Direct Killing as Intentional Killing

In situations where a pregnancy threatens the lives of mother and child, could it be permissible to remove a non-viable baby from the mother’s womb, knowing that the removal will bring about the baby’s death? Is this always an instance of “direct abortion”? Or is it similar, in its moral analysis, to removing a developing embryo in procedures intended to save a mother’s life in cases of ectopic pregnancy?

To specify some instance of killing as “direct” means that the killing is willed as an end or a means. A synonym for direct is “intentional.” A hotly contested question of moral philosophy is the relationship between what one intends and what one causes as a result of what one intends. Some argue that if one deliberately chooses some behavior that immediately and foreseeably causes some effect, one necessarily intends that effect. So any act that brings about death in an immediate way involves intentional killing.

Others argue, reasonably in my judgment, that one intends precisely and only what one resolves (or sets oneself) to bring about by some piece of behavior. This includes two things: (1) some end for the sake of which one acts—that which one seeks for its own sake; and (2) some means (call it a close-in end) that one resolves to bring about as a way of realizing the end one seeks. One intends, as Aquinas clearly states, both one’s end and one’s means (as a close-in end). Together they form one’s intelligible proposal for action. So in writing this essay for the sake of clarifying a disputed question, I intend both to clarify a disputed question (end) and to write this essay (means).

Now whatever else follows from this chosen end and means (e.g., sore eyes, headache, etc.), but what forms no part of my intelligible proposal for action—what I do not resolve to bring about, and indeed may very much wish to avoid but find unavoidable—is properly speaking a side effect of my action, however immediately or foreseeably or “directly” it may follow from my act in the physical order. So the man who always sleeps poorly when he travels, who knows he will sleep poorly, whose sleeping poorly follows “directly” as a result of his traveling, and who nevertheless intends to travel, still does not intend to sleep poorly when he travels. It is a foreseeable, perhaps an inevitable, side effect of his choice to travel.

In many instances, we have little problem agreeing that we don’t intend everything we knowingly and immediately cause. A woman goes on business and leaves her three children at home with her husband, which causes her children to be lonely and her husband to be overwhelmed. But in traveling she doesn’t intend that her children feel lonely or her husband anxious. Every time we start our car we pollute the environment, but it would be wrong to say that we intend the pollution as a “means” to the “end” of arriving someplace by car, even though we foresee that it will happen and that it happens as an immediate result of our starting and driving our car. We may take cod-liver oil in the morning, which causes us to have fishy burps for the rest of the day, but no one would dare say that the burps were part of our proposal for action, or that we intend the distasteful effect as either an end or a means.

In each case, we accept the harms as foreseeable side effects of our proposals: To travel on business (means) to fulfill professional duties (end); to drive our car (means) to get to work (end); and to take cod-liver oil (means) to stay healthy (end). We intend good ends and good means and tolerate the harmful side effects.

These examples are straightforward. But when the effect immediately and foreseeably caused is the death of an innocent person, the insight that the death need not be intended, even if foreseen, can seem less clear. Yet the principle I’ve highlighted applies consistently across the board.

For example: An F/A-18 fighter pilot shoots down a loaded passenger airplane. The physical description of the act includes causing the deaths of 235 passengers. Does he intend their deaths? He may or may not; to determine that we’d need to know more than just the physical description. If he’s a terrorist who shoots down the plane to teach those American dogs a lesson, then his end is hurting the United States, and his means is killing innocent passengers and destroying costly property; his end is achieved precisely through his means. The proposal that moves him is killing innocents for the sake of hurting the United States; and he intends both.

But if, on the command of the joint chiefs, he shoots down the plane because there is compelling reason to believe that it’s being used by terrorists as a guided missile, his intended end is to protect Americans on the ground, and his means is to remove the plane from the sky by blowing it up. He knows the passengers will die as an immediate and physically “direct” result of his act; the deaths are—in the physical order—causally connected to his choice. But the deaths per se contribute nothing to protecting Americans, and this is what he and the joint chiefs are interested in. So the deaths are not part of the intelligible proposal that moves them to act, not as ends choiceworthy for themselves, nor as means choiceworthy for anything else. If they could get the plane out of the sky without causing the deaths, they would. But they judge they cannot. So the pilot shoots down the plane for the sake of protecting lives, and they (the pilot and joint chiefs) accept the deaths of the passengers as a tragic side effect—foreseen, physically caused but morally unintended; the deaths are (in Aquinas’s language) praeter intentionem.

I think most reasonable people would agree that the F-18 pilot and joint chiefs are not, morally speaking, mass killers, which they certainly would be if they ordered and carried out the intentional killing of hundreds of innocent people, either as an end sought for its own sake, or as a means to some further end, no matter how apparently beneficial.

Now two behaviors, both of which bring about the death of an unborn baby, may embody different intents. If a pregnant woman is so ill that the strain of carrying a child to viability would kill her, it is sometimes medically possible to remove the child from her womb. The removal would involve terminating the pregnancy before viability. But it is reasonable to judge that the removal need not involve intentional killing. Preserving the mother’s life would be the end, sought for its own sake; and removing the child from her uterus would be the means—removal, not death.

The doctor knows the child will die as a result of the removal; the death is—in the physical order—causally connected to the doctor’s choice to remove the child. But the death per se contributes nothing to saving the mother, and this is what he is interested in. So death forms no part of his intelligible proposal, not an end choiceworthy for itself, nor a means choiceworthy for anything else. If the doctor could save the mother and save the baby, he would (and should). But (let’s presume for the sake of argument) he judges he cannot.

The judgment defended here presupposes that a doctor who carries out a proposal to save a mother’s life by virtue of a removal procedure, who foresees that the child will die, but has no interest in the baby’s death per se, only his removal; who has no life-affirming alternative for the child, and the only life-affirming solution for the mother is to remove her child; and who removes the child and brings about death, is a different kind of person from a doctor whose proposal is to assist some pregnant woman precisely by getting her baby dead.

They are different kinds of persons because they perform different acts as they carry out different proposals: the one, a proposal to remove a non-viable child to save the mother; the other, to kill that child for the mother’s benefit. I believe that collapsing the order of physical causality (namely, the fact that both bring about death) with the order of intention distorts a central—perhaps the central—truth of moral life: namely, that morality in the first place is about forming oneself into a particular kind of person precisely by formulating and executing intelligible proposals. The moral life is not primarily about the harms or benefits I cause, but about the actions I perform.

So the first moral concern in the example above is not identifying what outcome is produced by the physical behavior, but rather what the doctor does. Does he remove the baby in order to kill him? Was getting the baby dead part of the doctor’s intelligible proposal? The clear answer seems to be no. Again, does he kill the baby in order to save the mother? Is it through the baby’s death that his proposal to save the mother is realized? Again, the answer seems clearly to be no. The baby’s death contributes nothing per se to the mother’s recovery; and so it accomplishes nothing that the doctor is interested in. Like not sleeping when one travels, polluting the environment, burping from taking cod-liver oil, and bringing about the deaths of the 235 passengers, the harm is a foreseen and tolerated, but unintentional side-effect.

A caricature of the account I am presenting is not uncommon in the literature. It summarizes my argument as follows: “All I need to do is mentally ‘direct my attention’ away from some effect of my action, and voilà, the action becomes ‘unintentional.’ And then I describe my act in terms of my wishful mental directing.”

But the view I am defending is not about mental self-persuasion. Intending is not mental directing, but rather willing in response to some intelligible proposal. One only acts on what one is interested in; and one is interested in some end for its own sake and some means one believes is suitable for bringing one’s end about. And one intends only what is part—really and entirely—of this ends-means nexus. What falls outside it is praeter intentionem.

Does saying that something is unintentional imply that it’s morally legitimate? Certainly not. If there is not a iusta causa (or, in more traditional language, a “proportionate reason”) to tolerate (though never to intend) the harm, then the act would be wrong on that account. If, for example, an unborn baby is viable, then causing its death, even unintentionally,would be unjust—unfair—to the child; and the act would be immoral.

In almost all actual procedures that cause an unborn child’s death (e.g., the fifty million abortions the Guttmacher Institute recently estimated have occurred in the United States since Roe), death is intended, i.e., chosen, as a means to some end: usually, to avoid the burdens of parenthood. In such cases, the child’s death is part of the agent’s proposal for action: It is chosen as contributing to the goal(s) being sought.

Moreover, even in those few cases where the death of an unborn child really is only a side effect, justice requires that it be proportionate to the end being pursued, as in Aquinas’s account of self-defense and other applications of the principle of double effect. So it is reasonable to suppose that the only justifiable ground for performing a procedure that unintentionally brings about a baby’s death is to save the mother’s life, where otherwise both mother and child would die.

This is not proportionalism, which argues that evil may be done if doing so promises some “greater” good. This is an application of double effect reasoning, which argues that bad consequences sometimes may be unintentionally caused, if there is very serious reason for accepting the harm.

My analysis is consistent with a resolute and public witness to the moral value of every human life and to the intrinsic immorality of direct abortion—and euthanasia, embryo-destructive research, suicide, and other offenses against the good of human life. At the same time, it integrates this witness, more consistently than competing views, with the principle that one may accept proportionate harms as unintended side effects of morally good acts, while articulating the meaning of “side effects” in conformity with right reason.