Oxford Uehiro Prize in Practical Ethics: Global Warming & Vegetarianism: What should I do, when what I do makes no difference? By Fergus Peace

This essay received an Honorable Mention in the Graduate Category of the Oxford Uehiro Prize in Practical Ethics 2017

Written by University of Oxford student, Fergus Peace

  1. The Problem of Cumulative Impact

In large, integrated societies, some of the most important moral challenges we face can only be resolved by large-scale collective action. Global poverty and climate change are problems which won’t be solved unless large numbers of people act to address them.

One important part of our response to these problems is to avoid fallacious ‘futility thinking’, a cognitive bias which makes people less likely to act when they see the problem as being too large for them to solve. You aren’t going to end world poverty alone, but that doesn’t mean there’s nothing you should do about it. Your individual donations can make an enormous difference.

Other problems, however, are more philosophically and practically challenging. Sometimes morally significant outcomes are driven by an aggregate which your individual action is powerless to meaningfully affect. In these cases, it’s not just that your individual action won’t completely solve the problem: it won’t do any moral good at all.

Consider a few examples.

  • Voting: No election of any real size is decided by a margin of one vote, so it’s true of your vote that it makes no difference: if you don’t vote and your candidate loses, your vote wouldn’t have made them win; if you do vote and they win, withdrawing your vote wouldn’t have made them lose.
  • Vegetarianism: Butchers don’t respond to every small change in their customers’ purchasing; wholesalers don’t respond to every change in one butcher’s purchasing; abattoirs and farms don’t respond to every change in wholesale orders. If you don’t buy pork today, your supermarket will order just as much meat next month – animal suffering won’t be reduced at all.
  • Climate Change: By a similar chain, through suppliers, wholesalers and power generation companies, your turning off a laptop or light overnight won’t at all impact how much coal or gas gets burned. And a small change in the total amount of greenhouse gases emitted won’t affect any of the morally significant outcomes in any case.

This leaves each one of us, as an individual, in an ethically puzzling position. Though I clearly have reason to wish that millions of people tomorrow take public transport to work instead of driving, it’s hard to see why I should have any reason at all to do so myself. What we should do as legislators or policymakers is not touched by this problem. But if, as we surely want to say, we also have duties to respond to these pressing global problems as individuals, then the apparent irrelevance of our individual actions is troubling.

Schematically, these are cases in which many people φ-ing will produce a morally significant outcome, O, but each individual instance of φ-ing makes no difference to whether (or to what extent) O is realised. Explaining why each person nonetheless has reason to φ is what I will call the problem of cumulative impact.[1] There have been a number of proposed solutions to this problem. Some have suggested that all cumulative impact cases have a ‘threshold’ structure, in which one or a few individuals acts make a truly vast difference, and therefore that expected utility explains why I have reason to φ.[2] Others claim that the reason to φ arises as a matter of fairness, since you shouldn’t force others to bear all the burden of responding to our moral problems.[3] Other suggestions, too complex to fairly summarise, posit the existence of another kind of moral relationship between action and outcome, which doesn’t require you to have made a difference.[4]

I don’t have space in this essay to address these solutions, but none are fully adequate responses to the problem of cumulative impact. To paint in extremely broad strokes, all of them ultimately founder on the reality of how little impact each individual action has: so little that expected utility calculations may not come out the way we want them to[5], that it’s not clear how considerations of fairness can even become relevant[6], and that the reasons generated by putative other kinds of moral relationship are extremely weak.[7] This worrying moral problem is still, therefore, in search of a solution.

  1. A Puzzle About Imperfect Duty

Leave aside the practical issue for a moment, and let’s turn to a longstanding puzzle in normative ethical theory. Part of moral philosophy since at least the early 17th century is a distinction between ‘perfect’ and ‘imperfect’ duties. No very precise statement of this distinction is widely accepted, but roughly speaking imperfect duties admit a greater degree of latitude in how you can go about fulfilling them.

To illustrate with traditional examples, the duty to repay a debt is perfect. It requires you to give a particular amount of money, to a particular person, at a particular time. The duty to give to charity, however, is imperfect: you must give to charity, but you’re not required to give today rather than tomorrow, or to give to Oxfam or UNICEF or any charity in particular.

The puzzle about imperfect duty is twofold. Firstly, there is great disagreement about how the distinction should be understood, even in broad terms: George Rainbolt identifies eight different glosses of it, which don’t appear to line up with each other. Disagreement about proposals for a more precise formulation are, unsurprisingly, also widespread.[8] But secondly, it’s not very clear why there should be such a thing as imperfect duty. If a type of action is morally significant enough to ground a duty, why should there be such pure optionality about whether to do actions of that type on every particular occasion? W.D. Ross didn’t think any such action could really be a duty.

This paradox is eased slightly by attending to imperfect duty’s structure.[9] It might be the case that you can choose not to give to a charity today for no reason other than that you feel like it, but it’s not the case that you can opt out of doing your duty simply because you feel like it. It’s just that you don’t have a duty to act on any particular occasion – only to do acts of that type a certain number of times, or with a certain frequency. Only slightly, though: this still leaves unexplained why there should be a duty to do actions of type T periodically, if it’s morally indifferent whether you do a T-type action on each particular occasion the opportunity arises.

I’m going to propose that we simply accept this. Someone has an imperfect duty to perform some type of action when it’s morally significant that they perform some pattern of actions over time, but unimportant how they instantiate that pattern, and whether they do any particular action which could form part of it. This is a new way of understanding imperfect duty, which I’ll call the pattern-centred conception. Since imperfect duty is a technical notion invented by philosophers, this conception of the distinction will stand or fall on nothing other than its usefulness as an analytic category in moral theory. (As a historical matter, I think there is decent evidence for attributing a version of this conception to Kant; but I won’t argue that here.)

The main philosophical merit of this proposal will become clear in the next section. For now, we can illustrate the proposal with one of Kant’s imperfect duties, the duty to develop one’s talents. The sort of self-perfection that interests Kant is achieved by consistent pursuit of it over time. It doesn’t matter whether I do an hour’s piano practice tonight: this single hour of practice won’t make a difference to whether I adequately develop my talents. A pattern of action, however – practicing for an hour five times every week – could clearly make such a difference. So it matters morally that I perform such a pattern, but not whether I practice on any particular evening; that, on my view, is the heart of imperfect obligation.

  1. Imperfect Duty & Cumulative Impact

Let’s circle back to the problem of cumulative impact. ‘Cumulative impact’ cases, in which many acts together matter morally while no particular one does, come in horizontal and vertical varieties. Ordinarily the focus is on horizontal cumulative impact, referred to as ‘collective action’: a matter of many individuals’ contemporaneous decisions aggregating into an important outcome. This is an important aspect of the phenomenon, and the only characterisation for cases like voting. But there are also cases of vertical cumulative impact: when a single individual’s decisions over time aggregate into something significant. The pattern-centred conception of imperfect duty is well-suited to address this latter category.

Many cumulative impact cases are both horizontal and vertical. It would make a difference if everyone cycled rather than drove today; but it would also make a difference if I cycled rather than drove every day. My decisions over time have an aggregate impact which is morally significant, and which I can affect through consistent patterns of action even if no particular choice matters. If protecting living conditions for future generations or reducing animal suffering are important goals, then we each have a duty to enact behaviour patterns which greatly reduce our carbon emissions and meat consumption. Hence the duties to reduce carbon emissions or not eat meat are imperfect.

With these cases in hand, it becomes even clearer how the pattern-centred conception solves the core dilemma about imperfect obligation. How could it be a matter of moral indifference whether I eat meat on some particular occasion? Because it is a literal matter of indifference; nothing of moral importance will change as a result of that choice. Then how could there be a duty to eat less meat in general? Because that behaviour pattern does affect something of moral significance. The dilemma no longer has any bite.

On this conception, imperfect duty illustrates the point about morality that Warren Quinn’s ‘puzzle of the self-torturer’ makes about prudential rationality: to properly understand the reasons bearing on our actions, we must view them as parts of patterns or strategies, not only singly.[10]

  1. Conclusion

Imperfect duty is not a total solution to all the problems associated with cumulative impact. For one thing, it doesn’t offer anything for purely horizontal cases like voting. Moreover, I don’t want to overstate the problems of other responses to the problem which have been proposed. In fact, one particular merit of recognising imperfect duties to reduce energy and meat consumption is how this proposal can combine with – and improve the power of – other ways of responding to the problem of cumulative impact.

As we saw above, the core difficulty for all these approaches was that individual actions are so insignificant that it’s hard to connect them to the outcome in any way that could matter morally. If we are taking patterns of behaviour over time as the unit of analysis, rather than individual actions, that problem becomes less acute. My energy consumption over the course of my life is quite large, so the probability that my contribution crosses some tipping point is higher, and the expected utility of reducing it therefore greater. Since it’s greater, it’s also more plausibly described as a contribution to a collective project of fighting global warming, which means that considerations about fairness get a grip, in a way they can’t with respect to individual actions.

These different solutions to the problem of cumulative impact can thus complement each other. The key is to break with the tradition of focusing ethical analysis on individual acts alone. Turning our moral attention to the patterns of behaviour we enact over time shows that we do, after all, make a difference and that we should recognise the moral duties which flow from that fact.



Budolfson, Mark Bryant, forthcoming. ‘The Inefficacy Objection to Consequentialism and the Problem with the Expected Consequences Response’, Philosophical Studies.

Cullity, Garrett, 2000. ‘Pooled Beneficence’, in Imperceptible Harms and Benefits, M. Almeida (ed.). Kluwer Academic Publishers.

Hill, Thomas E, Jr., 1971. ‘Kant on Imperfect Duty and Supererogation’, Kant-Studien 62(1), pp55-76.

Kagan, Shelly, 2011. ‘Do I Make a Difference?’, Philosophy & Public Affairs 39(2), pp105-141.

Murphy, Liam, 2003. Moral Demands in Nonideal Theory. Oxford University Press.

Nefsky, Julia, 2011. ‘Consequentialism and the Problem of Collective Harm: A Reply to Kagan’, Philosophy & Public Affairs 39(4), pp364-395.

Nefsky, Julia, 2015. ‘Fairness, Participation, and the Real Problem of Collective Harm’, in Oxford Studies in Normative Ethics, Volume 5. Oxford University Press.

Nefsky, Julia, ms. ‘How You Can Help, Without Making A Difference’.

Quinn, Warren S., 1990. ‘The Puzzle of the Self-Torturer’, Philosophical Studies 59(1), pp79-90.

Rainbolt, George, 2000. ‘Perfect and Imperfect Obligations’, Philosophical Studies 98(3), pp233-256.

Regan, Donald, 1980. Utilitarianism and co-operation. Oxford University Press.

Sinnott-Armstrong, Walter, 2005. ‘You Ought To Be Ashamed Of Yourself (When You Violate An Imperfect Moral Obligation’, Philosophical Issues 15, pp193-208.

Statman, Daniel, 1996. ‘Who Needs Imperfect Duties?’, American Philosophical Quarterly 33(2), pp211-224.

[1] It’s also aptly called ‘the problem of collective action’, but for reasons that will become clear I prefer to make a slightly different emphasis.

[2] Kagan, 2011.

[3] Cullity, 2000.

[4] Nefsky, ms. Regan, 1980 arguably also fits in this category.

[5] Nefsky, 2011; Budolfson, forthcoming.

[6] Nefsky, 2015.

[7] I have pressed this criticism in other work.

[8] For some of the debate about how to precisely formulate the perfect/imperfect distinction, see Rainbolt, 2000; Statman, 1996; Sinnott-Armstrong, 2005.

[9] Hill, 1971, p63.

[10] Quinn, 1990.

  • Facebook
  • Twitter
  • Reddit

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use the <em>, <strong> and <blockquote> tags. Links have been disabled to combat spam.

Notify me of followup comments via e-mail. You can also subscribe without commenting.


Subscribe Via Email