The nature of women's rape fantasies: an analysis of prevalence, frequency, and contents.
Women's rape fantasies pose a special challenge for researchers, as there is something about these fantasies that does not seem to make sense. Why have a fantasy about an event that, in real life, would be repugnant and traumatic? Current evidence indicates that there is nothing abnormal or even unusual about women having rape fantasies (Critelli & Bivona, 2008; Leitenberg & Henning, 1995). For example, romance novels, which are extremely popular forms of literature written for women, allow the reader to participate in a structured fantasy. In these novels, rape of the lead character is a common theme (Thurston, 1987).
The level of confusion on this topic can be illustrated by noting that the two leading explanations of rape fantasy stand in direct contradiction to each other. One states that rape fantasies allow women with high sex guilt to avoid the blame and anxiety that would accompany a consensual sexual fantasy. Because the fantasy involves force, she cannot be blamed for its sexual content (Moreault & Follingstad, 1978). The other leading explanation is that rape fantasies are just the natural expression of an open, accepting, and guilt-free approach to sexuality (Pelletier & Herold, 1988).
The scientific challenge for psychology is to find out why women have rape fantasies and to determine how these fantasies inform our understanding of female sexuality. The foundation for this work lies in determining what goes into a rape fantasy. What type of force is typically used? How much resistance is offered? Is the non-consent real or token? Does non-consent change over the course of the interaction? What is the perpetrator's primary motive? What is the dominant emotion at the end of the fantasy? This investigation is designed to answer these questions.
Before dealing with the typical contents of rape fantasies, we will clarify our use of terms. Following Leitenberg and Henning (1995), we use the term sexual fantasy to refer to almost any conscious mental imagery or daydream that includes sexual activity or is sexually arousing. Sexual fantasies are acts of imagination rather than direct observations of external events or thoughts directed at solving a problem (Singer, 1966). A sexual fantasy can be an elaborate story or only a fleeting thought (Wilson, 1978). It can involve memories of past events or be completely imagined; and it can be intentionally imagined, occur spontaneously, or be stimulated by other events. Consistent with usage in this area of research, the term fantasy does not necessarily imply a desired experience or a pleasurable event.
Rape fantasies are a subset of sexual fantasies. The term rape fantasy follows legal definitions of rape and sexual assault (Corpus Juris Secundum, 2002). As such, it refers to women's fantasies that involve the use of physical force, threat of force, or incapacitation to coerce a woman's self-character in a fantasy into sexual activity against her will. As rape includes the use of force or incapacitation to coerce sex against a woman's will, rape fantasy also includes each of these components. In this sense, rape fantasy is a behaviorally accurate descriptor for these types of fantasies. At the same time, this terminology can be misleading, as it may connote a realistic depiction of violent stranger rape, which in reality is not typical of most actual rapes (Koss & Oros, 1982). In addition, many rape fantasies are not realistic depictions of rape. They are often abstracted, eroticized portrayals that emphasize some aspects of actual rape and omit or distort other features (Kanin, 1982). With this clarification, the term rape fantasy is used in this investigation.
The Content of Rape Fantasies
Kanin (1982) reported the only systematic empirical observations of rape fantasy content. He asked women to describe their rape fantasies and classify them as either sexual, fearful, or a combination of both. Kanin's results indicated that 50% of the women who had rape fantasies reported that these fantasies were completely fearful, 29% reported they were completely sexual, and 21% reported that they were a combination of fearful and sexual. Note that the estimate for completely fearful fantasies is higher than that suggested by other research (Maltz, 1995; Strassberg & Lockerd, 1998), and Kanin's methods were relatively unique. Rather than embedding a rape fantasy item within a sexual fantasy checklist, he asked a lengthy, solitary, and strongly worded item on rape fantasy (see the discussion later). In addition, participants completed their questionnaires in an unusual research setting--a dormitory without experimenter supervision.
After reinterpreting participants' self-descriptions, Kanin (1982) concluded that women's rape fantasies fall into two discrete categories: sexual and fearful. In sexual fantasies, a woman is aggressively approached by an attractive man, who is typically a friend or lover. She gives only token verbal resistance, if any, to a desired sexual encounter. Kanin described these as highly charged, aggressive seductions. Thus, Kanin suggested that women's self-identified rape fantasies that involved sexual arousal should be thought of more as aggressive seductions rather than as rapes. In fearful fantasies, the man is likely to be older, less attractive, and a stranger; and the woman is more likely to see herself as an innocent victim. These fantasies contain coercive and painful violence and no sexual arousal. Kanin did find systematic differences between fearful and sexual fantasies, but there was also considerable overlap with regard to the man's status as a friend or stranger, male age, male attractiveness, and perceived female innocence.
The information gathered by Kanin (1982) is valuable, but his methodology may have led to misleading conclusions. After inspecting the descriptions of the fantasies, Kanin reclassified the self-identified combination participants into the sexual category, although four of these cases were described as "too vague to classify," and the women themselves reported these as rape fantasies that involved both fear and sexual arousal. In contrast, we believe that if a rape fantasy involves the use of force to coerce sex against the will of a woman's self-character, it should qualify as a rape fantasy, regardless of fear or sexual arousal, just as a similar event in real life would qualify as rape. Further, in Kanin's study, 54% of fantasies that were characterized as exclusively fearful contained physically attractive male rapists, suggesting an element of sexual attraction. Rather than conclude that "women's fantasies are all negative or all positive" (p. 116), it may be more accurate to describe these fantasies on a continuum.
Although a number of other research studies have been conducted on rape fantasies, these are limited to anecdotal statements of content. For example, on the basis of informal observations from interviews, Hariton (1973) suggested that the male perpetrator in rape fantasies was often presented as ardent and faceless; the woman offered only token non-consent, and she was often incapacitated by being unconscious, tied up, asleep, or intoxicated. In addition, forced sex fantasies were highly erotic, often leading to orgasm.
Haskell (1976) distinguished between realistic fantasies of rape in which there is no sexual arousal and erotic rape fantasies with high female sexual arousal. She suggested that in erotic rape fantasies, the woman displays only token resistance; and the man uses minimal force, is attractive, and has high levels of sexual desire for her. He is motivated by uncontrollable longing and the desire to enhance her pleasure.
Gold, Balzano, and Stamey (1991) had participants write narratives of their frequent sexual fantasies. Those who included a force fantasy reported feeling more frightened, guilty, disgusted, and less happy after having the fantasy than did other women. Those with a force fantasy may have been describing the type of aversive rape fantasy noted by Kanin (1982). It is also possible that some of these fantasies were negatively toned flashbacks of previous sexual assault, as such flashbacks are common (Foa, Rothbaum, Riggs, & Murdock, 1991; Kramer & Green, 1991; Maltz, 1995).
Strassberg and Lockerd (1998) examined logs of sexual fantasies and suggested that most force fantasies are desired by the women, and that women do not voluntarily engage in erotic fantasies of realistic rape. Bader (2002), on the basis of clinical experience, suggested that some erotic rape fantasies do involve a powerful, ruthless, uncaring brute and that these fantasies are highly exciting for some women. His interpretation was that rape by this type of man relieves dominant women of the need to feel worried or guilty that a full expression of their own sexuality might threaten or overwhelm an actual partner who is believed to be sensitive and weak.
Taken as a whole, current literature on the content of rape fantasies is incomplete and contradictory. There is agreement that mental imagery of realistic violent rape is almost never an erotic experience (Bond & Mosher, 1986) and that nearly all women have no interest in acting out a realistic fantasy of rape (Gold et al., 1991). Beyond that, there is considerable disagreement. Several researchers suggest that the non-consent and resistance in erotic rape fantasies is token, but this has not been empirically demonstrated. Many other questions about the content of rape fantasies also require an empirical answer. For example, how typical is it that the perpetrator is faceless or the woman incapacitated? How often are rape fantasies highly arousing? How often is the man concerned with enhancing the woman's pleasure? How often is the perpetrator selfish and uncaring? How often does the fantasizer have a negative emotional reaction after having the fantasy? How often are rape fantasies flashbacks of previous sexual assault? In addition, what types of verbal or physical abuse are used against the woman? What proportion of rape fantasies, if any, involve erotic reactions to realistic depictions of rape? Many of these questions can be answered for rape fantasies as a whole and for the various types of rape fantasy, if a meaningful typology can be constructed.
This study seeks to provide a more thorough and systematic description of the typical contents of women's rape fantasies and to identify major types of rape fantasy. In particular, it tests Kanin's (1982) descriptive theory that rape fantasies are accurately described by two discrete types against an alternative model: that rape fantasies exist on an erotic-aversive continuum.
The Prevalence and Frequency of Women's Rape Fantasies
A number of studies have assessed the prevalence of women's rape fantasies. All have assessed prevalence with a single item, typically as part of a larger sexual fantasy checklist. With regard to the wording of the rape fantasy item, two main variants have appeared: those using "overpowered or forced" and those using the more explicit language of "rape." At this time, research studies have not included multiple wordings for rape fantasy items within the same study, so wording differences have been confounded with sample differences. This precludes drawing any meaningful conclusions regarding the impact of item wording on participant response, which is one goal of this investigation.
Overall, there are nine reports of the percentage of women who have had rape fantasies (Critelli & Bivona, 2008), and these estimates range from 31% (Person, Terestman, Myers, Goldberg, & Salvadori, 1989) to 57% (Kanin, 1982). Estimates are evenly distributed, with a median of 42%. However, there are reasons to believe that current findings underestimate the actual prevalence, as past research has not assessed the various acts that would constitute rape such as rape by a man, rape by a woman, forced oral sex, and forced anal sex.
As for frequencies of rape fantasies, estimates have varied. Strassberg and Lockerd (1998) found that 55% of women have had rape fantasies, and these women estimated that their rape fantasies occurred an average of three times a month, with a median of once a month. Shulman and Horne (2006), in an Internet survey, reported that 10% of women have rape fantasies once a month or more. Two studies (Hsu et al., 1994; Person et al., 1989, respectively) have estimated the percentage of women who have had rape fantasies within the last 3 months as 22% and 20%. With the exception of Strassberg and Lockerd, the previous estimates are fairly consistent, but it is difficult to compare these rates across studies, as researchers have not consistently distinguished between prevalence and frequency. In addition, it would be useful for researchers to report mean and median frequencies per month or per year for women who have had rape fantasies.
When measuring frequency of rape fantasies, most previous research has used Likert-type scales with nonquantitative markers. For example, Hariton and Singer (1974) asked participants to indicate how often they had each fantasy during intercourse on a scale from 1 (never) to 5 (every time I have intercourse). It is unclear exactly what frequency would be represented by points 2 through 4 on this scale. This study evaluated rape fantasy prevalence and frequency through the use of a sexual fantasy checklist that included the full range of forced sexual acts generally considered to constitute rape and sexual assault. Further, rape fantasies were measured on a scale with a true zero and specific quantitative frequency markers.
Participants were 355 women over the age of 18 from two universities in the Southwestern United States, who received research credit for their participation. Most (81%) were between the ages of 18 and 21 (M=20.04, SD=2.41). The racial and ethnic composition of the sample included 59% Caucasian-White; 16% African American-Black; 13% Hispanic-Latina; 6% biracial; 5% Asian, Asian American, or Pacific Islander; and 1% American Indian or Alaskan Natives. Most participants were single (90%), heterosexual (91%), and Christian (78%).
All measures used in this study were put into a secure Internet site, and participants completed the measures online while in a research lab in the presence of a researcher. Participants completed demographic information, a sexual fantasy checklist, a sexual fantasy log in which they described their most coercive sexual fantasy, and a measure assessing history of sexual victimization. Participants also completed personality measures that were not used for these analyses.
Sexual fantasy checklist. A checklist was adapted for this study from one originally developed by Hariton and Singer (1974). The original items were based on interviews with women, as well as on clinical case studies and fictional literature. This measure has continued to be adapted and used in sex research (Knafo & Jaffe, 1984; Moreault & Folingstad, 1978; Strassberg & Lockerd, 1998).
For this study, items were added to improve the assessment of rape fantasies including items to assess the occurrence of forced sex by a man; forced sex by a woman; forced anal sex; forced oral sex; and forced sex by incapacitation due to drugs, alcohol, or sleep. Other changes included correcting ambiguous or confusing wording, removing redundant items, and separating out items with multiple parts. The checklist contained 80 items. The 8 items used to measure rape fantasy appear in Table 1.
Sexual fantasy log. A sexual fantasy log was constructed for this study. Previous studies have used sexual fantasy logs that asked participants to describe one of their most frequent or favorite sexual fantasies. This format was modified for this study because, given instructions to describe a frequent sexual fantasy, some women who have rape fantasies may not select a rape fantasy for presentation in the log. This log asked participants to write a sexual fantasy that they have had which involved a use of sexual aggression or sexual coercion against them. The log then prompted participants to answer a series of questions for the beginning, middle, and end of the fantasy. The questions asked for a description of the characters in the fantasy; their interaction and sexual acts; and the thoughts, feelings, and motives of the characters.
Participants also provided self-ratings for the beginning, middle, and end of the fantasy for their own level of sexual arousal--ranging from 0 (no sexual arousal) to 4 (extreme sexual arousal); for the self-character's level of sexual arousal (same scale); and for the self-character's level of non-consent--ranging from 0 (total consent) to 4 (extremely strong non-consent). Participants provided overall ratings of the level of sexual aggression in the fantasy--ranging from 0 (no sexual aggression) to 4 (extreme sexual aggression), the level of perpetrator attractiveness--ranging from 0 (not at all attractive) to 4 (extremely attractive), and the fantasy's level of aversiveness or unpleasantness for the fantasizer--ranging from 0 (not at all aversive) to 4 (extremely aversive). As suggested by Hazen (1983), participants also rated both self-character and perpetrator at the end of the fantasy as winner, loser, neither, or both.
To be categorized as a rape fantasy, the following criteria were required: the use of force or incapacitation against the will of the self-character in the fantasy at the time of vaginal, anal, or oral contact or penetration, regardless of any sex acts that may have taken place consensually. The will of the self-character was determined by the participant rating the self-character as having nonconsent greater than zero. As with Kanin (1982), fantasies were also coded as rape fantasies if there was extreme non-consent at the time of attempted penetration.
Two independent raters coded fantasy contents using pre-established scoring criteria for each category. For ratings in which the two primary raters disagreed, ratings of a third independent rater were used. Interrater reliabilities were computed using Cohen's kappa (k) and percentage of agreement. In general, consensus estimates of 70% or greater are preferred for percentage of agreement. Cohen's kappa, however, is a chance-corrected measure of agreement with different guidelines for interpretation (Stemler, 2004). Kappa values of 0.41 to 0.60 are considered moderate, and values above 0.60 are substantial (Landis & Koch, 1977).
Interrater reliability as measured by percentage of agreement ranged from 0.71 to 1, and kappa ranged from 0.40 to 1. Most categories (89%) had a kappa above 0.50 and a percentage of agreement above 0.80. Eighty percent had a kappa above 0.60, and 87% had a percentage of agreement above 0.90. The content categories appear in Table 2.
History of sexual aggression. The sexual experiences survey (Koss & Oros, 1982) is a commonly used measure to identify people from the general population that have experienced sexual aggression or sexual victimization. It is composed of 13 yes-no items and has been found to have an internal consistency of .74 and a 2-week test-retest reliability of .93. In this study, Cronbach's [alpha] = .82.
Kanin's question. Participants also provided a yes-no response to the solitary rape fantasy item used by Kanin (1982):
Since college entrance, did you ever have fantasies (daydreams) about being raped? More specifically, did you ever have thoughts where you put yourself in the place of a woman being raped so that you were able to visualize the rape situation and the rapist? Here we are not referring to those fleeting apprehensions of rape that occur when you find yourself walking alone at night or in strange neighborhoods. (p. 115)
Sexual Fantasy Checklist
The checklist was used to assess the percentage of women who have rape fantasies and the frequency of these fantasies. Women who gave a non-zero response to any of the eight items listed in Table 1 were categorized as having had a rape fantasy. Sixty-two percent of participants reported having a rape fantasy.
Reponses to Items 1 through 5 on Table 1 were combined to form an overall estimation of the frequency of rape fantasy. In combination, these five items cover the range of variations in content found in the legal definition of rape. Further, these specific items are non-overlapping in content (e.g., forced by a man and forced by a woman are separate items); therefore, a combined frequency of these items would be less likely to double-count the same fantasy than would a combination of all eight. The two items on forced oral sex were included in the combined frequency, whereas the item on anal sex was not because we believed that the phrasing of "forced to surrender sexually" would not appear to include oral sex, whereas it would include anal sex. The wording "overpowered or forced" was used to calculate frequency because it is somewhat more neutral than "rape."
To determine the overall rape fantasy frequency, participant responses to each of the five rape fantasy items were converted to a scale reflecting number of times the participant had the fantasy per year. These five frequencies were then summed, and the total was converted back to the 7-point frequency scale. The mean for all participants was 1.52 (SD = 1.62), and the mean for women who had rape fantasies was 2.43 (SD = 1.42). The distribution of rape fantasy frequencies is presented in Figure 1.
Participants were grouped as having either high-frequency or low-frequency rape fantasies based on the combined rape fantasy frequency score. This was done by grouping women with total combined frequency scores of 1 (less than once a year), 2 (a few times a year), or 3 (about once a month) into the low-frequency category. Women with total combined frequency scores of 4 (about once a week), 5 (four or five times a week), or 6 (several times a day) were grouped into the high-frequency category. Using this methodology, 38% of participants never had a rape fantasy, 49% have rape fantasies with low frequency (once a month or less), and 14% have rape fantasies with high frequency (once a week or more).
[FIGURE 1 OMITTED]
To calculate the proportion of rape fantasies for each participant, the participant's combined yearly frequency of rape fantasies was divided by the participant's yearly frequency of consensual sexual fantasies, plus the yearly frequency of non-rape coercive sexual fantasies, plus the combined yearly frequency of rape fantasies. The frequency of consensual fantasies was obtained by using the participant's response to the following item from the checklist: "Engaging in sexual activities that are consensual." The frequency of non-rape coercive sexual fantasies was obtained by using the participant's response to the following item from the checklist: "Being forced to expose my body."
On the yearly scale, the mean frequency of rape fantasy for women who have had a rape fantasy was 49.81 (SD = 136.18), and the median was 3.50. The discrepancy between mean and median is due to a small number of women who had rape fantasies with very high frequency. Because the women with very high frequency had such a large impact on the mean, the median is a better indicator of central tendency for the average person.
The mean proportion of rape fantasies for women who have rape fantasies was 0.20 (SD = 0.27), and the median was 0.06. The mean proportion for women who have high-frequency rape fantasies was 0.45 (SD = 0.25), and the median was 0.49.
Sexual Fantasy Log
A number of content areas were coded for each rape fantasy log (see Table 2). In addition, participants provided self-ratings for the fantasizer's level of sexual arousal for the beginning (M=2.02, SD=1.40), middle (M=2.61, SD=1.41), and end of the fantasy (M:=2.18, SD=1.55); and the self-character's level of sexual arousal for the beginning (M=1.93, SD=1.50), middle (M=2.45, SD=1.50), and end (M=2.10, SD=1.59). Level of non-consent was also rated for the beginning (M=3.24, SD=1.32), middle (M = 2.85, SD = 1.48), and end (M = 2.76, SD = 1.67). In addition, overall self-ratings were given for the level of sexual aggression (M=2.64, SD= 1.06), level of perpetrator attractiveness (M= 2.49, SD = 1.51), and level of aversiveness (M= 1.21, SD = 1.40).
Eighty-six (24%) of the fantasy logs qualified as rape fantasies. Rape fantasies were then categorized as aversive, erotic, or erotic-aversive based on responses to the following two self-report items: the fantasizer's highest level of sexual arousal in response to the fantasy--ranging from 0 (no sexual arousal) to 4 (extreme sexual arousal), and the level of aversiveness of the fantasy--ranging from 0 (not at all aversive) to 4 (extremely aversire). Sexual arousal and aversiveness showed a strong negative correlation (r = -.70, p < .001), supporting the use of an erotic-aversive continuum. Rape fantasies that had a zero response on aversiveness and a non-zero response on sexual arousal were considered exclusively erotic. Rape fantasies that had a zero response on sexual arousal and a non-zero response on aversiveness were considered exclusively aversive. Rape fantasies that had a non-zero response on both sexual arousal and aversiveness were considered to have both erotic and aversive elements.
A continuum between exclusively aversive and exclusively erotic was formed by subtracting aversive scores from erotic scores. Figure 2 displays the results of the continuum: 9% of the fantasies were exclusively aversive, 45% were exclusively erotic, and 46% had both erotic and aversive elements. For fantasies with a score of 0 on the continuum, the aversiveness score was the same as the score for sexual arousal: 1 participant had scores of 0 on both aversiveness and sexual arousal, 1 had scores of 1, 1 had scores of 2, 1 had scores of 3, and 2 had scores of 4.
[FIGURE 2 OMITTED]
History of Sexual Aggression
Among the sample participants, 78% had some history of sexual coercion (answered yes to at least 1 of the 13 items on the scale). The mean total coercion score was 2.87 (SD = 1.78). Twenty-one percent reported that they had experienced acts that would constitute rape.
Wording of Rape Fantasy Items
In response to the checklist item, "being overpowered or forced by a man to surrender sexually against my will," 52% of the participants reported they had the fantasy, whereas 32% reported they had the fantasy of "being raped by a man." This difference was significant: z(353) = 7.50, p < .001.
In response to the item, "being overpowered or forced by a woman to surrender sexually against my will," 17.2% reported they had the fantasy, whereas 9.3% reported they had the fantasy of "being raped by a woman." This difference was significant: z(352) = 4.50, p < .001.
In response to the extended rape question used by Kanin (1982), 25% reported they had the fantasy. This is in contrast to 57% of the participants who reported having the fantasy in Kanin's study, and it was less than one half of the percentage responding affirmatively to the "overpowered or forced by a man" wording in this study.
Research Question: What is the Typical Content of Women's Rape Fantasies?
All rape fantasy logs. Content results for rape fantasies are listed in Table 2. Of all the sexual fantasy logs written, 24% (n = 86) were rape fantasies.
Self-character and fantasizer sexual arousal scores were compared. A paired samples t test was conducted to evaluate whether the fantasizer's highest level of sexual arousal (M= 2.84, SD = 1.43) was significantly different from the self-character's highest level of sexual arousal (M=2.76, SD=1.50). The means did not significantly differ: t(86)=0.91, p = .37.
Major racial and ethnic groups were compared on frequency of rape fantasy, proportion of rape fantasies, erotic-aversive continuum scores, and on each of the log content categories. No significant differences or trends were observed.
Correlations were computed between the fantasizer's highest level of sexual arousal and level of non-consent for the beginning (r=-.46, p<.01), middle (r=-.46, p<.01), and end (r=-.55, p<.01) of the fantasy. The more non-consensual the rape fantasy, the less arousing it was to the fantasizer.
A one-way analysis of variance (ANOVA) was conducted to evaluate the relation between change in the self-character's consent and the maximal level of sexual arousal for the fantasizer. The independent variable, change in consent, had three levels (resistant to willing, willing to resistant, resistant throughout). The ANOVA was significant, F (2, 79)=12.41, p<.001. Post hoc comparisons were made using Dunnett's C test, and pairwise differences among the means revealed that the resistant to willing fantasies were higher in sexual arousal (M=3.45, SD = 0.74) than the resistant throughout fantasies (M=1.97, SD = 1.73). The willing to resistant fantasies were not significantly different from the other types (M= 2.80, SD = 1.23).
Aversive rape fantasies. Of all the rape fantasies written, 9% were completely aversive. In all aversive rape fantasies, self-character non-consent was real; and in nearly all, the self-character was resistant throughout the interaction (91%).
Women who had been raped comprised 21% of the sample. Of the aversive rape fantasies, 36% were written by women who had been raped. Further, there was a negative point biserial correlation between history of being raped and the continuum score (r=-.23, p < .05); those who had been raped wrote logs that were more aversive than erotic.
After the systematic and quantitative evaluation of aversive rape fantasy content, these fantasies were reviewed a second time to look for overall patterns and to provide a more experiential sense of rape fantasy content. The three types of rape fantasy were, however, defined as in the prior quantitative analyses. An initial observation was that aversive rape fantasy logs were unusually brief. The other rape fantasies were two to three times longer than the aversive fantasies.
A closer inspection of the aversive fantasies revealed that a "dark alley" prototype occurred with some frequency. In this type of aversive rape fantasy, the fantasizer imagined a scene in which the self-character was attacked and raped while she was in a vulnerable situation, such as walking alone to her car at night or while sleeping alone in her apartment. These fantasies did not appear to be a reliving of a traumatic situation. Rather, they appeared to be the result of thinking, "What if something bad happens to me?" The following is an example. (Minor spelling and grammatical errors were corrected for ease of reading.):
I am always out with my friends walking somewhere or going to a party when it is really dark outside. My character becomes intoxicated and a strange man runs up behind me and covers my mouth and tells me not to say anything. I can just remember thinking that I am going to get raped and all I see are tears. My character wants to leave but the mysterious man will not let her because he wants to rape her. My character begins to get raped and abused by this mysterious person and all I remember is a lot of crying and sadness. There is not much said except for my character telling the man to stop. I have never actually seen what happens, it always goes blank or to a different scenario. My character feels used, disgusting, and sad. The male character always leaves my character in tears. My character is left alone in the rain and has to walk home. My emotional reaction to it was that I became fearful of going places alone and at night. My self-character used different kinds of resistance like hitting, kicking, screaming, and pleading with the attacker.
Erotic rape fantasies. Of all the rape fantasies written, 45% were completely erotic. In a large majority of these fantasies, the non-consent was feigned or token (85%); and in over three-fourths, the self-character's level of consent changed during the fantasy from resistant to willing (77%).
An informal evaluation of the erotic rape fantasies revealed that a "not right now" prototype scenario came up frequently. In this scenario, the self-character was excited by the idea of the potential sexual interaction but expressed non-consent for reasons, such as a fear of getting caught or not wanting sex with a forbidden partner (e.g., friend's boyfriend). The following is an example:
This friend of mine comes over and immediately shoves me against the wall, pinning my hands over my head and kisses me passionately. The guy in my fantasy is my current boyfriend but I started having this dream last year when we were good friends but both dating other people. But he doesn't look as sweet as he normally does; he looks hungry for me. He does all the initiating. I tell him to stop, that it's wrong and we can't do this. He says he doesn't care; he cannot wait another minute. He's thinking that he has to have me immediately. His motivation is satisfying his own sexual hunger. I am thinking that this is wrong but it feels so good. While my hands are still pinned over my head he uses his other hand to tear off my clothes not caring if they rip. He undresses himself and shoves his body against mine shoving his tongue into my mouth. He tells me he finds me irresistible and that he doesn't care if we are both with other people. I tell him it's wrong and we can't do this. He tells me he knows I want him; he can tell from the way 1 look at him and touch him when we're together. We're both naked and he kisses me all over my body. He is still only motivated with fulfilling his desire. I am begging him to stop, telling him it's wrong and that we can get caught any minute. He picks me up and screws me against the wall. At first it hurts but it feels so good that I can't help but enjoy it. When we're done he leaves because he knows my boyfriend is going to be over soon. He tells me how much he loves my body and how I please him like no woman ever has before and that he would give anything to be with me. I am torn between the pleasure and knowing that it's morally wrong.
Erotic--aversive rape fantasies. Of all the rape fantasies written, 46% were erotic-aversive. A large majority of these fantasies contained genuine non-consent by the self-character (80%).
An informal inspection of the fantasies in the erotic-aversive category revealed a common prototype: "partner goes too far." In this fantasy, the self-character initially consented and seemed to enjoy the sexual interaction. Then, the partner wanted to engage in a sexual behavior that the self-character did not want, and the interaction became non-consensual. The following is an example:
The fantasy begins with both characters in bed looking at each other as though they are in love. Both characters are sexually attractive. The male character is dominating and the female character is passive. The two characters are in the bed about to engage in sexual intercourse. The male begins to kiss on the female's neck. He climbs on top of her as she opens her legs so he can fit in between them. He begins to kiss her all the way to her navel. Then he begins to give her oral sex. He puts his penis in her and begins to penetrate. He then turns her over and continues to have sex with her from behind. He then whispers in her ear that he wants to try new things with her. She already knows that he wants anal sex but she is not ready for that. She tells him no. He whispers again that it won't hurt, "I'll go slow and take my time." She still says no. He then takes his penis out of her vagina and forces it in her anus anyway. She screams in pain and he continues despite her discomfort. As he reaches his climax he pulls out and tells her, "See I told you it wouldn't hurt." She cries herself to sleep. The female character is wondering if she was just raped.
Comparisons of fantasy content. To determine how the three types of rape fantasy (erotic, aversive, erotic-aversive) differed from each other, a series of chi-squares were computed. Because multiple comparisons were made, the criterion for statistical significance was set at .01.
Type of non-consent and rape fantasy group were related, [chi square](2, N=86)=39.97, p<.001. Real nonconsent was more typical of aversive rape fantasies than of erotic rape fantasies, with feigned or token nonconsent more typical of erotic rape fantasies than of aversive or erotic-aversive fantasies.
Change in type of consent and rape fantasy group were related, [chi square](4, N=82)=35.65, p<.001. The change in consent from resistant to willing was more typical of the erotic rape fantasies than of the other two types, and the change from willing to resistant was more typical of the erotic-aversive rape fantasies than of the other two types. Being resistant throughout was more typical of the aversive than of the erotic rape fantasies.
Level of verbal resistance and rape fantasy group were related, [chi square](4, N=85)=17.46, p<.01. No verbal resistance and low-level verbal resistance were more typical of the erotic and erotic-aversive rape fantasies than of aversive fantasies. High verbal resistance was more typical of aversive than of the erotic fantasies.
Level of physical resistance and rape fantasy group were related, [chi square](4, N = 85)= 22.18, p < .001. No physical resistance was more typical of erotic than of aversive rape fantasies, and high-level physical resistance was more typical of aversive than of erotic fantasies.
Fantasizer reaction and rape fantasy group were related, [chi square](4, N=84)=32.54, p<.001. A positive reaction was more typical of the erotic than of the erotic-aversive or aversive rape fantasies, and a negative reaction was more typical of the aversive than of the erotic rape fantasies. A mixed positive and negative reaction was more typical of the erotic-aversive fantasies than of the other two types.
The presence of any aggressive act inflicted on the self-character and rape fantasy group were not significantly related, [chi square](2, N= 85) = 4.98, p > .05. Rape fantasy groups were also compared on the inclusion of physical attack; again, the groups were not significantly different, [chi square](2, N = 85) = 4.88, p > .05.
A one-way ANOVA was conducted to compare perpetrator's level of attractiveness in the three types of rape fantasy. The ANOVA was significant, F(2, 82) = 28.62, p<.001. The strength of the relation between rape fantasy group and perpetrator level of attractiveness, as assessed by [[eta].sup.2], was strong, with type of rape fantasy accounting for 41% of the variance of the dependent variable. Follow-up tests were evaluated as pairwise differences among the means. Because the variances among the three groups were not equal, post hoc comparisons were made using Dunnett's C test, which does not assume equal variance among the three groups. Both the erotic group (M=3.14, SD=1.09) and the erotic-aversive group (M= 2.59, SD = 1.37) included a more attractive partner than did the aversive group (M= 0.9, SD = 0.30), ps < .05.
Additional continuum analyses. A correlation was computed between frequency of rape fantasy and aversiveness of rape fantasy. The correlation was significant (r=-.39, p<.01), indicating that as frequency of rape fantasy increased, aversiveness decreased. A second correlation was computed between frequency of rape fantasy and erotic-aversive continuum scores (r = .42, p<.01), indicating that women with more frequent rape fantasies had fantasies that were more erotic than aversive.
These results suggest that measuring rape fantasies on a checklist that includes the various sex acts that would constitute rape provides a more comprehensive and more accurate assessment of rape fantasies than relying on a single item. In addition, the inclusion of prompts and self-ratings for the logs produced unusually complete rape narratives and valuable interpretive information.
These results indicated that 62% of women have had a rape fantasy, which is somewhat higher than any previous estimate, and about 20 percentage points higher than the median of previous estimates. This relatively high prevalence estimate is the result of the more comprehensive assessment of rape fantasies in this study.
These results also indicated that for women who have had rape fantasies, their median frequency was about four times per year, with 14% of participants reporting that they had rape fantasies at least once a week. Of the two studies that reported specific frequencies, these results are somewhat higher than those of Shulman and Horne (2006) and considerably lower than those of Strassberg and Lockerd (1998).
Somewhat unexpectedly, results also indicated that for women who have had rape fantasies, 20% of their sexual fantasies involved forced sex. This relatively high proportion was largely due to a small number of women who had rape fantasies with high frequency, as the median percentage of sexual fantasies that involved forced sex was 6%. For the 14% of participants who reported having rape fantasies at least once a week, the mean proportion of sexual fantasies involving forced sex was 45%, with a median of 49%. Thus, for a sizable minority of women, rape fantasies comprise nearly one half of their sexual fantasies, suggesting that the rape theme has considerable appeal for millions of women in the United States.
These results indicate that wording of rape fantasy items does make a difference. In this study, 52% of the sample reported having the fantasy, "being overpowered or forced by a man to surrender sexually against my will," whereas only 32% reported the fantasy, "being raped by a man." This 20% discrepancy could be partly due to the stigma related to the word "rape," with some women not wanting to put their fantasy into this category even if it technically fit the criteria. In addition, the label of rape may be used when fantasies are more unpleasant or when they are more violent, and this label may also suggest assault by a stranger to a greater extent than does the phrasing "overpowered or forced to surrender sexually." Although Kanin's (1982) strongly worded single item produced a prevalence estimate of 57% in his sample, the estimate for this item in this study was only 25%. It is possible that this difference is due to the data for the original study being collected in an unsupervised setting. At this time, it appears that using the term rape to assess rape fantasies inappropriately suppresses prevalence estimates.
Rape Fantasy Content
Twenty-four percent (n = 86) of the fantasy logs were determined to be rape fantasies. Although Kanin (1982) categorized the content of women's rape fantasies as being either completely erotic or completely fearful or aversive, the results of this study indicate that this conclusion is inaccurate. Figure 2 shows that rape fantasies vary on a continuum between aversive and erotic, with many fantasies falling between these two extremes. In this study, 9% of rape fantasies were completely aversive, 45% were completely erotic, and 46% were both erotic and aversive.
Several researchers (Hariton, 1973; Haskell, 1976; Kanin; 1982) suggested that the non-consent in rape fantasies or sexual rape fantasies was a "token no." In this study, participants were prompted to clarify the nature of the self-character's non-consent. These results support the token no for erotic rape fantasies but not for aversive rape fantasies, and not for the 46% of rape fantasies that were erotic-aversive, for which 80% showed genuine non-consent by the
self-character. Thus, rape fantasies often have high levels of sexual arousal when there is true non-consent. The highest levels of arousal, however, occurred in rape fantasies for which non-consent was feigned or token.
Aversive rape fantasies. Although previous research suggested otherwise (Kanin, 1982), these results indicate that aversive rape fantasies make up only a small proportion of the overall occurrence of rape fantasies. Descriptive information on the aversive rape fantasies suggests that they have a different tone and quality than the erotic and erotic-aversive fantasies. By definition, these fantasies produced no sexual arousal for the fantasizer.
In all aversive rape fantasies, the self-character's expression of non-consent was genuine, and this is in sharp contrast to the erotic rape fantasies. In nearly all aversive rape fantasies, the self-character remained resistant throughout the fantasy, which again was more common in aversive than in erotic rape fantasies.
In nearly one half of the aversive rape fantasies, the self-character was physically attacked in such a way that bodily damage could have been caused; and in nearly one hall the self-character offered a high level of verbal resistance, such as screaming or crying. In a majority of cases, there was a high level of physical resistance such as scratching, biting, and kicking. As would be expected, high levels of verbal and physical resistance were more common in aversive than in erotic rape fantasies.
In over one half of aversive rape fantasies, the perpetrator was either a stranger or "faceless." Although the perpetrator was a relative in 18% of aversive fantasies, the perpetrator was never a relative in the other two types of rape fantasy. The perpetrator in aversive fantasies was less attractive than the perpetrator in the other rape fantasies, and the only perpetrator motive identified in aversive rape fantasies was to hurt or degrade the self-character.
In over one half of aversive rape fantasies, the self-character was described as the loser. For the fantasizer, the large majority of aversive rape fantasies generated negative feelings such as guilt, shame, and embarrassment, which is similar to findings from Gold et al. (1991). Having negative feelings in response to the rape fantasy was more common for aversive than for erotic rape fantasies. These negative feelings may have resulted from the aversive experience of the fantasy itself and from reactions to having a fantasy that may seem socially inappropriate to some women. As aversive rape fantasies do seem to be quite different from erotic and erotic-aversive rape fantasies, they will most likely require a different type of explanation. Aversive rape fantasies appear to operate as attempts to deal with the fear of actual rape by gaining some sense of control over rape situations and rehearsing how one might deal with actual rape (Gold et al., 1991; Gold & Clegg, 1990).
Women who had been raped comprised 21% of the sample, and 36% of aversive rape fantasies were written by women who had been raped. The correlation between rape status and erotic-aversive continuum scores revealed that women who had been raped rated their rape fantasies as more aversive or less sexually arousing than did other participants.
Erotic rape fantasies. Erotic rape fantasies, which comprised 45% of the rape fantasy logs, are highly erotic fantasies that included forced sex at some point in the story. By definition, fantasies within this group were not at all aversive to the fantasizer.
In a large majority of erotic rape fantasies, the nonconsent was feigned or token (i.e., not a real attempt to end the sexual interaction in the fantasy). In about three-fourths of erotic rape fantasies, the self-character's level of consent changed from being initially resistant to willing, and this type of change in consent was also more common in erotic than in the other two types of rape fantasy.
As with the other rape fantasies, the self-character was physically restrained by the assailant in a majority of these fantasies. Although physical attack was somewhat less common in erotic than in aversive rape fantasies, it did occur in nearly one-fifth of these fantasies. As would be expected, having little to no verbal or physical resistance was more common in erotic than in aversive rape fantasies.
In nearly one half of these fantasies, the perpetrator was a partner or ex-partner; and in one-fifth, he was an acquaintance. The perpetrator in these fantasies was rated as highly physically attractive, more attractive than the perpetrator in the aversive rape fantasies. In nearly one half of the erotic rape fantasies, the perpetrator was perceived as being driven either by his physical or romantic attraction toward the self-character.
A majority of these fantasies were a positive experience for the fantasizer, and this positive reaction was more common for erotic than for the other types of rape fantasy. As these were erotic fantasies that were self-rated as not at all aversive, it is not clear why the percentage of those reporting a positive experience was not even higher. It may be the case that when the reaction here was not positive, this may have been due to guilt from having a fantasy that seemed socially inappropriate.
Hazen (1983) suggested that rape fantasies involve a struggle between an assailant and a potential victim in which it is relevant to consider who is the winner and who is the loser. At one level, it is a struggle over sex, but the woman's non-consent may be feigned or token. At another level, the woman may be seeking a victory that is not about whether sex occurs, but about what happens emotionally between the protagonists. In either case, when one is the winner, the other may not necessarily be the loser. Although the perpetrator was typically described as the winner of the conflict, in a large majority of erotic rape fantasies, the self-character was also described as being either the winner or as neither the winner nor the loser. In other words, in erotic rape fantasies, the perpetrator is often the winner, but the self-character is not usually the loser.
Erotic-aversive rape fantasies. The rape fantasies containing both erotic and aversive elements made up 46% of the rape fantasy logs. These rape fantasies were a mixture of fear, excitement, and sexual arousal. They were both sexually arousing and somewhat unpleasant.
In a large majority of erotic-aversive rape fantasies, the non-consent was real for the self-character. These fantasies showed a broad pattern of changes in consent. In over one-third, the self-character's level of consent changed from non-consent to consent; in over one-third, the self-character was resistant throughout; and in less than one-fourth, the self-character initially consented and later became resistant. The change in consent from willing to resistant was more common in erotic-aversive fantasies than in the other types of rape fantasy.
In one-fourth of these fantasies, the self-character was physically attacked in such a way that bodily damage could have been caused. In about one-fourth, there was either verbal intimidation by the perpetrator or abusive language toward the self-character. A majority contained minimal verbal and physical resistance.
In nearly one half of these fantasies, the perpetrator was a partner or ex-partner; and in nearly one-fourth, the perpetrator was a stranger. The perpetrator in these fantasies was typically rated as moderately attractive. In one-fourth, the perpetrator was perceived as being driven either by his physical or romantic attraction toward the self-character; and in less than 10%, the perpetrator wanted to hurt or degrade the self-character.
In a large majority of cases, either the fantasizer had both a positive and a negative emotional reaction to the fantasy or a negative emotional reaction. As would be expected, the mixed positive and negative reaction was more common in erotic-aversive than the other types of rape fantasy.
These results are limited to university students. It would be important to replicate these findings with older women and with younger women
outside the academic community. Research on non-coercive sexual fantasies does show a decrease in frequency with age (Leitenberg & Henning, 1995), and we would expect a similar pattern with rape fantasies.
Rape fantasies appear to exist on an erotic-aversive continuum, with three main types of rape fantasy: aversive, erotic, and erotic-aversive. We believe that any attempt to explain rape fantasies without distinguishing among these types is unlikely to succeed, as these types are distinct in content, consent, and tone. Aversive rape fantasies often included genuine non-consent, high levels of verbal or physical resistance, and they generated negative emotions for the fantasizer. Erotic rape fantasies often included feigned non-consent, low levels of verbal or physical resistance, and they generated positive emotions for the fantasizer. Erotic-aversive rape fantasies typically included genuine non-consent, low levels of verbal or physical resistance, and they generated both positive and negative emotions for the fantasizer.
We believe that these results represent a step forward in understanding what is meant by a "rape fantasy" and what the typical sex acts, aggressive behaviors, emotions, and motivations are in the various types of rape fantasy. The continuing challenge for psychology is to work toward a comprehensive explanation of why many women have rape fantasies, particularly erotic and erotic-aversive fantasies. The enigmatic quality of rape fantasies suggests that this domain may have important implications for advancing the understanding of women's sexuality.
Bader, M. J. (2002). Arousal. The secret logic of sexual fantasies. New York: St. Martin's Press.
Bond, S. B. & Mosher, D. L. (1986). Guided imager of rape: Fantasy, reality, and the willing victim myth. Journal of Sex Research, 22, 162-183.
Corpus Juris Secundum. (2002). Volume 75. St. Paul, MN: West Group.
Critelli, J. W. & Bivona, J. M. (2008). Women's erotic rape fantasies: An evaluation of theory and research. Journal of Sex Research, 45, 57 70.
Foa, E. B., Rothbaum, B. O., Riggs, D. S., & Murdock, T. B. (1991). Treatment of post-traumatic stress disorder in rape victims: A comparison between cognitive-behavioral procedures and counseling. Journal of Consulting and Clinical Psychology, 59, 715 723.
Gold, S. R., Balzano, B. F., & Stamey, R. (1991). Two studies of females' sexual force fantasies. Journal of Sex Education & Therapy, 17, 15-26.
Gold, S. R. & Clegg, C. L. (1990). Sexual fantasies of college students with coercive experiences and coercive attitudes. Journal of Interpersonal Violence, 5, 464-473.
Hariton, E. B. (1973). The sexual fantasies of women. Psychology Today, 6, 39-44.
Hariton, E. B. & Singer, J. L. (1974). Women's fantasies during sexual intercourse: Normative and theoretical implications. Journal of Consulting and Clinical Psychology, 42, 313-322.
Haskell, M. (1976, November). Rape fantasy. Ms, 84-96.
Hazen, H. (1983). Endless rapture. New York: Scribner.
Hsu, B., Kling, A., Kessler, C., Knapke, K., Diefenbach, P., & Elias, J. E. (1994). Gender differences in sexual fantasy and behavior in a college population: A ten-year replication. Journal of Sex & Marital Therapy, 20, 103-119.
Kanin, E. J. (1982). Female rape fantasies: A victimization study. Victimology: An International Journal, 7, 114-121.
Knafo, D. & Jaffe, Y. (1984). Sexual fantasizing in males and females. Journal of Research in Personality, 18, 451-462.
Koss, M. P. & Oros, C. J. (1982). Sexual experiences survey: A research instrument investigating sexual aggression and victimization. Journal of Consulting and Clinical Psychology, 50, 455-457.
Kramer, T. L. & Green, B. L. (1991). Posttraumatic stress disorder as an early response to sexual assault. Journal of Interpersonal Violence, 6, 160-173.
Landis, J. R. & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33, 159-174.
Leitenberg, H. & Henning, K. (1995). Sexual fantasy. Psychological Bulletin, 117, 469-496.
Maltz, W. (1995). Unwanted sexual fantasies. Contemporary Sexuality, 29, 1-11.
Moreault, D. & Follingstad, D. R. (1978). Sexual fantasies of females as a function of sex guilt and experimental response cues. Journal of Consulting and Clinical Psychology, 46, 1385-1393.
Pelletier, L. A. & Herold, E. S. (1988). The relationship of age, sex guilt, and sexual experience with female sexual fantasies. Journal of Sex Research, 24, 250-256.
Person, E. S., Terestman, N., Myers, W. A., Goldberg, E. L., & Salvadori, C. (1989). Gender differences in sexual behaviors and fantasies in a college population. Journal of Sex and Marital Therapy, 15, 187-198.
Shulman, J. L. & Horne, S. G. (2006). Guilty or not? A path model of women's forceful sexual fantasies. Journal of Sex Research, 43, 368-377.
Singer, J. L. (1966). Daydreaming. New York: Random House.
Stemler, S. E. (2004). A comparison of consensus, consistency, and measurement approaches to estimating interrater reliability. Practical Assessment, Research, & Evaluation, 9(4). Retrieved March 14, 2007, from http://PAREonline.net/getvn.asp?v=9&n=4
Strassberg, D. S. & Lockerd, L. K. (1998). Force in women's sexual fantasies. Archives of Sexual Behavior, 27, 403-414.
Thurston, C. (1987). The romance revolution: Erotic novels for women and the quest for a new sexual identity. Urbana: University of Illinois Press.
Wilson, G. D. (1978). The secrets of sexual fantasy. London: Dent.
Jenny Bivona and Joseph Critelli
Department of Psychology, University of North Texas
Correspondence should be addressed to Jenny Bivona, Department of Psychology, University of North Texas, P.O. Box 311280, Denton, TX 76203. E-mail: firstname.lastname@example.org
Table 1. Sexual Fantasy Checklist: Rape Fantasy Items Frequencies How Often Do You Fantasize About 0(%) 1(%) 2(%) 3(%) 4(%) Each of the Following? 1. Being overpowered or forced 48.2 17.2 13.5 10.4 5.9 by a man to surrender sexually against my will. 2. Being overpowered or forced 82.8 10.2 1.7 2.8 1.7 by a woman to surrender sexually against my will. 3. Being overpowered or forced 75.8 13.2 5.9 2.8 1.4 into a sexual act against my will because I was incapacitated due to drugs, alcohol, or sleep. 4. Being overpowered or forced 72.6 12.4 5.1 6.5 3.1 by a man to give him oral sex. 5. Being overpowered or forced 90.9 3.4 4.0 1.4 0.3 by a woman to give her oral sex. 6. Being overpowered or forced to 83.9 5.6 7.3 1.7 1.4 have anal sex. 7. Being raped by a man. 67.8 14.1 8.2 6.5 2.3 8. Being raped by a woman. 90.7 4.8 2.8 1.4 0.0 Frequencies How Often Do You Fantasize About 5(%) 6(%) M SD Each of the Following? 1. Being overpowered or forced 4.5 0.3 1.98 1.49 by a man to surrender sexually against my will. 2. Being overpowered or forced 0.6 0.3 0.54 1.11 by a woman to surrender sexually against my will. 3. Being overpowered or forced 0.8 0.0 0.70 1.11 into a sexual act against my will because I was incapacitated due to drugs, alcohol, or sleep. 4. Being overpowered or forced 0.3 0.0 0.90 1.26 by a man to give him oral sex. 5. Being overpowered or forced 0.0 0.0 0.27 0.72 by a woman to give her oral sex. 6. Being overpowered or forced to 0.0 0.0 0.50 0.96 have anal sex. 7. Being raped by a man. 1.1 0.0 1.04 1.29 8. Being raped by a woman. 0.3 0.0 0.26 0.72 Note. N = 355. Means and standard deviations are shown for women who have had the fantasy. Scale ranges from 0 (never), 1 (less than once a year), 2 (a few times a year), 3 (about once a month), 4 (about once a week), 5 (four or five times a week), to 6 (several times a day). Table 2. Contents of Rape Fantasy Logs Percentage of All Rape Fantasy Percentage Logs Containing of Erotic This Type of Rape Fantasy Content Content (n = 86) Logs (n = 35) Consensual sexual acts (a): Kissing 30 34 Breast stimulation 7 9 Self-character masturbating 6 9 partner Partner masturbating the 9 14 self-character Oral sex given by self-character 6 6 to partner Oral sex received by 13 17 self-character from partner Vaginal sex 50 72 Anal sex 5 9 Non-consensual sexual acts (a): Kissing 23 32 Breast stimulation 13 17 Self-character masturbating 2 3 perpetrator Perpetrator masturbating 12 17 self-character Oral sex given by self-character 14 23 to perpetrator Oral sex received by 12 14 self-character Vaginal sex 63 54 Anal sex 11 11 Aggressive acts inflicted on self-character (a): Slapped 6 9 Hit 6 3 Bitten 8 6 Hair pulled 11 20 Kicked 0 0 Thrown 12 17 Pushed 22 32 Clothes ripped off 15 26 Any aggressive acts 42 56 Aggressive acts inflicted on perpetrator (a): Slapped 1 3 Hit 5 6 Bitten 6 3 Hair pulled 0 0 Kicked 5 0 Thrown 0 0 Pushed 24 20 Clothes ripped off 0 0 Nature of non-consent: Real for self-character 57 15 Feigned or token for 43 85 self-character Change in self-character's level of consent during the fantasy: Initially resistant, later willing 49 77 Initially willing, later resistant 12 0 Resistant throughout 35 14 Other 4 9 Type of force used by the perpetrator (a): Verbal intimidation 13 11 Abusive language 8 9 Physical restraint 65 63 Physical attack 24 17 Weapon present 2 0 Maximal verbal resistance used by self-character: No verbal resistance 34 43 Low level of verbal resistance 48 57 High level of verbal resistance 18 0 Maximal physical resistance used by self-character: No physical resistance 31 40 Low level of physical resistance 57 57 High level of physical resistance 13 3 Perpetrator's identity: Partner or ex-partner 46 49 Stranger 21 11 Acquaintance 14 20 Authority figure 7 9 Faceless individual 7 3 Relative 3 0 Other 2 9 Perpetrator's motives (a): To hurt or degrade self-character 5 0 Physical attraction to 13 20 self-character Romantic attraction to 15 20 self-character Fantasizer's attitude at end of fantasy: Positive (i.e., feeling enjoyable, 38 66 glad to have the fantasy) Negative (i.e., feeling guilty, 33 6 shameful, embarrassed) Both positive and negative 28 23 Neutral 1 3 Percentage of Percentage of Erotic-Aversive Aversive Rape Rape Fantasy Fantasy Logs Content Logs (n = 40) (n =11) Consensual sexual acts (a): Kissing 33 0 Breast stimulation 8 0 Self-character masturbating 5 0 partner Partner masturbating the 8 0 self-character Oral sex given by self-character 8 0 to partner Oral sex received by 13 0 self-character from partner Vaginal sex 45 0 Anal sex 3 0 Non-consensual sexual acts (a): Kissing 13 37 Breast stimulation 13 0 Self-character masturbating 3 0 perpetrator Perpetrator masturbating 10 0 self-character Oral sex given by self-character 8 9 to perpetrator Oral sex received by 10 9 self-character Vaginal sex 73 55 Anal sex 13 0 Aggressive acts inflicted on self-character (a): Slapped 5 0 Hit 10 0 Bitten 13 0 Hair pulled 2 10 Kicked 0 0 Thrown 8 9 Pushed 18 9 Clothes ripped off 5 17 Any aggressive acts 39 17 Aggressive acts inflicted on perpetrator (a): Slapped 0 0 Hit 3 9 Bitten 8 9 Hair pulled 0 0 Kicked 5 18 Thrown 0 0 Pushed 30 18 Clothes ripped off 0 0 Nature of non-consent: Real for self-character 80 100 Feigned or token for 20 0 self-character Change in self-character's level of consent during the fantasy: Initially resistant, later willing 38 0 Initially willing, later resistant 25 0 Resistant throughout 38 91 Other 0 9 Type of force used by the perpetrator (a): Verbal intimidation 15 9 Abusive language 10 0 Physical restraint 68 64 Physical attack 25 46 Weapon present 5 0 Maximal verbal resistance used by self-character: No verbal resistance 25 46 Low level of verbal resistance 50 9 High level of verbal resistance 25 46 Maximal physical resistance used by self-character: No physical resistance 28 18 Low level of physical resistance 63 27 High level of physical resistance 10 55 Perpetrator's identity: Partner or ex-partner 48 18 Stranger 23 36 Acquaintance 15 9 Authority figure 8 0 Faceless individual 8 18 Relative 0 18 Other 0 0 Perpetrator's motives (a): To hurt or degrade self-character 8 9 Physical attraction to 10 0 self-character Romantic attraction to 15 0 self-character Fantasizer's attitude at end of fantasy: Positive (i.e., feeling enjoyable, 18 0 glad to have the fantasy) Negative (i.e., feeling guilty, 40 81 shameful, embarrassed) Both positive and negative 43 18 Neutral 0 0 Note. n=86. (a) Non-mutually exclusive category. All other categories are mutually exclusive.