I. INTRODUCTION
The notion of sufficiency seems to play a useful guiding role in our everyday decision-making. We recognize the instrumental importance of having had enough sleep, having enough money and setting aside enough time. However, attempts to develop sufficiency as a fundamental moral and philosophical ideal have been widely regarded as unsuccessful.Footnote 1 Summarizing much of the critical literature, we can say that according to their critics principles of sufficiency are implausible because they sometimes require benefiting the better-off by small amounts rather than benefiting the worse-off by large amounts;Footnote 2 they are indifferent to objectionable inequalities;Footnote 3 and they appeal to a threshold when no such threshold can be specified in a non-arbitrary and unambiguous manner.Footnote 4 These criticisms strongly suggest that the prospects for sufficientarianism in distributive ethics are not good.
This article argues that we should be optimistic about the prospects for sufficientarianism because a clarified account of sufficiency principles reveals that these objections can be avoided and a further examination reveals that the main claims of sufficiency principles are plausible. In section II, I outline the two main versions of sufficiency principle and the powerful objections to which they are vulnerable. This highlights sufficientarianism's poor reputation. In section III I make explicit the minimum claims of distinctive sufficiency principles. I show that such principles state that there is a shift in our reasons to benefit people once they have secured enough. Understood in this way, I argue, sufficiency principles avoid the main objections that have brought them into disrepute and so their prospects should be re-examined.
In section IV, I argue that we can re-examine the prospects for sufficientarianism by examining the plausibility of the shift. I claim that we can examine the plausibility of the shift by examining whether we have the sorts of reasons that would support a shift. In section V, I show that some of our strongest reasons of distributive ethics seem to be of this kind. This supports optimism about the prospects for sufficientarianism.
II. THE STATE OF SUFFICIENCY
Put simply, sufficientarians claim that securing enough of some goods is of special importance. This special importance has been construed in two main ways, both of which are vulnerable to powerful objections. Some sufficiency principles express the headcount claim.Footnote 5
The Headcount Claim: we should maximize the number of people who secure enough.
These sufficiency principles assess distributions solely in terms of the number of people who have secured enough in each distribution. Benefits to those who do not reach the threshold do not improve the assessment of the distribution. Critics have pointed out that these sufficiency principles are vulnerable to what I term the excessive upward transfers objection because they recommend that benefits should be transferred upwards from the worse-off to the better-off.Footnote 6
The Excessive Upward Transfers Objection: sufficiency principles are implausible because, amongst those below the threshold, they require benefiting the better-off by tiny amounts at the expense of large benefits to the worse-off.Footnote 7
To illustrate this point consider a threshold of 100 units, where 100 units represents being very well-off and 1 unit represents being extremely badly off. A version of headcount sufficientarianism will hold that we should benefit the person with 99 units by 1 unit at the expense of benefiting the person with 1 unit by 98 units, but this seems to be the wrong answer. In this case it seems that we should benefit the very badly off person by 98 units. The availability of such counterexamples renders these views relatively implausible since rival principles would favour benefiting the worse-off in such cases.
The special importance of securing enough has also been construed in a second way.Footnote 8 Some sufficientarians have expressed what Paula Casal identifies as the negative thesis.Footnote 9
The Negative Thesis: once everyone has secured enough, no distributive criteria apply to benefits (though wholly aggregative criteria may apply).
There are at least two versions of this view. One version holds that all distributions in which everyone has secured enough are equally good, even when distributions vary in terms of their total level of advantage. The other version holds that distributions in which everyone has secured enough can still be judged better or worse by wholly aggregative criteria.Footnote 10 This second view will hold that when everyone has secured enough in more than one distribution the distribution with the greatest total of advantage is to be preferred.
Critics have argued that these sufficiency principles are implausible because they are incapable of condemning some regressive policies, which recommend imposing greater costs on the worse-off than the better-off, when both groups have secured enough. These critics advance what I term the indifference objection.
The Indifference Objection: sufficiency principles are implausible because they are objectionably indifferent to inequalities once everyone has secured enough.
To illustrate the force of this objection consider Harry Frankfurt's version of the negative thesis, which sets the sufficiency threshold at the level of contentment.Footnote 11 So long as all individuals are content it does not matter, on his view, if the worse-off bear greater costs than the better-off. But this is implausible. Imagine that there are only super-contented millionaires and much poorer persons who are content in our society. Even assuming that the tax burden will not push members of either group below the contented threshold, it seems that we should not be indifferent about who should bear the greater costs in this situation. The super-contented billionaires should bear the costs and indifference is implausible in such cases.
We should also note that negative sufficientarians can avoid this problem only by inviting another powerful objection. Negative sufficientarians could set the threshold so high that very few if any individuals could be said to have secured enough. This makes indifference appear less objectionable because, to put things crudely, we do not seem to have reason to care about the inequalities between trillionaires, or their welfare equivalents, though these inequalities may be far greater than the inequalities, which do disturb us, between those who are less well-off. This modification, however, deprives the view of any distinctive guidance where we need it. Negative sufficientarians who set a very high threshold will offer identical guidance to the form of distribution that applies to distributing benefits below the threshold. Since the threshold makes no difference to the guidance offered in those realistic circumstances, sufficiency principles with such a high threshold are dispensable.
In light of these objections to the two main types of sufficientarianism we can say that the prospects for sufficientarianism seem bad, but it is important to understand whether sufficiency principles must be vulnerable to these objections. In the next section I will identify and clarify the claims sufficiency principles must make and I will show that the objections that have brought sufficiency into disrepute can be avoided and so we should re-examine the prospects for sufficientarianism.
III. A ROLE FOR SUFFICIENCY IN DISTRIBUTIVE ETHICS
To see if sufficientariansim can avoid the objections levelled at it we must identify the claims that sufficientarians must make. Paula Casal has done more than anyone to elucidate the general structure of sufficientarianism and the role of sufficiency in distributive ethics, so I begin by noting a number of Casal's insights.Footnote 12 Casal notes that sufficientarians express the positive thesis.
The Positive Thesis: it is important that people live above a certain threshold, free from deprivation.Footnote 13
The term ‘deprivation’, however, seems to suggest a low threshold, but a sufficiency threshold may be high.Footnote 14 For instance, Roger Crisp's version of sufficientarianism does not claim that there is a threshold at the point of deprivation in this sense. Crisp's threshold is set by appeal to compassion felt by an impartial spectator.Footnote 15 There is no reason to think that only circumstances involving deprivation would elicit compassion from an impartial spectator. We often feel compassion for those who are unlucky in love or lose their dog. Whether they are free from deprivation or not does not affect our reaction.
It should also be noted that sufficiency principles must claim that we have weighty reasons to secure ‘enough’. If securing enough is a less than weighty demand then sufficiency principles would be trivial and lightweight principles that did not make a significant difference in any situation for which we require guidance. However, securing enough need not be the weightiest distributive demand. For example, securing basic civil liberties might be more important than securing sufficient wealth, at least in reasonably favourable conditions, but once no greater equal liberty for all can be secured, the sufficiency of wealth principle would become operative and will have an important role to play in guiding policy. The sufficiency of wealth principle would still provide us with important guidance even if a principle of liberty is lexically prior.
Sufficiency principles must also state that securing ‘enough’ is a non-instrumentally weighty demand. Sufficiency principles are dispensable if securing enough is only important as a means to meet the demands of purely aggregative, egalitarian or prioritarian principles. For example, we might favour policies to ensure that people have secured enough of some things – enough education, enough wealth, or enough health are worthy policy goals – but if these policies are grounded in a non-sufficientarian distributive principle, perhaps a prioritarian principle of welfare, there would be no need for sufficiency principles to account for these requirements of distributive ethics. Since instrumentally weighty sufficiency principles can be accounted for just as well without reference to sufficiency principles those principles cannot usually affect the prospects for sufficientarianism. It is worth noting that if there are instrumental sufficiency principles which, given certain reasonable empirical assumptions about how societies work, are foreseeably fixed requirements of realizing the demands of a fundamental, sound, non-sufficientarian principle then these instrumental sufficiency principles would contribute to the prospects for sufficientarianism in distributive ethics. However, when testing the prospects for sufficientarianism we should, at least in the first instance, set the bar higher and examine whether there are any non-instrumental sufficiency principles. In light of these considerations I would restate the positive thesis thus.
The Positive Thesis: We have weighty non-instrumental reasons to secure at least enough of some good(s).
Casal also identifies the negative thesis discussed above, which expresses the indifference many sufficientarian views have towards inequalities above the threshold.Footnote 16 Casal rejects the negative thesis because it renders sufficientarian views vulnerable to the indifference objection. But in searching for the minimum claims sufficiency principles must make it is worth considering a weaker version of the negative thesis, which I term the diminution thesis.
The Diminution Thesis: once people have secured enough our reasons to benefit them further are weaker.
Those expressing the diminution thesis can avoid the indifference objection because they can claim that once people have secured enough we still have reasons to care about the worse-off, though they are weaker. This seems to capture what sufficientarians are getting at in claiming that sufficiency has special importance. Together the positive thesis and the diminution thesis hold that for equally sized benefits it is less important to benefit someone who has secured enough than someone who has not secured enough. However, we should not characterize sufficientarianism in this way, because on this characterization sufficientarianism will not provide distinctive guidance. Prioritarians claim that our reasons to secure further benefits always become weaker as the beneficiary becomes better-off.Footnote 17 Thus, however much is determined to be enough, prioritarians can always agree that it is more important to benefit the worse-off who have not secured enough than the better-off who have secured enough, at least where equally sized benefits are concerned. Prioritarians will reject that this is because securing enough is important, but the guidance offered by prioritarians and sufficientarians will be identical and so equally plausible. For this reason the diminution thesis and the positive thesis are not jointly sufficient claims of distinctive sufficiency principles.
It could be argued that the positive thesis contains the distinctive idea of sufficiency principles: the threshold. However, to provide a full account of the character of distinctive sufficiency principles we need to explain the idea of a threshold more explicitly than the positive thesis allows us to on its own. I shall turn to this task presently.
Sufficiency thresholds seem to amount to a change in our reasons to benefit people once they have secured enough. We should treat those who have secured enough differently from those who have not secured enough in virtue of their position relative to the threshold. One way of explaining this thought is that there is a change in the nature of our reasons, and not merely a diminution in their weight, once enough is secured. Consider the following examples of intuitively sufficientarian views that fit this mould. A sufficientarian view may claim that leximin, which attaches absolute priority to benefits to the worst-off group or person, applies to benefiting those who have not secured enough and that weighted prioritarianism applies to further benefits.Footnote 18 The diminution thesis cannot explain why there is a shift in the criteria that apply in this example, though the diminution thesis is compatible with such a view.
Consider another view that states that we should give weighted priority to the worse-off below the threshold and once people have secured enough we should distribute equally.Footnote 19 The diminution thesis is incompatible with this sufficientarian view since there is a shift in the nature of our reasons that is not also a diminution of their weight and so we should reject it as a necessary claim of sufficiency principles.
The principles in these examples seem clearly sufficientarian and they seem so in virtue of the different distributive criteria that apply to those who have secured enough and those who have not. The sufficiency threshold, then, seems to mark a shift in the nature of our reasons to benefit people further. This intuitive thought can be formally expressed by what I term the shift thesis.
The Shift Thesis: once people have secured enough there is a discontinuity in the rate of change of the marginal weight of our reasons to benefit them further.
The shift thesis and the positive thesis together render sufficientarianism distinctive from prioritarianism because they state that once people have secured ‘enough’ there is a discontinuity in the rate of change of the marginal weight of our reasons to benefit them further. This means that once a person has secured enough the relationship between our reasons to benefit her and how well-off she is changes. Whether she has secured enough or not affects the rules that determine the weight of our reasons to benefit her.
Some, but not all, prioritarians claim that there is no such shift and that priority to the worse-off diminishes at a continuous rate.Footnote 20 The recommendations of prioritarian principles can be represented by a smooth curve on a graph plotting the relationship between the moral importance of one additional unit of benefit and how well-off the recipient is (see fig. 1). A shift will disrupt this smooth curve. Indeed, if we should distribute equally once people have secured enough then we could not plot the guidance entirely on a graph like this:
Figure 1. Uniform prioritarianism
Now, it is worth pausing to show that the shift thesis renders sufficiency principles distinctive from other possible prioritarian views. The broadest possible characterization of prioritarianism states that if A is worse-off than B then we have weightier reasons to give an equally sized benefit to A than B. This characterization of prioritarianism is compatible with manifestations of the discontinuity that I have called a shift (see fig. 2). This may lead us to think that the shift does not render sufficiency principles distinctive from these views. Those prioritarian views compatible with a shift I term non-uniform prioritarianism. Non-uniform prioritarians work out the importance of benefiting those with at least a certain level of advantage differently from those with less than that level of advantage. To be determinate, non-uniform prioritarian principles must state the different rates at which priority diminishes before and after this shift.
Figure 2. Non-uniform prioritarianism
Non-uniform prioritarians cannot explain the shift in terms of merely giving priority to benefiting the worse-off. Different rates of change are required to explain why the importance of benefits changes in this way. The only explanation available to them states that there is a discontinuity in the rate of change of the marginal weight of our reasons to benefit people once they have secured ‘enough’. This is the shift thesis. In order to account for such a change, then, non-uniform prioritarians must appeal to a sufficiency principle.Footnote 21
The role of sufficiency in non-uniform prioritarianism is but one among many possible roles for sufficiency principles in distributive ethics. Shifts are also compatible with views that hold that sufficiency is important alongside the difference principle, responsibility-sensitive egalitarianism or strict equality.Footnote 22 When a shift marks a change from priority to equality, once people have secured enough, sufficiency principles are more obviously distinctive and so no additional argument is necessary.
Together the positive thesis and the shift thesis comprise the minimum claims of any distinctive sufficiency principle. Articulating these claims together draws our attention to the set of sufficiency principles that determine the prospects for sufficientarianism and enables us to see that sufficientarians and their critics should be arguing over the plausibility of a particular shift. Characterizing sufficientarianism in this way also enables us to see that sufficientarianism can avoid the main objections to it and this motivates a re-examination of the prospects for sufficientarianism.
The indifference objection stated that sufficiency principles are implausible because they are objectionably indifferent to inequalities once everyone has secured enough. However, shifts are compatible with a wide range of distributive criteria once everyone has secured enough. For example, sufficientarians can care solely about equality once everyone has secured enough. They may still set the threshold too high or too low, but the dilemma is not as great a problem because there is no need to claim indifference once people have secured enough. This shows that the indifference objection is of limited significance for examining the prospects for sufficientarianism.
The excessive upward transfers objection stated that sufficiency principles are implausible because, amongst those below the threshold, they recommend benefiting the better-off by tiny amounts at the expense of large benefits to the worse-off. However, sufficientarians can avoid this objection by attaching priority to the worse-off below the threshold. For example, sufficientarians can claim that the difference principle, which holds that we should maximize the position of the least well off, applies to benefits to those who have not secured enough.Footnote 23 Employing this strategy means that equally sized benefits would always go to the worse-off. This would also enable them to avoid the upward transfers objection. This objection also does little to dim the prospects for sufficientarianism.
The foregoing discussion has identified the minimum claims that distinctive sufficiency principles will make and has shown that the objections commonly made of sufficientarianism can be avoided. Since these objections have so far constituted the examination of the prospects for sufficientarianism we should look to re-examine the prospects for sufficientarianism by using new tests. In the next section I will show how we can examine the plausibility of particular sufficiency principles and the prospects for sufficientarianism generally.
IV. RE-EXAMINING THE PROSPECTS FOR SUFFICIENTARIANISM
The prospects for sufficientarianism depend upon sufficiency principles being plausible where distinctive. What is distinctive about sufficiency principles is the shift articulated jointly by the shift thesis and the positive thesis. To test the plausibility of the shift we can identify the conditions that would support a shift. To do so I will focus on the claims about reasons made by the positive thesis and the shift thesis. If we have these kinds of reasons then we should conclude that there is likely to be a shift and that the prospects for sufficientarianism are good. There may be other ways of testing the prospects for sufficientarianism, for example, sufficiency principles might be well suited to resolving value clashes, providing guidance in non-ideal situations and may more readily be the object of consensus than their rivals, but I focus on this method because I believe it is the clearest way of identifying a sufficiency threshold and is the most promising line of enquiry.
Recall the positive thesis,
The Positive Thesis: We have weighty non-instrumental reasons to secure at least enough of some good(s).
For there to be a shift there must be at least some non-instrumental, weighty reasons to secure at least enough of some good(s). Now, recall the shift thesis.
The Shift Thesis: once people have secured enough there is a discontinuity in the rate of change of the marginal weight of our reasons to benefit them further.
One way of accounting for the discontinuity in the rate of change of the marginal weight of our reasons to benefit people is by appealing to satiable reasons that cease to apply to benefits once people have secured enough. If some weighty non-instrumental reasons for benefiting people only apply up to a certain point then the overall profile of reasons we have to benefit people will change at that point. Those who have not secured enough can call on the weight of more and different reasons than those who have secured enough. To account for the shift, then, we must have some reasons that are satiable.
We can also note that to support distinctive sufficiency principles these reasons must be non-egalitarian and sated at a low enough level to provide us with some guidance for real situations. I will now clarify the features of reasons that support a shift.Footnote 24
Non-instrumental
To support a shift the reason(s) must be non-instrumentally weighty. As I discussed when clarifying the positive thesis, if securing enough is important only because securing enough serves some other distributive principle, such as principles of priority or equality, then we can account for the shift without reference to a sufficiency principle. As such, instrumental reasons cannot be the hallmark of an indispensable sufficiency principle.
Reasons that support a shift need not be fundamentally weighty, however. Non-instrumental reasons can be sufficient to establish an indispensable role for a sufficiency principle, in combination with other features. If securing enough of something is necessary to satisfy the demands of a fundamental reason then we would have a non-instrumental reason to secure enough. We may have a fundamental reason to establish, say, a society of equals of which sufficient health might be a permanent part. In this case the reason to secure enough health would not itself be fundamental, but the sufficiency principle it supported could not be omitted from a complete and sound theory of distributive ethics.
Satiability
Joseph Raz distinguishes between satiable and insatiable principles in the following passage.
Satiable principles are marked by one feature: the demands the principles impose can be completely met. When they are completely met then whatever may happen and whatever might have happened the principles cannot be, nor could they have been, satisfied to a higher degree.Footnote 25
Satiable reasons behave like satiable principles. Instrumental satiable reasons are familiar from ordinary discussion. Once a person has secured ‘enough’ money to buy one more pint of ale he cannot, with any force, ask for more on those grounds. Once a person has secured enough money for a bus ticket, ‘for a bus ticket’ is no longer a reason to give her more money. Our reasons to benefit people change when that person is no longer deficient in the relevant respect. There may be strong claims for benefits beyond the application of that reason, we need not be upper-limit sufficientarians, but such claims must be made using a different profile of reasons, which alters our overall reasons to benefit people further. The task for the sufficientarian is to find similar, but less familiar, satiable reasons that are non-instrumental.
Avoid a high threshold
In order to offer distinctive guidance sufficiency principles must be supported by non-instrumental satiable reasons, but these reasons must not be sated at a point so high that they would set a very high threshold. As we saw at the end of section II, if a sufficiency principle has a high threshold then its guidance will be identical to the guidance provided by the distributive criteria that apply prior to the shift in realistic circumstances. One way of specifying how high is too high might be in terms of the circumstances of justice. We might say that the reason must be sated within moderate scarcity where benefits are neither extremely scarce, such that no competing claims can be met, nor superabundant, such that all claims can be met.Footnote 26
Non-egalitarian
Another requirement of reasons that support a shift is that they must not be sated only by an exactly equal distribution of the good in question. If enough just meant enough for equality, then the sufficiency principle would not be distinctive from principles of equality. On some understandings distributive equality is not a satiable demand.Footnote 27 However, equality can be understood as a satiable demand and so I must stress that satiable egalitarian reasons cannot support a sufficientarian shift.
This highlights the forms of egalitarianism with which sufficiency principles are incompatible. Sufficiency principles are incompatible with egalitarian principles that claim that all that is required is exactly equal shares of benefits, but they are compatible with egalitarian ideals like the ideal of a society of equals or egalitarian principles, which claim that we should secure equality only once we have secured sufficiency.Footnote 28 Rivalry with equality is not essential to the sufficientarian position.
Weighty
Finally, to support a shift the same reason(s) to secure enough must also be weighty. If securing enough is not a weighty demand the sufficiency principle it supports would be a trivial, lightweight consideration that does not make a significant difference to the overall guidance. However, sufficiency need not be the weightiest distributive demand for the prospects for sufficientarianism to be good. Such reasons must be weightier than trivial but need not have absolute weight.
Summary
I shall term reasons that meet all of the above criteria sufficientarian reasons. The re-examination of the prospects for sufficientarianism can be understood as a search for sufficientarian reasons. This is because sufficientarian reasons support a shift, which only sufficiency principles can explain. Sufficiency principles supported by these reasons will be indispensable. The foregoing section helps us to see that the debate about the place of sufficiency in distributive ethics is a debate about the plausibility of the shift and sufficientarian reasons. In the next section I will suggest that some of the most compelling reasons to alter the distribution of benefits and burdens are sufficientarian reasons and so we should be optimistic about the prospects for sufficientarianism.
V. IS THE SHIFT PLAUSIBLE?
A number of egalitarians have cited sufficientarian reasons in support of their view.Footnote 29 Both Thomas Scanlon and John Rawls argue that we have weighty fundamental reasons to eradicate deprivation, remove objectionable differences in power and avoid stigmatizing differences in status.Footnote 30 These claims have also been defended, more recently, by Martin O'Neill.Footnote 31 Many of our most powerful intuitions about injustice draw on situations in which the demands of these reasons are unmet. These include our intuitions about exploitation, humiliation and poverty. Since explanation of these intuitions is a key factor in determining the plausibility of an account of distributive ethics these reasons can be ignored only by the most implausible principles and theories. Interestingly, these reasons seem not only weighty and fundamental, as their proponents claim, but also non-egalitarian and satiable and therefore support shifts.
The relief of severe deprivation is a sufficientarian reason. Our reason to ‘eradicate severe deprivation’ applies to benefits to those who suffer from it and does not apply to those who do not. Deprivation can be eradicated by ensuring that everyone has at least a certain bundle of goods that coincides with its removal, whether this is equal or not. Indeed, it seems odd to think that deprivation can be eradicated only by an exactly equal share of some good. This reason, it seems, is satiable consistent with inequality and therefore supports a distinctive sufficiency principle: people should secure ‘enough’ to avoid suffering from severe deprivation.
Stigmatizing differences in status and unacceptable differences in power can be removed, in principle, consistent with inequalities in power and status. Typically citizens do not have equal shares of power and yet it seems that stigmatizing differences do not, and need not, exist between citizens in such circumstances. An unequal division of responsibility, and as such power and status, seems fully compatible with full citizenship, and may be required where some individuals are better able to serve the rest of society by taking up positions of greater responsibility. Rawls's principle of fair equality of opportunity, for instance, regards as unjust inequalities in career prospects arising from social class but permits them on the grounds of talent and motivation.Footnote 32 Hence, some inequalities in status and power are permitted for Rawls. So these reasons can be sated consistent with inequality too and support a sufficiency principle emphasizing the importance of securing ‘enough’ control over one's life, ‘enough’ autonomy or ‘enough’ to avoid appearing in public without shame.Footnote 33 It is implausible to think that only equal shares of some good are demanded by such complex reasons.
Given the prevalence of sufficientarian reasons it seems as if the prospects for sufficientarianism are not only better than has been thought but are in fact rather good. Moreover, many egalitarians should welcome this conclusion since sufficientarian reasons are often reasons to oppose the most pernicious inequalities.
VI. CONCLUSION
In this article I have identified the set of sufficiency principles that determine the prospects for sufficientarianism by clarifying the positive thesis and identifying the shift thesis as together comprising the minimum claims of distinctive sufficiency principles. This re-characterization made explicit what sufficientarians and their critics must argue over: the plausibility of the shift. Moreover, so understood, sufficiency principles can avoid the objections that have brought them into disrepute by avoiding the negative thesis and the head-count claim. I then devised a framework that can be used to re-examine the prospects for sufficientarianism. This involved identifying sufficientarian reasons, which support the kind of shift only sufficiency principles can account for. Finally, I argued that because we seem to have some sufficientarian reasons we should be optimistic about the prospects for sufficientarianism.
I believe that the foregoing arguments and explanations have shown that moral and political philosophers must look again at what sufficiency principles have to offer by focusing on the plausibility of the shift. One way they may do so is by arguing for or against the proposition that we have sufficientarian reasons.Footnote 34
