We appreciate Norenzayan et al.’s nice collection of theory and evidence for processes fostering human cooperation in large-scale groups. However, we are not convinced by the proposed connection between large-scale cooperation and prosocial gods for several reasons. First, the proposed group-functional aspects of prosocial religions or gods seem not to prevent conflicts within such religious groups. Second, general group processes foster prosocial tendencies within groups independent of religion. Third, the proposed surveillance and punishment mechanisms also seem to work without gods. Finally, gods as enhancers of prosocial tendencies would have to be more effective in enhancing prosocial tendencies than alternatives in order to spread via cultural evolution.
In contrast to enhancing prosocial tendencies, a shared common religion does not prevent conflicts between groups defined by language, ethnicity, or nation. Other group identities (e.g., nation, kingdoms, ethnic groups) are sometimes more important for collective action than religious identities. For example, in World War I, Christians and Muslims allied (Germany and the Ottoman Empire) to fight other Christians and Muslims (e.g., French, British, Egypt). Here, an allegiance to the same Big God does not prevent war, and allegiances to different Big Gods are no obstacle for alliances. Moreover, European history is full of wars in which Christians fought against other Christians. Besides being unable to prevent conflicts derived by other identities, prosocial religions even instigate conflicts within the same faith. For example, religious ideas promote conflict between Shiites and Sunnis about who the real Muslim is, and Protestants and Catholics fought hundreds of years about who the right Christian is. Juergensmeyer (Reference Juergensmeyer2008) even suggests that there is an inevitable conflict between orthodox groups and more secular groups within all religions. Hence, prosocial religions do not produce happily united cooperative groups. They fail to prevent conflicts within, and sometimes they even instigate conflicts. The prosocial aspect of Big God religions thus seems too weak to establish and maintain high levels of cooperation within religious groups.
Second, as Norenzayan et al. suggest, Big Gods may serve as templates of identification within a group, but there are many other effective templates. Groups can substitute shared ideologies (e.g., communism), nonomnipotent leaders (e.g., dictators), general ideas (e.g., human rights), a shared feature (e.g., skin color), or even experimentally created categories that an individual identifies with (Tajfel & Turner Reference Tajfel, Turner, Worchel and Austin1979) for Big Gods. Experimental data in social psychology demonstrates that the psychological processes Norenzayan et al. propose effectively attach to any group identity. For example, arbitrary and ad hoc created groups in psychological laboratories trigger the formation of common norms (Sherif Reference Sherif1965), favorable treatment of fellow in-group members over out-group members (Tajfel et al. Reference Tajfel, Billig, Bundy and Flament1971), expectation about reciprocal behavior among in-group members (Yamagishi & Kiyonari Reference Yamagishi and Kiyonari2000), and positive interdependence between group members (Platow et al. Reference Platow, Grace and Smithson2012). Hence, all salient group memberships can promote cooperation between its members, and in-group identification ensures that cooperation is maintained. This is true for small face-to-face groups, as well as for broad categories defining a group (Brown & Brewer Reference Brown, Brewer, Gilbert, Fiske and Lindzey1998). If people are attached to their group, they display prosocial behavior toward their fellow group members. It does not matter if the group is defined by a common religion or by any other group identity.
Third, punitive Big Gods are believed to increase norm compliance by threatening to punish deviants. However, if the belief in a punishing god would be enough to stabilize cooperation, actual punishment of unfair or uncooperative individuals in and outside of the laboratory would not be necessary. Nonetheless, people engage in punishment, even if an investment of their own resources is necessary (Fehr & Gächter Reference Fehr and Gächter2002). In addition to first-person punishment, where the victim punishes, there are other sources of social control, such as third-person punishment and institutionalized punishment (e.g., the police), that account for norm compliance. Further, surveillance does not need to be performed by a higher entity, but cooperation is enhanced by the presence of others (Dawes et al. Reference Dawes, McTavish and Shaklee1977) or even by exposure to a pair of stylized eyes (Haley & Fessler Reference Haley and Fessler2005). In many experimental studies, the mere option of punishment increases cooperative behavior among participants (Fehr & Fischbacher Reference Fehr and Fischbacher2004a). Sometimes, people even prefer groups in which deviants can be punished by fellow group members over groups in which no punishment option is available (Gürerk et al. Reference Gürerk, Irlenbusch and Rockenbach2006). Despite the evidence that punishment is effective in fostering group cooperation, it could increase social distance and conflict within and between social groups (Turner Reference Turner2005). Forcing people to do something enhances resistance and reactance to such pressure (Brehm & Brehm Reference Brehm and Brehm1981). Hence, punishment comes at some considerable costs of social conflict and resistance. In contrast, positive interactions (e.g., incentives, respect, etc.) also enhance the motivation to cooperate (Balliet et al. Reference Balliet, Mulder and Van Lange2011). Positive interactions as a means to convince group members about what is right fosters their intrinsic motivation and thereby enhances group-serving behavior as effectively as punishment and coercion (Turner Reference Turner2005). Although Big Gods may also play a role in these more positive ways to enhance prosocial tendencies, they are not necessarily more effective than other rewarding structures. Hence, there is no need for a Big God or other punishing instances if people conform to what they think is right and not because they want to avoid punishment.
In summary, prosocial religions constituting common group identities seem to be connected to large-scale cooperation through more general social psychological processes. However, in our understanding, alternative identities are just as good and often even better in fulfilling the purpose outlined by Norenzayan et al. For a proper argument in favor of the relation between the two developments, one would have to demonstrate that prosocial religions are more successful in promoting cooperation than other identities. Otherwise, we suggest substituting prosocial religion with any important identity and leaving the spread of Big Gods as a still unresolved puzzle.
You have
Access
We appreciate Norenzayan et al.’s nice collection of theory and evidence for processes fostering human cooperation in large-scale groups. However, we are not convinced by the proposed connection between large-scale cooperation and prosocial gods for several reasons. First, the proposed group-functional aspects of prosocial religions or gods seem not to prevent conflicts within such religious groups. Second, general group processes foster prosocial tendencies within groups independent of religion. Third, the proposed surveillance and punishment mechanisms also seem to work without gods. Finally, gods as enhancers of prosocial tendencies would have to be more effective in enhancing prosocial tendencies than alternatives in order to spread via cultural evolution.
In contrast to enhancing prosocial tendencies, a shared common religion does not prevent conflicts between groups defined by language, ethnicity, or nation. Other group identities (e.g., nation, kingdoms, ethnic groups) are sometimes more important for collective action than religious identities. For example, in World War I, Christians and Muslims allied (Germany and the Ottoman Empire) to fight other Christians and Muslims (e.g., French, British, Egypt). Here, an allegiance to the same Big God does not prevent war, and allegiances to different Big Gods are no obstacle for alliances. Moreover, European history is full of wars in which Christians fought against other Christians. Besides being unable to prevent conflicts derived by other identities, prosocial religions even instigate conflicts within the same faith. For example, religious ideas promote conflict between Shiites and Sunnis about who the real Muslim is, and Protestants and Catholics fought hundreds of years about who the right Christian is. Juergensmeyer (Reference Juergensmeyer2008) even suggests that there is an inevitable conflict between orthodox groups and more secular groups within all religions. Hence, prosocial religions do not produce happily united cooperative groups. They fail to prevent conflicts within, and sometimes they even instigate conflicts. The prosocial aspect of Big God religions thus seems too weak to establish and maintain high levels of cooperation within religious groups.
Second, as Norenzayan et al. suggest, Big Gods may serve as templates of identification within a group, but there are many other effective templates. Groups can substitute shared ideologies (e.g., communism), nonomnipotent leaders (e.g., dictators), general ideas (e.g., human rights), a shared feature (e.g., skin color), or even experimentally created categories that an individual identifies with (Tajfel & Turner Reference Tajfel, Turner, Worchel and Austin1979) for Big Gods. Experimental data in social psychology demonstrates that the psychological processes Norenzayan et al. propose effectively attach to any group identity. For example, arbitrary and ad hoc created groups in psychological laboratories trigger the formation of common norms (Sherif Reference Sherif1965), favorable treatment of fellow in-group members over out-group members (Tajfel et al. Reference Tajfel, Billig, Bundy and Flament1971), expectation about reciprocal behavior among in-group members (Yamagishi & Kiyonari Reference Yamagishi and Kiyonari2000), and positive interdependence between group members (Platow et al. Reference Platow, Grace and Smithson2012). Hence, all salient group memberships can promote cooperation between its members, and in-group identification ensures that cooperation is maintained. This is true for small face-to-face groups, as well as for broad categories defining a group (Brown & Brewer Reference Brown, Brewer, Gilbert, Fiske and Lindzey1998). If people are attached to their group, they display prosocial behavior toward their fellow group members. It does not matter if the group is defined by a common religion or by any other group identity.
Third, punitive Big Gods are believed to increase norm compliance by threatening to punish deviants. However, if the belief in a punishing god would be enough to stabilize cooperation, actual punishment of unfair or uncooperative individuals in and outside of the laboratory would not be necessary. Nonetheless, people engage in punishment, even if an investment of their own resources is necessary (Fehr & Gächter Reference Fehr and Gächter2002). In addition to first-person punishment, where the victim punishes, there are other sources of social control, such as third-person punishment and institutionalized punishment (e.g., the police), that account for norm compliance. Further, surveillance does not need to be performed by a higher entity, but cooperation is enhanced by the presence of others (Dawes et al. Reference Dawes, McTavish and Shaklee1977) or even by exposure to a pair of stylized eyes (Haley & Fessler Reference Haley and Fessler2005). In many experimental studies, the mere option of punishment increases cooperative behavior among participants (Fehr & Fischbacher Reference Fehr and Fischbacher2004a). Sometimes, people even prefer groups in which deviants can be punished by fellow group members over groups in which no punishment option is available (Gürerk et al. Reference Gürerk, Irlenbusch and Rockenbach2006). Despite the evidence that punishment is effective in fostering group cooperation, it could increase social distance and conflict within and between social groups (Turner Reference Turner2005). Forcing people to do something enhances resistance and reactance to such pressure (Brehm & Brehm Reference Brehm and Brehm1981). Hence, punishment comes at some considerable costs of social conflict and resistance. In contrast, positive interactions (e.g., incentives, respect, etc.) also enhance the motivation to cooperate (Balliet et al. Reference Balliet, Mulder and Van Lange2011). Positive interactions as a means to convince group members about what is right fosters their intrinsic motivation and thereby enhances group-serving behavior as effectively as punishment and coercion (Turner Reference Turner2005). Although Big Gods may also play a role in these more positive ways to enhance prosocial tendencies, they are not necessarily more effective than other rewarding structures. Hence, there is no need for a Big God or other punishing instances if people conform to what they think is right and not because they want to avoid punishment.
In summary, prosocial religions constituting common group identities seem to be connected to large-scale cooperation through more general social psychological processes. However, in our understanding, alternative identities are just as good and often even better in fulfilling the purpose outlined by Norenzayan et al. For a proper argument in favor of the relation between the two developments, one would have to demonstrate that prosocial religions are more successful in promoting cooperation than other identities. Otherwise, we suggest substituting prosocial religion with any important identity and leaving the spread of Big Gods as a still unresolved puzzle.