THE REJECTION OF HYPOTHESIS

In the fourteenth Lettre philosophique, Voltaire capped his attack on Descartes's philosophy with the withering charge that it was nothing more than a "roman ingénieux."24 Voltaire's famous rejection of Descartes in favor of Newton and Locke involved more than a preference for a physics of attraction over one of vortices or for a psychology of the tabula rasa over one of innate ideas. What Voltaire admired about the revolutionary English thinkers, at least as much as their results, was their intellectual method—or what he took to be their method. Descartes uncovered the errors of antiquity, we are told in the thirteenth Lettre philosophique, only to replace them with errors of his own invention. For, like many other thinkers, Descartes was blinded by the esprit systématique.25

From the point of view of empiricist philosophy, the esprit systématique —or the esprit de système as it came to be called—is generally characterized as a taste for rationalistic metaphysics, with its attendant habit of intricate reasoning from dubious first principles.26 But in Voltaire's mind, and in the minds of many of his contemporaries, these faults were intimately connected to the fault of hypothesizing.27 Hypothesizing and speculative reasoning in fact were often

seen not as two different sorts of deviations from responsible method but as one and the same. To hypothesize was precisely to engage in speculative reasoning, to fall into the esprit de système. This general conception of improper intellectual method is clearly reflected in Johnson's definition of hypothesis: "A supposition; a system formed upon some principle not proved."28

From the perspective of the late-twentieth-century philosophy of science, this conflation of hypothesis and speculative metaphysics seems odd. For hypothesis formation has come to be seen as the fundamental imaginative component of scientific discovery. Nor was the eighteenth century unanimous on the issue. In 1765, for example, we find the author of the Encyclopédie article "Hy-pothèse" insisting on the indispensability of hypotheses:

Les hypotheses doivent donc trouver place dans les sciences, puisqu'elles sont propres á faire dècouvrir la vérité et à nous donner de nouvelles vues. . . .
Il y a deux excès à èviter au sujet des hypotheses, celui de les estimer trop, et celui de les proscrire entièrement. Descartes, qui avait établi une bonne pattie de sa philosophie sur des hypothèses , mit tout le monde savant dans le goût de ces hypothèses, et l'on ne fut pas longtemps sans tomber dans celui des fictions. Newton et surtout ses disciples, se sont jettés dans l'extrémité contraire.29

Hypotheses therefore must play a role in the sciences since they allow us to discover the truth and afford us new perspectives. . . .
With regard to hypotheses, there are two extremes to be avoided, that of valuing them too much and that of banishing them altogether. Descartes, who based a good portion of his philosophy upon hypotheses, gave the entire learned world a taste for these hypotheses, which gave way before long to a taste for fictions. Newton, and especially his disciples, have fallen into the opposite extreme.

For Voltaire, however, whose admiration for Newton was much less qualified than that of the anonymous encyclopedist, hypothesizing was not part of responsible scientific method; it was its utter antithesis. Francis Bacon's Novum Organum had proposed an inductive method that would move gradually, step by step, from the particulars of experience to more and more general explanatory axioms. One would diligently begin with observation and experimentation, and the general explanations would eventually emerge as the reward for diligence. This careful procedure, based on sense

perception and the resistance of any temptation to hypothesize, was seen to have borne its greatest fruit in Newton's optics and celestial mechanics. For Newton was an avowed Baconian, who had explicitly declared "hypotheses non fingo." To hypothesize was to ignore the lessons to be learned from observing reality, and to presume to invent one's own reality.

Since this pure-induction model of scientific method has come to be seen as a distortion of what really goes on in scientific discovery,30 contemporary scholars have set about to argue that Bacon wasn't really an inductivist31 and that when Newton seemed to be rejecting the feigning of hypotheses, he must have really meant something else.32 But in the case of Voltaire, at least, any such scholarly revisionism would seem to be doomed to failure. For whatever Newton really meant, Voltaire seems to have taken him literally and to have seen his superiority over Descartes to reside precisely in his refusal to "feign hypotheses." Writing against the Cartesian theory of tourbillons, in his Elémens de la philosophie de Newton, Voltaire asserts:

Il n'y a pas la moindre experience, pas la moindre analogie dans les choses que nous connaissons un peu, qui puisse fonder une pré-somption légère en faveur de ce tourbillon de matière subtile: ainsi de cela seul que ce système est une pure hypothèse, il doit être rejeté.33

There is not the slightest experimental evidence, not the faintest analogy among things with which we are even slightly familiar, that would allow us to grant even the slimmest credibility to this vortex of subtle matter: this theory must be rejected for the simple reason that it is purely hypothetical.

A few pages later, to explain why hypotheses lead into error, why they need to be rejected, Voltaire employs the metaphor of the search for the source of a river:

N'allons donc point d'abord imaginer des causes et faire des hypotheses: c'est le stir moyen de s'égarer; suivons pas h pas ce qui se passe réelement dans la nature. Nous sommes des voyageurs arrives à l'embouchure d'un fleuve: il faut le remonter avant que d'imaginer où est la source.34

Let us not proceed then by first imagining causes and making hypotheses; that is the sure way to go astray; let us follow step by

step that which actually happens in nature. We are travelers who have arrived at the mouth of a river; our task is to follow the river to its source, not to imagine where the source may be.

The notion that responsible intellectual method should resemble the tracing of a river from its mouth to its source illustrates the seductiveness, for Voltaire, of the inductive model, according to which results are gradual and inevitable. It follows that the rejection of hypothesis is the rejection of imaginative guesswork, which attempts, presumptuously, to leap over the painstaking efforts of a conscientious investigator.

Consistent with this view of things, Voltaire's rejection of Descartes in the Lettres philosophiques centers on the taint of imagination. Descartes was born with a vivid and powerful imagination, Voltaire tells us, which "ne put se cacher même dans ses ouvrages philosophiques."35 This would account for the English view of the difference between Descartes and Newton (as it is characterized by Voltaire)—"le premier était un rêveur. . . . l'autre était un sage"36 —an opinion that is soon followed by the one quoted above, that Descartes's philosophy was nothing more than a "roman ingénieux."