Objectification and Domination

  • Aleksy Tarasenko-Struc (Indiana University)


I resolve a tension between two prominent strands of feminist social critique. On the first, the domination of women consists largely in their objectification, and the objectifying character of such domination primarily explains why it is wrong. On the second, some salient forms of domination have a distinctively intersubjective dimension that makes them crucially unlike our standard modes of relating to objects. Yet in that case, how could characterizing these acts as objectifying capture why they are wrong? Focusing on domination that seeks recognition from the subordinate, I argue that each strand contains half the truth, weaving these together. The first is correct to point out that the concept of objectification is necessary for capturing the wrong of recognition-seeking domination, and that acts can be objectifying and ‘subjectifying’ all at once. The second strand is right to insist that domination of this sort has an irreducibly subjectifying character in light of which the concept of objectification is insufficient for clarifying its wrongness. The general lesson is that an account of the wrong of recognition-seeking domination is adequate only if it does justice to both its objectifying and subjectifying aspects.

How to Cite:

Tarasenko-Struc, A., (2021) “Objectification and Domination”, Ergo an Open Access Journal of Philosophy 8: 14. doi:



Published on
13 Dec 2021
Peer Reviewed
Christ, it was better than
hunting bear
which don’t know why
you want him dead.

—Robert Hayden1

1. The Concept of Objectification: Necessary yet Insufficient

Feminists have traditionally taken there to be an important relationship between the domination of women and their objectification—that is, their being treated or seen as things or objects. According to the radical feminists Andrea Dworkin and Catharine MacKinnon, the sexual objectification of women by men is the primary mechanism of women’s subordination: men regard women as suitable for sexual use and exercise socially sanctioned control over them so as to make them conform to men’s view of them; in so doing, men make women into objects, wrongly harming them by denying them their humanity.2 More recently, Martha Nussbaum and Rae Langton have advanced general theories of objectification animated by a sensitivity to latent connections between earlier feminist accounts of sexual objectification and complementary discussions within the Kantian and Marxian traditions.3 Despite their differences, these feminist philosophers agree on the following point, both among themselves and with their radical-feminist forebears: the domination of women is objectifying, and its objectifying character plays an ineliminable role in shaping its distinctive moral profile. They have argued, variously, that sadistic rape (Nussbaum 1995: 252–53, 279–81; Langton 2001a: 234), sadism generally (Langton 2001b: 336–42), and silencing (Langton 2001a: 229) objectify, and that acts such as these are morally objectionable because they involve treating a person as an object.

The thematization of this alleged resemblance between the domination of persons and our ways of relating to objects sits awkwardly with another strand of feminist social critique, which tends, by contrast, to emphasize the interpersonal or intersubjective core of much domination (Manne 2016; 2018: ch. 5; Mikkola 2016: ch. 6). In Down Girl, for example, Kate Manne discusses the case of Elliot Rodger, the perpetrator of the Isla Vista killings, who outlined his motives for the planned mass murder in an infamous video. Rodger clarified that he was motivated by a sense of vengeful aggrievement directed primarily against conventionally attractive women, who, in his view, had long refused to provide him with the love, sex, and affection that he so clearly deserved. Manne comments:

What is striking about these sentiments is that they not only presuppose but seem to hinge on the women’s presumed humanity … . Rodger ascribes to these women subjectivity, preferences, and a capacity to form deep emotional attachments (love, as well as affection). And he attributes to them agency, autonomy, and the capacity to be addressed by him. (2018: 149–50)

We can agree with Manne that it’s significant that Rodger’s intended targets, the women of the Alpha Phi Sorority House, were also his intended addressees and that, at one point, ‘[h]e was … speaking to, not of, them, second-personally’ (2018: 149). For that reason, Rodger’s acts seem more ‘subjectifying’ or ‘humanizing’ than objectifying. For one, as Manne points out, his sentiments presuppose an awareness of his victims as human beings or persons—an awareness not just that they are persons but more specifically that they are subjects with minds and a range of capacities that are plausibly constitutive of personhood, such as autonomy, understanding, and the capacity to form personal relationships. Even more strikingly, not only does Rodger appear to be aware that the women whom he vowed to attack are bona fide bearers of these capacities, he also seems to care about their attitudes, especially their attitudes toward him. Indeed, the fact that he sought to communicate his feelings to them at all may indicate that he even cares non-instrumentally about his victims’ view of him. Domination of this kind therefore begins to look like a way of treating a person as a subject.

The problem is that these two descriptions of gender-based domination—that it is both objectifying and ‘subjectifying’—appear to be in tension with one another, even if they are not strictly inconsistent. Many acts of domination (such as mass murder, torture, and sadistic rape) are intelligible only on the assumption that the dominator believes that the subordinate is a person and that he is aware of, and even concerned with, the subordinate’s experiences and feelings. But if all that is true of the dominator, how could he possibly count as treating the subordinate as a thing? The more we come to appreciate the deeply intersubjective character of domination, in other words, the less it appears to resemble our normal orientation toward mindless objects, and the more tenuous, apparently, is the case for comparing acts of domination to our treatment of things. We seem justified in suspecting that the concept of objectification is inapplicable to acts of this kind—perhaps to any act—and, thus, that it is incapable of clarifying why they are wrong.

In this essay, I show that each of these views contains an important insight about the relationship between certain forms of domination and our treatment of things. My discussion will be restricted to the kind of interpersonal domination that Rodger sought for himself, which is also the kind that seems to present an especially hard case for the idea that gender-based domination is wrongly objectifying: what I call ‘recognition-seeking domination’. So, after getting the target phenomenon in view (§2), I argue that the concept of objectification is not only applicable to acts of recognition-seeking domination but that reference to their objectifying features is, in fact, necessary for explaining why they are wrong (§3). Thus, I propose to resolve the tension just identified by locating a familiar, apt, and morally relevant sense in which we can treat or regard others as objects—one that does not exclude the possibility of, at the same time, treating or regarding them as subjects or even persons. In partial vindication of the ‘humanizing’ conception of gender-based domination, however, I contend that the concept of objectification is also insufficient for capturing the special wrong of recognition-seeking domination (§4). Such acts are wrong at least partly in virtue of their irreducibly ‘subjectifying’ features. The general lesson, then, is that an account of the wrong of recognition-seeking domination is adequate only if it does justice to both its objectifying and its ‘subjectifying’ or ‘humanizing’ aspects.

2. Recognition-Seeking Domination as Essentially Communicative

To properly characterize the main tension between the two strands of feminist social critique, I begin by presenting a view that is representative of the first, objectification-focused strand: the theory of Martha Nussbaum and Rae Langton (hereafter: the ‘Nussbaum-Langton Theory’). This view makes for a fitting starting point not only because it is probably the most well-known, most influential, and most complete theory of the phenomenon, but also because, in light of its commitments, it appears to most clearly generate the tension with which this essay is concerned.

The Nussbaum-Langton Theory is a view primarily about the concept of objectification. It is defined by three key claims along with a proposed catalogue of forms of objectification. The first claim is that our idea of objectification is a cluster concept that is applicable to a wide and assorted range of activities. Nussbaum proposes that these seven related notions are invoked by our use of the concept, each of them corresponding to a distinct kind of objectifying act or attitude:

  1. Instrumentality: treating someone as a tool for one’s purposes

  2. Denial of autonomy: treating someone as lacking in autonomy

  3. Inertness: treating someone as lacking in agency (or activity)

  4. Fungibility: treating someone as replaceable

  5. Violability: treating someone as lacking in boundary integrity

  6. Ownership: treating someone as owned, bought, sold

  7. Denial of subjectivity: treating someone as something whose thoughts and feelings may be discounted4

What do these modes of objectification have in common? This brings us to the second key claim. Nussbaum claims that each entry in her catalogue is ‘a feature of our treatment of things, though of course we do not treat all things as objects in all of these ways’ (1995: 257). Langton is even more explicit in her embrace of this approach, aspiring to extract a general theory of objectification from our everyday idea of an object and from an associated contrast between persons and things (2001b: 163). So, the second key claim of the Nussbaum-Langton Theory is that there is a conceptual connection between objectifying treatment and our standard dealings with mindless objects. Specifically, the boundaries of the concept of objectification are fixed by the fact that the forms of treatment falling under it bear a (morally significant) likeness to our characteristic ways of relating to things. That is what unifies the above modes of objectification into a single category.

The third and final plank of the Nussbaum-Langton Theory is that the moral valence of an objectifying act—whether it is permissible or wrong—is highly sensitive to the context in which it is performed. In particular, the fact that an act is objectifying (in some respect) does not always make it wrong; there are morally neutral and even positive modes of treating persons as things, in addition to wrongful ones.5 One of Nussbaum’s examples of benign objectification is that of using one’s lover’s stomach as a pillow while lying in bed. As it involves the use of a person’s body, this act is, in fact, objectifying. Yet it is morally unobjectionable as long as certain background conditions are met—for instance, as long as the arrangement is consensual, not harmful to the lover, and part of an otherwise non-instrumentalizing relationship. The third claim implies that, strictly speaking, the objectifying feature of an act does not always make it wrong on its own; it is the act’s objectifying feature plus the absence of the relevant background conditions. Still, the fact that an act is objectifying, in any of the above respects, does make it at least prima facie wrong, morally risky—wrong unless these background conditions obtain.

So, on the Nussbaum-Langton Theory, (1) the concept of objectification is a cluster concept, (2) this concept is applicable to acts and attitudes resembling our characteristic ways of relating to mindless objects, yet (3) to apply the concept to an act/attitude is not to thereby register that it is wrong. Rather, an objectifying act is prima facie wrong, and whether it is permissible or wrong depends crucially on whether or not certain background conditions are met.

Why should the kind of domination that Elliot Rodger sought for himself be thought to present a hard case for the Nussbaum-Langton Theory? To answer this question, we need to look more closely at Rodger’s acts, paying special attention to the structure of his motives. Notice, first, that it is natural to call these acts of domination, in one common, serviceable sense of the term: acts that consist in the morally objectionable application of physical or psychological force to an individual, in an exercise of superior power over his will and/or emotions.6 Now acts of this kind are sometimes performed as a means to some end that is independent of the acts themselves.7 Suppose that, weapon in hand, you issue a threat to me (‘Your money or your life!’) in order to compel me to hand my wallet over to you, and that you do so simply in order to increase your purchasing power. Your act of threatening me, brandishing the weapon, would then be a means—maybe one among others—of bringing it about that you have the money. For the relation between the act (your threatening me, brandishing a weapon) and its end (your obtaining the money) is an extrinsic, causal one—a relation between distinct existences.

Based on his spoken and written testimony, it is reasonably clear that Rodger did not kill and terrorize people for the sake of some end that was independent, to that extent, of the acts themselves. He acted so as to get revenge, but his acts were not, strictly speaking, a means to—not merely causally connected with—some further end: revenge. Rather, terrorizing and killing his victims constituted or realized his revenge. And his desire for revenge was intrinsic, not purely instrumental: he wanted revenge, apparently, for its own sake.8

But Rodger’s acts of killing were non-instrumental in a further respect: he seemed to be motivated by a non-instrumental desire for others to hold certain attitudes toward him. For he also wanted his victims (and bystanders) to recognize his acts as exercises of power and to have certain attitudes toward him—such as fear and awe—in the wake of that recognition. He would not have been fully satisfied had he simply planted a bomb, undetected, in the Alpha Phi Sorority House and detonated it remotely under a cloud of anonymity. Much of the language of Rodger’s manifesto, as well as the very fact that he created and shared it in the first place, makes the recognition-seeking character of the killings especially evident: ‘You will finally see that I am in truth the superior one. The true alpha male,’ he laughs, ‘now I will be a god compared to you.’ By his own lights, he would have succeeded in achieving his final end—revenge—only if his victims (and bystanders) came to hold these attitudes. And their holding the attitudes in question is a constituent, or specific realization, of that end, not a means of achieving it; he did not want them to hold these attitudes because their doing so brought about a separate end of his.

I will say that Rodger’s acts of killing were acts of recognition-seeking domination: acts of domination motivated by a non-instrumental desire for the subordinate to hold certain attitudes toward the dominator himself, to see the dominator in a particular light.9 As I am using the term, then, recognition-seeking domination (hereafter: ‘RSD’) is essentially second-personal or dyadic in that the dominator is motivated to act so as to gain recognition from the subordinate herself. Officially, then, my definition excludes forms of domination in the general neighborhood that are also recognition-seeking, such as acts of domination performed on a subordinate so as to gain recognition from a third party—white Americans lynching African-Americans in the Jim Crow South, for example.10 While these kinds of acts are indeed related to RSD, as I define it, my focus in this paper will largely be on the second-personal variety of the phenomenon, as this variety seems to most clearly threaten the very idea of objectification. Before I explain why, I want to consider what else might be included in the category of interest.

The structure of Rodger’s motives also belongs to many typical acts of sadism and abuse. Consider, first, one of the purest forms of sadism—torture for its own sake.11 The torture of political prisoners, for example, is a particularly terrible demonstration of the torturer’s absolute power over the victim, not uncommonly sought for its own sake rather than for the sake of some further end. Many of those who practice torture non-instrumentally, in these contexts, do not simply want to make another person physically suffer for its own sake. These torturers also (non-instrumentally) want to be recognized by the victim as making him suffer, and to be seen by him as more powerful or of a higher status relative to him, as confirmed by their ability to inflict suffering on him with impunity. Moreover, these torturers want the victim to acknowledge their superior power to them, nonverbally (through cries and groans) and/or verbally (through pleas and lamentations).12 The torturers whom I have in mind seek not only to prove their absolute power by inflicting physical suffering on their victim but also to prove it to the victim and to procure the victim’s acknowledgment of this power, for its own sake. A similar motivational structure belongs to less extreme instances of sadism, such as bullying and milder forms of revenge, insofar as these are done from a non-instrumental motive. For like torture for its own sake, these activities are undertaken in order to get their victims to recognize, and acknowledge, the superior power of the one engaging in them, for its own sake.

Yet even some acts that do not appear to have been done from a sadistic motive count as acts of RSD, too. Here is Sandra Lee Bartky’s evocative description of an imaginary (but, I expect, for many people perfectly familiar) episode of street harassment:

While it is true that for these men I am nothing but, let us say, a ‘nice piece of ass,’ there is more involved in this encounter than their mere fragmented perception of me. They could, after all, have enjoyed me in silence. … But I must be made to know that I am a ‘nice piece of ass’: I must be made to see myself as they see me. There is an element of compulsion in this encounter, in this being-made-to-be-aware of one’s own flesh; like being made to apologize, it is humiliating. It is unclear what role is played by sexual arousal or even sexual connoisseurship in encounters like these. What I describe seems less the spontaneous expression of a healthy eroticism than a ritual of subjugation.13

What is striking about Bartky’s encounter is that her harassers’ treatment of her appears to have an interactive, ‘I-You’ structure.14 For their aim appears to be that she recognize, and somehow acknowledge, their superior power or status—power that they exercise in seeking to compel her to relate to them in ways befitting of a ‘nice piece of ass’ (coyly? lustfully? silently?). Bartky’s harassers are trying to effect a ‘meeting of minds’, of a sort: they succeed in their efforts only if she comes to see herself through their gaze and regards herself as they regard her—as an object for sexual use, before actual or potential users. Street harassment of this kind seeks a relation of mutual recognition between the two parties, then, in that the aim of the interaction is a reciprocal acknowledgment of these roles relative to each other. If my interpretation is correct, Bartky’s harassment does constitute RSD, as it was motivated by a non-instrumental desire that she see them in a certain light and that she respond accordingly.

Rodger, too, sought a twisted sort of meeting of the minds. He wanted to be feared by his victims, and indeed to be regarded by them—in the nexus of his ‘righteous’ anger and their terror—as a powerful, vengeful God slaughtering the pathetic herd of humanity. He sought to occupy this nexus of attitudes with his victims for its own sake, not just because doing so facilitated the realization of separate ends. That suggests that he cared non-instrumentally about his victims’ perspectives—their thoughts and feelings, at least about him. What Manne’s discussion of Rodger alerts us to is the fact that acts of RSD are essentially communicative, in the broad sense. I take communication with other subjects to include not just speech but also the nonlinguistic expression of emotions and other attitudes—centrally, the expression of attitudes to others. RSD’s communicative dimension, then, consists in (a) the fact that its agent’s end is to establish a relation of mutual recognition between himself and the other individual, in the form of a nexus of attitudes shared among agent and patient, and in (b) the fact that the agent pursues this intersubjective relation for its own sake, not for the sake of some independent end.15

Now, however, our original tension reappears with renewed force. How, exactly, could acts such as Rodger’s be wrongly objectifying, resembling our treatment of things, if they are motivated by a desire to stand in relations of mutual recognition with others for its own sake? The challenge is especially acute for the Nussbaum-Langton Theory due to its insistence that the objectifying treatment of persons is conceptually connected to our characteristic treatment of mere things. For RSD seems to be indisputably subjectifying: because of its dyadic, second-personal character, it seems to constitute a way of treating a subject as a subject. So, given that domination of this kind appears to be subjectifying in one key respect, we might begin to wonder whether it is objectifying at all—or, if it is, whether that alleged fact has any significant place in an explanation of what makes it wrong. Indeed, skepticism about the viability of the concept of objectification lurks in the background. Perhaps the concept is incoherent. Perhaps there is simply no ‘there’ there, and protests against being ‘treated as a thing’ are nothing more than empty rhetorical devices. The idiom of objectification would then be comparable to colloquial objections to being ‘treated like dirt’ or ‘treated like trash’—interjections that expresses generic moral outrage without picking out a distinctive kind of morally objectionable conduct.

3. Recognition-Seeking Domination as Wrongly Objectifying

To determine whether the puzzle is real or only apparent, we will need answers to two questions. First, are acts of RSD objectifying? Do they count as treating persons as things? Second, even supposing that these acts were objectifying, would their allegedly objectifying features explain why they are wrong? And if they would be explanatory to some extent, how much explanatory work would they do, exactly? One strong position is that appealing to RSD’s objectifying features never, or anyway rarely, explains why it is wrong—perhaps because no such features exist in the first place. A moderate position, on the other hand, is that while reference to these features does contribute to an explanation of the wrongness of these acts, its explanatory power is only partial.16 In this section and the next, I rebut the strong position and argue in favor of a moderate position. In my view, while the concept of objectification is limited in its capacity to explain the wrongness of RSD, it is nevertheless required to fully capture the phenomenon’s moral texture. So, the arguments of this essay function, effectively, as a partial, qualified defense of the Nussbaum-Langton Theory and as an exploration of that theory’s limitations and scope.17

3.1. The Very Idea of Objectification: Two Challenges

One recent defender of the strong position is Mari Mikkola, who provides grounds for doubting that the concept of objectification is capable of doing important normative work, particularly in feminist normative theorizing. In a rebuttal to John Gardner and Stephen Shute’s (2007: 16) view that rape is wrong because it is the ‘sheer use’—hence, the dehumanization—of a person, Mikkola cites martial rape, the practice of rape as a weapon of war, as a noteworthy counterexample:

[M]artial rapists do not appear to view the affected women as thing-like. They are viewed as persons with goals, life plans, and a desire for well-being; martial rape as a practice is aimed at precisely thwarting these aspects of women’s lives, thereby destroying whole communities and making it extremely hard for people to rebuild them. The victims of the practice are not treated as inert things simply to be destroyed, like dwellings. … the kind of mere use in martial rape that violates others presupposes that the affected subjects are persons with life plans and particular social roles. And the wrong committed by the rapist is precisely aimed at exploiting that in order to violate the affected persons.18

Mikkola’s critique comprises two points. The first point, expressed in the passage above, is that the practice of martial rape and other forms of dehumanization cannot be wrongly objectifying because its perpetrators do not regard their victims as things or as thing-like. Martial rapists are manifestly not in the epistemic situation of, say, the man who mistook his wife for a hat in Oliver Sacks’s famous case study, nor are other dehumanizers. On the contrary, martial rapists do regard their victims as human beings; they must do so, in fact, because their aim assumes a view of the victims as persons, not things. On that basis, Mikkola concludes that martial rape is not wrong in virtue of its instrumentalizing character or its objectifying character more generally.

Mikkola’s second point is related: treatment such as martial rape cannot be wrongly objectifying because its victim is not literally turned into a thing by that sort of treatment. Indeed, if we thought of this treatment as literally objectifying, she claims, it would be hard for us ‘to make sense of resistance. On the face of it, things cannot resist what is done to them or fight back. … this would make social change, emancipation, and empowerment simply impossible’ (Mikkola 2016: 177, see also 176, 148, fn. 2).

Mikkola is not the first philosopher to question the suitability of the idiom of objectification on these grounds. In a similar spirit, Ann Cahill charges theories of objectification (particularly the Nussbaum-Langton Theory) with supposing—implausibly—’that the objectified woman is rendered a non-person, lacking in autonomy and subjectivity’ (2011: 27). In defense of this charge, Cahill notes that in objectifying images of women ‘the objectified woman is not constituted … as an object, but rather as a particular kind of subject: one who has (among other characteristics) little to no concern for the respect she is due from herself and others on account of her personhood.’ ‘Because of this persistence of (at least the impression of) subjectivity,’ however, she claims that ‘the notion of objectification cannot quite illuminate the ethical wrongs in these images of women’ (2011: 28, cf. 31). Later in the book, this critique is extended with a view to challenging similar accounts of sexual violence (2011: 127–42, esp. 133–37). There Cahill denies that rape is an act of objectification, for ‘the [rape] victim is never just an object to the assailant; her pain and anguish are not only recognized, but in fact actively sought, and they must be the kind of pain and anguish which a human person experiences.’19 So, we find both points here, too: allegedly objectifying acts are never, in fact, wrongly objectifying, both because they do not literally make their patients into things and because agents do not regard patients as things, strictly speaking.

Taken together, the two points voiced by both Cahill and Mikkola constitute challenges for the Nussbaum-Langton Theory and any similar conception of objectification. I will now argue, nevertheless, that the challenges may be met, and that their apparent force derives from a set of assumptions that the theory does not accept and that are highly implausible on independent grounds. Thus, my aim is first to show that acts of RSD are indeed objectifying in a morally relevant sense. Then, with the challenges disarmed, I will contend that the concept of objectification is necessary for capturing the wrongness of domination of this sort. RSD is wrong partly because it is objectifying in key respects. One caveat before I begin. In seeking to establish this conclusion, I focus on martial rape as an example of RSD. I do so not under the assumption that all instances of the practice are recognition-seeking but because one characteristic motive of martial rapists is to punish and humiliate the victim; in such cases, the act does have a second-personal element, even if motives for martial rape vary widely and may sometimes be banal.

3.2. Disarming the Challenges: The Negative Argument

Start with the second challenge: that martial rape and similar acts cannot be objectifying because they do not literally make their victims into (mindless) objects—except, perhaps, in a subset of cases, in the uninteresting sense of killing their victims. I know of no philosopher who accepts as a condition on objectifying treatment that it effect the transformation in question, and it is implausible, in any case; for one, that condition would rule out sexually objectifying behavior from the outset.20 Why insist that only an act of metaphysical magic count as treating someone as an object?

Fortunately, there is a less literal but nonetheless apt sense in which persons may be ‘made into things’. Persons may be made into things in the sense that they may be treated as it is characteristic and fitting to treat things, in light of things’ nature and value—or, perhaps, in the sense that they may be treated in ways that are analogous, in some key respect(s), to the ways that things are (characteristically, fittingly) treated. We may also add that to count as objectifying, the treatment in question must mark out or define our concept of a kind of object, as to-be-treated-in-the-relevant-way. Treating someone as a tool for our purposes appears to fit well with this formulation, as it crucially resembles our standard, appropriate treatment of a certain kind of object—a tool—and our concept of a tool is defined by the fact that a tool is to-be-used.

While this formulation does not quite constitute a complete definition of objectifying treatment, I believe that it tracks our intuitions concerning what counts as treatment of this kind. Suppose, for example, that you induce me, through blackmail, to crouch down on the floor so that you can sit on my body as you would on a chair. Here it seems perfectly appropriate to say that you have treated me as a thing or made me into a thing. For you have used me as you would an object—a chair.21 Notice, too, that in the case just described, it also seems perfectly appropriate to say that you are treating me as an object regardless of your motives. Imagine that in an act of RSD you use me as you would a chair so as to make me suffer for its own sake. Even then, you still count as treating me as a thing, it seems—again, despite the obvious fact that you do not literally make me into a chair, or, indeed, into any other object. To treat someone as an object, we need not literally transform her into an object. So much for the second challenge.

Let’s unpack the first challenge: that martial rape does not count as objectifying instrumentalization because martial rapists do not see their victims as things or as thing-like. In Mikkola’s view, these rapists do not count as treating their victims as things because they regard them as ‘persons with goals, life plans, and a desire for well-being,’ and she seems to believe that this view of the women excludes the possibility of seeing them as things. Cahill makes a similar claim about objectified women more generally, including victims of rape. Hence, Cahill and Mikkola both appear to be committed to the following thesis:

Seeing-Treating Link: A treats B as an object only if A sees B as an object.

How credible is Seeing-Treating Link? While the thesis certainly holds some intuitive weight, it nevertheless seems to founder on the possibility of inadvertently objectifying treatment.22 For instance, a man may fully respect a woman as an end-in-herself yet make a remark that—given the broader social context—sexually objectifies her, contrary to his own intentions. We would still classify his treatment of her as objectifying, I submit, despite the fact that—ex hypothesi—he does not see her as an object but as an independent, autonomous subject. (More generally, a person can do wrong despite having unobjectionable attitudes and motives.)

For the sake of argument, however, I am prepared to grant Seeing-Treating Link. More worrying, in my view, is that Cahill and Mikkola both seem to assume the following principles:

Exclusion Principle 1: A sees B as an object only if A does not believe either that B is a person or that B has some person-specific property/properties.

Exclusion Principle 2: A treats B as an object only if A does not care about B’s person-specific properties (e.g., B’s mental properties).23

Yet neither Nussbaum (1995: 279–81) nor Langton (2001a: 234; 2001b: 336–42) accepts these exclusion principles, as evinced by their explicit discussions of sadistic forms of objectification. In any case, these principles are simply dubious on their face. I will start with the state of seeing someone as a thing. Consider sexual objectification, particularly as it is embodied in the sexually objectifying gaze. When Romeo sexually objectifies Juliet in this way, the foreground of his experience of her is occupied by her bodily appearance and its various potentialities for satisfying him sexually, while her perspective—her thoughts, feelings, and aims—is relegated to the background of his attention.

To the extent that Romeo exhibits these perceptual saliences and patterns of attention, it is natural to claim that Romeo sees Juliet as a thing, in a familiar sense of the phrase: his overall experience of Juliet is centered on a property (or set of properties) of hers that is of a kind that is shared with mindless objects—such as her purely physical or instrumental properties—at the expense of attention to her person-specific properties. In that respect, seeing someone as a thing resembles one of our paradigmatic modes of regarding objects, whose physical and instrumental properties we do tend to focus on and whose value is normally grounded in such properties. Moreover, sexually objectifying others in this manner does appear to be prima facie wrong—minimally, because in taking up this sort of objectifying stance toward another person, we are at greater risk of treating that person in a way that is dangerously heedless of her perspective.24

At any rate, it is clear that Romeo could count as sexually objectifying Juliet, in this sense, yet not be in the grip of an ontological mistake; he need not (and cannot) harbor false metaphysical beliefs about her, to the effect that she is, in the most literal sense, a mindless object, like a sex doll. Seeing another person as a thing, then, is consistent with registering, on an intellectual level, that she is a person—and often assumes it, as the case of sexual objectification demonstrates. Thus, Exclusion Principle 1 is false.25 The same argument applies, mutatis mutandis, to treating someone as a thing. Treating a person as a thing is fully compatible with believing that she is, in fact, a person. Return to my earlier example. If you coerce me into allowing you to use me as a chair, you will certainly have to believe that I am a person with a range of person-specific properties, such as the power to speak and to understand the world. Nor do you need to believe that I lack any such property in order to be able to use me in this way. Finally, Exclusion Principle 2 is false as well. If you force me to allow you to sit on me to make me suffer for its own sake rather than for some further end, you have an intrinsic concern for at least some of my person-specific properties—namely, my suffering. Yet you still treat me as a thing in the sense that you still use me as you would an object—again, a chair.

So, we can count as seeing someone as an object even if we believe that she is a person with a full suite of person-specific properties. We can likewise count as treating someone as an object even if we care about some such properties of hers. There is no in-principle obstacle, then, to admitting that martial rapists regard their victims as objects and as a person at the same time, or in admitting that martial rapists treat or use their victims as objects. A martial rapist might see a particular victim as an object insofar as his attention, particularly at the time of action, is trained on such features of hers as her physical appearance and sexual functions, or perhaps her capacity to advance the military goals of the regime (through his rape of her). His taking this view of her is compatible with, and even presupposes, his holding the belief that she is a person.

A martial rapist can be said to treat his victim as an object even though he attends to her mind—her suffering and humiliation—and cares non-strategically about her mental properties; after all, he is moved by a non-instrumental desire to make her suffer or to thereby send her a humiliating message. For rape is instrumentalizing, and thus objectifying, by its very nature. It is the use of a person—of her body and sexual nature—as a tool for the rapist’s sexual gratification. Rape consists of the rapist appropriating the victim’s body, in its sexual capacity, as a masturbatory tool, using it as a sex aid for his own satisfaction. But martial rape is no different in this respect. The fact that it is perpetrated in such a sinister spirit does not disqualify it as objectifying treatment any more than your using me as a chair to make me suffer for its own sake disqualifies that act. These forms of treatment are instrumentalizing and thus objectifying. But martial rape is objectifying in other respects as well, for in addition to its instrumentalizing character, it involves overriding the victim’s capacity for choice (autonomy-denial), violation of her bodily integrity (violability), and disregard of her thoughts and feelings (subjectivity-denial).

These considerations suffice to disarm Cahill and Mikkola’s skeptical challenges. The concept of objectification applies to at least some forms of distinctively subjectifying domination. Clearly, though, my argument falls short of establishing that martial rape, much less RSD generally, is wrong because it is objectifying. So, Mikkola might still be correct in her analysis of martial rape—that it is not wrong in virtue of any objectifying feature(s). She is, admittedly, correct to think that martial rape is not wrong only on these grounds. The question is whether martial rape is wrong partly due to its objectifying features, at least. Is it wrong at least partly due to its instrumentalizing character, for instance? We may worry that even if martial rape is instrumentalizing, that fact nevertheless does not explain why it is wrong. The worry is even more pressing for the Nussbaum-Langton Theory, according to which instrumentalization or use of a person is not necessarily wrong and may even be positively good. If using people isn’t always wrong, then how, exactly, could it play a role in an account of an act’s wrongness?

3.3. Disarming the Challenges: The Positive Argument

My goal here is only to isolate the objectifying element in martial rape and to explain how that element contributes to the act’s wrongness, not to give a complete rendering of its moral profile.26 I will argue that one of the wrong-making features of martial rape—indeed, of rape generally—is its peculiar instrumentalization of the victim, although there are certainly others. Hence, a harm-based view of martial rape’s wrongness that makes no mention of its instrumentalizing character is incomplete, as is a corresponding consent-based view. I will discuss each type of view in turn.

Start with a harm-based view. On this sort of view, martial rape is wrong because it consists in, or involves, the infliction of severe physical, psychological, and relational harms on the rape victim, her family, and her community/group. Now it is undeniable that martial rape is harmful along these dimensions, that the harm suffered by victims is severe, and that this fact plays an essential role in its moral profile. I doubt, however, that we can fully understand the harm of martial rape in abstraction from its instrumentalizing character. The act is harmful at least partly in virtue of being instrumentalizing in the specific way that it is. Which way is that?

Before we can answer this question, we need a definition of instrumentalization or use. Recall that on the Nussbaum-Langton Theory the fact that an act falls into one or more categories of objectifying behavior implies that it is prima facie wrong—wrong absent relevant background conditions—rather than that it is wrong, period. On this theory, then, instrumentalization is prima facie wrong: there is a moral presumption against using someone, but the default moral objection to use may be defeated by other features of the circumstances of action, thereby rendering the act in question permissible or even morally good. With that in mind, A instrumentalizes or uses B (in my sense of these terms) in φ-ing if and only if, and because,

  1. A’s φ-ing involves acting on some property/properties of B’s, particularly B’s physical or mental capacities (or both);

  2. A φ-s so as to serve some (putative) interest, or satisfy some purpose, of A’s; and

  3. A φ-s with disregard for B’s perspective on his act, particularly B’s thoughts, feelings, and/or aims.27

Hence, instrumentalization (or use), in my sense, differs importantly from Kant’s notion of treating someone merely as a means (or what we might call mere use), which is wrong in itself. We count as instrumentalizing or using a bus driver, for instance, to the extent that we engage his labor and skills to serve an interest of ours (e.g., that of getting across town quickly and safely), while not attending much to his thoughts and feelings about the whole arrangement. Whether our doing so is mere use, and therefore wrongly objectifying, however, depends on the broader context of his employment—on, say, whether he works consensually, is adequately compensated for his labor, and so forth. While using a bus driver in this more innocuous way is arguably permissible, it is still prima facie wrong—wrong absent these background conditions.

With this definition of use in mind, I contend that the harm of rape partly consists in a kind of use: it is the harm of having one’s sexual nature used by another person (instrumentalization), in a context in which one’s consent is absent (autonomy-denial). Rape is wrong partly, yet crucially, for the same reason. In my view, rape is harmful not just because of the physical damage that it does to the person raped; martial rape is no exception, even though the physical damage that it brings is profound. And although rape is standardly traumatic and therefore psychologically damaging as well, we must understand the trauma in question as a subjective response to the objective violation expressed in the act of rape. In other words, to the extent that rape is psychologically harmful, it is harmful in virtue of what it means to be raped.

To begin to understand the relevant meaning(s) of an act of rape, I want to turn to Gardner and Shute’s sketchy but suggestive remark that ‘it is the social meaning of consensual sexual penetration which the rapist exploits by subverting it,’ and to their complementary picture of rape as the degradation of a paradigmatically intersubjective relation to another person—sex—into a subject-object relation.28 This picture needs to be filled in. Now I am not pretending that the following elaborations constitute a complete theory, or that they accurately represent Gardner and Shute’s views of the matter. I suggest that sex is an intersubjective relation because a/the constitutive aim of sexual union is the achievement of a kind of mutual recognition that is characteristically expressed through action directed at the body of another; somewhat less abstractly, a/the constitutive aim of sexual union is the joint expression of mutual sexual desire, of the partners for each other, in shared action affirming the other in their embodied subjectivity.

It is not difficult to see why rape is a subversion of sexual union, so conceived, and therefore a degraded form of an essentially intersubjective relation. Briefly, the rape of a person does not count as a shared action, much less as the joint expression of mutual sexual desire. For a sex act constitutes the joint expression of sexual desire (in action) only if that expression is, at a basic level, freely given, and because it does not seek the partner’s voluntary expression of desire, rape—by its structure—undermines the very conditions of that expression.29 Martial rape further undermines these conditions in that the martial rapist’s aim of inflicting suffering on the victim presupposes her unwillingness and aversion to the act, which is sought for its own sake. One core meaning of an act of rape, then, is that it is the degradation of an essentially intersubjective relation into a subject-object relation, mutual recognition (and affirmation) of embodiment into mere use of another person’s body.30 Martial rape is a particularly twisted corruption of the former. Pace Gardner and Shute, though, I maintain that the psychological harm of rape consists crucially in the subversion of this relation, of which the sexual use of the victim’s body is one key ingredient. Hence, it is, to a great extent, the trauma of having one’s sexual nature used by another person against one’s own will, thereby serving as the object in a subject-object relation that is, effectively, a parody of an essentially intersubjective relation.31 It is this meaning of rape that martial rapists intend to exploit in order to humiliate and harm their victims.

But then the physical and psychological harms of rape, martial or otherwise, cannot fully explain the wrongness of rape.32 For—to retrace our steps—the psychological harm is itself a response to the objective violation in which rape consists; that violation is best understood in terms of the meaning of an act of rape; and an act of this kind carries the meaning that it does because it constitutes the instrumentalization of a person’s sexual nature without her consent. The wrongness of martial rape is not fully grounded in its physical and psychological harms.

Let’s leave harm-based views behind, then, and turn to consent-based views. Now on the latter sort of view, rape is wrong centrally because it is nonconsensual sex. This view meshes well with a general picture of permissible sexual activity according to which any sex act is wrong, other things being equal, if and only if it is nonconsensual.33 Thus, on a consent-based view, martial rape is wrong partly because it is nonconsensual sex and partly because it is the intentional infliction of extreme harm on the victim, her family, and her community/group.

My counterproposal is that rape, including martial rape, is wrong partly because it is the sexual use of a person without her consent. What recommends this view over the alternative? Importantly, the consent-based view appears to capture the wrong of martial rape without any reference to its supposedly instrumentalizing character or any other objectifying features. Thus, we seem to get a cleaner account of the wrong of rape, martial rape included, by foregrounding the absence of consent. So, why bring objectification—instrumentalization—into the analysis at all? We may worry that the question is even more urgent for the Nussbaum-Langton Theory, which holds that the fact that an act is objectifying in some respect does not normally make it wrong by itself. If an act’s instrumentalizing character does not by itself explain its wrongness, then isn’t that feature of it morally innocuous as such? And if so, isn’t it morally idle, so that the absence of consent must be what is doing the normative work in cases of rape?

First, a clarification is in order. It is true that the Nussbaum-Langton Theory distinguishes between forms of instrumentalization that are sometimes permissible and those that are always wrongful—use and mere use, as I have dubbed them. But it does not follow that (sexual) instrumentalization is morally innocuous, that an act is never wrong at least partly in virtue of its (sexually) instrumentalizing character. Again, using a person (particularly her body or mind) may not always be wrong, but it is still prima facie wrong, on this view—wrong unless relevant background conditions are in place. In this respect, sexual use is like immobilizing a person—tying them up, say. Tying someone up is not wrong in and of itself: with the consent of the person tied up (e.g., as part of a sexual encounter, rehearsal for a play, etc.), it is normally permissible, other things being equal. Nevertheless, immobilizing a person in this way is prima facie wrong, wrong absent relevant background conditions. For acting in this way is indeed morally risky, as it has the effect of putting that person under our power, thereby exposing him to significant harm; by default, then, tying someone up requires a special moral justification and counts as wrong where there is none.

The immobilization of persons, depriving them of this kind of elementary control over their bodies, is not morally innocuous, then, even if it is sometimes morally permissible. And when an act of tying someone up is wrong (when it is nonconsensual, for example), it seems more natural to claim that it is wrong in virtue of rendering the person powerless without her consent, not simply in virtue of the absence of her consent. It is wrong in virtue of both features. These same points apply, mutatis mutandis, to the sexual use of another person. While sexual use of another is not wrong in and of itself, sexual use is not morally innocuous, either; it is morally risky—a standard site of moral risk, in fact. It seems more natural to claim that rape is wrong because it is sexual use without the raped person’s consent, not just because it is nonconsensual sex.

At any rate, to approach the issue more directly, one reason to believe that sexual use is one aspect of the wrong of rape, including martial rape, is that there are morally objectionable yet perfectly consensual sex acts whose moral character is obscured by the consent-based view yet illuminated by my counterproposal.34 That is, my own view identifies the wrong-making feature(s) common to cases of rape and some cases of wrongful consensual sex—for example:

Complicit Lover: Romeo and Juliet want to have sex with each other. However, Juliet wants to have sex with Romeo only in order to cause herself suffering, as part of a pattern of self-destructive behavior. Romeo knows this but does not mind, and Juliet knows that he knows. They have sex.

Sexist Lover: Romeo and Juliet want to have sex with each other. But Romeo wants sex with Juliet largely because he believes that women are ‘for’ sex, that it is only fitting that they be sexually used by men. Juliet knows this but does not mind. They have sex.

Heedless Lover: Romeo and Juliet want to have sex with each other, and after a date they have sex. Yet Romeo pursues his own sexual gratification exclusively, heedless even of whether his sexual advances are wanted or unwanted by Juliet, so long as she consents.35

Sulky Lover: Romeo wants to have sex with Juliet, but she has no inclination to have sex with him for its own sake. Still, Juliet knows that he will sulk through their whole date if she does not agree to have sex. So, noticeably reluctant, she concedes. They have sex.36

Let’s assume that in these cases both parties give consent. Does Romeo nevertheless act wrongly? Of the four cases, Romeo’s acts in Complicit Lover seem most clearly wrong. Even with Juliet’s consent, it is morally objectionable for him to have sex with her when doing so also involves knowingly playing a role in her own self-destruction. In my view, the best explanation for that act’s wrongness is that he is sexually using her, in a context in which his use of her brings her significant harm, and it is wrong to sexually use someone when such use would bring her significant harm. However, the consent-based view might agree that Romeo acts wrongly yet chalk its wrongness up to its harmfulness alone; after all, plausibly enough, it is wrong to significantly harm people, even with their consent. So, it is not immediately clear that my counterproposal more effectively captures the wrong done in Complicit Lover, although this case does cut against the view that consent is sufficient for permissible sex. What of the other cases? Romeo’s acts in Sexist Lover, Heedless Lover, and Sulky Lover strike me as wrong as well, and in these cases the aspect of sexual use is more manifest and more clearly morally objectionable by itself—indeed, even if his acts cause Juliet little suffering and she voluntarily agrees to the arrangement. Briefly, Romeo acts wrongly in Sexist Lover because in treating Juliet as a tool for his own sexual gratification in such a sexist spirit, he contributes to her subordination by socially designating her specifically—and women generally—as primarily ‘for’ sexual use by men, in the context of a broader pattern of interactions that socially designate men as sexual users and women as to-be-sexually-used.37 In Heedless Lover, too, Romeo treats Juliet wrongly by sexually using her with such disregard for her own desires and preferences, in the context of a kind of interaction—a date—that is arguably geared toward the shared exploration of mutual desire, particularly mutual sexual desire. Romeo’s pursuing sex with Juliet in Sulky Lover strikes me as wrong on similar grounds, although in this case there is an additional hint of manipulation.

Thus, in each case, Romeo acted wrongly, even though—ex hypothesi—Juliet consented. The best explanation for the wrongness of these acts is that Romeo treated Juliet as a tool for his own sexual gratification, with disregard for her perspective, and this element of sexual use—shading into subjectivity-denial—is among the wrong-making features of the acts in question. Crucially, the same sort of element is also among the wrong-making features of acts of rape, although in rape this element is importantly different because the sexual use of the victim takes place without the victim’s consent, just as in martial rape it is importantly different because the sexual use of the victim is oriented to the infliction of physical and psychological harm on her.

Thus, my counterproposal enables us to see the morally significant resemblance between (a) wrongful yet consensual sex and (b) rape, and between (c) rape as a weapon of war and (d) rape purely as a means of sexual gratification—a resemblance that the consent-based view renders invisible.

The case for my conclusion is further strengthened by the fact that the Nussbaum-Langton Theory promises to provide a unified explanation of the wrongness of other acts that intuitively appear to be similar to martial rape in morally relevant respects, including other acts of RSD. For example, instrumentalization is a defining feature of the institution of slavery, as, of course, is disregard for the enslaved person’s capacity for choice (autonomy-denial). These essential features of slavery, analogous as they are to our treatment of mere objects, do partly make enslavement the wrong that it is; slavery is not just wrong because it causes the enslaved person extreme suffering, destroys his person-specific capacities, or unduly restricts his freedom. More generally, at first glance, the enslaved person, the exploited sweatshop worker, the rape victim, the target of street harassment, and the one patient cut up to save five others have been mistreated in broadly similar ways, even if there are important differences between these cases. Instrumentalization is one common moral thread in these acts and practices, unifying them with one another and with other forms of conduct that involve neither domination nor any desire for recognition. My counterproposal vindicates our intuitive sense that all of these disparate acts are wrong at least partly, albeit not exclusively, on similar grounds. For they are all ways of dehumanizing persons by treating them as it is standard and fitting to treat mere instruments.38

Let’s survey the terrain covered so far. I have argued that martial rape, like rape generally, is wrong at least partly because it is objectifying. It is wrong partly because it is the sexual use of another person’s body (instrumentalization) without that person’s consent (autonomy-denial). That is also partly why rape, including martial rape, is harmful in the distinctive way that it is. Thus, the concept of objectification is not only applicable to martial rape but also necessary for capturing the wrongness of that practice, insofar as it is recognition-seeking. Because martial rape is a representative example of recognition-seeking domination, it follows that the concept is also necessary for capturing the wrongness of RSD more generally.

In closing, I want to mention just one further advantage of foregrounding the concept of objectification. Privileging the concept better accords with one vital yet insufficiently appreciated aspect of the epistemology of morality: the fact that we come to know how persons are to be treated, in part, through an awareness of the contrast between how persons are treated and how things—such as tools and obstacles—are treated. Our initiation into the moral point of view is, I submit, marked by our learning that persons are not to be treated as mere things—not to be used, bought and sold, disposed of; not to be treated merely as objects of beauty; and so forth. These platitudes define our normatively loaded conception of ourselves as persons, which is framed by—in J. M. Bernstein’s phrase—a ‘simple ethical geometry’ that elevates persons above mindless objects (2015: 134). By drawing on this self-conception, we come to see some of the mistreatment that we experience at the hands of others as wrong and to formulate corresponding complaints against such treatment. Thus, in giving priority to the concept of objectification in our analysis, we stand to anchor our account of the wrongness of RSD in our conception of ourselves as persons, into which the contrast between (the fitting treatment of) persons and things is built.39

4. Communication and the Special Wrong of Recognition-Seeking Domination

Having mounted a qualified defense of the Nussbaum-Langton Theory, I now want to shift gears. For even in light of the arguments of the last section, it may still be difficult to shake the suspicion that critics of the concept of objectification have a point. It may seem that RSD is subjectifying at its core, that theorists of objectification have been insufficiently sensitive to this aspect of it, and that any account that fails to acknowledge this aspect is incomplete.40 I will argue that there is a grain of truth to this suspicion. My contention is that the communicative dimension of RSD makes it not only essentially but also irreducibly subjectifying, and that this dimension is morally significant in and of itself, giving the phenomenon its distinctive shape.

Before I pointed out that RSD is subjectifying in that it has an essentially communicative dimension. We now need to determine whether this dimension is reducible to any mode(s) of objectification countenanced by the Nussbaum-Langton Theory. A simple reductive view is that the communicative dimension of RSD is nothing over and above a form of instrumentalization: the dominator uses the subordinate as a tool for his purposes in that he uses her for her recognitive attitudes toward himself or for the satisfaction that these attitudes might bring him.

The problem with this simple reductive view is that non-instrumental communication with others—communicating with others yet not for the sake of any independent end—is a paradigmatically subjectifying activity, while instrumentalization is a paradigmatically objectifying one. Non-instrumental communication is paradigmatically subjectifying because it is essentially intersubjective: it is one of the most direct modes of encountering another subject’s perspective, and, importantly, it is an activity undertaken to satisfy our human need to make contact with a subjectivity distinct from ourselves—to transcend, albeit in a limited fashion, the boundaries between self and other. Certainly, it is difficult to think of another human activity that more clearly counts as an intersubjective relation, given both its dyadic structure and the fact that it is tied to our characteristically human desire for mutual recognition—a general and pervasive human motive that moves the recognition-seeking dominator as well. Acts with communicative purport of this kind are therefore deeply, although not always exclusively, subjectifying: they are among the quintessential ways of treating another subject as a subject.

Because a tool is an archetypical object, however, treating someone as a tool is paradigmatically objectifying. Of course, the same act can be both objectifying and subjectifying all at once, and this idea is, I think, crucial for understanding the inner dynamic of RSD. But to insist that a paradigmatically subjectifying activity is reducible, without remainder, to a paradigmatically objectifying one is to risk erasing the distinction between the two categories. Although an act may also have features that render it objectifying at the same time, the dimension of non-instrumental communication is textbook subjectification: if anything is subjectifying, this form of communication is, for the reasons adduced above. So, if non-instrumental communication turns out to be fully reducible to a paradigmatically objectifying activity such as instrumentalization, the objectification– subjectification distinction collapses, for, in that case, there is simply nothing for subjectification to be. Subjectification must at least be notionally distinct from objectification if we are to be able to make sense of the latter. If the objectification-subjectification distinction collapses, in other words, there is nothing else for objectification to be, either, since this notion depends on a contrast between objectifying behavior/attitudes and our quintessential modes of relating to other subjects as such. The arguments of the last section should make us doubt that the concept of objectification is empty.41

Hence, as a conceptual matter, the communicative dimension of RSD cannot be fully assimilated to instrumentalization, lest the objectification-subjectification distinction evaporate. While acts of RSD may be instrumentalizing, then, they are, at best, an impure case of instrumentalization, in virtue of being motivated by a quintessentially subjectifying desire: the non-instrumental desire to communicate with another subject. These considerations justify us in rejecting the simple reductive view. Consider, then, the more sophisticated reductive view, on which the communicative dimension of RSD is instead reducible to its instrumentalizing character in combination with, variously, its autonomy-denying, subjectivity-denying, and/or violating character, as the case may be. Could the communicative dimension of RSD be reduced without remainder to instrumentalization plus other modes of objectification, as on this view?

The sophisticated reductive view faces a different but related problem. While acts of RSD do typically have autonomy- and subjectivity-denying features, they also have the contrary, more purely subjectifying features. Sadistic rape exemplifies this dual character, as Langton shows us:

In this sort of case, it’s not that [the sadistic rapist] 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. (2001a: 234)

In sadistic rape, she points out, as in standard non-sexual torture, autonomy is both affirmed in one respect and denied in another—and, in fact, affirmed so that it can then be denied. And far from ignoring the victim’s perspective, sadistic rape is also subjectivity-affirming as well as -denying, since it essentially includes attention to, indeed non-instrumental concern for, that perspective. Consider one of Nussbaum’s examples of objectification: in an early scene in Laurence St. Clair’s hardcore pornographic novel Isabelle et Véronique, the novel’s protagonist, Macrae, is seized by a sudden desire to ‘violate … desecrate, destroy’ and so brutally rapes a sleeping woman, Isabelle, presumably in order to make her suffer for its own sake.42 There is plausibly an element of subjectivity-denial here, as his act is based on a failure of empathy and imagination. Still, it is, strictly speaking, false that Macrae’s treatment of Isabelle shows that he is ‘completely indifferent about how Isabelle is feeling or thinking,’ as Lina Papadaki (2012: 23–24) claims. In a perfectly apt sense, the perspective of a victim of sadistic rape does matter to the rapist—all the more so if the act is motivated by a non-instrumental desire to secure her recognition. Isabelle’s experiences are, disturbingly, taken into account by Macrae, in that he has a non-instrumental desire that she suffer and perhaps recognize him as making her suffer. This may be only an impurely subjectifying relation, tainted as it is with objectification, but it is subjectifying nevertheless. Thus, the sophisticated reductive view is false: like sadism more generally, RSD cannot be purely objectifying. There is, again, a stubbornly subjectifying remainder.

From these reflections, it follows that the communicative purport of RSD is irreducible to any combination of categories in Nussbaum’s table. Nor is it advisable to simply add another category of objectification to the table to cover the remainder. From the standpoint of the Nussbaum-Langton Theory, this category would be gerrymandered, as it were, failing to preserve the conceptual connection to our treatment of objects, and the concept of objectification would then begin to look alarmingly amorphous. The communicative purport at the core of domination of this kind is not fruitfully understood in terms of the treatment of things, and there are rather strict limits to any apt comparison with the sheer use or wanton destruction of objects.

Is the communicative dimension of acts of RSD itself morally significant, however? Start with an observation: the moral profile of an act is at least sometimes shaped by its aim or motive. For example, there appears to be a morally significant difference between you coercing me into giving you my wallet purely in order to increase your purchasing power, on the one hand, and you threatening me as a way of dominating me for its own sake, on the other. Granted, you act wrongly in both cases. But there is something especially sinister about the second act, and an account of that act’s wrongness would be inadequate if it failed to mention its aim or motive. Indifferent instrumentalization is morally different from (non-recognition-seeking) sadism.

Now imagine a different case: you coerce me not only to dominate me for its own sake but also to make me see you as more important (and myself as less important), as evinced by your power to so dominate me. This is an act of RSD, then, and I will take it to be representative of acts of this kind. Does the communicative dimension of this act make a moral difference by itself, so that an account of its wrongness is incomplete without mention of this dimension? My intuition here is that while it is morally bad for you to subject me to coercion so as to dominate me for its own sake, without any non-instrumental desire to affect how I regard you (or myself), it is morally worse for you to treat me in this way so as to force me, for its own sake, to acknowledge this treatment as proof—or enactment—of your superiority (and my inferiority).

So, the communicative dimension of this act of RSD does seem to make it more distinctively sinister than its non-recognition-seeking counterpart. How might that be? One possibility is that the former act is morally worse than the latter act because, in virtue of the former’s communicative dimension, it simply causes the victim more psychological suffering—it is an especially painful form of humiliation, perhaps—and thus causes more harm overall.

It is not obvious to me that acts of RSD always or generally cause more psychological suffering, but apart from that, the description just provided does not seem to faithfully portray the special nature of the wrong itself. For one, the additional layer of psychological suffering inflicted by this act of RSD—humiliation before the humiliator—is not a brute sensation, like an itch or a headache. Here again, rather, the psychological suffering ineluctably shadows or tracks the objective violation embodied in that act of domination. When you dominate me for its own sake but do not seek my recognition, I feel humiliated by this treatment largely in virtue of the nature of that act and its social meaning. When you dominate me as a way of securing my recognition for its own sake, however, I will tend to feel particularly humiliated by the act because you have thereby made me regard, and perhaps acknowledge, you as more powerful than myself, as evinced by the humiliating treatment to which you have subjected me. But if the act of RSD is particularly humiliating, causing greater suffering, the best explanation is that the act itself is distinctively wrongful. For in its effort to make the subordinate regard herself (as thus violated) through her dominator’s eyes, the act of RSD aims to subvert the subordinate’s capacity for mutually recognitive interaction—the very capacity that is central to her identity as a human subject. To borrow an expression from David Sussman, RSD is the most direct way of involving another subject in her own violation: the other is made to participate in the wrong done to her by her exercising a capacity that is an especially deep, significant feature of her subjectivity.43

Other things being equal, an act of RSD is therefore a special sort of wrong, a more intimate variety of violation than purely instrumental domination or non-instrumental domination that is not recognition-seeking. I would even go so far as to claim that RSD constitutes wronging par excellence, an idea supported by the fact that recognition-seeking torture and rape appear to be the kinds of acts that we are most readily disposed to call evil. And it is not difficult to see why that is: they constitute, I submit, the most direct assault on their victims’ value as persons. The recognition-seeking character of RSD, then, makes a moral difference in its own right, and not just in virtue of causing its victim a greater degree of psychological suffering; indeed, that suffering is largely a reflection of the prior, essentially communicative violation committed.44 This conclusion comports with our sense that the kind of act committed by Elliot Rodger and his ilk has a moral profile that differs, in morally significant respects, from the moral profiles of both its purely instrumentalizing counterpart and its intrinsic yet non-recognition-seeking counterpart. Rodger’s acts seem specially wrong compared with, say, his killing the same people in secret by detonating a bomb in the Alpha Phi Sorority House, purely from an intrinsic desire to harm them. My view explains why. What Rodger did is specially wrong because he sought to subvert his victims’ capacity for mutual recognition so as to involve them in their own violation; this dimension would have been absent from his act’s non-recognition-seeking counterpart.

Let me summarize the argument of this section. RSD is essentially communicative in the sense that it constitutively aims to establish a relation of mutual recognition (of a sort) between its agent and its patient, for its own sake. This fact is significant, I have tried to show, because the communicative dimension of this kind of domination is irreducible to the categories of objectification countenanced by the Nussbaum-Langton Theory—irreducible to each individually as well as to any combination of these, actual or potential. This communicative dimension is morally significant as well: it makes RSD wrong in a special way, if not always more wrong than other acts. Because the special wrongness of RSD is at least partly grounded in its communicative dimension and this dimension is irreducible to any category of objectifying treatment, it follows that its objectifying features do not fully ground its wrongness. Thus, application of the concept of objectification only goes so far in clarifying the wrong of recognition-seeking domination.45 An additional moral category is needed to complete the picture.

5. Concluding Remarks: The Tension Revisited

I have argued that the concept of objectification is necessary but insufficient for capturing the special wrong of recognition-seeking domination, as exemplified by the acts of many mass murderers, martial rapists, and the like. The general lesson of my discussion is that an account of the wrong of RSD is adequate only if it exhibits a sensitivity to both its objectifying and its subjectifying or humanizing aspects. Thus, properly understood, these different descriptions of RSD are not contradictory, but complementary. Indeed, both descriptions are compulsory, for domination of this kind most plausibly consists in treating someone as an object as a way of treating her as a subject. The foregoing arguments have at least dispelled the air of paradox standardly associated with this combination of descriptions and have shown how each represents a significant aspect of the phenomenon. However, I have not provided a full account of the wrong of RSD that would clarify how, exactly, the objectifying and subjectifying aspects of such acts are related, much less the contribution that each aspect makes to RSD’s moral profile.

Although I take myself to have resolved the original tension, I must admit that there is still a tension in the vicinity. But it does not belong to the combination of objectifying and subjectifying descriptions of recognition-seeking domination. Rather, it belongs to the phenomenon itself. The tension is not conceptual but psychological, and it is located at the level of the recognition-seeking dominator’s attitudes toward the subordinate: the dominator’s desire to make his victim into an object yet to have her recognize, and perhaps also acknowledge, his status with respect to her, thereby confirming that she is, in fact, a subject. While these attitudes are not logically inconsistent, they are in a kind of conflict with one another. An adequate moral psychology of inhumanity must uncover the basis and psychological structure of the conflict.


I presented this essay as a colloquium at Oakland University as well as at a workshop at Northwestern University, the 2018 meeting of the Illinois Philosophical Association, the Minds of Our Own conference at MIT, the 2019 Bay Area Feminism and Philosophy Workshop, and the 2019 Rocky Mountain Ethics Congress. I am grateful to audiences at these venues for discussion. I was also fortunate to have many excellent commentators read and discuss my work. A special thanks goes to (in order of appearance) Kyla Ebels-Duggan, Susan Dieleman, Jennifer Foster, Wendy Salkin, and Sumeet Patwardhan. For particular, helpful exchanges, I would like to thank Ásta, Olivia Bailey, Erin Beeghly, Sharon Berry, Talbot Brewer, Colin Chamberlain, Shlomo Cohen, Mary Clayton Coleman, Ryan Davis, Jeremy Fix, Sahar Heydari Fard, Sally Haslanger, Jonathan Jenkins Ichikawa, Tony Laden, Patricia Marechal, Rebecca Mason, Oded Na’aman, Keshav Singh, David Livingstone Smith, Joan Straumanis, Aaron Yarmel, Eric Wiland, and Robin Zheng. Apart from my official commentators, Selim Berker, Yongming Han, Céline LeBoeuf, Benjamin Lee Nelson, Grant Rozeboom, and three anonymous referees all read drafts of the essay and provided quite useful feedback which prompted many necessary revisions. Finally, I must acknowledge a special debt to Christine Korsgaard for her characteristically insightful and detailed commentary, which helped the piece find its final form.


  1. Hayden (1985: 15).
  2. See MacKinnon (1989: 123–30), Dworkin (2000: 30–31). Sally Haslanger (2002: 223–32) follows MacKinnon’s view closely. For a recent view in a similar mold, see also Jütten (2016).
  3. Nussbaum (1995; 2007), Langton (2001a; 2001b), and Papadaki (2010; 2012; 2015; 2017). The concept of objectification also appears in Kant’s discussion of sexual desire and marriage (1991: 96–98; 1997: 155–59), as well as in Marx’s analysis (1988: 69–84) of the estrangement of laborers under capitalism.
  4. Nussbaum (1995: 257). To Nussbaum’s catalogue Langton adds three further categories of objectification:

    8. Reduction to body: treating someone as primarily to be identified with their body or body parts

    9. Reduction to appearance: treating someone primarily in terms of their sensory appearance

    10. Silencing: treating someone as lacking the capacity for speech

    See Langton (2001a: 228–29). For the purposes of this paper, I remain agnostic about whether Nussbaum’s catalogue is enhanced by Langton’s amendments and whether every item that appears in it deserves its place. However, I will argue that at least some of the behaviors/attitudes in the catalogue are indeed legitimate forms of objectification. Finally, a terminological point: I refer to both acts and attitudes as ‘activities’, whereas I use the term ‘treatment’ to pick out only acts. I mostly focus on the wrongness of objectifying acts rather than attitudes.
  5. Nussbaum (1995: 265, 275–78). This third claim must be interpreted carefully. For example, Alan Soble (2017: 306) alleges that on Nussbaum’s view ‘small sexually vulgar chunks of a couple’s relationship, small pieces of noxious sexual objectification, are permissible in virtue of the larger or more frequent heavenly chunks of mutual respect that comprise their relationship.’ I do not think that this reading of Nussbaum can be correct. On a more charitable interpretation, Nussbaum’s point is that an act of objectification is right or wrong, morally good or morally bad, partly—but crucially—in virtue of the context in which it is performed, where she takes the context to principally include facts about the nature of the relationship between agent and patient. So, for her, it is not the case that the acts in question are morally bad or wrong in and of themselves but mysteriously rendered permissible by other, rightful or morally good acts of the agent’s (‘treating you badly today’ but still treating ‘you admirably the whole day yesterday’). Rather, on the Nussbaum-Langton Theory, the moral qualities of an objectifying act/attitude are not totally fixed in advance of a specification of the background context. For criticism of this aspect of the theory, see Papadaki (2010: 29–31; 2017: 392–93).
  6. While I believe that this formulation comports quite well with at least one strain of vernacular usage, nothing in my discussion will depend on the truth of that belief, and I concede that the word ‘domination’ (and its cognates) may be used in other ways. Specifically, my use of the term differs from the more standard conception of domination privileged in the neorepublican tradition: domination as subjection to another’s power to interfere with our aims on an arbitrary basis. See, in particular, the work of Philip Pettit (1996; 1997; 2012). Take the above definition as stipulative, then: this is the sort of phenomenon that I am interested in.
  7. Here I am assuming, of course, that there is a coherent distinction to be drawn between acts that are performed for the sake of an independent end and acts that are not, where the latter includes both acts that are identical with their ends (i.e., exercising simply for its own sake) and acts that realize or constitute their ends (i.e., greeting someone as a way of showing her respect). The former distinction is so familiar and apparently useful, however, that the assumption does not seem out of order.
  8. In other words, his desire to terrorize and kill them was a realizer desire, as it was a desire for an end that counted as a specific realization of the end for which he had an intrinsic desire: revenge. His desire to terrorize and kill was not an instrumental desire merely. I borrow the distinction between instrumental, intrinsic, and realizer desires from Arpaly and Schroeder (2014: 6–9).
  9. My definition also excludes acts of ‘symbolic’ domination performed on the subordinate not in order to gain her (or anyone’s) recognition but purely for the sake of the symbolic significance of the act itself (e.g., a man raping a woman while she is unconscious, for the thrill of so demeaning her). My thanks to Sahar Heydari Fard for this example.
  10. I am grateful to Jeremy Fix and David Livingstone Smith for pressing me to more clearly distinguish this form of domination from the one with which this essay is principally concerned.
  11. This psychological aspect of torture is evident in the testimony of some victims of torture. For example, Jean Améry (1980: 35), who was tortured by the Nazis, characterizes his torturers as motivated by a peculiar kind of sadism, a desire for ‘total sovereignty’—to wield ‘control of the other’s scream of pain and death,’ becoming ‘master over flesh and spirit, life and death.’ In a similar vein, Jacobo Timerman (1981: 37–38; see also 60–61), a journalist tortured by the Argentine junta during the Dirty War, writes of the mutually recognitive bond between torture victim and torturer: ‘Both parties seem to feel need of the other: for the torturer, it is a sense of omnipotence, without which he’d find it hard perhaps to exercise his profession—the torturer needs to be needed by the tortured; whereas the man who’s tortured finds in his torturer a human voice, a dialogue for his situation, some partial exercise of his human condition…’ Thanks to Patricia Marechal for making me aware of Timerman’s piece. For a description of the psychological dynamics of torture, see also Scarry (1985: 28–38, 51–59).
  12. As Elaine Scarry (1985: 19–20) points out, torture of this kind characteristically includes protracted, apparently fruitless bouts of interrogation.
  13. Bartky (1990: 27). Street harassment typically has a further recognitive dimension as well: it is motivated by a desire for recognition from one’s fellow harassers.
  14. This is one of the main themes of Hegel’s (1977: 112–19) dialectic of the master and the slave. An interpersonal dynamic of this kind is at the heart of Sartre’s (1992: 471–75) description of intersubjectivity (‘being-for-others’) as essentially conflictual, marked by a basic tension between an objectifying and ‘subjectifying’ stance toward another person.
  15. Cf. Jütten (2016: 44): behaviors such as street harassment ‘reduce women to sex objects in the eyes of their harassers and invite others to see them in this way. More generally, sexual harassment communicates to women that they lack control over their self-presentation; there is nothing that they can do in order to avoid being reduced to a sex object if the harassers enjoy reducing them to one for their own titillation.’
  16. Specifically, on the moderate position, the explanatory power of referencing an act’s objectifying features and the absence of relevant background conditions does not constitute a full explanation of the act’s wrongness.
  17. Thus, in this paper I do not propose to fully answer every objection to the Nussbaum-Langton Theory or to theories of objectification more generally. But there are three objections that I want to acknowledge at the outset. First, Alan Soble (2002: 51–54) alleges that the idiom of objectification should be jettisoned because it is based on the false presupposition that human beings or persons have a distinctive value—dignity—that elevates us above mere things. Second, Nancy Bauer (2015: 35–36) argues that theories of objectification such as Nussbaum’s make the concept so generic that they threaten to rob the idea of sexual objectification of its explanatory power and social-critical bite. Third, Ann Cahill (2011: 24–27) charges theories of objectification with marginalizing, indeed vilifying, our material, embodied nature and our intrinsic relationality by making autonomy our defining feature. I want to respond very briefly to these objections. First, Soble’s discussion misplaces the burden of proof in the debate, as the idea that people have a value higher than, and different in kind than, mere objects is so central to modern moral consciousness that a powerful argument is needed to dislodge it, and we find no such argument in his discussion. See also the response to Soble in Halwani (2018: 254–55). Second, Bauer’s insistence that sexual objectification is not illuminatingly understood as part of a general phenomenon of objectification is problematic, as it reduces our awareness of the resemblances between sexual objectification and apparently similar phenomena, like the sheer use of exploited workers or slaves. Finally, contra Cahill, I do not believe that the concept of objectification, construed as on the Nussbaum-Langton Theory, is committed to a picture of persons as disembodied souls whose most significant feature is our autonomy (construed atomistically)—just to the claim that there are characteristic ways of relating to mindless objects that are morally objectionable when applied to persons, at least absent certain background conditions.
  18. Mikkola (2016: 157; see also 8, fn. 2). As she points out, there is even a sense in which this practice involves an affirmation of its victims’ humanity, rather than utter disregard for it.
  19. Cahill (2011: 133). Problematically, Cahill seems to take sadistic rape as the model for rape more generally. Beyond that, however, Cahill’s denial that (sadistic) rape is wrong partly in virtue of its being objectifying is at odds with the descriptions offered by some survivors of rape, who do present the ordeal in explicitly objectifying terms. Elizabeth Smart, who was kidnapped from her home at the age of fourteen and raped repeatedly for nine months, has this to say (Connor 2015) about her rapist’s treatment of her: ‘To her I was a slave, and to him I was an object … I didn’t feel human.’
  20. To be fair to Mikkola and Cahill, Dworkin and MacKinnon do sometimes speak of objectification in ways that seem to suggest that, in their view, treating a woman as an object turns her into an object, in the most literal sense of the phrase. For example, Dworkin (2000: 30) writes that objectification ‘occurs when a human being, through social means, is made less than human, turned into a thing or commodity, bought and sold.’ Cf. MacKinnon (1989: 124). Nevertheless, it is more accurate to interpret them as holding that (1) objectification damages women’s characteristically human capacities, thereby making them more object-like, and that (2) objectification involves according to women (something like) the social status of an object. Indeed, this interpretation gains support from the fact that the passage of Dworkin’s (2000: 31, my emphasis) just cited ends thus: ‘ … those who can be used as if they are not fully human are no longer fully human in social terms; their humanity is hurt by being diminished.’
  21. I want to thank Shlomo Cohen for this example.
  22. This category of objectifying treatment is highlighted in Papadaki (2010: 34–35), among other work.
  23. Cahill (2011: 28) explicitly charges theorists of objectification with embracing a claim that looks very much like the exclusion principles, alleging that ‘feminist criticisms of objectification are flawed inasmuch as they draw a clear opposition between subject-hood and object-hood: … they assume, for the most part, that while one is an object, one is not a subject, and vice versa.’ And while Raja Halwani does not embrace the exclusion principles, he still explicitly defines objectification (and sexual objectification in particular) in a way that excludes the possibility of being treated as an object and as a person all at once. ‘At its core,’ writes Halwani (2018: 241–42), ‘objectification is treating or considering a person as only an object,’ and sexual objectification is treating or regarding someone ‘only as a sex object.’ For Halwani, the ‘only’ is crucial, ‘because without it we would treat someone else simultaneously as an object and as person,’ making objectification ‘morally innocuous’ and therefore incapable of capturing the wrongness of apparently objectifying behavior or attitudes.
  24. Pace Kant, however, I doubt that sexual desire characteristically leads those who are in its grip to treat their (prospective) partners merely as means to sexual gratification rather than as ends-in-themselves; in the context of a relationship informed by respect and concern, for instance, sexual activity is morally unobjectionable. Kant’s view of sexuality is discussed sympathetically by Herman (1993: particularly 59–65) and Halwani (2018: 245–59).
  25. The point is more general. You count as treating me as a chair even if you know full well that I’m not a chair but a person, just as you count as treating me as a child even if you know full well that I’m an adult. Normally, it is the features of these acts that determine whether or not they are instances of treating a person as an X or a Y, not the agent’s metaphysical beliefs about that person.
  26. In particular, it is very plausible that martial rape’s genocidal character—the fact that it aims to destroy a particular ethnic group—makes it wrong in the special way that it is. For an excellent analysis of the wrongness of martial rape along these lines, see Miller (2009: esp. 513–19).
  27. Thus, my definition of use differs somewhat from Samuel Kerstein’s (2009: 166) definition of use as an agent’s intentionally doing something to someone ‘in order to secure (or as a part of securing) one of his ends.’ Interestingly, my definition is closer to Alex Guerrero’s (2016: 780; cf. 779) definition of using a person merely as a means, since one of his conditions is that the agent ‘not consider [the patient’s] views with respect to’ the act in question, or ‘weigh or consider the effects’ of doing that action on the patient. Discussion of Guerrero’s argument is beyond the scope of this essay.
  28. Gardner and Shute (2007: 22). Cf. Baber (2002: 304), who claims that one of the harmful aspects of rape is that it is a violation of our interest in not being used as mere means.
  29. For a kindred view, see Begon (2019: 350–52).
  30. I do not claim that the meaning of an act of rape is exhausted by this description, that there is just one core meaning. or that rape is psychologically harmful only in virtue of carrying the meaning that I have identified. My point is only that the meaning characterized above is one core meaning which explains a crucial part of the humiliation—and thus psychological harm—of rape. More obviously, rape is standardly psychologically harmful for all sorts of reasons, e.g., because it is physically harmful and/or involves being made radically powerless, etc.
  31. On this point, Jean Hampton (1999: 135) is right: ‘a rapist whose rape conveys this attitude [that women are ‘its’] expresses the idea that women are even lower than chattel—mere “objects” who are there to be used whenever the male feels the need to do so… . The woman is used as though she is an object, and so she is thought to be one.’ Hampton also claims, plausibly, that rape expresses a threatening message to other women, even those who have not experienced rape: ‘ “You are the sort of creature who is for me, and who will be used by me if I choose to do so.” ‘ See also Frye and Shafer (1977: esp. 340–42).
  32. These considerations give us reason to suspect that something is missing in Alan Wertheimer’s (2003: 112) account of the harm of rape as residing in the victim’s distinctive ‘distress and psychological damage’. See Wertheimer (2003: 89–118).
  33. See, e.g., Primoratz (2001), Marino (2008), and Mappes (2017). For critical exploration of the idea that consent is the sole criterion of permissible sex, see MacKinnon (1982), Klepper (1993), Hampton (1999), Morgan (2017), West (2017), and Gavey (2019: ch. 5).
  34. Of course, disregard for a person’s choices and ends—by using her body, in its sexual capacity, without her consent—is arguably objectifying in a further respect as well: it constitutes autonomy-denial, the treatment of a person in a way that overrides or bypasses her capacity for choice. But I will not press the point here.
  35. This case is adapted from Klepper (1993); see his Case 1 (1993: 480–81).
  36. I have in mind some of the cases, discussed by Nicola Gavey (2019: 128–55), that occupy a ‘gray area’ between rape and permissible sex.
  37. I concede that instrumentalization cannot constitute the full explanation of why this familiar form of sexual objectification is wrong. In particular, I agree with Jütten that the sexual objectification of women is wrong in virtue of imposing an autonomy-undermining social meaning on them in the context of gender-based oppression. See Jütten (2016: esp. 37–40); cf. Burgess-Jackson (1999: 108). Nevertheless, the sexual objectification of women expresses this deleterious social meaning precisely because it is the sexual use of a woman that socially designates her as to-be-sexually-used.
  38. I take my argument to cut against Bauer’s (2015) view that we should scrap the general concept of objectification and attend to sexual objectification instead. See fn. 17, above.
  39. These considerations are relevant to evaluating Cahill’s proposed alternative to the concept of objectification. Cahill (2011: 32) argues that we find a conceptual tool that is better suited to feminist normative theorizing in the notion of what she calls derivatization. By her definition, to derivatize someone is ‘to portray, render, understand, or approach a being solely or primarily as the reflection, projection, or expression of another being’s identity, desires, fears, etc.’ There will be substantial overlap between these concepts, of course, since the value of an object for us is standardly determined by its capacity to serve our needs and interests, but, in any case, I think that the concept of derivatization is a valuable one. Nevertheless, because the concept of objectification occupies a more secure place in our moral epistemology, I do not find the concept of derivatization to be a suitable replacement.
  40. While Nussbaum and Langton would presumably disagree that the fact that an act of domination has some combination of objectifying features fully explains why it is wrong, in reading their work one gets the impression that they are confident that the concept of objectification is capable of doing extensive normative work. So, while I am not exactly affirming a claim that they explicitly deny, I am highlighting a limit to the concept’s explanatory power. My argument therefore serves less as criticism of the Nussbaum-Langton Theory and more as critique, in the older sense of the term: determining the scope and limits of a concept in order to prove its legitimacy.
  41. I take this argument to show that the prospects for both a conceptual reduction and meta-physical reduction are dubious. Formulated as a conceptual reduction, the simple reductive view holds that the concept of intersubjective engagement of this kind is identical with, or constituted by, the concept of using someone for (the pleasure brought by) their recognitive attitudes. Cast as a metaphysical reduction, by contrast, this view is that the reduction holds at the level of the corresponding worldly properties, not concepts. The point of the argument is to bring out that conceptual considerations make it doubtful that either the concept or the property of intersubjective engagement of this kind can be reduced to the relevant (allegedly) instrumentalizing concepts or features, on pain of incoherence.
  42. See St. Clair (1989: 2–4), Nussbaum (1995: 252–53, 280–81).
  43. David Sussman (2005: esp. 30) argues that torture is wrong in virtue of perverting the victim’s moral relations to herself and to the torturer, thereby making the victim participate in her own violation (a wrong that Sussman calls a ‘humiliation’).
  44. It does not follow that acts of RSD are always morally worse (or morally worse in virtue of their recognition-seeking character) than acts that are not recognition-seeking, only that this character makes some moral difference.
  45. Thus, Michael Rea (2019: 12, fn. 19) is wrong when he claims, in response to an earlier version of this essay, that this part of my argument ‘goes through only on the mistaken assumption that objectification is inherently mind-suppressing, or mind-insensitive.’ No such assumption is at work in this argument.


Améry, Jean (1980). At the Mind’s Limits (Sidney Rosenfeld and Stella P. Rosenfeld, Trans.). Indiana University Press.

Arpaly, Nomy, and Timothy Schroeder (2014). In Praise of Desire. Oxford University Press.

Bartky, Sandra L. (1990). On Psychological Oppression. In Femininity and Domination (22–32). Routledge.

Baber, H. E. (2002). How Bad is Rape? In Alan Soble (Ed.), The Philosophy of Sex (4th ed., 303–16). Rowman and Littlefield.

Bauer, Nancy (2015). How to Do Things with Pornography. Harvard University Press.

Begon, Jessica (2019). Sexual Perversion: A Liberal Account. Journal of Social Philosophy, 50(3), 341–62.

Bernstein, J. M. (2015). Torture and Dignity. University of Chicago Press.

Burgess-Jackson, Keith (1999). A Theory of Rape. In Keith Burgess-Jackson (Ed.), A Most Detestable Crime: New Philosophical Essays on Rape (92–117). Oxford University Press.

Cahill, Ann (2011). Overcoming Objectification: A Carnal Ethics. Routledge.

Connor, Tracy (2015, November 2). A Decade After Her Rescue, Elizabeth Smart Says, ‘I Didn’t Feel Human’. NBC News.

Dworkin, Andrea (2000). Against the Male Flood: Censorship, Pornography, and Equality. In Drucilla Cornell (Ed.), Oxford Readings in Feminism: Feminism and Pornography (19–41). Oxford University Press.

Frye, Marilyn and Carolyn Shafer (1977). Rape and Respect. In Mary Vetterling-Braggin, Frederick Elliston, and Jane English (Eds.), Feminism and Philosophy (333–46). Rowman and Littlefield.

Gardner, John and Stephen Shute (2007). The Wrongness of Rape. In Offences and Defences (1–32). Oxford University Press.

Gavey, Nicola (2019). Just Sex?: The Cultural Scaffolding of Rape (2nd ed.). Routledge.

Guerrero, Alexander (2016). Appropriately Using People Merely as Means. Criminal Law and Philosophy, 10, 777–94.

Halwani, Raja (2017). Casual Sex, Promiscuity, and Objectification. In Raja Halwani, Alan Soble, Sarah Hoffman, and Jacob M. Held (Eds.), Philosophy of Sex (7th ed., 401–20). Rowman and Littlefield.

Halwani, Raja (2018). Philosophy of Sex, Love, and Marriage: An Introduction (2nd ed.). Routledge.

Hampton, Jean (1999). Defining Wrong and Defining Rape. In Keith Burgess-Jackson (Ed.), A Most Detestable Crime: New Philosophical Essays on Rape (118–58). Oxford University Press.

Haslanger, Sally (2002). 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 (209–53). Westview Press.

Hayden, Robert (1985). Night, Death, Mississippi. In Frederick Glaysher (Ed.), Collected Poems (15–16). Liveright Publishing.

Hegel, Georg W. F. (1977). The Phenomenology of Spirit (A. V. Miller, Trans.). Oxford University Press.

Herman, Barbara (1993). Could It Be Worth Thinking about Kant on Sex and Marriage? In Louise M. Antony and Charlotte Witt (Eds.), A Mind of One’s Own: Feminist Essays on Reason and Objectivity (53–72). Westview Press.

Jütten, Timo (2016). Sexual Objectification. Ethics, 127(1), 27–49.

Kant, Immanuel (1991). The Metaphysics of Morals (Mary J. Gregor, Trans.). Cambridge University Press.

Kant, Immanuel (1997). Lectures on Ethics. Peter Heath and J. B. Schneewind (Eds.). Cambridge University Press.

Kerstein, Samuel (2009). Treating Others Merely as Means. Utilitas, 21(2), 163–80.

Klepper, Howard (1993). Sexual Exploitation and the Value of Persons. The Journal of Value Inquiry, 27(3–4), 479–86.

Langton, Rae (2001a.) Autonomy Denial in Objectification. In Sexual Solipsism: Philosophical Essays on Pornography and Objectification (223–40). Oxford University Press.

Langton, Rae (2001b). Sexual Solipsism. In Sexual Solipsism: Philosophical Essays on Pornography and Objectification (311–56). Oxford University Press.

MacKinnon, Catharine (1982). Feminism, Marxism, Method, and the State: An Agenda for Theory. Signs, 7(3), 515–44.

MacKinnon, Catharine (1989). Toward a Feminist Theory of the State. Harvard University Press.

Manne, Kate (2016). Humanism: A Critique. Social Theory and Practice, 42(2), 389–415.

Manne, Kate (2018). Down Girl: The Logic of Misogyny. Oxford University Press.

Mappes, Thomas (2017). Sexual Morality and the Concept of Using Another Person. In Raja Halwani, Alan Soble, Sarah Hoffman, and Jacob M. Held (Eds.), Philosophy of Sex (7th ed., 273–92). Rowman and Littlefield.

Marino, Patricia (2008). The Ethics of Sexual Objectification: Autonomy and Consent. Inquiry: An Interdisciplinary Journal of Philosophy, 51(4), 345–64.

Marx, Karl (1988). Economic and Philosophic Manuscripts of 1844 and the Manifesto of the Communist Party (Martin Milligan, Trans.). Prometheus Books.

Mikkola, Mari (2011). Dehumanization. In Thom Brooks (Ed.), New Waves in Ethics (128–49). Palgrave-MacMillan.

Mikkola, Mari (2016). The Wrong of Injustice: Dehumanization and Its Role in Feminist Philosophy. Oxford University Press.

Miller, Sarah Clark (2009). Moral Injury and Relational Harm: Analyzing Rape in Darfur. Journal of Social Philosophy, 40(4), 504–23.

Morgan, Seiriol (2017). Dark Desires. In Raja Halwani, Alan Soble, Sarah Hoffman, and Jacob M. Held (Eds.), Philosophy of Sex (7th ed., 349–70). Rowman and Littlefield.

Nussbaum, Martha (1995). Objectification. Philosophy & Public Affairs, 24(4), 249–91.

Nussbaum, Martha (2007). Feminism, Virtue, and Objectification. In Raja Halwani (Ed.), Sex and Ethics: Essays on Sexuality, Virtue, and the Good Life (7th ed.). Palgrave Macmillan.

Papadaki, Evangelina (2010). What is Objectification? Journal of Moral Philosophy, 7(1), 16–36.

Papadaki, Evangelina (2012). Understanding Objectification: Is There Any Special Wrongness Involved in Treating Human Beings Instrumentally? Prolegomena, 11(1), 5–24.

Papadaki, Evangelina (2015). What is Wrong about Objectification? In Thom Brooks (Ed.), Current Controversies in Political Philosophy (87–99). Routledge.

Papadaki, Evangelina (2017). Sexual Objectification. In Raja Halwani, Alan Soble, Sarah Hoffman, and Jacob M. Held (Eds.), Philosophy of Sex (7th ed., 381–99). Rowman and Littlefield.

Pettit, Philip (1996). Freedom as Antipower. Ethics, 106(10), 576–604.

Pettit, Philip (1997). Republicanism: A Theory of Freedom and Government. Clarendon Press.

Pettit, Philip (2012). On the People’s Terms: A Republican Theory and Model of Democracy, Cambridge University Press.

Primoratz, Igor (2001). Sexual Morality: Is Consent Enough? Ethical Theory and Moral Practice, 4(3), 201–18.

Rea, Michael C (2019). Representational and Attitudinal Sexual Objectification. Feminist Philosophy Quarterly, 5(4), 1–23.

Sartre, Jean-Paul (1992). Being and Nothingness (Hazel Barnes, Trans.). Washington Square Press.

Scarry, Elaine (1985). The Body in Pain. Oxford University Press.

Soble, Alan (2002). Pornography, Sex, and Feminism. Prometheus Books.

Soble, Alan (2017). Sexual Use. In Raja Halwani, Alan Soble, Sarah Hoffman, and Jacob M. Held (Eds.), Philosophy of Sex (7th ed., 293–320). Rowman and Littlefield.

St. Clair, Laurence (James Hankinson) (1989). Isabelle and Veronique: Four Months, Four Cities. Blue Moon Books.

Sussman, David (2005). What’s Wrong with Torture? Philosophy & Public Affairs, 33(1), 1–33.

Timerman, Jacobo (1981). Prisoner without a Name, Cell without a Number. University of Wisconsin Press.

Wertheimer, Alan (2003). Consent to Sexual Relations. Cambridge University Press.

West, Robin (2017). The Harms of Consensual Sex. In Raja Halwani, Alan Soble, Sarah Hoffman, and Jacob M. Held (Eds.), Philosophy of Sex (7th ed., 371–79). Rowman and Littlefield.