A test of functional theories of religion in a non-western sample

Abstract

Scholars of religion have long pondered why humans believe in supernatural agents and participate in religious rituals. Some propose that religious features evolved as adaptations for a range of functions, such as alleviating insecurity, structuring social support and cooperating in large groups. Most work in this area has been conducted in Western, Educated, Industrialised, Rich and Democratic (WEIRD) countries, in times of relative resource abundance. This thesis uses religious survey and behavioural economic methods to investigate variation in religiosity and cooperation and test a range of functional explanations for religion in non-WEIRD societies, and in times of relative abundance and crisis. Fieldwork in 2014 in Christian and indigenous Kastom religious groups on the island of Tanna, Vanuatu, revealed that belief in an increasingly punitive deity predicted higher monetary offers to outgroup members but not coreligionists. Further fieldwork in 2015 allowed the same measures to be collected following the devastation of Cyclone Pam. Comparison of pre-vs post-cyclone data showed a general decrease in prosociality and more parochial giving to religious ingroup. Post-cyclone giving depended on the level and nature of affectedness; property damage predicted reduced prosociality and parochial giving, whereas exposure to others in distress predicted higher offers to coreligionists and outgroups. Cyclone experience did not predict changes in post-cyclone religiosity and religiosity did not buffer against perceived food insecurity. However, greater personal commitment to one's moralistic god predicted giving less to outgroup members and more parochial giving after the cyclone. Collaboration with the Cultural Evolution of Religion Consortium allowed investigation into functional explanations for putative gender differences in religiosity across a global sample of 14 societies. Women generally showed greater religiosity towards moralistic (but not local) gods. However, the clearest mediator of this gender gap was formal education, consistent with a general process of secularization rather than proposed functional accounts. Overall, the research presented in this thesis suggests individual experiences during a resource shock calibrate prosociality towards religious ingroups and outgroups. Individual religiosity predicted cooperation beyond the religious ingroup in times of plenty and more parochial giving in times of need. Less support was found for religion's function as a coping mechanism, be it in response to resource shocks or as an explanation of gender differences in religiosity.