1 Fairness and Non-Compliance
In certain kinds of situations, moral obligations are very naturally and plausibly
2understood in the first person plural – as owed by us. For example, if my spouse and I
together promise our son to take him to his friend‟s house on Monday afternoons then
absent some special circumstances we owe our son a weekly trip to his friend‟s house on Mondays. Equally plausible is the idea that when obligations are held in common in this way the burdens of meeting those obligations should be fairly divided. Taking our son to his friend‟s house might well be burdensome for both of us, forcing each of us to leave work early, etc. All else being equal, it would be unfair in such circumstances for one spouse always to be forced to take the son to his friend‟s house. Of course, all else is
often not equal, and the fact that one spouse usually discharges some other comparably burdensome and jointly held obligation might make this fair after all. The main point, however, is that when obligations are held in common and discharging those obligations is burdensome those burdens ideally ought to be divided fairly. Just what constitutes a fair division of such burdens in the ideal case in which everyone does their fair share is a good question in roughly what John Rawls famously called “ideal theory,” the relevant
3idealization being that everyone does their fair share. For purposes of the present paper I
shall remain neutral about how fairness itself is best characterized and hence about what a fair division in the ideal case would be. Some philosophers have argued that fairness should not really be understood as a substantive value which is fundamentally distinct from the values of consistent and impartial application of rules, fidelity and
1 Note removed to preserve anonymity. 2 For further discussion of the importance of the first-person plural for moral deliberation, see Postema 1995.
proportionality (see especially Hooker 2000: 45-55). In what follows I assume that fairness can matter morally in its own right, but this may be compatible with taking fairness in these contexts to consist primarily in proportionality (of welfare to required sacrifice, e.g.). So my argument may even be compatible with these reductionist accounts of fairness, though I shall not explore that question here. Rather than rely on a particular conception of fairness and its importance, I instead rely on what I take to be commonly held moral intuitions about which outcomes would be fair, but those intuitions are compatible with a wide range of background theories about the nature of fairness.
Sadly, on any plausible conception of fairness people all too often do not do their fair share. The non-compliance of others can leave the conscientious agent with a difficult and practically pressing moral question - what should I do when some of those with whom I jointly hold a moral obligation fail to do their fair share in discharging that obligation? The question is of quite general interest, but is especially urgent in the case of the jointly held obligation of those in the affluent West to provide much-needed and highly cost-effective assistance to those in less fortunate circumstances suffering in what
4 In MacNamara‟s terms, such Robert MacNamara characterizes as, “absolute poverty.”
people lead lives, “so characterized by malnutrition, illiteracy, disease, squalid surroundings, high infant mortality and low life expectancy as to be beneath any reasonable definition of human decency.” (MacNamara 1978: iii) When hundreds of millions of people must live on less than the equivalent of $1.00 per day and unsurprisingly lead lives that are nasty, brutish and short, it is very plausible to suppose that those of us in the industrialized West who are grotesquely wealthy by global
3 See Rawls 1971. 4 See McNamara 1976: 14 and Singer 1979: chapter 8.
standards owe it to those so much worse off to help them. The philosophical case for this conclusion has been made at length, and I shall not repeat the main moves of that
5dialectic here. The point here is that this obligation is very plausibly understood as one which is jointly held. Ideally, the burdens of fulfilling this obligation would be divided fairly, with very wealthy people like Bill Gates devoting a higher percentage of his income and wealth to meeting the obligation than an average middle class citizen of the United Kingdom, for example. In ideal conditions in which every one did their fair share to discharge this duty of beneficence, the burden of compliance on any one person certainly would not require any one person to give virtually everything to charity, and on some accounts the burdens may be quite small. However, our actual conditions are in this sense far from ideal. Most people in affluent nations give nothing or next to nothing to help those in absolute poverty, and hence fall far short of doing their fair share to meet this jointly held obligation. This puts the problem of how an individual should act in conditions of partial compliance into very stark relief. On the one hand, philosophers like Peter Unger argue that in these circumstances the morally decent agent must give away virtually everything to help those in such great need, for their needs are so much greater and more urgent than our own. Unger makes an impressive case for this conclusion, emphasizing the fact that a very small donation to the right charity would have a very good chance of preventing a premature death. Moreover, as Peter Singer, Peter Unger and others have argued the best explanation of our the soundness of certain strongly held intuitive judgements about other saving lives in other cases (like the hackneyed case of a small child drowning in a shallow pond) seem to point inexorably to the conclusion that the demands of beneficence in our actual circumstances are very high. Finally, Unger
5 See, for example, Singer 1979: chapter 8 and Unger 1996.
provides an interesting debunking explanation of why people find this conclusion so hard to accept; if that explanation is correct then our resistance to the conclusion that the demands of beneficence are very great indeed are not well founded. As compelling as these arguments are, they seem simply to ignore the issue of fairness which seems so
6obviously relevant when we are thinking about ideal theory. This can seem very
implausible. If how much I should do in circumstances of full compliance is a function of what would be my fair share, then surely considerations of fairness should play some role in determining my obligations in conditions of partial compliance. Why, the conscientious agent might well ask, should I be expected to pick up the enormous slack left by the non-compliance of others? Surely, the most one could reasonably demand is that I do my fair share in discharging this obligation. If others fail to do their fair share then they, and not I, am to blame for that. Doing their fair share as well as my own might be morally virtuous but it can hardly be my duty. These intuitions are powerful, and Liam Murphy has recently defended a set of moral principles which are meant to vindicate those intuitions. Very roughly, Murphy argues that in conditions of partial compliance a person is never morally required to sacrifice so much that he would thereby end up less well-off than he would have been under full compliance from then
6 Of course, Singer and Unger could acknowledge the value of fairness and argue that the demands of beneficence trump considerations of fairness in some way, and some of their examples might well be deployed to bolster this idea. Still, so far as I know they do not confront the issue of unfair demands directly. Furthermore, even if this trumping suggestion is right (and I do not at this stage mean to suggest it is), it remains important to get the overall map of the overall moral terrain correct, including even moral reasons which are trumped by other considerations in the particular case of duties of beneficence to those in absolute poverty. Moreover, in other situations in which the stakes are not nearly so high as in the case of absolute poverty it will not be nearly so plausible to suppose that other reasons trump considerations of fairness, which gives us a more general reason to become clear about how fairness functions in these contexts. Thanks very much to Keith Horton for prompting me to discuss this point in more detail.
7 Murphy calls his principle the “collective principle,” since it emphasise the onwards.
idea that our obligations are held in common, and I shall here follow his terminology. The suggestion that we should think about obligations in conditions of partial compliance in terms of fairness is very plausible, and Murphy is right to emphasise the philosophical importance of this often neglected idea.
However, the collective principle‟s interpretation of the role of fairness itself seems problematic. For just as a conscientious agent in the affluent West can reasonably complain about the unfairness of being forced to do more than his fair share, those living in absolute poverty can reasonably complain that it is unfair for them to be forced to bear the entire brunt of the non-compliance of those who do not give their fair share. Why, after all, should those who are already so much worse off be forced to bear all of the negative consequences of the non-compliance of those who do not do their fair share when others who are much better off could in effect absorb some of that burden by giving more? While the moral principles advocated by Singer and Unger seem to ignore the unfairness imposed on the wealthy who aim to do the right thing when others fail to comply, the collective principle seems to ignore the unfair burdens imposed on those
8living in absolute poverty.
The moral seems to be that in conditions of partial compliance unfair burdens may simply be inevitable. What is not inevitable is any particular distribution of those unfair burdens. Unfairly imposed burdens can in general themselves be more or less
7 See Murphy 2000. The characterization given in the text roughly paraphrases Murphy at Murphy 2000: 87. Murphy takes Parfit‟s discussion of these issues in Reasons and Persons as his starting point, though
Parfit himself does not endorse Murphy‟s conclusion. See Parfit 1984. Thanks to Brad Hooker for
helpfully reminding me of the etiology of Murphy‟s views in Parfit‟s work. 8 Keith Horton, in an unpublished manuscript, very helpfully refers to this as the „tu quoque‟ reply to Murphy. The basic point against Murphy is also made in passing in Hurley 2003.
fairly distributed. For example, a cruel teacher may unfairly single out two particular children to always bear the burden of cleaning the blackboard while the other children head off to recess. This burden is unfairly imposed in that these two children have done nothing to warrant singling them out in this way. It does not follow, however, that this unfair burden cannot be more or less fairly distributed between them. All else being equal, if one child regularly did 90% of the work while the other child is idly daydreaming then this would be a paradigmatically unfair division of the unfairly imposed burdens shared by the two children. My suggestion here is that this is how we should approach the issue of fairness in conditions of non-compliance with jointly held moral obligations. I agree with Murphy that fairness should supply an upper limit on an individual‟s obligation in conditions of partial compliance, but that limit should not
correspond to the burden the agent would be forced to bear under conditions of full compliance. Instead, fairness dictates that in conditions of partial compliance no agent can be required to do so much to make up for the non-compliance of others that he thereby bears more than his fair share of the burdens of non-compliance, or so I shall argue. The approach advocated here is a kind of compromise between unconstrained beneficence (Singer and Unger) beneficence as constrained in conditions of partial compliance by fair shares under full compliance (Murphy). Like Murphy‟s account and unlike Singer‟s and Unger‟s account, the account offered here does give weight to the idea of unfairness. However, like Singer‟s and Unger‟s account and unlike Murphy‟s
account, the account given here also insists that we may nonetheless sometimes have a moral duty to pick up at least some of the slack of those who do not comply. To put the point another way, on Singer‟s and Unger‟s account the perspective of those in absolute
poverty seems to be completely dominant in determining our duties in conditions of non-compliance, while on Murphy‟s account the perspective of the affluent seems much more
9dominant. The account developed here attempts to give due weight to both of these perspectives. Ultimately, our actual moral obligations in the case of assistance to those in absolute poverty may turn out on the account offered here to be much closer to the extreme demands advocated by Unger and Singer than the much more moderate demands advocated by Murphy, though I shall not try to settle this difficult (and in large part empirical) issue here. Nonetheless, the account on offer is in an important sense a compromise between the two positions, and the accounts will diverge in their recommendations in a wide range of cases. Finally, I should emphasise that the account developed here extends well beyond the particular case of helping those in absolute poverty. The account on offer addresses the more general issue of fairness-based limits of our duties in cases of partial compliance with jointly held moral obligations, and that issue can arise in a very wide range of contexts. I have chosen to focus on the case of our obligations to those in absolute poverty simply because that case is particularly morally important.
The idea that we can never be morally obligated to do more than our fair share in discharging our collectively held moral duties has a lot of intuitive support. To reject this idea seems to put the morally decent at the mercy of the morally lazy and irresponsible, since their non-compliance forces the rest of us to make greater sacrifices. Furthermore,
9 Murphy himself no doubt would not characterize his position in such stark terms, and presumably would emphasize the importance of recognizing the agency of others. Nonetheless, this perspective does seem to give greater weight to considerations which do in fact work to the advantage of the affluent and which presumably would naturally seem more salient to the affluent for that reason.
the idea that we are never obligated to do more than our fair share in such cases is compatible with the idea that it would be morally virtuous or supererogatory to do more, so it need not reject the intuition that someone who does more than his fair share may thereby exhibit moral virtue. In spite of its initial plausibility, though, this idea cannot be sustained. Before explaining in more detail why this idea is indefensible, it is important to be clear about the dialectic. The suggestion is not that considerations of fairness provide moral reasons for us not to do more than our fair share in discharging a collectively held moral duty. The idea is rather that considerations of fairness limit our other obligations by somehow functioning to transform what would otherwise be a moral reason underwriting a moral obligation into a moral reason which merely underwrites supererogation. In the example of helping those in absolute poverty, the moral reason to give more is that it would help those in grave need – a reason of beneficence. The fact
that giving more would constitute my giving more than my giving more than my fair share is meant not to entirely defeat that moral reason of beneficence (so that it no longer functions as a reason at all), but rather to transform it into the sort of reason which generates supererogation rather than moral obligation. Exactly how we should understand the distinction between reasons which ground obligations and reasons which ground supererogation is itself a very difficult topic which for present purposes must be set to one side. The main point is that the issue here is whether considerations of fairness can transform moral reasons in this way in conditions of non-compliance. Different moral theories will provide different accounts of how fairness might function in this way.
One theory (though not the only one) which makes this potential role of fairness easily intelligible is moral contractualism of the sort defended by T. M. Scanlon. On
Scanlon‟s account, roughly, an action is morally required if and only if it would be required by principles for the general regulation of behaviour which nobody could reasonably reject. Scanlon allows that moral considerations like fairness can themselves provide reasonable grounds for rejection. Within the contractualist framework, we might interpret the proposed role of fairness as providing those who would have to do more than their fair share with grounds on which they could reasonably reject any moral principle requiring to do more – the fact that such principles would themselves be unfair
to them would provide the basis for reasonably rejecting such a principle. On this account considerations of fairness provide a moral reason after all, but not a moral reason not to do more than one‟s fair share. Rather, considerations of fairness provide a moral
reason to reject a principle for the general regulation of human behaviour which would require one to do more. The arguments developed here certainly shall not presuppose Scanlon‟s contractualism. Nonetheless, the contractualist framework provides a very
helpful framework within which to present some of the main ideas, so I shall at some points return to that framework for expository purposes.
When others default on their fair share of a collectively held moral obligation, some unfair burdens are inevitable, except in the unlikely event that others doing more than their fair share just happens to be no burden at all. For barring this unlikely coincidence of duty and self-interest, either one of two forms of unfairness will result. First, some of the other people who hold this collectively owed moral duty might pick up the slack of those who default. As Murphy‟s discussion emphasises, this is unfair to those who are forced to pick up the slack. Second, though, if others do not pick up the slack of those who default on their fair shares then a burden will be imposed on those to
whom the collective duty is owed, and it is hard to resist the conclusion that this burden is unfairly imposed. In the case of our duty to help those in absolute poverty, the failure of others to do their fair share in conjunction with the failure of others to pick up the slack of their non-compliance in effect means that those in absolute poverty must bear the unfair burden of misery, poverty and malnutrition which ought to have been alleviated. In conditions in which some unfairness is inevitable no matter what we do, all else being equal we should aim for the lesser of the evils. How, though, might we determine the lesser of the evils in these kinds of cases?
The relevant evil is unfairly imposed burdens, and this evil is inevitable because of the non-compliance of others. Of course, we presumably should do our best to encourage or even in some cases force those who do not comply to do so, but in the relevant cases this is either impossible, unduly costly or objectionable on other moral grounds. When non-compliance is inevitable, we are faced with the problem that an unfair burden inevitably will be imposed upon someone – either those to whom the duty
is owed or those who pick up the slack of those who default (the „or‟ is inclusive). Our question, then, is how we should deal with a situation in which the imposition of unfair burdens is inevitable. One of the examples briefly discussed in the introduction provides a clue as to how we should deal with such situations. Recall the case of the teacher who unfairly singles out two children from the rest of the class to clean the blackboard. These children have by hypothesis done nothing to deserve this burden, but there is no feasible way for them to avoid it either; parental intervention, we may assume, is unavailable to them because their parents trust the teacher‟s disingenuous rationale for their punishment. The case is one in which the non-compliance of one party (the teacher) makes the