Eight lectures, delivered to paying general audiences at the Lowell Institute in Boston over November and December 1906 and again at Columbia in January 1907, revised and published in 1907. James dedicated the book to John Stuart Mill.

The audience matters more than it usually does. He was not writing for philosophers, he was explaining to a room, and the consequences run both ways: it is the most readable book in the tradition and it is the one where the formulations are loosest, which is why it made pragmatism famous and why it gave its critics everything they needed.

What reading it is like

Short — around 250 pages — and genuinely easy, which for a philosophical classic is rare enough to be worth saying plainly. There is no apparatus, no technical vocabulary, and the sentences are built to be heard rather than studied. James was the best prose writer in American philosophy and this is the book where that is most obvious.

Where to start: Lecture II. If you read one thing from the pragmatist tradition it should be that lecture, and the squirrel is in the first three pages of it. Lecture I sets up the framing and is worth having. Lectures VI and VII are the contested ones. Lecture VIII is where he says what he actually wanted the method for.

Lectures III, IV and V apply the method to standing metaphysical disputes and are the most dated, because the disputes were live in 1907 and mostly are not now.

The squirrel, which is the method

James opens the second lecture with an argument he says he walked in on during a camping trip. A squirrel is on one side of a tree trunk and a man on the other. The man moves around the tree trying to see the squirrel; the squirrel moves too, keeping the trunk between them, so the man never catches sight of him.

Does the man go round the squirrel?

The party had split and was deadlocked. James's answer is that it depends entirely on what you mean by going round, and that both sides are right on their own reading and are not actually disagreeing about anything. If going round means passing north, east, south and west of the squirrel, the man does. If it means being successively in front of, to the right of, behind and to the left of him, the man does not — because the squirrel turns with him.

That is the pragmatic method in operation rather than in description. Take a dispute; work out what practical difference each answer would make; if no difference can be found, the dispute is verbal and the work is to show it. It is the same instruction Peirce gave in 1878, applied to a case anyone can follow.

Three pages, and the whole first half of the book is downstream of it.

Tender-minded and tough-minded

Lecture I sets up the frame the rest is arranged against, and it is a temperamental diagnosis rather than an argument.

Philosophy, James says, divides into two casts of mind. The tender-minded are rationalist, intellectualist, idealistic, optimistic, religious, free-willist, dogmatical. The tough-minded are empiricist, sensationalistic, materialistic, pessimistic, irreligious, fatalistic, skeptical.

His claim is that people want the facts of the second and the values of the first, and that the existing options force a choice. Pragmatism is offered as the thing that does not — a method with the empiricist's respect for evidence and no prior commitment against the tender-minded conclusions.

It is a sales pitch, and he is open about it being one. It is also the most honest account anyone has given of why philosophical positions track personality, and it is the part of the book least often attacked, because it is hard to deny.

The sentence that caused the trouble

In Lecture II, and again in VI, James says that true ideas are those we can assimilate, validate, corroborate and verify; that truth happens to an idea; that it is made true by events. And, fatally, that the true is the name of whatever proves itself to be good in the way of belief — and that we can say of a truth that it has cash-value in experiential terms.

The cash-value phrase is a lecturer's image and it did more damage to his reputation than any argument against him. Read generously, the claim is about what it is for a belief to be true: not standing in a static relation to a fact, but working — leading through experience, fitting with what else you hold, surviving contact with what happens.

Read as written, it makes truth a property of what a belief does for the person holding it, and that is what Russell attacked in William James's Conception of Truth in 1910, and Moore and Bradley from their own directions. James answered at length in The Meaning of Truth (1909), which is the companion volume and reads as a man discovering that the phrase he was pleased with has cost him ten years.

The standing treatment of that objection is on William James. What belongs here is that the book is where it happened, and that the loose formulation and the fame are the same fact.

Lecture VIII, which is the point

The final lecture is on religion, and it is what the method was for.

James argues that where a choice is live, forced and momentous, and the evidence cannot settle it, a person is entitled to decide according to what he calls their passional nature. The position is continuous with The Will to Believe (1896) and with The Varieties of Religious Experience, and it is the destination the whole book has been moving toward.

This is the honest objection to the book taken as a whole. A method for settling disputes by their practical consequences, which arrives at a conclusion licensing religious belief on the ground that believing it pays — the critic's suspicion that the machinery was built to reach the destination is not unreasonable, and James does nothing in Lecture VIII to allay it.

Key passages

Lecture I — the tender-minded and tough-minded taxonomy; the claim that philosophy is driven by temperament.

Lecture II — the squirrel; the statement of the pragmatic method; truth as what happens to an idea; the cash-value passage.

Lecture III — the method applied to substance, materialism and design.

Lecture IV — the one and the many, which James calls the most central of all philosophical problems. See the one and the many.

Lecture V — common sense as a set of inherited concepts that worked, and the question of whether anything better is available.

Lecture VI — Pragmatism's Conception of Truth. The most attacked chapter in American philosophy.

Lecture VII — Pragmatism and Humanism; the claim that we add to reality in describing it.

Lecture VIII — Pragmatism and Religion; the meliorist position, between optimism and pessimism.

What is contested

Whether the theory of truth is defensible at all. Russell's objection has never been answered to general satisfaction, and the two subsequent traditions split on it: the Peircean recovery, most forcefully by Cheryl Misak, holds that the defensible core belongs to Peirce and that James's formulations are the reason pragmatism spent a century dismissed as relativism; Rorty's neo-pragmatism from the 1980s embraces the Jamesian version and accepts the consequence.

Whether the method does any work. The squirrel is a genuinely verbal dispute and the method resolves it cleanly. Critics argue that real philosophical disagreements are not like that, and that where the method has been applied to one — free will, substance, design — it has tended to report that the dispute is verbal in a way its practitioners found convenient.

How much of it is James's temperament. He opens by arguing that philosophical positions track personality, and then advances a position that fits his own unusually well. He would not have regarded this as an objection.

Its relation to Dewey. Dewey built a systematic philosophy on similar ground, took the instrumental account of ideas much further, and largely avoided the vocabulary that got James into trouble — which is one reason he was still being read as a live philosopher when James was being taught as a stylist.