Moral argument

The moral argument for the existence of God is a family of arguments that reason from the existence of objective moral facts — binding moral values, obligations, or moral knowledge — to the existence of God as the ground or source of morality. Unlike cosmological arguments, which proceed from features of the physical universe such as contingency or the beginning of time, moral arguments take as their starting point features of the normative order: the apparent reality of objective right and wrong, the experience of moral obligation, or the human capacity for moral knowledge. The central contention shared by all versions of the argument is that a transcendent, personal moral lawgiver provides a better explanation for these moral phenomena than any purely naturalistic alternative.⁸

Moral arguments have a long philosophical pedigree, stretching from Immanuel Kant's postulate of God as a condition of the highest good in the late eighteenth century, through C.S. Lewis's argument from moral awareness in the mid-twentieth century, to the sophisticated analytic formulations of Robert Adams, William Lane Craig, Paul Copan, and Mark Linville in contemporary philosophy of religion. The argument remains a subject of active philosophical debate, with defenders arguing that theism alone can ground the objectivity and normativity of morality, and critics contending that secular frameworks — including robust moral realism, constructivism, evolutionary ethics, and error theory — can account for moral phenomena without invoking a divine being.^{4, 8}

Kant's moral argument

The philosophical use of morality as evidence for God's existence received its first rigorous treatment in the work of Immanuel Kant. In the Critique of Practical Reason (1788), Kant argued that the moral law commands us to pursue the summum bonum — the highest good, understood as a state in which perfect virtue is proportionally rewarded with happiness. Since the attainment of the highest good requires a proportioning of happiness to virtue that is beyond the power of finite moral agents to guarantee, practical reason requires us to postulate the existence of God as the being capable of ensuring that the moral order is ultimately fulfilled. Kant was careful to distinguish this practical postulate from a theoretical proof: he did not claim to have demonstrated God's existence through speculative reason, but rather that the commitment to morality rationally requires the postulation of God as a condition of the intelligibility of the moral enterprise.¹

Kant's argument also included the postulate of the immortality of the soul, since the progressive approximation toward perfect virtue that morality demands could not be completed within a finite lifetime. Together, the postulates of God and immortality constituted what Kant called the "postulates of pure practical reason" — beliefs that are not theoretically demonstrable but that rational moral agents are justified in holding because they are necessary conditions of the coherence of moral practice. Kant explicitly rejected the idea that morality depends on God for its authority; the moral law is autonomous and binding regardless of whether God exists. Rather, God is postulated as the guarantor that the moral enterprise is not ultimately futile — that virtue and happiness can be united in the highest good.^{1, 8}

Kant's moral argument was not, therefore, an inference from the existence of moral facts to God as their metaphysical ground. It was an argument that rational moral agents, committed to the pursuit of the highest good, have a practical rational need to believe in God. This distinction is important because later moral arguments would take a more robustly metaphysical form, arguing not merely that belief in God is rationally needed for moral coherence but that God's existence is required to explain the objective reality of moral values and duties.

C.S. Lewis and moral awareness

In the twentieth century, C.S. Lewis developed an influential moral argument aimed at a popular audience. In Mere Christianity (1952), based on BBC radio talks delivered during the Second World War, Lewis began from the observation that human beings universally appeal to a standard of behaviour that they expect others to recognise — a "Law of Human Nature" or "Moral Law" that is not simply a description of how people do behave but a prescription of how they ought to behave. Lewis argued that this law cannot be reduced to social convention, biological instinct, or evolutionary adaptation, because it stands in judgement over all of these. When we say that one set of moral ideas is better than another, Lewis contended, we are measuring both against a real standard — and the existence of that standard is evidence of a "Mind" behind the universe that is intensely interested in right conduct.²

Lewis's argument proceeded by elimination. The Moral Law is not mere herd instinct, because it sometimes tells us to act against our strongest instincts (for example, the instinct of self-preservation) in favour of weaker ones (such as the impulse to help someone in danger). It is not social convention, because we judge some conventions to be morally better than others and could not do so without a standard external to convention. And it is not a tautology or a truism, because people frequently violate it and feel genuinely guilty when they do — which would make no sense if the "law" were merely a description of how people happen to behave. Lewis concluded that the best explanation for this universal moral awareness is a transcendent moral intelligence — a being that is "not any one of the observed facts but a reality which makes them."²

Lewis's formulation was philosophically informal compared to Kant's, but its influence was considerable. It brought the moral argument to millions of readers outside academic philosophy and anticipated the structure of later analytic formulations, particularly the emphasis on the universal human experience of moral obligation as data requiring explanation.

Contemporary formulations

The contemporary analytic form of the moral argument was shaped decisively by the work of Robert Adams. In a series of influential essays beginning in the 1970s and culminating in his systematic treatment Finite and Infinite Goods (1999), Adams developed a modified divine command theory in which moral obligations are grounded in the commands of a loving God, while the good itself is identified with the nature of God. Adams argued that theism provides the most satisfying account of the "thickness" of moral properties — their action-guiding force, their objectivity, and their connection to human flourishing — and that purely naturalistic accounts leave these features unexplained.^{3, 11}

William Lane Craig subsequently formulated the moral argument as a deductive syllogism and has defended it extensively in academic publications and public debates, making it one of the most discussed arguments in contemporary philosophy of religion:⁴

P1. If God does not exist, objective moral values and duties do not exist.

P2. Objective moral values and duties exist.

C. Therefore, God exists.

The argument is logically valid: it takes the form of modus tollens (if not-P then not-Q; Q; therefore P). The question of its soundness depends entirely on whether P1 and P2 are adequately supported. Craig distinguishes the locus of moral values (God's nature) from the locus of moral duties (God's commands): values are anchored in what God is, while obligations are determined by what God commands, and the commands are constrained by the nature. This allows Craig to combine a divine nature theory of the good with a divine command theory of the right.⁴

Paul Copan has contributed to the defence of the moral argument through both popular and academic work, arguing that the widespread human intuition of objective moral truths — the sense that the Holocaust was objectively evil, that torturing innocents for amusement is genuinely wrong regardless of cultural context — constitutes powerful evidence for P2, and that naturalistic worldviews lack the ontological resources to account for these objective moral truths. Copan has emphasised the self-defeating character of moral relativism, arguing that relativistic positions inevitably smuggle in objective moral assumptions when making claims about tolerance, fairness, or the wrongness of imposing one's values on others.¹⁷

Mark Linville has developed a particularly rigorous version of the moral argument in his contribution to The Blackwell Companion to Natural Theology (2009). Linville's formulation combines the ontological argument (that naturalism cannot ground objective moral facts) with an epistemological argument (that naturalism cannot account for reliable moral knowledge). Drawing on evolutionary psychology, Linville argues that if human moral beliefs are the product of unguided evolutionary processes, then our moral faculties were shaped by natural selection for reproductive fitness rather than for tracking moral truth. On naturalism, the correlation between our moral beliefs and objective moral facts would be an unexplained coincidence. Theism, by contrast, can explain this correlation by appealing to a God who designed the evolutionary process to produce beings with reliable moral intuitions.¹⁸

The premise that objective moral values exist

The second premise of the moral argument — that objective moral values and duties exist — is the more intuitively accessible of the two premises, and its defence draws primarily on moral experience and moral intuition. The most direct argument for P2 is an appeal to moral experience: human beings, across cultures and historical periods, have experiences of certain actions being genuinely right or wrong, not merely approved or disapproved by their society but objectively so. The conviction that enslaving human beings is objectively wrong, that gratuitous cruelty to the helpless is genuinely morally prohibited — these appear to be claims about the way things actually are, not merely expressions of personal preference. To deny P2 is to hold that there is no objective moral difference between a universe in which the Holocaust occurred and one in which it did not. Craig argues that the denial of objective moral values is so counterintuitive that it places a heavy burden of proof on the moral sceptic.⁴

Philosophical support for P2 also comes from the position known as moral realism — the view that there are objective moral truths that exist independently of human beliefs or attitudes. Michael Huemer has argued that our moral intuitions provide prima facie justification for moral beliefs in the same way that perceptual experiences provide prima facie justification for beliefs about the physical world. Just as we are justified in believing that there is a table before us unless we have strong reason to doubt our perception, we are justified in believing that gratuitous cruelty is wrong unless we have strong reason to doubt our moral intuition.⁵

It is important to note that P2 is not a specifically theistic claim. Moral realism is defended by philosophers across the theistic-atheistic spectrum, and many secular moral realists affirm that objective moral truths exist while denying that they require a divine ground. The dialectical significance of this point is that defenders of the moral argument can appeal to the broad philosophical support for moral realism in defence of P2 and then argue, via P1, that this moral realism is best explained by theism.^{5, 8}

God as the best explanation

The first premise of the moral argument asserts a conditional: if God does not exist, then objective moral values and duties do not exist. Defenders argue that on a naturalistic worldview — one in which the universe is ultimately composed of nothing more than matter, energy, and the physical laws governing them — there is no adequate foundation for objective moral normativity.^{4, 8}

The case for P1 proceeds along several lines. First, if human beings are the products of unguided evolutionary processes, then the moral sentiments evolution has instilled in us were selected because they enhanced reproductive fitness, not because they track objective moral truths. On this view, human morality is a biological adaptation, and there is no reason to suppose it reflects an objective moral reality any more than our aesthetic preferences reflect an objective aesthetic reality. As J.L. Mackie argued, the existence of objectively prescriptive moral properties would be "queer" — utterly unlike anything else in the natural world — and the best explanation of our moral experience is that we are projecting subjective attitudes onto a morally neutral world.^{7, 19}

Second, proponents argue that naturalism lacks the resources to account for moral obligation specifically. Values might plausibly exist as brute facts even in a godless universe, but obligations involve a relation between a person and a prescribing authority. To be morally obligated to do something is to be subject to a binding moral command, and commands require a commander. On a naturalistic worldview, there is no prescribing authority capable of issuing universally binding moral imperatives, and so the concept of moral duty becomes metaphysically groundless. Adams developed this point in detail, arguing that moral obligations are constituted by the commands of a loving God and that without such a commander, the category of obligation collapses into mere prudential recommendation.^{3, 11}

Third, defenders contend that even if one could identify natural facts that correlate with moral facts, the normativity of morality remains unexplained. Why ought one to promote well-being? Why is the fact that an action causes suffering a reason not to perform it, rather than merely a description of one of its effects? The transition from descriptive "is" statements to prescriptive "ought" statements — what philosophers call the is-ought gap or Hume's guillotine — presents a challenge for any naturalistic ethics. Theists argue that God's nature, as the paradigm of goodness, and God's commands, as the source of obligation, provide the missing normative bridge.^{4, 6}

Comparison of moral argument variants^{1, 3, 4, 8, 18}

Evolutionary debunking arguments

A major contemporary challenge to the moral argument comes from evolutionary debunking arguments, which contend that the evolutionary origins of human moral intuitions undermine either the case for objective moral truths (attacking P2) or the inference from moral facts to God (attacking P1). The most developed version of this challenge is due to Sharon Street, who has argued that evolutionary forces have profoundly shaped our evaluative attitudes — our dispositions to judge certain things as good, bad, right, or wrong. If natural selection has been the primary influence on our moral beliefs, then one of two conclusions follows. Either moral truths are independent of our evolved attitudes, in which case there is no reason to think our attitudes track those truths (since selection pressures operated on the basis of reproductive fitness, not moral truth), and our moral beliefs are rendered epistemically unreliable. Or moral truths are not independent of our attitudes, in which case morality is not objective in the sense required by P2 but is instead a construction of evolved psychology.¹⁰

Street characterises this as a "Darwinian dilemma" for moral realism: either our moral beliefs are largely mistaken (because unguided evolution would not have tracked moral truth), or morality is not objective. Either way, the moral argument faces difficulty. If moral realism is undermined, P2 is false. If moral beliefs are unreliable, P2 cannot be justifiably asserted even if it happens to be true.¹⁰

Defenders of the moral argument have responded in several ways. One response, advanced by Craig and Linville, is that the evolutionary debunking argument actually supports the first premise rather than undermining the argument as a whole. If evolution alone cannot account for the reliability of our moral beliefs, then atheistic moral realism faces a serious epistemological problem: how can unguided evolution produce creatures whose moral intuitions reliably track mind-independent moral truths? Theism, by contrast, can explain the correspondence by appealing to a God who designed human beings with the capacity for genuine moral knowledge. The evolutionary debunking argument, on this reading, is an argument not against moral realism per se but against naturalistic moral realism — which is precisely what P1 asserts.^{4, 18}

A second response challenges the debunking argument's key assumption that natural selection is the sole or primary influence on our moral beliefs. Theists have argued that if God exists and has designed the evolutionary process, then the evolutionary origins of moral intuitions are not epistemically threatening, because the process was guided rather than blind. Even if empathy and fairness were selected for their fitness-enhancing properties, a designing God could have ensured that the traits selected also correspond to genuine moral truths.⁸

Moral realism without God

Perhaps the most direct philosophical challenge to the moral argument is secular moral realism — the position that objective moral truths exist but do not require God for their existence or explanation. If secular moral realism is viable, then P1 is false: objective moral values and duties can exist without God.

Non-naturalist moral realism holds that moral properties are sui generis — a distinct category of reality that is neither reducible to natural properties nor dependent on a divine source. On this view, moral facts are brute facts about the normative structure of reality, much as mathematical facts are brute facts about abstract structure. David Enoch has defended a version of robust moral realism that treats moral facts as mind-independent, non-natural truths that we access through a faculty of rational intuition analogous to mathematical insight. Enoch argues that we are entitled to presuppose the existence of moral truths because they are indispensable to our practical deliberation — just as we are entitled to presuppose the existence of mathematical truths because they are indispensable to our theoretical reasoning.²¹

Erik Wielenberg has developed the most sustained challenge to P1 under the heading of "godless normative realism." In Value and Virtue in a Godless Universe (2005) and Robust Ethics (2014), Wielenberg argues that objective moral facts can exist as brute, necessary truths without any grounding in God. Just as the truth of mathematical propositions requires no external foundation — it is simply a necessary truth that 2 + 2 = 4 — so the wrongness of torturing innocents for amusement is a necessary moral truth that holds in all possible worlds. Wielenberg contends that the demand for a "ground" of moral truths reflects a misunderstanding: some truths are foundational and need no further explanation. He further argues that supervenience relations between natural and moral properties — for instance, the relation between the natural fact of causing gratuitous suffering and the moral fact of wrongness — can be brute necessary connections that require no divine legislator to establish them.^{16, 22}

Wielenberg has also responded to the epistemological version of the moral argument by proposing a "third-factor explanation" of the reliability of human moral beliefs. On this account, a natural fact (such as the capacity for suffering) both causes our moral belief (that causing suffering is wrong) and makes the corresponding moral proposition true. The natural fact serves as a common cause that explains the correlation between moral belief and moral truth without invoking either God or a cosmic coincidence. This response has been the subject of considerable debate, with theistic philosophers arguing that it fails to explain why the relevant supervenience relations hold in the first place.¹⁶

Naturalist moral realism holds that moral properties are identical to or supervene upon natural properties. On this view, "goodness" can be analysed in terms of well-being, flourishing, or some other natural property. Gilbert Harman's influential work raised the question of whether moral facts play any explanatory role in our best theories of the world, arguing that natural facts alone suffice to explain our moral observations without invoking irreducible moral properties.⁹

Defenders of the moral argument respond to secular moral realism by pressing two questions. First, what explains the existence of these free-floating normative facts? If moral properties are neither physical nor mental but belong to a sui generis category, their existence seems metaphysically mysterious. Second, how do human beings, as physical organisms, gain epistemic access to this abstract moral realm? These questions, proponents contend, receive natural answers on theism but remain unanswered on secular moral realism.^{4, 8}

Constructivism and error theory

Constructivism denies P2 as the moral argument's defenders understand it while preserving a robust role for moral reasoning. On constructivist accounts, moral truths are not discovered but constructed by rational agents through idealised procedures of practical deliberation. Christine Korsgaard, the most prominent contemporary constructivist, argues that normativity does not require an external metaphysical ground — whether divine or Platonic — because it arises from the structure of practical reason itself. A rational agent, reflecting on the conditions of her own agency, discovers that she must value her own humanity and, by extension, the humanity of all rational beings. Moral obligations are thus self-imposed through the exercise of rational autonomy, not imposed by a divine commander.²⁰

Korsgaard's Kantian constructivism follows Kant in grounding morality in the formal requirements of practical reason rather than in any substantive metaphysical claim about the nature of the universe. The categorical imperative — the requirement to act only on maxims that one could will to be universal laws — generates genuine moral obligations without appeal to God, Platonic Forms, or brute normative facts. The obligations are objective in the sense that they bind all rational agents regardless of their desires, but they are not objective in the sense of being mind-independent features of the universe that exist independently of rational agents. Defenders of the moral argument respond that constructivism yields a notion of objectivity that is too thin: if moral obligations are constructed by rational agents, then they are contingent on the existence of rational agents and would not exist in a universe without them, which falls short of genuine moral objectivity.^{20, 8}

Contractarian approaches, such as that of David Gauthier, offer a related alternative. Gauthier argued that moral norms can be grounded in the agreements that rational agents would reach under idealised bargaining conditions, deriving morality from rational self-interest without any metaphysical foundations beyond the existence of rational agents with interests. Critics of the moral argument who adopt contractarianism accept that morality does not require a transcendent ground, while defenders respond that contractarianism fails to account for the unconditional character of genuine moral requirements — obligations that bind agents regardless of whether compliance serves their interests — and that it excludes from moral consideration beings who fall outside the scope of the bargaining process.¹⁵

Error theory represents the most radical response to the moral argument. As developed by J.L. Mackie in Ethics: Inventing Right and Wrong (1977) and defended more recently by Jonas Olson and Richard Joyce, error theory accepts the moral argument's conditional (P1) but denies P2: there are no objective moral values or duties. All moral judgements that purport to state objective moral truths are systematically false. Mackie argued that objective moral properties would be metaphysically "queer" — utterly unlike any other properties in the natural world — and that the simplest explanation of our moral experience is that we project subjective attitudes onto a morally indifferent universe. The "error" in error theory is the error of ordinary moral discourse, which speaks as though moral claims are objectively true when in fact they are not.^{19, 23}

Error theory poses a distinctive challenge to the moral argument because it concedes P1 (morality does require God to be objective) while rejecting P2 (morality is not, in fact, objective). Defenders of the moral argument respond that error theory is self-defeating in practice: no one, including the error theorist, actually lives as though moral claims are uniformly false. The phenomenology of moral experience — the sense that torturing innocents is genuinely wrong, not merely disapproved — is so powerful that denying it carries a heavier burden of proof than affirming it. Craig has argued that the error theorist's position is less plausible than either of the moral argument's premises, making the argument's conclusion more reasonable to accept than its negation.^{4, 12}

The companions in guilt strategy

The companions in guilt strategy is an important dialectical move in the debate over the moral argument. The strategy challenges the claim that moral facts are uniquely problematic for naturalism by pointing to other categories of objective, normative, or abstract facts that atheists and theists alike accept without invoking God as their ground. If these "companion" facts are structurally analogous to moral facts and do not require a theistic explanation, then moral facts do not require one either — and P1 of the moral argument is undermined.^{8, 16}

The most common companion cited is mathematics. Mathematical truths are objective, necessary, mind-independent, and non-natural. They are not physical objects, they are not reducible to empirical observations, and yet they are binding in the sense that rational agents cannot coherently deny them. If mathematical truths can exist as brute, necessary features of reality without requiring a divine mathematician, then (the argument runs) moral truths can exist as brute, necessary features of reality without requiring a divine moral lawgiver. Wielenberg has pressed this analogy with particular force, arguing that the supervenience of moral properties on natural properties is no more mysterious than the supervenience of mathematical properties on physical structures.^{16, 22}

Other companions in guilt include logical truths (the laws of logic are prescriptive norms for reasoning, yet they are not typically grounded in God), epistemic norms (the norm that one ought to believe what the evidence supports is objective and action-guiding, yet it is not usually attributed to divine command), and aesthetic truths (if there are objective facts about beauty or artistic merit, they seem to exist without divine grounding). The strategy does not require that all of these companions are perfectly analogous to moral facts; it requires only that some of them are sufficiently similar to cast doubt on the claim that moral facts are uniquely in need of a theistic foundation.⁸

Defenders of the moral argument have responded to the companions in guilt strategy in several ways. Craig and others have argued that the analogy between moral and mathematical truths is weaker than it appears. Mathematical truths are descriptive — they describe relations between abstract objects — whereas moral truths are prescriptive — they impose obligations on agents and carry with them the force of genuine "oughtness." The prescriptive, action-guiding dimension of morality, theists contend, is categorically different from the descriptive character of mathematical truths and does require an explanation in terms of a personal will or commanding authority. Adams has argued that the relational, personal nature of moral obligations — the fact that to be morally obligated is to be obligated to someone — distinguishes moral normativity from logical or mathematical necessity and points specifically to a personal God.^{3, 4}

Others have responded by accepting the challenge and extending the theistic account. Some theistic philosophers have argued that God does ground mathematical and logical truths as well as moral truths, locating all necessary truths in the divine intellect. This "theistic activism" about abstract objects avoids the companions in guilt problem by denying that any category of necessary truth is ungrounded, but it generates its own philosophical difficulties regarding the relationship between God and abstract objects.¹⁴

The Euthyphro dilemma

The oldest and most discussed objection to theistic accounts of morality is the Euthyphro dilemma, which originates in Plato's dialogue Euthyphro (c. 399–395 BCE). In its monotheistic formulation, the dilemma asks: is something morally good because God commands it, or does God command it because it is morally good? The first horn implies that morality is arbitrary — God could have commanded cruelty, and cruelty would then be good. The second horn implies that moral goodness is independent of God, existing as an external standard to which even God must conform, in which case the moral argument loses its force because moral values require no theistic explanation.¹³

The most influential theistic response, developed by Adams, Alston, and Plantinga, rejects the dilemma as a false dichotomy by proposing a third option: moral goodness is grounded not in God's commands (first horn) nor in a standard external to God (second horn) but in God's nature. God does not arbitrarily decide what is good, nor does God conform to an independent moral standard; rather, God's own nature — God's essential goodness, love, justice, and holiness — is the standard of goodness itself. God's commands flow necessarily from God's nature, so they are neither arbitrary nor external. The question "Could God have commanded cruelty?" receives a negative answer: since God's nature is essentially good, God could not command cruelty any more than a necessarily truthful being could lie.^{3, 14}

Critics have responded that this third option faces its own difficulties. If God's nature is the standard of goodness, then in what sense is it meaningful to call that nature "good"? The claim threatens to become either circular or substantive but requiring further justification. Others have argued that locating moral goodness in a divine nature rather than in divine commands does not fully resolve the structural problem, because one can still ask whether God's nature is good in some independent sense or only trivially "good" by self-definition.⁷

Current state of the debate

The moral argument remains one of the most actively debated arguments in contemporary philosophy of religion, in part because it engages questions at the intersection of metaethics, epistemology, philosophy of mind, and evolutionary biology. The debate has become increasingly technical, with both defenders and critics deploying sophisticated tools from analytic philosophy.

On the theistic side, the argument has been refined in several directions. Craig continues to defend the deductive formulation in both academic and public settings. Linville's combined ontological-epistemological version has added a new dimension by pressing the question of moral knowledge alongside the question of moral ontology. Adams's framework, grounding values in God's nature and duties in God's commands, remains the most philosophically developed theistic metaethics and has been widely adopted by defenders of the moral argument.^{3, 4, 18}

On the secular side, the most significant recent development has been the emergence of robust secular moral realism as a serious philosophical programme. Wielenberg's godless normative realism and Enoch's indispensability argument for moral realism represent sustained attempts to show that moral objectivity does not require theistic foundations. These programmes have shifted the debate from the question of whether secular morality is possible to the more nuanced question of whether theism or secular realism provides the better explanation of the features of morality — its objectivity, its normativity, its knowability, and its authority over rational agents.^{16, 21}

The evolutionary debunking literature continues to generate productive exchanges. Street's Darwinian dilemma has been taken up by defenders of the moral argument as evidence that naturalism and moral realism are in tension, while secular realists have developed increasingly sophisticated responses, including Wielenberg's third-factor account and Enoch's pre-established harmony defence. The question of whether evolutionary explanations of moral belief undermine moral knowledge or merely explain its psychological origins remains unresolved.^{10, 16}

The philosophical assessment of the moral argument ultimately depends on one's prior metaethical commitments. Those who accept moral realism but find secular accounts of its grounding inadequate may find the argument compelling. Those who accept secular moral realism will regard P1 as unsubstantiated. Those who deny moral realism altogether — error theorists and expressivists — will regard P2 as false. Constructivists will challenge the sense of "objective" operative in the argument, arguing that morality can be authoritative and rational without being mind-independent in the metaphysically robust sense the argument requires. What the argument accomplishes, regardless of one's assessment of its soundness, is to force a serious engagement with the question of what grounds morality — a question that any comprehensive philosophical worldview must address.^{8, 12}