Some moral principles require agents to do more than their fair share of a common task, if others won't do their fair share – each agent's fair share being what she would be required to do if all contributed as they should. (Act-) consequentialist principles of beneficence provide one important example. Such principles require agents to promote the good as much as they can, roughly speaking. If some agents do not act on such a principle, then my acting on it will involve doing more than my fair share of good-promoting – my fair share being how much good-promoting I would have to do if everyone promoted the good as much as they could. And this seems to provide a strong basis for objecting to such principles. For it seems unfair that I should be required to do more than my fair share of good-promoting, just because others won't do their fair share.
This is an important part of moral theory, but surprisingly little progress has been made with it. One symptom of this is that there is not yet agreement even on what kind of objection it is. It seems natural to say that it is unfair to require individuals in such circumstances to do more than their fair share, as I have just done, and others have done.Footnote 1 The philosopher who has written most about this issue, however, Liam Murphy, argues that it is not unfair to do so, at least in the standard sense of that term. In this article, I give Murphy's reasons for saying this, explain why I think he's wrong, and then say a little about why this might be important.
Murphy focuses on a standard consequentialist principle, which he calls the ‘optimizing principle of beneficence’. This principle requires agents to promote the good as much as they can, roughly speaking.Footnote 2 In situations of partial compliance with this principle – situations in which some agents do not act on this principle – my acting on it will involve doing more than my fair share of good-promoting, if one takes my fair share of good-promoting to be how much good-promoting I would have to do if everyone promoted the good as much as they could.Footnote 3 Is it unfair to require me to do more than my fair share in this sense? Murphy writes:
under partial compliance with the optimizing principle of beneficence, each agent is required to fulfill what would be his own fairly allocated responsibility under full compliance plus (speaking roughly) some of the responsibility that properly belongs to noncomplying agents. Since each agent faces exactly the same requirement – and in substance, not just in form – no standard issue of fairness can arise . . .Footnote 4
And he goes on (pp. 90–1):
the principle itself would be unfair, in the standard sense, only if, for example, some but not all people were required to make sacrifices that would leave them in a worse position than they would be under full compliance. The idea of fairness, in its standard or central sense, is comparative or distributional; and the optimizing principle of beneficence, since it treats all agents substantively the same way under partial compliance, cannot be said to impose responsibility under partial compliance in a way that is unfair in that central sense.Footnote 5
One can see Murphy's point. Certainly, one standard way in which a principle can be unfair is by treating some agents differently from others in a context in which fairness requires that they be treated in the same way.Footnote 6 And the optimizing principle of beneficence is not unfair in that sense.Footnote 7 Acknowledgement of this point hardly removes the sense that it is unfair to require agents to do more than their fair share, though. So what is going on?Footnote 8
The first thing to note is that in order for a principle to be fair, it is not sufficient for it to tell each relevant agent to do the same thing (in a context in which it is appropriate to treat all agents in the same way).Footnote 9 We also need to take into account what it tells each agent to do. This might be easiest to see if we begin with another kind of case. Imagine that a prisoner, Jake, agrees to do extra work for a month in return for extra rations. He keeps his side of the bargain, but at the end of the month the prison warden tells Jake that he has changed his mind, and that Jake will now have to do extra work for another month in order to get the extra rations. Jake objects that it is unfair for the warden to change the terms of the agreement as he goes along in this way. The warden replies, ‘But I treat every other prisoner in the same way. And so whatever objection you might have to my behaviour, you can't really claim that I'm treating you unfairly – at least not in the standard sense of that term.’
Would this reply satisfy Jake – or us? I take it as uncontroversial that it would not. Changing the terms of an agreement in the kind of way the warden does is a paradigm case of unfair behaviour. And this is because it contravenes a material principle of fairness: that agreements are to be honoured.Footnote 10 Because it contravenes this principle, the warden's behaviour toward Jake is unfair. And because it is unfair for the warden to treat Jake in this way, doing so remains unfair even if the warden treats every prisoner (substantively) the same way. As Joel Feinberg puts it in a related context, ‘If we treat everybody unfairly, but equally and impartially so, we have done each an injustice that is, at best, only mitigated by the equal injustice done all the others.’Footnote 11 If Murphy were right, the very idea of treating everyone unfairly would not make sense (at least in the standard sense of the term).
Similar considerations apply to the optimizing principle, and to other principles that require agents to do more than their fair share of some common task, when some won't do their fair share. (Call such principles ‘more-than-fair-share principles’.) Such principles also contravene a material principle of fairness: that the burdens of contributing to a common task are to be distributed equitably amongst all those who should contribute to that task – all those who should contribute, not only those who do contribute.Footnote 12 Such a distribution assigns each individual their fair share. Principles that require agents to do more than that contravene this principle. And consequently, such principles are unfair, even when they treat all agents (substantively) the same way.Footnote 13
In general, then, principles and practices that infringe material principles of fairness are unfair, even if they do not infringe any formal principles of fairness – even if, for example, they treat all relevantly similar agents the same way. And they are unfair in a perfectly standard sense of the term. This is evidenced at the intuitive level by the fact that it is perfectly natural to call such principles and practices unfair. And at the theoretical level, there is simply no good reason for saying that principles and practices that infringe material principles of fairness can only be unfair in a sense of the term that is non-standard.
We can, then, vindicate the intuition that it is unfair, in a perfectly standard sense of that term, to require agents to do more than their fair share when other agents won't do their fair share. And we can also state clearly why it is unfair to do so – because doing so contravenes the standard and uncontroversial material principle of fairness outlined above. This doesn't answer all our questions about this objection to such principles, of course, but it does at least enable us to put it in the right normative category. And as I shall now show, it may also help us to put the objection in its most plausible form.Footnote 14
It is striking in this context that Murphy argues for an absolute constraint against requiring agents to do more than their fair share of beneficence.Footnote 15 This position is highly counter-intuitive, because of cases in which the cost of doing more than one's fair share is trivial, and the benefits to others are great. (Call such cases ‘low-cost high-benefit cases’.) In one standard example, two passers-by (Jane and John) come upon two children drowning. Saving each would involve paying some small cost. Jane and John are both morally required to save one child, on the assumption that the other will also save one. If John refuses to do so, though, what is Jane morally required to do? If there is an absolute constraint against requiring agents to do more than their fair share of beneficence, and saving the second child would mean that Jane does more than her fair share, then we would have to say that she is not required to save the second child. And this is indeed what Murphy says.Footnote 16 This is highly counter-intuitive, though. Indeed, it is so counter-intuitive that such cases apparently lead many to assume that the objection to more-than-fair-share principles that we have been discussing (call it the ‘Fair Shares Objection’) has no role at all in determining our obligations of beneficence.
Thinking of the objection in the kind of way I have suggested, as being based in a standard way on considerations of fairness, suggests a different picture, and one that enables us to give the intuitively right answer in low-cost high-benefit cases without implying that that objection has no relevance at all to our obligations of beneficence. Two related features of such considerations are particularly relevant here: first, that objections to principles based on considerations of fairness can be more or less strong, depending on various factors including how great the consequences of such unfairness are; and second, that (in part for this reason) considerations of fairness are naturally and standardly taken to be decisive in some contexts and not in others, depending on how strong the objection based on such considerations is in the context in question, and on what other considerations are in play. Given these features, it would be very odd to argue for an absolute constraint against principles requiring agents to do more than their fair share, if one thinks of the objection to such principles as being based on considerations of fairness. For to do so would be to take such considerations to have an invariant and overriding force, no matter how weak or strong those considerations are in the particular context in question, and how weak or strong any countervailing considerations are. It would be much more natural to take that objection to be scalar in its force – weaker or stronger in different contexts – and in part for that reason as being decisive on some occasions and not on others, depending inter alia on how strong the objection is in the context in question, and what other considerations are in play.
Such an interpretation, I suggest, is far superior to the constraint model. On the one side, it enables us to give the intuitively right answer in the kind of case sketched above. For in that case, the cost to Jane of rescuing the second child is small, and therefore it is natural to interpret the objection to her doing so (on the picture just sketched) as being very weak. And if the objection to her doing so is very weak, then it is presumably easily overridden by the fact that a life is at stake. At the same time, though, this needn't commit us to the view that the Fair Shares Objection lacks any role at all in determining our obligations of beneficence. For that objection will be stronger when (what one might call) the more-than-fair-share costs (the costs beyond those that a principle requiring one to do one's fair share would impose) are higher. And for this reason, that objection may yet be decisive when the costs are high enough, even if the benefits to others are still great.Footnote 17 On this alternative model, then, we can give the intuitively right answer in low-cost high-benefit cases without ruling out the possibility that the Fair Shares Objection is decisive in other cases. And that, I suggest, makes the objection harder to dismiss.
I will not pursue these issues further here.Footnote 18 For now, my point is just that this alternative and (I suggest) superior understanding of the Fair Shares Objection is the natural one if one thinks of that objection as being based straightforwardly on (standard) considerations of fairness. And for this reason, it may in fact make a big difference whether or not one thinks of it in this way.
