of the game: there must already be fully determinate rules defining legitimate roles, available
moves, efficient strategies, etc. Therefore if the historical a priori were conceived on the model
of conventionality operative in games of this type, then it would be possible to ascribe to
180
Foucault a theory of institutions as systematic totalities of constitutive rules. But the criterial
level is not susceptible to a similar analysis. There was no convention, in this sense, which
determined that prisons should be where we lock up the delinquent, or that asylums be the places
where we should treat the mad. There could be no agreement grounding the appearance of these
practices as normative. Hence the determination of which criteria are capable cannot be
assimilated to that of contractually binding rules, which can be written down and offered for
explicit consent. In line with the mode of analysis I described in relation to the History of
Madness, we could say that that social relations come to be perceived on the model of
contractual relations itself presupposes historically transformable criteria both governing and
immanent in the practices to which Foucault refers. In that sense even institutions are intelligible
against the background of criteria. So, for instance, the confinement of the figures of unreason
and the way it restructures the social space is intelligible only in relation to the experience of
madness Foucault describes.
Yet the rules with which Foucault is concerned are not only descriptions of regularities of
behavior in a given context. They govern perceptions, actions, and statements without, however,
being explicit prescriptions or causal connections. This partially explains why there can be no
empirical verification of Foucault‘s claims, since the rules he seeks to describe are those which
govern what counts as empirical verification.209The necessity at issue, however, is neither eidetic
nor transcendental. We may no longer want to call this a rule—in which case historical a priori,
or episteme, or conditions of acceptability, or dispositif, or criteria will do. The crucial point, by
any other name, is that they presuppose neither anchoring in intentions of speakers nor
integration in a totality of relations.
209 I describe why this is so in the second part of Chapter 2.
181
But all of this goes only so far, since Foucault‘s historical analysis still implies putting
the text in context. Derrida‘s challenge should not be confused with the view according to which
statements could be true or false, or speech acts happy or unhappy, independently of any context.
The upshot of the claim is rather that there is no absolute, foundational context; and that would
extent equally to contexts absolute for a given group at a given time and place. And this is very
much what the historical a priori looks like. So, for instance, the meaning and value of
Descartes‘s cogito cannot be specified and exhausted by any determinate historical context, even
though Descartes is very much of his time. What follows is that even though there are
institutions, i.e. finite systems of rules constitutive of practices, the internal difference of any
system with itself makes full determination impossible in principle. In that sense, that which
makes any system possible and its full closure impossible exceeds every context. Foucault, then,
would fail to capture precisely the status of this principle, which is referred to the quasitranscendental
status of citationality, or iteration: ―ultimately [en dernière instance] there is
always a police and a tribunal ready to intervene each time that a rule [constitutive or regulative,
vertical or not] is invoked in a case involving signatures, events, or contexts…. If the police is
always waiting in the wings, it is because conventions are by essence violable and precarious.‖210
However—and this is the counter-challenge—there are practices which are compulsory
without necessarily implying full determination. The force of discursive and nondiscursive
practices that is salient for Foucault is not that which concerns justification of performances by
invoking an institutional framework. That is what the reduction of validity and rationality is
supposed to make visible. Nor is it modeled on the brute effectivity of what is physical. But only
by appealing to a system of rules which specify the conditions of success of speech acts—part of
210 LI, p. 105.
182
which involves institutional elements—can we do justice to speech acts as events capable of
creating or defining new types of behavior and new modes of perception. Against foundationalist
attempts to ground our practices in an exhaustive system of reflectively articulated rules, it may
be helpful to articulate quasi-transcendental ―conditions of their impossibility‖; but Foucault‘s
appeals to historical a priori are not motivated by foundationalist ambitions, not even historically
modulated ones. He is not interested in grounding meaning and value on some putative presence,
but in describing how meaning and value emerge from historically situated practices.211
And there is meaning and value calling for such description, to the extent that what
counts as acceptable scientific statement, or right type of treatment, change historically.212 And
that change can only be understood contextually. The concepts which Derrida deploys, such as
citationality, shakes foundationalist pretensions by establishing the in principle impossibility of
capturing unreason in the conventional structure of any institution; but they pass over the fact
that, in a given time and place, only some people are perceived as in need of locking up, and are
then locked up. Or rather, they treat this latter fact as only secondary and derivative relative to
the quasi-transcendental structures making it both possible and impossible. And the Foucauldian
211 That is why experience, as I want to use it, does not imply Derrida‘s remarks in relation to its concept when he
says : ―[T]he notion of experience, even when one would like to use it to destroy metaphysics or speculation,
continues to be, in one or another point of its functioning, fundamentally inscribed within onto-theology: at least by
the value of presence, whose implication it can never reduce by itself. Experience is always a relationship with a
plenitude, whether it be sensory simplicity or the infinite presence of God‖ (OG, p. 283). Or again, ―For
[empiricism] at bottom, has ever committed but one fault: the fault of presenting itself as a philosophy. And the
profundity of the empiricist intention must be recognized beneath the naïveté of certain of its historical expressions.
It is the dream of a purely heterological thought at its source. A pure thought of pure difference. Empiricism is its
philosophical name, its metaphysical pretention or modesty. We say the dream because it must vanish at daybreak,
as soon as language awakens‖ (WD, p. 151).
212 The risk of ―talking past one another‖ is perhaps greatest at this point, for Derrida anticipates a move similar to
what I insist on here: ―For it might be said:…[y]ou cannot deny that there are also performatives that succeed, one
has to account for them: meetings are called to order…people say: ‗I pose a question‖; they bet, challenge, christen
ships, and sometimes even marry. It would seem that such events have occurred….I‘ll answer: ‗Perhaps‘. We should
first be clear on what constitutes the status of ‗occurrence‘ or the eventhood of an event that entails in its allegedly
present and singular emergence the intervention of an utterance…‖ (LI, p. 17), where ―utterance‖ brings back the
structure of citationality or the logic of iteration. The ―to the extent that‖ in my claim marks my anticipation of this
anticipation, and the beginning of its explication.
183
rejoinder is that such structures are too far removed from the density of historical practices to
understand their specificity, and thereby advance a critical interrogation of our actuality.
The symmetry between the challenges presented by Derrida and Habermas results from
their insistence that the transcendental standpoint must have primacy over the empirical one by
right. Hence the necessarily oblique nature of the responses I advanced: I cannot place myself
outside the space of reasons without thereby impugning my rights, since every right presupposes
that one is always already placed inside reason‘s space. From this perspective, the space of
confinement Foucault describes can only appear as a self-defeating criticism of reason in the
name of a better reason. And since the figure of that ―better reason,‖ in Foucault‘s story, appears
as madness itself, beyond the institutionalizing work of reason, Foucault‘s refusal to offer
explicit justification of the standpoint from which that history can be written, or his refusal to
refer the structure of exclusion to a quasi-transcendental (non-)principle of iteration, necessarily
appears as either naivety or willful rebellion.
That necessity is what motivates the insistence of Habermas and Derrida. Against this
perspective, I insist that Foucault‘s refusal is also motivated: the history of madness is
simultaneously the history of reason. Therefore, if the genesis of the space of reasons is
inextricably bound up with the spaces of confinement, the history of that genesis cannot be
written from the transcendental standpoint alone. The genitive in ―history of madness‖ should be
understood both subjectively and objectively. But the subjective sense refers not to madness
itself which would finally speak its primitive purity, but rather to the flipside of the limits
criterial for our own space of reasons. If the description of that entanglement necessarily appears
as both subjective and objective, that necessity is inscribed not in eidetic or transcendental
structures but is the result of historically situated antagonistic relations. Therefore, Foucault‘s
184
articulation of experience as the matrix of context-bound, but no less constitutive, rules of
formation is more attuned to the historical singularity of events and the critical interrogation of
their conditions of emergence.
However, the force of the challenge of Derrida and Habermas implies that Foucault must
say more about two related problems: 1) Could one give up the transcendental, or rather the
space of reasons transcendentally grounded? What justification could there be for that
renunciation? How does one frame the recoil consequent on that abandonment? 2) Why call that
experience? The response (but probably not the solution) to both problems demands that one say
more about who or what could count as the subject of this experience.
185
…it is not by wishing to escape it that one goes
beyond reality. (MIP, 84)
My name is “L. W.”And if someone were to dispute
it, I should straightaway make connections with
innumerable things which make it certain.
(Wittgenstein, On Certainty)
4. Experience of Freedom: Speaking for Oneself.
If I am correct in claiming that the aim of Foucault‘s investigations is to bring to light the set of
historically specific practices criterial for what counts as knowledge of objects and the proper
treatment of subjects, there still remains the question of whether (and why) that should (or could)
be called ―experience‖. The question becomes more urgent since the works in which Foucault
offers his most explicit definitions of experience are volumes two and three of The History of
Sexuality, where what is at issue is ―a history of the experience of sexuality,‖ (UP, 4) but where
almost everything Foucault says concerns what people have said about sexuality. Compared
with, for instance, Merleau-Ponty‘s acute and fine-grained descriptions of the lived experience of
sexuality, Foucault is particularly vulnerable to criticism if it turns out that he is privileging
discursive practices and, as a result, never saying anything except repeating what other people
said about everything. Therefore, his account of subjectivity, and how it fits into his accounts of
discursive and nondiscursive practices, needs further specification.
The recourse to subjectivity is also necessary in order to evaluate the epistemic and
normative status of the histories Foucault writes. In chapters two and three I raised the question
of whether it might be possible to formulate a discourse at once historical and critical without
reproducing the incoherent oscillations characteristic of the analytic of finitude; and I suggested
that a plausible way to articulate that possibility would be to give up transcendental reflection.
186
But my strategy so far has been negative in that, I have insisted on the motivation behind that
refusal by comparing it favorably with some other alternatives I consider to be particularly
strong, and I displayed the intelligibility of the kind of historical discourse consequent on that
refusal by offering a reading of the History of Madness. But a more direct engagement with
subjectivity is required in order to explicate the critical import of Foucault‘s discourse to the
extent to which, in modernity, criticism is motivated by, carried out in the name and grounded on
the capacities of the subject, the fundamental one being its spontaneity. The question, bluntly, is:
Am I free to give up the kind of reflection constitutive of modernity without thereby rendering
freedom itself meaningless and impossible?
I propose to formulate my answers to these two questions through direct engagement
with the thought of two thinkers, Merleau-Ponty and Adorno. A critical encounter with Merleau-
Ponty is motivated by two main reasons: First, in the Phenomenology of Perception he
thematizes the experience of sexuality in a section titled ―The Body in its Sexual Being,‖ and the
preface to the same work offers a succinct discussion of his appropriation of phenomenology.213
Second, his influence on the early Foucault is unmistakable, both in terminology and
methodology. On the other hand, Adorno, probably more than any other philosopher in the
twentieth century—with the possible exception of Derrida—has been sensitive to the recoil of
critique on the critic and lodged his thinking in the very space of that complicity between who
criticizes and what is criticized. A critical encounter with a significant moment of his work,
namely, the reciprocal mediation of subject and object, enables a sharper definition of whether
Foucault can claim any sort of legitimacy and efficacy for his practice of history.
213 Maurice Merleau-Ponty, Phenomenology of Perception, (New York: Routledge, 1962) pp. vii-xxi and pp. 154-
174 (hereafter PP). My reference to Adorno will draw on a number works, which I cite in the second part of this
chapter.
187
A word of caution is in order: my interest in these comparisons is not to establish
structural similarities and differences. The similarities should not be overstated, since it is not
surprising that anti-foundationalist strategies, which refuse the syntheses of both idealism and
positivism, should share some premises and conclusions; whereas the differences usually amount
to little more than showing that one thinker is not the other, and vice versa. I emphasize those
claims in both thinkers which, despite their proximity to Foucault, could be construed as
challenges to the account I have proposed so far, and for that very reason can sharpen the
contrast and clarify the image I want to define. Since (at least) Hegel, the devil is truly in the
detail, and I think closer attention helps advance the critical dimension of Foucault‘s account of
experience and subjectivity.
4.1 Talking Sex
There is good reason to claim that Merleau-Ponty is the philosopher who strove the most to do
justice to corporeal existence, that is, to the human being as an embodied being. Not only are his
phenomenological descriptions meticulous, but also he eschews mere polemic in favor of a
judicious incorporation of opposing viewpoints in trying to understand how corporeality can be
the mute basis of both the subjective and the objective aspects to reality. But for that very reason,
his language is slippery, no sooner offering than withdrawing, asserting and denying. Since that
movement is the very richness of his descriptions, and since it cannot be separated from their
content and validity claim, I propose to do a close reading of two sections of Phenomenology of
Perception: ―Preface‖ and ―The Body in its Sexual Being‖.
The former is a very concise articulation of his creative appropriation of key Husserlian
claims, and thus it sheds light on some of the methodological issues which I have introduced in
188
Foucault‘s relation to phenomenology and existentialism. The latter is a thematic investigation of
sexuality, and it thus offers, almost in strict counterpoint to that of Foucault, one way in which
sexuality as experience can be understood. It therefore raises the question of why Foucault‘s
alternative account could or should be called experience.
4.11 The Body and its Sexuality
The thought I want to understand is the following, and it is worth quoting at length:
All that we are, we are on the basis of a de facto situation which we appropriate to ourselves and
which we ceaselessly transform by a sort of escape which is never an unconditioned freedom.
There is no explanation of sexuality which reduces it to anything other than itself, for it is already
something other than itself…our whole being. Sexuality, it is said, is dramatic because we
commit our whole personal life to it. But just why do we do this? Why is our body, for us, the
mirror of our being, unless because it is a natural self, a current of given existence, with the result
that we never know whether the forces which bear us on are its or ours—or…rather that they are
never entirely either its or ours. (PP, 171)
The central problem of the passage is what could account for our investment in sexuality, why it
seems to have such a strong hold on our everyday engagements and theoretical reflections. One
reason is presented and dismissed: We commit our whole personal life to it; it is something we
do which accounts for the fact that sexuality becomes the scene where actions and passions are
enacted and which arouses pity and terror. This, however, cannot be a sufficient reason since it
does not explain just why we should do this. Unless something in sexuality itself elicited this
response by soliciting our attention, our investment in and anxiety in the face of sexuality would
be no different from those which are generated by my inability to solve, say, a mathematical
equation. But sexuality seems to permeate my entire existence, so much so that my dread (or its
189
absence) before a mathematical equation is itself to be referred to the structures of my sexual
existence.
If my gloss on the question Merleau-Ponty raises is accurate, then his answer is
intelligible only in terms of his reinscription of Husserlian themes in an existential register. In the
Preface he provides a succinct presentation of these themes. Without doing too much injustice to
his language, it is possible to capture them under five headings that articulate how one should
understand the phenomenological recasting of: 1) essences, 2) reduction, 3) the role of
description in relation to explanation, 4) the ―true Cogito‖, 5) intentionality.
The goal of phenomenological description which governs the ―project‖ is ―to elucidate
the primary function whereby we bring into existence, for ourselves, or take a hold upon, space,
the object or instrument, and to describe the body as the place where this appropriation
occurs‖(PP, 154). But since this work of phenomenological ―elucidation‖ is typically understood
as bringing essences to light, and since Husserl is typically charged with having severed the
connection between essence and existence, Merleau-Ponty first undermines that objection by
claiming that, contrary to common understanding, phenomenology ―puts essences back into
existence‖ (PP, vii). This involves understanding both the human being and the being of the
world in terms of their facticity. However, since we do not go about defining essences in our
everyday activities, one must pass from the fact of existence to its essence. The putative
contradiction entailed by this movement is then resolved through a reformulation of the
phenomenological reduction.
According to Merleau-Ponty‘s interpretation, the reduction is transcendental and eidetic.
It does lead to Wesenschau, the intuition of essences, by abstracting from the facticity of what is
190
given to consciousness, but this is only a necessary step in order to make existence a theme.
Directly relevant to my purposes is the way in which Merleau-Ponty seems to interpret the
objection with respect to the separation of essence from existence as entailing reduction to
discourse. For he says, by way of discounting it, that seeking the essence of consciousness is not
escaping into ―the universe of things said‖ (PP, xv): ―Whatever the subtle changes of meaning
which have ultimately brought us, as a linguistic acquisition, the word and concept of
consciousness, we enjoy direct access to what it designates…. For we have the experience of
ourselves…and it is on the basis of this experience that all linguistic connotations are assessed‖
(Ibid.).
Moreover, he explicitly denies that the reduction is a reduction to transcendental
consciousness. Since existence is not the same as awareness of existing, it cannot be reduced to
the series of apperceptive syntheses which I could reconstruct retrospectively. Therefore the
distinction between the transcendental unity of apperception, as the formal unity of
consciousness presupposed by determinate experience of objects, on the one hand, and the
empirical self to which everything else about the individual is to be referred, on the other, is
untenable. On the contrary, both questions of fact and of validity are to be referred to our ―direct
and primitive contact with the world‖ (PP, vii). Reduction is simply the ―slackening [of] the
intentional thread‖ (PP, xiii) whereby I suspend my theoretical commitment to truth and naïve
belief in the independent existence of the world214.
214 Merleau-Ponty refers to Eugen Fink‘s formulation of the reduction as wonder in the face of the world. He
therefore appropriates the reduction as precisely the gesture through which the world is revealed as strange:
―Reflection does not withdraw from the world towards the unity of consciousness as the basis of the world; it steps
back to watch the forms of transcendence fly up like sparks from a fire‖ (Ibid). He also claims that therein lies the
difference between the Husserlian and the Kantian transcendental. The latter is charged with reducing the world to
the immanence of the subject—for the very reason that it ―makes use of our relation to the world‖—whereas the
191
Therefore, what is philosophically interesting cannot be explained on the causal model
prevalent in natural scientific explanations. Neither psychology nor sociology can provide an
explanation of individual and collective generation of meaningful experience, since science itself
is possible on the basis of the world as it is directly experienced. Explanation is to be replaced by
description, for ―I am not the outcome or the meeting-point of numerous causal agencies which
determine by bodily or psychological make-up. I cannot conceive myself as nothing but a bit of
the world, a mere object of biological, psychological or sociological investigation‖ (PP, viii).
Scientific explanations can account for everything except their own status and meaning,
and the latter can only be provided by referring science itself to the matrix of all meaning and
value, namely, the life-world as the antepredicative, or pre-reflective, context prior to the
reflective division between subject and object. Relative to the life-world, science is a secondorder
abstraction ―as is geography in relation to the country side in which we have learnt
beforehand what a forest, a prairie or a river is‖ (PP, ix). Thus ―I am the absolute source‖ since
my existence cannot be explained as exhaustively determined by an antecedent causal chain,
physical or social (PP, viii-ix).
This invocation of an absolute source, however, is to be distinguished from the idealist
appeal to subjectivity as first principle. According to Merleau-Ponty, idealism misconstrues
actual experience when it seeks to find its conditions of possibility: since conditions of
possibility are distinct from what they condition, and since experience on this interpretation is
nothing other than the unity of my representations, we only ever get reconstructions. The subject
of this experience is the formal unity of consciousness providing the syntheses which give
former shows ―the unmotivated upsurge of the world‖—precisely because it suspends our familiar acceptance of the
world. (PP, xiii-xiv)
192
coherence to representations. In contradistinction to this reconstruction, phenomenological lived
experience remains ―within the object‖.
Since reflection always starts from and presupposes an unreflective experience, what is
needed is not construction but description of the antepredicative unity of meaning. Moreover,
perception is not a synthesis or a predication, the model for which is provided by determinate
judgment. Therefore what descriptions reveal behind reflective thematizations are not the
constituting acts of consciousness, but rather ―my actual presence to myself‖: ―In the silence of
primary consciousness can be seen appearing not only what words mean, but also what things
mean: the core of primary meaning round which the acts of naming and expression take place‖
(PP, xv).
Finally, Merleau-Ponty‘s recasting of Husserlian themes in an existential register defines
the chief achievement of phenomenology as its having united ―extreme subjectivism and extreme
objectivism in its notion of the world or of rationality‖ (PP, xix)215. The key concept through
which the subjective and objective poles are related is that of intentionality. He distinguishes
between two types (PP, xviii): intentionality of act—which characterizes judgments and
voluntary taking up of positions216—and operative intentionality: ―that which produces the
natural and antepredicative unity of the world and of our life, being apparent in our desires, our
evaluations and in the landscape we see, more clearly than objective knowledge, and furnishing
the text which our knowledge tries to translate into precise language‖ (PP, xviii). It is not a
question, then, of formulating objective laws, but ―a certain way of patterning the world‖ (Ibid).
215 I discussed the prominence of this conjunction in Foucault in the first part of chapter three.
216 Merleau-Ponty‘s gloss on this is: ―the only intentionality discussed in the Critique of Pure Reason‖ (Ibid.)
193
Earlier he had asserted that perception is ―not…an act, a deliberate taking up of a
position,‖ (PP, x-xi) but rather the background from which all acts stand out. In that context he
even uses the spatial term ―field‖ in order to characterize this prereflective background
presupposed by all reflective positions. Now he relates it explicitly to the theme of history:
―Whether we are concerned with a thing perceived, a historical event or a doctrine, to
‗understand‘ is to take in the total intention…the unique mode of existing expressed…in all the
events of [say] a revolution…‖ (PP, xviii).
His explication of what this ―patterning,‖ or ―field,‖ involves is instructive in relation to
Foucault‘s account of experience: ―that formula which sums up some unique manner of behavior
towards others, towards Nature, time and death‖ (PP, xviii). He goes on, however, to claim that:
―Considered in the light of its fundamental dimensions, all periods of history appear as
manifestations of a single existence, or as episodes in a single drama—without our knowing
whether it has an ending. Because we are in the world, we are condemned to meaning, and we
cannot do or say anything without its acquiring a name in history‖ (PP, xix). The fundamental
dimension, then, is that of existence, the structures of which give meaning and value to each
gesture and each habit. As examples, Merleau-Ponty offers lapsing into silence due to fatigue or
using a hackneyed platitude: they may appear accidental, but they express lack of interest, and to
that extent they indicate the adoption of a definite position in relation to a situation. And therein
lie their meaning.
4.12 A Critical Ontology?
At first blush, then, these five general themes manifest many points of contact with the account
of experience I have argued to be operative in Foucault. Its conception of essence is not separate
194
from existence; therefore, it cannot be charged with (a typical) essentialism. Eidetic reduction is
supposed to lead not to transcendental subjectivity as first principle stopping the infinite regress
of reasons—in the manner of the Agrippan trilemma—but rather to the pre-reflective level
presupposed by all reflective activities. And to the extent that this level provides the field, or
fundamental structure, within which gestures and habits, words and behaviors, acquire meaning
and value, it resembles a criterial dimension. Moreover, it eschews natural scientific causal
explanations and objective laws in favor of description. Finally, the perception of which
Merleau-Ponty speaks is not the doing of this or that individual but the result of pre-personal,
universal and necessary structures217.
But a basic difference in orientation brings to light what is unique about Merleau-Ponty‘s
―experience‖. He approvingly quotes the Husserl of Cartesian Meditations: ―It is that as yet
dumb experience…which we are concerned to lead to the pure expression of its own meaning‖
(PP, xv).218 And when he defends phenomenology as ―the study of essences,‖ what he takes to be
the misunderstanding or criticism to be countered by putting essences back into existence is
precisely the reduction to discourse at the expense of what is lived through. For him the virtue of
phenomenology is precisely its avoidance of mistaking discursive significance for the whole of
significance. Accordingly, even language must be referred back to that mute but meaningful
domain with which we are always already and directly acquainted. The aim of description, once
217 Ultimately, of course, Merleau-Ponty will ―situate‖ these structures at the level of corporeal existence as the zero
degree of what is subjective and what is objective, as their point of indiscernibility. I address this difference shortly
when I compare his account of sexuality with that found in Foucault.
218 Cartesian Meditations, op. cit., p. 33
195
the necessary parentheses are put in place, is to realize ―in breaking the silence, that which
silence wished for and yet could not obtain.‖219
Hence even though Merleau-Ponty‘s rejection of a transcendentally constituting
consciousness clearly debars appeal to already constituted intentions independently of language;
and despite the fact that, by the same token, he does not claim that linguistic expression is a
matter of providing merely external cover for intention; all sense as well as nonsense are referred
to the embodied subject as the site of ―the unmotivated upsurge of the world‖: ―Thought is not in
objective time and space but it is nonetheless not without place in the phenomenological world‖
(PP, xiii). Therefore, it is possible to claim that, for Merleau-Ponty, experience is still a question
of a phenomenological silence and pre-linguistic seeing.
But if the pre-reflective level, which provides a foundation220, however incomplete and
lodged in the ambiguity of things themselves, is to be construed as anticipation of all meaning
which is brought to linguistic expression only derivatively, then the subject of this experience is
a transcendental subjectivity transposed into a hermeneutic register. The structures so described
will have the same claim to universality and necessity as the Kantian a priori, to the extent to
which cultural and historical formations, to which Merleau-Ponty is exceptionally sensitive, are
still ―grounded‖ in the subject for whom phenomenology is an essential possibility: ―We shall in
ourselves, and nowhere else, find the unity and true meaning of phenomenology‖ (PP, viii).
219 Merleau-Ponty, Le visible et l‟invisible (Editions Gallimard, 1964) (hereafter VI) See also p. 18, where he
identifies the task of philosophy as leading to their expression ―the things themselves, from the depth of their
silence.‖ All translations from this text are mine.
220 Merleau-Ponty says: ―Phenomenology, as a disclosure of the world, rests on itself, or rather provides its own
foundation‖ (PP, xx-xxi).
196
In comparison with this, when Foucault defines experience ―as the correlation between
fields of knowledge, types of normativity, and forms of subjectivity in a particular culture‖ (UP,
4), he could only be mistaking the map for the territory, just as geography is always a secondary
abstraction in relation to my embodied sense of left and right, up and down. But perhaps one
could formulate the stakes of Foucault‘s definition better through the inflection he gives to the
question, which initially bears a striking surface resemblance to the one raised by Merleau-
Ponty: ―Why has sexuality been so widely discussed…? What were the effects of power
generated by what was said? What are the links between these discourses, these effects of power,
and the pleasures that were invested by them? What knowledge (savoir) was formed as a result
of this linkage? The object, in short, is to define the regime of power-knowledge-pleasure that
sustains the discourse on human sexuality in our part of the world‖ (HS, 11).
The central problem, then, is that of understanding why we feel so invested in our
sexuality and how it appears to permeate all aspects of our activity. However, when Foucault
wants to account for the fact that sexuality is spoken about, he refers this not to our embodied
and mute contact with things themselves, but, as a ―discursive fact,‖ to ―who does the speaking,
the positions and viewpoints from which they speak, the institutions which prompt people to
speak about it and which store and distribute the things that are said‖ (Ibid.). Hence, the
background against which our reflective activities acquire meaning is not the antepredicative or
pre-reflective structures of human existence, but the ―regime‖ of power-knowledge and the
―polymorphous techniques of power‖ which it articulates. In short, the question is how my
problems have been historically formed in a context of cognitive and practical relations, and
discursive and nondiscursive practices, such that I have come to recognize my identity in my
197
sexuality. Before a final encounter with Merleau-Ponty, the methodological assumptions which
frame Foucault‘s question need further clarification.
In a late commentary on the aims of his histories, Foucault specifies that which defines
the History of Sexuality as the study of ―the constitution of the subject as its own object: the
formation of the procedures by which the subject is led to observe itself…to recognize itself as a
domain of possible knowledge. At issue…is the history of ‗subjectivity,‘ if by that word is meant
the way in which the subject experiences itself in a truth game in which it has a relation to itself‖
(CC, 316).221 This self-relation through which the subject constitutes itself is then spelled out
further as entailing ―various practices (self-examination, spiritual exercises, avowal, and
confession) to apply the game of truth and falsehood to [oneself]….‖ Such self-constitution is the
process whereby a subject comes to recognize itself as a subject of desire, or pleasure, or lust,
where this recognition at once involves insertion of the subject as an object in truth games.
The games of truth are the ―rules according to which what a subject can say about certain
things involves the question of truth or falsehood‖ (DE, 4:632); therefore when Foucault defines
sexuality as a singular experience constituted by historically analyzable practices that involve
three sets of rules, besides those which operate at the level of formation of discursive knowledge
and those which operate at the level of the exercise of power, there is a third set which
determines the forms in and through which the subject constitutes itself. But if it is true that the
two meanings of the word ―subject,‖ namely ―subject to someone else by control and
dependence, and subject tied to his own identity by a conscience or self-knowledge,‖ both
―suggest a form of power which subjugates and makes subject to [assujettit]‖ (MF, 212), then it
221 This is the entry Foucault wrote on his work under the pseudonym ―Maurice Florence‖. I refer to the translation
provided in the Cambridge Companion.
198
is difficult to see what could possibly distinguish the self-relation specific to this set of rules
from those characterizing discursive and nondiscursive practices.
I think the difficulty is in part a result of the almost decade-long period separating the
first and the second volumes of the History of Sexuality and the shift in emphasis between the
two. The first volume, the French title of which is ―La volonté de savoir,‖ belongs to what is
commonly called the genealogical period and where the emphasis is on the passivity of the
subject in relation to the systems of power and knowledge which constitute it. Hence the
language in the passage cited above: ―the subject is led to observe itself‖. The term Foucault uses
most commonly to refer to this aspect of constitution is ―assujettissement‖. But in the same
passage, Foucault connects this with self-relation and recognition, and this, it would seem,
cannot simply be a matter of ―subjection to,‖ to the extent that it strives to refer to an aspect of
constitution where some reflective activity on the part of the subject is possible, and in fact
necessary, in order to constitute oneself as a subject. The term Foucault uses, not without some
ambiguity, to capture this register is ―subjectivation‖.222Foucault‘s discussion, then, of ―forms of
subjectivity‖ as a dimension of experience must somehow capture both activity and passivity in
the way the subject makes itself an object. And ―recognition‖ is to be the name of this process
whereby individuals constitute themselves as subjects.
222 I shall use the transliteration ―subjectivation‖ when I refer to this aspect. An alternative translation is
―subjectification,‖ which could have been fine for my purposes, but ―subjectivation‖ too has entered common usage
in discussions of Foucault, and it helps me to better draw the distinction I have in mind. For two concise occurrences
of ―subjectivation,‖ see UPfr, pp. 41, 44. ―Assujettissement,‖ which I shall translate as ―subjection,‖ is used in The
Use of Pleasure but as one aspect of Foucault‘s reconfiguration of ethics. His definition, however, does not appear
to be very clear in view of the distinction. He says: ―…mode of subjection [assujettissement], that is to say, the
manner in which the individual establishes his relation to [the] rule and recognizes himself as linked [comme lié] to
the obligation of its realization‖ (UPfr, 38). I suggest one way of disambiguating this usage in the discussion of
ethics in the second part of this chapter.
199
I want to motivate this distinction not as a return to spontaneous subjectivity after what
would be its earlier reduction to an effect of systems of knowledge and power, but rather as a
recourse, made necessary because a reflective moment of self-relation is required to make
intelligible any appeal to subjectivity. From this perspective, it matters little whether Foucault
was always aware of this—I believe he was, since all of his dismissals through the 1960s are of a
particular conception of subjectivity—or he realized its urgency only later. The main question is
whether these two dimensions of activity and passivity can be thought together in the framework
I have constructed, or whether Foucault‘s rearticulation of experience is without recourse in any
form of recognizable subjectivity.223Since the major contention against Merleau-Ponty‘s account
circles around his appeal to a pre-reflective and direct access which the (embodied) subject has
to itself, what is involved in Foucault‘s account of self-relation becomes central.
Discipline and Punish proposes two poles of analysis, that of subjection and of
objectification. Having offered a ―nominalist reduction of philosophical anthropology,‖ Foucault
brackets scientific, natural or otherwise, appeals to the essence of man as what could be
explained through the formulation of objective laws; he now describes the correlation between
the techniques of subjection and the processes of objectification. In chapter three I offered one
interpretation of how the reciprocal relation of mutual implication which holds between
discursive and nondiscursive practices can be understood. The correlation between subjection
and objectification becomes intelligible in the light of that interpretation:
223 It should be clear, then, that whether Foucault‘s contracting of HIV has something to do with his ―turn to the
subject,‖ or whether the late works are indicative of his ―going soft,‖ whatever interest they may have, are external
to the line of reasoning I want to pursue here. It is undeniable, even for someone sympathetic to Foucault, that
volumes two and three of the History of Sexuality, relative to the fire and brimstone style of the earlier works, are
somewhat soporific. I want to claim, however, that their motivation lies in Foucault‘s attention to a reflective
moment which is presupposed by subjectivity. For a contrasting view on this point, see Slavoj Zizek, The Ticklish
Subject: The Absent Centre of Political Ontology, (London: Verso, 1999)
200
Relations of power are not in a position of exteriority with respect to other types of relationships
[e.g. economic, epistemic, sexual] but are immanent in the latter; they are the immediate effects of
the divisions, inequalities, and disequilibriums which occur in the latter, and conversely they are
the internal conditions of these differentiations. (HS, 94, my italics).
‗Truth‘ is linked in a circular relation with systems of power which produce and sustain it, and to
effects of power which it induces and which extend it. A‗regime‘of truth. This regime is not
merely ideological or superstructural. (PK, 133, my italics)
New objects are constituted for knowledge through techniques of subjection, which are in
turn extended and supported by normalizing knowledge. Foucault places special emphasis on the
development of examination as a specific form of inquiry and its extension across institutions.224
Through the diffuse application of examination in hospitals, military barracks, schools, and job
interviews, individuals are disciplined, i.e. made into docile bodies capable of satisfying
imperatives of production. And this disciplinary subjection results in the production of
knowledge which objectifies the individual further.
The ―genealogy of the modern soul,‖ then, is a question of showing the reality of the soul
as ―the present correlative of a certain technology of power over the body‖: ―[the soul] is
produced…around, on, within the body by the functioning of a power that is exercised on those
punished---…in a more general way, on those one supervises, trains and corrects…‖ (DP, 29).
The disciplinary individual, unlike the juridical subject of legal contracts, is the product of
relations of power ―governing our gestures‖ and ―dictating our behavior‖; and even though the
relation to truth is integral to the constitution of subjects, this truth is ―a system of ordered
procedures for the production, regulation, distribution, circulation, and operation of statements…
224 DPfr, pp. 201, 207-20, 260-64
201
[which is] linked in a circular relation with systems of power which produce and sustain it…‖
(FR, 74).
The body, which, for Merleau-Ponty, is the point of non-differentiation of the subjective
and the objective, is caught up in a regime of truth, where material subjection and ideal
objectification mutually support and reinforce one another. On this account, the matrix of
practices that provides the intelligibility and orientation of theoretical formulations and reflective
actions is not the pre-reflective opening of corporeal existence but the techniques of corporal
punishment. Confinement of human beings in the rigidly structured spaces of hospitals,
classrooms, factories, and prisons (DP, 227-28), makes possible the emergence of ―submissive
subjects…and a dependable body of knowledge built up around them‖ (DP, 295). The body,
then, is permeated by relations of force which constitute individuals as objects of knowledge and
subjected to power. Similarly, the first volume of the History of Sexuality brackets the question
of what the reality of sex may be in its natural existence or corporeal lived-through experience
(vécu), but shows it as ―an ideal point made necessary by the deployment of sexuality and its
operation‖ (HS, 155).225
Therefore, the criticism which I formulated from the perspective of Merleau-Ponty‘s
phenomenological questioning, namely that Foucault mistakes the map for the territory, has to be
turned around: there is no domain of lived-through signification where ―[t]hings murmur
225 The French text has: ―…made necessary par le dispositif de sexualité…‖ (p. 205). In chapter two I suggested
that the concepts Foucault marshals during this period may be brought closer together in the light of how I read the
historical a priori as involving both discursive and nondiscursive practices. This reading required the modification of
Foucault‘s own claims about the historical a priori as involving solely discursive systems and their autonomous
regulation. The concept of ―dispositif‖ (apparatus) can also be assimilated to that of ―regime,‖ to the extent to which
it designates ―…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‖ (PK, 196). What
lends further support for this reading is Foucault‘s specific designation of the discursive element in ―regime of truth‖
as the ―discursive regime‖. Also see his definition of ―discipline‖ in OD, pp. 222-23.
202
meanings our language has merely to extract… [and where] this language was already
whispering to us of a being of which it forms the skeleton‖ (OD, 228). If a regime of truth
mediates our knowledge of ―things‖ as well as of our ―selves,‖ then there can be no immediate
intuitive grasp of already constituted meanings prior to their discursive uptake.
Of course, that this is so does not admit of direct proof, in the manner in which Merleau-
Ponty could consistently suggest that we should look and see, verify, in ourselves the truth of
phenomenology; what Foucault can do is to show the ways in which this conception is one
among several ways of ―eliding the reality of discourse‖ (OD, 227) and offer an alternative
conception: Discourses are ―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 language [langue] and to speech.
It is this ‗more‘ that [must be described]‖ (AK, 49).
So Merleau-Ponty is correct in claiming that neither language as a formal system of
differential elements nor the reflective selection and formulation of sentences by a speaking
subject is sufficient to grasp the actuality of signification; but he is wrong to locate the latter in
our mute contact with the world. Or, in the terms of my initial formulation of the question,
sexuality does elicit our response by soliciting our attention, but this is because ―relations of
power had established it as a possible object; and conversely, if power was able to take it as a
target, this was because techniques of knowledge and procedures of discourse were capable of
investing it‖ (HS, 98).
Merleau-Ponty‘s attention to sexuality is motivated by ontological questions, or perhaps
the ontological question: ―If…we want to bring to light the birth of being for us, we must…look
203
at that area of our experience which clearly has significance and reality only for us, and that is
our affective life‖ (PP, 154). Sexual experience, then, reveals how being begins to exist for us
through desire and love, in order to gain insight into how being exists for us in general. Not
unlike Foucault‘s analyses in the first part of Mental Illness and Psychology, sexual pathology
provides initial traction on this question.226 His description of how sexual anomalies are lived as
an incapacity shows the inadequacy of explanations in terms of conditioned reflex reactions or
ideational associations: ―Pathology brings to light, somewhere between automatic response and
representation, a vital zone in which the sexual possibilities of the patient are elaborated, in the
same way…as are his motor, perceptual and even intellectual possibilities‖ (PP, 156). The sexual
meaning and value of external stimuli derive from the ―structure of perception or erotic
experience‖ (Ibid), and it is this structure itself which becomes etiolated or impoverished in
pathological sexuality.
The normal cases, however, are those where the body is not perceived as an object: it is
―subtended‖ by a sexual schema, ―which is strictly individual‖ (Ibid). This structure of
perception is in turn explicated as the power of projection of a sexual world. For the pathological
subject, it is the world itself which has ceased to speak to him sexually, or ―because the patient
no longer asks, of his environment, this mute and permanent question which constitutes normal
sexuality‖ (PP, 156, my italics).
226 There is a curious appeal to cases regarded as pathological which runs through many twentieth century
reflections on knowledge. Interestingly, it is to be found in Heidegger‘s emphasis on the hammer‘s breaking down
as the condition of possibility of its becoming present-at-hand (or its Vorhandenheit mode of being), as well as in
the privileging of sexual anomalies as revelatory of the normal functioning of the human mind and body in naturalist
psychology and physiology. See Martin Heidegger, Being and Time (New York: Harper & Row, 1962) pp. 48, 97f,
114, 116, 200.
204
Hence the phenomenological description of sexual experience and its structures is a
question of referring sexuality to that power which, below the level of conscious awareness and
reflection, synthesizes stimuli and reaction (or behavior) into a situation. And this power of
synthesis is located at the level of the body as lived. In The Visible and the Invisible, Merleau-
Ponty writes ―my body, as director [metteur en scène] of my perception,‖227 which captures the
active role of the body as perception. To be sure, this activity is not that of transcendental
subjectivity, and Merleau-Ponty, in a beautiful phrase, aims at ―the passivity of our
activity‖.228He criticizes all efforts that link up passivity and activity only to end up prioritizing
one over the other: ―either [this priority] extends passivity to their ensemble, which amounts to
detaching us from being, for in the absence of a contact of self to self, I am given over to an
organization of my thoughts, the premises of which are masked from me, and to a mental
constitution which is given to me as a fact; or it restores activity [which reduces the world to
thought of world]‖.229
Moreover, his explicit eschewal of prioritizing one term over the other, in favor of the
relation of reversibility which binds them together,230 takes the form of the resolution of an
antinomy the key to which is his notion of ―flesh‖.231‖The flesh of the visible‖ does not partake
of an anthropological investigation, but rather it designates the flesh of being [l‟être charnel],
which, as ―presentation of a certain absence, is the prototype of Being, of which our body,
227 VI, p. 23
228 Ibid., p. 221. The following discussion of the theme of activity and passivity are based on pp. 65-66.
229 Ibid, p. 65.
230 ―…between sound and meaning [sens] and speech and what it wants to say [ce qu‟elle veut dire]…‖ (Ibid, 188).
231 Ibid, pp. 177, 181-82, 190, 198.
205
sensing sensible [le sentant sensible], is a variation…and the constitutive paradox of which is
already in all that is visible…like my body is at one stroke both phenomenal and objective.‖232
Therefore, Merleau-Ponty resolves the antinomy of activity and passivity by installing
their perpetual reversibility in this ―degree zero of Being‖233 that is my body. The dichotomy
between subject and object is not cancelled through a reduction, and it is not raised through a
sublation. Instead, perception as ―the background from which all acts stand out, and is
presupposed by them‖ (PP, xi), is the field of this perpetual interplay between the subjective and
the objective, intention and expression. And the dramatic actions and passions of my sexuality
are only derivative in relation to the ontological structures of my affective life through which
Being is revealed through my body. From this perspective, ―language is only a regional
problem…if one considers the secondary and empirical operation of translation, coding and
decoding….‖234
But prior to all formal systems and conventional meanings, there is parole parlante, as
the ―assumption‖ of convention as nature by the being who lives in language; and which refers
to a langage opérant ―which does not need translation into significations and thoughts
[because]…like a weapon, an action, an insult and a seduction, it brings to light all the deep
relations of experience [vécu] where it is formed.‖235 Thus, there is a language interior to that
phenomenological domain of silence and which is both activity and passivity. The relation
between lived experience and this language (langage opérant) is not that between transcendental
condition and empirical conditioned. It is rather similar to that between what is implicit and what
232 Ibid, p. 177.
233 Ibid, p. 150. His invocation of this phrase occurs here as an explicit contrast with the ―primitive power to
contemplate, pure gaze that fixes things…and essences….‖
234 Ibid, p. 165.
235 Ibid.
206
is explicit, to the extent to which it is the ―apparition of something there where there was nothing
and no other thing.‖236There may be no pre-constituted intention lodged in a transcendental
space, but the prevalent model of language for Merleau-Ponty remains that of expression.
Sexuality, therefore, can only be conceived as a modality of expression of that intentionality
which pertains essentially to my body as living.
If the central insight of Merleau-Ponty‘s answer to the question of sexuality is that the
life of the body (flesh) and the life of the soul are in a reciprocal relation of expression, locked in
a perpetual reversibility of the subjective and the objective, then the central insight of Foucault is
that, contrary to the infinite and indefinite movement of intention and expression, knowledge is
produced and subjects are locked up in a reciprocal relation of implication between subjection
and objectification. And if Merleau-Ponty refers this movement to the zero degree of Being, or
again, to leben as the ―primary process from which…it becomes possible to erleben this or that
world…‖ (PP, 160), Foucault refers it to the regime of truth as the generative matrix of the
formation of subjects and knowledge.
Merleau-Ponty‘s project is by no means a typically foundationalist one, since ambiguity
and indeterminacy are constitutive of existence (PP, 169), and are therefore inherent in things.
But he nonetheless offers the lived experience of corporeal existence as the prereflective ground
of all reflective formulations and actions. The result is that the ontology of ourselves—here of
the body in its sexual being—becomes the hermeneutic activity wherein meaning is anticipated,
confirmed, and disappointed, only to start over again. And because ―man is not a machine,‖ but
236 Ibid.
207
the dialectical exchange between autonomy and dependence, his phenomenological descriptions
resonate with our intuitions about how we relate to our sexuality and its activity.
However, it is precisely Merleau-Ponty‘s inscription of these intuitions in an ontological
register237 that creates problems when interrogated from the standpoint of Foucault‘s experience.
For consider Merleau-Ponty‘s appropriation of psychoanalysis:
Whatever the theoretical declarations of Freud may have been, psychoanalytical research is in
fact led to an explanation of man, not in terms of his sexual substructure, but to a discovery in
sexuality of relations and attitudes which had previously been held to reside in consciousness.
Thus the significance of psychoanalysis is less to make psychology biological than to discover a
dialectical process in functions thought of as ‗purely bodily‘, and to reintegrate sexuality into the
human being. (PP, 157-158)
The hermeneutic aspect of psychoanalysis, which reads symptoms not as effects of
antecedent causes, but as meanings to be deciphered, enables Merleau-Ponty to enlist it in the
service of bringing to light the fundamental dimensions of existence. It is from the perspective of
these dimensions that nothing appears as accidental: ―There is no doubt at all that we must
recognize in modesty, desire and love in general a metaphysical significance…‖ (PP, 166); or
again, ―The importance we attach to the body and the contradictions of love are…related to a
more general drama which arises from the metaphysical structure of my body, which is both an
object for others and a subject for myself‖ (PP, 167).
Once grasped in the primary ontological structures through which sexuality is interpreted
as a free projection of existence, no event can be conceived as externally determined because
―the symptom is overdetermined‖. With the result that, escape from reductionist naturalism is
237 Or better, his provision of an ontological framework for these intuitions, where I intend no dismissal by calling
them ―intuitions,‖ but want to admit a prima facie counter-intuitive thrust which pertains to Foucault‘s conception.
208
purchased by interpreting, say, ―frigidity‖ as the expression of ―a refusal of orgasm, of
femininity or of sexuality‖ (PP, 158, my italics). Or again, loss of speech becomes the refusal of
communal existence, or intersubjectivity, and hysteria the flight from situation.
In chapter three, I discussed some of the virtues and flaws of this account of experience.
Here I want to focus on one key aspect in the light of what I said in relation to the regime of truth
and subjectivity. It is that the ―operative intentionality‖ of the experience of corporeal existence
cannot bear the weight Merleau-Ponty places on it when he ontologizes the categories through
which subjects come to relate to their sexuality. What motivates his argument is the desire to
steer between idealism and empiricism—which is what Foucault also wants to do—but he
pursues it in the direction of an ontological framework which does the work of a priori synthesis,
gathering external stimuli and subjective behavior into the cohesion of a situation.238
It is true that no datum is given to consciousness without appearing as already integrated
into a pattern, but his interpretation of what this patterning involves restores the transcendental
activity of a subject, (only) in accordance with which what is given must be taken. So much so
that forgetfulness, hysteria, frigidity, and all the other categories of sexual behavior, are
interpreted as so many ways in which the subject expresses its fundamental projection of a
meaningful world.239Merleau-Ponty is explicit in not reducing this intentional act to a conscious
and voluntary one—and to that extent he is in the neighborhood of Foucault‘s ―intentional and
nonsubjective‖ power relations (HS, 94)—but he underestimates and misconstrues the historical
constitution of the categories through which identities are forged and sexuality is experienced.
238 Merleau-Ponty approvingly refers to Freud‘s comparison of ―the accident occurring from outside [with] the
foreign body which, for the oyster, is merely the occasion for secreting a pearl‖ (PP, 158, n. 2): strictly speaking,
then, no event is externally determined.
239 ―Forgetfulness is therefore an act‖ (PP, 162).
209
One should rather say that:
The power relation that underlies the exercise of punishment begins to be duplicated by an object
relation in which are caught up not only the crime as a fact to be established according to
common norms, but the criminal as an individual to be known according to specific
criteria….The processes of objectification originate in the very tactics of power and of the
arrangement of its exercise. (DP, 101)
From the level of Merleau-Ponty‘s ontological interrogation, the perspective reflected in
this constitution of the individual as criminal can only appear as an empirical fact, similar to the
operation of translation which he stated could only be secondary and derivative relative to parole
parlante. Therefore, despite the fact that he wants to articulate a hermeneutical conception of
activity and passivity through their reciprocity, he ends up privileging the moment of activity.
―If,‖ he claims, ―we conceive man in terms of his experience, that is to say, of his distinctive way
of patterning the world…a handless or sexless man is as inconceivable as one without the power
of thought‖ (PP, 170). He thereby elides the specific power of constitution operative in
discourses which create the objects of which they speak. The operative intentionality of the body,
non-voluntary and non-conscious though it is, misses the way in which the individual ―whom we
are invited to free, is already in himself the effect of a subjection much more profound than
himself….The soul is the effect and instrument of a political anatomy‖ (DP, 30). In short, what
the ontologizing of sexuality, by referring it to the body in its sexual being, cannot grasp are the
criterial discursive and nondiscursive practices and their transformations.
It would be premature and presumptuous to say that nothing interesting can be said about
the ―unmotivated upsurge of the world,‖ or that its ontological structures cannot be defined. But
it is important to distinguish Merleau-Ponty‘s ontological referral of sexuality to ―this mute and
permanent question which constitutes normal sexuality‖ (PP, 156) from the ―historical ontology
210
of ourselves‖ (MF, 237), which Foucault claims to be doing and which refers that question itself
to the regime of truth through which sexuality is constituted as something to be questioned as to
its truth and in relation to which we feel compelled to seek knowledge.
A useful way of specifying what this involves and why Foucault feels the need to write
its history is through comparison with Martha Nussbaum‘s proposal to construct an account of
non-relative virtues240. Starting from Aristotle‘s claim that ―all human beings seek not the way of
their ancestors, but the good,‖241 she argues against the view according to which virtue ethics
would entail a denial of any ―transcultural norms, justifiable by reference to reasons of universal
human validity, by reference to which we may appropriately criticize different local conceptions
of the good.‖242Her response to this interpretation is to reconstruct Aristotle‘s account of the
virtues as involving, for each candidate virtue, the isolation of a sphere of human experience that
figures in any human life and in which any human being will have to make some choices or act
in some ways rather than others. The reconstruction of the list of virtues includes: the experience
of fear of damage, especially death, correlated with the virtue of courage; experience of bodily
appetites and their pleasures, correlated with moderation.243
Since the construction of the list starts from putatively universal experiences involving
choice, it is not possible to say that ―a given society does not contain anything that corresponds
to a given virtue. [Nor is it an open question] whether a certain virtue should or should not be
included in [a particular agent‘s] life.‖244 People will argue about what an appropriate choice in
240 Martha Nussbaum, ―Non-Relative Virtues,‖ in Moral Relativism: A Reader (Oxford University Press, 2001) pp.
199-226
241 Aristotle, Politics, 1268A39 ff.
242 Ibid., p. 200.
243 For the full list, see ibid. pp. 202-203.
244 Ibid., p. 203.
211
fact is, but they will then be arguing about the same thing and only advancing competing
specifications of the same virtue. She claims that the reference term of each virtue is fixed by its
relevant sphere of experience, which she calls ―grounding experience‖.245 For instance,
experiences of harm, deprivation and inequality fix the referent of justice. A ground experience,
then, is a domain of human existence in which choice is both non-optional and problematic
(since there is de facto disagreement about the specification of what counts as a just act).
Her reconstruction emphasizes the strict division between the demarcation of grounding
experiences and the concrete inquiry which specifies what counts as appropriate within each
experience. This division enables her to admit a degree of historicity of virtues while grounding
their universality. So, Aristotle‘s inclusion of megalosuchia—which is more Greek than
Christian, to the extent that, for the latter, the appropriate attitude towards one‘s own worth is
one of humility—in the list of virtues is to be replaced by a more neutral name which stands for
the universal experience of the problem concerning one‘s own worth. Christian humility and
Greek megalosuchia would then be two competing concrete specifications, but they would only
be ―rival accounts of one and the same thing.‖246
Nussbaum considers three possible objections to this reconstruction of non-relative
virtues. First, the problem may be the same across its cultural and historical incarnations, but this
does not guarantee that there will be a single universally valid solution. Second, there are no
grounding experiences, since all such putatively primitive experiences are culturally and
historically constructed. Third, the construction of virtues suffers from a blinkered historical
245 Ibid.
246 Ibid., p. 206. She then provides a historical justification of this division by tracing the movement from Aristotle
through the Stoics to the Christian fathers which transformed greatness of soul, through the doctrine of the
worthlessness of externals, into the Christian denial of the body and of the worth of terrestrial existence.
212
sense in not being able to imagine a radical utopian society in which the virtues would be
superfluous.247
Her response to the first objection is: that the virtues are context-sensitive ―does not
imply that it is right only relative to, or inside, a limited context….It is right absolutely,
objectively, anywhere in the human world to attend to the particular features of one‘s context.‖248
Thus some interpretative and contextual differences may be admitted, but if the same context
were to be reproduced, the same solution would be the right one. She responds to the second
objection by conceding that ―there is no ‗innocent-eye,‘ no way of seeing the world that is
entirely neutral and free of cultural shaping;‖ but she insists that it must be possible to identify
common features of humanity, two features of which will be ―mortality and the body‖.249
Whatever the different cultural interpretations of death across history may be, the fact of
our mortality shapes every aspect of more or less every human life. Moreover, prior to any
cultural mediation, we are familiar with our embodied existence, since we experience hunger,
thirst and desire: ―The experience of the body is culturally influenced; but the body itself, prior
to such experience, provides limits and parameters….‖250Finally, the third objection, the
paradigmatic statement of which criticizes Aristotle‘s inclusion of generosity as a failure to
imagine a form of social organization without private property, is referred to the tragic structure
according to which ―all forms of life, including the imagined life of a god, contain boundaries
247 For the first objection, see p. 207; for the second one, see pp. 208-210; and for the third one, see pp. 210-211.
248 Ibid., p. 213.
249 Ibid., pp. 215-216, 218.
250 Ibid., p. 218. What follows, on the next page, is a similar claim about pleasure and pain.
213
and limits.‖251Therefore it is unrealistic and dangerous, based on human experience, to expect the
creation of a social arrangement in which no limits would be needed.
Nussbaum‘s argument is exceptionally sensitive to the demands of a context-specific
understanding of virtue, and strictly speaking, she is not offering an ontological account of
experience. But her line of reasoning tracks that of Merleau-Ponty to the extent that she wants to
insist on the fact of our embodiment which, paradoxically, cannot be reduced to any other fact.
Although in her response to the second objection she withdraws the suggestion that grounding
experiences could be a pre-linguistic bedrock; and although her response to the third objection
concedes that structural and historically modifiable power relations are endemic features of
social contexts; she nonetheless wants to insist on the fact of our body as the site of the universal
and necessary experiences of pain and pleasure, of desire and its frustration, and of hunger, thirst
and death. She insists on this fact, I believe, because she thinks that the alternative is either a
formalism about the virtues which makes ethics more rational than its actuality can support, or a
relativism without recourse. And I want to claim that this insistence is mistaken, from a
Foucauldian perspective, not because there is no such fact, but because it never performs the role
which Nussbaum or Merleau-Ponty would have it do.252
In The Use of Pleasure, Foucault defines the task of the history of thought as articulating
―the conditions in which the human being ―problematizes‖ what he is, what he does, and the
mode in which he lives‖ (UPfr, 18), and he links ―problematization‖ to a set of practices. Even
more explicitly, he argues: ―Problematization doesn‘t mean representation of a pre-existing
251 Ibid., 222.
252 There is also the paradoxical repetition, given the Aristotelian background, of a quasi-transcendental distinction
between form and content in the conceptual pair ―grounding experience‖ and its ―concrete specification,‖ with the
result that the universality of problem is purchased at the expense of its radical separation from any substantive
considerations.
214
object, nor the creation by discourse of an object that doesn‟t exist. It is the totality of discursive
and non-discursive practices that introduce something into the game of the true and the false and
constitute it as an object for thought‖ (PPC, 257, my italics). It is not, therefore, a question of
denying the existence of human embodiment, its appetites, and its mortality; but rather denying
the explanatory and justificatory role assigned to them.
If Foucault is right that, for instance, Greek antiquity made a very different problem of
the appetites and their management, and consequently, did not take the gender of the partner to
be a very significant element in their reflection on sexual ethics or hygiene, then it does not
explain much of anything to refer to the ―experience of homosexuality‖ in ancient Greece—for
there was none in a sense that is salient for modernity; nor does it justify what appropriate
attitude we should have with respect to homosexuality. In other words, the appeal to the
universality and necessity of ground experiences, or of structures of embodied comportment, or
again of the ontological framework subtending my corporeal existence, does not perform any
epistemological role, and it does not provide any normative purchase. Therefore, Foucault‘s
reconfiguration of experience in terms of discursive and nondiscursive practices cannot (should
not) be taken as a prima facie reason for denying the right to call that experience. Nothing of the
softness of a caress, the abrasion of a desire, or the intensity of hunger, is lost by referring
experience to the conditions mediating the subject‘s self-relation.
4.2 Am I Free Not to Take Up a Standpoint?
Merleau-Ponty refers experience to the ―patterning‖ work which free existential projection does
prior to the reflective problematizations of and positions toward actions. This work seems not
215
unlike that performed by the historical a priori or the regime of truth. But if my argument in the
preceding section is correct, a crucial difference is that, from a Foucauldian perspective, the
operative intentionality of embodied existence cannot bear that weight or ground that function. If
anything only ever becomes a problem demanding a reflective solution within the context of
practices, then intentional analysis—however broadly it is understood—which would be
grounded on the free projection of the subject, cannot account for the degree to which
subjectivity itself is constituted by practices. And appeal to existence or Being as the
fundamental, all-mediating background of all determinate meaning and value only occludes that
fact.
However, it could legitimately be argued that if the argument succeeds, then it is only by
shifting the emphasis to the pole of passivity at the expense of the pole of activity; and it is
precisely their reciprocity which motivates Merleau-Ponty to refer them to the degree zero of
Being that is my body: for ―I never become quite a thing in the world‖ (PP, 165). In this light,
Foucault‘s distinction between subjectivation and subjection would appear to be swallowed by
the implacable operation of ubiquitous power relations. And if that were the case, his claim to be
formulating a non-reductive concept of experience would ring hollow.
One could insist that it is the ontological conception of experience which reduces it to a
quasi-transcendental subjectivity modulated hermeneutically. But that insistence on its own
would remain somewhat disingenuous. There is a quip according to which an alcoholic is
someone you do not like who drinks as much as you do; the philosophical charge of
reductionism is a bit like that. Hence Foucault has to say more about what subjectivation is, not
only to clarify what experience would be, but also to motivate the possibility of a discourse that
is both historical and critical. In modernity more than in any other period, the peculiar activity of
216
critique is conducted in the name and grounded on the capacities of the spontaneous subject. And
so, if I am not a bit of the world, Foucault‘s grappling with the dilemma of whether I am free to
give up the kind of reflection constitutive of modernity without thereby rendering freedom itself
meaningless and impossible, needs further unpacking.
4.21 Subjects without Experience
In one of his most explicit discussions of the trajectory his histories followed, Foucault says:
If by thought is meant the act that posits a subject and an object in their various possible relations,
a critical history of thought would be an analysis of the conditions under which certain relations
between subject and object are formed or modified, to the extent that these relations are
constitutive of a possible knowledge. (CC, 314)
Two features stand out from this admittedly laconic passage. First, contrary to what could be
called the predominant strategy in the twentieth century of avoiding the subject-object
dichotomy, Foucault appears to situate his practice of critical history squarely within their
reciprocity. Second, the language he uses—―positing act,‖ ―constitution‖—is reminiscent of
German idealism, a mode of philosophy to which his explicit references are notoriously sketchy.
These two features raise a question: Is Foucault‘s definition of thought one more anthropological
repetition of the transcendentally constitutive subject?
217
A preliminary step toward an answer goes through Foucault‘s negations. The conditions
of possibility of relating subject and object are not formal conditions of relation to a general
object X—which seems to target the Kantian formulation; and they are not empirical conditions
which enable a subject to become conscious of an already given object—which seems to target
empiricist formulations. Rather, ―[t]he question is one of determining what the subject must be,
what condition is imposed on it, what status it is to have, and what position it is to occupy in
reality or in the imaginary, in order to become the legitimate subject of one type of knowledge or
another…it is a matter of determining its mode ‗subjectivation‘‖ (CC, 315, modified). His
examples suggest that the mode of subjectivation is the way in which the subject relates to itself
under the particular description and normative status constituted by different types of knowledge:
subjectivation is different according to whether knowledge is that of exegesis of sacred texts,
observation of mentally ill patients, etc.
However: ―…the question is also and at the same time one of determining under what
conditions something can become an object of possible knowledge, how it could be
problematized as an object to be known, to what procedure of division it could be subjected, and
what part of it is considered pertinent…its mode of objectivation‖ (Ibid, modified). It is, then, the
mutual implication of subjectivation and objectivation, which is then referred to the ―emergence
of truth games‖. And critical history is to be the history of the conditions of emergence of these
games, in which it is a matter neither of the discovery of truth nor of its occultation, but of
―veridictions‖.253
253 I discuss conditions of emergence (existence) in relation to conditions of possibility in chapter two, part two.
218
This may be another instance of Foucault‘s penchant for multiplying concepts beyond
necessity. But I think there are two mitigating factors. First, he refers this linkage between forms
of subject and types of object to ― [what] has constituted for a time, a space, and particular
individuals, the historical a priori of a possible experience‖ (Ibid, 315). Therefore, the new
conceptual articulation is consistent with the strategy I have pursued in relation to experience.
Second, the mutual implication of subjectivation and objectivation, to the extent that it tracks that
of subjection and objectification, is motivated by Foucault‘s realization that critical discourse,
whether his own or that the history of which he writes, presupposes a reflexive moment that
allows the possibility of self-relation.254
It is important to note a restriction which Foucault imposes on the domain of his critical
historiography: what is at issue is only ―those truth games in which the subject itself is posited as
an object of possible knowledge: what are the processes of subjectivation and objectivation that
allow the subject to become, as subject, an object of knowledge?‖ (Ibid, modified, my italics).
Therefore, the object domain is to coincide with that of the human sciences.255If ―man‖ is that
object which, as transcendentally constitutive and empirically determined, is also subject, then
the human sciences aim at knowledge of just that conjunction. But since the analytic of finitude
shows the epistemological dead ends to which that conception leads; and since attention to the
regime of truth establishes the normalizing role knowledge plays; the foundational bedrock
254 I addressed the issue of whether this should be called his ―belated realization‖ or his ―explication‖ of what was
already in his early work in the preceding section. What I want to emphasize in this context is that, belated or not,
there is a reflexive moment in his definition of experience, and the question is whether he has any right to it.
255 Which is consistent with the concrete analyses of the History of Madness (psychology), The Birth of the Clinic
(clinical medicine), The Order of Things (psychology and sociology), Discipline and Punish (criminology), The
History of Sexuality (psychoanalysis and psychology). This assignation is somewhat schematic and ignores
considerable overlap, but part of the claim is precisely that for the human sciences there cannot be a strict and
positive demarcation of object-domain, since the object ―man‖ does not exist. (I discuss what that denial entails in
chapter two, part one.)
219
claimed by the human sciences is criticized by Foucault as epistemically unsound and politically
suspect. Hence his attempt to bring to light the very historicity of forms of experience as
processes of subjectivation and objectivation without seeking to ground those processes on an a
priori conception of truth or human nature.
He specifies three methodological implications of this shift to the historicity of forms of
experience: First, a nominalism in relation to anthropological universals. This nominalism is
motivated by the two-pronged criticism of the human sciences.256 Second, the avoidance of
transcendental reflection invoking the constitutive subject in relation to which the conditions of
possibility of objectivity as such are determined. Third, displacement of enquiry onto an
investigation of practices, ―approaching one‘s study from the angle of what ―was done‖ (Ibid,
318).257
These three claims entail three ways in which Foucault‘s critical history could potentially
be neither history nor critique. First, nominalism in relation to human nature entails skepticism
with respect to any nontemporal truth of the subject which could ground possible critique (of the
human sciences in particular and of practices in general). Second, circumventing transcendental
subjectivity entails reduction to pure objectivity and return to naïve positivism.258 Third, recourse
to practices as an ―ultimate‖ or ―fundamental‖ in some sense threatens to unleash either a
transcendentalism with no boundaries or the causal determinism of empiricist materialism. I
256 For a detailed discussion of ―methodological nominalism,‖ see Flynn, op. cit., pp. 31-48. I think the qualification
of Foucault‘s nominalism as a methodological and not a metaphysical commitment succeeds against those attempts
which read Foucault‘s claims as ontological. But as Flynn points out, an absolute distinction between method and
object of enquiry is not tenable in Foucault‘s historiography. I want to add that even methodological commitment to
individuals alone is vulnerable to the type of criticism I will formulate on the basis of Adorno‘s theory of mediation.
257 These three ―methodological choices‖ are articulated in CC, pp. 317-318.
258 See AK, p. 205ff on one formulation of this difficulty.
220
want to push these implications as far as they will go and see whether and in what sense
Foucault‘s reconfiguration can be the basis of a possible response. For he says:
[T]he practices—ways of doing things—that are more or less regulated, more or less conscious,
more or less goal-oriented, through which one can grasp the lineaments of both what was
constituted as real for those who were attempting to conceptualize and govern it, and of the way
in which those same people constituted themselves as subjects capable of knowing... [and]
modifying the real. These ―practices,‖…simultaneously as modes of acting and of thinking… [are
the key to] a correlative constitution of the subject and the object. (CC, 318)
I propose to reach a clearer understanding of practices, which, as ―modes of thinking‖
and of ―acting,‖ constitute the subject and the object in a relation of mutual implication, through
a comparison with Adorno‘s account of subject-object mediation. The reference to the latter is
not arbitrary but is motivated by the fact that Adorno too retains the subject-object relation as the
ultimate matrix for an analysis of practices and as the basis of a critical theory. However, since it
could be said that he is more sensitive than Foucault is to the implications of critique‘s recoil on
the critic, his very proximity to what I have argued so far provides the occasion for a sharper
definition of the contours of ―practice‖ and the evaluation of the critical status of Foucault‘s
histories.259
259 There are few studies of what a Foucault-Adorno ―confrontation‖ might look like. Those which do compare the
two almost invariably start from the acknowledgement that they are both concerned with the relationship between
rationality and power; move through Foucault‘s claim that ―if [he] had known the philosophers of [the Frankfurt
School] … [he] would have been so captivated by them that [he] wouldn‘t have done anything else but comment on
them‖ (EW, 3:274); and finally conclude by Foucault‘s ―repudiation‖ of both reason and the subject contrary to their
critical preservation by the latter. This is not so much false as not very informative, and I hope to complicate this
picture, however briefly, in what follows. On Foucault and Adorno, see Axel Honneth, ―Foucault and Adorno: Two
Forms of the Critique of Modernity,‖ The Fragmented World of the Social: Essays in Social and Political
Philosophy (State University of New York Press, 1995) pp. 121-131; Peter Dews, op. cit.; David Couzens Hoy,
―Power, Repression, Progress: Foucault, Lukes, and the Frankfurt School‖ Foucault: A Critical Reader, (Basil
Blackwell, 1986) pp. 123-149. On Foucault‘s own attempt at differentiating his project from that of the ―Frankfurt
School,‖ see EW, 3:272-73; 3:299-300; 3: 328-329.
221
It could be said of Adorno‘s discussion of experience what he said in relation to Proust,
that ―[he] brushed experience against the grain: But ‗it‘s all completely different‘ would remain
stamped with the impotence of the exotic, if its force were not also that of ‗this is how it
is‘.‖260He inscribes experience in a structure of subject-object mediation, which he explicitly
characterizes as dialectical, but where ―[t]he pendulum of the dialectic has come to a
standstill.‖261He insists on the possibility of unreduced/undiminished experience where the
reciprocity between subject and object would exclude the domination of one over the other; but
he fully recognizes its historical impossibility: ―One might say that experience is the union of
tradition with an open yearning for what is foreign. But the very possibility of experience is in
jeopardy. The break in the continuity of historical consciousness… [leads to] a historical moment
that…is ready to subscribe to the status quo, even by mirroring it where it opposes it.‖262I want
to first provide an account of the epistemological underpinning of some of his claims, before I
turn to a critical evaluation of Foucault.
Brian O‘Connor, in explicating Adorno‘s appropriation of Kant‘s thesis with respect to
things-in-themselves, argues that what is at work is the articulation of a critical materialism
which implies ―both that the subject does not passively receive meanings from the object, and
that the activity of the subject is circumscribed by the determinate independence of the
object.‖263This critical maintenance of both activity and passivity within the structure of
260 Theodor W. Adorno, ―Short Commentaries on Proust,‖ in Notes to Literature, vol. 1 (New York: Columbia
University Press, 1991), p. 181.
261 Adorno, ―Trying to Understand Endgame,‖ Ibid, p. 269. Also see ―Introduction to Benjamin‘s Schriften,‖ Notes
to Literature, vol. 2, p. 228.
262 Adorno, ―In Memory of Eichendorff,‖ Ibid, p. 55.
263 Brian O‘Connor, Adorno‟s Negative Dialectic, Philosophy and the Possibility of Critical Rationality,
(Cambridge: MIT Press, 2004) p. 20. O‘Connor provides a lucid account of subject-object mediation as the structure
of unreduced experience in Adorno. However, his thesis that Adorno‘s strategy is transcendental generates the risk
that there is little to distinguish this from Kant: what in Kant is mediated priority of subject becomes, in Adorno, the
mediated priority of object. Contrary to this, I want to argue that Adorno cannot provide a normative justification of
222
mediation is in response to the ―subjectivism‖ of idealism and the ―objectivism‖ of positivism.
Mediation, Adorno argues, is that relationship whereby subject and object ―reciprocally permeate
each other,‖ or again, ―constitute one another as much as—by virtue of such constitution—they
depart from each other.‖264Therefore, subject and object are not two independent substances
which are then placed in a relationship with one another, but they are constituted through their
very mediation.
But just as crucial for Adorno‘s account is that the epistemological separation of subject
and object is not merely a function of philosophical confusion: ―[their] separation is both real
and illusory: true, because in the cognitive realm it serves to express the real separation, the
dichotomy of the human condition, a coercive development. False, because the resulting
separation must not be hypostasized, not magically transformed into an invariant.‖265Failure to
appreciate the mutually mediated status of the subject and the object in any account of
experience will lead to contradictions in the very attempt to theoretically ground that account.
So, for instance, the idealist hypostatization of the transcendental subject must
simultaneously posit its existence as an object. The subject does not exist in a private space
outside of any conditioning by its environment and no sense can be given to ―constitution‖
without reference to empirical existence: ―The solidity of the epistemological I, the identity of
self-consciousness, is visibly modeled after the unreflected experience of the enduring identical
the thesis he wants to push, and the attempt to provide one for him, admirable though it is, inevitably ends up
transforming what is unique in Adorno into an expanded version of idealism. One must rather give up the
transcendental and take responsibility for the consequences. What responsibility may here mean is not easy to
articulate, but I hope to motivate its possibility through a comparison with Foucault‘s account of experience. On
O‘Connor‘s transcendental reading, see ibid. passim, but especially p. 55.
264 Theodor W. Adorno, Negative Dialectics, (London: Routledge, 1973) (Hereafter ND) pp. 139 and 174
respectively.
265 Theodor W. Adorno, ―Subject and Object,‖ in The Adorno Reader, (Malden: Blackwell Publishers, 2000) p. 139
(Hereafter, SO).
223
object. Even Kant essentially relates it to that experience‖ (Ibid, 148). But symmetrically, the
object itself cannot be thought without the subject either: ―After an elimination of the subjective
moment, the object would come diffusely apart like the fleeting stirrings and instants of
subjective life‖ (Ibid, 149). In other words, what Adorno is grappling with here through the
notion of the reciprocal mediation of subjectivity and objectivity can be seen as articulating two
moments: First, the object cannot be reduced to its conceptual uptake, or to the meaning-giving
activities of the subject; second, there is no immediate access to the nonconceptual properties of
the object, on which one must nonetheless insist.266
The second moment is Adorno‘s thesis regarding ―the priority of the object‖:
―Subjectivity changes its quality in a context which it is unable to evolve on its own. Due to the
inequality inherent in the concept of mediation, the subject enters into the object altogether
differently from the way the object enters into the subject. An object can be conceived only by a
subject but always remains something other than the subject…‖ (ND, 183).267It is beyond the
scope of this study to unravel what the meaning and force of this thesis may be268; but two points
relevant to my purposes could be drawn from it. First, part of what Adorno is getting at is that
objects have irreducible nonconceptual properties which are nonetheless meaningful. That is to
say, the subject‘s conceptual activity is not the only source of meaning of experience. Second,
266 What Adorno says, à propos the phenomenological critique of empiricist sense-datum theory of perception,
shows the dual demands on which he insists: ―But if Gestalt theory correctly objects to Hume and the psychology of
association that ‗there are‘ no such things as unstructured, more or less chaotic ‗impressions‘ isolated from one
another at all, then epistemology must not stop there. For data of the sort that epistemology cites Gestalt theory as
appropriately describing simply do not exist. Living experience is just as little acquainted with the perception of a
red ‗Gestalt‘ as it is with the ominous red percept. Both are the product of the laboratory.‖ (Against Epistemology: A
Metacritique, [Cambridge: MIT Press, 1983], p. 159.) (Hereafter, AE)
267 Or again: ―What is known through consciousness must be something; mediation aims at the mediated. But the
subject, the epitome of mediation, is the How—never the What…‖ (SO, 142).
268 For a fine-grained attempt at making sense of the thesis of priority using the sources of contemporary ―analytic
philosophy,‖ see Jay M. Bernstein, Adorno: Disenchantment and Ethics, (New York: Cambridge University Press,
2001), Chapter 6, and ―Re-enchanting nature,‖ in Reading McDowell On Mind and World, (New York: Routledge,
2002)
224
the object gains its conceptual qualities only through its socially and historically specific
mediation by the subject. The first point marks a divergence from Foucault‘s account and brings
Adorno into a qualified proximity to Merleau-Ponty to the extent that he too offers a ―thicker‖
account of embodied experience imbued with epistemological and justificatory functions.269I
want to focus on the second point.
Adorno‘s explication of what the social and historical dimension of conceptual mediation
involves is: ―[The] immanent generality of something is objective as sedimented history. This
history is in the individual thing and outside it; it is something encompassing in which the
individual has its place….The history locked in the object can only be delivered by a knowledge
mindful of the historic positional value of the object in its relation to other objects‖ (ND, 163).
Sedimented in the object, then, is the history of the meanings which have been deposited therein
through intersubjective practices.
But, unlike Husserl‘s appeal to sedimentation of meaning in the life-world, across (or
precisely on the basis) of which it is possible in principle to recover the original self-evident
intuitions that would secure the rationality of knowledge, Adorno refers this to the social totality,
―something encompassing in which the individual has its place,‖ and where the ―positional
value‖ of the object is determined. Moreover, even though the object‘s meaning and value is
determined by the social totality—the context of discursive and nondiscursive practices—no one
individual can be taken as a constitutive source. Social meaning and value sedimented in the
object confronts the individual as independent of his activity. Thereby, a degree of passivity is
incorporated in the very concept of social totality.
269 I discussed one possible response to what this may entail for the concept of experience I am arguing for in the
preceding section.
225
Characteristically, however, Adorno‘s affirmation of this claim is not without a negation:
What shows up faithfully in the doctrine of the transcendental subject is the priority of the
relations—abstractly rational ones, detached from the human individuals and their
relationships—that have their model in exchange. If the exchange form is the standard social
structure, its rationality constitutes people; what they are for themselves, what they seem to be to
themselves, is secondary. They are transformed beforehand by the mechanism that has been
philosophically transfigured as transcendental. (SO, 141, my italics.)270
[But] society is immanent in experience, not an allo genos. Nothing but the social self-reflection
of knowledge obtains for knowledge the objectivity that will escape it as long as it obeys the
social coercions that hold sway in it, and does not become aware of them. Social critique is a
critique of knowledge, and vice versa. (SO, 143, my italics.)
I want to tease out what I take to be the three implications of what Adorno is claiming
here: First, social reality is constituted by historical practices, and to that extent the shape of
reality—―actuality‖ in the Hegelian sense—is determined through human activity, conceptual
and otherwise. Second, this social reality, and the totality of relations which it comprises, is
constitutive for any given individual, to the extent that the history ―locked in objects‖ is not a
function of willful projection and cannot be modified through voluntary individual action.
Finally, the claims of (1) and (2) are referred to and intelligible through the structure of
reciprocal mediation of the subject and the object, which structure, moreover, is that of
experience.
These three claims can then be brought to bear on the three points I outlined above in
relation to Foucault‘s account of subjectivity, namely, nominalism with respect to
270 Earlier on the same page, he writes: ―[Psychological individuals] have little to say in the world, having on their
part turned into appendages of the social apparatus and ultimately ideology. The living human individual, as he is
forced to act in the role for which he has been marked internally as well, is the homo oeconomicus incarnate, closer
to the transcendental subject than to the living individual for which he immediately cannot but take himself‖ (my
italics).
226
anthropological universals, the repudiation of transcendental subjectivity, and the critical history
of practices. First, two points of convergence: the meaning and value of objects can only be
understood in terms of the relational social context in which they are located, and the subject is
implicated in that very knowledge of objects. Moreover, the structure of experience is precisely
this historically specific and determinate mutual implication of subject and object.
However, from an Adornian perspective, nominalism with respect to the subject appears
as the subject‘s dissolution into the totality of relations constitutive of it. And if that is correct,
then Foucault would be breaking out of the structure of experience as mediation, thereby
generating epistemological contradictions and normative confusions:271―Nominalism denies
society in concepts by disparaging it as an abbreviation for individuals‖ (SO, 151); and ―the very
term ‗particular person‘ requires a generic concept, lest it be meaningless‖ (Ibid, 139). In other
words, Foucault‘s nominalism could be charged with dissolving the very individual it seeks to
understand.
Furthermore, despite Adorno‘s trenchant criticism of transcendental subjectivity; and
despite his claim that far from being a fixed essence, subjectivity itself is determinable (not only
determining) and is transformed along with objectivity; he nonetheless insists that any rational
comprehension of this circular relation and transformation be understood in the medium of
conceptual reflection: ―One of the motives of dialectics is to cope with [what is evaded] by
usurping a standpoint beyond the difference of subject and object—the difference that shows
how inadequate the ratio is to thought. By means of reason, however, such a leap will fail. We
cannot, by thinking, assume any position in which that separation of subject and object will
271 An influential version of such an objection is articulated by Nancy Fraser, ―Foucault on Modern Power:
Empirical Insights and Normative Confusions,‖ Praxis International 3 1982
227
directly vanish, for the separation is inherent in each thought. It is inherent in thinking itself‖
(ND, 85).272 From this perspective, it might appear that the standpoint from which Foucault is
articulating his own discourse entails invocation of an immediacy, from which he is debarred
rationally.
Finally, the displacement of critical interrogation on to the domain of practices—as
defined by Foucault—could be charged with ―sociologism,‖ which would parallel the
epistemological problems of psychologism. The parsing of this criticism, from an Adornian
perspective, is doubly complicated because of Adorno‘s own commitment to referring both
objective and subjective formations to the context of their social and historical genesis. On the
one hand, ―[e]mpirical social research cannot evade the fact that all the given factors
investigated, the subjective no less than the objective relations, are mediated through
society.‖273Therefore, only a dialectical investigation of society would be adequate to the internal
contradictions and antagonisms of society.
He too is careful not to appeal to relations of production, or class conflict, or market
exchange, as the explanatory bedrock of society‘s antagonisms; nevertheless, from the
perspective of his insistence on social mediation of all determinations, Foucault‘s claim to be a
―happy positivist,‖ merely describing individual power relations would appear to mistake as
natural or ontological what is in fact his own methodological assumption of giving reality only to
atomic individuals. Even more forcefully, Foucault‘s appeal to a historically specific regime of
truth, only relative to which the validity claims it makes possible could be intelligible and must
272 The passage ends with: ―Today, as in Kant‘s time, philosophy demands a rational critique of reason, not its
banishment or abolition.‖ If what I claim in what follows is correct, however, Adorno‘s own ―rational critique of
reason‖ is far from unproblematically marked off from what its simple abolition would amount to.
273 Adorno, ―Sociology and Empirical Research,‖ in Adorno Reader, op. cit. p. 189.
228
be evaluated, could appear an instance of that ―childish relativism [which] would deny the
validity of formal logic and mathematics and treat them as ephemeral because they have come to
be‖ (ND, 40).
The reason why this charge is doubly complicated is that, Adorno himself writes as if the
law of noncontradiction, for instance, could be exhaustively referred to the context of its genesis:
Genetically logic presents itself as an attempt at integration and solid ordering of the originally
equivocal—a decisive step in demythologization. The law of non-contradiction is a sort of taboo
which hangs over the diffuse. Its absolute authority…directly originates in the imposition of the
taboo and in the repression of powerful counter-tendencies. As a ‗law of thought,‘ its content is
prohibition: Do not think profusely. (AE, 80)
There appears to be an almost perfect isomorphism between this passage and the one where
Foucault claims that:
[I]n every society the production of discourse is at once controlled, selected, organized, and
redistributed according to a certain number of procedures, whose role is to avert its powers and its
dangers, to cope with chance events, to evade its ponderous, awesome materiality. (OD, 216)
And when Foucault continues, in the same essay, to list the rules of exclusion which are
constitutive of the production and distribution of discourse, he includes, along with forms of
prohibition and the division between reason and madness, the opposition between the true and
the false; and so, he would seem to be guilty of the same charge which Adorno both raises and
seems to commit.
Whatever the implications of this ambivalence for Adorno may be, there is one recourse
for motivating and justifying the critical thrust of his reflections which is not available to
Foucault. In Excursus I of Dialectic of Enlightenment titled ―Odysseus or Myth and
Enlightenment,‖ Adorno traces the contradictions of modern social organization and the
229
domination inherent in intersubjective relations to the repression of inner nature presupposed by
the very development of subjectivity, or that aspect of it which claims self-identity.274The
emergence of self-consciousness through the denial of its dependence on nature entails the
internalization of external coercion by the subject. Adorno calls this process ―the introversion of
sacrifice,‖ and it is what makes the process of enlightenment a dialectical process.
Not unlike Nietzsche‘s account, in the Second Essay of the Genealogy of Morals, of the
development of ―conscience‖ through the ―turning inward‖ of those instincts which are
prohibited from external discharge275, Adorno invokes the organization and limitation of drives
presupposed by the development of a self-identical ego. This limitation is then internalized and
reproduced as the internal division of the self from itself: ―The antithesis of thought to whatever
is heterogeneous to thought is reproduced in thought itself, as its immanent contradiction‖ (ND,
146) Therefore, Adorno can provide qualified justification for his critical practice which, as
immanent critique, diagnoses the contradictions resulting from the theoretical and practical
efforts to assert the full autonomy of the subject.
Moreover, Adorno‘s insistence on the inevitability of subject-object mediation enables
him to offer qualified normative justification for his critique. The development of industrial
capitalism increases the repressive constraints imposed on the subject and further impoverishes
274 Max Horkheimer and Theodor W. Adorno, Dialectic of Enlightenment, Philosophical Fragments, (Stanford:
Stanford University Press, 2002) p. 40ff. The book is co-authored with Horkheimer, but I ignore what that might
entail in what follows.
275 Friedrich Nietzsche, ―On the Genealogy of Morality,‖ in ‗On the Genealogy of Morality‟ and Other Writings
(Cambridge University Press, 2006) ―Second Essay: ‗Guilt‘, ‗bad conscience‘ and related matters,‖ p. 35ff
230
its spontaneity; and to that extent self-preservation, for the sake of which the domination and
denial of external nature was undertaken, becomes self-renunciation.276
But spontaneity remains a potential of subjectivity. The dialectical account of the natural
historical emergence of the subject entails both that instinctual life contains a moment of
sensuous happiness and that their limitation has a moment of freedom (because enjoyment
presupposes a minimum of independence from nature). And since subjectivity is simultaneously
freedom and domination, repression and happiness, Adorno avoids positing a static opposition
between what would be the spontaneous diffusion of the instincts and the repressive identity of
the subject: identity contains an ineliminable moment of non-identity, and the non-identical,
however indeterminately, provides that in the name and on the basis of which critique operates.
This recourse to a speculative account of the natural historical emergence of the subject
―grounds‖ two series of mutually conditioning claims. On the one hand, Adorno is in full
agreement with Foucault that the type of power relations in contemporary societies produces
socialized subjects which are objects: their spontaneity—let alone autonomy—does not count as
a determining instance of social life. On the other hand, ―[t]he undifferentiated state before the
subject‘s formation was the dread of the blind web of nature, of myth‖ (SO, 140). Therefore
dissolution of the subject would result in a slavery worse than that operative in socialized
society.
Because his account of subjectivity insists on the necessity of holding onto this tension
and not resolving it in favor of either pole of the opposition, the potential of a utopian resolution
276 Where self-renunciation assumes the paradoxical form of narcissism, since the subjects so constituted demand—
and in some instances find—immediate satisfaction of desires and needs produced and sustained by the ―culture
industry‖. Adorno‘s detailed discussion of this theme is in The Culture Industry (Routledge, 2001).
231
is immanent in the very dialectical relation between the individual and society. Since the latter
can only reproduce itself through the former, resistance is possible in principle, however
attenuated it may be. Adorno refuses to determine the conditions of what such a resolution would
be, since, to the extent that the attempt would presuppose identification on the basis of present
categories complicit with domination, it would betray its object in its very assertion.
Nevertheless, without the invocation of its (logical and practical) possibility, subjectivity would
dissolve into the bundle of social relations constitutive of it, and critique would remain without
recourse.277
Hence even though Adorno refrains from specifying what cognitive and practical
relations without domination would be like, he secures his right to call domination by its name
by appealing to the possible sublation of the structure of subjectivity which, up until now in
history, has been one of internal and external domination of difference and non-identity.
Therefore if emancipation is possible, its path passes through the subject: ―If [the subject] were
liquidated rather than sublated in a higher form, the effect would be regression—not just of
consciousness, but a regression to real barbarism‖ (SO, 140).278In comparison with this, when
277 His ambivalent assessment of Benjamin‘s writings could then be brought to bear on the risk which I concede here
attaches to Foucault‘s experience: ―Before his Medusan glance, man turns into the stage on which an objective
process unfolds. For this reason [his] philosophy is no less a source of terror than a promise of happiness,‖ Prisms
(The MIT Press, 1983) p. 235; see also p. 233
278 I emphasize ―up until now in history‖ in order to convey the same point at which I hinted earlier when I called in
question a ―transcendental‖ reading of Adorno‘s immanent critique. It is beyond the scope of this study to fully
justify the stakes of this claim, so I only note that: Adorno does not (or cannot) inscribe subject-object mediation
which underwrites his argument here in any structure resembling the Kantian interpretation of the a priori nor its
historical transposition through a Hegelian appeal to a teleological conception of history. The appeal to universality
and necessity, which would be the essential features of such a move, were he to make it, cannot be grounded through
immanent critique. Hence his appeal to the context of social practices and the qualification of his ―theses‖ by
history. It is this dimension which brings his ―negative dialectics‖ in proximity to Foucault‘s ―historical ontology‖:
because the sublation of which he speaks cannot be achieved through a determinate negation, his ―dialectic‖ cannot
be assimilated to its Hegelian version—at least to its standard interpretation; just as, because Foucault refers
theoretical and practical formations to historical practices which constitute them, his ―ontology‖ cannot be
assimilated to its Heideggerian version—at least to its standard interpretation. Adorno‘s playing off of Kant against
232
Foucault claims that the individual ―whom we are invited to free, is already in himself the effect
of a subjection much more profound than himself‖ (DP, 30), he could be charged with pulling
the rug from beneath his own feet, to the extent to which it seems to eliminate the only normative
ground he could claim for a critical history.
Moreover, the dialectical process of mediation which Adorno calls experience can
account for its ―withering‖ under conditions of advanced capitalism and administration without
denying that it could be undiminished: ―Not least to blame for the withering of experience is the
fact that things, under the law of pure functionality, assume a form that limits contact with them
to mere operation, and tolerates no surplus, either in freedom of conduct or in autonomy of
things, which would survive as the core of experience, because it is not consumed by the moment
of action.‖279Or again, the disappearance of the continuity of historical experience turns it into ―a
timeless succession of shocks, interspaced with empty, paralyzed intervals,‖ and thereby
destroying the space of experience as ―the lag between healing oblivion and healing recollection‖
(Ibid, 54).
Thus, it is not that Adorno opposes a diminished experience to an undiminished one, but
rather, in accordance with the structure of mediation, he argues for an internal connection
between a form of subjectivity which would preserve the moment of reflective unity without
denying its dependence on alterity, or its own inherent non-identity, and its historical distortion
through mechanisms of domination. In the light of this ―withering,‖ when Foucault approvingly
identifies the effective concept of experience in ―Nietzsche, Bataille, and Blanchot‖ as having
the ―function of wrenching the subject from itself, of seeing to it that the subject is no longer
Hegel, and vice versa, is mirrored by Foucault‘s playing off of Heidegger against Marx, and vice versa; and both
moves are strategic.
279 Adorno, Minima Moralia: Reflections on a Damaged Life, (New York: Verso, 2005) p. 40.
233
itself, or that it is brought to its annihilation or dissolution‖ (EW, 3:241), he could only be
ratifying a social process that has already accomplished such a ―desubjectivation‖ (Ibid).
Therefore, what one could call Adorno‘s ―subjects without experience,‖ which can only
be discerned reflectively through the aporetic thinking of what experience has become, confronts
Foucault‘s ―experience without subjects‖ (EW, 3: 248), which articulates the limits constituted
by historical practices. And from the perspective of immanent critique, Foucault‘s appeal to
―limits‖ appears as either a hypostatization of contemporary social relations, i.e. as an
ontological affirmation of the reification of subjectivity; or an ungrounded appeal to immediacy,
i.e. as the affirmation of a pure essence grasped in intuition.280For if the subject is entirely
constituted by power, and since constitution presupposes some kind of subject, power itself
seems to become a metasubject; and if the ―normalization‖ of subjects in modernity through
psychiatric or incarcerational practices is bad in some sense, this could only be called by that
name on the basis of what these subjects would be in the absence of such normalization or
distortion.281 Theoretical justification and normative grounding, then, appear to be unavoidable,
if Foucault‘s histories are also going to be critical; and the price he pays for giving up the
transcendental is either full adaptation to the very coercive mechanisms he describes, or the loss
of the critical dimension by no longer speaking the language of reason.
280 For example, Foucault‘s claim to be writing a history of madness itself in the 1961 Preface to the History of
Madness; or the enigmatic appeal to ―a different economy of the body and its pleasures‖ (HS, 159) ; but also see FR,
pp. 296-297.
281 Different versions of these charges are formulated in Habermas, Philosophical Discourse of Modernity, op. cit.;
Charles Taylor, ―Foucault on Freedom and Truth,‖ in Foucault: A Critical Reader (Wiley-Blackwell, 1991)
(hereafter FCR).
234
4.22 Subjects, Objects, and Mere Plants
I want to take up the discussion of ―subjectivation‖ I introduced above and complicate it by
incorporating the force of these objections, which I believe acquire urgency when framed by
Adorno‘s negative dialectics and its proximity to and difference from historical
ontology.282When Foucault introduces ―subjectivation‖ in the context of his History of Sexuality,
what he is interested in are ―the conditions in which the human being ―problematizes‖ what he is,
what he does, and the mode in which he lives‖ (UP, 18)This is then taken as constituting the
domain of ―ethics‖ (UP, 36-45). He distinguishes this from ―morals,‖ which, as a broader
category, studies the moral code comprising the rules which specify what is and is not permitted,
what has positive or negative value, what is tolerated or shunned. In relation to this code, it is
possible to study people‘s actual behavior as it conforms to this code, or not; or one could offer
philosophical justifications for the rules contained in it. Unlike the former, which can be called a
―sociology of morals,‖ and the latter, which is moral philosophy,283Foucault is interested in the
way in which the individual constitutes herself as a moral subject of her own actions.
Ethics as ―relation to self‖ comprises four dimensions284: First, ―ethical substance,‖
which is that part of oneself or one‘s behavior taken as the relevant domain for ethical judgment:
for instance, whether the emphasis is placed on feelings, desire, intention, or reason. Second,
―the mode of subjection,‖ which is the way in which someone is incited or invited to recognize
her moral obligation: for instance, whether obligation is conceived as divine law, rational
282 Part of my argument takes up the one I formulated in relation to Habermas in chapter three, part two.
283 On this distinction, see Arnold Davidson, ―Archaeology, Genealogy, Ethics,‖ FCR, p. 228.
284 ―Rapport à soi‖; see UPfr, passim, but in particular, pp. 13, 40, 44, 239.
235
imperatives, or social habit. Third, ―practice of self,‖285 which is the type of work the individual
is to perform on herself in order to become an ethical subject: for instance, the kind of asceticism
implied by self-deciphering examination in monastic communities, or self-help books in
contemporary societies. Fourth, ―telos,‖286 which is the kind of being to which we aspire when
we behave morally: for instance, purity, immortality, or self-mastery.
Therefore, Foucault inscribes ―relation to self,‖ the reflexive dimension which I claimed
above as motivating his discrimination between subjection and subjectivation, within the
fourfold articulation of ethical substance, mode of subjection, self-forming activity, and telos of
the self. What this articulation allows is the comprehension of how the same prohibitions may
give rise to different types of ethical self-constitution, or of the different types of activity and
passivity entailed by similar moral codes.
I want to advance the following claim: When Foucault defines experience as the
correlation between fields of knowledge, types of normativity, and forms of subjectivity, and he
specifies the last dimension as subjectivation; the reflexive relation to self thereby articulated on
the dimensions of knowledge and power is all one can (or should) ask for from a discourse that
claims to be both historical and critical. This point is at once simple and complex:
If now I am interested…in the way in which the subject constitutes himself in an active fashion,
by the practices of self, these practices are nevertheless not something that the individual invents
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. (FF, 11)
Experience, in this configuration, becomes the name for epistemic and practical engagements of
the subject with the object. But the ―opposition‖ between activity and passivity, which I pursued
285 ―Pratique de soi,‖ or self-forming activity; see UPfr, pp. 12, 42, 44, 84.
286 ―Téléologie du sujet moral‖; see UPfr, pp. 39, 54.
236
throughout this chapter, cannot be resolved in favor of one term over the other. Foucault‘s
strategy is to write the history of the determinate relations between subjects and objects, and
those relations, as constituted by discursive and nondiscursive practices, is the structure of
experience.
However, Foucault‘s reference to the subject and the object is not motivated by a
metaphysical belief in their primacy; it rather involves a strategic negotiation. So when he
defines thought as ―the act that posits a subject and an object in their various possible relations‖
(CC, 314), and continues to argue that ―[t]hought is not what inhabits a certain conduct and gives
it its meaning; rather, it is what allows one to step back from this way of acting or reacting, to
present it to oneself as an object of thought and question it as to its meaning, its conditions and
its goals‖ (EW, 1: ); he is including the possibility of a critical subject in the very definition of
experience. What thereby emerges is neither unique nor novel, to the extent that the capacity to
―step back‖ from one‘s activity—to acquire a minimum distance with respect to it—in order then
to interrogate it, is an old philosophical chestnut. What is interesting in this formulation is that no
ground for this capacity is admitted other than the very context of practices in which the critical
subject is always already enmeshed. And ―mesh‖ is an apt metaphor because the effort to give
primacy to either activity or passivity, in order to provide a theoretical foundation and a
normative ground for one‘s considered judgments, not only results in contradictions of its own,
but also results in an illusory conviction of critical efficacy.
It could be said, however, that this reference to practices still skirts the issue: either it
simply replaces other foundational terms, such as discourse, or power, or regime, or dispositif—
237
in which case it reproduces the confusion between the transcendental and the empirical;287 or it
simply refuses to enter the space of reasons—in which case there is no reason why anyone
should take Foucault‘s histories seriously. It appears as if Foucault cannot even escape between
the horns of this complex dilemma, to the extent to which the disjunction seems exhaustive of
the possibilities: it is categorial for modernity. I argued for one way in which the disjunction
may be denied at the end of the previous chapter. Here I want to push the difficulty even further,
and I think it goes as far back as (at least) Aristotle. The passage is wonderful, and it is worth
quoting at length.
[H]e whose subject is being must be able to state the most certain principle of all things. This is
the philosopher, and the most certain principle of all is that regarding which it is impossible to be
mistaken; for such a principle must be both the best known…and non-hypothetical….It is, that
the same attribute cannot at the same time belong and not belong to the same subject in the same
respect; we must presuppose, in the face of all dialectical objections, any further qualifications
which might be added….For it is impossible for anyone to believe the same thing to be and not to
be, as some think Heraclitus says; for what a man says, he does not necessarily believe….Some
indeed demand that even this [principle] shall be demonstrated, but this they do through want of
education. For it is impossible that there should be demonstration of everything: there would be
an infinite regress, so that there would still be no proof. But if there are things of which one
should not demand demonstration these persons cannot say what principle they regard as more
indemonstrable than the present one….We can, however, demonstrate negatively even that this
view is impossible, if our opponent will only say something; and if he says nothing, it is absurd to
attempt to reason with one who will not reason about anything, in so far as he refuses to reason.
For such a man, as such, is seen already to be no better than a mere plant.288
It is sufficient, then, that someone say anything, for the principle of noncontradiction to
be demonstrated negatively: even the denial that the principle of noncontradiction is true would
287 For example, ―power‖ would be the transcendental condition of possibility, and ―biopower‖ or ―pastoral power,‖
would be its empirical and historically varying content.
288 Aristotle, Metaphysics IV. 3, 1005b8-34, IV. 4, 1006a5-22 (Complete Works, vol 2 (Princeton University Press,
1984)
238
be tantamount to presupposing its truth. Therefore, in any debate with reason, one has already
taken the side of reason, unless one decides to be ―a mere plant‖. So much so that, Aristotle
seems to deny the very possibility of anyone‘s actually being able to deny the principle. There is
a slippage or shift which occurs about midway through the first half of the passage: the principle,
which concerns properties of things (―subjects‖), is applied to beliefs about things. If Heraclitus
in fact affirmed the contrary of the principle, as ―some think,‖ then he could not have really
believed it, ―for it is impossible for anyone to believe the same thing to be and not to be‖.
Therefore, one can say or do something determinate only if one has either affirmed or denied
something; and if one denies the principle of noncontradiction, he would have to both affirm and
deny that very denial.289Everyone knows this—if she knows anything at all—but it is the
philosopher who can state it.
Against the background of this ―negative demonstration,‖ Foucault appears to be either a
plant, or, if he insists that he is saying something, he could not really believe what he says—or
he says what he does not really believe. Granted that this dilemma imposes itself with this
peculiar force on its opponent: just why must it be a question of being ―for‖ or ―against‖ the
principle of noncontradiction?290In other words, why should referring the validity of such a
principle to the context of social behavior in which it originates and operates be equivalent to its
denial?
There are, of course, the familiar and plausible arguments against psychologism and
sociologism. But I do not think that Foucault‘s invocation of a ―fundamental experience,‖
289 In fact, Aristotle makes this point with even more force: ―It follows that all would then be right and all would be
in error, and our opponent himself confesses himself to be in error—And at the same time our discussion with him is
evidently about nothing at all; for he says nothing. For he says neither ‗yes‘ nor ‗no,‘ but both ‗yes‘ and ‗no‘; and
again he denies both of these and says ‗neither yes nor no‘‖ (Ibid., IV. 4, 1008a8-33).
290 Foucault‘s grappling with this issue in the context of a reading of Kant and the Enlightenment is in PT, p. 110.
239
comprising practices, entails either of these positions: he is not reducing validity to mental states,
and he is not reducing it to ―collective states‖. The claim is rather that validity is only ever
formulated in statements, and statements are only ever produced in socially and historically
determinate contexts. In other words, the production of statements that could become candidates
for validity is a social practice and cannot be understood independently of the needs and desires
of those who issue them; but furthermore, those very needs and desires are themselves
interpreted through historically transformable practices. This means only that validity cannot be
situated outside the processes of subjection and objectification I described above.
If one generalizes from this relation between validity and practices to questions of
objectivity, it could be said that neither the constitution of the subject nor the formation of the
object can be placed outside the conditions of their emergence. And a Foucauldian rejoinder to
the dilemma would be that only on the basis of external assumptions about objectivity can one
insist on its inevitability. For instance, both the motivation and the ultimate support for
Aristotle‘s negative demonstration is his substance-attribute metaphysics. That is to say, his
conception of reality constrains his conception of thought, and vice versa: it is the belief in their
mutually grounding harmony which sustains Aristotle‘s effort. More generally, insistence on
absolute objectivity no less than on absolute subjectivity must account for how such an absolute
could ever be recognized by empirical subjects. From this perspective, it is precisely a concept
such as ―regime of truth‖ which avoids questionable appeals to immediacy.
If my argument is successful, it establishes not that questions of rationality, or even of
truth, are only a function of the brute effectivity of physical constraints, but that the statement
and evaluation of objectivity, as an act of thought, presuppose criteria of what counts as
objective (or rational). In chapter two I argued for one way in which such criteria can be
240
understood. Here I only want to emphasize that these criteria are neither a function of individual
psychology nor sociology—as theoretical disciplines, these themselves presuppose just those
criteria which they claim to ground—but are immanent in the very practices which they define.
And that means: it is not necessary to insist that the only sense which objectivity can have
is one where it has an independent constitution, requiring nothing but secondary recognition and
reflective ratification. Foucault may not be able to justify or refute a principle such as that of
noncontradiction; but neither can Aristotle: its demonstration, as demonstration, presupposes the
very conclusion it claims to justify. If one insists, however, that Aristotle is not simply begging
the question, since the question is not one of offering a positive demonstration for the principle,
then it could be argued, a fortiori, that Foucault is not refuting what we take to be our most basic
principles. In fact he is trying to show precisely why they have such a strong grip on us. And to
that extent he is ―accounting for‖ exactly that which remains unaccounted in any insistence on
absolute objectivity or subjectivity.
Therefore, when Foucault claims that ―the exercise of power creates objects of
knowledge, makes them emerge….The exercise of power perpetually creates knowledge, and
inversely, knowledge entails effects of power‖ (DE, 2: 753), what is at stake is the production of
true knowledge, and not knowledge that would be false on that account. Moreover, his claim is
not that ―power‖ as such is bad, for the simple reason that ―power as such does not exist‖ (EW,
3: 336). He is only interested in how ―intentional and nonsubjective‖ (HS, 94) power relations
structure and restrict the field of what can be done, which does not imply that they determine
what will or must be done. So, for instance, when Beatrice Han argues that ―The Foucauldian
analysis of subjectivity… [oscillates] in a contradictory manner, between a definition of
subjectivity as ‗self-creation‘ [and] the need…to go back to the practices of power of which
241
subjects are not masters and are usually unaware‖ (BTH, 172), this is because she insists that
subjectivation and subjection can only be understood as absolute activity and absolute passivity,
respectively. But if the two poles are mediated by discursive and nondiscursive practices, then
the insistence appears as a false dilemma.
I want to take one example from Foucault‘s description of how apparently secondary and
empirical techniques of notation, registration, and statistical measurement were effective in
constituting the ―individual‖:
One is no doubt right to pose the Aristotelian problem: is a science of the individual possible and
legitimate? A great problem needs great solutions perhaps. But there is the small historical
problem of the emergence, toward the end of the eighteenth century, of what might generally be
termed the ‗clinical‘ sciences, the problem of the entry of the individual (and no longer the
species) into the field of knowledge; the problem of the entry of the individual description, of the
cross-examination…into the general functioning of scientific discourse. To this simple question
of fact, one must no doubt give an answer lacking in ‗nobility‘: One should look into these
procedures of writing and registration…and a new type of power over bodies. [It is] in these
‗ignoble archives‘, where the modern play of coercion over bodies, gestures and behavior has its
beginnings. (FR, 202-203)291
Foucault‘s interest, then, is not in the individual as the putative raw material for molding
through subjection; and when he adds that subjectivation concerns the ―forms and modalities of
the relation to self by which the individual is constituted and recognizes himself as subject‖ (UP,
12), he is not positing the existence of an ahistorical substratum which would assume different
historical shapes according to mechanisms of subjection and modes of subjectivation. The claim
291 In relation to Adorno‘s thesis concerning the ―withering of experience,‖ it should be noted that where Adorno
sees a decline in individuality through more intensified adaptation to the anonymity of administered social and
economic life, Foucault sees progressive individuation through a different type of power relations. Adorno‘s claim
of decline, however, is that of the spontaneity of subjects, and that is compatible with what Foucault is claiming
here. But the Foucauldian rejoinder is that, the diagnosis of this state of affairs does not imply the postulation of
what spontaneity itself, outside the normalizing mechanisms of modern disciplines, might be.
242
rather entails that there is no unconditioned ground or foundation for the different selfinterpretations
through which individuals problematize what they say and do in historically
restricted fields of possible actions. Subjectivity so construed is, not unlike most hermeneutic
conceptions, relative and without an absolute center. But unlike most hermeneutic conceptions,
no appeal to a deep meaning or truth of the subject, which would always be anticipated in
advance and which would always occlude itself, is at work.
However, even these oblique ways in which the possibility of a critical history, or
―critical ontology of ourselves‖ (FR, 50), is motivated leave one somewhat dissatisfied. One
rhetorical device on which I relied in order to motivate that possibility and its intelligibility
betrays that unease: several times, in order to discount through acknowledgement a particular
implication suggested by some Foucauldian claim, I wrote ―At first blush this may appear as….‖
But why blush in the first place?
The reason is that, it seems almost self-evident that any talk of domination presupposes a
prior definition, however indeterminate, of freedom: in the absence of something spontaneous,
neither domination nor freedom seems to have sense.292 Therefore, Foucault‘s refusal to
distinguish between ideology and legitimate knowledge, or the absence of a principled
distinction between good power and bad power, seems to undermine any critical thrust his
histories could have. Adorno, faced with a similar difficulty, opts for an immanent critique which
strives to hold up the idea of reason itself to its historical development as instrumental
rationality, without pretending that this idea could be secured through the regulative ideal of
communicative rationality.
292 And Foucault does want to draw a distinction between power relations and domination: ―The problem is…to
acquire…the practice of the self, that will allow us to play [the] games of power with as little domination as
possible‖ (EW 2: 298).
243
Rather, his admission of a qualified naturalism—only accessible through the negative
mediation of the subject by the object and vice versa—which struggles to make visible the
dominated (suppressed) moment of non-identity (sensuous particularity) through the
contradictions which that suppression itself generates, at least lodges his critique in the refuge
that contradictions provide. And since this means that resistance, however attenuated, is always
possible, ―the standpoint of redemption‖ (MM, 247) in face of despair provides some normative
justification for the claims of critique, while accepting, without subterfuge, the critic‘s
complicity with what she criticizes. Hence the inimitable and perpetual oscillations between her
responsibility and her guilt, possibility and impossibility, exemplified in Adorno‘s own writings:
To gain [critical] perspectives without velleity or violence, entirely from felt contact with its
objects—this alone is the task of thought. It is the simplest of all things, because the situation
calls imperatively for such knowledge, indeed because consummate negativity, once squarely
faced, delineates the mirror image of its opposite. But it is also the utterly impossible thing,
because it presupposes a standpoint removed, even though by a hair‘s breadth, from the scope of
existence, whereas we well know that any possible knowledge must not only be first wrested
from what is…but is also marked, for this very reason, by the same distortion and indigence
which it seeks to escape. (Ibid.)
But what if not even ―consummate negativity‖ reflects the mirror image of its opposite
(any more), and what if the absolute, negative or otherwise, elicits only a yawn, even when the
word is uttered with ―no doubt‖?293 The predicament in which Foucault finds himself is perhaps
more similar to this than that which negative dialectics articulates. However, the flipside of this
―deflation‖ of the absolute, if that is what it is, is a paradoxically more sanguine assessment of
actual possibilities. In relation to Habermas‘s ideal speech situation, Foucault claims that: ―I
293 Samuel Beckett, ―Endgame,‖ Dramatic Works, The Grove Centenary Edition, (New York: Grove Press, 2006) p.
93
244
have always had a problem insofar as he gives communicative relations…a function that I would
call ‗utopian‘. The idea that there could exist a state of communication that would allow games
of truth to circulate freely, without any constraints or coercive effects, seems utopian to me. This
is precisely a failure to see that power relations are not something that is bad in itself, that we
have to break free of‖ (EW 1: 298). It is therefore not a question of deflating utopian notions
because of their impracticability—that would amount to misunderstanding the very critical
function for the sake of which they are invoked; rather, it is motivated by the recognition that a
society without power relations would not be a society with unconstrained intersubjectivity, but
the total elimination of all meaningful activity.
At the end of the passage I quoted from Minima Moralia, Adorno states: ―[B]eside the
demand thus placed on thought the question of the reality or unreality of redemption itself hardly
matters‖ (MM, ibid, my italics). Granted that Foucault‘s difference from the insistence on the
refuge which the standpoint of redemption provides does not stem from its reality or unreality;
still, how can he make sense of that demand itself, since it is an inherently normative concept?294
It is perhaps the peculiar force of this question which accounts for Foucault‘s inscription
of his ―critical ontology‖ in the lineage of Kantian critique and the Enlightenment. For Kant had
―found it necessary to deny knowledge in order to make room for faith‖ (CPR, Bxxx). The three
questions, which it is the task of critical philosophy to answer, are: What can I know? What must
I do? What may I hope? (CPR, A804-5/B832-3). I delineated what the Kantian answer to the first
question is, and how it conditions and sustains modern attempts to articulate the activity and
passivity of the subject, in chapter one. However, Kant himself goes beyond the perspective
294 I offer one part of a possible answer to this question in the conclusion; here I focus on one of its dimensions in
order to bring the discussion of this chapter to a conclusion.
245
opened therein. The antinomies establish the impossibility of transcendent knowledge; but the
compulsion under which the subject seeks such knowledge is not referred to its irrationality or
ignorance: ―our unquenchable desire to find a firm footing beyond all bounds of experience‖
(CPR, A 796/B824) is inscribed into the very constitution of reason. Moreover, the very
possibility of moral value itself seems to require a standpoint beyond possible experience. A
brief attention to the Groundwork for the Metaphysics of Morals is sufficient to adumbrate the
basic dimensions of this requirement relevant to what I want to say here.295
For Kant all rational-activity is rule-governed activity. What this means is that we need to
distinguish between arbitrary ―performances‖ from genuine actions. A genuine action is one
which is done for the sake of an end; it is action in view of an end. This also means that the agent
is able to formulate his action in terms of a rule. A maxim, in Kantian parlance, is precisely such
a rule: it is the subjective principle which articulates the different components of an action and
serves as the basis of one‘s actions. The question Kant asks himself is what makes an action have
moral worth. And his answer, rather bluntly put, is that the agent must choose to be guided in the
selection of her ends by her desires.296 Therefore, the requirement of reflexivity which Kant
imposes on experience finds its parallel in his conception of practical agency.
For an action to have moral worth, the agent must make it her maxim, say, to tell the truth
because it is her duty to do so; or because it is what is required of her as a rational moral agent;
or again, because it makes a claim on her. Such an agent is said to act for the sake of duty, or
from duty. It can in fact be the case that by telling the truth the agent may be accomplishing other
ends, such as her happiness or that of the community, but the important thing is that if the action
295 Immanuel Kant, Groundwork of the Metaphysics of Morals (Cambridge University Press, 1998) (hereafter
GMM).
296 See GMM, pp. 11-16 and 19.
246
is to have moral worth, then it must be done for the sake of duty: the agent chooses to tell the
truth, because she conceives of it as a moral obligation.297
The upshot of this insistence on reflective agency is Kant‘s claim that if we are to have
genuine moral obligations, then we must conceive of ourselves as autonomous beings.298 And the
invocation of autonomy is comprehensible as a consequence of Kant‘s Copernican revolution,
which I traced out in chapter one. If critical philosophy debars positive theoretical knowledge of
God as unavailable for beings of discursive intellects and sensible intuitions, then traditional
religious grounds for introducing moral value into the world can no longer have force. By the
same token, no immediate intuition of value survives the disenchantment of critique. Moreover,
since Kantian nature is only the formal unity of representations, where that unity is a function of
mechanical causality (as a category of the understanding), objective teleology, that is, conceiving
the existence of natural purposes independently of the constitution of subjectivity, also falls
under the critical bar.
Only a ―utilitarian‖ understanding of morality would seem to be available to the human
animal in a world thus flushed of its own normativity. But Kant, as always, surprises one‘s
expectations by arguing that what matters in morality is not happiness but acting so as to become
worthy of happiness.299 Since any appeal to mechanical determination of the will by the causality
of desires I happen to have can only be a posteriori—and therefore contingent and particular; and
since the very concept of duty implies that it is unconditional (hence a priori, which implies its
necessity and universality); action, if it is meaningful at all, can only be conceived under the
condition of freedom as self-determination.
297 Ibid., p. 15
298 Ibid., p. 55
299 See CPR, A808/B836; GMM, pp. 12-14 and 19-20.
247
It is therefore, and again, a question of the foundational division of the transcendental and
the empirical which motivates and justifies Kant‘s reconfiguration of what constitutes morality.
Nothing empirical is sufficient to generate the unconditional necessity of the moral law, because
the content of experience exhibits only a series of conditioned objects. And it is precisely the
confusion of the transcendental with the empirical which condemns all attempts to justify
morality either transcendently—by invoking God, say—or empirically—by appealing to
contingent desires subjects happen to have.
Justification then, to follow the line traced out in chapter one, is based not on the object—
some independent conception of the good—but the subject—the reflexive capacity of the will. I
may value many things: health, wealth, good-looks, etc. But if I conceive of myself as
determined solely and immediately by means of such empirical ends—Kantian ―inclination‖—
then I would be no different from a leaf suspended on a branch which cannot act otherwise, once
the relative strength of forces acting on it is given: ―pull of gravity is greater than push (up by the
branch)‖ equals ―I fall‖. Therefore what the Kantian transcendental turn opens up is the necessity
entailed by how a good will must be determined: the rational agent, if ―she‖ is to be rational, or
an agent, must abstract from all personal desires and inclinations as possible motives for ―her‖
actions, and focus on the thought of duty alone as the sole determining factor for ―her‖ will.
When I tell the truth not because of some desire or inclination (even ones we would think
of as legitimate ones such as the desire to make others happy), but because I take truth-telling to
be what is morally required of me, as making a demand and claim on me, this is like taking truthtelling
as a law. And since I have abstracted from all personal desires in the formulation of my
maxim, it has the form of a law, valid for everyone. What this means for the motive of action, for
what it is in such a case that ―moves‖ me to tell the truth is: the thought that my maxim has the
248
form of a law is the sole incentive for my action. And that is acting solely out of respect for the
law.
From the perspective of the transcendental standpoint—that which such a law, the
―categorical imperative,‖ presupposes—it matters little whether freedom as self-determination is
in fact actual: the reality of transcendental freedom is a ―merely speculative question, which we
can leave aside so long as we are considering what ought or ought not to be done‖ (CPR,
A801/B829). Whether I am in fact free or not, in the sense implied by this account of morality,
the categorical imperative is valid for me as rational being.
I want to emphasize the degree to which Kant‘s introduction of moral value and the
good—normativity beyond the effectivity of physical constraint or utility—through the moral
law (and not vice versa), is bound up with the necessity to hold the transcendental and the
empirical standpoints apart.300 The sublime moment of Kantian ethics is that, though it is made
possible and necessary by the critical limits imposed on theoretical knowledge—which can only
be of phenomena—its necessity requires the postulation of what transgresses those limits—the
postulation of noumena. And the very ―empty‖ impersonality of the categorical imperative,
regardless of whether my freedom is actual, is what makes rational reflection an expression of
freedom.
Kant is offering an independent conception of and justification for morality—
independent of both religion and the pursuit of happiness. His view is based on a notion of
human dignity that is a function of the human being as a rational free agent.301 According to this
300 This is probably not a controversial claim about Kant, but it is important to the intelligibility of the Foucauldian
position I have been pursuing.
301 See GMM, p. 23
249
conception most of our desires (inclinations)—including the one for happiness, which in fact can
be seen as a blanket term for all our empirical ends—are discounted, or rather, are rendered
irrelevant because a moral law cannot be binding for a rational free agent simply in virtue of the
contingent desires she happens to have. Thinking of one‘s self as deciding under the sway of
such desires turns such a self into the leaf that gets pushed and pulled in every which way by the
forces acting on it now and again. It is precisely our capacity to abstract from such desires, and
make it our maxim whether or not to be determined by them, that opens up the space for
freedom.
Schematically put, Kant is saying: you are not simply determined to act in this or that
way because you happen to have a desire to achieve this or that; you choose to make it your
maxim to be so determined.302 To use truth-telling as an example again: it is not that you tell the
truth simply because you happen to want to contribute to communal happiness; you choose to
make communal happiness the kind of end requiring your action. This is what makes us
autonomous beings for Kant. And autonomy so understood becomes self-determination, selflegislation,
or again giving the law to oneself.
The categorical imperative tells you that you must act only in accordance with a maxim
that can be universalized, i.e. that can be seen as a universally valid law binding for all rational
beings. It is precisely this capacity, for Kant, that makes the moral law something which is
legislated by the rational agent ―herself‖: the law on the basis of which we act is not found
written outside on some granite block by God; and it is not laid down by our corporeal nature,
which imposes the force of this or that desire; but, as the categorical imperative, it is written by
302 For a good account of what the radical reflexivity of Kant‘s concept of autonomy entails (in the framework of
Lacanian psychoanalysis), see Alenka Zupancic, The Ethics of the Real: Kant, Lacan (Verso, 2000).
250
us as rational beings. Therefore, allowing myself to be determined by any other end is
heteronomy, that is being determined by something external, other.
It was already Hegel who leveled the charge that Kant‘s ethics lacks any substantial
content.303 The objection is that no substantial conclusions about how to act follow from a formal
principle of rationality: a purely rational will, so completely isolated from all corporeal desires
and concrete circumstances of action, will have no basis for making any decisions about how to
act. Why, we may wonder, Kant so radically deprives the will of any determining motivation
other than respect for the law; why would anybody regard any action that is not done strictly for
the sake of duty, but rather in some way motivated by other desires, as inherently suspicious,
lacking genuine moral worth?
The answer is that, since, for Kant, all empirical desires are contingent and subjective,
only by reflectively detaching myself from them, that is detaching myself from my present
interests and concerns as this particular individual, can I attain an objective, universal
standpoint—the only standpoint that can generate objective laws binding for all rational agents.
And what the categorical imperative tells us is precisely to regard ourselves as one moral agent
among others, untainted by the subjective coloring of our desires and interests.
The upshot of the objection, then, is to insist on the fact that, from such a detached
perspective, one will have no motivations left for acting in this rather than that way. However,
one could argue, it is precisely the formality, and hence the substantive emptiness of the
303 For a defense of Kantian formalism transposed into a linguistic, communicative register by Habermas, see
―Morality and Ethical Life: Does Hegel‘s Critique of Kant apply to Discourse Ethics?‖ Moral Consciousness and
Communicative Action, op. cit. In chapter three, I offer an objection to the Habermasian strategy involved here
without necessarily impugning its formalism. In this section, however, I am interested in the formal aspect of the
transcendental/empirical divide, as it relates to the problem of justification and its standpoints.
251
categorical imperative that Kant considered to be the supreme achievement of his conception of
ethics.304 Then a Kantian defense might be that the point is not to generate motivations, but to
offer rational justification for the ones we already have, from a perspective that is outside these
motivations.305 It is precisely this concern with justification (and more broadly, legitimation)
which would condemn the Foucauldian definition of ethics—as the articulation of ethical
substance, mode of subjection, self-forming activity, and telos of the self—as hopelessly
empirical, and therefore contingent: not only can it offer no self-grounding, thereby establishing
the universality and legitimacy of some fundamental principles, it cannot even make a principled
distinction between right and wrong.
But perhaps one should reverse the terms of the argument: either it is not possible to offer
such justifications from the objective standpoint—because it is so abstract—or the putative
objective standpoint will smuggle precisely those motivations into its perspective which it ends
up endorsing, in which case it would be self-deception, though a comforting one, to consider it
objective.
The Kantian insistence on the constitutive divide between the transcendental and the
empirical is set up to secure the self-grounding of reason that is critique. That self-grounding
depends on: first, the invocation of the conditions for the possibility of a unified, implicitly selfconscious
subject of thought and action; and second, on the essential unity and teleology of
304 Habermas takes just this achievement to be important in the essay referred to above, passim. The stronger version
of the claim would be: it is precisely the formality of the moral law which makes the concrete richness of
substantive identities possible in modernity—as opposed to, say, the stifling imposition of societal roles in antiquity
or medieval Europe.
305 Christine Korsgaard, Sources of Normativity (Cambridge University Press, 1996) See pp. 12-13, 15, and 61.
252
reason.306In the light of what I said above, I want to claim that Foucault provides us with the
conditions of actuality of self-relation without self-grounding. This is what accounts for both his
proximity to and difference from the Kantian lineage:
[W]e have to give up hope of ever acceding to a point of view that could give us access to any
complete and definitive knowledge of what may constitute our historical limits. And from this
point of view, the theoretical and practical experience we have of our limits and of the possibility
of moving beyond them is always limited and determined….But that does not mean that no work
can be done except in disorder and contingency. (FR, 47)
Therefore, the exclusive and one-sided emphasis on the problems of justification and
legitimation occludes another, no less important question: ―…if the Kantian question was that of
knowing what limits knowledge has to renounce transgressing…the critical question today has to
be…in what is given to us as universal, necessary, obligatory, what place is occupied by
whatever is singular, contingent, and the product of arbitrary constraints?‖ (FR, 45). And I want
to add that the import of this question is ethical to the extent to which it refuses to accept the
terms of the challenge according to which suffering will matter only in so far as it is
argumentatively justified.
The ―historical ontology of ourselves‖ (Ibid.), contrary to what one might expect from an
ontology, is an ethics. The articulation of the criterial discursive and nondiscursive practices
which determine the historically specific limits of what we think and say and do, is also the
opening up of the space of freedom (as possible transgression). If, however, the charge is that the
sheer contingency of the space thereby made possible is devoid of any normative foundation, and
hence cannot even justify why and which suffering should matter, the Foucauldian rejoinder is:
306 CPR, Axiii: ―In fact pure reason is such a perfect unity that if its principle were insufficient for even a single one
of the questions that are set for it by its own nature, then this [principle] might as well be discarded, because then it
would not be up to answering any of the other questions with complete reliability.‖
253
suffering should not require justification. If one persists in demanding a justification for the
force of this ―should not‖, on account of the constitutive divide between the transcendental and
the empirical, then the response would be: no amount of argumentation in the world could pick
up the slack generated by that divide. And if one does not already accept the non-transcendental
necessity of that ―should not‖, ethics could only be the (sometimes subtle and sometimes crass)
cruelty of the imperative: Justify your despair!
254
They said to me, Here‟s the place, stop, raise your
head and look at all that beauty. That order! They
said to me, Come now, you‟re not a brute beast,
think upon these things and you‟ll see how all
becomes clear. And simple! They said to me, What
skilled attention they get, all these dying of their
wounds.
And the horizon? Nothing on the horizon?
What in God‟s name could there be on the horizon?
(Beckett, Endgame, pp. 150, 113)
Limit-Experience: Hegel (without the Absolute) Again?
In defense of the historical function of Foucault‘s work, given its problematic status in relation to
the standards of academic historiography, Garry Gutting claims that the History of Madness
could be called a ―Phänomenologie des Kranken Geistes,‖ and adds: ―not entirely facetiously.‖307
I want to advance that comparison entirely with a straight face. Two factors immediately speak
against such a proposal: First, Foucault‘s explicit references to Hegel in general, and the
Phenomenology of Spirit in particular, are few and far between; and when he does thematize
Hegel and ―the dialectic‖, his remarks are almost always negative.308Second, his methodological
reflections tend to privilege qualities that are opposed to the properties commonly attributed to
307 Gutting, ―Foucault and the History of Madness,‖ Cambridge Companion to Foucault, op. cit., p. 66
308 For his relatively less hostile but still sketchy discussions of the Phenomenology of Spirit, see L‟herméneutique
du sujet, cours au Collège de France. 1981-1982 (Gallimard/Le Seuil, 2001), pp. 30, 467, 505-506; HMfr pp. 69,
437, 642-643, 659. His negative remarks in general make me think that when Foucault says ―Hegel‖ or ―dialectic‖,
most of the time he is thinking ―Sartre‖ and ―historical materialism‖. It is no wonder that he was influenced by
Althusser, whose interpretation of Marx excoriated the humanist appropriation of Hegel: see, Louis Althusser, For
Marx (Verso, 2006).
255
dialectical thought: spatial and not temporal, contingent and not necessary, empirical and not
transcendental.309
But Foucault also writes that any distance traveled in relation to Hegel ―assumes that we
are aware of the extent to which Hegel, insidiously perhaps, is close to us; it implies a
knowledge, in that which permits us to think against Hegel, of that which remains Hegelian. We
have to determine the extent to which our anti-Hegelianism is possibly one of his tricks directed
against us, at the end of which he stands, motionless, waiting for us‖ (AK, 235). I think
Foucault‘s sensitivity to this difficulty is motivated by more than his desire to pay tribute to Jean
Hyppolite.310
My defense of Foucault‘s account of experience in terms of rule-governed practices,
following the logic of the analytic of finitude and the demands of the Agrippan trilemma,
involved two strategies: a) since it is not possible to place oneself explicitly outside the space of
reasons without thereby (and implicitly) situating oneself inside it, I tried to show how the
concept of experience need not be construed in terms of the absolute separation of the inside
from the outside, b) I claimed that insistence on this dynamic as a transcendental requirement
occludes dimensions of practices that become visible through a Foucauldian account. My
description of what I take to be Hegel‘s transposition of ―deductive‖ function into history and his
309 Which is why, in one sense, it is so easy to (dialectically) co-opt Foucault. In Bataille‘s succinct expression, it is
impossible to escape the Hegelian system, since the system makes the absence of a relation into a relation—and
Bataille himself tries to find a ―language‖ for ―inner experience‖that will not immediately devolve into the Hegelian
immediate (L‘expérience intérieure [Gallimard, 2002] pp. 96, 127-145). Moreover, because opposition is the
paradigmatic relation in dialectical thinking, the series of binary terms I list here would, if we follow the logic of
sublation, become partial moments of an ultimate synthesis. It is this logic that I want to render problematic.
310 The citation is from Foucault‘s inaugural lecture at Collège de France, The Order of Discourse, and Foucault was
replacing Hyppolite‘s former chair.
256
move from self-consciousness to Spirit would then seem to follow squarely these two strategies.
And therefore Foucauldian genealogy would appear Hegelian Erinnerung by other means.311
What, then, are the differences that make a difference? The claim I want to advance in
conclusion is that experience is already limit-experience, and what that entails is intelligible
through its articulation in terms of rule-governed criterial practices, which I have tried to
develop. Moreover, I think that the critical thrust of Foucault‘s historical discourse lies therein.
That critique is intimately connected to a worry about and the demarcation of limits is not
news: if Kant is right to frame the question of experience in terms of our rights, rather than its
fact, then nothing empirical or transcendent can limit reason‘s claims from the outside; therefore
the only legitimate philosophical path—one which would go beyond common sense, without
messing about with neurons, as Rorty says312—becomes that of reason‘s self-critique. The
reflexivity of conscious experience, which Kant presupposes and seeks to vindicate
transcendentally, determines from the outset the direction that critique will take: the problem of
knowledge becomes that of self-knowledge. Hence the peculiar circularity of transcendental
arguments, by means of which we vindicate our rights to that which we already (take ourselves
to) possess in fact. However, if this movement does not merely beg the question, it is because
―possible experience,‖ the centrality of which I emphasized in chapter one, turns out to be the
limiting concept.
311 My discussion of Foucault‘s division of his ―methodologies‖ over the course of his trajectory avoided their strict
separation—as archaeology, genealogy, and problematization—in favor of their articulation in terms of discursive
and nondiscursive practices. In conclusion, I will use ―genealogy‖ as the name for this enquiry which seeks to trace
the contours of an ensemble of historically constitutive and therefore criterial practices.
312 See chapter 1, introductory and concluding remarks, of the present study.
257
The techniques of examination, then, which occupy Foucault so centrally in Discipline
and Punish,313 are not too far removed from what for Kant is the essence of critique: the selfexamination
of reason, whereby its claims concerning that which falls inside the limits of
possible experience are justified, but only at the price of renouncing its rights on what falls
outside such limits. And transgression is to be revealed and punished by the same movement
which reason must but cannot forego, through the tangle of contradictions any claim to
knowledge beyond possible experience generates. The theme of self-reference is thus inscribed
at the core of the critical project. The Critique of Pure Reason is the tribunal, and it is reason
itself which issues the verdict on its own limits. And as antinomial, the contradictions that accrue
to transgression indicate not that reason comes up against something other which it cannot grasp,
but that reason is at odds with itself because it seeks ―firm footing‖ there where there are no
grounds.314
One must be cautious, however, not to move too quickly from the centrality of
examination to the problem of freedom and domination. Just as something in our experience
makes us suspicious toward the intricate balancing act through which Kant links up our freedom
to the status of our epistemic and practical rights as rational beings of a certain kind; we must
pose the question why to what Foucault says in relation to a key moment in the history of
madness, which will be repeated in different configurations in so many limit-experiences: ―Freed
from the chains that had ensured that it was a pure object of the gaze, madness was paradoxically
stripped of its essential liberty, which was that of solitary exaltation; it became responsible for
what it knew of its truth, and was imprisoned in its own gaze, which was constantly turned back
313 DPfr, pp. 201, 217ff, 220, 260-264
314 Kant says: ―[our] unquenchable desire to find a firm footing beyond all bounds of experience‖ (CPR, A
796/B824), and he writes about ―the humiliation reason feels‖ when its pure use yields no positive results and when
it therefore ―requires discipline‖.
258
on itself, finally chained to the humiliation of being an object for itself‖ (HM, 499). Why should
it be humiliating to be my own object, when an entire philosophical tradition locates the very
exercise of my freedom there?
It is difficult to answer this question, not least because it concerns the nodal point where
questions of epistemology and questions of morality intersect. Perhaps, by way of getting an
initial traction on it, one should reverse the question and take up again the line I traced in chapter
one from Kant to Hegel: Why should it be liberating, or an expression of autonomy, to determine
oneself? Or rather, why does freedom find its definition in autonomy? The short answer, which I
developed at the end of chapter four in Kantian terms, is that, according to the idealist conception
of subjectivity, nothing else could count as its determining ground without its having already
determined it as such: i.e., nothing could count as a reason unless the subject takes it as a reason,
on pain of heteronomy, or determination through another (or any other).
However, and already in the idealist conception itself, autonomy so understood comes
under a certain amount of strain which it cannot contain, since self-relation is possible only
through relation to some other. The Hegelian rejoinder I adumbrated expresses the pressure
which the subjects‘ placement in a social situation and in relation to one another creates. If selfconsciousness
presupposes recognition, and if genuine mutual recognition requires a genuinely
universal basis of recognition, then Spirit must be conceived as a community of mutually
recognizing individuals.315
315 PS, pp. 104-111.
259
In Hegel‘s language, this transformation of self-consciousness points towards an ―I that is
a We and a We that is an I,‖ that is, Spirit316. And the fundamental institutions of such a
community will be legitimate, because they will embody universal recognition. Hence the force
of Hegel‘s transformation of the Kantian opening: cognitive experience is itself a social
institution, and rationality is ultimately explicable in terms of recognition. The subject of
experience, which determines for itself its own fundamental practices, becomes Spirit, i.e., a
collective, socially self-determining, self-realizing subject.
If, as Hegel says, ―[self-consciousness] can achieve satisfaction only when the object
itself effects the negation within itself; and [if] it must carry out this negation of itself in itself‖
(PS, 109), the dialectical relation thereby articulated expresses a necessary presupposition of
(individual) subjectivity: the object cannot ground the subject, because its negation entails its
destruction, therefore only another subject can ground the subject, since it alone can potentially
negate itself without destroying itself. And recognition or acknowledgement is precisely the
Hegelian name for this non-destructive self-negation. Moreover, its necessity and universality is
underwritten by the determinate negation of the alternative moves available to agents who must
confront the implications of their sociality. Therefore, Hegel, if successful, would achieve an
immanent grounding of subjectivity and the legitimacy of its epistemic and practical activities,
not through a denial of its dependence on some other, but precisely by insisting on the mutual
implication of dependence and independence.
Phenomenology then is the narrative of the strategies that seek to provide a satisfactory
relation to self in and through its relation to another. And Hegel‘s conception of determinate
316 Ibid., pp. 110.
260
negation expresses how these strategies must fail for internal reasons, and in so doing, bring to
light the conditions required for full satisfaction. One moment is key in relation to Foucault: if
we take the strategies to be essentially and basically relations of power, Hegel‘s ―looking on‖
already appears to have anticipated their emergence and disappearance: to the extent that ―I‖
succeed in wresting recognition from ―my‖ other without recognizing it in return, ―I‖ lose the
very basis of recognition. Only as freely offered can acknowledgment do the work which mere
consumption of an external object could not; but a coerced recognition soon devolves into mere
consumption. That is, I lose myself along with my other.
From this perspective, the ―humiliation of being my own object‖ that Foucault describes
in the above quotation must also be a criticism of one possible mode of self-relation; and as
criticism, it must explicate and defend its own presuppositions. Furthermore, it cannot do this
without at the same time adumbrating what could be in some sense a better form of self-relation.
And it must do this because Hegel seems to have anticipated even that strategy which would
refuse it, as when he provides the famous description of the Stoic figure: ―Whether on the throne
or in chains, in the utter dependence of its individual existence, its aim is to be free, and to
maintain that lifeless indifference, which steadfastly withdraws from the bustle of existence,
alike from being active as passive, into the simple essentiality of thought.‖317
317 PS, p. 121 And of course Hegel‘s claim that such a shape of consciousness could only appear in a time of
universal fear and bondage and culture finds its echoes in not uncommon interpretations of the mood of post-war
Paris. One could insist that Foucault, in the passage at issue, is only describing a particular configuration, namely the
birth of the asylum in contrast with Classical confinement, and that his concern is precisely that asymmetrical
relationship between psychiatrist as medical authority and the patient as object of knowledge (HM, 499f). However,
the question then becomes why that description should have any value beyond its own boundaries, and more
importantly, why that relationship should seem paradigmatic of epistemic and practical interactions in modern
societies at large.
261
Foucault the Modern stoic? Perhaps. In any event, a more fruitful way of reading
Foucault‘s claim passes through his stated aim of conducting a ―critical ontology of ourselves as
a historico-practical test of the limits we may go beyond…as work carried out by ourselves on
ourselves as free beings‖ (FR, 47, my italics). This ―critical ontology‖ is a genealogy because ―it
will separate out, from the contingency that has made us what we are, the possibility of no longer
being, doing, or thinking‖ (Ibid, 44) the ways in which we do. And such a 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‖ (Ibid, 50, my italics).
The critical ontology of ourselves, then, which Foucault proposes at the end of his
trajectory, resonates with the history of limits he proposed at its beginning:
We could write a history of limits—of those obscure gestures, necessarily forgotten as
soon as they are accomplished, through which a culture rejects something which for it
will be the Exterior; and throughout its history, this hollowed-out void, this white space
by means of which it isolates itself, identifies it as clearly as its values. (HM, xxix)
To interrogate a culture about its limit-experiences is to question it at the confines of
history about a tear [déchirure] that is something like the very birth of its history. (Ibid.)
I want to underline three terms, namely, critical work, genealogy, limit-experience, and
briefly discuss each one as the site of a contestation between Foucault and Hegel.
When Foucault identifies his histories as part of ―a critical ontology of ourselves‖—as
work which ―we perform on ourselves‖—he might be already within the ambit of the progression
Hegel narrates in the Phenomenology of Spirit. For the very principle of that narration is the
labor of the negative, which, moreover, consciousness is said to perform on itself, to the extent
that ―[c]onsciousness provides its own criterion from within itself, so that the investigation
262
becomes a comparison of consciousness with itself‖ (PS, 53); and therefore, ―consciousness
suffers this violence at its own hands: it spoils its own limited satisfaction‖ (Ibid., 51). This
constitutes the main ―methodological‖ improvement of Hegelian phenomenology over against
what he calls ―argumentation,‖ which knows only how to say ―[t]hat something is not the case,‖
and hence remains only negative insight: ―In order to have a content once again, something new
must be taken over from elsewhere….[and therefore] this insight is the negative that fails to see
the positive within itself‖ (PS, 36).
Speculative thought, however, accounts for the very negativity of its object, and thereby
goes beyond it: ―the negative belongs to the content itself, and is the positive, both as the
immanent movement and determination of the content, and as the whole of this process‖ (Ibid).
Determinate negation, therefore, is the positive result emerging from the process of negation.318
In less lofty terms, every determinate claim presupposes some fundamental criterion of what is
really there to judge, or what counts as right action, or again, some implicit presupposition about
what any claim-making involves; but, short of the absolute standpoint, each such presupposition
generates contradictions or, if we ascribe to Hegel a weaker conception of dialectical transitions
than he at times seems to want, at least inconsistencies, for purely internal reasons.
Or, even more bluntly put, our classifications of nature and identifications of ourselves
change on the basis of the internal failure of previous categorizations of the same: new criterion
presupposes internal inconsistency of the previous ones, which presuppose undischarged
318 See also p. 51, where Hegel argues to the effect that skepticism which fails to attain the speculative standpoint
and takes its result as merely false, ―pure nothingness,‖ abstracts ―from the fact that this nothingness is specifically
the nothingness of that from which it results. For it is only when it is taken as the result of that from which it
emerges, that it is…the true result; in that case it is itself a determinate nothingness, one which has a
content….[Hence] in the negation the transition is made through which the progress through the complete series of
forms comes about of itself [my emphasis].‖
263
assumptions, which eventually lead to ultimate consensus as the only basis for objective
justification. Therefore, there are no meta-rules grounding our first order rules, but rationality is
self-correcting.
Thus Hegelian phenomenology achieves the truly presuppositionless standpoint, not by
excluding every claim which admits of the slightest doubt, and not by reducing every claim to
the immanence of transcendental subjectivity, but by including, in advance and exhaustively, the
totality of presuppositions. It displays the immanent rationality of Spirit by articulating the
possible strategies through which subjects aim at full satisfaction and fail.
Failure, then, is already accounted for within Hegel‘s account of what would count as
success. The conceptual revisions—where ―conceptual‖ refers to a criterial level—which
consciousness ―performs,‖ are the self-revision of thought because their principle is the internal
discrepancy between what consciousness takes as its standard of objectivity and the object itself.
What consciousness implicitly presupposes as its standard of objectivity is not identical with
what it takes to be its objects, and so long as that identity, which the absolute standpoint knows
to be the identity of subject and object, is not achieved, each shape of consciousness will exceed
itself and lead to another one with new content, because it is already divided from itself.
One could say, therefore, invoking Adorno, that critical work, understood dialectically
―is the consistent sense of non-identity. It does not begin by taking a standpoint. My thought is
driven to it by its own inevitable insufficiency, by my guilt of what I am thinking‖ (ND, 5); but
add that, for Hegel, the labor of the negative as speculative is precisely the accumulation of all
shapes of consciousness. What makes absolute knowledge absolute, that is unconditional, is just
its inclusion of all conditions that fall short of grasping the identity of subject and object. Hence
264
the overcoming of dissatisfaction through mutual recognition is achieved when we come to
realize our complicity with it: if we look rationally at the world, the world will look rationally
back.319 The immense power of Hegel‘s thought drives from his refusal to make cognitive and
practical satisfaction a question of immediate intuition—say, of some one thing called the
absolute—rational or otherwise, but the sublation (Aufhebung) of one-sided and internally
limited partial standpoints.
Therefore, the absolute standpoint is complete not because there is nothing more to learn
empirically, but because it is the very inclusion of exclusion, i.e., the recognition that mutual
recognition is all the ground necessary and sufficient to legitimate the practices, cognitive or
otherwise, of self-realizing and self-determining subjects. That is why the dialectic of
dependence and independence I discussed in chapter one ―ends‖ only when the ―slave‖ becomes
free-citizen through the negativity of work. And the work of history is the achievement of such a
fundamental like-mindedness between subjects who recognize one another as recognizing each
other.
So when Foucault claims that ―[t]he idea of bios as a material for an aesthetic piece of art
is something which fascinates me,‖ (FR, 348) and that it might be possible to articulate a relation
to oneself where the main task would be ―to build [my] existence as a beautiful existence‖ (Ibid,
319 Hegel, Lectures on the Philosophy of World History (Cambridge University Press, 1981) p. 29 The full
translation by Nisbet reads: ―Whoever looks at the world rationally will find that it in turn assumes a rational aspect;
the two exist in a reciprocal relationship.‖
265
354), he might be giving voice to the ―beautiful soul‖320 denounced by Hegel, precisely because
an aesthetics of existence would shun the power of the negative:
But that…what is bound and actual only in its context with others, should attain an
existence of its own and a separate freedom—this is the tremendous power of the
negative….Death, if that is what we want to call this non-actuality, is of all things the
most dreadful, and to hold fast to what is dead requires the greatest strength. Lacking
strength, Beauty hates the Understanding for asking of her what it cannot do. But the life
of Spirit is not the life that shrinks from death and keeps itself untouched by devastation,
but rather the life that endures and maintains itself in it. It wins its truth only when, in
utter dismemberment, it finds itself. It is this power...only by looking the negative in the
face, and tarrying with it. This tarrying with the negative is the magical power that
converts it into being. This power is what we [call] Subject [as] that being or immediacy
whose mediation is not outside of it but which is this mediation itself. (PS, 19) (My
italics)
I want to claim that what Foucault‘s invocation of an aesthetics of existence involves is
not a denial of or ―shrinking from‖ negation, rational or otherwise, but the denial of the negation
of negation. That is, Foucault refuses to grant negativity a primacy in how we come to revise our
classifications and identifications of the world and of ourselves. The reason why ―aesthetics‖ so
understood nonetheless involves a negative moment is that it is inseparable from
problematization: ―Problematization doesn‘t mean representation of a pre-existing object, nor the
creation by discourse of an object that doesn‘t exist. It is the totality of discursive and nondiscursive
practices that introduce something into the game of the true and the false and
constitute it as an object for thought‖ (PPC, 257). That is, constitute it as an object for moral
reflection, scientific knowledge, and political analysis. Therefore, ―[t]he work of the history of
320 See PS, pp. 383ff, 400, 406-407
266
thought would be to rediscover at the root of…diverse solutions the general form of
problematization that has made them possible‖ (EW, 1:118).
So, for instance, it is a question of determining how sexuality becomes an object of
knowledge and that domain where we seek the truth about ourselves, when in Greek and Roman
antiquity it occupied, more or less, the same level as that of diet and health. Two points need
underscoring: First, Foucault‘s ―return to the Greeks‖ is not in order to retrieve some form of
ideal or new paradigm from what would be the genuinely free practices of antiquity (relative to
the disciplinary aspect of modernity). Foucauldian ―history of the present‖ stretches further back
in time because the dimension of self-relation in experience, and not that of power and
knowledge, becomes the explicit object of enquiry later on his trajectory. Therefore, antiquity
becomes interesting only to the extent that the organization of social life is not as rigidly
organized by disciplinary and normalizing discursive and non-discursive practices.
In this sense, Deleuze is on the mark in his ―negations‖ when he says that:
―A process of subjectivation, that is, the production of a way of existing, can‘t be equated
with a subject, unless we divest the subject of any interiority and even any identity.
Subjectivation doesn‘t even have anything to do with a ―person‖: it is a particular or
collective individuation characteristic of an event….It is a mode of intensity, not a
personal subject. It is a specific dimension without which we can‘t go beyond knowledge
or resist power….‖ (Deleuze, Negotiations, pp. 98-99, modified)
Subjectivation aims to trace the contours of ―some event‖ which cannot be that of the interiority
of subjectivity, whatever its modalities. But when Deleuze continues to say that subjectivation is
a question of the constitution of ourselves as a self ―beyond knowledge and power,‖ the claim
goes awry in its assertion of the positivity and possibility of a region ―beyond power and
267
knowledge‖.321In that sense, there is nothing beyond power and knowledge. But rather than
ontologizing that nothing as the Dasein, or for-itself, or transcendental subject, Foucault inserts it
into the interstices of the articulation of knowledge and power, as that self-relation which is not
sufficiently consistent or substantial for it to provide a self-grounding.
Second, Foucault‘s problematization of Greek antiquity does not constitute a return
because his concern is not to establish a metahistorical continuity: ―What must be grasped is the
extent to which what we know of [the generality of problems in a tradition]…constitute[s]
nothing but determinate historical figures, through a certain form of problematization that
defines objects, rules of action, modes of relation to oneself‖ (FR, 49). Therefore, it is not a
question of anthropological invariants or chronological variations; nor is it a ―deduction‖ or
justification of transformations. In that sense, it differs from Hegel‘s interest in Greek antiquity.
When Hegel reads Sophocles, for instance, he is concerned with showing how the Greek ethical
life had to collapse under the weight of its own contradictory presuppositions, because it could
not account for and provide the free individual subjectivity which it nonetheless demanded.322
When Hegel approvingly quotes Sophocles to the effect that ―[b]ecause we suffer, we
acknowledge we have erred‖ (PS, 284), this forms one of the dialectical reversals through which
Spirit is led to a progressively more inclusive and adequate conception of what individuality
demands; that is, to a more satisfactory conception of itself. Foucault‘s ―genealogy of the modern
subject,‖ however, is not interested in legitimation, which is the primary concern for Hegel‘s
recourse to history. The realization of freedom through the dialectical implication of the
321 I think Deleuze is partially aware of this in ―Desire and Pleasure,‖ FI, pp. 183ff. Also see Foucault (University of
Minnesota Press, 1988) p. 94ff In line with what I have said about ―exprerience,‖ I propose a somewhat different
interpretation of ―Foucault as cartographer,‖ which Deleuze discusses on p. 23ff
322 See, among others, PS, pp. 284f.
268
universal and the particular, as Hegelian experience, is simultaneously the retrospective
justification of the path Spirit has traversed. Since the changing relation between subject and
object requires that the subject redefine who it is and what its object is in response to its failed
attempts at relating itself to its other, experience is at once a recognition and a misrecognition.
And the negative thrust of this dialectical movement accentuates the failure of any isolated
moment to arrive at a satisfactory coherent self-understanding and articulation.
However, the teleological development of Hegelian experience means that speculative
discourse, which articulates the positive moment of this process, comes to full objectivity only
when experiences are given an appropriate narrative order in memory. Narrative recollection, the
Hegelian Erinnerung, exemplified by the path traversed by consciousness in the Phenomenology
of Spirit, is open to contingency, while retrospectively and at the same time imposing a closure.
It is therefore the sublation of contingency and necessity.
The narrative recollection which is the culmination of experience is also a collective
recollection. Hegel moves away from Kant‘s transcendental subjectivity, with its formal
interpretation of categorial conditions, toward the achievement of some fundamental ―likemindedness‖
as the condition of knowledge. We thereby shift away from a transcendental
deduction toward an account of the genesis whereby such like-mindedness is achieved as a
result. And it is this genetic account that bears the weight of justification, since we can no longer
appeal to transcendental conditions. This move then entails the reconfiguration of the
problematic of self-consciousness. Hegel‘s claim is that we must be able to reconstruct why
participants in intersubjective situations would come to take this or that practice as constraining
possible judgments about objects, facts, and legitimate political institutions.
269
The necessity of each transition between different shapes of consciousness is a function
of his reconfiguration of objectivity as Wirklichkeit: the practice at issue will have been
established if our reconstruction displays a rational inevitability in the development of such
progressive self-consciousness. This reconfiguration transforms the pursuit of cognitive
experience in general, and knowledge in particular, into participation in a social practice that is
rule-governed, collective, and teleological. In order to evaluate the rationality of such practices
one must consider such self-consciously held criteria as social norms. And retrospectively
justified criteria will constitute the basis for mutual recognition.
There is a sense in which Foucault‘s historiography, the History of Madness, say, would
be a Phenomenology of Spirit without the last chapter; or a sense in which The Order of Things,
for instance, would be a ―socialized‖ Science of Logic, listing the unfolding of the most
fundamental concepts structuring social reality—actuality: not ―Being, Nothing, Becoming…‖
but ―Resemblance, Order, History….‖ That would be Hegel‘s infinite proximity to any inquiry
which historicizes the transcendental and conceives of knowledge and action on the basis of
social institutions, i.e., as social practice. But genealogy‘s refusal of the standpoint of
legitimation also constitutes an infinite distance to the extent that it is not so much a historicizing
of the transcendental as its renunciation. History then becomes ―the concrete body of becoming;
with its moments of intensity, its lapses, its extended periods of feverish agitation, its fainting
spells; and only a metaphysician would seek its soul in the distant reality of an origin‖ (EW, 2:
373).
If ―truth or being lies not at the root of what we know and what we are but the exteriority
of accidents‖ (Ibid., 374), then no amount of retrospection will pick up the contingency of the
moment of emergence and convert it to a rational necessity: ―As it is wrong to search for a
270
descent [Herkunft] in an uninterrupted continuity, we should avoid accounting for emergence
[Entstehung] by appeal to its final term; the eye was not always intended for contemplation, and
punishment has had other purposes than setting an example. These developments may appear as
a culmination, but they are merely the current episodes in a series of subjugations‖ (Ibid., 376).
This ―hazardous play of dominations‖ (Ibid., 376-377) takes place in a non-place,
because there is no common place where the adversaries confront one another. However, I think
it would be a mistake to take such an invocation of the ―endlessly repeated play of dominations‖
as a celebration of violence: it rather designates the unavailability of a ground, i.e., an epistemic
or practical reason, which could justify or serve as a basis for mutual recognition. In other words,
it should be understood as framing our interactions by a concept of interpretation that is deprived
of the recourse to an absolute standpoint: there are no criteria for our criteria, but only further
practices; that is, only interpretations of interpretations.
Historical inquiry then becomes ―wirkliche Historie‖ (Ibid., 379f): It refuses totalization
and closure; it does not seek subjective recognitions or reconciliation; and it gives up teleological
imposition of unity. But perhaps the best way to describe the stakes of ―effective history‖ is:
―The dead. The body count. We don't like to admit the war was even partly our fault 'cause so
many of our people died. And all the mourning's veiled the truth. It's not ‗lest we forget,‘ it's ‗lest
we remember.‘ That's what all this is about — the memorials, the Cenotaph, the two minutes'
silence. Because there is no better way of forgetting something than by commemorating it.‖323
The genealogist, then, is the one who will say ―That‘s what all this is about,‖ or: ―[he]
will know what to make of this masquerade…he will push [it] to its limit and prepare the great
323 Alan Bennet, History Boys: A Play (Faber & Faber, 2006)
271
carnival of time where masks are constantly reappearing. No longer the identification of our faint
individuality with the solid identities of the past, but our ‗unrealization‘ through the excessive
choice of identities‖ (EW, 385-386). And so it will be an ―experimentation on ourselves‖ (Ibid.,
388).
But if history displays only the endless installation of ―[violence] in a system of rules and
thus proceeds from domination to domination,‖ what sense could one give to ―That‘s what all
this is about,‖ and how can one understand the call to experimentation with the limits that define
us, without contingency collapsing into mere arbitrariness? Does not the very concept of rule
rule out experimentation so understood? In short, is it only a question of will?
Foucault‘s giving up of the transcendental standpoint is not a repudiation of reflection; it
is rather motivated by the conviction that the moment of self-relation entailed by reflection
cannot be anchored in any unreflected given. And that includes appeals to immediate intuition, or
will, or power, where these are taken as merely given to thought from the outside. There is no
unconditioned standpoint, transcendental or empirical, which would provide a refuge for thought
outside of the practices constitutive of its forms. And that also implies the inseparability of the
forms of thought from what is thought. Therefore, the abandonment of transcendental reflection
is at the same time a radicalization of self-reflection.
If we simply insist on the separation between the transcendental and the empirical as
constitutive for thought as such, then Foucault‘s histories must conform to the following schema:
a hyper-empiricism, which historicizes all content, and a hyper-transcendentalism, which blindly
posits a transcendental form (episteme, power, dispositif, regime, etc.). And the latter would be
272
worse than any we find in German idealism, since it is cut off from any reflective relation to the
subject (hence its blindness).
Against this insistence, one must offer a different type of reflection on limits and rules,
and their normative hold on how we think and speak and act. I have attempted to show how this
different type of reflection is not the opposite of what would be its transcendental counterpart;
and I have tried to understand why it nonetheless appears, from the transcendental perspective, as
mere stammering or stumbling, that is, somehow self-defeating. You cannot respond to Zeno by
walking from here to there; or set up an actual race between the tortoise and Achilles—though
that would be fun; or again, refute Berkeley by kicking a stone.324 But you can formulate
alternative ways of thinking about experience such that it would already be limit-experience.
Kant turns geographer in a beautiful section of the Critique of Pure Reason titled ―On the
impossibility of a skeptical satisfaction of pure reason that is divided against itself‖325:
The sum total of all possible objects for our cognition seems to us to be a flat surface,
which has its apparent horizon, namely that which comprehends its entire domain and
which is called by us the rational concept of unconditioned totality. It is impossible to
attain this empirically, and all attempts to determine it a priori…have been in vain. Yet
all questions of our pure reason pertain to that which might lie outside this horizon or in
any case at least on its borderline. (CPR, A759-760/B788-789)
―The famous Hume,‖ Kant continues, ―one of these geographers of human reason,‖
―[expelled these questions] outside the horizon of human reason, which however he could not
determine‖ (Ibid.). Hume, then, is at fault for merely limiting ―our understanding without
324 A wonderful presentation of these paradoxes are: ―The Perpetual Race Of Achilles and the Tortoise‖ and ―A New
Refutation of Time,‖ in Jorge Luis Borges, Selected Non-Fictions (Penguin, 2000).
325 CPR A758-769/B786-797, where it is a question of ―the discipline of pure reason in polemical use‖ no less.
273
drawing boundaries for it,‖ (A767/B795) whereas the critique of pure reason should prove from
principles ―not merely the limits but rather the determinate boundaries of [itself]‖
(A761/B789).326
Hume is right when he claims that the judgment according to which the sunlight‘s
illumination of the wax also melts it, could only be learned from experience; but he goes awry
when he ―falsely infer[s] from the contingency of our determination in accordance with the law
the contingency of the law itself, and he [confuses] going beyond the concept of a thing to
possible experience [which is a priori]…with the synthesis of the objects of actual experience,
which is of course always empirical‖ (A766/B794). In other words, that it is the sun‘s warmth
which melts the wax is contingent and particular, i.e., empirical; but that something must have
preceded its melting in accordance with a constant law is necessary and universal, i.e., a priori.
Because skepticism fails to appreciate this distinction, it can never be ―itself satisfying for
questions of reason, but [it is] preparatory for arousing its caution and…securing it in its rightful
possessions‖ (A769/B797).
The clearest expression of how bounds differ from limits is found in Prolegomena to Any
Future Metaphysics:327 ―Bounds (in extended beings) always presuppose a space existing outside
a certain definite place and inclosing it; limits do not require this, but are mere negations which
affect a quantity so far as it is not absolutely complete. But our reason, as it were, sees in its
326 Kant calls the accomplishment of this drawing of boundaries the third step of pure reason, where the first step is
its ―dogmatic childhood‖ and the second step is skeptical, ―sharpened by experience‖; the third step, however,
―pertains only to the mature and adult power [männlichen] of judgment.‖ It would be interesting to trace out this
attachment of the dialectic to ―maturity,‖ since Plato too is careful to restrict it by age and experience (see Republic,
trans. G. M. A. Grube (Hackett Publishing Company, 1992) p. 205 What is curious is that Kant explicitly contrasts
this drawing of boundaries with what would be the ―censorship of reason,‖ when Plato‘s appeal is in tandem with
the ―need‖ for censorship.
327 Kant, Prolegomena to Any Future Metaphysics (Hackett Publishing Company, 1977), ―On the Determination of
the Bounds of Reason,‖ 350-365 (I cite the marginal pagination.)
274
surroundings a space for the cognition of things in themselves, though we can never have
determinate concepts of them and are limited to appearances only‖ (352). Therefore, ―in all
bounds there is something positive‖ (354), and we know this because ―transcendental ideas have
urged us to approach [the limits], and thus have led us…to the spot where the occupied space
(viz., experience) touches the void (that of which we can know nothing)‖ (Ibid.). The bounds of
pure reason, then, as limits of pure reason in a positive sense, can be conceptually determined
and known.
Critique is that activity of thought which already leads to that ―point or line of contact‖
(Ibid., 353) between the inside and the outside, and therefore, already in Kant, the determination
of our rightful claims to knowledge require spatial metaphors, to the extent that the very
reflexivity of reason implies talk of an inside and an outside. In other words, Kant is already a
cartographer, since he recognizes the necessity of mapping the boundaries of knowledge. If such
a mapping is possible, however, experience, in some sense, already points beyond itself, not
because we could have experience (sensuous intuition) of that which transcends experience, but
because our reason inherently forms concepts of totalities of conditions. The bounds of
experience cannot be grasped from the outside—since that space is void, uninhabited, or
inhabited only by illusory monsters—but we can know them from the inside. How then do I
place myself on the borderline?
For Kant this requires an intricate balancing act—and it is possible to see the beginning
of the interminable oscillations of the analytic of finitude here, at the limits of critique. I want to
mention two moments: First, ―[e]xperience, which contains all that belongs to the sensory world,
does not bound itself; it only proceeds [from one conditioned thing to another]…But the setting
of a boundary to the field of experience…is still a cognition which belongs to it even at this
275
point, and by which it is neither confined within the sensible nor strays beyond the sensible, but
only limits itself..to the relation between what lies beyond it and what is contained within it‖
(Ibid., 360) In other words, experience does not tell me where it ends—its boundary is not
empirical—but I can know a priori that and where it ends.
Second, I can know the outside only from the inside because the space of experience is
constituted by our categories, where ―we‖ are rational beings with discursive intellects and
sensible intuitions. Even though the inside and the outside, the transcendental and the empirical,
touch only liminally, they are anchored in the double status of man as transcendentally
determining and empirically determined, that is, as subject and object of knowledge. Hence
Kant‘s separation of the transcendental and the empirical standpoints as two irreducible
dimensions of human finitude is linked up with the anthropological theme, to the extent that each
standpoint is both attached to and distinct from its counterpart.
Kant says: here are the edges of the map, and outside there are only monsters. But the
necessity of the antinomies is inseparable from the drawing of boundaries: ―metaphysics leads us
toward bounds in the dialectical attempts of pure reason (..stimulated thereto by the nature of
reason itself)‖ (Ibid., 353, my italics). The Hegelian rejoinder is not far off: ―The main point that
has to be made is that antinomy is found not only in the four particular objects taken from
cosmology, but in all objects of all kinds, in all representations, concepts, and ideas‖ (EL, 42)
What this implies is:
The being that is kept firmly distinct from the determinacy, being-in-itself, would be only
the empty abstraction of being. In being-there the determinacy is one with being and is at
the same time posited as negation; this determinacy is limit….Thus, otherness is not
something-indifferent outside it, but its own movement. In virtue of its quality, something
276
is first finite and secondly alterable, so that the finitude and alterability belong to its
being. (EL, 148)
The limit, therefore, is not something external to what it limits, to the extent that as a
determinate something, it is what it is only within, that is, on the basis of its limit. From this
perspective, then, the Kantian attempts at inscribing critique within the stable framework of a
series of dualities—concept/intuition, intelligible/sensible, etc.—is futile, since liminality
permeates existence. Limit, therefore, is both constitutive of what it limits and its negation. It is
dialectical, i.e., as constitutive, it is a nothing that is; or again, something is in itself the other of
itself.328
Dialectical thought internalizes the transgression of limits as alteration—one form of
which will be conceptual revision—because ―the inner contradiction with which being-there is
burdened from the start, and which drives it beyond itself‖ ensures that ―[s]omething becomes an
other, but the other is itself a something, so it likewise becomes an other, and so on ad infinitum‖
(Ibid., 149). But as speculative, the infinity implied by this movement is distinct from
―spurious/negative‖ infinity, which is the infinite progression of ―ought‖ that is at the core of
Kantian—and Fichte‘s—―moral progress,‖ the asymptotic approach to fulfilling the commands
of duty—with happiness deferred to the afterlife—or the infinite extension of scientific
knowledge—without ever knowing ―things in themselves‖. (Ibid, 150-152)
Genuine infinity, however, which it is the task of speculative logic to articulate, is not this
movement from one mediation to its other and back, ―one damn thing after another,‖ as one
might say; rather, it ―remains at home with itself in its other‖ (Ibid.) The teleology of reason, in
328 And lest someone begs to differ, Hegel adds: ―Those who are too squeamish toward the finite achieve nothing
real at all, but remain in the realm of the abstract and peter out‖ (Ibid.)
277
its historical transposition, is this movement of Spirit, through figures of its self-alienation, until
it arrives at the satisfaction of self-reconciliation. It is in this sense that Hegel can confidently
claim that ―every genuine philosophy is idealism‖ (Ibid, 152), to the extent that genuine
satisfaction—epistemological, practical, and aesthetic—is not possible unless one comes to
accept the ideality of every finite determination.
Moreover, this movement is rationally motivated, to the extent that it is the achievement
of a ―we‖: a community of mutually recognizing free individuals, who determine for themselves
and through the inadequacy of their past practices, a shared form of life, the institutions of which
structure the practical unity of their social existence. There is no other to community so
understood, since its very historical achievement is supposed to have already accounted for
anything that could even count as its other. If, as Hegel says, ―something is first finite and
secondly alterable, so that the finitude and alterability belong to its being‖ (EL, 148); and if,
moreover, it is precisely for that reason that it can be infinite; then, otherness as such cannot be
comprehended on the basis of an exclusion: exclusion is always already inclusion; such is the
trick of reason.
I want to situate Foucault‘s limit-experience on this line, but ―beneath the sun of the great
Nietzschean quest, [to] confront the dialectics of history with the immobile structures of the
tragic‖ (HM, xxx).329The strange ―histories‖ Foucault writes, then, are the site of this
329 ―Limit-experience‖ itself has a convoluted history, and its interpretation in relation to Foucault usually elicits
comparison with Karl Jaspers and his concept of Grenzsituationen, (Psychologie der Weltanschauungen [Berlin:
Springer, 1919], cited in Eugene Webb, Worldview and Mind: Religious Thought and Psychological Development
[University of Missouri Press, 2009] p. 15; Battaille and ―inner experience,‖ (L‟experience interieure [Gallimard,
1978]). These comparisons often draw on Foucault‘s own brushes with ―limit-experiences‖: a not very successful
discussion is in James Miller, The Passion of Foucault (Harvard University Press, 2000), and a more nuanced and
careful discussion can be found in Martin Jay, ―The Limits of Limit-Experience: Bataille and Foucault‖ in Cultural
Semantics: Keywords of Our Time (University of Massachusetts Press, 1998) pp. 62-79. Jay also mentions the
invocations of similar ―experiences‖ during the inter-war period in Germany, specifically Ernst Jünger‘s Kampf als
278
confrontation: ―…in the history of madness [sic.] I was investigating the way in which a culture
can determine in a massive, general form the difference that limits it, I am concerned [in The
Order of Things] with how…a culture experiences the propinquity of things, how it establishes
the tabula of their relationships and the order by which they must be considered‖ (OT, xxiv).
And he can write those histories because ―[f]rom the limit-experience of the Other to…the
conceptions of the Same, what is available to archaeological analysis is…the threshold that
separates us from Classical thought‖ (Ibid.)
When Foucault invokes Borges as having motivated The Order of Things, and starts by
quoting the ―Chinese encyclopedia,‖ which divides animals into ―(a) belonging to the Emperor,
(b) embalmed, (c) tame…(k) drawn with a very fine camelhair brush, (l) et cetera, (m) having
just broken the water pitcher…‖ (OT, xv)330, this is because the fundamental question of a
history of the same concerns the ground on (the basis of) which we establish the validity of our
classifications, as ―when we say that a cat and a dog resemble each other less than two
greyhounds do, even if both are tame or embalmed, even if both are frenzied, even if both have
just broken the water pitcher‖ (OT, xix). What justification do we have when we designate that
―table of categories‖ impossible?
What Borges‘s ―fable‖ shows, in the guise of a merely exotic taxonomy, are the limits of
our own classifications, the network of necessities according to which we cannot think that. This
innere Erlebnis (1922), p. 73. Heidegger, of course, has something to do with all such occurrences of ―limitexperience‖
in the twentieth century; see, for instance, his discussions of anxiety in section 40 (p. 228ff) and of
authentic being-towards-death in section 53 (p. 304ff) in BT, op. cit. Habermas‘s hasty charges of
Lebensphilosophie are perhaps intelligible against this backdrop. I want to propose a different interpretation— in
line with the concept of experience I developed—one according to which experience is already limit-experience, and
not its static and confining other. By the same token, limit-experience is more mundane than some of its lyrical
evocations might suggest.
330 Borges‘s account is in ―John Wilkins‘ Analytical Language,‖ Selected Non-Fictions, op. cit. pp. 229- 233
279
confrontation with limits is the source of Foucault‘s ―laughter,‖ which ―shatters all the familiar
landmarks of [his] thought‖ (OT, xv). But why laugh?
I have referred the massive number of heteroclite and contradictory events constituting
our history to the level of fundamental experience; and I proposed that this be understood in
terms of the spatially and temporally indexed, rule-governed practices criterial for a group of
people. My contention has been that this provides a better grid of historical intelligibility, where
―better‖ means: capable of discerning the historical singularity of events and critical
interrogation of their conditions of emergence. I want to add: it also explains the uneasiness we
feel before this ―laughter,‖ since it seeks to articulate a type of affirmation which is grounded
neither in the recognition of truth understood as correspondence to an independent reality, nor in
the mutual recognition of subjects understood as rational autonomous beings. It forgoes both the
quasi-transcendental presuppositions of communicative action and consensus, and the dialectical
development and rational reconstruction of a self-realizing community.
Gillian Rose claims that ―[n]either positive nor negative, [Foucault‘s] affirmation is
without determination or characteristic; it does not represent an encounter with the power of
another but an ecstasy of blind laughter or blinding tears, which…is simply that old familiar
despair.‖331But I think she is mistaken, for two reasons: First, this affirmation does not deny
reflection, or self-relation, or even negation, but only a particular conception of negation which
seeks to recover ―what we have lost over the last half-century…in the second degree, by means
of the analysis of…analyses‖ (AK, 202). Therefore, it is neither a question of what one could call
writing the history of reason in terms of a reason without history, nor the denial of subjectivity,
331 Gillian Rose, Dialectic of Nihilism: Post-Structuralism and Law (Wiley-Blackwell, 1991) p. 207
280
but of ―[freeing] history from the grip of phenomenology,‖ as well as denying satisfaction to ―all
transcendental narcissism‖ (Ibid., 203). In other words, it is only the denial of the dilemma
according to which a particular conception of rationality would seal itself against all empirical
conditioning in advance, and say: ―either it does not reach us or we claim it‖ (Ibid., 206).
Second, there is something of despair in Foucault‘s laughter; he says as much himself:
―The uneasiness that makes us laugh when we read Borges…‖ (OT, xviii) And just what should
be so bad about ―that old familiar despair‖? The ―uneasiness‖ is related to the ―distress of those
whose language has been destroyed: loss of what is ‗common‘ to place and name‖ (Ibid., xix).
Some aphasiacs, Foucault informs us, are consistently unable to arrange colored skeins of wool
into any coherent pattern on top of a table:
[A]s though that simple rectangle were unable to serve in their case as a homogenous and
neutral space in which things could be placed….Within this space in which things are
normally arranged and given names, the aphasiac will create a multiplicity of tiny,
fragmented regions in which nameless resemblances agglutinate things into unconnected
islets....But no sooner have they been adumbrated than all these groupings dissolve
again…so the sick mind continues to infinity, creating groups, then dispersing them
again….‖ (Ibid., xviii)
I know of no better description of how Foucault‘s historiography itself patterns its ―material,‖
and therefore of the experience of reading his histories. Foucault the aphasiac?
Perhaps. But perhaps this form of historical aphasia is precisely our fundamental
experience. The interstices separating and juxtaposing our categories is where monstrosity lurks.
These monsters are not exactly on the order of what is fictional—fabula, or fabulous—since that
has its own place in our categories. Rather, what is liminal about experience, thereby making it
already limit-experience, is the utter contingency of the practices, discursive and non-discursive,
281
that sustain our classifications and identifications. No transcendental ground, not even a
historically modulated one, can account for the ways in which we think and act and speak, if
―accounting‖ is understood as rational legitimation. In that sense, there is no accounting for the
ways in which we count—on one another, but also on the world.
If it is Borges‘s fiction which helps us to this ―realization,‖ it is not surprising that there is
something fictional about Foucault‘s histories. For it is possible to assert that, not unlike Borges,
he too ―dispenses with the least obvious, but most compelling of necessities; he does away with
the site, the mute ground upon which it is possible for entities to be juxtaposed‖ (OT, xvii). But
if history so understood does something more than what we, perhaps too easily, call ―fiction,‖ it
is because it makes it possible to see ―a worse kind of disorder than that of the incongruous, the
linking together of things that are inappropriate…[namely] the disorder in which fragments of a
large number of orders glitter separately in the dimension, without law or geometry, of the
heteroclite…in sites so very different from one another that it is impossible to find a place of
residence for them, to define a common locus beneath them all‖ (Ibid., xvii-xviii)
I said: ―discursive and nondiscursive practices‖. For it is also a question of force and of
power. My claim has been that an understanding of experience as an articulation of two ―types‖
of practices, where their articulation is neither a logical, nor linguistic, nor even causal
determination, provides a more honest description of human action precisely because it refuses
to read more rationality into it than is to be found. This articulation concerns rather ―that less
apparent syntax which causes words and things (next to and also opposite one another) to ‗hold
282
together‘‖ (OT, xviii).332I have tried to relate this ―syntax‖ to Wittgenstein‘s ―grammar‖.
However, even though I argued against a transcendental reading of Wittgenstein, the opposite
risk of that engagement is that critique may then appear only a matter of ―this is how we go on‖;
and critical work would be restricted to the description of the moves ―we‖ make in different
language games, and if ―they‖ do not move like us, then they are not one of us.333
There are times when that may be all we can or even should say and do. At such times the
Foucauldian rejoinder against strategies of exclusion is not to ―cry normalization,‖ but to insist
that we realize that that is what we are doing, and accept responsibility for the consequences. It
therefore has more critical force than an argument which would transcendentally secure the
legitimacy of that exclusion by ―demonstrating‖ the irrationality of those who are so excluded,
without accounting for the constitutive role our social sanctions play in their very creation. That
insistence takes two forms: epistemic, which I discussed in terms of conditions of acceptability
and conditions of predication; and normative, which I discussed in terms of power relations and
their strategies. Demonstration, then, is complicit with the formation and exclusion of monsters,
which does not entail that we would thereby be excused from making distinctions in order to
collapse truth and falsehood, right and wrong, into a new night in which all cows are black.
332 A very good formulation of what I take to be a similar point from another angle is found in Michel Serres, ―The
Geography of the Incommunicable: Madness,‖ in FI: ―…the work of Foucault is in no way a history (or a chronicle)
of psychiatry, in as much as the recurrent exploration…does not bring presciences to light. It is an archaeology of
the subject who is sick in the most profound sense, that is to say, more than a generalized etiology, in that it reveals
the conditions of knowledges indissolubly linked to conditions of sickness….This Folie et Déraison is thus…a
history found in the mirror of the asylum‘s microcosm, disfigured certainly, silent and pathetic, but rigorously
organized….This hallucinatory mirror does not open a space of virtual images, but rather discovers the originary
terrain of cultural processes, the forgotten latencies in human works.‖ pp. 55-56
333 Rorty, for instance, opts for an unquestioned separation between the private and the public in order to resolve a
similar difficulty: ―The vocabulary of self-creation is necessarily private…the vocabulary of justice is necessarily
public and shared,‖ Contingency, Irony, Solidarity (Cambridge: Cambridge University Press, 1989) p. xiv. The
separation, however, results in the impoverishment of both (what are then called) ―aesthetic creation‖ and
―moral/legal action‖.
283
Nonetheless, there is something tragic about this, and even tragedies have a denouement.
But perhaps tragedy here should only be understood in its contrast with both dialectics and
rational reconstruction: it is the denunciation of that rationality which magically transforms
differences into oppositions, in order then to make negation determinate by the teleology of a
closed system. In short, it is the refusal of facile and final syntheses. The work so constructed ―is
not a critical history which aims to demonstrate that behind…so-called knowledge there is only
mythology, or perhaps nothing at all…. [But it is rather] the problematization of something
which is real, but that problematization is something which is dependent on our
knowledge…techniques, social relations and economical processes‖ (FL, 418).
A Bradleyan epigraph from Minima Moralia reads: ―Where everything is bad it must be
good to know the worst‖ (MM, 83). But why? What good could come out of knowledge, the
content of which would be exhausted in its reflection of the badness of the totality? Conversely,
how could the totality be all bad, so long as it still allows space for at least one statement
―everything is bad‖? Perhaps it is against the backdrop of this difficulty that Foucault‘s
disclaimer becomes intelligible: ―My point is not that everything is bad, but that everything is
dangerous‖ (FR, 343); and his appropriation of the critical tradition as ―the patient labor giving
form to our impatience for liberty‖ (Ibid., 50). A historical critique of this type is necessarily
unsatisfactory, since it disavows both transcendental and empirical grounding; and therefore, it is
necessarily problematic and fragile, since it depends only on what it criticizes, making its
cooptation possible in principle.
But perhaps this fragility is precisely what harbors its indeterminate negativity. Kant
invoked the transcendental as precisely that dimension of thought which, in a world where one
no longer speaks with gods or sees natural purposes, makes freedom theoretically possible and
284
practically necessary. But its grounding in a formal normative framework—it matters little
whether that is inscribed in a doctrine of faculties or the norms of communicative action—
secures the enlightenment of good reasons only to blind us to how we can also be blinded by
good reasons. And that is because it thereby renders secondary and derivative the question of
good life. Or rather, that question is no sooner raised than forgotten, formal and ideal, like the
point of its mark. Freedom too is of this world, and if it is not a thing, this is not because its
possibility is secure in principle, nor because its actuality is visible in fact, but because we keep
on making a problem of what we say and think and do, without knowing whether it will continue
or whether it has ended already.
285
Bibliography
Adorno, Theodor W., The Adorno Reader, Brian O‘Connor ed., (Wiley-Blackwell, 2000).
---. Critical Models: Interventions and Catchwords (Columbia University Press, 1998)
---.Culture Industry, 2nd ed. (Routledge, 2001)
---. and Max Horkheimer, Dialectic of Enlightenment (Stanford University Press, 2007)
---. Notes to literature volume one, trans. Sherry Weber Nicholson, ed. Rolf Tiedeman (New
York: Columbia University Press, 1991)
---. Notes to literature volume two, trans. Sherry Weber Nicholson, ed. Rolf Tiedeman (New
York: Columbia University Press, 1992)
---. Minima Moralia: Reflections on a Damaged Life (Verso, 2006)
---. Negative Dialectics, 1st ed. (Routledge, 1990)
---. Prisms, (The MIT Press, 1983)
Allison, Henry E. Kant's Transcendental Idealism: An Interpretation and Defense (Yale
University Press, 2004)
Althusser, Louis, For Marx (Verso, 2006)
Aristotle, Complete Works of Aristotle, Volume 2 (Princeton University Press, 1984)
Armstrong, T. ed, Michel Foucault : Philosopher (1st ed. Routledge, 1991)
Austin, J.L. How to Do Things with Words: The William James Lectures Delivered in Harvard
University in 1955, 2nd ed. (Oxford University Press, 1976)
Bataille, Georges. L' Experience Interieure (Gallimard-Jeunesse, 1978)
Beckett, Samuel. The Dramatic Works of Samuel Beckett: Volume III (Grove Press, 2006)
Bennett, Alan, The History Boys: A Play (Faber & Faber, 2006)
Bennington, Geoffrey, Derrida (Editions du Seuil, 2008)
Bernstein, J. M. Adorno: Disenchantment and Ethics (Cambridge University Press, 2001)
Borges, Jorge Luis. Borges: Collected Fictions, 1st ed. (Penguin, 1999)
---. Borges: Selected Non-Fictions, 1st ed. (Penguin, 2000)
---. The Book of Imaginary Beings, (Penguin Classics, 2006)
Cavell, Stanley. The Claim of Reason: Wittgenstein, Skepticism, Morality, and Tragedy (Oxford
University Press, 1999)
Davidson, Arnold I. ed. Foucault and His Interlocutors (University of Chicago Press, 1998)
Deleuze, Gilles. Foucault (University of Minnesota Press, 1988)
286
---. Negotiations 1972-1990, (Columbia University Press, 1997)
Derrida, Jacques, Edmund Husserl's "Origin of Geometry": An Introduction (University of
Nebraska Press, 1989)
---. Limited Inc, 1st ed. (Northwestern University Press, 1988)
---. Of Grammatology, (The Johns Hopkins University Press, 1998)
---. Writing and Difference, (University of Chicago Press, 1978)
Descartes, Rene, The Philosophical Writings of Descartes: Volume 2. (Cambridge University
Press, 1985)
Dews, Peter, Logics of Disintegration: Poststructuralist Thought and the Claims of Critical
Theory (Verso, 2007)
Dreyfus, Hubert L. and Paul Rabinow, Michel Foucault: Beyond Structuralism and
Hermeneutics, 2nd ed. (University of Chicago Press, 1983)
Gros, F., Foucault et la folie (Presses Universitaires de France, 1997)
Fichte, Johann Gottlieb, The Science Of Knowing: J.g. Fichte's 1804 Lectures On The
Wissenschaftslehre (State University of New York Press, 2005)
Flynn, Thomas R., Sartre, Foucault, and Historical Reason, Volume Two: A Poststructuralist
Mapping of History (University Of Chicago Press, 2005)
Foucault, Michel. Surveiller et punir, Naissance De La Pris (Gallimard, 1998)
---. Aesthetics, Method, and Epistemology, Essential Works vol. 2 (New Press, 1998)
---. Discipline & Punish: The Birth of the Prison (Vintage, 1995)
---. Ethics: Subjectivity and Truth, Essential Works vol. 1 (New Press, 2006)
---. Histoire de la folie a l'age classique (Gallimard, 1972)
---. History of Madness, trans. Jean Khalfa (Routledge, 2009)
---. Introduction to Kant's Anthropology, trans. Kate Briggs (Semiotexte, 2008)
---. L'Ordre du Discours (Gallimard, 1971)
---. Maladie mentale et personalité (Paris: Presses Universitaires de France, 1954)
---. Mental Illness and Psychology (University of California Press, 2008)
---. Order of Things: An Archaeology of the Human Sciences (Routledge, 200)
---. Politics, Philosophy, Culture: Interviews and Other Writings, 1977-1984 (Routledge, 1990)
---. Power/Knowledge: Selected Interviews and Other Writings, 1972-1977 (Pantheon, 1980)
---. The Archaeology of Knowledge & The Discourse on Language (Pantheon, 1982)
---. The Final Foucault (The MIT Press, 1988)
---. Foucault Live: Interviews, 1961-84. (Semiotext(e)), 1996)
---. The Foucault Reader (Pantheon, 1984)
---. L'Hermeneutique du sujet: Cours au Collège de France (Seuil, 2001)
---. The History of Sexuality, Vol. 1: An Introduction (Vintage, 1990)
---. The History of Sexuality, Vol. 2: The Use of Pleasure (Vintage, 1990)
---. The History of Sexuality, Vol. 3: The Care of the Self (Vintage, 1988)
---. Histoire De La Sexualite 2: L'Usage Des Plaisirs (Gallimard, 1984)
287
---. The Politics of Truth (Semiotext, 2007)
---. Power, Essential Works vol. 3 (New Press, 2001)
Franks, Paul W. All or Nothing: Systematicity, Transcendental Arguments, and Skepticism in
German Idealism (Harvard University Press, 2005)
Gardner, Sebastian, Kant and the Critique of Pure Reason (Routledge, 1998)
Gutting, Gary. Michel Foucault's Archaeology of Scientific Reason: Science and the History of
Reason (Cambridge University Press, 1989)
---. ed., The Cambridge Companion to Foucault (Cambridge University Press, 1994)
Habermas, Jürgen, Knowledge and Human Interests (Beacon Press, 1972)
---. The Philosophical Discourse of Modernity: Twelve Lectures (The MIT Press, 1990)
---. The Theory of Communicative Action, Volume 1: Reason and the Rationalization of Society
(Beacon Press, 1985)
---. The Theory of Communicative Action, Volume 2: Lifeworld and System: A Critique of
Functionalist Reason (Beacon Press, 1985)
Hacking, Ian, Historical Ontology (Harvard University Press, 2004)
Han, Béatrice, Foucault's Critical Project: Between the Transcendental and the Historical
(Stanford University Press, 2002)
Hegel, G. W. F., Hegel: Elements of the Philosophy of Right (Cambridge University Press, 1991)
---. Lectures on the Philosophy of World History (Cambridge University Press, 1981)
---. Science of Logic (Routledge, 2004)
---. Phenomenology of Spirit (Oxford University Press, 1979)
---. The Encyclopaedia Logic: Part 1 of the Encyclopaedia of Philosophical Sciences with the
Zusatze (Hackett Publishing Company, 1991)
Heidegger, Martin, Being and Time (HarperOne, 1962)
---. Hegel's Concept of Experience (Harper and Row, 1970)
Henrich, Dieter, Between Kant and Hegel: Lectures on German Idealism (Harvard University
Press, 2008)
Honneth, Axel, (The Fragmented World of the Social: Essays in Social and Political Philosophy.
State University of New York Press, 1995)
Hoy, David ed., Foucault: A Critical Reader (Wiley-Blackwell, 1991)
Husserl, Edmund, Cartesian Meditations: An Introduction to Phenomenology (Martinus Nijhoff
Pub., 1977)
---. Crisis of European Sciences and Transcendental Phenomenology, trans. David Carr
(Northwestern University Press, 1970)
---. Formal and Transcendental Logic (Martinus Nijhoff, 1966)
288
---. Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy: First
Book: General Introduction to a Pure Phenomenology (Springer, 1983)
---. Logical Investigations, Vol. 2 (New edition. Routledge, 2001)
Inwood, Michael, A Hegel Dictionary (Wiley-Blackwell, 1992)
Jay, Martin, Cultural Semantics: Keywords of Our Time (University of Massachusetts Press,
1998)
---. Songs of Experience: Modern American and European Variations on a Universal Theme
(University of California Press, 2006)
Kant, Immanuel, Critique of Pure Reason (Cambridge University Press, 1999)
---. Groundwork of the Metaphysics of Morals (Cambridge University Press, 1998)
---. Practical Philosophy (Cambridge University Press, 1999)
---. Prolegomena to Any Future Metaphysics That Will Be Able to Come Forward As Science
with Kant's Letter to Marcus Herz (Hackett Pub Co, 2002)
Korsgaard, Christine M, The Sources of Normativity (Cambridge University Press, 1996)
Lear, Jonathan Freud (Routledge, 2005)
---. ―The Disappearing 'We'.‖ Proceedings of the Aristotelian Society, Supplementary Volumes
58 (1984): 219-258.
McDowell, John, Mind and World (Harvard University Press, 1996)
Merleau-Ponty, Le visible et l‟invisible, suivi de notes de travail (Gallimard, 1991).
---. Phenomenology of Perception (Routledge, 2002)
Miller, James, The Passion of Michel Foucault (Harvard University Press, 2000)
Moser, Paul K. and Thomas L. Carson, eds. Moral Relativism: A Reader (Oxford University
Press, USA, 2000)
Nancy, Jean-Luc, The Experience of Freedom (Stanford University Press, 1994)
Neuhouser, Frederick, Fichte's Theory of Subjectivity (Cambridge University Press, 1990)
Nietzsche, Friedrich, 'On the Genealogy of Morality' and Other Writings (Cambridge University
Press, 2006)
O'Connor, Brian, Adorno's Negative Dialectic: Philosophy and the Possibility of Critical
Rationality (The MIT Press, 2005)
Pippin, Robert B. Hegel's Idealism: The Satisfactions of Self-Consciousness (Cambridge
University Press, 1989)
---. Modernism as a Philosophical Problem: On the Dissatisfactions of European High Culture
289
(Wiley-Blackwell, 1999)
Plato, trans. G. M. A. Grube, and C. D. C. Reeve, Republic, second ed. (Hackett Publishing
Company, 1992)
Rorty, Richard, Contingency, Irony, and Solidarity (Cambridge University Press, 1989)
---. Philosophy and the Mirror of Nature (Princeton University Press, 1981)
Rose, Gillian, Dialectic of Nihilism: Post-Structuralism and Law (Wiley-Blackwell, 1991)
--. Hegel Contra Sociology (Verso, 2009)
Roudinesco, Elizabeth, et al., Penser la folie: Essais sur Michel Foucault (Galilee, 1992)
Schelling, Friedrich Wilhelm Joseph Von. Unconditional in Human Knowledge: Four Early
Essays (Associated University Press, 1980)
Searle, John R., Speech Acts: An Essay in the Philosophy of Language (Cambridge University
Press, 1970)
Smith, Nicholas, ed. Reading McDowell: On Mind and World (Routledge, 2002)
Welton, Donn, ed. Critical and Dialectical Phenomenology (State University of New York
Press, 1987)
Wittgenstein, Ludwig, On Certainty (Wiley-Blackwell, 1991)
---. Philosophical Investigations, 4th ed. (Wiley-Blackwell, 2009)
Zizek, Slavoj, The Ticklish Subject: The Absent Centre of Political Ontology (Verso, 2009)
Zupancic, Alenka, Ethics of The Real: Kant, Lacan (Verso, 2000)
290