the consequences are bad enough. However, I do not think that this “dramatically lesser evil” argument is the best way to morally analyze most cases of torture, though it may be the way to analyze the so-called Dostoevskian ones in which an innocent who has no responsibility for the threat facing the world must be tortured or killed to save millions.4 In cases where the person to be tortured is morally responsible for the threat, there could be other grounds than that torturing him is a “dramatically lesser evil” done to prevent the threat from coming to fruition. For example, Danto describes someone who is refusing to give information that he has that could save many people. This person might not have started the threat facing them; however, he is “letting them die” when he could easily save them (and so has what is called “negative responsibility” for the event). However, if letting die is morally different from killing, this could argue against his having sufficient responsibility for the threat to justify torturing him to prevent it. By contrast, suppose the very person who presented the threat to the world, for example, Hitler, was the person who would have to be tortured to stop the threat from coming to fruition. The argument for the moral permissibility of torturing him is not that it is the dramatically lesser evil, but that Hitler’s wrong act makes him responsible for the threat, and so he is liable to be harmed in order to save his victims. Call this the Responsibility Argument. I think it may succeed.
5 The Responsibility Argument need not be restricted to cases involving an enormous loss to the world. Consider ordinary cases in which a villain threatens to kill one other person (e.g., he has set a bomb that will go off unless we torture him for the information to stop it). I agree with Danto that we should not rely on the extreme case (in which many lives are at stake) to argue about the ordinary case, but this is consistent with it being correct to use the same type of Responsibility Argument here as in the extreme case: In order to save his potential victim once he has finished setting his bomb, the villain could be liable to suffer at least the sorts of harms that it would be permissible to impose on him while he is establishing his threat in order to stop him doing so.9 Using the Responsibility Argument to justify sometimes torturing in such non-extreme cases would not signify “the devaluation of value.”
Notes
1 1 See Danto 1981.
2 2 See Wollheim 2012.
3 3 These cases are due to James Rachels (see Rachels 1975).
4 4 It is also possible that, as Wollheim says in the case of art and non-art, it is because killing and letting die differ per se that a particular killing is morally equivalent to a particular letting die. For more on this see Kamm 1996.
5 5 Danto 1974.
6 6 Oddly, Danto goes on to say that despite assassination’s “categorical extremity” it is not really “beyond good and evil” and it is not true “that nothing morally useful can be said about political assassination” (Danto 1974, 434–5).
7 7 Some of what I say here is based on my discussion of torture in Kamm 2011.
8 8 See Nagel 2012, 74. However, it may rather be true that one of the acts is right even though it wrongs someone. For example, I have argued that even though it is permissible to turn a trolley from five toward one other person, that person may be wronged by our doing this. See, for example, my discussion in Kamm 2016.
9 9 See Kamm 2011 for additional steps in the argument.
References
1 Danto, Arthur C. 1974. “A Logical Portrait of the Assassin.” Social Research 42(3): 426–38.
2 ––––. 1981. Transfiguration of the Commonplace. Cambridge, MA: Harvard University Press.
3 Kamm, F. M. 1996. Morality, Mortality, Vol. 2. New York: Oxford University Press.
4 ––––. 2011. Ethics for Enemies: Terror, Torture, and War. New York: Oxford University Press.
5 ––––. 2016. The Trolley Problem Mysteries. New York: Oxford University Press.
6 Nagel, Thomas . 2012. “War and Massacre.” In Mortal Questions, 53–74. Cambridge: Cambridge University Press.
7 Rachels, James . 1975. “Active and Passive Euthanasia.” The New England Journal of Medicine 292: 78–80.
8 Wollheim, Richard . 2012. “Danto’s Gallery of Indiscernibles.” In Danto and His Critics, edited by Mark Rollins, 2nd ed., 30–9. Hoboken, NJ: Wiley-Blackwell.
9 A Normative Perspective on Basic Actions
CAROL ROVANE
Are there such things as basic actions? Is the idea of a basic action philosophically important? Arthur Danto’s novel, and deservedly influential, contribution to the philosophy of action is to raise these questions and then to answer them with a most definite yes.
I want to consider how we should answer these questions if we bring a normative orientation to bear in the philosophical study of agency. Such an orientation has become a familiar fixture of philosophical work on the topic of agency. But I want to situate Danto’s work in relation to a particular elaboration of it that has emerged at Columbia University, where Arthur presided for so long as a senior philosophical figure. This particular normative orientation – which I will characterize as thoroughgoingly normative – governs Isaac Levi’s work on decision theory and rational choice, as well as Akeel Bilgrami’s work on self-knowledge and the metaphysics of agency and value, and also my own work on agent-identity, self-constitution, and the metaphysics of value.
This thoroughgoingly normative orientation invites us to look more closely at aspects of Danto’s views that have not drawn nearly as much philosophical attention as his conception of basic actions, which has enjoyed such enormous influence. The aspects I have in mind concern the agent’s first-person point of view. Although Danto’s scattered remarks about this are fascinating, I do not find in his work a sufficiently elaborated account of three aspects of it: what is epistemically distinctive about the sort of first-personal knowledge that agents possess; what is metaphysically distinctive about the sorts of things that agents know from a first-person point of view; what is a first-person point of view. In what follows, I explore all of these matters, both from Danto’s point of view and from the thoroughgoingly normative point of view of his Columbia colleagues.
1 What Danto Says about the Agent’s First-Person Point of View
Let me remind readers how Danto defines basic actions. His initial definition is negative: when we perform a basic action there is nothing else that we do in order to cause it. But he also specifies, more positively, that a basic action is something we do just like that, at will. Two correlative claims about non-basic actions follow: when we perform a non-basic action, there is something else that we do in order to cause it; in such cases, we do one thing, the non-basic thing, by doing another thing, which may or may not be basic.
Danto claims that each agent has a repertoire of basic actions, which roughly speaking is a set of capacities to perform various kinds of basic actions. He claims further that agents know that they are able to perform the various kinds of basic actions that belong to their repertoires, but they do not know how
