Berkeley and the role of hypothesis in natural philosophy

Abstract

The issue of the epistemological status of hypotheses postulating unobservable entities became prominent with the advent of modern science, in the 17th century. The basic reason is that such entities were widely employed by the new scientific theories in the explanation and prediction of natural phenomena, whereas empiricist epistemology, which at that time became very popular among philosophers and scientists, formed a clearly inhospitable background for unobservable elements in general. This paper examines the stands adopted, and the proposals made on this topic by George Berkeley, one of the most important critics of the philosophical foundations of natural science. It is shown that in the Principles of human knowledge he put forward a new notion of scientific explanation, according to which science could dispense with any hypotheses about unobservable entities whatsoever. A decade later, in De motu, Berkeley modified his philosophical system in order to accommodate Newtonian mechanics, which he considered "the best key" to natural science, by advocating an instrumentalist interpretation of the hypotheses on the existence of forces, which are of central importance in that theory. Finally, Berkeley's last book, Siris, is to a large measure dedicated to the discussion of a series of hypotheses about unobservable fluids, some of which Berkeley apparently defended as literal descriptions of unobservable layers of reality. In the last section of the paper an effort is made to show that this new position is not altogether incompatible with the main tenets of Berkeley's epistemology and metaphysics.