The Miseries of Life Hume and the Problem of Evil

Abstract

My topic is Hume's treatment of the problem of evil in the Dialogues and elsewhere in his philosophical writings. The aim is to provide an overall view of Hume's position which also takes account of the historical debate associated with the problem of evil. Critical and interpretative issues will also be addressed. We shall see that Hume is concerned mainly with a particular form of the evidential argument from evil which appears especially damaging to theistic belief in so far as it calls into question traditional views of the nature of God.

Introduction

Hume's principal discussion of the problem of evil occurs in parts 10–11 of the Dialogues.¹ It is therefore not surprising that treatments of Hume's views have tended to focus on this discussion and, in particular, on the contributions of Philo as a participant in the Dialogues who articulates the problem of evil in part 10.² I aim in this paper to examine the place of this discussion within the Dialogues as a whole and also its relation to Hume's philosophical writings more generally. In doing so I will be considering in detail the different aspects of Hume's treatment of the problem of evil and their significance within the context of the historical debate associated with this problem. The result is a synoptic view of Hume's position which also engages with issues raised in the secondary literature. It will emerge from my discussion that Hume is principally concerned with a particular form of the evidential argument from evil as it relates to the project of natural theology.³ While Hume treats this argument as one that bears primarily on the question of the nature of God we shall see that it also has significant implications for what is involved in accepting the existence of God. In fact, the argument from evil appears more damaging to theistic belief, on Hume's view, than any shortcomings in the argument from design.

I shall proceed as follows. I begin in section 1 by noting some central themes of the Dialogues, and in section 2 I relate these to the discussion of the problem of evil which takes place in parts 10–11. In sections 3 and 4 I focus on Hume's position in regard to the problems of, respectively, moral evil and natural evil. Section 5 is concerned with some of the more general philosophical questions arising from Hume's employment of a form of the evidential argument from evil. I conclude by noting the implications of Hume's treatment of the problem of evil for the dispute between theism and atheism and for natural religion.

1. Some Central Themes in the Dialogues

There is an important continuity between discussion of the topic with which parts 10–11 of the Dialogues are concerned—human misery and wickedness and their implications for the existence of a deity which supposedly possesses moral attributes like mercy and benevolence—and discussion in preceding parts of the attempt to establish the existence of a supreme creator by arguments belonging to natural religion. The preamble to the Dialogues sets the scene for the succeeding debate by indicating that the really difficult or obscure question at issue in natural religion concerns the nature of the deity—his attributes and plan of providence—rather than his being or existence (DNR Prologue 5; 128).⁴ This is reflected in the skeptical stance adopted by Philo in the opening parts of the Dialogues.⁵ Thus, in part 1 Philo does not directly challenge the religious hypothesis itself but, rather, emphasises that speculation about the powers and operations of the universal spirit may be taking us beyond the reach of our faculties (DNR 1.10; 135). In similar vein, Demea suggests that what is at issue is the nature of God and emphasises the mystery of the essence of the divine mind and its attributes (DNR 2.1, 3.11; 141, 155–56). Philo agrees that while the being of God is unquestionable and self-evident, his attributes lie beyond our comprehension. The point of ascribing perfections like wisdom and knowledge to the divine being is essentially to enable us thereby to express our adoration of God; but we should not suppose that his perfections have any resemblance to those of human beings (DNR 2.3; 142).⁶

In part 4 there is an exchange in which Cleanthes represents Demea as a mystic, someone who emphasises the incomprehensibility of the deity; while Demea responds by labelling Cleanthes as an anthropomorphite, someone who represents the deity as being similar to a human mind and understanding (DNR 4.2; 158–59).⁷ This provides us with a useful pair of philosophical labels for referring to these different approaches to understanding the nature of the divine mind. Philo aligns himself with Demea in opposition to anthropomorphism, in particular regarding the comparison between, on the one hand, the divine mind and the plan of the world it is supposed to contain and, on the other, the mind of a human architect and his ideas for the plan of a house. To this extent Philo is prepared to refer to himself as a mystic (DNR 6.6, 10.27; 172, 199), though his understanding of this notion differs in important respects from that of Demea.⁸ Perceived as proponents of mysticism, Demea and Philo face the accusation that their position is really indistinguishable from atheism in so far as a mind to which we are apparently unable to attribute thought, reason, will, and so on, is no mind at all. In part 5 Philo pursues his disagreement with Cleanthes who appeals to the principle that like effects prove like causes in order to establish that the works of nature exhibit a design (art, contrivance) for which a mind must be responsible (DNR 5.1–4; 165–66). Philo responds by raising questions about the kinds of property which may be ascribed to any such mind on this basis. For one thing, given the principle that the cause ought to be proportioned to its effect, the attributes of the mind supposedly responsible for the natural world cannot apparently be considered infinite. But secondly, Philo anticipates the discussion of Dialogues 10–11 by suggesting that the many "inexplicable difficulties in the works of nature" also prevent us from ascribing perfection to the deity. Finally, there is a question about the unity of the deity. Without some antecedent proof that there is one deity possessing the attributes required for producing the universe, the question remains as to why these attributes should be united in one subject (DNR 5.9; 168).

Philo continues to emphasise, in part 6, the absolute incomprehensibility of the divine nature and the mysticism to which we are committed in this respect as a result of our limited experience (DNR 6.6; 172). In part 8 he responds to Cleanthes' claim that nature provides us with proof of benevolent design by raising a more general problem for anthropomorphism: viz., that thought is known to have influence on matter only where, contrary to the religious hypothesis, there is a reciprocal influence (DNR 8.11; 186). Finally, while Demea suggests in part 9 that argument a priori may be employed to establish truths about the divine nature, he is reminded by Cleanthes of Philo's previous claim about the impossibility of proving any matter of fact by means of such argument (DNR 9.5; 189).

We see, then, that a crucial assumption underlying the discussion of parts 1–10 of the Dialogues is that the question essentially at issue concerns the nature of God.⁹ From this perspective we may distinguish anthropomorphism, which ascribes attributes to the divine mind on analogy with those which belong to human minds, and mysticism, which questions the legitimacy of this analogy and stresses the incomprehensibility of God.¹⁰ So far as the latter is concerned—at least, as represented by Philo—there are particular difficulties with the supposed infinity, perfection and unity of the divine attributes. This provides the background for the discussion of the problem of evil which occurs in parts 10–11 of the Dialogues.

2. The Problem of Evil in Dialogues parts 10–11

The reality of human misery on which both Philo and Demea agree at the beginning of part 10 provides Philo with a further line of attack on Cleanthes' anthropomorphism: in particular, concerning the ascription of moral attributes to the deity.¹¹ The context is provided by "Epicurus's old questions" about reconciling the existence of God as a supposedly benevolent and all-powerful being with the occurrence of evil (DNR 10.25; 198).¹² In what respects may divine benevolence and mercy, for example, be seen to resemble human benevolence and mercy? Mysticism is once again contrasted with anthropomorphism. Nature presents us with a "mixture of phenomena" so far as human misery and happiness are concerned, and we are unable to comprehend the attributes from which they are derived. While Demea appeals to the providence of the deity as providing a solution to the dilemmas posed by Epicurus, Cleanthes claims that the idea of divine benevolence can be supported only by denying the misery and wickedness of men (DNR 10.31; 200). Philo concludes his contribution to this part of the discussion by suggesting that even if human pain and misery are compatible with the infinite power and goodness of the deity, there is still the apparently insuperable problem of proving (or establishing) the existence of such divine attributes on the basis of our experience of the mixed phenomena with which nature presents us (DNR 10.35; 201). There is, it appears, no view of human life from which any such inference may be drawn.

Philo and Cleanthes appear to agree about the difficulty of answering "Epicurus's old questions" regarding the compatibility of evil with divine omnipotence and benevolence. But Hume is nevertheless clearly aware of theodicies which are employed in response to the Epicurean dilemmas. He recognises a form of argument in reply to Epicurus which disputes the assumption that an infinitely powerful and benevolent deity would not allow the existence of any kind of suffering. There is a reference in the first Enquiry to the view that nature is ordered with perfect benevolence so that every "physical ill" is an essential part of the system which even the deity himself would be unable to remove without causing some greater ill or excluding a greater good (EHU 8.34; SBN 101).¹³ Hume's familiarity with Leibniz's Essais de Théodicée is illustrated by Philo's reference to Leibniz in the present context (DNR 10.6; 194).¹⁴

Given the arguments of Dialogues 10–11, it is not difficult to envisage Hume's own reaction to a theodicy which appeals to the idea that the best of all possible worlds requires human suffering on the scale with which we are familiar. There is, in particular, Philo's insistence that unless we are already convinced of the existence of a supremely benevolent and powerful intelligence then the appearances of things simply fail to accord with the kind of world we might expect such a being to have created (DNR 11.2–4; 203–205). In fact the arguments which Philo goes on to deploy suggest that we can rather easily imagine ways in which a supreme being could contrive matters so as to make this a happier world without apparently compromising the benefits to be derived from a universe which is conducted by general laws.¹⁵ In any event, Cleanthes' agreement with Philo about the force of the Epicurean argument explains why he should accept that, "[i]f you can make out the present point, and prove mankind to be unhappy or corrupted, there is an end at once of all religion" (DNR 10.28; 199). Above all, the existence of human suffering poses a direct threat to the attribution of moral attributes like benevolence to the deity. To paraphrase Philo's rendering of Epicurus, if the deity is able but not willing to prevent evil then he is malevolent.

While Demea appeals to the after-life ("some future period of existence") as the time at which the evils of this life will be rectified by God's providence,¹⁶ Cleanthes rejects this as a hypothesis which is unsupported by the "apparent phenomena" (DNR 10.30; 200). To the extent that human suffering is considered to be a pervasive feature of our experience we are deprived of any basis for supposing that the deity does exercise benevolence or justice in this way.¹⁷ Cleanthes is therefore required to maintain that human suffering, to the extent that it exists, is in fact considerably outweighed by pleasure and happiness.¹⁸ Philo responds to this with a subtle treatment of the reality of evil as represented by the predominance of pain over pleasure. He distinguishes, for example, between the apparent frequency with which pain occurs as compared with pleasure and the degree to which we are affected by pain as opposed to pleasure (DNR 10.32; 200).¹⁹ But he also recognises the difficulty of arriving at any objective measure of the comparative frequency with which these different sensations occur and points to this as a difficulty for anyone who rests his system of religion on a particular claim about their occurrence (DNR 10.33; 200–201; cf. Stewart, "An Early Fragment," 167). In any case, as a final point against a position like that of Cleanthes, it would seem that any misery at all, however much it might be exceeded by human happiness in this life, is sufficient to block the inference to the existence of a perfectly benevolent deity (DNR 10.34; 201; cf. Stewart "An Early Fragment," 168).

There is another aspect of Hume's approach to evil in the form of human suffering which is worth mentioning.²⁰ Hume observes that while the theist may find a problem here for belief in a supreme intelligence, the kinds of events which cause such suffering are a source of the vulgar belief in a being of this kind (NHR 153–54)²¹. The "anxious fear of future events" which is the chief source of religious belief may even be described as "the natural state of religion, when surveyed in one light" (NHR 176). In somewhat similar fashion Philo suggests that "just representations of the misery and wickedness of men" provide "the only method of bringing everyone to a due sense of religion" (DNR 10.2; 193). There is also a particular feature of Hume's philosophy which would require him to reject a position of the kind taken by Cleanthes in so far as maintaining divine benevolence requires him "to deny absolutely the misery and wickedness of man" (DNR 10.31; 200). This has to do with Hume's account of virtue and vice and the moral sentiments with which they are associated. In brief, we approve of those characters which contribute to the happiness of others and disapprove of those which tend to their unhappiness. Thus, the notion that "everything is right with regard to the WHOLE" is at odds with our recognition of a real distinction between virtue and vice (EHU 8.35; SBN 102).

Philo goes on in Dialogues 11 to respond to Cleanthes' suggestion that natural and moral evil might be reconciled with a deity conceived as both benevolent and powerful provided that the deity is only finitely perfect in these respects (DNR 11.1; 203).²² Such a being would have to allow some evil in order to avoid still greater evil. The essence of Philo's response consists once more in the suggestion that allowing for the consistence of a very powerful, wise, and benevolent deity with the imperfections of the world, still we are unable to make an inference to the existence of such a being from the world as we experience it (DNR 11.4; 205). This reflects a line of reasoning which runs throughout Philo's contributions to parts 10 and 11 of the Dialogues. On the one hand, Philo is unable to find any flaw in the arguments attributed to Epicurus. After providing a brief summary of these arguments he comments that "Nothing can shake the solidity of this reasoning" except our recognition that "these subjects exceed all human capacity" (DNR 10.34; 201). It is by faith alone that the conjunction of infinite benevolence with infinite power and wisdom can be discovered (DNR 10.36; 202). On the other hand, Philo goes on to acknowledge the possibility of solutions to the problem of regarding the world with its apparent misery and disorder as the workmanship of a sublime and benevolent being (DNR 11.2; 204). There may therefore be good reasons why providence does not intervene to prevent suffering; yet while this supposition "may be sufficient to save the conclusion concerning the divine attributes … surely it can never be sufficient to establish that conclusion" (DNR 11.8; 207; italics in original). Again, if the goodness of the deity could be established a priori then natural evil might in some unknown way be reconcilable with this position.²³ But while our experience of suffering may be compatible with this supposed attribute of the deity it cannot prove its existence (DNR 11.12; 211). This indicates that while Philo is unwilling to concede that the issue of consistency or compatibility has been resolved by any of the available theodicies, his principal concern is with the problem of evil as it bears on the attempt to establish by argument a posteriori the existence of a God who may be compared to human mind and intelligence (DNR 2.7–8; 144). In other words, he is concerned with the implications of human suffering for the attempt to establish by such argument that God exists as a being whose attributes may be compared with those of human minds.

3. The Problem of Moral Evil

The distinction between natural and moral evil is mentioned at various points in part 11 of the Dialogues (DNR 11.1, 11.10, 11.16; 203, 209, 212). No explicit account of this distinction is offered, but it seems clearly to be assumed that natural evil involves suffering which arises from causes associated both with "the economy of the animal creation" (DNR 11.6; 205) and also that of the physical world, while moral evil is a product of human vice (or "wickedness" as it is referred to in both parts 10 and 11).²⁴ While discussion in the Dialogues is principally focused on the topic of natural evil, it would clearly be of interest to consider what is said both here and elsewhere to indicate Hume's position on this alternative aspect of the problem of evil. So far as the Dialogues is concerned, Philo's few references to moral evil appear to suggest that it may be treated in much the same way as natural evil. Thus, he claims that both natural and moral evil largely result from a particular deficiency of human nature—a disposition to idleness—of which we could readily have been "cured" by an inconceivably powerful creator (DNR 11.10; 208–209).²⁵ According to Philo, then, this vice may be blamed for many, at least, of those other vices which give rise to actions from which human beings suffer at the hands of others. Having set out his response to the various causes of natural evil, Philo goes on to remark: "What I have said concerning natural evil will apply to moral, with little or no variation" (DNR 11.16; 212). In each case there are apparent implications for the moral attributes to be ascribed to the divine being: respectively, those of benevolence and rectitude. Philo qualifies what he has said only to the extent that there may be a special problem about moral evil in so far as it is more predominant over moral good than natural evil over natural good. Bearing in mind the feelings of approval and disapproval evoked in us by, respectively, moral good and evil, the question which then arises is how we can ascribe moral sentiments as we experience them to the supreme being.

It may at first appear surprising that the Dialogues contain no discussion of the distinctive issues associated with the problem of moral evil, that is, the problem of reconciling the suffering caused to human beings by the vicious actions of others with the existence of a just and omnipotent God. What, for example, of the traditional free-will defence on behalf of theism? In short, this consists in the claim that human beings have been created by God with the capacity for freely choosing between good and evil, and that the suffering caused by vicious actions is the responsibility of human beings thus exercising free choice. This apparently enables us to account for moral (and perhaps in some cases natural) evil while maintaining the existence of a perfectly good and all-powerful God. Hume was obviously well aware of this response to the problem of evil as witnessed, for example, by his early notes on King's Essay on the Origin of Evil.²⁶ More generally, the free-will defence is associated with philosophers like Samuel Clarke who are among the targets of Hume's critique of the notion of freedom of the will (in T 2.3.2; SBN 407–12), even if they are not mentioned by name. Quite apart from Hume's rejection of the idea that we are able to "feel that the will itself is subject to nothing" (T 2.3.2.2; SBN 408; cf. EHU 8.22n18; SBN 94n1), discussion in Section 8 of the first Enquiry makes clear Hume's own view of the theological implications of the relation between will and action (EHU 8.32–36; SBN 99–102). If we accept that this relation is a causal one then it appears that it is subject to the same "laws of necessity" which govern the operations of matter. In the case of material objects or events, we suppose that the relation of cause and effect applies throughout, even if there are many instances of "irregular events" whose causes we are unable to identify because we are ignorant of the "springs and principles" on which they ultimately depend. While the vulgar may be content to attribute the uncertainty of events to an uncertainty in their causes, philosophers adopt the maxim that "the connexion between all causes and effects is equally necessary, and that its seeming uncertainty in some instances proceeds from the secret opposition of contrary causes" (EHU 8.13; SBN 87). If we apply such considerations to the case of voluntary actions it would seem that there must be a "continued chain" of causes linking our volitions to God as the original cause of all (EHU 8.32; SBN 99).²⁷

This important line of argument generates a dilemma for the theologian: either voluntary human actions possess no wickedness, in so far as their ultimate cause is considered to be wholly good; or, if they are vicious, then God must bear the blame for them. In the context of the Dialogues, Philo presents an abbreviated version of the argument of the first Enquiry in raising the question of how the anthropomorphite is to account for vice:

You must assign a cause for it [vice], without having recourse to the first cause. But as every effect must have a cause, and that cause another; you must either carry on the progression in infinitum, or rest on that original principle, who is the ultimate cause of all things.

(DNR 11.17; 212)

The first horn of the dilemma, with its implication that human actions have no moral turpitude given the supposed goodness of their cause, is associated with the discredited view that the whole is ordered with perfect benevolence. Any such speculation is impossible to maintain in face of the sentiments of praise and blame which represent our natural response to certain actions and characters. If, however, we accept that human actions are in some instances blameworthy, we have also to recognise that there can be no extenuating circumstances in an ultimate cause which is supposed to differ from us in being infinitely wise and powerful. The problem that remains, of how God can be the ultimate cause of our actions without being the author of moral turpitude, is dismissed by Hume in the first Enquiry (with mock piety, perhaps) as one of those "sublime mysteries" which illustrate the limitations of natural reason (EHU 8.36; SBN 103).

There is of course a fundamental reason, even if is not spelled out in the Dialogues, why Hume would reject the free-will theodicy as a response to the problem of moral evil. This lies with his repudiation of the very notion of the will as something that is "free" in the sense of being uncaused. His discussion of this topic in the Treatise (T 2.3.1–2; SBN 399–412) and the first Enquiry (EHU 8; SBN 80–103) is sufficiently familiar that it is unnecessary to go into the details here. In the former context Hume takes himself to have established that "all actions of the will have particular causes" (T 2.3.2.8; SBN 412)—in particular, ones which belong not only to the circumstances of the agent, but also to such features as his motives and temper (T 2.3.1.4, 11; SBN 400, 403). Similarly, in the first Enquiry, having affirmed that "nothing exists without a cause of its existence" (EHU 8.25; SBN 95), Hume goes on to argue that necessity, as an essential part of the definitions of cause, belongs to the will of human beings as much as to objects themselves (EHU 8.27; SBN 97). We must accordingly reject any notion of liberty that reduces human actions to mere chance events by representing the will itself as being undetermined. The "free-will defence" is therefore to be rejected as the product of confusion regarding the notions of liberty and necessity. The only meaningful conception of liberty— "liberty of spontaneity" (T 2.3.2.1; SBN 407) or "hypothetical liberty" (EHU 8.23; SBN 95)—represents us as being free in so far as we are able to act or to refrain from acting in accordance with our choices or volitions. But our recognition that what we choose to do (or to refrain from doing) is itself the result of other causes pertaining both to ourselves and our circumstances, leads via the acceptance that nothing exists without a cause to the theological dilemma referred to above.

4. The Problem of Natural Evil

The principal topic of discussion in Dialogues part 11 is natural evil and the various circumstances from which it arises. In brief, Philo identifies four causes or circumstances on which the ills associated with nature depend, none of which appear to be necessary and unavoidable. These range from the occurrence of pain throughout the "animal creation," through the general laws of nature which leave us vulnerable to such "accidents" as tempests and ill health and the "great frugality" with which powers and faculties are distributed among both ourselves and animals, to the "inaccurate workmanship" of the springs and principles of nature as displayed not only by extremes of weather, for example, but also by defects of both body and mind (DNR 11.5–11; 205–210). Philo's position is that an inconceivably powerful creator would have been able to spare us the resulting ills without depriving us of any benefits that might be associated with them. While he also acknowledges that if it were possible to establish the goodness of the deity a priori then these various circumstances might in some way be reconciled with that attribute, Philo insists that there are no grounds for inferring goodness from the ills of the universe when it appears that they might so easily have been remedied (DNR 11.12; 211).

It appears, therefore, that even if the attributes of the deity are considered to be finite, human suffering still provides an obstacle to the anthropomorphic conception of divine benevolence. This leads Philo to consider Manichaeanism as a hypothesis which would attempt to account for the mixture of good and ill with which life presents us.²⁸ Philo rejects this hypothesis on the ground that the universe reveals to us uniformity and an agreement of parts rather than marks of the supposed combat of a malevolent and a benevolent being. The true conclusion, according to Philo, is that the original cause or source of all things is indifferent to the opposing principles—heat and cold, pain and pleasure, good and ill—which underlie the operations of nature (DNR 11.14; 212).²⁹ So far as the attributes of this cause are concerned, the most probable hypothesis is that they include neither goodness nor malice.³⁰ According to Philo, the case of moral evil leads to a similar conclusion: it leaves us no reason to infer rectitude in God as this quality is found in human beings (DNR 11.16; 212).³¹

Philo's position here might be thought to betray some inconsistency in so far as he apparently abandons the skeptical stance adopted in parts 1–10 of the Dialogues. The skepticism endorsed by Philo in these preceding parts is characterised at the outset as being of the "speculative" kind (DNR 1.9; 134). Given the limits of human reason even in regard to the data of ordinary experience, it seems impossible to rely on it when engaging with topics remote from common life and experience. Thus, according to Philo, when we speculate about such matters as the creation and formation of the universe, or "the powers and operations of one universal spirit," we can scarcely fail to recognise that we are dealing with objects which are beyond our grasp. So long as we remain under the influence of common life then it is difficult, if not impossible, to maintain our skepticism; but when we turn to the topics with which the Dialogues are concerned there is a kind of balance between the arguments of the theologian and those of the skeptic: one which represents the "triumph of scepticism" (DNR 1.11; 136). This is the spirit in which Philo responds to "experimental theism" by offering various alternative hypotheses regarding the origin of the natural world, arguing that experience "can afford no probable conjecture concerning the whole of things" (DNR 7.8; 177). The skeptical conclusion drawn by Philo is that "A total suspense of judgement is here our only reasonable resource" (DNR 8.12; 186–87). How, then, is it possible for Philo to endorse what might be described as the "indifference hypothesis" regarding the original source of things having previously insisted on the limits of reason in this context and the suspense of judgement to which it leads?³²

It must be admitted that this rather contentious issue can scarcely be resolved within the context of a more general discussion of Hume's treatment of the problem of evil in the Dialogues. In fact, a number of questions arise here. Is Philo's adoption of the indifference hypothesis to be taken at face-value? If so, is this inconsistent with Philo's position earlier in the Dialogues? And supposing that there is an inconsistency, how should it be accounted for? So far as the first question is concerned, it has been suggested that Philo is doing no more than to cast doubt on Cleanthes' attempt to reconcile both natural and moral evil with the existence of a first cause of the universe which is perfectly (if only finitely) good (Pike, 100). This interpretation, however, appears to be at odds with Philo's explicit endorsement of the indifference hypothesis as "the most probable" of those available and not merely one which provides a possible alternative to these other hypotheses. For similar reasons, it is difficult to accept that Philo has done no more in parts 10 and 11 than to argue that Cleanthes has failed to establish that we can have knowledge of the moral attributes of the Deity.³³ But accepting that Philo does mean to endorse a negative hypothesis regarding God's possession of such attributes, there remains the question of whether this represents a real change in his stance towards Cleanthes' experimental theism. One reason for supposing that this must be so is the fact that the mysticism adopted by Philo, with its emphasis on the "absolute incomprehensibility of the divine nature" (DNR 6.6; 172), seems clearly to be incompatible with his apparent commitment in part 11 to a hypothesis about the divine nature.³⁴ It may be true that Philo is warranted in adopting a different approach to the moral properties of the Deity as compared with the natural properties with which earlier parts of the Dialogues are concerned.³⁵ But this does not by itself remove the appearance of conflict between Philo's mysticism and his endorsement of the indifference hypothesis.

If there is a significant difference between Philo's position on the claims of theism about the moral qualities of God and his position on attempts to establish the existence of God as a designer of the universe, then it may be suggested that this reflects a difference in Philo's approach in parts 10 and 11 as compared with the earlier parts of the Dialogues. In brief, the latter difference is supposed to amount to the following. When dealing with Cleanthes' argument from design Philo is content to generate hypotheses which are at least as plausible as that of Cleanthes; but he does so in order to argue that the only reasonable response is to suspend judgement. But in part 11, where what is at issue is Cleanthes' hypothesis of a finite benevolent Deity, Philo is able to proceed "more scientifically" by showing that our experience is compatible with only one hypothesis regarding the Deity, namely, that of indifference.³⁶ Even if we accept this, however, the point remains that Philo's mysticism should prevent him from endorsing any hypothesis about the divine nature, whatever difficulties he finds in Cleanthes' attempt to maintain his anthropomorphic conception of God given the apparent counter-evidence provided by human suffering.

While this whole topic no doubt deserves extended discussion in its own right, I suggest that if Philo has, in part 11 of the Dialogues, exceeded his brief as a speculative skeptic, then a plausible explanation is provided by an exchange between Philo and Cleanthes which occurs in part 1. We have seen that while Philo concedes that skepticism of the speculative kind yields to the demands of common life, it triumphs by leading to a suspense of judgement when we engage in theological reasoning about the creation of the universe and the powers of an infinite spirit.³⁷ According to Cleanthes, however, the doctrine and practice of speculative skepticism are just as likely to diverge "in the most abstruse points of theory as in the conduct of common life" (DNR 1.12; 136). Assent will occur in proportion to the evidence even in "the most abstruse corners of science," as in the case of Newton's explication of the rainbow where we are dealing with a topic—the nature of light—which might be considered too refined for our comprehension. Cleanthes goes on to claim that the arguments employed in common life and in science are of a similar nature and contain the same evidence (DNR 1.16; 137). This is why, according to Cleanthes, Philo should be prepared to attend to the evidence when dealing with theological and religious issues rather than withholding assent simply because of the remoteness of such matters.

If Cleanthes is right then we should expect that in spite of Philo's official doctrine (speculative skepticism), and the suspense of belief to which it should lead, in practice he will assent to certain hypotheses which arise in a theological context in accordance with the evidence. This is precisely what happens in part 11 where the evidence itself—the existence of human suffering—appears beyond dispute (notwithstanding Cleanthes' remark in the previous part of the Dialogues that in order to support divine benevolence it is necessary "to deny absolutely the misery and wickedness of man," before he goes on to qualify this by claiming that natural and moral evil are compatible with the existence of a finitely perfect deity). In light of this evidence Philo identifies four hypotheses regarding the first causes of the universe which he evidently takes to be exhaustive.³⁸ Philo's assent to one of these hypotheses—that of the indifference of these "first causes" —is a result of eliminating the remaining alternatives on the basis of the available evidence. All this is in contrast to his treatment of Cleanthes' design argument, where he claims that the systems or hypotheses of cosmogony may be multiplied almost indefinitely (DNR 8.1; 182) and that the experiential data are unable to make any one of them more probable than the others (DNR 7.8; 177).

While this helps to explain Philo's adoption of the indifference hypothesis in part 11, the fact remains that he has here departed from the speculative skepticism which governs his contributions to parts 1–8 of the Dialogues and the associated commitment to a mystical incomprehensibility thesis regarding the divine mind. He is, on the whole, more careful in the final part of the Dialogues where he suggests that while the atheist may be forced to admit that the first principle of order in nature bears some "remote inconceivable analogy … to the economy of human mind and thought," the theist must equally allow that there is "a great and immeasurable, because incomprehensible, difference between the human and divine mind" (DNR 12.7; 218; italics in original). Philo does go on to suggest that we have reason to infer that "the natural attributes of the Deity have a greater resemblance to those of man, than his moral have to human virtues" (DNR 12.8; 219), so that while "the cause or causes of order in the universe probably bear some remote analogy to human intelligence" this analogy cannot be extended to any other qualities of the mind (DNR 12.33; 227; italics in original). But this still falls short of endorsing the indifference hypothesis adopted in the previous part of the Dialogues.

5. Hume and the Evidential Argument from Evil

It is natural to consider the relation between the argument from evil we find in the Dialogues and other forms of the evidential argument.³⁹ We should note that the evidential argument may be directly concerned with the question of the existence of God. The gist of the argument in this form is that evil provides counter-evidence to belief in the existence of God as traditionally conceived, or that it makes the existence of such a being unlikely or improbable. The argument, so understood, might also be taken as an attempt to establish that belief in the existence of God is, in light of the evidence, unreasonable. While it is evident that the argument as it occurs in the Dialogues has a more limited objective, concerned as it is with the nature (more especially, the moral nature) of God, it seems clear that some of the same kinds of philosophical issue arise in each case. How, for example, should we understand claims about what is improbable in this context? And how might theism respond to such claims? So far as the former question is concerned, it seems clear that the notion of probability involved here must be an epistemic one, in so far as it is concerned with the question of what it is reasonable to believe regarding the existence or nature of God. The position for which Philo argues is that the miseries of life undermine the attempt of experimental theism to establish the existence of a deity possessing moral as well as natural attributes—and, indeed, that our experience of evil warrants belief in the alternative "indifference" hypothesis.

It is a striking feature of this debate about God's moral nature in parts 10–11 of the Dialogues that it is focused on the issue of divine benevolence. When Philo first addresses the problem that human vice and suffering appear to raise for an anthropomorphite like Cleanthes, his list of the moral attributes ascribed to the deity also includes justice and mercy (DNR 10.24; 198). This, however, is one of just a relatively small number of references to the notion of divine justice or rectitude in the Dialogues (see also DNR 4.2, 10.29, 11.16, 12.8; 159, 199, 212, 219), in spite of the fact that this notion provides the basis for a number of theodicies which attempt to reconcile our sufferings with divine goodness. We might take as an important example, and one which would have been available to Hume, the idea developed by Butler of probation: that is, that our present life is a "state of trial" with regard to the future world (Butler, 131–38). From this point of view, the sufferings and temptations we encounter provide us with an opportunity to exercise virtue and resist vice. This, in turn, reflects the idea that God's moral governance has to do with more than producing happiness or preventing pain: we are to be rewarded or punished in accordance with the rule of distributive justice.⁴⁰ Whatever we make of Butler's position in particular, it points to a strategy that always seems available to theism in response to the evidential argument. In brief, this is to raise the possibility that whatever sufferings we undergo there may be some greater good which an omnipotent being could achieve only by permitting them to occur. And one such good might be to provide appropriate punishment for those cases in which we fail to resist vice. Even the sufferings of the apparently innocent might be explained in this kind of way in so far as original sin prevents any of us from being truly innocent.⁴¹

Butler's notion of divine justice seems clearly to depend on acceptance of a future life in which the distribution of reward and punishment can ultimately be achieved. If the present government of God is one in which virtue is usually rewarded with happiness and vice punished with misery, then this, according to Butler, is reason to believe that we shall be punished or rewarded hereafter (Butler, 98–109). More generally, our present life may therefore be considered a probation for a future one. As we have seen, however, this notion of a future state in which supreme justice will be exercised is condemned as an arbitrary supposition or hypothesis both in part 10 of the Dialogues (DNR 10.30; 199–200) and also in section 11 of the first Enquiry (EHU 11.17–23; SBN 138–42). But there is another factor to be taken into account here, and this is Butler's treatment of moral evil as something that God permits because it is a necessary consequence of a greater good: namely, human freedom. The system of reward and punishment by which we are divinely governed reflects the fact that we are free rather than necessary agents (Butler, 174–76). To the extent that our freedom extends to the choices we make between virtuous and vicious actions, and not merely to our ability to act accordingly, Butler appears to be committed to a version of the free-will defence against the problem of evil;⁴² and this, as we know, is something that Hume has independent reason to reject.

The evidential argument may invite us to choose between rival hypotheses regarding the explanation of human wickedness and suffering. In Hume's case, as we have seen, the choice lies between regarding this, on the one hand, as something that is permitted by a deity, as traditionally conceived, in light of the greater good that may thereby be achieved or the greater evil prevented; and on the other hand, regarding evil as something that reflects the indifference, or absence of moral concern, of any supreme being that may be responsible for our existence and that of the world around us. This kind of approach may be found in those versions of the evidential argument which are directly concerned with the implications of suffering for the very existence of God.⁴³ The suggestion may therefore be that in explaining the existence of evil we are faced with a choice between theism as a hypothesis and the "Indifference Hypothesis," where the latter denies that our circumstances are the result of the morally motivated actions of a supernatural being (even allowing for the possible existence of any such being). And a proponent of the evidential argument may then claim that the Indifference Hypothesis explains the facts about our experience of pain and pleasure better than theism does; while this, in turn, is to be understood as the claim that the antecedent probability of these facts is greater on the assumption that the Indifference Hypothesis is true than on the assumption that theism is true (Draper, 14). There is at least the hint of this approach in some of Philo's remarks in the Dialogues:

It must, I think, be allowed, that, if a very limited intelligence, whom we shall suppose utterly unacquainted with the universe, were assured, that it were the production of a very good, wise, and powerful Being, however finite, he would, from his conjectures, form beforehand a different notion of it from what we find it to be by experience; nor would he ever imagine, merely from these attributes of the cause, of which he is informed, that the effect could be so full of vice and misery and disorder, as it appears in this life.

(DNR 11.2; 203–204; italics in original)

Philo's essential point is that unless we are antecedently convinced of the existence of a supreme intelligence that is both benevolent and powerful, then the appearances of things will fail to provide any rational basis for belief in such a being. This would, of course, be disputed by theodicies of the kind already mentioned: for example, by appeal to the idea that suffering is an inevitable consequence of the divine gift of free will as something which has a value outweighing the evil of pain. But even waiving the objections which would lead Hume to reject a theodicy of this kind, there is still the problem of accounting for pain for which we are not morally responsible, quite apart from those apparently gratuitous pains suffered by non-human animals which presumably cannot be justified by reference to any possible moral value they might possess.⁴⁴

How, then, should we assess the nature of Hume's contribution to the debates associated with the evidential argument from evil? We have seen that Hume, in the Dialogues, is concerned in particular with the implications of human suffering for experimental theism with its reliance on experience as a source of evidence for the natural and moral properties of a divine creator. It is in this context that Philo's indifference hypothesis appears to undermine the position of a theist like Cleanthes who is apparently committed to establishing the goodness of God on the basis of experience. As Philo puts it, Cleanthes as a representative of this approach must "prove" such attributes of the deity as power and goodness from the mixed moral phenomena we encounter (DNR 10.35; 201). So far as the problem of moral evil is concerned, while this topic plays little part in the Dialogues it is evident that Hume would be committed to rejecting the free-will theodicy which attempts to find an explanation for human wickedness that would be consistent with its being permitted by an omnipotent deity of perfect rectitude. In fact, as we have seen, Hume finds in the appeal to the role of human volition in moral evil a further dilemma for theism, namely, that either against all appearances moral evil would have to be considered non-existent in view of the supposed goodness of the ultimate cause of things; or that, on the other hand, in so far as moral evil does exist, its ultimate cause must be considered responsible. Not only this, but Hume responds to what might be described as the "future state" theodicy both in Section 11 of the first Enquiry as well as in the Dialogues; and here, once again, there is a specific problem for experimental theism in appealing to experience as a source of knowledge of such divine moral attributes as that of justice. The dilemma it encounters in this case is that either virtue is rewarded and vice punished in this life, so that we do not need to rely on divine justice to rectify any anomalies in this respect; or, if virtue is not always rewarded or vice punished then we are deprived of any experiential grounds for ascribing justice or rectitude to the deity (EHU 11.22; SBN 141–42). It is also true to say, however, that Hume recognises the essential limitations of the argument from evil: in particular, that if we have independent reason (for example, on the basis of a priori argument) for accepting the existence of an all-powerful and perfectly good deity, then the problem of evil is thereby circumvented. (We may not know how a deity of this kind could permit human suffering, but we can be sure that there must be a reason why not to do so would result in the loss of a greater good or the existence of a greater evil).

Nevertheless, Hume's general philosophical perspective would suggest that theism is able to derive little comfort from this last consideration. In terms of the distinction established at the beginning of section 4 of the first Enquiry, the question of God's existence and nature belongs to the area of reasoning about matters of fact rather than reasoning about relations of ideas (cf. EHU 4.1; SBN 25). Thus, Demea's attempt in part 9 of the Dialogues to employ argument a priori to establish the existence of a first cause whose attributes are infinite, is met by the rebuff (on the part of Cleanthes) that it is absurd to attempt to demonstrate any matter of fact by such means (DNR 9.5; 189). The argument given for this by Demea echoes that provided by Hume in the first Enquiry: namely, nothing that is distinctly conceivable implies a contradiction; whatever we conceive to be existent we can also conceive as non-existent; there is therefore no being whose non-existence implies a contradiction; and consequently, there is no being whose existence is demonstrable (cf. EHU 4.2, 12.28–29; SBN 25–26, 163–64). This in turn emphasises the importance of the problem of evil for the only kind of theism which Hume appears to take seriously, where in order to establish the existence and nature of God as anything more than a matter of faith, appeal must be made to arguments derived from experience.

Conclusion

Hume evidently recognises the problem of evil as a threat to traditional theism and, in particular, to the ascription to God of moral attributes (like justice, mercy, and benevolence) in addition to the natural attributes of an original and all-powerful cause.⁴⁵ The difficulty of making any inference to the moral attributes of God on the basis of our experience of human pain and suffering bears directly on the issue with which the participants in the Dialogues appear principally to be concerned, namely, that of the nature, as opposed to the existence, of God. While it is important to distinguish these issues, we have seen that consideration of the kinds of attribute which may legitimately be ascribed to God has a crucial bearing on what is involved in accepting the existence of such a being. For what is at stake here is the choice between two very different conceptions of God: those of mysticism and anthropomorphism. If, as Hume apparently concludes in his discussions of the problem of evil, we are bound to reject the latter, then this has profound implications for the dispute between theism and atheism. Even if, as Philo concedes in part 12 of the Dialogues, the analogy of the works of nature to the products of human artifice allows us to infer a proportional analogy in their causes (DNR 12.6; 216–17), the question of whether the deity should be called a mind or intelligence becomes a merely "verbal controversy." The dispute between theist and atheist is essentially one about the terms in which we characterise the supreme cause: the former being forced to concede the "incomprehensible" difference between the human and the divine mind, and the latter being forced to concede that the ultimate principle of universal order bears some "remote inconceivable analogy" to human thought or intelligence (DNR 12.7; 218). The difference between theist and atheist becomes, essentially, one of emphasis.⁴⁶

There remains, however, a crucial distinction between the natural attributes of the deity—those which might be required to account for the works of nature, such as power and intelligence—and his moral attributes (in particular, his supposed benevolence and justice).⁴⁷ As we have seen, so far as the supposed moral attributes of God are concerned, the gist of Philo's position is that we have no basis for supposing that such a being would possess this kind of concern for us as inhabitants of the natural world.⁴⁸ This, indeed, is reflected in the absurdity, according to Philo, of ascribing human passions to the deity (DNR 12.31; 226), bearing in mind that our moral lives are, on Hume's view, inseparable from passion and sentiment. Once more we find a rejection of all but the most limited kind of anthropomorphism in favour of the emphasis of mysticism on the incomprehensibility of the divine nature.⁴⁹ To this extent the problem of evil, as presented by Philo, casts doubt on the possibility of providing a concept of God with any real religious content; and, in doing so, it also undermines Cleanthes' project in the Dialogues of finding in experience a basis for the claim that "the Author of nature is somewhat similar to the mind of man" (DNR 2.5; 143). In this respect, the problem of evil appears to represent for Hume as great an obstacle to the project of natural theology as any of the deficiencies in the arguments by which it seeks to establish the truth of the religious hypothesis.