Leibniz et la Méthode de la Science (review)

BOOK REVIEWS 305 they approach the book with the following subtitle in mind: A Not-So-Revisionist Intellectual Biography of His Scientific Career. ENRIQUE CH~VEZ-ARvIZO University of Reading Frangois Duchesneau. Leibniz et la M(thode de la Science. Paris: Presses Universitaires de France, x993. Pp. vi + 413 . Paper, FF 958. The philosophical topic of method has always been central to the enterprise of science, for science has always had the task of mediating among, and integrating, experimental results, mathematical models, and high-level principles that express metaphysical assumptions . During the twentieth century, the topic of method has been central to debates in the philosophy of science. Francois Duchesneau finds current debate stagnating between the poles of critical realism and constructive empiricism, due to the unreflective acceptance of a view of method that is Newtonian on the one hand and Kantian on the other. Invoking Imre Lakatos's notion of a changing and yet rationally coherent series of research programs, Duchesneau suggests that philosophers look again and more deeply at the history of science, to see how methodology appears in sciences that are alive and growing. This kind of detailed study, like that undertaken by Duchesneau himself in his recent La Dynamique de Leibniz (Paris: J. Vrin, 1994), may reveal aspects of scientific methodology that have escaped contemporary debate. As Duchesneau remarks in the introduction to Leibniz et la M~thode de la Science, the choice of Leibnizian science as a case study may be especially fruitful for philosophers. Not only was Leibniz an important physicist who revised Cartesian kinematics into a dynamics worthy of the name, providing a powerful and expressive mathematical language for dynamics through the "characteristic" of differential equations, but he was also a great philosopher who reflected upon his own scientific methodology. Moreover, his method and his reflections upon method diverge in significant ways from the method of his contemporary Newton and the later reflections of Kant upon scientific (primarily Newtonian) method. For Leibniz, method is not an autonomous rational procedure which one could describe in a general epistemological analysis independent of scientific activity; science itself furnishes models and schemata of rational procedure. Nor is it a kind of induction shaped by empirical observation: it depends on a theory of knowledge and metaphysical assumptions about the reality that underlies phenomena. Method is thus both internal and external to science, and indeed these two dimensions of method cannot be disentangled in the way that science constantly surpasses itself as it develops new resources for exact and thorough description, and for maximal systematic unity in the discovery of underlying causes. Thus, Leibniz et la M~thode de la Science should be of great interest to philosophers and historians of science. It will also appeal to those with a serious interest in Leibniz, for it takes up a number of important questions of interpretation that have occupied 300 JOURNAL OF THE HISTORY OF PHILOSOPHY 35;2 APRIL 1997 students of Leibniz's philosophy in the last half-century. Chapter One, "La m~thode d'invention," traces the evolution of Leibniz's early project of compiling an encyclopedia of human knowledge to the more mature project of a general science consisting of an ars judicandi and an ars inveniendi, in which the emphasis is more upon a unified order than upon the diversity of phenomena. In his discussion of Leibniz's general science, Duchesneau is careful to keep the binary opposition arsjudicandi/ars inveniendi distinct from another important methodological distinction, analysis/synthesis, for both "arts" involve analysis as well as synthesis. Given the combinatory nature of knowledge for Leibniz (in the empirical as well as the formal sciences), analysis and synthesis must always be complementary. What we find in empirical science as practiced by Leibniz is the conjunction of a combinatory analysis of qualitative properties and their connections with models based upon a conceptualization of possible formal causes for those qualities: he always supposes a continuity or correspondence between the surface order of phenomena, and the order of (causal) sufficient reasons. Analogy here is a most important instrument, for Leibniz brings arrays of facts into line with abstract models by means of a presumed analogy central to his metaphysics...