Is Ethics Possible Without God?

Abstract

This essay defends the position that ethics must be grounded in God, where the notion of ‘God’ is understood as a transcendental source of normativity, though not necessarily a personal being who ‘commands’ moral behavior. The essay argues that the true debate is between the naturalistic reduction of ethics and the idea of a transcendental ground for moral normativity. I claim that only the latter can provide a sufficient basis for morality.

1. Introduction

In this essay I defend the position that a genuine ethics is not possible without God. This may seem a reactionary view for the modern secularized world built on a separation of church and state. Many intellectuals today subscribe to the Rawlsian view that religion is not an appropriate basis for public policy because it cannot be defended with reasons that are publicly available. Moreover, the idea that ethics requires belief in God also seems to go directly against the obvious fact many atheists and agnostics in fact are highly moral (and, of course, many religious believers have dubious moral standards). Further, this position contradicts the increasingly common view that one need not be religious to be moral. According to a 2022 poll, 65% of Americans agree with the claim that it is not necessary to believe in God to be moral; only 34% hold the contrary. Similar views are held in most developed countries; Sweden represents the extreme, with 90% agreeing with that claim.1 However, I will argue that my thesis is perfectly compatible with a liberal and secular order and does not require an affiliation with any particular religion or with any religion at all. I will suggest that the real issue at stake is the debate between naturalism and non-naturalism: is the physical world the sum of all there is, or does a transcendent moral value constitute part of reality as well?

2. What Does It Mean to Say That Ethics Is Not Possible Without God?

This question as stated is multiply ambiguous. To begin, what does it mean for ethics to be ‘possible’? One way of addressing this question is whether ethical behavior is more likely in a world with God than in a world without him. However, this is a sociological question rather than a philosophical one, and indeed one of dubious relevance. Even if religious believers were more likely to be moral, that might just be because of a fear of hell (hardly a praiseworthy moral motivation). And even if religious believers were less likely to be moral than atheists, what exactly would that demonstrate? As has often been noted, there are many confounding variables here, for example, the fact that people who are more troubled may be attracted to religion in the first place, and the fact that atheists tend to be more likely to be well-off (a factor that make it easier to be moral). It is thus unclear how relevant this question is to the current debate, and I will ignore it.

What then could it mean for ethics not to be possible in a world without God? Let us imagine a possible world in which there is no god, and in which all people know that fact. Is the question whether anyone could ever act morally? Or whether morality could be widely accepted? Surely we can imagine a possible world in which there is no god and yet at least some people, and even everyone, behaves in a perfectly moral fashion. I cannot think of any reason why moral behavior should be logically impossible in any world, godless or not. Indeed, in our own world there are plentiful examples of atheists who are highly moral people. So the question as to whether in a world without God ethics would not be possible seems trivial and uninteresting.

There is however a more interesting way of interpreting the question. The question is not about mere obedience to ethical rules, but about whether one has a good reason to be moral. For there are many non-moral reasons to be moral. Kant in the Groundwork gave the example of the shopkeeper who refrains from the temptation to cheat his customers, out of the sole motivation of ensuring that he did not get caught and lose business as a result (Kant 1981). This person, Kant plausibly claimed, is not truly ethical. To be ethical is not merely to follow the correct moral rules, but to do so for the right reason. And it is a familiar feature of our experience that most of the time there is, as in the shopkeeper case, a strong self-interested reason to be moral. Those with a reputation for lying and cheating will likely be socially isolated, and society is structured so that there are many formal and informal penalties (and often severe punishments) for immoral behavior. This situation would be the same in a world without a god. Yet we also want to say, with Kant, that genuine ethics requires a motivation that transcends mere self-interest. The truly ethical person is one who acts morally because it is the right thing to do, not merely because it serves his interests.2 Even in a world without a god, the same alignment of self-interest and morality would exist, and so most people would likely be moral most of the time. But there would be a big difference. If self-interest is the only motivation to be moral, then there is the problem of what people would—and should—do when the two fail to coincide. Why be moral, if it goes against self-interest? And for that matter, even when the two do coincide, as in the shopkeeper case, true ethics requires a motivation that goes beyond self-interest. So the question that we need to ask is: can there be an ethics that is based in something that goes beyond self-interest, in a world without a god?

We come then to the final ambiguity: what do we mean by ‘God’? This issue will I believe prove to be the most contentious, yet it is the key issue on which my argument will rest. Sterba (2024) writes that he is referring to ‘traditional theism,’ yet traditional theism is not to my knowledge a well-defined category, nor does he attempt to explain it. Perhaps he means something like folk theism, but it would be odd to claim that ethics requires belief specifically in folk theism. Reference to ‘traditional’ theism also raises the question as to just which tradition is he referring to. It seems to me hopelessly parochial to limit this debate to just the Western tradition and in particular the Judeo-Christian-Islamic tradition of personal monotheism, as if people from other religious traditions cannot be moral persons. We will need a substantially broader notion of God, including one that does not require a personal deity (as in Daoism for example) nor a strict monotheism.

3. Ethics as an Ultimate End

It is important to state clearly the nature of the problem with respect to explaining moral obligation: i.e., why be moral? We have argued that a genuine form of ethical behavior is only possible if there is a reason to be moral that goes beyond self-interest. The question is not merely one of ‘motivation’ but of the normativity of ethics: ethical requirements are supposed to bind us even when we lack the desire to be moral. The very idea of normativity is a problem for a naturalist worldview, as has often been observed (De Caro and MacArthur 2010). For the naturalist, the ground of moral obligation must be found within the natural world: e.g., in self-interest or happiness or, in Sterba’s account, reason. But this effort involves a major challenge, for the nature of moral obligation is quite extraordinary. It is supposed to in general override any other reasons and goals.3 That is most obviously true for self-interest, but it goes beyond that: thus even if there is a duty to obey the law, it is widely held to be suspended when the law itself is unjust. So ethics involves not only a duty for its own sake (a categorical rather than instrumental duty) but an ultimate duty: one that overrides other obligations and desires. The challenge then is to explain this ultimate obligation: where could it come from?

4. The Divine Command Theory

The simplest and crudest answer to this question is the Divine Command Theory, now widely viewed as discredited.4 Unfortunately, it is too often assumed that any theory that makes ethics dependent on God must assume the Divine Command Theory (Sinnott-Armstrong in his book Morality Without God seems to make this assumption).5 The Divine Command Theory purports to provide an explanation of the source of ethics, the basis for our knowledge of ethics, and the reason to be ethical. It is however unsatisfactory on each count, as Plato showed in the famous Euthyphro problem. If God is the source of ethics merely through his command, then that means that ethics has no rational basis but is merely the arbitrary will of a powerful being; if on the other hand, ethics comes not from the sheer command of God but proceeds from his intrinsic goodness, then the explanation is empty, for it does not account where goodness comes from. The idea that divine revelation is the source of our knowledge of ethics is also unsatisfactory; we expect religious believers not to mindlessly obey the commands of scripture but to assess and interpret religious revelation in light of independent objective ethical standards. Indeed, even within religion one of the conventional means of assessing the legitimacy of a revelation is that it expresses acceptable moral ideals. Finally, the Divine Command Theory gives us a wholly unsatisfactory account of moral motivation—essentially one that makes ethics a matter of pure authority (because God said so), and that enforces ethics with the threat of punishment. Again, this is illegitimately appealing to self-interest as a basis for ethical behavior.

But if we reject the Divine Command Theory, just what is supposed to be the relation between God and ethics? After all, it is widely taken that the Euthyphro objection to the theory summarized above can be taken as demonstrating that morality must in some sense be prior to and independent of God. So God seems to become no more than a mediator between us and morality, playing no essential role. Indeed, that would seem to demonstrate that ethics is entirely possible without God. Thus the Euthyphro Dilemma is often taken to suggest that ethics must be based on reason not religious belief. And this is just what Sterba suggests: that ethics can be based entirely on reason, so that an atheist has just as much reason to be moral as a religious believer. To that argument we now turn.

5. Can Reason Be the Basis for Morality?

A widely influential argument, most famously developed by Immanuel Kant and embraced by Sterba, is that ethics can be grounded in rationality (let us call this the Kantian Argument). On this position, to be immoral is to be irrational; to the extent one is a rational being, one will necessarily be moral. The argument can be developed in a number of ways, but here is one version: to be immoral is to place one’s own interests (or those of one’s group) above those of others. But there is no rational basis for arbitrarily preferring one person’s interest over that of another. Sometimes there is a rational basis for a difference in treatment: it is legitimate to give Milo the Wrestler a larger portion of food than other people, since his much greater size and caloric needs dictate that he needs more food than others. But the nature of immorality is to arbitrarily favor oneself over others, that is for no reason other than mere self-preference. I cannot claim a larger portion of food merely because that is what I want. And this is, strictly speaking, irrational, in that one cannot give a reason why one’s value should be higher than that of similarly situated others. Since other human beings are no different than me in any relevant respect, then reason requires that their interests be treated with equal weight as one’s own. The immoralist is then simply being irrational.

The argument is, though not universally accepted, to my mind persuasive in showing that ethics has a rational basis, and I will assume for purposes of this argument that it works.6 Sterba claims that this argument proves that ethics has a wholly secular basis, not requiring any religious commitment. One does not need to religion to be moral; all one needs is a commitment to rationality, and hence ethics is possible without God. However, the Kantian Argument has a fatal flaw with respect to the present debate. Its strategy is to explain the claim of ethics on us by grounding it in reason. But this assumes that reason has a normative and even ultimate claim on us, i.e., that reason itself is a transcendental value. Recall that the problem is explaining why ethics has an ultimate claim on us. Explaining this in terms of reason does not work, because rather than explaining the basis of ethics, it merely pushes the question back: why should reason have ultimate authority over us?

Thus anyone trying to explain within a naturalistic perspective why ethics has normative force over us will face the very same problem with respect to reason: why should reason have normative force (let alone ultimate normative force)? On a Darwinian account, reason is merely a tool, a device that made humans an extremely successful species. There is nothing ‘special’ about the faculty of reason, any more than any other tool, such as echolocation in bats or a powerful sense of smell in polar bears. Reason evolved because it promoted human survival and reproduction. There is no basis for giving reason a transcendental value any more than ethics. In particular, there is no reason to ever sacrifice self-interest to the claim of reason, i.e., no reason to be moral. To do so would be to make a fetish of reason, to misunderstand what it is for. It exists, for the Darwinian, because it promotes self-interest, and to use it to restrain self-interest makes no sense (other than restraining short-term self-interest to promote long-term self-interest). Thus reason cannot ground ethics after all, within the naturalistic framework. The immoralist can simply respond: if it is ‘irrational’ to favor myself over others, then so be it. But why should I care what rationality requires, unless it promotes my own interests? Indeed, what is truly rational is serving my own desires.

Hence the dominant definition of reason among social scientists today (and among numerous philosophers as well) is instrumental: reason is a means to an end but not an end in itself. A leading scholar of rationality in the field of psychology defines reason as ‘the optimization of the individual’s goal fulfillment.’7 The normative force of reason (and of ethics) is explained entirely in terms of means-end rationality: if you want a given end, then you should (rationally) will the means to that end. Hence reason is not itself a value to be pursued; it has no intrinsic normative force, and certainly cannot dictate our ultimate aims. This is the tradition deriving from David Hume, the father of modern naturalist philosophy. For Hume, reason is merely a ‘slave’ to the passions. It is not hard to find real-world examples of this instrumental use of reason. For instance, when a Supreme Court seat became vacant in February 2016, Senate Majority Leader Mitch McConnell declared that he would not allow the Democratic president to fill it, since a presidential election was coming up in November 2016 and the American people should have a ‘voice’ in this decision via the election process. When a Republican president won the election, he was able to fill the seat. Then when another vacancy occurred in September 2020, McConnell reversed course and insisted on the Republican president’s right to fill the vacancy, despite there being a presidential election in just two months. For McConnell and the Republicans, consistency in one’s principles is subordinate to political expediency. McConnell’s behavior can be condemned as unprincipled and unjust. But from the standpoint of instrumental reason, it was the only rational thing to do, as it optimized McConnell’s ‘goal fulfillment.’

Kant attempted to reverse the Humean idea and make the passions subservient to reason, so that reason is not merely a tool to satisfy the desires. But then he needed an explanation of the normative power of reason, why we should elevate it into a binding obligation. Kant attempted to answer it by arguing that ‘our existence has a different and far nobler end, for which, and not for happiness, reason is properly intended, and which must, therefore, be regarded as the supreme condition to which the private ends of man must, for the most part, be postponed.’8 Kant claims that instinct would be a much better guide to happiness than reason, and therefore reason must have a higher purpose than happiness. Therefore, reason is the highest purpose of human life, not happiness, and therefore we must pursue reason (and morality as well) even at the cost of our own happiness. However, Kant was writing 100 years before Darwin, and today his reasoning about why nature implanted reason in us would not be accepted. The very idea that there are purposes in nature, let alone that reason is the highest purpose of man, conflicts with the modern naturalistic view.9 And indeed anyone endorsing the Kantian Argument today must face this dilemma: if reason is to provide a ground ethics, it must constitute a transcendental value incompatible with a naturalistic framework. For the naturalistic view of reason is merely an instrumental one; reason is merely a mean to achieve some other goal, and hence cannot give us the basis to support morality as an ultimate value, one that binds us regardless of our desires. It seems that Kant is giving us not a naturalistic argument but a transcendental one, rooted in his religious conviction that there is a ‘higher’ purpose to human existence than mere happiness.10

6. The Transcendental Basis of Ethics

As Charles Taylor observes, modern ethical theory largely avoids the question of how ethics can have intrinsic and even ultimate normative force; in modern ethics ‘the basis of ethics is seen as something obvious, and there seems no call to examine the understanding of the incomparably higher underlying all this, much less raise the question whether it points to something transcendent.’11 But if the Divine Command Theory cannot explain how God could be the source of ethics, what alternative is there to explaining the transcendental power of ethics? It is however not widely recognized that Plato aimed to refute the Divine Command Theory precisely in order to give us a more sophisticated account of the origin of ethics in God. We will call this account, in a loose sense, Platonism. For Platonism, as I am using the term, moral values12 are objective but non-physical realities, and they are grounded in an ultimate moral ideal that Plato calls The Good. The Good is the ultimate source of human ethics. In this last section of the paper, I will give a brief defense of a Platonist view, though a full and comprehensive defense is beyond the scope of this project.

So far we have a fairly minimal form of Platonism, one that asserts the existence of objective moral values, on which even many agnostics could agree.13 More controversial is the question of the relation between this ideal realm of values and the observable physical world. Obviously this is an enormous and difficult topic and we can only touch on it here. For it seems that it is not enough merely that there exists a realm of values, if the physical world is separate, self-contained and wholly autonomous. It is arguable that there must be some sort of relation between the two if the physical world is to be ‘governed’ in some sense by value, i.e., if value is normative at least for rational physical beings like us. Plato insists on a very strong relation, one in which the physical realm is subordinate to the ideal realm. For Plato, the Good is not merely an existent ideal, but the source and origin of all that is (Plato 1992, Republic 6, p. 50914). We may also draw on John Leslie’s neoplatonic idea of ‘ethical need’ as a basis for creation (Leslie 1989, p. 165). Leslie argues that the best explanation for the existence of the world is an ethical one: that the existence of the world is good.15 The fact that ethical need is the basis for existence imparts to the world an intrinsic purpose: that it exists in order to manifest and express goodness. This fact entails that morality is a requirement for creatures like us.

Leslie’s argument is based on inference to the best explanation; for him, the naturalist has no explanation for the existence of reality. But just as important, and more relevant to the present debate, is what we might call the Argument from Moral Normativity. If morality is to have an ultimate normative claim on us, there must be a higher, non-physical basis for it. The naturalist cannot provide such a ground, and the inevitable implication for naturalism is moral nihilism (the view that morality does not have any normative claim on us, let alone an ultimate one). Naturalist philosopher John Mackie concedes that moral objectivity, if it existed, would need an explanation and that a divinity would be a prime candidate for such an explanation:

If we adopted moral objectivism, we should have to regard the relations of supervenience which connect values and obligations with their natural grounds as synthetic; they would then be in principle something that a god might conceivably create; and since they would otherwise be a very odd sort of thing, the admitting of them would be an inductive ground for admitting also a god to create them. There would be something here in need of explanation, and a being with the power to create what lies outside the bounds of natural plausibility or even possibility might well be the explanation we require.

(Mackie 1982, p. 118)

Mackie’s strategy is to reject moral objectivism. His defense of naturalism and of an ethics without ‘transcendental necessities’ (Mackie 1982, p. 199) ends up endorsing moral relativism and what is effectively moral nihilism. Mackie acknowledges that self-interest and ethics will not always coincide, and when they do not, there is no reason to choose ethics over self-interest (ibid., p. 190). He defensively insists that we can still be moral because ‘nearly all of us do have moral feelings’ (ibid., p. 191). It is unclear what his evidence is for this claim or just what ‘nearly all’ means. In fact, I would guess that most people believe that ethics is normative for us not because they ‘feel’ like being moral but because they view ethics as a fundamental obligation in the sense discussed above. In any case, Mackie’s conclusion is more troubling than he acknowledges. For on his view there is no reason to privilege moral feelings over non-moral ones; all that matters is the strength of the feeling. We have no rational basis for believing that kindness is any better than cruelty. We have no reason to teach our children that they should be moral; on Mackie’s view, we just follow our desires, whether those be moral or not. And what about those who lack such moral feelings? We cannot demand that they be moral, nor can we punish them for their immoral behavior. If we inflict harm on them to protect ourselves, we cannot claim our actions are morally justified punishment, since no such objective justification exists. If we are being honest, the most we can say is that might makes right. We inflict harm on them when we are stronger than they are, but not if they are stronger than us: and neither one is justified. It is all a matter of power.

For the naturalist, it is inexplicable that a completely self-existent physical world, in which human behavior evolved by wholly natural physical processes, would produce beings that see morality as normatively binding, let alone as an ultimate purpose. Indeed, the naturalist should view this idea as simply a mistake; morality is merely a tool and it is irrational to subordinate self-interest to morality. For those who believe in morality as objective and normatively valid, the reasonable (and even necessary) inference is that the physical world is subordinate to a higher, ideal realm, one which imparts a moral purpose to it. This is the Platonist position. And this is, I would argue, the very essence of religion. Indeed, William James’ famously defines religion in just this way: ‘Were one asked to characterize the life of religion in the broadest and most general terms possible, one might say that it consists of the belief that there is an unseen order, and that our supreme good lies in harmoniously adjusting ourselves thereto.’16 That is, the heart of religion is belief in a higher moral order and the necessity of conforming our behavior to that order.

7. Platonism and God

It may be objected here that the Platonist concept of the Good is quite distinct from the conventional notion of God. Mackie holds that Plato gives us at best a substitute for God, but not a personal God.17 Many will agree that the impersonal ideal of the Good should not be considered the same as God.18 Obviously, this issue is directly relevant to the present debate, since the very issue is whether ethics requires belief in God, though again it raises very difficult issues that can only be briefly addressed here. This question is presumably why Sterba insists on limiting his claim to belief in ‘traditional theism,’ though I criticized that limitation above as both obscure and arbitrary. Can one hold that, even if ethics requires belief in the Good, it does not strictly require belief in God?

The objection however mistakenly assumes without argument that the only ‘real’ notion of deity is a personal one. Yet for Plato, arguably the Good is simply a name for God. To quote Lovejoy (1976): ‘there can be little doubt that the Idea of the Good was the God of Plato; and there can be none that it became the God of Aristotle, and one of the elements or ‘aspects’ of the God of most of the philosophic theologies of the Middle Ages, and of nearly all the modern Platonizing poets and philosophers.’19 Plato’s role as religious reformer is often neglected, but it was his conscious purpose to replace the traditional and crudely anthropomorphic Greek myths of the gods with a more philosophically sophisticated conception of the deity and of the moral law, i.e., with the ‘God of the philosophers.’ Moreover, as we noted earlier, many religions involve impersonal deities: Daoism, Hinduism (which involves both a personal and a higher impersonal deity), Buddhism (on some interpretations), and even Christianity (in its mystical tradition). It would be ethnocentric to insist that the only legitimate conception of God is the Judeo-Christian Islamic one. Further, sophisticated theologians in the Western tradition recognize that the very concept of ‘personality’ when applied to God is at best an analogy rather than a literal truth.20 Nor, for that matter, is it even a well-defined concept; once we get past the vulgar conception of God as an old bearded man on a throne, the distinction between personal and impersonal deity is far from clear. In the Platonist tradition, the Good is no less the creator of the world, and creates out of love, just as does the Christian God, although of course these terms are used merely analogically.21

Consider, for example, just how God can function as a moral guarantor for the world, which is precisely the question at issue here. It is far from obvious that a personal deity is necessary for this role (indeed, perhaps the most systematic example of a supernatural moral guarantor is the doctrine of karma, which is based on impersonal laws not personal action).22 One way to think of the very inadequacy of the Divine Command Theory is that it attempts to make divine moral authority into a purely personal matter: as if the decision as to what moral rules to mandate were a matter of individual will, and the enforcement of morality as a personal matter as well. If the essential difference between personal and impersonal deity is that the former acts on a ‘will,’ it is unclear just what this adds to our conception of divinity. For either the divine will is wholly arbitrary, or it acts necessarily out of love of goodness. In any case, I have argued that the question in debate is whether ethics requires a transcendent purpose to reality, and the issue of whether this transcendent purpose must issue from a personal being narrows the debate unnecessarily.23

It is of course possible to be a Platonist in a weaker sense: one can believe in objective moral ideals that exist independently, but without commitment to the idea that they are grounded in an ultimate entity (the Good) that has creative power responsible for the existence of the world. In fact, this is a common position (e.g., Enoch 2011). But as Mackie argues, it then becomes mysterious just how these transcendent values have any claim on us or why these ideals do or should have any influence on the physical world, in other words, why we should be ethical. For such a view, it is simply a brute fact that these ideals should be obeyed. But this account, as Mackie rightly suggests, would be a very strange anomaly in an otherwise naturalistic worldview, i.e., one in which the physical world is entirely explicable in terms of physical or natural processes. Weak Platonism would seem to be an unstable halfway position, giving up on a full naturalism yet giving no causal role to the non-natural moral realm.

It would be equally odd if only one type of creature (us) in this physical realm were responsive to the Good. It is for this reason that (strong) Platonism insists on a causal role for the Good in creating and sustaining the world, and this causal responsibility in turn entails that the world itself is good and is guided by the good. Such a position is of course must face the objection that abstract objects cannot have causal powers. But of course ‘impersonal’ is not the logical equivalent of ‘abstract.’ A stone, for example, is an impersonal entity, but it is no less concrete than a human person. The Good is lacking in the traits that make it a person (however those are defined), but it does not follow that it is ‘merely’ an abstract being. Second, the traditional personal deity is generally conceived of as an eternal, non-physical being, and hence faces no less of a problem in explaining how God can be causally efficacious; it is widely held that a non-spatial entity cannot have causal power over a spatial one (Taliaferro 2015, p. 773). Third, the notion that abstract entities are somehow less real as well as less causally efficacious is a product of the early 20th century thought and the influence of positivism, and it risks begging the question against Platonists for whom the Forms are more real and have more causal power than physical things. As Falguera et al. (2022, II.i) put it, ‘Plato’s Forms were supposed to be causes par excellence’. Finally, some philosophers have made a strong case against the conventional position that abstract objects cannot have causal power (Taliaferro). Obviously, this is a vast and difficult topic, and a full discussion of it is far outside the scope of this paper. However, it seems to me at the very least that we cannot dismiss the idea of the Good having causal powers simply in virtue of its being impersonal.

I believe that Sterba is correct to insist that the theodicy problem is crucial for the present debate about ethics. If there is no reason to believe that the world as a whole is guided by the moral good, then Platonism must fail. Sterba claims that ‘my logical argument from evil has shown that such a God is logically impossible, given all the evil that exists in our world’ (Sterba 2024). Now as I have said, I do not know what ‘traditional theism’ means. However, I have taken Sterba’s challenge to be against any form of religious belief, including Platonism, the one I have defended here. It is notable furthermore that Platonists have their own solution to the theodicy problem, the ‘plenitude’ solution, the idea that the infinite and perfect Good necessarily gives rise to the imperfect, finite beings of this world with all their attendant limitations and vulnerabilities, i.e., the ‘evils’ of this world, in that a world with a maximum variety of beings is better than one with simply the one perfect being. Obviously, a full defense of that idea is far outside the scope of this essay. I also disagree with his belief that even if the world is shown not to be guided by the Good, then ethics can still be normative for us. Here again I think Mackie is correct; it makes no sense in a wholly naturalistic framework to hold that ethics (or reason either) has an ultimate claim on us. For ethics to be possible, i.e., for it to be binding on us in the very strong sense that we need, it requires that we believe in a transcendental Good that is an active force in the world. In this sense, I claim that ethics is not possible without belief in God. If Sterba’s challenge is only directed at believers in personal deities, then my conclusion is not at odds with his, though as I have said I doubt that his goal is meant to be so narrowly defined. My argument has been aimed at the broader defense of a naturalistic ethics against any form of transcendentalism, which I think is a more interesting question. I have tried to demonstrate that Sterba’s defense of ethical naturalism does not work, and that a Platonist account of ethics provides the necessary defense of ethics that naturalism cannot.