Human Knowledge

Overview

About five hundred and forty pages, Bertrand Russell’s last big book, published in 1948, asking how the confident claims of science can be justified from the scraps of private experience on which they rest. In six parts it surveys the world of science, language, science and perception, scientific concepts, probability, and the postulates of scientific inference.

Its conclusion is that induction alone cannot ground science and that five postulates, such as the quasi-permanence of things and the existence of causal lines, must be assumed because animals and humans in fact rely on them and survive.

What happens

The world of science. Russell’s first part surveys the world of science as commonly understood, establishing the confident public claims whose actual epistemological foundation the book’s remaining parts will subject to sustained critical examination.

Language. Russell examines language, treating linguistic structure as bearing directly on how scientific concepts get formed and communicated, connecting his epistemological project to his own earlier logical and linguistic concerns.

Science and perception. Russell addresses science and perception together, examining how the confident public claims of science actually relate to the genuinely limited private perceptual experience from which they must somehow be constructed.

Scientific concepts. Russell surveys scientific concepts themselves, examining how these abstract theoretical constructs relate to and depart from the immediate experiential evidence that originally motivated their formation and continued use.

Probability. Russell treats probability at length, addressing how uncertain inference can nonetheless support confident scientific conclusions, a question directly bearing on his eventual argument about induction’s genuine limits.

The postulates of scientific inference. Russell concludes with the postulates of scientific inference, proposing five specific postulates, including the quasi-permanence of things and the existence of causal lines, that must be assumed beyond pure induction.

How to read it

Read Russell’s central question, how confident scientific claims can be justified from the scraps of private experience on which they actually rest, as exposing empiricism’s genuine foundational vulnerability with unusual directness and sustained analytical rigor.

Pay close attention to how the conclusion that induction alone cannot ground science represents a significant late concession from a philosopher who had spent much of his career working within broadly empiricist assumptions about scientific justification.

Notice how the proposed five postulates, including the quasi-permanence of things and the existence of causal lines, offer a distinctive solution justified not by further argument proving their truth but by animals and humans relying on them and surviving.

Consider that as Russell’s last major philosophical work, this book represents his final considered attempt to address the fundamental relationship between science’s confident public claims and the genuinely limited private experience from which all such claims must somehow be constructed.

This book completes a long philosophical arc worth tracing back to Russell’s much earlier Problems of Philosophy: the confident epistemological program sketched there confronts, decades later, the harder limits this final, more sobered treatment openly concedes.

Questions it raises

• Does Russell’s conclusion that induction alone cannot ground science successfully identify a genuine limit on empirical justification, or does this concession understate resources available to more sophisticated accounts of inductive and probabilistic reasoning developed after his own writing?
• How persuasive are the five proposed postulates, justified by animals and humans relying on them and surviving rather than by independent logical demonstration? Does grounding epistemology in practical survival adequately answer skeptical challenges to scientific knowledge’s foundations?
• What does surveying language, perception and scientific concepts together, before addressing probability and the postulates, suggest about how Russell understood the relationship between these different dimensions of the broader epistemological problem he was addressing?
• How should a reader weigh this book’s status as Russell’s last major philosophical work against his earlier, more confident treatments of similar epistemological questions? Does this final treatment represent genuine intellectual humility, or simply a different emphasis within a consistent underlying position?
• Given that this book was published in 1948, how do its arguments about science’s foundations relate to subsequent developments in philosophy of science, particularly later debates about theory-ladenness of observation and the underdetermination of theory by evidence?
• What does this book’s considerable length and six-part systematic structure suggest about the scope Russell believed the question of scientific justification actually required? Does this comprehensive treatment succeed in addressing the problem from every necessary angle?