John Mackie describes Hume’s analysis of the causal relation as ‘[t]he most significant and influential single contribution to the theory of causation’,1 yet there is surprisingly little general agreement among philosophers as to how it is to be understood. Few question the importance of Hume’s analysis, and the depth of its influence on later developments in philosophy is hard to deny, but even those sympathetic to the general drift of his account differ widely in their appreciation of its aims and merits. Hume’s long, digressive, and, on occasion, somewhat careless exposition of his theory is brought to a head in one of the most difficult and obscure passages of his philosophy. The problem of how to resolve its apparent and oft-noted inconsistency has inspired a large and controversial secondary literature, much of which, regrettably, either ignores or misrepresents Hume’s intentions in order to extract from his argument answers to philosophical questions which he almost certainly did not mean to address. Hume’s account of causality begins with a search for the impression from which the idea of necessity arises and ends, notoriously, some ninety-five pages later, with his two non-equivalent ‘definitions’ of causation. These pages are among the best-known and most influential in early modem philosophy. In summation of them, Hume promises to ‘collect all the different parts of this reasoning, and by joining them together form an exact definition of the relation of cause and effect, which makes the subject of the present enquiry’ (T.169). He defines cause, as a philosophical relation, as ‘[a]n object precedent and contiguous to another, and where all the objects resembling the former are plac’d in like relations of precedency to those objects, that resemble the latter’, and, again, as a natural relation, 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’ (T.170). Hume offers similar definitions in the Enquiry, appearing, once again, to identify causation with, on the one hand, constant conjunctions of phenomena, and, on the other, the association of ideas which is caused by these natural regularities and from which, according to Hume, the idea of necessity arises. The problem of how to reconcile these two disparate definitions of the same crucial concept is the central exegetical problem of Hume’s theory of causation.
Few commentators have believed that Hume succeeds in giving an ‘exact definition’ of causation. The main objections to his account are well-known. I will deal with some of them in the course of this chapter. For the moment, I note only that the definitions are neither directly equivalent, nor are they what philosophers call co-extensive.2 It is easily seen that they are not synonymous since the sums of qualities to which they refer are not identical. The definitions are not, in other words, intensionally equivalent. But it might still be argued that although the two definitions are not intensionally equivalent, they are, nevertheless, co-extensive, that is to say, that where one is satisfied the other is satisfied also. But it is clear that this is not so. There are evidently circumstances in which a pair of events could satisfy one of the definitions without satisfying the other. It is evident, for example, that not every instance of constant conjunction produces a psychological association of ideas. A constant conjunction of two events could satisfy the first definition without ever having been observed to do so, without, that is to say, satisfying the second definition. On the other hand, the second of the two definitions might equally well be satisfied without the first. Hume writes that we sometimes declare two objects to be causally related after only a few ‘experiments’, or, in some cases, on the basis of ‘one experiment’ of a kind (T.105). In cases like these, a psychological association results from an experience of the conjunction of objects, but the objects do not instantiate the kind of regularity that would satisfy the first definition. Hume, therefore, presents two definitions of the same important concept which are neither intensionally nor extensionally equivalent. There are objects which would be causes according to one definition, but not according to the other. However, according to Hume, the definitions differ only in presenting ‘different views of the same object’ (T. 169–70). It is apparent that they are intended to relate to the same concept and that Hume means to endorse them both. But how can both be correct as definitions of the same concept? As a good empiricist, Hume attempts to clarify our idea of the causal relation by considering the impressions which give rise to it. But as Hume’s two definitions refer to different groups of impressions, to which different ideas correspond, it seems obvious that they must define two distinct, different complex ideas. The difficulty of reconciling the conflicting strands of Hume’s theory has led commentators to treat it either as comprising two accounts of causation – an ontological account of the nature of causation ‘in the objects’ and a psychological account of the genesis of belief in causal connections – or as meaning to endorse as correct only one of the two definitions. In this chapter, I will argue that there are strong textual grounds for questioning the adequacy of both these approaches.
Commentators have responded in different ways to the challenge of making sense of these two definitions within a unified theory. Some have argued that only the first of the two definitions represents Hume’s real view of causation.3 The first definition has often been used to support the view that Hume was a regularity theorist about causation. On this interpretation, Hume argues that it is a mistake to include in the analysis of causation the relation of necessity. He presents instead the ontological thesis that causation is nothing but the constant conjunction of phenomena. In nature, Hume is supposed to have argued, ‘one thing just happens after another’.4 This view of causality has proved attractive to modem empiricists such as John Mackie. Hume has offered what might now be termed a reductive analysis of the meaning of causation, designed to show that propositions about causes can be satisfactorily analysed into ones about regularity of succession. The second definition can be read simply as a psychological reflection on causation so-defined. However, as I remarked earlier, this view does not sit comfortably with the text. Hume writes that, apart from contiguity and succession, ‘[t]here is a NECESSARY CONNEXION to be taken into consideration’ in defining the causal relation, and that, furthermore, this relation ‘is of much greater importance, than any of the other two above mentioned’ (T.77). He later writes that ‘[a]ccording to my definitions, necessity makes an essential part of causation’ (T.407). The obvious difficulty of squaring a reductive interpretation with these and other passages, in which Hume writes of ‘conceal’d’ and ‘secret’ powers in nature, has led some commentators to question the orthodox sceptical reading of Hume on causation. Among them are those philosophers who argue that Hume is a realist or ‘sceptical-realist’ about causes or causal connections.5 These commentators argue that, far from adopting a regularity theory of causation, Hume believes in ‘causal power or natural necessity’.6 Belief in objective causal connections has an important role to play in Hume’s philosophical outlook and an appreciation of this fact is central to the interpretation of his philosophy. Hume does not deny that there exist necessary connections between causes and effects or that we can form some conception of such connections. He means to say only that these connections are hidden from us and that the ideas we form of them are, at best, relative or non-specific.
Proponents of this new view face well known difficulties in reconciling their realist reading of Hume’s remarks about causal powers and forces with his theory of meaning, which seems to deny that talk of this sort can make sense. Such objections, while perhaps not fatal to the realist view, nevertheless represent compelling grounds for caution. One possible response is to deny that the empiricist theory of meaning has as important a role to play in Hume’s philosophy as has traditionally been thought. But, as we will see, this is a view which is difficult to sustain against the weight of textual evidence to the contrary. In this chapter, I argue for the rejection of both the orthodox Humean and the sceptical realist views of Hume on causation, and for their replacement with something which I believe is more sensitive to the breadth of Hume’s intentions in Part III. Hume’s account of the idea of necessity, of which the two definitions are the undoubted climax, emerges as part of his sustained study of the transition the human mind makes from observed to unobserved matters of fact. He is interested in the idea of necessity because of the role the idea had been thought to play in justifying causal inferences. Although Hume abandons the search for an impression which might serve in this foundational role, he continues to search for the origin of our idea of necessary connection. He goes on to show that the idea of necessity the mind possesses is the product of the causal inferences it makes. Hume presents his two definitions in summary of the evidence which has led him to conclude that our idea of causality arises not from any perception of necessity ‘in the objects’, but from an ‘unintelligible’ instinctive operation of the mind. His theory of causation cannot, therefore, be properly understood other than as the summation of the whole argument of Part III. It is only when taken out of this context that Hume’s theory appears to identify causation with mere regularity of succession. Within the context of his broader study of the causal relation, it is clear that Hume does not believe that constant conjunction is alone sufficient to give us an adequate idea of causality. It is essential, therefore, that we view Hume’s account of the idea of causation as part of his more general study of inference from experience.
Hume’s intention for Part III as a whole is to ‘explain fully’ the relation of cause and effect, and, that we might ‘reason justly’, to trace to its source the idea of causation ‘concerning which we reason’ (T.74). There follows a lengthy, and highly digressive, search for the causal origin of the idea of causation, and, in particular, the idea of necessity, which makes so significant a part of it. Our idea of causality is the idea of a complex relation, composed of succession, contiguity in space and time, and necessary connection. Hume makes clear that of these three relations it is the enigmatic relation of necessity that really interests him. This is because of the role which necessity had been thought to play in justifying inferences from cause to effect and assuring us of the existence of objects and events which we are not observing. Hume’s account of the idea of necessity builds on the argument concerning induction or probable reasoning which was our main concern in Chapter Three. I argued there that Part III, as a whole, is devoted to the question of the nature of our reasonings concerning matters of fact. Hume is interested in the causal relation because this is the relation by which the mind is engaged when it forms its conclusions concerning matters of fact. It is, he says, the only relation ‘that can be trac’d beyond our senses, and informs us of existences and objects, which we do not see or feel’ (T.74). Of the three natural relations or principles of association to which Hume refers, we find, by experience, that belief ‘arises only from causation, and that we can draw no inference from one object to another, except they be connected by this relation’ (T.107). He proposes to consider our idea of the causal relation. What, he asks, are the features apparent to us when we apply the concept in identifying a relation as causal? When we examine any instance of causal interaction we always find the relations of contiguity and succession to be present. But an object may be contiguous and prior to another without being considered its cause. There is another feature to be taken into consideration and this proves to be of greater importance than either of the other two. The crucially important relation is, of course, necessity or necessary connection. Contiguity and succession will never, by themselves, be enough to give us the idea of causation. However, taking any single instance of interaction we observe no necessary connection ‘in the objects’. All we see is one event followed by another. It is part of Hume’s negative argument in Section VI that we do not perceive any necessity in causal relationships which could provide the basis for inferences from experience.7 There is nothing in any object’ consider’d in itself, which can afford us a reason for drawing a conclusion beyond if (T.139). Early on, however, Hume speculates that it may, in the end, appear that ‘the necessary connexion depends on the inference, instead of the inference’s depending on the necessary connection’ (T.88). And, as we will see, this is precisely how things turn out.
Hume presents his lengthy discussion of probable argument as a detour justified by the light it might shed on the nature of the troublesome but essentially important idea of necessity. Hume’s definition of cause as a natural relation refers to the ‘determination’ of mind which is the source of the idea of necessary connection which, in turn, according to Hume, ‘makes an essential part of causation’. This associative ‘habit or determination’ to transfer the past into the future in causal inference arises not from a single instance of interaction but from a multiplicity of cases. It is the constant conjunction of related events that turns out to be the indirect source of our idea of necessity. In Section VI, Hume writes of the mind, in these circumstances, being ‘determin’d’ not by reason, ‘but by certain principles, which associate together the ideas of these objects, and unite them in the imagination’ (T.92). Hume treats the faculty of imagination as one other relevant causal factor in his mechanistic theory of the generation of belief. Convinced that, even after the observation of the constant conjunction of objects, reason cannot determine us to draw any inference concerning unobserved matters of fact, he turns to the other faculty of mind which, he believes, can explain why we continue to have recourse to it. It is, he says, the natural ‘principles of union’ among ideas which ‘make us pass from one object to another, even tho’ there is no reason to determine us to that transition’ (ibid.). The nature of causal association and belief is Hume’s main topic in Sections VII, VIII, and IX of Part III. Cause and effect, Hume argues, differs from those other forms of natural association, resemblance and contiguity, in terms of the strength or fixity of the association and the force and vivacity of the idea which is the conclusion of the inference. The degree of vivacity transmitted from the impression to the associated idea is, uniquely, sufficient to make that idea a belief. An impression, according to Hume, ‘not only transports the mind to such ideas as are related to it, but likewise communicates to them a share of its force and vivacity’ (T.98). In causal inference, the idea is not only ‘more vivid and intense’, but is, by nature, fixed, and of a more settled order (T.108). Hume finds that the influence of contiguity and resemblance, when compared to that of cause and effect, is ‘very feeble and uncertain’ (T.109). The strength of the influence of cause and effect alone warrants the use of the term ‘determination’ in description of it. Hume uses the term to denote the operation of a causal factor which, in effect, forces a particular perception upon the mind. When the mind is faced with a constant conjunction of events, it is ‘determin’d’, by the principles of union among ideas, to infer the existence of one from the appearance of the other (T.156). In these circumstances, the mind moves naturally and easily from the impression to the idea. Belief, in cases where we have found ‘the same impression’ to be ‘constantly conjoin’d with some other impression’, arises ‘immediately, without any new operation of the reason or imagination’ (T.102). The thought passes smoothly from impression to idea, ‘without any choice or hesitation’ (T. 110).
In cases where the correlation of past events is perfect, and the argument, therefore, ‘entirely free from doubt and uncertainty’, the mind is determined, even ‘forced’, to make the transition in question (T. 124–5). Hume also considers cases where the evidence for a belief is ‘attended with uncertainty’ (T.124), and the transition unconstrained. The sections ‘Of the probability of chances’ and ‘Of the probability of causes’ show how, when our experience of correlations is ‘mixed’ or ‘imperfect’, the ‘full and perfect’ habit of the past ‘gets broke into pieces’, and an ‘imperfect belief is produced (T. 134–5). In such cases, a lesser degree of vivacity is transferred from impression to idea, and the conclusion of the inference is, on that basis, less probable. Hume also argues that, where the correlation is imperfect, the mind is not determined, but merely influenced, by what it observes. Since Hume has explained belief in terms of custom, founded on the perceived resemblance of pairs of events, it is not, he thinks, strange that ‘the want of resemblance shou’d overthrow what custom has establish’d, and diminish the force of the idea, as much as that latter principle encreases it’ (T.114). A ‘contrariety of experiments’ produces a change not only in the vivacity of a belief, but in the strength of the transition to that, now ‘imperfect’, belief. Probabilities are thus to be thought of as sometime proofs, diminished in influence on the discovery of contrary experiments. It is, therefore, the ‘full and perfect’ regularities in our experience which account for the ‘fixt and unalterable’ state of the ideas we get by causal inference, each of which, Hume says, takes its place in the imagination ‘as something solid and real, certain and invariable’ (T.110). The same regularity explains the feeling of determination or constraint which attends the kind of causal association produced by a constant conjunction. We are not only determined, but, crucially, for Hume, we feel ourselves to be determined. It is this feeling which, in due course, Hume will show to be the source of the elusive idea of necessity.
Hume continues to develop his associationalist causal thesis in Sections VII and VIII, and in the sections dealing with probability. It ought to be noted that Hume’s account of causation, and of necessity, in particular, is itself causal. Hume is naturally unapologetic about this aspect of his account, having made it clear from the beginning that he intended to present a reflexive causal explanation of the basic human tendency to think of events in our experience in causal terms. Causal inference is one example of a natural process which can be investigated using the inductive methods of the natural sciences. The conclusion of Hume’s argument concerning causal inference prepares the ground for the positive account of causation and the idea of necessity which will occupy him for the remainder of Part III. In the last chapter, I argued that, in Section III of Part III, we find Hume abandoning the notion that any idea or relation of ideas could rationally ground our causal inferences, and instead presenting an account of our idea of the causal relation which is, in Fogelin’s words, ‘transparently causal’.8 He goes on to offer an explicitly causal explanation of how and why the idea of necessity arises, and a definition of causation as a natural relation which contains the causal term ‘determines’.9 Hume’s procedure has seemed to some critics circular, and, indeed, this could well be so, if Hume is to be read as offering a definition of cause in the ‘strict logical sense’ in which a contemporary philosopher would understand it. Hume’s doubts as to the completeness of his definitions, and their appeal to ‘objects foreign to the cause’, suggest that he did not think of them in this way. But even if we do construe them as definitions in this sense it is not clear that Hume has that much to worry about. He still has a number of options open to him in avoiding the charge of circularity. He can, for example, argue that what occurs in the definition is not the term but the impression of determination,10 or he can identify cases of determination by appealing to the feature of constant conjunction mentioned in the first definition.11 I do not propose to enter any more deeply into this debate or to defend Hume on these or other similar grounds. What I think is clear is that Hume does not regard the presence of the term as a defect of his definition, nor does he attempt in any way to disguise it causal nature. It is, in Baier’s words, ‘a part of the self-referential subtlety of the whole account’.12
Near the start of Section VI, Hume anticipates that it will ‘appear in the end, that the necessary connexion depends on the inference, instead of the inference’s depending on the necessary connexion’ (T.88). His suggestion is that the ‘new-discover’d relation’ of constant conjunction will somehow supply the worrying deficit in our idea of cause. But when he turns to the question of how this third relation can produce the idea of necessity he is faced with the problem of explaining how, from ‘the mere repetition of any past impression’, a new original idea, such as that of a necessary connexion, can arise (ibid.). This was the problem with which Hume introduced his discussion of causal inference, only to leave off giving a direct answer until the closing sections of Part III. By the end of Section XIII, the idea of necessary connection is still to be accounted for. Hume has indicated that without the observation of constant conjunction the idea would never arise, but how can ‘mere repetition’, which, as Hume admits, can generate nothing new in the events themselves, produce any new idea, other than that of repetition itself? In Section XIV, Hume argues that it can do so only by generating a new idea in the mind, an idea which is itself the copy of an ‘internal impression’ of reflection.
Hume argues that the transition from observed to unobserved matters of fact occurs only once we have observed not one but a multiplicity of similar cases of causal interaction. Causal beliefs are acquired by custom, as a result of regularities in our experience. Constant conjunction produces an association of ideas, causing the mind to pass from the impression of a cause to ‘the conception and belief of the usual effect’ (T.656). The belief in question arises naturally and unreflectively, without choice or hesitation. It is an important part of Hume’s thesis that we do not choose to believe as we do. Unbroken constancies in our experience produce a ‘habit or determination to transfer the past to the future’ (T.134), such that we cannot ‘without a sensible violence’ survey them in any other way (T.125). As we have seen, Hume thinks that we would never adjudge a sequence causal unless we believed that the cause in question necessitated its effects. It is therefore commonly supposed that besides contiguity, succession and constant conjunction, there is, in any instance of causality, a necessary connection between cause and effect, and that the cause possesses a causal ‘power’, ‘force’, or ‘energy’,13 and it is this supposition, of course, that Hume is interested in explaining (T.78). What, he asks, is our idea of necessity, ‘when we say that two objects are necessarily connected together’ (T.155). In accordance with the central principle of his empiricism – that every idea is derived from some corresponding impression of sensation or reflection – Hume begins again to look for the impression from which the idea of necessity is derived. In no single instance of cause and effect is it possible to discover any relation between the objects, except contiguity and succession. Hume, therefore, enlarges his view ‘to comprehend several instances’, adding constant conjunction to his list of relations. At first sight, he admits, ‘this seems to serve but little to my purpose. The reflection on several instances only repeats the same objects; and therefore can never give rise to a new idea. But upon farther enquiry I find, that the repetition is not in every particular the same, but produces a new impression, and by that means the idea, which I at present examine’ (ibid.).
After a frequent repetition of similar instances, Hume finds ‘that upon the appearance of one of the objects, the mind is determin’d by custom to consider its usual attendant, and to consider it in a stronger light upon account of its relation to the first object’ (T.156). The repetition is not therefore ‘in every particular the same’ as it produces a new impression of reflection, a feeling of determination, which, according to Hume, ‘affords me the idea of necessity’ (ibid.). This impression arises from the one circumstance in which classes of cases differ from singular cases. Following the observation of a constant conjunction of phenomena, the mind is determined to move from one idea or impression to the idea of its usual conjunct, and it is from our ‘internal impression’ of the feeling of constraint which accompanies this operation that the idea of necessary connection arises. Since reasoning a priori can never give rise to it, the idea of necessity must be derived from experience, and, as there is no impression of sensation which could produce the idea, it can only be derived from some internal impression of reflection. There is, furthermore, no impression of reflection ‘which has any relation to the present business, but that propensity, which custom produces, to pass from an object to the idea of its usual attendant. This therefore is the essence of necessity’ (T.165). We are faced then with a choice between conceding that we have no idea of necessity or necessary connection, and concurring in Hume’s view that these terms and their synonyms can mean ‘nothing but that determination of the thought, acquir’d by habit, to pass from the cause to its usual effect’ (T.657).
Hume’s conclusion needs to be considered with care. He has argued that while we have no impression or idea of necessity as an extra relation ‘in the objects’ already related by contiguity, succession and constant conjunction, we do, nevertheless, have an impression and an idea of necessity as that determination or feeling of determination of the mind to pass from cause to effect and from effect to cause. The ‘several instances of resembling conjunctions’ which produce the idea of necessary connection ‘are in themselves totally distinct from each other, and have no union but in the mind, which observes them, and collects their ideas’ (T.165). The repetition of similar instances does, however, give rise to a new impression. This impression is the effect of the observation of instances of resembling conjunction, and is ‘nothing but an internal impression of the mind, or a determination to carry our thoughts from one object to another’ (ibid.). To some extent, what Hume wants to say is clear, but there is some ambivalence in the way in which he says it. Hume argues that the idea of necessity arises from an impression we get when our minds are caused or ‘determined’ to infer the existence of one object from that of another. However, in the passage last quoted, and in numerous others, Hume seems to want to say that the experience of constant conjunction gives rise to the idea of necessity by producing a new internal impression of necessary connection which can be identified with the determination of the mind ‘to carry our thoughts from one object to another’. Since repetition can produce nothing new in the objects themselves, it must be from some new impression in the mind that the idea arises, and this new impression, Hume suggests, is an ‘internal impression’ of the determination of the mind to pass from one perception to its usual attendant. This, however, is not a finding with which Hume can rest comfortably, and, as I will now argue, it is not entirely consistent with what he says elsewhere.
Hume believes that the generation of this new idea is explained by the ‘determination of the mind’ that occurs after the repeated experience of conjunction and the observation of one of the conjuncts. The conjunction together with the present impression of an object causes the mind to get the idea of its usual attendant and, when this happens, the idea of necessity is produced. This much of what Hume has to say seems clear. The waters start to look muddier when he comes to account for the content of the impression from which this idea arises. The difficulty stems from his apparent claim that the impression of necessity which gives rise to the idea is an impression of mental determination, that is to say, of one event in the mind causing another. Hume writes of the determination of the mind in passing from an impression of one object to the idea of, or belief in, another with which it has been regularly conjoined. This describes a typical Humean cause-and-effect pairing between the constant conjunction and an impression of one of the conjuncts, and the belief that its usual attendant will occur. The problem with identifying this determination with the necessity of which, Hume says, we have an internal impression, is that he makes clear elsewhere that there can be no impression of any one mental or physical event causing another. Yet this is an identification which Hume does, on occasion, appear to make. He says, quite explicitly, that the ‘determination of the mind to pass from one object to its usual attendant’ must ‘be the same with power and efficacy’ (T.165). Necessity, he writes, ‘is nothing but an internal impression of the mind, or a determination to carry our thoughts from one object to another’ (ibid.). He also says, in passage from the Enquiry, that the customary transition of the imagination from one object to another is the ‘sentiment or impression from which we form the idea of power or necessary connexion’ (E.75). Necessity, Hume seems to suggest, is just the transition or determination from which the idea arises. This is not only troubling, but, as Stroud points out, is, taken literally, quite incoherent.14
Stroud writes that Hume appears to be arguing that, since the idea of necessity arises as a result of one mental event causing another, and since, according to the first principle of Hume’s empiricism, the cause of every idea is an impression, the one event’s causing another is the impression from which the idea of necessity is derived.15 This view of Hume’s intentions seems to be borne out in the above passage from the Enquiry in which he appears to argue that since one mental event causing another is the cause of the idea of necessity, it must be an impression of mental causation which is the source of the idea. His position is incoherent because, as he has shown, it can make no sense to talk of one event’s causing another being an impression. The determination or transition of the mind from ‘one object to its usual attendant’ does not directly furnish the mind with a new perception in addition to its perceptions of the two distinct mental events in the relation. When Hume writes of the customary transition from one mental event to another he can have in mind only the regular and orderly conjunction of events in the mind. We can be aware of the occurrence of pair of events, and of the order in which they occur, and trace up the impressions to which these ideas correspond, but we can have no impression, and hence, no idea, of the one causing the other.
Hume does, however, on other occasions, seem to want to say that in certain circumstances, specifically, in cases of mental causation, we do have an impression of the causal or necessary connection between the events in the relation. A number of passages appear to support the reading that, for Hume, we do sometimes get an impression of necessity ‘in the objects’, though only where the objects in question are mental, and that it is this impression that is the source of our idea of necessity. For instance, Hume says that in cases of causal inference we ‘immediately feel a determination of the mind’ (T.165). In the Enquiry, he writes that ‘we feel in the mind’ the ‘connexion’ by which ‘the mind is carried by habit, upon the appearance of one event, to expect its usual attendant, and to believe that it will exist’ (E.75). This would appear to suggest that, for Hume, we really do get an impression of the necessary connection between two mental events. If this was found to be Hume’s official view, it would be a most surprising, not to say, perplexing, find. Hume expressly denies that we can apprehend any necessary connections, whether between physical or mental objects or events. He argues that there is no independent impression of power accompanying the motions of the body or the acts of the mind. Shortly before restating his two definitions of cause in the Enquiry, Hume writes of the ‘pretension’ that we ‘are every moment conscious of internal power’ when we ‘feel … by the simple command of our will, we can move the organs of our body, or direct the faculties of our mind’ and that from this apprehension we ‘acquire the idea of power or energy’ (E.64). When we examine the supposition we find that the ‘influence of volition over the organs of the body’ is a ‘fact, which, like all other natural events, can be known only by experience, and can never be foreseen from any apparent energy or power in the cause, which connects it with the effect, and renders one the infallible consequence of the other’ (E.64–5). To apprehend a necessary connection between mental events would be to apprehend a ‘fact’ which would, so to speak, cast a writ over all future events of the sort, and, for Hume, no apprehension of any fact can have this remarkable consequence. Hume concludes ‘that our idea of power is not copied from any sentiment or consciousness of power within ourselves, when we give rise to animal motion’, nor are we ‘conscious of a power or energy in our own minds, when, by an act or command of our will, we raise up a new idea, fix the mind to the contemplation of it, turn it on all sides, and at last dismiss it for some other idea’ (E.67). He clearly means to reject out of hand the suggestion that introspection on events in the mind can give us the impression of necessity denied us when we considered objects and events in the physical world. This is, of course, unsurprising since it was Hume’s failure to find any relation of necessary connection between causally related objects that gave rise to the problem of the source of the idea of necessity in the first place, and which, furthermore, prompted him to seek an explanation of our erroneous attributions of necessity in the natural operations of the mind. The idea of necessity is caused by an impression or feeling of determination, but that impression cannot be understood as the perception of the causal or necessary connection between two events in the mind.
To be fair to Hume, this is not the only attempt he makes to characterise the content of the impression from which the idea of necessity is derived. Nor is it the most plausible or consistent account he offers. But it is important to bring out why, on occasion, Hume feels pushed into describing the content of the impression in these terms. What Hume wants to say is, quite simply, that when the mind is caused to infer an effect from a cause, it has, in addition to the enlivened idea of the effect, a certain feeling of constraint or determination in the transition, which it then projects beyond that source, onto the events themselves. This avoids saying either that we have an impression of one event’s causing another, or that the impression from which the idea of necessity arises is itself an impression of the necessary connection between mental events. Hume, of course, would not want to say either of these two things. The impression or feeling of determination which gives rise to the idea is not an impression of some additional relation discovered ‘in the objects’ of every causally related pair, but arises in the mind only when repeated observation causes us to infer the existence of one object from the observation of another. Hume tells us that in such cases the thought is ‘always determin’d to pass’ to the idea of the related object, and that it is the feeling of determination or constraint accompanying the transition which is the source of the idea of necessity. We get a feeling of the inevitability of the transition which we project onto ‘the objects’ in the form of an expectation that, given a constant conjunction of events, and a present impression of one of the conjuncts, the other must occur. Hume’s problem is to characterise this feeling of inevitability or determination as an impression without seeming to concede that in this case we are aware of one event being necessitated by another. He wants to say that in causal inference we feel a determination of the mind to pass from one object to the other and that it is this feeling which explains our belief that, given the cause, the effect must inevitably follow, but he needs to be able to account for the content of this feeling of determination in terms of a theory of ideas, according to which an idea must be derived from a correspondent impression. He finds himself pressed into saying that while we have no impression of necessity as such, we, nevertheless, have an impression of the inevitability of a happening which is something similar to a direct experience of necessary connection. Since, according to Hume, the idea of necessity arises from a determination of the mind, the impression from which the idea derives must, it seems, be an impression of the determination of the mind. But, as we have seen, Hume cannot consistently argue that the impression in question is a perception of one mental event causing another, or that it is an impression of the necessity of the connection between the events. What Hume needs is some way of expressing the thought that we experience the transition from cause to effect as inevitable without saying that we have a direct impression of the inevitability of the transition to the effect. He has argued that we get the idea of necessity because our minds are naturally disposed to react in certain ways to certain features of our experience, features which include contiguity in space and time, succession and constant conjunction, but not necessary connection. He wants to explain how and why the mind is led, following an observed constant conjunction, and the appearance of one of the conjuncts, ‘to expect its usual attendant, and to believe that it will exist’ (E.75), and then to project this expectation and belief onto a world which lacks the relevant feature of necessity. But that explanation has to be given within the constraints of the theory of ideas, and, as we have seen, it is his commitment to the theory of ideas that leads him to give an account of the content of the impression from which the idea arises which is, at best, confused, and which, at worst, violates at least two other important commitments.
Hume’s central thought is clear enough. He says that in describing a relationship as causal we are projecting a response we have to certain features of our experience onto an individual instance of conjunction. These features are contiguity, succession, and constant conjunction. They do not, of course, include necessary connection. We ascribe this further relation to objects and events only because of the way in which the mind responds to the above-mentioned features. The idea of necessity which we project beyond its origin in the mind onto the objects themselves arises because of the mind’s natural tendency to react in this way to the observation of events between which these three relations hold. It is this idea which we commonly mistake for ‘that very circumstance in the cause, by which it is enabled to produce the effect’ (E.67–8), that ‘very quality, which makes [causes] be follow’d by their effects’ (T.156). We have no idea of necessity as such. There is no fourth relation among the objects, mental or physical. We get the idea of power or necessity from ‘what we feel internally’ when the mind is led ‘upon the appearance of one event, to expect its usual attendant’ (E.75). It is this impression or feeling of the inevitability of the transition which we project back onto the world in describing one event as causally responsible for the other.
Hume concludes his account of ‘the essence of necessity’ by writing that ‘[u]pon the whole, necessity is something, that exists in the mind, not in objects; nor is it possible for us ever to form the most distant idea of it, consider’d as a quality in bodies. Either we have no idea of necessity, or necessity is nothing but that determination of the thought to pass from causes to effects and from effects to causes, according to their experienc’d union’ (T. 165–6). He does not mean to say that the relation of necessity holds only among mental events or that when we talk of events being necessarily connected we are saying something about events in our own minds. His point is that when we attribute necessity to a relation we do so not because, as we suppose, we have experienced some such quality in the objects, but because of the natural tendency of the mind to react to certain features of its experience in certain ways. The idea of necessity is derived from an impression of reflection which arises when the experience of constant conjunctions of phenomena causes the mind to infer cause from effect. Our tendency to project this relation onto the objects is explained by the ‘great propensity’ the mind has ‘to spread itself on external objects, and to conjoin them with any internal impressions, which they occasion’ (T.167). The same propensity ‘is the reason, why we suppose necessity and power to lie in the objects we consider, not in the mind, that considers them’ (ibid.). This belief that our idea of causal power or necessity is an idea of that quality in the objects ‘which binds the effect to the cause, and renders one an infallible consequence of the other’ (E.63) is clearly false since ‘it is not possible for us to form the most distant idea of that quality, when it is not taken for the determination of the mind, to pass from the idea of an object to that of its usual attendant’ (T.167). We have no idea or impression of the necessity with which we suppose two events to be connected. The idea of necessity which the mind possesses, and which makes an essential part of the idea of causation, is caused by an impression or feeling of determination which takes place in the mind when it is caused to infer an effect from a cause. It is this idea that we mistakenly ascribe to the objects and events of the physical world. Hume claims to have uncovered the ‘most violent’ of all the paradoxes which he will have occasion to advance, that the ‘efficacy or energy of causes is neither plac’d in the causes themselves, nor in the deity, nor in the concurrence of these two principles; but belongs entirely to the soul, which considers the union of two or more objects in all past instances. ‘Tis here that the real power of causes is plac’d, along with their connexion and necessity’ (T.166).
Hume is, finally, in a position to ‘collect all the different parts of this reasoning, and by joining them together form an exact definition of the relation of cause and effect’ (T.169). He now presents his two definitions which, as we saw, he believes present two different views of the same object (T.170). He defines cause in two ways, once, as a philosophical, and, again, as a natural relation:
We may define a Cause to be ‘An object precedent and contiguous to another, and where all the objects resembling the former are plac’d in like relations of precedency and contiguity to those objects, that resemble the latter.’ If this definition be esteem’d defective, because drawn from objects foreign to the cause, we may substitute this other definition in its place, viz. ‘A Cause is an object precedent and contiguous to another, and so united with it, that the idea of the one determines the mind to form a more lively idea of the other.’ (ibid.)
The first definition is a definition of causality as a philosophical relation because it considers the relation as ‘a comparison of ideas’, grasped through reflection on past experience of the relations which hold between events in a causal sequence. The second definition considers cause as a natural relation, which is to say, as an association of ideas. Two events might be said to be naturally related if they produce a ‘union among our ideas’, such that the thought of one carries the mind naturally to the thought of the other. The first definition refers to constant conjunctions of phenomena, the second to the psychological association of ideas which the experience of constant conjunction produces, and which gives rise to the idea of necessary connection. As I noted at the outset, not only are the two definitions not directly equivalent, referring, as they do, to different sets of impressions, but they are not even co-extensive. It is not difficult to conceive of sequences of events which would be adjudged causal on the basis of one definition, but not the other. It would seem that, although Hume appears to mean to endorse both definitions, they cannot both be correct, in the sense that they cannot both express the full meaning of the term. This has led to controversy among commentators, many of whom have believed that only one of the two definitions is to be understood as the correct one, alone offering the ‘precise’ definition of cause and effect Hume promises. The first definition, in particular, has often been used to support the view that Hume means to define causation as mere regularity of succession. In this section, I will consider this and a number of other influential views of Hume’s intentions, venturing an alternative which, I believe, can help resolve the debate.
It has been common for commentators to argue that the first definition alone represents Hume’s true view of causation as mere constant conjunction.16 J.A. Robinson, in his influential article, argues that, of the two definitions, only the first is to be regarded as the correct one, since only the definition of cause as a philosophical relation can be read as embodying Hume’s analysis of causation ‘as nothing more than an instance of general uniformity of concomitance between two classes of particular occurrences, and as quite independent of any association of ideas which may or may not exist in human minds’.17 The second ‘definition’ of cause as a natural relation is not strictly a definition at all, but ‘simply a restatement of the proposition that the (already defined) cause-effect relation is a natural relation, in a somewhat elliptical formulation’.18 Hume has argued that we are mistaken in supposing the causal relation to involve anything more than mere regularity of succession, and has sought to explain the origin of this ‘philosophical error’. His theory of causation should, on this view, be read as a metaphysical thesis about what causation is ‘in the objects’. The causal relation, according to Robinson, is just a philosophical relation, and to define causation as a philosophical relation is, simply, to define it. Hume’s theory can be aptly characterised as a reductive analysis of the concept of causality, according to which, propositions about causes can be satisfactorily analysed into ones about mere regularity of succession. All we can either understand or mean by causation is the regular succession of events.
Given a certain reading, a number of the key passages of Section XIV can be made to support the view that Hume identifies causation with constant conjunctions of phenomena, but there are equally, if not more, compelling textual grounds for rejecting the view. Hume, for example, continues to insist, both before and after his statement of the two definitions, that necessity ‘makes an essential part of causation’ (T.407). Hume not only argues that a correct definition of causation must contain some reference to necessary connection, but adds that this relation is of much greater importance than either of the other two relations referred to (T.77). It is scarcely credible, on the evidence of the text, that Hume used the second definition merely to restate that the already-defined causal relation is a natural relation.19 In so far as we confine ourselves to cause as a philosophical relation, we find only contiguity, succession, and constant conjunction, but Hume clearly means to say that this alone is not sufficient to give us the idea of causality. He writes that “tho’ causation be a philosophical relation, as implying contiguity, succession, and constant conjunction, yet ‘tis only so far as it is a natural relation, and produces a union among our ideas, that we are able to reason upon it, or draw any inference from it’ (T.94). Hume goes on to describe the two definitions as each presenting a ‘different view of the same object’ (T.170), and, in his concluding of Section XIV, reiterates his insistence on both statements being considered as definitions of the same relation (T.172). There is no direct textual evidence for the view that the second definition is no more than a psychological comment on causation as defined in the first, and to characterise his intentions in this way is, I believe, to seriously misrepresent them. The notion of causality comprises the notions not only of contiguity, succession and constant conjunction, but also of necessary connection. It is for this reason that the task of giving a correct account of the idea of causation proves especially difficult. The notion that constant conjunction can give us the idea of causality is rejected, for there must also be a natural association of the ideas, from which the idea of necessity can arise. There are very strong textual grounds for rejecting the view that Hume identified causation with constant conjunction. Hume’s claim that the idea of causation cannot possibly arise through a philosophical relation is reiterated in the Enquiry, where he writes that it is not possible ‘to define a cause, without comprehending, as a part of the definition, a necessary connexion with its effect’ (E.95). The first definition contains no direct or indirect reference either to necessary connection, or to the psychological association of ideas from which the idea of necessity arises. It would be surprising, not to say, astonishing, to find that Hume believed that this statement alone correctly defined causation.20
Hume proposes to examine the idea of causation. He argues that our idea of necessity arises not from what we perceive in any instance of conjunction, but from the impression we get when we infer one conjunct from the other. He does not argue that we have no idea of necessity. Nor does he deny that there might be ‘several qualities both in material and immaterial objects, with which we are utterly unacquainted’, and which it might please us to call ‘power or efficacy’ (T.168). He says only that we have no idea of causal power or necessity as that quality of causes ‘which makes them be follow’d by their effects’ (T.156). Necessity, according to Hume, must be understood as something ‘that exists in the mind, not in objects’ (T.165). This does not, of course, mean that, for Hume, we perceive the necessity of the connection between events in the mind. He means that we get the belief that a cause determines or necessitates its effects only because we feel ourselves to be determined when we infer one from the other. It is this idea of necessity or determination, arising from our natural tendency to infer one of two constantly conjoined objects from the other, that we commonly mistake for the cause of these constancies. Hume wants to retain the connection between causation and necessity, but he sees that the idea of necessity the mind possesses cannot be of ‘the very force or energy of the cause’ (E.69), but must instead arise from an internal impression of reflection, from ‘something that exists in the mind’.
Robinson’s suggestion was that this projectivism of Hume’s somehow committed him to the sceptical view that all there is to causation is constant conjunction. Hume is supposed to have argued that in nature ‘one thing just happens after another’.21 This view of Hume’s intentions has been vigorously challenged in recent years, notably in the influential writings of American scholar John P. Wright.22 If the above interpretation is to hold water it must be the case, Wright argues, that Hume infers the conclusion that there is no necessity in the objects from the premise that we have no idea of necessity as such. Wright further argues that Hume could only make that inference on the basis of the tacitly-held assumption that our ideas represent reality adequately. This assumption, Wright believes, can be attributed to Malebranche, to whom Hume’s account of causation owes much, and it is to Hume’s critical examination of Malebranche, he argues, that we must look in order to ascertain whether or not it is an assumption Hume shares. Wright argues that while Hume agrees with both Malebranche’s idea of ‘true’ causation as involving the direct perception of necessity, and his conclusion that sense experience furnishes us with no such perception, he nevertheless resists the conclusion that there is no power or necessary connection in the objects. Hume, Wright believes, ‘operates in terms of a belief in real physical causation’, and takes this and the representative theory of perception to be the ‘important ontological beliefs underlying science’.23 His rejection of the conclusion that causation in nature just is regularity of succession is explained by his rejection of the controversial assumption that our ideas adequately represent reality.24
The argument Hume presents in the section ‘Of the idea of necessary connexion’ is, according to Wright, directed ‘against contemporary theories which sought “to rob nature, and all created beings, of every power”’ (E.71).25 Such an interpretation would make Malebranche’s theory of causation a likely target. Malebranche argues that apparent causal relations between natural phenomena are not ‘true causes’ after all, but only occasions on which the ‘one single cause that is truly a cause’ has acted in effecting change.26 All apparent causal relations, or ‘secondary causes’, in nature, and all acts or volitions of mind, are ‘occasional causes’.27 The only true, ‘primary’ cause, according to Malebranche, is God. This is the argument Hume criticises in the Treatise (T. 159–161). In Wright’s view, Hume intends to attack not only Malebranche’s doctrine that causal power, being absent from matter, ‘must lie in the DEITY’ (T.159), but the assumption Wright alleges Malebranche makes that all ideas, whether of matter or of God, are adequate to the reality they represent. Some such view may be what Malebranche has in mind when he writes that a true cause ‘is one such that the mind perceives a necessary connection between it and its effect’.28 It is on the basis of this assumption that Malebranche is supposed to have argued that there is no ‘power or efficacy in any object’ (T.160), and this, of course, is the conclusion that Hume is supposed to be resisting.
According to Wright, while Hume endorses Malebranche’s understanding of true causation as involving necessary connection, he nevertheless rejects his claim that our ignorance of causes implies that they do not exist. Since Hume does not agree that ideas adequately represent the nature of reality, he can resist the otherwise irresistible conclusion that causation in the objects is just constant conjunction. Malebranche believed that God is the only true or primary cause since it is only between God’s will and its effects that we get a perception of necessary connection. Hume rejects this premise. He argues in the Treatise that, the principle of innate ideas being allowed to be false, the supposition of a deity ‘can serve us in no stead, in accounting for that idea of agency, which we search for in vain in all the objects, which are presented to our senses, or which we are internally conscious of in our own minds’ (ibid.). The same course of argument which led Malebranche to the conclusion that matter is not ‘endow’d with any efficacious principle, because ‘tis impossible to discover in it such a principle’, ought also to have led him to ‘exclude it from the supreme being’, and, so, to conclude that we ‘have no adequate idea or efficacy in any object’ (ibid.). In other words, the same assumption which caused Malebranche to deny causal power or efficacy to matter, ought also to have caused him to conclude the same with respect to God. Thus, the appeal to God cannot help us avoid the conclusion that there is no causal power, either in primary or secondary causes. This unwanted conclusion can be avoided only if we forsake the assumption that ideas reveal the ‘real power’ or nature of causes.
Hume argues that the occasionalist theorist of causation is committed, on the basis of the assumptions to which he subscribes, to an account of causation in nature as constant conjunction. According to Wright, Hume attempts to avoid this negative conclusion by rejecting one of the assumptions. He saw that this course of reasoning led to the ‘entirely Pyrrhonian’ thought that ‘there are no real forces in nature’.29 Nevertheless, in Wright’s view, Hume accepts ‘the basic Cartesian requirement of a “true” cause, namely, that there must be a necessary connection between cause and effect’.30 This is a surprising claim which, in my view, reflects Wright’s misapprehension as to where the brunt of Hume’s attack on the Cartesian theory of causation lies. Wright argues that when Hume writes of causal power or efficacy he means much the same as Malebranche did in characterising causation in terms of necessary connection. But this seems to ignore Hume’s own claim that his argument is to examine and subject to criticism the idea of causal power which these philosophers made use of. Hume writes that there is no question which has caused greater controversy among philosophers than ‘this concerning the efficacy of causes, or that quality which makes them be follow’d by their effects’, complaining that, before entering into these disputes, ‘it wou’d not have been improper to have examin’d what idea we have of that efficacy, which is the subject of the controversy’ (T.156). Hume wants to show that, contrary to the expectations of the Cartesians, we have no conception of what it would be for one event to necessitate another. This is not, of course, to show that Wright is mistaken either in thinking that Hume rejected the assumption that ideas adequately represent reality, or in stressing the important link between Malebranche’s theory of causation and Hume’s argument that we can have no direct perception of necessary connection. But Hume clearly means to place under suspicion, if not to reject outright, the notion that we have any idea of necessity as Malebranche thinks of it. He rejects the idea that the mind has access to the one instance of true causation, between God’s will and its effects, from which, according to Malebranche, the idea of necessity might yet arise. We have no impression and hence no idea of this sort of objective necessary connection between causes and their effects. Hume does not, of course, deny that there are causal qualities in objects with which we are ‘utterly unacquainted’, but these qualities, if such exist, can have no part to play in explaining how the idea of necessary connection is produced. The idea of necessary connection that the mind possesses arises not because we perceive the necessity in the relation of the objects of the inference, but because of the impression or feeling of determination we get when we infer one from the other. If this is the only idea of necessity we possess, then we can have no idea of necessity as Malebranche thought of it. Once we have understood this point, and Hume’s reasons for making it, the realist or sceptical realist view of Hume on causation begins to look increasingly difficult to uphold.
Wright’s reading of Hume on causation has been questioned, and, I think, effectively debunked, in an important article by Martin Bell.31 Bell argues that Wright’s exegesis of the crucial passages from the Treatise attributes to Hume a significant misunderstanding of an important doctrinal point in Malebranche. Having made the mistake in his own reading of Malebranche, Bell argues, Wright is quick to ascribe the same erroneous view to Hume. Hume is supposed to have been critical of Malebranche because, on the assumption that we are ‘perfectly acquainted’ (T.159) with the essence of both matter and God through our ideas, he is committed to the view that there is no causal power or necessity either in matter or in the supreme being.32 The problem for this view, according to Bell, is that this is not an assumption to which Malebranche subscribes. Although Malebranche did hold the view that we know bodies through our ideas, and Hume does, Bell admits, resist this view, ‘he did not hold that we know God through our ideas of him, and he would have denied the premise that Wright thinks Hume attributed to him’.33 Malebranche argues that although we know God ‘through himself and ‘by a direct and immediate perception’, our knowledge of the deity in this life is ‘very imperfect’. Our knowledge of corporeal things, on the other hand, is gained ‘through their ideas, i.e., in God, since only God contains the intelligible world, where ideas of all things are located’.34 Malebranche would not, therefore, be committed to the view, supposedly attributed him by Hume, that because we have no idea of the necessary connection between God’s will and its effects, there is no true causation in reality. There is strong evidence for thinking that Hume consulted Malebranche’s writings throughout his composition of these sections of the Treatise,35 It seems unlikely that a reader as astute and attentive as Hume would have misread Malebranche on a piece of doctrine which is central to his thought. Bell suggests an alternative reading which avoids ascribing Wright’s error of interpretation to Hume. According to Bell, when Hume argues that the ‘course of reasoning’ Malebranche follows will lead, once the doctrine of innate ideas has been discounted, to the conclusion that there is no causal power or necessity, he has in mind not ‘the use of the allegedly Cartesian premise that the mind has ideas adequate to the essence of both body and god’, but ‘the use of the doctrine that if the mind cannot perceive necessary connections when it consults its ideas of putative causes and effects, then these are not true, real causes’.36 Hume contends that if, as he argues, the doctrine of innate ideas is false, then Malebranche, if consistent, would be forced to deny that God is a true cause, on the grounds that it is not possible to discover in God any ‘efficacious principle’ (T.160). Given that the idea of God cannot be innate, but must, like all ideas, be taken from experience, the conclusion must follow for occasionalist philosophers like Malebranche, ‘precisely because they hold that to discover true causation is to consult one’s ideas and perceive that relation of ideas which they mean by “necessary connection”’.37
Hume’s concern, on this reading, is not with the avoidance of the regularity outcome that threatens the ‘Cartesian’ theory of causation, but with the account Malebranche gives of the ideas the mind possesses. Malebranche makes the mind’s perception of necessary connection the criterion for the existence of true causation or causal power. Hume objects that, given the falsity of the theory of innate ideas, and the failure of the mind to perceive any necessary connection in experience, Malebranche would be committed to the denial of true causation, whether in matter or the deity. To avoid so ‘absurd and impious’ an opinion, Malebranche and the Cartesians need only conclude ‘from the very first, that they have no adequate idea of power or efficacy in any object; since neither in body nor spirit, neither in superior nor inferior natures, are they able to discover one single instance of it’ (T.160). Far from endorsing ‘the Cartesian criterion of power’, as Wright suggests he does, Hume argues that we have no adequate idea of causal power or efficacy so understood, ‘in any object’. Since all ideas derive from experience and in no experience are we able to discover a single instance of it, we should admit, from the first, that this idea is one which we do not possess.
It can now be seen that some of the passages Wright quotes in support of his thesis do not support it in way he believes them to. Wright argues that by rejecting the Cartesian assumption that our ideas reveal ‘the essence or true nature’ of reality, Hume is able ‘to reject the Cartesian argument which leads to the conclusion that there is no power or force in material events’.38 In writing that ‘necessity makes an essential part of causation’, Hume is supposed to have in mind something of which, in fact, he believes we have no clear or adequate idea at all. But the idea of necessity Hume has in mind here is not the necessity of Malebranche and the Cartesians, but an idea drawn from the impression or feeling of determination we get when we infer cause from effect. By using the Cartesian criterion of true causation, Hume reaches the same conclusion as Malebranche: that we have no idea of causation in corporeal objects. But rather than argue for the existence of real causes in spite of our ignorance of them, Hume wants to deny that we have any idea of power or necessity so understood. Not only does the mind not perceive the sort of relation between ideas that would make possible a priori certain inferences from cause to effect, but it has no useful conception whatever of what it would be to apprehend a relation of this sort. The manner in which bodies operate on each other remains, Hume believes, ‘entirely incomprehensible’ to us (E.72). Nothing is more evident, Hume writes, ‘than that the human mind cannot form such an idea of two objects, as to conceive any connexion betwixt them, or comprehend distinctly that power or efficacy, by which they are united’ (E. 161). Hume continues to insist that necessity makes an ‘essential part’ of our idea of causation, but the idea of necessity we possess is not an idea of some quality of objects. Hume does not deny that there may be qualities of objects of which we are ignorant, and, he writes,
if we please to call these power or efficacy, ‘twill be of little consequence to the world. But when, instead of meaning these unknown qualities, we make the terms of power and efficacy signify something, of which we have a clear idea, and which is incompatible with those objects, to which we apply it, obscurity and error begin to take place, and we are led astray by a false philosophy. This is the case, when we transfer the determination of the thought to external objects, and suppose any real intelligible connexion betwixt them; that being a quality, which can only belong to the mind that considers them. (T.168)
Hume rejects the Cartesian notion of necessary connection, in favour of an idea arising from happenings in the mind which are the result of experienced constant conjunction. We have no idea of the sort of necessary connection between events which would ground a priori inferences and, so, give sanction to our inductive practices. The realist or sceptical realist view faces difficulties in reconciling its claims concerning Hume’s realism about causal connections with his avowed commitment to an empiricist theory of meaning which seems to deny that talk of these unknown and unknowable properties can make sense. A number of attempts have been made to deal with this difficulty. I will now consider one of the most prominent.
Galen Strawson, like Wright, believes that causal realism is a significant factor in Hume’s philosophical outlook. He argues, correctly, I think, that Hume is not concerned to deny that there are unknown properties in matter, underlying and explaining the regularities to which the mind, in inferring cause from effect, reacts, but adds, more controversially, that Hume’s commitment to this view is indicative of his willingness to bend, and even sacrifice, important elements of the empiricist theory of meaning. Strawson rests his argument on a distinction drawn from the Treatise which, he believes, allows Hume to talk intelligibly of causal powers without committing himself to the inconsistency of supposing that our minds possess an idea which he argues we cannot have. He points to a passage in which Hume seems to argue that there are some things of which we cannot conceive, of which we have no idea, yet which we can very well suppose to exist. Towards the end of his discussion of scepticism with regard to the senses, in Part IV, Hume writes of external objects that ‘we may well suppose in general, but ‘tis impossible for us distinctly to conceive, objects to be in their nature any thing but exactly the same with perceptions’ (T.218). The point can be read as a reiteration of his claim, made much earlier in the Treatise, that the farthest we can go in conceiving of external objects, ‘when suppos’d specifically different from our perceptions, is to form a relative idea of them, without pretending to comprehend the related objects. Generally speaking we do not suppose them specifically different; but only attribute to them different relations, connexions and durations’ (T.68). We are able to form ‘relative ideas’ of things specifically different from those we comprehend directly, and so, according to Strawson, can talk intelligibly of causes in nature as something more than mere constant conjunctions of phenomena.39 Strawson argues that Hume strives, quite self-consciously, to leave room, within his theory, for us to mean more by causation in the objects than regularity of succession. We can thus form a relative idea of natural necessity, and think coherently of it, without having any ‘descriptively contentful’ idea of what it is.40 We have, of course, no direct conception of what it is for one event to cause or necessitate another, but we may, nevertheless, suppose there to be something, of which we have no distinct idea, which is responsible for the regularities in question. We have what Hume, in the specialist language of eighteenth century empiricism, calls a ‘relative’ idea, of whatever it is ‘in virtue of which reality is regular in the way it is’.41 What we get is, in Strawson’s words, ‘a referentially efficacious but in a sense contentless and hence “merely relative” idea of something x’ in cases where one has an idea of a something ‘and one can refer to it only as, say, “whatever it was caused this mess”. One has no positive conception of the nature of x’.42
The main problem for Strawson’s view concerns the lack of good textual evidence for ascribing it to Hume. As a number of commentators, David Pears and Simon Blackburn among them, have shown, not only is there little direct textual support for the view, and none at all where we are led to most expect it, but, as we will see, the interpretation is in direct conflict with other aspects of Hume’s analysis.43 Hume does, of course, draw the distinction between specific and relative ideas, but he does so only in the context of his analysis of our idea of external existence, and, then, with little, if any, positive emphasis, and no acknowledgment of the importance the distinction is supposed to have had in his philosophy. He tells us that the farthest we can go in forming a conception of external objects is to form a ‘relative idea’ of them. We may ‘suppose in general’, but can never distinctly conceive, these objects to be specifically different from the perceptions which represent them. In neither passage does Hume make out what could be termed a positive case for the use of relative ideas in forming a relational understanding of an object, where we lack a direct conception of it. In the earliest of the passages, the context for Hume’s comment is given by his insistence upon the impossibility ‘for us so much as to conceive or form any idea of any thing specifically different from ideas and impressions’. Let us ‘chase our imagination to the heavens, or to the utmost limits of the universe’, he writes, we never ‘can conceive any kind of existence, but those perceptions, which have appear’d in that narrow compass’ (T.67–8). While Hume allows that we may ‘suppose in general’ something to exist which is specifically different, he is typically scathing about the results: ‘Generally speaking we do not suppose them specifically different; but only attribute to them different relations, connexions and durations.’ (T.68)
In the second passage, Hume is equally dismissive of the claim that relational ideas add to our understanding of external objects: ‘Philosophers deny our resembling perceptions to be identically the same, and uninterrupted; and yet have so great a propensity to believe them as such, that they arbitrarily invent a new set of perceptions, to which they attribute these qualities.’ (T.218) Once again, the emphasis is on the impossibility of our advancing ‘a step beyond ourselves’ towards an understanding or conception of anything specifically different from our perceptions. In the only other direct reference to relative ideas in the Treatise, Hume complains that ‘as every idea is deriv’d from a preceding perception, ‘tis impossible our idea of a perception, and that of an object or external existence can ever represent what are specifically different from each other. Whatever difference we may suppose betwixt them, ‘tis incomprehensible to us; and we are oblig’d either to conceive an external object merely as a relation without a relative, or to make it the very same with a perception or impression’ (T.241). Where Hume does, however briefly, entertain the notion of relational as opposed to specific ideas, he is scathing about the results and usefulness of such suppositions. He makes no direct use of the distinction between specific and relative ideas in his discussion of causation, nor does he refer to it in any of the relevant sections of Part III. This is understandable since the distinction might be said to work more naturally where the relative or relational idea in question is an idea of an object, rather than of a further relation. The idea of causation is the idea of a relation. So, the first thing to note is that the matter is not as straightforward as we might suppose to be the question of ‘whatever it was caused this mess’. Hume’s concern is not with the objects of the relation, but with their actual connectedness, something of which, for Hume, we have, and can have, no experience.
Although, as Strawson admits, it is a consequence of the theory of ideas that we can never form a ‘genuinely contentful conception’ of external objects, ‘it is still an intelligible supposition that there should be such things’.44 It is on the basis of such a supposition that Strawson thinks Hume is able to ‘mean something like Causation, at least in the sense of genuinely referring to it’,45 and to formulate a relative idea of something ‘in virtue of which reality is regular in the way it is’.46 Strawson characterises Hume’s realism as the claim ‘that there is something ‘external’ or ‘out there’ just in the sense of being independent of, or something other than, our perceptions – something which somehow gives rise to our perceptions, and is the reason why they are as they are’.47 As I have argued, it is no part of Hume’s brief to deny the existence of either causes or external objects. Hume nowhere argues the dogmatic point that there are no unknown properties operating in objects. But he does deny that these properties can explain or play any role in our tendency to treat the relation between two objects as necessary. Unless we recognise the impossible demands Hume places on the conception of the sort of thing Strawson has in mind in writing of the ‘fundamental forces’48 which underlie the phenomena, then we will not have understood the nature of Hume’s scepticism about causation or his reasons for dismissing out of hand the results of the supposition of relative ideas, and this, I suggest, would be a fundamental error of interpretation.
Strawson’s position, though influential, has been much criticised, most notably by Simon Blackburn. Blackburn argues that Strawson is betrayed into underestimating the theoretical pressures on the kind of fact that would, as he puts it, ‘soothe away inductive vertigo’, by a failure to distinguish adequately between what he calls a ‘thick nexus’ and a ‘thick straightjacket’.49 The distinction concerns two distinct things which might be asked of a ‘thick’ concept of causation.50 On the one hand, when we think of a causally related pair of events, ‘we want there to be a further fact than (mere) succession, or even mere regular succession of these kinds of event’, a fact making it so that ‘when the first happens the second must happen’.51 Blackburn terms this the desire for a causal nexus. On the other hand, when we shift our gaze to the ‘whole ongoing course of nature’, we feel that the pattern of regularities would be too much of a coincidence unless there was something in virtue of which it is regular in the way it is. We want there to be some ultimate cause, or ‘thick straightjacket’, whose ‘existence at one time guarantees constancies at any later time’.52 This is what Hume terms an ‘ultimate connexion’ (T.91). Blackburn is quick to point out how peculiar any fact that dispels this ‘inductive vertigo’ must be. It would have to be something the ‘continuing efficacy’ of which was subject ‘to no possibility of change or chance of failure’.53 For this reason, it is important to separate out the desire for a causal nexus, which is subject to possible change, from the desire for a straightjacket, which is not.
Blackburn claims that Strawson conflates the desire for a thick nexus with the desire for a thick straightjacket, and, overlooking the very special demands on a straightjacket, thinks he can straightforwardly point to what he terms ‘fundamental forces’ as a fact with the sort of potency required to ‘soothe away inductive vertigo’. Strawson argues that the concept of causation should be included ‘in the class of fundamental, non-sensory properties of reality’ which ‘we attribute to objects and which are essentially constitutive of our fundamental (pre-scientific) conception of their nature’.54 But he fails to see that something further is required if these ‘fundamental forces’ are to serve as the straightjacket Blackburn contrasts with the ‘ongoing flukes’ of the Regularity Theory. While Strawson uses the single term ‘Causation’ to refer to both a nexus and a straightjacket, Hume is more sensitive to the sort of distinction Blackburn has in mind. He writes in the Enquiry that experience ‘only shows us a number of uniform effects, resulting from certain objects, and teaches us that those particular objects, at that particular time, were endowed with such powers and forces’. When an object with similar sensible qualities appears,
we expect similar powers and forces, and look for a like effect … But this is surely a step or progress of the mind that wants to be explained. When a man says, I have found, in all past instances, such sensible qualities conjoined with such secret powers: And when he says, Similar sensible qualities will always be conjoined with similar secret powers, he is not guilty of a tautology, nor are these propositions in any respect the same. (E.37)
Hume recognises that whatever the nexus or connection between events at a particular time, it is something that may not hold at other different times. Even were experience to decide ‘that that very object, which produc’d any other, was at that time endow’d with such a power … [it] can never prove, that the same power must continue in the same object or collection of sensible qualities; much less, that a like power is always conjoin’d with like sensible qualities’ (T.91). All our ideas being distinct and separable, it is conceivable, and, therefore, for Hume, possible, that the same ‘powers and forces’ which operated on an object at one point in time, will not operate in the future. The appeal to past experience ‘decides nothing in the present case’ and, at the utmost, can prove only that a connection held between events at one time.
Hume makes the above distinction in order to underline his conviction that ‘if there be any suspicion that the course of nature may change…all experience becomes useless, and can give rise to no inference or conclusion’ (E.37). But it also lends support to Blackburn’s contention that, for Hume, the ‘ongoing regularity and constancy even of a thick nexus between one kind of event and another is just as much a brute contingent regularity as the bare concatenation of events’.55 In both cases, there is something that is capable of engendering the sense of ‘inductive vertigo’ to which Blackburn refers. Nevertheless, despite the clarity with which Hume expresses the distinction between the desire for a nexus and the desire for a straightjacket, he frequently talks of the perception of a causal nexus as though any such perception could make possible a priori certain inferences concerning unobserved matters of fact. He writes as though the idea of power or causation could only be an idea of necessary connection where ‘“necessary” implies that it belongs to the kind of which a priori knowledge is possible’.56 This aspect of Hume’s procedure has been criticised by both Craig and Strawson. Strawson argues that one reason for Hume’s confidence in his claim that an impression of necessary connection is unavailable ‘is that his conception of what something would have to be like in order to count as an idea of Causation or power or necessary connection in the objects is so demanding that it turns out to be simply obvious that nothing could ever count as such an idea, or such an impression’.57 Craig complains that it ‘looks as if he has just ignored what the modem reader will think of as an obvious prima facie possibility, that there is a necessity, stronger than concomitance but weaker than the deductive, and it is of this that we are seeking the impression and idea’.58 Craig’s accusation is that Hume stipulatively rules out the possibility of an apprehension of necessary connection by making the conditions for such an apprehension arbitrarily and unmeetably strict. It seems that Hume fails to consider the very possibility which the above-mentioned passages point to – that a causal nexus which holds at one time need carry no implications for any other.
Blackburn suggests an answer to the puzzle. Hume, he writes, must have seen ‘that nothing would really count as apprehension of a particular “must” unless it carried with it implications of uniformity for the general case’.59 To see that one event has to follow another at one particular time is, in other words, to see that it must always do so. Anyone apprehending a thick nexus at one time will also apprehend something which has general implications. If an observation cannot be said to do this for the general case, that is, for the whole possible course of nature, then nothing at all can be said of it as far as guaranteeing outcomes goes. One could not see that one event must happen, given the other, without seeing ‘something with general implications’, and this, for Hume, makes it difficult to see ‘how a particular nexus could be an object of observation’. How, Blackburn asks, could any time-limited observation apprehend ‘something that essentially casts its net over the whole of space and time?’.60 Anyone apprehending a thick nexus, a particular ‘must’, will also be apprehending something with general implications, the impossibility of events turning out otherwise, and this, of course, makes the demands on the observation of a particular nexus extraordinary.61 Hume rules out the possibility of the apprehension of any thick causal nexus not because he arbitrarily stipulates that full apprehension of a nexus will make possible a priori certain judgments about unobserved matters of fact, but because he saw that to apprehend any kind of connection implying that one event must follow another is to apprehend a fact with implications for the entire course of nature.
Hume argues that the perception of the necessary connection between events would ‘amount to a demonstration, and wou’d imply the absolute impossibility for the one object not to follow, or to be conceived not to follow upon the other’ (T. 161–2). A demonstration of this sort is, for Hume, impossible, since the ideas of cause and effect are distinguishable and hence separable by the thought and imagination (T.18). No perception can have the consequence of making a priori knowledge of matters of fact possible. Hume is concerned to show that we have and can have no idea of that ‘quality’ in the objects which makes causes be followed by their effects, and which ‘renders the one an infallible consequence of the other’ (T.156; E.63). We have no impression or idea of necessary connection as a relation between events, and no conception at all of what it would be for one event to necessitate another. As we have seen, Hume’s account leaves no room for the lesser claim ‘to have apprehended a particular, but not necessarily timeproof, thick connexion’.62 He thought that to apprehend a thick connection between events was to apprehend something with the consequences of a straightjacket, something which, per impossible, has a priori implications for all unobserved cases of a sort. Strawson’s conflation of the desire for a nexus with the desire for a straightjacket, and his failure to recognise the extra demands on a straightjacket, explain his appeal to ‘fundamental forces’ as a means of dispelling our fears about the collapse of the ongoing course of nature. Hume is at pains to make clear that such a supposition can in no way help us since we can have no idea at all of the sort of fact which would alleviate our concerns about the ongoing course of nature. He does not deny that there are ‘ultimate springs and principles’, ‘unknown’, ‘secret’, and ‘totally shut up from human curiosity’, but these are not the causes and principles with which science ought to concern itself (E.30). The natural scientist has no choice but to eschew speculation about ‘ultimate connexions’. They remain ‘entirely incomprehensible’ (E.33). If we please to call these qualities ‘power or efficacy, ‘twill be of little consequence to the world’ (T.168), for such suppositions have no role to play in understanding. Natural science can hope only ‘to reduce the principles, productive of natural phenomena, to a greater simplicity, and to resolve the main particular effects into a few general causes, by means of reasonings from analogy, experience and observation’ (E.30). The ‘most perfect’ natural philosophy, Hume says, ‘only staves off our ignorance a little longer’ (E.31).
The mind does possess an idea of necessity, derived from the impression it gets when it is caused to infer an effect from its cause. Hume’s account of how we come to project this idea beyond the mind, and onto the objects, explains the view of the ‘generality of mankind’ that ‘they perceive the very force or energy of the cause, by which it is connected with its effect, and is for ever infallible in its operation’ (E.69). They little suspect the falseness of their view since they have acquired ‘such a turn of mind, that, upon the appearance of the cause, they immediately expect with assurance its usual attendant, and hardly conceive it possible that any other event could result from it’ (ibid.). Hume describes how the mind changes when exposed to the constant conjunction of events so that the impression of one of the conjuncts produces the idea of its ‘usual attendant’. When this happens the mind has an impression or feeling of determination which it projects onto the events in the form of an attitude of expectation that one event will be followed by the other. In none of this do we perceive in the objects anything more than the regular succession of events. We have the choice, Hume writes, ‘either to assert, that nothing can be the cause of another, but where the mind can perceive the connexion in its ideas of the objects: Or to maintain, that all objects, which we find constantly conjoin’d, are upon that account to be regarded as causes and effects’ (T.248). But the idea of two objects constantly conjoined cannot by itself give rise to the idea of causation. We only think of the objects as causally related because the observation of constant conjunction produces a ‘union in the imagination’, a tendency to infer one from the other. It is because of the feeling of determination which accompanies the inference that we think of the relation as a necessary or determined one. Hume believes that necessity is an essential ingredient of the idea of causation. But the origin of the idea of necessity cannot be explained with reference to the objective relations of events in the world. Necessary connection is ‘something that exists in the mind’. It is this idea that the mind ‘spreads upon the world’, forming habits of expectation, which it then projects beyond their source in the mind, onto the ‘entirely loose and separate’ objects of its experience (E.74).
It can easily appear that Hume is offering two theories of causation, one to do with the evidence for a causal inference, and one to do with its effects on the mind.63 But Hume’s intentions are quite different. He is interested in explaining how it is the mind comes to see and describe the world in terms of causal connections, inspite of a reality which has none of the representational features our causal vocabulary leads us to expect. The mind gets its idea of necessity because of the feeling of determination that accompanies the inference from one object or event to its usual conjunct. The idea of necessity is an effect of those regularities. And it is because of that idea that we treat their relation as necessary. These are the two main aspects of the causal story that Hume traces up to his two definitions of causation. I argued earlier that Hume intends to endorse both definitions. Hume prefaces their statement with the promise that he is ‘to collect all the different parts of this reasoning, and by joining them together form an exact definition of the relation of cause and effect’, adding that ‘[t]here may two definitions be given of this relation’ (T.169). They together form what Hume terms ‘a precise definition’ of causation. It would be remarkable, in these circumstances, for Hume to mean to endorse only one of the two definitions, while giving no indication as to which one. Nevertheless, the fact that the definitions are neither directly equivalent nor coextensive means that Hume appears to contradict himself in presenting two ‘exact’ definitions of the same concept which do not apply to all the same objects. If Hume’s intention is to state the full and accurate meaning of the term then his definitions are contradictory. They cannot both be correct as definitions of the same notion. But there is reason to believe that Hume did not think of them as definitions in the strict logical sense of contemporary philosophy. Neither gives an ‘exact’ and ‘precise’ definition, as a modem day philosopher or lexicographer would understand it. Hume, indeed, admits that ‘it is impossible to give any just definition of cause, except what is drawn from something extraneous and foreign to it’ (E.76). Instead, the two definitions are intended to distinguish and encapsulate the two sides of the causal story sketched above.
The first definition of cause as a philosophical relation describes the regularities to which the mind reacts in inferring an effect from a cause. The second definition concerns causation as a natural relation. Hume explains that although causation is a philosophical relation, implying contiguity, succession and constant conjunction, ‘yet ‘tis only so far as it is a natural relation, and produces an union among our ideas, that we are able to reason upon it or draw any inference from it’ (T.94). The idea of constant conjunction will never be enough to give rise to the idea of causation. There must also be the natural associative tendency of the mind to infer one object from the other. The second definition emphasises the changes that take place in the mind when it is caused to infer an effect from a cause. It is because the mind feels itself to be determined in the transition that it treats a sequence as necessary or determined. No new feature or relation in the objects is revealed or plays any part in this change. But once the experience of constant conjunction has given rise to the idea of necessary connection, the mind treats the relevant sequences as causal or necessary ones, acting and forming expectations in accordance with its projections. The definitions, though not definitions in the strict logical sense, nevertheless sum up, definitively, the two distinguishable aspects of Hume’s positive theory of causation. One concerns the objective states of affairs which cause the mind to infer effect from cause; the other the changes in the mind which explain our treatment of processes as necessary, causal ones. Objects or events which are observed to satisfy the first definition will give rise to the association of ideas appealed to in the second definition. It is only because of this association that we get the idea of necessity. We would not otherwise regard any process or sequence of events as causal or necessary. It is for this reason that necessity is said to make ‘an essential part of causation’ (T.407). The first definition, while stating all the relations which are observed to hold between events which we believe are causally related, nevertheless fails to account for all that we mean in describing a relation as causal. It cannot, therefore, be taken to be all that Hume understands causality to be. There is also necessary connection to be taken into consideration (T.77). Hume admits that any definition of causation, his own included, will be drawn ‘from something extraneous and foreign to it’ (E.76). Although both definitions draw on circumstances foreign to the cause, we cannot attain a ‘more perfect’ definition, pointing out the circumstance of the cause which gives it a connection with its effect, since ‘[w]e have no idea of this connexion, nor even any distant notion of what it is we desire to know, when we endeavour at a conception of it’ (E.77). What is clear from a reading of Part III as a whole is that Hume intends his two definitions to distinguish and encapsulate the two main elements of his account of the idea of causation – the objective relations of events and the mind’s response to them – and to show their interdependence. This is not, Hume acknowledges, to give a full or strict definition of causality, but it is, he suggests, as far as we can go towards an understanding of it.
1John Mackie, The Cement of the Universe, Oxford, Oxford University Press, 1974, p.3.
2See, for example, J.A. Robinson, ‘Hume’s Two Definitions of Cause’, The Philosophical Quarterly, vol. 12, 1962; and Mackie, op.cit., pp.3–4.
3See Robinson, op.cit.; and Terence Penelhum, Hume, London, Macmillan, 1975. A similar view is ascribed to Hume by Thomas Reid, Essays on the Active Powers of the Human Mind, in The Works of Thomas Reid, 6th edition, ed. William Hamilton, Edinburgh, Maclachlan and Stewart, 1863, p.627. Reid describes Hume’s position as follows: ‘I know of no other author before Mr. Hume, who maintained, that we have no other notion of a cause, but that it is something prior to the effect, which has been found by experience to be constantly followed by the effect.’
4A.J. Ayer, The Central Questions of Philosophy, Harmondsworth, Penguin, 1973, p.183.
5See, for example, Edward Craig, The Mind of God and The Works of Man, Oxford, Clarendon Press, 1987; John P. Wright, op.cit.; and Galen Strawson, op.cit.
6Galen Strawson, op.cit., p.ix.
7This thesis constitutes the kernel of Hume’s theory of causation. Hume argues that we can have no perception of necessity or necessary connection ‘in the objects’ on the grounds that such a perception would ‘amount to a demonstration, and wou’d imply the absolute impossibility for the one object not to follow, or to be conceived not to follow upon the other’ (T. 161–2). He was not, however, the first philosopher to argue against the possibility that necessary connections could be given in experience. The seventeenth-century French philosopher Nicolas Malebranche argued, on similar grounds, that we have no intelligible notion of necessary connection between mind and body, suggesting instead that the mind perceives a necessary connection ‘only between the will of an infinitely perfect being and its effects’, op.cit., p.450. Malebranche believed that since all causal connections are necessary, and no other cause is necessarily followed by its effect, the only true cause is the will of God. What we think of as causation in nature, according to Malebranche, just is the regular succession of objects or events (ibid., pp.449–51). Natural causes ‘are only occasional causes that act through the force and efficacy of the will of God’ (ibid.).
8Fogelin, op.cit., p.48.
9See Norman Kemp Smith, The Philosophy of David Hume, London, Macmillan, 1941, p.401. Smith was the first commentator to give special emphasis to the causal nature of Hume’s theory of causal belief
10As argued by Wade L. Robison, ‘Hume’s Causal Scepticism’, in David Hume: Bicentenary Papers, Edinburgh, Edinburgh University Press, 1977, p. 159.
11See Garrett, op.cit., p. 112.
12Baier, op.cit., p.99.
13T. 157. Hume writes: ‘I begin with observing that the terms of efficacy, agency, power, force, energy, necessity, connexion, and productive quality, are all nearly synonimous.’
14See Stroud, op.cit., pp.80–1.
15Ibid., p.80.
16See, for example, J.A. Robinson, ‘Hume’s Two Definitions of “Cause”’, The Philosophical Quarterly, vol. XII, 1962, pp. 163–171; Nicholas Capaldi, David Hume: The Newtonian Philosopher, Boston, Twayne, 1975, pp.95–129; and A.H. Basson, David Hume, Baltimore, Penguin Books, 1958, pp.73–6.
17J.A. Robinson, op.cit., p. 167.
18Ibid.
19T.J. Richards, ‘Hume’s Two Definitions of “Cause”’, The Philosophical Quarterly, vol. XV, 1965, pp.247–253, criticises Robinson on these grounds.
20Beauchamp and Rosenberg, op.cit., p. 16, argue that the second definition could just as plausibly be taken as the only true definition. There is limited textual evidence for this view also. In the Enquiry, Hume writes that when we say that one object is connected with another ‘we mean only that they have acquired a connexion in our thought, and give rise to this inference, by which they become proofs of each other’s existence’ (E.76).
21A.J. Ayer, op.cit.
22See Wright, op.cit., pp. 123–186. A similar defence of Hume is laid out by Galen Strawson, op.cit. Strawson writes that ‘even if there were good grounds for thinking that Hume’s theory of ideas licensed the strange claim that all we can suppose a thing to be is what we can detect or experience or know of it, because we cannot manage to mean anything more than what we can detect or experience or know of it when we think or talk, the following problem would remain: the claim that causation is nothing but regular succession, which makes a positive ontological assertion, is violently at odds with Hume’s scepticism with respect to knowledge claims about what we can know to exist, or know not to exist, in reality’ (p.276).
23Wright, op.cit., p. 13.
24Ibid., p. 129.
25Ibid., p. 135.
26Malebranche, op.cit., p.451.
27Ibid., p.450.
28Ibid.
29Wright, op.cit., p. 127.
30Ibid.
31Martin Bell, ‘Hume and Causal Power: The Influences of Malebranche and Newton’, The British Journal of the History of Philosophy, vol. 5. no. 1, 1997, pp.67–86.
32Ibid., p.77.
33Ibid., pp.77–8.
34Malebranche, op.cit., pp.236–7.
35See Charles J. McCracken, Malebranche and British Philosophy, Oxford, Clarendon Press, 1983, pp.257–8. McCracken cites a number of passages from the Treatise which seem to be closely modelled on passages from Malebranche, suggesting, he says, ‘that Hume not only kept the Search in mind, as he wrote on causality, but that he even had it open for consultation while writing’.
36Bell, op.cit., p.78.
37Ibid., pp.78–9.
38Wright, op.cit., p. 143.
39Strawson, op.cit., p.279.
40Ibid.
41Ibid.
42Ibid., p.281.
43See Pears, Hume’s System: An Examination of the First Book of His Treatise, Oxford, Oxford University Press, 1990, pp.91–2; and Blackburn, ‘Hume and Thick Connections’, Essays in Quasi-Realism, Oxford, Oxford University Press, 1993, pp.94–110.
44Strawson, op.cit., pp.281–2.
45Ibid., p.284.
46Ibid., p.279.
47Ibid., p.282.
48Ibid., p.91.
49Blackburn, op.cit., pp. 100–101.
50Ibid., p.94. Blackburn describes a thick concept of the dependence of one object on another as ‘any concept of one event producing another, or being necessarily a cause or consequence of another, and that involves something in the events beyond their merely being kinds of events that regularly occur together’.
51Ibid., pp.97–8.
52Ibid., p.98.
53Ibid.
54Strawson, op.cit., pp.254–5.
55Blackburn, op.cit., p.99.
56Craig, op.cit., p.97.
57Strawson, op.cit., p. 110.
58Craig, op.cit.
59Blackburn, op.cit., p.99.
60Ibid.
61Ibid., pp.99–100.
62Ibid., p.100.
63A view that Pears, op.cit., p. 117, ascribes to Beauchamp and Rosenberg.