Abstract
Previous research has shown that there is higher tolerance of violence against women in cultures with salient gender‐specific honor norms, especially when the violence occurs in intimate relationships and in response to threat to male honor. The present cross‐cultural study (N = 398) extended these findings to sexual aggression (i.e., marital rape) by comparing participants from a culture that emphasizes honor (Turkey) and participants from cultures without strong honor traditions (Germany and Britain). Turkish participants blamed the victim and exonerated the perpetrator more than did German and British participants. In all cultural groups, participants blamed the victim and exonerated the perpetrator more when the husband's reputation was threatened than in the absence of such threat, and in all cultural groups, men blamed the victim and exonerated the perpetrator more than women. Yet, the effect of masculine reputation threat and this pattern of gender differences were somewhat more pronounced among Turkish than German or British participants. Results exploring the predictive role of honor norms at the individual level beyond rape myth acceptance and traditional gender role attitudes revealed that honor norms were the primary predictor of rape perceptions and blame attributions in Turkey (an honor culture), but not in Germany and Britain (dignity cultures) where rape myth acceptance was the strongest predictor. These results provide insights into the cultural factors influencing marital rape judgments in ways that may undermine victim's well‐being and fair handling of rape cases, and highlight the domains most urgently in need of potential intervention.
Keywords: culture of honor, marital rape, rape myths, reputation threat, traditional gender role attitudes, victim and perpetrator blaming
1. INTRODUCTION
Sexual aggression, particularly by an intimate partner, is a severe problem worldwide. Even though rape is considered as a heinous act and a crime, it is often the victim rather than the perpetrator who is blamed and held responsible for the rape (e.g., van der Bruggen & Grubb, 2014). For instance, a large‐scale survey conducted across Europe found that 27% of young men and 20% of young women think that the female victim is partially responsible for the rape in some situations such as when voluntarily going home with someone, or wearing revealing or sexy clothing (European Commission, 2016). These findings are alarming given that blaming rape victims for their plight can significantly undermine their recovery process as well as legal handling of rape cases (Krahé, Temkin, Bieneck, & Berger, 2008).
Judgments about sexual aggression are influenced by a broad range of factors comprising contextual variables and characteristics of the victim, the perpetrator, the observer, and the interaction between these factors (Temkin & Krahé, 2008; van der Bruggen & Grubb, 2014). A relevant observer characteristic is the broader culture as it applies to gender‐related norms, values, and beliefs (e.g., Grubb & Turner, 2012). Although it seems likely that many cultural variables influence judgments of sexual aggression (e.g., traditional gender role beliefs, rape myths), one important cultural variable that is much less studied is culture of honor. We propose that culture of honor seems likely to encourage tolerant judgments of men's sexual aggression against women, especially in situations when the sexual assault occurs within an intimate relationship (Vandello & Cohen, 2008) and when the victim's pre‐rape behavior is perceived as threatening to the perpetrator's reputation (Canto, Perles, & San Martín, 2017). In the present research conducted in three cultures (Turkey, Germany, and Britain), we aimed to investigate blame attributions as a function of honor culture (at the society level), the degree of perceived threat to the male perpetrator's reputation (due to female partner infidelity), and gender of the observer in the context of marital rape.
1.1. Cultures of honor versus dignity
The honor‐face‐dignity culture framework proposes that cultures differ based on their different approaches to managing social order and relationships, and for defining the grounds for self‐worth (Leung & Cohen, 2011). According to this framework, honor cultures are a specific form of collectivism, prevalent in the societies of circum‐Mediterranean, Middle East, and Latin America, and are different than face‐based collectivism which dominates the East Asian societies (as we focus on honor and dignity cultures, we will not discuss face cultures any further).
In honor cultures, self‐worth strongly depends on one's reputation and his/her own assessment of what others think. Thus, people are socialized to vigorously defend their reputation, which often requires engaging in aggressive behaviors in response to insults or other threats to their reputation. Because one's reputation is influenced by the company one keeps, people are concerned with protecting not only their own reputation, but also the reputation of their close others such as family and friends. In contrast, in dignity cultures, which dominate societies of Northern/Western Europe and Northern America, self‐worth is viewed as inherent and inalienable, which is not affected by the perceptions and actions of others as much as in honor cultures. Thus, one's sense of worth in his/her own eyes and in the eyes of others is relatively impervious to insults and affronts, and people are not required to vigilantly and aggressively defend their own and family's reputation (Leung & Cohen, 2011).
Culture of honor (vs. dignity) distinction seems especially relevant for studying cultural differences in justification and tolerance to aggression. Studies demonstrated that individuals from a variety of honor cultures are more accepting of male‐to‐male aggression as well as male‐to‐female aggression than those from dignity cultures, but only when the aggression occurs in response to perceived reputation threats, such as insults (e.g., Cohen, Nisbett, Bowdle, & Schwarz, 1996) or partner infidelity (e.g., Vandello & Cohen, 2003).
Based on a large body of literature that classifies Turkey, a Mediterranean/Middle‐Eastern country, to have strong honor values and traditions (e.g., Uskul & Cross, 2019), in this study, we used Turkey to represent an honor culture. Furthermore, prior studies have shown Western European countries to have strong dignity values (e.g., Guerra, Giner‐Sorolla, & Vasiljevic, 2013). Therefore, we used Germany and Britain, from which comparative data are scarce, to represent dignity cultures. Although Germany and Britain represent dignity cultures, we still explored potential differences between the two given that past research revealed differences between German‐ and English‐speaking countries on beliefs that serve to justify rape (e.g., Gerger, Kley, Bohner, & Siebler, 2007) and that the two countries differ in theoretically relevant macro‐level variables, such as levels of gender equality and gender development (UNDP, 2019).
1.2. Culture of honor and tolerant perceptions of physical aggression against partners
Although detection or suspicion of infidelity is a leading cause of men's partner‐directed aggression all around the world (Buss & Duntley, 2011), this type of aggression against women is particularly high in traditional honor cultures such as the Middle East (Kulczycki & Windle, c). In honor cultures, social norms and values emphasize the importance of female chastity and loyalty, and regard these as issues as part of men's reputation (Vandello & Cohen, 2008). A man's ability to exert control over his female partner is seen as an important component of his reputation, and her infidelity represents the ultimate act that damages his reputation and brings shame upon him, his family, and even the community.
Consistent with these notions, cross‐cultural studies conducted with North and South American samples have found that individuals from honor cultures (Brazilians, Chileans, U.S. Southerners, and Hispanic Americans) perceived a woman's infidelity as reflecting more negatively on her husband's reputation and manliness compared to individuals from dignity cultures (U.S. Northerners and Anglo‐Saxon Canadians). Moreover, individuals from these honor cultures perceived a man who hit his unfaithful wife as manlier and excused the man's violent behavior toward his wife (Dietrich & Schuett, 2013; Vandello & Cohen, 2003; Vandello, Cohen, Grandon, & Franiuk, 2009). Importantly, a man's aggression towards his wife was only excused by honor culture individuals when conflicts were infidelity‐related, not when conflicts were unrelated to infidelity (e.g., spending too much money).
1.3. Culture of honor and tolerant perceptions of rape
Besides physical aggression, few studies have shown that culture of honor is implicated in sexual aggression. Data comparing honor and dignity states within the United States showed that men in honor states are more likely to engage in sexually coercive behaviors toward women than in dignity states (Brown, Baughman, & Carvallo, 2018). In addition, there is some empirical support for the association between endorsement of honor norms and tendency to blame the victims and condone the perpetrators (e.g., Canto et al., 2017; Saucier, Strain, Hockett, & McManus, 2015). Although these studies are informative, they did not examine blame attributions in the context of marital rape when the victim's threat to the husband's reputation before rape varies in intensity. The current study conducted in Turkey, Germany, and Britain goes beyond these single‐culture studies and extends initial findings by systematically examining judgments of marital rape as a function of honor culture (both at the society and individual levels), degree of threat to the husband's masculine reputation before rape as well as observer gender.
Numerous studies have shown that marital rape is perceived as less severe than other forms of rape and also less likely to be considered a crime (e.g., Ben‐David & Schneider, 2005; Rebeiz & Harb, 2010). There is also evidence that perpetrators are attributed less, and victims are attributed more blame when the rape occurs between spouses than when the victim and the perpetrator are strangers or acquaintances (e.g., Krahé, Temkin, & Bieneck, 2007). Furthermore, investigating the effect of culture of honor on blame attributions in the context of marital rape seems important, given that culture of honor leads to tolerant judgments of men's aggression against women especially in intimate relationships and when the perpetrator experiences a threat to his reputation (Vandello & Cohen, 2008).
Related to this point, studies show that the victim's behavior before or during the sexual assault has a considerable impact on blame attributions. For example, victim blaming increases when the victim's behavior is at odds with the stereotypical notions of female respectability, such as dressing “provocatively” in skimpy clothes (Whatley, 2005). In the same vein, whether the victim's pre‐rape behavior is perceived as threatening to the perpetrator's reputation may also influence blame attributions, whereby individuals may attribute more blame to the victim and less blame to the perpetrator in the presence of such a threat. Although, we would expect this outcome to hold in both honor and dignity cultures, reputation threat's effect on victim blaming may be especially pronounced in honor cultures.
1.4. Observer gender and other cultural predictors of marital rape perceptions
Beyond victim's pre‐rape behavior, another significant factor which influences rape blame attributions is the observers' gender. Studies have shown that men blame the victim more and the perpetrator less than do women, regardless of the relationship between the parties (see van der Bruggen & Grubb, 2014, for a review). Thus, we expect men to blame victims more and perpetrators less than women across both honor and dignity cultures. Yet, given that men generally endorse gender‐based honor values more strongly than women, and that this gender difference seems especially pronounced in honor cultures (e.g., Vandello et al., 2009), higher victim blaming and lower perpetrator blaming by men than women may be stronger in honor than in dignity cultures.
Notably, victim blaming and tolerant judgments of perpetrators are viewed as extensions of traditional gender roles, in which men are expected to be dominant, powerful, and sexually assertive, whereas women are expected to be submissive and passive (Newcombe, Van Den Eynde, Hafner, & Jolly, 2008). The relationship between traditional gender role attitudes and more assignment of blame to rape victims has been demonstrated in numerous studies (e.g., Grubb & Turner, 2012). Blaming the victim and exonerating the perpetrator has also been associated with higher rape myth acceptance in a large body of research (see van der Bruggen & Grubb, 2014, for a review). However, there is little knowledge regarding the degree to which culture‐related factors such as traditional gender role attitudes, rape myth acceptance, and honor norms, contribute to rape blame attributions in honor versus dignity cultures. Thus, a secondary aim of this study was to advance understanding of the cultural factors that influence rape blame attributions by examining the predictive strength of these three factors in honor versus dignity cultures.
1.5. The present study
The primary aim of the present study was to investigate blame attributions as a function of a culture of honor (at the society level), the degree of perceived threat to the male perpetrator's masculine reputation (due to female partner infidelity), and gender of the observer in the context of marital rape. Following previous research (e.g., Dietrich & Schuett, 2013; Vandello & Cohen, b), we operationalized masculine reputation threat as female partner infidelity and used vignettes to experimentally vary the severity of the masculine reputation threat experienced by the husband before the rape by presenting the partner's sexual infidelity as publicly known, unknown, or not having committed infidelity. Based on the research discussed above, we derived the following hypotheses:
H1: We expect a main effect of culture such that individuals from an honor culture (Turkey) would attribute more blame to the victim (wife), less blame to the perpetrator (husband), perceive the crime as less severe, and be less likely to identify the incident as rape than individuals from dignity cultures (Germany and Britain).
H2a: We expect a main effect of threat such that individuals across the three cultures would attribute more blame to the victim, less blame to the perpetrator, perceive the crime as less severe, and be less likely to identify the incident as rape when marital rape occurs after a threat to the perpetrator's masculine reputation than in the absence of such threat.
H2b: We expect a culture and threat interaction such that the effect of masculine reputation threat (vs. absence of threat) on increased victim blaming and decreased perpetrator blaming, perceived crime severity, and rape identification would be especially strong for individuals from an honor culture, whereas individuals from dignity cultures would not be affected by the presence or absence of reputation threat as strongly.
H3a: We expect a main effect of gender such that across the three cultures, men would attribute more blame to the victim, less blame to the perpetrator, perceive the crime as less severe, and be less likely to identify the incident as rape than would women.
H3b: We expect a culture and gender interaction such that the higher victim blaming, less perpetrator blaming, less perceived crime severity, and less likelihood of rape identification observed among men than women would be especially pronounced in an honor culture compared to dignity cultures.
In addition to hypotheses H1–H3, we explored the predictive role of individual‐level endorsement of honor norms, traditional gender role attitudes, and rape myth acceptance simultaneously to highlight the unique cultural dynamics influencing rape blame attributions in each culture (Turkey, Germany, and Britain).
2. METHOD
2.1. Participants
We conducted a power analysis in G*Power to determine sample size. Our power analysis yielded a total required sample size of 335 (~112 per cultural group) for a power of .90 to detect a small effect size of d = .30 with a multivariate analysis of variance (MANOVA) test on four response variables, three independent variables and four covariates, using an alpha of .05.
Initially, 498 participants (154 Turkish, 178 German, and 166 British) completed the survey entitled “A study on sexual problems in relationships.” Excluding one participant who did not specify her/his gender, 19 participants with a migration background from the German sample, 35 participants who were not White‐British from the British sample, and further excluding those who failed to pass attention check questions (22 Turkish, 15 German, and 8 British participants) left a final sample of 398 participants (132 Turkish, 143 German, 123 British). Participants were a mixture of students and community members recruited through college participant pools and social media platforms (students formed 55% of the Turkish, 69% of the German, and 67% of the British samples). In addition, British participants were also recruited through Prolific Academic. The main demographic characteristics of participants from each cultural group are reported in Table 1. Cultural groups differed in age, degree of religious upbringing and current religiosity, political orientation, and socioeconomic status (results of these tests are available in the online the Supporting Information Material [SM]), thus we reported the results controlling for these demographic variables given that these variables may influence rape victim and perpetrator perceptions.
Table 1.
Culture | Gender | N | Age a | Religion (%) | Religiosity b | Political orientation c | Socioeconomic status d | |||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Childhood | Current | |||||||||||||
n | % | M | SD | Range | M | C | N | O | M (SD) | M (SD) | M (SD) | M (SD) | ||
Turkey | Men | 61 | 46.2 | 25.34 | 4.41 | 19–40 | 60.7 | 3.3 | 36.1 | 0 | 2.72 (0.99) | 2.20 (1.24) | 59.43 (28.08) | 52.25 (19.02) |
Women | 71 | 53.8 | 25.89 | 6.33 | 18–50 | 53.5 | 0 | 42.3 | 4.2 | 2.37 (1.03) | 1.80 (0.95) | 75.68 (18.58) | 58.69 (15.80) | |
132 | 100 | 25.64 | 5.51 | 18–50 | 56.8 | 1.5 | 39.4 | 2.3 | 2.53 (1.02) | 1.98 (1.11) | 68.17 (24.74) | 55.71 (17.59) | ||
Germany | Men | 59 | 41.3 | – | – | – | 0 | 49.2 | 49.2 | 1.7 | 2.28 (1.13) | 2.05 (1.14) | 66.73 (19.76) | 61.51 (17.01) |
Women | 84 | 58.7 | – | – | – | 1.2 | 41.0 | 51.8 | 6.0 | 2.14 (1.07) | 2.05 (0.97) | 67.74 (19.26) | 57.58 (17.01) | |
143 | 100 | – | – | – | 0.7 | 44.4 | 50.7 | 4.2 | 2.20 (1.09) | 2.05 (1.04) | 67.32 (19.41) | 59.20 (17.06) | ||
UK | Men | 72 | 58.5 | 23.06 | 5.40 | 18–43 | 0 | 25.0 | 75.0 | 0 | 2.15 (1.04) | 1.44 (0.85) | 63.44 (21.56) | 48.35 (17.36) |
Women | 51 | 41.5 | 19.57 | 2.28 | 18–34 | 0 | 35.3 | 60.8 | 3.9 | 2.41 (1.15) | 1.51 (0.86) | 63.14 (23.66) | 51.65 (22.01) | |
123 | 100 | 21.61 | 4.70 | 18–43 | 0 | 29.3 | 69.1 | 1.6 | 2.26 (1.09) | 1.47 (0.85) | 63.32 (22.36) | 49.72 (19.41) |
Abbreviations: C, Christian (all denominations); M, Muslim; N, no religion; O, other religions (e.g., Judaism, Buddhism).
Due to a programming error, we failed to collect data on German participants’ age, yet they were in the undergraduate student age‐group.
Degree of religiosity (1 = not at all, 5 = extremely), Childhood refers to degree of religiosity during childhood and Current refers to current degree of religiosity.
Political orientation (0 = extremely conservative, 100 = extremely liberal).
Socioeconomic status (0 = low/working class, 100 = high/upper class).
2.2. Design and procedure
Participants were randomly assigned to one of the three reputation threat conditions (high, low, no threat; see SM for the vignettes), which entailed a rape scenario between a married couple. Thescenarios were created with the aim of varying the perceived severity of the masculine reputation threat experienced by the husband before the sexual assault incident. Both the high and low threat conditions used a scenario including a husband whose wife commits sexual infidelity and verbally derogates his sexual competence. The only difference between the two conditions was that in the high threat condition, the wife's infidelity was publicly known in the community, whereas in the low threat condition the wife's infidelity was only known by those involved. We included the two threat conditions to explore any potential differences between a masculine reputation threat that involves wide publicity versus the private nature of the wife's infidelity on threat perceptions and rape blame attributions. In the no threat condition, the wife was presented as hanging out with their female neighbor whom the husband is not fond of. In all three conditions, the husband finds out about the infidelity (high and low threat condition) or the friendship with the female neighbor (no threat condition). In his rage, he rapes his wife. After reading their assigned scenario, participants were asked about their perceptions of the husband and the wife, and to evaluate their behaviors.
The study protocol and all instruments were approved by the institutional review board of the first author's university. The German and Turkish versions of the questionnaire were created by careful back translation procedures, comprising the translation from the English original to German and Turkish by bilingual native speakers of German/Turkish and back translated by bilingual speakers fluent in the respective languages.
2.3. Measures
Participants completed the measures in the order described below. Unless otherwise stated, all were answered on 7‐point scales ranging from (1) not at all to (7) very much.
2.3.1. Manipulation check for scenarios
To test if our scenarios were successful in manipulating the degree of masculine reputation threat, we measured participants' perceived reputation threat and perceived masculinity threat by asking them to rate the extent to which they think the wife's behavior before the sexual incident was damaging to her husband's social‐image, reputation, and honor (three items; α TR = .90; α GER = .87; α UK = .96; α All = .93) and the extent to which it was a threat to the husband's masculinity and a challenge to his manhood (two items; r SB‐TR = .87; r SB‐GER = .92; r SB‐UK = .93; r SB‐All = .91).
2.3.2. Victim and perpetrator blaming
Seven items measured the extent to which participants blamed the victim (e.g., “How much do you think the wife is to blame for what happened?”) and seven corresponding items measured the extent to which participants blamed the perpetrator (e.g., “How much do you think the husband is to blame for what happened?”; taken after Abrams, Viki, Masser, & Bohner, 2003). The scores on these items were averaged to create composite measures of victim blaming (α TR = .88, α GER = .87, α UK = .91, α All = .92) and perpetrator blaming (α TR = .80, α GER = .70, α UK = .83, α All = .80).
2.3.3. Perceived crime severity
Participants were asked two items to rate the extent to which they think the husband should be held criminally liable for what he did (1 = not at all, 7 = absolutely) and the appropriate punishment he deserves (1 = no punishment, 2 = fine, but no prison, 3 = 1–7 years in prison, 4 = 8–15 years in prison, 5 = 16–20 years in prison, 6 = more than 20 years in prison). The two items were averaged to create a composite measure of the husband's perceived crime severity (r SB‐TR = .82; r SB‐GER = .72; r SB‐UK = .63; r SB All = .73).
2.3.4. Rape identification
Participants were asked one item to rate the extent to which they think this incident meets the legal definition of rape (1 = not at all, 7 = absolutely).
2.3.5. Gender‐based honor norms
Participants were asked to rate their agreement with gender‐specific honor norms (1 = strongly disagree, 7 = strongly agree) using the 9‐item Honor Endorsement Index (HEI; Vandello et al., 2009). Five items were specific to male honor (e.g., “A man must defend his honor at any cost”), and four items were specific to female honor (e.g., “A woman must protect the family's good reputation.”). As done by the original authors, we averaged all items to create a total honor endorsement score (α TR = .87, α GER = .85, α UK = .90, α All = .88) with higher scores indicating higher endorsement of gender‐based honor norms.
2.3.6. Traditional gender role attitudes
Gender role attitudes were measured with a simplified version of the Attitudes Toward Women Scale (AWS; Spence & Helmreich, 1972) developed by Nelson (1988). Participants were asked to rate 22 statements (e.g., “It sounds worse when a woman swears than when a man does”; 1 = strongly disagree to 5 = strongly agree). Items were averaged to create a composite score (α TR = .91, α GER = .85, α UK = .90, α All = .90). Higher scores indicate a more traditional gender role orientation.
2.3.7. Rape myth acceptance
An 11‐item short version of the Acceptance of Modern Myths About Sexual Aggression Scale (AMMSA; Gerger et al., 2007) was used to measure contemporary rape myth acceptance (e.g., “Many women tend to misinterpret a well‐meant gesture as a sexual assault”; 1 = completely disagree to 7 = completely agree). Items were averaged to create a composite score (α TR = .89, α GER = .93, α UK = .92, α All = .90). Higher scores indicate higher acceptance of rape myths.
3. RESULTS
In total, only little data was missing on the studied variables (0.5% or less). Assuming a missing at random (MAR) mechanism, the list‐wise deletion procedure was applied in our analyses. Table 2 presents means and standard deviations on all dependent variables (controlling for demographic variables: religious upbringing, current religiosity, political orientation, and socioeconomic status). Tables S1–S3 in the SM present bivariate correlations separately for each cultural group.
Table 2.
Turkey | Germany | Britain | ||||
---|---|---|---|---|---|---|
Threat | No threat | Threat | No threat | Threat | No threat | |
Gender | M (SD) | M (SD) | M (SD) | M (SD) | M (SD) | M (SD) |
Perceived reputation threat | ||||||
Male | 5.65 (1.68) | 3.18 (1.86) | 2.87 (1.14) | 1.39 (0.46) | 4.52 (1.78) | 2.13 (1.62) |
Female | 4.72 (1.68) | 3.00 (2.37) | 3.05 (1.39) | 1.29 (0.70) | 3.93 (1.69) | 1.41 (0.82) |
Total | 5.10 (1.73) | 3.11 (2.06) | 2.97 (1.29) | 1.33 (0.62) | 4.28 (1.76) | 1.83 (1.37) |
Perceived masculinity threat | ||||||
Male | 5.58 (1.71) | 3.15 (1.98) | 3.87 (1.89) | 1.72 (0.81) | 4.56 (1.74) | 2.13 (1.41) |
Female | 4.88 (1.92) | 2.74 (1.99) | 4.79 (1.63) | 1.56 (1.20) | 3.87 (1.82) | 1.35 (1.10) |
Total | 5.16 (1.86) | 2.98 (1.97) | 4.40 (1.80) | 1.62 (1.07) | 4.28 (1.80) | 1.80 (1.33) |
Victim blaming | ||||||
Male | 5.46 (1.00) | 4.21 (1.74) | 2.78 (1.35) | 1.83 (0.78) | 3.92 (1.46) | 2.29 (1.25) |
Female | 4.44 (1.19) | 2.48 (1.27) | 2.47 (1.09) | 1.37 (0.49) | 2.84 (1.41) | 1.46 (0.85) |
Total | 4.86 (1.22) | 3.49 (1.77) | 2.61 (1.21) | 1.53 (0.64) | 3.48 (1.52) | 1.94 (1.16) |
Perpetrator blaming | ||||||
Male | 4.00 (1.34) | 4.52 (1.43) | 4.76 (1.04) | 5.56 (1.01) | 4.46 (1.28) | 5.66 (0.72) |
Female | 4.45 (1.16) | 5.51 (1.14) | 5.27 (0.90) | 6.08 (0.58) | 5.08 (1.26) | 6.10 (0.46) |
Total | 4.27 (1.25) | 4.93 (1.40) | 5.05 (0.99) | 5.89 (0.79) | 4.72 (1.30) | 5.85 (0.66) |
Crime severity | ||||||
Male | 3.30 (1.75) | 4.15 (1.36) | 4.44 (1.09) | 4.19 (0.99) | 4.66 (0.90) | 4.70 (0.75) |
Female | 4.53 (1.41) | 5.15 (1.11) | 4.77 (0.67) | 4.91 (0.82) | 4.62 (1.31) | 5.00 (0.85) |
Total | 4.03 (1.66) | 4.56 (1.34) | 4.63 (0.89) | 4.66 (0.94) | 4.64 (1.08) | 4.83 (0.80) |
Rape identification | ||||||
Male | 4.38 (2.43) | 5.08 (2.04) | 5.90 (1.57) | 5.39 (1.61) | 6.33 (1.27) | 6.48 (0.73) |
Female | 5.72 (1.99) | 5.94 (1.92) | 6.08 (1.26) | 6.09 (1.42) | 6.24 (1.56) | 6.59 (1.06) |
Total | 5.18 (2.27) | 5.44 (2.01) | 6.00 (1.40) | 5.84 (1.52) | 6.29 (1.38) | 6.52 (0.88) |
HEI | ||||||
Male | 4.53 (1.33) | 2.59 (1.19) | 2.86 (1.21) | |||
Female | 3.08 (1.32) | 2.34 (0.94) | 2.61 (1.36) | |||
Total | 3.75 (1.51) | 2.44 (1.05) | 2.76 (1.28) | |||
AWS | ||||||
Male | 2.15 (0.79) | 1.62 (0.47) | 1.68 (0.52) | |||
Female | 1.50 (0.46) | 1.37 (0.26) | 1.56 (0.58) | |||
Total | 1.80 (0.71) | 1.47 (0.38) | 1.63 (0.54) | |||
AMMSA | ||||||
Male | 3.16 (1.28) | 2.89 (1.27) | 3.13 (1.20) | |||
Female | 2.08 (0.96) | 2.19 (0.93) | 2.35 (1.25) | |||
Total | 2.58 (1.24) | 2.47 (1.13) | 2.80 (1.27) |
Note: Means and SDs for HEI, AWS, and AMMSA are for all threat conditions combined.
Abbreviations: AMMSA, Acceptance of Modern Myths about Sexual Aggression Scale; AWS, Attitudes Towards Women Scale; HEI, Honor Endorsement Index.
3.1. Preliminary analysis
3.1.1. Manipulation check for scenarios
To check if our manipulation of masculine reputation threat worked, we conducted a 3 × 3 (Culture × Threat; culture: Turkey, Germany, Britain; threat: high, low, no threat) MANOVA on perceived reputation and masculinity threat variables, controlling for demographic variables (degree of religious upbringing and current religiosity, political orientation, and socioeconomic status). Relevant pairwise contrasts were examined using Sidak correction.
The analysis revealed a significant multivariate main effect of threat, F(4, 764) = 46.39, p < .001, ηp 2 = .20, and multivariate interaction effect of culture × threat, F(8, 764) = 2.35, p = .017, ηp 2 = .02. The univariate interaction effect was nonsignificant for perceived reputation threat, F(4, 382) = 1.57, p = .181, ηp 2 = .02, and for perceived masculinity threat, F < 1. The univariate main effect of threat was significant for both perceived reputation threat, F(2, 382) = 80.85, p < .001, ηp 2 = .30, and perceived masculinity threat, F(2, 382) = 90.82, p < .001, ηp 2 = .32. The wife's behavior before the sexual incident was perceived as significantly more threatening to the husband's reputation and masculinity in both high and low threat conditions compared to the no threat condition, ps < .001. As expected, the high threat condition was perceived as more threatening to the husband's reputation than the low threat condition (p = .013), but high and low threat conditions did not differ from each other on perceived masculinity threat (p = .992). The pairwise contrasts tests examining simple effects of threat within each culture and of culture within each threat condition are presented in the SM.
Overall, these results suggest that both threat conditions led to a higher threat perception than the no threat condition, but high threat condition did not warrant a higher threat perception than the low threat condition. Thus, we collapsed the high and low threat conditions into a single threat condition.
3.1.2. Culture and gender differences in endorsement of honor norms, traditional gender role attitudes, and rape myths
To verify that culture of honor is indeed more salient in Turkey than in Germany and Britain, and to examine culture and gender differences on endorsement of honor norms, traditional gender role attitudes, and rape myth acceptance, we conducted a 3 × 2 (culture × participant gender) MANOVA on these scales, again controlling for the same demographic variables. Again, relevant pairwise contrasts were examined using Sidak correction. The analysis showed significant multivariate effects of culture, F(6, 770) = 21.78, p < .001, ηp 2 = .15, and gender, F(3, 384) = 17.42, p < .001, ηp 2 = .12, as well as a significant culture × gender interaction, F(6, 770) = 3.61, p = .002, ηp 2 = .03.
For the endorsement of honor norms, significant main effects of culture, F(2, 386) = 53.30, p < .001, ηp 2 = .22, and gender, F(1, 386) = 21.95, p < .001, ηp 2 = .05, as well as a significant culture × gender interaction, F(2, 386) = 6.21, p = .002, ηp 2 = .03, were revealed. As expected, Turkish participants endorsed significantly higher levels of honor norms than did German and British participants (ps < .001), confirming the honor culture conceptualization of the Turkish society. An unexpected finding was that British participants endorsed significantly higher levels of honor norms than did German participants (p = .017), indicating differences within the prototypical dignity cultures too (we discuss this finding later in the Discussion section). Furthermore, Turkish men endorsed significantly higher levels of honor norms than did Turkish women (p < .001), but German men and women (p = .210), and British men and women (p = .205) did not differ.
With respect to traditional gender role attitudes, significant main effects of culture, F(2, 386) = 16.98, p < .001, ηp 2 = .08, and gender, F(1, 386) = 34.47, p < .001, ηp 2 = .08, as well as a significant culture × gender interaction effect, F(2, 386) = 4.35, p = .014, ηp 2 = .02, were found. Turkish participants had higher traditional gender role attitudes than did German and British participants (ps ≤ .016), and British participants had higher traditional gender role attitudes than did German participants (p = .024). Furthermore, men had higher traditional gender role attitudes than did women in Turkey and Germany (ps ≤ .002), but British men and women did not differ (p = .156).
On rape myth acceptance, only a significant main effect of gender emerged, F(1, 386) = 47.06, p < .001, ηp 2 = .11, with men accepting rape myths more than women.
3.2. Results testing the confirmatory hypotheses
To test our hypotheses, we conducted a 3 × 2 × 2 (Culture × Threat × Participant gender; culture: Turkey, Germany, Britain; threat: absence vs. presence) MANOVAs on the rape blame attribution variables, controlling for the same demographic variables. Relevant pairwise contrasts were examined using Sidak correction.
The multivariate main effect of culture was significant, F(8, 756) = 19.89, p < .001, ηp 2 = .17. The univariate main effect of culture was significant for all dependent variables, victim blaming, F(2, 380) = 89.06, p < .001, ηp 2 = .32; perpetrator blaming, F(2, 380) = 17.92, p < .001, ηp 2 = .09; crime severity, F(2, 380) = 5.02, p = .007, ηp 2 = .03; rape identification, F(2, 380) = 11.95, p < .001, ηp 2 = .06. As predicted by H1, Turkish participants blamed the victim more (ps < .001) and the perpetrator less (ps < .001), perceived the crime as less severe (ps ≤ .070), and were less likely to identify the incident as rape (ps < .008) compared to German and British participants. British participants also blamed the victim more than did German participants (p = .014), but no other differences between two cultures were observed (ps ≥ .132). These results supported H1 (direct tests showing that these differences between Turkey and Germany/UK on rape blame attributions are indeed explained mainly by individual differences in endorsement of honor norms are presented in Tables S4 and S5 in the SM).
The multivariate main effect of threat was significant, F(4, 377) = 32.27, p < .001, ηp 2 = .26. Significant univariate main effects of threat were obtained for victim blaming, F(1, 380) = 110.08, p < .001, ηp 2 = .23, perpetrator blaming, F(1, 380) = 54.90, p < .001, ηp 2 = .13, and crime severity, F(1, 380) = 4.77, p = .030, ηp 2 = .01, but not for rape identification, F < 1. As predicted by H2a, the victim was blamed more and the perpetrator was blamed less in the threat condition than in the no threat condition (ps < .001). Furthermore, the perpetrator's crime was perceived as less severe in the threat condition than in the no threat condition (p = .030). In sum, H2a was supported with regard to victim and perpetrator blaming and perpetrator's crime severity.
The multivariate interaction effect of Culture × Threat was marginally significant, F(8, 756) = 1.80, p = .073, ηp 2 = .02. Significant univariate interaction effects of Culture × Threat were obtained for crime severity, F(2, 380) = 3.64, p = .027, ηp 2 = .02, but not for victim blaming, F(2, 380) = 1.74, p = .176, perpetrator blaming, F < 1, and rape identification, F(2, 380) = 1.66, p = .191. Despite the nonsignificant interaction effects on the majority of the variables, we still examined the planned contrasts to test our directional prediction (H2b) that only Turkish (not German and British) participants should assign more blame to the victim, and less blame to the perpetrator, perceive the perpetrator's crime as less severe, and have lower rape identification in the threat condition than in the no threat condition. Mirroring the main effect of threat found among the whole sample, participants within each culture blamed the victim more and the perpetrator less in threat condition than in the no threat condition (ps ≤ .001). Importantly, however, only Turkish participants perceived the perpetrator's crime as less severe in the threat condition than in the no threat condition (p = .001), but German and British participants' perceived crime severity did not change between the threat conditions (ps ≥ .347). Ratings on rape identification did not differ as a function of threat in any culture (ps ≥ .219). Overall, H2b was supported only with regard to perceived crime severity.
The multivariate main effect of gender was significant, F(4, 377) = 11.51, p < .001, ηp 2 = .11. Significant univariate main effects of gender were obtained on all dependent variables, victim blaming, F(1, 380) = 39.84, p < .001, ηp 2 = .10; perpetrator blaming, F(1, 380) = 20.14, p < .001, ηp 2 = .05; crime severity, F(1, 380) = 16.50, p = .001, ηp 2 = .04; and rape identification, F(1, 380) = 4.66, p = .032, ηp 2 = .01. As predicted by H3a, men compared to women blamed the victim more and the perpetrator less, perceived the perpetrator's crime as less severe, and were less likely to identify the incident as rape.
A marginally significant multivariate interaction effect of Culture × Gender also emerged, F(8, 756) = 1.79, p = .076, ηp 2 = .02. The univariate Culture × Gender interaction was significant only for victim blaming, F(2, 380) = 3.05, p = .049, ηp 2 = .02, but not for perpetrator blaming, F < 1, crime severity, F(2, 380) = 2.25, p = .107, and rape identification, F(2, 380) = 1.05, p = .351. In Turkey, men blamed the victim more, the perpetrator less, perceived the perpetrator's crime as less severe, and were less likely to identify the incident as rape compared to women (ps ≤ .047). In Britain, men blamed the victim more and the perpetrator less than did women (ps ≤ .010), but no gender differences emerged for crime severity and rape identification (ps ≥ .525). In Germany, men blamed the perpetrator less and perceived the perpetrator's crime as less severe compared to women (ps ≤ .007), but no gender differences emerged for victim blaming and rape identification (ps ≥ .075). Overall, these results provided support for H3b by showing that the expected pattern of gender differences in rape blame attributions is especially pronounced in Turkey (an honor culture), whereas in Germany and Britain (dignity cultures) the gender difference is still there, but less pronounced. Multivariate Gender × Threat, F < 1, and Culture × Gender × Threat interactions, F(8, 756) = 1.09, p = .366, were not significant.
3.3. Results testing the exploratory hypothesis
We explored whether individual‐level endorsement of honor norms predicts rape perceptions and blame attributions beyond rape myth acceptance and traditional gender role attitudes in each cultural group (Turkey, Germany, Britain). We conducted three‐step hierarchical regression analyses with participant gender, threat, and demographic variables in Step 1, traditional gender role attitudes (AWS) and rape myth acceptance (AMMSA) in Step 2, and honor norms (HEI) in Step 3. Table 3 presents the hierarchical regression results on the outcome variables separately for each culture.
Table 3.
Victim blaming | Perpetrator blaming | Perpetrator's crime severity | Rape identification | |
---|---|---|---|---|
Turkey | ||||
Gender a | .17* | .02 | −.01 | .09 |
Reputation threat b | .49*** | −.28** | −.19** | −.08 |
Religious upbringing | .04 | .03 | .04 | .11 |
Current religiosity | −.08 | .01 | −.10 | −.17 † |
Political orientation | −.09 | .11 | .19* | .10 |
Socioeconomic status | −.08 | −.04 | .07 | .05 |
AWS | .04 | −.25 † | −.08 | −.23 † |
AMMSA | −.10 | .01 | −.22* | −.15 |
HEI | .46*** | −.27* | −.27* | −.28* |
R 2 (Step 2) | .419 | .272 | .502 | .459 |
R 2 (Step 3) | .500 | .299 | .528 | .487 |
ΔR 2 | .081*** | .027* | .026* | .029* |
Germany | ||||
Gender a | −.01 | −.10 | −.09 | .00 |
Reputation threat b | .44*** | −.38*** | .02 | .07 |
Religious upbringing | .05 | −.03 | .11 | −.01 |
Current religiosity | .00 | −.04 | −.18* | −.09 |
Political orientation | −.03 | −.07 | .01 | .01 |
Socioeconomic status | −.01 | .03 | −.04 | −.04 |
AWS | .05 | −.07 | −.24* | −.04 |
AMMSA | .46*** | −.48*** | −.32** | −.29* |
HEI | .07 | .07 | .03 | −.18 † |
R 2 (Step 2) | .483 | .414 | .331 | .186 |
R 2 (Step 3) | .486 | .416 | .332 | .203 |
ΔR 2 | .003 | .001 | .000 | .017 † |
Britain | ||||
Gender a | .19** | −.13 † | .11 | .13 |
Reputation threat b | .45*** | −.42*** | −.07 | −.06 |
Religious upbringing | .12 | −.12 | −.08 | ‐.07 |
Current religiosity | −.01 | −.11 | .03 | −.08 |
Political orientation | −.03 | .03 | .12 | .14 |
socioeconomic status | −.04 | .09 | .10 | .10 |
AWS | .03 | .03 | −.17 | −.23* |
AMMSA | .37*** | −.31** | −.36** | −.31** |
HEI | .17* | −.10 | −.12 | −.01 |
R 2 (Step 2) | .531 | .398 | .326 | .311 |
R2 (Step 3) | .547 | .404 | .334 | .312 |
ΔR 2 | .016* | .006 | .008 | .000 |
Abbreviations: AMMSA, Aceptance of Modern Myths about Sexual Aggression Scale; AWS, Attitudes Towards Women Scale; HEI, Honor Endorsement Index.
1 = female, 2 = male.
0 = no threat, 1 = threat (high and low threat conditions combined).
p < .10.
p < .05.
p < .01.
p < .001.
In Turkey, introducing HEI in Step 3 significantly increased the amount of variance explained for all outcome variables (ps ≤ .032). In detail, HEI was positively linked to victim blaming and negatively linked to perpetrator blaming, crime severity, and rape identification. In contrast, in Germany, HEI did not explain further variance when entered in Step 3 (ps ≥ .096), and in Britain, HEI explained variance in victim blaming (p = .048), showing a positive association, but not in the remaining outcome variables (ps ≥ .246).
With respect to AMMSA, in Turkey, it only predicted crime severity (p = .024), showing a negative association, but it did not predict other outcome variables. However, in Germany and Britain, AMMSA was significantly linked to all outcome variables; AMMSA was positively linked to victim blaming and negatively linked to perpetrator blaming, crime severity, and rape identification (ps < .032). Furthermore, AWS explained no variance in Turkey and only little variance in Germany and Britain: It was negatively linked to crime severity in Germany, and to rape identification in Britain.
4. DISCUSSION
Cultural norms, particularly gender‐based honor norms, seem crucial for understanding rape blame attributions, especially when rape occurs in intimate relationships and the victim is perceived as threatening the perpetrator's masculine reputation before rape. However, research investigating the factors that make victim blaming and lenient perceptions of rape more likely has paid little attention to these factors. Addressing this gap, the present study conducted in Turkey, Germany, and Britain examined rape perceptions and blame attributions as a function of honor culture (at the society level), the degree of perceived threat to the male perpetrator's masculine reputation (due to female partner infidelity), and gender of the observer in the context of marital rape. It also examined the predictive role of individual‐level endorsement of honor norms, traditional gender role attitudes, and rape myth acceptance simultaneously to highlight the unique cultural dynamics influencing rape blame attributions in honor versus dignity cultures.
Cultural psychology research and theory distinguishes between honor and dignity cultures, with Western European countries generally described as dignity cultures and Turkey as an honor culture (Guerra et al., 2013; Uskul & Cross, 2019). In the present study, we demonstrated that Turkish participants endorsed significantly higher levels of honor norms than did German and British participants, confirming these notions. However, we also found that British participants endorsed significantly higher levels of honor norms than did German participants, showing variation in the strength of honor norms between the two dignity cultures. This finding is interesting, given that in both cultural samples, we excluded participants (migrants or participants with non‐White ethnicity) who could have been socialized with honor norms, and controlled for societal differences in the demographic variables that correlated with honor norms (religious upbringing, current religiosity, political orientation, and socioeconomic status). A potential explanation may be that the UK has a lower country‐level gender equality ranking than Germany (UNDP, 2019). Although our three‐country data did not allow for testing the correlation between country‐level gender equality and mean scores on honor endorsement, we did find that traditional gender role attitudes ‐ a measure of gender (un)egalitarian attitudes ‐ were stronger in the British than in the German sample. Furthermore, when traditional gender role attitudes were controlled, mean differences on honor norms between the UK and Germany became nonsignificant (p = .266), providing some support for the gender egalitarianism argument.
Our first hypothesis addressed honor versus dignity cultural differences. As predicted, compared to participants from dignity cultures (Germany and Britain), participants from an honor culture (Turkey) blamed the victim more and the perpetrator less, perceived the crime as less severe, and were less likely to identify the incident as rape. As reported in the SM, these cultural differences were indeed explained mainly by differences in endorsement of gender‐based honor norms. Our findings were consistent with Vandello and Cohen (2003) who studied physical aggression and found that individuals from an honor culture (Brazilians) perceived physical aggression against women more justified than individuals from a dignity culture (Northern Americans). Cross‐cultural data examining the impact of cultural norms on judgments about rape victims and perpetrators is generally scarce. Previous studies considering ethnicity or nationality showed, for instance, that South African college students blamed the rape victim more than did college students from Australia (Heaven, Connors, & Pretorius, 1998), and Asian and Hispanic Americans were more likely to attribute blame to the rape victim than Caucasian Americans (Jimenez & Abreu, 2003; Lee, Pomeroy, Yoo, & Rheinboldt, 2005). Although South Africans or Hispanic Americans can be considered as honor culture groups, unlike our study, these previous studies did not address the influence of honor norms directly.
Furthermore, we examined potential differences between German and British participants. Participants from the two cultures did not differ on perpetrator blaming, crime severity, or rape identification, but British participants blamed the victim more than did German participants. As shown in the mediation results presented in the SM (see Figure S1 and Table S6), gender‐based honor norms, traditional gender role attitudes, and rape myth acceptance were all potential mediators explaining higher levels of victim blaming by British participants.
Regarding our second hypothesis, as expected, in all three cultures participants in the threat condition attributed more blame to the victim, less blame to the perpetrator, and perceived the perpetrator's crime as less severe compared to the absence of threat condition. These results suggest that, just as threat to masculine reputation through female infidelity can lead observers to have more condoning attitudes toward physical abuse against partners (Vandello & Cohen, 2003), such a threat can also translate into more negative evaluations of marital rape victims and exoneration of perpetrators. Furthermore, providing some support for our prediction that reputation threat's effect on rape blame attributions would be especially pronounced in an honor culture (vs. dignity cultures), we found that masculine reputation threat led only Turkish participants (not German and British participants) to perceive the perpetrator's crime as less severe.
A notable finding (presented in the SM) was that even in the absence of reputation threat, Turkish participants perceived the wife's behavior as significantly more threatening to the husband's reputation and masculinity than did German and British participants, and they also blamed the victim more and perpetrator less compared to German and British participants. Together, these findings suggest that not just sexual infidelity, but any behavior that may be against the husband's wishes may be perceived as a threat to husband's masculine reputation and reason for blaming the female victim in a society with strong honor culture traditions such as in Turkey.
With respect to our third hypothesis concerning observer gender differences, we found that men blamed the victim more and the perpetrator less than women across all three cultures. Consistent with previous research (van der Bruggen & Grubb, 2014), men also perceived the perpetrator's crime as less severe and were less likely to identify the incident as rape compared to women. Furthermore, in line with previous studies (e.g., Vandello et al., 2009), we found that these patterns of observer gender differences were overall more pronounced among honor culture participants.
Finally, we explored whether individual‐level endorsement of honor norms predicts rape attributions beyond rape myth acceptance and traditional gender role attitudes in Turkey, Germany, and Britain. Our results showed that honor norms were a significant predictor of rape perceptions and blame attributions in an honor culture (Turkey), whereas it played a rather negligible role in dignity cultures (Germany and Britain). By contrast, rape myth acceptance was the most important predictor of rape perceptions among German and British participants, whereas in the Turkish sample it was only negatively associated with perceived crime severity. Attitudes toward traditional gender roles had only a marginal influence on rape perceptions in all three cultural groups. These results are in line with previous studies showing a positive relationship between honor norms and blaming of rape victims (Canto et al., 2017; Saucier et al., 2015). Yet, the current cross‐cultural study makes a significant contribution to the literature by examining the impact of honor norms on marital rape perceptions beyond rape myth acceptance and traditional gender role attitudes. In addition, because previous research on the impact of rape myth acceptance was conducted mainly with North American and Western European samples, other cultural processes and mechanisms, particularly those related to honor norms, have been largely neglected. Hence, the present findings reveal new insights into rape perceptions across different cultural groups, suggesting that gender‐based honor norms may be more powerful in predicting rape perceptions than rape myths in cultures in which honor represents a core value.
4.1. Limitations, future research, and implications
As any other study, the current study is not without limitations. One limitation is that the threat manipulation failed to produce differences between the high and low threat conditions in the Turkish and German samples. This suggests that sexual infidelity, regardless of whether it is known widely in the community or only by those involved, is generally enough to warrant the same masculine reputation threat perception. Furthermore, we used a single conceptualization of masculine reputation threat ‐ sexual infidelity ‐ given that across cultures, this type of threat is the leading cause of partner rape (see Buss & Duntley, 2011), and it is predominantly used to induce masculine reputation threat in the literature (e.g. Dietrich & Schuett, 2013; Vandello & Cohen, 2003; Vandello et al., 2009). However, masculine reputation can be threatened in non‐sexual ways as well (e.g., verbally derogating one's manhood, romantic rejections). Future research should use different kinds of masculine reputation threats other than sexual infidelity to examine if the findings generalize across other conceptualizations of threat.
It is also important to note that the results obtained from a single honor culture (Turkey) are not generalizable to all other types of collectivistic cultures. Turkey is considered a collectivistic culture with strong honor values, and collectivistic cultures with strong face values (e.g., Japan) may not embody the same beliefs and attitudes toward women and rape as do honor cultures. Future research should aim to replicate these findings with a diverse range of cultures to better understand the unique cultural factors that contribute to tolerant rape perceptions in honor and dignity cultures, and also expand these investigations to other types of collectivistic cultures.
This study focused on highlighting the importance of gendered honor norms as a factor that makes victim blaming and tolerant perceptions of marital rape more likely, especially in an honor (vs. dignity) culture, and showed the unique contribution of gendered honor norms above and beyond traditional gender role attitudes and rape myth acceptance. Although these latter two constructs are key to understanding rape blame attributions, other variables such as just world beliefs and social dominance orientation have also been shown to account for negative rape victim perceptions (Lambert & Raichle, 2000; Sakallı‐Uğurlu, Yalçın, & Glick, 2007). Future research in this area may benefit from examining these ideological variables that seem relevant in both honor and dignity cultures in combination with honor norms, traditional gender role attitudes, and rape myth acceptance.
An additional limitation concerns our samples, comprising of mostly college students which are not representative of their respective nation's population, as college students are more educated and tend to have more liberal attitudes than the average population. Although nonrepresentative samples can yield valid conclusions (Straus, 2009), future studies should seek representative or random samples, and investigate current or prospective criminal justice workers whose attitudes and perceptions of marital rape victims and perpetrators have direct consequences for fair handling of marital rape cases (see Krahé et al., 2007, 2008).
Despite these limitations, our findings have important implications for acts and decisions of any third‐party observer of marital rape, including victims' family and friends, police, judges, prosecutors, and health personnel. Given that rape inflicts profound social and psychological damage, victims' well‐being and recovery as well as fair handling of rape cases can be significantly undermined to the extent that laypersons or legal professionals think masculine reputation threat through female infidelity is a viable reason to blame the victim and minimize perpetrator's liability for partner rape, as our study demonstrates. Our findings highlight the domains of danger most urgently in need of potential intervention and amelioration, especially in cultures where male–female relationships and behavior are organized around honor norms. In addition, our findings demonstrate that the key cultural predictors of blame attribution to marital rape victims and lenient perceptions of perpetrators are not identical in honor and dignity cultures. This suggests that interventions that aim to reduce negative attitudes toward rape victims may be more effective if they focus primarily on addressing honor norms in honor cultures, whereas myths and stereotypes about sexual aggression may be the most important cultural factors that need to be addressed in dignity cultures.
CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.
AUTHOR CONTRIBUTIONS
P. G. and I. S. contributed equally to this manuscript.
Supporting information
ACKNOWLEDGMENTS
We would like to thank Susan Cross, Barbara Krahé, and Ayse Uskul for their helpful comments and suggestions throughout the preparation of this article. Ethics was granted from the University of Kent. During the data collection and analysis phase Pelin Gul was affiliated with the University of Kent, UK, and during the manuscript writing phase she was affiliated with Iowa State University, USA. Currently she is at the University of Twente, the Netherlands. Isabell Schuster was affiliated with the University of Potsdam during data collection and the first manuscript writing phase. Currently she is at the Free University of Berlin, Germany.
Gul P, Schuster I. Judgments of marital rape as a function of honor culture, masculine reputation threat, and observer gender: A cross‐cultural comparison between Turkey, Germany, and the UK. Aggressive Behavior. 2020;46:341–353. 10.1002/ab.21893
REFERENCES
- Abrams, D. , Viki, G. T. , Masser, B. , & Bohner, G. (2003). Perceptions of stranger and acquaintance rape: The role of benevolent and hostile sexism in victim blame and rape proclivity. Journal of Personality and Social Psychology, 84, 111–125. 10.1037/0022-3514.84.1.111 [DOI] [PubMed] [Google Scholar]
- Ben‐David, S. , & Schneider, O. (2005). Rape perceptions, gender role attitudes, and victim‐perpetrator acquaintance. Sex Roles, 53, 385–399. 10.1007/s11199-005-6761-4 [DOI] [Google Scholar]
- Brown, R. P. , Baughman, K. , & Carvallo, M. (2018). Culture, masculine honor, and violence toward women. Personality and Social Psychology Bulletin, 44, 538–549. 10.1177/0146167217744195 [DOI] [PubMed] [Google Scholar]
- Buss, D. M. , & Duntley, J. D. (2011). The evolution of intimate partner violence. Aggression and Violent Behavior, 16, 411–419. 10.1016/j.avb.2011.04.015 [DOI] [Google Scholar]
- Canto, J. M. , Perles, F. , & San Martín, J. (2017). Culture of honour and the blaming of women in cases of rape. Revista de Psicologia Social, 32, 80–107. 10.1080/02134748.2016.1250488 [DOI] [Google Scholar]
- Cohen, D. , Nisbett, R. E. , Bowdle, B. F. , & Schwarz, N. (1996). Insult, aggression, and the Southern culture of honor: An “experimental ethnography”. Journal of Personality and Social Psychology, 70, 945–960. 10.1037/0022-3514.70.5.945 [DOI] [PubMed] [Google Scholar]
- Dietrich, D. M. , & Schuett, J. M. (2013). Culture of honor and attitudes toward intimate partner violence in Latinos. SAGE Open, 3, 1–11. 10.1177/2158244013489685 [DOI] [Google Scholar]
- European Commission . (2016). Gender‐based violence Special Eurobarometer 449 Report. Retrieved from http://ec.europa.eu/commfrontoffice/publicopinion/index.cfm/ResultDoc/download/DocumentKy/75837
- Gerger, H. , Kley, H. , Bohner, G. , & Siebler, F. (2007). The acceptance of modern myths about sexual aggression scale: Development and validation in German and English. Aggressive Behavior, 33, 422–440. 10.1002/ab.20195 [DOI] [PubMed] [Google Scholar]
- Grubb, A. , & Turner, E. (2012). Attribution of blame in rape cases: A review of the impact of rape myth acceptance, gender role conformity and substance use on victim blaming. Aggression and Violent Behavior, 17, 443–452. 10.1016/j.avb.2012.06.002 [DOI] [Google Scholar]
- Guerra, V. M. , Giner‐Sorolla, R. , & Vasiljevic, M. (2013). The importance of honor concerns across eight countries. Group Processes & Intergroup Relations, 16, 298–318. 10.1177/1368430212463451 [DOI] [Google Scholar]
- Heaven, P. C. L. , Connors, J. , & Pretorius, A. (1998). Victim characteristics and attribution of rape blame in Australia and South Africa. Journal of Social Psychology, 138, 131–133. 10.1080/00224549809600363 [DOI] [Google Scholar]
- Jimenez, J. A. , & Abreu, J. M. (2003). Race and sex effects on attitudinal perceptions of acquaintance rape. Journal of Counseling Psychology, 50, 252–256. [Google Scholar]
- Krahé, B. , Temkin, J. , & Bieneck, S. (2007). Schema‐driven information processing in judgements about rape. Applied Cognitive Psychology, 21, 601–619. 10.1002/acp.1297 [DOI] [Google Scholar]
- Krahé, B. , Temkin, J. , Bieneck, S. , & Berger, A. (2008). Prospective lawyers' rape stereotypes and schematic decision making about rape cases. Psychology, Crime & Law, 14, 461–479. 10.1080/10683160801932380 [DOI] [Google Scholar]
- Kulczycki, A. , & Windle, S. (2011). Honor killings in the Middle East and North Africa: A systematic review of the literature. Violence Against Women, 17, 1442–1464. 10.1177/1077801211434127 [DOI] [PubMed] [Google Scholar]
- Lambert, A. J. , & Raichle, K. (2000). The role of political ideology in mediating judgments of blame in rape victims and their assailants: A test of the just world, personal responsibility, and legitimization hypotheses. Personality and Social Psychology Bulletin, 26, 853–863. 10.1177/0146167200269010 [DOI] [Google Scholar]
- Lee, J. , Pomeroy, E. C. , Yoo, S. K. , & Rheinboldt, K. T. (2005). Attitudes toward rape: A comparison between Asian and Caucasian college students. Violence Against Women, 11, 177–196. 10.1177/1077801204271663 [DOI] [PubMed] [Google Scholar]
- Leung, A. K. , & Cohen, D. (2011). Within‐ and between‐culture variation: Individual differences and the cultural logics of honor, face, and dignity cultures. Journal of Personality and Social Psychology, 100, 507–526. 10.1037/a0022151 [DOI] [PubMed] [Google Scholar]
- Nelson, M. C. (1988). Reliability, validity, and cross‐cultural comparisons for the simplified attitudes toward women scale. Sex Roles, 18, 289–296. 10.1007/BF00288291 [DOI] [Google Scholar]
- Newcombe, P. A. , Van Den Eynde, J. , Hafner, D. , & Jolly, L. (2008). Attributions of responsibility for rape: Differences across familiarity of situation, gender, and acceptance of rape myths. Journal of Applied Social Psychology, 38, 1736–1754. 10.1111/j.1559-1816.2008.00367.x [DOI] [Google Scholar]
- Rebeiz, M. J. , & Harb, C. (2010). Perceptions of rape and attitudes toward women in a sample of Lebanese students. Journal of Interpersonal Violence, 25, 735–752. 10.1177/0886260509334410 [DOI] [PubMed] [Google Scholar]
- Sakallı‐Uğurlu, N. , Yalçın, Z. S. , & Glick, P. (2007). Ambivalent sexism, belief in a just world, and empathy as predictors of Turkish Students' attitudes toward rape victims. Sex Roles, 57, 889–895. 10.1007/s11199-007-9313-2 [DOI] [Google Scholar]
- Saucier, D. A. , Strain, M. L. , Hockett, J. M. , & McManus, J. L. (2015). Stereotypic beliefs about masculine honor are associated with perceptions of rape and women who have been raped. Social Psychology, 46, 228–241. 10.1027/1864-9335/a000240 [DOI] [Google Scholar]
- Spence, J. T. , & Helmreich, K. L. (1972). The attitude toward women scale: An objective instrument measure attitude toward the right and role of women in contemporary society. JSAS Catalogue of Selected Document in Psychology, 2, 66–67. [Google Scholar]
- Straus, M. A. (2009). The national context effect: An empirical test of the validity of cross‐national research using unrepresentative samples. Cross‐Cultural Research, 43, 183–205. 10.1177/1069397109335770 [DOI] [Google Scholar]
- Temkin, J. , & Krahé, B. (2008). Sexual assault and the justice gap: A question of attitude. Oxford: Hart. [Google Scholar]
- UNDP . (2019). Human development report 2019. Beyond income, beyond averages, beyond today: Inequalities in human development in the 21st century. Retrieved from http://hdr.undp.org/sites/default/files/hdr2019.pdf
- Uskul, A. K. , & Cross, S. E. (2019). The social and cultural psychology of honour: What have we learned from researching honour in Turkey? European Review of Social Psychology, 30, 39–73. 10.1080/10463283.2018.1542903 [DOI] [Google Scholar]
- Vandello, J. A. , & Cohen, D. (2003). Male honor and female fidelity: Implicit cultural scripts that perpetuate domestic violence. Journal of Personality and Social Psychology, 84, 997–1010. 10.1037/0022-3514.84.5.997 [DOI] [PubMed] [Google Scholar]
- Vandello, J. A. , & Cohen, D. (2008). Culture, gender, and men's intimate partner violence. Social and Personality Psychology Compass, 2, 652–667. 10.1111/j.1751-9004.2008.00080.x [DOI] [Google Scholar]
- Vandello, J. A. , Cohen, D. , Grandon, R. , & Franiuk, R. (2009). Stand by your man: Indirect prescriptions for honorable violence and feminine loyalty in Canada, Chile, and the United States. Journal of Cross‐Cultural Psychology, 40, 81–104. 10.1177/0022022108326194 [DOI] [Google Scholar]
- van der Bruggen, M. , & Grubb, A. (2014). A review of the literature relating to rape victim blaming: An analysis of the impact of observer and victim characteristics on attribution of blame in rape cases. Aggression and Violent Behavior, 19, 523–531. 10.1016/j.avb.2014.07.008 [DOI] [Google Scholar]
- Whatley, M. A. (2005). The effect of participant sex, victim dress, and traditional attitudes on causal judgments for marital rape victims. Journal of Family Violence, 20, 191–200. 10.1007/s10896-005-3655-8 [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.