We consider three hypotheses about relatedness and well-being including the hive hypothesis, which says people need to lose themselves occasionally by becoming part of an emergent social organism in order to reach the highest levels of human flourishing. We discuss recent evolutionary thinking about multilevel selection, which offers a distal reason why the hive hypothesis might be true. We next consider psychological phenomena such as the joy of synchronized movement and the ecstatic joy of self-loss, which might be proximal mechanisms underlying the extraordinary pleasures people get from hive-type activities. We suggest that if the hive hypothesis turns out to be true, it has implications for public policy. We suggest that the hive hypothesis points to new ways to increase social capital and encourages a new focus on happy groups as being more than collections of happy individuals.Question: What is the difference between society and the sun? Answer: If you really want to, you can stare directly at the sun. But to see society, you must use special glasses. Social scientists generally use one of two kinds: glasses that reveal atoms (individuals) and glasses that reveal networks (groups of connected individuals). Psychologists and some economists seem to prefer looking at individuals. We model people as agents who have beliefs and desires and who act to maximize the satisfaction of their desires given their beliefs. We revel in demonstrations that people sometimes do not maximize, and we advance our sciences by bringing in unconscious desires, discounting curves, and errors in the reasoning processes by which people make inferences from their beliefs.When we put on the atomizing glasses, a research agenda and a humanitarian project appear before us: we must fully understand the JONATHAN HAIDT is Associate Professor and J. PATRICK SEDER and SELIN KESEBIR are graduate students in the Department of Psychology, University of Virginia.