Magazine: POLITICAL THEORY, August 1990
THE CRITIQUE OF IMPURE REASON
-----------------------------
Foucault and the Frankfurt School
I.
Following Foucault's own example, commentators have generally paid much
more attention to his break with previous forms of critical social
theory than to his continuities with them. It is not surprising that a
thinker of his originality, having come intellectually of age in postwar
France, would eventually come to assert his intellectual identity in
opposition to the varieties of Marxism prevalent there. But for purposes
of developing a critical theory adequate to the complexities of our
situation, focusing only on discontinuities can become
counterproductive. In fact, viewed at some remove from the current
debates, what unites Foucault with neo-Marxist thinkers is as
significant as what divides them. This is particularly true of the group
of theorists loosely referred to as the Frankfurt School, to whom he did
not address himself in any detail. Let me begin by noting certain broad
affinities between Foucault's genealogy of power/knowledge and the
program of critical social theory advanced by Max Horkheimer and his
colleagues in the early 1930s and recently renewed by Jurgen
Habermas.(n1)
1. Both Foucault and the Frankfurt School call for a transformation cum
radicalization of the Kantian approach to critique. The intrinsic
"impurity" of what we call "reason"--its embeddedness in culture and
society, its entanglement with power and interest, the historical
variability of its categories and criteria, the embodied, sensuous, and
practically engaged character of its bearers--makes its structures
inaccessible to the sorts of introspective survey of the contents of
consciousness favored by early modem philosophers and some twentieth-
century phenomenologists. Nor is the turn to language or sign systems an
adequate response to this altered view of reason; all forms of
"linguistic" or "discursive idealism" rest on an indefensible
abstraction from social practices. To explore the "nature, scope, and
limits of human reason," we have to get at those practices, and that
calls for modes of sociohistorical inquiry that go beyond the
traditional bounds of philosophical analysis. The critique of reason, as
a nonfoundationalist enterprise, aims at grasping structures and rules
that transcend the individual consciousness. But what is supraindividual
in this way is no longer understood as transcendental; it is
sociocultural in origin.
2. Correspondingly, both Foucault and the Frankfurt School reject the
Cartesian picture of an autonomous rational subject set over against a
world of objects which it seeks to represent and, through representing,
to master. Knowing and acting subjects are social and embodied beings,
and the products of their thought and action bear ineradicable traces of
their situations and interests. The atomistic and disengaged Cartesian
subject has to be dislodged from its position at the center of the
epistemic and moral universes, and not only for theoretical reasons: It
undergirds the egocentric, domineering, and possessive individualism
that has so disfigured modern Western rationalism and driven it to
exclude, dominate, or repress whatever is different. Thus the
desublimation of reason goes hand in hand with the decentering of the
rational subject.
3. More distinctive perhaps than either of these now widely held views
is that of the primacy of the practical over the theoretical that
Foucault shares with the Frankfurt School. A reversal of the traditional
hierarchy was already proposed by Kant, only to be retracted by Hegel;
it was then reinstated by the young Marx but soon faded into the
background of scientific socialism. Once we have fumed our attention
from consciousness to culture and society, however, there is no good
reason why knowledge and representation should enjoy the privilege over
values and norms that Western philosophy has accorded them. Moreover, if
knowledge is itself understood as a social product, the traditional
oppositions between theory and practice, fact and value, and the like
begin to break down, for there are practical, normative presuppositions
to any social activity, theorizing included. Like other practices,
epistemic practices too have to be comprehended in their sociocultural
contexts. In this sense, the theory of knowledge is part of the theory
of society, which is itself embedded in practical contexts, and in
rather distinctive ways. It is in recognition of the peculiarly
reflexive relation of thinking about society to what is being thought
about that leads Foucault to characterize his geneology as "history of
the present." Situated in the very reality it seeks to comprehend, and
relating the past from the practically interested standpoint of an
anticipated future, it is anything but a view from nowhere. And though
Western Marxism has repeatedly succumbed to the siren calls of a
scientific theory of history or a speculative philosophy of history,(n2)
it has usually found its way back to a similar notion of practical
reflexivity. In this version of critical social theory, there is an
essentially prospective dimension to writing the history of the present
in which one is situated; and the projected future, which gives shape to
the past, is not a product of disinterested contemplation or of
scientific prediction but of practical engagement; it is a future that
we can seek to bring about.
4. With suitable changes in terminology, much of the foregoing could be
said of philosophical hermeneutics as well. It, too, takes seriously the
fact that reason, in its cognitive employment as well, is embedded in
sociocultural contexts, mediated by natural languages, and intrinsically
related to action. It, too, maintains that speech and action occur
against immeasurable, taken-for-granted backgrounds, which are
historically and culturally variable and which can never be brought
fully to conscious awareness. And yet genealogy is as distinct from
hermeneutics as is critical social theory. Despite some very real
differences on this point, neither wishes to leave to the participants
and their traditions the final say about the significance of the
practices they engage in. Both see the need for an objectivating
"outsider's" perspective to get beyond shared, taken-for-granted
meanings and their hermeneutic retrieval. Foucault's way of creating
distance from the practices we live by is to display their "lowly
origin`" in contingent historical circumstances, to dispel their
appearance of self-evident givenness by treating them as the outcome of
multiple relations of force. From the start, critical social theory was
also based on a rejection of what Marx viewed as the specifically
"German ideology," and Horkheimer called "the idealist madness," of
understanding ideas solely in terms of other ideas. It has insisted that
the full significance of ideas can be grasped only by viewing them in
the context of the social practices in which they figure, and that this
typically requires using socio-historical analysis to gain some distance
from the insider's view of the participants. Genetic and functional
accounts of how and why purportedly rational practices came to be taken
for granted play an important role in both forms of the critique of
impure reason.
5. In neither perspective, however, does this mean simply adopting the
methods of the established human sciences. Both Foucault and the
Frankfurt School see these as particularly in need of critical analysis,
as complicitous in special ways with the ills of the present age. There
are, to be sure, some important differences here, for instance, as to
which particular sciences are most in need of critique and as to how
total that critique should be.(n3) But there are also a number of
important commonalities in their critiques of the epistemological and
methodological ideas in terms of which we have constituted ourselves as
subjects and objects of knowledge. Furthermore, both are critical of the
role that the social sciences and social scientifically trained
"experts" have played in the process of "rationalization." They see the
rationality that came to prevail in modern society as an instrumental
potential for extending our mastery over the physical and social worlds,
a rationality of technique and calculation, of regulation and
administration, in search of ever more effective forms of domination.
Inasmuch as the human sciences have assisted mightily in forging and
maintaining the bars of this "iron cage," to use Max Weber's phrase,
they are a prime target for genealogical and dialectical critique.
6. As ongoing practical endeavors rather than closed theoretical systems,
both forms of critique aim at transforming our self-understanding in
ways that have implications for practice. It is true that Foucault
persistently rejected the notions of ideology and ideology-critique and
denied that genealogy could be understood in those terms. But the
conceptions of ideology he criticized were rather crude, and the
criticisms he offered were far from devastating to the more
sophisticated versions propounded by members of the Frankfurt School. It
is, in fact, difficult to see why Foucault's efforts to analyze "how we
govern ourselves and others by the production of truth," so as to
"contribute to changing people's ways of perceiving and doing things,
"(n4) do not belong to the same genre. On this reading, in both
genealogy and critical social theory, the objectivating techniques
employed to gain distance from the rational practices we have been
trained in afford us a critical perspective on those practices. Making
problematic what is taken for granted- for instance, by demonstrating
that the genesis of what has heretofore seemed to be natural and
necessary involves contingent relations of force and an arbitrary
closing off of alternatives, or that what parades as objective actually
rests on prescriptions that function in maintaining imbalances of power--
can weaken their hold on us. Categories, principles, rules, standards,
criteria, procedures, techniques, beliefs, and practices formerly
accepted as purely and simply rational may come to be seen as in the
service of particular interests and constellations of power that have to
be disguised to be advanced, or as performing particular functions in
maintaining power relations that would not be subscribed to if generally
recognized. Because things are not always what they seem to be, and
because awareness of this can create critical distance--because, in
particular, such awareness can undermine the authority that derives from
presumed rationality, universality, or necessity--it can be a social
force for change. Whether or not this is so, and the extent to which it
is so, is, in the eyes of both Foucault and the Frankfurt School, not a
question of metaphysical necessity or theoretical deduction but of
contingent historical conditions. That is, the practical significance of
critical insight varies with the historical circumstances.
If the foregoing comparisons are not wide of the mark, Foucault and the
Frankfurt School should be located rather close to one another on the
map of contemporary theoretical options. They hold in common that the
heart of the philosophical enterprise, the critique of reason, finds its
continuation in certain forms of sociohistorical analysis carried out
with the practical intent of gaining critical distance from the
presumably rational beliefs and practices that inform our lives. This
would certainly place them much nearer to one another than to other
varieties of contemporary theory, including the more influential
varieties of textualism. Why, then, have the oppositions and differences
loomed so large? At least part of the explanation (but only part) is
that the disagreements between them are no less real than the
agreements. Though genealogy and critical social theory do occupy
neighboring territories in our theoretical world, their relations are
rather combative than peaceable. Foucault's Nietzschean heritage and the
Hegelian-Marxist heritage of the Frankfurt School lead them to lay
competing claims to the very same areas:
1. While both seek to transform the critique of reason through shifting
the level of analysis to social practice, Foucault, like Nietzsche, sees
this as leading to a critique that is radical in the proper sense of
that term, one that attacks rationalism at its very roots; whereas
critical social theorists, following Hegel and Marx, understand critique
rather in the sense of a determinate negation that results in a more
adequate conception of reason.
2. While both seek to get beyond the subject-centeredness of modern
Western thought, Foucault understands this as the "end of man" and of
the retinue of humanist conceptions following upon it; whereas critical
social theorists attempt to refashion notions of subjectivity and
autonomy that are consistent with both the social construction of
individual identity and the situated character of social action.
3. While both assert the primacy of practical reason and acknowledge the
unavoidable reflexivity of social inquiry, Foucault takes this to be
incompatible with the context-transcendence of truth claims and the
pretensions of global social theories; whereas the Frankfurt theorists
seek to combine contextualism with universalism and to construct general
accounts of the origins, structures, and tendencies of existing social
orders.
4. While both refuse to take participants' views of their practices as
the last word fin understanding them, critical social theorists do take
them as the first word and seek to engage them in the very process of
trying to gain critical distance from them; whereas the genealogist
resolutely displaces the participants' perspective with an externalist
perspective in which the validity claims of participants are not engaged
but bracketed.
5. While both are critical of established social sciences and see them
as implicated in weaving an ever tighter web of discipline and
domination, Foucault understands this to be a general indictment of the
human sciences (genealogy is not a science but an "anti-science");
whereas critical social theorists direct this critique against some
forms of social inquiry, while seeking to identify and develop others
that are not simply extensions of instrumental rationality.
6. Finally, while both see the critique of apparently rational practices
as having the practical purpose of breaking their hold on us, Foucault
does not regard genealogy as being in the service of reason, truth,
freedom, and justice--there is no escaping the relations and effects of
power altogether, for they are coextensive with, indeed constitutive of,
social life generally; whereas Frankfurt School theorists understand the
critique of ideology as working to reduce such relations and effects and
to replace them with social arrangements that are rational in other than
an instrumentalist sense.
II.
With this broad comparison as a background, I would like now to take a
closer and more critical look at Focault's radical critique of reason
and of the rational subject in the context of the theory of power he
developed in the 1970s. For purposes of defining what is at issue
between him and the Frankfurt School, I shall use Habermas's attempt to
renew Horkheimer's original program as my principal point of reference.
As noted above, Foucault's genealogical project can be viewed as a form
of the critique of reason. Inasmuch as modem philosophy has understood
itself to be the most radical reflection on reason, its conditions,
limits, and effects, the continuation-through-transformation of that
project today requires a sociohistorical turn. What have to be analyzed
are paradigmatically rational practices which cannot be adequately
understood in isolation from the sociohistorical contexts in which they
emerge and function. Foucault is, of course, interested in the relations
of power that traverse those practices and their contexts. He reminds us
repeatedly that "truth is not the reward of free spirits" but "a thing
of this world" that is "produced only by virtue of multiple forms of
constraint."(n5) Analytical attention is redirected to the rules,
prescriptions, procedures, and the like that are constitutive of
rational practices, to the relations of asymmetry, nonreciprocity, and
hierarchy they encode, and to the ways in which they include and exclude,
make central and marginal, assimilate and differentiate. This shift in
focus makes us aware that there is something like a politics of truth
and knowledge already at this level of analysis.(n6) Irrationality,
incompetence, deviance, error, nonsense, and the like get marked off in
various ways from their opposites; people and practices get valorized or
stigmatized, rewarded or penalized, dismissed or vested with authority
on this basis. But genealogical analysis does not confine itself to the
political aspects of rules and regulations "internal" to discursive
practices. It also examines the "external" relations of theoretical
discourses--especially the discourses of the "sciences of man"--to the
practical discourses in which they are "applied"--the discourses of
psychologists, physicians, judges, administrators, social workers,
educators, and the like--as well as to the institutional practices with
which they are interwoven in asylums, hospitals, prisons, schools,
administrative bureaucracies, welfare agencies, and the like. As soon as
one tries to comprehend why a particular constellation of rules and
procedures should define rational practice in a given domain,
consideration of the larger sociohistorical context becomes unavoidable.
"Each society," as Foucault puts it, "has its regime of truth,"(n7) and
genealogy is interested precisely in how we govern ourselves and others
through its production. Focusing especially on the human sciences--the
sciences of which "man" is the object--the examines the myriad ways in
which power relations are both conditions and effects of the production
of truth about human beings. In areas of inquiry ranging from psychiatry
and medicine to penology and population studies, he uncovers the
feedback relations that obtain between the power exercised over people
to extract data from and about them--by a variety of means, from
observing, examining, and interrogating individuals to surveying and
administering populations--and the effects of power that attach to the
qualified experts and licensed professionals who possess and apply the
knowledge thus gained. According to Foucault, the sciences of man not
only arose in institutional settings structured by hierarchical
relations of power, they continue to function mainly in such settings.
Indeed, what is distinctive of the modern disciplinary regime, in his
view, is just the way in which coercion by violence has been largely
replaced by the gentler force of administration by scientifically
trained experts, public displays of power by the imperceptible
deployment of techniques based on a detailed knowledge of their targets.
>From Foucault's perspective, then, the human sciences are a major force
in the disastrous triumph of Enlightenment thinking, and the panoptical
scientific observer is a salient expression of the subject-centered,
putatively universal reason-which that thinking promotes. By tracing the
lowly origins of these sciences in struggle and conflict, in
particularity and contingency, in a will to truth that is implicated
with domination and control, genealogy reveals their constitutive
interconnections with historically changing constellations of power:
"[Plower and knowledge directly imply one another.... The subject who
knows, the objects to be known, and the modalities of knowledge must be
regarded as so many effects of these fundamental implications of power-
knowledge and their historical transformations."(n8)
Although Habermas agrees with Foucault in regarding truth as "a thing of
this world," he distinguishes between fundamentally different cognitive
approaches marked by different configurations of action, experience, and
language.(n9) He does this with the aim of resisting the identification
of instrumental and strategic rationality with rationality tout court.
To construe sociocultural rationalization merely as the growing hegemony
of techniques of power and control, of domination and administration, is
not so much erroneous as partial. That reading does not grasp the
selectivity of capitalist modernization, its failure to develop in a
balanced way the different dimensions of rationality opened up by the
modern understanding of the world. Because we are as fundamentally
language-using as tool-using animals, the representation of reason as
essentially instrumental and strategic is fatally one-sided. On the
other hand, it is indeed the case that those types of rationality have
achieved a certain dominance in our culture. The subsystems in which
they are centrally institutionalized, the economy and government
administration, have increasingly come to pervade other areas of life
and make them over in their own image and likeness. The resultant
"monetarization" and "bureaucratization" of life is what Habermas refers
to as the "colonization of the life world."(n10)
This picture of a society colonized by market and administrative forces
differs from Foucault's picture of a disciplinary society in, among
other ways, targeting for critique not the Enlightenment idea of a life
informed by reason, as such, but rather the failure to pursue it by
developing and institutionalizing modalities of reason other than the
subject-centered, instrumental ones that increasingly shape our lives.
The two pictures do overlap in a number of areas. For instance, both
focus on the intrication of knowledge with power that is characteristic
of the sciences of man. But Foucault regards this analysis as valid for
all the human sciences, whereas Habermas wants to distinguish
objectivating (e.g., behavioral) approaches from interpretive (e.g.,
hermeneutical) and from critical (e.g., genealogical or dialectical)
approaches. The interests that inform them are, he argues, fundamentally
different, as are, consequently, their general orientations to their
object domains and their characteristics logics of inquiry. From this
perspective, only purely objectivating approaches are intrinsically
geared to expanding control over human beings, whereas other approaches
may be suited to extending the intersubjectivity of mutual understanding
or to promoting reflective distanciation from taken-for-granted beliefs
and practices.
There is a broad agreement between Foucault and Habermas that the
expansion of the welfare state is increasingly dependent on the
generation and application of expert knowledge of various sorts. In this
regard, Foucault's account of the interrelation between social
institutions geared to normalization, on one hand, and the growth of
knowledge suited to that purpose, on the other, parallels Habermas's
account of the interconnection between the administrative colonization
of the lifeworld and the rise of objectivating social science. Here, too,
the differences have chiefly to do with how all-inclusive this critical
perspective can claim to be. Foucault extrapolates the results of his
analyses of knowledge generated in the more or less repressive contexts
which he singles out for attention to the human sciences in general. One
consequence of this is his clearly inadequate account of hermeneutic
approaches; another is his inability to account for his own genealogical
project in other than actionistic terms--genealogy ends up being simply
another power move in a thoroughly power-ridden network of social
relations, another intervention meant to alter the existing balance of
forces.
In the remainder of this section, I would like to make clear the price
that Foucault pays for his wholesale critique of impure reason by taking
a closer look at two key elements of his metatheory of genealogical
practice: the ontology of power and the representation of the subject as
an effect of power.
Power: Ontology versus Social Theory
The differences between Foucault's genealogy and Habermas's critical
social theory are misrepresented by the usual opposition between the
nominalistic particularism of the former and the abstract universalism
of the latter. In his Nietzschean moments, Foucault can be as
universalistic as one might like, or dislike. While he insists that he
wants to do without the claims to necessity typical of foundationalist
enterprises, he often invokes an ontology of the social that treats
exclusion, subjugation, and homogenization as inescapable
presuppositions and consequences of any social practice. And while he
targets for genealogical analysis social institutions that are clearly
marked by hierarchies of power, his own conception of power as a network
of relations in which we are all, always and everywhere, enmeshed,
devalues questions of who possesses power and with what right, of who
profits or suffers from it, and the like. (These are questions typical
of the liberal and Marxist approaches that he rejects.) What we gain
from adopting this conception is a greater sensitivity to the
constraints and impositions that figure in any social order, in any
rational practice, in any socialization process. In this expanded sense
of the term, "power" is indeed "a productive network that runs through
the whole social body."(n11) Giving this insight an ontological twist,
one could then say with Foucault that "power produces reality, it
produces domains of objects and rituals of truth,"(n12) or,
alternatively, that "truth is not the product of free spirits" but is
"produced by virtue of multiple forms of constraint."(n13) It is
undeniable that any "regime of truth" involves privileging certain types
of discourse, sanctioning certain ways of distinguishing true from false
statements, underwriting certain techniques for arriving at the truth,
according a certain status to those who competently employ them, and so
forth. In this sense, there is indeed a "political economy" of truth, as
there is of any organized social activity; this insight is the principal
gain of Foucault's generalization of the concept of power.
There are also losses incurred in generalizing and ontologizing the
concept of power: Having become more or less coextensive with constraint,
it becomes all too like the night in which all cows are black. Welcoming
or denouncing someone, putting someone at ease or into prison,
cooperating with or competing with someone--these are all equally
exercises of power in Foucault's conceptualization. If his aim is to
draw attention to the basic fact that patterned social interaction
always involves normative expectations and thus possible sanctions, this
is a rhetorically effective way of doing so. But the costs for social
theory of such dedifferentiation are considerable. Distinctions between
just and unjust social arrangements, legitimate and illegitimate uses of
political power, strategic and cooperative interpersonal relations,
coercive and consensual measures--distinctions that have been at the
heart of critical social analysis--appear only marginally, if at all. If
there were no possibility of retaining the advantages of Foucault's
Nietzschean move without taking these disadvantages into the bargain, we
would be faced with a fundamental choice between different types of
social analysis. But there is no need to construe this as an either/or
situation. We can agree with Foucault that social action is everywhere
structured by background expectations in terms of which we hold one
another accountable, that deviations from these are sanctionable by
everything from negative affective responses and breakdowns of
cooperation to explicit reprimands and punishments, and that our
awareness of this differential accountability is a primary source of the
motivated compliance that characterizes "normal" interaction.(n14) And
we can agree with him that the modern period has witnessed a vast
expansion of the areas of life structured by instrumental, strategic,
and bureaucratic forms of social interrelation. None of this prevents us
from then going on to mark the sociologically and politically crucial
distinctions that have figured so centrally in the tradition of critical
social theory. Nancy Fraser has stated the issue here with all desirable
clarity:
The problem is that Foucault calls too many different sorts of things
power and simply leaves it at that. Granted, all cultural practices
involve constraints. But these constraints are of a variety of different
kinds and thus demand a variety of different normative responses....
Foucault writes as if oblivious to the existence of the whole body of
Weberian social theory with its careful distinctions between such
notions as authority, forge, violence, domination and legitimation.
Phenomena which are capable of being distinguished via such concepts are
simply lumped together.... As a consequence, the potential for a broad
range of i formative nuances is surrendered, and the result is a certain
normative one-dimensionality.(n15)
The Subject: Deconstruction versus Reconstruction
Foucault has related on various occasions how "people of [his]
generation were brought up on two forms of analysis, one in terms of the
constituent subject, the other in terms of the economic-in-the-last-
instance."(n16) As we have seen, he worked himself free of the latter by,
among other things, drawing on Nietzsche to develop a "capillary"
conception of power as coextensive with the social. In working free of
the former, he was able to call on the assistance of structuralist
semiotics to argue for the priority of systems of signification over
individual acts thereof. Even after he distanced himself from
structuralism by taking as his point of reference "not the great model
of language and signs, but that of war and battle,"(n17) he retained
this order of priority in the form of the "regimes," the interconnected
systems of discourses, practices, and institutions that structure and
give sense to individual actions. From the perspective of the
genealogist, the subject privileged by phenomenology is in reality not
the constituens but the constitutum of history and society; and
phenomenology itself is only a recent chapter in the long tradition of
subjectivism. At the core of that tradition is a hypostatization of the
contingent outcome of historical processes into their foundational
origin--not in the sense, typically, of a conscious creation, but in
that of an alienated objectification of subjective powers, which has
then to be con-sciously reappropriated. This latter figure of thought is,
for Foucault, the philosophical heart of the humanist project (including
Marxist humanism) of mastering those forces, without and within, which
compromise "man's" aotonomy and thus block his true self-realization.
Like Horkheimer and Adorno in the Dialectic of Enlightenment, Foucault
sees this as inherently a project of domination, a project that defines
modern Western man's domipeering relation to otherness and difference in
all forms.
Foucault's reaction to this perceived state of affairs is, I shall argue,
an overreaction. Owing in part to the continued influence of
structuralist motifs in his genealogical phase, he swings to the
opposite extreme of hypostatizing wholes--regimes, networks, dispositifs,
and the like--over against parts, thus proposing to replace an abstract
individualism with an equally abstract holism. To argue that "the-
individual is not to be conceived as a sort of elementary nucleus, a
primitive atom," it is not necessary to maintain that the individual is
merely "one of the prime effects of power."(n18) One might defend
instead the less radical thesis, that individuation is inherently linked
to socialization: We become individuals in and through being socialized
into shared forms of life, growing into preexisting network of social
relations. To this, one might want to add the equally familiar idea that
the forms of individuation and individuality characteristic of the
modern world are not to be found at all times and in all places, but
have come to be historically in conjunction with changes in the
structures of social relations. From this perspective, Foucault's claim
that the individual who is an effect of power is at the same time "the
element of its articulation" or "its vehicle,"(n19) might be construed
as advancing the common sociological view that social structures are
produced and maintained, renewed and transformed only through the
situated actions of individual agents. But this view entails that agency
and structure are equally basic to our understanding of social practices,
and that is decidedly not Foucault's approach. He wants to develop a
form of analysis that treats the subject as an effect by "accounting for
its constitution within a historical framework." If this were only a
matter of "dispensing with the constituent subject," of avoiding all
"reference to a subject which is transcendental in relation to the field
of events," the disagreement would be merely terminological.(n20) It is
not, however, only the constituent, transcendental subject that Foucault
wants to do without; he proposes a mode of inquiry that makes no
explanatory reference to individual beliefs, intentions, or actions.
Genealogy, he advises us, "should not concern itself with power at the
level of conscious intention or decision": It should refrain from posing
questions of the sort: "Who has power and what has that person in mind?"
The focus should instead be on "how things work at the level of ongoing
subjugation, at the level of those continuous and uninterrupted
processes" through which "subjects are gradually, progressively, really
and materially constituted."(n21) Again if this were merely an argument
for the need of supplementing an internalist view of social practices
with an externalist one, of balancing an account of agency with an
account of structure, of integrating a microanalysis of social practices
with a structural analysis of persistent patterns of interaction, or
with a functional analysis of their unintended consequences, or with an
institutional analysis of the normative contexts of individual action,
there would be no incompatibility in principle between genealogy and
approaches operating with some concept of agency. But Foucault does not
want to supplement or balance or integrate; he wants to replace. And the
results of this either/or thinking are no happier here than in the
traditional theories he criticizes.
There is no hope of arriving at an adequate account of social
integration if the only model of social interaction is one of
asymmetrical power relations and the only model of socialization is that
of an intrusion of disciplinary forces into bodies. Nor can we gain an
adequate understanding of most varieties of social interaction by
treating agents simply as acting in compliance with preestablished and
publicly sanctioned patterns--as what Foucault calls "docile bodies" or
Garfinkel calls "cultural dopes." We have to take account of their own
understandings of social structures and their own reflexive use of
cultural resources for making sense. This is no less true of the types
of setting that most interest Foucault; as Goffman and others have made
so abundantly clear, interpreting social situations, understanding what
is expected in them, anticipating reactions to conformity and deviance,
and using this knowledge for one's own strategic purposes are basic
elements of interaction in disciplinary settings, too.(n22) These
elements open up space for differential responses to situations, the
possibility of analyzing, managing, and transforming them. Furthermore,
the same competence and activity of agents is required for an adequate
analysis of the rule-following practices central to Foucault's notion of
power-knowledge regimes. Social rules are neither fully spelled out nor
algorithmically applicable; what they usually call for is not mere
conformity but competent practical reasoning to deal with contingencies
as they rise. Since rules do not define their own application, rule-
following is always to some degree discretionary, elaborative, ad hoc.
Each new application requires the agent's judgment in the light of the
specifics of the situation.(n23)
One could go on at length in this vein. The point is simply to indicate
how deeply the conceptual framework of agency and accountability is
ingrained in our understanding of social practices. Foucault cannot
simply drop it and treat social practice as anonymous, impersonal
processes, for even normalization makes no sense apart from agents who
have the possibility, at least in principle, of resisting it. To be sure,
Foucault does insist on the interdependence of the notions of power and
resistance;(n24) but he refuses to link the latter to the capacity of
competent subjects to say, with good reason, "yes" or "no" to the claims
made on them by others. As a result, he is hard put to say just what it
is that resists; most often he alludes to something like "the body and
its pleasures."(n25) But that only plunges us deeper into just the sorts
of conceptual tangles he wants to avoid. For it is Foucault, after all,
who so forcefully brought home to us just how historical and social the
body and its pleasures are. But when the need arises, as in the present
context, he seems to conjure up the idea of a presocial "body" that
cannot be fitted without remainder into any social mold. This begins to
sound suspiciously like Freud's instinct theory and to suggest a
refurbished model of the "repressive hypothesis" that Foucault so
emphatically rejected.(n26)
If treating the subject merely as "an effect of power"--which must
itself then be conceptualized as a subjectless network--undercuts the
very notions of discipline, regime, resistance, and the like that are
central to genealogical "theory," it raises no less havoc with
genealogical "practice." Who practices genealogical analysis? What does
it require of them? What promise does it hold out to them? If the self-
reflecting subject is nothing but the effect of power relations under
the pressure of observation, judgment, control, and discipline, how are
we to understand the reflection that takes the form of genealogy? Whence
the free-play in our reflective capacities that is a condition of
possibility for constructing these subversive histories? Foucault
certainly writes as if his genealogies advanced our self-understanding,
and in reading them we repeatedly have the experience of their doing
just that. Can we make any sense of this without some, perhaps
significantly revised, notion of subjects who can achieve gains in self-
understanding with a liberating effect on their lives? Charles Taylor
captured this point nicely, when he wrote that
`power' belongs in a semantic field from which `truths end `freedom'
cannot be excluded. Because it is linked with the notion of imposition
on our significant desires and purposes, it cannot be separated from the
notion of some relative lifting of this restraint.... So `power'
requires `liberty', but it also requires `truth'--if we want to allow,
as Foucault does, that we can collaborate in our own subjugation....
Because the imposition proceeds here by foisting illusion upon us, it
proceeds by disguises and masks.... The truth here is subversive of
power.(n27)
This metatheory, deriving from our Enlightenment heritage and shared by
the Frankfurt School, seems to make better sense of Foucault's practice
than his own. If that is so, we may learn more from inquiring, as
Foucault himself finally did in the 1980s, how his work develops and
enriches the critical tradition extending from Kant through the
Frankfurt School than from insisting that it has brought it to an end.
III.
In his first lecture of 1983 at the College de France, Foucault credited
Kant with founding "the two great critical traditions between which
modern philosophy is divided." One, the "analytic philosophy of truth in
general," was a target of Foucault's criticism from the start. The other,
a constantly renewed effort to grasp "the ontology of the present," he
acknowledged as his own: "It is this fore, of philosophy that, from
Hegel, through Nietzsche and Max Weber, to the Frankfurt School, has
formed a tradition of reflection in which I have tried to work."(n28)
This belated affirmation of what he calls the "philosophical ethos," of
the Enlightenment signals important changes in Foucault's understanding
of his critical project.(n29) In this final section, I want briefly to
characterize those changes in respects relevant to our discussion and
then critically to examine their consequences for Foucault's treatments
of the subject and power.
Perhaps the clearest indication of Foucault's altered perception of the
Enlightenment tradition can be found in his reflections on Kant's 1784
essay concerning the question: Was ist Aufklarung?(n30) He regards that
essay as introducing a new dimension into philosophical thought, namely,
the critical analysis of our historical present and our present selves.
When Kant asked "What is Enlightenment?," writes Foucault,
he meant, What's going on just now? What's happening to us? What is this
period, this precise moment in which we are living? Or ID other words,
What are we? as Aufklarer, as part of the Enlightenment? Compare this
with the Cartesian question, Who am I? as a unique but universal and
unhistorical subject? For Descartes, It is everyone, anywhere, at any
moment. But Kant asks something else: What are we? in a very precise
moment of history?(n31)
>From Hegel to Habermas, Foucault continues, this question has defined a
way of philosophizing which he, Foucault, has adopted as his own. What
separates this way from a universally oriented "analytic of truth" is an
awareness of being constituted by our own history, a resolve to submit
that history to critical reflection, and a desire thereby to free
ourselves from its pseudo necessities.
As I argued above, Foucault could and should have said the same of the
genealogy he practiced in the 1970s; but it became clear to him only in
the 1980s that his form of critique also belongs to what Taylor called
the "semantic field" of Enlightenment discourse. "Thought," he later
tells us, "is what allows one to step back from [a] 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. Thought is freedom
in relation to what one does, the motion by which one detaches oneself
from it, establishes it as an object, and reflects on it as a
problem."(n32) Freedom, in turn, is the condition and content of
morality: "What is morality if not the practice of liberty, the
deliberate practice of liberty? . . . Liberty is the essential condition
of ethics. But ethics is the deliberate form assumed by liberty."(n33)
By releasing us from a state of "immaturity," critical thinking makes
possible a "practice of freedom," oriented toward a "mature adulthood"
in which we assume responsibility for shaping our own lives.(n34)
To be sure, behind all these Kantian formulae there lies a considerably
altered critical project. Foucault stresses that faithfulness to the
Enlightenment does not mean trying to preserve this or that element of
it but attempting to renew, in our present circumstances, the type of
philosophical interrogation it inaugurated--not "faithfulness to
doctrinal elements, but rather the permanent reactivation of an attitude,
that is, of a philosophical ethos which could be described as a
permanent critique of our historical era."(n35) Since the Enlightenment,
this type of reflective relation to the present has taken the form of a
history of reason, and that is the form in which Foucault pursues it: "I
think that the central issue of philosophy and critical thought since
the eighteenth century has always been, still is, and will, I hope,
remain the question: What is this Reason that we use? What are its
historical effects?" Of course, in our own day, we have to add: "What
are its limits and what are its dangers? How can we exist as rational
beings, fortunately committed to practicing a rationality that is
unfortunately crisscrossed by intrinsic dangers?"(n36) As noted earlier,
Foucault's genealogical histories stress the local and contingent
aspects of prevailing forms of rationality rather than their
universality. In one way, this is continuous with Kant's linking of
enlightenment and critique: When we dare to use our reason, a critical
assessment of its conditions and limits is necessary if we are to avoid
dogmatism and illusion. On the other hand, genealogy is a very different
way of thinking about conditions and limits:
[I]f the Kantian question was that of knowing what limits knowledge has
to renounce transgressing, it seems to me that the critical question
today has to be turned back into a positive one: 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? The
point, in brief, is to transform the critique conducted in the form of
necessary limitation into a practical critique that takes the form of a
possible transgression.... [C]riticism is no longer going to be
practiced in the search for formal structures with universal value, but
rather as an historical investigation into the events that have led us
to constitute ourselves and to recognize ourselves as subjects of what
we are doing, thinking, saying.... [lit will not deduce from the form of
what we are what it is impossible for us to do and to know; but it will
separate out, from the contingency that has made us what we are, the
possibility of no longer being, doing, or thinking what we are, do, or
think. It is not seeking to make possible a metaphysics that has finally
become a science, it is seeking to give new impetus, as far and wide as
possible, to the undefined work of freedom.(n37)
As this passage suggests, Foucault's critical histories of the
"practical systems" of rationality that "organize our ways of doing
things"(n38) are at the same time genealogies of the subject of these
rational practices, investigations into the ways in which we have
constituted ourselves as rational agents. And their point is not to
reinforce established patterns but to challenge them. Genealogy is
"practical critique": It is guided by an interest in the "possible
transgression" and transformation of allegedly universal and necessary
constraints. Adopting an experimental attitude, it repeatedly probes the
"contemporary limits of the necessary" to determine "what is not or no
longer indispensable for the constitution of ourselves as autonomous
subjects."(n39)
Let us turn now to the two topics on which we criticized Foucault's
earlier self-understanding: the subject and power. This will enable us
to focus our account of the theoretical shifts in his later work and to
determine more precisely where they leave him in relation to Habermas.
Power Again: Strategic and Communicative Action
My criticisms of Foucault in part II of this article fumed on his one
dimensional ontology: In the world he described, truth and subjectivity
were reduced in the end to effects of power. He escapes this
reductionism in the 1980s by adopting a multidimensional ontology in
which power is displaced onto a single axis. Referring to Habermas in
his first Howison Lecture at Berkeley in Fall 1980, he distinguishes
three broad types of "techniques": techniques of production, of
signification, and of domination.(n40) To this he adds a fourth, namely,
techniques of the self, which subsequently becomes the principal axis of
analysis in the second and third volumes of his History of Sexuality.
These same four dimensions are distinguished (as "technologies") in the
seminar he conducted at the University of Vermont in Fall 1982,(n41) and
the first three of them are elaborated (as "relations") in the Afterword
(1982) to Dreyfus and Rabinow, Michel Foucault: Beyond Structuralism and
Hermeneutics, where, referring once again to Habermas, he notes that
they are not "separate domains" but analytically distinguishable aspects
of social action that "always overlap" in reality.(n42) Thereafter,
Foucault settled on a three-dimensional ontology, not unlike Habermas's
tripartite model of relations to the objective world, to the social
world, and to ourselves.
In volume II of the History of Sexuality, for example, he works with a
distinction between fields of knowledge, types of normativity, and forms
of subjectivity, with three correlated axes of analysis: discursive
practices, relations of power, and forms in which individuals recognize
themselves as subjects.(n43) What immediately strikes one in comparing
this scheme with Habermas's is that normatively structured social
relations are, as a matter of course, construed as relations of power.
Earlier, when rules and norms constitutive of rational practices were
regarded simply as technologies for "governing" and "normalizing"
individuals, this is what one would have expected. But now we have to
wonder what has been accomplished by distinguishing the three
ontological dimensions if we are still left with a reduction of social
relations to power relations. Part of the answer, I think, is a shift of
attention from relations of domination to strategic relations. I want to
suggest, in fact, that Foucault's final ontology tends to equate social
interaction with strategic interaction, precisely the equation that
Habermas seeks to block with his concept of communicative action.
The most elaborate explication of his later notion of power appears in
Foucault's Afterword to the first edition of the Dreyfus and Rabinow
study. There, he construes the exercise of power as "a way in which
certain actions modify others," a "mode of action upon the action of
others," which "structures[s] the possible field of [their] action."(n44)
The relationship proper to power is neither violence nor consensus but
"government," in the very broad sense of "guiding the possibility of
conduct and putting in order the possible outcome."(n45) Viewed in this
way, says Foucault, power is "coextensive with every social relationship,
"(n46) for "to live in society is to live in such a way that action upon
other actions is possible and in fact ongoing."(n47)
Foucault's matter-of-course treatment of social relations as power
relations is less startling once we realize that he now defines the
latter in more or less the same terms that the sociological tradition
has used to define the former. What makes actions social is precisely
the possibility of their influencing and being influenced by the actions
and expectations of others. On Foucault's definition, only actions that
had no possible effects on the actions of others--that is, which were
not social--would be free of the exercise of power. What is at stake
here? Is this merely a rhetorical twist meant to sharpen our awareness
of the ways in which our possibilities of action are structured and
circumscribed by the actions of others? In part, perhaps, but there is
also a metatheoretical--or, in Foucault's terms, ontological--issue
involved. His conceptualization of social interaction privileges
strategic over consensual modes of "guiding the possibility of conduct
and putting in order the possible outcomes."
To see how this is so, we must first take a brief look at his
distinction between power and domination. Whereas earlier, situations of
domination-- asylums, clinics, prisons, bureaucracies, and the like--
were treated as paradigms of power relations generally in the panoptical
society, now they are clearly marked off as a particular type of power
situation:
When one speaks of "power", people think immediately of a political
structure, a government, a dominant social class, the master facing the
slave, and so on. That is not at all what I think when I speak of
"relationships of power." I mean that in human relations, whatever they
are--whether it be a question of communicating verbally . . . or a
question of a love relationship, an institutional or economic
relationship-- power is always present: I mean the relationships in
which one wants to direct the behavior of another.... These relations of
power are changeable, reversible, and understandable.... Now there are
effectively states of domination. In many cases, the relations of power
are fixed in such a way that they are perpetually asymmetrical and the
margin of liberty is extremely limited.(n48)
Thus Foucault now distinguishes "relationships of power as strategic
games between liberties" in which "some people try to determine the
conduct of others" from "the states of domination . . . we ordinarily
call power."(n49) The idea of a society without power relations is
nonsense, whereas the reduction to a minimum of states of domination--
that is, fixed, asymmetrical, irreversible relations of power--is a
meaningful political goal. "Power is not an evil. Power is strategic
games.... To exercise power over another in a sort of open strategic
game, where things could be reversed, that is not evil.... The problem
is rather to know how to avoid . . . the effects of domination."(n50) In
short, whereas "games of power" are coextensive with social relations,
"states of domination" are legitimate targets of political struggle
aimed at freeing up space for open strategic games. "The more open the
game, the more attractive and fascinating it is."(n51)
It is difficult to judge just how far Foucault would have been willing
to take this line of thought. It leads in the end to conceptualizing
social relations as strategic relations and social interaction as
strategic interaction. It would be ironic indeed if his wholesale
critique of modern social theory should finally end in an embrace of one
of its hoarier forms.(n52) But rather than rehearsing the standard
criticisms of game-theoretical approaches to the general theory of
action, I shall remark only on one key issue that separates Foucault
from Habermas.
There are, at least on the face of it, ways of influencing the conduct
of others that do not fit very neatly into the model of strategic games.
Habermas's notion of communicative action singles out for attention the
openly intended illocutionary effects that speech acts may have on the
actions of others.(n53) Establishing relations through the exchange of
illocutionary acts make it possible for speakers and hearers to achieve
mutual understanding about their courses of action, that is, to
cooperate rather than compete in important areas of life. Foucault,
however, treats even the consensus that results from raising and
accepting validity claims---claims to truth, rightness, sincerity, and
so forth--as an instrument or result of the exercise of power.(n54)
Though he avoids any direct reduction of validity to power in his later
work, his definition of power ensures that every communication produces
it: "Relationships of communication," he writes, "produce effects of
power" by "modifying the field of information between parties."(n55) Of
course, if producing effects of power amounts to no more than
influencing the conduct of others, we have here a sheep in wolf's
clothing. Habermas's notion of noncoercive discourse was never intended
to refer to communication that is without effect on the behavior of
others! Foucault comes closer to the real issue when, in an apparent
reference to Habermas, he criticizes the idea of dissolving relations of
power in a "utopia of a perfectly transparent communication."(n56) He
elaborates: "The thought that there could be a state of communication
which would be such that the games of truth could circulate freely,
without obstacles, without constraint, and without coercive effects,
seems to me to be Utopia."(n57) This takes us back to our discussion of
rational practices in part II, and particularly to the idea that "truth
is produced by virtue of multiple forms of constraint." As we saw there,
the point at issue cannot be whether there are "games of truth" without
the constraints of rules, procedures, criteria, and the like. And it
does not seem to be whether constitutive constraints could possibly
obligate participants in a symmetrical and reciprocal manner.(n58) Thus
the problem must arise at the level of the "networks of practices of
power and constraining institutions" in which socially established
"games of truth" are always embedded.(n59) Foucault allows that the link
between games of truth and relations of power need not, in and of itself,
impair the validity or efficacy of what results from them.(n60) So the
question appears to be whether what Habermas calls communication free
from domination, in which claims to validity are decided on the basis of
the force of the better argument, can actually be realized in practice.
And that seems to be a matter of more or less rather than all or
nothing. If this is so, Habermas's idea of national discourse would make
as much sense as a normative ideal as Foucault's notion of a level
playing-field. It would be utopian only in the sense that the full
realization of any regulative ideal is utopian.
The Subject Again: Autonomy and Care of the Self
Foucault's growing emphasis on the "strategic side" of the "practical
systems" that organize our ways of doing things--the freedom we have to
act within, upon, or against them--is not the only way the individual
comes to the fore in his later thought.(n61) His balancing of the
"technological" with the "strategic" in conceptualizing power is
accompanied by a shift of attention from "subjectification" via
"individualizing power" to "self-formation" via "care of the self." This
shift occurred between the publication of volume I of the History of
Sexuality in 1976 and the publication of volumes II and III in 1984. As
Foucault explains it, earlier, in Discipline and Punish and similar
writings, he had been concerned with "techniques for `governing'
individuals" in different areas of life. When he turned his attention to
the genealogy of the modem subject in the History of Sexuality, there
was a danger of "reproducing, with regard to sexuality, forms of
analysis focused on the organization of a domain of learning or on the
techniques of control and coercion, as in [his] previous work on
sickness and criminality."(n62) And this is indeed what we find
happening prior to his work on volumes II and III. In volume I he could
still describe the aim of his study:
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.... What is at issue, briefly, is ... the way in which sex is
`put into dis-course'.... [M]y main concern will be to locate the forms
of power, the channels it takes, and the discourses it permeates in
order to reach the most tenuous and individual mode of behavior.(n63)
In the Tanner Lectures delivered at Stanford three years later (1979),
one still finds a treatment of individuality in relation to
"individualizing power," that is, to "power techniques oriented toward
individuals and intended to rule them in a continuous and permanent
way."(n64) What Foucault calls "pastoral techniques," from Christian
examination of conscience and cure of the soul to contemporary methods
of mental health, are analyzed there as instruments for "governing
individuals by their own verity."(n65) And "governmentality" apparently
continued to serve as the general perspective on individualization in
the years immediately following.(n66)
By 1983, however, the perspective had dearly shifted. In an interview
conducted by Dreyfus and Rabinow in April of that year, Foucault, hard
at work on the later volumes of History of Sexuality, announces that
"sex is boring," and that he is interested rather in techniques of the
self.(n67) Clarifying that remark, he goes on to draw a clear
distinction between technologies of the self geared to normalization and
ethical techniques aimed at living a beautiful life.(n68) What the
Greeks were after, he says, is an aesthetics of existence: "The problem
for them was "the techne of life . . . how to live . . . as well as
[one] ought to live," and that, he tells us, is his interest as well:
"The idea of the bios as material for an aesthetic piece of art is
something which fascinates me."(n69) Accordingly, he now characterizes
the third axis of genealogical-archeological analysis as directed not
toward modes of normalizing subjectification but toward "the kind of
relationship you ought to have with yourself, rapport a soi, which I
call ethics, and which determines how the individual is supposed to
conduct himself as a moral subject of his own action."(n70) Elsewhere,
this is described as a shift from the investigation of "coercive
practices" to the study of "practices of freedom," "exercises of self
upon self by which one tries . . . to transform one's self and to attain
a certain mode of being."(n71) And this "care of the self," which
establishes a form of self-mastery, is now said to be a sine qua non of
properly caring for others, that is, of the art of governing.(n72)
According to Foucault, the search for an ethics of existence that was
stressed in antiquity differed fundamentally from the obedience to a
system of rules that came to prevail in C!Christianity. "The elaboration
of one's own life as a personal work of art, even if it obeyed certain
collective canons, was at the center, it seems to me, of moral
experience, of the will to morality in Antiquity; whereas in
Christianity, with the religion of the text, the idea of the Will of God,
and the principle of obedience, morality took on increasingly the form
of a code of rules."(n73) To be sure, there are "code elements" and
"elements of ascesis" in every morality, prescriptive ensembles of rules
and values as well as ways in which individuals are to form themselves
as ethical subjects in relation to them.(n74) Nevertheless, some
moralities are more "code oriented" and others more "ethics oriented."
In the former, the accent is on code, authority, and punishment, and
"subjectivation occurs basically in a quasi juridical form, where the
ethical subject refers his conduct to a law, or set of laws, to which he
must submit"(n75); in the latter, the main emphasis is on self-formative
processes that enable individuals to escape enslavement to their
appetities and passions and to achieve a desired mode of being, and "the
system of codes and rules of behavior may be rather rudimentary [and]
their exact observance may be relatively unimportant, at least compared
with what is required of the individual in the relationship he has with
himself."(n76) Whereas histories of morality have usually focused on the
different systems of rules and values operative in different societies
or groups, or on the extent to which the actual behavior of different
individuals or groups were in conformity with such prescriptive
ensembles, Foucault's History of Sexuality focuses on the different ways
in which "individuals have been urged to constitute themselves as
subjects of moral conduct," on the different "forms of moral
subjectivation and the practices of the self that are meant to ensure
it."(n77) This choice is motivated in part by his diagnosis of the
present state of morality: "If I was interested in Antiquity it was
because, for a whole series of reasons, the idea of morality as
disobedience to a code of rules is now disappearing, has already
disappeared. And to this absence of morality corresponds, must
correspond, the search for an aesthetics of existence."(n78) Thus the
problem of our present, and of our present selves, to which Foucault's
later work is oriented is an "etho-poetic" one: how to revive and renew
"the arts of individual existence."
This certainly constitutes a major shift from his earlier emphasis on
networks or field of power in which individuals were only nodal points,
and his methodological injunction to do without the subject and modes of
analysis that rely on it. Both the ethical subject and the strategic
subject are now represented as acting intentionally and voluntarily(n79)-
-within, to be sure, cultural and institutional systems that organize
their ways of doing things. But they are not simply points of
application of these practical systems; they can critically and
reflectively detach themselves from them; they can, within limits,
modify them; and they can, in any case, make creative use of whatever
space for self-formation they permit or provide. This model now enables
us to make sense of the possibilities of resistance and revolt which,
Foucault always insisted, are inherent in systems of power. It corrects
the holistic bias we found in his work of the 1970s. The question I
would now like to raise is whether he has gone too far in the opposite
direction and replaced it with an individualistic bias.
Although the later Foucault refers appreciatively to Kant's ideas of
maturity and autonomy, he gives them a very different twist. In "What Is
Enlightenment?," for example, his analysis of Kant's notion of
Mundigkeit is immediately followed by a discussion of Baudelaire's
attitude toward modernity: "Modern man, for Baudelaire, is not the man
who goes off to discover himself, his secrets, his hidden truth; he is
the man who tries to invent himself. This modernity does not `liberate
man in his own being': It compels him to face the task of producing
himself."(n80) In this respect, Baudelaire's attitude is Foucault's own;
but it is certainly not Kant's.(n81) The representation of autonomy as
aesthetic self-invention eliminates the universality at the heart of his
notion--the rational Wille expressed in laws binding on all agents
alike. This is, of course, no oversight on Foucault's part. As we saw,
he distinguishes code-oriented moralities, in which a quasi juridical
subject refers his or her conduct to a set of laws, from ethics-oriented
moralities, in which general rules of behavior are less developed and
less important than individual self-formation. There can be no doubt as
to how he ranks them: "The search for styles of existence as different
from each other as possible seems to me to be one of the points on which
particular groups in the past may have inaugurated searches we are
engaged in today. The search for a form of morality acceptable to
everybody, in the sense that everybody should submit to it, strikes me
as catastrophic."(n82) In the context of his history of sexuality, it is
Christianity that serves as the paradigm of a code-oriented morality:
"The Church and the pastoral ministry shared the principle of a morality
whose precepts were compulsory and whose code was universal."(n83) And
this, it seems to me, is what motivates the either/or approach expressed
in the lines quoted earlier: Universal morality is construed materially
and not formally, that is, in a pre-Kantian manner.
Contemporary neo-Kantians treat justice and the good life as
complementary not opposed concerns. Thus Habermas differentiates the
type of practical reasoning proper to questions of what is morally right
from that concerned with what is ethically prudent.(n84) If questions of
justice are involved, fair and impartial consideration of conflicting
interests is called for; when questions of value arise, deliberation on
who one is and who one wants to be is central. Like Kant, Habermas
regards matters of justice, rather than matters (specifically) of the
good life or of individual self-realization, to be the proper domain of
universalistic morality. This is not to say that ethical deliberation
exhibits no general structures of its own; but the disappearance of
value imbued cosmologies and the disintegration of sacred canopies have
opened the question "How should I (or we, or one) live?" to the
irreducible pluralism and individualism of modern life. To suppose that
it could be answered once and for all, that moral theory could single
out one form of life right for everyone, is no longer plausible. This
does not, however, eliminate the need for a general theory of a more
restricted sort: A theory of justice that reconstructs the moral point
of view from which competing interest and valuebased claims can be
fairly adjudicated. Like Kant, Habermas understands this type of
reasoning to be universal in import; however, he replaces the
categorical imperative with the idea that for general norms to be valid
they have to be acceptable to all those affected by them, as
participants in practical discourse.
I cannot go into the details of that approach here, but enough has been
said, perhaps, to indicate that Foucault's representation of universal
morality, geared at it is to substantive codes, misses the point of
formal, procedural models: namely, to establish a general framework of
justice within which individuals and groups may pursue differing
conceptions of the good or beautiful life. Although Foucault does not
address himself to this most general level of morality, he cannot do
without it. When asked on one occasion if the Greek arts of existence
present a viable alternative to contemporary conceptions of the moral
life, he responded that "the Greek ethics were linked to a purely virile
society with slaves, in which the women were underdogs whose pleasure
had no importance."(n85) That is to say, Greek ethics were tied to
unjust practices and institutions. And when asked on another occasion
whether consensus might not serve as a regulative principle in
structuring social relations, he replied:
I would say, rather, that it is perhaps a critical idea to maintain at
all times: to ask oneself what proportion of non-consensuality is
implied in such a power relation, and whether that degree of non-
consensuality is necessary or not, and then one may question every power
relation to that extent. The farthest I would go is to say that perhaps
one must not be for consensuality, but one must be against non-
consensuality.(n86)
And, as we have seen, Foucault proposes as a goal for political practice,
the transformation of states of domination into open and symmetrical
(fair?) strategic games. In these and other contexts, it is clear that
Foucault conceives the "elaboration of one's life as a personal work of
art" to be limited by considerations of justice: I should not make the
misery of others the condition or consequence of my happiness, their
want of my plenty, their degradation of my elevation, their unfreedom of
my freedom, the ugliness of their lives the basis of the beauty of mine.
This is the unmistakable orientation of Foucault's analyses, and it is
an orientation that calls for its own reflective elaboration. In that
form, universalistic morality is not opposed to, but is a presupposition
of, the search for a personal ethics, if that search is to be open to
everyone.
A similar ease could be made for Sittlichkeit, or shared ethical life.
The overly individualistic optic of Foucault's later work brings culture
and society into view primarily as constraining networks of imposed
rules, prohibitions, values, standards, identities, styles, and so
forth. He does, of course, recognize that self-invention is not a ex
nihilo, that it works with material from the sociocultural environment.
But community and tradition, shared forms of life and collective
identities, common fates and the common good play no central role in the
practice of liberty as he presents it. The problem, it seems to me, lies
precisely in his antithetical conceptualizations individual freedom and
social interaction. Whereas Kant's autonomy consists in a
deindividualized respect for universal law, Foucault's liberty appears
to consist in a desocialized aesthetics of individual existence. As any
operation of the other upon the self is conceived to be an exercise of
power--in which the other governs my conduct, gets me to do what he or
she wants--liberty can only consist in operations of the self upon the
self--in which I govern or shape my own conduct. The one-dimensional
view of social interaction as strategic interaction displaces autonomy
outside of the social network. There are, of course, post-Kantian
alternatives to this in which individual freedom includes reasoned
agreement to the norms of common life, individual identity is formed and
maintained in reciprocal relations with others, and group memberships
contribute to self-fulfillment. Foucault's aesthetic individualism is no
more adequate to the social dimension of autonomy than was the
possessive individualism of early modern political theory.
The same problem turns up in a different form in Foucault's views on the
relation of ethics to politics and society: "The idea that ethics can be
a very strong structure of existence, without any relation to the
juridical per se, to an authoritarian system, a disciplinary structure,"
he tells us, "is very interesting."(n87) "For centuries," he continues,
we have been convinced that between our ethos, our personal ethics, our
everyday life, and the great political and social and economic
structures there were analytic relations, and that we couldn't change
anything, for instance, in our sex life or in our family life, without
ruining our economy, our democracy and so on. I think we have to get rid
of this idea of an analytical or necessary link between this and other
social or economic or political structures.(n88)
And a bit further on, he asks rhetorically: "But couldn't everyone's
life become a work of art?"(n89) In his earlier work, Foucault himself
gave us ample grounds for answering that question in the negative under
existing social, economic, and political conditions. The problem is not
with "analytic and neccessary links" but with de facto empirical
interdependencies between structures and events at the personal and
societal levels. The existence of such interconnections does not, of
course, mean that "we couldn't change anything" in our individual lives
without changing society as a whole. But it does mean that the
conditions of individual existence will be different at different
locations in the social system, and that the possibility of making one's
life into a work of art will be differently distributed. As Hans-Herbert
Kogler has put this point:
The sociocultural resources and opportunities for developing an
autonomous personality are inequitably distributed, and this cannot be
evened out by an ethical choice of self.... That approach leaves fully
unanswered the question of how we might possibly criticize contexts that
themselves render impossible [autonomous] modes of subjectivation.(n90)
Viewed from the perspective of critical social theory, Foucault's later
framework of interpretation lies at the opposite extreme from his
earlier social ontology of power. Then, everything was a function of
context, of impersonal forces and fields, from which there was no escape-
-the end of man. Now, the focus is on "those intentional and voluntary
actions by which men not only set themselves rules of conduct but also
seek to transform themselves . . . and to make their life into an
oeuvre"(n91)--with scant regard for social, political, and economic
context. Neither scheme provides an adequate framework for critical
social inquiry. The ontology of power was too reductive and one-
dimensional for that purpose; the later, multidimensional ontology still
depicts social relations as strategic relations, thus forcing the search
for autonomy, so crucial to the critical tradition, onto the private
path of a rapport a soi. On the other hand, there is the undeniable
power of Foucault's historical-critical studies themselves. Their
strengths are often weaknesses of previous critical social theory, their
nominalism, descriptivism, and historicism not only a complement but a
counterweight to its emphasis on the general, the normative, and the
theoretical. However universal critical theory may be at the level of
concepts and principles, it must, if it is to pursue its practical
interest, reach finally to the variable, contingent, "transformable
singularities" that so occupied Foucault. Abstract conceptions of
justice and the good life remain just that--abstract conceptions-- until
they take on the flesh and bone of concrete historical existence. And
while critical theorists have long recognized that the critique of
impure reason has to eventuate in sociohistorical studies of specific
"practical systems," there have been far too few studies of that kind.
In this regard, Foucault's investigations into the cultural and
historical contexts in which the sciences of man arose and developed,
and his studies of the formation of the moral-rational subject are a
valuable counterpoise to the globalizing discourses about
rationalization that have tended to preponderate. Moreover, his
relentless scrutinizing of the impositions, constraints, and hierarchies
that figure in the emergence and functioning of rational practices
challenges critical theorists to go further than they have in
detranscendentalizing the guiding conception of reason, truth, and
freedom.
Genealogical histories have proved to be an extremely effective means of
problematizing what is taken for granted, of re-turning attention to
what has long been neglected, of raising politically relevant questions
about matters that are usually viewed apolitically. In shaping that
approach, Foucault devoted himself single-mindedly to issues about which
he cared a great deal. Too often his single-mindedness found expression
in an either/or stance toward existing frameworks and modes of critical
inquiry. I have tried to suggest that the strengths of genealogy are
better viewed as complementary to those of classical critical theory.
The point is not to choose between them but to unite them in
constructing theoretically informed and practically interested histories
of the present.
NOTES
(n1.) See Horkheimer's inaugural lecture (1931) as Director of the
Institut fur Sozial-forschung, "The State of Contemporary Social
Philosophy and the Tasks of an Institute for Social Research," in S.
Bronner and D. Kellner, eds., Critical Theory and Society (New York,
1989), 25-36; and his contributions to the Zeitschrift fur
Sozialforschung from the early 1930s, some of which have been collected
in Horkheimer, Critical Theory (New York, 1972) and Horkheimer, Selected
Essays (MIT Press, Forthcoming). Habermas's renewal of this program is
elaborated in The Theory of Communicative Action, vols. I and II (Boston,
1984, 1987). The comparison that follows would look quite different if
its reference point were the version of critical theory developed by
Horkheimer and Adorno in the 1940s, particularly in their Dialectic of
Enlightenment (New York, 1972), which is very close in spirit to the
genealogy of power/ knowledge Foucault practiced in the 1970s. It is
that period of Foucault's work, by far the most influential in the
English-speaking world, which is the other point of reference for the
comparison in this section. The ethic of the self, which he developed in
the 1980s, will be discussed in part III. I will not be dealing with the
first phase(s) of his thought, which came to a close around 1970-71 with
the appearance of "The Discourse of Language" (printed as an appendix to
The Archeology of Knowledge [New York, 1972],215-237), and "Nietzsche,
Genealogy, History" in Language, Countermemory, Practice (Ithaca, 1977),
139-164.
(n2.) This includes the early Horkheimer.
(n3.) The differences are as great among the various members of the
Frankfurt School at the various stages of their careers.
(n4.) M. Foucault, "Questions of Method," in K. Baynes, J. Bohman and T.
McCarthy, eds., After Philosophy (Cambridge, MA, 1987), 100-117, at 112.
(n5.) M. Foucault, "Truth and Power," in Foucault, Power/Knowledge (New
York, 1980), 109-133, at 131.
(n6.) The application of sociological and ethnographic approaches to the
natural sciences has led to similar conclusions.
(n7.) "Truth and Power," 131.
(n8.) M. Foucault, Discipline and Punish (New York, 1979) 27-28.
(n9.) See J. Habermas, Knowledge and Human Interests (Boston, 1971).
(n10.) See. J. Habermas, The Theory of Communicative Action, vol. II.
(n11.) M. Foucault, "Truth and Power," 119.
(n12.) Discipline and Punish, 114.
(n13.) "Truth and Power," 131.
(n14.) Cf. Harold Garfinkel, Studies in Ethnomethodology (Cambridge,
1984).
(n15.) N. Fraser, "Foucault on Modern Power: Empirical Insights and
Normative Confusions," in Fraser Unruly Practices (Minneapolis, 1989),
17-34, at 32.
(n16.) "Truth and Power," 116.
(n17.) "Truth and Power," 114.
(n18.) "Two Lectures," in Power/Knowledge, 78-108, at 98.
(n19.) Ibid.
(n20.) Cf. "Truth and Power," 117.
(n21.) "Two Lectures," 97.
(n22.) See Erving Goffman, Asylums (New York, 1961).
(n23.) See John Heritage, Garfinkel and Ethnomethodology (Cambridge,
1984), 103-134.
(n24.) See, for instance M. Foucault, The History of Sexuality, vol. l
(New York, 1978), 95-96.
(n25.) Ibid., 157.
(n26.) For a discussion of this problem, see David Michael Levin, The
Listening Self (London and New York, 1989), 90ff.
(n27.) Charles Taylor, "Foucault on Freedom and Truth," in David Hoy,
ea., Foucault: A Critical Reader (Oxford, New York, 1986) 69-102, at 91-
93.
(n28.) A revised version of part of the lecture was published as "The
Art of Telling the Truth," in Michel Foucault, Politics, Philosophy,
Culture (New York, 1988), 86-95, at 95. Foucault sometimes writes as if
the analytic of truth in general -- that is, the traditional concerns
with knowledge, truth, reality, human nature, and the like--should be
abandoned as a lost, but still dangerous, cause. At other times, he
represents it as a still viable research orientation, which, however, he
chooses not to pursue. See, for example, "The Political Technology of
Individuals," in L. H. Martin, H. Gutman, and P. H. Hutton eds.,
Technologies of the Self: A Seminar with Michel Foucault (Amherst, 1988),
145-162, at 145. In either case, the fact that he pursues his "ontology
of the present and of ourselves" in separation from any (explicit)
"analytic of truth" constitutes a major difference from Habermas, whose
diagnosis of the present is linked to a continuation of the critical
project Kant inaugurated with his three Critiques. I will not be able to
explore that difference here.
(n29.) In emphasizing the changes in Foucault's self-understanding in
the 1980s, I am taking issue with commentators who stress the continuity
with earlier work, usually by treating Foucault's later redescriptions
of it as accurate accounts of what he was "really" up to at the time.
The frequent (and varied) redescriptions he offers are, in my view,
better read as retrospectives from newly achieved points of view.
Foucault himself was often quite open about the changes. See, for
instance, the three interviews conducted in January, May, and June 1984:
"The Ethic of Care for the Self as a Practice of Freedom," in The The
Final Foucault, James Bernauer and David Rasmussen, eds. (Cambridge, MA,
1988), 1-20; "The Concern for Truth," in Politics, Philosophy, Culture,
255-267 (at 255 he says, "I changed my mind" after the publication of
vol. l of The History of Sexuality); and "The Return of Morality," in
Philosophy, Politics, Culture, 242-254 (where he says essentially the
same thing at 252-253). From his published writings, see, for instance,
the introduction to vol. II of The History of Sexuality: The Use of
Pleasure (New York, 1985), especially "Modifications," 3-13.1 find this
straightforward acknowledgement of a "theoretical shift" hermeneutically
more satisfactory than any of the attempts to read his earlier work as
if it had been written from the perspective of the 1980s. For an
overview of the development of Foucault's thought and the distinctive
features of the last phase, see Hans-Herbert Kogler, "Frohliche
Subjektivitat. Historische Ethik und dreifache Ontologie beim spaten
Foucault," in E. Erdmann, R. Forst, and A. Honneth, eds., Ethos der
Moderne--Foucaults Kritik Aufklarung (Frankfurt, 1990). For a somewhat
different view, see Arnold I. Davidson, "Archeology, Genealogy, Ethics,"
in Foucault: A Critical Reader, 221-233.
(n30.) Translated as "What Is Enlightenment?" in Lewis White Beck, ea.,
Kant on History (New York, 1963), 3-11. Foucault's fullest treatment can
be found in a posthumously published text with the same title, in Paul
Rabinow, ea., The Foucault Reader (New York, 1984), 32-50. In the 1980s,
he repeatedly expressed his appreciation of Kant's essay. In addition to
text just cited, see his Afterword to Hubert Dreyfus and Paul Rabinow,
Michel Foucault: Beyond Structuralism and Hermeneutics (Chicago, 1982),
"The Subject and Power," 145-162, at 145; and "Structuralism and
Poststructuralism: An Interview with Michel Foucault," Telos 55 (1983);
195-211 at 199,206.
(n31.) "The Subject and Power," 216.
(n32.) "Polemics, Politics, and Problematizations," in The Foucault
Reader, 381-390, at 388. Compare Foucault's remark in the introduction
to vol. II of the History of Sexuality that the object of these studies
is "to learn to what extent the effort to think one's own history can
free thought from what it silently thinks, and so enable it to think
differently" (p. 9).
(n33.) "The Ethic of Care for the Self," 4.
(n34.) The concept of "mature adulthood" (Kant's Mundigkeit) is
discussed in "What Is Enlightenment?," 34-35, 39. Dreyfus and Rabinow
deal with this topic in "What Is Maturity? Habermas and Foucault on
`What Is Enlightenment?'," in Foucault: A Critical Reader, 109-121. But
their representation of Habermas's position is misleading on key points,
for example, as regards his views on "phronesis, art, and rhetoric" (p.
111), on authenticity (p. 112), and on reaching agreement (pp. 119-120).
(n35.) "What Is Enlightenment?," 42; see also "The Art of Telling the
Truth," 94-95.
(n36.) "Space, Knowledge, and Power," in The Foucault Reader, 239-256,
at 249.
(n37.) "What Is Enlightenment?," 45-46. Foucault sometimes takes a line
closer to Habermas's, for instance, when he explains that "singular
forms of experience may perfectly well harbor universal structures," in
the original preface to The History of Sexuality, vol. 11, in The
Foucault Reader, 333-339, at 335. But, characteristically, he
immediately goes on to say that his type of historical analysis brings
to light not universal structures but "transformable singularities"
(p.335). As we saw in part II, it nevertheless relies on an interpretive
and analytic framework comprising universalistic assumptions about the
structure of social action. As l shall elaborate later, the same holds
for his later investigations as well, but the framework has been altered
in important respects.
(n38.) "What Is Enlightenment?," 48.
(n39.) Ibid., 43. Foucault explicitly gives preference to "specific" and
"partial" transformations over "all projects that claim to be global or
radical" and "any programs for a new man." (pp. 46-47). Cf. Habermas's
remarks in Knowledge and Human Interests, 284-285. There are many
similarities between the Foucault of "What is Enlightenment?" and the
earlier Habermas, who pursued "an empirical theory Of history with a
practical intent." Cf. my account of this phase of Habermas's thought in
The Critical Theory of Jurgen Habermas (Cambridge, MA, 1978), chaps. 1,
2, and 3.
(n40.) Manuscript, 7. He is apparently referring to the scheme that
Habermas proposed in his inaugural lecture at Frankfurt University in
1965 (printed as the appendix to Knowledge and Human Interests, 301-317)
and subsequently altered. On page 313 Habermas characterized his three
dimensions of analysis as labor, language, and domination (Herrschaft).
(n41.) Technologies of the Self, 18-19. It is clear, however, that he
has not yet fully disengaged from the power ontology, for all four types
of technologies are said to be "associated with" domination, and he
characterizes his new field of interest as "the technologies of
individual domination." One year later, he would no longer talk that
way. See nt. 66.
(n42.) "The Subject and Power," 217-218.
(n43.) The Use of Pleasure, 4. A version of this appears already in his
discussions with Dreyfus and Rabinow al Berkeley in April 1983, "On the
Genealogy of Ethics: An Overview of Work in Progress," in The Foucault
Reader, 340-372, at 351-352. It is elaborated in the original preface to
vol. II, 333-339, as a distinction between fields of study, sets of
rules, and relations to self.
(n44.) "The Subject and Power," 219-221.
(n45.) Ibid., 221.
(n46.) Ibid., 224.
(n47.) Ibid., 222.
(n48.) "The Ethic of Care for the Self," 11-12.
(n49.) Ibid., 19. The categories of power, domination, and strategy are
of course used earlier as well, but not with the same meanings. In vol.
I of The History of Sexuality, for instance, "states of power" are said
to be generated by virtue of the inequality of force relations (p.93),
and power is said to be exercised in nonegalitarian relations (p.94);
"major dominations" arise as the hegemonic effects of wide-ranging
cleavages that run through the social body as a whole (p.94), while
strategies are embodiments of force relations (p.93, with example on pp.
104-105).
(n50.) Ibid.,18.
(n51.) Ibid., 20.
(n52.) Foucault's three-part definition of "strategy" in "The Subject
and Power," 224-225, is conventional enough. It is said to designate (a)
means-ends rationality aimed at achieving some objective, (b) playing a
game with a view to gaining one's own advantage, and (c) the means to
victory over opponents in situations of confrontation.
(n53.) See "What Is Universal Pragmatics?," in Habermas, Communication
and the Evolution of Society (Boston, 1979), 1-68, esp. 59-65; and The
Theory of Communicative Action, vol. I., 273-337.
(n54.) See "The Subject and Power," 220.
(n55.) Ibid., 218.
(n56.) "The Ethic of Care for the Self," 18.
(n57.) Ibid.
(n58.) See Foucault's discussion of the "morality that concerns the
search for truth" in "Polemics, Politics, and Problematizations," 381-
382, where he describes what are essentially symmetry conditions among
dialogue partners. See also his account of the role that communication
with others played in the care of the self, in The History of Sexuality,
vol. III, 51-54. The reciprocity of helping and being helped by others
which he describes there hardly accords with his official view of social
relation as strategic relations.
(n59.) "The Ethic of Care for the Self," 17.
(n60.) Ibid., 16. It is not clear, however, what validity means in this
context, whether, for example, it means more than being produced
according to the constitutive rules of the games and thereby fulfulling
the intended function of the game.
(n61.) Foucault draws a distinction between the "strategic" and
"technological" sides of "practical systems" in, for instance, "What Is
Enlightenment?," 48.
(n62.) The original preface to The History of sexuality, 337-339.
(n63.) The History of Sexuality, vol. I,11.
(n64.) "Omnes et Singulatim: Towards a Criticism of Reason," in The
Tanner Lectures on Human Values, vol. II, Sterling McMurrin, ed. (Salt
Lake City, 1981), 225-254, at 227.
(n65.) Ibid., 240.
(n66.) For example, in his Howison Lectures delivered at Berkeley in
Fall 1980, he describes his project as an investigation of the
historical constitution of the subject which leads to the modem concept
of the self (manuscript, lecture 1, 4) and goes on to say that he is now
focusing on the "techniques of the self" by which "individuals effect a
certain number of operations on their own bodies, on their souls, on
their own thoughts, on their conduct" (p. 7). But though he clearly
distinguishes such techniques from "techniques of domination," they have
to be understood precisely in relation to them (p.7). The "point of
contact" between the two is government: "When I was studying asylums,
prisons, and so on, I insisted too much on the techniques of
domination.... But that is only one aspect of the art of governing
people in our societies.... [Power] is due to the subtle integration of
coercion technologies and self technologies.... Among [the latter],
those oriented toward the discovery and formulation of the truth
concerning oneself are extremely important" (p. 8). Accordingly, in the
closing passage of his lectures, he asks rhetorically whether the time
has not come to get rid of these technologies and the sacrifices linked
to them (lecture II, 20). In the first part of "The Subject and Power,"
208-216, which was delivered as a lecture at the University of Southern
California in Fall 1981, the way in which we turn ourselves into
subjects is described as an element in the "government of
individualization" (p.212). At the same time, however, Foucault notes
the increasing importance of struggles against "forms of subjection"
through "individualizing techniques" (p. 213), against shaping
individuals to ensure their integration into the modem state (p. 214).
And he concludes with a line that could serve as the epigraph of his
last studies: "[T]he political, ethical, social, philosophical problem
of our days is not to try to liberate the individual from the date and
from the state's institutions, but to liberate us both from the state
and from the typo of individualization which is linked to the state. We
have to promote new forms of subjectivity through the refusal of this
kind of individuality which has been imposed on us for several
centuries" (p. 216). But he does not yet have the categories that later
make sense of this project. In the seminar on "Technologies of the Self"
offered at the University of Vermont one year later (Fall 1982),
"governmentality is still identified as the point of "contact between
the technologies of domination and those of the self." The new focus is
characterized as "technologies of individual domination!" (p. 19). In a
lecture on "The Political Technology of Individuals" presented on the
same occasion, however, Foucault distinguishes questions concerning how
"we directly constitute our identity through some ethical techniques of
the self" from questions concerning "the political technology of
individuals," but this thought is not developed there.
(n67.) "On the Genealogy of Ethics," 340. The subordination of his
interest in sexuality as such to a broader problematization of
techniques of self-formation is clearly stated in those volumes. It is,
he writes, with sexual behavior as a "domain of valuation and choice,"
with the ways in which "the individual is summoned to recognize himself
as an ethical subject of sexual conduct" that the later studies are
concerned (vol. II, 32). Thus his analyses of "prescriptive discourses"
about diatetics, household management, erotica, and so forth focus on
the mode of subjectivation presupposed and nourished in the
corresponding practices. Very briefly: the genealogy of "desiring man"
as a self-disciplined subject" is Foucault's key to the genealogy of the
"subject of ethical conduct" (vol. II, 250-251); and this is itself an
element in a more comprehensive "history of truth" (vol. II, 6).
Similarly, in analyzing parrhesia, or truth-telling, in Antiquity,
Foucault conceives of the genealogy of the parrhesiastic subject, the
truth-teller, as part of the "genealogy of the critical attitude in
Western philosophy." Discourse ant Truth: The Problematization of
Parrhesia, transcription by Joseph Pearson of a seminar given at the
University of California, Berkeley, Fall 1983,114. These connections
suggest the continuing relevance of Foucault's work to what I referred
to as the critique of impure reason.
(n68.) Ibid.,341.
(n69.) Ibid., 348.
(n70.) Ibid., 352.
(n71.) "The Ethic of Care for the Self," 2-3.
(n72.) Ibid., 6-7. The connections between governmentality, care of the
self, and strategic interaction are suggested on pages 19-20 of the same
interview: "[I]n the idea of governmentality I am aiming at the totality
of practices by which one can constitute, define, organize,
instrumentalize the strategies which individuals in their liberty can
have in regard to each other. It is free individuals who try to control,
to determine, to delimit the liberty of others, and in order to do that,
they dispose of certain instruments to govern others. That rests indeed
on freedom, on the relationship of the self to self and the relationship
to the other." On the relation between self-mastery and the mastery of
others in antiquity, see The History of Sexuality, vol. II, 73ff.
(n73.) "An Aesthetics of Existence," in Philosophy, Politics, Culture,
47-53, at 49. Foucault's later studies abound in comparisons between
ethical practices in Antiquity and in Christianity. See, for example,
The History of Sexuality, vol. II., 92, 136-139, and vol. III, 68,
140ff., 165, 235ff. These comparisons, so patently unfavorable to
Christianity, bespeak Foucault's own commitment to an ethopoetics of
existence. In my view, an analysis of the normative and evaluate
presupposition underlying Foucault's later work--carried out along the
lines of Fraser's analysis of his earlier studies (see n. 15)-would
reveal them to be more or less the ones he openly espoused in the
lectures, interviews, and methodological asides of the last period. ID
neither phase should the absence of explicit value judgments in his
sociohistorical studies obscure the presence of implicit value
orientations underlying them. (Cf. Weber's distinction between
Werturteile and Wertbeziehungen.) But the large gap between the official
and the operative frameworks of evaluation in the 1970s was considerably
closed in the 1980s.
(n74.) The Use of Pleasure, 25-26.
(n75.) Ibid., 29.
(n76.) Ibid., 30.
(n77.) Ibid., 29.
(n78.) "An Aesthetics of Existence," 49. See also "On the Genealogy of
Ethics," 343; and "The Concern for Truth," 262-263.
(n79.) See The Use of Pleasure, 10.
(n80.) "What Is Enlightenment?," 42.
(n81.) Nor, for that matter, is it Socrates', Plato's, or Aristotle's.
There is more than one way to take issue with Foucault's notion of, an
ethics of self-invention. I will be stressing Kant's connection of
autonomy to a rational will, but problems could also be raised from the
standpoint of the ethics of community, character, virtue, and the like.
(n82.) "The Concern for Truth," 253.
(n83.) The Use of Pleasure, 21.
(n84.) See his Moral Consciousness and Communicative Action (Cambridge,
MA, 1990) and The Theory of Communicative Action, vol. II, 92-111.
(n85.) "On the Genealogy of Ethics," 44. The masculinist and dominative
orientation of Greek ethics is stressed throughout vol. II of The
History of Sexuality. See, for example,82-86,146-151, 69-77, 215-225.
(n86.) "Politics and Ethics: An Interview," in The Foucault Reader, 373-
380, at 379.
(n87.) "On the Genealogy of Ethics," 348.
(n88.) Ibid., 350.
(n89.) Ibid.
(n90.) "Frohliche Subjektivitat," manuscript, 29.
(n91.) The Use of Pleasure, 10.
~~~~~~~~
By THOMAS McCARTHY, Northwestern University
Thomas McCarthy is Professor of Philosophy at Northwestern University.
He is author of several books, including The Critical Theory of Jurgen
Habermas (MIT Press, 1978) and general editor of the MIT Press Studies
in Contemporary German Social Thought.
Copyright of Political Theory is the property of Sage Publications Inc.
and its content may not be copied without the copyright
holder's express written permission except for the print or
download capabilities of the retrieval software used for
access. This content is intended solely for the use of the
individual user.
Source: Political Theory, Aug90, Vol. 18 Issue 3, p437, 32p.
Item Number: 9703116693
