ARTICLE

THOMAS AQUINAS ON HOMICIDAL SELF-DEFENSE

by Thomas A. Cavanaugh

The locus classicus of double-effect reasoning (DER) — often referred to as the “principle of double effect” — is Aquinas’ discussion of homicidal self-defense found in S.T. IIaIIae, q.64, a.7. In article seven, Thomas asks whether it is licit to kill a man in self-defense. He offers a number of objections against the liceity of so acting. St. Augustine voices two objections. The first comes from his epistle to Publicola; the second Thomas takes from Augustine’s De Libero arbitrio. There Augustine asks:

How are they free from sin in the sight of divine providence who, for the sake of these contemnible things have taken a human life? (q.64, a.7, ob.2)

Aquinas notes that among the slight goods which men may forfeit against their wills, Augustine includes corporeal life. Augustine appears to rule out homicidal self-defense.

Aquinas interprets Augustine as not permitting the intentional taking of an aggressor’s life. Thomas has noted earlier in his discussion of war (q. 40, a.1) that Augustine thinks it licit for one charged with the public good to take life during a war. Accordingly, in q. 64, a.7, Aquinas considers the bailiff and the soldier to be agents who may in self-defense and as public officials intentionally take the life of an aggressor. Thus, in q. 64, a.7, the self-defense of particular interest is that of the private individual, as such, taking the life of an assailant.

The corpus of q.64, a.7 reads:

Nothing prevents one act from having two effects, of which only one is intended, the other being praeter intentionem. Now moral acts receive their character according to that which is intended, not, however, from that which is praeter intentionem, since this is accidental, as is evident from what has been said earlier. Thus, from the act of self-defense, two effects may follow: one, the conservation of one’s own life; the other, the death of the aggressor. Since what is intended is the conservation of one’s own life, such an act is not illicit: it is natural for each thing to preserve itself in existence for as long as it is able. Nevertheless, some act proceeding from a good intention may be rendered illicit if it is not proportioned to the end. Thus, it would not be licit if someone defending his own life were to use more force than necessary. But, if he repels force with moderation, his defensive act will be licit: for, according to the jurists, it is licit to repel force by force, with the moderation of a blameless defense. Nor is it necessary for salvation that a man forego an act of moderate force in order to avoid the death of another: since one is more responsible to care for one’s own life than someone else’s. But, since to kill a man is not licit except for the public authority acting for the sake of the common good (as is evident from what was previously said [article 3]), it is not licit for a man to intend to kill another man in order to defend himself, except for those who have public authority. These, intending to kill a man in self-defense, refer this to the public good. This is evident in the case of a soldier fighting an enemy, and in the case of a minister of the judge fighting against thieves. Nevertheless, even these would sin if they were moved by private animosity.

What does Thomas mean by the phrase “praeter intentionem?”

Aquinas says that what is praeter intentionem is per accidens. In IaIIae, q.12, Aquinas considers intention. In q.12, a.1, Aquinas claims, “intention, just as the very word implies, means to tend to something.” Since the will moves the powers of the soul to their appropriate ends, it is evident, Thomas asserts, that intention is an act of the will. He argues that intention is the act of the will with respect to the end “as the term towards which something is ordained” (IaIIae, q.12, a.1, ad 4).

According to Thomas, we will the end, we choose the means, and we intend the complex, end-through-means. Using Aquinas’s example, when we intend health, we intend health-by-means-of-medicine. We choose medicine-for-the-sake-of-health. How does Aquinas understand the agent’s intention to relate to the goodness of the agent’s act?

Thomas offers an elaborate account of the goodness and badness of human actions in IaIIae, qs. 18, 19, 20, and 21. For the sake of understanding his statement in q.64, a.7, it is not necessary to articulate his entire analysis. Nevertheless, what he has to say about the relation of the intention of the end to the moral analysis of the goodness or badness of an act requires attention.

Aquinas proposes a complex analysis of actions. Each aspect relates variously to the others. This reflects the Dionysian dictum that goodness is integral; evil, the lack of such integrity, vitiates what otherwise is morally good. For the moral assessment of an act, three aspects of the act require attention, as Aquinas argues: what is being done (the deed or object), the circumstances in which it is done, and the end or reason it is done (IaIIae, q.18, a.1).

According to Thomas, of the aspects of an action which make up its integral goodness or its disintegrated badness, the intention of the end is a necessary, but not a sufficient condition for a complete analysis of the action’s ethical status.

In the light of his account of the relation between intention and the ethical analysis of action, it becomes clear what Aquinas means in q.64, a.7 when he asserts that, “moral acts receive their character according to what is intended, not according to what is praeter intentionem, for this is per accidens.” Clearly, Thomas asserts that what is praeter intentionem is not essential to establishing the agent’s action as good or as bad.

If the assailant’s death results from a private individual’s justified act of self-defense and the death is neither intended nor accidental, how is the death positively characterized? In q.64, a.7, Aquinas proposes and contrasts two cases of homicidal self-defense, that of a public official and that of a private individual. Aquinas holds that an officer of the polity — a police officer in contemporary terms — can intend to take the life of his aggressor as long as he uses minimal force, refers the slaying to the common good, and does not harbor animosity against the attacker (q.64, a.7). In the case of a private individual’s justified homicidal self-defense, Thomas accepts the slaying of the assailant as long as it results from the use of minimal force and is not intentional.

Say that I am a private individual. Both I and my assailant have swords. We begin to fight with them. I realize that my aggressor has far greater endurance than I and that the only way I can preserve my life is to kill him, say by cutting off his head. According to Thomas, I cannot do so because I cannot intentionally kill him. Thus, if this case were to obtain, then, according to Thomas, I, as a private individual, would not be permitted so to defend myself. If I were an officer of the state, however, executing my role as such, and I were in this same situation, Aquinas holds it permissible for me intentionally to take the life of the aggressor by cutting off his head.

Thus, while in both cases the force used must be minimal — not more than is necessary for the preservation of one’s life — this corresponds to a larger set of possible responses in the case of the officer of the state. For, according to Thomas, the officer of the state may proximately intend to take his assailant’s life. Therefore, he may use a neck-severing sword stroke, for it is proportioned to this end.

In the case of a private individual, however, minimal force does not include cases in which such force corresponds to an intention to take the life of the attacker. For, according to Thomas, the private individual cannot intentionally take the assailant’s life. Thus, the private individual cannot use means proportioned to the taking of the aggressor’s life. A neck-severing sword stroke is such a means. Such a sword stroke is proportioned to the preservation of one’s own life only insofar as it is proportioned to the taking of the aggressor’s life. Therefore, according to Thomas, a private individual cannot use such a sword stroke.

I have assumed that Aquinas would permit the use of a weapon, and even a potentially deadly one, a sword. A weapon is an instrument. As an instrument it admits of characteristic ends. One defending his own life with a sword may not maintain that the assailant’s death results accidentally from the employment of a sword. One of the ends to which sword-makers fashion swords is the taking of human life. Presumably, a sword not fit for the taking of another’s life is not much of a sword.

The use of a sword contrasts with pushing an attacker, who then stumbles on the curb, falls, and dies of a broken neck. Characteristically, pushing, shoving, pulling, scratching, biting, kicking, gouging, and generally being a great nuisance to an aggressor do not result in his death. If death were to result from such acts, it would result accidentally. Because death does not characteristically result from the ingenious deployment of teeth, nails, knees, elbows, and fists, one’s attacker could not charge one with endangering his life by so defending oneself.

If one were to use a sword, however, the attacker could claim that his life had been endangered. This is significant for two reasons. First, although the one defending himself by means of a sword may not intend to take the life of the aggressor, he is willing to risk taking the aggressor’s life. Second, if intending to take another’s life differs from knowingly endangering another’s life, then there is something else besides either the assailant’s death resulting intentionally or the death resulting accidentally; namely, there is the assailant’s death resulting as a risked consequence.

Accidental homicide differs from homicide which results from having endangered life. In an accidental killing, the agent inculpably does not foresee the death. When death results from having knowingly endangered someone’s life, however, the agent foresaw the death as a possible consequence of his action. Accordingly, when one kills someone accidentally, one is not ethically responsible for his death; when one kills someone whose life one has knowingly endangered, one is ethically responsible for his death. Thus, killing someone accidentally importantly differs from killing someone whose life one has knowingly endangered.

Does intentionally killing someone differ from killing someone as the result of risking his life? When one intends to take another’s life, one certainly endangers his life. Indeed, being the object of someone’s intention to kill is probably the most extreme case of having one’s life endangered. Does one intend to take another’s life if one endangers another’s life? For example, does one intend to take one’s own life when one endangers one’s own life?

Soldiers, stuntmen, race-car drivers, police officers, firefighters, and construction workers knowingly endanger their lives. Do they intend their own deaths? Perhaps some of them do and perhaps some of them ought not so to endanger their lives even if they do not intend to take them. In any case, it would indeed be an eccentric theory of intention which concluded that anyone who knowingly imperilled his life intended his death. Similarly, there is no reason to say that knowingly jeopardizing another’s life is to intend the other’s death.

Chancing the assailant’s life is precisely what I do if I do not intend to take his life, but I knowingly risk it in defense of my own life. I choose to risk his life rather than to forfeit my own, and such a choice on my part is ethically assessable. That the assailant’s death characteristically might follow from my using a sword in defense of my life indicates that I am more willing to preserve my life than I am to forego hazarding the assailant’s.

As I understand Aquinas, he proposes that a private individual may not intend to take the life of an assailant, while he may knowingly risk the assailant’s life by defending himself with such force that the aggressor’s death, if it results, would be one of the foreseeable characteristic consequences of the self-defensive act.

This interpretation may strike some as novel. Nevertheless, it accords with what Aquinas himself implies when he asserts that:

[T]he act of fornication or of adultery is not ordered to the conservation of one’s own life out of necessity as is the act from which sometimes (quandoque) follows homicide (q.64, a.7, ad 4).

Here, Aquinas does not appear to consider the foresight of an inevitable consequence, for such a consequence would not be said to follow “sometimes.”

I have argued that in q.64, a.7, Aquinas uses “praeter intentionem” to refer to a risked consequence. In his discussion of homicidal self-defense, he does not extend this concept to inevitable results. In fact, in his response to the fourth objection, he limits his use of the distinction to cases in which the result follows sometimes, and thereby excludes from his justification cases in which the harm follows always.

Aquinas’s account of DER is not simply that it is ethically in the clear to risk causing the death of one’s assailant insofar as one does not intend to kill one’s assailant. This is only the first condition of his account: the foreseen risked consequence is not intended. Like contemporary accounts of DER, Aquinas argues that in addition to this first condition, a second condition must be met for the risking of the harm to be justified. He argues that:

It is not necessary for salvation for a man to forego an act of moderate defense in order to avoid the death of another, since a man is more responsible to provide for his own life than for that of another (q.64, a.7).

Thomas asserts that one has a greater obligation to watch over one’s own life than to do so over another’s. Thus, when it comes to preserving lives, ceteris paribus, one is more obliged to preserve one’s own than another’s. Of course, covered by the “other things being equal” clause are such factors as the role one has with respect to the other’s life at risk. For example, a captain of a sinking ship may be more obliged to care for a passenger’s life than for his own.

In his account of a private individual’s justified homicidal self-defense, Thomas presents the seeds of DER as it is presently understood. Yet, in one important feature, his contribution differs from contemporary double-effect reasoning. That feature became evident in Thomas’s use of “quandoque” to characterize the assailant’s death as risked.

It requires a considerable, and, as I would argue, ultimately untenable interpretive stretch to attribute to Thomas the application of “praeter intentionem” in his treatment of a private individual’s act of homicidal self-defense to cases in which the assailant’s death is foreseen as resulting inevitably. Accordingly, one cannot attribute contemporary double-effect reasoning to Thomas tout court. Thomas does not use DER to justify a private individual’s homicidal self defense in cases in which the aggressor’s death is foreseen as inevitable. He does use it in cases in which the assailant’s life was risked. What is the value of noting this difference between Aquinas’s account and contemporary accounts of DER?

If one does not note this difference, then one will attribute to Aquinas an idiosyncratic account of intention which he does not have. For example, Jeff McMahan, following the customary interpretation, says:

Aquinas ... assumes that it is possible for one to foresee with certainty that one’s act will kill one’s assailant without intending the killing as a means of self-defence. ... To illustrate [this] view, consider:

Self Defence 1: One’s only defence against an unjust and potentially lethal attack is to shoot the attacker at close range with a flame-thrower.1

McMahan thinks that Aquinas and “the followers of Aquinas” hold that this case is an instance of self-defense justified by DER.2 If one thinks that a defender can shoot one’s attacker at close range with a flame-thrower, and that this is not intentional, then one seems to rely on a very narrow conception of what it is to intend a means. But, if one relies on such an account, then how will one argue, for example, that a terror bomber cannot drop bombs on noncombatants without intending their deaths? If one can use DER in the case of self-defense presented by McMahan, then one seems able to use it in terror bombing as well. As McMahan notes, such an account of DER, “results in an unacceptably permissive doctrine.”3 Thomas’s account of DER concerns the risking of harm. Accordingly, Thomas does not rely on an idiosyncratic account of intention which turns DER into the laxist account parodied by Pascal in the Provincial Letters and rightly rejected by McMahan.

Thomas A. Cavanaugh is assistant professor of philosophy at the University of San Francisco. The above piece is a shortened and revised version of the article “Aquinas’s Account of Double Effect,” which appeared in The Thomist in January, 1997 (Vol. 61, no. 1, pp. 107-121).





NOTES
1    McMahan, “Revising the Doctrine of Double Effect,” Journal of Applied Philosophy, 11.2 (1994): 202.
2    Ibid., 211.
3    Ibid., 212.