Introduction
Hume begins his Natural History of Religion (NHR) with two questions:
As every enquiry, which regards religion, is of the utmost importance, there are two questions in particular, which challenge our attention, to wit, that concerning its foundation in reason, and that concerning its origin in human nature. (NHR Intro.)Footnote 1
An answer to the second question ‘is the subject of our present enquiry’ (NHR Intro.). As for the other question, whether religion has a foundation in reason, Hume says the answer is obvious:
The whole frame of nature bespeaks an intelligent author; and no rational enquirer can, after serious reflection, suspend his belief a moment with regard to the primary principles of genuine Theism and Religion. (NHR Intro.)
Of course, this answer isn't obvious at all. Hume's Philo of the Dialogues concerning Natural Religion delivers what is perhaps the most clever, thorough, and relentless critique of the argument from design ever articulated in the English language.Footnote 2 What, then, are we to make of this apparent endorsement?Footnote 3 I intend to persuade you that we should make very little of it.
Hume's seemingly peculiar treatment of the design argument is not indicative of a radical, or even modest, shift in his overall epistemic evaluation of it. Rather, his handling of it fits quite nicely with exactly what he purports to do in the NHR, that is, advance a social-scientific examination of the causes of religious belief.Footnote 4 Upon closer inspection, Hume's treatment of the design argument is hardly a puzzling distraction from the main argument of the NHR, but rather a fully integrated part of his examination of the psychology of believers.
In this article I argue that Hume's focus on the design argument is on its impact on the psychology of religious believers, and not its rational acceptability. Hume might suggest, perhaps even strongly, that if anything could serve as the rational foundation of religious belief, it is the design argument. However, he doesn't take up the task of actually making this case. He is far too focused on the real purpose of the NHR: a thorough examination of the social and psychological causes of religious belief. This article has two parts. The first part briefly explicates some recent interpretations of the NHR, specifically of Hume's treatment of the design argument. This helps to explicate some of the broader points of Hume's NHR and to situate the view I defend within the interpretative landscape. The second part advances a detailed textual analysis of Hume's treatment of the design argument in the NHR and argues that every mention of it advances primarily an observation about the psychology of religious believers, rarely and only superficially commenting on the rational acceptability of their beliefs.
Hume's NHR interpreted
What is the overall purpose of the NHR? The title piques one's curiosity – is it truly a natural history?Footnote 5 Does Hume engage in a new type of project or is it somehow continuous with his previous works?Footnote 6 Is it a work in social science, epistemology, or both? If both, what, roughly, is the distribution? There is little consensus on the answers to these questions.
P. J. E. Kail argues that the NHR as a whole is little more than a thinly veiled full-scale attack on the rationality of all religious belief, an integral part of a confederated strategy that includes the DCNR and some of Hume's essays on religion, and that his handling of the design argument in the NHR is a crucial tactic in achieving this epistemological end.Footnote 7 On the other hand, Lorne Falkenstein argues that Hume is solely concerned with the social and psychological processes by which religious beliefs are formed in the NHR and does not pursue a hidden anti-religious epistemological agenda.Footnote 8 Jennifer Smalligan Marušić, somewhere in between, argues that the NHR advances ‘serious epistemological claims, but not claims that purport to undermine the rationality of all religious belief’.Footnote 9 Instead, she claims, Hume surgically demonstrates the incompatibility of genuine theism and superstitious monotheism.
Kail, Falkenstein, and Marušić all agree that Hume insincerely or marginally endorses the design argument in the NHR, but they disagree about his motivation for doing so.Footnote 10 Falkenstein's ultimate aim is to show that the problems invited by assuming that the NHR has a critical epistemological purpose (that of attacking the rationality of theism) are of such a magnitude that they outweigh the interpretative benefits. His take on Hume's treatment of the design argument is straightforward. If the NHR is part of an attack on the rationality of genuine theism, Falkenstein argues, then it is difficult to make sense of Hume's declaration ‘that the design argument offers an ‘invincible’ proof for genuine theism’.Footnote 11 His endorsement, if sincere, makes no strategic sense.Footnote 12 Furthermore, Hume's sincerity is dubious given the open manner in which he criticizes the argument in the DCNR and the Enquiry. Why would Hume suddenly develop a fear of expressing views critical of the design argument having already laid out his cards elsewhere? Thus, the most plausible interpretation given the facts available is ‘that ‘Hume's remarks were insincere but motivated mainly by a desire to turn his readers’ attention away from the negative point that there has been no general consent to genuine theism, and towards the more positive claim that popular religions have been more superstitious in character’ (Falkenstein (Reference Falkenstein2003), 13). Hume simply did not want to engage his readers on the particular point of the design argument in the NHR, so he feigned agreement with the accepted view. Think of Hume's endorsement as a focusing device.
According to Kail, the conclusion of Hume's account of the causes of religious belief in the first three sections of the NHR ‘is suspension of belief’ because ‘the account ‘rationally destabilizes’ belief concerning the content it explains’ (Kail (Reference Kail2007), 194). Hume's account argues that religious belief is based on an epistemically destabilizing source ‘awareness of which provides a reason to suspend the belief thus sourced’ (ibid., 198). What's the source in question? As a response to the anxiety produced by unpredictable natural conditions and their ignorance of causes, humans posited gods. Because of the ‘universal tendency among mankind to conceive all beings like themselves, and to transfer to every object, those qualities, with which they are familiarly acquainted, and of which they are intimately conscious’ (NHR 3.2), those gods were anthropomorphic. Polytheism is built on a foundation of fear and anxiety and held up by the faulty pillars of anthropomorphism. Beliefs acquired because they remove our anxiety about the future display ‘motivated irrationality’ (Kail (Reference Kail2007), 199). Monotheism derives its core content, belief in an invisible intelligent power, from this destabilizing source. If the foundation of religious belief is faulty, believers who want to be rational must seek out a better one. According to Kail, Hume suggests that there may be some reasons ‘out there’, but they are ‘reasons which we know Hume undermines elsewhere’ (ibid., 197). Think of Hume's endorsement as a lure.
Marušić makes a compelling case that Hume pursued a very targeted mission of displaying the incompatibility of genuine theism and superstitious monotheism in the NHR. Unlike Falkenstein, she does take the NHR to have a critical epistemological purpose. Unlike Kail, she argues that Hume finds polytheism and monotheism epistemically problematic for different reasons. The problem with polytheism concerns the formation of belief based on anthropomorphizing: ‘the defect in the process by which polytheistic beliefs are formed occurs in our coming to believe in the ideas resulting from the anthropomorphizing tendency of the imagination, and not merely in our having the ideas’ (Marušić (Reference Marušić2012), 727). The problem with monotheism, according to Marušić, emerges in the following passage from the NHR:
It appears certain, that, though the original notions of the vulgar represent the Divinity as a limited being, and consider him only as the particular cause of health or sickness; plenty or want; prosperity or adversity; yet when more magnificent ideas are urged upon them, they esteem it dangerous to refuse their assent. Will you say, that your deity is finite and bounded in his perfections; may be overcome by a greater force; is subject to human passions, pains, and infirmities; has a beginning, and may have an end? This they dare not affirm; but thinking it safest to comply with the higher encomiums, they endeavor, by an affected ravishment and devotion, to ingratiate themselves with him. As confirmation of this, we may observe, that the assent of the vulgar is, in this case, merely verbal, and that they are incapable of conceiving those sublime qualities, which they seemingly attribute to the Deity. Their real idea of him, notwithstanding their pompous language, is still as poor and frivolous as ever. (NHR 7.1)
She focuses on the penultimate sentence. The epistemic problem with monotheism is not that the relevant beliefs aren't justified, as is arguably the case with polytheism. The monotheist, she infers from NHR 7.1, doesn't really believe what she professes to believe. ‘Rather,’ she claims, ‘Hume seems to hold that there is something wrong with the claims that the monotheist makes: His claims are not expressions of belief but the result of attempts to flatter a particular god’ (Marušić (Reference Marušić2012), 722).
Marušić doesn't think that the NHR puts pressure on contemporary religious belief by connecting it to the faulty foundation of polytheism, but she does think that it puts pressure on contemporary religious belief. It does so, she argues, ‘by exploiting fundamental differences between genuine theism and popular religion’ (ibid., 730). Genuine theism and superstitious monotheism are, she claims, in a ‘curious evidential situation’ (ibid., 731). To explain, she draws significantly upon the following passage from the NHR:
The doctrine of one supreme Deity, the author of nature, is very ancient, has spread itself over great and populous nations, and among them has been embraced by all ranks and conditions of men: But whoever thinks that it has owed its success to the prevalent force of those invincible reasons, on which it is undoubtedly founded, would show himself little acquainted with the ignorance and stupidity of the people, and their incurable prejudices in favour of their particular superstitions. Even at this day, and in EUROPE, ask any of the vulgar, why he believes in an omnipotent creator of the world; he will never mention the beauty of final causes, of which he is wholly ignorant: He will not hold out his hand, and bid you contemplate the suppleness and variety of joints in his fingers, their bending all one way, the counterpoise which they receive from the thumb, the softness and fleshy parts of the inside of his hand, with all the other circumstances, which render that member fit for use, to which it was destined. To these he has been long accustomed; and he beholds them with listlessness and unconcern. He will tell you of the sudden and unexpected death of such a one: The fall and bruise of such another: The excessive drought of this season: The cold and rains of another. These he ascribes to the immediate operation of providence: And such events, as with good reasoners, are the chief difficulties in admitting a supreme intelligence, are with him the sole arguments for it. (NHR 6.1)
According to this passage, she argues, Hume associates genuine theism with the design argument and superstitious monotheism with the denial of it. Instead, superstitious monotheists subscribe to divine providence, which is actually incompatible with the design argument. If God designed the world to operate in accordance with natural laws, as the design argument has it, then God's immediate operation in the world would occur outside of, and perhaps even in violation of, these laws. For Hume, she argues, ‘the argument from design rules out particular providence’ (Marušić (Reference Marušić2012), 732). Marušić claims that her interpretation allows us to take a neutral stance on Hume's remarks on the design argument. Perhaps he is sincere in his avowal; perhaps not.Footnote 13 What he really wants to do with the argument, she maintains, is to pit it against superstitious monotheism. This poses a dilemma for religious apologists:
If the design argument is really successful, then popular monotheism is shown to be false. On the other hand, if the religious apologist denies that the design argument is a successful argument for genuine theism, she seems to have thrown out the baby with the bathwater. (ibid., 735)
Marušić suggests that this schism puts pressure on the rational acceptability of popular religion.
Taking stock, there is consensus in the literature around the claim that Hume insincerely or marginally endorses the design argument, but there isn't consensus around what explains why he does so. Kail argues that Hume's insincere endorsement of the design argument is a trap for unwitting believers. He dangles it as a potentially stabilizing source having shown that the current foundation for religious belief is destabilizing, knowing full well that it, too, is faulty. Falkenstein argues that it is a focusing device meant to keep Hume's reader's attention on his socio-psychological claims about popular religion's foundation in superstition. Finally, Marušić argues that his marginal endorsement is a schism inciter, pitting superstition monotheism against genuine theism.
The focus on whether, to what extent, and why Hume endorses the design argument has had, I fear, a distorting effect on our overall appreciation of the broad arc of this important text. The NHR is a socio-psychological examination of the causes of religious belief and the design argument is only ever marginally endorsed as means to assess its efficacy in producing religious belief. Hume assigns it some epistemic merit only to argue that despite any merit it has, it isn't causally responsible for religious belief. I have no objection to calling Hume's initial evaluation of the design argument in the NHR an ‘endorsement’, where all that is meant is that it has some evidence in its favour and therefore some epistemic merit. I do, however, have serious concern that by focusing so narrowly on the epistemic status of the design argument we are losing sight of just how dedicated Hume is to his main objective in the NHR, showing that religious belief, in all of its manifestations, is rooted in the passions. The arguments I'll offer in the next section are complementary to Falkenstein's view. Like Falkenstein, I believe it is a mistake to assign a critical epistemological purpose to the NHR.Footnote 14 However, his arguments focus on making the negative case against assigning a critical epistemological role to the NHR. I will make the positive case for seeing Hume's treatment of the design argument as part and parcel of his overall socio-psychological explanation of religious belief.
The design argument in the NHR
My argumentative strategy is as follows: examine carefully each passage in which the design argument is mentioned in the NHR and persuade you that Hume's primary purpose in bringing it up is to explain that it doesn't play an important role in producing religious belief. That is, Hume's motivation for bringing it into the discussion is to assess its status as a potential socio-psychological cause of belief. Let's start with the Introduction:
The whole frame of nature bespeaks an intelligent author; and no rational enquirer can, after serious reflection, suspend his belief for a moment with regard to the primary principles of genuine Theism and Religion. (NHR Intro.)
This is Hume's response to the question, ‘what is religion's foundation in reason?’ But read carefully. He claims that there is evidence in favour of design, ‘the whole frame of nature bespeaks an intelligent author’, but he does not claim that this evidence is decisive. He simply offers that there is evidence for design; that evidence might very well be defeasible. We know that in the DCNR he does consider counter-evidence and competing hypotheses, so it is plausible that Hume meant only to assign some marginal degree of epistemic justification to the design argument. So far, my take on this passage is compatible with all of the aforementioned.
However, the remainder of the passage has been taken by nearly every commentator who has considered it to mean that a rational enquirer cannot disbelieve that the primary principles of genuine theism include the design argument.Footnote 15 This passage seems to refer to the first question Hume delineated – what is religion's foundation in reason? – and it sounds like an endorsement – genuine theism, as opposed to bad superstition, is rooted in the design argument. However, this passage can be plausibly interpreted as a descriptive claim about what causes and sustains a particular variety of religious belief or about what adherents to a certain belief system regard as being foundational or constitutive principles. In this descriptive sense, Hume might think that the design argument is among the principles of genuine theism. This is compatible with Hume thinking that the principles of genuine theism, while explanatorily efficacious or constitutive of a belief system, aren't rationally justified. One could certainly claim, for example, that racial purity was a primary principle of Nazism and that the idea of racial purity explained membership in and was a core doctrine of that belief system, and at the same time reject the principle. So, allowing that Hume thinks that the design argument is the foundation of genuine theism and that there is some evidence for it is compatible with Hume ultimately finding the design argument rationally unacceptable. When he writes that no rational enquirer can suspend his belief with regard to the primary principles of genuine theism, he doesn't mean that he can't suspend his belief about the truth of the principles of genuine theism. Instead, Hume is just suggesting that he can't suspend his belief about what the principles of genuine theism are.
Another way of interpreting this passage is to take Hume as leaving open the content of the principles. When he writes that no rational enquirer can suspend his belief with regard to the primary principles of religion he leaves a hostile reader in a position to assume that he means the design argument, but a more careful reader might conclude that the principles are precisely what Hume claims they are in the main text of the NHR, the passions! This passage might reasonably be interpreted as Hume's way of cleverly answering the ‘first question’. What is religion's foundation in reason? It has no foundation in reason, if foundation is understood in the common way to mean the lowest, or first, part of an edifice. Religion is founded on the passions. Perhaps there are arguments that can be deployed to support it antecedently, but its foundation is the passions, and Hume thinks that no rational enquirer can disbelieve this fact.
The next passage considers the possible causal impact of the design argument on early believers:
Nothing could disturb this natural progress of thought, but some obvious and invincible argument, which might immediately lead the mind into the principles of pure theism, and make it overleap, at one bound, the vast interval which is interposed between the human and the divine nature. But though I allow, that the order and frame of the universe, when accurately examined, affords such an argument; yet I can never think, that this consideration could have an influence on mankind, when they formed their first rude notions of religion. (NHR 1.5)
‘But though I allow’ might be read as ‘though I'll allow for the sake of argument that the design argument might serve as an obvious and invincible argument for theism’.Footnote 16 The point being argued is whether the design argument could plausibly influence the belief-forming mechanisms of early theists. Hume's position is that, even granting that the design argument is rationally compelling, it is unlikely that it did. Here, as in subsequent passages, Hume marginally endorses the design argument only to set up his real point: the design argument is ineffective in producing religious belief.
The next passage is perhaps the strongest evidence in favour of interpreting Hume's treatment of the design argument as a sincere endorsement:
Were men led into apprehension of invisible, intelligent power by a contemplation of the works of nature, they could never possibly entertain any conception but of one single being, who bestowed existence and order to this vast machine, and adjusted all its parts, according to one regular plan or connected system. For though, to persons of a certain turn of mind, it may not appear altogether absurd that several independent beings, endowed with a superior wisdom, might conspire in the contrivance and execution of one regular plan; yet is this a merely arbitrary supposition, which even if allowed possible, must be confessed neither to be supported by probability or necessity. All things in the universe are evidently of a piece. Everything is adjusted to everything. One design prevails throughout the whole. And this uniformity leads the mind to acknowledge one author; because the conception of different authors, without the distinction of attributes or operations, serves only to give perplexity to the imagination, without bestowing any satisfaction on the understanding. (NHR 2.2)
Hume suggests that once a person thinks about the order of nature she ‘could never possibly’ consider any explanation except that a supreme being designed the universe and discusses at length how obvious it is that the universe exhibits order. This makes a strong case, but this passage must be interpreted in conjunction with the paragraph that follows it, where Hume explains that the first believers would not have been moved to religious belief by an argument from design. Hume goes on to explain the true cause of their belief:
On the other hand, if leaving the works of nature, we trace the footsteps of invisible power in the various and contrary events of human life, we are necessarily led into polytheism and to the acknowledgement of several limited and imperfect deities. Storms and tempests ruin what is nourished by the sun. The sun destroys what is fostered by the moisture of dews and rains. War may be favourable to a nation, whom the inclemency of the seasons afflicts with famine. Sickness and pestilence may depopulate a kingdom, amidst the most profuse plenty. The same nation is not, at the same time, equally successful by sea and land. And a nation, which now triumphs over its enemies, may anon submit to their more prosperous arms. In short, the conduct of events, or what we call the plan of a particular providence, is so full of variety and uncertainty, that, if we suppose it immediately ordered by intelligent beings, we must acknowledge a contrariety in their designs and intentions, a constant combat of opposite powers, and a repentance or change of intention in the same power, from impotence or levity . . .
We may conclude, therefore, that, in all nations, which have embraced polytheism, the first ideas of religion arose not from a contemplation of the works of nature, but from a concern with regard to the events of life, and from the incessant hopes and fears, which actuate the human mind. (NHR 2.3–4)
Notice the sharp turn in the third paragraph of Section 2: ‘On the other hand . . .’. This turn of phrase is commonly deployed to communicate that the author will now consider counter-evidence, which is exactly what Hume does. He is contrasting two lines of reasoning, one in support of design and one in support of chaos, without explicitly endorsing either one of them. Furthermore, we might regard Hume's use of ‘could not possibly’ as referring to psychological necessity. And, when he writes that ‘persons of a certain turn of mind’ might not find it absurd that the design argument actually supports polytheism, the ‘turn of mind’ of which he writes is a rational state of mind. He remains neutral in this passage on which position is epistemically superior. Although he suggests that the polytheistic take on the design argument is not better than the monotheistic take (he calls the supposition arbitrary and claims that it isn't supported by probability or necessity), he doesn't explicitly claim that it is any worse. He claims, descriptively, that certain considerations lead the mind to certain beliefs – a psychological process – and that certain other beliefs perplex the imagination and give no satisfaction to the understanding – again, a claim about human cognition, not about rational acceptability.
The next mention of the design argument is nestled in an explanation of superstition and the tendency to anthropomorphize deities:
Ignorant of astronomy and the anatomy of plants and animals, and too little curious to observe the admirable adjustment of final causes; they remain still unacquainted with a first and Supreme Creator, and with that infinitely perfect spirit, who alone, by his almighty will, bestowed order on the whole frame of nature. Such a magnificent idea is too big for their narrow conceptions, which can neither observe the beauty of the work, nor comprehend the grandeur of its author. (NHR 3.4)
As in NHR 2.2, Hume explains that it is implausible that ignorant early believers would have had the requisite background to comprehend the concept of fine-tuning and to deploy it as a premise in an argument in favour of intelligent design. Instead, they assume that their gods are more like humans than the Abrahamic omni-god associated with the design argument. Again, the main point concerning the design argument is that it is ineffective in producing religious belief.
Let's consider the next two passages that mention the design argument together:
Whoever learns by argument, the existence of invisible intelligent power, must reason from the admirable contrivance of natural objects, and must suppose the world to be the workmanship of that divine being, the original cause of all things. (NHR 5.2)
But whoever thinks that it has owed its success to the prevalent force of those invincible reasons, on which it is undoubtedly founded, would show himself little acquainted with the ignorance and stupidity of the people, and their incurable prejudices in favour of their particular superstitions. (NHR 6.1)
Notice that Hume refers to ‘Whoever learns’ in NHR 5.2 and ‘whoever thinks’ in NHR 6.1. It is reasonable to conclude that Hume makes the subject purposefully nebulous in these paragraphs. By referring to ‘whoever’ is compelled by the design argument, Hume leaves open the question of whether one should be, all things considered. He is simply outlining what one would be committed to if one were compelled by the design argument. We might also seize upon Hume's use of ‘suppose’ in the first of the two quotations. He writes that a design argument proponent would have to suppose, a psychological term, that the world is the craftsmanship of a divine being, and not that she would have to know or be justified in believing it.
In the second of the two quotations, ‘on which it is undoubtedly founded’ is a problematic clause for me, as the scope is unclear. I propose that we read it as claiming that whoever thinks that the design argument is founded on invincible reasons is ignorant of common people's stupidity and prejudice. I see ‘on which it is undoubtedly founded’ as coming out of the design argument proponent's mouth. But one might argue that Hume is speaking in his own voice in this clause. In response to this I offer two thoughts. First, if the text itself doesn't offer us a compelling reason for choosing one interpretation of the clause over the other, then we have, at worst for me, a draw. Second, if my analyses of the other texts in which the design argument is discussed in the NHR are compelling, then the text does supply us with a reason to prefer my reading of the clause.
The design argument is mentioned a second time in Section 6:
Thus [the inventions of new strains of adulation] proceed; till at last they arrive at infinity itself, beyond which there is no farther progress: And it is well, if in striving to get farther, and to represent a magnificent simplicity, they run not into inexplicable mystery, and destroy the intelligent nature of their deity, on which alone rational worship or adoration can be founded. (NHR 6.5)
This passage seems to claim that one could found rational worship on an intelligent deity, a relatively strong epistemic claim. But we could plausibly interpret Hume as tipping his hand regarding the relative merits of arguments for monotheism. If anything could serve as a foundation for theism, it would be the design argument, but this is not to say that it does. We might also read ‘rational’ in ‘rational worship’ as ‘having some foundation in reason’, which leaves open the possibility that, all things considered, worship founded on the design argument is not supported by the best available reasons. Taken this way, the passage advances a relatively weak epistemic claim: of the arguments available for monotheism, the design argument offers the only defensible foundation, but this does not entail that it is indefeasible. Besides, what Hume is really describing here is the tendency of the passions to subdue reason. He is observing that even if rational worship were founded on the design argument, it cannot withstand, psychologically speaking, the force of the passions.
The design argument is not mentioned again until Section 15:
Though the stupidity of men, barbarous and uninstructed, be so great, that they may not see a sovereign author in the more obvious works of nature, to which they are so familiarized; yet it scarcely seems possible, that anyone should reject that idea, when once it is suggested to him. A purpose, an intention, a design is evident in every thing; and when our comprehension is so far enlarged as to contemplate the first rise of this visible system, we must adopt, with the strongest conviction, the idea of some intelligent cause or author. The uniform maxims too, which prevail throughout the whole frame of the universe, naturally, if not necessarily, lead us to conceive this intelligence as single and undivided, where the prejudices of education oppose not so reasonable a theory. Even the contrarieties of nature, by discovering themselves every where, become proofs of some consistent plan, and establish one single purpose or intention, however inexplicable or incomprehensible. (NHR 15.1)
This is a tough passage, as Hume makes some modal claims that are difficult to parse: ‘it scarcely seems possible’ that someone could reject the argument upon first hearing it and ‘we must adopt’ and with ‘the strongest conviction’ belief in an intelligent cause. However, these claims can be plausibly interpreted as concerning the psychology of believers. A belief can be psychologically irresistible without being logically or rationally compelling. Given human nature, we are forced to accept certain beliefs regardless of their rational acceptability. Just as early humans were forced to adopt polytheistic beliefs in response to the unpredictability of their environment, some modern humans, given their background beliefs and education, must adopt certain beliefs when they consider the design argument, at least for some time. This fits nicely with the second half of the passage. When Hume writes that uniformity ‘naturally, if not necessarily, lead[s] us to conceive’ certain beliefs, he seems to be claiming that belief in design is, at least for some, a psychologically natural position, but not a logically necessary one. Finally, the last sentence counts strongly against interpreting this passage as a sincere endorsement of the rational acceptability of the design argument: ‘Even the contrarieties of nature, by discovering themselves every where, become proofs of some consistent plan, and establish one single purpose or intention, however inexplicable or incomprehensible.’ Here Hume claims that we are psychologically compelled to take evidence that should count against design as evidence for it. From a rational point of view, this behaviour is ‘inexplicable’ and ‘incomprehensible’, but some agents find it psychologically nearly irresistible.
The last occurrence of the design argument in the NHR is by far the most sarcastic treatment of it:
What a noble privilege is it of human reason to attain the knowledge of the Supreme Being; and from the visible works of nature, be enabled to infer so sublime a principle as its supreme Creator? But turn the reverse of the medal. Survey most nations and most ages. Examine the religious principles, which have, in fact, prevailed in the world. You will scarcely be persuaded, that they are anything but sick men's dreams: Or perhaps will regard them more as playsome whimsies of monkies in human shape, than the serious, positive, dogmatical asseverations of a being, who dignifies himself with the name of rational. (NHR 15.6)
The main point here is really that religious belief isn't, in most cases, explained by the workings of a rational argument. That's not what explains why believers believe, and he makes the category of believers fairly deep and broad. If we look at most nations and most ages, the most prevalent belief-explaining principle is not reason. He isn't here claiming that the design argument supports a kind of rationally superior belief. Rather, he is claiming that one would be hard pressed to find any religious beliefs with an initial causal relationship to the design argument. Again, the design argument is ineffective in producing religious belief.
That concludes our textual analysis of Hume's treatment of the design argument. When we line up all the passages wherein Hume discusses the design argument, his primary aim is to set it up for assessment as a potential belief-producing mechanism. His assessment consistently reveals that it is ineffective.Footnote 17 Hume may marginally endorse the argument, but probably only to incorporate it into his overall examination of the socio-psychological causes of belief. It may also be the case that in associating it with reason, Hume is primarily concerned to explain its origin – rather than deriving from the imagination, the design argument is primarily a product of the faculty of reason. We can use our faculty of reason to produce arguments that are irrational for us to believe, so a claim's origin in reason doesn't guarantee its rationality, permissibility, acceptability, or compulsory force. Taken this way, Hume's treatment of the design argument is oriented towards explaining its origin, the faculty of reason, which is exactly what we should expect to see in the NHR.
Hume does make some epistemic claims about the relative merits of the design argument, but these claims are fairly weak. He writes of it as though it is the best of the options available, but, of course, this leaves open the possibility that he finds none of the options to be, all things considered, rationally compelling. The design argument is perhaps the most rationally compelling of all of the arguments for monotheism, and this shouldn't surprise us – Hume would find an a posteriori argument more compelling than an a priori argument. And Hume might grant this in a particularly generous way for precisely the reasons Falkenstein and Kail offer. However, the NHR gives us no reason to believe that Hume sincerely or strongly endorses the design argument as providing a rationally compelling foundation for monotheism. Further, Hume can recognize the design argument as having some foundation in reason and still maintain that the passions bear the primary responsibility for religious belief; and Hume can maintain that the design argument has a foundation in reason, but also maintain that religious belief has no foundation in the design argument. As he writes in the Treatise, ‘Reason is, and ought only to be the slave of the passions’ (T 2.3.3.4).Footnote 18 Whatever the merits of the design argument, and we have reason to think Hume finds them to be outweighed, it isn't a foundation for monotheism for two reasons. First, whatever its rational basis, there is a compelling case that Hume finds it defeasible. Second, the design argument simply is not the kind of thing that can serve as a foundation of religious belief because it isn't efficacious by itself in producing or maintaining religious belief. Hume maintains, despite what his clever language in the Introduction might induce us to believe, that religion has no foundation in reason. This is the ‘most obvious, at least, the clearest solution’ (NHR Intro.) to the question, ‘what is religion's foundation in reason?’Footnote 19
