Wait, in what sense are they squaring off? How are their positions about this topic in conflict with one another?

Lockwood’s unwillingness to recognize distinctions between kinds, or degrees of, sexual assault, seems implausible to me. Of course there might be an individual who’s more traumatized by an unwanted kiss than another individual is traumatized by forcible penetration, to use her example. But these unusual exceptions don’t undermine an otherwise obvious distinction. Consider: A is so tough and B so fragile that B is more traumatized by a soft slap than A is by knife stabbing. Does that mean knife stabbings don’t generally involve a greater degree of violence than slaps?

It’s a consequence of Lockwood’s view that it would be no more disturbing to learn that 20% of women were forcibly penetrated than to learn that 20% were forcibly kissed. No one actually believes this.

JRS, I don’t think that’s a consequence of Lockwood’s view. It might be a consequence that it should be no more disturbing, but that’s not to say that is would be no more disturbing. I think, though, the consequence of Lockwood’s view is rather that this sort of limited description is not enough to determine the kind of harm done. And in distinguishing the kind of harm done, it’s not just a matter of how the wrongful action interacts with the toughness or weakness of the one acted upon–it can also be a function of, e.g., the type of relationship between a victim and assailant (for example, being assaulted by someone you trust may be more traumatic than being assaulted by a stranger in virtue of common psychological dispositions), social dynamics and context (e.g., because of cultural norms of masculinity men who are assaulted will face certain challenges women will not), and so on.

Thanks to both Lockwood and Tooley for doing this!

If it would make you feel any better, I know some victims of rape who think that to not take Lockwood’s view seriously would be to do them, and other victims of sexual violence, a disservice (which is not to say, of course, that you’d have to agree with it).

Kathryn, if your previous post is intended as a defense of Lockwood’s claim that data on sexual violence shouldn’t distinguish between forcible penetration and unwanted kisses, then it relies on the following premise:

– If an act of kind x is neither necessarily more harmful nor less harmful than an act of kind y (because, when certain factors are present it’s more harmful, when other factors present, it’s less), then we shouldn’t distinguish between x and y in our data.

This premise implies that our research shouldn’t distinguish between any harmful acts. So in particular, we shouldn’t distinguish between sexual assault and assault.

Maybe Lockwood misspoke and meant to say that we can draw such distinctions in our research. Maybe she instead meant to reject certain value judgments that we could make based on this research, for example, the assumption that, if it turned out that the 20% of women who have experienced sexual assault according to studies were forcibly penetrated, this would be worse than if they had received unwanted kisses. The reason for resisting this assumption would be that, as you’ve pointed out, there are distant possible worlds in which an unwanted kiss is more harmful than a forcible penetration.

The problem with this is that it’s obvious that the assumption we’re told to reject is true in the real world, given the realities of forcible penetrations and of unwanted kisses. There’s no need to avoid an assumption about the real world because of what happens in distant possible worlds–because there’s *some* set of conceivable, if far-flung, factors that makes the unwanted kiss worse.

I’m not saying there aren’t highly upsetting forcible kisses. But I am saying that, in the real world, the harm caused by the practice of unwanted kissing absolutely pales in comparison to the harm caused by the practice of forcible penetration. It’s the denial of this reality that I’m uncomfortable taking seriously.

Heidi, right, you’re not refusing to distinguish kinds of sexual assault. But you are refusing to draw the distinctions Tooley mentions–for example, the distinction between forcible penetration and unwanted kisses. That’s what my arguments concern. And my objection is only to that–I don’t have any objections to, for example, including certain non-penetrative sexual acts as instances of rape.

I take your point about my underestimating the damage that an unwanted kiss can cause. However, you mischaracterize my view when you attribute to me the claim that people who have been harmed by an unwanted kiss are unusual. What I said was that it’s unusual for an unwanted kiss to be more harmful than an act of forcible penetration. And that’s presumably relevant to classifying kinds of sexual violence.

Also, I disagree with you about what my analogy implicitly said. I added that the person receiving the slap was especially fragile to make it plausible that this could harm them more than a stabbing harmed someone else. The fragility wasn’t the analogy with unwanted kissing vs. forcible penetration; the analogy was that both are cases in which, typically, one kind of act is far more harmful than the other. And since it’s obvious that we should generally distinguish stabbings from slaps, so we should generally distinguish forcible penetrations from unwanted kisses.

I can see, though, that it could have been easy to read my post as you did, so I’ll just say that I don’t think that anyone who was harmed by an unwanted kiss is fragile.

Sorry, my conclusion follows from the premise together with the premise (which I left implicit) that sexual assaults are neither necessarily more harmful nor necessarily less harmful than plain old assaults. I think this premise is obviously true.

JRS, I take it that since Heidi wrote, “your proposed conclusion that we shouldn’t distinguish between sexual assault and assault” doesn’t follow, that she meant not that it doesn’t follow by way of modus ponens on the premise you thought my earlier comment relied on, but rather that it didn’t follow from a modus tollens on the implied premise in your larger conditional claim: “[I]f your previous post is intended as a defense of Lockwood’s claim that data on sexual violence shouldn’t distinguish between forcible penetration and unwanted kisses, then it relies on the following premise: If an act of kind x is neither necessarily more harmful nor less harmful than an act of kind y (because, when certain factors are present it’s more harmful, when other factors present, it’s less), then we shouldn’t distinguish between x and y in our data.”

The style of argument of giving two specific cases doesn’t address my question regarding the *tendency* of the two kinds of acts to compromise sexual agency or cause harm. (Heidi, you mischaracterize my view by leaving out the reference to tendency.)

On further thought, though, I’m less confident that forcible penetration tends to compromise sexual agency more than unwanted kisses.

So while I accept your points (1) and (2), Heidi, I don’t think effects on sexual agency are the *only* features of a crime that are relevant to its classification, for data collection, law, or moral evaluation. In particular, the *tendency* of a certain kind of act to cause a certain degree of harm is relevant. This is particularly so when we’re unable to acquire statistics about the degree of harm caused by all the instances of these kinds.

Example: it’s worth distinguishing armed assault from unarmed assault. If we were to find that in a county without law L, there were 7 assaults per 1000 persons and they were all armed, and that after passing law L, there were still 7 assaults per 1000 persons, we would be reasonable to assume that L was a good law. (We would perhaps want further data on, eg, rate of hospitalization. but even without that, the data on the two kinds of assaults would justify our conclusionabout L.) Why would this assumption be reasonable: because armed assault *tends* to be more harmful than unarmed assault. So it’s useful to distinguish between them. The analogy to forcible penetration vs. unwanted kisses is clear.

Of course, there are some unarmed assaults that are more harmful than some armed assaults. But, as I’ve repeated too many times now, an act doesn’t have to be necessarily different in some respect than another in order to be classified differently. It just has to *tend* to be different in this respect.

Oops! In the example I give above, after passage of law L, there are 7 assaults per 1000 persons, *and they are all unarmed*.

I agree that unwanted kisses and forcible penetration both compromise sexual agency to the same extent. Insofar as they both are wholly non-consensual sexual acts they both _completely_ deny sexual agency to the victim.

What’s unclear in the exchange between JRS and Prof. Lockwood above is whether Prof. Lockwood thinks sexual agency is the only factor relevant to classification of the act for legal and moral purposes. It might even be right to classify unwanted kisses as “sexual assault” but seems wrong to punish that as severely as you would sexual assault through forcible penetration. When I think about why, it does seem like something about tendency to cause harm and the severity of the harm caused. To ridiculously oversimplify and use just one measure: PTSD diagnoses. I would suspect that the diagnosis is more common among victims of forcible penetration than unwanted kisses. Just like a physical assault where the victim walked away with relatively minor injuries is prosecuted differently than one that leaves the victim hospitalized, I would think it would be appropriate to punish unwanted kisses in a manner less severe than forcible penetration.

So, Prof. Lockwood, does your view that they should both be classified as “sexual assault” involve the view that individuals prosecuted for each should be prosecuted and/or punished in similar (if not identical) ways?

JRS, I said implied premise not suppressed premise. I am already aware of what a suppressed premise is, but thank you for that anyway–it would have been helpful were I not. As to your second point, that’s why I said implied and not suppressed.

JRS, apologies for the confusion. I wasn’t using implied in any technical tense, rather in it’s ordinary sense; as in, your comment implied a premise like, “If your conclusion is ‘this sort of limited description is not enough to determine the kind of harm done’ then you must be relying on the premise ‘[insert the premise you said I was relying on here]’, and then Heidi was offering reasons to do modus tollens on that conditional.

JRS, I’m laughing because I genuinely cannot tell if you’re trolling me or if we are really having this much of a difficult time understanding one another. If you’re not trolling me, I hope you think that’s funny too and aren’t offended by my finding it so humorous. In case, no, I didn’t mean “your” to be referring to me Kathryn, but as a kind of generic. If there’s an argument X, along lines Y, that concludes Z, it must involve premise W. And I thought this because earlier you wrote, “Kathryn, if your previous post is intended as a defense of Lockwood’s claim that data on sexual violence shouldn’t distinguish between forcible penetration and unwanted kisses, then it relies on the following premise: If an act of kind x is neither necessarily more harmful nor less harmful than an act of kind y (because, when certain factors are present it’s more harmful, when other factors present, it’s less), then we shouldn’t distinguish between x and y in our data.” You seemed to be.”

As to the substantive point, obviously Heidi can speak for herself, but one might think certain token harmful acts (which are no only in far off possible worlds) might indicate that where we draw the distinctions between kind could be rethought to include additional factors–that would be true even if your tendency point is right. If you lived in the Marvel Universe, for example, you might not want to distinguish between various kinds of violence purely in terms of whether they are, e.g., a slap or stabbing, as a stabbing committed by an ordinary human, Bruce Banner or Ant Man (particularly in his shrunken state) would be remarkably less violent than a slap committed by the Hulk, Colossus, Rhino, Thor, etc.

No, that makes sense, it’s just that if the reference was meant to be preserved I wouldn’t be making any sense–and I hope that I don’t appear quite that nonsensical. 🙂 But, you’re right this isn’t the substantive issue, so I’m happy to leave it too.

In response to Rob at 24: the problem of sexual assault on campus is not specific to heterosexual members of campus communities. prohibition on liaisons — and yes, prohibitions on liaisons between faculty and students should extend to all.

“what I’m calling for is a more fine-grained approach to prosecution and punishment, one which is more sensitive to the harm that the victim actually experiences, rather than the harm that the victim “should” experience in the eyes of non-victims, many of whom do not understand the potential harm of acts that compromise sexual agency.”

This sounds reasonable to me. The reason I was earlier making somewhat inflammatory claims about your views was that what you said in the debate was stronger than this. Tooley brought up the distinction between, e.g., forcible penetration and unwanted kissing not in the context of prosecution or punishment, but rather in the context of collecting data for the purpose of measuring the extent of the problem of sexual assault on college campuses. It seems to me reasonable to deny that we need this distinction for prosecution and punishment (though I haven’t fully thought it through). But I don’t see how the distinction could possibly be irrelevant to assessing the extent of the problem of sexual assault.

To see why, here’s a variation on the kind of things I’ve said a few times. I’ll try to focus on the core issues so we don’t get distracted by red herrings, as we did before.

We know that roughly 20% of women are sexually assaulted in a sense of “sexual assault” that includes both forcible penetration and unwanted kissing. One possibility is that 3/4 of these women (15% of women) were forcibly penetrated and 1/4 (5%) suffered unwanted kisses. Another possibility is that 1/4 were forcibly penetrated (5% of women) and 3/4 suffered unwanted kisses (15%). Both possibilities are terrible. But it seems obvious to me that the scale of the problem of sexual assault would be much more catastrophic if the first possibility were actual than if the second were. Why? For the reasons indicated by myself and Anon Grad: in the actual world, it’s overwhelmingly likely that more harm is caused to female victims of sexual assault in the first possibility than in the second. If that’s right, then by distinguishing between the two kinds of sexual assault in our research, we learn something about the scale of the problem that we wouldn’t otherwise learn. Therefore, we should draw the distinction in our research.

Furthermore, suppose that we started distinguishing forcible penetration from unwanted kisses in social scientific research and we find that the first possibility is actual. Then we pass a law as a result of which we move into the second possibility (and nothing else of significance changes). Isn’t this a good law? Granted, it leaves work to be done (we should try to reduce the unwanted kisses too). But it seems to have brought about an improvement. If that’s the case, then, again it must be worth distinguishing between unwanted kisses and forcible penetration for research purposes (the purpose for which the question originally arose in your debate with Tooley).

Do you disagree with these lines of reasoning? If so, with which of my assumptions or inferences?

This last example is structurally similar to my previous one (comment #19) involving a country before and after a law that leaves the rate of assaults unchanged but that “converts” armed assaults into unarmed assaults. So if you have time, I’d also like to hear whether, if you think it’s not important to know the frequency of forcible penetration as opposed to unwanted kissing, is it likewise not important to know the rate of armed vs. unarmed assault for purposes of evaluating law L?

Dr. Lockwood,

Is there not a tension between your calls (1) for a categorical prohibition against liasions between faculty and students regardless of gender and/or sex, and (2) for “a more fine-grained approach to prosecution and punishment”? Is it your view that the gender and/or sex of the parties don’t contribute distinctively enough toward shaping the nature and/or severity of harm? Are there data or some other source of good reasons to believe that, say, homosexual liasons between faculty and students are similiarly enough (or identically) impacted by the other variables which contribute to the risky power differentials that concern you?

Thanks for a very interesting presentation, Heidi.

You opened your talk by handing several items to Michael, who put them on the table for you. You later said that doing this was a violation of his consent and autonomy, and you also said that the harm done by violating another’s consent is to be measured not by the nature of the action or by the intention of the agent, but by the feelings of the recipient (at least, you seemed to be saying that someone who kissed a student (s)he was supervising would be doing something far worse than someone who actually penetrated someone’s sex organs without another’s consent, so long as the student who received the kiss was far worse affected).

This seems to imply that your treatment of Michael could have been worse than the action of a pervert who jumps out of the bushes with a gun and demands that someone perform some sex act, so long as Michael had been badly enough affected and for long enough by your handing him the objects and so long as the person attacked by the pervert in the bushes managed to brush off the experience fairly easily. You seem to acknowledge that this follows from your view, since you made a point of apologizing to Michael and attempting to justify your handing him the items by saying that you needed to make the point somehow.

But if you are right, this apology and justification might not be enough. For Michael might have been extremely offended by your taking advantage of him like that. He might have felt humiliated in public by being treated as your servant whose role it was to bring the objects to the table. He might have felt taken advantage of by extending his courtesy to you and then having you walk all over it to make a point. Perhaps brooding on these transgressions of his good will toward you might lead him to feel withdrawn, depressed, and anxious for a long time. And while many would say that he would be reacting inappropriately and tell him to get over it, this option doesn’t seem open to you, since on your view the wrong is correlated to the subjective response of the person affected, not the nature of the act itself.

Imagine if someone in a similar situation to yours had tried to make a point about sexual harassment by unexpectedly groping a fellow panelist, actually reaching under the panelist’s pants in the process. After such an outrageous act, it couldn’t possibly suffice to give an apology and explain that it was necessary to illustrate a point about the nature of assault. I can’t imagine that anyone would dispute this, including you. Moreover, even if the panelist happened for some reason to find it very funny and not to mind, that would not exonerate the groping. Any reasonable person would understand that this is just an unacceptable way of treating someone who has not given consent. But in the case of handing Michael various objects, you (and we) could see that there was nothing unacceptable involved. The only conclusion must be that you, also, acknowledge that there is a world of difference between handing someone objects to put on a table and groping someone. What makes these acts objectionable is not purely in the mind of the receiver, contra your thesis.

She may have confused Table 1 and Figure 1 since Figure 1 in that report *is* about those who report to the police, and she didn’t seem to have a copy of the report in front of her. In any case, as to whether or not Tooley is right that these data undermine the idea that sexual assault is prevalent in college, that very report he cites gives us good reason to think maybe not:

“The NCVS is one of several surveys used to study rape and sexual assault in the college-age population. In addition to the NCVS, two recent survey efforts used in research on rape and sexual assault of college students are the National Intimate Partner and Sexual Violence Survey (NISVS) and the Campus Sexual Assault Study (CSA). Each of these surveys has a different purpose and methodological approach than the NCVS. Depending on which of the three data sources are used, researchers will generate different estimates of the prevalence and frequency of rape and sexual assault victimization. Some of these differences include— „

Survey context and scope. In a 2014 report on the measurement of rape and sexual assault in the NCVS and other federal surveys, the National Research Council (NRC) of the National Academies of Sciences described survey context as “a major contributor of differences in the estimates of rape and sexual assault” across different data collections.9 Survey context refers to how the survey is described to respondents through notification letters, survey questions, or the interviewer. The NCVS is an omnibus survey designed to collect information on experiences with a broad range of crimes. It is likewise presented to respondents as a survey about criminal victimization. Because victims of rape or sexual assault may not consider their victimization a crime, this context could discourage or suppress recall and reporting of those incidents.10 Additionally, because the NCVS covers a wide range of criminal victimization, the number of screening questions related to rape and sexual assault are limited.

In comparison, the NISVS focused on sexual violence, stalking, and intimate partner violence and was presented as a survey collecting data on a range of behaviors that impact public health. This public health perspective may encourage respondents to recall and report on experiences that they may not typically think of as criminal victimization. It also may result in the collection of incidents that may not be considered criminal behavior. Similarly, the CSA study focused specifically on rape and sexual assault, also from a public health and safety perspective. „

Definitions of rape and sexual assault. The NCVS, NISVS, and CSA define rape and sexual assault slightly differently. The NCVS definition is shaped from a criminal justice perspective and includes threatened, attempted, and completed rape and sexual assault against males and females (see Methodology). Penetration due to coercion is included in the definition of rape, but the survey does not specifically ask about incidents in which the victim was unable to provide consent because of drug or alcohol consumption. Because the NISVS is focused on rape and sexual assault from a public health perspective, the scope of sexual violence included in NISVS is broader than the definitions used in the NCVS. In NISVS, sexual violence includes threatened, attempted, or completed rape, including incidents in which the victim was unable to provide consent due to drug or alcohol use; forced penetration of another person; sexual coercion, which includes nonphysical pressure to engage in sex; unwanted sexual contact, including forcible kissing, fondling, or grabbing; and noncontact unwanted sexual experiences, which do not involve physical contact. The CSA definition of rape and sexual assault included unwanted sexual contact due to force and due to incapacitation, but excluded unwanted sexual contact due to verbal or emotional coercion.

Longitudinal versus cross-sectional design. Both the CSA and NISVS were cross-section data collections administered to the sample a single point in time that asked about events that occurred during a specified reference period. The NCVS is a longitudinal survey administered seven times to the same sampled household, with questions asked about events occurring since the last interview. Longitudinal surveys like the NCVS have the advantage of bounding the reference period and ensuring that events occurring outside of that reference period are not included in estimates. Since research has suggested that traumatic events, such as rape and sexual assault, may be particularly prone to telescoping (i.e., the reporting of events occurring outside of a reference period as though they occurred within the specified period), unbounded surveys may have artificially high incident rates due to events occurring outside of the reference period being telescoped in. „

Question wording. The language and ordering of questions in a survey may affect whether a respondent indicated that an incident occurred. The three surveys used different approaches to asking about experiences with rape and sexual assault. The NCVS used a two-phased approach to identifying incidents of rape and sexual assault. Initially, a screener was administered, with cues designed to trigger the respondent’s recollection of event and ascertain whether the respondent experienced victimization during the reference period. The screener questions directly focused on rape and sexual assault were— – (Other than any incidents already mentioned), has anyone attacked or threatened you in any of these ways:… (e) any rape, attempted rape, or other type of sexual attack; y Incidents involving forced or unwanted sexual acts are often difficult to talk about. (Other than any incidents already mentioned), have you been forced or coerced to engage in unwanted sexual activity by (a) someone you didn’t know before, (b) a casual acquaintance? OR (c) someone you know well?

Even if the respondent did not respond affirmatively to these specific screeners on rape and unwanted sexual contact, the respondent could still be classified as a rape or sexual assault victim if a rape or unwanted sexual contact was reported during the stage-two incident report.

Unlike the NCVS which used terms like rape and unwanted sexual activity to identify victims of rape and sexual assault, the NISVS and CSA used behaviorally specific questions to ascertain whether the respondent experienced rape or sexual assault. For example, one question on the NISVS survey read: y When you were drunk, high, drugged, or passed out and unable to consent, how many people ever… – had vaginal sex with you? By vaginal sex, we mean that {if female: a man or boy put his penis in your vagina} {if male: a women or girl made you put your penis in her vagina}. – {if male} made you perform anal sex, meaning that they made you put your penis into their anus? made you receive anal sex, meaning they put their penis into your anus? – made you perform oral sex, meaning that they put their penis in your mouth or made you penetrate their vagina or anus with your mouth? – made you receive oral sex, meaning that they put their mouth on your {if male: penis} {if female: vagina} or anus?

Questions on the CSA used similar behaviorally specific cues to identify victims of rape and sexual assault. „

Survey mode. The NCVS, NISVS, and CSA used different modes of administration. The CSA study was a self-administered web survey sent via email to students at the participating colleges. Self-administered surveys are not subject to interviewer effects, but may result in lower response rates or confusion over question wording that could otherwise be clarified by an interviewer. The NISVS was a random digit dialing telephone survey. Telephone surveys exclude respondents without a phone, may be subject to sampling bias because of multiple phones associated with particular households or individuals, and may be subject to nonresponse bias due to low response rates. The NCVS used a multimode design that begins with an initial in-person interview, followed by telephone follow-ups every 6 months for the 3.5 years the household is in the sample. While respondents can develop rapport with the interviewer and familiarity with the survey questions, the NCVS may be more subject to interviewer effects than the CSA or NISVS.”

So where does this one-in-five statistic come from? Well, there seem to be two different “one-in-fives,” one for women during college and one for American women across their lifetimes. For those who (like me) are not social scientists or statisticians, a couple of the Washington Post’s Face Checker pieces are extremely helpful. Here are the links.
http://www.washingtonpost.com/blogs/fact-checker/wp/2014/12/17/one-in-five-women-in-college-sexually-assaulted-an-update/
http://www.washingtonpost.com/blogs/fact-checker/wp/2015/02/12/obamas-claim-that-one-in-five-american-women-have-been-raped/

JRS @ 35: I disagree with your line of reasoning because I reject the soundness of the argument. Specifically, I don’t accept the assumption that “the scale of the problem of sexual assault would be much more catastrophic if [a significantly higher percentage of sexual assaults were due to forcible penetration.”
I am of course not denying that forcible penetration can cause very significant harm. I’m simply denying that it necessarily causes *more* harm than forms of sexual assault that involve mental coercion or force rather than physical coercion or force.
Part of the reason I’m denying this is because I have empirical evidence to the contrary. In my experience working with victims, I have actually found the opposite; when someone a victim knows — particularly someone in a position of relative power — employs mental coercion or manipulation to achieve sexual ends, the victim experiences a sort of PTSD with both physical and mental triggers, and is often unable to get to the point where s/he feels safe enough to openly engage in discourse or an exchange of ideas.
To use a thought experiment: would you rather be assaulted by a stranger in an alley, and then deal with PTSD and anxiety in connection with strangers, alleys, etc., or would you rather be, as Colin McGinn so colorfully put it, “mindfucked” by a philosophy faculty member, and then deal with loss of trust in philosophers and philosophy itself, and perhaps even an inability to become mentally close, to trust anyone again?
Obviously both are horrific. But I’ll take the first over the second.

Johnny @ 36: Yes, your remark, “Agency has to do with a broader notion of personal identity, where that identity can be (to a more or less degree) pushed upon, molded, or even determined by others who wield more power than us” is a good start.

Let me try to say a little bit more.

First, it’s worth noting that there are various feminist critiques of the classic liberal theories of autonomy developed by theorists such as Dworkin and Feinberg. The basic idea behind these critiques is that the notion of autonomy should be guided by observations of the lives of diverse groups of actual persons, rather than, for example, theorizing about the overly-intellectualized ideal of an autonomous person as an isolated individual or cold and detached rational calculator.

Thinking about actual lives and actual persons very quickly leads us to the realization that there are various social forces which seem to restrict the autonomy of individuals. Institutional or cultural androcentrism, for example, may shape not just the choices that are available to women, but also the tendency of women themselves to value practices, norms, and characteristics socially understood as male — and to place these attributes, as well as the men who exemplify them, at the center of institutional or cultural practices.

So the puzzle is: how do we make sense of autonomy in light of a deeper understanding of the impact of socialization and interconnectedness?

Some feminists argue that socioeconomic influences should be seen as restricting or qualifying autonomy, rather than eliminating it. So, for example, Diana Meyers argues that autonomy is a multifaceted, refinable competence — a gradable attribute, rather than a binary all-or-nothing. Later structural and post-structural feminists (e.g., KP Addelson) introduce the notion of power into the puzzle by focusing not on vague, undirected “social construction,” but rather on the ways in which the complex mosaic of intersecting practices of social construction are shaped by power — and in particular, the sorts of power that enable one group to socialize another group in a manner that encourages or requires subordination. In recognizing that the sort of self-direction that is required for autonomy doesn’t occur through a process of extracting the pre-social self from a nexus of power (this would be futile), some of these feminists make the shift from autonomy to agency (e.g., Hilde Lindemann, who argues for a narrative conception of agency).

The reason that I’m encouraging a shift from violations of sexual autonomy to violations of sexual agency in our understanding of how to assess harm in cases of sexual misconduct is because power is an obvious factor in the practices of sexualized dominance which shape both women and men. Feminists such as MacKinnon have described women — paradigmatically — as being subjected to sexual assault or threats of assault, harassment, and pornography. And these practices are often accompanied by insidious arguments and linguistic constructions that hold victims responsible for the harms they suffer… or dismiss the harms as private matters… or even try to justify the harms in terms of the perpetrators’ civil rights. To restrict our attention to just those forms of sexual misconduct which violate sexual autonomy (e.g., forcible non-consensual penetration) is to overlook the fact that sexualized power structures are grounded not merely in violations of sexual *autonomy*, but rather in more insidious restrictions on an individual’s sexual agency.

Note that I’m not calling for an abstract reconception of sexual autonomy or an (impossible) overnight reworking of the various social, political, and economic influences that combine to create the sorts of power at play in sexual relationships. Nor am I calling for the elimination of power from sex. What I *am* calling for is a heightened sensitivity — both in our cultural understanding, and in our policies and investigations — of the ways in which power can compromise sexual agency and make it impossible for a superficially autonomous agent to genuinely consent. As I indicated in the debate, sometimes yes means no.

Random Viewer @ 39: This is a great question — your thought experiment beautifully illustrates precisely the point I am trying to make.

Giving the sock to Michael was an extremely risky move. As one member of the audience pointed out, he might well have had a foot fetish, in which case it would (under very slightly different circumstances) have been natural for him to interpret the inexplicable gift as a sexual advance. I took the risk — and will accept criticism for my reliance upon moral good luck — because I felt it would be a helpful and memorable illustration of several points I wanted to make.

One of these points is that many sexual predators engage in a process called “grooming,” in which they test and stretch the limits of the object of their attention by incrementally pushing (and then sometimes retreating from) the boundaries of the acceptable. What I was doing was attempting to simulate the beginning of the process of grooming Michael. It might well have progressed beyond just giving him my phone, the markers, and the sock, to something more explicit and unacceptable, involving, say, unwanted sexual contact.

In other words, you have nicely drawn out the point that I was making. Most of us would *not* intuitively find the act of giving him the sock clearly problematic. I think this intuition is wrong. Our intuitions fail us here for the same reason that we are beguiled into accepting the sock, or worse. After all, what could possibly be wrong with a sock, right? (I didn’t, as you note, engage in unwanted sexual contact.) And then what could possibly be wrong if I touched Michael’s back, right? (It’s not, after all, sexual contact.) And so on. Almost all of the victims that I know who were subjected to unwanted sexual contact were not randomly and suddenly groped. The move to physical touching comes after a long string of questionable but acceptable or at least explicable/dismissible odd behavior.

Although the act of public sock-giving was not problematic in isolation, it can be problematic when understood as part of a larger process of grooming behavior — which would most likely culminate in the kind of behavior you describe, or worse.

Michael asked to keep the sock, by the way. I was initially conflicted about this — I of course didn’t want to add insult to injury by asking for a return of the gift, but it did mean that I’d have to wear slightly dirty socks on my trip to DC the following day. Then he explained that he was asking to keep it because he wanted to take a picture of it and put it on his web site with the caption “I was socked by Lockwood.” And… well, how could I say no to that?

Heidi,
Thanks, for your response. You ask me to answer some questions. As I’ll explain below, I think you’ve changed the case I described. I think that indicating the ways in which you’ve changed my case should answer the question you have for me. I’d also like to hear about whether you agree that the changes to the case that you made are significant for the reasons I discuss and why. Of course, if agree that the changes are significant, then I’m still interested in your take on my original case of the two possibilities (15%/5% vs. 5%/15%).

Before I get to that, though, you say: “I’m simply denying that it [forcible penetration] necessarily causes *more* harm than forms of sexual assault that involve mental coercion or force rather than physical coercion or force.” The suggestion is that you’re denying something I’ve asserted.

I wish I could once and for all convey that I’m *not* assuming that forcible penetration *necessarily* causes more harm than unwanted kisses. It seems like every response to my comments attributes this view to me, no matter how many times I disavow it, but I’ll try once more. (Compare a feminist claim about some tendency of men to do x and the response, “not *all* men do x”.)

Now, on to the changes to the case I presented. You move from unwanted kisses (my example) to “forms of sexual assault that involve mental coercion or force rather than physical coercion or force”. This move would be licit only if the likelihood that unwanted kisses will be of this kind (call this likelihood ‘P(u)’) is not significantly less than the likelihood that a forcible penetration will be highly harmful (call this likelihood ‘P(f)’). Why this necessary condition on your move? Because if we avoid the distinctions you want to avoid, for research purposes, then it’s people who have received unwanted kisses generally, and not jut people who have been mindfucked, who will be lumped in with people who have been forcibly penetrated as victims of sexual assault.

So, our main question: is P(f) significantly greater than P(u)? It seems obvious to me that it is. In my experience (like you, I have to rely on it), it’s pretty common for a few undergrads to go out, get drunk, and for one to kiss another who doesn’t want the kiss. Or to rub up against another who doesn’t want to be rubbed up against while dancing at a party or bar. In both cases, the kiss/rubbing is generally somewhere between a nuisance and an infuriating affront; generally not a traumatic event. It frequently doesn’t involve mindfucking, but just someone having a really strong crush and losing too many inhibitions. I’m assuming, by contrast, that an insignificant number of forcible penetrations are merely nuisances, affronts, or something of the same degree of harm.

Like you, I’m partly relying on what I’ve personally experienced here. But what I’m suggesting also makes sense. There are many more opportunities to do unwanted kisses/rubbing up against a friend or a stranger you’re dancing with at a bar than someone you have power over–unless students socialized/drank with their professors/TA’s as much as they do so with each other–which they clearly don’t (for one thing, there are way more students than profs/TA’s). Also, the less harmful kinds of unwanted kisses are less risky. (Again: this is not to suggest it doesn’t happen, but only that it’s less common than the other kind of unwanted kissing.)

Regarding the personal evidence you (Heidi) cite working with victims of sexual assault: people who have experienced the kinds of (*generally*) less harmful kissing/rubbing I just described won’t (*generally*) report what happened to a sexual assault officer, and so won’t influence this personal evidence. Your personal evidence may give you a sense of the relative harmfulness of mindfucking and forcible penetration; but it doesn’t give you evidence about P(u), about how many unwanted kisses involve mindfucking. (Maybe you’ve had other experiences that do bear on this; I’m only responding to the specific evidence you described.)

So if my assumptions are true (and the empirical evidence you’ve discussed so far doesn’t contradict them), then it follows that forcible penetrations are more harmful in the aggregate than unwanted kissing (*not* that every possible forcible penetration is more harmful than every possible act of mindfucking). And we should therefore distinguish between the two in our research.

I was violently assaulted. Then pursued and stalked for years. PTSD is a clinical diagnosis reserved for extremity level experiences such as near death (as commenter above 52 explains). An unwanted kiss is not the same as being assaulted and stalked. This is an extract from my medical diagnosis: “I assessed X on … X’s scores, on the Beck Anxiety Inventory, indicated X was suffering from clinically severe levels of anxiety. X’s scores on the Beck Depression Inventory … were of clinical severity … The Clinical Outcomes in Routine Evaluation – Outcome Measure (Core-OM) recorded the following findings: impairment to wellbeing of clinical severity, clinical severity for problems and symptoms, clinical severity for functioning; no clinically signficant risk of harm to self and others;, and clinically severe distress levels. … scores above 36 on this scale [the Impact of Events scale] indicate the presence of trauma symptoms … X scored 84 out of 88.”

Leaving trauma out, let’s take the example of depression. Someone with a seemingly stress free life–upper middle class, great parents, solid social support network–could develop severe depression while someone living in extreme poverty working two jobs to feed their kids doesn’t. It doesn’t mean the former is weak or shouldn’t be depressed; it just means there’s a lot about the brain we don’t understand yet. And regardless, the former individuals depression is real and valid and it requires treatment. Same with a traumatized individual, regardless of what someone else thinks of their trauma.

On a separate note: The CDC released a report in Oct. 2014 that supports the 1 in 5 statistic.

Let me correct you, Heidi. As someone who has been violently assaulted, I know the symptoms (hypervigilance, flashbacks, triggering, sleep disturbance and so on), because I have them. I know the criteria, because my therapist is an expert on PTSD, a specialist in treating trauma. According to the American Psychiatric Association’s guidance on DSM-5, PTSD is based on the criterion:
“The diagnostic criteria for the manual’s next edition identify the trigger to PTSD as exposure to actual or threatened death, serious injury or sexual violation.”
http://www.dsm5.org/Documents/PTSD%20Fact%20Sheet.pdf