In this paper, I argue that existing debates about Kant on the particular laws of nature segments into two kinds of approach: the “derivation” approach, and the “best system” approach. Further, I argue that both approaches share a common assumption, namely, that an account of our knowledge of the laws of nature, or the epistemology of laws, plays a role in Kant’s most general account of what it is to be a law. And I argue that this assumption is false. That makes way for a new approach — which I call the “necessitation” approach, and defend here.