The new physiology was closing in on the mind. Through the 1860s the German laboratories measured nerve conduction, mapped function onto brain tissue, and reduced what had been called the soul to a sequence of reflexes; Darwin had already made human beings continuous with animals. If that account was complete, a decision was only the feeling that accompanies a process already settled, and moral effort was theatre. The men best placed to feel the force of this were the ones trained inside it. William James was one, and the argument nearly finished him. What he did about it became a philosophy: no description of the machinery, however good, gets to decide what a person is entitled to believe.
He arrived there by a route with no straight lines in it. He studied painting under William Morris Hunt at Newport, switched to Harvard's Lawrence Scientific School in 1861, sailed up the Amazon with Louis Agassiz in 1865 and came home after eight months, sick and unconvinced by collecting. He took his medical degree in June 1869 and never practised a day of medicine. Then came the collapse: months of depression and suicidal thought, made worse by the conviction that a determined universe left him nothing to push against. In the spring of 1870 he read Charles Renouvier on freedom of the will and recorded in his diary the decision that ended the paralysis — his first free act would be to believe that his will was free.
That could have stayed a private repair, but he generalised it instead. In The Will to Believe, published in 1897, he argued that when a question is genuinely live, cannot be dodged, matters enormously, and cannot be settled by evidence either way, refusing to decide is itself a decision, and one that carries its own risk of being wrong. He was answering W. K. Clifford, who had ruled in 1877 that it is wrong, always and everywhere, to believe anything on insufficient evidence. James thought the rule smuggled in a preference: the fear of error over the hope of truth.
Meanwhile he was building American psychology from nothing. He taught the first experimental psychology course at Harvard in 1875–76, and liked to point out that the first lecture on psychology he ever heard was one he gave. He contracted The Principles of Psychology in 1878, promising it in two years, and delivered it in 1890 — a decade late and, by his own description, a bloated mass. It described consciousness as a stream rather than a chain of separate ideas, one in which the transitions and relations are as real as the parts they join.
The claim he is remembered for followed from all of it. An idea's meaning is the difference it makes to experience; truth is not a property an idea sits on but something that happens to it as it works. Stated that flatly, it drew fire at once. Charles Peirce, who had coined the term, renamed his own version pragmaticism in 1905 to get clear of James; Arthur Lovejoy counted thirteen incompatible pragmatisms in 1908; Bertrand Russell later ridiculed the whole thing.
James died of heart failure at Chocorua, New Hampshire, on 26 August 1910, his metaphysics unfinished. The ridicule held for half a century, and then Richard Rorty and Hilary Putnam brought pragmatism back, Putnam arguing that Russell had attacked a caricature.