Abstract
In recent years, there has been increasing pressure on men to prevent sexual violence. This study uses data from 25 semi-structured interviews to explore how heterosexual undergraduate men have responded to cultural and organizational pressure to seek consent. Participants answered questions about their recent sexual experiences and attitudes toward campus sexual consent policies. Findings indicate that while participants condone key elements of sexual consent, they do not consistently apply reliable strategies to seek consent. Instead, they use ambiguous social cues that are common in both consensual and nonconsensual sexual interactions, which reinforce the notion that consent is unclear.
Sexual violence is widespread in the United States. In recent years, college students’ experiences of sexual assault have received particular attention. Recent research estimates that between 19% and 23% of undergraduate women experience either an attempted or completed sexual assault during college (Cantor et al., 2015; Krebs et al., 2007) and as many as 11% of undergraduate men commit a rape before graduation (Swartout et al., 2015). To address the high rates of violence, college campuses have done something few other organizations have—they have formally instituted prevention measures, many of which require men to take a proactive approach to sexual assault prevention. These changes represent a stark departure from typical narratives of adult sexual violence, which place the burden of preventing violence on women (Bedera & Nordmeyer, 2015; DiBennardo, 2018). Among the most widespread changes targeting men is a movement to redefine sexual assault using an affirmative consent standard that requires all parties—including, and perhaps particularly, men—to actively seek consent from their sexual partners through clear verbal or physical cues as a way to minimize miscommunication.
The theory that sexual assault is simply miscommunication has been explored in depth by scholars across disciplines. Although college students’ negotiations of consent may appear clumsy or confused (Hirsch & Khan, 2020), the sexual scripts employed become clearer through the use of a gendered lens (Muehlenhard et al., 2016). Despite evidence that men and women use different cues to convey sexual desire (Beres, 2010; Hickman & Muehlenhard, 1999; Jozkowski et al., 2014), academics have long refuted the claim that young men cannot understand women’s sexual signals (Hickman & Muehlenhard, 1999; Kitzinger & Frith, 1999). Instead, many have suggested that men’s alleged misinterpretation of women’s sexual signals is better described as the systematic marginalization of women’s voices (Abbey, 1982, 1987; Osman, 2006) that can be exacerbated in specific social environments, particularly those common to college campuses (Armstrong et al., 2008; Hirsch et al., 2019; Martin, 2016; Wade, 2017). Researchers have found that men who engage in hostile masculinity are especially likely to claim to misinterpret a partner’s sexual cues (Jacques-Tiura et al., 2007), and reinterpretation of a victim’s mundane actions as indicators of consent is a common tactic used by rapists to justify their violence (Scully & Marolla, 1983). In sum, these findings suggest that traditional consent acquisition practices are embedded in a gendered hierarchy that prioritizes men’s sexual interests and allows men to dictate what is and is not consent. The popular notion that most campus sexual violence is miscommunication reinforces and naturalizes these gendered hierarchies surrounding consent-seeking practices.
Ideally, campus affirmative consent policies and the associated trainings would challenge these gender hierarchies and put an end to the myth that sexual violence is the result of miscommunication. To put it simply, those accused of sexual assault could no longer revert to an excuse that they could not tell that they had traumatized a sexual partner; they would be expected to have known in advance that they had clear consent. However, there are many reasons to anticipate that affirmative consent programming would fall short of these goals, particularly because men are commonly viewed—and accepted—as resistant to sexual assault prevention efforts (Bedera & Nordmeyer, 2015; Messner, 2016; Rich et al., 2010). This study seeks to understand how men have responded to increased organizational and cultural pressures to take a more proactive response in sexual violence prevention through seeking clear affirmative consent in their sexual encounters.
Literature Review
Redefining Sexual Consent
Over the past decade, the definition of consent has evolved on college campuses. Previously, a victim would need to offer evidence that sexual consent had been explicitly revoked to categorize a sexual encounter as assault. Instead, feminist reformers recommended affirmative consent policies (e.g., Friedman & Valenti, 2008), which require all parties to proactively ensure that sexual contact is consensual. While affirmative consent policies vary between institutions, they generally include three main criteria: (a) consent must be given knowingly; (b) consent must be given voluntarily; and (c) consent must be given affirmatively (as early statewide examples, see California State Legislature, 2014; New York State Senate, 2015). Notably, affirmative consent policies may privilege verbal methods of obtaining consent, but nearly all also permit enthusiastic participation in sexual action as acceptable evidence of a partner’s consent, so long as that enthusiastic participation is ongoing throughout the entirety of a sexual encounter. While these policies are gender-neutral, cultural heterosexual scripts largely place the burden of initiating sex (and, therefore, seeking consent) on men, leading many to interpret affirmative consent trainings as primarily directed at men (e.g., Jozkowski & Peterson, 2013).
Affirmative consent policies have become common on college campuses, with 80% of colleges defining sexual assault using an affirmative consent standard (Armstrong et al., 2019). At the time of data collection, all U.S. universities accepting federal funding were mandated to provide sexual assault prevention training, with affirmative consent training as one of the recommended and widely accepted ways to meet this burden (U.S. Department of Education Office for Civil Rights, 2014). On any campus that has adopted an affirmative consent policy, obtaining sexual consent is not a recommended practice, but rather is mandatory for all students. For these reasons, this study uses affirmative consent as the standard by which to evaluate men’s consent-seeking practices. The goal of this study is not to critically analyze affirmative consent as an ideology, but rather to assess whether men changed their gendered attitudes and behaviors based on an institutionally mandated feminist initiative.
Men, Consent, and Ambiguity
Despite enthusiasm from anti-violence experts about men’s capacity to end sexual violence (e.g., Flood, 2011; Messner et al., 2015), research on men’s engagement in anti-violence and other feminist activist efforts is divided. Even in evaluating similar programs, researchers have come to disparate conclusions about men’s willingness to engage in sexual assault prevention (Murphy, 2009; Piccigallo et al., 2012), perhaps related to the heterogeneity expected in men’s responses to such programs (Malamuth et al., 2018). As sexual violence becomes a more widely discussed topic on college campuses, it is possible that men will use anti-violence rhetoric as a way to assert their masculinity and sexual prowess without changing their behavior (Pascoe & Hollander, 2016), and even men who actively participate in prevention programming may get the impression that their own behavior should be exempt from scrutiny (Messner, 2016) or might unintentionally reify gendered hierarchies that allow violence to occur (Bridges, 2010; Bridges & Pascoe, 2014; Connell & Messerschmidt, 2005; Masters, 2010). For these reasons, men’s successful engagement with affirmative consent programming might be more complicated than their stated desires to prevent sexual assault and pursue purely consensual sex.
In particular, sexual ambiguity offers all men benefits in traditional heterosexual sexual parings. Ambiguity in narratives of sexual encounters allows men to exaggerate their sexual prowess in verbal performances of masculinity (Currier, 2013) and receive more sexual pleasure than their partners during casual sex (Armstrong et al., 2012). Although these advantages do not come from participating in sexual violence, the ambiguity they require and their contributions to a larger sexual script cannot be separated from violence (Simon & Gagnon, 1986). A sense of ambiguity makes bystanders less likely to intervene in a sexual assault they witness (Pugh et al., 2016) and leads many to discredit survivors’ claims of victimization (e.g., Stepp, 2007) or engage in victim blaming (Frese et al., 2004; Randall, 2010; Ryan, 1976; Siefkes-Andrew & Alexopoulos, 2018). Ambiguity can also make it more difficult for survivors to categorize what happened to them (Peterson & Muehlenhard, 2004; Smith & Martinez, 1995; Warshaw, 1988), effectively blocking victims from reporting or accessing resources (Hlavka, 2014; Holland & Cortina, 2017). Perhaps most importantly, ambiguity permits subjective interpretation of the presence or absence of consent, which researchers have suggested leads men to label ambiguous events as nonconsensual at a drastically lower rate than women (King et al., 2020). In short, the normalization of ambiguity in sexual interactions and narratives about them—whether consensual or not—can make the tasks of defining and responding to violence harder.
Data and Method
Participants
The data for this study come from 25 semi-structured interviews conducted during 2015 with undergraduate men at a large university in the mid-Atlantic region. The university they attend had adopted an affirmative consent policy and implemented mandatory trainings for students about sexual consent, including a sexual assault awareness seminar at first-year student orientation and an online course about sexual violence that students of all class standings must complete every semester. Students may also engage in additional education about sexual consent through engagement with student clubs, fraternities, or certain classes available on campus. The university’s sexual violence prevention training has received national recognition for its rigor. In comparison with men in the general public, the men in this study have received an exceptionally high amount of formalized training on the topics of sexual violence and consent-seeking. While this study does not specifically evaluate any program on campus, the data collection site was selected based on the many opportunities participants would have had to receive both formal and informal education about approaches to preventing sexual violence in their sexual encounters.
I recruited heterosexual men of traditional college age, as they are the members of campus most likely to commit sexual violence and, therefore, whose behavior requires the greatest degree of modification through affirmative consent trainings (Swartout et al., 2015). I recruited participants who met these demographic criteria and who described themselves as sexually active from six lower division courses from different disciplines 1 with at least 50 students. Students enrolled in these classes were invited in person 2 by a mixed-sex pair to participate in a study about college men’s sexual behavior and were informed that they would receive a US$10 gift card for their participation.
Participants’ ages ranged between 18 and 24 years, with a mean age of 20 years. Of the participants, 17 identified as White, three identified as Black, two identified as Asian American, two identified as non-White Latino, and one identified as biracial. Participants reported a variety of religious backgrounds, including Catholic, Jewish, Protestant, Muslim, and Latter-day Saints. Of the participants, four belonged to a fraternity. None of the participants currently played on an official athletic team; however, one participant had previously competed on an athletic team at another university and nine participants participated in intramural athletic organizations.
Methods
I used a semi-structured interview approach to explore the participants’ experiences with sex in long-term relationships and hookups. To avoid socially desirable responses, I did not use the word “consent” during interview sections in which participants described their recent sexual encounters or regretted sexual events. Instead, I used multiple lines of questioning about how participants could tell their partners wanted to engage in sexual activity with them at various points during each interaction and specifically probed for any verbal cues that helped them gauge a partner’s willingness to engage in sexual activity. Specifically, I asked three questions: (a) What was your first clue that your partner wanted to [sexual activity that occurred] with you? (b) When did you know for sure that [sexual activity that occurred] would happen? (c) While you were [sexual activity that occurred], how could you tell she wanted to [sexual activity that occurred] with you? I only used the word “consent” in a final cluster of questions about the participants’ attitudes toward affirmative consent policies and their willingness to seek consent according to affirmative consent requirements.
To guide the portions on their reported sexual behaviors, I asked participants about their most recent sexual encounter in a long-term relationship and in a hookup, although many participants also told me about earlier experiences that informed their attitudes toward their most recent encounter. Drawing on the methods of Armstrong and colleagues (2012), I let the participants use their own definitions of relationship and hookup. When discussing recent sexual events, participants typically chose to talk about events that ended in intercourse. Some participants had not engaged in intercourse (particularly during hookups), but instead told me about when they had engaged in kissing or other sexual acts. Interviews lasted between 24 and 76 min with a median length of 48 min. 3
All interviews used for this study were conducted by a woman. To understand the effect of my gender on my participants, I employed a male undergraduate research assistant to conduct supplementary interviews. The research assistant was trained over the course of 3 months, including conducting mock interviews. I reviewed recordings of his interviews regularly to ensure that his demeanor and interview style matched my own. In general, the substance of the participants’ comments was similar between the two sets of interviews; however, the participants were more reluctant to give detailed accounts of their past sexual encounters when discussing them with another man. This was particularly true for the sections of the interview on sexual encounters occurring in the context of a romantic relationship, perhaps indicating their unfamiliarity or discomfort with discussing issues of emotional intimacy with other men, which other interviewers have written about in detail (e.g., Orenstein, 2020). Accordingly, interviews conducted by a man were significantly shorter and the data are less detailed, producing thin description. They are not included in the data for this study. In contrast, many of the participants working with a female interviewer commented about how they felt unencumbered during their interviews. Many participants commented that they felt comfortable sharing information about their sex lives that they had never told anyone before. Only one participant expressed discomfort after an interview with a woman—he felt embarrassed by how much he divulged about his feelings toward his ex-girlfriend—but did not wish to withdraw his interview from the data, nor was that portion of his interview relevant to the analysis presented in this article.
During my analysis, I used NVivo qualitative analysis software to code for the cues participants reported to use to attribute consent (or willingness to engage in sexual interaction) to their partners. I separated these codes into two separate categories based on in which portion of the interview they were mentioned by a participant (i.e., reflection on their previous behaviors or in their ideal strategies for seeking consent) and used open-coding to categorize the cues (e.g., verbal, physical, or other; ambiguous or unambiguous). As there is no widely accepted scholarly definition of ambiguity in sexual interactions, I defined ambiguity as the absence of the three primary pillars of consent as found in state laws impacting collegiate policies: (a) consent must be given knowingly; (b) consent must be given voluntarily; and (c) consent must be given affirmatively (California State Legislature, 2014; New York State Senate, 2015). For example, a signal was interpreted as ambiguous if it could be seen as sexual unknowingly (e.g., not wearing a bra in public, drinking alcohol) or involuntarily (e.g., increased heart rate, increased breath rate). For a signal to be coded as unambiguous, it had to meet all three criteria (e.g., reciprocated kissing, initiating a sexual act). I chose to use an affirmative consent standard because it matched the definition of consensual sex used by the participants’ university. All participants had received training on affirmative consent and were required by their college to exclusively engage in sexual interactions that met an affirmative consent standard. I analyzed the participants’ reported attitudes toward affirmative consent policies and practices, which I compared with an affirmative consent standard using the same methods.
I also found it useful to analyze data based on whether a cue was sexual or nonsexual in nature and present findings using this distinction. I define “nonsexual” as any action that is common to nonsexual situations (e.g., making eye contact, talking), and all other actions were designated “sexual.” Although previous scholars have argued that many nonsexual actions can become sexual in certain contexts, I found this distinction useful in assessing ambiguity, particularly as woman are likely to disagree with men’s designation of many of their actions as sexual (e.g., wearing a short skirt as “asking for” sex) (Harding 2015).
To deepen analysis, I separated the data based on the context of the sexual encounter described by a participant (i.e., hookup or relationship sexual events) and based on the ages and class standing of the participant, but found no meaningful differences between any of these categories that were relevant to this article. I also separated the data based on the questions asked by the interviewer. I found two differences—participants were more likely to mention kissing and moaning once sexual activity had begun; however, the distribution of the type of cue used (e.g., ambiguous or unambiguous; physical or verbal) did not vary between questions, and all cues referenced more than once were used as responses to multiple questions. Accordingly, I present the findings in the aggregate. I use pseudonyms to maintain the confidentiality of participants. I made small edits for clarity.
In the sections that follow, I describe the participants’ attitudes toward affirmative consent policies and efficacy in defining consent and producing ideal strategies for unambiguous consent-seeking. Afterward, I compare their idealized beliefs about consent-seeking with their reported behaviors. To ease comparison for the reader, I will begin each section with the frequencies at which each code was applied; however, the analyses that follow will be focused on the qualitative differences between participants’ idealized beliefs and self-reported behaviors.
Findings
Defining Consent
In general, the participants in this study endorsed affirmative consent as an ideology and they demonstrated relative ease in producing a definition of consensual sexual behavior. Among participants who struggled to define consent, they seemed to trip over their words, but ultimately understand its key principles—i.e., that consent must be knowing, voluntary, and affirmative. As an example, Ethan, a White 20-year-old, paused and interrupted himself, but ultimately defined consent: Um . . . If I had to define consent, um, I would say to give consent would be to, um, say that you’re able and willing to, um, engage in whatever they’re proposing, um, and that they have your permission to engage in it. If that makes sense.
Most commonly, the participants defined consent accurately and concisely. Similar to the findings from previous studies (Jozkowski et al., 2014), the participants in this study typically defined consent either as an explicit agreement to have sex or as sex wanted by all parties. In the former group, Elliot, a White 20-year-old, emphasized the importance of communication when he defined consent as “[A] verbal or nonverbal cue that the other person wants to continue this particular act or initiate this particular act.” In the latter group, Luis, a Latino 18-year-old, had a simpler interpretation of consent based on mental state. He said, “The fact that both sides are willing to do the same stuff.” Participants like Elliot offered definitions of consent that had a clear course of action for implementation; however, even approaches like Luis’s reflect an understanding of affirmative consent.
Few participants rejected or contested affirmative consent as a whole. Typically, participants who questioned affirmative consent did so under the assumption that affirmative consent required an explicit verbal discussion about sexual intentions—an assumption not reflected in any state and few campus policies (Armstrong et al., 2019). As an example, Will, a White 19-year-old, defined consent as follows: Consent on both sides is—now, if I say it’s a verbal agreement, then I would have by my definition violated my own definition of sexual consent, so I’m not going to say verbal. I’m gonna say it’s common sense. Common sense, but it’s hard to articulate. It is when both partners—or multiple partners—have mutually agreed either by physical or verbal indications that they are willing to participate in sexual intercourse.
Ultimately, Will embraces the ideology of affirmative consent, calling it “common sense” and clearly defining consensual sex as a mutual willingness to participate in sexual events. Only a single participant, Michael, a White 20-year-old, clearly resented affirmative consent, wanting instead a return to a sexual system in which “if a girl does not want to do something, then she’ll let you know [emphasis added].” One additional participant, Ray, a Black 24-year-old, refused to define consent, but insisted that “It’s important that you try to do as much [as you can] to ensure that it’s consent” and provided multiple examples of situations in which he had ended a flirtation or sexual interaction because he believed he did not have clear consent, adding, If you’re unsure and you’re trying to figure [it] out, I would ask her. Ask. ’Cause sometimes it’s like she ripped her clothes off or whatever. That could be consent, but at the same time, a girl could rip her clothes off and then hop in the shower and it’s like, “No, I didn’t want to have sex with you. I wanted to hop in the shower.” . . . It’s nothing that I personally think can [have] a definition for it. It’s something you have to keep in mind at all times.
As a group, the participants demonstrated that they understood the core tenets of affirmative consent policies and actively endorsed affirmative consent as an idea.
Consent-Seeking Strategies
The participants were equally as enthusiastic in their description of tactics to seek consent from their partners. All but one insisted that they applied what they had learned on campus about affirmative consent to their sexual encounters. Of the strategies mentioned by participants when speaking broadly and explicitly about their consent-seeking practices, 34% referred to verbal signals that were unambiguous and explicitly sexual (e.g., asking for consent), 22% referred to physical signals that were explicitly sexual (e.g., enthusiastic participation in kissing), 19% referred to signals that hinted to a partner’s state of mind (e.g., knowing a partner was comfortable), and 16% relied on the absence of a signal that a partner had revoked consent (e.g., not asking to stop). Most participants claimed to use multiple signals to determine whether or not their partner(s) had consented to a sexual action, such as enthusiastic participation in physical sexual activity and the absence of signals that consent had been revoked.
Most commonly, participants insisted that they explicitly asked for affirmative consent. Participants insisted that they explicitly verbally communicated with their partners about their sexual intentions—and most said they did so every time they had sex. Ethan, a White 20-year-old, provides a clear example. He explained, “My personal favorite [way to get consent] is to straight up just ask them. Because I feel like that’s the quickest, easiest, and with the least amount of gray area.” Later he added, “I always ask for consent.” Will, a White 19-year-old, similarly mused, “I’ve never been in a hookup where I haven’t asked [for consent].”
Participants also commonly cited looking to a partner’s physical cues as a way to ensure they had their partners’ consent, referring to a partner’s enthusiastic participation in a sexual act. As an example, Josh, a White 23-year-old, described, “[I make sure] they’re always giving something back. Like a confirmation in some sense. Like if you’re making out with them, they’re not struggling to push you away. They’re kissing you back.” Other participants took more conservative stances, suggesting that the best strategy to seek consent nonverbally is to allow a partner to initiate the sex acts they want. As his primary strategy, Tal, an Asian American 19-year-old, explained, “When I’m in situations where I do end up hooking up with someone, they’re usually the one to initiate it with me, you know? And because of that, I’m pretty sure they consent to it.” In each type of strategy based on physical signals, participants captured the requirement for enthusiasm and agency in a partner’s behavior. The strategies relying on verbal and physical signals of consent met the standards set by affirmative consent policies.
Although nearly all participants condoned affirmative consent in theory, there was a minority whose ideal strategies for consent-seeking would not meet an affirmative consent standard. Instead, they made vague comments about a partner’s state of mind or a lack of indicators that consent was absent. There were very few participants who exclusively mentioned these types of strategies. Only four participants provided only indicators of an absence of consent as their ideal strategies for consent-seeking. For example, Michael, the sole participant who had disparaged the affirmative consent standard, said, “I just always try and, you know, make sure my partner is comfortable with what we’re doing and the second I hear, ‘I’m not comfortable with this’, or, ‘This isn’t doing it for me’, I’ll stop.” Similarly, Zachary, a biracial 22-year-old, articulated, “I expect [my partners] to say no when they don’t want to. And yeah, the absence of the no confirms [that I had consent].” These participants made up a small minority who did not produce at least one strategy of consent-seeking that met an affirmative consent standard, but an important one. They most clearly represent a group of men who still place the burden of sexual assault prevention on women, requiring their partners to say “no” or “stop” rather than taking an active role themselves in ensuring that their sexual interactions are consensual.
Whatever their methods to gauge consent, the participants insisted that they used them reliably. In response to the question, “Have all of your sexual experiences been consensual?” only two answered no. It was common for participants to reference consistent use of the strategies they mentioned as evidence that they had never sexually assaulted anyone.
Consent-Seeking in Previous Encounters
While the participants as a group condoned affirmative consent policies and claimed to consistently apply clear strategies for consent-seeking, their descriptions of gauging a partner’s willingness to engage in their recent sexual encounters relied primarily on ambiguous cues that would not meet an affirmative consent standard. Of the signals referenced by participants describing their most recent sexual encounters, 39% referred to ambiguous and nonsexual physical signals (e.g., eye contact, dancing together), 20% referred to physical signals that were explicitly sexual and varied in degree of ambiguity (e.g., kissing), 13% referred to explicit verbal conversations about sex (e.g., asking for consent), 8% referred to verbal conversations not on the topic of sex (e.g., invitations to cook together), 4% referred to an absence of signals of revoked consent (e.g., telling a participant to stop), and 3% referred to signals that hinted at a partner’s state of mind (e.g., being comfortable). Figure 1 visualizes the participants’ ideal strategies to seek consent in comparison with their reports of consent-seeking behaviors in their previous sexual encounters.
Participants’ ideal strategies to get consent and reported indicators of consent in recent sexual experiences.
The starkest disparity between the participants’ ideal and actual consent-seeking behaviors came from their reports of engaging in explicit conversations about consent. Although explicit conversations about sexual expectations comprised a third of the ideal strategies participants claimed to use, they made up only 13% of the signals they recalled drawing upon in their recent sexual encounters. When an explicit dialogue about consent did occur, a woman usually initiated it. In three-quarters of the cases in this data set, it was the participant’s partner who asked for consent. When a man initiated a conversation about sexual expectations, he usually did not ask for consent, but instead spoke as if his partner’s consent was implied. As an example, Ethan, a White 20-year-old, described the beginning of his most recent hookup: “I was like, ‘Hey, we don’t have to do anything if you don’t want to. Or if we start and you change your mind, we can stop’. And she was like, ‘Okay.’” By telling his partner that she could revoke consent at any time, Ethan recreated a traditional framework in which his partner became responsible for preventing an act of sexual violence during their encounter. Ethan did not continue to verbally check in with her, but instead expected she would speak up if she felt violated by him.
Even in the exceptional cases in which a participant initiated a discussion of consent in the form of a question, he would quickly add that he did not seek consent in this form each time he had sex. For example, Ray, a Black 24-year-old, described how he and his girlfriend decide whether or not to have sex: So if we’re chillin’, she could be like, “Want to have sex?” And I’ll be like, “Yeah, let’s have sex.” Or I’ll be like, “Want to have sex?” And she’ll be like, “Yeah, let’s have sex.” [But] that’s not all the time or whatever.
Participants like Ray used explicit verbal communication about sexual expectations as one way to seek consent, but they did so much less consistently than they had claimed they would when speaking about their ideal strategies.
Instead of consent-seeking through explicit verbal signals, participants most commonly reported relying on their partners’ physical cues, many of which were ambiguous and often nonsexual in nature. These ambiguous cues were actions that could reasonably take place between two people who have no romantic or sexual relationship (e.g., smiling at each other) or could reasonably affect an individual in a nonsexual scenario (e.g., elevated heart rate), making them inherently ambiguous. Accordingly, these cues did not clearly meet the standard of enthusiastic participation required by an affirmative consent framework of being knowing, voluntary, and affirmative indicators of interest in sexual activity. Table 1 lists the most common ambiguous physical and nonsexual signals invoked by participants.
Common Physical Nonsexual Signals of Sexual Consent.
Ambiguous physical and nonsexual signals made up 39% of the cues participants reported using to attribute consent to their partners. As a group, these cues were numerous and varied greatly, making it especially unlikely that women would be aware that their actions could be—and, in these scenarios, were—interpreted as indicators of consent. Out of 224 physical and nonsexual cues mentioned, most were only referenced once or twice. As an example, Jaylen, a Black 18-year-old, used a string of his partner’s physical nonsexual actions to explain how he became certain that sex would occur between them: Everything she did screamed sex. . . . She wore no bra. She clearly knew what she was doing. She put on some great music. I love music, so she knew every right thing to do. . . . [Before that day, we had gone to] this party and we were riding back. We had a small little car with seven people and she chose to sit on my lap. . . . We all get home together and she chilled with me the whole night. . . . [The day we had sex, she was] twerking on my bed, laying down, stretching. Oh my gosh. She’s a dancer, so she really got on my bed, hit a split, and said, “Oh my god. I’m so sore.”
It is entirely possible—and even probable—that Jaylen’s partner consented to have sex with him, but the indicators he invoked in describing the encounter as consensual are not reliable evidence of that. Jaylen’s partner might have stretched before him as a way to entice him, but she might have simply needed to stretch after dance practice. She probably passed many men on her way to meet Jaylen at his dorm, but her decision not to wear a bra was certainly not a sexual invitation to each of them. Similar to Jaylen, other participants invoked signals like receiving a specific emoji in a text message or a woman brushing up against them at a party as signs of consent. The participants sexualized these otherwise mundane interactions to craft their narratives of consent, ultimately suggesting that any woman who engages in polite discourse with them might have consented to engage in sex or could be blamed for sending mixed signals.
As evidence of the problematic nature of these ambiguous signals, consider the way participants in this study thought about moaning. Out of the ambiguous physical nonsexual cues, moaning was by far the most commonly cited by participants as a signal of consent (n = 25). Once sexual contact had begun, many participants relied solely on a combination of moaning and involuntary physiological actions as indicators of ongoing consent. When operating from a presumption of pleasure, the participants’ narratives appear consensual and mutually enjoyable. However, the same excerpts also would not be unusual in a description of sexual violence as indications of pain. For example, Josh, a White 23-year-old, described his partner’s behavior during a recent sexual event with, “I always think moaning is a good [sign]. [And] increased heartrate and then obviously when women get more wet.” Moaning can be a signal of discomfort and increased heart rate and vaginal lubrication can be a response to trauma (Levin & van Berlo 2004). Similarly, Alex, a White 20-year-old, gave a similarly ambiguous description of his most recent hookup when he said, “She was moaning a lot and her eyes were closed and she gave me—I had a lot of scratches on my back.” Scratching, too, can be a signal of pleasure or pain. As with Jaylen, it is entirely possible that Josh and Alex’s experiences were entirely consensual, but it is concerning that the narratives of consent are similar to narratives of sexual abuse.
Participants also commonly brought up eye contact as an indicator of consent (n = 11). When pressed for further details, most participants struggled to explain what differentiated ordinary eye contact—like the eye contact we made during our interviews—from sexualized eye contact that indicated consent. Many made no such distinction at all. Ray, a Black 24-year-old, gave the most attention to the subject. After giving “eyes” as his one-word answer about how he knew his most recent hookup would take place, I asked him to elaborate. He went on, “It’s like an interest—a curiosity. It’s like when you look at someone with a curiosity to know more about them. . . . I can see them like, ‘Hmmm. What is that guy like in bed?’” I asked Ray to physically demonstrate what this type of eye contact would look like. He narrowed his eyes slightly. The difference was nearly imperceptible to me even when I explicitly looked for it. He explained simply, “It’s subtle.” Few participants used eye contact as a sole indicator of consent the way that Ray did, but even those who used many different cues often relied exclusively on other ambiguous signals, such as laughing at a joke, listening carefully, or not looking for an excuse to leave a conversation. None of these signals meet the standards of affirmative consent that the participants endorsed.
Even when participants’ use of physical signals referred to explicitly sexual interaction, the cues were often still applied liberally by the participants. Most often, kissing or the removal of clothing indicated to the participants that they had consent for most—if not all—sexual actions, including vaginal intercourse. As with the participants who used verbal cues, they believed that a partner’s participation in any explicitly sexual interaction promised consent for the duration of the sexual encounter. They did not report seeking ongoing indicators of consent. For example, Josh, a White 23-year-old, said that he “knew” his girlfriend was willing to have sex with him when she “took off her clothes and jumped on top of me and started making out with me.” Similarly, Andrew, an Asian American 19-year-old, believed he had consent to vaginal sex during his most recent hookup when “clothes started coming off.” As with other ambiguous signals, the removal of clothing in front of another person is often correlated with an interest in sex, but many sexual assaults also take place following the voluntary removal of clothes and even after consenting to some sexual activity earlier in the encounter. Some participants who mentioned kissing or removal of clothing as an indicator of consent also commented on enthusiastic participation in other sexual events (e.g., asking for a condom, initiating oral sex), but this was rare.
Ultimately, only one participant described using physical signals as indicators of consent in the same way in describing ideal tactics for seeking consent and in reflection on a recent sexual encounter. Ethan, a White 20-year-old, described, It wasn’t me doing everything. She kind of took turns where I’d be in control and then she’d be in control. [At one point] I was on top and then she . . . flipped me over so I was on my back and then [she] got on top. . . . I assume if she had felt uncomfortable or wanted to stop, when I backed off to allow her to take over, she wouldn’t have.
Ethan’s narrative of his most recent hookup included a key element of affirmative consent that the other men did not: He described allowing his partner to share control throughout the entirety of their encounter, including after sex had begun. It is this element that makes his partner’s consent clear in his narrative. In addition, this counterexample most clearly demonstrates the role of gender and masculinity in other men’s use of ambiguous cues for consent. By keeping their interactions ambiguous, men maintain control over their partners throughout the sexual encounter. They maintain their ability to designate which of their partners’ actions are sexualized and construct a narrative of the sexual encounter in which all of a woman’s actions indicate sexual interest and, accordingly, a man’s successful enactment of masculinity.
Overall, the participants’ narratives of consent-seeking and attribution in their recent sexual experiences were different from the ideal strategies they shared when speaking generally about their beliefs about affirmative consent. The participants did not engage in explicit verbal conversations about consent or sexual expectations as often as they asserted they would, instead relying largely on ambiguous physical actions to indicate consent.
Discussion
This study explores the way college men conceptualize and apply notions of affirmative consent in the wake of changing policies addressing sexual assault prevention on campus as an indicator of men’s broader attitudes about their role in addressing sexual violence across society. Overall, the men in this study endorsed affirmative consent, could clearly define the concept, and claimed to regularly apply specific and unambiguous strategies to ensure their sexual encounters were consensual; however, their narratives about consent attribution in their most recent sexual encounters did not reflect the ideologies they claimed to hold. While participants claimed to primarily obtain consent through explicit verbal communication about sexual expectations or physical cues akin to enthusiastic participation in sexual activity when asked about “affirmative consent,” the participants’ reflections on their past sexual experiences suggested that they more often attributed consent to a partner’s ambiguous physical nonsexual actions.
It is extremely unlikely that every man in this study who attributed consent to a partner’s ambiguous physical actions did so because they lacked compelling evidence that their previous sexual interactions were consensual. More likely, the participants in this study looked to ambiguous signals as a way to demonstrate their sexual skill and maintain power in their sexual encounters. Ambiguity in sexual narratives allows men to embellish their sexual prowess and, accordingly, demonstrate their masculinities (Currier, 2013). In this case, the participants’ narratives hint that their partners were eager to engage in sexual activity with them from the moment of first eye contact. Still, whether or not the men in this study committed acts of sexual violence is not all that is at stake in their varied applications of consent-seeking techniques. Equally as important are the narratives they endorse and normalize by relying on ambiguous signals to tell the stories of their sexual exploits. By sexualizing their partners’ ambiguous and mundane actions, the men in this study make it more difficult to distinguish stories of consensual sex and sexual violence. Previous studies have found that sexual assailants use similar ambiguities to deny that they have committed acts of sexual assault and resist accountability (Scully & Marolla, 1983). Furthermore, men’s widespread willingness to accept mundane interactions like eye contact, moaning, or holding a conversation as evidence of consent upholds rape myths and creates a social context supportive of victim blaming (Hlavka, 2014; Peterson & Muehlenhard, 2004; Smith & Martinez, 1995).
Just as the participants’ use of ambiguous sexual signals does not condemn them, it cannot absolve them either. The men in this study only spoke highly of affirmative consent practices when asked explicitly about “affirmative consent.” When we discussed their previous sexual experiences using more neutral words, the participants’ narratives included few references to consent-seeking behaviors that fit the affirmative consent model. These findings are consistent with Pascoe and Hollander’s (2016) work in which they argued that the label of “rapist” is stigmatizing to young men, indicating their failures of performing masculinity. Accordingly, men might tease each other with allegations of sexual assault as a way to disparage one another’s masculinity and bolster their own, regardless of whether or not the teasers use violence to obtain sexual access to women. The participants’ widespread endorsement of “affirmative consent” might be subject to social desirability biases, particularly now that discussions of consent have become linked with sexual assault prevention. It is possible that some of the men in this study overstated their commitment to sexual assault prevention and seeking consensual sex.
Accordingly, the findings from this study cast past academic research on college students’ gendered norms and sexual practices in a new light. On the topic of sexual consent, Jozkowski and Peterson (2013) and Jozkowski and colleagues (2014) found that college men more commonly rely on physical cues to indicate consent than college women. This study corroborates college men’s dependence on physical cues and offers more detail about what young men consider to be physical signals of consent, which often have an ambiguous connection to sexual activity. Furthermore, this study suggests that men may overstate their self-reported commitment to affirmative consent, especially in response to questions that use the word “consent” explicitly. Accordingly, this study provides evidence that affirmative consent trainings on campus have not successfully eliminated sexual violence. This is unsurprising, given that the sexual assault rate has not appeared to fall after the implementation of these findings and that sexual violence is not the result of mere miscommunication (Abbey, 1982; Abbey, 1987; Hickman & Muehlenhard, 1999; Kitzinger & Frith, 1999; Osman, 2006). Still, the failures of affirmative consent training are worth consideration when moving toward a new model of sexual assault prevention. The findings from this study make a strong case that sexual assault prevention should explicitly discuss gender inequities in sexual interactions, rather than take the currently popular gender-neutral approach. It would be a mistake, however, to assume that the findings presented here are not an improvement for the men in this study. It is entirely possible that men who have received no affirmative consent training are even more likely to ascribe consent to their sexual partners’ ambiguous signals.
This study also offers insight into gendered differences in interpretations of sexual initiation. In their work on college hookups, England and colleagues (2007) found that heterosexual men were more likely to attribute sexual initiation to women than the heterosexual women in the study. The findings from this study suggest that some college men interpret women’s mundane actions such as continuing a conversation as consent to—and perhaps initiation of—a sexual encounter. In these scenarios, it is possible that men are the only ones who consider an interaction sexualized, allowing them greater control over sexual activity that may occur, despite attributing sexual initiation to their partners. These findings have practical implications for those adjudicating sexual assault complaints, as those accused commonly claim the event in question was consensual or even initiated by the complainant. The data from this study exemplify how claims of initiation may be contested and connected to a gendered hierarchy. Expanding administrators’ understandings of gender dynamics in consent negotiations is crucial, particularly as many campus policymakers struggle to understand the nuances of consent themselves, encouraging students to draw upon power-laden cultural norms (Degiuli & Nowotny, 2020).
There are limitations to this study. Although interviews are useful in understanding cultural narratives, the findings from this study rely heavily on self-reported data, which are subject to bias and limit analysis to the narratives crafted. Matched pair interviews would be a useful step forward in triangulating the signals used by consenting sexual partners because both partners’ accounts would be included in the data. Longitudinal research would be illuminating in attempts to better understand the effect of individual affirmative consent policies. Future research may focus on asking more explicitly about sexual events that did not end in vaginal intercourse or sexual events that the participants opted to stop out of concern that their partner was not enthusiastically consenting. It is possible that men draw upon different culture scripts that are more similar to those espoused by affirmative consent policies when facing a potential partner’s clear disinterest. In addition, it is clear from the participants’ rhetoric that they tend to conflate consent and pleasure; however, not all sex is intended to be pleasurable. Future research should investigate how men make sense of their obligation to seek consent during sexual interactions initiated for other reasons, such as sex to conceive a child or sex intended solely to please a partner. Finally, the findings from this study are specific to the United States where affirmative consent trainings have been mandated at public universities. The ideology that affirmative consent is the appropriate standard for designating which sexual events are and are not sexual violence is not shared internationally. In general, future research should investigate how men who have not had exposure to ideals of affirmative consent make sense of their obligation to prevent sexual violence—both off campus and in other national contexts.
In sum, the findings from this study indicate that recent attempts to educate men about affirmative consent and their responsibility to prevent sexual violence have likely reached them, but may not have led them to change their sexual behaviors or the way they invoke ambiguous signaling in narratives of their sexual encounters. Even when young men condone affirmative consent and claim to apply its teachings, they still rely on ambiguous and nonsexual physical cues as evidence that their partners consent to sexual activity. It is the use of these cues that reproduce the cultural notion that consent is unclear.
Footnotes
Acknowledgements
The author would like to thank Rashawn Ray, Dana Fisher, Elizabeth Armstrong, Erin Cech, Fatma Müge Göçek, Alford Young, and Lisa Wade for their support; Matt Elgin, Patteson (Rosie) Delk, and Lejla Bajgoric for all of their hard work as research assistants; and the affiliates of the University of Michigan Gender and Sexuality Workshop and the Department of Sociology’s Professional Writing Course (F16) for their feedback. The biggest thank you to Kristjane Nordmeyer, Jonté Jones, William Rothwell, Heidi Gansen, Kathryn Holland, Jackie Cruz, Betsey Blair, and Colin Eichinger for being incredible friends and extra sets of eyes.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the National Science Foundation Graduate Research Fellowship and the University of Maryland’s Dean Research Initiative.
