series of papers. At the same time a suggestion by Whitehead inspired him to apply logical techniques to the analysis of perception; the result was a set of lectures delivered at Harvard and subsequently published as Our Knowledge of the External World (1914). This book, together with a paper entitled ‘The Relation of Sense-Data to Physics’ published in the same year, represents an excursus by Russell into something like phenomenalism. ‘Phenomenalism’ is the view that perceptual knowledge can be analysed in terms of our acquaintance with the fundamental data of sensory experience. (I say ‘something like phenomenalism’ because although Russell half a century later described these views as phenomenalistic, in the original writings they are not unambiguously so; this point is discussed in the appropriate place below.) Four years later, in another series of lectures, Russell applied his analytic method to objects and our talk of them. He called this the ‘Philosophy of Logical Atomism’. At the same time he published what is in effect a popular version of Principia Mathematica , setting out the basic ideas of the philosophy of mathematics. This book is entitled An Introduction to Mathematical Philosophy (1918).
In the 1920s Russell sought to extend and improve the application of his analytic techniques to the philosophy of psychology and physics. The first fruit of this was The Analysis of Mind (1921) in which his version of quasi-phenomenalism is applied to the analysis of mental entities. The second was The Analysis of Matter (1927), where Russell seeks to analyse the chief concepts of physics, such as force and matter, in terms of events. The argument of this book is strongly realist; Russell did not think it feasible to analyse the basic concepts of physics without admitting that certain entities exist independently of perception of them, which marked the end of any dalliance with phenomenalism. It might also be described as a ‘return’ to realism, because Russell had been committed to a rather swingeing form of realism before writing Our Knowledge of the External World .
Having made this return journey from a version of phenomenalism or something close to it, Russell reconsidered problems which he now felt had not been properly dealt with under his phenomenalist assumptions. The result was An Inquiry Into Meaning and Truth (1940) where he again discusses the relation of experience to contingent knowledge, and Human Knowledge (1948), where, among other things, he returns to a matter left inadequately discussed in earlier writings: the important question of non-demonstrative (non-deductive) inference, of the kind generally supposed to be employed in science.
7. Frontispiece of The Principles of Mathematics , published in 1900, with the premiss that mathematics and logic are identical.
Each of these phases in the development of Russell’s thought merits extended discussion, to be found in the works cited in Further reading below. In the following sections I give a summary account of them.
The rejection of idealism
Idealism takes a number of variant forms, but its basic tenet is that reality is fundamentally mental. ‘Idea-ism’ would be a more informative version of the name. It is a technical term of philosophy, and has nothing to do with ordinary senses of the English word ‘ideal’. In one of its forms, as held by Bishop Berkeley, idealism is the thesis that reality ultimately consists of a community of minds and their ideas. One of the minds is infinite, and causes most of the ideas; Berkeley identifies it as God. In later views of the kind espoused by T. H. Green and F. H. Bradley, both of whom were much influenced by German idealism, the thesis is that the universe ultimately consists of a single Mind which, so to speak, experiences itself. They argue that our finite, partial, and individual experience, which tells us that the world consists of a plurality of independently existing entities –