Foucault and Ethical Universality

Christopher Cordner

University of Melbourne

Foucault’s resistance to a universalist ethics, especially in his later writings, is well-
known. Foucault thinks that ethical universalism presupposes a shared human essence,
and that this presupposition makes it a straitjacket, an attempt to force people to
conform to an externally imposed ‘pattern’. Foucault’s hostility may be warranted for
one – perhaps the usual – conception of ethical universality. But there are other
conceptions of ethical universality that are not vulnerable to Foucault’s criticism, and
that are ethically and culturally important. I set out one such conception, and show
why it matters. Paul Patton has argued that Foucault is best read as grounding his
analyses of power in a ‘conception of human being’ traceable to Nietzsche. I explain
why this does not amount to the ethical universalism that I sketch below.

Foucault was hostile to the idea of a universal ethics. He recognized that a
commitment to ‘universality’ is one possible ethical orientation among
others, and that it has at times been an influential orientation.

1

But he clearly

resisted any suggestion that ethics is inherently universal, and equally clearly
wanted no truck with such a commitment himself. He thought this
commitment was, and still is, tyrannical and crippling. As he put it in a
late interview: ‘The search for a form of morality acceptable to everybody in
the sense that everyone should submit to it strikes me as catastrophic’.

2

In

Foucault’s view the conviction of universalism generate the will to impose
uniformity on all.

At the cultural level, Foucault thinks this pairing of attitudes finds

expression in the normalizing operation of disciplinary power. Conceptions
of normality and sanity, for example, operate to exclude difference, and even
co-opt people into denying their own difference. They straitjacket people, but
they do so in part by suborning people into straitjacketing themselves. As
Barry Allen puts it:

The point of discipline is not to force people to do what you want, but to make them
into the kind of people you want; not to make people do what you want them to do, but
to make them want to do it, and to do it as you want them to, with the desired tools,
efficiency and order.

3

The normalizing forces of disciplinary power shape the conviction that there
is a shared human essence, and then operate to confine people in accordance
with it. But there is no such essence. The search for a universal ethic is
‘catastrophic’ just because it seeks to impose an illusory sameness on the

Inquiry, 47, 580–596

DOI 10.1080/00201740410004313

# 2004 Taylor & Francis

important reality of human difference. Foucault thinks that universal ethics is
always predicated on an already-given conception of the range of human
capacities, so that its application excludes all sorts of humanly important
possibilities, including many that have yet to appear. As Paul Patton puts it:
‘(Foucault) 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.’

4

This, very briefly, I take to be the

background of Foucault’s resistance to ethical universality.

Some of Foucault’s hostility to ethical universalism is warranted, but in my

view not all of it. In this paper I want to sort out what is warranted from what
is not. More specifically, I want to clarify what kind or form of ethical
universality, if any, it might still be important to recognize. I shall argue that
there is an important form of ethical universality, different from the
conception Foucault opposes, that any decent ethics must acknowledge. To
say that this conception of ethical universality is different from the conception
Foucault opposes leaves open whether the conception is compatible with
Foucault’s outlook – and beyond that, whether it might even be retrieved from
his thought. I shall return to these questions at the end.

Foucault assumes that ethical universality involves commitment to

universal norms of behaviour. I share much of his opposition to ethical
universalism so conceived, but there is a distinction within the domain of
behavioural-norm universalism that Foucault seems to overlook. Arguably,
a less objectionable form even of this kind of universalism can be spelled
out once this distinction is recognized. We can get at the distinction via
Foucault’s discussion of a shift in Stoic thinking from ‘an idea of aesthetics of
existence’ to a ‘universalist’ orientation. Foucault illustrates the former
orientation in the following way:

For example, you find in Isocrates a very interesting discourse, which is supposed to
be held with Nicocles, who was the ruler of Cyprus. There he explains why he has
always been faithful to his wife: ‘Because I am the king, and because as somebody
who commands others, who rules others, I have to be able to show that I am able to
rule myself.’ And you can see that this rule of faithfulness has nothing to do with the
universal and Stoic formulation: ‘I have to be faithful to my wife because I am a
human and rational being’. In the former case it is because I am the king!

5

It is true that Nicocles (as described by Foucault) thinks that something
specific about his situation imposes the requirement of faithfulness on him
and not everyone. But it does not follow that there is no universality, still less
that Nicocles is giving expression to an ‘aesthetics of existence’. Note the
words Foucault gives to Nicocles: ‘Because I am the king, and because as
somebody [my italics] who commands others … I have to show that I am able
to rule myself’. Nicocles’ conception of himself as ‘somebody’ who
commands others is what provides his rationale for faithfulness. But that is

Foucault and Ethical Universality

581

clearly a description that others could, and in fact do, satisfy. The reason on
which he acts he thus recognizes as applicable to any ‘somebody who
commands others’. That is to say, the reason is universal in form. Nicocles
acts on it because he recognizes that this universal reason applies to him as
‘somebody who commands others’.

6

Nicocles’ ethical orientation is clearly not shaped by a conception of

himself as ‘just anybody’. But it need not therefore be a matter of merely
‘personal’ or ‘aesthetic’ choice, without implications for anyone else. Those
are not the only possibilities. What is salient to Nicocles is that he is ‘a king’.

7

His thought seems to be that any king would be obligated as he is; and that is
a ‘universal’ thought, with clear implications for others – any others who are
kings. The universalism thus expressed is not predicated on what Bernauer
and Mahon call a ‘least [i.e., lowest] common denominator’ of human
dispositions, capacities and situations.

8

They use that phrase to describe the

behavioural-norm universalism Foucault rejects. But behavioural-norm
universalism can recognize that various differences in individual capacity
and situation make a difference to behavioural requirements (the ‘rules’ a
person considers binding). Such a situation-sensitive and capacity-sensitive
universalism does not impose a straitjacket on human beings regardless of
their differences of situation and capacity, since it can specify at least some
ethical requirements in terms of how one stands with respect to these
‘variables’. It is ‘a person of such and such a kind, in such and such a
situation’ – for example, as one who is a king – that such ethical requirements
bind. Ethical requirements can in fact be made ever more specific and
situation-sensitive, while remaining in this sense universal. What this shows
is that the generality (or otherwise) of ethical requirements is not the same as
their universality (or otherwise).

Does anybody really think that what is ethically required of us is always

independent of our specific circumstances and capacities? Foucault some-
times cites Kant as thinking this, but Kant does not hold that all the specific
ethical requirements are derivable from the ‘least common denominator’ of
capacities and situations. He distinguishes between perfect duties, which are
indeed like this, and imperfect duties, which are not.

9

Perfect duties include

prohibitions on murder and lying, for example, and on Kant’s view these
prohibitions are indeed required simply by the ‘least common denominator’
of our shared rational nature. In the case of our imperfect duties, by contrast –
for example, the duty to perfect one’s talents and the duty to help others
– Kant held that what one is to do in order to fulfil these duties varies
according to individual capacity, circumstances and preferences. In the
domain of imperfect duties, ‘universal norms of behaviour’ cannot be laid
down for all to follow.

Of course Kant’s ethics may still be too restrictive for Foucault. Foucault

might, for example, reject all universal norms of behaviour. But in any case,

582

Christopher Cordner

Foucault is less concerned with the claims of philosophers than with the
normalizing effects of disciplinary power. Here in practice is a domain of
‘universal norms of behaviour’ imposed on human beings in all sorts of social
and institutional contexts. ‘You may not do this or that because it is insane, or
abnormal or perverted or ….’. Disciplinary power in effect seeks to extend the
reach of ‘universal norms’ a very long way into our lives and activity. Even if
Kant formally speaking makes room for individual variation in ethical self-
realization, via the category of imperfect duty, that may still leave far too
much scope to the avaricious tentacles of disciplinary power. As Foucault
might perhaps have put it had he used Kant’s terms: ‘imperfect’ duty, by
means of which we give individual ethical shape to our being, covers a much
larger territory than Kant allowed.

So far I’ve been arguing that a capacity – and situation – modulated ethical

universalism perhaps need not fall to Foucault’s objections against
universalism. As a matter of fact, however, Foucault’s opposition to
universalist ethics extends even to modulated versions such as this one. He
seems to resist any suggestion that ethics is inherently universalist, even a
relatively concessive, situation-sensitive one. To put the point in terms of his
example of Nicocles: once being ‘a king’ can make a difference to what is
ethically required, then so indeed can being ‘this king’. The ethically relevant
specificity of one’s situation can reach all the way down to its uniqueness, and
then, it may seem, every vestige of rule-universality has disappeared.

I shall not dispute this claim here, since my interest is in whether, even if

we grant it, we have thereby forsworn all commitment to ethical universality.
Before pursuing that question, however, it is worth noting some thinkers from
a different philosophical background who broadly share Foucault’s views
about universal ethical ‘norms of behaviour’. Stuart Hampshire said that,
ethically speaking, no-one can avoid going ‘lop-sided to the grave’ – with
humanly rich ethical potentialities undeveloped.

10

He did not mean that this

showed a regrettable failure in each of us to develop all of our capacities as we
might have done if we had been less lazy or more imaginative, or perhaps
lucky enough to have been born into a richer or more tolerant culture. He
meant that there are indefinitely many ways in which human beings can
develop, and can come to live rich, vital and ethically good lives, and that
each of these ways will, at many points, involve excluding possibilities that
could be realized in the living of a different rich, vital and good life. In
reflecting upon George Orwell’s essay about Gandhi, Peter Winch makes a
similar point. He contrasts one aspect of Gandhi’s ‘asceticism’ – his ‘turning
away from types of human relationship that involve emotional entanglement
with other individuals’ – with Orwell’s belief in the need to accept the
dangers of such entanglement.

11

The difference between Gandhi and Orwell

is not merely contingent – as if, given world enough and time, each might
well have added the other’s practices and outlooks to his own. Rather, as

Foucault and Ethical Universality

583

Winch puts it, ‘(t)he two moral conceptions are not merely different and
mutually exclusive, but … they are opposed to each other: the one involves a
rejection of the other.’

12

But it does not follow from this that Gandhi and

Orwell must each think that everybody else ought to do and be like he does
and is. While each recognizes that his orientation excludes the other’s, that
does not require each to think of the other as ‘mistaken’ or ‘wrong’. Either of
them might think that, but Winch’s point is that this would be a further
thought, not something already contained as a matter of pure logic in his own
ethical orientation. Winch summarizes what he takes to be the lesson of his
example in these terms:

There is no reason to suppose that two such men would ever reach agreement on what
divides them, whatever arguments were adduced … But equally there is no reason to
think that failure to agree must be a sign of deficient understanding or of bad faith in
either one of the disputants.

13

What is thus recognized by Hampshire and Winch I would not (nor would

they) describe in Foucault’s terms as ‘the search for styles of existence as
different from each other as possible’. That would make it sound too much a
matter of ‘mere’ choice – as if Gandhi, being the man he was, might instead
just as well have gone for what Orwell valued – and also too much as if the
motive is to be as different as one can from anyone else. Still, the basic
emphasis in Hampshire and Winch is similar to Foucault.

14

Does this emphasis negate all ethical universality? I do not believe so. A

certain conception of ethical universality is still available to us.

15

I want to

show what it involves – and beyond that, why it matters that we acknowledge
it. To show this we need to shift our attention a little. It is well-known that
for Foucault the domain of morals involves much more than whatever
‘code’ of proscriptions and requirements – norms of behaviour – a person
might adhere to at any cultural moment. It also involves the four aspects
of what Foucault called ‘rapport a` soi’, the ‘relation to oneself’ in one’s
refrainings and doings.

16

This adds to ‘what it is right to do’ a concern for

what Charles Taylor calls ‘what it is good to be’.

17

For Foucault ethics

involves flexibility in how human beings relate to themselves in whatever
norms of behaviour they may acknowledge. (This is also the territory of
ascesis – the ‘work on oneself’ whose ethical importance Foucault
highlights.) Rapport a` soi is the complex of ways in which, through our
culturally situated freedom, norms of behaviour get woven into living a
distinctive ethical life.

This is an important enrichment of what belongs to ethics. But Foucault’s

resulting ‘picture’ of ethics still, I believe, misses something very important.
What that is can be described in different ways. One of them highlights the
ethical significance of how one thinks of those others vis-a`-vis those in whom

584

Christopher Cordner

one may recognize various duties. That does not seem to belong either to the
‘contents’ of the ethical code – the particular ensemble of prohibitions and
requirements that a person recognizes – or to any of the aspects of rapport a`
soi. Reflection upon this omission will reveal an important conception of
ethical universality missing from Foucault’s work.

Someone whose own child is murdered will usually be personally

devastated. While she may be moved, sometimes even deeply upset, by
news of distant children being murdered, she will likely not be utterly
devastated in the same way. She would scarce be human if this were not so.
This difference in response does not indicate narcissism, or callousness, or
lack of ethical concern. But now suppose that a mother who suffers huge grief
and devastation at the murder of her own child went on to say, about children
similarly murdered in another country: ‘Of course that was not terrible in
the way the murder of my child was. My child was appallingly violated,
suffered terrible cruelty and injustice. It is not the same for those children,
or for their parents.’ That would show a fundamental failure of ethical
understanding. The parent who is not personally grief-stricken at the death of
children distant from her need not, and usually will not, show this failure
of understanding. She can appreciate that any child who is thus murdered
suffers as terrible a violation and outrage as that suffered by her own child.
And precisely that appreciation shows her assumption of an ethical
universalism. (This appreciation could also be described as an expression
of spontaneous respect for all those others more or less distant from her.)
Were she to lack this appreciation in the way I described, her attitudes and
outlook could not but strike us as at best shallow and condescending, or worse
deeply racist and dehumanizing. But this implies that a certain kind of
universalism of outlook is a condition of any ethical orientation that is to be
taken seriously.

While this is in fact a very simple point, it is easily overlooked in

discussions of ethics. Part of the reason is that ethical universalism is usually
conceived in the way discussed earlier, as a matter of exceptionless norms of
behaviour – of what actions are to be done, or forbidden, or permitted. (As we
saw, this is how Foucault conceives of ethical universality.) But nothing about
what norms of behaviour this mother should adopt vis-a`-vis those other
murdered children follows from anything I have said. That she shows a
universal conviction that all children thus murdered have suffered the same
violation and injustice is one thing; what is to be done in relation to any of
them is quite another.

18

Recognition of this important kind of ethical

universalism carries, that is to say, no commitment to ‘universal norms of
behaviour’. This form of universalism is engaged, instead, at the level of how
others are acknowledged.

19

It is perfectly compatible with Foucault’s – and

Winch’s and Hampshire’s – rejection of what I called behavioural-norm
universalism.

Foucault and Ethical Universality

585

Of course, as I have already suggested, the universalism that often thus

shows in the ways people spontaneously respond, does not always show there.
Raimond Gaita writes of a woman he knew who had recently lost her child.

(She) was watching a television documentary on the Vietnam War which showed the
grief of Vietnamese women whose children were killed in bombing raids. At first she
responded as though she and the Vietnamese women shared a common affliction.
Within minutes, however, she drew back and said, ‘But it is different for them. They
can simply have more’.

20

Gaita describes this as ‘a racist remark’, and he continues:

(The woman) did not mean that whereas she was sterile they were not. Nor did she
mean that as a matter of fact Vietnamese tended to have many children. Hers was not
an anthropological observation. She meant that they could replace their dead children
more or less as we replace dead pets …. (W)e would misunderstand (her) sense of the
difference between herself and Vietnamese women if we thought that she did not
believe they suffered. After all … she sees their "frantic grief" and hears their
anguished wailing. Her sense of difference centres on how she conceives what
suffering may mean to the Vietnamese. She fails to see that it could mean anything
deep and she takes this attitude to their entire inner life, to all their hopes, fears and
joys whatever these may centre on.

21

This woman clearly does not think of those Vietnamese in the way she

thinks of ‘her own kind’. She cannot recognize them as able to be violated in
the way she and her kind are capable of being violated. Whatever were to be
done to them, she could not recognize as involving the same injustice,
degradation, humiliation, outrage that could be visited on her and her own
kind, including her children. That is simply the other side of her inability to
see the same kind of depth in their sufferings as she can find in the sufferings
of her own kind.

In similar vein James Isdell, the Protector of Aborigines in Western

Australia at the beginning of the last century, said in defending the
government policy of the day: ‘I would not hesitate for one moment to
separate any half-caste from its Aboriginal mother, no matter how frantic her
momentary grief might be at the time. They soon forget their offspring’.

22

Isdell recognizes that these aboriginal mothers feel ‘grief’ when their children
are taken from them. He could even acknowledge that by all behavioural
standards they show (for a time at least) as much grief as women belonging to
‘his own kind’. What he denies is that in their grief they suffer as would
women of his own kind if their children were taken. Or to put it another way,
he cannot find the same meaning in their grief that he finds in the grief of
women of his kind. It is tempting to suppose that this is because he believes
they will ‘soon forget their offspring’. But it is really the other way around.
His conviction that they soon forget their offspring does not explain his denial
that their suffering means what the suffering of women of his own kind would
mean. Rather, he has that conviction only because he cannot find any depth in

586

Christopher Cordner

their grief. One natural description of Isdell’s limited response to those
aboriginal women is that they are not fully present to him as human beings in
his ‘thinking’ of them. His racist condescension towards them precludes their
being so. That is to say, his failure (or refusal) to let them be fully present to
him is the form his racist condescension takes.

It is important to distinguish how Isdell thinks about ‘Aboriginal women’

from how he might think about this or that particular member of his own kind.
He might, for example, think of his next door neighbour as a singularly stupid
woman, as shallow and thoughtless, and as quite unable to understand what
being a ‘good mother’ might involve. Isdell might even think that she treats
her children as if she could just ‘have more’. That might sound akin to how he
thinks of aboriginal women. But it is very different. He thinks of his
neighbour as having specific failings which in fact she might not have.
Perhaps, indeed, she still might overcome them. It is a contingent fact that she
is stupid and shallow; given grace or luck, she might have been otherwise. To
put this point slightly differently: he thinks of his neighbour as occupying a
conceptual space within which certain categories and their contraries find
application – the categories of, for instance, (a certain kind of ) moral
intelligence and moral stupidity, (a certain kind of ) depth and shallowness.
Where his neighbour actually stands in that conceptual space is a matter of
contingent fact. She is stupid and shallow, but she might have been otherwise.
That is the sense he implicitly makes of her failings.

But that is not how he thinks of aboriginal women. In not having a certain

kind of intelligence and depth they are not, to him, stupid and shallow at all,
any more than his dog is. As with the dog, he simply finds those categories to
have no intelligible application to these women. For him they simply do not
belong to a kind who could be judged under those categories. He does not
think that way about his neighbour. That is to say, the difference between
those of his own kind who are wise and deep, and those of his kind who are
not – his stupid neighbour, for instance – is one kind of difference. The
difference between his neighbour and those aboriginal women he takes to be a
very different kind of difference. In her stupidity and shallowness
his neighbour still inhabits the same space of conceptual possibilities as
do those of his own kind who are wise and deep, while the aboriginal women
do not.

The significance of this difference can be registered in various ways. Isdell

acknowledges that his neighbour is still fully his fellow human being in a way
he implicitly denies the aboriginal women are. What counts for this
acknowledgement is not where someone is judged to stand in a certain space
of conceptual possibilities, but what the space of conceptual possibilities is
within which judgments about her fall. So Isdell’s very judgments of his
neighbour as stupid, shallow and thoughtless locate her within a conceptual
space that also contains the possibility of a certain kind of moral intelligence

Foucault and Ethical Universality

587

and depth. That he sees another as belonging within that space is crucial to his
recognizing her as fully human – or, we could also say, as his moral equal.
Those aboriginal women he simply does not see as belonging within that
space. For Isdell, and also for the woman watching the Vietnamese on
television, some kinds of people simply do not count as fully fellow human
beings. Their ethical sensibility thus involves a radical and ethically troubling
failure of universality. This conception of universality, I suggest, is a
condition of any ethical orientation we can respect.

Nothing in Foucault’s four aspects of rapport a` soi, however, excludes

orientations lacking such universality. Rapport a` soi can involve ways of
‘thinking’ of others that are shallow, condescending and demeaning. Foucault
did of course try to argue that rapport a` soi can take forms that do entail care
for others.

23

But even if his arguments on that score are granted force, they

still do not reach far enough. Suppose rapport a` soi does sometimes involve
care for some others of the kind Foucault describes. Still it does not exclude
(for example) the kind of racist condescension I described. If that is so, then
ethics reaches not just beyond ‘norms of behaviour’ but also must reach
beyond care of the self even when that entails some degree and kind of care
for others.

24

So far as Foucault’s conceptual framework for ethics does not

recognize this, it seems insensitive to a kind of universal acknowledgment of
human beings that is a precondition of serious ethical ‘treatment’ of them or
engagement with them. This universal dimension shows itself neither in the
determination of what to do, nor in the complex ‘work on oneself’ that
Foucault thinks ethically central, but rather in fundamental ways in which
others are ‘thought’.

What is at issue in this conception of universality is realizing a kind of

equality with others – all others – which depends on seeing them under the
aegis of certain fundamental possibilities of human meaning. These include
the possibility of depth and shallowness of thought and feeling, the possibility
of various kinds of intelligence and stupidity, of vulnerability to various kinds
of degradation, humiliation and outrage, and the possibility of answering as
well as making various kinds of claims upon others. These are all elements or
aspects of our seeing others as capable of the same kinds and depths of human
fulfilment and suffering, happiness and misery, joy and despair, as we
ourselves are – whatever their cultural differences from us. This sense of
equality does not seek to impose a single set of norms of behaviour on people,
or presuppose that human beings have a timelessly fixed range of capacities.
And it is compatible with recognizing an indefinite variety of activities and
cultural forms and patterns of behaviour, including many that have not yet
appeared. Failure to acknowledge this sense of universality is not confined to
‘extreme’ racism of the sort I described. It has been an enduring feature of
colonialist condescension, for example, for a very long time.

To many, however, what I have been saying will seem exactly the wrong

588

Christopher Cordner

response to such baleful attitudes. A common view is that these are the result
of ethical universalism, and that what is needed instead is the ‘recognition of
difference’. Interpreters of Foucault have often supposed that once a universal
moral code is rejected, all that is left is the importance of recognizing
difference. Few commentators seem willing to press Foucault’s thought
beyond an anti-foundationalism that reaches no further than the importance of
negative freedom – the individual’s ability to work on herself (or perhaps at
the cultural level the ability of members of a society to work on themselves)
without interference.

25

I am saying, in effect, that difference can be taken

seriously only when it is recognized as expressive of a certain kind of human
significance, and this depends on its being seen under the aegis of those
possibilities of human meaning I mentioned. That is to say, a certain universal
background – although emphatically not one that seeks to straitjacket people
in universal norms of behaviour – is a condition of an ethically robust and
respectful acknowledgment of difference.

But am I right to say that Foucault’s work overlooks this kind of conception

of ethical universality? Or, even if he does not explicitly recognize it, might
his work perhaps contain resources out of which something like this
conception could be constructed? One who arguably thinks so is Paul Patton
in his seminal essay, ‘Foucault’s Subject of Power’.

26

Patton argues that

Foucault’s work needs, and in fact ‘presuppose(s) a certain conception of
human being’.

27

A conception of human being is indeed a form of

universalism about human nature. Patton tries to elicit from Foucault’s work
a form of universality that is compatible with Foucault’s rejection of
behavioural-norm universalism. Does the ‘conception of human being’ that
Patton claims to find in Foucault reflect (something like) the ethical
universalism I have described? Has Patton in effect already identified in
Foucault’s thought what I have criticized Foucault for missing? Let me say
why I think the answer to these questions is, ‘no.’

The conception of human being that Patton finds in Foucault certainly is

incompatible with a set of prescriptions for human behaviour that would
adequately reflect our ‘nature’. That is because the ‘nature’ in question is an
autonomy whose exercise issues in our ‘becoming other than what we are’.
Through such ‘becoming’, we will be creatures who escape being fixed by any
determinate norms of behaviour formulated to answer to our ‘real nature’. If the
norms are then reformulated to reflect what we have now become, the problem
simply recurs. With each exercise of our autonomy we outstrip the norms
fashioned to answer to what we were. Our ‘universal’ nature, then, is our
sharing of something – autonomy – that makes it impossible to reflect what
we are in universal norms of behaviour, and ‘catastrophic’ to try to do this.

In that respect, the universalism that Patton finds in Foucault does indeed

echo the universalism I sketched. But there are significant differences too.
Some of them owe to the specific roles Patton sees this ‘conception of human

Foucault and Ethical Universality

589

being’ playing in Foucault’s thought. According to Patton, this conception
contains the resources for responding to two different concerns among
Foucault’s critics. The first concern is how, given the pervasive way in
which power is held in Foucault’s earlier work to constitute the subject,
there is so much as the possibility of resistance to domination. The second
concern pertains to the lack of any clearly articulated framework in which to
place and justify (to use Jeremy Moss’s phrase) ‘the obvious normative
engagement of Foucault’s texts’.

28

Patton’s straightforward response to the

first concern is that, ‘in their attempt to exercise their capacity for autonomous
action, those subject to relations of domination will inevitably be led to
oppose them …. (S)uch resistance follows from the nature of particular
human beings. It is an effect of human freedom’.

29

Patton develops his more

complex response to the second concern initially by reference to the work of
C. B. Macpherson. The conception of autonomy he employs in connection
with both concerns has its roots in a broadly Nietzschean understanding of
power.

I am not here interested in exploring those specific concerns and whether

the conception of human being Patton identifies does indeed make possible a
good response to them. My present interest is in comparing the universalism
implicit in that conception with the kind of universalism that I earlier
sketched and argued was necessary to any ethical orientation worthy of
respect. Briefly: I think the materials out of which that conception of human
being is woven are intrinsically ill-suited to reflecting the universalist
conception I sketched. James Isdell could readily recognize, after all, that
aborigines resist restrictions of their freedom and that they seek to ‘maximize
their feeling of power’, without this in the least disturbing his deep racist
condescension towards them; and this need show no inconsistency or failure
of reason on his part. The same is true of the woman watching the Vietnamese
on television. What, after all, is so special about ‘the feeling of power’, at
least in its Nietzschean version, since even (non-human) animals have it? For
Nietzsche, all life is will-to-power. Recognizing in other human beings the
fact of ‘the feeling of power’ and their attempt to maximize it, therefore
seems consistent with thinking of some of them as in some ways more ‘like
animals’ than like us, as Isdell and that woman do. At least when autonomy
is tied to ‘the feeling of power’ in the way Patton suggests, the simple fact is
that no elaboration on the theme of autonomy, or on what is involved
in recognizing another as possessing it, need lead to the kind of human
acknowledgment of another that is involved in the universalist orientation I
described earlier.

But have I perhaps overlooked the steps Patton interposes between ‘the

feeling of power’ and the experience of autonomy? He argues that in our
(human) exercise of autonomy we are embedded in ‘enduring social and

590

Christopher Cordner

institutional relations … [and] … also … frameworks of moral interpre-
tation’.

30

He tells us that,

… the experience of autonomy depends upon the larger networks of practice and
social relations within which individuals act, [and] also upon the interpretative
frameworks in terms of which they judge the success or failure of their acts …

31

Patton’s point is that ‘the feeling of power’ becomes the experience of
autonomy only in a socio-cultural setting, and such settings include ‘the
interpretative frameworks’ in which human beings judge their own actions.
This makes the human feeling of power dependent on a capacity for
reflection in a way (other) animals’ feelings of power are not. For this reason,
anyone who thinks of (some) other human beings as merely ‘like animals’ in
respect of their ‘feelings of power’ is missing what is distinctive about the
human feeling of power, and thereby missing something crucial to human
autonomy.

Let all that be granted. My criticism two paragraphs ago did not depend on

denying it. Isdell (and also the woman watching the Vietnamese on
television) could perfectly well recognize that the aborigines he speaks of
make use of such ‘interpretative frameworks’ and that they experience their
‘feeling of power’ against the background of ‘larger networks of practice and
social relations’ of a kind in which (other) animals are not embedded. But
still none of that need disturb his racist condescension (even if it does mean
he cannot see them simply as ‘like animals’). The point is that he simply
cannot see the same depth in the ways in which they inhabit and give
expression to those interpretative frameworks, and in the networks of practice
and social relations in which they are woven, as he finds in the interpretative
frameworks and networks of practice of others ‘of his own kind’. As Raimond
Gaita puts it, in relation to how a certain kind of racist thinks of ‘blacks and
Asians’:

‘They’ can do and feel almost everything we can except not as we do, not as deeply as
we do. We grieve, but they ‘grieve’, we are joyful, they are ‘joyful’, we love and they
‘love’, we feel remorse, they feel ‘remorse’, and so on. And through this denigration
of their inner lives runs a denigration of how it is possible for them to think and reflect
… The inner lives of blacks and Asians are placed in inverted commas by white
racists because they cannot believe there could be any depth in them.

32

This racist outlook can still ascribe to those it denigrates everything Patton
says about human autonomy. Those people are recognized as woven into
‘larger networks of practice and social relations’ and as embedded in
‘interpretative frameworks’. It is just that in their case these things are not
acknowledged by the racist as having the depth they have in the lives of ‘us’
and ‘our kind’.

But that is to say, very precisely, that the ‘feeling of freedom’ and the

concept of autonomy Patton distils from it do not themselves afford a

Foucault and Ethical Universality

591

conception of human being that is expressive of the sense of universality I
sketched earlier. As I put it before, that sense involves seeing another as
occupying a certain space of possibilities within which alone he or she can be
acknowledged as fully one’s fellow human being. To see another this way
involves seeing her, for example, as able to be humiliated in certain sorts of
ways, as one whose life is capable of certain sorts of meaning and who is able
to understand her own life as having or lacking such meaning, as someone one
could seriously wrong and who could occasion one’s serious remorse. To see
another in this complex of ways is to see her as one whose life is capable of a
certain kind of depth – precisely the kind of depth that Isdell denies to the
lives of the aborigines he speaks of, and that Gaita’s racist likewise denies to
‘them’. It may be that this presupposes ascribing to her the kind of autonomy
Patton speaks of, but a sense of others’ lives as capable of such depth takes us
well beyond that ascription. This is the universality, or commonality, which
any ethical orientation must acknowledge on pain of otherwise being
seriously deficient, and Patton’s terms do not take us to it.

What then guarantees that one ‘sees’ another’s life as capable of such

depth? Nothing will guarantee that, though a good deal more can be said about
what goes into realizing the kind of universality – or human commonality
with others – whose importance I have tried to highlight. Here is one thought.
The James Isdell who said what I quoted him as saying could not possibly
have heard aboriginal music or singing as having the poignancy, richness,
resonance, power that he may have found in the singing of ‘his own kind’. He
could not have heard or felt in their stories and myths, or in their ritual
celebrations of birth or marriage or death, the kind of significance and moving
gravity he heard in the stories myths and rituals of ‘his own people’. Those
‘could nots’ are logical in force. For him to come to hear and feel those things
differently – as having the same kind of depth and power he finds in the
stories myth and rituals of his own people – would be for him to come
to realize the human commonality of those people with him and his kind;
and his radical condescension could not survive that realization. This is
the ‘level’ so to speak, at which our human universality is realized, and must
be realized if we are not to engage with others in a way that risks radically
demeaning them.

This conception of universality has, as I emphasized earlier, nothing to do

with enjoining universal ‘norms of behaviour’. In that respect it is indeed akin
to the ‘conception of human being’ that Patton claims to identify in Foucault.
And beyond that, it no doubt has some link with what Patton means by
autonomy. But it cannot be understood, or made sense of, through that
concept alone, even when the concept is given the ‘rich’ interpretation Patton
gives it. The problem I find in what Patton says could be summarized this
way: Kant thought rationality distinguished human beings from other
creatures, and he in turn identified our possession of rationality with freedom.

592

Christopher Cordner

Patton sees Foucault as re-fashioning something like this Kantian move,
but in broadly Nietzschean terms, within the domain of history and
contingency: ‘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)’.

33

But this resolution does not go far enough. It overcomes one Kantian error,

while replicating another. There is no single ‘condition of moral action’,
transcendental or otherwise; and the reliance on ‘autonomy’ as such a
condition in the end still distorts the character of our shared humanity with
others. But the point is not quite that there are instead ‘multiple’ conditions of
moral action. It is that the ‘ground’ of moral action is not the recognition of
either one or any number of distinctive empirical facts or features of human
beings. Draw up any catalogue you like of the great variety of such facts and
features that different philosophers have highlighted: linguistic capacity, the
ability to reason, or to generalise (Locke: ‘Brutes abstract not’), the ability to
reflect, or to deliberate, or perhaps the capacity to laugh or to weep, or to tell
stories; and so on. Recognition of any or even all of these capacities will never
force a person to acknowledge that depth of humanity in the other to which I
have pointed. That is so whether those features are ‘transcendental’ or
‘empirical’. Conversely, only when others are humanly significant for us in
the ways I have described does their autonomy matter to us. Kant was
mistaken to think that respect for their autonomy was the ground of our moral
acknowledgment of others. It is closer to the other way around: our ‘moral
acknowledgment’ of others – in the forms I sketched – is the ground of
valuing their autonomy. Such moral acknowledgment involves the realizing
of our common humanity with others.

34

This is the fundamental level at

which the universality of ethical understanding is manifest; and it does not at
all confine us in the straitjacket of universal behavioural norms.

We need ways of thinking our common humanity that not only escape

philosophical distortion but also avoid abetting the forces of conformity and
oppression. Sometimes, of course, what is needed more than a reminder of
that commonness is an insistence on difference and individuality. John Stuart
Mill rightly thought his was such a time;

35

so did Foucault. But even such

insistence must presuppose a sense of common humanity if it is not to be
vulnerable to the kind of shallow, potentially brutal disrespect I spoke of. In
this essay I have returned to that presupposition because I think it is what now
– in our present cultural circumstances – needs revivifying, bringing to mind
once again. If Foucault thought that any such presupposition involved belief
in a rationalistic structure that could generate a ‘universal code’, he was (as
I argued earlier) mistaken. Patton may be right that Foucault did not think
this, and that he invoked a different kind of ‘conception of human being’.
But, suggestive though it is, this conception still seems to have serious

Foucault and Ethical Universality

593

limitations.

36

I have tried to give sense to a more robust conception that still

acknowledges the force of Foucault’s resistance to universal norms of
behaviour.

37

N O T E S

1 Foucault identifies Kantian and Stoic thought as two important examples of ethical

universalism. See ‘On the genealogy of ethics’, in P. Rabinow (ed.), The Foucault Reader
(Penguin Books, 1984), pp. 353–54.

2 ‘Final interview’, Raritan Vol. 4 (1985), p. 12.
3 B. Allen, ‘Foucault and modern political philosophy’ in J. Moss ed., The Later Foucault

(London: Sage Publications, 1998), p. 174.

4 P. Patton, ‘Foucault’s subject of power’ in Moss (ed.), op. cit., p. 65.
5 ‘On the genealogy of ethics’, op. cit., p. 354.
6 Michael Hardimon (‘Role obligations’, Journal of Philosophy XCI, 1994, pp. 333–63) and

John Simmons (‘External justification and institutional roles’, Journal of Philosophy XCIII,
1996, pp. 28–36) discuss the source of the normative force of ‘institutional obligations’. The
line I have taken on Foucault’s example may seem to commit me to John Simmons’ claim
that, ‘(I)nstitutional obligations acquire force only by being required by external moral rules’
(p. 30). But whether I am so committed depends on how external, as it were, such rules can
aspire to be without losing their applicability to various contexts that are supposed to fall
under them. I do not take a stand on that question here.

7 Foucault actually says ‘the king’, but the point he goes on to make in fact turns only on

Nicocles’ sense of himself as a king – a ‘somebody’ who rules others – and not on his being
uniquely this king.

8 J. Bernauer & M. Mahon, ‘The ethics of Michel Foucault’ in G. Gutting ed., The Cambridge

Companion to Foucault (Cambridge: Cambridge University Press, 1994), p. 153.

9 I. Kant, The Metaphysics of Morals trans. M. Gregor (Cambridge: Cambridge University

Press, 1996) p. 153 (Ak. 6:390).

10 S. Hampshire, Morality and Conflict (Oxford: Blackwell, 1983), p. 158.
11 P. Winch, ‘Ethical relativism’ in Winch, Trying to Make Sense, p. 187.
12 Ibid.
13 Op. cit., p. 188.
14 Hampshire and Winch do not infer, as Foucault sometimes does, that if there are no universal

‘norms of behaviour’ then the aim of ethics is ‘to create [oneself] as a work of art’. (Foucault,
‘On the genealogy of ethics’, p. 351. See also ‘What is Enlightenment?’ in Rabinow (ed.),
op. cit., pp. 40–42.) In my view this is a mark in their favour. Their ways of ‘going on’ from
Foucault’s insight about ethical norms of behaviour sidestep Foucault’s own occasional
‘aestheticizing’ of ethics. To put that point another way: ethics does not have to become
‘aesthetics of the self’ just because it is not tied to universal norms of behaviour.

15 Habermas is another who would acknowledge at least much of what Winch and Hampshire

say, while also resisting the conclusion that all ethical universality therefore dissolves. The
legitimate differences between outlooks that Winch and Hampshire emphasise, Habermas
would locate within the domain of ethical judgment, while moral judgment remains
universalist. (See, for example, J. Habermas, Justification and Application trans. C. Cronin,
Cambridge, MA: MIT Press, 1993; and J. Habermas, Between Facts and Norms trans. W.
Rehg, Cambridge, MA: MIT Press, 1996.) Could the conception of universality – and of its
relation to legitimate individual ethical difference – that is developed below then be mapped
onto Habermas’s contrast between morality and ethics? I make two observations relevant to
that question. First, Habermas’s universalist emphasis remains on norms of action whereas
mine is on ways in which others are acknowledged or ‘thought’. Only, I believe, if
universality is realised at this level of ‘experiencing’ others, will the Habermasian
commitment to taking into account the interests of all affected when framing norms of action
be able to express a deep and abiding sense of our common humanity with others. (To put
that point slightly differently: my emphasis is more phenomenological than Habermas’s, and

594

Christopher Cordner

in my view is therefore philosophically and ethically more fundamental.) But secondly,
regardless of whether that claim is found compelling, unresolved questions remain about the
character and basis of Habermas’s contrast between morality and ethics. (See, for example,
T. McCarthy, ‘Practical discourse’, in his Ideals and Illusions Cambridge, MA: MIT Press,
1991 pp. 181–99; and J. Heath, Communicative Action and Rational Choice Cambridge,
MA: MIT Press 2001, pp. 227–42.)

16 See M. Foucault, ‘On the genealogy of ethics’, in Rabinow (ed.), op. cit., pp. 340–72;

M. Foucault, The History of Sexuality, Vol. 2 The Use of Pleasure trans. R. Hurley (New
York: Vintage Books, 1990), pp. 25–32. A useful summary of Foucault’s account of rapport
a` soi is to be found in Arnold Davidson, ‘Archaeology, genealogy, ethics’, in D. Hoy (ed.),
Foucault: A Critical Reader (Oxford: Blackwell, 1986), pp. 221–34.

17 C. Taylor, ‘Iris Murdoch and moral philosophy’, in M. Antonaccio & W. Schweiker (eds.)

Iris Murdoch and the Search for Human Goodness (Chicago: University of Chicago Press,
1996), pp. 1–3.

18 Of course there is a serious question here about whether we (in rich western countries) in fact

do enough – in the form of foreign aid and in various other ways – for those who suffer
elsewhere. The answer is that too often we do too little.

19 It is worth noting that this is the level at which ethical universality is most fundamentally

engaged in Kant’s moral philosophy. Acknowledging every other as an end in himself is for
Kant the basic moral orientation (more precisely, this disposition answers to one of the
several formulations of the categorical imperative. See I. Kant, Groundwork of the
Metaphysic of Morals, ed. & trans. H. J. Paton (Harper Torchbooks, 1964.) Kant goes on to
specify various universal maxims of duty, but these are all derivative from that basic
orientation.)

20 R. Gaita, A Common Humanity (Melbourne: Text Publishing, 1999), p. 57.
21 Ibid.
22 Quoted in Gaita, ibid.
23 See especially ‘The ethic of care for the self as a practice of freedom’, in J. Bernauer

and D. Rasmussen (eds.), The Final Foucault (Cambridge, Mass.: MIT Press, 1988), pp. 7–9.

24 For further criticism of Foucault along these lines see my ‘Ethics, the Other and Rapport a`

Soi’ (forthcoming); and B. Smart, ‘Foucault, Levinas and the Subject of Responsibility’ in
Moss (ed.), op. cit., pp. 78–92.

25 See for example J. Rajchman, ‘Ethics after Foucault’, Social Text Vol. 13/14 (1986),

p. 186; and D. Palmer, ‘On refusing who we are’, Philosophy Today, Winter 1998,
pp. 402–10.

26 P. Patton, ‘Foucault’s Subject of Power’ in Moss (ed.), op. cit., pp. 64–77.
27 Op. cit., p. 64.
28 ‘Introduction: The Later Foucault’, in Moss (ed.), op. cit., pp. 1–17.
29 Patton op. cit., p. 73.
30 Ibid.
31 Op. cit., p. 75.
32 Gaita, op. cit., p. 63.
33 Op. cit., p. 73.
34 It is worth noting that the conditions of application of Rorty’s concept of solidarity fall well

short of what I have argued to be involved in such realizing.

35 See J. Mill On Liberty (Harmondsworth: Penguin Classics, 1985), especially Ch. 3: ‘Of

individuality, as one of the elements of well-being’.

36 I express this critical comment cautiously, because I acknowledge that I have not shown

definitively that Foucault’s work is incompatible with recognition of the kind of ethical
universality whose character and importance I have tried to clarify. I have instead argued
that on one powerful and widely shared interpretation – Patton’s – Foucault’s work lacks the
resources to generate such a conception. That finding is logically consistent with the
following two claims. First, that there is some other cogent reading of Foucault’s work on
which it does contain such resources; and secondly, that even if that is not so, still what is
most fundamental in Foucault’s thought is not strictly incompatible with the kind of

Foucault and Ethical Universality

595

universality I have explicated. My argument at least throws out a challenge to those who
would seek to make the first claim. To those who would assert the second, what I say can be
taken as urging not a correction of his views but a significant addition to them.

37 I wish to thank Jeremy Moss for helpful discussion of the issues explored in this paper.

Received 16 June 2004

Christopher Cordner, Department of Philosophy, University of Melbourne, Old Quad,
Building 150, Victoria 3010, Australia. E-mail: [e-mail]

596

Christopher Cordner