Darwinian evolution has to provide an explanation for cooperative beha
viour. Theories of cooperation are based on kin selection (dependent o
n genetic relatedness)(1,2), group selection(3-5) and reciprocal altru
ism(6-9). The idea of reciprocal altruism usually involves direct reci
procity: repeated encounters between the same individuals allow for th
e return of an altruistic act by the recipient(10-16). Here we present
a new theoretical framework, which is based on indirect reciprocity(1
7) and does not require the same two individuals ever to meet again, i
ndividual selection can nevertheless favour cooperative strategies dir
ected towards recipients that have helped others in the past. Cooperat
ion pays because it confers the image of a valuable community member t
o the cooperating individual. We present computer simulations and anal
ytic models that specify the conditions required for evolutionary stab
ility(18) of indirect reciprocity. We show that the probability of kno
wing the 'image' of the recipient must exceed the cost-to-benefit rati
o of the altruistic act. We propose that the emergence of indirect rec
iprocity was a decisive step for the evolution of human societies.