We best understand Rule Consequentialism as a theory of pattern-based reasons, since it claims that we have reasons to perform some action because of the goodness of the pattern consisting of widespread performance of the same type of action in the same type of circumstances. Plausible forms of Rule Consequentialism are also pluralist, in the sense that, alongside pattern-based reasons, they recognise ordinary act-based reasons, based on the goodness of individual actions. However, Rule Consequentialist theories are distinguished from other pluralist theories of pattern-based reasons by implausible claims about the relative importance of act-based and pattern-based reasons in different cases. Rule Consequentialists should give up these claims. They should either embrace some other pluralist pattern-based view, or reject pattern-based reasons altogether. Note, though, that these arguments apply only to compliance-based, rather than acceptance-based, versions of Rule Consequentialism. This suggests that these two kinds of theory are more different from each other than we might previously have realised.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
Tax calculation will be finalised during checkout.
See Hooker (2000, chapter 4).
See, for example, the papers in Hooker et al. (2000).
This means that the forms of Rule Consequentialism I shall criticise cannot use Hooker’s way of escaping the traditional collapse/incoherence dilemma, since that relies at crucial points on appeals to the costs of acceptance. See Hooker (2000, p. 94). However, I believe they can still escape this dilemma. I shall not try to show this here, since my purpose is not to defend compliance-based versions of Rule Consequentialism (but see note 5 below for some relevant comments). I am grateful to Nicholas Day for discussion of these issues.
To make the claims about reasons more plausible, we have to restrict them to claims about moral reasons. Presumably someone could have a (normative) prudential or aesthetic reason for action without that being related in the ways described to a set of rules. So, an interesting feature of Rule Consequentialism is that it seems to be committed to distinguishing in some fairly sharp way between moral reasons and other normative reasons.
Won’t Simple Rule Consequentialism contain just one rule: Act Consequentialism? No. We should formulate Rule Consequentialism in a way that captures its appeal, and part of that is to have different implications than Act Consequentialism. Simple Rule Consequentialism may contain rules that are practically equivalent with Act Consequentialism in a world of universal compliance with those rules (Regan (1980) aside). But so long as the content of the rules is not a statement of Act Consequentialism, its rules should not be practically equivalent with Act Consequentialism in other worlds, including our one. Provided we remember this, we need not adopt acceptance-based Rule Consequentialism to avoid practical equivalence with Act Consequentialism.
For simplicity, I will ignore the possibility of tied optimal sets of rules.
For discussion of this, see Hooker (2000, pp. 75–80).
This is one way of drawing the contrast between rightness and reasons, and it seems the most natural way for Rule Consequentialism to draw it. Note that Act Consequentialists typically draw it differently, by reference to the contrast between good in one way and good overall.
One result of this is that there may be no way of judging rightness without relying on intuitions about the upshot of conflicts between reasons. See Hooker (2000, pp. 131–134).
It is no part of my argument here to defend the existence of pattern-based reasons, though I try to do so elsewhere (Woodard 2007; see also 2002). Instead, I claim here merely that Rule Consequentialism is best understood as a theory according to which there are such reasons. If that is correct, Rule Consequentialists who deny the existence of pattern-based reasons are incoherent, whether or not that denial is correct.
Indeed, one can think of act-based reasons as a species of pattern-based reason, in which the action concerned is identical with the pattern whose rightness or goodness provides the reason. For ease of discussion, however, I shall speak of ‘pattern-based reasons’ in what follows as if the category were properly contrasted with that of act-based reasons.
See note 10 above.
I assume that we should evaluate theories in ethics partly in terms of their practical implications, and partly in terms of the plausibility of their component doctrines. As Parfit writes, “We have intuitive beliefs, not only about which acts are wrong, but also about which principles or theories might be true. So, as well as having plausible implications, a successful principle or theory must be in itself plausible. Only such a principle or theory could support our more particular moral beliefs” (2007, p. 264, emphasis in original).
Susan Hurley formulated this idea, though in a slightly different form. She writes: “Someone’s reason in acting may ... be that his act bears a constitutive relationship to a valuable form of agency. His contribution to its realization is not a causal one, but that of a part to a whole; it is hardly any less of a contribution, or irrational, on that score. If anything, the relation of part to whole seems more immediate than that of cause to effect” (Hurley 1989, p. 148). Notice that she emphasises being part of a valuable form of agency, whereas I emphasise performing an action that is part of a valuable pattern. One context in which this idea seems particularly salient is when people care about not being complicit in wrongdoing. On complicity, see Kutz (2000).
There is a debate in deontic logic between ‘actualists’ and ‘possibilists’. For a recent discussion, see Zimmerman (1996, chapter 6).
Parfit argues that Kantian Contractualism might justify Rule Consequentialism (2007, pp. 255–260; passim). However, despite appearances this is not an alternative rationale for the focus on rules in Rule Consequentialism, since Parfit’s only reason for preferring Rule Consequentialism to Act Consequentialism at the relevant stage in his argument (pp. 255–258) has to do with the two views’ practical implications. It is not a matter of the independent plausibility of the two views’ doctrines.
For example, see Hooker (2000, pp. 98–99).
Versions of Rule Consequentialism could support exceptionless prohibitions even if they contain disaster prevention rules, so long as they do not claim that every other rule could be broken permissibly to prevent disaster.
Simple Rule Consequentialism would include a disaster prevention rule only if its other rules are not practically equivalent to Act Consequentialism in every world. See note 5 above.
I am grateful to Alan Carter and others at the BSET Annual Conference for discussion here.
Hooker (2000, p. 80).
Hooker (2000, pp. 83–85).
Ridge (2006, p. 248). The statement of Variable Rate Rule Utilitarianism in this paper suggests that zero compliance should also be included, but Ridge corrected that in his presentation ‘Climb Every Mountain?’, at the University of Reading, 3rd November 2006.
See Hooker (2000, p. 163).
He also identifies the “realm of creation”, which concerns questions about which people to bring into existence. We can set this aside, however, since it does not bear on our present discussion. See Mulgan (2001, pp. 170–172).
Of course, Mulgan’s own view includes a pattern-based element, since Collective Consequentialism is a pattern-based theory. The point of replacing Collective Consequentialism with Rule Consequentialism is not to introduce pattern-based reasons, but to make his proposal directly relevant to our topic.
We can understand Ridge’s proposal as being pluralist in the relevant sense, as I explained in Section 3. However, if we understand it in that way it does not yield a clear explanation of why act-based reasons should sometimes predominate, since its concern with the consequences of single acts (or very attenuated patterns) is not identifiable with any feature of the final code (in contrast to the devices of a disaster prevention rule, or Mulganesque division of realms). It forms one part of the complex evaluation of each candidate code.
Some versions of Rule Consequentialism might claim that the act-based reasons exist all along, but only become important when a disaster is on the cards; others will claim that they only exist when a disaster is on the cards. The former claim is more plausible, since it is difficult to understand why the causal properties of actions, for example, should support reasons only when a disaster is on the cards. It is somewhat easier to suppose that these properties always support reasons, but that these reasons are important only when a disaster is on the cards.
The fact that the disaster-prevention rule embodies a tacit theory according to which the considerations grounding act-based reasons govern the interaction between pattern-based and act-based reasons may be a residue of truth in the old objection that Rule Consequentialism is incoherent.
For example, it gets the right answer in cases in which doing only one’s fair share of disaster prevention would have very bad consequences. In contrast, Murphy’s Collective Principle of Beneficence seems to get this sort of case wrong. See Murphy (2000, pp. 127–134).
Brad Hooker suggested to me that there is a sense in which acceptance-based Rule Consequentialism might be pattern-based: it explains reasons in terms of the goodness of patterns of motivation. This is a different species of pattern-based theory, which merits further investigation.
Hooker B (2000) Ideal code, real world. Clarendon, Oxford
Hooker B, Mason E, Miller D (eds) (2000) Morality, rules, and consequences. Edinburgh University Press, Edinburgh
Hurley SL (1989) Natural reasons. Oxford University Press, New York
Kutz C (2000) Complicity. Cambridge University Press, Cambridge
Mulgan T (2001) The demands of consequentialism. Clarendon, Oxford
Murphy LB (2000) Moral demands in nonideal theory. Oxford University Press, New York
Parfit D (2007) Climbing the mountain. http://individual.utoronto.ca/stafforini/parfit/parfit_-_climbing_the_mountain.pdf. Cited 21 March 2007
Regan D (1980) Utilitarianism and cooperation. Clarendon, Oxford
Ridge M (2006) Introducing variable-rate rule-utilitarianism. Philos Q 56:242–253
Woodard C (2002) Group-based reasons for action. Ethical Theory Moral Pract 6:215–229
Woodard C (2007) Reasons, patterns, and cooperation. Routledge, New York (in press)
Zimmerman MJ (1996) The concept of moral obligation. Cambridge University Press, Cambridge
For their helpful comments and questions, I am very grateful to audiences at the University of Nottingham and at the Annual Conference of the British Society for Ethical Theory, University of Bristol, 9–11 July 2007. I am especially grateful to Roger Crisp and Nicholas Day.
About this article
Cite this article
Woodard, C. A New Argument Against Rule Consequentialism. Ethic Theory Moral Prac 11, 247–261 (2008). https://doi.org/10.1007/s10677-007-9083-5
- Act Consequentialism
- Rule Consequentialism
- Act-based reasons
- Pattern-based reasons