Presupposition and Consent

I argue that ‘consent’ language presupposes that the contemplated action is or would be at someone else’s behest. When one does something for another reason — for example, when one elects independently to do something, or when one accepts an invitation to do something — it is linguistically inappropriate to describe the actor as ‘consenting’ to it; but it is also inappropriate to describe them as ‘not consenting’ to it. A consequence of this idea is that ‘consent’ is poorly suited to play its canonical central role in contemporary sexual ethics. But this does not mean that nonconsensual sex can be morally permissible. Consent language, I’ll suggest, carries the conventional presupposition that that which is or might be consented to is at someone else’s behest. One implication will be a new kind of support for feminist critiques of consent theory in sexual ethics.


Introduction
I asked my friend Alice, who is is an expert on presupposition, to read this paper and give me comments. She's a busy woman with lots of demands on her time, but because she's a generous friend, she read my paper. One might ask: (1) Did Alice consent to reading this paper?
Under the circumstances (1) is a fine question, and its answer is 'yes'. If I'd forced her into reading the paper, or threatened her, then the answer would be 'no'. In cases like this, we use 'consent' language to mark whether people are doing things because they've freely agreed to them, or are forced or coerced. So far, so obvious. Now consider Bob, who reads this paper for a very different reason. Bob is a philosopher interested in the role of consent in sexual ethics, who heard someone mention my new paper. Since Bob is interested in learning what other scholars have to say on the subject, Bob emailed me and asked to read my paper. I sent it to him, and he read it. One might ask: (2) Did Bob consent to reading this paper? 1 One sometimes hears the advice that sexual consent must be enthusiastic. Read strongly and literally as the claim that if one is not enthusiastic when one says 'yes', then one has not consented, this advice implies that Alice did not consent, and that these sexual interactions are therefore nonconsensual. I take these cases, and the many other real-world cases relevantly like them, to refute that strong view of consent. (It is probably not a view that many people intended literally.) Compare Fischel (2019, pp. 2-3).
2 One difference between these sexual activities is that the kiss is in an intuitive sense something Bob did to Alice, while the blowjob is something Alice did to Bob. (Whether this distinction would survive deep philosophical scrutiny is not obvious; each involved active participation by the other. But there is at least an intuitive difference here.) I'll discuss this distinction further in §6 below. experiencing the sexual activity he wants. 'Consent' feels, at least to some, like appropriate language for the sex cases, but not for the reading case. Why?
One possible answer is that sexual consent is a very different kind of beast from other kinds of consent. A different kind of answer -the one I favour -is that a particular kind of consent-based sexual ethics shapes many of our intuitions about sexual consent, in a way that leads to a distorted recognition of consent. To bring this out, I'll focus first on just what it is that is strange about (2); I'll go on to argue that, contrary to orthodoxy, there is something similarly strange about (5) and (6). This will amount to novel support for longstanding feminist critiques of consent-based sexual ethics.

Presupposition and Consent
This section explores the sense in which it's strange to ask whether someone 'consented' to something they chose for themselves -not in response to someone else's request or suggestion. Let's focus for now on nonsexual cases where the 'consent' language feels off. Bob read this paper because he's interested in the topic. I ran a marathon because I wanted to test my body's limits. My friend flew to Australia because she wanted to have a grand adventure.
None of these cases involved coercion or manipulation. (They could have -but they didn't.) We made those choices as freely and autonomously as anyone does anything. We didn't bend (not even voluntarily) to anyone else's will. Nobody asked or instructed me to run that marathon; I did it for me. This is part of the reason why the 'consent' language seems ill-fitting for these actions. Saying I consented to that marathon makes it sound like someone else asked me to run it. Saying Bob consented to reading my paper makes it sound like he did it at someone else's behest. (Compare Alice, who did consent to read my paper, as a favour to me.) Still, it also doesn't seem right to say that Bob didn't consent to reading my paper. That makes it sound like somebody else made him do it. Likewise for the idea that I didn't consent to the marathon, or that my friend didn't consent to going to Australia. In other words, both (7) and (8) seem pretty strongly to suggest that someone other than me was trying to get me to run that marathon: (7) I consented to the marathon. This pattern is characteristic of presupposition. The use of certain language conventionally presupposes that certain assumptions are taken for granted; if S presupposes that p, an utterance of S doesn't make sense unless we assume that p.
To take a famous kind of example, sentence (9) presupposes that China used to stockpile metals: 4 3 The English verb 'consent' can take a noun phrase or an infinitival verb-phrase -one can speak of 'consenting to a marathon', or of 'consenting to run a marathon'. I do not perceive any difference in acceptability, in any of the data I'm citing, between these kinds of uses; it also sounds odd to say I 'consented to run the marathon', or to say that I didn't. It is likewise fine to speak either of Alice 'consenting to a kiss', or 'consenting to be kissed'. So I do not draw a significant distinction between these grammatical forms, and I use data of both kinds to indicate that I intend my discussion to generalize to both kinds of sentence. Thanks to a referee for drawing my attention to this distinction. I also intend my discussion to extend to the adjective 'consensual' and the noun 'consent' -see fn. 28. 4 I take this example from Beaver andGeurts (2012, p. 2434).
If we don't collectively think or assume that China used to stockpile metals, and I say (9), my utterance is infelicitous. (If you didn't previously have an opinion on the matter, you might feel some conversational pressure to pretend like you'd known all along that China used to stockpile metals.) And one of the interesting features of presuppositions is that presuppositions tend to "project" from various embedded uses of the language. For example, they tend to project from negation, attitude ascriptions, and questions. Note that all of the following, just like (9), carry the presupposition that China used to stockpile metals: (10) China has not stopped stockpiling metals.
(11) Canadian intelligence agencies believe that China has stopped stockpiling metals.
Projection from such embedded contexts is one of the distinguishing characteristics of presupposition, as opposed to entailment or implicature. 5 Other common constructions that carry presuppositions include: • definite descriptions. 'The dog is hungry' presupposes that there is a unique salient dog.
• cleft constructions. 'It was I who loosened the bolt' presupposes that someone loosened the bolt.
• factive verbs. 'Bob is sorry that Alice is busy' presupposes that Alice is busy.
• gender pronouns. 'She is waiting for you' presupposes that the person indicated is female. 6 All of these presuppositions, too, project from various embeddings. (Note that 'she isn't waiting for you' and 'I think Bob might not be happy that she is waiting for you' also presuppose that the indicated person is female.) That presuppositions project from embeddings like negation and questions suggests a response to our puzzle about odd questions like (2) from §1: (2) Did Bob consent to reading this paper? This is an odd question, if, as assumed, Bob read this paper an expression of his own will, and not at anyone else's behest. We don't want to say to say (13), because it suggests that he was doing it at someone else's request: (13) Bob consented to reading this paper. 5 These three projections are discussed in Beaver andGeurts (2012, pp. 2434-5). Beaver and Geurts also discuss others: projection from the antecedent of conditionals ('if China has stopped stockpiling metals, the price will go down'), possibility modals ('China might have stopped stockpiling metals'), and probability adverbs ('China has probably stopped stockpiling metals'). 6 The first three forms are widely discussed in the literature. Eckert and McConnell-Ginet (2013, p. 171) discuss gender pronouns as presuppositional. Thanks to Jennifer Foster.
But it's even odder to say (14), because that suggests he was forced or coerced: (14) Bob didn't consent to reading this paper.
These data are well-explained by the hypothesis that (13) presupposes, falsely, that someone else was trying to get Bob to read this paper. That same presupposition would be predicted to project from the negation (14), and the question (2). 7 So all three utterances involve a presupposition failure, which is why they are defective. 8 More generally, when one consents, one is yielding to another's will. One may do so freely, as when Alice agreed to read my paper. (Or, getting back to the sexual cases, as when Alice agreed to give Bob a blowjob.) But this is different from when one does things out of one's own accord. Consent presupposes the former; it is typically a response to a request, an instruction, a command, or the like. When you are considering doing something at another's behest, the question of consent is an appropriate one. When you are considering doing something for your own reasons, the question doesn't even arise.
I'll say a bit more about what it takes to satisfy the presuppositions of 'consent' in §4 below. First, however, I'd like to connect the thoughts above to the other main topic of this paper: feminist critiques of consent theory in sexual ethics.

Feminist Critiques of Consent Theory
The language of 'consent' is pervasive in contemporary discourse about sexual ethics. It is common in philosophical and other academic discourse, 9 and completely inescapable in discussions of sexual ethics in popular liberal media. This is why I call the orthodox approach to sexual ethics 'consent theory'. 10 But there have been serious critiques of consent theory, challenging whether consent should be considered so central a question as the contemporary consensus assumes.
Some of these critiques develop the complaint that consent is a low bar for morally permissible sex; they worry that focusing exclusively on consent obscures other morally important phenomena. For example, Linda Martín Alcoff has argued that the wrongfulness of many sexual assaults of children cannot be explained by consent theory, since many children are wrongfully sexually assaulted by virtue of acts to which they consent, i.e., "where consent was produced by structural conditions of economic dependency, or was caused by emotional 7 This is one reason to prefer the presupposition strategy to an alternative explanation in terms of pragmatic implicature as per Grice (1975). Implicatures do not in general project from questions and negations; in Grice's famous thought experiments, while 'he has excellent handwriting' famously implicates that he is a poor student, and 'there is a gas station around the corner' implicates that it is open, 'he does not have excellent handwriting' and 'is there a gas station around the corner?' do not carry the same implicatures. Moreover, implicatures are cancellable in a way presuppositions are not -'he has excellent handwriting, and he is an excellent student' is fine, but 'he consented to read my paper, and he asked for it because he really wanted to read it' is not. Thanks to Chelsea Rosenthal for helpful discussion here. 8 We needn't decide what to say about the semantics of sentences involving presupposition failures. We might say with Frege that they fail to express meaningful contents, or with Russell that they are false. Or we might say something else. I remain neutral on these options, insisting only that they are not true.
10 In using this label, I do not assume that there is a specific doctrine, subscription to which is necessary and sufficient for being a consent theorist; one embraces consent theory to the degree to which consent plays a central role in one's sexual ethics. I'll say a bit more about the contours of consent theory in §9 below.
confusion, or was given in an attempt to help or protect others in the family." 11 Audrey Yap (2019) says something quite similar about consensual sexual relationships involving institutional power-over relationships, like relationships between professors and students. These critiques of consent theory are interesting and important, but retain the orthodox insistence that consent is necessary for ethical sex. 12 There have also been more radical feminist critiques of consent theory, suggesting that the emphasis on sexual consent is gendered, heteronormative, or similarly problematic. Here, for example, is Carole Pateman: Consent must always be given to something; in the relationship between the sexes, it is always women who are held to consent to men. The "naturally" superior, active, and sexually aggressive male makes an initiative, or offers a contract, to which a "naturally" subordinate, passive woman "consents." An egalitarian sexual relationship cannot rest on this basis; it cannot be grounded in "consent." Pateman (1980, p. 164) 13 Stereotypes of sexual consent certainly involve a man asking for a woman's consent, as in my examples of Bob and Alice in §1. 14 In one sense this an understandable reflection of the gendered pattern of sexual violence. But the norm is surprisingly robust -academic discussions of consent very often reinforce the pattern, even in nonsexual examples, even though most academics (and all student services websites) pay at least lip-service to the idea that consent is important among all parties to sexual encounters, regardless of gender. 15 So one feminist concern about consent is that it reinforces sexist heterosexual stereotypes.
A related feminist concern about consent theory is that it does not leave room for an important role for women's sexual desire or agency. See e.g. these passages: If marriage and intercourse were clearly beneficial to women, if they were experiences that women, as self-interested, autonomous, rational beings, perceived as appealing and desirable, we would not speak of women's consent, but rather of their desire. Cahill (2001, p. 174) 11 Alcoff (2018, p. 81). Many consent theorists will resist the characterization of such cases as consensual; see §9 below. Alcoff herself seems ambivalent about the 'consent' language; sometimes in such contexts, but not always, she puts the word in scare quotes.
12 See also West (2009). 13 This passage is discussed in both Cahill (2001, p. 174) and Alcoff (2018, p. 83). Pateman's broader point is somewhat different from mine; she ultimately argues that genuine consent requires a fuller freedom and equality than women enjoy. This passage, which fits well into my broader point, actually fits somewhat awkwardly into hers.
14 This is part of a much broader stereotype -consider also the heteronormative conventions about proposing marriage, or asking someone out on a date. 15 To note one rather striking example, in Larry Alexander's influential (2014) "The Ontology of Consent", I count eleven total thought experiments related to whether consent is or is not present. Of these, eight specify, by name, pronoun use, or both, that a man (or boy) is seeking an action that would be a violation of a woman's (or girl's) rights if she didn't agree to it: John wants to take Jane's car, Al tries to poison Alice, Jim wants to borrow Jane's lawn mower, D (male pronouns) wants to borrow V's (female pronouns) car, D (male pronouns) wants to have sex with V (female pronouns), Sam (male pronouns) wants to cross Sally's property, Edith "mentally accepted" sex with Ed, D (male pronouns) inserts his penis into V's (female pronouns) vagina. Two more examples specify female victims of gender-unspecified actors -one person threatens a victim (female pronouns) to have sex, and another threatens W (female pronouns) to engage in prostitution. There is only one example in Alexander's paper that definitively does not involve the question of a woman consenting to a man -it is that of "Tom the bully" (male pronouns, acting in the stereotypical male role), who wants to move "Billy the 90-pound weakling" (male pronouns, acting in the stereotypical female one). This is a particularly vivid illustration, but my sense from the broader literature is that it is indicative of a wider pattern.
The concept of consent relies for its social appeal on the assumption that it stands in for desire. Whenever its use in the sexual arena is questioned, which is mighty seldom, the response is to wave the flag of desire. This is consent's credibility cover. Consent can be considered to include authentic desire, but the term is never used in that context in real life, and nothing limits it to that in law. In social reality, the crucible of meaning, sex that is actually desired or wanted or welcomed is never termed consensual. It does not need to be; its mutuality is written all over it in enthusiasm. Consenting is not what women do when they want to be having sex. Sex women want is never described by them or anyone else as consensual. No one says, "We had a great hot night, she (or I or we) consented." ... My point is, when a sexual interaction is equal, consent is not needed and does not occur because there is no transgression to be redeemed. MacKinnon (2016, pp. 450 and 476) (Not everyone -especially not all of today's younger people -share MacKinnon's linguistic intuitions here. I'll discuss this further in §9 below.) Quill Kukla (2018) rehearses some of the worries above, but also gives a helpful theoretical explanation for why consent theory can distort our understanding of sex. Consent, Kukla says, is a performative speech act that, by its nature, is a response to requests. (This is closely related to my own suggestion in §2.) But, they observe, our near-exclusive focus on consent and refusal when we talk about sexual negotiation has had a deeply distorting and damaging impact on our understanding of sexual ethics and communication. Good-quality sexual communication requires that we do much more with language than request, agree to, and refuse sex. Kukla (2018, p. 75) Kukla goes on to emphasize other ways, besides one person consenting to another's request, that sex might be initiated. For example, someone might accept someone's invitation, or someone's offer of a sexual gift.
I agree with the central elements of Kukla's paper, but I wish to emphasize, as they do not, their radical consequences. Kukla writes at times as if they are advancing the kind of critique I attributed to Alcoff and Yap above -one that complains that consent is just too low a bar for ethical sex. 16 But I think their critique, properly understood, implies not only that consent is insufficient for ethical sex -it's not necessary either. (I do not think this means that nonconsensual sex can sometimes be OK! In §5 I'll explain how the considerations from §2 prevent that implication.) Perhaps because of its obvious clash with feminist orthodoxy, Kukla does not emphasize, as I do, the implication that consent is not a necessary condition for ethical sex. But the argument for this is simple. Consent to Φ, Kukla observes, echoing Pateman, Cahill, and MacKinnon, only makes sense under the assumption that Φ is something someone else is 16 E.g., "Consent, including completely autonomous, unmanipulated consent, is never going to be sufficient to make sex go well -we can consent to all sorts of lousy sex, including demeaning, boring, alienated, and unpleasantly painful or otherwise harmful sex." Kukla (2018, p. 72) In their subsequent (2020) paper, Kukla seems to be working with a different notion of consent that is closer to the 'inflationary' one I'll discuss in §9.
trying to get you to do. 17 But your having ethical sex does not require that someone else is trying to get you to do something. Therefore, your having ethical sex does not require your consent. This sounds like a radical repudiation of contemporary orthodoxy about sexual ethics. (Just how radical a repudiation this is depends on some sensitive questions about the relationship between language and ethics. I'll discuss this issue in §9 below.) Kukla is somewhat circumspect about the implications of this approach, writing that [s]exual activity is only ethical if everyone involved has communicated successfully that they want to engage in it and is doing so autonomously and willingly. If this broad point is all we mean by saying that all sex must be consensual, then consent is a necessary condition for the ethical acceptability of any sexual activity. But this seems to me to be an unfortunate way of describing the principle, since 'consent' is the name of a speech act with a specific pragmatic shape. If I make clear that I would like to accept an invitation or a gift, then I have communicated my autonomous desire to participate; it strains the pragmatics of language to call this 'consent'. At least for the purposes of philosophical analysis, it seems to me worth using the term 'consent' more narrowly and giving a more fine grained analysis of the ethics of sexual discourse. 18 So Kukla emphasizes that standard normative verdicts about particular instances of sex need not be challenged by this framework -the issue is how to talk about them with regard to consent, not whether particular patterns of behaviour are OK. So far, I agree. But this passage also carries an implication I reject: the idea that ultimately, we are discussing a relatively technical linguistic matter, disconnected somewhat from contemporary moral issues. I read this implication, for instance, in the last sentences quoted, which limits the suggestion to one concerning how we theorists should use 'consent' language. I will argue below that, given the presuppositions implicit in 'consent' discourse in the ordinary vernacular, there is good reason to advocate for reform of public discourse on sexual ethics, not merely academic theoretical talk. 19 4 Acting at Another's Behest Consent to Φ, I suggest, presupposes that someone else is trying to get one to Φ. Most paradigmatically, consent is a response to requests, but the language of consent can be appropriate in other cases too. Here is an example where it makes sense to talk about whether or not there is consent, even though there is no request: Suppose that Calvin is in 17 How exactly to spell out what it is for someone to be 'trying to get you to do' something is a tricky question. As Kukla (2018) puts it, consent is always a reaction to a request. As I'll explain in §4, I think this is too restrictive -one can consent to someone instructing one to Φ, or ordering one to Φ, or perhaps even one's insinuating that they want one to Φ too. But I agree with Kukla that we do not "consent" to invitations or gift offers, and we do not "consent" to do things when we choose them autonomously, at nobody else's behest.
18 Kukla (2018, p. 92), emphasis added. Thanks to two referees for encouraging me to discuss this passage and contrast it with my position. 19 Here are two more differences between Kukla's discussion and mine. First, Kukla does not seriously engage with what I take to be the central objection to the feminist critiques of consent theory -the idea that they imply, implausibly, that sex without consent can be permissible. The detailed discussion of presuppositions and projection from negation, I think, is important to fully understand the view that Kukla and I share. Compare my remarks on John Gardner's related approach in §8. Second, Kukla says that consent is specifically a response to requests. As I'll discuss in §4, I think 'consent' language is appropriate to a broader range of circumstances.
Moe's path, and Moe wants him to move. Moe might put his hands on Calvin's shoulders and push him in the direction he wants him to go. In such a case, the presuppositions of (15) are met: (15) Calvin consented to move out of Moe's way.
I say the presuppositions of (15) are met; that doesn't mean it's true. Whether this is so depends on Calvin: if he recognizes Moe's desire and freely decides to cooperate in fulfilling it, it is; if he resists, it is false, whether his resistance is successful, or he is overpowered. So too if he moves reluctantly, out of the fear of further possible violence. Some cases in between will be difficult to categorize. But the presuppositions are met by Moe's attempt to move him.
So consent doesn't presuppose requests in particular; the kind of will that consent involves deferring to can be manifested in various ways. (The same would be true if Moe had issued a command: "move!" Then Calvin could consent, or not, to Moe's demand.) By contrast, one might also receive suggestions and invitations that are expressive of someone else's desires or interests, where the presuppositions of consent are not satisfied. Return to Bob, who read my paper. In §1 I said he decided by himself that he wanted to read my paper, and asked me for permission to do so; let's now suppose instead that I knew of Bob's long-standing interest in the topic, and thoughtfully offered to send him my paper and let him read it. He gratefully accepted my offer, and read my paper. The strange (2) is no less strange for my having been the one who raised the idea: (2) Did Bob consent to reading this paper?
Or suppose I hear about a hiking trail and ask you if you want to join me. This is an invitation. An invitation is or involves, at least to some degree, an expression of the inviter's will -typically, by inviting you, I manifest my desire that you come. But, at least in many cases, if I say "hey, I was thinking about hiking the Baden Powell this weekend, would you like to come?", that's not a request. 20 More to the point, if you welcome my invitation and accept it, and we go hiking together, the presuppositions of (16) are not met: (16) You consented to hike the Baden Powell Trail with me.
To be sure, there are possible cases of invitations that will satisfy the presuppositions of (16) -these are generally less positive experiences than the ones I have in mind. If you feel pressured by my invitation, or worry that some bad consequence might follow from declining, and so reluctantly agree to come, then the presuppositions of (16) will obtain. But if you are my friend and equal, and accepted my invitation because you like hanging out with me (or because you're indifferent to me personally, but like hiking) the presuppositions of (16) fail.
I've said that 'consent to Φ' language presupposes that Φ is or would be at someone else's behest. This is meant to cover requests, commands, and the like, and to contrast with cases in which someone decides to Φ on their own, or in response to an offer or a free invitation. If someone requests that you Φ, or orders you to Φ, or attempts physically to make you Φ, the behest presupposition is satisfied, and it makes sense to talk about whether you consent to Φ. If you decide to Φ all on your own, or if someone offers to let you Φ or invites you to Φ, the behest presupposition is typically not satisfied, and the question of your consenting to Φ is infelicitous. 21 I am now in a position to clarify more precisely how I think consent relates to ethical sex. The question of consent is obviously applicable to sexual assault: when one person forces or coerces another into unwanted sexual contact, they are imposing their will on their victim in a way that satisfies the presuppositions of the 'consent' language, which is why it is true to describe sexual assault as sexual contact without consent. So consent can play a significant role in explaining what goes wrong in sexual assault.
But I do think there can be cases of permissible sex where the presuppositions of 'consent' are not satisfied. Indeed, many of the paradigms of good sex are like this. Most sexespecially most good sex -does not arrive from one person acquiescing to another's behest. Sexual negotiations that involve suggestions, invitations, brainstorming, the exploration of shared ideas and fantasies, etc., needn't involve satisfying the presuppositions of 'consent'. (Asking if someone is interested in trying a new sex toy, for example, needn't be a request any more than asking if someone is interested in going hiking this weekend need be. 22 ) The granting of sexual requests -which is consent -can certainly be a part of ethical sexual negotiation, but it is far from the whole story. 23

Explaining Away Orthodox Intuitions
Its inconsistency with consent-emphasizing language in contemporary society will make this picture a tough pill for many to swallow. Many will complain, first, that it is repugnant to allow that sex without consent can be permissible; that nonconsensual sex is a very serious wrong. Second, they will attribute the problems identified for consent theory to an overly simplistic view of consent: they may agree that professors may not have sex with their students, but argue that the power differentials in such cases are inconsistent with valid consent. On the other hand, on this view, one will typically hold that cases of unproblematic sex like Alice and Bob's, where I think "consent" is an inappropriate label, are in fact consensual. In this section and the next, I'll respond to the first reply; in §9 I'll react to the latter. I say, against orthodox consent theory and everything you'll read on your university's student services website, that consent is not a necessary condition for morally permissible sex. Some may find this somewhat shocking, because they think it suggests that I think it can be OK to have sex without consent. But I don't think that either. 21 'Consent' may carry other presuppositions as well. As Samia Hesni pointed out to me, the language of 'consenting to Φ' has a tendency to communicate that Φ is in some way out of the ordinary, or beyond what would typically be expected of someone. And as Cat Saint-Croix pointed out to me, 'consenting to Φ' has a tendency to communicate that Φ is the kind of thing we think might be harmful. The idea that 'consent' presupposes features along these lines, in addition to a reaction to someone else's behest, strikes me as plausible, but further exploration of it is beyond my present scope. In §6 I will discuss the idea that 'consent' presupposes that something is being done to one.
22 As a referee points out, it can under certain circumstances be felicitous to describe responses to suggestions with 'consent' language, as in 'she suggested that they holiday in the Maldives, and he consented'. I think this is because 'suggestions' are sometimes rather request-like, and sometimes much more neutral. The 'consent' language can be felicitous, but it does introduce the presupposition that he went along in deference to her -not that it was something they decided collectively.
This has the air of a contradiction, but it's a paradox that the machinery rehearsed in §2 is well-equipped to resolve. I have argued that the language of 'consenting' to Φ presupposes that somebody else is trying to get one to Φ. And since this presupposition projects from under negation, saying one doesn't consent to sex presupposes that the sex is at someone else's behest, just as much as saying that one does consent does.
Recall that in my thought experiments from §1, Bob read this paper and got a blowjob, both at his own behest. When it came to the former, it wasn't too hard to see that saying he consented to read my paper wasn't right, but saying he didn't consent to it wasn't right either. My diagnosis was that each carries the false presupposition that someone else was trying to get him to read the paper.
I've found, by asking colleagues and students, that intuitions are murkier about the sex case. Many people are happy to endorse (17) and (18), even though each event happened at Bob's own behest: (17) Bob consented to the kiss.
(18) Bob consented to the blowjob.
Both on general grounds, and in light of the feminist critiques marshalled in §3, I think we should reject both (17) and (18) as carrying a false presupposition, just as we rejected (13). Someone only consents to what they are asked, instructed, or otherwise externally encouraged to do. But that doesn't mean we should accept (19) and (20): (19) Bob didn't consent to the kiss.
(20) Bob didn't consent to the blowjob.
For this pair carries the same false presuppositions the previous, affirmative pair did. Rejecting consent as a necessary condition for good sex doesn't require allowing that nonconsensual sex can be okay. The entire discourse of 'consent' carries presuppositions that fail in many cases of good sex. There can be good sex that we can't correctly describe as 'consensual', even though it isn't nonconsensual either -these are cases where the shared presuppositions of any 'consent' language fail.

Comparison to Cousin Views
A number of recent authors have suggested that 'consent' language carries substantive presuppositions, and that this is relevant for sexual ethics. This section situates my proposal with respect to other recent discussions in the literature, and emphasizes the similarities and differences. Readers more interested in the direct questions about sexual ethics, and less interested in how my proposal fits into the broader literature, should feel free to skip ahead to §7.
I have suggested that 'consent' presupposes that something is done at someone else's behest. The closest extant view to mine of which I am aware is Quill Kukla's (2018), which I have already been discussing. Another idea is that 'consent' presupposes an active-passive dichotomy: it only makes sense to talk about consenting to things that are done to one. Here, for instance, is another passage from the same Catherine MacKinnon paper cited in §3: Intrinsic to consent is the actor and the acted-upon, with no guarantee of any kind of equality between them, whether of circumstance or condition or interaction, or typically even any interest in inquiring into whether such equality is present or meaningful, at least in the major definition of the most serious crime. MacKinnon (2016, p. 440) Similarly, John Gardner (2018, p. 60) writes that "the question of consent belongs to sex individualistically, even solipsistically, conceived, to sex conceived as something that one person does to another (even if, in the course of their sexual encounter, the individuals concerned take scrupulously equitable turns in being the doer and the done to)." And Tanya Palmer (2017, p. 13) writes that "[c]onsent is ... a minimal, reactive form of participation. In the specific context of sexual activity, a consent framework implies that sex always involves one (active) person doing something to another (passive) person." Call the idea that it only makes sense to talk about consent to that of which one might be a passive recipient, the passivity presupposition. This is a distinct idea from the behest presupposition that I have defended. Theorists have not always kept these ideas separateperhaps because many stereotypes of heterosexual sex satisfy both presuppositions. 24 But they can come apart. If I ask someone to punch me in the stomach -perhaps I am testing my constitution -if they do so, they do something to me, but at my behest. Likewise, when Bob asks Alice for a blowjob, it is something she does, but at his request.
I defend the behest presupposition, but I think there is something to the passivity presupposition, too. At least in many cases, talking about whether S consents to Φ will convey the idea that Φ is something done to S, rather than something S does themself. And in some cases, 'consent' language seems appropriate when the passivity presupposition is satisfied, even though the behest presupposition fails. (Consider consent to a tattoo, or to a medical procedure.) Although I recognize that such cases put pressure in the direction of the passivity presupposition and against the behest one, I am impressed by the force of some of the other kinds of cases I've discussed, which pull against it. It is important to ask, for instance, whether Alice consented to give Bob that blowjob, since she did it at his behest. 25 As for tattoos, I want to bite the bullet and suggest that it is something of a misnomer to describe the customer of a tattoo parlour as consenting to be tattooed, if they're not having it done at someone else's behest. The legalistic framework in which contemporary businesses operate has a tendency to normalize this kind of language -much as orthodox consent theory does the same for sexual consent -but I do want to suggest that there is something linguistically off about talk of consenting to the tattoo that one has picked out for oneself. (In light of artistic license, one might consent to some of one's artist's particular choices.) Something quite similar goes for consent to medical procedures (although here a more robust legalistic idiolect has certainly developed). This needn't mean that the label of a 'consent form' is inapt -one does, when one asks for a tattoo, consent to waiving certain rights and liability claims. And this waiver is indeed granted at the parlour's request. 26 There is a strand of thought that assimilates sexual consent to this kind of legalistic framework of liability waivers, but this, I think, is far from desirable. 27 Language is imprecise and constantly changing. It may be that there is not one clear and coherent notion of consent at play in all of these examples. Perhaps the behest presupposition tends to be more dominant in some contexts, while the passivity presupposition prevails in others. Nevertheless, the behest presupposition, I think, is strong enough, and active enough, in a wide enough example of cases to be philosophically interesting, and, I think, of moral significance when it comes to the best language to discuss sexual ethics. I will return to this issue in §9 below.
In his (2018) paper, John Gardner invokes the passivity presuppostion in support of a conclusion similar to mine. I will contrast my approach more specifically with his in §8. First, however, I would like to clarify the sense in which I wish to reject consent as a necessary condition for permissible sex.

What is it for Consent not to be Necessary?
Language that carries substantive presuppositions, especially those corresponding to stereotypes we may assume by default, makes careful statements of general theories that suspend those presuppositions difficult. In this section I will clarify more precisely what I mean when I say that consent is not a necessary condition for permissible sex.
If consent were a necessary condition for permissible sex, then every example of permissible sex would be one in which there is consent to that sex. 'Consent' language, I have argued, requires that one is doing something at someone else's behest; since there can be permissible sex that doesn't involve one person yielding to another's request -the kind of sex where 'consent' language won't even come up -we cannot maintain that consent is necessary for ethical sex.
As highlighted above, the sex in question won't properly be characterized as nonconsensual sex either. I assume that 'consensual' just means 'with consent', and carries the same presuppositions, and that 'nonconsensual' is the negation of 'consensual', and also carries the same presuppositions. 28 So the part of the consent theory that says that all nonconsensual sex is morally impermissible, is perfectly consistent with everything I have said. (And 26 Liquid Amber, one of the local tattoo parlours in my neighbourhood, asks its clients to sign a 'Client Waiver, Release & Consent to Tattoo' form. But the actual text of the form, interestingly enough, does not use the word 'consent' as applied to the tattoo. Most of the form concerns acknowledgement of risks and the waiver of liability -a request the parlour makes of the client. The only time 'consent' occurs, other than in the title, concerns the client's 'consent' to the use of their photographs in promotional material. (As of 3 June 2020, that form is available at https://www.liquidambertattoo.com/thewaiver.) Some 'consent forms' do discuss 'consent to receive a tattoo,' but this is not by any means universal.
27 Cf. . 28 This assumption is not mandatory, and I concede that the linguistic data is less than obvious here. I am not sure whether 'Bob read the paper consensually' is as infelicitous as 'Bob consented to read the paper.' However, given the close etymological relation, and the apparently analytic inferences, between consensual and with consent, I don't find a view that posits significant presuppositional differences between these words to be very plausible. (A possible view like this is described Chadha (2020, p. 3, fn. 9): "On that way of talking, a sexual encounter can be consensual without the individuals involved giving consent.".) This way of talking does not strike me as plausible or attractive. Certainly the orthodoxy to which I am responding does not draw a clear distinction between 'consensual' and 'with consent'. The same goes for the noun form of 'consent': I assume the naive connection between the noun and the verb: 'consent' is what happens when one 'consents'. Thanks to Gretchen Ellefson and an anonymous referee for encouraging me to discuss this point.
indeed, I endorse it.) That is to say, in any situation where the question of consent is even applicable, if one doesn't agree to the sexual contact being asked of one, but that contact occurs anyway, a serious moral wrong -sexual assault -has occurred.
One might be tempted to express this thought in terms of a distinction between 'nonconsensual sex' and 'sex without consent', holding that the former is always bad, but that the latter can be fine when the presuppositions of 'consent' are not met. My view would be simpler to state, were this so. Unfortunately, in ordinary English, I think 'sex without consent' carries the same presuppositions that 'nonconsensual sex' does. Presuppositions project from a wide variety of grammatical locations, and I think that embedded within the 'without' preposition is among them. (Even raising the question of whether it's OK for me to go to the party without my wife presupposes that I have a wife.) So just as I think nonconsensual sex is always wrong, so too do I think that sex without consent is always wrong. I don't draw a distinction between those locutions.
What I do think can be fine is sex that occurs without the presuppositions of 'consent' being satisfied -as when, for example, sex partners mutually undertake a sexual activity out of a shared desire, or when one partner freely accepts another's sexual invitation. Since some of these are cases where sex is morally permissible, but where 'consent' language is inapplicable -and so a 'consent' ascription would not be true -I want to say that consent is not a necessary condition for permissible sex. I am in effect attempting to use the 'is not a necessary condition for' operator in a way such that the presuppositions of 'consent' don't project from within it. 29 This operator is not a colloquial expression in conversational English; it is a philosopher's term of art; I hope it is not too much of a misuse to employ it in this way.
An analogy that may help to illustrate the logical space I am exploiting. Saying someone 'stopped' doing something presupposes that they were previously doing it. So (21) presupposes that Rohit was previously texting his friend: (21) Rohit stopped texting his friend.
Texting while driving is dangerous, so one might be tempted to sign up to a norm like this one: (22) Rohit must stop texting his friend, before driving.
But this rule is only appropriate if we assume the presupposition of (21) and (22): that Rohit was previously texting his friend. Note that Rohit does not comply with (22) if he drives while never having texted anyone; so if we do not want to assume that he was, (22) is a bad rule. And if we set out a universal generalization of (22) like (23) Everyone must stop texting their friend, before driving, we'd be presupposing that everyone who might drive was previously texting their friend. Stopping texting one's friend is not a necessary condition for safe driving, because it is consistent with safe driving that one was not previously texting one's friend. 30 29 If the presupposition did project, then I would reject the sentence, 'Consent is a necessary condition for morally permissible sex,' but also reject 'Consent is not a necessary condition for morally permissible sex.' Then one would have to ascend metasemantically to express the key idea; one would say that there can be cases of morally permissible sex such that the sentence, 'that sex involved consent' would not be true. 30 One might challenge my claim that (23) presupposes that everyone was previously texting their friend. Perhaps, as a referee suggested to me, only a weaker presupposition holds, to the effect that generically, This example feels a bit artificial, in part because it rather mysterious why anyone might assume that everyone who might drive was previously texting their friend. But consider a similar case where the presupposition is more tempting -and where it reinforces a harmful stereotype. Here is objectionable advice that one might read in an academic lifestyle guidebook: (24) Every professor should be grateful for the support his wife gives him.
One problem is extensional: (24), like (23), miscategorizes certain cases. Someone who doesn't have a wife cannot satisfy this norm. But there is also a deeper problem: the norm itself reinforces sexist, heteronormative, amatonormative, and mononormative stereotypes, by presupposing, of every professor, that (a) they are male, and (b) they have a unique wife. Even if we grant for the purpose of argument that any professor who has a wife should be grateful to her, we should think that (24) is a bad rule. The statement and reinforcement of a norm like this will explicitly censure ungrateful husband professors, but it will implicitly exclude professors who are unmarried, married to someone who isn't a woman, and/or non-male. 31 In exactly the same way, setting out a rule like (25) invokes the presupposition that everyone 32 who might be having sex would be trying to get their partner to do it at their behest: (25) Everyone who has sex should get consent first. This is a harmful assumption to presuppose, insofar as we wish to resist rape culture. I don't think it can be OK for anyone to have sex without consent -but I reject (25) for the same reasons as I reject (23) and (24). I reject consent as a necessary condition for ethical sex because I think it carries that harmful presupposition, and miscategorizes cases where the presupposition fails -not because I think nonconsensual sex can be OK.
The consent theorists' mistake here, I think, derives from an over-emphasis on the wrongmaking feature of sexual assault. They are right that the lack of consent is a critical piece of the picture for the wrongness of sexual assault, but they conclude from thismistakenly in my view -that the presence of consent is necessary for, and explanatory of, the permissibility or goodness of good sex. This would be a good inference only on the assumption that the question of consent is relevant -and the presuppositions of consent language are met -in all sex. But, I have argued, this is not so.

Comparison: Teamwork Sex
As I mentioned in §6 above, John Gardner (2018) has recently defended a view similar to mine. Like me, Gardner argues that 'consent' language carries substantive presuppositions that ill-suit its central role in contemporary sexual ethics. Like me, Gardner is interested would-be drivers were previously texting their friend. I do not agree with this suggestion; I think the 'everyone' in (23) makes it explicit that all drivers who conform with the norm must have met that condition, namely having stopped texting their friend. But the weaker claim is actually enough for the purposes that follow, so I'll keep that idea live in this section in the next two footnotes. 31 And even if, per my referee's suggestion, the presupposition were merely a generic one, as opposed to a universal one, to the effect that a normal professor is male and has a wife, the norm would be bad for very similar reasons: it marginalizes professors who do not fit the stereotype.
32 Or normal people.
in venerating a kind of sexual activity that is inconsistent with the presupposition. But Gardner's discussion is different from mine in important ways. Most obviously, Gardner grounds his critique of consent theory in the passivity presupposition, not the behest presupposition. He writes: By consenting, one is placing oneself in the position of patient and the other in the position of agent, so far as what is consented to is concerned. From there, one can quickly see that the question 'was there consent?' presupposes an asymmetry of exactly the kind that I suggested is not to be found in good (teamwork) sex. It presupposes that the sexual activity was not fully agent-agent symmetrical. Gardner (2018, p. 58) "Teamwork sex", according to Gardner, is sex that the partners literally do together, as a kind of collective agent. The problem with consent theory, according to Gardner, is that it cannot make sense of what is good about sex that one person isn't doing to another. Sex that happens at one person's behest may often be sex that that person does to another, as opposed to a team action, but it needn't be: team actions can be initiated by individual requests. (Suppose I ask you, 'will you please dance the tango with me?', and, as a favour to me, you agree, and we dance. We perform the dance as a team, but you took your part in it at my behest. On my view, 'consent' language is perfectly appropriate for describing our dance; on Gardner's it won't be.) So, because we posit different presuppositions to 'consent' language, our views make different predictions.
Another point of similarity between Gardner's approach and mine is that like me, Gardner wishes to reject the orthodox claim that consent is necessary for morally permissible sex. His rationale is similar to mine. 33 In one important respect, Gardner is also much more ambitious than I am: the "teamwork" that Gardner esteems, he considers to be a necessary condition for good sex. I do not make any such claim, either for Gardnerian teamwork nor for my corresponding notion of acting at no one else's behest. I have emphasized that many paradigms of good sex are collaboratively elected, or initiated as the result of an open invitation, in a way that fails the behest presupposition. But I make no blanket condemnation of the consensual granting of sexual requests. Undoubtedly the world would be better if there were more sex inconsistent with these presuppositions, and less that matches them. But I see no reason to say that no one should ever ask for sexual favours from one's partner, or that, if such a request is granted, the resulting sex is bad. 34 Sex, like most nice things, can be appropriately 33 Gardner writes: "Here I am advancing the more explosive proposition that, when the sexual going is good, consent is also unnecessary. Before you explode, bear in mind that my case proceeds, not from the thought that consent is too high an expectation for our sex lives, but rather from the thought that it is too low an expectation. Ideally, I suggest, the question of consent does not arise between sexual partners." Gardner (2018, p. 60) Unlike me, Gardner does not explain or engage with presupposition and its characteristic projection from various embeddings. But some of the details of presupposition are essential for understanding the view. Note that on its face, Gardner's suggestion that the reason consent is unnecessary for permissible sex is that it is "too low an expectation", crumbles under logical scrutiny. Being too weak a standard could never be a reason to deny that it is not a genuine requirement. Absent some discussion of the mechanics of presupposition, and its projection from negation, Gardner's suggestion here looks like a non sequitur.
34 Thanks to Carrie Jenkins, Cat Saint-Croix, and Mona Simion for thoughtful discussion here.
established in any of a variety of ways.

Inflationary Notions of Consent
What of consent theorists' positive strategy? Consent theorists do not typically reject sentences like (18) as suffering from a presupposition failure; on the contrary, it is standard to interpret a sincere, sober, uncoerced request for a blowjob as a strong indication of consent to a blowjob. 35 Likewise, consent theorists are typically happy to describe the kinds of sexual negotiation I discussed in §4, where the presuppositions of 'consent' are not satisfied, as consent-involving. Indeed, the rhetoric of 'enthusiastic' consent is designed to valourise just these kinds of cases.
There is a general tendency in discussions of contemporary sexual ethics to treat 'consent' as a label for whatever it is that we think makes sex morally permissible. 36 In response to ethical worries about, say, sex that an employee agrees to have with her employer, there is a temptation to describe the power imbalance as inconsistent with genuine consent. In response to ethical worries about someone reluctantly agreeing to unwanted sex, there is a temptation to describe enthusiasm as necessary for consent. And in response to considerations like the ones I've offered, which I take to suggest that there are cases of ethical sex where the question of consent doesn't even get off the ground, the consent theorist will say there is consent.
Words change their meanings over time, and there is a tradition of attempting to use 'consent' language in the inflationary way described. Based on my observations and informal polls, today's younger adults are much less likely than those in MacKinnon's generation to feel the force of her suggestion, quoted in §3, that "[s]ex women want is never described by them or anyone else as consensual." Perhaps the language is shifting. I have seen some evidence that this tendency is particularly pronounced in some queer communities.
Nevertheless, I have reservations about the attempt to salvage consent theory by altering the contours of the notion of 'consent'. Here are two. 37 First, if one understands consent in a broad way, encompassing all relevant moral constraints on ethical sex, the dictum that sex requires consent becomes tautological. It is neither informative, nor a helpful guide to decision-making, to be told that for sex to be permissible, it needs to have all the features that are required for sex to be permissible. Consent is supposed to explain moral features of sex, not merely redescribe them. On the broad understanding of consent, describing an instance of sex as nonconsensual would just be another way of saying that it is a sexual violation. It is of no use in identifying such violations, to someone who doesn't already know sexual right from sexual wrong.
Second, presuppositions can be both stealthy and sticky. If I am right that 'consenting to Φ' language typically carries the presupposition that someone else is trying to get you to 35 Cf. Pallikkathayil (2019, p. 4): "I might ... ask you to move my car so that I can avoid a parking ticket. A presupposition of this request is that I consent to you moving my car." (Pallikkathayil is not, I think, using 'presupposition' in the same specific linguistic sense I am -note that the consent communicated would not project from e.g. negation: 'I am not requesting that you move my car' certainly doesn't convey consent to move my car.) 36 Or perhaps more precisely, whatever it is that makes sex qua sex morally permissible. Sex might be 'consensual' in even the strong sense contemplated, yet still immoral because e.g. it breaks a promise.
37 In addition to the more central reservations given in the main text, I also worry that it is somewhat ad hoc, requiring a sui generis notion of sexual consent, divorced from the more general notion of consent. The idea that affirmative enthusiasm is necessary to consent to reading a philosophy paper is not remotely plausible, so holding that it is for sex requires special treatment of sexual consent. Compare Gardner (2018, p. 59) on hairstylists et al.
Φ, we should expect the language of consent to have a tendency to continue to communicate that thought, even at an implicit level, even as we start to use the language in a different way. The idea of deference to another is baked in deep to the idea of consent, at the level of the social imagination. 38 So unless and until the language changes completely, such that the presupposition is gone altogether, one expects the invocation of 'consent' language as a general norm about sex to continue to convey and normalize the assumption that sex is something that one person asks for and another grants or refuses. This will reinforce the sexist stereotypes described in §3 and §7, and potentially contribute to bad norms. 39 For this reason, it is often much easier to change which words one uses than it is to change what presuppositions those words carry. Although language is ultimately conventional, the conventions run deep. The English 'he', for instance, presupposes that the referent is male. In cases where the gender of the person in question is unknown or unspecified, one might, with a nod to gender-inclusivity, attempt to use this word without its presupposition, intending to convey no assumption about the gender of the doctor with a sentence like, 'when you see the doctor, give him this note.' Some 20th-century English style guides supported the gender-neutral 'he' on such grounds. 40 As feminist-informed contemporary style guides point out, this was a mistake. The maleness presupposition of the pronoun 'he' wasn't suspended, just because some speakers and writers decided it should be possible to use that word in that way. At best, the presupposition was suppressed, making its operation harder to see. This was not progress. Better to reform the actual use of the language, using different gender-neutral pronounsin contemporary English, 'they' is by far the most successful and available. 41 If one wishes to stop communicating what is presupposed by a given bit of language, it is far easier to use different language than it is to change the meaning of the language to strip it of the presupposition.
Many European languages include a distinction between formal and informal secondpersonal pronouns; these are used to express familiarity or distance, and to reflect relationships in the social hierarchy. In German one addresses a friend or a child as 'du,' but a customer or social superior as 'Sie'. In such a language, it is quite difficult to speak to someone without communicating something about their perceived social status. Insofar as a person or a society might wish to avoid this eventuality, one might prefer not to have these hierarchical structures encoded in the presuppositions of commonly-used words. In theory, one possible strategy would be to collectively decide, as a society, that the words in question will no longer carry such a message. As a practical matter, this is not possible -at least 38 See Medina (2013) or Yap (2017). Thanks to Stacey Goguen for discussion. 39 Compare MacKinnon (2016, pp. 455-6) and Gardner (2018, pp. 68-9). My point here is also analogous to some of the discussion from feminist discussions of pragmatics and pornography, in e.g. Langton and West (1999). Langton and West argue that pornography is a kind of speech act that can subordinate women by normalizing women's subordination; McGowan (2004, pp. 108) points out that this can happen in a particularly insidious way when it does so despite neither the speaker (the producers') nor the hearers intending or even recognizing this effect. I think the accommodation of the stealthy presuppositions of 'consent' language are excellent candidates for manifestations of conversational 'exercitives' in McGowan's sense. Thanks to Cam Gilbert for drawing my attention to this parallel. 40 Fowler (1926, p. 648), for instance: "Their should be his; & the origin of the mistake is clearly reluctance to recognize that the right shortening of the cumbersome he or she, his or her, &c., is he or him or his though the reference may be to both sexes." Huddleston and Pullum (2002, p. 492) gives an overview of the debate, including arguments that this androcentric usage was at best 'purportedly' gender-neutral. 41 Some English speakers use neologisms like 'zie', but they remain marginal. In other languages, novel pronouns have proven much more successful -the Swedish 'hen,' I am told, has achieved widespread use in just the past decade. Preliminary evidence suggests its use diminishes androcentric bias -see Tavits and Pérez (2019).
not without generations of committed use. Just saying, 'we'll continue to use Sie to address some people, but we'll use it in a way that does not carry the honorific presupposition,' is not a viable strategy.
It is not surprising, then, that when, in the 1960s, Swedish society collectively decided, on egalitarian political grounds, to dispense with their formal second-person pronoun, they did so by changing the use, not by changing the connotations of the words. Swedish society made a rather deliberate decision to dispense with the formal 'Ni', employing the informal 'du' for all singular second-person uses. Linguistic reform along these lines is certainly possible, even when tied to specific political ideologies, as in the case of the Swedish 'du'. 42 So I do not hold much optimism for the strategy of retaining the central use of 'consent' language in sexual ethics, while attempting to dispense with its presuppositions. Presuppositions are just not so easily dispensed with.
And when the presupposition remains at a hidden level, it can lead to significant harms. Some contemporary discourse about consent is used in a way that does not require that one is acting at someone else's behest, but this is far from universally the case. Quite a lot of contemporary discourse around consent pretty clearly does retain presuppositions -or even more explicit commitments -along such lines. My own university's student services website, for example, includes this language: We know -it's hard to make the first move, but when you do, it's important to ask for consent first. Sexual consent means that the person who initiates sexual activity is responsible for making sure that their actions are okay with their partners. 43 This is explicit in its assumption that consent is about someone's response to a sexual overture. (It is implicit in its assumption that sexual encounters are begun by one person requesting another's permission.) Implicitly or explicitly, consent theory continues to normalize the idea that sex is something that happens when one partner agrees to give it to another.
Worse, the idea that the responsibility rests only with 'the person who initiates sexual activity' implies that there can be no sexual violation against someone who initiated, if sexual violation is always a matter of consent. This language literally suggests that 'she started it' could be an adequate defence to a sexual misconduct charge. 44 This is part of the reason why, as I alluded to in my discussion of Kukla in §3, I think these arguments, while focused on linguistic matters, are not merely verbal in the sense of having limited normative significance. The choice of which words to use to discuss sexual ethics has practical implications. I said above that the project of this paper has radical implications with respect to sexual ethics. Someone might think, as a referee for this paper suggested to me, that this is something of an overstatement -that my proposal is at 42 Thanks to Anna-Sara Malmgren for discussion of Swedish. 43 https://students.ubc.ca/ubclife/consent-clear, accessed 3 June, 2020. 44 This would presumably follow even if, after starting it, she changed her mind and decided she wanted to stop, and her subsequent protestations were ignored. This confusion is bad enough when given to students as educational material, as cited above, but it can also appear in official policies, which compounds the harms. In 2019 the University of British Columbia Board of Governors proposed an amendment to that university's sexual misconduct policy, which would have codified this 'initiator' language into official policy. This language implies, potentially disastrously, that if one party 'initiates' sexual contact, then changes their mind, their lack of consent is irrelevant for the purpose of identifying sexual misconduct! Udwadia (2020). At the time I am writing, I do not know whether this new language will be ratified. (I submitted a comment to University Council outlining this and other problems with the proposed amendments in January 2020.) most "terminologically radical", not "normatively radical". While it is true that it is not my project to recategorize orthodox moral classification of sexual activity -I am not saying that lots of sex that consent theorists thought was good is in fact bad, or vice versa -I do take seriously normative considerations about the use of language itself. Words are powerful; consequently, arguments about which words to use can have significant normative implications. 45

Conclusion: Modern Love and Consent
There is at least some anecdotal evidence that the current contemporary emphasis on affirmative consent is actively reinforcing some of these sexist norms. I will close with such an anecdote. As a referee reminds me, a curated anecdote alone is of limited evidential value, but I think it can help illustrate the kinds of possible harms that I am describing -and some readers may share my sense that this story is representative of something broader.
In a 2018 Modern Love column, Courtney Sender (2018) describes a perceived cultural difference in a sexual encounter with a younger man: We had met on Tinder. I was nearly 30 and he was 24, but our age gap somehow seemed a lot larger than five years. Not because he acted especially young. It was more that when it came to sex and foreplay, he acted so differently from guys my age, asking for my consent about nearly everything.
She goes on to describe bemusement at repeated requests for permission, at more stages of their encounter than she'd expected. She theorized that "[s]omewhere in our five-year age gap, a dramatic shift must have taken place in sexual training." Sender describes becoming more used to the verbal requests for permission, and coming to perceive them as an expression of care. But in a subsequent sexual encounter, when she tried to model his behaviour in her actions towards him, he balked. She asked his permission to take off his belt.
His hips were arcing toward me, but I paused. Learn, I told myself. I said, "Is this O.K.?" He was taken aback. "I ask you that," he said.
"Why?" "Because I'm the one who could make you do something you don't want to do," he said. "Not vice versa." At least sometimes, contemporary consent theory seems to be communicating the thought that consent implies a kind of imbalance of interests and autonomy. Insofar as we do not wish to further normalize such an imbalance, feminist theorists, buttressed by my observa-