From Kepler's laws, so-called, to universal gravitation: empirical factors (Q2537954)

This study is concerned with the intertwining roles of conjecture and empirical fact in the development of planetary theory from Kepler to Newton. It divides into three parts. The first argues that Keplerian planetary theory is basically hypothetical. Since Voltaire's ``Éléments de la philosophie de Newton'' (1738), it has been customary to speak of ``Kepler's laws'', and to regard the elliptical orbit, equable description of areas, and proportionality between the squares of the periods and cubes of the mean radii, as truths established empirically by Kepler, ``deduced strictly from observations'', to quote Smart's ``Celestial Mechanics''. Before the publication of Newton's \textit{Principia}, by contrast, astronomers spoke not of ``Kepler's laws'' but of ``Kepler's hypothesis'', and this was in fact the more appropriate locution. The main reason is that, with the \(2'\) to \(4'\) errors in Tycho's observations, the ellipticity of the Martian orbit did not admit of persuasive verification on the basis of radius vector determinations alone. As Newton will put it in 1686, ``Kepler knew ye Orb to be not circular but oval & guest it to be Elliptical''. What Kepler could verify with satisfying precision was the combination `ellipsis-cum-lex-superficierum', where the so-called area law was not an empirically established truth but rather a deduction from Kepler's hypothetical and indeed mistaken celestial mechanics. The logical situation here -- the unavoidable mixture of conjecture with fact -- was recognized by Kepler himself and by a number of astronomers who followed him in the decades before the appearance of Newton's \textit{Principia}. The second section deals with attempts of mid-century astronomers to revise or replace the Keplerian foundation, and in particular with Ismael Boulliau's attempt to derive the elliptical path and mode of motion of the planets from a basis that is at once Aristotelian, Copernican, and anti-Keplerian. Boulliau's hypothesis turned out (after some confusing alterations) to be equivalent to assuming the planetary motion to be angularly uniform about the superior or empty focus. This ``simple elliptic hypothesis'' had the distinction of proving to be false, when compared with Tycho's observations of Mars. By 1670 it was thus widely known that on the assumption of an elliptical orbit, the calculus for the motion needed to approximate closely to Kepler's area rule. The elliptical orbit nevertheless remained hypothetical, and Newton will claim to be the first to establish it on a sound basis. The third part of the study is concerned with the emergence of Newton's law of universal gravitation. One view had been that Newton has all his important ideas on this subject in 1665 or 1666, and then, either because of a faulty value for the radius of the Earth (Pemberton, Voltaire), or because of the lack of a certain theorem, having to do with the attraction of a sphere of gravitating matter on an external particle (Cajori), delayed for nearly two decades to develop these ideas. Another view has been that Newton obtained all his important ideas on the subject of gravitation from Robert Hooke in letters of 1679, and then delayed for five years to develop these ideas (Vavilov, Lohne). The present paper argues, on the basis of the available Newtonian documents from the years before the ``Principia'', that Newton did not hit on his argument for universal gravitation before the summer of 1684, and therefore -- being a stickler for as much certainty as the subject matter admitted of -- was not satisfied to proceed on the basis of this assumption before that time. The important step had to do with the deduction from the proportionality of weight and mass (as verified by pendulum experiments) of the mutual gravitational attraction of all bodies accessible to observation. If the reading her given of the sequence of Newtonian manuscripts is correct, the law of universal gravitation began as a surmise, a conjecture, that had to make its way against other speculative alternatives. It came to be confirmed by an inductive argument, but this argument was not merely a simple act of extrapolation or generalization, but contained a logical `tour de force' which had occurred to no one, so far as the evidence shows, before 1684. After the publication of the ``Principia'', the subtlety and care of Newton's argumentation from phenomena to universal gravitation, and from there to phenomena again, including Kepler's so-called laws as approximate results, was generally lost from sight. With substantial assistance from Voltaire, an uncommon incomprehensibility gradually, by habituation, became a common incomprehensibility, and men came to think of universal gravitation as one of the obvious inductive truths. It nevertheless remained hypothesis, impressively confirmed, it is true, but with consequences, as we know today, that are slightly wrong.