The strategic dimension of pro-social behavior 1 Running head: THE STRATEGIC DIMENSION OF PRO-SOCIAL BEHAVIOR Helping to improve the group stereotype: On the strategic dimension of pro-social behavior

In three studies we consider a basis for inter-group helping. Specifically we show that group members may help others in order to disconfirm a stereotype of their own group as mean. Study one shows that Scots believe they are seen as mean by the English, resent this stereotype, are motivated to refute it, and believe out-group helping is a particularly effective way of doing so. Study two shows that increasing the salience of the English stereotype of the Scottish as mean leads Scots to accentuate the extent to which Scots are depicted as generous. Study three shows that increasing the salience of the stereotype of the Scots as mean results in an increase in the help volunteered to out-group members. These results highlight how strategic concerns may result in out-group helping. In turn, they underscore the point that helping others may be a means to advance a group’s interest. The strategic dimension of pro-social behavior 3 Helping to improve the group stereotype: On the strategic dimension of pro-social behavior “Nothing in human nature is so God-like as the disposition to do good to our fellow creatures”. In these words, the novelist Samuel Richardson (writing in 1742) sums up a common assumption about helping: pro-social behavior reflects an underlying prosocial impulse. That is, we help out of a desire to do good to others and we refrain from helping when dispositions or circumstances dull that desire. However, a wealth of anthropological analyses of gift exchange (e.g., Malinowski, 1922; Mauss 1907/1957) show the giving of goods or services can reflect other motivations. For example, Mauss describes customs of giving that are less to do with helping others and more to do with demonstrating one clan’s superiority over others in a local group hierarchy. Such analyses alert us to the point that acts of giving may be inter-group in nature and need to be analysed through reference to the inter-group relations within which they take place. This logic is illustrated in recent social psychological research by Nadler and Halabi (Nadler, 2002; Nadler & Halabi, 2006). They show that members of a powerful group may provide help to members of powerless groups as a means of maintaining the dependency of their subordinates. Equally, members of powerless groups may reject help from powerful groups especially when they are challenging inter-group inequalities. The wider implication is that helping can be grounded less in a desire to alleviate the plight of the recipient than in an attempt to improve the lot of the donor. It does not always reveal a ‘God-like’ disposition. The strategic dimension of pro-social behavior 4 This paper explores further the group interests that inter-group helping may service. We seek to show that helping others – especially out-group others – may be increased where group members wish to ameliorate the perceptions held about one’s group by others. Specifically, we explore whether telling Scots that the English consider them to be mean results in increased helping towards representatives of a third-party national out-group (the Welsh). Before explaining the context to our studies and our hypotheses, we outline the grounds for studying helping as a group phenomenon and then consider the strategic side of group behavior. Helping as a group process From its inception, helping research addressed collective phenomena. Much attention focused on the idea that the larger the group, the greater the diffusion of responsibility such that each individual feels less obligation to act (Latané & Nida, 1981). Yet, from early on, it was clear that the presence of fellow bystanders can, under certain conditions, increase intervention, especially when other bystanders help (Bryan & Test, 1967). What seems critical is the extent to which people see themselves as belonging to a common social group (Darley, Teger and Lewis, 1973; Horowitz, 1971; Rutkowski, Gruder & Romer, 1983). Where they do, they will be influenced by the actions of their fellows, whether in the direction of intervention or non-intervention. Where they don’t, the actions of one bystander will have little impact upon that of others (Levine, Cassidy, Brazier & Reicher, 2002). In short, what matters is not so much whether bystanders constitute a physical group but rather whether they constitute a psychological group (Turner, Hogg, Oakes, Reicher & Wetherell, 1987), and, if they are a psychological The strategic dimension of pro-social behavior 5 group, whether their norms support or oppose intervention (Horowitz, 1971; Rutkowski et al., 1983). A similar argument can be made concerning the relationship between bystanders and victims. People are more likely to feel concern for those seen as belonging to the same group as themselves and hence are more likely to help them (Dovidio, Gaertner, Validizic, Matoka, Johnson & Frazier, 1997; Levine, et al., 2002; Sturmer, Snyder & Omoto, 2005). What is more, when the boundaries of in-group inclusion are drawn more broadly we become more inclusive in terms of who we will help (Levine, Prosser, Evans & Reicher, 2005). Work on the psychology of volunteerism shows the importance of common group membership is not limited to face-to face-helping but extends to broader forms of social solidarity. The more that people identify with a particular community the more they are likely to volunteer for organizations which provide support for that community (Simon, Sturmer and Steffens, 2000). Such collective action may have greater impact than individual interventions in cases of emergency. For example, whereas acts of individual bravery undoubtedly saved individual Jews from the Nazi holocaust the large-scale collective mobilizations in countries such as Bulgaria saved many more (Genov & Baeva, 2003). Analysis of the key documents addressed to the Bulgarian population shows that both definitions of shared group membership and in-group norms were important in the rescue of Bulgarian Jews (Reicher, Cassidy, Wolpert, Hopkins & Levine, 2006; Todorov, 2001). Jewish people were consistently construed as a Bulgarian (in-group) minority rather than a religious or ethnic out-group and Bulgarian identity was associated with norms opposing oppression. The strategic dimension of pro-social behavior 6 There is therefore archival, survey and experimental research demonstrating the theoretical and practical importance of group processes for helping. This evidence shows how the cognitive salience of category membership and category content affects the support we give to others. The strategic side of group helping Reicher et al. (2006) identify one further argument used to promote the rescue of Bulgarian Jews: if Bulgaria wished to be regarded as a ‘civilized nation’ it could not be complicit in mass slaughter. This argument has less to do with defining the boundaries to in-group inclusion and group norms, and more to do with advancing group interests. It therefore fits with a growing literature stressing the need to consider the strategic dimension to group action (see Klein et al, in press; Reicher, Spears and Postmes, 1995). Of particular interest is the idea that actors may seek to induce others to see their group as they themselves wish it to be seen. For example, Klein & Azzi (2001) show that group members, when confronted with an out-group depiction of their in-group will confirm positive aspects of this stereotype yet disconfirm the negative aspects. Our work develops this logic. It explores how group helping behaviors may be a way of responding to specific types of perceived negative out-group stereotypes (in saying this, we do not dispute that there may be other ways of doing so, but our concern here is to do with meta-stereotype disconfirmation as a basis for helping and not with the bases of meta-stereotype disconfirmation per se). While helping has positive moral connotations which may serve to ameliorate negative out-group stereotypes in general, we expect that the meanings associated with such actions will be particularly effective in The strategic dimension of pro-social behavior 7 disconfirming stereotypes that portray the in-group as having specific anti-social characteristics: being selfish or egocentric. This is not to say that there may not be other ways of disconfirming such stereotypes. Our point is rather that one reason that people help is to demonstrate their generosity to doubting others. In turn this implies that people are likely to help more when confronted with such out-group doubts. Furthermore, we suggest that the motive to disconfirm ‘mean’ stereotypes of the in-group impacts specifically upon out-group helping – something of particular note since the group processes discussed above impact uniquely on help accorded to other in-group members. This difference is because help, as we conceptualise it here, is an act of communication which means that we must consider the communicational value of different forms of helping. Thus, helping “one’s own” is weakly diagnostic of a group’s qualities: such help is to be expected from any group and can easily be construed as selfish and sectarian. By contrast, helping out-group members is more noteworthy and diagnostic. This is readily apparent in the biblical parable of ‘The Good Samaritan’ in which a Samaritan helps a Levite at a time when Samaritans were regarded by Jews as a rival ethnic out-group. That the act of helping depicted here is inter-group is crucial to the parable’s power. Such helping is unexpected and cannot be dismissed as an example of group members simply looking after ‘their own’. Drawing these points together, we propose that out-group helping may be used to challenge negative out-group stereotypes in general and “mean” hetero-stereotypes in particular. We predict that, where group members are made aware that a significant outgroup stereotypes them as mean, they will emphasize their helpfulness, specifically through increasing levels of helping to a third party out-group. However, not all group The strategic dimension of pro-social behavior 8 members will be equally motivated to act in this way. How people respond to negative characterizations of their group (and other threats to collective identity) depends upon their degree of group identification, with high identifiers being more motivated to act in ways that defend/promote the in-group (Branscome, Ellemers, Spears & Doosje, 1999). It therefore follows that our predicted effect should be greater amongst high identifiers. The present research Our research explores how the helping behavior of Scots is affected when their reputation for being mean is made salient. The stereotype of the Scots as mean is widespread and features as an illustrative example in Allport’s 1954 classic “The Nature of Prejudice”. ‘A Scotsman who is penurious’, he writes, ‘delights us because he vindicates our prejudgement’ (p. 22). This stereotype continues to have popular currency and continues to annoy Scottish people. This is well illustrated in the way Scots react to the jokes told about their meanness. For instance one website reports a number of such jokes and then retorts that Scots give more to registered charities per head of population than any other part of the UK . Throughout our research we also sought to capitalize on the inter-group relationship between the Scots and the English. Scotland and England are united in a single state, yet although Scots may define themselves as British, a Scottish identification remains potent and is bound up with inter-group comparisons with the (historically more powerful) English rival (Hopkins & Moore, 2001). Scots often complain that the English are arrogant and ignorant about their northern neighbors and resent the way they are The strategic dimension of pro-social behavior 9 characterized (or more accurately, think they are characterized) by the English (Reicher & Hopkins, 2001). In our research we encouraged our participants, who were young Scots, to reflect upon the Scottish-English relationship and the degree to which they are seen by the English as being mean. In the first Study, we investigated whether Scots believed themselves to be seen as mean by the English (that is we investigated what may be termed their meta-stereotype, see Vorauer, Main and O’Connell, 1998). We also investigated participants’ beliefs about how they could refute this image. In the second Study, we manipulated the salience of an English stereotype of the Scots as mean and examined how this affected participants’ depiction of the Scottish in-group. In the third, and main Study, we included a behavioral measure of helping and explored how in-group and out-group helping behavior was affected by manipulating the salience of an English stereotype of the Scots as mean. Our hypotheses (and the studies that address them) are as follows: H1: Young Scots believe that they are viewed as mean by the English, believe that this stereotype not to be justified, are motivated to challenge the stereotype and believe that out-group helping is particularly effective in doing so (Study 1). H2: The more the mean meta-stereotype is salient, the more Scots will seek to present the Scots as generous (Study 2) and (H2a), in particular, the more they will behave helpfully to a third-party out-group (Study 3).