1. As was explained (in 6‑B), choices are related to human goods in two ways. First, in choosing one determines oneself in respect to human goods (see S.t., 1–2, q. 1, a. 3; q. 26, a. 2; q. 28, a. 6; q. 86, a. 1, ad 2). One establishes one’s moral identity as open to the good in its fullness or as open only to limited aspects of goodness, not to integral human fulfillment. Second, there is a causal relationship: In choosing, one sets oneself to bring about certain states of affairs (see S.t., 1–2, q. 17, a. 1). In these states of affairs, human goods are to some extent realized or manifested. In more figurative language, by choosing one makes one’s soul and also sets the course of one’s life. In deciding to play golf, for example, one integrates the good of play rightly or wrongly into one’s moral self, while at the same time setting oneself to perform the playing of the game.
2. These two relationships to the goods are distinct. They can coincide, but they need not. One can choose to look at a beautiful sunset and then look at it. But one can also see the sunset without choosing to see it (the second relationship to the good exists without the first) or choose to see it but be prevented (the first relationship without the second).
3. There are different kinds of actions, distinguished by the simplicity or complexity of various elements of the action as a whole.
4. In the simplest kind of action, the only relevant good actually affected is that actualized in the state of affairs brought about in carrying out the choice. If, for instance, one chooses to play and plays golf simply for the sake of play, the only human good directly affected is the good of play. One determines oneself in respect to it in choosing, and one realizes it in the playing of the game.
5. Slightly more complicated in outward structure, yet morally the same, is a case where the deed one chooses to do is the cause of a desired effect. One chooses, for example, to take medicine for the sake of health; the only relevant good is health, not the taking of medicine. Here the complexity of the cause-effect sequence does not affect the act’s moral simplicity. From the point of view of medical technique, taking medicine is a means to health. From a moral point of view, the act is as simple as playing golf for the sake of play. There is a real difference between the cause and the effect, but it is morally unimportant, since the two are humanly inseparable and both are included in one actuation of the will—the choice to do a health-giving action.5 The choice determines the person only in respect to the single good: health.6
6. Some actions, however, are morally more complicated. Although carrying out a proposal will have an effect upon a human good, one makes this choice for the sake of some good other than that involved in the carrying out of the choice and the unfolding of its humanly inevitable effects. For instance, one might choose to play golf, even though one detests the game, for the sake of making a business deal.7
7. In such cases, the action or omission one chooses is a means from the existential point of view; thus the means-end distinction has ethical significance here. One has some responsibility not only in regard to the good one pursues, but also in regard to the goods one willingly affects in using the means one chooses. Still, it makes a difference whether the effect on a human good of the use of the means is or is not part of one’s proposal. Effects of carrying out a proposal, if foreseen but not included in the proposal, are side effects. (The voluntariness involved in accepting side effects will be considered in question F.) But the impact on a human good which is part of one’s chosen proposal is not a side effect. Even though one’s action affects this good only as a means one is using to attain one’s end, still one determines oneself—not simply in respect to one’s end, but in respect to any good affected as means.
8. For example, the United States, in order to protect freedom and peace, has adopted a policy of nuclear deterrence which includes a real and continuing choice to destroy an enemy’s noncombatant population if deterrence should fail.8 This policy involves a fourfold relationship to human goods. If deterrence works, the goods which are its end, peace and freedom, are realized. But if execution of the policy becomes necessary, this also will affect a human good: life. Furthermore, the moral self of everyone who cooperates in and supports this policy is also doubly determined. Love of peace and freedom is part of one’s identity; but so, too, is irreverence toward human life, implicit in the readiness to kill for peace and freedom’s sake. (In misunderstanding human goods and misconstruing the nature of morality, proportionalists overlook the “soul-making” aspect of one’s choice in cases of this sort, as was explained in 6‑B and 6‑G.)9
In addition to the object of the human act done by choice, classical analyses mention the “end of the agent.” The present analysis clarifies this conception, both by distinguishing between the two ways in which one’s choice is related to a good, and by making explicit the relevance of both the good which pertains to the end and the good affected by an act chosen as a means.
Obviously, the end of the agent is very important, especially insofar as it is that to which one commits oneself, for in this aspect the end constitutes the person acting. But in choosing and acting one also constitutes oneself by one’s relationship to a good which is relevant only to a means adopted in pursuit of an ulterior end. One who chooses to kill as a means wills to be a killer in wanting the particular death, not of course for its own sake, but for the sake of the ulterior good for which killing is chosen. The ulterior good is intended both as that to which one commits oneself and as that which one hopes to realize by killing.
One who chooses to kill one person to save the life of another constitutes himself or herself both by the intentional killing and by the intended saving of life. The inconsistency involved can exist in one’s moral self only if one makes some sort of distinction between life worth saving and life which is expendable. In making such a distinction, one is qualifying one’s basic love of life, limiting it to love of life of a certain quality, or something of the sort.
It is worth recalling that the good to which one determines oneself and in which one thereby shares is wholly distinct from the concrete good brought about by one’s action only in the case of goods pertaining to domains other than the existential. In the existential domain, to determine oneself to a true good is to begin to effect what one wills; to seek some satisfaction in a mere appearance of the good is to constitute oneself accordingly. For example, one who gives a gift out of friendship thereby begins to realize the good of friendship; one who does favors only for selfish purposes brings into existence a manipulative relationship.
5. I have made this point in previous works. Joseph A. Selling, “The Problem of Reinterpreting the Principle of Double Effect,” Louvain Studies, 8 (1980), 53, n. 17, criticizes it (without citing any particular work) by saying: “Grisez will not allow for the existence of a cause-effect relationship in anything other than his indivisible—read physical—process,” and providing examples of causes and effects which are not in an indivisible process. But Selling has missed my point, and the claim that I do not allow such relationships is false. My point rather is that some cause-effect relationships need not be means-end relationships in a moral sense. Selling’s entire argument in this article presupposes the acceptance of proportionalism, for he maintains that only the agent’s intention of an end and proportionality determine morality (57). Since proportionalism has been refuted (6‑F) independently of the analysis of action, analyses of action which assume proportionalism may be rejected precisely on that ground. Selling also refers to his general theory as “an ethics of responsibility” (62), as if proportionalism has an exclusive claim on that title.
6. W. Van der Marck, O.P., Love and Fertility: Contemporary Questions about Birth Regulation, trans. C. A. Jarrott (London: Sheed and Ward, 1965), 42–60, says that the means-end distinction simply is between a piece of outward behavior and the total human act determined by the good one intends. In saying this, he makes two mistakes, both of which can be seen by analyzing examples. First, he fails to notice that the mere outward behavior cannot in any case be regarded as a means or act (in any sense), for just as such it can be picked out only by a naturalistic description which is altogether neutral with respect to categories such as “means.” This point is clear from the work of empiricists who attempt to reduce action to categories acceptable in their metaphysics; they are compelled to introduce beliefs and wants to distinguish even the most basic acts. See, for instance, Arthur C. Danto, Analytical Philosophy of Action (Cambridge: Cambridge University Press, 1973), 18 and 72–78. Second, Van der Marck assumes that all acts are as simple as those in which the means-end distinction is merely technical, and thus overlooks the more complex case in which one adopts a certain proposal by choice, realizing it to bear upon certain human goods in certain ways and willing that bearing so far as it is included within the proposal, but directs the entire action to some other purpose—for example, one plays golf for the sake of making a sale. Here, regardless of what one thinks about its morality, the chosen means is such from an existential point of view, and an analysis which leaves no room for this is obviously inadequate.
7. This distinction explains why descriptions of acts from a moral point of view can vary within limits in characterizing by consequences what one is doing, but cannot plausibly vary beyond certain limits. Once a description includes a bearing upon some human good, a redescription specified by further consequences which conceals that bearing will seem implausible, especially if doing the act as involving that bearing is morally arguable, even if acceptable. See Eric D’Arcy, Human Acts: An Essay in their Moral Evaluation (Oxford: Clarendon Press, 1963), 10–39.
8. See Robert S. McNamara, The Essence of Security: Reflections in Office (New York: Harper and Row, 1968), 52–53; The Organization of the Joint Chiefs of Staff, United States Military Posture for FY [fiscal year] 1983 (Washington, D.C.: U.S. Government Printing Office, 1982), 19. For quotations from and references to other statements of the threat and a history of the evolution of American strategic doctrine, see Donald M. Snow, Nuclear Strategy in a Dynamic World: American Policy in the 1980s (University, Ala.: University of Alabama Press, 1981), 48–85, with notes, 253–56.
9. Richard A. McCormick, S.J., accepted the view of Bruno Schüller, S.J., that with respect to nonmoral goods (such as human life) one need no more “approve” the evil (destruction of life) when one chooses to kill as a means to some ulterior end than when one accepts death freely as a side effect. Thus for nonmoral goods both authors denied the significance of the distinction between direct and indirect killing, sterilizing, and so on. (“Approve” is equivocal to emotion and will; Schüller gained some plausibility from this since one’s emotional attitude can easily be the same in the two cases although the volitional self-disposition is very different.) See McCormick, in Doing Evil to Achieve Good: Moral Choice in Conflict Situations, ed. Richard A. McCormick, S.J., and Paul Ramsey (Chicago: Loyola University Press, 1978), 254–62. McCormick admitted that this position makes it difficult to understand why the direct-indirect distinction is vital in the case of causing another’s sin, where both authors wish to retain it (258–59). The explanation of the difficulty is that they and others who have embraced proportionalism overlook the existential character of moral acts in general and so do not see the self-determination with respect to basic goods in every free choice. With respect to substantive goods, they think of upright choices as if they were nothing but efficient means of bringing about their realization in concrete instances. But they do see that one cannot will that another do evil even in only a particular instance without willing—and so being—morally evil oneself.