construction of the physical world. In this book he abandoned sensibilia as fundamental constituents of the world in favor of events, which were ‘neutral’ because intrinsically neither physical nor mental.
In 1916 Russell was dismissed from Cambridge on political grounds and from that time on had to earn his living by writing and public lecturing. His popular lectures, ‘The Philosophy of Logical Atomism’ (1918), were a result of this. These lectures form an interim work, looking back on the logical achievements of 1905–10 and emphasizing their importance for philosophy, while taking stock of the problems raised by Wittgenstein’s criticisms of the multiple relation theory. In 1919 Russell’s philosophy of mind underwent substantial changes, partly in response to those criticisms. The changes appeared in ‘On Propositions: What They Are and How They Mean’ (1919) and The Analysis of Mind (1921), where the influence of contemporary trends in psychology, especially behaviorism, is evident. Russell gave up the view that minds are among the fundamental constituents of the world, and adopted neutral monism, already advocated by Mach, James, and the American New Realists. On Russell’s neutral monism, a mind is constituted by a set of events related by subjective temporal relations (simultaneity, successiveness) and by certain special (‘mnemic’) causal laws. In this way he was able to explain the apparent fact that ‘Hume’s inability to perceive himself was not peculiar.’ In place of the multiple relation theory Russell identified the contents of beliefs with images (‘imagepropositions’) and words (‘word-propositions’), understood as certain sorts of events, and analyzed truth (qua correspondence) in terms of resemblance and causal relations.
From 1938 to 1944 Russell lived in the United States, where he wrote An Inquiry into Meaning and Truth (1940) and his popular A History of Western Philosophy (1945). His philosophical attention turned from metaphysics to epistemology and he continued to work in this field after he returned in 1944 to Cambridge, where he completed his last major philosophical work, Human Knowledge: Its Scope and Limits (1948). The framework of Russell’s early epistemology consisted of an analysis of knowledge in terms of justified true belief (though it has been suggested that he unintentionally anticipated Edmund Gettier’s objection to this analysis), and an analysis of epistemic justification that combined fallibilism with a weak empiricism and with a foundationalism that made room for coherence. This framework was retained in An Inquiry and Human Knowledge, but there were two sorts of changes that attenuated the foundationalist and empiricist elements and accentuated the fallibilist element. First, the scope of human knowledge was reduced. Russell had already replaced his earlier Moorean consequentialism about values with subjectivism. (Contrast ‘The Elements of Ethics,’ 1910, with, e.g., Religion and Science, 1935, or Human Society in Ethics and Politics, 1954.) Consequently, what had been construed as self-evident judgments of intrinsic value came to be regarded as non-cognitive expressions of desire. In addition, Russell now reversed his earlier belief that deductive inference can yield new knowledge. Second, the degree of justification attainable in human knowledge was reduced at all levels. Regarding the foundation of perceptual beliefs, Russell came to admit that the object-knowledge (‘acquaintance with a sensedatum’ was replaced by ‘noticing a perceptive occurrence’ in An Inquiry) that provides the non-inferential justification for a perceptual belief is buried under layers of ‘interpretation’ and unconscious inference in even the earliest stages of perceptual processes. Regarding the superstructure of inferentially justified beliefs, Russell concluded in Human Knowledge that unrestricted induction is not generally truthpreserving (anticipating Goodman’s ‘new riddle of induction’). Consideration of the work of Reichenbach and Keynes on probability led him to the conclusion that certain ‘postulates’ are needed ‘to provide the antecedent probabilities required to justify inductions,’ and that the only possible justification for believing these postulates lies, not in their self-evidence, but in the resultant increase in the overall coherence of one’s total belief system. In the end, Russell’s desire for certainty went unsatisfied, as he felt himself forced to the conclusion that ‘all human knowledge is uncertain, inexact, and partial. To this doctrine we have not found any limitation whatever.’ Russell’s strictly philosophical writings of 1919 and later have generally been less influential than his earlier writings. His influence was eclipsed by that of logical positivism and ordinary language philosophy. He approved of the logical positivists’ respect for logic and science, though he disagreed with their metaphysical agnosticism. But his dislike of ordinary language philosophy was visceral. In My Philosophical Development (1959), he accused its practitioners of abandoning the attempt to understand the world, ‘that grave and important task which philosophy throughout the ages has hitherto pursued.’ See also FREGE, LOGICAL CONSTRUCTION , LOGICISM , PERCEPTION , SET — THEORETIC PARADOXES , SET THEORY, THEORY OF DESCRIPTIONS , TYPE THEORY , WHITEHEA. N.G. & D.B.M.