People discussing science and religion usually frame their conversations in terms of essentialist assumptions about science, assumptions requiring the existence (but not the specification) of criteria according to which science can be distinguished from other forms of inquiry. However, criteria functioning at a level of generality appropriate to such discussions may not exist at all. Essentialist assumptions may be avoided if science is understood within a broader context of human practices. In a philosophy of practices, to label a practice as “scientific” is to make a practically motivated provision for a way of speaking. Charles Taylor and Joseph Rouse have produced complementary philosophies of practice that promote this kind of understanding. In this essay I review the work of Taylor and Rouse, identify apparent residues of essentialism that each seems to harbor, and offer a resolution to some of their disagreements. I also criticize a form of essentialism commonly employed in Christian circles and outline an anti‐essentialist view of science that may be helpful in science‐and‐religion discussions.