On its face, Leibniz's argument for primitive concepts seems to imply that unless we can analyze non-primitive concepts into their primitive constituents, we cannot grasp them. This implication, together with Leibniz's belief that we do conceive of some non-primitive concepts, entails that we can analyze some non-primitive concepts into their primitive components. However, Leibniz claims elsewhere that we are incapable of doing this. To resolve this inconsistency, I argue that, for Leibniz, grasping a concept is not an all-or-nothing affair; instead we possess concepts in degrees. When reinterpreted in line with this, Leibniz's argument can avoid the inconsistency.