© Copyright by T. William Altermatt, 2001 CHIVALRY: THE RELATION BETWEEN A CULTURAL SCRIPT AND STEREOTYPES ABOUT WOMEN BY T. WILLIAM ALTERMATT B.A., Millersville University of Pennsylvania, 1995 A.M., University of Illinois at Urbana-Champaign, 1998 THESIS Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Psychology in the Graduate College of the University of Illinois at Urbana-Champaign, 2001 Urbana, Illinois ABSTRACT Chivalry is a cultural script prescribing the preferential treatment of women by men in the contexts of protection and provision. Although it is often regarded as a polite and prosocial script for male-female relations, chivalry is sometimes suspected of undermining efforts to promote equal status for women because it suggests that women require more help than men do. In this report, I examined the relation between chivalry and two beliefs about women: the belief that women are more virtuous than men and the belief that women are less agentic than men. Endorsement of the chivalry script was found to be significantly positively correlated with the belief that women are more virtuous than men and the belief that women are less agentic than men. In addition, chivalrous men tended to show preferential treatment only to women who appeared to be high in virtue and low in agency. Finally, participants who merely observed a woman receiving chivalrous treatment perceived her to be significantly less independent than participants who observed the same woman when she did not receive chivalrous treatment. iii ACKNOWLEDGMENTS I am very grateful to my dissertation committee chair, Joe McGrath, and research advisor, Dov Cohen, without whose attention, encouragement, knowledge, and guidance this dissertation would not have been possible. I would also like to thank the other members of my dissertation committee – Ulf Bockenholt, Incheol Choi, Eva Pomerantz, and Janice Juraska – for their insights and suggestions that helped to make this dissertation better than it otherwise would have been. Finally, I would like to acknowledge the tireless support and inspiration of my best friend, editor, research consultant, and wife, Ellen Altermatt. iv TABLE OF CONTENTS CHAPTER 1: INTRODUCTION...................................................................................................1 CHAPTER 2: A REVIEW OF THE LITERATURE ON CHIVALRY AND BELIEFS ABOUT WOMEN....................................................................................................7 CHAPTER 3: RESEARCH PRELIMINARY TO THE DISSERTATION EXPERIMENT........................................................................................................................18 CHAPTER 4: OBSERVING CHIVALRY...................................................................................79 CHAPTER 5: INTEGRATION..................................................................................................112 TABLES......................................................................................................................................119 FIGURES.....................................................................................................................................155 APPENDIX A: A HISTORY OF CHIVALRY...........................................................................158 APPENDIX B: VIGNETTES AND TRAITS FROM STUDY 2................................................163 REFERENCES............................................................................................................................168 CURRICULUM VITAE..............................................................................................................176 v CHAPTER 1: INTRODUCTION Chivalry is a concept that has undergone considerable change in meaning since its origin in the Middle Ages. Derived from the French cheval (“horse”), chivalry initially referred to a group of warriors on horseback, what we now call the cavalry (a word with similar etymology). Over the centuries, chivalry began to connote a code of appropriate behavior expected of men. In its modern usage, chivalry typically refers to a pattern of behavior characterized by gallantry toward women. In this report, I treat chivalry as a “cultural script” – a set of expectations for behavior that is shared by members of a culture (Triandis, Marin, Lisansky, and Betancourt, 1984). Cultural scripts describe the sequence of activities that are expected under certain social conditions (Harris, Schoen, & Hensley, 1992). Like the script for a play, the chivalry script describes a set of behaviors expected from particular actors toward particular targets in a particular context. In the chivalry script, preferential treatment is expected from men toward women in the context of protection and provision. The chivalry script leads men to behave toward women in a way that is different from the way they would treat other men – a pattern of behavior that fits the definition of discrimination. Unlike most forms of discrimination, chivalry is associated with benevolent rather than hostile intentions toward the target group, women. Peter Glick and Susan Fiske have explored this unusual situation in their theory of Ambivalent Sexism (Glick & Fiske, 1996). According to their theory, there are two forms of sexism: hostile and benevolent. Whereas hostile sexism involves negative beliefs about women and an adversarial approach to male-female relations, benevolent sexism involves “a set of interrelated attitudes toward women that are sexist in terms of viewing women stereotypically and in restricted roles but that are subjectively positive in feeling tone (for the perceiver) and also tend to elicit behaviors typically categorized as prosocial 1 (e.g., helping) or intimacy-seeking (e.g., self-disclosure)” (Glick & Fiske, 1996, p. 491). Glick and Fiske (1996) found that the correlation between measures of benevolent and hostile sexism is approximately 0.43 among male student samples and slightly higher among female student samples (Glick & Fiske, 1996, p. 502), suggesting that the positive and negative attitudes about women assessed by their scales tend to co-occur. Based on the correlation between the two types of sexism, Glick and Fiske (1996) propose that they should be considered simultaneously, as ambivalent sexism, in which responses to the target group (women) are a mix of positive and negative attitudes. The distinction between chivalry and ambivalent sexism is between a set of attitudes and a cultural script. Whereas ambivalent sexism is a set of attitudes about women, chivalry is an organized knowledge structure containing information about behaviors – when and to whom they should occur, and what activities are involved (e.g., door-opening, giving up one’s seat, etc.). Glick and Fiske (1996) describe the relation between chivalry and ambivalent sexism as the relation between a set of cultural conditions and the attitudes that emerge from those conditions. In their discussion of how levels of hostile and benevolent sexism vary across cultures, the authors use the example of a chivalrous society to illustrate how the ratio of benevolent to hostile sexism can be very high (Glick & Fiske, 1996, p. 492). To summarize, chivalry is a set of expectations for behavior that is shared by members of a culture. It is characterized by the themes of men protecting and providing for women, and it is related to hostile and benevolent sexism in that it describes one set of cultural conditions likely to produce a high ratio of benevolent to hostile sexism.1 1 In contrast to chivalry, some cultures exhibit a script of institutionalized misogyny. Women of the Yanamano Indian culture, who live on the border between Brazil and Venezuela, are described as “among the most brutalized in the world” (Triandis, 1994, p. 129). In their culture, wives are obtained through raids, and it is prestigious for a man to beat and even stab his wife in public (Triandis, 1994). Thus, chivalry is only one of many possible cultural scripts prescribing the proper relation between men and women. 2 In this report, I examine the hypothesis that chivalry is related to two beliefs about women: the belief that women are more virtuous than men, and the belief that women are less agentic than men. In Chapter 2, I provide a theoretical context for this hypothesis by reviewing some of the history, theory, and research on chivalry and beliefs about women. Chapters 3 and 4 describe three empirical approaches to testing the central hypothesis. The first two approaches involved four studies conducted in collaboration with Dov Cohen and Tina Johnson, and are discussed in detail in Chapter 3. The third approach involved an experiment conducted independently by me, and comprises the independent research component of this dissertation. It is discussed in Chapter 4. The three approaches are introduced below. The goal of the first approach was to develop a measure of the strength of endorsement of the chivalry script and to examine whether the endorsement of chivalry was correlated with two beliefs about women: that they are more virtuous than men and that they are less agentic than men. This approach tells us nothing about the causal direction of the relation between chivalry and these beliefs. It cannot tell us, for example, whether chivalry leads to certain beliefs about women, whether certain beliefs about women lead to chivalry, or whether some third factor (such as a belief in traditional sex roles) leads to both. What it does show is that individuals who endorse chivalry also hold beliefs about women’s virtue and agency relative to men. This would suggest that chivalry is not only consistent with these two beliefs, but that the three sets of beliefs in some sense “go together.” Another way to test whether chivalry is related to beliefs about women is to observe what happens when these beliefs are disconfirmed. If a chivalrous man believes that women are more virtuous and less agentic than men, what happens when he encounters a woman who is very high in agency or very low in virtue? Does he treat her differently than a man who does not endorse 3 chivalry? The second approach was to examine whether the endorsement of chivalry is related to reactions to women who violate or conform to expectations of high virtue and low agency. In this approach, virtue and agency are treated as experimental variables, manipulated to be displayed at either high or low levels by a stimulus person. According to this approach, we would expect a man who endorses chivalry to give preferential treatment to a woman only when she is high in virtue and low in agency. When she appears to be low in virtue or high in agency, chivalrous individuals are expected to respond with indifference or disapproval rather than preferential treatment. This approach contributes two pieces of information for understanding chivalry. First, if individuals high in chivalry are more likely than individuals low in chivalry to distinguish among women based on how well they conform to expectations of virtue and agency, then there is more support for the idea that chivalry is related to beliefs about women’s virtue and lack of agency. Specifically, chivalry could be said to be contingently related to beliefs about women; it is more likely to occur when the target is low in agency and high in virtue. Second, the finding that chivalry is not indiscriminate in its provision of preferential treatment is itself noteworthy. It calls into question the image of chivalry as an altruistic pattern of behavior by showing that certain conditions must be met in order for a woman to receive special treatment. To the extent that the receipt of chivalrous behavior is regarded as rewarding by women, the existence of these conditions has implications for how chivalry might actively promote a decrease in agentic behavior and an increase in virtuous behavior among women. If chivalrous treatment is rewarding and received only when women behave in a virtuous and non- agentic manner, then the likelihood of virtuous and non-agentic future behavior will increase. Thus, if chivalry is related to a discrimination among women based on their apparent virtue and 4 agency and the receipt of chivalry is regarded as rewarding, then it could be said to promote a lack of agency and a premium on virtue. The third approach deals with the consequences of chivalrous behavior. Imagine two scenarios in which a woman is riding on a crowded bus and does not have a seat. In one scenario, a male passenger offers her his seat and she accepts. In the other, no offer is made. Would observers of these two scenarios form different impressions of the actors? Would the female who accepted the offer, for example, appear more dependent than the female who never received an offer? This approach offers two features not present in the first two approaches. First, it treats chivalry as an experimental variable rather than an individual difference variable, enabling a test of the causal relation between chivalrous behavior and third-party impressions. Second, it takes the perspective of an observer of chivalrous behavior. This perspective is important because the observation of public behavior is one method of receiving information about social roles. In her social roles theory (1987), Alice Eagly proposes that beliefs about sex differences are influenced by observations of the roles typically occupied by men and women. Observations of chivalrous behavior may contribute to the belief that women are less agentic than men are because in the chivalry script, men play the role of provider while women play the role of recipient. In summary, the first approach examines whether two sets of cognitions – the chivalry script and beliefs about women’s virtue and agency – tend to co-occur. The second approach tests whether chivalrous men distinguish among women based on virtue and agency. The third approach investigates the effect of chivalrous behavior on the impressions that are formed of actors in a dyadic interaction. The methods and results of these three approaches are discussed in detail in Chapters 3 and 4. In the next chapter, I review the empirical and historical literature 5