Hume Texts Online


Of the Association of Ideas.

E 3.1, SBN 23

IT is evident, that there is a principle of connexion between the different thoughts or ideas of the mind, and that, in their appearance to the memory or imagination, they introduce each other with a certain degree of method and regularity. In our more serious thinking or discourse, this is so observable, that any particular thought, which breaks in upon the regular tract or chain of ideas, is immediately remarked and rejected. And even in our wildest and most wandering reveries, nay in our very dreams, we shall find, if we reflect, that the imagination ran not altogether at adventures, but that there was still a connexion upheld among the different ideas, which succeeded each other. Were the loosest and freest conversation to be transcribed, there would immediately be observed something, which connected it in all its transitions. Or where this is wanting, the person, who broke the thread of discourse, might still inform you, that there had secretly revolved in his mind a succession of thought, which had gradually led him from the subject of conversation. Among different languages, even where we cannot suspect the least connexion or communication, it is found, that the words, expressive of ideas, the most compounded, do yet nearly correspond to each other: A certain proof, that the simple ideas, comprehended in the compound ones, were bound together by some universal principle, which had an equal influence on all mankind.

E 3.2, SBN 24

Though it be too obvious to escape observation, that different ideas are connected together; I do not find, that any philosopher has attempted to enumerate or class all the principles of association; a subject, however, that seems worthy of curiosity. To me, there appear to be only three principles of connexion among ideas, namely, Resemblance, Contiguity in time or place, and Cause or Effect.

E 3.3, SBN 24

That these principles serve to connect ideas will not, I believe, be much doubted. A picture naturally leads our thoughts to the original[2]: The mention of one apartment in a building naturally introduces an enquiry or discourse concerning the others[3]: And if we think of a wound, we can scarcely forbear reflecting on the pain which follows it[4]. But that this enumeration is compleat, and that there are no other principles of association, except these, may be difficult to prove to the satisfaction of the reader, or even to a man's own satisfaction. All we can do, in such cases, is to run over several instances, and examine carefully the principle, which binds the different thoughts to each other, never stopping till we render the principle as general as possible[6]. The more instances we examine, and the more care we employ, the more assurance shall we acquire, that the enumeration, which we form from the whole, is compleat and entire.

E 3.n2, SBN 24


E 3.n3, SBN 24


E 3.n4

Cause and effect.

E 3.n6, SBN 24

For instance, Contrast or Contrariety is also a connexion among Ideas: But it may, perhaps, be considered as a mixture of Causation and Resemblance. Where two objects are contrary, the one destroys the other; that is, is the cause of its annihilation, and the idea of the annihilation of an object, implies the idea of its former existence.


Of the Association of Ideas.

This section originally continued significantly beyond its current ending, starting with an added final sentence in paragraph 3. Footnote 6 also occurred later on in this section, as shown here. The additional material varied between the editions from 1748 to 1772, but was deleted in the final 1777 edition. The text below is taken from the 1772 edition.
E 3.3v

[... is compleat and entire.] Instead of entering into a detail of this kind, which would lead into many useless subtilties, we shall consider some of the effects of this connexion upon the passions and imagination; where we may open a field of speculation more entertaining, and perhaps more instructive, than the other.

E 3.4

As a man is a reasonable being, and is continually in pursuit of happiness, which he hopes to attain by the gratification of some passion or affection, he seldom acts or speaks or thinks without a purpose and intention. He has still some object in view; and however improper the means may sometimes be, which he chuses for the attainment of his end, he never loses view of an end; nor will he so much as throw away his thoughts or reflections, where he hopes not to reap some satisfaction from them.

E 3.5

In all compositions of genius, therefore, it is requisite, that the writer have some plan or object; and though he may be hurried from this plan by the vehemence of thought, as in an ode, or drop it carelessly, as in an epistle or essay, there must appear some aim or intention, in his first setting out, if not in the composition of the whole work. A production without a design would resemble more the ravings of a madman, than the sober efforts of genius and learning.

E 3.6

As this rule admits of no exception, it follows, that, in narrative compositions, the events or actions, which the writer relates, must be connected together, by some bond or tye: They must be related to each other in the imagination, and form a kind of Unity, which may bring them under one plan or view, and which may be the object or end of the writer in his first undertaking.

E 3.7

This connecting principle among the several events, which form the subject of a poem or history, may be very different, according to the different designs of the poet or historian. Ovid has formed his plan upon the connecting principle of resemblance. Every fabulous transformation, produced by the miraculous power of the gods, falls within the compass of his work. There needs but this one circumstance in any event to bring it under his original plan or intention.

E 3.8

An annalist or historian, who should undertake to write the history of Europe during any century, would be influenced by the connexion of contiguity in time and place. All events, which happen in that portion of space and period of time, are comprehended in his design, though in other respects different and unconnected. They have still a species of unity, amidst all their diversity.

E 3.9

But the most usual species of connexion among the different events, which enter into any narrative composition, is that of cause and effect; while the historian traces the series of actions according to their natural order, remounts to their secret springs and principles, and delineates their most remote consequences. He chuses for his subject a certain portion of that great chain of events, which compose the history of mankind: Each link in this chain he endeavours to touch in his narration: Sometimes unavoidable ignorance renders all his attempts fruitless: Sometimes, he supplies by conjecture, what is wanting in knowledge: And always, he is sensible, that the more unbroken the chain is, which he presents to his reader, the more perfect is his production. He sees, that the knowledge of causes is not only the most satisfactory; this relation or connexion being the strongest of all others; but also the most instructive; since it is by this knowledge alone, we are enabled to controul events, and govern futurity.

E 3.10

Here therefore we may attain some notion of that Unity of Action, about which all critics, after Aristotle, have talked so much: Perhaps, to little purpose, while they directed not their taste or sentiment by the accuracy of philosophy. It appears, that, in all productions, as well as in the epic and tragic, there is a certain unity required, and that, on no occasion, can our thoughts be allowed to run at adventures, if we would produce a work, which will give any lasting entertainment to mankind. It appears also, that even a biographer, who should write the life of Achilles, would connect the events, by shewing their mutual dependence and relation, as much as a poet, who should make the anger of that hero, the subject of his narration[5]. Not only in any limited portion of life, a man's actions have a dependence on each other, but also during the whole period of his duration, from the cradle to the grave; nor is it possible to strike off one link, however minute, in this regular chain, without affecting the whole series of events, which follow. The unity of action, therefore, which is to be found in biography or history, differs from that of epic poetry, not in kind, but in degree. In epic poetry, the connexion among the events is more close and sensible: The narration is not carried on through such a length of time: And the actors hasten to some remarkable period, which satisfies the curiosity of the reader. This conduct of the epic poet depends on that particular situation of the Imagination and of the Passions, which is supposed in that production. The imagination, both of writer and reader, is more enlivened, and the passions more enflamed than in history, biography, or any species of narration, which confine themselves to strict truth and reality. Let us consider the effect of these two circumstances, an enlivened imagination and enflamed passion; circumstances, which belong to poetry, especially the epic kind, above any other species of composition: And let us examine the reason, why they require a stricter and closer unity in the fable.

E 3.11

First. All poetry, being a species of painting, brings us nearer to the objects than any other species of narration, throws a stronger light upon them, and delineates more distinctly those minute circumstances, which, though to the historian they seem superfluous, serve mightily to enliven the imagery, and gratify the fancy. If it be not necessary, as in the Iliad, to inform us each time the hero buckles his shoes, and ties his garters, it will be requisite, perhaps, to enter into a greater detail than in the Henriade; where the events are run over with such rapidity, that we scarcely have leisure to become acquainted with the scene or action. Were a poet, therefore, to comprehend in his subject, any great compass of time or series of events, and trace up the death of Hector to its remote causes, in the rape of Helen, or the judgment of Paris, he must draw out his poem to an immeasurable length, in order to fill this large canvas with just painting and imagery. The reader's imagination, enflamed with such a series of poetical descriptions, and his passions, agitated by a continual sympathy with the actors, must flag long before the period of the narration, and must sink into lassitude and disgust, from the repeated violence of the same movements.

E 3.12

Secondly. That an epic poet must not trace the causes to any great distance, will farther appear, if we consider another reason, which is drawn from a property of the passions still more remarkable and singular. It is evident, that, in a just composition, all the affections, excited by the different events, described and represented, add mutual force to each other; and that, while the heroes are all engaged in one common scene, and each action is strongly connected with the whole, the concern is continually awake, and the passions make an easy transition from one object to another. The strong connexion of the events, as it facilitates the passage of the thought or imagination from one to another, facilitates also the transfusion of the passions, and preserves the affections still in the same channel and direction. Our sympathy and concern for Eve, prepares the way for a like sympathy with Adam: The affection is preserved almost entire in the transition; and the mind seizes immediately the new object as strongly related to that which formerly engaged its attention. But were the poet to make a total digression from his subject, and introduce a new actor, nowise connected with the personages, the imagination, feeling a breach in the transition, would enter coldly into the new scene; would kindle by slow degrees; and in returning to the main subject of the poem, would pass, as it were, upon foreign ground, and have its concern to excite anew, in order to take part with the principal actors. The same inconvenience follows in a less degree; where the poet traces his events to too great a distance, and binds together actions, which, though not entirely disjoined, have not so strong a connexion as is requisite to forward the transition of the passions. Hence arises the artifice of the oblique narration, employed in the Odyssey and Aeneid; where the hero is introduced, at first, near the period of his designs, and afterwards shows us, as it were in perspective, the more distant events and causes. By this means, the reader's curiosity is immediately excited: The events follow with rapidity, and in a very close connexion: And the concern is preserved alive, and, by means of the near relation of the objects, continually increases, from the beginning to the end of the narration.

E 3.13

The same rule takes place in dramatic poetry; nor is it ever permitted, in a regular composition, to introduce an actor, who has no connexion, or but a small one, with the principal personages of the fable. The spectator's concern must not be diverted by any scenes disjoined and separated from the rest. This breaks the course of the passions, and prevents that communication of the several emotions, by which one scene adds force to another, and transfuses the pity and terror, which it excites, upon each succeeding scene, till the whole produces that rapidity of movement, which is peculiar to the theatre. How must it extinguish this warmth of affection, to be entertained, on a sudden, with a new action and new personages, no wise related to the former; to find so sensible a breach or vacuity in the course of the passions, by means of this breach in the connexion of ideas; and instead of carrying the sympathy of one scene into the following, to be obliged, every moment, to excite a new concern, and take part in a new scene of action?

E 3.14

To return to the comparison of history and epic poetry, we may conclude, from the foregoing reasonings, that, as a certain unity is requisite in all productions, it cannot be wanting in history more than in any other; that, in history, the connexion among the several events, which unites them into one body, is the relation of cause and effect, the same which takes place in epic poetry; and that, in the latter composition, this connexion is only required to be closer and more sensible, on account of the lively imagination and strong passions, which must be touched by the poet in his narration. The Peloponnesian war is a proper subject for history, the siege of Athens for an epic poem, and the death of Alcibiades for a tragedy.

E 3.15

As the difference, therefore, between history and epic poetry consists only in the degrees of connexion, which bind together those several events, of which their subject is composed, it will be difficult, if not impossible, by words, to determine exactly the bounds, which separate them from each other. That is a matter of taste more than of reasoning; and perhaps, this unity may often be discovered in a subject, where, at first view, and from an abstract consideration, we should least expect to find it.

E 3.16

It is evident, that Homer, in the course of his narration, exceeds the first proposition of his subject; and that the anger of Achilles, which caused the death of Hector, is not the same with that which produced so many ills to the Greeks. But the strong connexion between these two movements, the quick transition from one to another, the contrast[6v] between the effects of concord and discord among the princes, and the natural curiosity which we have to see Achilles in action, after so long a repose; all these causes carry on the reader, and produce a sufficient unity in the subject.

E 3.17

It may be objected to Milton, that he has traced up his causes to too great a distance, and that the rebellion of the angels produces the fall of man by a train of events, which is both very long and very casual. Not to mention, that the creation of the world, which he has related at length, is no more the cause of that catastrophe, than of the battle of Pharsalia, or any other event, that has ever happened. But if we consider, on the other hand, that all these events, the rebellion of the angels, the creation of the world, and the fall of man, resemble each other, in being miraculous and out of the common course of nature; that they are supposed to be contiguous in time; and that being detached from all other events, and being the only original facts, which revelation discovers, they strike the eye at once, and naturally recall each other to the thought or imagination: If we consider all these circumstances, I say, we shall find, that these parts of the action have a sufficient unity to make them be comprehended in one fable or narration. To which we may add, that the rebellion of the angels and the fall of man have a peculiar resemblance, as being counterparts to each other, and presenting to the reader the same moral, of obedience to our Creator.

E 3.18

These loose hints I have thrown together, in order to excite the curiosity of philosophers, and beget a suspicion at least, if not a full persuasion, that this subject is very copious, and that many operations of the human mind depend on the connexion or association of ideas, which is here explained. Particularly, the sympathy between the passions and imagination will, perhaps, appear remarkable; while we observe that the affections, excited by one object, pass easily to another object connected with it; but transfuse themselves with difficulty, or not at all, along different objects, which have no manner of connexion together. By introducing, into any composition, personages and actions, foreign to each other, an injudicious author loses that communication of emotions, by which alone he can interest the heart, and raise the passions to their proper height and period. The full explication of this principle and all its consequences would lead us into reasonings too profound and too copious for this enquiry. It is sufficient, at present, to have established this conclusion, that the three connecting principles of all ideas are the relations of Resemblance, Contiguity, and Causation.

E 3.n5

5. Contrary to Aristotle, Μνθος δ' εστιν εις, ουχ, ωσπερ τινες οιονται, εαν περι θ' ενα η. Πολλα γαρ, και απειρα τω γενει Συμβαινει, εξ ων ενιων εξ εστιν εν. Ουτως δε και πραξεις, ενος πολλαι εσιν, εξ ων μια ουδεμια γινεται πραξις, &c. Κεφ. η.

E 3.n6v

6v. Contrast or contrariety is a connexion among ideas, which may, perhaps, be considered as a mixture of causation and resemblance. Where two objects are contrary, the one destroys the other, i.e. is the cause of its annihilation, and the idea of the annihilation of an object implies the idea of its former existence.

An Enquiry concerning Human Understanding (1748, 1777)

prepared by Peter Millican

What is now known as Hume’s first Enquiry was first published in 1748, by Andrew Millar of the Strand, London, under the title Philosophical Essays Concerning Human Understanding. It is described in My Own Life as a recasting of the “first part” of the Treatise, though it is not confined to topics from Treatise Book 1: the long Section 8 “Of Liberty and Necessity” reworks Treatise 2.3.1-2 (and incorporates T as E 8.13), while Sections 11 and 12, on Miracles and the Design Argument, have no precedent. Perhaps it is best thought of as a presentation of the core—and some of the more striking applications—of Hume’s inductive epistemology and causal metaphysics: the heart of his theoretical philosophy. It thus elaborates what the Abstract described as the “Chief Argument” of the Treatise, while adding an explicit focus on religious topics (which Hume had deleted from the Treatise for reasons of prudence). Although much shorter than the Treatise, and omitting most of the detailed associationist psychology (e.g. from the discussions of belief, probability, and the external world), the Enquiry expands the central philosophical discussions on induction (Section 4), free-will (Section 8), and scepticism (Section 12) while also polishing significantly the treatment of causation (Section 7). It is therefore an indispensable source for Hume’s epistemology and metaphysics, although most scholars have tended to ignore Hume’s own request—expressed in the “Advertisement” of 1775—that the Enquiry should be taken (together with the other works in his final edition of Essays and Treatises Volume 2) as the authoritative statement of his mature “philosophical sentiments and principles”, thus supplanting the Treatise.

A second edition of the Philosophical Essays appeared in 1750, and this was reprinted in 1751 and 1753, the latter in the form of volume II of Hume’s four-volume Essays and Treatises on Several Subjects. This arrangement was retained in the third edition of 1756, at which point volume II was the only one of the four to be reissued. The next edition of the Essays and Treatises, in 1758, combined the constituent works into a single volume, and here Hume permanently changed the title of his Philosophical Essays to An Enquiry Concerning Human Understanding, thus recognising its systematic nature alongside the Enquiry Concerning the Principles of Morals. In the four-volume 1760 and 1770 editions of the Essays and Treatises, the Enquiry appeared in volume III followed by A Dissertation on the Passions. In the two-volume editions of 1764, 1767, 1768, 1772, and 1777, it appeared at the beginning of volume II, followed in order by A Dissertation on the Passions, An Enquiry Concerning the Principles of Morals, and The Natural History of Religion.

Only in three cases have substantive changes been made to the copytext here, where typographical changes were clearly required and could be identified by reference to other editions: these involve the insertion of “[is]” within 3.3 n. 6, “[and]” within endnote [B], and the substitution of “reasoning” for “reasonings” in the first line of endnote [H].

Footnotes have all been numbered, as with all of the editions on this site. This leads to one small complication: Beauchamp’s note 5 from the 1772 edition is absent from the 1777 text followed here, but for ease of cross reference I have numbered the notes as Beauchamp does. Thus note 5 only appears in the Textual Variants tab in section 3.

Display Options

By default, the edited version of the text is shown (see the notes on the Edited Versions). When displaying the edited text, check the “Show Changes” option to see the editorial interventions we have made. Deletions will appear with a line through them, and additions underlined; by hovering your mouse over the change, you will see a brief explanation of the edit. Our interventions are minimal, consisting mostly of changes sanctioned by Hume himself, together with corrections of just a few very obvious errors.

Alternatively, you can deselect the “Edited Version” option to see the original edition, a faithful reproduction of the copytext save for some systematic and insignificant changes intended to make the text easier to read and navigate (see the notes on the Original Editions). Note that this affects the text against which your search queries are tested as well, though the option can also be set on the Search page.

Page numbers from standard editions are shown alongside each paragraph. For paragraphs that range over more than one page, page breaks can be displayed in the text as a pipe symbol (|), by checking “Show Page Breaks”.