Skip to main content

Response to Chapter 15 of Russ Shafer-Landau’s book Whatever Happened to Good and Evil? “Does Ethical Objectivity Require God?” Part VIII

By David Baggett

We’re discussing Russ Shafer-Landau (SL), and his critique of theistic ethics. He started with the Euthyphro Dilemma, and then uses analogies to make his point better. He asks us to envision a referee at a sporting match. A good referee is good in virtue of following the rules of the game, rather than making up new rules willy-nilly. A good referee can cite reasons for his calls, and reasons that aren’t merely ad hoc, made up on the spot, lacking rationale.

He admits it may sound odd, or mildly blasphemous, to liken God to a sports referee, but he doesn’t think there’s much harm in it. “The Divine Command Theory has us picture a God who controls our game in its entirety, making up all the rules, perhaps continually, and having no need to cite any reasons on their behalf.” For what other reasons could there be? “If there are not moral rules or reasons prior to God’s commands, then there is nothing God could rely on to justify the divine commands. So any choice is arbitrary.” Had God chosen differently, “we’d be saddled with a morality that encourages torture, pederasty, perjury, and all sorts of other things we now recognize to be evil.”

Recall, though, that on a view like that of Adams’, God typically commands something that’s good. He may have had plenty of reasons to provide the additional moral reasons to perform a particular action that we already had moral reasons to perform. The goodness of the action is one reason for God to command it, and the additional motivation for us that the command would provide is another, and those are just two examples. DCT makes an action right, not good, to the thinking of leading DCT’ists today. Presumably, in his infinite wisdom and knowledge, God has compelling reason to issue the command, rendering an already good action morally obligatory. But this is not to say that he couldn’t have done otherwise, at least on some occasions. It’s plausible to many, including me, that at least some of God’s commands are contingent. Not all of them follow ineluctably with necessity from his nature; he retains, at least with respect to certain actions, to command them or not to command them. The goodness of the action isn’t affected, but rather whether it’s obligatory or not. Perhaps God might even speak to me personally, commanding me to perform an action, that otherwise wouldn’t be obligatory—like help a particular homeless person. It becomes my duty once he issues the command.

Another important point to remember here is that if we’re dealing with a God of perfect love, there are some things God simply would never command. They would be inconsistent with his character. To say God is essentially loving, for these words to retain their meaning, is to suggest that some actions—those that are irremediably hideous and treacherous, for example—are ruled out. The ascription of love and goodness to God has determinate content, ruling some things out. So though God may retain a measure of divine prerogative in issuing various commands, there are still some commands outside his character he would never command. In fact, it’s right to say he can’t, in the sense, to put it into the terms of modal logic, there’s no metaphysically possible world in which he does issue such a command. As the delimiter of possible worlds, on an Anselmian conception, there are likely worlds and states of affairs we can vaguely conceive of or imagine that nevertheless don’t constitute genuine possibilities.

Cover for 

Whatever Happened to Good and Evil?

Now, when we say God is good, SL thinks the only sense we can make of such an ascription is that God follows the moral rules. But this is where the long tradition of analogical predication in the history of the Christian church may prove handy. When we say God is good, we’re not saying God is good in exactly the same sense that we attribute goodness to people. Human beings may be good to one degree or another, but God is, on a view like that of Adams’, goodness itself, the paradigm, the exemplar, the archetype of the good. Ultimate goodness is a person, not a set of principles. In fact, there’s something deeply intuitive about making persons the locus of goodness. States of affairs may be pleasant or unpleasant, but aren’t morally good or bad. People are. It makes sense to think of persons as the primary subjects of goodness, but no merely human person is perfectly good. God, though, almost by definition, is perfectly good. Whether we predicate perfect goodness of God or identify God with goodness, or both, God’s goodness is nonnegotiable on Anselmianism. But his goodness isn’t univocal with our own; ours is the imperfect wheel; his is the perfect circle. There’s relevant resemblance, but also infinite distance, as God is perfect and we are far from it.

So this isn’t equivocation, but analogical predication, with which we can still meaningfully, in a sort of analogically extended sense, ascribe goodness, indeed perfect goodness, to God. If A. C. Ewing was right—and I think he was—this is also consistent with God functioning at the foundation of ethics, for the source of the good is also most plausibly taken to be perfectly good. Obviously, though, all of this is a far cry from SL’s simplistic and minimally charitable analogies and caricatures.

SL anticipates that some will object and say God’s command of rape or torture is impossible. “A good God would never allow such a thing.” Right enough, SL replies. “But what does it mean to be good? If the Divine Command Theory is correct, then something is good just in case it is favored by God. But then look what happens: to say that God is good is just to say that God is favored by God.” That’s not very informative, and in fact wouldn’t preclude a self-loving being from issuing hideous commands.

True enough, except note that SL is offering a DCT account of goodness, having earlier confined it to rightness. This may not have been intentionally duplicitous; he may have just used rightness as a generic term for morality, a penumbral term under which falls both goodness and rightness. But for present purposes, the distinction is a crucial one. DCT nowadays is nearly always delimited to deontic matters, rightness rather than goodness. For extended accounts of how and why God is aptly thought of as good, see the work of Evans, Hare, Adams, etc.

SL is convinced he knows exactly from what an ascription of goodness to God must derive: “A good God, like a good referee, is one who plays by the rules. When we speak of God as morally good—indeed, as morally perfect—what we really mean is that God cannot fail to uphold and respect all moral rules.” SL seems to be operating on the assumption that a perfect God either is perfect in virtue of following all the moral rules or is a vacuous conception because it means he can change the moral rules at will. But surely those don’t exhaust the alternatives. Recall the earlier point that God indeed can’t change the moral rules at will; there are indeed constraints on his behavior if he’s perfect; it’s just that the constraints happen to be entirely internal to his character. They’re a feature of his perfection. A God who could commit suicide, deny himself, or lie would be imperfect. The constraints don’t threaten his omnipotence or sovereignty, but help reveal it. Recall that on an Anselmian picture God possesses all the great-making properties to the maximally compossible degree, which admit of intrinsic maxima.

SL is convinced the analogy is close between referees and games, on the one hand, and God and morality on the other. But I am not. SL’s insistence is on a God who is not the ultimate reality, but distinctly secondary. He refuses to acknowledge relevant disanalogies between human referees and the divine, and he thinks that constraints on God’s actions necessitate that morality doesn’t find its foundation or locus in God. He does much of this by illegitimately assuming the only theistic ethic on offer is a radically voluntarist version of DCT, and he ignores the illuminating good/right distinction in the process.

Again, he argues that if the moral character of torture is fixed prior to God’s reaction to it, then God is not the author of the moral law. But the moral character of an action is not just based on divine commands. Its goodness or badness traces to a different foundation (on Adams’ view, and that of most DCT’ists). The action may already have lots of moral features to it besides being obligatory, permissible, or forbidden. Its moral hideousness, for example, might already obtain. And God’s command against an action in certain cases, I’ve argued, isn’t contingent, but necessary, meaning such commands couldn’t have been otherwise. This actually makes good sense of necessary moral truths even in deontic matters—and a better explanation of them, to my thinking, than what (nontheistic) nonnaturalists can offer. This resonates nicely with Plantinga’s suggestion in “How to be an Anti-Realist” that the necessary truths can offer an insight into God’s unchanging character.

In the next blog, at long last, I’ll wrap up my response to this chapter of SL’s.

The Third Option to the Euthyphro Dilemma

By Frederick Choo

In general, Divine Command Theory (DCT) says that “If God commands X, then X is a moral obligation for us.” I will limit my discussion of DCT to moral obligations and prohibitions, which are used synonymously with rightness and wrongness. These are deontic properties which is distinct from goodness, which is axiological. For example, something can be good to do, such as becoming a lifeguard to save lives, but we do not have a moral obligation to do so. So I will use DCT as a theory of rightness that presupposes a theory of the good.

The Euthyphro Dilemma (ED) is often raised against DCT. For example, in the case of rape Walter Sinnott-Armstrong asks, “Did God have any reason to command this? If not, his command was arbitrary, and then it can’t make anything morally wrong. On the other hand, if God did have a reason to command us not to rape, then that reason is what makes rape morally wrong. The command itself is superfluous. Either way, morality cannot depend on God’s commands.” In short, the ED says:


(1) God has no reasons for His commands,


(2) God has reasons for His commands but these reasons are sufficient by themselves in explaining moral obligations.

Embracing (1) leads to objections such as God’s commands being arbitrary which makes morality arbitrary. Furthermore, this means that God’s commands could possibly be what we consider abhorrent, such as commanding that we ought to torture babies solely for fun resulting in a moral obligation to do so. Any objection to this that says God has reasons is a move away from (1).

Embracing (2), shows that actions are morally obligatory prior to and independent of God’s commands, making God at most an epistemic authority who is just conveying His perfect moral knowledge to us. However DCT proponents want God’s commands to explain moral obligations instead.

From the ED, I think a third option is clear, which DCT proponents can well affirm:

(3) God has reasons for His commands but these reasons are not sufficient by themselves in explaining moral obligations without God’s commands.

God just needs good reasons to make an act morally obligatory. An act itself does not have the property of being morally obligatory prior to God’s command, but can have other relevant properties, such as being morally good or even “non-moral considerations ultimately based in God’s nature.” God’s commanding however adds certain properties that make the act obligatory. To use an analogy, let us think of other obligations. Consider a legal obligation not to smoke in a certain area when implemented by law. For the obligation to arise, there must be good reasons behind why it is implemented by law. Yet those reasons by themselves are not sufficient to give us legal obligations unless it is actually implemented by law. Hence a legal obligation arises because it is implemented by the law and there are good reasons for it being implemented. Likewise, DCT proponents say that a moral obligation arises because it is commanded by God and God has good reasons to command it.

One objection to (3) is based on a principle that moral properties strongly supervene on non-moral properties necessarily. Matthew Jordan says, “The doctrine of global moral supervenience, the uncontroversial thesis that any two possible worlds that are identical in all non-moral respects must be identical in all moral respects, implies that moral truths – at least the most fundamental ones – are metaphysically necessary.” So moral obligations are in some way determined and fixed by their non-moral properties. How exactly does moral supervenience amount to an objection to (3) exactly?

In “An Essay on Divine Authority”, Mark C. Murphy argues that DCT “must be false, for it, in conjunction with a very weak and plausible claim about God’s freedom in commanding, entails that the moral does not supervene on the non-moral.” To show this, he argues that according to voluntaristic versions of DCT, where God is free to choose what to command, there can be two possible worlds exactly the same in their natural features, but God gives different commands and thus we have different moral obligations in two possible worlds that have the same natural features. This seems to violate the supervenience of the moral on the non-moral, since two worlds with the same natural features should have the same moral obligations.

How may a proponent of a voluntaristic version of DCT reply? C. Stephen Evans points out that for the theist, non-moral properties can include both natural and supernatural properties. Supernatural properties are “properties possessed because what has the properties has a certain kind of relation to God,” such as “being commanded by God”, “being preferred by God,” or “being pleasing to God” or “being conducive to a better relation to God.” If an act is commanded by God, then it will have the further properties mentioned, such as “being conducive to a better relation to God” which is a non-moral property. These non-moral properties may even be linked to natural properties such as “being conducive to the agent’s happiness.” If a relationship with God is conducive to our happiness, and such a relationship requires that we follow what He commands, then the property of “being commanded by God” would be one that could alter the moral status of an act, especially for those who think that the moral status of an act is linked to whether the act is conducive to an agent’s happiness. Hence on DCT, it is both natural and supernatural properties that make up non-moral properties which moral properties supervene on. If so, then there can be two worlds alike in all their natural properties but differ in their supernatural properties, and hence moral properties can be different as it supervenes on both. So moral supervenience along with God’s freedom does not amount to an objection against (3).


Evans, C. Stephen. God and Moral Obligation. Oxford, United Kingdom: Oxford University Press, 2013.

Jordan, Matthew Carey. ““Theism, Naturalism, and Meta-Ethics”.” Philosophy Compass 8, 2013, 373-380.

Miller, Christian B. “Euthyphro Dilemma.” In Blackwell International Encyclopedia of Ethics, edited by Hugh LaFollette. Malden, MA: Wiley-Blackwell, 2013

Murphy, Mark C. An Essay on Divine Authority. Ithaca, N.Y.: Cornell University Press, 2002.

Sinnott-Armstrong, Walter, “Why Traditional Theism Cannot Provide an Adequate Foundation for Morality”, in Is Goodness without God Good Enough: A Debate on Faith, Secularism and Ethics, edited by Robert K Garcia and Nathan L King, 101-115. Lanham: Rowman & Littlefield Publishers, 2008.

Smith, Michael. The Moral Problem. Oxford: Blackwell, 1994.
Image: “Socrates” by Oscar Anton. CC License. 

Chapter 4, Part II, of C. Stephen Evans’ God and Moral Obligations: “Objections to Divine Command Theory”:

Summary by David Baggett

God and Moral Obligation by C. Stephen Evans 

In this chapter Evans raises and attempts to answer several common objections to Divine Command Theory. This post will cover the last four objections discussed.

The Prior Obligations Objection

Wainwright calls this objection the Cudworth objection, since it can be found in the writings of Ralph Cudworth. It’s rooted in the worry that DCT is too narrow and does not account for all genuine moral obligations. A full-fledged DCT holds that all moral obligations are divine commands. The prior obligations objection argues that there must be some moral obligations that are not grounded in divine commands because they hold antecedently to or independently of divine commands. Specifically, the claim is that humans have a moral obligation to obey God. This obligation is not itself grounded in God’s commands. There’s a prior obligation to obey God that explains why God’s commands can create new obligations. So there must be some moral obligations other than those that are created by God’s commands.

Defenders of DCT could say that these “prior obligations” to obey God are not actual obligations but just hypothetical ones: To say that we have a prior obligation to obey God is just another way of saying that we are obligated to obey God if God issues commands, but there is no actual obligation until a command is issued. The conditional proposition itself can be understood as simply spelling out the meaning of the claim that God has moral authority. To say that God has moral authority is just to say that he has the right to issue commands that ought to be obeyed. The sense of prior oughtness should not cause the DCT’ist worry, Evans thinks.

Notice, too, that this objection could apply to any moral theory, but such objections are wrongheaded. But the critic might object and suggest there is nothing hypothetical about the fact that we ought to obey God. It’s a standing obligation, not a hypothetical one. But to this Evans replies by reminding readers of his earlier distinction between sorts of ought statements—not all of them are moral obligations. We may even have moral reasons to obey God without those reasons constituting moral obligations.

One final reply Evans offers is this, which can supplement what’s already been said: perhaps, since the Bible does command us to obey God, God can convert the antecedent fact that humans ought to obey God’s commands into an actual moral duty, in what Evans calls a “bootstrapping” manner. On this view it is both true independently of God’s commands that one ought to obey God, but also true that “one ought to obey God” is in fact a moral obligation, because God has commanded us to keep his commands.

At worst, the objection would show that some obligations aren’t rooted in God’s commands, whereas others may be.

The Supervenience Objection

Murphy offers a version of this objection. It’s directed against an identity version of DCT, according to which God’s commands just are moral obligations. God’s prerogative in issuing commands, combined with an identity version of DCT, Murphy argues, entails the falsehood of the supervenience of the moral on the nonmoral. The idea behind this notion of supervenience is that the moral properties of things are supposed to be fixed by their non-moral properties. Moral properties, such as being obligatory, do not “float free of their other properties. So it can’t be the case that two actions are exactly alike except that one of the acts is morally obligatory and the other is not obligatory. Evans agrees with Murphy on this, and that if DCT entailed the falsehood of supervenience, that would be a problem for DCT.

Now, why does supervenience appear to be threatened? Because if God has discretion in his commands, then there could be two possible worlds exactly alike in their natural features, but in which God gives different commands. If moral obligations are determined by God commands, then it follows that actions in one of those worlds will be exactly like those in the other except for being morally obligatory. But this violates the supervenience of the moral on the non-moral.

It is true there’s a difference in the two situations: in one world God issues one command, and in the other a different command. But if God’s commands are moral obligations (on the identity thesis), the difference is a moral difference, but there’s no difference in nonmoral or natural properties, and thus a violation of supervenience.

Now, one response might be to reject divine discretion here, but Evans isn’t willing to do that. A second possible reply is to appeal to weak supervenience rather than strong supervenience. According to strong supervenience, “property x supervenes on property y” = a claim that in any possible world whatever has property y will also have property x. On weak supervenience, instead, “property x supervenes on property y” = it is necessarily the case that anything that has property y in some particular world will also have property x in that world. Since weak supervenience confines its requirement to a single possible world, Murphy’s transworld constraint is removed and supervenience is easier to satisfy. But Evans doesn’t go that route, opting instead for a way to satisfy both strong and weak supervenience.

To do so, Evans points out the distinction between non-moral properties and natural properties. There may be non-moral but supernatural properties: such as the property of being pleasing to God—in each situation, what it pleases God to do might be different. So there might be a difference in nonmoral properties after all, and supervenience is thereby preserved.

Consider two situations identical in all their natural properties, but God commands X in one of them but not the other (either in different worlds or in the same world). The command/noncommand is a moral property, but other supernatural properties are non-moral, such as: “being pleasing to God” or “being preferred by God.” This preserves the supervenience intuition: “no difference in moral properties without some difference in non-moral properties.”

Mysterious Relationship Objection

What is relation between divine commands and moral obligation? Three possibilities are:

  1. God’s commands are the cause of moral obligations.
  2. Moral obligations supervene on God’s commands.
  3. God’s commands are identical to moral obligations.

Having discussed identity and supervenience variants of DCT, Evans considers a causal version. Murphy argues that the problem with a causal version of DCT is that such a view does not allow for God’s commands to be at least partially constitutive of one’s reasons for actions. It’s true that if God commands us to do X, then we are obligated to do X, and thus have reason to do it. But it’s not God’s command that gives us that reason. A bully who threatens to beat you up unless you do X gives you a reason to do X, but the bully doesn’t have moral authority over you.

In reply, Evans suggests that the proponent of a causal version of DCT should say that it is the fact that God has the requisite authority that gives him the causal power to establish moral obligations. The special sort of causation at play is the authority of God that makes God different from the bully.

Promulgation Objection

Recipients of God’s commands, according to some critics, must know that the commands come from God; otherwise no obligation is generated. I may be obliged to loan a car to my friend to whom I owe a favor, but an anonymous note saying to leave my car with its keys available produces no duty to do so. Likewise with God’s commands if we don’t know where they came from.

How does Evans reply? He rejects the idea that for God to successfully communicate a command to humans, he must communicate the command in such a way that it is obvious that the command comes from God. Two helpful distinctions can be brought to bear here: (1) We must distinguish between the recognition of a moral obligation and the recognition of the moral obligation as a divine command; and (2) one must distinguish a recognition of a moral obligation from an explanation of the existence of a moral obligation.

Believers and unbelievers alike can recognize various moral obligations. It might be a function of God’s grace for God to conceal that he’s behind obligations initially in order to mitigate judgment on those who disregard them.

Find the other chapter summaries here. 

Image: “Inside Pieterskerk, Leiden” by Canadian Pacific. CC License.