In lieu of an abstract, here is a brief excerpt of the content:

Reviewed by:
  • Isaac Newton’s Scientific Method: Turning Data into Evidence about Gravity and Cosmology by William L. Harper
  • Katherine Dunlop
William L. Harper. Isaac Newton’s Scientific Method: Turning Data into Evidence about Gravity and Cosmology. Oxford-New York: Oxford University Press, 2011. Pp. xvii + 424. Cloth, $75.00.

Not a full treatment of Newton’s scientific method, this book discusses his optical research only in passing (342–43). Its subtitle better indicates its scope: it focuses narrowly on the argument for universal gravitation in Book III of the Principia. The philosophical project is to set out an “ideal of empirical success” realized by the argument.

Newton claims his method is to “deduce” propositions “from phenomena.” On Harper’s interpretation Newton’s phenomena are patterns of data, which are used to measure “parameters” by which the theory explains them. An example is Kepler’s Area Rule. Its fit to the data confirms the hypothesis that the force maintaining a body in orbit is directed toward the center, but (according to Harper) treating phenomena as measurements makes “more informative evidence” out of data than merely predicting them (3). Newton proves that if the rate at which areas are swept out is increasing, the force is directed forward of the center, and if the rate is decreasing, it is directed backward. Thus an increase or decrease of the rate measures the direction (relative to the center) of the force, and these “systematic dependencies” offer a “compelling sort of explanation of the uniform description of areas by the centripetal direction of forces” (43). [End Page 489]

This process of deduction warrants provisional acceptance of propositions (e.g. that in a certain system, the force maintaining a body in orbit is directed toward the center). Testing these regularities gives direction to research. For instance, deviations from the Area Rule prompt us to find another body influencing a system. An example of the progress in theory spurred by such discrepancies is Newton’s quantitative account of the variational inequality of the moon’s orbit, which explains violations of the Area Rule in terms of the sun’s action on the earth–moon system. Through iterations of this process, propositions accepted as approximately correct are rendered more exact. Newton’s deliberate employment of successive approximations defuses the classic objection that the consequences of universal gravitation contradict the premises from which it is derived.

On Harper’s view, the basic way in which Newton’s method makes more informative evidence out of data than the hypothetico-deductive is by treating phenomena as measurements. So precision is central to the ideal of success. But at a crucial juncture—the identification of terrestrial gravity with the force keeping the moon in orbit—Newton’s claim to precision appears spurious. Newton appeals to the agreement between pendulum measurements of gravity at the earth’s surface and his computation of its strength from the data of the lunar orbit, with the inverse-square law assumed. Newton fails to clearly justify his choice and correction of an estimate of the moon’s distance or his calculation of the sun’s contribution to its attraction toward earth. His entreaties to the Greenwich astronomer John Flamsteed foster suspicion that his data are selected for their fit with theory. Newton argued that for Flamsteed to publish his lunar observations with Newton’s theory would demonstrate “their exactness and make you readily acknowledged the Exactest Observer that has hitherto appeared” (The Correspondence of Isaac Newton, IV.87 [Cambridge UP, 1967]). Flamsteed retorted “theories do not commend observations, but are to be tried by them” (quoted in An Account of the Rev. John Flamsteed, Francis Baily, 152 [Lords Commissioners of the Admiralty, 1835]).

Harper offers a powerful defense of Newton’s reasoning. Harper argues that by calculating a value from theory for the earth–moon distance, Newton could show that the calculated value for surface gravity continues to agree with measurement even as the orbit assumed for the moon is complexified to account for various effects. This is relevant for a philosophical understanding of Newton’s goals, because it brings home that precision is to be achieved through successive approximations, not by a single inference. Harper...

pdf

Share