Does God Destroy Our Duty of Compassion?

Stephen Maitzen

The great majority of believers in God would answer my title question with a resounding “On the contrary!” Far from destroying our duty of compassion, they’d say, the existence of God gives us a duty to act compassionately toward others. Some would go so far as to say that it also works in reverse: a duty of compassion requires the existence of God—that is to say, we have a duty to act compassionately only if God exists, only if the atheistic view of an uncaring, unguided universe is false.

I claim, however, that we have a duty of compassion only if the God described by traditional theism doesn’t exist. Although it would be easy, I won’t rely on any of the many passages in the Bible and the Qur’an that seem to portray God as commanding cruelty rather than compassion. Never mind how unfavorably the supreme being might be portrayed in the scriptures. I’ll argue that the very existence of a supreme being—a being perfect in power, knowledge, and goodness—would undermine the main basis of our duty to act compassionately.

This issue matters for many reasons, among them the fact that atheism has an unfairly bad reputation. A recent survey showed that Americans distrust atheists more than any other group, much to the surprise of those who conducted the survey (see Penny Edgell, et al., “Atheists as ‘Other’: Moral Boundaries and Cultural Membership in American Society,” American Sociological Review 71 [2006]). This distrust apparently comes from the widespread belief that atheism is bad for morality and that atheists are therefore morally unreliable. On the contrary, I argue that theism is what’s bad for morality, and it’s theists who can’t be trusted if they truly believe what theism implies about our moral duties.

In a nutshell, my message is this: the better God is, the worse we are allowed to be, and the easier morality is on us. If God is as good as theism claims he is,* then it turns out that we don’t have the duty of compassion that (in our morally best moments) we thought we had.

The key premise in my argument is that exploitation is always an imperfect, or defective, kind of behavior. It follows, then, that no perfect God could possibly exploit people, not even people he creates and who therefore owe him their very existence. What do I mean by “exploit”? Let me give an example: God exploits people if he allows them to undergo intense, undeserved, and involuntary suffering that isn’t for their own net benefit.

Suppose that God, although having the knowledge and power to prevent it, allows a child to experience terrible suffering—not because the child will ultimately benefit from it but for some other reason or perhaps for no reason at all. The suffering is intense, the child doesn’t deserve to undergo it (as punishment, say), and the child hasn’t volunteered for it (as you might volunteer for the pain of donating bone marrow). In allowing the suffering, God exploits the child.

The treatment is what’s important here, not the exact term we use for it, so you can reject my assumption that we actually use the word exploit in this way. My claim is that no supreme being could treat the child in the way I’ve described, whichever label we choose for that treatment. There’s something less than perfect about letting a child suffer terribly for the primary benefit of someone else—whether for the benefit of a bystander who gets a hero’s chance to intervene or for the benefit of a child-abuser who gets to exercise unchecked free will. If you doubt the previous sentence, consider whether you would dream of letting a child you love suffer abuse in order to secure either of those benefits.

But is it always wrong, you might ask, to exploit people? Don’t we sometimes justifiably use innocent people for the benefit of others? If the child in my example contracts an untreatable, fatal, and highly contagious disease, might we not justifiably quarantine him or her if that is the only (or best) way to prevent the spread of the disease? Might we not justifiably isolate the child in a way that benefits others at the expense of the child? Don’t we justifiably perform triage, letting some patients suffer so we can attend to more urgent cases? Yes. But these practices reflect our imperfection: it’s only limitations in our knowledge and power (in this case, medical) that make us resort to triage or quarantine. We regret having to do it; we wish we had the resources to make these practices unnecessary.

A perfect God, however, isn’t subject to our limitations in knowledge or power, or indeed to any real limitations in knowledge or power. So no perfect God has an excuse for exploitation, even if we sometimes do. Moreover, if God were to face an actual moral dilemma, a case in which he does something morally wrong no matter what he does, then he wouldn’t count as morally perfect on the plausible assumption that “morally perfect” implies “does nothing wrong.” If I’m right, then God can’t possibly allow a child’s intense, undeserved, and involuntary suffering unless the suffering is necessary (or if not strictly necessary then optimal) for the child’s overall benefit.

Yet many children endure intense, undeserved, and involuntary suffering every day, all over the world. What should we do when we encounter them? Obviously, we ought to act compassionately toward them. In many cases, at least, we have a duty to prevent or relieve their suffering. Their suffering is very bad for them, which is the most important reason we ought to act with compassion in the first place. But wait. If God exists, then that suffering must be needed—somehow, even if we can’t see how—for the overall benefit of those very sufferers. In that case, what happens to the moral duty we thought we had to prevent or relieve their suffering? It disappears.

If a perfect God exists, then any suffering that occurs is suffering that God allows to occur, because any perfect being has the power to prevent any occurrence, including any case of suffering. If perfection also rules out exploitation, as I’ve argued it does, then God allows the suffering of children only if those children ultimately benefit from the suffering. The word from matters there. It’s not enough if God merely compensates children for the suffering he lets them endure; the suffering itself must be necessary (or at least optimal) for their greater good.

Why? Because compensation doesn’t justify an action. I can compensate you after harming you, and indeed the courts may require me to compensate you, but no amount of compensation will justify my harming you. The only justification for my harming you would be my need to harm you in order to stop you from harming me or some innocent third party. Likewise, then, whenever God allows a child’s undeserved involuntary suffering: the suffering must be needed, or at least optimal, for the child’s overall benefit. Otherwise God can’t justifiably permit it.

But if the suffering is needed, or optimal, for the child’s overall benefit, then it’s like the pain from a needle when the needle is the only (or best) way to deliver a vaccine. If so, then we never have a duty to prevent the suffering of children, just as we don’t have a moral obligation to prevent painful vaccinations when they’re beneficial. Even those who oppose childhood vaccinations wholesale do so because they think vaccines do more harm than good. We don’t think it’s compassionate to try to prevent vaccinations just because needles hurt, even if compassion tells us to comfort the
child during the process as best we can. By the same token, we never have an obligation to prevent terrible suffering by children if a perfect God makes sure the suffering is for their own good. We have, at most, a duty to comfort them while they suffer.

If we never have a moral obligation to prevent suffering by children, then which moral obligations do we have? None, as far as I can see. How can we be morally obligated to refrain from theft, fraud, bigotry, or slander if we lack the even more basic obligation to prevent the suffering of children? If we lack a moral obligation to prevent even the worst suffering by children, then morality falls apart or at best it becomes frivolous because it no longer concerns the most serious kinds of harm.

Theism causes a related problem as well. Suppose I’m wrong, and we can reconcile God’s existence with a duty on our part to prevent at least some suffering. Theism still encourages a bizarre “reverse triage”: the worse an innocent person’s suffering, the more reason theism gives us for thinking that the suffering must be needed for the sufferer’s own good, and hence the less reason we have to prevent the suffering. We ought to prevent mild suffering first, extreme suffering later. Far from shoring up our moral outlook, adding God turns it upside down.

Some might try to answer this argument by rejecting its key premise, by countering that a perfect God can allow a child to suffer for the primary benefit of others. Oxford philosopher Richard Swinburne, for instance, says that God has moral permission to exploit any human being because all human beings owe their existence to God, God is on balance their benefactor, and furthermore “being of use is a good for the victim” who gets used (Richard Swinburne, “Theodicy, Our Well-Being, and God’s Rights,” International Journal for Philosophy of Religion 38 [1995]). Does Swinburne’s response work?

Imagine that I clone a child into existence from a single one of my skin cells, and imagine that I treat the child splendidly for all but the final minute of his or her life. But during that final minute, I allow someone to abuse the child to death in order to show onlookers just how revolting child abuse is and thereby deter them from ever abusing a child. Think of it as aversion therapy. The child owes his or her existence to me (via my use of technology), and I am on balance his or her benefactor, treating the child well for all but the final minute of life. Moreover, his or her horrific death isn’t purely gratuitous; it serves as an object lesson for the benefit of others, not only deterring some potential child-abusers but also protecting children they might otherwise have abused. Nevertheless, in this story I behave imperfectly, to say the least! Even granting Swinburne’s premises, his conclusion doesn’t follow. His defense of exploitation on the part of a perfect God therefore fails.

Because God’s existence would destroy the main reason for compassion, theism is in that way bad for morality—at least any morality we’d recognize. Must we turn the tables, then, and conclude that theists can’t be trusted? Fortunately, no. It’s obvious that plenty of theists can be counted on to act compassionately, as can plenty of atheists. With some tragic exceptions—such as fundamentalist parents who let their children die painfully rather than seek medical treatment that would cure them—when theists confront suffering they accept the duty (or at least feel the urge) to respond with compassion. So, of course, do atheists.

People instinctively try to relieve terrible suffering they directly encounter, and they try to prevent it if they see it coming. To judge by their behavior, they typically believe that terrible suffering isn’t for the sufferer’s overall benefit. They evidently don’t stand aside thinking, “This suffering must be needed for the victim’s greater good” or “Some supernatural agent will prevent this child’s suffering if I don’t.” As I’ve argued, however, theists can think that way only by failing to see what their theism implies.


* I use masculine pronouns for God but only in keeping with the monotheistic tradition; I don’t mean to imply that God has a gender, nor does my argument depend on Gods having a gender.

Stephen Maitzen

Stephen Maitzen is the W. G. Clark Professor of Philosophy at Acadia University.

The great majority of believers in God would answer my title question with a resounding “On the contrary!” Far from destroying our duty of compassion, they’d say, the existence of God gives us a duty to act compassionately toward others. Some would go so far as to say that it also works in reverse: a …

This article is available to subscribers only.
Subscribe now or log in to read this article.