I Cartesian Science

Did Descartes Teach a ‘Philosophy of Science’ or Implement ‘Strategies of Natural Philosophical Explanation’?

John Schuster

Introduction: The ‘Clarkean’ Consensus about Cartesian Physical Explanation

By the late 1970s, anyone examining the state of Descartes studies could have noticed surprising new findings about the structure of explanation in Descartes’ scientific and natural philosophical works. Over the previous decade, Gerd Buchdahl, A. I Sabra, Laurens Laudan, and Desmond Clarke had contributed to the fall of the scholars’ myth that Rene Descartes both intended and sometimes practised a top-down deductivist physics—that he somehow drew out from metaphysical truths not only the principles of his physical explanations but also actual empirical details. These scholars collectively established what Descartes actually held concerning the status of the corpuscular-mechanical explanatory models deployed in his natural philosophy. In this chapter, I shall term this the new ‘consensus’ about these matters. It has held at the highest levels of Cartesian scholarship for two generations. While not questioning this consensus as to fundamentals, this chapter articulates the consensus in new directions in order to address hitherto unexamined questions about the work of the mature Descartes qua systematic, corpuscular-mechanical philosopher of nature in Le Monde and in the Principia philosophiae.1

Clarke et al. undermined Descartes’ occasional, overblown claims to have been able to deduce—on some mathematical ideal of ‘demonstration’—his system of natural [1]

philosophy from absolutely certain metaphysical principles. This folklore had arisen from the strictly deductivist tone of Descartes’ abortive method and from some of his more offhand, rhetorically motivated, statements about the issue—statements that belied his actual practices in ‘doing’ natural philosophy. In his mature work, Descartes increasingly came to see that neither the details of particular explanatory models, nor the facts to be explained, could be deduced from metaphysics. Rather, he held that we may know with certainty, from metaphysical deduction, that the essence of matter is extension, but we cannot deduce from this truth more detailed explanatory models (concerning corpuscular sizes, shapes, arrangements, and motions) that can explain various phenomena. The best one can say is that such models should not contradict metaphysically derived certainties. Hence corpuscular-mechanical explanatory models have a necessarily hypothetical character—with the caveat that they must arguably not be inconsistent with metaphysical truths. Available evidence, and in particular the facts to be explained, also bear on the formulation of such detailed explanatory models and in the assessment of their ‘goodness’ in regard to explanatory power and scope of application.[2] [3]

  • [1] On Le Monde and the Principia philosophiae as systems of natural philosophy, see John Schuster,Descartes-Agonistes: Physico-Mathematics, Method and Corpuscular-Mechanism, 1619-33 (Dordrecht,2013), chs 10, 11, 12.
  • [2] As early as 1980,1 tried to characterize this consensus, which I believed was crucial to any future workon Cartesian natural philosophy from a steadfastly historical and developmental perspective. John Schuster, ‘Descartes’ mathesis universalis: 1618-1628’, in Stephen Gaukroger (ed.), Descartes: Philosophy,Mathematics and Physics (Brighton, 1980), 41-96, at p.75. I argued that this new consensus view ofCartesian explanation applied to Descartes’ career from the construction of Le Monde, his first system ofcorpuscular-mechanical natural philosophy, after the collapse in 1628-9 of his dream of method with thefailure of the later portions of the Regulae.
  • [3] Gerd Buchdahl, Metaphysics and the Philosophy of Science. The Classical Origins: Descartes to Kant(Cambridge, Mass., 1969).
< Prev   CONTENTS   Source   Next >