Scientific Inference and the Earth’s Interior: Dorothy Wrinch and Harold Jeffreys at Cambridge

Teru Miyake

Introduction

That philosophical issues can, at times, have a profound influence on the development of a science, is by now a familiar idea to historically-minded philosophers of science. Studies of such influences have been limited, however, to a few sciences, with by far the most work being done on physics. I am quite confident, for example, that only a handful of philosophers are at all aware of a connection between the development of the field of seismology in the early decades of the twentieth century and the school of philosophy centered around Bertrand Russell and G. E. Moore at Cambridge. The main aim of this paper is to bring this connection to light through an examination of the work of Dorothy Wrinch and Harold Jeffreys, each of whom were students of this tradition of philosophy, and who went on to do important work in several different fields of science.

Since this is a topic that is unfamiliar to most philosophers, and the motivations for this research might be obscure, I will explain my reasons for undertaking this research. I was originally led to this work through an interest in the epistemology of contemporary geophysics. Put simply, I was interested in the question of how it is that we can say we know things about the deep interior of the earth, in the face of what looks to be a severe underdetermination problem. Almost all the information we have about the earth’s deep interior comes from observations of seismic waves at the earth’s surface. The number of such observations we can make is necessarily finite, and indeed limited by the number and magnitude of the earthquakes that occur. On the other hand, the earth’s interior is immense, and likely to be extremely complicated—the number of degrees of freedom of the earth’s interior is practically infinite. This underdetermination was explicitly pointed out in the late 1960s by the geophysicists George Backus and Freeman Gilbert (Backus and Gilbert 1968), and

T. Miyake (*)

Nanyang Technological University, Singapore, Singapore e-mail: This email address is being protected from spam bots, you need Javascript enabled to view it

© Springer International Publishing AG 2017

F. Stadler (ed.), Integrated History and Philosophy of Science, Vienna Circle Institute Yearbook 20, DOI 10.1007/978-3-319-53258-5_7

this realization led to the development of techniques for making inferences in the face of uncertainty that are now known as the theory of inverse problems (Miyake 2011).

Of course, it is not as if the problem was unknown to geophysicists until then. Indeed, it is in grappling with this very problem that Harold Jeffreys, one of the founders of the field of seismology, came to work on the foundations of probability, eventually writing two seminal works on the topic (Jeffreys 1931, 1939). Of particular interest for philosophers is a series of papers that Jeffreys co-authored in the late 1910s and 1920s with Dorothy Wrinch, a philosopher and mathematician who would later go on to make significant contributions to mathematical biology. At the time, she was a student of Russell’s, and an active member of philosophical circles at Cambridge. This is, then, an interesting case where budding scientists who were deeply familiar with the cutting edge of epistemology tried to take these philosophical ideas and apply them towards their own scientific work, the most significant being the work that would occupy Jeffreys for the rest of his life—that of inferring properties of the deep interior of the earth from seismological data. It turns out that Wrinch and Jeffreys found the then-current theories of epistemology lacking, and tried to develop their own ideas—some of which would end up in Jeffreys’s later work on the foundations of probability.

This paper will be a brief examination of this work by Wrinch and Jeffreys, focusing on the following question: Exactly what was it that they found inadequate about the epistemological theories of the time, and how did they try to overcome such problems? In the rest of this paper, I will first give very short biographical sketches of Wrinch and Jeffreys, since many readers are likely unfamiliar with them. I will then provide a couple of examples of Wrinch and Jeffreys’s confrontations with ideas coming out of Cambridge philosophy at the time, and the way in which they reacted to them.