1. As has been explained, a duty must be left unfulfilled if its fulfillment would require doing something forbidden by a norm derived from the eighth mode of responsibility. The same is true if any other norm seems to require an act excluded by an absolute norm. By definition, nonabsolute norms yield to absolute ones.
2. Today, many try to override norms derived from the eighth mode. They argue, for instance, that the choice to kill an unwanted child is sometimes justified as a lesser evil. This is an example of the proportionalism criticized in chapter six; what was shown there need not be repeated here.
3. In many cases, apparent conflicts are removed when the morally right course, previously ignored because it is unappealing, is accepted as a practical possibility. For example, persons who have divorced and remarried need not really choose between committing adultery and renouncing their responsibilities to their second family. They can choose instead to live together in celibacy, in accord with the moral truth that they have no marital rights but do have familial responsibilities.3
At other times a certain creativity would resolve apparent conflicts. Thus, a person seemingly compelled either to lie or violate a serious obligation of secrecy might simply refuse to answer questions, while pointing out that the refusal was a matter of policy (for example, in the case of the confessional) and could not be taken as indicating anything for or against the proposition about which questions were being asked.
4. In other cases, an apparent conflict of responsibilities involving the eighth mode can be dissolved by clarifying how relevant norms apply to the choices under consideration. It is virtually impossible to do anything without bringing about some humanly bad effects, and these can often be foreseen. But, as was explained (in 9‑F), there is a difference between what one brings about as a chosen means to one’s end and what one brings about as a freely accepted side effect of carrying out one’s choice. The eighth mode of responsibility precludes only a choice to destroy, damage, or impede a human good. It does not rule out accepting foreseen side effects which it would be wrong to choose. Hence, such side effects may be accepted, unless some other mode of responsibility requires the contrary.
Even the best acts bring about some bad effects, at least indirectly. For example, preaching the gospel brings about the effects of belief and disbelief, and so leads to division, which ultimately leads to conflict and persecution. Going for an automobile ride may add to the traffic death toll and certainly adds to pollution (which causes various diseases) while using up an energy resource, and so forth. Even the most innocent acts use time and energy which thereby are not available for service to various human goods. Thus, in an indirect way, an hour’s meditation contributes to the misery of those whom an hour’s labor could have helped.
This situation is not a consequence of our fallen human condition, nor is it essentially a result of our finitude.4 Even God could not create without bringing about foreseen bad consequences; he knew that freedom could and would be abused. His proposal, however, was not that creatures should sin, but that they should be able to love freely. He accepted sin as an unwanted side effect of the creaturely freedom by which love was refused (see S.c.g., 3, 71).
According to the seventh and eighth modes of responsibility, it is wrong to choose to destroy, damage, or impede any instance of an intelligible human good. One is tempted to make such choices either out of a nonrational hostility or out of a nonrational preference for some good which is to be realized through the act chosen. To bring about humanly bad effects out of such nonrational motives always is wrong.
But there are other modes of responsibility. The fifth, for instance, is concerned with fairness. If a gardener uses a poison to protect a crop with a possible side effect of poisoning a neighbor’s children, the acceptance of the possible ill effect probably is unfair. The gardener would object if someone took such a risk with his or her children. Likewise, an upright person will not accept foreseen bad effects to a real good in acting for a merely apparent one, in violation of the sixth mode of responsibility, although such bad effects normally are no part of one’s proposal.
Therefore, the problem of when one may act in a way which one foresees will have humanly bad effects can be solved only by considering all of the intelligible features of that about which one is deliberating, referring to all the modes of responsibility, and articulating a norm for the act under consideration. The act will be right if and only if in choosing it one violates none of the modes of responsibility.
Thus, if one’s proposal is to destroy, damage, or impede any of the basic human goods, the act will be excluded by the seventh or eighth mode of responsibility. If one is inclined to accept the bad effects out of partiality and would not accept them otherwise, one violates the fifth mode. If the bad effects are consequent on an omission to which one is inclined out of laziness, one violates the first mode. And so forth. If the foreseen bad effects are accepted without violating any mode of responsibility, one simply accepts them as incidental to the good one is doing. One does not need any reason for so accepting them, except the reason provided by the intelligible goodness of that for which one acts.
5. There might appear to be a conflict of responsibilities involving the eighth mode when, for example, one must choose between removing a nonviable ectopic embryo and allowing the pregnancy to continue, with a definite threat to the mother’s life and little hope for the embryonic child’s survival. In choosing to remove the ectopic embryo, however, one need not be choosing to kill it, in violation of the eighth mode of responsibility. The embryo’s death is not a means and is in no way helpful to solving the problem. Rather, one can make this choice—to remove the ectopic embryo—simply to remove its threat to the mother’s life, accepting the embryonic child’s certain death only as an inevitable side effect.
6. Apparent conflicts of this sort arise because acts conventionally classified in one way (for example, as killing an attacker) must be morally classified in another way (for example, as stopping a violent attack) (see S.t., 1–2, q. 1, a. 3; q. 18, a. 5). The behavior involved in bringing about another’s death in self-defense is outwardly similar to that involved in morally impermissible acts in which one chooses to kill and carries out the choice. However, one who brings about another’s death in self-defense need not be choosing to kill in violation of the eighth mode of responsibility (cf. S.t., 2–2, q. 64, a. 7). If there is no choice to kill, the morality of what is done depends upon fairness and other modes of responsibility. One’s act will be right if one violates no mode of responsibility.
The Catholic teaching condemning all direct killing of the innocent certainly excludes any choice to kill them. There also is a received teaching condemning direct abortion, and “direct abortion” is taken to mean the expulsion of a nonviable fetus otherwise than as a side effect of some other legitimate act. Usually direct abortion has been considered a species of direct killing of the innocent. While I do not wish to deal with specific normative problems here, I admit that my analysis points to the permissibility of certain operations which classical moralists would have excluded. I do not think this position is in significant conflict with received Catholic teaching.5 However, if my theory and the Church’s teaching should in a particular case lead to inconsistent conclusions, I would follow and urge others to follow the Church’s teaching rather than my theory. If the Church’s teaching is open to legitimate refinement in the details of its application, the refinement must be completed by those who exercise teaching authority in the Church.
7. Since problems of this sort are dissolved by analyzing the human act more closely and applying norms to it, it is misleading to formulate the solution in language suggesting that justifiable exceptions are being made to absolute moral norms. For instance, the moral rightness of removing an ectopic embryo and performing certain other operations which result in the death of the unborn should not be summarized by saying: It is wrong to kill the unborn except when the mother’s life is at stake. To choose to kill is always wrong, and choosing anything else while freely accepting another person’s death usually violates one of the modes, especially fairness. But sometimes it is morally right to choose to do good and accept bad side effects.
8. Classical moral theology dealt with this sort of problem by the so-called principle of double effect.6 Double effect is not a normative principle but a somewhat cumbersome attempt at clarifying what one is morally responsible for in freely accepting side effects which it would be wrong to choose.
9. In most formulations of double effect, one stated condition is that there be a proportionate reason for accepting the bad side effect. Some proportionalists maintain that this shows that classical moral theology was committed to proportionalism. If so, however, the classical moralists would not have required a double effect analysis. They did so because in fact they held that certain things are always wrong, regardless of ulterior good consequences.
When the classical moralists required a “proportionate reason” for freely accepting bad side effects, they implied that the good sought and the evil accepted could be rendered commensurate. They did not say how this might be done. But the commensuration they required can be explained without admitting the commensuration of premoral goods and bads the proportionalist requires. For one can say that the reason for accepting bad side effects is “proportionate” if their acceptance does not violate any of the modes of responsibility. For example, by this criterion one who risked the death of healthy children in medical experiments would lack a proportionate reason, for to take such a risk would be unfair.
In a case of a pregnant woman with a cancerous uterus, the analysis proposed here leaves open the question of the morality of surgical treatment which would lead to the child’s death. If both mother and child were likely to die without the treatment, then the choice to treat with its foreseen bad consequences certainly can be accepted without unfairness. But if the child has a chance for survival if the treatment is not initiated, it is not clear that the choice to carry on treatment to save the mother does not involve partiality.
When a soldier on a battlefield shoots an enemy, the proposal executed by this performance need not include killing. (Of course, very often the shooting is precisely to kill; dead bodies are viewed as a good means to ending the war.) A soldier who is involved in a defensive action against an unjust attack could be trying only to inactivate the attackers and limit the injustice. The shooting at the enemy could be effective for its real objective if the shot missed, but caused the enemy to surrender or to flee and cease engaging in the unjust action; it also could be effective if the enemy was wounded only minimally but sufficiently to be incapable of further participation in the unjust action. This example begins to indicate both the possibility and the limits of justifiable war on the theory presented here.7
3. See John Paul II, Familiaris Consortio, 74 AAS (1982) 184–86; Eng. ed. (Vatican City: Vatican Polyglot Press, 1981), 160 (sec. 84).
4. As is maintained by some Protestants, for example: Helmut Thielicke, Theological Ethics, vol. 1, Foundations, ed. William H. Lazareth (Philadelphia: Fortress Press, 1966), 279–97.
5. I do not accept direct killing in any instance, but do question whether everything which has been considered direct abortion is direct killing. In the latter part of the nineteenth century, an extensive and rather unsatisfactory controversy developed among Catholic moralists, all of whom rejected direct killing, concerning certain problems in this area. The Holy See ended the controversy by determining what might and might not be taught. An extensive summary of this controversy: John R. Connery, S.J., Abortion: The Development of the Roman Catholic Perspective (Chicago: Loyola University Press, 1977), 214–303. Have the fine determinations of what counts as direct killing been proposed universally by the magisterium as judgments to be held definitively? I do not think so, and if they have not, they are open to refinement (as will be explained in 35‑G). But I do consider these judgments normative unless they are reconsidered by the magisterium itself.
6. For a historical summary, see Joseph T. Mangan, S.J., “An Historical Analysis of the Principle of Double Effect,” Theological Studies, 10 (1949), 40–61; J. Ghoos, “L’Acte à Double Effet: Etude de Théologie Positive,” Ephemerides Theologiae Lovaniensis, 27 (1951), 30–52. Ghoos is unjustifiably negative about the work of others; he does not show Mangan’s study to be in error, although he does add material useful for additional insight into the history.
7. See Germain Grisez, “Toward a Consistent Natural-Law Ethics of Killing,” American Journal of Jurisprudence, 15 (1970), 91–94, for further discussion of the limits of killing in war.