Assessing the Effects of Violent Erotica on Attitudes of Sexual Assault
L.M. Allen
Hampden-Sydney College
Since the creation of the Presidential Commission on Pornography and Obscenity in 1970, a debate has raged on whether or not exposure to pornography has a negative impact on societal attitudes and behavior. Although a wealth of valuable research examining the effects of pornography has been conducted, this research has not yet yielded any definitive answers. It has been suggested that media depictions of violence and sexual violence have detrimental effects on a wide range of attitudes. The purpose of the present study is to investigate whether or not exposure to violent pornography has detrimental effects on an individual's attitudes toward date rape.
Past researchers have utilized two different approaches when examining the potentially detrimental effects of pornography on attitudes of behaviors involving sexual assault. First, there is research designed to assess the possibilities of increased aggression or decreased arousal resulting from exposure to sexually explicit material. Second, there are studies which examine the effects that sexually explicit materials have on attitudes regarding sexual assault. In other words, the first body of research focuses on behavioral or physiological changes while the second body of research focuses on attitudinal changes. Although both are valid methods of assessing the effects of sexually explicit materials, each approach investigates a different type of detrimental effect.
A study conducted by Donnerstein (1980) examined the aggressive effects of pornography. Subjects were exposed to one of the three different films; neutral, erotic, and violently erotic. They were then asked to deliver mild shocks to confederates in conjunction with incorrect responses on a given task. Subjects exposed to aggressive erotica showed an increase in aggression toward the confederates, particularly to female confederates, (i.e., these subjects delivered significantly higher numbers of shocks to the confederates for wrong responses than did subjects in the other two groups) (Donnerstein, 1980). This research suggests that exposure to violent erotic material is associated with increased aggression toward others, particularly toward women.
In a second study examining the relationship between aggression and erotica, Donnerstein and Linz (1986) proposed that repeated exposure to violent and sexually violent material caused a desensitization in men to violent and sexually violent depictions toward women. They observed that after each presentation of violent material, subjects scored lower on a physiological arousal test to the material. Thus, there are possible direct and maladaptive behavioral effects of exposure to depictions of violence and sexual violence.
The second type of research, that which examines attitudes, is more prevalent in examining the effects of exposure to sexually explicit material. For example, Linz, Donnerstein, and Penrod (1988) investigated the effects of emotional desensitization to films of violence against women and the effects of explicit and non explicit sexually degrading films on beliefs regarding rape and the sexual objectification of women. Specifically, they examined how male viewers were affected by three different types of pornography: (a) that which was overtly violent, (b) that which was not overtly violent, but was sexually explicit and arguably depicted women as sexually degraded objects, and (c) that which was not sexually explicit, but nearly always portrayed women as sexual objects. Subjects were asked to describe their attitudes toward women, toward alleged female victims of sexual assault, and toward an accused assailant in a reenactment of a legal trial. Subjects in the "R-rated violent" exposure group scored lower on measures of sympathy and of rape empathy than did other subjects. Linz et al. (1988) concluded that repeated exposure to sexually violent material emotionally desensitizes subjects not only to that material, but also to women who are victims of sexual assault.
Weaver (1987) found similar effects of desensitization as well as attitude change toward rape victims as a result of exposure to films with sexual and violent content. There were three phases to Weaver's study. In the first phase, male and female college students read a description of a man and woman who were either stereotypically submissive or promiscuous, and then were asked to describe the stereotypical person on a questionnaire consisting of twenty-four adjectives (e.g., yielding, innocent, irritating, permissive, virtuous). Next, subjects were exposed to one of five nonpornographic (R-rated) film clips: (1) neutral, (2) consensual sex, (3) female-instigated sex, (4) male-coerced sex, and (5) eroticized violence "slasher films." Finally, subjects participated in a person-perception task linked to the description of the man and woman presented to subjects in phase one, and a legal judgment task involving sexual assault and battery (e.g., numbers of years in prison the assailant should receive and the amount of money the victim should be awarded for damages). Results indicated that subjects in the male-coerced sex group and the eroticized violence group rated the non-permissive female peer as less pleasant and less innocent than did neutral subjects. Males in the consensual sex group and the female-instigated sex group perceived the non-permissive peers as more permissive than did neutral males. All subjects perceived females as sexually promiscuous and sexually submissive. Finally, the female-instigated sex group subjects recommended lower sentencing than the neutral or consensual sex groups (Weaver, 1987). This research supports the notion that exposure to violent or sexually violent material not only negatively affects the attitudes of subjects toward the victims of rape, but also enhances attitudes towards the rapist. These subjects viewed the rape victim as more promiscuous, and delivered a lighter sentence to the rapist.
Another study conducted by Intons-Peterson and Roskos-Ewoldson (1987) exposed college males to a fifteen minute clip of a commercially released film depicting either sexual violence, nonviolent sexuality, or neutral content, and tested subjects with the Rape Myth Acceptance (RMA) scale before and after the exposure. The RMA was developed by Deitz, Blackwell, Daley, and Bentley (1982) to assess an individual's empathy and sympathy toward rapists as well as his or her acceptance of myths such as, "all women subconsciously want to be raped." Subjects were then asked to watch a rape trial reenactment and complete a questionnaire regarding their attitudes toward the victim and the defendant. Intons-Peterson and Roskos-Ewoldson found that subjects who were exposed to the sexually violent film scored higher on the RMA (indicating more sympathy for the rapist) than did subjects who saw either the neutral or the nonviolent, sexual films.
In a similar study by Donnerstein, Berkowitz, and Linz (1986), college males were either angered or treated neutrally by a female and then exposed to one of three films - nonaggressive pornography, aggressive pornography, or nonsexual, violent material. These subjects were then asked to complete a questionnaire, which was compiled from the RMA scale, immediately following exposure to the films. Donnerstein et al. (1986) found that both exposure to aggressive pornography and nonsexually violent material resulted in an increase in scores on the RMA scale, with exposure to sexually violent material producing the highest scores on the RMA.
Malamuth, Haber, and Feshbach (1980) also studied the effects of sexually explicit and sexually violent material on attitudes toward rape, but also measured arousal to rape depictions. Subjects were exposed to either a nonviolent, sexual scenario or a sadomasochistic, sexual scenario. Males exposed to the sexually violent material showed an increased in arousal when subsequently presented with a rape scenario. These subjects also viewed the rapist as being less guilty.
Finally, Check (1985) exposed adult males to one of four levels of sexual explicitness to discern if self-reported likelihood to commit rape would be affected. Subjects viewed either sexually violent pornography, nonviolent dehumanizing pornography, nonviolent erotica, and a no-exposure control film clip. Subjects were tested on numerous scales sensitive to the attitudes of males toward rape and sexual violence (e.g., RMA scale and Adversarial Sex Belief scale). Subjects were also asked how likely they would be to commit rape or force a woman to do something against her will if there were no consequences for their actions. Though there was no main effect for the film content on this measure, simple effects comparisons showed that there were significantly higher scores on the likelihood to commit rape for subjects exposed to violent pornography than for subjects who were in the control group.
Hence, the research reviewed in the present study indicates that violent and sexually explicit material may produce undesirable changes in attitudes toward rape, the victims of rape, and the perpetrators of rape. The rape depictions involved in these studies, however, are of situations involving forcible rape carried out by a stranger. One might expect situations of acquaintance rape to be more familiar or pertinent to the college student population. Further, because some examples of acquaintance rape may be more ambiguous, one might expect to see an even greater desensitization effect.
Allen and Staunton (1993) examined the possibility of desensitization to victims of acquaintance rape resulting from exposure to violent and sexually explicit materials. After exposure to film clips which varied in terms of sexual explicitness and interpersonal violence, subjects read two vignettes involving interactions between men and women, one of which being an ambiguous date rape scenario. Subjects were then questioned about their feelings toward the characters in the film clips and vignettes. Our hypothesis was that subjects exposed to sexually violent materials would show the greatest degree (among the four conditions) of desensitization to date rape, evidenced by less sympathy toward the depicted victim and more empathy toward the depicted rapist. Our findings, however, indicated that subjects exposed to violence without explicit sexuality showed the greatest desensitization, not violence coupled with sexuality. These results were unexpected and may have been due to the particular methodology used.
It is possible that there were differences between the patterns of results in the different "violent" conditions because the intensity of the material used in our film clips was not equivalent across all conditions. Specifically, the level of violence in the two film clips that contained violence may not have been equal in intensity. Films which are equal in intensity should be used in the three conditions containing sex and violence. Additionally, the films should depict interactions of males and females, as women were not featured in some of the violent films used in the Allen and Staunton (1993) research.
Second, the non-date rape vignette could be eliminated. The original intent of Allen and Staunton (1993) was to hide the sexual attitudes aspect of the experiment from subjects. In retrospect, this deception was ineffective and unnecessary because the non-date rape vignette was in such sharp contrast to the films and date rape vignette. Using such an arbitrary filler for deception provided subjects with little more than a distraction. In addition, a longer and more detailed questionnaire might obtain a better measure of empathy and sympathy.
Finally, it might be useful to obtain pre-treatment and post-treatment measures of attitudes toward rape, which would enhance sensitivity of measures of change. One possible instrument is the Rape Empathy Scale (RES; Deitz, Blackwell, Daley, and Bentley, 1982), designed to measure empathy that individuals have toward rapists. The RES could yield a more accurate measure for two reasons; (a) the RES has been proven to be an effective measure in prior research, and (b) assessment of both pre- and post-exposure attitudes is a more sensitive method of measuring change.
Despite the slight variations in methods of the research studies examining this topic, one conclusion has been most frequently reached. Specifically, exposure to violent sexually explicit material is associated with attitudinal differences regarding women and acts of sexual assault. It is evident from the research conducted by Donnerstein, Penrod, Linz, and others that subjects exposed to sexually violent material show a desensitization to sexual assault. The present study proposes that if male subjects are exposed to violent or sexually violent material, they will show desensitization to an ambiguous date rape scenario. Namely, by showing less empathy and sympathy for fictitious rape victims and more empathy for fictitious rapists than those in the non-violent conditions as measured by the attitudinal questionnaire.
Method
General Design
The experiment was a 2 x 2 x 2 factorial design. There were two between groups independent variables: degree of violence and degree of sexual explicitness. Each of these had two levels: present or absent. This created four testing groups: (1) violent and sexually explicit, (2) violent without sexual explicitness, (3) sexually explicit without violence, and (4) nonviolent and nonsexually explicit. In addition, pre- and post-treatment measures of rape empathy were taken, creating test period as a repeated measures variable.
Participants
Fifty-three college-age men drawn from introductory psychology classes at Hampden-Sydney College served as participants. The subjects were selected on a volunteer basis and given credit toward their course for participating. Subjects were randomly assigned to experimental treatment conditions before arriving for the experimental session. Between nine and fifteen subjects were assigned to each of the treatment conditions.
Apparatus and Materials
Pre/post Test At least one week prior to and immediately following film exposure subjects completed the Rape Empathy Scale (RES; Deitz, Blackwell, Daley & Bentley, 1982).
Film Exposure Subjects viewed one of four types of film clips: a PG-rated nonviolent and nonsexually explicit film clip (A Room with a View and Tempest), an R-rated sexually explicit nonviolent film clip (Wild Orchid, 9 1/2 Weeks, and Basic Instinct), an R-rated violent and nonsexually explicit film clip (Unlawful Entry, Hear No Evil), and an R-rated violent and sexually explicit film clip (The Accused, I Spit on Your Grave, and Vice Squad). Each of the four film clips were 15 minute segments of commercially released films transferred to a standard videotape. Each was chosen by the experimenter on the basis of content. Specifically, male versus female conflict be it verbal, physical, or otherwise.
Following film exposure subjects were asked to read and complete an attitudinal questionnaire. The questionnaire consisted of eighty question pertaining to subjects' attitudes toward the characters in the films. There were five measures extracted from the questionnaire: (1) six questions addressed subjects' degree of empathy toward the males depicted in the film, (2) six questions measured subjects' degree of empathy toward the females depicted in the film, (3) four questions measured subjects' degree of sympathy toward the males depicted in the film, (4) four questions measured subjects' degree of sympathy toward the females depicted in the film, and (5) two questions assessed subjects' likelihood to commit acts similar to those depicted in the films. Subjects also read a short vignette depicting an ambiguous date rape scenario. The vignette was followed by the same questionnaire that followed the film pertaining to the subjects' attitudes towards the actions taken in the vignette. All questions were in a six point Likert format and measured the subjects' attitudes toward both the man and the woman in the film clips and the vignette.
Procedure
A separate experimenter administered the pre-treatment RES in conjunction with a separate set of surveys being done at least one week before film exposure. This pre-exposure measure was conducted ostensibly by informing subjects that they were completing the RES for an experimenter from another institution in conjunction with her research. Subjects reported to a lecture classroom for participation in this study, and were told that the purpose of the research was to examine differing interpretations which may result from visual versus written presentations of interactions between men and women. The introduction was followed by a fifteen minute film clip exposure. Subjects watched the film, after which the vignette and questionnaire were handed out. Subjects were given fifteen minutes to complete the questionnaires on the film clips, read the vignette, and complete the questionnaire on the characters in the vignette. To prevent subject bias, the post-treatment RES was then readministered in a manner disguising any connection to the present research. Specifically, the experimenter informed subjects immediately after completion of the questionnaire on the characters in the vignette and all the materials were collected that the original RES results from study from the other institution had been damaged. The subjects were then requested to take the RES again to replace the missing tests. A thorough debriefing was conducted immediately thereafter lasting about twenty minutes. This period was used to explain the true purpose and implications of the present research. In conjunction with the debriefing, subjects watched a fourteen minute video on the issue of acquaintance rape, and researchers were available to field questions or concerns.
Results
Several dependent measures were obtained from the questionnaire and the RES scale. First, from the questionnaire, subscores were obtained from each subject by computing individual averages on the six empathy toward the rapist questions, six empathy toward the victim questions, four sympathy toward the rapist questions, four sympathy toward the victim questions, and two likelihood to commit similar acts questions. Second, the male sympathy scores were subtracted from the female sympathy scores and the male empathy scores from the female empathy scores. These "difference scores" were calculated to determine the degree to which individuals vary in comparison of responses to males and females. (A larger difference between males and females would indicate more sympathy or more empathy.) Third, the RES was scored by adding the subjects' responses on the e scale, so that a high RES score indicates high empathy toward the female, and a low score indicates high empathy toward the male. (Because not all of the subjects took the RES the first time, only the 36 subjects who took it both times were included in this analysis.) The measures of sympathy, empathy, and likelihood were analyzed individually using a series of 2 (violence) X 2 (sexual explicitness) ANOVAs. Finally, the pre- and post-treatment RES scores were analyzed for all subjects who had taken both using a 2 x 2 x 2 ANOVA as described above, but with the addition of RES test period treated as a repeated measures factor.
Figure 1 shows the mean level of empathy toward the male depicted in the vignette, as a function of the level of violence and degree of sexual explicitness. An examination of this figure shows that subjects who saw the violent, nonsexually explicit films had the highest empathy ratings, and those in the nonviolent, nonsexually explicit condition had the lowest empathy ratings toward the male character. However, the mean ratings of empathy toward the male tend to be low and the differences small.
Fig.1. Mean empathy rating for male character
The ANOVA revealed that neither the main effect for violence nor for sexual explicitness were significant, but there was a significant violence x sexual explicitness interaction for the degree of empathy toward the male character, (F (1, 49) = 4.742, p < .05). Simple effects comparisons on this interaction showed differences approaching significance between the violent, nonsexually explicit and nonviolent, nonsexually explicit conditions (p = .051) and between the violent, sexually explicit and violent, nonsexually explicit conditions (p = .062).
Figure 2 depicts the mean level of empathy toward the female character depicted in the vignette. Higher levels of empathy toward the females were found than toward the males. Mean empathy ratings appear uniform across the conditions except for a lower level of empathy toward the woman in the violent, nonsexually explicit condition. An ANOVA showed neither significant main effects nor a significant interaction, suggesting that the manipulations had no reliable effect on empathy towards the woman in the vignette.
Fig. 2. Mean empathy rating for female character
Figure 3 depicts the mean level of sympathy toward the male character depicted in the vignette. An examination of this figure shows that subjects who saw the violent, nonsexually explicit films had the highest sympathy ratings, and those in the nonviolent, nonsexually explicit condition had the lowest sympathy ratings. The analysis of these data showed neither significant main effects nor a significant interaction, suggesting that the manipulations had no reliable effect on sympathy towards the man in the vignette.
Fig. 3. Mean sympathy rating for male character
Figure 4 depicts the mean level of sympathy toward the female character depicted in the vignette. This figure shows higher levels of sympathy, overall, toward the females than was seen toward the males. An examination of this figure shows that subjects who saw the violent, nonsexually explicit films had the lowest sympathy ratings, and those in the nonviolent, nonsexually explicit condition had the highest sympathy ratings. The analyses of these data showed neither significant main effects nor a significant interaction, suggesting that the manipulations had no reliable effect on sympathy towards the woman in the vignette.
Fig. 4. Mean sympathy rating for female character
Figure 5 shows the subjects' responses to the questions measuring likelihood to commit an act of sexual assault like the one depicted in the vignette. The figure reveals very low levels of likelihood to commit this act across all conditions, but a higher likelihood rating in the violent, nonsexually explicit condition.
Fig. 5. Mean rating of likelihood of committing acquaintance rape
The statistical analysis revealed that there was a significant main effect for the level of violence, (F (1,49) = 4.186, p < .05), and a significant violence x sexual explicitness interaction, (F (1, 49) = 5,683, p < .05). Simple effects comparisons on the scores involved in the interaction revealed an effect for the level of violence in the nonsexually explicit conditions, (p < .05), and an effect for level of sexual explicitness in the violent conditions, (p <.001). Thus, likelihood to commit sexual assault was greater in the violent nonsexually explicit condition that in the other conditions.
Comparing Figures 1 and 2, and figures 3 and 4, it is clear that the subjects were more empathetic and sympathetic toward the female character. However, and as stated earlier, a large difference in rating would suggest a greater sensitivity to the victim in the vignette, and a small difference would suggest less sensitivity to the victim. Thus, the differences between each subjects' rating of empathy and sympathy was calculated.
The mean difference (female - male) for the empathy ratings is shown in Figure 6. This figure shows that the violent, nonsexually explicit condition exhibits the lowest difference score for empathy while the nonviolent, nonsexually explicit condition exhibits the highest difference score. The analyses of these data showed only a significant interaction for male - female empathy difference, (F (1,49) = 4.769, p < .05). Simple effects comparisons showed an effect for the level of violence in nonsexually explicit conditions, (p < .05), and for the level of sexual explicitness in the violent conditions, (p < .05).
Fig. 6. Mean difference in empathy ratings (Female - Male)
Figure 7 illustrates the difference between mean responses on male sympathy questions and female sympathy questions. This figure shows that the violent, nonsexually explicit condition exhibits the lowest difference score for sympathy while the nonviolent, nonsexually explicit condition exhibits the highest difference score. The analyses of these data showed neither significant main effects nor a significant interaction, suggesting that the manipulations had no reliable effect on differences between male and female sympathy means.
Fig. 7. Mean difference in sympathy ratings (Female - Male)
Finally, the RES scores for each of the treatment conditions for both testing periods are shown in Figure 8. (It should be noted that only 36 subjects that completed the experiment took the pre-test, thus only those subjects' scores are included in this analysis.) All post-treatment measures as illustrated are lower than pre-treatment measures. Surprisingly, all conditions did not score approximately equal on the pre-treatment RES scoring. The scores in the nonsexually explicit groups decreased from pre- to post-treatment by the same amount. Conversely, the sexually explicit groups decreased more from the pre to post, with the violent, sexually explicit condition decreasing most from pre- to post-treatment.
Fig. 8. RES score as a function of test period, for four different types of video exposure
The pre- and post-tests were analyzed with a 2 x 2 x 2 factorial ANOVA, with violence and sexual explicitness treated as independent groups factors, and test period treated as repeated measures. Only the main effect for the test period was significant, (F (1,36) = 9.472, p < .01). Although neither the violence x test period (p = .42) nor the sexual explicitness x test period (p = .08) interactions were significant, it was decided to perform separate tests for simple effects because a priori predictions for lower responses on the RES for each of these had been made. Specifically, the violent and sexually explicit conditions were expected to lead to lower RES scores than nonviolent and nonsexually explicit conditions. According to these analyses, there was a significant simple effect over the test periods for the subjects who viewed the violent films (p < .05), and a significant simple effect over test periods for the subjects who viewed the sexually explicit films (p < .001).
Discussion
The results of the experiment partially support the prediction. I proposed that subjects exposed to violence and violence coupled with sex would show less empathy and sympathy for fictitious rape victims and more empathy for fictitious rapists than those in the nonviolent conditions. The results indicate that subjects exposed to the violent, nonsexual condition were more empathetic toward the fictitious rapist than were the other subjects. Also, those subjects showed a much smaller difference in empathy ratings between the male and female characters. The subjects exposed to the violent, nonsexually explicit condition were more likely to perform the act of sexual assault depicted in the ambiguous date rape scenario than subjects in the other conditions. Finally, levels of rape empathy decreased more from pre- to post-test for subjects exposed to violent material and subjects exposed to sexually explicit material.
These findings support the idea that exposure to violent (and possibly sexually explicit) material raises a person's tolerance to those who commit acts of sexual assault, and is associated with a desensitization of those acts.
A comparison of the differing means of empathy for the male and empathy for the female across groups revealed that subjects in the violent, nonsexual condition showed a smaller difference in their ratings of the male and female than did subjects in the other three groups. I other words, although there was a general tendency for all subjects to have a higher empathy score overall toward the victim, those subjects in the violent, nonsexual condition varied significantly less in their empathy for the victim and the rapist. This suggests that the violent content of this condition may cause subjects to "belittle" the trauma of the victim and perhaps identify more with the rapist.
In addition, subjects who were exposed to the violent, nonsexually explicit material indicated that they were more likely to commit an act similar to the one in the vignette than did any of the other subjects. Although none of the subjects responded in a manner indicating a likelihood to replicate what they read, those in the strictly violent condition showed more of a willingness to act violently than did any other subjects. Higher likelihood scores suggest that these students had less difficulty imagining themselves in the same situation as the rapist than did the other subjects. This supports the idea that violent material may create a level of callousness or desensitization to acts of date rape on a personal level.
Although it was predicted that the results would show that subjects exposed to violence coupled with explicit sexuality would experience the greatest attitudinal desensitization toward the sexual assault in my ambiguous date rape scenario, and that sexual explicitness alone would not cause such a desensitization, the results did not show this effect. Subjects exposed to violence without explicit sexuality showed the greatest desensitization. The failure to find a greater desensitization among subjects exposed to sex coupled with violence runs contrary to past research. For example, Donnerstein, Berkowitz, and Linz (1986) found that subjects exposed to sexually violent material showed a greater desensitization than those exposed to nonviolent sexuality and violent nonsexuality.
It is possible that a greater desensitization in the subjects exposed to violence coupled with explicit sexuality was not found because the intensity of the material used in the film clips was unequal across all conditions. Specifically, the level of violence in the two violent film clips may not have been equal in intensity. As noted earlier, this may have been a problem in the research by Allen and Staunton (1993). An attempt was made in this study to make the content of the films more comparable. Unlike the violent, nonsexually explicit condition in the Allen and Staunton research that consisted primarily of conflicts involving men only, the violent film clips in the current study were selected so that all depictions of violence would involve men and women. In addition, the violent, sexually explicit condition used by Allen and Staunton (1993) was taken entirely from one movie. The violent, sexually explicit condition in the present study used clips from three different films to insure that different characters were viewed, thus avoiding a decrease in intensity which might occur from watching the same individuals. However, it is important to note that no specific pre-testing was done on the materials to judge the equivalence of violence or sexual explicitness across conditions.
Another possible explanation of the lack of effect for the violent, sexually explicit condition is that the film starts with a rape scene. This may have produced demand characteristics or sensitized the subjects to the issue being investigated. Specifically, the subjects may have deduced from the fact that a rape scenario was both watched and read, that rape attitudes were being assessed. If this were the case, then responses to the vignette for the violent, sexually explicit condition might be rendered invalid.
Finally, the coupling of violence and sex may have been disturbing to the extent that subjects were actually sensitized. If the contents of the violent, sexually explicit film clips induced anxiety or anger from the subject, then the predicted effects would certainly not be obtained. Many commercially released films depicting sexual violence are not as graphic as the materials used in the present study. The fact that subjects may not have been accustomed to the intensity of the sexual violence presented in this experiment could have weakened any effects from the violent, sexually explicit condition.
Although empathy measures exhibited a number of predicted effects, there were no effects for the measure of sympathy. There are two possible explanations for this failure to replicate previous research. Perhaps the selected films had no effect on an individual's sympathy or the instruments used in this experiment were not sensitive to that construct.
Another possible difficulty with this research involves the nature of the ambiguous date rape scenario. Perhaps the vignette was not as ambiguous as the author intended. Although the female character acted in a manner that was arguably suggestive, she was, without question, violated. The male character did force sex on the female character against her will. A more ambiguous scenario might include a woman who concedes to sex, but only after a high level of pressure by a male character. These situations occur more frequently, and are thus arguably a larger problem than forcible rape situations (even among people who have had sex with each other before).
A final note on the findings of this study is that although significant effects were not found for all measures, there was a clear and consistent pattern in the results. The violent, nonsexually explicit condition always showed the greatest desensitization, the nonviolent, nonsexually explicit condition always showed the least desensitization, and the remaining two groups were always immediate and approximately equal.
Possibilities for future research could be to evaluate the desensitizing effects of exposure to varying levels of violence, or to study what aspects of violence cause desensitization (i.e., arousal, gore, explicitness, etc.). It would also be important to study the long term effects of this attitudinal desensitization in order to determine what repercussions it might have on society as a whole.
An important issue for society is whether or not exposure to violent material affects the behavior of those exposed. This research did not measure the behavior of the subjects, but it did measure their perception of how likely they would be to commit an act like the one portrayed in the ambiguous date rape scenario. Subjects in the violent, nonsexually explicit exposure group were more likely to state that they would commit acts of sexual assault if no consequences were involved than subjects in the other exposure groups. If these results can be generalized to society as a whole, then we should be aware of the possibility that exposure to violent material may lead to sexual violence.
Emotional desensitization to acts of violence is potentially problematic for our society, for as media presentation of violence or sexual violence increases, so might actual acts of violence or sexual violence. The current research (regarding date rape specifically) is of particular importance to problems which are rampant on college campuses across the nation. If the findings can be generalized to a population, one must accept that a population saturated by materials containing violence and sexual violence will then become desensitized to acts of violence and sexual violence. College aged males who are frequently exposed to violent and/or sexually explicit materials may have a greater risk of desensitization (particularly to date rape because it is often viewed as ambiguous). If a society is callused to violence, then the society may be less likely to respond to the needs of those affected by violence. For example, police will be less likely to intervene in situations of sexual assault or politicians may be less likely to pass legislation protecting the rights of victims of assault. It is anticipated that the present study will assist in the ongoing examination of the link between the emotionally desensitizing effects of pornography and sexual assault in America.
References
Allen, L.M., & Staunton, M.L. (1993). Unpublished manuscript. Hampden-Sydney College, Hampden-Sydney, VA.
Check, J.V.P. (1985). The effects of violent and nonviolent pornography. Cited from Linz, D.G., (1989). Exposure to sexually explicit materials and attitudes toward rape: a comparison of the study results. The Journal of Sex Research, 26, 50-84.
Deitz, S.R., Blackwell, K.T., Daley, P.C., & Bentley, B.J. (1982). Measurement of empathy toward rape victims and rapists. Journal of Personality and Social Psychology, 43(2), 372-384.
Donnerstein, E. (1980). Aggressive erotica and violence against women. Journal of Personality and Social Psychology, 39(2), 269-277.
Donnerstein, E., Berkowitz, L., & Linz, D. (1986). Role of aggressive and sexual images in violent pornography. Cited from Linz, D.G., (1989). Exposure to sexually explicit materials and attitudes toward rape: a comparison of study results. The Journal of Sex Research, 26, 50-84.
Donnerstein, E., & Linz, D. (1986). Mass media, sexual violence, and male viewers. American Behavioral Scientist, 29(5), 601-618.
Intons-Peterson, M.J., & Roskos-Ewoldsen, B. (1987). Mitigating the effects of violent pornography. In S. Gubar and J. Hoff-Wilson (Eds.), For Adult Users Only: The Dilemma of Violent Pornography. Indiana University Press, Bloomington.
Linz, D., Donnerstein, E., & Penrod, S. (1988). Effects of long-term exposure to violent and sexually degrading depictions of women. Journal of Personality and Social Psychology, 55, 758-768.
Malamuth, N., Haber, S., & Feshbach, S. (1980). Testing hypotheses regarding rape: sexual violence, sex differences, and the 'normality' of rapist. Journal of Research in Personality, 14, 121-137.
Weaver, J.B. (1987). Effects of portrayals of female sexuality and violence against women on perceptions of women. Unpublished doctoral dissertation, University of Indiana, Bloomington.