American Catholic Philosophical Quarterly

While our common sense seems to tell us that intention matters to ethical evaluation, there is considerable disagreement among ethicists regarding why and how it matters. In this article I argue that intention matters to act evaluation in much the way that the principle of double effect (PDE) implies. First, I identify five propositions—one epistemological and four ethical—that the proponent of PDE holds regarding the ethical relevance of intention. Second, I give two general arguments for the ethical relevance of intention. Third, I offer preliminary arguments for each of the five propositions outlined in the first section. Together, these general and more specific arguments are meant to place the burden of proof on one denying intention’s relevance to act evaluation. Fourth and finally, then, I show that recent critiques of PDE and of the ethical relevance of intention fail to carry that burden of proof.

This essay consists of some clarifying remarks on the doctrine of double effect (DDE). After providing a contemporary formulation of the doctrine we put special emphasis on the distinction between those aspects of an action plan that are intended and those that are merely foreseen (the I/F distinction). Making use of this distinction is often made difficult in practice because salient aspects of the action plan exhibit a felt “closeness” to one another that is difficult if not impossible to articulate with the precision we might like. The essay goes on to examine an especially adroit criticism of DDE best articulated by J. J. Thomson. We conclude with a brand new double effect case (new to the philosophical literature anyway) taken from medicine and Roman Catholic pastoral ministry.

Two hard cases have generated controversy regarding the application of the principle of double effect in recent years. As regards the first, the case of the conjoined twins of Malta, there has been considerable convergence: most natural law ethicists seem to agree that separation of the twins was morally permissible. By contrast, the so-called “Phoenix case,” involving an abortion at a Catholic hospital for a woman with pulmonary arterial hypertension, has become a touchstone of disagreement between defenders of the so-called “new” natural law theory, and more “traditional” Thomists. I argue in this essay that, contrary to widespread opinion, the two cases were alike in the following respect: in neither need the principal agents have intended the death of anyone.

This paper considers the problem of closeness in the ethical use of intention. In section I, attempts inspired by Anscombe to use a “coarse grained” understanding of intention, to deal with certain difficult cases, are rejected. In section II it is argued that the difficult cases can be addressed using other moral principles. In section III a more detailed account of intention is set out, analysing intention as a reason for action, and in section IV two paradoxes apparently created by this account are addressed: on the contrast between intentions and intentional action, and the difference between killing a group of people together or individually. In section V another set of cases is considered, to test how this account of intention handles the intention of harm. Section VI considers the objection that an agent may cause what is a harm without intending it as a harm.

Essential to the doctrine of double effect is the idea that agents are prohibited from intending evil as a means to a good end. I argue in this paper that some recent accounts of intention from proponents of double effect (such as by Michael Bratman and John Finnis) cannot sustain this prohibition on harmful means. I outline two ways to gerrymander intention that mark these accounts. First, intention is construed in such a way that an agent intends only those states of affairs that she cares about or finds motivating for their own sake. Second, intention is construed in such a way that what counts as intended is determined sufficiently by the agent’s beliefs.

The core of the double effect rule supposes the existence of a kind of impermissible action whose impermissibility is determined by its including the intention of a bad result. How can the reality of actions having this tight connection between intending bad results and impermissibility be justified? None of the obvious justifications is promising. But the conditions of human agency provide a justification for the centrality of intention within the impermissible actions double effect addresses. The human power to avoid intentional actions is robust, but not the power to avoid unintended bad results. Supposing there is a normative case for indefeasible prohibitions (which the rule does not establish but needs if it is to have application), limiting them to intentional actions is warranted, since the prohibition can be complied with. But when unintended bad results are not avoidable, such a prohibition would demand the impossible.

The traditional distinction between the agent’s intention and the effects which she merely permits would seem to allow for a re-description of the act in terms of the agent’s overall good aims. This paper argues that Aquinas understands the relation between the agent’s choice and her overall intention in a different and more persuasive way. His analysis of the object and the end of the act is complicated, but once the relevant distinctions have been sorted out, it is apparent that he does not hold that a particular action can be described, or much less morally evaluated, in terms of the agent’s overall good intentions. On the contrary, he insists that the object of the agent’s immediate choice is always morally relevant, and can be morally decisive for assessing the overall value of the act.

This paper reconsiders whether Aquinas is rightly read as a double-effect thinker and whether it is right to understand him as concurring with Paul’s dictum that evil is not to be done that good may come. I focus on what to make of Aquinas’s position that, though the private citizen may not intend to kill a man in self-defense, those holding public authority, like soldiers, may rightly do so. On my interpretation, we cannot attribute to Aquinas the position that aiming to kill in self-defense is prohibited where so aiming is the only way to stay alive. Instead, for the private citizen though not for the public authority, it is aiming to kill as an end in itself, over and above the aim of saving one’s life, that is prohibited. Accordingly, we also cannot attribute to Aquinas the third condition of the principle of double effect in its textbook formulation.

Many philosophers assume that the principle of double effect (PDE) is meant to cover trolley cases. In fact, trolley cases come from PDE’s critics, not its defenders. When philosophers stretch PDE to explain intuitions about trolley cases, they define intended effects too broadly. More importantly, trolley cases make poor illustrations of PDE because they focus attention away from the agent and onto the victim. When philosophers lose sight of the agent, some intuitions that fit PDE survive, but the rational basis of these intuitions collapses. I avoid these problems by defending a minimalist, agent-based version of PDE. My version is minimalist because I do not try to turn PDE into a complete checklist that explains intuitions about every case. It is agent-based because I consider the agent’s perspective to define intentions and to make moral judgments.

If viable, DER justifies certain individual acts that—by definition—have two effects. Presumably, it would in some fashion (at the very least, redundantly) justify policies concerning the very same acts. By contrast, acts that sometimes have a good effect and sometimes have a bad effect do not have the requisite two effects such that DER can justify them immediately. Yet, a policy concerning numerous such acts would have the requisite good and bad effects. For while any one such act would lack the relevant two effects, a series of such acts and a policy governing such a series would have them. This paper addresses DER’s justification of policies that apply to such acts. It shows that there are certain acts which DER mediately justifies by justifying policies (having the requisite two effects) concerning them. Thus, it recommends the larger topic of DER’s bearing on policy.