Consequentialism and Virtue

Some consequentialists may opt for eliminativism about virtue. After all, Jeremy Bentham (1843), a prominent consequentialist, eliminates “natural rights” from his broader moral theory calling them “nonsense upon stilts” precisely because his moral theory does not offer a good account of them. Why not think the same about virtue? We thank Folke Tersman for this question. In our view, there is a theoretical cost for consequentialists who eliminate virtue, because they ignore a central part of our moral experience.

Philosophers in the consequentialist tradition including Francis Hutcheson, David Hume, Jeremy Bentham, Henry Sidgwick, and G. E. Moore have embraced ideas that resemble the BCV to varying degrees. Those interested in this tradition should see the chapters on Hutcheson, Hume, and Mill in this volume; also see Bradley (2017) for a nice discussion of Moore’s account of virtue.

Thomas Hurka’s (2001, 3–4) account of virtue as an intrinsically good attitude toward an intrinsic good is not essentially consequentialist: “It [this account of virtue] can be accommodated within … consequentialism. The account … can be extended in a deontological setting.” Furthermore, Driver’s view is purely consequentialist in a way that others are not. For example, Rosalind Hursthouse (1999) and Linda Zagzebski (1996) offer views of virtue that have a consequentialist necessary condition alongside at least one non-consequentialist necessary condition.

See Halbig’s contribution in this volume.

If the claims made by some social psychologists and philosophers that character traits do not exist (“We need to abandon all talk of virtue and character, not find a way to save it by reinterpreting it” (Harman 2000, 224)) or that external circumstances wholly swallow the influence of character traits are true claims, then there would be no virtues or vices on Driver’s view. We, however, are skeptical about the truth of those claims; and most social psychologists and philosophers are skeptical about them too (see Miller 2014, 199–200). In fact, most acknowledge that we at least have local character traits that make a difference to behavior (Doris 2002, 64); more recently, Gilbert Harman (2003, 92) has agreed: “people may differ in certain relatively narrow traits”.

Driver (2012) does not think that luck can affect every aspect of the moral life; for example, she argues that even the consequentialist can consistently hold the view that two identical reckless drivers are equally blameworthy even though one of them kills a pedestrian. The idea is that the killer driver’s wrongdoing is more serious than the merely reckless driver’s wrongdoing, but they deserve the same degree of blame. For an introduction to the problem of moral luck and various attempts to solve it, see Hartman (2017, ch 1).

For an argument that counterfactual consequences are practically significant, see Hartman (2015, 89–92).

We thank Erik Carlson for pressing this point.

At one point, Driver (2001, 95) equates “systematically” with “overall,” which we do not think is representative of her position on this issue.

Hurka’s (2001) theory of virtue and vice is compatible with act consequentialism, and it makes virtue and vice intrinsically good and bad in ways that avoid the two counterintuitive implications and this theoretical objection.

There is an analogous objection in the moral responsibility literature to views that collapse the intuitive distinction between being bad and being blameworthy (see Hartman 2020, 109).

Driver’s view functionally collapses into the view that only the consequences produced by a particular trait count toward its virtue status if there is only one instance of each type of trait instantiated in an actual context. This collapse may be the case given the complex nature of our character traits if traits are individuated in very fine-grained ways (see Miller 2014).

This move is not available to consequentialists who specify the good in a way that precludes the prevention of suffering from being good. We thank Vuko Andrić for this point.

We can, however, see other motivations for contrastivism. For example, one might adopt it to circumvent our objection that Driver’s account cannot make the intuitive distinction between a meager good producing trait and a virtue. One might also adopt it to circumvent an objection raised in conversation by Krister Bykvist: if someone’s being a good or bad person is determined by their virtues and vices, then everyone in Horribleland is a bad person, which is a counterintuitive result.

For explication of various conceptions of luck, see Church and Hartman (2019).

We thank the participants of higher seminars at the University of Gothenburg, Stockholm University, and Uppsala University for questions and comments, and we especially thank Felix Timmermann, Vuko Andrić, and Krister Bykvist for extensive comments.