The Oddness of Ethics Without God: Why Morality Is Not “At Home” in a Non-Theistic World

Abstract

Despite non-theistic proposals on offer to ground moral realism, theism still offers a far more stable, robust, and well-integrated metaphysical context for objective moral facts and moral values. Moral realism is not truly “at home” within such non-theistic frameworks, which include Cornell school-type naturalistic moral realism or non-naturalistic, non-theistic Platonist moral realism. Theism’s deep connection to moral facts and moral values is more natural, unifying, and basic than these alternatives. Furthermore, theism offers a more comprehensive explanatory account for other central features of our human experience, as well as key features of the universe in which we find ourselves.

1. Introduction

Moral realism maintains that objective moral facts exist, that they are independent of human beliefs, opinions, and cultures.1 But does God have anything to do with those moral facts? In response, here is what a range of philosophers might say:

A. Certain philosophers would say, roughly, “Hypothetically, yes, but there is no God. So moral facts do not exist; moral claims are false” (e.g., naturalists Friedrich Nietzsche, Bertrand Russell, J.L. Mackie).

B.–D. Others say, “No, moral facts exist, but God is not necessary to ground those moral facts.” These would include:

B. Naturalists like David Brink and Richard Boyd;

C. Atheistic Platonists such as Erik Wielenberg and Russ Shafer-Landau;

D. Theists like Richard Swinburne and Keith Yandell. In the case of Swinburne and Yandell, they would say that God adds greater depth and breadth to the moral landscape and that moral facts alone hardly plug all of the metaphysical holes required for a robust morality.

E. Still other philosophers would say, “Yes, God is necessary to ground those moral facts, which are not free-standing but are ontologically connected to God” (William Craig, David Baggett, and Jerry Walls).

In this essay, we will mention positions (B) and (D) only in passing while focusing on (A) naturalistic error theorists, (C) atheistic non-naturalistic moral realists, and (E) super-naturalistic moral realists. In my view, divinely grounded moral facts (in E) prove to be more stable, more robust, and better integrated with other features of moral reality than the alternatives. Not only does God ground moral facts, but he also unifies related features of moral reality such as human dignity, moral responsibility/free will, (self-)awareness, rationality, duties and so forth. God-less non-naturalistic moral realism proves to be less than “robust,” offering a more piecemeal or ad hoc justification by comparison and drawing on metaphysical resources that more strongly resemble theism.

In Part I of this essay, we set the context for moral realism against the backdrop of a naturalistic “Russellian” world (View A)—what Wilfrid Sellars called the “Scientific Image” (Sellars 1962).2 This view stands in contrast to broad naturalism, which affirms objective moral values such as human dignity, self-awareness, moral duties, personal responsibility, and so forth (the “Manifest Image”)—for example, in View B. Part II examines the representative position of Erik Wielenberg’s views that defend a God-less non-naturalistic moral realism (View C). Finally, the essay develops an overarching response to Wielenberg’s account, noting that it lacks robustness and strong explanatory power and arguing that moral realism is more at home in a personalistic, theistic world (View E). Whether the non-theist denies moral realism3 or attempts to present an account of moral realism without God, these views lack the robust explanatory power that theism more readily affords to account for the very evident moral world in which we find ourselves.

This overview is not comprehensive and, in places, tries to simplify and distill the complex. For example, it does not address naturalistic moral realism, as we mentioned. Moreover, the present essay does not offer detailed argumentation for all possible options in this debate; rather, it often draws on and cites the conclusions of noted proponents of the major relevant positions to help juxtapose the metaphysical resourcefulness—or lack thereof—of those compared viewpoints. I hope the reader will pardon my mere survey of basic positions and conclusions, going into further detail where I believe it is necessary.

2. A Strictly Naturalistic World: A No-Go for Objective Moral Values

J.L. Mackie maintained that objective morality is “queer”; it is odd if God does not exist and if nature is all the reality that there is (Mackie 1977). However, he noted, if objective moral values did exist, this would furnish evidence for God’s existence: “Moral properties constitute so odd a cluster of properties and relations that they are most unlikely to have arisen in the ordinary course of events without an all-powerful god to create them” (Mackie 1982)4—or, more precisely, to ground them. Mackie rejected both moral realism and non-cognitivism, opting instead for an error theory of morality.

Mackie’s stance is consistent with Bertrand Russell’s reflections on the implications of a universe without God:

That man is the product of causes that had no prevision of the end they were achieving. That his origin, his growth, his hopes and fears, his loves, his beliefs, are but the outcome of accidental collocations of atoms. That no fire, no heroism, no intensity of thought and feeling can preserve an individual life beyond the grave. That all the labours of the ages, all the devotion, all the inspirations, all the noonday brightness of human genius, are destined to extinction in the vast death of the solar system, and that the whole temple of man’s achievements must inevitably be buried beneath the debris of a universe in ruins. All these things, if not quite beyond dispute, are yet so nearly certain that no philosophy which rejects them can hope to stand. Only within the scaffolding of these truths, only on the firm foundation of unyielding despair, can the soul’s habitation, henceforth, be safely built.

(Russell 1963, p. 41)

Commenting on Russell’s view and noting Mackie’s, George Mavrodes expounds on the “queerness” of objective morality in a “Russellian” universe comprised solely of collocations of atoms. While a Russellian universe might produce in us feelings and beliefs of obligation (such as the feeling that we ought to protect our children), this is different from actually having such obligations that we ought to carry out. Mavrodes considered it extremely odd that objective moral obligation could arise in such a world (Mavrodes 1986, p. 219; see also Plantinga 2010).

Biologist Richard Dawkins takes the same Russellian view in light of our materialistic and deterministic origins: in a universe of electrons and selfish genes, the world “would be neither evil nor good in intention” but a place of “blind pitiless indifference (Dawkins 1995).

Mark Linville rightly notes that “the combination of metaphysical naturalism with moral essentialism forms an unstable compound” (Linville 2012, p. 131). He points to George Santayana, who persuaded Bertrand Russell to abandon his early commitment to moral realism. On the one hand, Russell’s moral realism implied that “[i]n the realm of eternal essences, before anything exists, there are certain essences that have this remarkable property, that they ought to exist, or at least that, if anything exists, it ought to conform to them” (Santayana 1957, pp. 152–53). On the other hand, Russell’s naturalism posed a problem: “What exists, however, is deaf to this moral emphasis in the eternal; nature exists for no reason” (Santayana 1957, p. 153). As we have seen, Russell’s universe of mere collocations of atoms “had no prevision of the end they were achieving,” including the human beings they produced (Russell 1963, p. 41). Russell rightly wondered at the dumb luck (or “strange mystery,” as he puts it) of how “ blind” nature should have “brought forth at last a child, subject still to her power, but gifted with sight, with knowledge of good and evil, with the capacity of judging all the works of his unthinking mother” (Russell 1963, p. 48). Indeed, this wonderment applies to all non-theistic efforts at justifying moral realism.

To avoid the ruthless metaphysical jaws of the radically stripped-down Scientific Image, those who want to remain naturalists must adopt a broad naturalism, which affirms the more humanizing Manifest Image with its moral realism, personal responsibility, intrinsic human dignity, moral intuitions, beauty, moral duties, self-consciousness, and meaning (e.g., representatives of View B). Given the unpromising beginnings of valueless, deterministic, mindless, nonrational material processes, the outcomes of human dignity, moral duties, etc., do seem surprising—indeed, shocking. There is no such surprise if God exists.

To put it another way, strict naturalists would claim that broad naturalism does a good deal of borrowing from a theistic metaphysical picture. It is this humanized view of life that strict naturalism increasingly rejects and marginalizes. As Huw Price puts it, “Like coastal cities in the third millennium, important areas of human discourse seem threatened by the rise of modern science…. The regions under threat are some of the most central in human life—the four Ms, for example: Morality, Modality, Meaning, and Mental” (Price 1997, p. 247). Broad naturalism runs contrary to the most consistent form of naturalism. As Jaegwon Kim views it: “[Naturalism is] imperialistic; it demands ‘full coverage’ … and exacts a terribly high ontological price” (Kim 1993, pp. 22–23).

Mackie, Russell, Mavrodes, and Dawkins deny moral realism because of its poor fit within the natural world.5 And we could add many others to this list.

3. Non-Naturalistic (God-Less) Moral Realism

While we could discuss naturalistic moral realism, which maintains that moral properties are natural properties (e.g., Boyd 1988; Brink 1984, 1989, 2006), and also some of the concerns that attend it (Horner and Moreland 2024; Baggett and Walls 2025), we put this to the side for our purposes.

Despite the charge that theists beg the question in their denial that moral properties are natural properties (the “just too different” claim),6 we note that theists are not alone here. We have cited philosophers like Mackie, who considered moral properties to be “queer”—that is, utterly different from natural ones—and that if they did exist, they would furnish the theist with a good argument for God. Philosopher Joel Marks, once a broad naturalist, had to give up on morality; it dawned on him that “atheism implies amorality” (Marks 2015). Nietzsche, Sartre, and other notables found that no God means no objective moral facts.

But while naturalism cannot account for morality, perhaps morality can be sustained through a nonnatural, transcendent Platonic realm—without reference to God. Thinkers like Wielenberg reject both the strict and broad naturalist perspectives on morality. In my estimation, Wielenberg and his ilk are on to something. A transcendent realm is needed to ground moral facts. They have embraced a two-tiered position: it takes a transcendent view when it comes to irreducible abstract moral facts, but a naturalistic view regarding the emergence of concrete, valuable human beings from materialistic and deterministic origins late in the history of the cosmos. As we will see, this two-tiered system creates tensions and even contradictions—ones easily resolved by theism.

Here in Part II, I provide an overview of Wielenberg’s position and offer a preliminary response—and then build on this critique in Part III. There, I will compare the broader metaphysical contexts of theism and non-naturalistic God-less moral realism. I will focus on Wielenberg, though similar criticisms apply to Shafer-Landau as well (on Shafer-Landau, see Moreland 2023). They would also apply to theist philosopher Richard Swinburne’s conviction that God is not necessary to secure moral realism (View D).7

3.1. An Overview

We could summarize ethical (God-less) non-naturalism as a combination of moral realism and broad naturalism with its emergent properties (see Horner and Moreland 2024; and Johnson 2021a).8 Wielenberg’s moral realism has three key features: (a) brute ethical facts, (b) the “making” relationship (that is, emerging moral properties are dependent on non-moral base properties), and (c) non-naturalism. For Wielenberg, the entities constituting moral realism are irreducible givens—metaphysically brute facts. But this is no worry for Wielenberg. Theism has its own brute fact—namely, God’s existence. So why give rational preference to the brute fact of God’s existence over moral facts as brute? Wielenberg claims that these brute moral facts “come from nowhere, and nothing external to themselves grounds their existence; rather, they are fundamental features of the universe that ground other truths” (Wielenberg 2014, p. 38).

Wielenberg makes a couple of distinctions—the two tiers. First, he distinguishes between (necessary) moral facts, which are transcendent, and moral values, which emerge through the evolutionary process. For example, human dignity emerges from, or supervenes on, a specific configuration of matter that has emerged from valueless biological and environmental naturalistic processes (Wielenberg 2010). Despite the valueless physical origins of humanity, he nevertheless claims: “From valuelessness value sometimes may come” (Wielenberg 2009, p. 40n).9

Secondly, Wielenberg distinguishes between natural facts and moral facts; for Wielenberg, a causal “making” relation exists between the two. This causal connection is unmediated by any natural laws, is metaphysically necessary, and thus holds in all possible worlds. He contrasts his “making relation” view with two other types of supervenience. One is a reductive view of supervenience, in which moral properties are entirely constituted by the base physical properties such that they can be reduced to them. For example, a sense of having moral duties is nothing more than the biologically hardwired impulse to survive and reproduce. Wielenberg further contrasts his view with a “resemblance” or “exemplar” view of supervenience. He applies this view to that of theists like Robert Adams (and thus named, A-supervenience in Adams’s honor): when a finite person’s character and actions (the “base properties”) resemble the goodness that belongs to the nature of the necessarily existent God, this moral property of goodness supervenes on those actions and character.

When we ask the important question, “Why does this making relation obtain?”, Wielenberg admits that this is merely a brute fact. But, as noted earlier, Wielenberg considers this type of causal relation that is assumed by many theists to be a brute fact as well: “[Within theism,] there is a distinctive and robust causal relation that holds between divine willing and its effects, and that this same sort of causal relation holds between non-moral and moral properties” (Wielenberg 2014, p. 18).

Thus, Wielenberg’s view of “robust causation” takes the relationship between non-moral and moral properties as a necessary connection, obtaining in all possible worlds. For Wielenberg, theists’ claim that God wills things into existence in creation is an example of how non-moral properties can have the capacity to produce moral ones.

3.2. A Preliminary Response

As noted, Wielenberg takes a commendable step both away from naturalism and toward theism by recognizing a transcendent realm. We can agree with him that naturalists attempting to locate moral facts in the physical world are misguided in their efforts to squeeze the water of moral facts from the rock of materialism. That said, however, we can make a few preliminary comments regarding moral facts and the making relation.

First, even though Wielenberg is happy to accept the label of “Platonist,” we can still ask: how does his view bring together abstract objects such as moral facts and concrete objects? As Peter van Inwagen observes, “The difference between any abstract object and any concrete object would seem to be the maximum difference any two objects could display” (Van Inwagen 2004, pp. 110–11). Reconciling this disparity is one of the conflicts created by Wielenberg’s two-tiered system. By contrast, Christian thinkers like Augustine took Platonist transcendentals and comfortably located them within a Christian (theistic) framework. While Plato’s Symposium referred to loving the form of the Good, this love involves no reciprocity or relationship. By contrast, Augustine’s personalized view of ultimate reality brought together Platonic ideals and located them within a Christian theistic framework: a personal God is the Good, and this makes possible the mutuality of loving the Good since the Good is personal (Rist 2002, p. 108). The tri-personal nature of trinitarian theism makes this grounding of goodness personalism all the more enriched and well-grounded (e.g., Johnson 2023a).

This personalistic divine grounding of moral facts is more basic, enriched, and organically connected than the brute, detached, abstract moral facts that hang in metaphysical midair; according to Wielenberg, these moral facts concerning persons exist even if human beings were never to appear on the stage. Clearly, abstract moral facts simply do nothing as they are causally effete (Baggett and Walls 2016, p. 137).

As it is, Wielenberg’s realm of abstract moral facts is disconnected from concrete persons—another clash in his two-tiered system. Why not simply consolidate and locate moral facts in the mind and character of a personal divine agent, who actually does have causal powers to create morally equipped creatures who can attend to those moral facts? There is a significant disanalogy between a concrete personal divine agent and abstract moral facts.

Second, despite Wielenberg’s embrace of transcendent moral facts, he is misguided in his optimism that value (e.g., human dignity) can “sometimes” emerge from valueless material processes (Wielenberg 2009, p. 40n)—yet another tension in his two-tiered view: the value presupposed by necessary moral facts is no guarantee that value will emerge, let alone value that coincidentally corresponds to those preexisting moral facts. Moreover, these moral facts are causally effete; it is hard to see how they have any bearing in a closed universe. The abstract moral facts and physical objects are somehow mysteriously connected. Wielenberg claims that goodness supervenes on two persons who love each other. Given the backdrop of materialism, determinism, and valueless, impersonal, undirected processes, Wielenberg is making quite a metaphysical leap.

On theism, no such leap or unwarranted optimism exists. Theism offers a natural context: from a concrete (i.e., non-abstract), supremely valuable divine personal being, concrete, valuable finite personal beings readily come. This is a natural and smooth transition from infinite value to finite value, from God to humans bearing God’s image, from one concrete object to another concrete object. In fact, Wielenberg appeals to theism to serve as a fair comparison of brute facts. But both God and the universe (and humans within it) are concrete objects, and thus, unlike Wielenberg, theists have no abstract-concrete dichotomy to bridge. Nor can Wielenberg offer us anything in experience to show how abstracta can causally influence concrete objects (William Lane Craig, Johnson 2021b).

Third, while Wielenberg speaks of the “making” relation between natural facts and moral facts, he faces considerable obstacles. I quote Angus Menuge at length:

An obvious objection is that the making relation is at least as mysterious as supervenience. Consider one of Mother Teresa’s supreme acts of kindness and a supreme act of cruelty by one of the Romanian communists that tortured Richard Wurmbrand. If we inspect the natural properties of these acts (those properties studied by the natural sciences), there seems to be no connection with any moral properties. In a godless world, there is no underlying telos according to which some natural properties conform to the way the world is supposed to be, while others do not. Rather, these properties simply are distributed in a certain way: there is no reason to think that the distribution underlying Mother Teresa’s act of kindness is morally good, or that the distribution underlying the communist torturer’s act of cruelty is morally bad.

Further, it does not seem that Wielenberg’s account is sufficient to answer a number of skeptical challenges. Suppose the communist torturer asserts that the cruelty of his act makes it right, and that the kindness of Mother Teresa’s act makes it wrong. If the making relation is a brute fact, there are no resources to refute immoralists and psychopaths who simply assert that the specific brute relations are different from those that a nice guy like Wielenberg assumes to obtain. The problem is that the making relation does not explain the particular patterns of covariance between non-moral and moral properties we may hope for. We need to know what it is about the natural properties of Mother Teresa’s act of kindness that makes it good, and what it is about the natural properties of the torturer’s act that makes it bad (Menuge 2016, p. 249).

It seems that Wielenberg’s two-tiered system creates serious internal challenges to his view. But even if Wielenberg’s defense of moral facts were successful, it would be a pyrrhic victory for God-less non-naturalistic moral realism. It still falls short in accounting for a robust moral realism. We explore this point in Part III.

4. Why Moral Realism Is More “At Home” Within a Theist Framework

We have seen that moral realism faces difficulties both on naturalism and with God-less non-naturalistic moral realism, as articulated by Wielenberg and others.

Here we explore a bit more of the metaphysical backdrop of a God-less non-naturalistic worldview and a theistic one.

4.1. More Natural, Unifying, and Basic

A comparison of the explanatory power of theism over against naturalism (or related views) can be assessed by considering largely agreed-upon phenomena—basic features of the universe and of our humanity (as has been done in Copan 2020; introduction in Copan and Taliaferro 2018; Copan, forthcoming). Then, when weighing the merits of the two options under consideration, we must consider which is more natural, more unifying, and more basic.

Natural: Which of the two worldviews is a more natural or less-surprising fit between the metaphysical background and the phenomena under consideration? Which worldview best anticipates—and offers a better metaphysical “home” for—such phenomena? Even if we allow for brute moral facts and begin with a strictly material, mindless, purposeless universe, where does this leave the God-less non-naturalistic moral realist? The moral facts she embraces are metaphysical danglers that seem quite unexpected—especially if human beings were never to appear on the scene. The climb up “Mount Improbable” points to the realization that humans barely made it on the scene after a long, tortuous process (Dawkins 1997). But if they did not beat those staggering odds, moral facts that pertain directly to them seem strange indeed.

Unifying: Which of the two worldviews best brings together wide-ranging, even disparate, features of reality to offer a unified explanation of them? The existence of a maximally good, powerful, intelligent personal being would readily unify these disparate features of the universe and also of human personhood. That being is integrally connected both to the existence of non-necessary, finite morally valuable human beings (made in the divine image) and to those preexisting necessary moral facts that pertain to those finite human beings. Moreover, that divine personal being is powerful and intelligent, which easily accounts for the beginning of a finely tuned, life-producing, stunningly beautiful universe. These otherwise-disconnected or disparate features of reality are far more difficult to unify on a God-less account of reality.

Basic: Which worldview offers a deeper explanation for these fundamental features of the universe (its existence, fine-tuning, etc.) and of our humanity (consciousness, rationality, and value)? As Alvin Plantinga puts it, theism offers a deeper, more stable explanation than naturalism or a God-less worldview: it “offers suggestions for answers to a wide range of otherwise intractable questions” (Plantinga 1995, p. 347). For example, the theist can do better—and go deeper—than a naturalist like Bertrand Russell, who, when asked how the universe came into existence, appealed to the brute fact of its existence: “the universe is just there and that’s all” (in Russell and Copleston 1964, p. 175). However, there is an underlying reason for the contingent universe’s existence—something independent and non-contingent. The universe’s energy, matter, space, and time began to exist a finite time ago. They are not eternal. As cosmologist Alexander Vilenkin notes, “The volume of the universe increases with time. Inflation cannot be eternal and must have some sort of a beginning…. The answer to the question, ‘Did the universe have a beginning?’ is, ‘It probably did.’ We have no viable models of an eternal universe” (Vilenkin 2015). Indeed, the probability of something coming from nothing—being from non-being—is exactly zero. Something non-physical must be the source of a universe whose energy/matter is winding down.

We could say the same about consciousness. While a broad naturalist or a Wielenbergian might claim that “consciousness is just there and that is all” or that “consciousness may sometimes emerge from non-conscious matter,” the theist can go one step further. With a transcendent source of the universe that is also personal and conscious, we can readily make metaphysical sense of the emergence not only of a universe but also of finite conscious beings: from consciousness, consciousness easily comes.

The existence of a supremely self-aware being takes us to a more natural, unified, and deeper, enriched metaphysical level than that furnished by naturalism or a Wielenbergian Platonist position. We could add that the latter positions have no resources to anticipate the universe’s beginning and its razor’s-edge fine-tuning—or the emergence of consciousness and of objective value (and we could add others).

4.2. A Comparison

When making comparisons with naturalism (and, again, this would largely apply to Wielenberg’s position), we can illustrate theism’s wider-ranging explanatory power in Table 1 (adapted from Copan and Taliaferro 2018, pp. ix–x). Theism’s metaphysical furniture better accounts for an array of disparate phenomena than naturalistic or non-theistic views of reality. The phenomena listed below are not surprising if God exists.

Table 1. Theism vs. naturalism.

Phenomena We Observe, Assume, or Recognize	Theistic Context	Naturalistic Context
Things exist. (Yet why does anything exist at all?)	God’s very nature requires his existence. God necessarily exists.	Each of the physical universe’s parts is contingent and thus cannot be self-existent. There is no necessity about the universe.
(Self-)consciousness exists.	God is supremely self-aware/self-conscious.	Consciousness was produced by mindless, non-conscious processes.
Personal beings exist.	God is a personal Being.	The universe was produced by impersonal processes.
We believe we make free personal decisions/choices, assuming humans are accountable for their actions.	God is a spirit and a free Being, who can freely choose to act (e.g., to create or not). The universe did not have to exist.	We are products of material, deterministic processes beyond our control.
Secondary qualities (colors, smells, sounds, tastes, and textures) exist throughout the world.	A personal God is a source of pleasure who gives capacities to his creatures to enjoy or take pleasure in.	The universe was produced from colorless, odorless, soundless, tasteless, and textureless particles and processes.
We trust our senses and rational faculties as generally reliable in producing true beliefs. The world is knowable.	A God of truth and rationality exists.	Naturalistic evolution is only interested in survival and reproduction, not truth. So, many beliefs would help us survive (e.g., the belief that humans have dignity and worth) but would be completely false.
Human beings have intrinsic value/dignity and rights.	God is the supremely valuable Being.	Human beings were produced by valueless processes.
Objective moral facts/values/duties exist.	God’s good character grounds moral values.	The universe was produced by non-moral processes.
The universe began to exist a finite time ago—without previously existing matter, energy, space, or time.	A powerful, previously existing God brought the universe into being without any pre-existing material. (Here, something emerges from something.)	The universe came into existence from nothing by nothing—or was, perhaps, self-caused. (Here, something comes from nothing.)
First life emerged.	God is a living, active Being.	Life somehow emerged from non-living matter.
The universe is finely tuned for human life (known as ‘the Goldilocks effect’—the universe is ‘just right’ for life).	God is a wise, intelligent Designer.	All the cosmic constants just happened to be right; given enough time and/or many possible worlds, a finely tuned world eventually emerged.
Beauty exists—not only in landscapes and sunsets but in ‘elegant’ or ‘beautiful’ scientific theories.	God is creative and capable of creating beautiful things according to His pleasure.	Beauty in the natural world is superabundant and in many cases superfluous (often not linked to survival).
We (tend to) believe that life has purpose and meaning. For most of us, life is worth living.	God has created/designed us for certain purposes (to love Him, others, etc.); when we live them out, our lives find meaning/enrichment.	There is no cosmic purpose, blueprint, or goal for human existence.
Real evils—both moral and natural—exist/take place in the world.	Evil’s definition assumes a kind of norm or standard (how things ought to be, but are not)—a standard of goodness by which we judge something to be evil. A good God’s existence supplies the crucial moral context to make sense of evil.	Atrocities, pain, and suffering just happen. This is just how things are; given naturalism, there is no cosmic ‘plan’ or a standard of goodness to which things ought to conform.
We have deep longings for security (relationship), significance (purpose), forgiveness/relief from guilt, and freedom from fear of death.	God has created us to find our satisfaction in him.	These longings are merely biological, being hardwired into us by naturalistic processes to enhance survival and reproduction.

To support this claim, one can readily quote naturalists and other non-theists themselves, who help reinforce theism by acknowledging how their own worldview has no metaphysical room for, say, objective beauty or human dignity or consciousness or even rationality. We quote noted naturalists/non-theists like Colin McGinn or Thomas Nagel regarding problems with consciousness. Regarding rationality, we could cite the naturalist philosopher Rosenberg (2012), who confessed, albeit in clunky fashion: “It’s a dirty little secret that among us naturalist philosophers that our biggest problem that looks like an inconsequential [view] outside philosophy—and that’s giving a naturalistic account of reason.”10 The same would apply to Wielenberg’s position.

When it comes to horrendous evils, these pose a problem for all worldviews, not simply theism. Arguments against God from evil presuppose both a departure from the way things ought to be and the value of human beings. But both this standard or norm and the reality of human dignity are at home within a theistic worldview, but not at home within a God-less one. We could add that theism guarantees that cosmic justice will be done, that the injustices in the world will not go unaddressed; a God-less worldview has no room for putting injustices to rights.

Theistic philosopher Stewart Goetz notes that, for the strict naturalist, consciousness is a physical phenomenon (and this would be Wielenberg’s view). That would mean that suffering—a conscious experience—cannot be reduced to physical states. What follows from this? Goetz puts it this way: “if you think that there is a problem of evil, then you must, if you are consistent, break ranks with contemporary philosophical orthodoxy” (Goetz 2009, p. 452). Alvin Plantinga similarly argues that horrifying wickedness makes better sense on theism: “Accordingly, if you think there really is such a thing as horrifying wickedness (that our sense that there is, is not a mere illusion of some sort), and if you also think that the main options are theism and naturalism, then you have a powerful theistic argument from evil” (Plantinga 1993, p. 73).

Indeed, when it comes to evil, the relevant question is: Which worldview has the best resources for responding to the problem of evil? As it turns out, theism offers more stable resources than its worldview competitors (see Nagasawa 2024; Copan, forthcoming).

We cannot go into any more detail on all of these features of the chart, however. (See, however, Copan (2012, forthcoming) for further development on this.) Over the years, my own list of naturalistic or non-theistic “supporters” for theism—those who acknowledge their own worldview’s gaps or metaphysical shortcomings in providing an adequate context for a wide range of phenomena—continues to grow.

4.3. On Wielenberg’s Moral Realism

Here I would like to point out specific features of Wielenberg’s own account that draw heavily on the resources of theism to make sense of his own God-less non-naturalistic moral realism. First, he assumes human beings have dignity. Second, he assumes many brute facts—not just one, as we see below. Third, he assumes personal responsibility despite his physicalism and determinism (“compatibilism”). But if my choices are the culmination of prior physical forces over which I have no control and that I cannot do otherwise because of those forces, then it is hard to see how I can be held personally responsible for my actions. Those actions are not up to me but to forces over which I have no control, and my choice is merely the culmination of those forces. And if he is correct about “the causal closure of the physical” (Wielenberg 2014, p. 15), how could the mental affect the physical since the mental is the product of prior physical forces? Fourth, he believes in moral duties, but these make better sense if connected to a personal authority or obligator than to mere moral facts. Let us examine these in turn.

First, Wielenberg takes for granted that humans have dignity (Wielenberg 2021), but his efforts to justify human dignity on secular grounds prove problematic. For starters, Wielenberg’s claim that a theistic account of human dignity is problematic is inadequate; he claims that different perspectives on the image of God allegedly conflict with each other. Is the image of God ontological, functional, or relational? All three views can readily be unified without pitting one against the other, which Wielenberg does. As it turns out, the ontological view (inherent human capacities and value) would be most fundamental, as it grounds both function (humans as stewards over creation and as worshipers of God) and relationality (humans made to relate to God and to fellow human beings). These are reconcilable perspectives, and they all presuppose human dignity.

Let us expound a bit more on Wielenberg’s specific view of human dignity. He maintains that those who have rights and duties do so because of certain functions they perform:

I propose the following answer: any being that can reason, suffer, experience happiness, tell the difference between right and wrong, choose between right and wrong, and set goals for itself has certain rights, including the rights to life, liberty, and the pursuit of happiness, and certain obligations, including the duty to refrain from rape (in typical circumstances). Having such cognitive capacities makes one have such rights and duties.

(Wielenberg 2014, p. 56)

This secular grounding for human dignity is what Wielenberg elsewhere calls psychological worth (Wielenberg 2021, p. 65). He includes two other categories of human worth:

potential worth, which pertains to infants and children: as they develop in their psychological capacities and functions, so does their psychological worth.

symbolic worth: these include humans lacking psychological and moral functions; here Wielenberg maintains that even a human corpse symbolizes someone who once had psychological worth.

(Wielenberg 2021, p. 66)

While the image of God entails the intrinsic worth of each human being regardless of function (see Copan 2021), Wielenberg takes a functionalist (rather than an essentialist) approach; according to his view, a person’s greater or lesser worth is based on his level of mental or moral function. This, though, would lead to the troubling conclusion that someone like Aristotle has greater worth than the average human citizen in any society (see Johnson 2023b, p. 8).11

Note what Wielenberg says: “Having such cognitive capacities [reasoning, choosing right from wrong, setting goals, etc.] makes one have such rights and duties” (Wielenberg 2014, p. 56), Now, when Wielenberg refers to “capacities” here, he refers to actual function rather than, say, latent or dormant human capacities in bound up with our essence or nature. On an essentialist (ontological) view, these capacities may not be actualized for one reason or another, but those capacities are nevertheless bound up with their human constitution. A person may be a fetus and simply needs time for those capacities to be realized—and it would be wrong to deprive that unborn person of future experiences, projects, and activities (Marquis 1989). Or a person may have dementia or Down syndrome, whose mental capacities may have a certain physical blockage from proper mental function; if that blockage were removed, then proper mental function would resume because the inherent capacities can be realized. Or a person may be comatose—or even sleeping—and may not at the moment be carrying out certain functions, but she certainly can exercise those capacities when conscious or wakeful, respectively. More fundamental and defining than function is ontology or nature.

A doctrine like the image of God—in which the value of each human being is rooted in divine personhood and value—avoids the implications of Wielenberg’s functionalist view of human value.12 Furthermore, we could add that Wielenberg’s list of worth-making functions—in a God-less world—could be viewed as mere evolutionary adaptations shaped by natural forces that provide certain advantages for survival and reproduction without going so far as considering these “better” or “superior” to other creatures (Menuge 2016, p. 250).13

Second, Wielenberg assumes not only one brute fact—the Platonic realm of moral facts—but rather an additional whole host of brute facts. I list these brute facts below (taken from Copan 2021, who also elaborates on each of them):

Brute Fact 1: Persons emerged from an impersonal universe.

Brute Fact 2: A universe, which began a finite time ago, is necessary for valuable human beings. (We need a universe.)

Brute Fact 3: The life-permitting conditions of the universe’s delicately balanced fine-tuning are necessary for valuable human beings. (We need a life-permitting universe.)

Brute Fact 4: Valuable human beings—along with all other living things—emerged from non-living matter. (We need a life-producing universe, as the existence of a life-permitting universe does not guarantee it will produce life.)

Brute Fact 5: The existence of valuable human beings requires a life-sustaining universe under highly tortuous conditions. (We need a life-sustaining universe, as the existence of a life-producing universe does not guarantee that life will be sustained.)

Brute Fact 6: A massive cosmic coincidence exists between pre-existing moral facts and the highly contingent emergence of morally valuable human beings.

Brute Fact 7: Consciousness somehow emerged from non-conscious matter.

Brute Fact 8: Morally responsible free agents emerged from materialistic, deterministic, valueless material processes.

Brute Fact 9: Rationality somehow emerged from non-rational processes.

Brute Fact 10: Objective value (e.g., valuable human beings) emerged from valueless processes.

These brute facts presupposed by Wielenberg are not brute facts under theism and are much more at home in a theistic world. For example, just consider the staggering statistical improbabilities stacked on other massive improbabilities—to get from nothing to a finite/contingent universe that happens to be life-permitting and life-producing and life-sustaining, finally arriving at Homo sapiens over a long, unrepeatable, tortuous process (see Copan 2021).14 So even if we grant Wielenberg his pre-existing moral facts without God, his moral realism hangs suspended in metaphysical midair.

Furthermore, God—a necessary being—is not just another brute fact. God by his very nature exists necessarily across all possible worlds. Ironically, Wielenberg argues for necessary moral facts that pertain to highly contingent, non-necessary human beings who (on a non-theistic account) made it onto the scene of existence against staggering odds. By contrast, connecting contingent, valuable human beings to a necessary, supremely valuable being makes far better—and more metaphysically organic—sense. If you are a betting person, the latter option is the far more secure and stable option.

Furthermore, something has always existed. Unless something can pop into existence uncaused out of nothing (which is impossible since being cannot come out of non-being or “no thing”), then it is inescapable that something has always existed. The universe is ontologically haunted (Willard 1990, p. 207). There is an eternally existent source of a contingent universe—an entity that is better positioned to qualify as necessarily existent and self-explanatory. The contingent universe and dependent human beings within it cannot be.

Third, Wielenberg’s own metaphysical position causes instability within his moral realism. Here we come to another tension created by Wielenberg’s two-tiered view. While the chart speaks of “free” agency—a libertarian freedom that seems to be necessary for human personal responsibility—we grant that Wielenberg believes in personal moral responsibility to act on the knowledge of moral facts. Even so, it is hard to see how Wielenberg’s adherence to reductive physicalism and “compatibilism” has any room for robust personal responsibility (see William Lane Craig, Johnson 2021b, p. 201). How can one be held morally responsible for acts one could not help performing? After all, these acts are not up to a person since they are the result of prior physical forces over which she has no control. And how can one live in moral alignment with those abstract Platonic moral facts affirmed by Wielenberg when one does not have the volitional wherewithal to rise above prior physical forces to do so? The “compatibilist” label does not diminish the overwhelming force of prior physical forces on one’s choices. Further, how can one develop character and make moral choices without a mind or self to exert an influence on the physical world?

In addition, Wielenberg holds that “moral properties themselves are epiphenomenal” (Wielenberg 2014, pp. 13–14). But this raises questions about whether they can be known—another challenge for Wielenberg’s two-tiered approach. Wielenberg sees cognition as comprising two “systems”: the adaptive unconscious (“System 1”) and the conscious mind (“System 2”). Given that S stands for the subject, N for a non-moral property, and M for a moral property. If x’s instantiation of N makes x instantiate M, S can reliably believe that M, given that:

(1)

System 1 reliably classifies x as N;

(2)

This causes S to believe that x is M;

(3)

The process type is reliable;

(4)

S has no undefeated defeaters for the belief that x is M (Wielenberg 2014, p. 94).

Menuge points out that there are “many worries” about Wielenberg’s account:

Most fundamentally, since M is epiphenomenal it is hard to see how S can possess the concept of M, and without that concept nothing can cause S to believe that x is M. But even if S does have the concept of M, the fact that System 1 reliably classifies x as N is no reason to expect that S will believe that x is M. In fact, it is not clear why S should have a specifically moral belief at all: for example, why should not x’s being N cause an emotional reaction but no moral judgment? Just because x’s being N makes x to be M is no reason to expect that x’s being N will cause S to believe that x is M.

(Menuge 2016, p. 251)

Naturalist David Papineau points out the problem with moral facts in a causally closed world. Moral facts that are nonnatural can make no difference to what happens in the physical world. We can neither rely on moral facts to inform human beings bound by physicalistic and deterministic processes. And why think that our moral intuitions connect to that transcendent realm of moral facts? The universe’s causal closure undermines this connection (Papineau 2020).

Papineau’s point is actually illustrated by Shafer-Landau. When asked by an interviewer, “What argument against non-natural moral realism do you find most troubling and why?” Shafer-Landau replied, “The argument can be put in many different ways but the core challenge is for nonnaturalists to explain how we can grasp a moral reality not of our own making. I know that genocide and malicious humiliation are wrong. You do, too. Explaining that is no easy task” (Shafer-Landau 2024). This conflict of the two tiers in Wielenberg’s framework applies to Shafer-Landau’s (on this, see Pycraft 2025).

Fourth, moral duties make sense because they are connected to personhood rather than to mere abstract moral facts. Jamie Dew and Paul Gould illustrate the point.

The issue between the theist and the Platonic Atheist boils down to the question of stopping points. Which is a better stopping point to ground moral facts: God or Platonic moral properties? We find the following consideration persuasive on behalf of theism. Consider, I (Paul) am under no obligation to my chair to weigh less than 500 pounds. If I did weigh 500 pounds and sat on my chair and it broke, I would not have wronged the chair. Granted, it would be unfortunate that the chair broke, but I am not obligated to the chair to not sit in it were I to weigh 500 pounds. (I would have obligations to myself and others not to weigh 500 pounds, but not to things such as chairs.) I do, however, have obligations to my students: to tell the truth, to treat them fairly, to not steal their money, and so on. We may ask, why am I obligated to my students but not my chair?

A plausible answer is this. We are obligated to persons, not things. But, according to Platonic Atheism, at rock bottom, my obligations are directed toward things: Platonic moral properties such as the property being just or being honest or being fair and the like. This story strikes many as implausible. We are obligated to people partly because value ultimately resides in persons. Theism accounts for this common moral intuition. Moral properties are ultimately grounded in the supreme person, God himself. The only brute fact is the fact that God exists and has the moral character he has. This, we submit, is the stopping place for explanation. Thus, theism, and not Platonic Atheism, best explains the fact of objective morality (Dew and Gould 2019, p. 234).

This personhood connection extends to the obligatory nature of divine commands: duties (the right) that are rooted in the divine character (the good). Moreland likewise connects the duties to personhood: “Moral laws are prescriptive and, as such, we discover them to have imperatival force and impose duties on us. However, imperatives and duty-impositions come from persons with wills” (Moreland 2023, p. 16). The existence of a good God provides just such a personal and practical authority over human beings to issue commands and to practically communicate them through moral intuitions, reason, conscience, moral exemplars, and so forth, such that humans can understand them (Himma 2025, p. 27).

Furthermore, while Wielenberg and other God-less non-naturalistic moral realists will argue that moral facts are necessary truths—whether that is actually the case—we can still say that a hierarchy of necessary truths exists. They are not all on the same level, and some are more basic than others. Indeed, moral truths stand in an asymmetrical relationship to God’s necessity (Copan, forthcoming). Moral facts may exist in all possible worlds, but so does a good and personal God, who grounds them. A personal God is more basic than the moral facts. Stephen Evans uses the example of prime numbers: We can rightly say: It is false that every prime number is odd is a necessarily true proposition. But the truth of that proposition is explained by a more fundamental necessary truth: 2 is both even and prime (Evans 2013, pp. 221–23). Mathematician Giuseppe Peano distilled five axioms as the basis for number theory: 2 + 2 = 4 is an inference from more basic mathematical axioms such as “0 is a natural number” and “there is always a successor number to a natural number”; “0 cannot succeed any [other] natural number”; etc. Again, certain necessary truths such as 2 + 2 = 4 are inferences from other more basic necessary axioms.

We can likewise speak of the moral fact that torturing innocents for fun is wrong, but this is rooted in more basic moral truths like “Humans have dignity and worth,” and even more basically, “Finite personhood is anchored in God’s infinite personhood” (Copan, forthcoming).

The upshot here is this: Moral realism is bound up with personhood, and abstract moral facts without any intrinsic causal connection to personhood seem odd indeed. Personhood includes not only moral value but also the capacity for rationality, for pursuing moral virtue, acting morally responsibly, and the like. Even if moral facts could exist without a good God, a great deal more is required to bring into existence contingent beings that, on naturalism, accidentally came into existence despite outrageous odds.

5. Concluding Remarks

We have offered a general overview of ethics without God. We have seen that a Russellian strict naturalism is incompatible with moral realism. As for the non-naturalistic God-less moral realism, Wielenberg (along with Shafer-Landau) is the best version offered. Nevertheless, it lacks metaphysical and explanatory robustness—not to mention raising conflicts within the two-tiered system of Platonism and concrete, valuable human beings.

So much more is required to offer a coherent and integrated moral realism—one that theism readily supplies. Thus, even if moral facts could exist apart from God, it is a pyrrhic victory. Theism’s metaphysical resources and explanatory power can incorporate those moral facts into a larger integrated whole. The existence of a good God readily and organically unites moral facts with human beings made in his image. Objective moral facts and the realization of finite value (e.g., human dignity) are rooted in the goodness of divine personhood, which not only provides a stable anchor for moral facts, but also for the existence and the moral constitution and activity of valuable human persons.