- Length
- 704 pagesNorton and Norton, A Treatise of Human Nature, Oxford, 2000
- Reading
- DemandingReadable with attention; assumes no other book
- Availability
- Many editionsIn print in several competing editions
A Treatise of Human Nature
A young man's attempt to do for the mind what Newton had done for the heavens, which arrived at conclusions so unsettling that its author spent the rest of his life restating them more gently. Nearly every problem in modern philosophy has a version of itself in this book.
- Date
- 1739-40
- Author
- David Hume
- Region
- Scotland and France
- Language
- English
Where it stands in the argument
- Position
- Exchange — Both: answers an earlier work and drew a later one
- Form
- Treatise
- Method
- Analytic — Clarification of language and logical form
- Transmission
- Continuous — Copied and read without a break in its own language
- Standing
- Live — Still argued in current philosophical work
- Years to the reply
- 25 years
- Years back to what it answers
- 26 years
Ignored on publication by Hume's own account, disowned by him in later life, and now the most cited book in modern philosophy after Kant's first Critique.
What the book says
Three books applying the experimental method to the mind itself. Everything in the mind is either an impression - a sensation or feeling as it occurs - or an idea, which is a fainter copy of one; and any term that cannot be traced back to an impression is empty. Hume then applies the test to the concepts philosophy relies on most, and they do not survive. Causation: watch one billiard ball strike another as often as you like and you see one motion followed by another, never a necessary connection, because there is no impression of one. What we call cause is constant conjunction plus a habit of expectation formed in us. Induction: no argument shows that the future must resemble the past, and reasoning that assumes it is circular. The self: look inward for the thing that has the perceptions and you find only more perceptions, never the owner. The second book treats the passions and argues that reason is and ought only to be the slave of them. The third turns to morals, argues that moral distinctions come from sentiment rather than reason, and observes that writers pass without warning from statements about what is to statements about what ought to be.
What it argues
That the mind is to be studied as a natural object, and that doing so shows how little of what we believe is founded on reason. The test is simple and ruthless: for any idea, find the impression it copies, and if there is none the idea is empty. Causation fails it, and with it the assumption that the world's regularities are underwritten by anything but our expectation. Induction fails it, since the only support for expecting the future to resemble the past is that it has done so, which is the thing in question. The self fails it. Morality fails a different test: no argument from facts alone yields an obligation, so moral judgement must rest on sentiment. What Hume offers instead of foundations is nature - we go on believing and expecting and judging because we are built to, and philosophy cannot stop us and should not try.
What it answered
Three Dialogues between Hylas and Philonous 1713George Berkeley
Berkeley directly and the whole rationalist project behind him. Hume's target is the assumption that reason establishes the framework we live by; his conclusion is that custom does. He is not a sceptic in the ancient sense, since he holds that the beliefs cannot be given up and need not be - a philosopher who doubts causation in the study plays backgammon afterwards and finds the doubts absurd. What he denies is that the beliefs have the rational credentials philosophy has claimed for them.
What answered it
An Inquiry into the Human Mind 1764Thomas Reid
Reid, in 1764, by rejecting the theory of ideas outright; Kant, who said Hume woke him from dogmatic slumber and whose entire critical philosophy is an attempt to secure what Hume had shown reason could not; and, in the twentieth century, the whole literature on induction, on causation, on personal identity and on the is-ought distinction, all of which cite him as the origin of the problem they are addressing. Very few of these count as refutations. The usual pattern is to accept that Hume identified something real and to argue about what follows.
How it argues
By a method he calls experimental, meaning that it starts from observation of how minds actually work rather than from what reason requires. The prose is clear and unhurried and the arguments are short; what makes the book hard is the accumulation, since each result depends on the copy principle established at the start. Hume is also willing to record where the argument leaves him stranded - the conclusion of book one describes a genuine crisis, philosophical despair relieved by dinner and company, and the appendix admits that his account of personal identity does not work and that he cannot repair it.
The moment
Written mostly in France, at La Flèche, where Descartes had been educated, by a Scot in his mid-twenties living cheaply on a small income after a breakdown he described to a physician in a remarkable letter. He published anonymously, expected a scandal, and got silence - the book, he wrote later, fell dead-born from the press. He never held a university post; he was rejected at Edinburgh and again at Glasgow on grounds of irreligion, and made his living as a librarian, tutor, secretary and finally as a very successful historian.
How it travelled
Slowly, then completely. The Treatise failed; the Enquiries and the Essays made Hume famous, and the History of England made him rich. The German reception through Kant put the Treatise's arguments at the centre of European philosophy even though Kant probably read the Enquiry rather than the Treatise itself. Nineteenth-century Britain treated Hume as a sceptic to be answered; twentieth-century philosophy treated him as a founder and rediscovered the Treatise as the fuller statement, which reversed his own judgement on his own book.
The strongest objection
Reid's, and it is not the crude one it is sometimes taken for. Reid does not appeal to common sense as a way of refusing to think; he argues that Hume's conclusions follow validly from a premise nobody has established - that the immediate objects of the mind are ideas rather than things - and that a chain of valid reasoning ending in absurdity is a reason to reject the premise. Since Descartes, Locke and Berkeley all share it, the whole modern tradition may have been reasoning from an assumption it never examined. Hume has no answer to this beyond the theory itself, and the appendix's admission that he cannot make personal identity work is evidence for Reid rather than against him.
Why read it
Book one, part three on causation, and the sections on personal identity and on scepticism at the end of book one, which together are the philosophical core. The conclusion of book one is also one of the few passages in philosophy where a writer describes what his own argument has done to him. Read the Abstract first; it is twenty pages and Hume wrote it because he knew the book was hard.
- First printed
- 1739
- First in English
- 1739
- Digitised
- Yes
An anonymous Abstract of 1740, now known to be Hume's own summary, explains the argument about causation more clearly than the book does. The Enquiries of 1748 and 1751 restate the positions Hume wanted to be judged by.
Written in English. All editions are public domain.
Where it stands
Live in a way almost nothing else here is: the problem of induction, the regularity theory of causation, the bundle theory of the self, and the is-ought gap are all named after positions in this book and are all argued about now.
The passage
Reason is, and ought only to be the slave of the passions
A Treatise of Human Nature, book II part iii section 3. Public domain, and Hume's own words.
Contested
Whether Hume is a sceptic, a naturalist, or a sceptical realist who thinks there are real necessary connections we cannot know is the central interpretive dispute and has been for fifty years. Whether the appendix's retraction concerns personal identity or something deeper is unresolved because Hume does not say. How far the Enquiries supersede the Treatise, as Hume asked, or simplify it, as most modern readers hold, is argued. And whether the is-ought passage means what it is universally taken to mean is disputed by a minority who read it in its immediate context rather than as a general principle.