Liberating Discretion: The Effect of Rape Myth Factors on Prosecutors’ Decisions to Charge Suspects in Penetrative and Non-Penetrative Sex Offenses

Abstract

In sexual assault cases, prosecutorial charging decisions may be influenced by legal factors like offense seriousness and convictability and extralegal rape myths. We use data on sexual assaults in Los Angeles, to test for the effects of victim behavior, victim credibility, and “real rape” stereotypes on the decision to file charges. We also test the liberation hypothesis, examining whether rape myths influence the charge decision more in less serious nonpenetrative cases then in penetrative cases. Results show that victim credibility and behavior, but not consistency with real rape stereotypes, affect charging decisions, even after controlling for legally relevant factors, and they influence prosecutors’ charging decisions equally in penetrative and nonpenetrative cases. Rape myths also influence the charging decision indirectly via victim cooperation. We conclude that rape myths are incorporated into the criminal justice system’s definition of and response to sexual violence, so cannot be addressed by changing case screening policies.

Keywords:

• Prosecutor’s charging decision
• sexual assault
• liberation hypothesis
• rape myths

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Suzanne St. George (Coble) is a doctoral student in the School of Criminology and Criminal Justice at Arizona State University. Her research focuses broadly onsexual violence, particularly how incidents of rape and sexual assault are perceived, and to what extent these perceptions affect decision making. She is also interested in neighborhood dynamics especiallyways to promote informal social control and collective efficacy among disadvantaged communities.Cassia Spohn is a Foundation Professor and Director of the School of Criminology and Criminal Justice at Arizona State University. She is the author or co-author of seven books, includingPolicing and Prosecuting Sexual Assault: Inside the Criminal Justice System. Her research interests include prosecutorial and judicial decision making, the intersections of race, ethnicity, crime and justice, and sexual assault case processing decisions.In 2013 she received ASU’s Award for Leading Edge Research in the Social Sciences and was selected as a Fellow of the American Society of Criminology.

Notes

1 In this paper, sexual assault refers to sexually motivated offenses involving physical contact. This can range from sexual battery, which includes touching or fondling the intimate parts of another person, to rape, which includes penile-vaginal penetration, sodomy, oral copulation, and penetration with an object.

2 This study focuses on prosecutors’ decisions to file charges in sexual assault cases involving female victims. Research on the charging decision in cases involving male victims is lacking, despite the low rate of reporting and high rate of attrition in sexual assault cases involving male victims. Extrapolating from the male rape myth literature, it seems likely that prosecutors’ charging decisions in male rape cases are influenced by myths and stereotypes about male rape in the same way that myths and stereotypes about (female) rape affect charging decisions in cases involving female victims. For example, the myth that “men cannot be raped” is consistent with the myth that defines “real rape” as one involving a sober female victim who is raped by a stranger at night with a weapon. Likewise, traditional gender roles dictate that men be physically fit and sexually aggressive. This prescription effectively denies victim status to men, because “real” men cannot be raped by either female perpetrators (they always want sex from women) or male perpetrators (they can successfully defend themselves). Rape myths then create barriers to successful adjudication of sexual assault cases regardless of victim gender and the gender-based discrimination associated with sexual assault case attrition pertains to male victims as well as female victims. In both cases, female and male victims who are perceived as violating their gender roles—that is, women who are not chaste or who engage in risky behavior and men who are victims of rape at all—are defined as illegitimate victims unworthy of adjudication.

3 With a few exceptions, most studies on charging decisions in sexual assault cases focus on felony offenses, which include rape and attempted rape. In the review of literature for this paper, “sexual assault cases” refers to rape and attempted rape, unless stated otherwise.

4 Comparative research uncovers similar findings (e.g. Frank, Hardinge & Wosicke-Correa 2009), as have studies in Israel (Ajzenstadt & Steinberg 2001), Europe (Boyne 2010; Jehle 2012; Kelly 2010; Lovett & Kelly 2009), Australia (Brereton 1997; Dylan, Regehr & Alaggia 2008; Lievore 2003), and Hong Kong (Cheung, Andry & Tam 1990).

5 As stated above, “real rape” describes incidents in which a sober woman is penetrated by a stranger, who resists vigorously, is severely injured, and who reports immediately. There is no question about the victim’s credibility in “real rape” cases. The most important difference in “simple rapes” is that the victim knows her assailant. They often also involve less or no physical resistance, “precipitation,” and delayed reporting. In rape mythology, the characteristics of simple rapes are used to question the victim’s credibility and to minimize the harm associated with the attack.

6 Intoxication may inhibit memory, thereby reducing the ability for a victim to accurately recall details of the rape incident or identify a suspect. So even though victim intoxication is legally irrelevant in the United States—it is unrelated to the criminal elements defining rape—it may still affect prosecutors’ assessments of evidence strength. Likewise, although most jurisdictions have removed corroboration requirements, the evidence corroborating the victim’s testimony obviously strengthens the case.

7 Rather than defining less serious cases as sexual battery and more serious cases as rape/attempted rape, the current study uses penetration to differentiate between more and less serious sexual assault cases for a number of reasons. First, non-penetrative sexual battery cases are wobblers—they can be charged as either misdemeanor or felony offenses—and so the legal severity of these cases is ambiguous. Second, whereas rape clearly differs from less serious attempted rape and sexual battery offenses based on the presence of penetration, sexual battery may overlap considerably with attempted rape depending on the amount and location of physical contact, nudity, and collateral injury.

8 The District Attorney (DA) and the prosecutors working under him/her cooperate with law enforcement to conduct criminal investigations and decide whether or not to file charges in cases referred to them—formally or informally—by the police.

9 Given the increasing use of plea bargaining by prosecutors to resolve criminal cases, some may question the relevance of juries’ beliefs and attitudes to the prosecutor’s decision to file charges. If the likelihood of going to trial is low, then prosecutors should be less concerned with how extralegal factors may affect juries’ decisions to convict suspects. However, in Los Angeles, prosecutors use a trial sufficiency standard (Jacoby, 1980) in filing charges in sexual assault cases. That is, the district attorney’s office will file charges only if there is sufficient evidence to prove the case beyond a reasonable doubt at a jury trial. Moreover, interviews with deputy district attorneys in Los Angeles confirmed that that concerns about victim credibility and the likelihood of conviction at trial motivated their charging decisions (see Spohn & Tellis, 2014, Chapter 3). Therefore, at least in the jurisdiction and sample of cases from which the data for this study are drawn, the downstream orientation toward juries’ decisions to convict or not is still a legitimate concern in the charging process.

10 Spohn and Tellis (2014) found that in LA there was a pre-arrest charge evaluation process in which detectives presented problematic cases to the DA for a filing decision. If the DA said they would not be filing charges, the suspect typically was not arrested, and the case typically was cleared exceptionally. Cases referred informally to the prosecutor in this manner are included in the sample. Only one case involving a pre-arrest consultation resulted in charges being filed so it was not possible to control for how the case was referred—formally or informally through pre-arrest consultation.

11 Garson (2012) explains that if the skew statistic falls between -2 and 2, then the distribution can be considered normally distributed, but that some researchers use the more stringent threshold of -1 to 1 if normality is critical. We use the more stringent threshold and transform variables when the calculated skew falls outside the -1 to 1 range.

12 Sexual battery and attempted rape are also legally distinct, especially in that sexual battery can be charged as either a misdemeanor or felony offense. These offenses are not disaggregated in this study, however, because sexual battery and attempted rape are not clearly differentiated in the narratives included in the case files. These incident types overlap considerably in terms of amount and length of contact, level of nudity, and violence involved. In at least some cases, battery seems to be differentiated from attempted rape only by how successful the victim was at escaping before the sexual activity could escalate. Given the difficulty in differentiating between battery and attempted rape, it makes more sense in this study to aggregate sexual battery and attempted rape together in comparison to completed rape based on the presence of penetration.

13 Whether the victim was willing to cooperate with the detective assigned to the case was determined from the case file. If the victim was uncooperative, it would be noted in the file by the investigating officer (IO). For example, the IO might have noted that he/she had attempted to contact the victim but the victim refused to talk to him/her (either via telephone or in person), that the victim stated that she did not want anything to happen to the suspect/that she did not want the suspect arrested, that the victim said (for a variety of reasons) that she did not want to take the case to court, that the victim stated that she was no longer interested in pursuing a criminal prosecution, or that the victim refused to participate in a pre-filing interview with the district attorney’s office.

14 Asian and Other racial groups were combined with White for two reasons. First, the number of cases involving Asian or Other victims and suspects was small—24 Asian victims and 16 Asian suspects, two Other victims and 10 Other suspects—making this group too small to adequately measure their effect on the outcome. Combining Asian and Other with White created a group of adequate size to include in the models. Second, some sentencing research indicates that Asian defendants are treated similarly to white defendants (Kutateladze et al., 2014; Johnson & Betsinger, 2009). It is reasonable to assume that similarities in treatment will be consistent across different case processing stages, including the charging decision.

15 While there are assumptions that must be met in order to identify binary logistic regression models, these assumptions are arbitrary in the sense that the cannot be tested (Long, 1997). As such no parametric assumption tests were performed.

16 More serious cases may result in more physical evidence or more prompt reports. They may also involve less concern about victim credibility, as prosecutors can overcome behavioral and credibility issues in more severe cases but not less severe ones.