Michael J.

Current
[email protected]

- - - - -

Paul Patton
Department of General Philosophy
The University of Sydney
Sydney
2006 Australia
(Currently: [email protected])

Foucault's Subject of Power

Political Theory Newsletter , Volume 6, No. 1, May 1994, pp.60-71.

"Three centuries ago certain fools were astonished because Spinoza

wished to see the liberation of man, even though he did not believe
in his liberty or even in his particular existence. Today, new
fools, or the same ones reincarnated, are astonished because the
Foucault who had spoken of the death of man took part in political
struggle."1

Criticism of Foucault returns constantly to two themes: first, his

descriptive analyses of power provide us with no criteria for judgment,
no basis upon which to condemn some regimes of power as oppressive or to
applaud others as involving progress in human freedom. As Nancy Fraser
puts this objection, "Because Foucault has no basis for distinguishing,
for example, forms of power that involve domination from those that do
not, he appears to endorse a one-sided, wholesale rejection of modernity
as such ... Clearly, what Foucault needs, and needs desperately, are
normative criteria for distinguishing acceptable from unacceptable forms
of power". (2) Second, critics complain that he offers no alternative
ideal, no conception either of human being or of human society freed from
the bonds of power. The lack of recourse to any philosophy of the subject
is often taken to explain the political weakness of Foucault's position:
thus, Habermas argues that it is because there is no conception of the
properly human subject in his work that Foucault is left with only the
"arbitrary partisanship of a criticism that cannot account for its
normative foundations". (3) For such critics, Foucault offers only a bleak
political horizon on which the subject will always be an effect of power relations,
and on which there is no possibility of escape from domination of one sort or
another. For others, such as Ian Hacking, the problem is not so much that Foucault
is pessimistic, it is that "he has given no surrogate for whatever it is that
springs eternal in the human breast". (4)
Despite the anti-humanism of Foucault's approach, I wish to argue
that his work does presuppose a certain conception of human being, and
that he does offer a surrogate for hope. What he offers is an analogue of
the grounds for Kant's belief in the possibility of human progress, that
is an historically grounded belief in the human capacity to transcend
limits to the autonomous use and development of human powers. After all,
his analyses of knowledge, power and sexual ethics are all concerned with
these modalities of cultural experience insofar as they affect human
social being. These analyses are undertaken with the aim of producing
critical effects upon present ways in which social reality is understood.
Foucault's characterization of these studies as contributions to
political struggles against individualizing technologies of power (5) would be
paradoxical, along the lines suggested by Deleuze's remark quoted above, in the
absence of some such belief.
It is true that Foucault makes no use of the traditional humanist
forms of critique. He argues that when philosophers invoke "man" as the
basis for their moral and political judgments, they invoke no more than
their own or others concepts of human nature, which are themselves the
products of particular, historically constituted regimes of truth. His
refusal to rely upon concepts of an essential human nature is evident in
his reluctance to use terms such as "ideology" or "liberation".
Nevertheless, his genealogies do presuppose a conception of the human
material upon which power is exercised, or which exercises power upon
itself. This human material is active; it is an entity composed of forces
or endowed with certain capacities. As such it must be understood in
terms of power, where this term is understood in its primary sense of
capacity to do or become certain things. This conception of the human
material may therefore be supposed to amount to a "thin" conception of
the subject of thought and action: whatever else it may be, the human
subject is a being endowed with certain capacities. It is a subject of
power, but this power is only realised in and through the diversity of
human bodily capacities and forms of subjectivity. Because it is a
"subject" which is only present in various different forms, or
alternatively because the powers of human being can be exercised in
infinite different ways, this subject will not provide a foundation for
normative judgment of the kind that would satisfy Fraser or Habermas: it
will not provide any basis for a single universal answer to the question,
"Why ought domination to be resisted?". (6) However, the theory of power
which frames this conception of the human subject does provide a means to
distinguish domination from other forms of power. Moreover, given certain
minimal assumptions about the nature of human being, and about the
particular capacities which human beings have acquired, Foucault's
conception of the subject does provide a basis on which to understand the
inevitability of resistance to domination.

Foucault's subject of power

In Discipline and Punish and The History of Sexuality Vol. I,

Foucault describes strategies of power whose object or target is
primarily the human body. Contrary to the view of critics such as Peter
Dews, this body is no mere inert matter upon which power is exercised and
out of which "subjects" are created. (7) It is a body composed of forces and
endowed with capacities. In Discipline and Punish, it is precisely in
order to address these bodily forces that the techniques of discipline
are deployed. In The History of Sexuality Vol. I, the body which has come
to be constituted as the bearer of a sex, during the course of European
history, is a body explicitly described as one capable of pleasures. The
strategy of pedagogic control of children's enjoyment of such pleasures
is one essential moment in the elaboration of the modern "experience" of
sexuality; the classification of adult pleasures into normal and
pathological is another. The two other grand strategies which Foucault
discusses also involve specific capacities of the female body, chiefly
those connected with reproduction and childbirth. The elaboration of what
we have come to take for granted as human "sexuality" thus may be
understood to have involved a certain classification, ordering and
finalisation of this range of capacities for being and doing certain
things. It is precisely because there is nothing natural about this
construction, and because it was fabricated upon an active body, a body
understood in terms of primary capacities and powers, that it was
accompanied by resistance. Similar remarks may be made about the
disciplined body.

How did so many critics manage to overlook this conception of the

body as subject of power? Perhaps because Foucault's writings during the
1970's tended to employ a language of bodies and forces in place of the
traditional terminology of political critique. Power relations were
characterized in terms of conflict or alliance between forces, engendered
on the basis of "the moving substrate of force relations" (8) which
constitutes the social field. While this language appeared to de-humanize
the social field entirely, abstracting from any notion of human agents or
agency, Foucault nevertheless sought to address the kinds of historical
phenomena which would ordinarily be regarded as the effects of human
agency by means of an impersonal, non-subjectivist language of strategy
and tactics. "Strategy" referred to the operationalisation of the social
field in particular ways, such as the attempt to produce an orderly,
obedient and productive population; "tactics" referred to the disposition
of forces employed to achieve strategic ends. Later discussions such as
"The Subject and Power " appear to revert to a more familiar language of
human agency: power relations are said to arise whenever there is action
upon the actions of others. (9) In other words, power relations are
conceived here not simply in terms of the interaction of impersonal or
inhuman forces, but in terms of action upon the action of "free" agents.
However, "free" means no more than being able to act in a variety of
ways: that is, having the power to act in several ways, or not being
constrained in such a fashion that all possibilities for action are
eliminated. Here too, the subject of power relations according to
Foucault is defined as a being endowed with certain capacities or
possibilities for action; in effect, a subject of power.

Nor are these two ways of conceptualising power relations as

different from one another as they might at first appear. Ordinarily,
what distinguishes an action from a mere bodily movement is the fact that
it is voluntary rather than involuntary motion, and that it is intended
to serve some purpose. Actions are intentional, goal-oriented movements
or dispositions of bodily forces. Strategies are likewise intentional,
goal-oriented movements or dispositions of forces. Strategies need not be
the work of a single strategist, but can just as well be the product of
more or less collective processes of calculation. Nothing that Foucault
says about the subject of power suggests that human agency is in
principle radically different. Nothing commits him to a voluntarist
rather than a Hobbesian or a Nietzschean notion of the will.

But Foucault is committed to the view that social relations are

inevitably and inescapably power relations. On his view, there is no
possible social field outside or beyond power, and no possible form of
interpersonal interaction which is not at the same time a power relation:
" ... to live in society is to live in such a way that action upon other
actions is possible - and in fact ongoing. A society without power
relations can only be an abstraction". (10) This view is sometime to imply that
domination is inevitable, according to Foucault, or that there is no possibility of
progress in human affairs in the sense that social
relations may become less oppressive. Such conclusions are based upon
misunderstanding Foucault's use of the concept of power. One source of
confusion is the failure to make the necessary distinctions between
power, power over and domination. In his later discussions of power,
Foucault does make these distinctions explicit, and in doing so refutes
the charge that his approach is incapable of distinguishing forms of
power that involve domination from those that do not.
Power and Domination

In order to make sense of Foucault's use of the term, "power" must

be understood in its primary etymological sense, as the capacity to
become or to do certain things. Power in this primary sense is exercised
by individual or collective human bodies when they act upon each other's
actions; in other words, to take the simplest case, when the actions of
one affect the field of possible actions of another. In this case, where
the actions of A have succeeded in modifying the field of possible
actions of B, we can say that A has exercised power over B. (11) "Power
over" in this sense will be an inescapable feature of any social
interaction. Moreover, characterised in these terms, with reference only
to Foucault's thin subject of power, it is a normatively neutral concept.
It involves no reference to action against the interests of the other
party. After all, there are many ways in which agents can exercise power
over other agents, only some of which might be detrimental to the
"interests" of the one over whom power is exercised: I can affect the
actions of another by providing advice, moral support, or by passing on
certain knowledge or skills. All of these will involve the exercise of
power over the other, but not necessarily in ways that the other will
find objectionable.

The exercise of power over others will not always imply effective
modification of their actions. Precisely because power is always
exercised between subjects of power, each with their own distinct
capacities for action, resistance is always possible: "where there is
power, there is resistance". (12) For this reason, it is only in exceptional
circumstances that A can be sure of achieving the desired effect on B. Only when
the possibility of effective resistance has been removed does the power relation
between two subjects of power become unilateral and one-sided: "A relationship of
confrontation reaches its term, its final moment (and the victory of one of the two
adversaries) when stable mechanisms replace the free play of antagonistic
reactions. Through such mechanisms one can direct, in a fairly constant manner and
with reasonable certainty, the conduct of others". (13) In such cases, we have
something more than the exercise of power over another, namely the
establishment of a state of domination: in these cases, "... the
relations of power, instead of being variable and allowing different
partners a strategy which alters them, find themselves firmly set and
congealed". (14) Bentham's Panopticon provides a model of such mechanisms
for controlling the conduct of others: the asymmetrical structure of
visibility which is the key to the architectural design maps onto the
fixed asymmetrical distribution of power which defines every system of
domination. Traditional family relations provide Foucault with another
illustration of the same structure of fixed and asymmetrical power
relations. Within the eighteenth and nineteenth century institution of
marriage the wife was not entirely deprived of power, she could be
unfaithful to the husband, steal money or refuse sexual access: "She was,
however, subject to a state of domination in the measure where all that
was finally no more than a certain number of tricks which never brought
about a reversal of the situation". (15)

Foucault is not the first to identify domination with stable and

asymmetrical systems of power relations. His definition does, however,
make it clear that such systems are secondary results, achieved within or
imposed upon a primary field of relations between subjects of power.
Moreover, as with the definition of "power over", his concept of
domination is non-normative. Domination allows more or less predictable
control of the actions of others. Beyond that, little is said about the
purposes for which such states are established and maintained. One
frequent purpose served by states of domination is to enable some to
extract a benefit from the activity of others: economic exploitation in
all its forms, from slavery through to the system of extraction of suplus
value which Marx identified as the secret of capital, depends upon such
systems of domination. C. B. Macpherson coined the useful term
"extractive power" in order to describe the capacity that some people
acquire to employ or make use of the capacities of others. He argues that
the system of private property and a free market in labour operates as a
mechanism for the continuous transfer of part of the power of the class
of non-owners to the class of owners. (16) In Foucault's terms, the
exclusive ownership of means of production amounts to a system of
domination which underpins the extractive power of a social class.

However, while extractive power may always presuppose some system of

domination, states of domination may occur in situations where the flow
of capacities or benefits is non-extractive. For example, Hobbes presents
the relation of subjects to sovereign power as one of domination, since
the sovereign has "the use of so much Power and Strength conferred on
him, that by terror thereof, he is inabled to forme the wills of them
all, to Peace at home and mutuall ayd against their enemies abroad". (17) In
Hobbes' account, the relationship of domination which obtains between
State and citizens is a condition of maintaining the rule of law. In this
case, the transfer of power precedes domination since it is the conferral
of power by parties to the social contract which constitutes sovereign
power. The purpose of this system of domination is not further extraction
but the enhancement of the powers of its subjects.

Pedagogic relations are another sphere in which a measure of

domination may be acceptable, at least during some part of the
educational process. Foucault uses this example in order to suggest that
asymmetrical power relations are not in themselves evil: "The problem is
... how you are to avoid in these practices - where power cannot not
play and where it is not in itself bad - the effects of domination will
make a child subject to the arbitrary and useless authority of a teacher,
or put a student under the power of an abusively authoritarian professor,
and so forth". (18) The qualifying clauses attached to the objectionable
cases of domination in these remarks suggest that other "effects of
domination" may not be objectionable. This indeed appears to be
Foucault's general position: the exercise of power over others is not
always bad, and states of domination are not always to be avoided.

Resistance, autonomy and freedom

Foucault does believe that the fact of widespread resistance to

forms of individualising power is evidence of the need for "a new economy
of power relations". (19) But what is meant by this phrase, and what is the basis
for such a recommendation? In global terms, to call for a new
economy of power relations is to invoke the possibility of a different
articulation of the forms of social and political domination, the forms
of reversible or non-coercive exercise of power over others, and
indvidual or collective capacities. It implies that, contrary to the
experience of European modernity, the enhancement of collective
capacities need not be linked to increase of domination. At the
individual level, a person's power to do or be certain things will also
be the result of a certain "economy", comprising relations to oneself,
relations to others, and relations to forms of discourse and modes of
thought which count as truth. These are in effect Foucault's three axes
of subjectification, and they serve to remind us that a minimal concept
of persons should refer to a body that is trained or cultivated in
certain ways, a set of relations to oneself and one's capacities (an
"ethics"), and a set of relations to modes of interepretation of one's
relations to self and others. Different powers may result from change
along any of these axes, or from changes in the larger networks of social
relations within which these personal capacities are exercised.

Recommendations such as this bring us back to the problem of the

lack of normative criteria in Foucault's work. To suggest as he does that
we need new forms of articulation of personal capacity, power over others
and mechanisms of domination appears to imply the possibility of
principles which might legitimize one "economy" of power relations as
better than another. Could Foucault adopt such principles while remaining
consistent with his theoretical anti-humanism? Rather, as the critics
have argued, he does not and cannot provide such criteria. It seems that
by confining himself to the very thin notion of human being as a subject
of power, Foucault deprives himself of the means to provide such
normative criteria. It does not follow from this, however, that he has no
basis upon which to distinguish between forms of power that involve
domination and those that do not. Nor does it follow that Foucault's thin
conception of human being cannot be filled out in a manner which explains
both resistance to domination and the possibility of transforming
existing economies of power. I suggest that Foucault does employ such a
robust conception of human being in his later work. However, far from
providing universal criteria which would allow us to distinguish
acceptable from unacceptable forms of action upon the action of others,
his approach exposes the limitations of the demand for such criteria. In
order to show how it does this, it will be helpful to pursue the
comparison with Macpherson.

Macpherson contrasts his concept of extractive power with another

concept which he calls "developmental power". The latter refers to an
individual's ability to use and develop his or her "essentially human
capacities". He then uses this concept in order to define a truly
democratic society as one which maximizes the conditions for the exercise
of developmental power. In other words, a truly democratic society is one
which seeks to maximize the ability of all to use and develop their
essentially human capacities. Leaving aside the question what capacities
are to count among the "essentially human", it is clear that the concept
of developmental power has a normative content. In effect, it provides
Macpherson with an ideal standard by which to judge the "democratic
quality" of any society. For he argues that the degree of deveopmental
power can be measured by reference to the presence or absence of
impediments to the use and development of human capacities by all members
of the society. On this basis, he is able to show that the structure of
social relations which gives one class extractive power in relation to
another class is incompatible with maximizing the developmental power of
those who are exploited in this way. More generally, Macpherson's concept
of developmental-power democracy provides a moral basis on which to
reject any system of domination which sustains a form of extractive
power. Any such system is incompatible with all being able to maximize
their ability to use and develop their own powers.

In fact, Macpherson's principle of maximizing developmental power

excludes all systems of domination and not only those which sustain forms
of extractive power. In order to show this, we need to reconsider his
concept of "essentially human capacities". While he does provide a list
of human capacities likely to be included among the essentially human
(capacity for rational understanding; for moral judgement and action;
aesthetic creation and contemplation etc), Macpherson is reluctant to
specify a determinate set of capacities which define human being. In
part, this is because he has a conception of human being as essentially
capable of development. The concept of developmental power refers to the
ability of individuals to use and develop their capacities. This implies
that new capacities might be developed, or that existing ones might be
developed in ways that cause revisions in what is considered to be
essentially human. Macpherson writes: "... the full development of human
capacities, as envisioned in the liberal-democratic concept of man - at least in
its most optimistic version - is infinitely great. No inherent
limit is seen to the extent to which ... human capacities may be
enlarged". (20)

Foucault's conception of human being in terms of bodies

(differentially) endowed with capacities for action is similarly
open-ended. He allows in the manner of Nietzsche that new human
capacities may come into existence as effects of forms of domination,
only to then become bases of resistance to those same forms of
domination. Deleuze takes this Nietzschean thought a stage further in
suggesting that the same forces which defined "man" have already begun to
connect with new, non-human forces: "Spinoza said that there was no
telling what the human body might achieve, once freed from human
discipline. To which Foucault replies that there is no telling what man
might achieve 'as a living being', as the set of forces that resist". (21) However,
Foucault's analyses of the different ways in which human beings are made subjects
expose one further form of domination which Macpherson does not address, perhaps
because of his focus upon extractive power.Determinate forms of subject may arise
as a result of historical
processes not directly connected with extractive power, as Foucault
argues with regard to modern sexuality in The History of Sexuality Vol.
I. Once established, such forms of subjectivity, or at least the forms of
knowledge, social relations, legal and other administrative arrangements
which sustain them, may amount to more or less fixed modalities of power
over individuals. As such, they constitute impediments to the ability of
some individuals to use and develop their human capacities in particular
ways, notably those identified as abnormal or deviant in a social,
medical or psycho-sexual sense. In this manner, the ways in which
certain human capacities become identified and finalized within
particular forms of subjectivity - the ways in which power creates
subjects - may also amount to a modality of domination.

In order to see that Macpherson is equally committed to including

such impediments among the limits to developmental power, we need only
consider the further capacity that he adds to his list of the essentially
human capacities, almost as an afterthought, namely the suggestion that
the exercise of human capacities, "to be fully human, must be under one's
own conscious control rather than at the dictate of another". (22) The loss of an
individual's ability to use his energies humanly, "in accordance with his own
conscious design", (23) plays a significant role in his account of the power which
is lost when individuals are forced to work under the control of others in order to
exercise their capacity for productive activity. In effect, the capacity for
relatively autonomous use and development of one's capacities is a meta-capacity, a
means of directing and experiencing the exercise of the other capacities of a
particular body or determinate subject. Examples of its employment might include
inventing and regulating one's use of a different economy of pleasures, or self-
consciously developing the attributes necessary to operate effectively in a given
political environment. As these examples suggest, there is no reason to expect that
such degrees of autonomy will be developed by individuals acting alone rather than
in the context of
movements for change in certain aspects of social life.

Foucault invokes the same meta-capacity for autonomous use and

development of human powers in his characterization of the ethos of
modernity in "What is Enlightenment?". Drawing on Kant's characterisation
of enlightenment as a process voluntarily embarked upon by some and aimed
at the removal of limits to the exercise of the human power of rational
self-determination, Foucault describes "modernity" as involving a
similarly self-critical attitude towards our present forms of social
being. Moreover, just as Macpherson suggests that increases in
developmental power may be negatively measured by the removal of limits
to its exercise, so Foucault's account implies that progress in this
critical task may be measured by the degree to which present limits to
what it is possible to do or be have been overcome. Criticism, both
theoretical and practical he says, "will be oriented toward the
"contemporary limits of the necessary", that is, toward what is not or is
no longer indispensable for the constitution of ourselves as autonomous
subjects". (24)

Modernity understood as an ethos of permanent self-criticism

presupposes the existence of possible subjects of such activity. Such
subjects will necessarily be free in the sense that their possibilities
for action will include the capacity to undertake this self-critical
activity which Foucault calls "work carried out by ourselves upon
ourselves as free beings". (25) So long as human capacities do in fact
include the power of individuals to act upon their own actions, we can
see that Foucault's conception of human being in terms of power enables
us to distinguish between those modes of exercise of power which inhibit
and those which allow the self-directed use and development of human
capacities. To the extent that individuals and groups acquire the
meta-capacity for the autonomous exercise of certain of their own powers
and capacities, they will inevitably be led to oppose forms of domination
which prevent such activity.

In this appeal to human autonomy, Foucault affirms a belief in human

freedom which appears to contradict his suspicion of modern humanism. How
then does his position differ from that of humanists such as Macpherson,
who treats the capacity for autonomous action as a defining property of
essentially human being, or the Critical Theorists who advocate the
commitment to autonomy as a universal moral ideal? This apparent
contradiction disappears once we take into account two features of
Foucault's position: first, the fact that the suspicion of humanism is
motivated above all by mistrust of the attempt to set limits to human
freedom. "What I am afraid of about humanism is that it presents a
certain form of our ethics as a universal model for any kind of
freedom". (26) Second, the fact that Foucault's appeal to a principle of
autonomy is not grounded in a metaphysical conception of human being as
essentially free but in an analytics of power. From at least "The Subject
and Power" onwards, Foucault suggests that freedom is the ontological
precondition of politics and ethics. However, this is an historical
rather than a transcendental ontology. Freedom here is not the
transcendental condition of moral action, as it is for Kant, but rather
the contingent historical condition of action upon the actions of others
(politics) and of action upon the self (ethics). Just as for Foucault
political power exists only in the concrete forms of government of
conduct, so freedom exists only in the concrete capacities to act of
particular agents. As a result, the subject of freedom is in effect a
subject of power in the primary sense of that term.

In this perspective, autonomy must be understood as a capacity to

govern one's own actions which is acquired by some people, in greater or
lesser degree, and in respect of certain aspects of their bodies and
behaviour. However it has been acquired and in whatever manner it is
distributed, this capacity for autonomous action is sufficient to explain
resistance to forms of domination. To the extent that domination enables
the direction of the actions of others, or even simply establishes more
or less fixed limits to the ways in which human capacities may be
exercised, then states of domination will always constitute limits to the
autonomy of those subject to them. In the attempt to exercise their
capacity for autonomous action, those subject to relations of domination
will inevitably be led to oppose them. It is not a question of advocating
such resistance, of praising autonomy or blaming domination as respective
exemplars of a good and evil for all, but simply of understanding why
such resistance does occur. Foucault does not think that resistance to
forms of domination requires justification. To the extent that it occurs,
such resistance follows from the nature of particular human beings. It is
an effect of human freedom.

Power and agency

The fact that human beings have acquired this capacity at all
presupposes the kinds of internal division within the self which
Nietzsche saw as resulting from the human will to power turned back
against its subject. The kinds of self-regulation of one's own body and
its sexual relations with others described in The Use of Pleasure are
evidence of the existence of such autonomy, however partial and
restricted in scope. The freedom of the subject in the Greek ethics of
moderation and self-mastery was, Foucault suggests, more than just an
emancipation from external or internal constraint: "in its full, positive
form, it was a power that one brought to bear on oneself in the power
that one exercised over others". (27) Here, as in many places, Foucault's
language recalls the Nietzschean origins of his conception of human being
in terms of power. In order to appreciate the more robust conception of
human being which informs Foucault's later work, and in order to see why
this leads him away from rather than towards normative criteria for
distinguishing acceptable from unacceptable forms of power, is useful to
look more closely at Nietzsche's conception of will to power.

Earlier, I suggested that the root concept of Foucault's concept of

power is the notion of capacity. For bodies with the complexity and
specific powers of human beings, power is the capacity for various kinds
of action upon oneself and others. What kinds of action a human body is
capable of will depend in part upon its physical constitution, in part
upon the enduring social and institutional relations within which it
lives. But the kinds of action of which a human body is capable will also
depend upon the moral relations which define its acts. Moral
interpretations of phenomena are among the most important means by which
human subjects act upon themselves and others: it is by such means that
one can arouse pity in others, or experience one's own actions as
cowardice or humility according to whether one lives in the moral culture
of ancient Greece or European christianity. By examples such as these,
Nietzsche draws attention to the interpretative dimension of human
action. The systems of knowledge and moral judgment which Foucault
studied in relation to mental illness, punishment and sexuality are no
less elements of the interpretative framework within which Europeans have
acted upon the action of others. In this sense, Foucault's history of
systems of thought involves a thought which "can and must be analyzed in
every manner of speaking, doing or behaving in which the individual
appears and acts as subject of learning, as ethical or juridical subject,
as subject conscious of himself and others. In this sense, thought is
understood as the very form of action ...". (28)

However, the peculiarity of human action is that it is not only

conscious but self-conscious: we are happy or sad according to whether
our actions produce a feeling that our power is enhanced or a feeling
that it is diminished. In other words, our own actions, and the actions
of others upon us, produce affective states and these affective states in
turn affect our capacity to act. In effect, there is a feedback loop
between the success or otherwise of one's attempts to act and one's
capacity to act. Nietzsche drew attention to the importance of this
self-reflective dimension of human action in insisting upon the primacy
of the "feeling of power" in his analysis of willing in Beyond Good and
Evil: "he who wills believes with a fair amount of certainty that will
and action are somehow one; he ascribes the success, the carrying out of
willing, to the will itself, and thereby enjoys the increase of power
that accompanies all success". (29) On the basis of such remarks, Mark
Warren argues that Nietzsche's theory of the will to power must be
understood primarily as an account of the conditions of the human
experience of agency. This is an historical rather than an a priori
account: given the emergence of self-consciousness in the human animal,
and given the relative weakness of this animal, Nietzsche claims that the
striving to achieve the feeling of power has become humankind's strongest
propensity: "the means discovered for creating this feeling almost
constitute the history of culture". (30) Nietzsche's historical account of the
human experience of power also functions as an argument for the
overriding importance of this experience. As Warren suggests, "In being
conscious and self-conscious, humans increasingly strive less for
external goals than for the self-reflective goal of experiencing the self
as agent". (31) This has important consequences for our approach to
politics: if the experience of autonomy depends upon the larger networks
of practice and social relations within which individuals act, but also
upon the interpretative frameworks in terms of which they judge the
success or failure of their acts, then maximizing autonomy requires
practices of government of self and others which effectively enhance the
feeling of power.

If we assume that Foucault's conception of human being as a subject

of power also includes the interpretative and affective dimensions of
agency as these are defined by Nietzsche, then some of the background
assumptions of his later work become clearer. First, his reliance upon
the experience of limits to freedom as the basis for social change. In
"The Subject and Power", Foucault takes as the starting point for the
analysis of power relations the existence of resistance to current
structures of domination: "opposition to the power of men over women, of
parents over children, of psychiatry over the mentally ill, of medicine
over the population, of administration over the ways people live". (32) No doubt,
the fact of resistance provides evidence that there is a capacity for relatively
autonomous action by individuals with respect to certain areas of social life.
However, Foucault's "thin" conception of human being as a subject of power provides
only the conceptual minimum required to describe the capacities of particular
situated, corporeal subjects. These will result from the techniques of formation
applied to the bodies of such subjects, as well as from the social relations within
which they live and act. In order to account for the experience of these systems of
power as forms of domination, as limits to individual's capacities for action,
Foucault must presuppose the existence of particular forms of self-interpretation
and the existence of something like the feeling of powerlessness. In other words,
he must suppose a fuller conception of human subjectivity which takes into account
both the interpretative and the self-reflective dimensions of human agency. Such a
conception is needed in order to explain both the feeling of power and the lack of
a sense of agency that is so often recorded as part of the experience of
oppression.

Second, if we accept Nietzsche's claim that forms of moral judgment

are among the most important means of self-interpretation, and his view
that what is important for human beings is the experience of the feeling
of power, or what Warren calls experiences of agency, then it follows
that effective moral values are dependent upon the conditions of such
self-experience. In other words, values are internal to types of
individual and social being, not independent of them. That is why
Foucault does not seek to provide universal moral norms or criteria of
evaluation, but instead offers a cautious recommendation of the Greek
practice of an "ethics of existence". (33) This might be read as a proposal for a
different economy of power with respect to our sexual being: an economy different
from that of the ancient Greek men, for whom
self-mastery and moderation in the use of pleasures was both conditioned
by and predicated upon relations of domination over others, notably women
and slaves; but also different from the modern regulation of sexual
conduct by means of legal and other institutional obligations, and by
means of discourses of truth about sexuality. It is a proposal for a
non-universalizable ethics whose importance in the present context lies
in the possibility that it might provide a "practice of freedom" which
enhances the feeling of power in a way which other liberated lifestyles
do not. Foucault's problem is not that of formulating the moral norms
that accord with our present moral constitution but rather the
Nietzschean problem of suggesting ways in which we might become other
than what we are.

Notes

* An earlier version of this paper was published in French in Sociologie

et Societes, Vol. XXIV, no.1, April 1992. For their helpful comments on
earlier drafts, I am indebted to Moira Gatens, Barry Hindess, and the
anonymous readers for Sociologie et Societes.

1. Gilles Deleuze, Foucault, translated by Sean Hand, Minneapolis:

University of Minnesota Press, 1988, p.90.

2. Nancy Fraser, "Foucault on Modern Power: Empirical Insights and

Normative Confusions", in Unruly Practices, Minneapolis: University of
Minnesota Press, 1989, pp.32-33.

3. Jurgen Habermas, The Philosophical Discourse of Modernity, translated

by Fredrick G. Lawrence, Cambridge, Massachusetts: MIT Press, p.276.

4. Ian Hacking, "The Archaeology of Foucault", in David C. Hoy, ed.,

Foucault: A Critical Reader, Oxford and New York: Blackwell, p.40.
Similar criticisms of Foucault are made by a number of the contributors
to this volume, including Hoy, Walzer, Dreyfus and Rabinow and of course
Habermas.
5. Michel Foucault, "The Subject and Power", Afterword to Michel
Foucault: Beyond Structuralism and Hermeneutics, by Hubert L. Dreyfus
and Paul Rab icago: University of Chicago Press, 1983, pp.208-216.

6. Fraser, Unruly Practices, p.29; cited approvingly by Habermas, The

Philosophical Discourse of Modernity, p.284.

7. See Peter Dews, Logics of Disintegration, London: Verso, 1987, p.156,

where Foucault is represented as proposing "that subjects are entirely
constituted by the operation of power"; also Scott Lash, "Genealogy and
the Body: Foucault/Deleuze/Nietzsche", Theory, Culture and Society, 2:2,
1984; and Nancy Fraser, "Foucault's Body A Posthumanist Political
Rhetoric?", in Unruly Practices.

8. Foucault, The History of Sexuality Vol. I, translated by Robert

Hurley, London: Allen Lane, 1979, p.93.

9. Foucault, "The Subject and Power", p.221.

10. Foucault, "The Subject and Power", p.222.

11. I have argued for the necessity of distinguishing between "power to"
and "power over", in order to rescue Foucault's remarks on power from the
charge of incoherence, in Paul Patton, "Taylor and Foucault on Power and
Freedom", Political Studies vol.XXXVII, no.2, June 1989.

12. Foucault, The History of Sexuality Vol. I, p.95.

13. Foucault, "The Subject and Power", p.225.

14. Foucault, "The ethic of care for the self as a practice of freedom",
Philosophy and Social Criticism, Vol.12, no.2-3, Summer 1987, p.114.

15. Foucault, "The ethic of care for the self as a practice of freedom",
p.123.

16. C.B. Macpherson, Democratic Theory: Essays in Retrieval, Oxford:

Clarendon Press, 1973, essay III, "Problems of a Non-Market Theory of
Democracy Hobbes, Leviathan, edited by C.B. Macpherson, Harmondsworth:
Penguin Books, p.227.

18. Foucault, "The ethic of care for the self as a practice of freedom",
p.129.

19. Foucault, "The Subject and Power", p.210.

20. Macpherson, p.62.

21. Deleuze, Foucault, p.93.

22. Macpherson, p.56.

23. Macpherson, p.66.

24. Foucault, "What is Enlightenment?", p.43.

25. Foucault, "What is Enlightenment?", p.47.

26. Foucault, "Truth, Power, Self", Technologies of the Self, ed. Luther
H. Martin et al, The University Massachusetts Press, 1988, p.15.

27. Foucault, The Use of Pleasure, translated by Robert Hurley, New

York: Pantheon Books, 1985, p.80.

28. Foucault, "Preface to The History of Sexuality, Volume II", in P.

Rabinow ed., The Foucault Reader, pp.334-5.

29. Nietzsche, Beyond Good and Evil, translated by R.J. Hollingdale,

Harmondsworth: Penguin, paragraph 19.

30. Nietzsche, Daybreak, translated by R.J. Hollingdale, Cambridge,

Cambridge University Press, paragraph 23.

31. Mark Warren, Nietzsche and Political Thought, Cambridge, Mass.: MIT
Press, 1988, p.138.

32. Foucault, "The Subject and Power", p.211.

33. Foucault, "On the Genealogy of Ethics: An Overview of Work in

Progress", in Michel Foucault: Beyond Structuralism and Hermeneutics, by
Hubert L. Dreyfus and Paul Rabinow, Chicago: University of Chicago Press,
1983, pp.229-237.