Leibniz’s ‘new science of dynamics’ is typically taken to have been completed in the late monadological metaphysics. On this view, stemming from Martial Gueroult and continuing in the recent interpretations of Robert Adams and Pauline Phemister, Leibniz accomplished his dynamics in his later account of physical forces as merely phenomenal modifications of monadic, metaphysical forces. This paper argues, by contrast, that Leibniz considered the dynamics to be an unfinished project: this is evident in statements from throughout his mature period until (...) his final months. Two possible reasons for the non-completion of the dynamics are discussed. The first is that his encounter with the superior physics of Newton’s Principia caused Leibniz to continually defer his own contribution. I argue, however, in favour of a second reason: that there are philosophical difficulties inherent in the science of dynamics that prevented its completion. These difficulties are found in the relation of dynamics to metaphysics, and in the more fundamental split within dynamics, into the doctrine of primitive and derivative forces. Through three late exchanges with de Volder, Wolff and Des Bosses, I show that the relation of primitive and derivative forces remained an unsolved – although not in principle unsolvable – problem for Leibniz. (shrink)