Hostname: page-component-74656f9884-8rzl6 Total loading time: 0 Render date: 2026-11-06T05:14:59.631Z Has data issue: false hasContentIssue false
  • English
  • Français

In Defence of the Priority View

Published online by Cambridge University Press:  31 August 2012

THOMAS PORTER
THOMAS PORTER*
Affiliation:
University of Manchestertom.porter@manchester.ac.uk
Rights & Permissions [Opens in a new window]

Abstract

In their paper ‘Why It Matters That Some Are Worse Off Than Others: An Argument against the Priority View’, Michael Otsuka and Alex Voorhoeve argue that prioritarianism is mistaken. I argue that their case against prioritarianism has much weaker foundations than it might at first seem. Their key argument is based on the claim that prioritarianism ignores the fact of the ‘separateness of persons’. However, prioritarianism, far from ignoring that fact, is a plausible response to it. It may be that prioritarianism disregards the fact of the ‘unity of the individual’. But even if this is true, that doesn't straightforwardly tell against prioritarianism as a view about distributive justice. In the end, Otsuka and Voorhoeve's argument relies on a non-decisive intuition that they appeal to early in their paper. Their conclusion, as a result, is not compelling.

Information

Type
Research Article
Information
Utilitas , Volume 24 , Special Issue 3: Prioritarianism , September 2012 , pp. 349 - 364
Copyright
Copyright © Cambridge University Press 2012

I. INTRODUCTION

In their paper ‘Why It Matters That Some Are Worse Off Than Others: An Argument against the Priority View’,Footnote 1 Michael Otsuka and Alex Voorhoeve argue that prioritarianism is mistaken.Footnote 2 I shall argue that their case against prioritarianism turns out to have much weaker foundations than it might at first seem. Their argument, as a result, is not decisive.

Prioritarianism is a view about distributive justice according to which benefiting a person matters more the worse off she is. Distinctively, the prioritarians in Otsuka and Voorhoeve's sights understand ‘worse off’ here in absolute terms: being worse off is a matter simply of having less than one might have, and not, in particular, a matter of having less than others. According to the prioritarians in question, then, for the purposes of judging the moral value of distributing benefits to you, how well off anyone else is makes no difference to how well off you are.

Otsuka and Voorhoeve make their case in three parts. The first part is an argument to the effect that prioritarianism yields the wrong judgement in situations involving the distribution of benefits to a single individual considered in isolation. The second part is an argument – seemingly independent of the first – to the effect that prioritarianism fails to change its strategy appropriately when we move from such one-person cases to cases involving the distribution of benefits among more than one individual. The third part comprises defences against four potential objections.

In what follows I first analyse the first two parts, raising objections to them that are not discussed by Otsuka and Voorhoeve in the third. With respect to the argument of the first part, I claim in section II that the intuition upon which Otsuka and Voorhoeve rely is not decisive. It is far from intuitively clear that prioritarianism gives the wrong answer in the one-person cases that they describe.

I then turn, in sections III and IV, to the argument of the second part of their paper. Here I argue that to the extent that Otsuka and Voorhoeve have an independent argument, based on the ‘separateness of persons’, against prioritarianism, they are unsuccessful. Prioritarianism, far from ignoring the separateness of persons, is a plausible response to that fact. It may be that prioritarianism fails to take into account justifications that are made available in one-person cases by the ‘fact of the unity of the individual’. In section V I argue that even if this is true, that doesn't straightforwardly tell against prioritarianism as a view about distributive justice. Too much still relies on the same non-decisive intuition that underpins their first argument.

II. THE FIRST ARGUMENT

The first argument in ‘Why it Matters’ begins with an example. We are presented with a one-person case in which the person in question has a 50 per cent chance of developing a very severe impairment (which leaves her unable even to sit up for much of the time) and a 50 per cent chance of developing a slight impairment (which limits the distance that she can walk without difficulty). She can receive only one of two treatments, and whichever she receives must be taken before either of these possibilities is realized. She is indifferent between the two treatments, so that on a standard conception of individual utility, the two treatments are of equal expected utilitarian value. The first treatment has an effect just in case she would otherwise develop the very severe impairment: it moves her from that very severe impairment to a merely severe impairment (now she will be able to sit up, although she will still need others’ assistance to move about). The second treatment has an effect just in case she would otherwise develop the slight impairment: it moves her thence to full health. In regard to this case, Otsuka and Voorhoeve claim that ‘it would be reasonable for you [a morally motivated stranger, as in all the cases that I'll discuss] to share her indifference’ between the two treatments, at least so long as we assume that ‘you are considering her fate in isolation from any consideration of how well off or badly off anybody else is (yourself included)’.Footnote 3

Prioritarians wouldn't agree. According to prioritarians, a given amount of utility has diminishing marginal moral importance, so that it's more morally valuable for someone with the very severe impairment to receive that amount than it is for someone with the slight impairment to receive it. Since the two treatments under consideration are, by hypothesis, of equal utility value, even the smallest degree of prioritarian weighting is enough to tip the balance in favour of the treatment for the very severe impairment. The treatment for the very severe impairment therefore has the highest expected moral value. So, you should, according to prioritarians, give the person that treatment.

To bring out further the contrast between their own and the prioritarian approaches, Otsuka and Voorhoeve point out that there will also be one-person cases in which ‘someone who holds the Priority View would conclude that he has decisive moral reason to supply . . . the treatment for the very severe impairment rather than the slight impairment even [when] the person would rationally prefer the treatment for the latter’ and without denying that the recommended treatment will have a lower expected utility value. Such cases arise when the greater utility of the treatment for the slight impairment is not sufficient for that treatment's moral value, on prioritarian weightings, to outweigh that of the treatment for the very severe impairment. They conclude that ‘The Priority View therefore unreasonably mandates provision of a treatment with a lower expected utility for the person concerned.’Footnote 4

Next, Otsuka and Voorhoeve present a case involving more than one person, but which is otherwise relevantly analogous. Prioritarians remain committed to the treatment for the severe impairment when presented with this multi-person case. The case is as follows. You are faced with a group of people, half of whom will develop the very severe impairment and the other half of whom will develop the slight impairment, and all of whom have the same preferences as the single individual in the first example. Otsuka and Voorhoeve's considered judgement about this case is that you should not be indifferent between administering to each person the treatment for the slight impairment and administering to each person the treatment for the very severe impairment. You should administer the latter. The prioritarian judgement in this case accords with both Otsuka and Voorhoeve's own considered judgement and the judgements of individuals surveyed in research that they cite.Footnote 5 All sides agree that morally motivated strangers should not be guided solely by the expected utility of the treatments (and so indifferent between them), but should provide the treatment for the very severe impairment.

In the multi-person case, then, prioritarians’ judgements accord with Otsuka and Voorhoeve's intuitions, which are also widely shared. Their judgements conflict, on the other hand, with Otsuka and Voorhoeve's intuitions about the one-person case. Otsuka and Voorhoeve cite no survey data at this point to show that these intuitions are also widely shared.Footnote 6 So, the argument should not be construed as an appeal to widespread intuition.

What, then, grounds their assertion, in the face of prioritarian denials, that it would be reasonable for you to share the indifference of the individual in question between the two treatments? (Or, in the variation of the one-person case, that it would be unreasonable for you give the individual the treatment that gives her the lower expected utility, as the prioritarian recommends?) They explicitly deny that this has anything to do with respect for individuals’ autonomy over matters that affect only themselves.Footnote 7 Presumably, then, the idea is that when we, the readers, are confronted with the one-person cases we shall find that we too share Otsuka and Voorhoeve's intuitions about them, and oppose, therefore, the prioritarian denial that the appropriate response is to be guided by the individual's expected utility alone.

It seems to me, however, that it's not at all counterintuitive to suppose that you ought to supply the treatment for the very severe impairment. This would be to disregard the person's own indifference between the two treatments or even her slight preference for the other treatment, indeed, but it would also be to save her from the possibility of a life spent ‘bedridden, save for the fact that [she] will be able to sit in a chair and be moved around in a wheelchair for part of the day if assisted by others’.Footnote 8 In these cases, in which we are to consider the person's situation in isolation from how well off she is relative to others, the most natural explanation for this judgement would be that although the treatment that it discommends is, from the point of view of the person involved, of equal or even of greater benefit, it's nevertheless simply more important to benefit her if she develops the more severe symptoms than it is to benefit her if she develops the milder ones – precisely because those more severe symptoms are more severe. This, of course, is also how prioritarians arrive at their judgement.

If Otsuka and Voorhoeve's intuitions about this case are not widely shared in the same way that their intuitions about the multi-person case are, and if prioritarians can offer a plausible account of the reasoning that their countervailing intuitions reflect, and if we can't identify any irrelevant distorting factors which might give us reason to discount intuitions on one side or the other, then it's not clear that prioritarians should feel under any pressure to renounce their prioritarianism about one-person cases.Footnote 9 And these antecedents appear to hold, for all that we've seen so far.

If I am right about this, then no clear case against the prioritarian approach arises from consideration of one-person cases. I conclude that Otsuka and Voorhoeve's first argument does not constitute grounds to reject prioritarianism after all. They have, however, a second argument to offer in response to this line of objection.

III. THE SECOND ARGUMENT

Otsuka and Voorhoeve begin their second argument with an arguendo concession. The concession is that prioritarians may be right to judge it a mistake to be guided solely by expected utility in the one-person case. They do not, however, contemplate conceding that this is because a morally motivated stranger ought to apply the same weighting criteria in one-person cases and multi-person cases alike. The concession is, rather, that prioritarians may be right to give ‘some extra weight to increases lower down the utility scale’,Footnote 10 so that in the one-person case the decision to treat is the right one.

But even if prioritarians are right about this, Otsuka and Voorhoeve go on to argue, it's still a mistake to apply the same weighting criteria in both types of cases. We are asked to imagine a one-person case in which the person in question faces a 50 per cent chance of a utility gain and a 50 per cent chance of a utility loss unless you intervene. If you intervene, the person will face no prospect of either gain or loss. Otsuka and Voorhoeve invite the prioritarian to specify the relevant utility gains and losses so that by her prioritarian lights the expected moral value of not intervening is just enough to outweigh the expected moral value of intervening. Then they ask the prioritarian to imagine an analogous two-person case in which – unless you intervene – there is a 50 per cent possibility that the first person will receive the gain in utility and the second person will suffer the loss in utility (these gains and losses being as they are in the one-person case). If you intervene, or if you don't intervene but the possibility just described isn't realized, neither will gain or lose. Otsuka and Voorhoeve suggest that the prioritarian ought to give greater prioritarian weight to losses in this case than she did in the one-person case. Since in the one-person case the moral value of non-intervention only just outweighed the moral value of intervention, this ‘shift’ in weighting will suffice to make intervention the appropriate course of action in the multi-person case.

But why ought there to be this ‘shift’? Otsuka and Voorhoeve explain as follows:

In the [two-person] case, there is no single person for whom the prospect of a greater gain is the flip side of exposure to the risk of a lesser loss and for whom the prospect of such gain might be worth exposure to such risk. Rather, if you do not intervene, there are two people: a first person who would face just a prospect of a gain, and who would, if this gain materializes, be better off than the second person, and a second person who would face just a risk of a loss, and who would, if this loss materializes, become worse off than the first person. It follows that rather than simply deciding whether the potential gain outweighs the potential loss to the same person [as in one-person cases], in a two-person case you must now decide whether the potential gain to the first person outweighs the potential loss to the second person, who would, if this loss materializes, be worse off than the first person. These differences between the one-person and the two-person case imbue the potential loss to a person with greater negative moral significance in the two-person case. You should therefore intervene in a two-person case to prevent the second person from facing the risk of loss, thereby also eliminating the first person's prospect of gain, even though this prospective gain is, by hypothesis, just large enough relative to the potential loss to justify refraining from intervention and letting the chips fall where they may in analogous one-person cases.Footnote 11

The first three sentences of this passage articulate claims that are obviously and uncontroversially true. But the claim in the fourth sentence and the conclusion that Otsuka and Voorhoeve draw in the fifth are not obviously or uncontroversially true, and it is open to the prioritarian to deny them.

Let me explain. It's true that in multi-person cases, prioritarians must decide whether gains to some outweigh losses to others. (As must proponents of any view about distributive justice.) Prioritarianism is typically formulated by reference to multi-person cases precisely in order to give an answer to this question. It offers a distinctive answer to the question when gains to some outweigh losses to others – an answer which gives voice, in one way, to the common supposition that it matters more, morally speaking, to benefit individuals the worse off they are. Unlike a view – such as the difference principle, on one reading – that gives absolute priority to benefits to the worst off, prioritarianism holds that sometimes losses for the worse off can be justified by appeal to the extent of the gains for the better off. But unlike utilitarians, who give no priority at all to benefits for the worst off, prioritarians think that it's not sufficient for such an appeal to succeed that the gains for the better off are greater than the losses for the worse off.

That, then, is the familiar prioritarian approach to justifying the potential losses for those who do worse in the kind of multi-person case that Otsuka and Voorhoeve describe. Now, the latter's argument is that this approach should be conceived as a response to the fact that the prospect of sacrifices by a worse off person that is correlated with the prospect of gains for the better off, in the relevant situations, is imbued with a ‘greater negative moral significance’ when the worse off and the better off are different people than when they are the same person. But prioritarians may simply deny this, insisting that their approach in multi-person cases is in fact simply a response to the fact that benefits to an individual who is worse off, in absolute terms, matter morally more than benefits to an individual (be she the same or a different individual) who is better off. And so far, Otsuka and Voorhoeve have offered no argument to the contrary. They have merely offered an alternative possible explanation that prioritarians are at liberty to eschew.

For all that we have seen, then, the force of Otsuka and Voorhoeve's argument derives from the assumption that everyone's judgements will undergo the same shift from the one-person case to the multi-person case. But prioritarians may deny that their judgements undergo this shift. And from a moral point of view, they may say, regardless of the desires of the individual involved, the potential losses and gains in a one-person case have just as great a significance as the potential losses and gains in the two-person case. The conclusion that Otsuka and Voorhoeve draw is not licensed. So, this argument moves us no further along than the first did.

IV. THE SEPARATENESS OF PERSONS

At the end of their presentation of the second argument, Otsuka and Voorhoeve say of the shift that they take to be justified when we move from one-person to multi-person cases that:

The Priority View . . . cannot countenance any such shift. Given that the separateness of persons renders such a shift appropriate, it follows that the Priority View is not adequately responsive to this morally significant fact.Footnote 12

It's true that the Priority View cannot countenance any such shift. But, as I've argued, the case hasn't yet been made that the shift is appropriate. Asked to imagine the point at which, in a one-person case, it becomes appropriate to take the chance of losses for the sake of potential gains, a prioritarian can say that she is eo ipso imagining precisely the point at which it becomes appropriate to take the chance of one person's losses for the sake of another's potential gains in a multi-person case. In the final sentence of the passage just quoted, however, Otsuka and Voorhoeve make a suggestion that might seem to constitute an objection to that denial. The suggestion is that ‘the separateness of persons renders [the] shift appropriate’. Could this be a reason to reject prioritarianism?

The ‘separateness of persons’ objection is most famous as an objection to utilitarianism. The problem with utilitarianism, it is said, is that it treats interpersonal balancing of losses and gains in the way that a rational decision procedure for an individual life would treat intrapersonal balancing of losses and gains. But it's not appropriate to treat interpersonal balancing in this way, because of the fact of the separateness of persons.

As David Brink argues, however, if that fact implies that (the prospect of) losses to a person can be compensated only by (the prospect of) gains to that very same person, then the only separateness-of-persons-respecting distributive options are Pareto improvements.Footnote 13 Moves to Pareto-incomparable distributions are prohibited, since by definition they involve losses for some in exchange for gains for others. But this prohibition, as Brink points out, is an implausibly restrictive condition to impose upon an acceptable distributive theory.Footnote 14

The fact of the separateness of persons is better interpreted as expressing the idea that ‘it is unacceptable to impose unjustified losses or burdens on one person in order that others may benefit’.Footnote 15 Competing theories of distributive justice can then be seen as employing competing conceptions of unjustified sacrifice. Utilitarians are said not to take the fact of the separateness of persons seriously because their conception of an unjustified sacrifice is inadequate. Someone's sacrifice of utility is justified, according to utilitarians, if it produces a net increase in total utility. But this doesn't discriminate between units of utility being enjoyed by those who have many and units of utility being enjoyed by those who have few. The familiar counterintuitive consequence is that utilitarianism would count a very badly off person's sacrifice of one unit of utility as justified if the alternative was for a very well off person to sacrifice two units of utility.

Egalitarian and prioritarian theories both avoid this consequence. Each reviews the sacrifices that a given distribution implies, in light of available alternatives, for those in it. Each asks whether those sacrifices give individuals in the distribution a justified complaint.

Comparative egalitarian theories of the type that Otsuka and Voorhoeve favour count a sacrifice unjustified, at least from the purely egalitarian perspective, when an alternative Pareto-optimal distribution is available in which the degree of inequality is smaller – either because ‘badness inheres in the relational property of some being less well off than others’Footnote 16 or because ‘one must justify any claim on resources in light of the comparative strength of the claims of others’.Footnote 17 For comparative egalitarians, a person's complaint under a given distribution is greater to the extent that she is worse off relative to others in that distribution. They may then aim to minimize the greatest complaint, or they may instead aim to minimize the total weight of complaints.

Prioritarian theories, meanwhile, count a sacrifice of benefits unjustified if an alternative distribution is available which realizes prioritarian value to a greater degree, where the prioritarian value of a unit of benefit is greater the worse off in absolute terms its recipient is. For such theories, a person's complaint is greater to the extent that alternative available distributions would have made her better off in non-comparative terms. The aim is then to try to minimize the total weight of complaints.Footnote 18

In practice, there may not be much to choose between the second egalitarian strategy and the prioritarian strategy, although there will normally be some situations in which the differences become apparent. However that may be, both prioritarianism and the egalitarian theories favoured by Otsuka and Voorhoeve are much more plausible responses than utilitarianism to the fact of the separateness of persons. Whichever theory is ultimately the more plausible overall, it should be clear that each avoids the fundamental problem with utilitarianism. Neither ignores the fact of the separateness of persons in multi-person cases.

Otsuka and Voorhoeve's claim that prioritarianism isn't adequately responsive to the fact of the separateness of persons can't plausibly be construed, then, as a challenge to the prioritarian strategy in multi-person cases. Although they might not agree with the prioritarian method for determining the strength of a person's complaint, it would be odd to suggest that that method, considered as an approach to multi-person cases, doesn't take the separateness of persons seriously in the way that utilitarianism doesn't take the separateness of persons seriously. So, the problem can't be that when prioritarians turn from one-person cases to multi-person cases, they adopt a strategy for the latter which isn't adequately responsive to the fact of the separateness of persons.

Instead, the problem must be supposed to lie in the absence of any change in strategy when prioritarians turn their attention from one-person cases to multi-person cases. Now, the strategy which they hold constant is, as we have seen, a plausible response to the separateness of persons in multi-person cases. So, if prioritarians’ strategy doesn't but ought to change when we turn from one-person cases to multi-person cases, that must be because their one-person strategy is faulty. However, there is no fact of the separateness of persons to fail to respond to in one-person cases. So, the objection cannot be, ultimately, that prioritarians pay insufficient attention to the separateness of persons. If anything, it must be instead that they pay insufficient attention to the unity of the individual by failing to change their strategy when they turn from multi-person cases to one-person cases. But what could justify this charge? The intuitions about one-person cases that Otsuka and Voorhoeve appeal to are not decisive, as we have seen.

V. THE UNITY OF THE INDIVIDUAL

Further on in ‘Why it Matters’, Otsuka and Voorhoeve say something (in response to different line of objection) that might substantiate the charge that prioritarians pay insufficient attention to the unity of the individual. They present two cases. In the first, a child has a 50 per cent chance of ending up with a disability and a 50 per cent chance of ending up able bodied. The child's parent must decide to move either to the city or to the suburbs before it is known how things turn out. Moving to the city will be good for the child if she ends up with a disability, but bad for her if she ends up able bodied. Moving to the suburbs will be good for her if she ends up able bodied, but bad for her if she ends up with a disability. The benefit to the child of moving to the suburbs if she ends up able bodied is substantially greater than the benefit to her of moving to the city if she ends up disabled.Footnote 19

The second case (due originally to Thomas Nagel) is a multi-person variant of the first. The parent has two children, one of whom is able bodied and the other of whom has the disability. The other details are held constant. As Otsuka and Voorhoeve point out, ‘[a] defender of the Priority View must maintain that one has just as much reason to move to the city in [the] one-child case as one does in the two-child case’.Footnote 20 But as with the one-person cases that they present at the beginning of their paper, they dispute this: one has, they claim, less reason to move to the city in the one-child case. The explanation that they offer here is as follows:

even if the child turns out to have a disability in our one-child case, one can justify a decision to move to the suburbs on the grounds that one was looking after that very same child's interest in flourishing in the event that he had turned out able bodied.Footnote 21

The point is that there is a particular kind of justification for exposure to the risk of a bad outcome that's available only in one-person cases. That justification appeals to the fact that the person who is exposed to the risk of the bad outcome and the person who faces the prospect of gain for the sake of which the risk of the bad outcome was not eliminated are the same person. By contrast, in the multi-person cases that I described earlier and in Nagel's original case, decisions which expose people to the possibility of facing losses can't be justified to those people by appeal to the notion that it is for the sake of the prospects of gains for them that they are thus exposed. The only available justifications appeal to the prospect of gains for others.

If the availability of this special ‘unity-of-the-individual’ justification in one-person cases required a change in strategy when we turn from the multi-person cases to the one-person cases, then the charge that prioritarians – who do not change their strategy – fail to pay sufficient attention to the unity of the individual would be substantiated.

However, it's not clear that the availability of the special justification does require a change in strategy. To see this, consider that the justification is available to us in other cases where it's not normally supposed to threaten a view about justice which doesn't change its strategy in light of the availability of the justification.Footnote 22 Specifically, it's also available when the individual in question deserves a particular loss in light of some moral wrong that she has committed. For example, suppose that Emma has killed someone and deserves, on some retributive theory of just punishment, to spend ten years in prison. You have the power to bring it about either that Emma spends ten years in prison, or else that she is exposed to a 50 per cent chance of being in prison for fifteen years and a 50 per cent chance of immediate freedom. In this one-person case, there is a unity-of-the-individual justification available for exposing Emma to the risk of longer imprisonment for the sake of the chance of immediate freedom. That justification appeals to the fact that the person who would be exposed to the risk of the bad outcome and the person who would face the prospect of gain for the sake of which the risk of the bad outcome was not eliminated are the same person: both are Emma. Presumably and plausibly, however, proponents of the retributive theory in question will not be persuaded that what you ought to do is to expose Emma to the gamble rather than sentence her to the ten years that she deserves. The supposed justification for exposing Emma to that risk simply shouldn't get any purchase in your deliberations about what to do here.Footnote 23

I suggest that it is open to prioritarians to view the unity-of-the-individual justification that's available in one-person cases, and invoked by Otsuka and Voorhoeve as grounds for deviating from the prioritarian approach in such cases, in the same way that proponents of the retributive theory of just punishment in Emma's case view it. From the mere fact that there is a justification available for deviating from the requirements of prioritarianism – a justification which is applicable in some cases concerning ‘what one person ought to give another, where that one person is not simply the other's agent’Footnote 24 – it can't be inferred that the deviation is morally required – or if it is, that it is a requirement of distributive justice.Footnote 25

Of course, some justifications for deviations from what a person takes to be the just course of action may be successful, and may justify taking the deviation to be in fact a requirement of justice. But that's not settled by the mere availability of the justification.Footnote 26 The question that prioritarians face, therefore, is as follows. Does the unity-of-the-individual justification warrant taking the deviation from prioritarianism that the justification recommends to be what justice in fact demands? To concede this would be to concede that the prioritarian view is mistaken.

Now, if prioritarianism were motivated solely by considerations that don't apply in one-person cases, then the unity-of-the-individual justification wouldn't conflict with the prioritarian motivation. It would be, in that case, hard to see how prioritarians could avoid the concession just described.Footnote 27 For whereas retributive theorists are motivated by considerations that both apply and give them grounds to disregard the unity-of-the-individual justification in one-person cases, nothing in the prioritarian motivation would give prioritarians reason to disregard that justification in such cases. For example, if prioritarians were motivated by concerns about how people fare in relation to others, then that motivation would give them no grounds to disregard the unity-of-the-individual justification in one-person cases, where relational considerations simply don't apply. For nothing would explain why the unity-of-the-individual justification in such cases should have no force as a matter of distributive justice.

But if the prioritarian view is motivated, as it might plausibly be supposed to be, by something like the thought that distributive justice is fundamentally concerned with protecting people from bad outcomes, understood in non-relational terms, then prioritarians do have grounds to disregard the unity-of-the-individual justification in one-person cases, at least so far as distributive justice is concerned. For acceptance of that justification as grounds for supposing that distributive justice requires the acts that it justifies would conflict with the prioritarian motivation. So, prioritarians can explain why the unity-of-the-individual justification in one-person cases should have no force from the point of view of distributive justice, just as retributive theorists do.Footnote 28

Since the prioritarian can explain her reactions to one-person cases in this way, we are thrown back once more upon our intuitions about one-person cases and the way that the prioritarian motivation applies in them. But these intuitions are not decisive, as we saw in section II. So, prioritarians can resist the present argument against their view.

It turns out to matter, then, that Otsuka and Voorhoeve's argument is not that prioritarians ignore the separateness of persons but that prioritarians ignore the unity of the individual.Footnote 29 For it seems right to think that taking the separateness of persons seriously is a necessary condition of any plausible view about distributive justice. If their argument defended a successful separateness-of-persons justification for deviations from the prioritarian approach in these cases, it would be hard to resist the conclusion that distributive justice itself demands those deviations, and so that prioritarianism is mistaken.

Indeed, the argument of sections III and IV of their paper presents matters as if the objection is a separateness-of-persons objection to the prioritarian approach in standardFootnote 30 multi-person cases: having conceded arguendo the legitimacy of a prioritarian approach in unity-of-the-individual cases, they argue that there should be – but by prioritarian lights can't be – a shift in approach when we turn to standard multi-person cases. Their appeal to the separateness of persons at this stage suggests that the problem is with the prioritarian approach in standard multi-person cases, and they propose a comparative egalitarian approach instead.Footnote 31

But, as I've argued, the prioritarian approach in these cases is adequately responsive to the separateness of persons. So, if a shift is appropriate, that must be because prioritarians get things wrong in unity-of-the-individual cases. As we've now seen, however, prioritarians can resist this charge too.Footnote 32

VI. CONCLUSION

Since it's plausible to think that prioritarianism pays appropriate heed to the fact of the separateness of persons in standard multi-person cases, and since it's also plausible to see the prioritarian motivation as warranting rejection of what I've called the unity-of-the-individual justification in cases where it applies, the plausibility of Otsuka and Voorhoeve's case against prioritarianism turns out to rest largely on their intuitions about the appropriate course of action for a morally motivated stranger in one-person cases. But, as I argued in section II, their intuitions don't count decisively. Even in response to the presentation of the one-person case that Otsuka and Voorhoeve themselves offer, prioritarians, at least, seem unlikely to share those intuitions. In light of all this, there's no reason to see Otsuka and Voorhoeve's own intuitions about these cases as grounds for egalitarianism rather than as reflections of their commitment to it.Footnote 33

References

1 Otsuka, Michael and Voorhoeve, Alex, ‘Why It Matters That Some Are Worse Off Than Others: An Argument against the Priority View’, Philosophy & Public Affairs 37.2 (2009), pp. 171–99CrossRefGoogle Scholar.

2 I use ‘prioritarianism’ and ‘the Priority View’ interchangeably.

3 ‘Why It Matters’, pp. 173–4.

4 ‘Why It Matters’, p. 178.

5 See ‘Why It Matters’, p. 174, and Nord, Erik et al. , ‘Incorporating Societal Concerns for Fairness in Numerical Valuations of Health Programmes’, Health Economics 9 (1999), pp. 25–393.0.CO;2-H>CrossRefGoogle Scholar.

6 Further on, Otsuka and Voorhoeve, ‘Why It Matters’, p. 192, n. 32, cite research that, they claim, ‘reports no significant difference between the treatments people would prefer’ when ‘they imagine themselves in a position of a third party thinking about the appropriate treatment for a single person considered in isolation’ as compared to ‘when considering treatments for themselves’. It's not clear, however, that the research (Ubel, Peter et al. , ‘Value Measurement in Cost-Utility Analysis: Explaining the Discrepancy between Rating Scale and Person Trade-Off Elicitations’, Health Policy 43 (1998), pp. 33–44CrossRefGoogle ScholarPubMed) supports this conclusion. The researchers report no significant difference between people's judgements when they rate (on a scale between death and ‘normal health’) conditions imagined to be conditions suffered by themselves and their judgements when they rate the same conditions, as well as the benefit of curing them, when these are imagined to be conditions suffered by others (considered in isolation). But someone's judgement giving a rating, on a scale between death and health, for a condition understood to be suffered by others cannot without argument be taken to be equivalent to her judgement about the morally appropriate treatment for such a person. At the very least, the possibility that the rating judgement should be treated as a kind of sympathetic first-personal judgement on behalf of the person being imagined must be ruled out. So far as I can see, nothing in ‘Value Measurement in Cost-Utility Analysis’ does so.

7 ‘Why It Matters’, pp. 187–8.

8 ‘Why It Matters’, p. 171.

9 Even if Otsuka and Voorhoeve's intuitions are widely shared, prioritarians can argue that their reasoning explains why these intuitions should be rejected. So, the point about intuitions would not be decisive on its own anyway.

10 ‘Why It Matters’, p. 179.

11 ‘Why It Matters’, p. 180.

12 ‘Why It Matters’, p. 180.

13 See Brink, , ‘The Separateness of Persons, Distributive Norms, and Moral Theory’, Value, Welfare, and Morality, ed. Frey, R. G. and Morris, Christopher W. (Cambridge, 1993), pp. 252–89, at 253–8CrossRefGoogle Scholar. Strictly speaking, moves to Pareto-indifferent and to Pareto-inferior distributions are also separateness-of-persons-respecting.

14 See ‘The Separateness of Persons’, pp. 257–8.

15 ‘The Separateness of Persons’, p. 258 (emphasis added).

16 ‘Why It Matters’, p. 183.

17 ‘Why It Matters’, p. 183.

18 Rawls's difference principle, on standard interpretations, understands complaints in the same way, but aims to minimize only the greatest complaint.

19 See ‘Why It Matters’, p. 188.

20 ‘Why It Matters’, p. 188.

21 ‘Why It Matters’, p. 188.

22 If ‘to justify’ is a success verb, so that the mere availability of a justification entails that it successfully justifies what it purports to justify, then insert ‘purportedly’ before the relevant instances of ‘justifies’ (and so on).

23 One reason that we might think that the justification shouldn't get any purchase in your deliberations is that punishment isn't concerned with the good of those being punished. But distributive justice, we might think, is concerned with the good of those among whom goods are distributed. That being so, it's much less plausible to suppose that the unity-of-the-individual justification shouldn't get any purchase in the prioritarian's deliberations. However, not all retributive theories of punishment take it to be unconcerned with the good of those being punished. Hegel, for example, holds that a criminal ‘is denied his honour’ unless he is punished (see Hegel, G. W. F., Elements of the Philosophy of Right, ed. Wood, Allen W. (Cambridge, 1991), p. 126Google Scholar). A concern with one's honour can plausibly be construed as a concern with one's good. Yet the unity-of-the-individual justification gets no more purchase in a Hegelian's deliberations than in those of a retributivist for whom retribution is unconcerned with the good of those being punished. Thanks to Andrew Williams for suggesting this line of response.

24 ‘Why It Matters’, p. 189.

25 I take prioritarianism to be a view about distributive justice, rather than, as would be implausible, a view intended to capture the whole of morality.

26 See n. 22 above about the way I understand ‘to justify’ for the purposes of this argument.

27 Assuming that prioritarianism itself is nevertheless supposed to apply in these cases, as I have been assuming.

28 This defence of prioritarianism commits prioritarians to the view that if in a one-person case one maximizes the relevant individual's expected utility as opposed to adopting the prioritarian approach, one commits an injustice, which may seem counterintuitive. I think that the counterintuitiveness in the prioritarian case can be explained, however, in part by suspicion of the idea that it could be unjust to act in a way that was preferred by all affected – a suspicion that is normally silenced in the punishment case by the deep hold that considerations of desert and retribution have on our thinking. It is not obvious that the suspicion is well founded: not only retributivists but also strong egalitarians and advocates of proportional justice are committed to seeing some acts as unjust even though they may be preferred by all affected. (Temkin, Larry has written extensively on this point. See for example his ‘Egalitarianism Defended’, Ethics 113 (2003), pp. 764–82CrossRefGoogle Scholar.) Note that although such views may have counterintuitive implications in ‘levelling down’ situations, the counterintuitiveness is not usually supposed to be simply the result of their taking this to be an issue of justice and injustice.

The counterintuitiveness in the prioritarian case may also be explained in part by the supposition that justice doesn't apply in what in n. 30 below I call unity-of-the-individual cases. Note, however, that the supposition that justice doesn't apply in cases involving only one individual (apart from the distributing stranger) doesn't license the conclusion that justice doesn't have application in unity-of-the-individual cases, since the latter cases need not involve only one individual (see n. 29 below).

29 Otsuka and Voorhoeve, ‘Why It Matters’, p. 175, n. 8, note that there are ways in which their one-person cases can be transformed into multi-person cases without producing any change in their intuitions about them. But the criticism that they offer of prioritarian recommendations in such cases remains, in effect, that prioritarians ignore the unity of the individual. So, the possibility of a multi-person presentation of their one-person intuitive case does not show that in the standard multi-person cases that I've been discussing, they have an objection to the prioritarian approach.

30 I'll use this qualification henceforth to distinguish the cases I mean from multi-person cases that are variants on the one-person cases in the way mentioned in n. 29 above. I'll refer to cases of the latter type (including the one-person cases) henceforth as unity-of-the-individual cases.

31 They repeat the emphasis on the separateness of persons at ‘Why It Matters’, p. 192.

32 Otsuka and Voorhoeve ‘Why It Matters’, pp. 185–6, countenance the possibility that considerations of autonomy might give us reason to depart from the prioritarian approach in their one-person case. They deal with this possibility not by arguing that prioritarianism is shown to be false by the appropriateness of the departure, but by pointing out that considerations of autonomy happen not to be present. That suggests that with regard to a one-person case in which considerations of autonomy were present, they wouldn't take it to be an objection to prioritarianism that the appropriate course of action was to respect the person's autonomy rather than to comply with prioritarian recommendations.

If this is right, then one might think that it's also no objection to prioritarianism if it's appropriate to depart from the prioritarian approach when a unity-of-the-individual justification is present. For that justification might be analogous to the autonomy-based justification. I have, of course, attempted to go further than this and show that prioritarians need not even accept that it's appropriate to depart from the prioritarian approach when the unity-of-the-individual justification is present. But even if I'm wrong about that, the foregoing suggests that prioritarianism can still be defended.

33 This article was presented as a paper in Manchester at a 2010 conference on the Priority View, organized by the Manchester Centre for Political Theory. Thanks to those who participated for helpful questions and comments. Special thanks to Michael Otsuka, Alex Voorhoeve and Jonathan Quong for their criticisms and suggestions.