Why physics still matters
Part 3 of 5: The nomological machine
In our everyday experience, we perceive a sharp contrast between the behavior of the heavenly bodies and that of the things around us here on earth. While the stars and planets exhibit regular and unchanging patterns of motion, which can be rigorously quantified, it is almost impossible to find any natural phenomenon on earth that can be described in this way. Contrary to what one sometimes reads in popular histories of science, modern physics did not originate with a decision to “start observing nature” — as opposed to following an a priori philosophy or theology. The ancients observed nature constantly, with much greater care than most modern scientists. The scientific revolution required a much more elusive intellectual insight: that the quantitatively reproducible motions required for a formally mathematical science of nature can be created by human industry.
The first steps in this direction were taken by scholars such as Jordanus of Nemore1 and John Buridan2 who investigated the motion of levers and projectiles in the great medieval universities. In the real world, every falling object follows its own unique trajectory, affected by air currents, the rotation of the object, and the precise way in which it was launched into the air. The ancient philosophers rarely attempted to describe such motion quantitatively, and assumed that the most that could be hoped for would be general qualitative trends, such as the fact that heavier things tend to fall faster (an empirically true observation when the objects are subject to air resistance, as they always are on earth). These Medieval scholars, however, realized that they could generate trajectories almost as reproducible as those of the planets, by conducting experiments under carefully controlled conditions. In this way, they gradually developed a geometric procedure for accurately predicting the location of a falling body dropped from rest under such conditions of isolation.3
At first glance, this method seems more like a practical art or engineering discipline than a speculative science. The object of study is not nature or a part of nature, but a set of quantities generated by human activity, describing an artificial scenario assembled by human intelligence. In fact, these experiments fit the definition of a “machine”: they are carefully assembled aggregates of materials designed by humans to reliably carry out a specific goal. But here, the goal is not any immediate practical end. This is a new and strange kind of machine, a machine for producing scientific knowledge of the phenomena of nature. From the Greek word for law (“nomos”), one could call it a “nomological machine,” a term coined by philosopher of science Nancy Cartwright to express precisely this concept.4
Working with such nomological machines led these Medieval scholars to an even more profound insight, which tied their research back to the science of heavenly phenomena which was its original inspiration. Since the heavenly bodies and the earth as a whole are presumably not subject to wind resistance, their motion should perfectly obey the law of inertia that holds only approximately for terrestrial bodies.5 The regular and reliable trajectories of these bodies no longer need to be attributed to their superior intrinsic qualities, or to guidance by angels. They may simply result from the extreme isolation of these objects from uncontrolled causal factors. This hypothesis was finally confirmed by Isaac Newton three centuries later, when he demonstrated that the very same mathematical relationships that characterize the motion of idealized projectiles on earth also describe the motion of all the planets of the solar system and all their observable moons.
Through the construction of nomological machines, all the phenomena of nature can now become the object of a formally quantitative science, the kind of science that was formerly restricted to astronomy alone. The universal scope of this effort enabled it to quickly gain possession of the name “physics,” taking it over from the philosophical discipline that studies the most general powers and properties of natural things. Unlike the old physics, this new science has a direct technological application: any practical machine constructed on the basis of established nomological machines should behave in a predictable and controllable way. But by abstracting the physical world down to a set of measurement results, the new physics surrenders the possibility of considering the old questions about the nature of time and space and the principles of motion.
See Pierre Duhem, Les Origines de la Statique (1906), pp. 208-212 in the linked translation. All references to Pierre Duhem in this series are taken from the appendix to Stanley Jaki, Scientist and Catholic: Pierre Duhem (2004).
See Pierre Duhem, Etudes sur Léonard de Vinci (1913), pp. 226ff. in the linked translation.
In a treatise published in 1545, the Spanish Dominican Dominic Soto reports the correct formula for computing the distance traveled by a falling body in a given amount of time (the same formula generally attributed to Galileo, who had not yet been born). He treats this formula as an accepted and well-known result, indicating that he probably learned it during his studies at the University of Paris, where it had already become part of the normal curriculum. See Pierre Duhem, Etudes sur Léonard de Vinci (1913), p. 230 in the linked translation.
Nancy Cartwright, “Where do laws of nature come from?” Dialectica 51:65-78, 1997. See also Edward Feser, Aristotle’s Revenge, 2019, pp. 177-190 for a recent reappraisal of Cartwright’s theory.
14th-century Parisian physicist John Buridan (1301-1359) seems to have been the first to write down this hypothesis: “At the creation of the world, God moved the heavens with motions identical to the ones with which they actually move. Therefore he imparted to them impetuses by which they continue to move uniformly. Not encountering any resistance contrary to them, these impetuses are never diminished nor destroyed … within this view it is not necessary to posit the existence of intellects that move the celestial bodies in an appropriate manner.” Taken from Pierre Duhem, Etudes sur Léonard de Vinci (1913), p. 228 in the linked translation.