Abstract
According to the “Textbook View,” there is an extensional dispute between consequentialists and deontologists, in virtue of the fact that only the latter defend “agent-relative” principles—principles that require an agent to have a special concern with making sure that she does not perform certain types of action. I argue that, contra the Textbook View, there are agent-neutral versions of deontology. I also argue that there need be no extensional disagreement between the deontologist and consequentialist, as characterized by the Textbook View.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
This case is deliberately schematic in order to guide our focus to the number of possible deaths and the ways that these may come about. So in discussing this case, and all those that follow, let us assume that “all else is equal.” (It does not turn out that one of the five is on the brink of discovering the cure for cancer, and so on.)
By saying that killings are worse than deaths, she may claim that you may not kill in the well-known case:
transplant. You can kill an innocent person and transplant her organs to save five others.
“A constraint on what?”, we might ask. Shelly Kagan portrays a constraint as a constraint on maximizing the good. Kagan (1989) reason is that deontologists recognize that there is a “pro tanto reason to promote the good” since they think you have a reason to help others. This may make it seem as if a deontologist starts with a consequentialist view, and then adds extras. This portrayal leaves them vulnerable to the charge that their additions are ad hoc, and posited only to save intuitions about cases. But deontologists need not think of their theories in this way. Indeed some, such as Judith Jarvis Thomson, explicitly deny that there is a property of goodness simpliciter. They ground an obligation to help others in other ways, such as in a virtue of benevolence (Thomson 1993). Still, the language of “constraints” can survive the observation that deontologists need not be fashioned in the image of consequentialists. Rather than positing “constraints on promoting the good,” deontologists could be seen as positing “constraints on action-types.” I will use the term “constraint” in this way to denote a deontic prohibition on certain action-types like killing. By contrast, a theory that simply claimed that there are morally relevant thresholds would not thereby count as positing a constraint in this sense. For example, consider the theory that an action is right just in case it produces more happiness than all other actions, unless it produces more than a certain threshold amount of happiness. This threshold theory would not count as positing a constraint in the sense of the word I intend. Thanks to Wayne Davis for prompting me to clarify this latter point.
The constraint would have to be time-relative as well as agent-relative if the constraint were to forbid an agent from killing to prevent the agent herself killing at other times. For discussion of time-relativity, see Smith (2009).
McNaughton and Rawling (1992).
See, for example, Kagan (1989), Scheffler (1994) and Pettit (2000). The claim that deontological theories are agent-relative (which is a key claim I argue against in this essay) is even more widespread. As Michael Ridge notes in his helpful discussion of agent-relative and agent-neutral reasons, “the characterization of deontological restrictions as agent-relative (or agent-centered) is close to being an orthodoxy.” Ridge (2008).
Jacobsen (2008, p. 165). Opponents of Dreier’s conception can still accept, as Jacobsen does, the most important point of Dreier’s, which was that some agent-relative theories have a teleological structure that gives a fundamental explanatory role to claims about consequences.
Anscombe (1958).
Thanks to Wayne Davis for suggesting this formulation of the principle.
In McNaughton and Rawling’s terminology, where “S” stands for “should (ceteris paribus) ensure that” and where square brackets mark out the content of a rule, the logical form of this rule is
(x) [x S (x does not kill to prevent more killings by others)]
Since there is an occurrence of “x” within the square brackets that is bound by the initial universal quantifier “(x)”, this is an agent-relative rule. This formalization aims to capture the intuitive idea that a deontic constraint against killing enjoins an agent to take special care that she not kill. McNaughton and Rawling (1991).
Thanks to Caspar Hare for posing these questions.
In McNaughton and Rawling’s terminology, this rule is
-
(x)
x S [(y) y does not kill to prevent more killings by others]
Since there is no occurrence of “x” within the square brackets bound by the initial universal quantifier “(x)”, this is an agent-neutral rule. This is a formalization of the intuitive idea that the content of an agent-neutral requirement does not give the agent any particular concern with her own killings (McNaughton and Rawling 1991).
-
(x)
We should not be misled into thinking that this is an agent-relative requirement simply because in some situations it requires the agent not to kill. Here she would simply be “ensuring… that something is true of her—but only insofar as she is one amongst many.” (McNaughton and Rawling 1991, p. 179).
I owe the point that agent-neutral consequentialist theories can posit constraints to Caspar Hare. In an unpublished manuscript, Hare considers an axiological principle like killings and rejects it as implausible. In its place, he suggests that an improvement would be to place a special importance on the actions in virtue of which two outcomes differ when ranking them from an agent-neutral perspective. He observes that by doing so the consequentialist can make the same extensional claims about which actions are right and wrong as a deontologist.
Since my intention is to provide a counterexample to a key component of the Textbook View, my criticism does not depend on the plausibility of the theory I offer as a counterexample. Still, we might independently wonder how plausible a theory with the claim, killings, is. I do not think such a theory is correct, but neither do I do think it is off the wall. Some people claim that the deontic status of certain actions depends on their causal relations to other actions. In light of this, it is not such a leap to make the teleological claim that an outcome’s value depends on the causal relations between its constituent actions. Thanks to Wayne Davis for prompting me to address this point.
This point concerns a simple case where the prevented harms are two further harms of the same type. But the point generalizes to cases that involve more complicated causal dependencies between tokens of action-types. For these more complicated cases, the consequentialist would need to supply other axiological principles. But we have a general recipe for seeing how she would do so in order to generate the same extensional results as a deontologist. According to the Textbook View, the deontologist posits principles with this form:
agent-relative general constraint. Each agent should ensure that she does not Φ to prevent a group of others from Ω-ing.
We can observe that for every principle with the above form, there is an axiological claim with the following form,
general axiological claim. If Φ-ing, Φ1, prevents Ω-ings, Ω1, Ω2,…, Ωn, then an outcome with Φ1 is pro tanto impersonally worse than an outcome with Ω1, Ω2,… ,Ωn.
A consequentialist can posit principles of the latter form in order to generate the same extensional results as a deontologist who posits principles of the former form.
Thanks to an anonymous reviewer for suggesting that I address this question.
My proposal is inspired by Daniel Jacobsen’s claim that Utilitarianism is a movement in the history of ideas. Jacobsen would resist this proposal since he does not think of consequentialism as such a movement, but instead as a “philosophers’ term of art, which means whatever philosophers have meant by it over the past half-century or so, when the term was coined and earned its place in the philosophical lexicon.” I am sympathetic to Jacobsen’s claim, but I think that the fundamental use for the term in the lexicon is to pick out the tradition in moral philosophy that I point to. Interestingly, Jacobsen claims that John Stuart Mill did not endorse an agent-neutral theory. Jacobsen (2008, p. 164).
References
Anscombe, G. E. M. (1958). Modern moral philosophy. Philosophy, 33, 1–19.
Bentham, J. (1903). An Introduction to the Principles of Morals and Legislation; G. E. Moore, Principia ethica, rev. ed., ed. Thomas Baldwin. Cambridge: Cambridge University Press.
Broome, J. (1991). Weighing goods. New York: Blackwell.
Dreier, J. (1993). The structure of normative theories. The Monist, 76, 22–40.
Feldman, F. (1995). Adjusting utility for justice: A consequentialist reply to the objection from justice. Philosophy and Phenomenological Research, 55(3), 567–585.
Jacobsen, D. (2008). Utilitarianism without consequentialism. Philosophical Review, 117(2), 159–191.
Kagan, S. (1989). The limits of morality. Oxford: Clarendon Press.
McNaughton, D., & Rawling, P. (1991). Agent-relativity and the doing-happening distinction. Philosophical Studies, 63(2), 167–185.
McNaughton, D., & Rawling, P. (1992). Honoring and promoting values. Ethics, 102(4), 835–843.
Nozick, R. (1974). Anarchy state and Utopia. New York: Basic Books.
Pettit, P. (2000). Non-consequentialism and universalizability. The Philosophical Quarterly, 50(199), 175–190.
Portmore, D. (2005). Combining teleological ethics with evaluator relativism: A promising result. Pacific Philosophical Quarterly, 86, 95–113.
Ridge, M. (2008). Reasons for action: agent-neutral vs. agent-relative. In: E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Fall Edition). http://plato.stanford.edu/archives/fall2008/entries/reasons-agent/.
Scheffler, S. (1994). The rejection of consequentialism. Oxford: Oxford University Press.
Schroeder, M. (2007). Teleology, agent-relative value, and good. Ethics, 117, 265–295.
Sen, A. (1982). Rights and agency. Philosophy and Public Affairs, 11, 3–39.
Smith, M. (2009). Two types of consequentialism. Philosophical Issues, 19(1), 257–272.
Thomson, J. J. (1993). Goodness and utilitarianism. Proceedings and Addresses of the American Philosophical Association, 67, 145–159.
Acknowledgments
For helpful comments and discussions, I would like to thank Wayne Davis, Daniel Halliday, Alejandro Pérez Carballo, Paulina Sliwa, Ekaterina Vavova, the participants in the 2011 Ethics in Society Post-doc Workshop at Stanford University, and an anonymous reviewer for Philosophical Studies. I would like to thank Caspar Hare in particular for invaluable discussions and feedback on multiple drafts of this essay.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Dougherty, T. Agent-neutral deontology. Philos Stud 163, 527–537 (2013). https://doi.org/10.1007/s11098-011-9829-8
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-011-9829-8