Philosophical Connections: Hume

Born in Edinburgh, the son of a Scottish landowner, David Hume was educated at Edinburgh University, where he studied arts (a course which included the classics, philosophy, and some elementary science). He left without taking his degree. His family wished him to enter the legal profession. But finding both this and banking uncongenial he devoted himself to the study of philosophy. In 1734 he went to France and wrote his Treatise on Human Nature. This was, however, poorly received when it was published (anonymously) in 1739-1740. "It fell dead-born from the press", he said. After thorough revision it was republished in separate parts over a period of time some years later. In 1752 he was appointed librarian of the Faculty of Advocates in Edinburgh, but despite two applications failed to secure a university post. He subsequently worked as private secretary to the British Ambassador in Paris. He also achieved distinction as a historian and was the first great modern writer on the philosophy of religion. Hume was a popular figure in literary circles — in France he was called "le bon David", in Scotland "Saint David".

[Sources: References to Hume's Enquiries are either to the relevant sections or to the marginal numbers used in the 1974; revision by P. H. Nidditch of Selby-Bigge's edition of 1893 (which did not appear in Hume's original editions).]

KNOWLEDGE

[1] As a preliminary to his overall programme to establish "a science of human nature", that is, a 'psychology' [Treatise, Introd.] — as far as possible along the lines of Newtonian physics [a], he thinks of man as a being who acts and reasons; and he is concerned to discover the nature and limits of human knowledge. He is critical of what he calls 'antecedent' scepticism, exemplified by Descartes' procedure of methodological doubt. [See Treatise, Book I, iv and v; Enquiry on Human Understanding, XII.] While it is useful to be generally cautious in one's claims Hume thinks of this as a 'mitigated', moderate form of scepticism, to doubt everything in the 'excessive' manner of Pyrrho, and then to claim, say, the 'cogito' as providing an indisputable or infallible basis for knowledge is untenable. Instead Hume utilizes a 'consequent' scepticism which involves an examination of our intellectual and sensory faculties — though this too, as will be seen, has its own dangers if carried to the extreme. However, he starts the Treatise by tracing the origins of our knowledge and criticizing traditional metaphysics [see also sec. 2] [b].

Knowledge, he says, derives only from sense-experience. He first of all distinguishes between impressions and ideas — both of which are mental contents [Treatise, I, i, 1]. Impressions are sensations, passions, and emotions; the having of an impression constitutes perception. Ideas are faint copies or images of these in our thinking or reasoning. Impressions being 'immediate' to the consciousness are stronger and more vivid or lively than ideas. Hume thus rejects the innate ideas of rationalist philosophers (though impressions, but not ideas, may be accepted as innate in the trivial sense that they occur at and subsequent to our birth). He goes on to make more distinctions [I, i, 1 and 2]. (1) Impressions may be of (a) sensation, (b) reflection [c]; and although in general impressions precede ideas, Hume says that some ideas may produce further impressions and thence more ideas of reflection (as when, for example, the idea of cold as pain produces in us an aversion to it). He allows that we may have an idea of a colour, say a particular shade of blue, without our having a prior impression; we have the idea from our impressions of shades on either side in the colour series, and the gap is filled by the imagination. (2) Both impressions and ideas may be (a) simple (a patch of colour and our idea of it), or (b) complex (for example, a city — its streets and buildings). An idea is simple if it cannot be analysed into any component distinct ideas. "Every simple idea has a simple impression, which resembles it and every simple impression a correspondent idea." Copying of impressions is achieved by either the memory or the imagination [I, i, 3]. Through the former impressions give rise to further impressions, for example, of reflection; while imagination produces faint images. Memory also preserves the order, position, and connection of our ideas: but the imagination combines ideas in any order. This leads on to a discussion of relations, which Hume regarded as 'the work of the mind'. He uses the term in two senses [I, i, 4 and 5]. (1) Natural relations: these are qualities by which ideas are connected in our imagination through the uniting principle of association. (2) Philosophical relations: these involve comparisons between any objects as a result of our will or choice [d]. The natural relations are resemblance, contiguity in time or place, and cause and effect. There are, however, seven philosophical relations: resemblance, identity, spatio-temporal relations, proportion in quantity or number, degrees of quality, contrariety, and causation.

In I, i, 6 he examines the idea of substance. This is not derived from any particular impression, whether of sensation or reflection. Rather we should regard it as nothing but a collection of simple ideas united by the imagination and to which we assign a given name [e]. It is because of this that we commonly think of these qualities as referring to an unknown 'something' in which they inhere, or at least as being connected by the relations of contiguity and causation. He contrasts this with ideas of modes (he gives as examples the ideas of a dance and of beauty), which represent qualities not supposedly united by such relations. Hume's rejection of the metaphysical concept of substance is paralleled by his critique of abstract general ideas ('universals') [I, i, 7]. Ideas, being copies of impressions, must be definite and determinate, and must also be of a particular impression. What then gives rise to the general idea? We observe resemblances between objects which justify our giving them the same name. When we hear the word the imagination produces an appropriate image, and is also ready to produce other resembling images. In this way the particular images can become general in their representation [f].

All human reasonings or inquiries involve either 'knowledge' or 'probability'. In the Enquiry Concerning Human Understanding Hume discusses this in terms of the distinction between 'relations of ideas' and 'matters of fact' [IV, 1]. (this is often referred to as 'Hume's fork'.) Relations of ideas are typically mathematical or logical and cannot be denied without contradiction (for example, 3 x 5 = 1/2 of 30). Hume says they are discovered by intuition (in the case of resemblance, contrariety, and degrees of quality) or by demonstration, that is, deduction (in the case of proportion); and this gives us a priori knowledge. Matters of fact (for example, that the sun will rise tomorrow), on the other hand, are discovered by observation and non-demonstrative inference and can be denied without contradiction. They are known a posteriori [g]. Observation involves the philosophical relations of identity and contiguity; while non-demonstrative inference, which is involved in all reasoning beyond immediate impressions, is founded on the relation of causation. What is the basis of this relation? This is a central issue in Hume's theory of knowledge and belief.

It is clear, Hume says [Treatise, I, iii, 2-6.], that the idea of cause must be derived from a relation between objects; we cannot find any quality common to the impressions we call 'causes'. He argues, firstly, that causes and effects are usually contiguous, that is, physically adjacent in space, either immediately or mediately (as in a chain). Further, a cause is temporarily prior to its effects. Of even greater importance is the idea of necessary connection. But where does this idea come from? In the Treatise Hume resolves this question into two. (1) Why do we suppose that it is necessary that everything that comes to exist should have a cause? (2) Why do we believe that particular causes necessarily have particular effects? What kind of inference is involved? As to (1), the principle is discovered neither by intuition nor by demonstration, it is not a relation of ideas; there is no contradiction in denying that something can begin to exist without a cause. (2) The causal inference must be grounded in experience not in any intuitive knowledge of 'essences'. And our belief rests on what he calls constant conjunction of particular instances (and thus invokes memory). What Hume means is that we observe event A on a number of occasions as being both contiguous with and prior to event B, and call A the cause and B the effect. It is from observation of these repeated impressions that the idea of necessity arises through the activity of the imagination. It is, says Hume, nothing more than an internal tendency, produced by custom, that is, through association, to pass from one impression to the other. As a natural relation (and here Hume gives a psychological analysis), a cause is defined as "An object precedent and contiguous to another and so united with it that the idea of the one determines the mind to form the idea of the other, and the impression of the one to form a more lively idea of the other" [Treatise, I, iii, 14]. As a philosophical relation (a conceptual analysis), cause is defined in terms of the placing of all the objects resembling an object in a like relation of priority and contiguity to those objects resembling the object to which it is precedent and contiguous [Ibid.]. His formulation of this 'philosophical' definition is similar (though simpler) in the Enquiry [VI, 2 (para. 60)]: "an object. followed by another, and where all the objects similar to the first are followed by objects similar to the second". But he concludes with a different formulation — what in today's terminology is called a counterfactual conditional: "Or in other words where, if the first object had not been, the second never had existed". Hume also says that it is only so far as it is a natural relation, and produces a union among our ideas, that we are able to reason about it or draw inferences from it [Treatise, I, iii, 6]. He stresses further that there is only one kind of cause: he rejects the traditional distinction between the four causes — efficient, formal, material, and final [I, iii, 14]. (In the Enquiry on Human Understanding [VI] he distinguishes between regular cause-effect sequences, such as the production of motion by gravity, which is a 'universal law', and chance sequences, such as the purgative effect of rhubarb. But he says that the latter are not to be attributed to any irregularity in nature; rather we should suppose there are 'secret causes' in the structures which have prevented the operation. Our reasonings about causation and custom remain the same. Hume's position is that if there are any causes in 'reality' we can have no knowledge of them) [h].

Hume also examines the ideas of space, time, and existence. Rejecting the notion of infinite divisibility of impressions and ideas (there are only minima of visible and tangible points, he says, beyond which no further division can be imagined) [Enquiry 124], he thinks of space and time (and thence the ideas of space and time) as the ways in which bodies are revealed to our senses [i] — as adjacent, through visual and tactile sensation, and as succession of perceptions in general [Treatise I, iii. 1-5]. As for existence, he argues that the idea of existence (and of external existence) is the very same as the idea of that which we conceive to be existent [I, ii. 6] [j]. This, for example, when we affirm God exists we are merely forming the idea of such a being [I, iii, 7].

[2] Hume distinguishes between knowledge and belief. As we have seen, knowledge is confined to relations of ideas, which although certain (by virtue of the logical connection between those ideas) tell us nothing about the world, and to those matters of fact, which are grounded in impressions (for example, 'This book is red') and are only probable (though it is not easy to be mistaken about them). Belief, on the other hand, is defined as "a lively idea related to or associated with a present impression" [I, iii, 7]. It is an act of the mind whereby assent is given to an idea (and Hume says that it differs from a 'fiction' only in the manner in which the idea is conceived). He also argues that judgement and reasoning (both of which involve the separating or uniting of different ideas — reasoning involving additionally the interposition of other ideas) are resolvable into conception, which is the "simple survey of one or more ideas" [ibid.]. We can thus perhaps say that the 'content' of judgements is the same as that of conceptions [a]. Belief is a consequence of habit; and it comes into play when we seek to pass beyond immediate experience. This would seem to give rise to a difficulty with causation, not least because Hume says [Enquiry, IV, 22] that all 'reasonings' (that is, inferences, including 'inductive' arguments) concerning matters of fact are founded on the relation of cause and effect. In appealing to our memory of constant conjunctions of past instances Hume presupposes that future instances — of which we obviously have no experience — will resemble those which we have experienced; in other words, that the course of nature is uniform [Treatise, I, iii, 6]. This assumption cannot be proved by demonstrative inference; no relation of ideas is involved. Nor can it be derived from probability, because although it involves appeal to experience it is itself founded on the presumption of a principle of the uniformity of resemblances between past and future objects [b]. Nevertheless, says Hume, we have to accept it, believe it, if we are to pursue our lives as rational agents. Similar considerations apply to his account of external bodies, minds, and God.

External bodies. [Treatise, I, iv, 2.] Given Hume's account of knowledge, it would seem that we can know nothing of objects existing externally to and independently of minds. All our perceptions, both of primary and secondary qualities, are in some sense 'internal' or within ourselves, and our knowledge is therefore confined to collections of impressions and ideas. Nevertheless, we have a natural belief in the continuous existence of objects; and since this belief is due neither to our senses nor to the reason, it must again be attributed to the imagination. And Hume draws attention to (a) the constancy, and (b) the coherence of some of our impressions as characteristics of our experience upon which the imagination can operate. Even when a sequence of impressions is interrupted in space and time, both imagination and memory step in, as it were, to fill the gaps, thereby supporting our belief in continuous existence and in the identity of particular bodies [c].

Minds. [I, iv, 5.] As we can have no coherent idea of substance (material or immaterial) in which our extended perceptions can inhere, clearly we can have no knowledge of any mental 'substance' or 'soul'. Furthermore, to regard my impressions and ideas as modifications of a soul 'substance' (as the 'theologians' do) must lead to atheism, because we cannot distinguish between perceptions and objects — which, according to Spinoza's "hideous hypothesis", are but modifications of the one substance. So what can be said of personal identity, a permanent 'self'? Hume says [I, iv, 6] that when he looks within himself all he encounters are sets of perceptions which appear, glide, pass away as if in a kind of theatre. Once more it is our memory which produces chains of images of past perceptions and enables us to become aware of causal connections. The imagination works on these sequences and gives rise to belief in continuity and identity However, Hume maintained that perceptions are "distinct existences" and that the mind never perceives any real connections between them. It also follows that while he did not reject the logical possibility of immortality, survival of a 'soul' could hardly be held consistently with his premisses [d].

As consequence of his philosophical assumptions and arguments it would seem that Hume has himself been led into an 'extreme' scepticism — despite his criticisms of Descartes. And indeed, in the Treatise [I, iv, 7], beset by doubts and uncertainties, he fancies himself "in the most deplorable condition imaginable, environed with the deepest darkness, and utterly deprived of every member and faculty". These clouds are dispelled only after he has given up his philosophizing and has sought amusement with his friends — conversation, a game of backgammon, after which he no longer has the heart to return to speculations which now seem "so cold, strained, and ridiculous". However, in the Enquiry [117ff.] his 'consequent' scepticism is more moderate. He recognises that, although we can never be immune from error, at some point we must make a stand, accept what is in the last analysis only probable — the existence of the external world, a continuous self, causation; for this is essential for our orderly day-to-day activities. Scepticism is beneficial to the extent that it encourages the cultivation of a non-dogmatic, critical stance [e]. Because our commonsense beliefs cannot be justified rationally in a strict sense it does not mean they are irrational in a narrow sense — in the way that superstitions and prejudices are. But what then is the difference between sensible and absurd beliefs? Hume [Treatise, I, iii, 7-9] differentiates between genuine belief and fancy. In genuine belief the idea assented to has a different 'feel' from that accompanying a fanciful or fictitious idea. It is characterized by force, vivacity, firmness, or steadiness. Genuine belief is thus an act of the 'sensitive' rather than the 'cogitative' part of our nature [1, iv, 11]. And he says that they may have either natural causes, involving causal relations and uniformities, or artificial.causes, arising from our education. However, either source may also produce irrational beliefs or fancies. Ultimately we must rely on our senses — faith in which seems to be a natural, blind and powerful instinct [Enquiry 118] — to test and support our reasonings.

ETHICS

[3] How do we discover the foundations of morals? Is it through our reason or through sentiment (feeling)? Arguments in favour of either are so plausible, Hume says [Enquiry concerning the Principles of Morals, section 1], that both may "concur in almost all moral determinations and conclusions". He therefore proposes to follow a simple method. This will consist of an analysis of "that complication of mental qualities, which form what, in common life, we call Personal Merit". What he intends to do in morals (which is for him an aspect of the wider study of man or human nature) is in fact comparable to what Newton did for physics: to follow the "experimental method, and deduce general maxims from a comparison of particular instances". We should consider only those arguments which are derived from experience, and "reject every system of ethics, however subtle or ingenious, which is not founded on fact or observation" .

He starts his examination of human nature by identifying the two social virtues, Benevolence and Justice. By benevolence [sec. II] Hume means "natural philanthropy" [V, 21] or "a feeling for the happiness of mankind and a resentment of their misery" [I, 1]. This affection for humanity is manifested in such qualities as mercy, sociability, generosity, and so on [II, 1]. In II, 2 he argues that part of the merit of such a person's actions lies in their utility, that is, other people derive happiness from them in so far as they offer love or friendship, or provide for those in need. Such utility also explains why such actions are so universally approved. Hume regards the "public utility" or "the true interests of mankind" as the primary means by which we may determine our duty [a]. And he says that many actions or life-styles which appear at first sight to be praiseworthy or reprehensible may subsequently turn out to deserve the opposite description when experience reveals the true consequences. Similar considerations apply to Justice, by which he means, roughly, the arrangements for determining the possession of the goods or property which will ensure an individual's happiness: "anything which it is lawful for him, and for him alone to use" [III, 3]. But in III, 1 Hume proceeds to show that, in the case of this social virtue, utility is the sole origin. In support of his claim he points out that there could be no place for justice in extreme situations such as either (a) a "golden age", when mankind lacked for nothing and lived in perfect harmony and tolerance, or (b) (with reference to Hobbes) a "state of nature" characterized by want, ignorance and savagery. So the rules of justice must depend on the condition men are placed in and on the 'public utility' that follows from their observance. Justice is an 'artificial' and conventional not a 'natural' virtue, the rules of which relate to the preservation of peace and order in society — and thereby to the public interest. [Treatise III, ii, 1] [b]. The ideas of property thus become necessary in a society which operates between extremes; and hence arise the usefulness, merit and moral obligation of justice. To support this account of justice and public utility Hume discusses [Enquiry III, 2; IV] particular laws both within a given state and between nations, and the rules or conventions which hold between individuals in matters of friendship, etiquette, and so on. There must even be honour among thieves if their "pernicious confederacy" is to be maintained [IV].

Hume now [V] raises the important question why we approve of the social virtues on account of their utility. What alternative accounts can be given of the origin of moral distinctions? They cannot all have arisen from education; such descriptions as 'honourable', 'shameful', 'lovely', 'odious' must have had their source in the "original constitution of the mind", if they were to be intelligible [V, 1 ]. Neither could morality be grounded in self-love or private interest; "the voice of nature and experience seems plainly to oppose the selfish theory" [ibid]. Moreover, he says, we often praise actions in other places or times which could not be remotely relevant to our self-interest. Sometimes we even approve of the actions of an adversary which could be contrary to our interests. Nevertheless, he argues [V, 2] that the interest of each individual (self-love) cannot be divorced from his concern for the general interest of the community [c].

Hume's position may still seem a little unclear. He has rejected self-love as the basis of morality but has stressed the interdependence of the individual's self-interest and that of society. And yet we may still approve of the actions of others even when they conflict with our interests. How does utility relate to the self? Why does public utility, as he puts it. "please"? His answer is to appeal to the notion of sympathy. What he means by this is explained in the Treatise on Human Nature [II, i, 11] in terms of the association of ideas. But in the Enquiry [V, 2] he thinks of sympathy as arising directly from a capacity we all possess of putting ourselves, by means of our imagination, in the place of another person and of praising or blaming him for exhibiting qualities which would arouse in us pride or humiliation respectively if we possessed them. In other words through sympathy we experience the sentiments of humanity and benevolence [d]. A man, says Hume [V, 1], cannot be indifferent to the happiness or misery of his fellow beings. Whatever promotes their happiness is good, what tends to their misery is evil. And we discover from our experience that utility is in all circumstances a source of approval and a foundation for morals [e]. Not surprisingly, Hume argues in section VI that the sentiment of humanity and the moral sentiment are originally the same because they are governed by the same laws, and are moved by the same objects. Furthermore, the fact that we approve of such qualities as temperance, patience, presence of mind, and so on, which serve the possessor alone without claim to any public value, cannot be attributed to any theory of self-love on our part, but rather supports a doctrine of disinterested benevolence [f] which ensures that there is no incompatibility in the community between morality and utility [see also Appendix II]. In VI, 2 Hume seeks to support his theory further by reference to our regard for "bodily endowments" and the "goods of fortune". And in Section VII he examines qualities which appear to be valued for the immediate pleasure they bring to their possessor rather than for their utility: but he argues that in all such instances social sympathy operates and that there is therefore no inconsistency with his general theory. Similar considerations apply to qualities "immediately agreeable to others" [VIII]. In section IX Hume summarizes his theory and also affirms that the individual's 'interested obligation' to virtue is a consequence of his self-interest — his regard for his own happiness in the practice of his moral duty. "The sole trouble which [virtue] demands, is that of just calculation, and a steady preference of the greater happiness" [228].

Hume can now [Appendix 1] return to the question raised at the beginning of the Enquiry, namely whether the foundation of morals is to be sought in reason or sentiment. He recognises that reason has a role to play in assessing the consequences of actions and determining their utility, but he asserts that it is through sentiment, or 'moral sense', that we gain insight into morality itself [g]. In support of his view that reason cannot be the sole source of morals he offers five "considerations".

(1) Reason, he says, can judge either of matters of fact or of relations. But in the case of certain 'crimes', for example ingratitude, it is the sentiment that determines their immorality. Morality cannot consist in the relation of actions to rules; to determine the "rule of right" reason would have to start from a consideration of those very relations themselves. 'Virtue' is thus defined as "whatever mental action or quality gives to a spectator the pleasing sentiment of approbation" (and 'vice' the contrary) [h].

(2) There is a distinction in method or procedure between 'speculative' reasoning and moral deliberations. In the former we consider what is known and infer from it something which was previously unknown, whereas in the case of the latter all the objects and their relations must be known so that we base our approbation or blame on the total situation.

(3) Moral beauty can be compared with natural beauty; in our apprehension of both, approval (or disapproval) arise from contemplation of the whole — and through the sentiments rather than by the intellectual faculties.

(4) If morality consisted merely in relations it would apply as much to inanimate objects as it does to moral agents.

(5) The ultimate ends of human actions can never be accounted for by reason. If you ask someone why he exercises he will say it is because he desires health, sickness is painful, and he hates pain. What more is there to be said? There can be no infinite progression; "something must be desirable on its own account, and because of its immediate accord or agreement with human sentiment and affection".

The bounds of reason and taste are thus easily ascertained. "The former gives us knowledge of truth and falsehood: the latter gives the sentiment of beauty and deformity, vice and virtue". (He is critical of attempts to account for such sentiments by "metaphysical reasonings" or by deductions from abstract principles [Enquiry I].) It is only taste that can become a motive for action [i], in so far as it gives pleasure or pain and therefore happiness or misery."Cool and disengaged" reason can do no more than direct the impulse received from appetite or inclination. Hume is thus setting out a more moderate version of the assertion to be found in the Treatise [II, iii, 3]:

One further important point remains to be noted. As a corollary to his subordination of reason to sentiment as the means of discovering virtue and vice, Hume had already, earlier in the Treatise [III, i, 1], drawn attention to an observation which he thinks may be of "some importance". In every system of morality he has encountered he has found that the author has moved imperceptibly from assertions about what 'is' or 'is not' to claims about what 'ought' or 'ought not' be done. It seems altogether inconceivable, he says, how this new relation can be a deduction from others of a different kind. Hume's point is that in so far as moral distinctions are derived from a 'moral sense' — one's own particular sentiment — they do not involve inferences of reason. The distinction of vice and virtue is neither founded on relations nor is perceived by reason. There can therefore be no inference from an 'is' to an 'ought' [j].

POLITICAL PHILOSOPHY

[4] Hume rejects the idea of an original 'state of nature' as being a fiction [a], arguing that primitive people join together, initially in sexual relationships, and then in larger societies, as they become aware of the value of community in providing remedies for the 'inconveniences' of life. Society enables them to increase their power and abilities, gives them security against chance adversities, and guarantees their possession of property. Utility is thus the fundamental concept in his political philosophy [ III, ii, 2]. Utility is also the basis of government: government is successful to the extent that it can establish and administer justice and run schemes for the common good [III, ii, 7] [b]. Nevertheless Hume says that society can exist without government (as American Indian tribes show) [ii, 8]. How then did government originate? Hume is not certain, but he says it may well have arisen out of inter-tribal conflict; it has the advantage of preserving peace and order. The idea of a formal 'social contract' is another fiction [c]. Allegiance to a government lies again in its utility [ii, 8-9]. But Hume is not concerned with any quest for an ideal or utopian society such as Plato's or More's. And generally he regards it as irrelevant whether authority in the state is sanctioned by long-term possession, conquest, or right — whether 'divine', inherited, or empowered by government [ii, 10; see also Idea of a Perfect Commonwealth; Of the Original Contract]. Finally, similar considerations of utility determine 'laws of nations' and inter-state relationships [ii, 11]. But he adds that the natural obligation to justice between nations is less strong that it is between individuals within a society; for it is the latter that the essentials of life are preserved. And he allows degrees of strength in moral obligation as between 'princes or ministers' and 'private' persons, but not a different morality altogether.

'METAPHYSICS'/ PHILOSOPHY OF RELIGION

[5] While it is arguable that 'metaphysical' assumptions underlie Hume's empiricism, he is an uncompromising critic of traditional metaphysics:

Not surprisingly, he is generally dismissive of supposed proofs for the existence of God, though he does not reject the possibility out of hand that the universe is attributable to some ultimate intelligence. We cannot prove that God exists from any a priori reasoning (such as variations of what we now call the Cosmological Argument). [See Enquiry, XI; Dialogues concerning Natural Religion.] We may attempt to argue from nature to God as an intelligent designer, a 'workman', but such a being lies beyond our experience [a]; and even if such an inference were valid, it could tell us no more about the cause than what is already known as effects in nature. Still less could it provide us with any expectation about the future course of events in nature or guidance for our own choices. The hypothesis of a God is therefore useless. He likewise rejects any argument which appeals to so-called miracles [Enquiry, X; Dialogues]. These are by definition 'extraordinary' occurrences, which are, in contrast to the 'ordinary' events of our normal experience, improbable. Indeed, they are breaches of laws of nature [b]. If we were to accept an event as being miraculous, strong evidence would be required [Dialogues, 113]. This means that the historical evidence adduced in favour of the alleged miraculous event as having occurred would have to outweigh the totality of the evidence available to us — based on our experience — which leads us to suppose that such events could not have happened [ibid., 127]. A miracle could be accepted only if it were even more miraculous that the evidence for it was false. Neither can we accept alleged miracles as a basis for religion, because firstly there is a greater probability that stories of miracles have been fabricated; and secondly all religions invoke miracles. But if not all religions are true, then clearly most of the supposed miraculous occurrences must be false. It might be said that men have a natural belief in God. But Hume argues that the evidence does not support this claim. Many primitive societies were animistic or polytheistic — and indeed were often for all intents and purposes atheistic. Hume dismisses the dogmas and beliefs of Christians as likewise so much superstition. What 'true' religion there is must concerned with no more than the recognition that any cause of order in the universe "probably bears some remote analogy to human intelligence". Only thus far would religious belief seem to be acceptable and God's existence in some limited sense be feasible. Otherwise he is severely critical of religion; it not only leads to fanaticism but also is actually harmful to morality in so far as religious people, he thinks, tend to act for reasons other than for the sake of virtue.

Another argument brings together the problems of human freedom and evil [Treatise, II, iii, 1,2; Enquiry, VIII]. Hume has argued that through the activity of the imagination we suppose the causal relation to be 'necessary' [see sec. 1]. But he now says that when we act we do so from choice. However, this does not mean our actions are uncaused, the products of chance (what he calls "liberty of indifference"). Rather he affirms "liberty of spontaneity", by which he means that we act freely without external constraint but as a consequence of our motives and in the light of our knowledge. Thus free actions are still caused, but caused by ourselves as agents [c]. Now if there is a God He must be the cause of our actions, and therefore either He must be responsible for moral evil or there are no evil actions. However, we can account for these in terms of human sentiment; our feeling determines what is good or bad. As for God being the cause of moral evil, Hume rejects the possibility as being beyond human reason [d]; it is, he says, perhaps ironically, a "sublime mystery". [See VIII, 2.]

CRITICAL SUMMARY

Hume's philosophy is a through-going empiricism; he has eliminated all rationalist or theistic elements still remaining in the thought of his immediate empiricist predecessors. Commentators disagree, however, as to how he should be interpreted. He has usually been regarded as promoting a scepticism which Kant, in particular sought to overcome. But it has been argued recently that he is more accurately to be seen as a 'naturalist' seeking to account for all our beliefs, about ourselves as well as the world, in terms of basic principles of human nature. There is probably no serious inconsistency between these two views. While seemingly depressed in the Treatise by the cul de sac into which his philosophical arguments have driven him, in the Enquiry concerning Human Understanding he clearly distinguishes between "excessive" (Pyrrhonian) scepticism and a more "mitigated scepticism" resulting from the correction of the former by common sense and reflection. The so-called 'gap' between evidence and conclusions can for all practical or everyday purposes be ignored. Nevertheless, there is a great deal in his philosophy which continues to be the subject of debate.

(1) His fundamental distinction between impression and ideas in terms of their quality is not as obvious as might be supposed. The criterion of vivacity is psychological and subjective, and not always supported by our experience. Moreover, he does not really consider the precise role played by the mind. Can we have bare sensations from which we derive ideas and beliefs? Some philosophers argue that sensation is inseparable from belief, both being interpenetrating aspects of perception as a whole. The view that 'private' sensibilia can provide a basis for knowledge (if Hume's thesis can be interpreted in this way) has also been criticized by most philosophers working within the 'ordinary language' analytic tradition.

(2) Truths of reason provide us with certain knowledge (though not about the world). Matters of fact give us only probable knowledge. Does this differ from belief? Hume's accounts of both probability and belief are again psychological. He talks of probability as a stronger belief opposed by a weaker and incompatible one; while belief is described in terms of the 'strength' and 'liveliness' of ideas arising from their association with present impressions. It is not clear how this process of association works. Do we have an innate propensity for it, or does it itself arise from experience?

(3) Probable knowledge is dependent on causation. Hume's analysis of this concept is particularly controversial. Are his two definitions (three if one includes what is in effect an appeal to counterfactual conditionals) compatible? The first seems to be a realist definition, the second is psychological and is consistent with there being no causation in nature at all. Either way Hume's analyses do not satisfactorily distinguish between 'genuine' causal and 'coincidental' regularities. As for the underlying principle of uniformity of nature, on Hume's thesis this has to be accepted pragmatically.

(4) Hume's account of personal identity might be criticized for providing an inadequate criterion of individuation, sameness, and continuity.

(5) The central concepts of Hume's moral philosophy have also given rise to much discussion. (i) Is benevolence approved of for itself, because we find it pleasing, or for its utility (which is determined by reason although our approval stems from sentiment)? The test of justice is solely utility, but it has been claimed that his interpretation of justice in terms of property is narrow, and that the two virtues taken together fail to reconcile private and public interest satisfactorily. It has also been suggested that Hume invokes utility for purely explanatory purposes rather than regarding it as a deontological principle on the basis of which one might decide which action one should pursue. (ii) While Hume is in effect critical of attempts to derive value judgements from factual premisses, it has been objected that in his own ethics he does just that in so far as he endows utility with moral worth. Against this it might be said that he nevertheless locates our perception of duty in sympathy, which is a sentiment. There is also a problem concerning his concept of 'interested obligation' in the light of his view that any move from an 'is' statement to an 'ought' statement is illegitimate. However, it can be argued that he is claiming no more than that it is reasonable for human beings possessing in common the sentiment of benevolence and a concern for self-interest to act accordingly to promote the greater happiness: this is what the rational man of sentiment can be expected to do if he is to be true to himself (as we might put it). 'Obligation' is thus to be understood as an expression of the individual's intrinsic humanity and integrity rather than an absolute moral imperative. Nevertheless this remains contentious.

(6) His criticisms of religion and, in particular, miracles have made a considerable impact on later thought. The main difficulty with his account, however, lies with the concept of evidence — how it is to be measured and assessed. Hume is saying that miracles are improbable not logically impossible. He might have added that what seems to us at a particular moment in history to be in breach of the laws of nature might at some time in the future be accommodated within the framework of scientific explanation.

READING

Hume: A Treatise on Human Nature (1739-40), ed. P. H. Nidditch; Enquiries concerning the Human Understanding (1748), and concerning the Principles of Morals (1751), ed. Nidditch; The Natural History of Religion (1757) and Dialogues concerning Natural Religion (1779), ed. in one volume by J. C. A. Gaskin. There are also Penguin editions of the Treatise and Dialogues.

Studies:

Introductory

A. J. Ayer, Hume.

Advanced

R. J. Fogelin, Hume's Skepticism.

J. C. A. Gaskin, Hume's Philosophy of Religion.

D. F. Pears, Hume's System.

G. Strawson, The Secret Connexion: Realism and David Hume.

B. Stroud, Hume.

Collections of essays

V. C. Chappell (ed.), Hume.

D. F. Norton (ed.), The Cambridge Companion to Hume.

CONNECTIONS

Hume

Note: (1) Hume's 'moral philosophy' involves the extension of Newtonian methodology to man. (2) He rejected all traditional metaphysics [see sec. 5]. In the following connections this is illustrated primarily by reference to Descartes (whose Meditations he had read.). (3) His criticisms of arguments for God's existence were directed primarily against the formulations of the Deist thinker Samuel Clarke (1675-1729), but they also constitute an implicit critique of the original arguments as presented by such thinkers as Anselm and Aquinas, as well as by their successors (references to Descartes, Locke, and Berkeley are therefore included below). (4) Hume's understanding of Pyrrhonic scepticism was probably acquired through a reading of Bayle's Dictionary (Pierre Bayle, 1647-1706, was himself an influential sceptic).

Philosophical Connections

Compiled by Anthony Harrison-Barbet

HUME