Abstract This paper re-engages with the science versus religion debate in the light of the last three decades of scholarly progress in the study of religion coming from the cognitive and evolutionary sciences of religion. I begin by asking (1) ‘if it is indeed possible to account for the origins of religious belief, its cultural transmission and evolution, and its maintenance—without any appeal to the real existence of any supernatural deity—then are we in a position to say we have explained it all away?’ If we are, how does this affect central themes in the philosophy of religion about God’s existence? In (2) I provide an outline of the cognitive and evolutionary approach to explaining religion. In section (3) I discuss abductive reasoning (3.1), problems with the accusation of having committed the genetic fallacy when answering (1) in favour of positive atheism (3.2), the need for first-order justifications of religious belief (3.3), and in (3.4) a peculiar divine attribute (i.e. omnipotent potential in a causal chain of events) that appears to immunise religious belief in a particular kind of supernatural deity from having been explained away.