Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

date: 23 August 2019

Feminism in Philosophy

Abstract and Keywords

This article explores how feminism might illuminate philosophy, and indeed vice versa. The aim is not so much to survey the immense continent of feminist philosophical research, as to display, and occasionally instantiate, some small parts of it. In thinking about how feminism has contributed to philosophy, the article considers it worth looking at two rather general ideas: the idea of dualism, and the idea of androcentrism. In thinking about how philosophy has contributed to feminism, it is also worth looking at one rather specific idea: the idea of treating someone as an object.

Keywords: feminism, feminist philosophy, dualism, androcentrism, psychological traits, psychoanalytic theory


It has been said (at least once) that philosophy leaves everything as it is. Feminists, on the other hand, don't leave everything as it is; we are always wanting things to be different, and better. So what exactly have feminists to do with philosophy? The idea that philosophy leaves everything as it is conveys a picture of philosophy as somehow neutral, cut off from the action, unreceptive to change from, and unproductive of change to, the world beyond its walls. If philosophy is inert, feminists might as well leave it alone. But we don't. Why not?

In thinking about why not, it's worth noting that the slogan could suggest something else: not that philosophy is inert (as Wittgenstein may have meant; 1958, §124), but that it's conservative. Philosophy is not unreceptive to the world outside; everything does not, after all, leave philosophy as it is. Philosophy is open to influence, bears the marks of its makers, shows signs of the soil from which it has sprung. Philosophy is made in a social world, and a world that excludes women might leave signs of that exclusion in philosophy, and make philosophy an agent of that exclusion. Philosophy might act as an obstinate weight, a resistance to change, ‘leaving everything as it is’ as a dragging presence in the social world, not a walledoff absence from it. Then it would not be irrelevant to feminism. If philosophy (p. 232) were to manifest a political illness, and help perpetuate it, then feminist diagnosis and remedy might be called for—if remedy is possible. Pessimists might view philosophy's prognosis as bleak enough to rule out remedy, but such pessimism, I'll be assuming, is unwarranted.

I shall be asking how feminism might illuminate philosophy, and indeed vice versa. My aim is not so much to survey the immense continent of feminist philosophical research, as to display, and occasionally instantiate, some small parts of it. In thinking about how feminism has contributed to philosophy, it will be worth looking at two rather general ideas: the idea of dualism, and the idea of androcentrism. In thinking about how philosophy has contributed to feminism, it will be worth looking at one rather specific idea: the idea of treating someone as an object.

These limitations seem narrow enough, I expect, and in practice they will be even narrower than they antecedently seem. In considering dualism, I shall focus on one argument by Susan James, in the metaphysics of personal identity and survival. In considering anthropocentrism, I shall focus on one argument by Elizabeth Lloyd, in the philosophy of science, prompted by a debate in evolutionary biology about the origin of the female orgasm. In considering the idea of ‘treating someone as an object’, I shall focus on one elucidation of it by Martha Nussbaum, according to which someone can be ‘treated as an object’ by (among other things) having their autonomy denied. In a constructive spirit, I try to take her account further, looking particularly at that issue of autonomy denial. If the goal of the earlier sections is chiefly to display what I take to be some fruitful conversations between feminism and philosophy, the goal of this last section goes further, in a way that (I hope) more actively contributes to that ongoing conversation. I shall be interested in a question that at first sight looks a nonsense: whether someone might be ‘treated as an object’ not just by having their autonomy denied, but by having their autonomy affirmed. As it turns out, the question is not a nonsense, and we shall see that it matters to a number of concerns, both philosophical and feminist.

In the end we shall see that it isn't quite true that philosophy ‘leaves everything as it is’, in either of the senses mentioned at the outset. Philosophy is not inevitably walled off, and irrelevant; and even if it is sometimes a dragging, conservative presence, it needn't be. Feminism brings a radicalizing energy to philosophical inquiry, and philosophy returns the favour, in its own way.

Many philosophers have ascribed to philosophy a power of demolishing prejudice, destroying ‘the habit of holding on to old opinions’, as Descartes's meditator put it (Descartes 1641: 34); and there is no doubt that pioneer feminists found liberation in philosophy, even Descartes's philosophy. Cartesian method requires close scrutiny of prejudice. Cartesian dualism, whatever its defects, requires all of us, women included, to identify ourselves as essentially thinking beings—a significant (p. 233) attainment for creatures overly identified with their bodies. The emphasis on method and on mind are both evident in Mary Astell's work, as she probed that old opinion of the ‘Natural Inferiority of our Sex, which our Masters lay down as…Self-Evident and Fundamental’. She wrote:

Error, be it as antient as it may, [cannot] ever plead Prescription against Truth. And since the only way to remove all Doubts, to answer all Objections, and to give the Mind entire Satisfaction, is not by Affirming,but by Proving, so that every one may see with their own Eyes, and Judge according to the best of their own Understandings, [the author] hopes it is no presumption to insist on this Natural Right of Judging for her self…Allow us then as many Glasses as you please to help our Sight, and as many good Arguments as you can afford to Convince our Understandings: but don't exact of us we beseech you, to affirm that we see such things as are only the Discovery of Men who have quicker Senses; or that we understand and Know what we have by Hear-say only; for to be so excessively Complaisant is neither to see nor to understand. (Astell 1700: 9–10)

Astell advanced the cause of feminism through philosophical method, showing not just absence of proof but outright contradiction in many ‘old opinions’ about women. ‘If all Men are born free,’ she asked, ‘how is it that all Women are born slaves’? (1700, p. 18.). And she drew from Cartesian dualism two feminist morals, a teleological and a practical. When a woman is a meditator, and discovers her nature as a thing that thinks, she discovers something about her purpose in life: ‘a Rational Mind is too noble a Being to be Made for the Sake and Service of any Creature. The Service [a woman] at any time becomes oblig'd to pay to a Man, is only a Business by the Bye. Just as it may be any Man's Business and Duty to keep Hogs’ (1700: 11).1 Hogs? The humour is classic Astell, but it has a point. A man whose essence is to think may still owe duties to hogs, and a woman whose essence is to think may still owe duties to a man. But swineherdly duties and wifely duties are ‘Business by the Bye’, and not what human beings are made for. Her practical conclusion was radical, and at the time unthinkable for most of her readers. If women are creatures whose essence is to think, there should be institutions to foster and develop women's thinking: in short, colleges for women.

Astell did her philosophy as a feminist, and she did her feminism as a philosopher, finding no tension between her philosophy and her feminism. That doesn't mean there was no tension, of course. Feminists in recent times have found in Cartesian philosophy a vivid example of a philosophy implicitly hostile to women; and in thinking about how feminism can contribute to philosophy, dualism is our first port of call.

(p. 234) 1. Dualism

The problem signs are there, say feminist critics, even in the method: that strange fantasy of isolation, the lonely stove-heated room (but who is tending the fire?), the meditator doubting the existence of his hands and body (but who is bringing the meals?). The signs are there, they say, in the fantasy of autonomy, the denial of human relationship, the priority of the ‘I’, the epistemological ambition. Look, look, I can do it all by myself! My existence, my essence, my knowledge—I can do them all on my own! (With just a little help from God.) It is above all in the dualism that the signs are manifest: the denigration of the senses, the denigration of matter, the divorce of mind from matter. So feminist critique of Descartes has taken many forms, exploring his metaphors, putting the meditator on the analyst's couch, reading reasons as symptoms, reading arguments as the workings of subconscious and deeply misogynistic desire (see Irigaray 1985; Flax 1983).

Whatever the merits of this critique of Descartes himself, one might be tempted to think it anachronistic. How much investment do philosophers have in dualism these days? In contemporary philosophy of mind, dualism is the exception, not the rule. After all, it's not as if one needs a feminist agenda to suspect Descartes might be wrong. There are plenty of reasons beyond political reasons, beyond psychoanalytic reasons, for supposing Descartes mistaken, and it is no accident that dualists are fewer than they were. Yet feminists maintain a continuing interest in dualism, extending well beyond its most famous historical champion. Why?

Feminists have been bothered by a kind of dualistic thinking that informs many philosophical projects, Cartesian or not. Feminists have claimed that there is something gendered about oppositions between body and mind, between emotion and reason. Mind and reason, they say, have been associated with the masculine; body and emotion with the feminine; and these associations, overt in the work of past philosophers, continue covertly even in philosophy today. Faced with a gendered opposition between, say, mind and body, it is not enough to say that women have minds too, as Astell did eloquently enough. One needs also to question the division that makes mind so different from body, and valorizes mind at the body's expense. (‘Minds, yes’, colleges for women, yes. But that doesn't have to mean ‘Bodies, no’.) So a feminist focus on Cartesian themes has contributed to a range of philosophical debates. A dubious strain of ‘methodological solipsism’ has been observed in political philosophy by a number of critics, including Alison Jaggar (1983): a dream of autonomy gone mad, with its assumption that something is only worth doing if you do it all by yourself. An analogous problem has been noted in moral epistemology by Karen Jones (1999)—an assumption that moral knowledge must be achieved all by yourself, not something that can be got ‘second-hand’, by way of testimony from others better situated than you are.

(p. 235)

Doing justice to the range of feminist contributions on this topic is impossible, so keeping to the principle that it's better to display than survey, let us take a look at a recent, and representative, argument, in metaphysics. Susan James (2000) has tried to show how dualistic thinking is present even in supposedly anti-dualistic arguments, tracing it in contemporary analytic work on personal identity and survival.2 On a traditional Cartesian way of thinking, a person at one time is identical with a person at a later time if and only if they are the very same mental substance. This Cartesian view has been opposed by Derek Parfit and others who say that what matters is not personal identity, but survival; and that a person may survive at a later time without being identical to what she was before. Survival, unlike identity, can come in degrees, so this on the face of things allows for a more flexible notion of the self. What interests James in this debate are the criteria philosophers use in thinking about survival. She observes how contributors to the debate have insisted upon a sharp distinction between two criteria, those of psychological continuity and physical continuity. To discover which of these is the relevant criterion for survival, we are asked to imagine the psychological continuity without physical continuity, and vice versa. To take the ‘transplant’ thought experiment: imagine that a person's memories and character are transplanted from one body into a different one. Where is the person now? The alleged intuitive answer is that the person is where the character and memories are, not where the previous body is. And so we are expected to conclude that the criterion for survival is psychological continuity, not physical continuity.

Now although this way of thinking is explicitly developed as an alternative to Cartesian dualism, its stark opposition between the physical and psychological makes it, in James's eyes, an immediate target of suspicion. The opposition helps to sustain a ‘symbolically masculine’ conception of personhood, says James. Transplant thought experiments tend to regard the body as an exchangeable container or receptacle for character or memory. Some philosophers acknowledge that this must be an oversimplification, that one cannot have body-swapping with just anybody, and still preserve psychological continuity: body-swapping between a man and a woman, between an emperor and a peasant, might fail to preserve all aspects of psychology. But James observes that those noting this problem tend to dismiss it. They say ‘Let us forget this’; or they stipulate that the swapped bodies be extremely similar; or they simply assert that character can survive large and disastrous alterations to body-type—that someone can still be brave, say, even if they lose the physical capacity to do daring things. James is sceptical. Not all character traits are so readily maintainable, she says: courage, perhaps, but what of physical dexterity, or delight in one's sexuality? Changing the body might well change these (p. 236) supposedly psychological traits. Furthermore, the assumption that continuity of memory can be had without continuity of body is at odds with the phenomenology of memory, she says: some memories, including memories of trauma and violence, are experienced at least partly in the body, or so survivors report (see e.g. Brison 1987). Applying the criterion of psychological continuity depends on a kind of laundering of memories and character traits, washing away all traces of body.

There are important questions, says James, about how body image, and recognition by others, can matter to who you are. According to some psychoanalytic theorists, a person's sense of their own body, as they take it to be represented by others (or as represented in mirror reflection), is a part or precondition of psychic integrity. And whether or not that is true, it may still be that body image and recognition by others both matter to one's sense of self—that who we are (or take ourselves to be?) can depend in part on who we are recognized and affirmed by others to be. Social recognition is mediated, at least in part, by perception of body. Suppose, says James, that the memories and character of a female fashion model are transplanted into the body of a male garage mechanic. What becomes of her body image now? Even on the assumption that friends and lovers believe her story, will they give her the kind of recognition and affirmation they did before? Some philosophers have insisted on the irrelevance of body to questions of this kind. Bodies are convenient tags for our recognition of other persons, and that's all. We use bodies, says Anthony Quinton, as

convenient recognition devices enabling us to locate the persisting characters and memory complexes…which we love or like. It would be upsetting if a complex with which we were emotionally involved came to have a monstrous or repulsive physical appearance …But that our concern and affection would follow the character and memory complex…is surely clear. (1975: 60)

James is again sceptical. That way of thinking fails to allow that a person's ability to sustain continuity might sometimes depend on other people recognizing and affirming them as embodied in a particular way.

When we imagine the memories and traits of a person being transplanted from one body to another, and imagine that as survival, we are imagining the person feeling much the same as before, committed, perhaps, to similar projects as before, committed to their own future just as before, and maintaining, perhaps, the very relationships they maintained before. But if the new body is sufficiently different, all that is surely up for grabs, says James. Many memories and traits will simply fail to survive the translation, the person may lose the recognition they enjoyed from others, may lose their projects, and lose emotional investment in the life they find themselves in. Some philosophers may insist there is still survival: but why go to such lengths to protect psychological continuity from the effects of the body and the rest of the world? Doesn't this begin to look just a little like Descartes's solipsistic meditator, so eager to wall off the world and the body, so eager to hold on to (p. 237) the mind as the real self? James concludes that when theorists of personal identity focus on psychological continuity as the stronghold of the self, ‘they secure only a self which would in other circumstances be regarded as pathologically disturbed’ (2000: 45): and in explaining why that could possibly seem attractive, we must go, she says, to the gendered structure of philosophical thinking.

2. Androcentrism

If ways of thinking are ‘androcentric’ when they reflect ‘an orientation geared to specifically or typically male interests or male lives’ (as Elizabeth Anderson 1995: 70 puts it3) then perhaps some kinds of dualistic thinking are androcentric. But if dualism is one possible manifestation of androcentrism, it is by no means the only one. One famous feminist contribution to ethics has been the charge of androcentrism brought by Carole Gilligan against the so-called ‘Justice’ perspec-tive, a standard of moral maturity described by the developmental psychologist Lawrence Kohlberg. The capacity to assess moral situations in terms of impartial universal principles, reasons, and rights was described by Kohlberg as the pinnacle of moral development, and by this standard (itself identified through study of allmale samples) girls and women were found to be less morally mature than boys and men.

Gilligan accepted (controversially) that women's ‘moral voice’ is different, citing the following kind of example. Confronted with the question ‘When responsibility to oneself and responsibility to others conflicts, how should one choose?’, a male subject, Jake, responds, ‘You go about one-fourth to the others and three-fourths to yourself’. Amy, on the other hand, responds,

Well, it really depends on the situation. If you have a responsibility with somebody else, then you should keep it to a certain extent, but to the extent that it is really going to hurt you or stop you from doing something that you really, really want, then I think maybe you should put yourself first. But if it is your responsibility to somebody really close to you, you've just got to decide in that situation which is more important, yourself, or that other person, and like I said, it really depends on what kind of person you are and how you feel about the other person or persons involved. (Gilligan 1982: 36–7)

Is Amy different? Yes. More long-winded? Hesitant? Perhaps. Immature? Decidedly not, according to Gilligan. Jake has a neat and crude rule, expressive of the ‘Justice’ (p. 238) perspective. The voice of Amy expresses an ethic of ‘Care’ rather than justice, according to Gilligan (1982: 36–7), emphasizing responsibility, relationships, and the particularity of situation—a different voice, but by no means inferior.4

Androcentrism is as much at home in epistemology and philosophy of science as it is in ethics, and here again an example will have to serve to illustrate the point. Looking at a case study in evolutionary biology, Elisabeth Lloyd (1993) has shown how androcentrism can influence the pursuit of scientific knowledge: it can influence the framing of a question, and also the search for an answer to it.5 The question she considers is one that is interesting and prima facie puzzling for evolutionary biologists, and indeed for anyone with an ounce of curiosity: namely, why do women have orgasms? In the context of evolutionary biology, the question is taken to imply a contrast, first between the situation of women and that of men; and second, between the situation of women and that of other female primates. Men have orgasms, but that presents no ‘why’ question for evolutionary biologists: it isn't hard to see why male orgasm might have ‘adaptive value’, as an obvious aid to reproducing one's genes, a powerful motivator for intercourse. But women don't need orgasms to reproduce. Moreover, the females of non-human primates get by well enough without them, or so it has been said. Hence the question: why do women have orgasms—given that women do have them (unlike other female primates), and don't need them (unlike men)?

In looking for answers, theorists looked for a link between orgasm and direct reproductive success, arguing that as a matter of fact women do need orgasms for reproductive success, or (more lamely) that at any rate orgasms can help.

Desmond Morris had one idea. Women walk upright, unlike other female primates, and that is why women have orgasms. As our hominid ancestors abandoned their knuckle-scraping ways and began to walk tall, they were faced with a new reproductive problem, posed by the simple force of gravity. The apes might take their post-coital promenade with impunity, precious sperm tucked safely inside; but for the biped human who too hastily arose for a stroll—alas, she risked losing all. Cue the female orgasm, as a helpful anti-gravity device:

If a female of our own species were so unmoved by the experience of copulation that she too was likely to get up and wander off immediately afterwards… [u]nder the simple influence of gravity the seminal fluid would flow back down the vaginal tract and much of it would be lost. There is therefore…a great advantage in any reaction that tends to keep the (p. 239) female horizontal when the male ejaculates and stops copulation. The violent response of female orgasm, leaving the female sexually satiated and exhausted, has precisely this effect. (Morris 1967: 79)

According to Lloyd, Morris's theory was taken up by a number of later theorists including Gordon Gallup and Susan Suarez (1983), who said in its defence that ‘the average individual requires about 5 minutes of repose before returning to a normal state after orgasm’.

There is a more widely accepted answer than Morris's anti-gravity theory, citing the adaptive role of orgasm in ‘pair-bonding’, which increases an organism's reproductive success. Offspring raised by heterosexually pair-bonded parents stand a greater chance of survival, as parents assist each other in child-rearing. Orgasm evolved for women because it supplied reward and motivation to engage in frequent intercourse, which in turn cemented the pair-bond, which in turn promoted repro-ductive success.

What, if anything, is androcentric about the shape of this question and its answers? To the extent that the question is supposed to be about why women, unlike other primates, have orgasms, it rests on a false presupposition. Unlike other primates? Female bonobos, pygmy chimpanzees, and stump-tailed macaques have wild sex lives: they scream with enjoyment as they practise what primatologists call ‘genito-genital rubbing’, they mount each other to stimulate themselves to orgasm. Such behaviour was for a considerable time not noticed as sexual at all, because it does not involve interaction with other males, does not correlate with hormonal oestrus, and is apparently of no reproductive significance. Instead such behaviour was seen as ‘ “greeting” behaviour, or “appeasement” behaviour, or “reassurance” behaviour, or “reconciliation” behaviour…or “food exchange” behaviour—almost anything, it seems, besides pleasurable sexual behaviour’ (Begemihl 1999, quoted in Saul 2003: 235).6 So androcentrism (among other biases) contributes to the shaping of a flawed question.7

What of the answers? If the question wrongly supposes that women are unlike other primates in experiencing orgasm at all, the answers wrongly suppose that women are like men in how they experience orgasm. The anti-gravity theory and the pair-bond theory both assume that women reliably achieve orgasm through penetrative heterosexual intercourse; the anti-gravity theory assumes in addition (p. 240) that women collapse, exhausted and appropriately horizontal, when they achieve it. Both sets of theorists blithely cited the then available clinical evidence: but that evidence flatly contradicts their assumptions, and on both counts, according to Lloyd.

The studies show that there is for women a significant ‘orgasm–intercourse discrepancy’, as Lloyd calls it. Only 20–35 per cent of women always (or almost always) experience orgasm with unassisted intercourse; around 30 per cent never do (Lloyd 1993: 144–5; she cites Hite 1976; Kinsey et al. 1953; and for cross-cultural comparison, Davenport 1977). Not to have an orgasm from unassisted intercourse is the experience of most women, most of the time, says Lloyd (1993: 144): ‘not to put too fine a point on it, if orgasm is an adaptation which is a reward for engaging in frequent intercourse, it does not seem to work very well’. Lloyd allows that the orgasm–intercourse discrepancy is compatible with there being some selective advantage after all to female orgasm. But there is at the very least a prima facie problem that the anti-gravity and the pair-bonding theorists need to address; for if a trait is supposed to have evolved as an adaptation, yet is rarely used in its supposedly adaptive context, that calls for an explanation. Androcentrism manifests itself in the fact that it did not cross their minds to address it, notwithstanding their access to the relevant studies.

The anti-gravity theory of female orgasm attributes to women in addition a distinctive experience of orgasm's after-effects: the female is ‘sexually satiated and exhausted’, she needs to stay ‘horizontal’, she needs ‘about 5 minutes’ of ‘repose’. Readers might not be wholly surprised to learn that the studies cited in support of these claims were studies of male sexual experience. (According to Lloyd 1993: 146, 148, Morris cites the Masters and Johnson 1966 studies of male sexual response; Gallup and Suarez cite the Kinsey et al. 1948 studies of male sexual response.) Not all women are satiated and exhausted after orgasm. On the contrary, 47 per cent in one survey did not feel that a single orgasm was always satisfying to them. Many women have (unlike men) the desire and capacity for more than one orgasm, without a significant break, sometimes (according to some studies) five or six within a matter of minutes, a capacity linked to the fact that all women tend, after orgasm, to return to the so-called ‘plateau’ phase of excitement, rather than the pre-aroused state (as men do) (for the 47 per cent figure, Lloyd cites Hite 1976: 417, and for the remainder Kinsey et al. 1953: 375–6 and Masters and Johnson 1966: 65).

It seems that androcentrism contributes to the shape of these flawed answers, then, as well as to the flawed question: for these explanatory hypotheses appear to reflect ‘an orientation geared to specifically or typically male interests or male lives’, as Anderson's phrase has it. We are left, though, with two questions, one for the philosopher of science, and one for the merely curious. I shall take a brief look at both.

If androcentrism can affect the practice of science at all these different levels, including the collecting of data, and the drawing up and evaluating of hypotheses, there is a question, first, about how philosophy of science should respond. Is there, (p. 241) or should there be, a distinctively feminist philosophy of science? Among feminists, various affirmative answers have been given, though all agree that the fact that women have been left out has hurt science (and hurt women); and the contribution of women can somehow help science (and help women). Feminist standpoint theory says that women can make a special contribution to knowledge: given their relative outsider-status and powerlessness, women occupy a relatively privileged epistemic position (comparable in certain respects to that of the workers, in Marxist theory)—a distinctive standpoint achievable, to a degree, by anyone willing to start their thought from women's lives (see e.g. Hartsock 1983; Harding 1993; Jaggar 1983). Helen Longino argues, however, that the issue is not so much the spe-cialness of the standpoint, but rather the need for diversity in a scientific community. Scientific objectivity itself is not a property of an individual scientist or a particular theory, but of a scientific community: it results from the critical interactions of its diverse groups and members (including women), with all their different assumptions and different interests. Louise Antony responds that even individuals can learn to become more objective, so objectivity cannot just be a property of communities. Moreover, the goal is not to eliminate bias (for bias is inevitable and in her opinion sometimes healthy), but to discover which biases are good because they lead to truth, and which are bad because they lead to error. Feminism can contribute to a kind of naturalized epistemology by helping to discover, through study and experience, which biases lead to truth and which to error; and it will be particularly alert to androcentric biases leading to error in, for example, the work of evolutionary theorists of women's sexuality.

For the merely curious, there remains the original question, why do women have orgasms? Lloyd's own view is that the demand for an adaptationist answer to the ‘why’ question is misguided. There is no direct teleological answer. The female orgasm is an ‘evolutionary freebie’. Women have orgasms because men need them to reproduce; and, from a developmental and embryological perspective, it is handy to make men and women somewhat alike. Men have nipples merely as a developmental side-effect of women's reproductive needs, but men don't need them. Women have orgasms merely as a developmental side-effect of men's reproductive needs, but women don't need them. Lloyd finds plausible this developmental hypothesis, which was initially proposed by Donald Symons (1979). Other feminists, though, have charged this hypothesis too with androcentrism. Look at that ‘merely’, they say. Surely that is being dismissive of female sexuality, denying the significance of female pleasure. Not necessarily, says Lloyd—indeed, not at all, unless you think (heaven forbid!) that significance is tied to reproductive advantage. The hypothesis is, on the contrary, emancipatory.

Think, after all, of our other evolutionary freebies. Take music, for instance: the ability we all have to make and enjoy music is ‘merely’ a happy by-product of cognitive abilities whose direct reproductive advantage lay elsewhere. But there's no real ‘merely’ about music: music is none the worse for its evolutionary by-product (p. 242) status, and some might say it is the better for it. Female sexuality would be none the worse either, and may be the better. As Lloyd says (1993: 149), ‘the realm formerly belonging to the reproductive drive would now be open to much, much more’.8 From this evolutionary perspective on orgasm, it may be the boys who want to make babies; but the girls just wanna have fun.

We have been thinking so far about how feminism might illuminate philosophy, focusing on two general ideas (and some particular applications) that have been central to feminist critique, namely the ideas of dualism, and of androcentrism. There is the converse question of how philosophy might illuminate feminism, a topic equally resistant to easy summary. But as we observed at the beginning, feminist philosophers seem to have had little trouble in putting together their feminism and their philosophy. Even Cartesian dualism could, in the hands of a Mary Astell, produce arguments for women's equality and women's education. More generally, feminist critiques that complain of dualism and androcentrism in philosophy very often draw upon other sorts of philosophy to serve their critique, and to construct better alternatives. Antony, as we just saw, finds a Quinean naturalized epistemology to be particularly congenial to a feminist philosophy of science, offering space for an empirical investigation into the biases that help or hinder.

Feminists have not hesitated to mine philosophy in their efforts to understand better the problems that bother them, as feminists. The idea of a speech act, for example, has been applied to topics that would not, in his wildest nightmares, have crossed J. L. Austin's own mind. The idea of speech that is not ‘only words’ (to borrow Catharine MacKinnon's phrase) has proved a fertile one for feminists keen to explore some hard and interesting questions. How is gender itself constructed? Judith Butler sees performativity in all social sayings and doings: beginning with that potent exclamation at birth—‘It's a girl!’—our utterances never merely report gender, but actively construct it (see e.g. Butler 1990, 1997). There are some specific political questions: how might pornography subordinate women? If speech is a kind of act, and pornography is speech, there is a question about the kind of speech act pornography is. Perhaps it is speech that, as illocutionary act, subordinates women. Under what conditions does someone have freedom of speech? If speech is illocution, then in Austinian terms, it involves not just noises but uptake, a kind of recognition on the part of hearers: so freedom of speech might require not just the minimal conditions for uttering words, but the conditions for being appropriately heard.9 Some feminists have used speech act theory to address a very basic question about language: what is the fundamental concept of linguistic meaning? Jennifer (p. 243) Hornsby (2000) argues that the most basic idea is that of saying something to someone, and it involves hearers as well as speakers. These last two applications of Austinian ideas connect with a more general feminist opposition to atomism: one should not assume that we are self-sufficient speakers any more than that we are self-sufficient Cartesian knowers.

In considering more closely how philosophy might illuminate feminism, I want to home in now, by way of illustration and more, upon an idea from moral philosophy that has been taken up with great effect by feminists.

3. Treating Someone as an Object

The thought that there might be something wrong with treating someone ‘as an object’ existed in philosophy long before feminism got hold of it, gaining one of its most famous expressions in Kant's work. For Kant, moral wrongdoing consists in the failure to treat humanity ‘always…as an end, never merely as a means’, a failure to respect that humanity ‘by virtue of [which] we are not for sale at any price’ (Kant 1785: 429; 1797: 435). This historically Kantian idea has gained new impetus in recent applications by feminist thinkers, who have observed its relevance to oppression, and to the varied ways that women might have been treated as a means only, and sometimes put up for sale.10 For feminists there is a focus on treatment encountered by women, and a claim that women's oppression (partly) consists in women's being treated ‘as objects’: Catharine MacKinnon says, ‘To be sexually objectified means having a social meaning imposed on your being that defines you as to be sexually used…and then using you that way’ (1991: 122).

That notion of sexual ‘use’ echoes the Kantian stricture against treating a person as a mere means or instrument. Kant himself thought that sexual relationships presented special dangers about the ‘use’ of the other person, as Barbara Herman (1993) has emphasized, pointing out the common ground shared by Kant and certain feminists.

Martha Nussbaum draws together the Kantian and feminist ideas in an instructive study of what it might be to treat someone as an object. Objectification is a cluster (p. 244) concept, on her way of thinking, in which the ideas of autonomy denial and instrumentality are at the core; but the cluster also includes related notions of inertness, fungibility, violability, ownership, and denial of subjectivity. Nussbaum's proposal deserves scrutiny, as an outstanding example of the kind of inquiry that is seamlessly feminist and philosophical; and I shall be wanting to ask whether the proposal does justice to the Kantian and feminist heritage Nussbaum claims for it, and if not, how it might best be augmented. I shall be drawing special attention to the idea of ‘treating’ in that notion of ‘treating someone as an object’. And while Nussbaum places the denial of autonomy at centre stage, I shall also investigate a possibility that smacks of paradox: whether one might treat someone as an object through affirming their autonomy.

This latter question is of philosophical interest, and of practical interest to feminists as well. One feminist application of the idea of objectification has been to pornography, which has been thought to deny women's autonomy, in its depiction of women ‘as objects, things or commodities’.11 However, other pornography has been thought to affirm women's autonomy, representing women as not in the least object-like or subordinate, but as active sexual agents; and the suggestion has been that this, surely, is politically innocent. Joel Feinberg (1985: 144) describes with nostalgia some ‘comic strip pamphlets’ of the 1930s and 1940s, which ‘portrayed …a kind of joyous feast of erotica in which the blessedly unrepressed cartoon figures shared with perfect equality. Rather than being humiliated or dominated, the women characters equalled the men in their sheer earthy gusto.’12 The pornographic film Deep Throat was hailed for representing women as sexually autonomous, its heroine described as ‘Liberated Woman in her most extreme form—taking life and sex on her own terms’.13 Such descriptions aim to be vindications, aim to distinguish a liberating from a subordinating pornography. Indeed feminist anti-pornography legislation allows for just such a distinction, labelling non-subordinating pornography as ‘erotica’. So in thinking about whether pornography might objectify women, it is tempting to think that the question must be about pornography that denies women autonomy—and not, surely, about pornography that affirms autonomy.

Well, the answer, as we shall see, is not so simple. Perhaps there really could be a liberating, autonomy-affirming pornography; certainly there are feminists who think the solution is not less pornography, but more—more pornography created by women, for women. Perhaps the 1930s and 1940s comic strips really were liberating; (p. 245) for present purposes we might as well accept Feinberg's rosy evaluation. But while allowing this, one can also allow that it is not the whole story. The description of Deep Throat as representing ‘Liberated Woman in her most extreme form’ must in the end be resisted, not because its affirmation of autonomy is a sham, but because, as I shall argue, objectification sometimes depends on affirmation of autonomy. This may sound strange at first, but it turns out to be a natural enough consequence of the plurality of ways in which someone can be treated as object.

Nussbaum herself begins with a plurality, naming seven features that form the cluster concept of objectification, seven ways to ‘treat an object as a thing’:

  1. 1. Instrumentality: one treats it as a tool of one's own purposes.

  2. 2. Denial of autonomy: one treats it as lacking in autonomy and self-determination.

  3. 3. Inertness: one treats it as lacking in agency and activity.

  4. 4. Fungibility: one treats it as interchangeable (a) with other things of the same type, and/or (b) with things of other types.

  5. 5. Violability: one treats it as lacking in boundary-integrity, as something that it is permissible to break up, smash, break into.

  6. 6. Ownership: one treats it as something that is owned by another, can be bought or sold, etc.

  7. 7. Denial of subjectivity: one treats it as something whose experience and feelings (if any) need not be taken into account. (Nussbaum 1995: 257)

    I take this to be a particularly helpful proposal about what ‘object’ amounts to, in the notion of ‘treating as an object’, and this is, in my view, at least half the story. The other half, as we shall see, rests not on what an ‘object’ is, but on what ‘treating as’ amounts to. Nussbaum's proposal gives content to the idea of objectification by teasing out a number of features associated with objecthood: an object lacks autonomy, lacks subjectivity, is inert, an appropriate candidate (sometimes) for using as a tool, exchange, destruction, possession.

    While she observes that not all objects are candidates for ‘objectifying’ treatment (a painting is not violable, fungible, a mere instrument), it is in the application to persons that the idea of objectification gets its chief point. And in the case of persons, treatment relating to autonomy becomes central, she thinks, partly because of its implications for other modes of object-making. When you treat a person as autonomous, that seems to imply not treating them just as instrument; it seems to imply treating them as not simply inert, not owned, not something whose feelings need not be taken into account. In other words, instrumentality, inertness, ownership, and denial of subjectivity each imply denial of autonomy.14 Instrumentality seems central too, but in a different way: many things are treated as lacking in (p. 246) autonomy which nonetheless ought not to be treated as mere means to one's ends—she mentions the examples of children, pets, and paintings. These examples suggest that while instrumentality may imply autonomy denial, the converse does not in general hold. Non-instrumental treatment is compatible with autonomy denial (young children), and with subjectivity denial (paintings). That said, when it comes to the treatment of adult human beings, the connections between instrumentality and autonomy denial are closer: non-instrumental treatment of adult human beings does imply treating them as autonomous. When it comes to adult human beings, she suggests that not treating someone as an instrument implies not denying their autonomy—and vice versa.15

    She applies this idea to sexual objectification, attending to a variety of texts, from the relatively highbrow (D. H. Lawrence, James Joyce, Henry James) to the low (Playboy, and a sadistic sample from the work of one ‘Laurence St Clair’); and she uses it to defend many aspects of the feminist understanding of women's subordination developed by Catharine MacKinnon and Andrea Dworkin, together with their associated critique of pornography. She develops at the same time a potentially surprising argument that some sorts of objectification, indeed some sorts of denial of autonomy, may be ‘a wonderful part of sexual life’—surprising, if the initial thought was that objectification is a distinctive moral failure.16

    If we are interested in whether Nussbaum's proposal does justice to the feminist philosophical understanding of ‘treating someone as an object’, it is worth asking whether there are aspects of the idea of an ‘object’ that are relevant to feminists, but absent from Nussbaum's proposal. With this in mind, the following could usefully be added to Nussbaum's list of seven:

  8. 8. Reduction to body: one treats it as identified with its body, or body parts.

    This idea is absent from the official list, notwithstanding its prominence in Nussbaum's illustrative examples. It appears in a clause of the famous feminist (p. 247) anti-pornography ordinance from which she takes inspiration, according to which (some) pornography treats women as ‘dehumanized as sexual objects, things, or commodities…reduced to body parts’.

  9. 9. Reduction to appearance: one treats it primarily in terms of how it looks, or how it appears to the senses.

    This is worth including for its importance, in different ways, to both Kantian and feminist thinking. It appears in Nussbaum's illustrative examples too, whether in the objectifying character of soft-core pornography, or in the relationship of a couple described by Henry James, who value each other in a purely aesthetic way, appropriate for fine paintings and antiques.

  10. 10. Silencing: one treats it as silent, lacking the capacity to speak.

    Speech is a distinctive capacity of persons, just as distinctive perhaps as autonomy and subjectivity. Admittedly, this was not a particularly important idea for Kant, but it has been important to feminists, who hold that women's subordination is partly constituted by the fact that women have been silenced, for example, by pornography. ‘Pornography makes [women's] speech impossible, and where possible, worthless. Pornography makes women into objects. Objects do not speak. When they do, they are by then regarded as objects, not as humans’ (MacKinnon 1987: 182). Leaving aside the question of what exactly this silencing amounts to, the idea of silence has been central to feminist thinking about women's situation, and it is worth adding independently to Nussbaum's list.

With these inclusions, the relevant idea of an ‘object’ turns out to be of something lacking in subjectivity and autonomy, something inert, something that is an appropriate candidate for using as a tool, exchange, destruction, possession, all as Nussbaum suggested; and in addition it is something that is silent, something that is just an appearance, just a body.

Teasing out a plurality of features associated with objecthood is, I suggest, only half the task. In this idea of ‘treating someone as an object’, we need to look not only at the notion of an object, but also at the notion of treatment. Here too we confront a plurality, albeit a different one. ‘Treat’ is a wide-ranging verb that has been functioning as a dummy, standing in for a host of different attitudes and actions. ‘Treating’ may be a matter of attitude or act: it may be a matter of how one depicts or represents someone, or a matter of what one (more actively) does to someone. Supposing, for simplicity, we restrict our attention to treatment relating to autonomy; supposing, too, we agree with Nussbaum that lack of autonomy is a salient feature of objects, and that ‘denying autonomy’ is therefore an aspect of ‘treating as an object’: the question remains as to what denying autonomy amounts to.17

On Nussbaum's description, denial of autonomy takes place when one treats something ‘as lacking in autonomy and self-determination’ (Nussbaum 1995: 257). (p. 248) Here are some examples she uses to illustrate autonomy denial. One permissibly denies autonomy when one treats a pen, a painting, one's pets, one's small children, as nonautonomous. Ownership is by definition incompatible with autonomy, so slaves are denied autonomy: and ‘once one treats as a tool and denies autonomy, it is difficult to say why rape or battery would be wrong, except in the sense of rendering the tool a less efficient tool of one's purposes’. Aspects of slavery anticipate the MacKinnon–Dworkin understanding of sexual objectification,

since [slavery] shows us how a certain sort of instrumental use of persons, negating the autonomy that is proper to them as persons, also leaves the human being so denuded of humanity, in the eyes of the objectifier, that he or she seems ripe for other abuses as well— for the refusal of imagination involved in the denial of subjectivity, for the denial of indi-viduality involved in fungibility, and even for bodily and spiritual violation and abuse, if that should appear to be what best suits the will and purposes of the objectifier.

The relationship between the Brangwens, described by D. H. Lawrence, involves ‘a mutual denial of autonomy’, ‘a kind of yielding abnegation of self-containment and self-sufficiency’. An example of hard-core sadistic pornography ‘represents women as creatures whose autonomy and subjectivity don't matter at all’, the woman's ‘inertness, her lack of autonomy, her violability’ is eroticized (Nussbaum 1995: 261, 264, 265, 273, 280).

Is autonomy denial a matter of attitude, or act, or both? Nussbaum's examples exploit a plurality of different ways of treating. An agent presumably believes the pen, the painting, the small child, are lacking in autonomy—that is why they then act as if they are lacking in autonomy—and act differently, depending on whether it is the child or the pen whose autonomy is ‘denied’. In buying the slave, an agent thereby ‘denies’ autonomy: but in what sense? The act may presuppose an attitude, of failing to regard the slave as autonomous. The act of buying may be an act of autonomy denial that violates the slave's autonomy, preventing him from having any choice in the matter. Or perhaps slavery does something even worse to the slave's autonomy—stifles it, or destroys it.

To make a start, in considering autonomy denial, non-attribution of autonomy needs to be distinguished from violation of autonomy. I take it that non-attribution is primarily a matter of attitude, while autonomy violation is something more—a more active doing, perhaps one that prevents someone from doing what they choose. The distinction between non-attribution and violation is obscured, surely, by allowing ‘autonomy denial’ to label both. The two are independent: there can be autonomy denial qua non-attribution, without autonomy denial qua autonomy violation—and vice versa.18 To illustrate the first possibility, there can be non-attribution of autonomy without violation of autonomy, in the ‘objective’ attitude (p. 249) described so well by P. F. Strawson (1974: 9), the attitude of the doctor or social scientist: ‘To adopt the objective attitude to another human being is to see him, perhaps, as an object of social policy; as a subject for what, in a wide range of sense, might be called treatment; as something…to be managed or handled or cured or trained.’19 Someone denying autonomy in this attitudinal way need not deny autonomy in other ways. A doctor viewing his patient from the objective stance may act in ways that override his patient's choices, but he may not: perhaps institutional procedures of securing informed consent will prevent autonomy violation, notwithstanding the shortfall in the doctor's attribution of autonomy. To illustrate the second possibility, there can be autonomy violation without failure to attribute autonomy, as in the classic examples from Kant's Groundwork,where autonomy is attributed, and violated. Someone who makes a lying promise to repay his friend's money does not suppose his friend lacking in autonomy, but does violate his friend's autonomy, on the Kantian view. An example more relevant to feminists might be sadistic rape, i.e. rape where non-consent is actively sought, rather than disregarded or ignored. In this sort of case, it's not that he doesn't listen to her saying ‘no’, he wants her to say ‘no’. Here there is violation of a woman's autonomy committed by someone who affirms her autonomy, attributes to her a capacity for choice—and desires precisely to overcome that choice, make her do what she chooses not to do.20 In sadistic rape, someone is ‘treated as an object’ in part by attributing autonomy to them in one way—precisely so that autonomy can be denied a different way. Autonomy attribution is a necessary feature of this way of treating as an object.21

Reflection suggests that someone violating autonomy not only can at the same time affirm autonomy but must affirm it, when the violation is deliberate: autonomy violation is not just compatible with autonomy attribution, but requires it, the one autonomy denial (violation) depending on the absence of the other (nonattribution). Deliberate violation of someone's choice presupposes attribution of a capacity for choice. This underlines the distinctness of the two ‘autonomy denials’ of non-attribution and violation: it shows how an act may be autonomy-denying in one way, and not another; and further, that autonomy denial is not just compatible (p. 250) with autonomy affirmation, but sometimes requires it. Here we have one way that objectification might actually depend on autonomy affirmation.

Non-attribution of autonomy, and autonomy violation, are two importantly different ways of denying autonomy: but still further possibilities are implicit in Nussbaum's discussion: for example, self-surrender of autonomy, and relatedly a demand for another's self-surrender of autonomy. These are features of Tom Brangwen's sexual experience, as described in The Rainbow, ‘a kind of yielding abnegation of self-containment and self-sufficiency’, offered by himself, and hoped for from his partner. This self-surrender, and the demand for it, are not failures to attribute autonomy, nor violations of autonomy. And while they are characterized as potentially ‘wonderful’ parts of sexual life, in conditions of mutuality and equality, an asymmetric version of the demand for surrender is less wonderful, being a feature of a different sadism, one that seeks surrender of autonomy, the ‘abjuration’ described unforgettably by Sartre (1943):

The spectacle which is offered to the sadist is that of a freedom which struggles against the expanding of the flesh, and which freely chooses to be submerged in the flesh. At the moment of abjuration …a freedom chooses to be wholly identified with this body; this distorted and heaving body is the very image of a broken and enslaved freedom.

The result sought by the sadist is that the other abjure herself, abjure her autonomy, freely choose to become thing-like. It misses something to place this demand for surrender under the generic label of ‘autonomy denial’, as if it were a version of what is going on when someone ‘denies’ autonomy to small children and inanimate objects.

Still further possibilities include the destruction or stifling of autonomy, perhaps implicit in the sexual slavery of The Story of O, described by Andrea Dworkin, and discussed by Nussbaum (1995: 269, quoting Dworkin 1974: 58). ‘O is totally possessed. That means that she is an object, with no control over her own mobility, capable of no assertion of personality, her body is a body, in the same way that a pencil is a pencil, a bucket is a bucket.’ This presents a deeper damage to autonomy, a snuffing out of the capacity for choice; or a stifling of that capacity, if it is prevented from growing in the first place.

We have been seeing how varied are the modes of ‘treating’, even when the aspect of ‘objecthood’, namely absence of autonomy, is held constant. This variation allows for the possibility of treatment that denies autonomy in one way, while affirming it in another. It allows for variation in moral significance: thinking of someone as lacking in autonomy is less invasive than violating or destroying their autonomy. Such variations are due not just to context-sensitivity (which Nussbaum rightly emphasizes), but to differences in what the agent is doing to autonomy itself: failing to attribute it, violating it, surrendering it, demanding that another surrender it, destroying it, stifling it.22

(p. 251)

We can observe that this plurality of modes of treatment probably extends to other listed features of objecthood. Subjectivity denial, for instance, may be failure to attribute subjective mental states; systematically attributing the wrong subjective mental states; manipulating someone's subjective mental states; even perhaps invading, destroying, or stifling their subjectivity. Silencing may be failure to attribute a capacity to speak, preventing someone from speaking, destroying or stifling someone's capacity to speak. The diverse ways of treating as an object link up with the diverse aspects of being an object, creating combinatorial possibilities whose surface we have barely begun to scratch.

In applying this to the question of pornography, we can begin with Nussbaum's comment on a violent sexual tale in the work of ‘Laurence St Clair’, material which in her view would fall clearly in the scope of MacKinnon's definition. Nussbaum agrees with MacKinnon that such material is objectifying, notwithstanding an ‘assuaging fiction’ that this violating treatment ‘is what she has asked for’—i.e. notwithstanding an ‘assuaging fiction’ that violation is the woman's choice. Nussbaum dismisses the affirmation of autonomy as, in this case, an ‘assuaging fiction’, but her remark brings us to the issue of pornography that ‘affirms autonomy’.

Recall that Deep Throat was hailed as presenting ‘Liberated Woman in her most extreme form, taking life and sex on her own terms’. Should that description be dismissed, as a mere ‘assuaging fiction’? It is tempting to say yes, that the autonomy attribution is a sham: Linda Lovelace is hardly affirming her autonomy when she embarks on a life driven by nothing more than an insatiable desire for throat sex. That is a possible option, but it is not the only one. There is a clear sense in which it is not a sham: Deep Throat attributes genuine choices to its fictional heroine. There is also the real Linda to consider: and here too the film may have attributed real choices, affirming the autonomy of the real Linda. It claimed to be ‘introducing Linda Lovelace as herself ’ (so the billing went), and in so doing, it depicted not only the fictional but the real Linda as acting autonomously. Extending further afield, there are women more generally to consider: and here again, if Linda is an iconic ‘Liberated Woman’, standing for all women, the film in a certain sense affirms the autonomy of women in the wider world.

Do these affirmations of autonomy (of the fictional Linda, the real Linda, or other women) settle the question of whether the film affirms autonomy tout court? No. There is scope for treatment that at once denies and affirms autonomy, as we have just seen. Paternalistic attitudes may deny autonomy through non-attribution, without denying autonomy through violation. Sadistic rape may affirm autonomy through attribution, while denying autonomy through violation. In general, autonomy affirmation is compatible with autonomy denial, once the plurality of possible modes of treatment is recognized. Even if Deep Throat were to affirm (p. 252) women's autonomy in all the above ways, that would not settle whether it affirms autonomy tout court.

The film does not affirm autonomy tout court. First, the claim that the film affirmed the real Linda's autonomy, even if true, requires instant qualification: for the film also denied her autonomy, as documented in her own testimony, Ordeal, and as more or less admitted by her one-time husband and pimp, Charles Traynor. Autonomy was attributed to her in the film, but denied her in life. In a remarkable set of televised interviews, Traynor admitted the autonomy denial: ‘She was pretty dumb. So everything she did, she had to be told how to do it, when to do it, and why she was doing it, and how to dress, and it just kinda rolled along like that, y'know…it was always a matter of telling her what to do.’23 What she was told to do was (among other things) to deny that she was being told what to do. The message was ‘Look as though it's your choice—or else’. Autonomy sells. Linda couldn't choose, but was more saleable if she looked like she could. She was a useful object, more useful if it looked like a subject. There is an odd sense in which her autonomy was itself commodified, thwarted in life but exaggerated, in fiction, for its cash value.

The film's autonomy attribution served autonomy violation. According to Linda's testimony in Ordeal, it took violence, rape, and death threats to make her play her role. Traynor admitted the violent relationship:

I was the dominant figure, she was a submissive figure, so if it reached the point where dominance had to take over, then dominance took over….If you argue to a point and somebody keeps pushing you, you know, fists are bound to fly. I don't mind somebody putting in their two cents worth, but I don't want them to argue with me to the point where I get upset or violently upset, and…yeah, that happened, on occasion….I think she didn't enjoy us getting in arguments with each other, but they say if you don't want me to get into an argument, don't argue with me.

Her interviews with the press, scripted by Traynor, convey not just violation of autonomy, but other kinds of autonomy denial, perhaps destruction or stifling of autonomy. Linda later said of those interviews:

I was just like a robot, I was told what to say and I said it, because if I didn't I was beaten, brutally.

Traynor admitted the scripting:

I schooled her on what to say. Always sound sexy. Always look cute. Wink at the camera. Wink at the interviewer …Always be titillating. You'd rather be having sex than doing anything. Y'know, it was just schooling, teaching her what to say, how to say it, and when to say it.

This is autonomy denial, and silencing too. What she was told to say was (among other things) to deny that she was being told what to say. Linda's own voice is (p. 253) silenced, in the scripted interviews, and silenced in a different way later on when her own testimony about abuse, in Ordeal, was sold as pornography.24

Finally, there is the autonomy of other women to think about. Linda is not just a woman, but woman, ‘Liberated Woman in her most extreme form, taking life and sex on her own terms’: there is autonomy attribution here, a vision of what autonomy is, not just for Linda, but for women in the wider world. But it can be argued that this autonomy affirmation serves autonomy denial, a false vision of autonomy being, after all, among the most potent enemies of autonomy. According to MacKinnon, and to testimony at the Minneapolis Hearings (see Minneapolis City Council 1988), the film legitimized a series of real-life autonomy violations, provoking an increase in throat rape (with associated suffocation), and an increase in unwanted and sometimes coercive attempts at throat sex. In affirming women's autonomy one way, and identifying that autonomy with sexual freedom, Deep Throat-style, it legitimized autonomy denial a different way, when the pornographer's image of women's choices was used to thwart real women's choices.

Some pornography, I conclude, might objectify even as it affirms autonomy: indeed, it might objectify through its autonomy affirmation, the way it objectifies depending on the distinctive way it affirms autonomy. The autonomy affirmation in Deep Throat (of a fictional Linda, a real Linda, or real women elsewhere) is one that serves autonomy denial (of a real Linda, and real women elsewhere). These denials of autonomy—the violations, silencings, or stiflings of autonomy—depend, substantially, on the affirmation of autonomy. That attribution of sexual autonomy to Linda was a structural feature of her oppressive circumstances, making abuse easier, hiding it, and hindering escape. That attribution of sexual autonomy to an iconic Liberated Woman, and thereby to other women, likewise facilitated violation of at least some other women's autonomy. Now whether Deep Throat is a typical or significant sample of autonomy-affirming pornography is a question I shall not, here, take time to address. But it is enough to suggest, I think, that pornography's way of affirming women's autonomy could, at least sometimes, be a way of denying women's autonomy; and there would be no paradox, at all, about that.

4. Concluding Regrets and Hopes

In arguing that feminism can illuminate philosophy and vice versa, it's best on the whole to try to show rather than tell—best, but also hardest, and I don't pretend to (p. 254) have succeeded in conveying much more than a glimmer. This chapter represents a very local and partial perspective on a vast and luxuriant continent, whose landscapes and borders it has barely begun to make visible, as its author is painfully aware. And daunted hopes aside, there is yet more trouble brewing, for we have ended up on dangerous ground. If feminist critique of philosophy has something to it, there are risks for feminist thinkers who value a two-way conversation, and who draw upon philosophy in trying to understand the very problems that bother them as feminists. Astell saw Cartesian philosophy as part of the solution to the problem she faced as a feminist: its method overturned prejudice, and its metaphysics cast women and men alike as essentially things that think. But if more recent feminist critique is right, dualism is not the solution, but the problem. Are all such feminist efforts fraught with danger? Audre Lorde (1984) famously said that ‘the master's tools will never dismantle the master's house’—not quite true, as your average burglar or vandal could tell you, but the metaphor captures a methodological suspicion that is not easy to lay to rest.

Take that idea of ‘treating someone as an object’, for instance, initially elaborated by Kant, and developed by feminists wanting to understand the nature of women's oppression. For Kant himself, the idea rested on a dualism more profound than that of Descartes—an unfathomable division between the realm of determined phenomenal objects, and timeless, noumenally free selves. Treating someone as an object is failing to treat them as a subject, it is to make them, somehow, part of the determined phenomenal order. Most of us feel no need for such dizzying metaphysics. But those of us who more modestly suppose there is something wrong with treating people as things—with treating women as things—must face the kinds of questions Astell's work has faced, and that Parfit's work faced in James's discussion. Whence comes this deep conceptual division between subjects and objects, between persons and things? And whence comes the valorization of mentality and freedom, the denigration of the physical, implicit in the idea that there is something inherently wrong with denying someone's subjectivity, denying someone's autonomy, reducing someone to their body? In short: doesn't the charge of dualism apply as much to feminist work on objectification as it applies to anything else?

This is an uncomfortable place to end up, and little scope remains here for a proper answer. But there appears, at least, to be some possible room for give and take. One could accept the complaint in part. Kant's visceral horror at the prospect of any abandoning of autonomy is surely a mistake, and Nussbaum's positive evaluation of certain kinds of self-surrender (and thus of certain kinds of objectification?) is surely right. On the other hand, to have any bite, a charge of dualism had better be made with sensitivity, asking what exactly is so bad about this ‘dualism’, as manifested in this idea, in this way. Does the badness consist in the history, and how such dualism has been used against women in the past? Does the badness consist in what it does to women here and now? Or does it consist not so much in what it does, as in what it leaves out? I can't guess right now the answers to these questions; but my hunch is that it will take some careful philosophy to find them.


Anderson, Elizabeth (1995). ‘Feminist Epistemology: An Interpretation and Defence’. Hypatia, 10: 50–84.Find this resource:

Astell, Mary (1700). Reflections Upon Marriage. London: John Nutt. Repr. in Patricia Springborg (ed.), Astell's Political Writings. Cambridge: Cambridge University Press, 1996.Find this resource:

Atherton, Margaret (1993). ‘Cartesian Reason and Gendered Reason’, in Louise Antony and Charlotte Witt (eds.), A Mind of One's Own: Feminist Essays on Reason and Objectivity. Boulder, Colo.: Westview Press.Find this resource:

Baumeister, Roy (1997). Evil: Inside Human Violence and Cruelty. New York: W. H. Freeman.Find this resource:

Begemihl, B. (1999). Biological Exuberance: Animal Homosexuality and Natural Diversity. London: Profile.Find this resource:

Brison, Susan (1987). ‘Outliving Oneself: Trauma, Memory and Personal Identity’, in Diana Tietjens Meyers (ed.), Feminists Re-think the Self. Boulder, Colo.: Westview Press.Find this resource:

Butler, Judith (1990). Gender Trouble: Feminism and the Subversion of Identity. London: Routledge.Find this resource:

——(1997). Excitable Speech: A Politics of the Performative. London: Routledge.Find this resource:

Caulfield, Mina (1985). ‘Sexuality in Human Evolution: What Is “Natural” in Sex?’ Feminist Studies, 11: 343–63.Find this resource:

Davenport, W. (1977). ‘Sex in Cross-Cultural Perspective’, in F. Beach (ed.), Human Sexuality in Four Perspectives. Baltimore, Md.: Johns Hopkins University Press.Find this resource:

Descartes, René (1641). Meditations on First Philosophy, trans. John Cottingham. Cambridge: Cambridge University Press, 1986.Find this resource:

Dworkin, Andrea (1974). Woman Hating. New York: E.P.Dutton.Find this resource:

Feinberg, Joel (1985). Offense to Others. New York: Oxford University Press.Find this resource:

Flax, Jane (1983). ‘Political Philosophy and the Patriarchal Unconscious’, in Sandra Harding and Merrill Hintikka (eds.), Discovering Reality. Dordrecht: Reidel.Find this resource:

Gallup, Gordon, and Suarez, Susan (1983). ‘Optimal Reproductive Strategies for Bipedalism’. Journal of Human Evolution, 12: 193–6.Find this resource:

Gilligan, C. (1982). In a Different Voice. Cambridge, Mass.: Harvard University Press.Find this resource:

Harding, Sandra (1993). ‘Rethinking Standpoint Epistemology: What Is ‘Strong Objectivity?’, in L. Alcoff and E. Potter (eds.), Feminist Epistemologies. London: Routledge.Find this resource:

Hartsock, Nancy (1983). ‘The Feminist Standpoint Theory: Developing the Ground for a Specifically Feminist Historical Materialism’, in Sandra Harding and Meryl Hintikka (eds.), Discovering Reality. Dordrecht: Reidel.Find this resource:

Haslanger, Sally (1993). ‘On Being Objective and Being Objectified’, in Louise Antony and Charlotte Witt (eds.), A Mind of One's Own: Feminist Essays on Reason and Objectivity. Boulder, Colo.: Westview Press.Find this resource:

Herman, Barbara (1993). ‘Could it Be Worth Thinking About Kant on Sex and Marriage?’, in Louise Antony and Charlotte Witt (eds.), A Mind of One's Own: Feminist Essays on Reason and Objectivity. Boulder, Colo.: Westview Press.Find this resource:

Hite, S. (1976). The Hite Report. New York: Macmillan.Find this resource:

Hornsby, Jennifer (1995). ‘Disempowered Speech’. Philosophical Topics, 23: 127–47.Find this resource:

—— (2000). ‘Feminism in Philosophy of Language: Communicative Speech Acts’, in Miranda Fricker and Jennifer Hornsby (eds.), The Cambridge Companion to Feminism in Philosophy. Cambridge: Cambridge University Press.Find this resource:

—— and Langton, Rae (1998). ‘Free Speech and Illocution’. Legal Theory, 4: 21–37.Find this resource:

(p. 256) Hrdy, Sarah Blaffer (1999). The Woman that Never Evolved. Cambridge, Mass.: Harvard University Press.Find this resource:

Irigaray, Luce (1985). Speculum of the Other Woman, trans. Gillian Gill. Ithaca: Cornell University Press.Find this resource:

Jaggar, Alison (1983). Feminist Politics and Human Nature. Totowa, NJ: Rowman & Allanheld.Find this resource:

James, Susan (2000). ‘Feminism in Philosophy of Mind: The Question of Personal Identity’, in Miranda Fricker and Jennifer Hornsby (eds.), The Cambridge Companion to Feminism in Philosophy. Cambridge: Cambridge University Press.Find this resource:

Jones, Karen (1999). ‘Second Hand Moral Knowledge’. Journal of Philosophy, 96: 55–78.Find this resource:

Kant, Immanuel (1797). Groundwork of the Metaphysic of Morals, trans. Mary Gregor. Cambridge: Cambridge University Press, 1997.Find this resource:

——(1797). Doctrine of Virtue, trans. Mary Gregor. New York: Harper & Row, 1964.Find this resource:

Keller, Evelyn Fox (1993). Refiguring Life: Metaphors of Twentieth-Century Biology. New York: Columbia University Press.Find this resource:

Kinsey, A. C., Pomeroy, W. B., and Martin, C. E. (1948). Sexual Behaviour in the Human Male. Philadelphia: W. B. Saunders.Find this resource:

—— —— —— and Gebhard, P. H. (1953). Sexual Behaviour in the Human Female. Philadelphia: W. B. Saunders.Find this resource:

Langton, Rae (1993). ‘Speech Acts and Unspeakable Acts’. Philosophy and Public Affairs, 22: 293–330.Find this resource:

——(1995). ‘Sexual Solipsism’. Philosophical Topics, 23: 149–87.Find this resource:

——(2004). ‘Projection and Objectification’,in Brian Leiter (ed.), The Future for Philosophy. Oxford: Oxford University Press.Find this resource:

Lloyd, Elisabeth A. (1993). ‘Pre-theoretical Assumptions in Evolutionary Explanations of Female Sexuality’, Philosophical Studies, 69: 139–53.Find this resource:

——(2005). The Case of the Female Orgasm: Bias in the Science of Evolution. Cambridge, Mass.: Harvard University Press.Find this resource:

Lloyd, Genevieve (1984). The Man of Reason: ‘Male’ and ‘Female’ in Western Philosophy. London: Methuen.Find this resource:

Lorde, Audre (1984). Sister Outsider: Essays and Speeches. Sydney: Crossing Press.Find this resource:

McGowan, Mary Kate (2003). ‘Conversational Exercitives and the Force of Pornagraphy’. Philosophy and Public Affairs, 31: 155–89.Find this resource:

Mackenzie, Catriona, and Stoljar, Natalie (eds.) (2000). Relational Autonomy: Feminist Perspectives on Autonomy, Agency and the Social Self. Oxford: Oxford University Press.Find this resource:

MacKinnon, Catharine (1987). Feminism Unmodified. Cambridge, Mass.: Harvard University Press.Find this resource:

——(1991). Toward a Feminist Theory of the State. Cambridge, Mass.: Harvard University Press.Find this resource:

Masters, W. H., and Johnson, V. (1966). Human Sexual Response. Boston, Mass.: Little, Brown.Find this resource:

Minneapolis City Council (1988). Pornography and Sexual Violence: Evidence of the Links. London: Everywoman. Transcript of Public Hearings on Ordinances to Add Pornography as Discrimination Against Women, Committee on Government Operations, 12–13 Dec. 1983.Find this resource:

Morris, Desmond (1967). The Naked Ape. London: Cape.Find this resource:

Nussbaum, Martha (1995). ‘Objectification’. Philosophy and Public Affairs, 24: 249–91.Find this resource:

O'Neill, Onora (1990). ‘Between Consenting Adults’, in O'Neill, Constructions of Reason. Cambridge: Cambridge University Press.Find this resource:

(p. 257) Perry, Ruth (1986). The Celebrated Mary Astell. Chicago: University of Chicago Press.Find this resource:

Quinton, Anthony (1975). ‘The Soul’, in J. Perry (ed.), Personal Identity. Berkeley: University of California Press.Find this resource:

Sartre, Jean-Paul (1943). Being and Nothingness, trans. Hazel Barnes. London: Methuen.Find this resource:

Saul, Jennifer (2003). Feminism: Issues and Arguments. Oxford: Oxford University Press.Find this resource:

Strawson, P. F (1974). ‘Freedom and Resentment’, in Freedom and Resentment and Other Essays. London: Methuen.Find this resource:

Symons, Donald (1979). The Evolution of Human Sexuality. New York: Oxford University Press.Find this resource:

Wittgenstein, Ludwig (1958). Philosophical Investigations, trans. G. E. M. Anscombe. Oxford: Blackwell.Find this resource:


(1) The passage continues: ‘he was not Made for this, but if he hires himself out to such an Employment, he ought conscientiously to perform it’. For more about Astell, see Perry (1986). On feminism and Cartesian rationalism, see Atherton (1993).

(2) In identifying this dualism, James is drawing upon the work of many other feminist theorists, most notably, perhaps, Genevieve Lloyd (1984). The reader is referred to James's first footnote (2000: 45) for other important references.

(3) See also Anderson's contributions on feminist epistemology in the Stanford Encyclopedia of Philosophy, and Saul (2003: 256).

(4) Discussed in Saul (2003: 208–9). Saul also reviews studies which suggest that most of the alleged differences disappear when adjustments are made for subjects' educational achievements. (I might add that, personally, I've never been quite sure why Jake's response is supposed to express Justice rather than, say, Egoism.)

(5) For extended development of the argument, see Lloyd (2005). Lloyd's example is also an important topic for Saul in her (2003,ch. 8: ‘Feminism, Science and Bias’), where it receives an excellent discussion, and my description here is partly indebted to that of Saul. For discussion of a different kind of androcentrism in biology, with a focus on metaphors of control and agency in the gene, see Keller (1993).

(6) Lloyd likewise discusses the invisibility to theorists of most sexual activity among female primates (1993: 141–2); and, where visible, its devaluation, e.g. by a researcher of female orgasm among stumptailed macaques, whose recording equipment was triggered by the male's increasing heart rate. When reminded that for this species the vast majority of orgasms occurred during sex among females alone, the reply was that he was only interested in the important orgasms.

(7) Other possible biases: the assumption that sex must be heterosexual (homosexual male behaviour may have been similarly ignored); and the adaptationist assumption that traits must themselves have adaptive advantage, rather than being, for example, side-effects of traits that have such advantage. These other biases are both discussed by Lloyd and Saul.

(8) Feminists who criticize the developmental view: Hrdy (1999: 165); Caulfield (1985).

(9) These arguments are both developed in Langton (1993), drawing upon ideas implicit in work by Catharine MacKinnon. See also Mary Kate McGowan (2003). The idea that free speech might include freedom to perform illocutionary acts is also developed by Jennifer Hornsby, e.g. in Hornsby (1995) and Hornsby and Langton (1998).

(10) The idea acquired its own self-contained verb, to ‘objectify’, and was taken up again by Marx, in his analysis of the oppression of the workers; and it has been used, in various forms, by continental philosophers, none of whom I know as well as I should, and few of whom had a particular interest in the situation of women. There is also an additional epistemological notion of objectification, in which projective belief plays a central role: for some feminist developments, see e.g. Haslanger (1993) and Langton (2004).

(11) From the definition drafted by MacKinnon and Andrea Dworkin, see MacKinnon (1987: 176). Note that if pornography is defined as subordination, the possibility of non-subordinating pornography is ruled as a matter of terminology. I use a more vernacular sense of the term here.

(12) Feinberg adds, ‘the episodes had no butt at all except prudes and hypocrites. Most of us consumers managed to survive with our moral characters intact’.

(13) The words of a commentator quoted in a documentary made by Mark Kermode and Russell Leven, The Real Linda Lovelace, first broadcast on Channel 4 in the UK, 26 Sept. 2002. At the time of writing, further details are available at Channel 4's website, <>.

(14) She allows that treating as autonomous may be compatible with fungibility and boundary violation, factors perhaps salient for ideologies of gay male promiscuity, and practitioners of consensual sado-masochism.

(15) There is an implicit biconditional: one treats an adult human being as mere means if and only if one denies their autonomy. One half of the biconditional is implied by Nussbaum's view that ‘treating an item as autonomous seems to entail treating it as noninstrumental’, i.e. (by contraposition) instrumentality entails autonomy denial. The other half of the biconditional is implied by her view that ‘treating as autonomous [may] be a necessary feature of the noninstrumental use of adult human beings’, i.e. non-instrumentality entails treating as autonomous, or (by contraposition) autonomy denial entails instrumentality, for adult persons (1995: 260, 261). This claim that instrumentality and autonomy denial are mutually entailing (for adult persons) is surely too strong. In terms of the distinctions made below, instrumentality entails autonomy denial construed as autonomy violation, it does not entail autonomy denial construed as non-attribution of autonomy. A liar or rapist may treat someone as a mere means, while attributing autonomy. In the other direction, there can be autonomy denial of either sort without instrumentality. A paternalistic doctor does not use the patient as a mere tool, or a means to his ends. When someone's autonomy is violated ‘for their own good’, this is not a use of the person. Nussbaum's examples suggest she wishes ‘autonomy denial’ to cover non-attribution and violation of autonomy, but neither of these have quite the intimate conceptual link with instrumentality she is after.

(16) She argues that, in a manner sensitive to context, in background conditions of equality, a mutual ‘denial of autonomy’ may be valuable, and she offers as an example the sexual self-surrender described by D. H. Lawrence in The Rainbow.

(17) Supposing too that we had an uncontested vision of what autonomy might be! For some feminist debate about this, see e.g. Mackenzie and Stoljar (2000).

(18) For a related discussion of the relationship between using someone and failing to treat them as a person, see O'Neill (1990: 105): ‘Making another into a tool or instrument in my project is one way of failing to treat that other as a person; but only one way.’

(19) Strawson would not wish to reduce such attitudes to, for example, metaphysical beliefs about whether the person is free or autonomous.

(20) I assume there are other kinds of rape in which the woman's unwillingness is incidental, not included in the content of what is desired; or even, as in negligent rape, unnoticed.

(21) That means, incidentally, that there is an important difference between the latter sexual example and those typically described by Kant, which involve autonomy violation in the service of some other purpose (for example, gaining money), achievable innocently by other means. Sadistic rape violates autonomy, not in the service of some other purpose (for example, achieving sexual pleasure), but partly for its own sake: autonomy violation partly constitutes the agent's purpose. For a discussion of a related contrast in the social psychology literature, see Baumeister (1997, chs. 4: ‘Greed, Lust, Ambition: Evil as a Means to an End’, and 7: ‘Can Evil Be Fun? The Joy of Hurting’), a psychological study of, roughly, a contrast between evil as means and as end.

(22) More thought is needed about how these interact, the relevance of repetition and pervasiveness, and institutional support; whether repeated and widespread autonomy violations may, for example, destroy autonomy; how a one-off autonomy violation (e.g. being lied to once) is vastly different from a systematic pattern that requires institutional support (e.g. slavery).

(23) This and the later quotations from Traynor and Marchiano are from interviews in the documentary made by Kermode and Leven, The Real Linda Lovelace.

(24) Ordeal was sold as pornography: this example of illocutionary disablement is discussed in Langton (1993), where a distinction is made between locutionary silencing due to threats (e.g. the scripted interviews) and illocutionary silencing (e.g. of Ordeal).