Eddington, Arthur Stanley (1882–1944)

views updated


Arthur Stanley Eddington was an English astronomer who was educated at Owens College, Manchester, and Trinity College, Cambridge, where he was Plumian professor of astronomy from 1913 to 1944. He never married, was socially rather diffident, and lived the quiet life of a Cambridge academic. He was elected a fellow of the Royal Society in 1914 and was knighted in 1930.

Eddington was one of the most brilliant theoreticians of his day, possessing an outstanding ability to survey complex and highly ramified subjects as wholes. His report to the Physical Society (1918) on the general theory of relativity, expanded into The Mathematical Theory of Relativity (London, 1923), contained important original contributions to the theory. Eddington's discovery of the mass-luminosity relation in stars and his explanation of white dwarf stars, which made possible the modern theory of stellar evolution, were published in The Internal Constitution of the Stars (London, 1926). These two books are considered to be his most substantial contributions to physics and astronomy. His interpretation of relativity theories led him to a belief in the profound importance of epistemology for physics. At first in semipopular books on modern physicsNature of the Physical World (London, 1928) and New Pathways in Science (London, 1935) being the most importantEddington argued for the view that physics could be almost entirely based upon investigations into the nature of sensation and measurement. A more elaborate and purely philosophical defense of his view was given in The Philosophy of Physical Science (London, 1939). Formal attempts actually to produce physics as derived in this way were presented in Relativity Theory of Protons and Electrons (London, 1936) and Fundamental Theory (London, 1946), published posthumously.

Eddington's real contributions to philosophy, if any, lie in his work on the epistemology of physics. However, he also defended idealism and mysticism, and he claimed that the indeterminacy of quantum physics solved the traditional philosophical problem of free will versus determinism in favor of free will. Particularly in his semipopular writings, Eddington was betrayed into philosophical excesses and, at times, gross confusion by a play of analogy and paradox, which, while part of his equipment as an immensely entertaining and brilliant writer, also served his love of mystery and obscurantism.

Selective Subjectivism

Eddington gave to his epistemological view the two names "selective subjectivism" and "structuralism." He accepted the causal theory of perception, and with this theory Eddington's own system stands or falls. From this theory it follows, first, that we know directly only the contents of our own consciousness (sense data) and, second, that these contents cannot be claimed to resemble elements of the objective world in any qualitative way. Our sensory apparatus selects from objective reality what we are able to observe and what is therefore the material for physical knowledge, just as, to use Eddington's own analogy, a net of a certain size mesh selects fish only of a size greater than the mesh. Just as we could generalize, prior to examining any catch of fish, about the size of fish the net would yield, so we can generalize in physics prior to the results of observation, merely by reflecting upon observational procedure, especially metrical procedures.

Despite distortions, mostly qualitative, in the picture that our senses thrust upon us, we may conclude that the picture has a structure in common with the unknowables that stimulate the senses. We notice patterns of recurrence in sensation, and it is the task of physics to elaborate the structure of these patterns. In particular, the structure of pointer-reading observations should be studied, since pointer readingsbeing merely observed coincidencesare minimally corrupted by the qualitative veils cast by our senses. However, Eddington denied the pointer readings directly represent anything objectively real.


Like Immanuel Kant, to whose system Eddington admitted that his own was distantly similar, he claimed that knowledge must conform to certain primitive rational patterns if it is to be intelligible. One of these forms of thought is that we believe in the existence of minds other than our own. The recognition of a common structure in the experience of many minds leads to a belief in an objective reality independent of these minds. There is no primitive belief in an objective reality. This route to the existence of an external world is an unobtrusive but significant part of Eddington's idealistic metaphysics.

Using the notion of structure as defined in the mathematical theory of groups, Eddington was able, out of highly generalized material from epistemology (for example, the claim that only relations between things are observable) and from the forms of thought, to build quite intricate group structures, for example, the structure found in Paul Dirac's mathematical specification of an elementary particle in an elementary state giving charge and spin. In addition to this a priori derivation of the formal structure of laws, Eddington also exploited the theory of groups in deducing a priori the basic natural constants, such as the gravitational constant and the fine structure constant, from various features of the group structure of the type of mathematics employed. In this, he compared himself with Archimedes, who deduced the nature of π from the axiom of Euclid, whereas previous determinations of its nature had relied upon merely empirical methods.

On this basis Eddington claimed that the mind fits nature into a pattern determined by the nature of the mind itself; that the discoveries made by the physicist are just what his sensory, intellectual, and metrical processes dictate that he shall find.

It is difficult not to share the general view that Eddington vastly overstated the extent to which convention enters into theory construction. Extensive criticism in this entry without more extensive elaboration of the complexities of his group structure derivations would be unjust. Some brief comments must suffice.

Eddington's view was that observation was required only for the purpose of identifying, on the one hand, the elements of the group constructed by pure mathematics with, on the other hand, the theoretical terms of, say, electromagnetism. It is far from clear where he thought the complete theoretical structure then stood from the point of view of its a priori status. If such "identification" demands that it be fully observed that the electromagnetic field is properly (that is, truly) described by Maxwell's equations, which have the group structure in question, then Eddington was requiring "observation" to add a very great deal more than he seems to have been prepared to admit.

Eddington fell into confusion that illustrates well his mistakes in general. This was his claim that the basis of the special theory of relativity may be deduced a priori because it depends on the fact that simultaneity of events at a distance from each other is not observable, that is, that it depends upon an epistemological fact. It is true that to decide a question about the simultaneity of spatially separated events, one must make assumptions as to the speed of the signals that inform one that the events have occurred. And it is also true that in the last resort these assumptions could be checked only if one could decide independently on the simultaneity of events spatially distant from each other. But this epistemological circularity is an insufficient basis for relativity theory. Moreover, further contingent facts, not deducible a priori (for example, the fact that in any inertial system light takes the same time round any closed paths of the same length, whatever their orientation) are required. Eddington claimed that the result of the Michelson-Morley experiment could have been foreseen on a purely epistemological basis. It seems quite clear that he was wrong.


"To put the conclusion crudelythe stuff of the world is mind-stuff," Eddington wrote in Nature of the Physical World. The idealist conclusion was not integral to his epistemology but was based on two main arguments.

The first derives directly from current physical theory. Briefly, mechanical theories of the ether and of the behavior of fundamental particles have been discarded in both relativity and quantum physics. From this Eddington inferred that a materialistic metaphysics was outmoded and that, in consequencethe disjunction of materialism or idealism being assumed exhaustivean idealistic metaphysics is required.

The second and more interesting argument was based on Eddington's epistemology and may be regarded as consisting of two parts. First, all we know of the objective world is its structure, and the structure of the objective world is precisely mirrored in our own consciousness. We therefore have no reason to doubt that the objective world, too, is "mind-stuff." Dualistic metaphysics, then, cannot be evidentially supported. (The conclusion appears to be a valid deduction from its premises.)

But, second, not only can we not know that the objective world is nonmentalistic, we also cannot intelligibly suppose that it could be material. To conceive of a dualism entails attributing material properties to the objective world. However, this presupposes that we could observe that the objective world has material properties. But this is absurd, for whatever is observed must ultimately be the content of our own consciousness and, consequently, nonmaterial. This last argument confuses, among other things, the supposition that the objective world has certain properties with the supposition of our observing that it has them.

See also Determinism and Freedom; Epistemology, History of; Idealism; Mysticism, History of; Kant, Immanuel; Popular Arguments for the Existence of God; Stebbing, Lizzie Susan; Subjectivist Epistemology.


additional works by eddington

Space, Time, and Gravitation. Cambridge, U.K.: Cambridge University Press, 1921.

Science and the Unseen World. London: Allen and Unwin, 1929.

The Expanding Universe. Cambridge, U.K.: Cambridge University Press, 1933.

works on eddington

Dingle, Herbert. The Sources of Eddington's Philosophy. Cambridge, U.K.: Cambridge University Press, 1954.

Stebbing, S. Philosophy and the Physicists. London: Metheun, 1937.

Whittaker, E. T. Eddington's Principle in the Philosophy of Science. Cambridge, U.K.: Cambridge University Press, 1951.

G. C. Nerlich (1967)