Adam Smith on Religion

Samuel Fleischacker for AdamSmithWorks


August 5, 2020
What were Adam Smith’s religious beliefs? Scholars disagree vehemently on this subject. Some say he was an atheist, or at least a radical religious skeptic like his friend David Hume. Others think he believed in some sort of God but moved away from Christianity as he got older. And still others think he was a traditional Christian all his life.1

The great differences among these views may in part reflect the religious or anti-religious commitments of the scholars who hold them. But they also reflect real difficulties in reading Smith. On the one hand, Smith refers to God throughout his Theory of Moral Sentiments (TMS), and included a footnote praising the Christian doctrine of vicarious atonement for sin in the first five editions of that book. On the other hand, Smith does not refer to God in his Wealth of Nations (WN), cut the passage on atonement from the last edition of his book, and criticizes Christian clerics in both books. He also wrote an effusive encomium to Hume upon the latter’s death, as a model of the “wise and virtuous man,” which brought upon him the fury of religious people who thought that an unbeliever could not possibly be virtuous.2

What should we make of all this? I propose a reading of Smith that can I think reconcile everything he said about religion.

To begin with, Smith never attempts to prove that God exists. Although he taught a course on natural theology at Glasgow, no record of those lectures has survived and Smith seems never to have tried to publish any part of those lectures. This last point is significant: every other subject Smith taught, from logic and belles-lettres to morality, jurisprudence and political economy, became the subject of a manuscript that he at least planned to publish. Natural theology alone seems not to have interested him enough to write about it, and there are only a handful of passages in his published writings that so much as allude to it. In that handful of passages, moreover, Smith restricts himself to describing what ancient peoples thought about the gods or what medieval universities taught about God, without endorsing any of these beliefs.

Neither in these passages nor anywhere else does Smith so much as mention the ontological or “first cause” arguments for God’s existence. At various points in TMS, Smith does seem to take for granted the truth of the design argument for God, however. He tells us, among other things, that the fact that we treat violations of justice as more serious than violations of beneficence reflects the wisdom of God rather than the wisdom of man (TMS II.ii.3.5), that the fact that we are naturally inclined to punish one another only for actions, not for unfulfilled intentions, “demonstrates the providential care of [nature’s] author” (II.iii.3.2), and, famously, that “When Providence divided the earth among a few lordly masters, it neither forgot nor abandoned those who seemed to have been left out in the partition” (IV.i.10): that the rich are led as if by “an invisible hand” to distribute much of their wealth to the poor. All of these claims suggest that we can see signs of divine design in nature. And Smith occasionally says things along these lines. “In every part of the universe we observe means adjusted with the nicest artifice to the ends which they are intended to produce,” he tells us, and he identifies this adjustment of means to ends with “the wisdom of God.” (II.ii.3.5). A bit later, he says explicitly that “every part of nature, when attentively surveyed, equally demonstrates the providential care of its Author.” (II.iii.3.2).

These remarks certainly lend credence to those who read Smith as a theist. Still, we need to be careful about inferring from Smith’s avowal of design in nature, in passages of this sort, that anything in his views depends on a belief that God designed nature. For in every case in which Smith speaks of God as the Author of nature, he also provides a wholly unmysterious, secular account of the feature of nature he is discussing. If we did not give justice priority over beneficence, our societies would not survive, Smith argues (II.ii.3.3-6). He also gives good naturalistic reasons for why we do not punish unfulfilled intentions (II.iii.3.2-3), and why the rich distribute much of their wealth to the poor (IV.i.10). At no point does he hold up a feature of our nature, or of nature in general, and ask how it could possibly have come into existence unless an intelligent being designed it.

But that is the characteristic move of those who try to argue to God’s existence from some feature of nature. Smith’s teacher Frances Hutcheson, for instance, had devoted an extended section of his first book to an attempt to show that it is “next to an absolute strict Impossibility” that beings as complex as plants and animals could come about without a designer. Nothing like this argument appears anywhere in Smith. Never does Smith say, “Look at the intricacy or fittedness of means to ends in this flower or aspect of human nature — surely that means that an omniscient and benevolent Being designed it.” Instead, writing for an audience that already accepted the argument from design, Smith makes use of that belief to help underwrite his claim that various features of human nature must have some good purpose. At the same time, he explains these good purposes on the basis of empirical facts that believers and unbelievers alike can recognize. So nothing he says in this regard requires his reader to believe in a benevolent Designer of nature, nor does he say anything that would convince readers to believe that nature must have a designer, if they did not already accept that idea.

Indeed, many of the passages in which Smith speaks of purposes in nature can be read as proto-evolutionary explanations, rather than appeals to a divine designer. Smith, along with Hume, has been claimed as an ancestor of Charles Darwin, someone to whom Darwin looked for a source of his idea that order could come about in nature without anyone planning it. And there clearly are elements in Smith’s account of the apparent purposiveness of nature that fit well with Darwinian modes of explanation. For instance, Smith tells us at several points that societies would not survive without certain features of human nature (II.ii.1-4, V.2.16). That is significantly different from saying, as would an Aristotelian or a Christian influenced by Aristotle, that we would not reach our highest moral or intellectual perfection, the final cause to which we aspire, without the feature of human nature in question. Of course Smith did not himself lay out Darwin’s theory of evolution. But Smith’s views do not depend on a teleological framework, and he goes to some lengths to show how one can accept his accounts of human nature without endorsing any such framework.

We should add that in WN there is not a single reference to an “Author of nature” nor a single case in which purposes or intentions are attributed to nature. Nor does the word “God” ever appear in the singular (Smith does talk twice about ancient beliefs in “the gods”), and the word “Deity” appears only in discussions of the place of theology in university curricula. Some scholars have used this fact as evidence that even if Smith believed in God when he wrote TMS, he abandoned that belief as he got older. I don’t think this is correct. When revising TMS for its final edition, a dozen years after publishing WN, not only does Smith keep almost all of the teleological language of the early editions, but he adds new passages using that language (VI.iii.30, for instance). It is nevertheless notable that Smith explicitly gives only secular, naturalistic explanations of the social structures he describes in WN (including the achievements he attributes to free markets), even when they might appear designed.

To sum up, Smith avoids engaging in natural theology, and never gives logical or metaphysical or scientific reasons for belief in God.

But that is not to say that Smith refrains from giving any reason for belief in God. Logical and metaphysical and scientific reasons are not the only kinds of reason. There are, for instance, moral reasons. Smith could for instance say that we should believe in God because only God’s commands can tell us how to be virtuous, or only God’s reward and punishment can keep us virtuous. Or he could say that morality needs no religious belief to sustain itself, but that it makes best sense within a religious framework. Smith nowhere makes the first of these moves, but there are several passages in which he makes the second. There are passages, that is, in which Smith anticipates Immanuel Kant’s famous argument that, while morality does not depend on religion, it can lead us into religion.

For Smith, the workings of our sentiments give rise to moral standards, and motivate us to live up to such standards, whether or not we believe in a God; no religious source is needed for us to learn morality nor is any religious sanction needed for us to heed it. But in “the natural course of things,” unjust and indecent people win out over just and decent ones. It is hard to countenance this fact unless the universe somehow compensates for such injustice. So our very moral impulses, Smith says — “the noblest and best principles” in human nature — lead us to believe that “the great Author of our nature ... will complete the plan which he himself has taught us to begin” and establish for us “a future state,” in which justice and decency are rewarded (III.5.10). People who lack such a belief need not be immoral — they may indeed, like David Hume, be models of virtue — but they will be prone to the “melancholy ... reflection” that they live in “a fatherless world.” By contrast, people who believe that God administers and directs the “great society of all sensible and intelligent beings” can see themselves as foot-soldiers in God’s army, when they act morally, carrying out God’s will on “the forlorn station of the universe” (VI.ii.3.3-4); they can see themselves as “co-operat[ing] with the Deity” (III.5.7). These reflections make overall sense of the project of being moral: they give morality a place in the universe, represent the universe as hospitable to morality. Religion is thus not necessary to morality, but religious reflections are a natural outgrowth of carrying out a moral life, and help support our moral commitments. “Religion enforces the natural sense of duty,” says Smith; those who believe in it “act under an additional tie, besides those which regulate the conduct of other men.” (III.5.13). For Smith, as for Kant, a moral faith in God is a reasonable consequence of a commitment to virtue, and a laudable addition to the other, purely secular reasons sustaining that commitment.

Note that in the passages I have just been discussing, references to God are not merely an extra, added on to otherwise naturalistic explanations. Elsewhere, Smith says that “Nature” or “the Author of Nature” intends such-and-such an end by way of a certain moral sentiment, but he also gives us a wholly naturalistic case for endorsing that sentiment which the reader can accept even if he or she does not believe in God. In the chapters I have just cited, however, Smith’s whole point is to show the importance of belief in God to our moral lives. Belief in God is crucial to our ability to see the natural order as a potentially just and decent one, Smith argues here; only if the natural order is nested within and complemented by a supernatural one do our moral actions contribute to a good universe. There is no way of re-stating this point without mentioning God. In these sections of TMS, Smith is arguing for religion, not merely using religious terms to present a point that he could put without religious coloring.

Note also that the reasons Smith gives us for belief in God support just a religion in which worship consists in moral action. If morality is something we can determine and follow for purely secular, naturalistic reasons, and if we come to a faith in God only as an outgrowth of our commitment to morality, then there can be no reason to suppose that God would demand of us non-moral rituals, or a belief in doctrines that are irrelevant to morality. And Smith, like Kant, in fact sees such rituals and doctrines as distracting people from the moral core of religion. He says that “the first duty” of religion should be “to fulfil all the obligations of morality,” and that “the natural principles of religion are ... corrupted by ... factious and party zeal” where people are taught instead “to regard frivolous observances, as more immediate duties of religion, than acts of justice and beneficence.” (TMS III.5.13) He rails against anyone who represents “the public and private worship of the Deity ... as the sole virtues which can either entitle to reward or exempt from punishment in the life to come.” (III.2.34). He notes that religion can be used to invert the notions of duty that ought to guide us — giving, as an example, people who are told to kill a virtuous man as an enemy of their religion — and says that “false notions of religion are almost the only causes that can occasion any very gross perversion of our natural sentiments in this way.” (III.6.12) And he calls for the doctrines of as many religions as possible to be reduced to a “pure and rational religion, free from every mixture of absurdity, imposture, or fanaticism” (WN V.i.g.8).

These views help support Smith’s harsh critique of clericalism in WN. The medieval Catholic church constituted, he says, “the most formidable constitution that ever was formed against the authority and security of civil government, as well as against the liberty, reason, and happiness of mankind.” (WN V.i.g.24). Catholics are, however, not his only target. Pastors elected by the people of their parish in Calvinist churches are “the most factious and fanatical of [their] order,” he says (WN V.i.g.36): the democracy of presbyterian clergy tends, ironically, to foster factionalism. At the same time, Smith tells us that the modest salaries that presbyterian clerics receive lead them in general to be “learned, decent, independent and respectable.” Why? Because “nothing but the most exemplary morals can give dignity to a man of small fortune.” (V.i.g.37-8). So clerics win Smith’s respect only when they represent a general human morality; the distinctively Christian rituals and doctrines they teach tend more to obstruct than to promote the true, moral religion. The less they represent Christianity’s distinctive rituals and doctrines, the more they represent just “the natural principles of religion,” the more, for Smith, they are to be praised. These views, too, align with Kant’s moral faith.

The moral faith reading I am proposing of Smith can, I think, account for everything he says about religion. Consider, for instance, the passage on atonement in the first five editions of TMS. Smith says there that the idea that we need some vicarious “intercession, ... sacrifice, ... atonement” to compensate for our wrongdoing in the eyes of a supremely good Being arises from our “natural sentiments” even without revelation: he endorses “the doctrines of revelation” because they “coincide, in every respect, with those original anticipations of nature.” (II.ii.3). Even here, then — even in the most Christian passage he ever wrote — Smith appeals to revelation only when he finds the ideas in it compatible with a purely naturalistic moral faith. That implies that supposedly “revealed” texts are actually human attempts to approximate the teachings that a moral God would want us to believe: and that all such texts, including the Christian Bible, should be re-interpreted in a moral light if they seem to teach anything else.

As for the passages in TMS that talk of an “Author of nature,” or impute benevolent and intelligent intentions to “Nature,” they too can be fitted in nicely with a moral faith. For once we have moral reasons to believe that an intelligent and benevolent Being has structured the universe, it is reasonable to look for signs of that Being’s designs in the natural world. We have no guarantee that we will find them, and even when we do, we should not think that they prove that God exists, but a moral faith nevertheless gives us reason to take up a teleological approach to nature. “We may admire the wisdom and goodness of God even in the weakness and folly of man,” says Smith (II.iii.3.2): we may admire the wisdom and goodness of God, even in folly, if we approach “every part of nature” from the perspective of a moral faith. Science alone will not force us to take up a teleological approach to nature. Nor do the teleological explanations we come up with count as evidence for God’s existence. But once we already assume that God exists, it makes sense to seek ways in which God’s designing hand may be manifested in natural phenomena. And it makes especially good sense to seek signs of divine intent in aspects of human nature that might otherwise disturb us morally, like our tendency to punish out of resentment, or to value praise over praise-worthiness — two contexts in which Smith is in fact especially prone to the use of such language (II.i.2.5, II.ii.3.5, III.2.29-31). For these aspects of ourselves can lead us to doubt whether we are really capable of virtue. Explanations that show something good about them — something that God might have intended — help relieve us of such doubts. Natural teleology can thus support our moral faith.

Finally, if Smith believed in God only on moral grounds, that would explain why he never so much as mentions logical or metaphysical or empirical proofs of God’s existence. At the same time, Smith never tries to undermine these or any other arguments for God’s existence, nor does he allude to his friend Hume’s attempts to do that; he also refers to God and God’s providential care for us, without apparent irony or hesitation, throughout TMS. All of this makes sense if Smith thought that there were reasons for believing in God, but not logical or metaphysical or scientific reasons.

I venture to say that the moral faith reading of Smith fits with every remark that Smith ever made about religion, in private or in public, and every act he took with religious import. That is not true of more skeptical readings of Smith on religion. They need to distort the chapters of TMS in which Smith argues for a belief in God, and to discount the many references to God that appear elsewhere in the book. It is also not true of readings of Smith that make him out to be an orthodox Christian. They need to distort or discount Smith’s encomium to Hume and harsh critiques of Christian churches. Only the moral faith reading enables us to avoid dismissing anything Smith himself said. It is also a coherent position, held by no less respectable a philosopher than Kant, and it makes for a fulsome explanation of Smith’s thought about religion. These are strong reasons to uphold it against its competitors.










  1. For Smith as atheist or radical sceptic, see Peter Minowitz, Profits, Priests, and Princes, (Stanford, 1994) and Emma Rothschild, Economic Sentiments, (Harvard, 2001); for Smith as non-Christian believer in God, see DD Raphael, The Impartial Spectator, (Oxford, 2009); for Smith as traditional Christian, see Paul Oslington, Adam Smith as Theologian, (Routledge, 2011).
  2. On this controversy, see Dennis Rasmussen (ed), Adam Smith and the Death of David Hume, (Lexington, 2018)