The classic construal of the problem of evil raises the question of the logical compatibility and evidential probability of God’s existence given the conspicuous prevalence of human and animal suffering. In this paper, I wish to address objections to God’s existence based on the contention that the prevalence of suffering is logically incompatible with an omniscient, omnipotent and perfectly good God. My response will focus on two philosophical assumptions that have driven the debate and shape the way in which theists have fashioned their replies to objections raised by the prevalence of suffering. Both assumptions concern the nature of God’s relation to moral goodness, namely: 1) that God’s status as a perfectly good being requires that he adhere to a standard of goodness existing independently of God, and that this standard places obligations upon God to act in cases of suffering unless a God-justifying reason exists to morally excuse God’s refusal to act; and 2) that theories of moral goodness that index God’s will as the source of obligations cannot adequately account for a non-arbitrary moral order and deprive God of moral praiseworthiness. By offering replies to these two assumptions – often held by theists and atheologians alike – I aim to provide an interpretation of divine command theory (DCT) that will eliminate the logical problem of evil while preserving a coherent picture of God’s relation to moral goodness.
While I think DCT can help us overcome the logical objections to God’s existence by clarifying the nature of God’s moral obligations, I concede that it cannot provide satisfactory explanations for why God permits suffering and I recognize the need to explore possible explanations in the form of free-will defenses and other theodicies. While I appreciate and approve of the human impulse to comprehend why God might allow suffering, I deny that these explanations must take the form of moral justifications. God’s actions cannot, given DCT, be anything other than morally blameless and our expectation that God adhere to the same standard of moral goodness imposed upon us is the result of the misapprehensions listed above.
I do not intend to offer an innovative refashioning of DCT, but rather aim to interpret its implications in a strict, uncompromising manner that some philosophers may find unsettling. Robert M. Adam and Marilyn M. Adams come closest to endorsing my position, I think, but I will argue that they stop short of adopting what I view to be the inescapable consequences of DCT. Their reservations needlessly complicate the discussion of the problem of evil and strengthen the objections to God’s existence raised by atheologians.
Before turning to the relevant philosophical literature, it may be helpful to consider the exact nature of the atheologian’s logical objections to God’s existence from evil. What does the atheologian object to when she cites the prevalence of suffering or cases of horrendous evils? I submit that she objects to the logical compatibility of suffering and God’s purported perfect goodness. How, she asks, can God be perfectly good when God allows innocent people to suffer and suffer horribly? For her, God is bound by moral duties to intervene in cases of suffering unless a moral justification exists that countervails against such demands. The relevant question for our discussion is whether God’s essential attributes – including perfect goodness – entails moral obligations. Does God’s moral perfection require God to perform certain acts in accordance with moral standards, or does it mean that God, as the originator of moral obligation, is morally blameless regardless of God’s actions? The atheologian who objects to the logical compatibility of God and suffering takes the former interpretation, while the divine command theorist, I argue, ought to take the latter.
By committing herself to the supposition that God’s perfect goodness entails moral obligations, the atheologian adopts one of the two horns of the Euthyphro Dilemma. For her, God loves goodness because it is good, not the other way around. Some independent standard of goodness exists, she claims, to which God must adhere, a situation that very closely parallels the structure of our own moral duties. Unlike finite human beings, however, God’s omnipotence ensures that God can always meet the demands of moral obligation, leaving God without excuse for moral failings. How then can God justify God’s failure to intervene in cases of suffering? For we typically recognize a standing obligation to personally intervene in cases of preventable suffering when we are able, and God’s infinite ability would appear to obligate God in every instance of suffering.
If God’s perfect moral goodness obligates God in the way described above, then the burden of justifying God’s failure to intervene falls to various defenses and theodicies. These must take the form of moral justifications, because on the atheologian’s reading God’s non-intervention threatens to violate God’s moral obligations and thereby God’s essential attribute as perfectly good. The theist responds by supplying plausible, God-justifying reasons to show how God manages to satisfy God’s moral obligations despite allowing for suffering.
This entire procedure, I submit, results from a confused understanding of God’s relation to goodness and ignores an alternative conception: DCT. Why think that God’s moral perfection demands that God adhere to an independent standard of goodness when divine command theorists hold that God is the originator of all obligations? I suggest that, on DCT, God’s perfect goodness ought to be understood as blamelessness, or the incapacity on God’s behalf to violate any moral obligations whatsoever. Given that one interprets DCT to mean that God creates moral obligations by issuing commands that do not themselves constrain God, the atheologian’s objection to God’s failure to intervene in cases of suffering no longer constitutes a moral objection, nor can the atheologian cite suffering as logical evidence against the existence of God. If God possesses no moral obligations then suffering does not jeopardize God’s perfect moral goodness and the problem of evil ceases to be a logical problem.
And yet the problem persists despite the work of many divine command theorists. Why? I suspect that the debate continues because divine command theorists have not availed themselves of the full power of their argument out of concern for the implications of embracing it unreservedly. To equate moral goodness with God’s will without including some provision that attenuates the specter of arbitrariness would, in their minds, open their theory to significant, philosophically valid criticism. There persists then an abiding worry on behalf of Christian ethicists over whether the most robust formulations of DCT satisfy a general requirement for moral stability and whether or not they succumb to unpalatable degrees of arbitrariness. This worry causes divine command theorists to withhold from adopting the strongest of the available arguments from DCT against the problem of evil and strengthens the atheologians argument against God. In what follows, I will show in what ways the divine command theorists needlessly restrict their arguments to accommodate fears over arbitrariness before offering some of my own suggestions on how to overcome the worry. First, it will be helpful to canvass some of the relevant literature.
In the following passage, Eleonore Stump offers a brief analysis of the two classic theories of God’s relation to goodness available to theists:
“There seem to be two main sorts of competing Christian theories concerning what is to count as good. Either God’s will is taken to create morality, so that whatever God wills is good just because he wills it: consequently, (TS) right actions are right just because God approves of them and wrong actions are wrong just because God disapproves of them. Or morality is taken to be grounded independently of God, so that God frames his will in accordance with those independently existing standards of goodness: consequently, (TO) God approves of right actions just because they are right and disapproves of wrong actions just because they are wrong” (DH, 183).
Stump’s distinction runs along familiar lines. (TO) describes something like natural law theory, in which the good permeates the whole of existence and constrains God as well as humankind, while (TS) describes the approach taken by divine command theorists. She goes on to articulate the respective liabilities of each position, but I will only cite the portion relating to (TS), or DCT:
“The problem with (TS) is that it constitutes a theological subjectivism in which, apparently, anything at all could turn out to be moral. So although (TS) makes a consideration of God essential to an evaluation of human actions, it does so at the cost of depriving that evaluation of its moral character; because it cannot rule out anything as absolutely immoral, (TS) seems to be a theory of religious morality which has dropped morality as we commonly understand it out of the theory. There have been some interesting attempts to resuscitate one or another version of (TS) in recent years; but despite these promising efforts version for (TS) are, I think still more widely known for their faults than for their virtues, and so I will leave such theories to one side” (DH, 184).
For Stump, as for so many other ethicists, there exists a gut-level aversion to (TS), or DCT. It simply strikes many people as intuitively problematic to equate morality with God’s will. It seems more likely that morality, if it exists at all, is grounded in something less prone to divine caprice and whim.
But (TO), or the natural law school of theistic ethics, faces significant problems of its own. If one subscribes to the theological doctrine of creation ex nihilo then the idea that morality co-exists independently of God proves impossible to accept. Certainly God, creator of all, factors more significantly in the establishment of moral obligations than (TO) would have us believe. If moral values exist prior to God’s creative act, then God merely lends support to those moral propositions, and I take this to be an insufficiently God-centered articulation of the origin of morality. No doubt some will find (TO) a more appealing option than (TS), but for now I put it aside in favor of assessing the threat to DCT posed by arbitrariness.
Robert M. Adams, who has worked hard to formulate a version of DCT that avoids the pitfalls Stump summarizes above, recognizes the full force of the objection from arbitrariness. “The gravest objection to the more extreme forms of divine command theory,” he writes, “is that they imply that if God commanded us, for example, to make it our chief end in life to inflict suffering on other human beings, for no other reason than that He commanded it, it would be wrong not to obey” (RRB, 233). Kierkegaard worried about a similar scenario, but with important distinctions, I think. His analysis of Abraham and Issac focused on what he referred to as the “teleological suspension of the ethical”, the experience of receiving a divine command that seemingly upheld the moral obligations of others while demanding that you violate your own. While Adams’ concerns are related to Kierkegaard’s, I take the latter case to be more troublesome. Determining precisely how to distinguish whether a command to violate a standing moral obligation emanates from God or some psychological defect poses significant epistemological problems; perhaps no a priori criteria can be established.
But Adams, unlike Kierkegaard, is less concerned about whether God can single out an individual and demand that she violate a standing obligation than he is worried about how moral obligations are established in the first place. It might follow from DCT that God possesses the authority to command individuals to violate their obligations and that such violations would be morally justified, but Adams’ worry is whether God can fix horrible actions as obligatory for everyone as the normative standard of ethical behavior. The distinction is important, I think, because Kierkegaard’s scenario occurs within the context of an established moral framework and considers the validity of commands to violate existing obligations. In Adams’ case, he considers the moral conditions that apply at the founding of moral obligations. This line of questioning should immediately strike the reader as odd. What moral objections could we raise against the prospect of God creating certain moral obligations before God creates them? Doesn’t the divine command theorist first require the establishment of ethical standards by God before being capable of adjudicating the ethical fitness of anything? Adams’ circular reasoning is symptomatic of his efforts to render DCT palatable to those who find the prospect of arbitrariness unsettling. Rather than adhere strictly to the simple formulation prescribed by DCT – that ethical wrongness is whatever is contrary to God commands – Adams complicates the theory by introducing unnecessary provisions.
Adams writes, “I prefer a less extreme, or modified, divine command theory, which identifies the ethical property of wrongness with the property of being contrary to the commands of a loving God” (RRB, 233). Building love into God’s essential nature provides Adams the sort of provision he feels is necessary to ward off undesirable degrees of arbitrariness and constrains the range of obligations God can deem proper. Adams stops short of claiming a logical objection to the possibility that God might command us to inflict suffering on others; there are no grounds on DCT for thinking such a prohibition obtains. God’s love is, then, a plausible and theologically sound option for introducing a normative principle intrinsic to God that might regulate God’s commands. A loving God, Adams argues, simply would not deem a command to inflict suffering on others a fitting obligation.
Where does this formulation of DCT leave the theist’s response to the problem of evil? It leaves it precisely where it was before: in need of God-justifying reasons why God would deviate from God’s loving character and allow suffering to persist unabated. For though Adams grudgingly admits that God might logically command cruelty for its own sake, he finds the prospect incredible. He writes, “The modified divine command theorist agrees that it is logically possible that God should command cruelty for its own sake; but he holds that it is unthinkable that God should do so” (MDC, 465). For Adams, God is constrained by God’s loving character in such a way as to make it virtually impossible for God to command cruelty and remain God. In other words, God still has to justify Godself for any aspect of creation that appears inconsistent with God’s loving character, and the conspicuous prevalence of suffering demands that theists supply plausible justifications for why a loving God does not intervene.
Adams vitiates the ability of DCT to adequately answer challenges to God’s moral authority by implying that God’s perfect moral goodness necessarily entails a loving character. This provision shifts the burden of the argument away from scrutiny of God’s essential attributes towards considerations of possible defenses and theodicies that might account for God’s apparent failure to always love as a loving God would. The abiding worry over God’s moral perfection results from this mistaken conflation of God’s essential attributes with God’s loving character. So long as God is understood to always act in conformity with God’s loving character the logical problem of suffering will remain troublesome.
But why think that God cannot act contrary to God’s loving nature? Why not instead think that God’s loving nature is coincidental to God’s moral perfection and not an essential attribute? Certainly a theist operating within one of the three Abrahamic religions will wish to maintain that God is a loving god, but why not distinguish between essential attributes, or those to which God cannot help but conform, and coincidental attributes that God may choose to exercise or suppress? It seems plausible that the God of DCT might exist as omnipotent, omniscient and perfectly blameless without possessing a loving character, in which case God’s love is for us a fortunate and welcome coincidence. Does not the fact that God possesses no moral obligations secure the question of God’s moral perfection without need of recourse to a loving character? Adams seems ready to admit as much, writing, “He [God] does not have a duty – at least not in the most important sense in which human beings have a duty. For He is not subject to a moral law not of His own making” (MDC, 472). If Adams stopped here, leaving God without any obligations, the prevalence of suffering on our planet would not pose a logical threat to God’s existence. Yet Adams wants to maintain that, while God does not possesses moral obligations, God possesses other kinds of normative tendencies relating to God’s loving character. He writes, “But there are other virtuous dispositions which God can have as well as men. Love, for instance. It hardly makes sense to say that God does what He does because it is right. But it does not follow that God cannot have any reason for doing what He does” (MDC, 472). We should certainly agree with Adams that God has good reasons for doing what God does. When God creates obligations that conduce to our flourishing, we can attribute this considerate act to God’s love for us. But must we insist that God is required to act lovingly simply to satisfy worries over arbitrariness?
I submit that such worries are exaggerated and result in unnecessary theological and philosophical commitments that exacerbate rather than mollify the objections raised by the problem of evil. The particular kind of arbitrariness at work in God’s will is distinct from the variety found in naturalistic occurrences of arbitrariness, where nothing at all accounts for the existence of one phenomenon over another. The latter variety rightly strikes us as utterly groundless, and the wrong kind of thing upon which to found a moral theory. But God’s will is a perfectly acceptable place to situate what may be unavoidable degrees of arbitrariness in the establishment of any ethical framework. Presumably an omnipotent, omniscient and perfectly blameless being has excellent reasons for commanding as God does, and the extent to which God’s actions are unbound by any necessity does not invalidate the authority of God’s pronouncements. It certainly has no bearing on whether God is morally entitled to God’s actions.
There is a related incentive for maintaining a distinction between God’s moral blamelessness and his coincidental loving character. If God is constrained by God’s love to always act as a loving being would act, then nothing about God’s actions can be rightly deemed praiseworthy. But if God’s love is not an essential attribute of God’s nature – in other words if God can choose when to act in conformity with God’s love and when not to – then we can appreciate the extent to which God voluntarily exercises love for us. I take it to be a necessary condition of God’s praiseworthiness that God be capable of choosing whether to act in accordance with God’s loving character. It would make little sense to refer to God, the author of all moral obligations and a being without moral duties, as “good” in the typical usage of the term. But we can refer to God as volitionally loving, and that provides excellent reason to consider God praiseworthy.
At the beginning of this paper I stated that I would offer responses to what I perceived to be two misapprehensions: 1) that God’s status as a perfectly good being requires that he adhere to a standard of goodness existing independently of God, and that this standard places obligations upon God to act in cases of suffering unless a God-justifying reason exists to morally excuse God’s refusal to act; 2) that theories of moral goodness that index God’s will as the source of obligations cannot adequately account for a non-arbitrary moral order and deprive God of moral praiseworthiness. I have attempted to show that DCT provides an answer to the logical problem of evil by denying that God possesses any moral duty to intervene in instances of suffering. Additionally, I argued that a voluntarist conception of moral obligation supplies an understanding of God’s praiseworthiness by citing God’s free choice in whether to command as a loving God would command. That any of God’s commands conduce to our flourishing is evidence that God exercises some degree of compassion and love toward us, even if we interpret flourishing to mean an eventual eschatological benefit. Far from being a hindrance to DCT, God’s radical freedom to command as God sees fit is a necessary condition of God’s praiseworthiness. Where do these conclusions leave the logical problem of evil?
At best, the atheologian can protest that God is not a loving God, for if God were loving God would intervene as any loving being would in cases of horrendous evil. But this misconstrues the Christian claim about God’s nature. Marilyn Adams writes, “It does the atheologian no good to argue for the falsity of Christianity on the ground that the existence of an omnipotent, omniscient, pleasure-maximizer is incompossible with a world such as ours, because Christians never believed God was a pleasure-maximizer anyway” (HE, 298). God need not be the kind of loving God that maximizes our pleasure in order to be God nor, on DCT, need God provide a moral justification for why God fails to intervene in cases of horrendous evils. Without obligations of either kind – moral or loving – the conspicuous prevalence of suffering cannot serve as a logical objection to God’s existence.
Admittedly, many theists and atheists alike will find this arrangement unsatisfactory. Those who cannot tolerate any degree of arbitrariness in the founding of moral obligations will balk at the idea of a coincidentally loving God without constraints of any kind. But I think my interpretation of DCT represents the purest rendition of the formula, a return to form necessitated by relatively recent concerns over arbitrariness that hamper its ability to counter arguments from evil. The will of God is the right place to locate such contingency, and God’s other attributes negate the worst of the negative implications of arbitrariness.
That said, I recognize the need to continue seeking good reasons for why God allows suffering to persist. While I deny that these reasons must take the form of moral justifications – for God needs no such justification – I agree that God is a rational being who acts for reasons and that we ought to consider what those reasons may be. But DCT takes the pressure off of free will defenses and theodicies to answer the logical problem of evil by bearing the philosophical burden of justification. True, the cold logical justification it provides does little to console or assuage our anguish in the face of the horror that often accompanies the human predicament, but it does negate the strictly formal objections to the compatibility of God and suffering.
DH: Stump, Eleonore. “Dante’s Hell, Aquinas’s Moral Theory, and the Love of God.” Canadian Journal of Philosophy. Volume 16, Number 2. June 1985, pp. 181 -198.
HE: Adams, Marily McCord Adams. “Horrendous Evils and the Goodness of God.” Proceedings of the Aristotelian Society, Supplementary Volumes, Vol. 63, 1989. 297- 323.
MDC: Adams, Robert M. “A Modified Divine Command Theory of Ethical Wrongness.” Religion and Morality. Garden City: Anchor Press, 1975. 318-47. Print.
RRB: Adams, Robert M. “Rationality and Religious Belief.” 1977. Philosophy of Religion: Selected Readings. 4th ed. New York: Oxford UP, 2010. 231-39. Print.
 Stump includes a footnote specifying that she is has in mind Robert M. Adam’s “A Modified Divine Command Theory of Moral Obligation.” More on this essay below.
 According to Adams, William of Ockham takes a similar, bite-the-bullet approach as I do.