Intergenerational Justice and Freedom from Deprivation

Abstract

Almost everyone believes that freedom from deprivation should have significant weight in specifying what justice between generations requires. Some theorists hold that it should always trump other distributive concerns. Other theorists hold that it should have some but not lexical priority. I argue instead that freedom from deprivation should have lexical priority in some cases, yet weighted priority in others. More specifically, I defend semi-strong sufficientarianism. This view posits a deprivation threshold at which people are free from deprivation, and an affluence threshold at which people can live an affluent life, even though their lives may be even further improved beyond that point. I argue that freedom from deprivation in one generation lexically outweighs providing affluence in another generation; in all other cases, freedom from deprivation does not have lexical priority.

1 Introduction

According to an important family of views about intergenerational justice, people should be free from deprivation, no matter when or where they are born.¹ If our actions cause future deprivation, and if we can act such that future generations would be free from deprivation instead, our actions are an injustice towards them. And similarly, if our actions bring prosperity to future generations but cause deprivation in this generation, they are unjust. Call this view intergenerational sufficientarianism.

The ideal of freedom from deprivation raises two questions about justice between generations. The first is whether this ideal should play any role at all in specifying what justice between generations requires. Though there is no consensus on what freedom from deprivation entails precisely, this question is often answered affirmatively, to the point where almost everyone accepts that freedom from deprivation should have at least some weight in intergenerational justice.² The second question, however, is how much weight freedom from deprivation should have. This question is at the heart of this article. Some theorists hold that freedom from deprivation should always trump other distributive concerns; other theorists hold that freedom from deprivation should have some but not lexical priority.³ I argue instead that freedom from deprivation should have lexical priority in some cases, yet weighted priority in others.

My view, semi-strong sufficientarianism, draws on two thresholds, where the qualification ‘semi-strong’ signals that freedom from deprivation has neither always (‘strong’) nor never (‘weak’) lexical priority.⁴ Its first threshold, the deprivation threshold, denotes the point at which people are free from deprivation but below which they are not. Subsequently, its affluence threshold denotes the point at which people can live an affluent life, even though their lives may be even further improved beyond that point. We can capture the core of my proposal as follows: freedom from deprivation in one generation lexically outweighs providing affluence in another generation; in all other cases, freedom from deprivation does not have lexical priority. This, I will argue, is the proper place for freedom from deprivation in intergenerational justice.⁵

This article is structured as follows. In Section 2, I introduce semi-strong sufficientarianism. In Section 3, I argue that semi-strong sufficientarianism offers the most plausible response to the lexicality objection, which maintains that freedom from deprivation should not have lexical priority in intergenerational justice. In Section 4, I discuss two objections to semi-strong sufficientarianism's response to the lexicality objection. In Section 5, I discuss the benefits of semi-strong sufficientarianism in relation to the demandingness of intergenerational justice. Section 6 concludes.

2 Semi-strong sufficientarianism

At the heart of semi-strong sufficientarianism are its thresholds and its way of assigning priority to people's distributive claims. Consider the thresholds first. The lower threshold denotes the point below which people live in deprivation, as Paula Casal famously put it.⁶ This idea of deprivation plays a fundamental role in many different and competing conceptions of intergenerational justice.⁷ As such, this deprivation threshold has strong intergenerational credentials. As Schuppert puts it, freedom from deprivation “exhibit[s] a form of moral urgency that is unmatched by other moral claims”.⁸

The affluence threshold denotes the point at which people can live an affluent life, even though their lives may be even further improved beyond that point. For my argument, it is important that the affluence threshold is not so high that no justice-relevant improvements are possible above that point. Otherwise, my account would hold that freedom from deprivation has lexical priority only vis-a-vis benefits that are not relevant to justice. Such a limited role for freedom from deprivation would fail to give due weight to the moral urgency of that ideal (see Section 5).

This affluence threshold too must have “intergenerational validity”.⁹ As Daniel Petz puts it, intergenerational thresholds must be “valid for different generations in the future and consistent over time-scales. As uncertainty about the properties and/or interests of future generations (particularly in the far future) increases, this is a difficult challenge and puts limits to the specificity of the threshold”.¹⁰ Without specifying its point exactly, then, and keeping in mind that justice-relevant improvements above the threshold should be possible, I believe that the high thresholds defended by other sufficientarians could serve as a blueprint for an intergenerational affluence threshold. David Axelsen and Lasse Nielsen, for example, define such a high threshold by drawing on “the ideal of freedom from duress” by which they mean “the freedom from significant pressure against succeeding in central aspects of human life”.¹¹ Another proxy may be Robert Huseby's maximal threshold, which “equals a level of welfare with which a person is content”¹² or, as he has defended more recently, a threshold above which people can live a “good life”.¹³ Alternatively, Yitzhak Benbaji proposes a ‘luxury threshold’ above which people “are so well off […] that every small benefit to them would be a luxury. They could use the resources allocated to them for another vacation, for consuming even better wines, or for having the honor of being listed as one of the richest people in the world”.¹⁴ Still another threshold would be the point above which people can live a fully flourishing life, as Ingrid Robeyns has proposed.¹⁵

The size of the gap between the deprivation threshold and the affluence threshold is an important feature of semi-strong sufficientarianism. On the one hand, the gap between these thresholds must be sufficiently large. This is to ensure that semi-strong sufficientarianism is a substantive improvement over single-threshold versions of intergenerational sufficientarianism, which are vulnerable to powerful objections.¹⁶ On the other hand, the gap between the deprivation threshold and the affluence threshold should also not be too large. Otherwise, the range of well-being levels (or another metric) with respect to which freedom from deprivation has merely weighted (rather than lexical) priority would be so large that benefits to the very well off, or even the incredibly well off, could outweigh freedom from deprivation.¹⁷ If this were the case, semi-strong sufficientarianism would fail to give due weight to freedom from deprivation. Hence, the gap between the deprivation threshold and the affluence threshold should be neither too small nor too large.

Let us briefly examine how semi-strong sufficientarianism relates to other sufficientarian views. The main normative role of the sufficientarian thresholds is usually presented by the positive thesis and negative thesis.¹⁸ The positive thesis highlights the moral importance of people reaching a certain sufficientarian threshold (e.g. a level of well-being or capabilities). The negative thesis marks the point above which no additional distributive requirements apply, and below which benefits have lexical priority. Some sufficientarians reject the negative thesis in favour of the shift thesis, which marks the point at which there is a discontinuity in the rate of change of the marginal weight of our reasons to benefit people further.¹⁹ As I will explain, semi-strong sufficientarianism accepts the positive thesis and the shift thesis but does not assert the negative thesis.

Semi-strong sufficientarianism accepts the positive thesis, at least with respect to the deprivation threshold, to highlight the moral importance of freedom from deprivation.²⁰ However, some sufficientarians might take issue with such a low threshold. For Huseby, for example, a low threshold, which he defines in terms of basic needs, “does not by any means exhaust the principle of sufficiency, but serves mainly to highlight the idea that insufficiency below subsistence level is morally more urgent than insufficiency above it, and that this should be taken into account when prioritizing below the maximal threshold”.²¹ Therefore, Huseby rejects that the positive thesis should be tied to such low threshold. Similarly, Liam Shields holds that the positive thesis should be concerned with a higher threshold than the deprivation threshold.²² Here, however, I will assume that it is of particular moral importance that people are free from deprivation, in the sense intended by the positive thesis. But as should become clear, I wholeheartedly agree that sufficientarianism is concerned with more than being free from deprivation, which is why semi-strong sufficientarianism also endorses the shift thesis.

Some sufficientarian views defend not one but multiple thresholds, and semi-strong sufficientarianism follows their lead. Benbaji, for example, accepts two similar thresholds, located at different levels, that are both of particular moral importance for people to reach and below which benefits have weighted priority.²³ Huseby has offered two versions of a multi-threshold view. In the earlier view, the “lower threshold […] does not refer to the positive thesis. It just emphasizes priority to the very badly off. The higher threshold, moreover, is intended to capture both the positive and the negative thesis”.²⁴ The later view accepts a lower threshold for the positive thesis (though this threshold is higher than the lower threshold in his initial view) and a higher threshold for the negative thesis.

Semi-strong sufficientarianism is similar to such views in that it accepts multiple thresholds. And it shares with Huseby's views in particular that its thresholds have different normative roles. However, semi-strong sufficientarianism rejects the negative thesis. Moreover, it ascribes normative roles to both the specific thresholds and to the combination of the thresholds. We can only explain the kind of priority that freedom from deprivation has according to semi-strong sufficientarianism by invoking both the deprivation threshold and the affluence threshold.

That brings us to the second element of semi-strong sufficientarianism: its way of assigning priority. We can capture this with the following two claims:

The strong claim. For any two generations G* and G^, freedom from deprivation in G* always outweighs providing affluence in G^.

The weak claim. In all cases not covered by the strong claim, when allocating goods between G* and G^, freedom from deprivation does not have lexical priority.²⁵

Saying that freedom from deprivation has lexical priority is commonly taken to mean that any improvement in freedom from deprivation outweighs any improvement above that threshold.²⁶ The strong claim, however, only gives lexical priority to freedom from deprivation if this is weighed against providing affluence in another generation. Such lexical priority can be grounded on two different commitments, which suggests that semi-strong sufficientarianism can be endorsed for distinct reasons. The first way for freedom from deprivation to lexically outweigh providing affluence is that the reason to promote freedom from deprivation in one generation trumps the reason to provide affluence in a different generation. This is true irrespective of the number of people that might become more affluent and irrespective of the total amount of benefits that could accrue to them. This is how lexical priority is typically understood.

The second way for freedom from deprivation to lexically outweigh providing affluence is that the reason in favour of the former silences or disables the reason in favour of the latter.²⁷ This reason for lexically prioritizing freedom from deprivation might appeal to certain non-sufficientarian intuitions about distributive justice, such as that weighing the interests of the very well off and those who live in deprivation as if they merit equal consideration is a sign of disrespect towards the worse off. If so, the reasons in favour of providing affluence need not be presented or can be ignored if they conflict with reasons to promote freedom from deprivation in a different generation.

Subsequently, the weak claim says that in all cases not covered by the strong claim, freedom from deprivation does not have lexical priority, and different distributive concerns, including freedom from deprivation, must be weighed. But the weak claim does not specify how such weighing should take shape. Therefore, it is open to different specifications. On a sufficientarian reading of the weak claim, for example, benefits below the threshold should have priority over benefits above it. And so, sufficientarians might endorse:

The sufficientarian weak claim. In all cases not covered by the strong claim, when allocating goods between G* and G^, benefitting the worse off people matters more if they are below the affluence threshold, and even more if they are below the deprivation threshold.

The sufficientarian weak claim says that benefits below the deprivation threshold should have priority. But this priority is weighted, rather than lexical, because if large benefits are at stake for people who are between the thresholds, these benefits can outweigh possible benefits to those who are below the deprivation threshold; and similarly for benefits between the thresholds or above the affluence threshold.

However, proponents of semi-strong sufficientarianism need not endorse the sufficientarian weak claim. They might hold, for example, that if the strong claim does not offer guidance, we should take into account egalitarian considerations (alternatively, we might rephrase the weak claim to take into account prioritarian, utilitarian or still other concerns).²⁸ Egalitarians might accept:

The egalitarian weak claim. In all cases not covered by the strong claim, when allocating goods between G* and G^, benefitting people matters more, the more such benefits promote equality.

The egalitarian weak claim holds that in cases in which freedom from deprivation does not have lexical priority, an egalitarian rather than a sufficientarian distribution must be pursued. Semi-strong sufficientarianism itself, however, is agnostic about the precise specification of the weak claim.

3 The lexicality objection

Freedom from deprivation is often considered to trump other distributive concerns.²⁹ A common objection to this is the lexicality objection.³⁰ This objection maintains that lexically prioritizing freedom from deprivation objectionably forgoes large improvements above that threshold, or improvements to many people above that threshold, only to realize minor improvements in freedom from deprivation.³¹ For this reason, critics maintain, we must reject the lexical priority of freedom from deprivation. In this section, I will argue that semi-strong sufficientarianism is not vulnerable to this objection.

Let us first consider some possible responses to the lexicality objection. Some people might hold that lexically prioritizing freedom from deprivation is a virtue rather than a vice. They might say, for example, that when someone reaches that threshold, this triggers a morally significant shift that justifies giving lexical priority to subthreshold benefits. But I agree with the lexicality objection that even if we have particularly strong reasons to be concerned with freedom from deprivation, giving it lexical priority comes with excessive and unjustifiable costs to those above the deprivation threshold (including those just above it).³² For this reason, freedom from deprivation should not always have lexical priority.

Alternatively, Lukas Meyer and Harald Stelzer propose ‘weak sufficientarianism’, which gives weighted rather than lexical priority to freedom from deprivation (or to some other sufficientarian value) and endorses prioritarianism above the threshold.³³ Weak sufficientarianism avoids implying that large benefits above the threshold, or benefits to many people above the threshold, are lexically outweighed by a concern for freedom from deprivation. In fact, the weaker the relative weight of freedom from deprivation, the further weak sufficientarianism moves away from the objectionable implications of giving lexical priority to freedom from deprivation. Weak sufficientarianism, then, aims to strike a balance between freedom from deprivation and other distributive concerns.

However, weak sufficientarianism is vulnerable to the non-lexicality objection. It must accept that benefits to people who are very well off can outweigh major improvements in freedom from deprivation, if the benefits to those who are very well off are sufficiently large. To illustrate, if a single-unit improvement in freedom from deprivation has a moral value of 100, and a single-unit benefit to someone above the affluence threshold has a moral value of 1, then if 101 people in affluence can be given a single-unit benefit, this outweighs a single-unit improvement in freedom from deprivation.

However, if we take seriously the claim that freedom from deprivation “exhibit[s] a form of moral urgency that is unmatched by other moral claims”,³⁴ we should reject sacrificing freedom from deprivation for the sake of providing benefits to those who are very well off.³⁵ The lexicality objection rightly shows that the unmatched moral urgency of freedom from deprivation does not justify lexical priority in all cases. But sufficientarians can still maintain that it justifies lexical priority in some cases, namely when the choice is between freedom from deprivation or providing affluence. In fact, this is exactly what is asserted by semi-strong sufficientarianism. This is not because improving the lives of the affluent lacks moral value; instead, it is because freedom from deprivation has particular moral urgency. For this reason, we should reject weak sufficientarianism's response to the lexicality objection and accept semi-strong sufficientarianism.

In response to this critique of weak sufficientarianism, a defender of that view might propose a multi-threshold version of weak sufficientarianism over the single-threshold version. For example, one might reject the strong claim and accept a revised version of the sufficientarian weak claim, modelled after Benbaji's view discussed in Section 2:

The revised sufficientarian weak claim. When allocating goods between any two generations G* and G^, benefitting the worse off people matters more if they are below the affluence threshold, and even more if they are below the deprivation threshold.

This multi-threshold view draws on two thresholds that both mark a point below which benefits have weighted priority. Compared to its single-threshold counterpart, it sets a higher bar for freedom from deprivation to be outweighed by benefits to those who are above the affluence threshold. This is because freedom from deprivation enjoys comparatively stronger priority over benefits above the affluence threshold than over benefits between the thresholds. However, this view too is vulnerable to the non-lexicality objection, as major gains above the affluence threshold can still outweigh freedom from deprivation.

To avoid this, one might endorse a multi-threshold view modelled after Huseby's view and say that all benefits below the affluence threshold have lexical priority, and not only the benefits that promote freedom from deprivation:

The revised strong claim. When allocating goods between any two generations G* and G^, benefitting the people below the affluence threshold has lexical priority.³⁶

By giving lexical priority to all benefits below the affluence threshold, this view avoids benefitting those above the affluence threshold at the expense of those below the deprivation threshold.

However, we must reject this view for the same reason as we must reject always giving lexical priority to freedom from deprivation. It would equally lead to forgoing large benefits above the affluence threshold, or benefits to many people above the affluence threshold, only to realize a minor gain to someone below the affluence threshold but above the deprivation threshold (even if that person is just below the affluence threshold). But this is objectionable. In Section 5, I will explain why it is important that the affluence threshold is not so high that no justice-relevant improvements are possible above that point. What matters here is that because such improvements are morally significant, we should not lexically prioritize benefits to those just below that threshold over those just above it.³⁷ Therefore, this multi-threshold view too should be rejected.

An alternative response to my critique of weak sufficientarianism is that the non-lexicality objection rests on an implausible empirical assumption about the number of people that will live in the future and the quality of life they are likely to have. The objection maintains that benefits to people who are very well off should not outweigh freedom from deprivation. But many future people might be living in deprivation, and so the theoretical possibility that weak sufficientarianism prefers providing affluence over freedom from deprivation might not materialize. However, though this might be true, it does not address an important reason for lexically prioritizing freedom from deprivation over providing affluence, namely that not doing so suggests that claims to affluence merit equal consideration as claims to being free from deprivation. I believe sufficientarians must reject this since freedom from deprivation, having unmatched moral urgency, is at the core of sufficientarian justice. Moreover, if we can avoid falling prey to the non-lexicality objection, this gives us a reason to revise our view.

That brings us to semi-strong sufficientarianism. This view agrees with weak sufficientarian views that freedom from deprivation has weighted but not lexical priority over benefits to people who are just above that threshold. And it agrees with lexical views that benefits to those above the affluence threshold can be lexically outweighed. Yet it maintains that only freedom from deprivation has lexical priority over benefits above the affluence threshold. And it holds that freedom from deprivation does not always lexically outweigh other distributive concerns. Because of this, the lexicality objection and its non-lexical counterpart lack force against semi-strong sufficientarianism. This makes semi-strong sufficientarianism preferable to alternative sufficientarian views about freedom from deprivation.

4 Two objections to semi-strong sufficientarianism's response to the lexicality objection

I want to consider two objections to my argument that semi-strong sufficientarianism offers the most plausible response to the lexicality objection. The first objection holds that the lexicality objection still applies to semi-strong sufficientarianism.³⁸ This is true. However, even the strongest possible version of the lexicality objection that applies to this view has little force. Note first that the strongest possible lexicality objection is not that semi-strong sufficientarianism prefers a distribution between generations in which everyone is only just above the deprivation threshold over a distribution in which one person is just barely below that threshold and all others (say, billions and billions) are ecstatic, living lives well above the affluence threshold. In that case, semi-strong sufficientarianism would prefer the latter distribution over the former. But what matters here is the underlying rationale. This preference is not grounded in the fact that in the latter distribution billions and billions of people are above the affluence threshold, as benefits above that threshold are lexically outweighed by concerns for freedom from deprivation. Rather, it is because freedom from deprivation has weighted but not lexical priority over benefits between the deprivation threshold and the affluence threshold, and in this scenario billions and billions of people would, in virtue of being well above that threshold, also reach that threshold.

To illustrate this point and to see how semi-strong sufficientarianism compares to other views, let us assume that the level of the deprivation threshold is 10 and the level of the affluence threshold is 100. Furthermore, suppose that we face the (unlikely) trade-off between benefitting a single person in generation G* or benefiting billions of people in generation G^ (or multiple generations G^, G`, … G’). Now consider the following three distributions, in which the numbers refer to levels of well-being (Table 1):

Table 1. Semi-strong sufficientarianism and the lexicality objection

Semi-strong sufficientarianism prefers distributions D2 and D3 over distribution D1. This is because in D2 and D3 billions of people are at the affluence threshold (or higher), whereas in D1 everyone lives at the deprivation threshold. For example, in comparing D1 and D2, the single-unit benefit to one person in D1 does not outweigh the benefits that billions of people receive in D2. However, semi-strong sufficientarianism prefers D3 over D2 because the single-unit benefit to one person in D3 lexically outweighs the multi-unit loss to billions of people in that distribution compared to D2. Instead of living ecstatic lives well above the affluence threshold, they are only at the affluence threshold. From the point of view of semi-strong sufficientarianism, however, this is preferable to the distribution in D2 because of the lexical priority of freedom from deprivation over providing affluence.

Advocates of weak sufficientarianism, utilitarianism and prioritarianism, among others, might object to this implication of semi-strong sufficientarianism because it shows that the view does not maximize utility and gives too much weight to the least advantaged. It forgoes large benefits to many beneficiaries above the affluence threshold only to achieve a minor benefit to someone below the deprivation threshold. (To make the point even more dramatic, we might even assume that billions of people suffer even larger losses, or that the benefit to one person in G* does not bring them to the deprivation threshold but only increases their well-being from 1 to 2.) And so, this, it seems to me, is the worst form of the lexicality objection that applies to semi-strong sufficientarianism.

Does this version of the lexicality objection render semi-strong sufficientarianism an implausible view? I do not think so, at least not for those who believe that freedom from deprivation should have significant weight in specifying what justice between generations requires. Some might think that the cost of protecting freedom from deprivation is too high. But if freedom from deprivation should have significant weight in intergenerational justice, semi-strong sufficientarianism advocates a plausible trade-off between freedom from deprivation and other distributive concerns that we should be willing to make.

The second objection to my argument that semi-strong sufficientarianism offers the most plausible response to the lexicality objection is that it falls prey to a different lexicality objection. Let us again assume that the level of the deprivation threshold is 10 and the level of the affluence threshold is 100. Now consider the following four people and their levels of well-being (Table 2):

Table 2. A revised lexicality objection

Person A is just below the deprivation threshold and person B is just above it. I argued that it is implausible to hold that A has lexical priority no matter the counterfactual benefits to B. But one might object that this gives rise to a new version of this objection. Person C is just below the affluence threshold, and person D is just above it. And so, semi-strong sufficientarianism gives lexical priority to A compared to D but not compared to C. This is true even though C's and D's life are almost equally good. And the opposite also holds: B only has weighted priority over D whereas A has lexical priority over D, even though A's and B's life are almost equally good. This, one might argue, shows that lexical priority, as advocated by semi-strong sufficientarianism, still comes with implausible implications.

This is indeed an implication of semi-strong sufficientarianism. But it is an implication that can be justified against the background of the challenges to both always and never lexically prioritizing freedom from deprivation. What drives the lexicality objection, in my view, is that in the canonical case discussed in Section 3, the person lacking freedom from deprivation and the person just above that threshold are almost equally well off, yet concerns for the former person have lexical priority over concerns for the latter person because only the former lives in deprivation. This is different from the example discussed here. If freedom from deprivation exhibits a form of moral urgency that is unmatched by other moral claims, and if, as I have argued, neither always lexically prioritizing freedom from deprivation nor never lexically prioritizing such freedom can be justified, then this speaks in favour of giving freedom from deprivation lexical priority in some cases, but weighted priority in others. The implication that minor changes in people's levels of well-being (or in another metric) around the thresholds can trigger shifts from weighted to lexical priority (and other way around) is simply an unavoidable implication of a view that is preferable to rival conceptions about freedom from deprivation in intergenerational justice.³⁹

5 Demandingness and intergenerational justice

According to Bernard Williams, people should not be “agents of the universal satisfaction system”,⁴⁰ and an account of intergenerational justice must express this idea in how it specifies the demands of justice, both for current and future generations. For that reason, theories of intergenerational justice must be neither underdemanding nor overdemanding. I will argue that this speaks in favour of semi-strong sufficientarianism compared to both non-sufficientarian conceptions of intergenerational justice and some common alternative versions of intergenerational sufficientarianism.

A theory of intergenerational justice is underdemanding if it does not address certain kinds of injustice. Failing to give proper weight to freedom from deprivation is a kind of underdemandingness. Any minimally plausible view about intergenerational justice should hold that freedom from deprivation should have at least some significant role in specifying how benefits and burdens between generations should be allocated. But worries about underdemandingness have also been raised against concerns for freedom from deprivation. Axel Gosseries, for instance, argues that intergenerational sufficientarianism allows for objectionable savings of resources for future generations, which leave the current generation, and the worst off in that generation in particular, with nothing but freedom from deprivation.⁴¹ This is because if justice is only concerned with freedom from deprivation, then anyone above the deprivation threshold, no matter how little or far above it, has no justice-relevant claim to additional benefits. Paradoxically, such sufficientarianism may also lead to objectionable dissavings, which leave future generations with just enough to be free from deprivation. This is because future generations too can, from the standpoint of justice, make justice-relevant claims to being free from deprivation but not to other distributive concerns.

This objection is important and applies to certain articulations of intergenerational sufficientarianism. Intergenerational sufficientarianism typically posits a relatively low threshold, such as a poverty threshold or a basic-needs threshold, which signals the point at which people are free from deprivation. If that is all that justice requires, the view is indifferent among all intergenerational distributions that are equally good with respect to the sole measure of securing freedom from deprivation. But this is a highly implausible view.⁴² There is more that is valuable from the standpoint of justice than freedom from deprivation, and a plausible principle of intergenerational justice must reflect this.

However, intergenerational sufficientarianism is only exclusively concerned with freedom from deprivation if it endorses the negative thesis above that threshold, and says that once freedom from deprivation is secured, no further distributive criteria apply. But semi-strong sufficientarianism is not committed to the negative thesis; and even views like Huseby's that endorse the negative thesis for a higher threshold are not committed to this. Semi-strong sufficientarianism can, following the idea that freedom from deprivation does not always have lexical priority, accept the need for principles that guide the distribution above its thresholds as long as these principles are compatible with how semi-strong sufficientarianism gives priority to freedom from deprivation. Therefore, it is more demanding than accounts of intergenerational sufficientarianism which accept a low threshold combined with the negative thesis above that threshold. This is because semi-strong sufficientarianism entails that what happens between its thresholds matters from the standpoint of justice, as does what happens above the affluence threshold. Even though it claims that, in some cases, freedom from deprivation has lexical priority, this does not exhaust all that is required from the standpoint of justice.

Hence, semi-strong sufficientarianism avoids underdemandingness in its specification of what justice between generations requires by (i) giving proper weight to freedom from deprivation and by (ii) saying that freedom from deprivation is not all that is needed from the standpoint of justice.

Additionally, semi-strong sufficientarianism also avoids being overdemanding: it does not, to paraphrase Williams, turn generations (or people above the affluence threshold within those generations) into agents of the universal satisfaction system. Semi-strong sufficientarianism does not always require maximal sacrifices from those above the affluence threshold. It is true that if a choice must be made between freedom from deprivation or providing affluence, freedom from deprivation has lexical priority. But when considering benefits to people between the thresholds or above the affluence threshold, lexical priority does not apply. Relatedly, semi-strong sufficientarianism does not always forbid burdening the worst off, even if they are below the deprivation threshold. Unlike some other views, semi-strong sufficientarianism maintains that burdens to those below the deprivation threshold can be justified in some cases, namely if sufficiently large benefits, or benefits to a sufficiently large number of people, between the deprivation threshold and the affluence threshold are at stake.

However, one might object that semi-strong sufficientarianism is vulnerable to a common overdemandingness objection in intergenerational justice, which would also apply to intergenerational utilitarianism, as well as prioritarianism and weak sufficientarianism.⁴³ Since the number of future people (in all future generations) is so vast, even a miniscule improvement in their well-being can outweigh the huge burdens these improvements impose on an earlier generation who can make the improvement happen.⁴⁴ Semi-strong sufficientarianism might be especially vulnerable to this sort of objection: a generation (e.g., the current generation) could be under a justice-based obligation (up to the point that it has decreased their own well-being to the deprivation threshold) to save resources for future generations and satisfy the justice-relevant claims to additional benefits for future people, even if those future people are well above the affluence threshold.

Some sufficientarians in particular might take issue with this implication, namely if they endorse the negative thesis. Advocates of that thesis might say that above the affluence threshold no additional distributive requirements apply. This means that current generations need not worry about providing benefits to future generations (or members of those generations) above the affluence threshold. Because of this, sufficientarian views that endorse the negative thesis are less demanding than sufficientarian views which reject that thesis, as well as views such as utilitarianism and prioritarianism. For those concerned with overdemandingness in intergenerational justice, this speaks in favour of views which endorse a high threshold above which the negative thesis applies.⁴⁵

In earlier sections, I argued that sufficientarian views which endorse the negative thesis (and the implied lexical priority to subthreshold benefits) fail to provide an adequate answer to the lexicality objection. However, I agree that because semi-strong sufficientarianism rejects the negative thesis, it is more demanding than sufficientarian accounts that accept it. That raises the question whether semi-strong sufficientarianism is too demanding, even if it is less demanding than views such as utilitarianism and prioritarianism.

In my view, we should prefer semi-strong sufficientarianism over the negative thesis, for two reasons. First, defending the negative thesis for the limits it sets on the demands of justice seems misguided. As I understand it, the reason for endorsing the negative thesis and saying that above some threshold no further distributive criteria apply, is that it expresses the (lack of) value of benefits above that threshold, rather than expressing how demanding a theory of distributive justice can or should be.⁴⁶ As Huseby puts it: “at some level of well-being, distributive concerns tend to peter out. At the very least, many do not care, from the point of view of justice, that two extremely well off individuals are not exactly equally well off”.⁴⁷ Or as Harry Frankfurt puts it: “if everyone had enough, it would be of no moral consequence whether some had more than others”.⁴⁸ If this reading of the main reason for endorsing the negative thesis is right, then those who accept the negative thesis because it makes a view less demanding accept it for the wrong reasons.

Second, and more importantly, semi-strong sufficientarianism gives freedom from deprivation a more central and demanding place in how it distributes benefits and burdens across generations than a view which endorses the negative thesis. The negative thesis entails that above-threshold benefits lack justice-relevant value. Note, first, that this is against how I have defined the affluence threshold, namely as denoting the point at which people can live an affluent life, even though their lives may be even further improved beyond that point. More importantly, however, a view which only lexically prioritizes freedom from deprivation over benefits that do not matter from the standpoint of justice does not seem to prioritize freedom from deprivation in a meaningful way. According to the negative thesis, maximally burdening those above the threshold is justified because there is not much of a trade-off in the first place, since benefits above that threshold lack justice-relevant value. But if freedom from deprivation has lexical priority only in such cases, it has insufficient moral weight. Sufficientarians should be more ambitious in their defence of freedom from deprivation. Therefore, semi-strong sufficientarianism holds that justice-relevant improvements above the affluence threshold are lexically outweighed by freedom from deprivation.

6 Conclusion

Justice between generations requires that people are free from deprivation, no matter when or where they are born. In this article, I have defended a novel version of such intergenerational sufficientarianism: semi-strong sufficientarianism. It posits a deprivation threshold at which people are free from deprivation, and an affluence threshold at which people can live an affluent life, even though their lives may be even further improved beyond that point. And it holds that freedom from deprivation in one generation lexically outweighs providing affluence in another generation; yet in all other cases, freedom from deprivation does not have lexical priority. I have argued that this is the proper place for freedom from deprivation in intergenerational justice.

My hope is that semi-strong sufficientarianism resonates with the increasing move towards eclectic and hybrid theories of distributive justice, which combine concerns for, among other things, equality, priority and sufficiency, into a single view.⁴⁹ The view sketched in this article offers the contours for such a theory in intergenerational justice, in which freedom from deprivation is sometimes but not always granted lexical priority.