John, Millum and Wasserman (Reference John, Millum and Wasserman2017) have argued that telic prioritarianism justifies morally acceptable discrimination against persons with disabilities in health-care resource allocation while cost-effectiveness analysis does not. In this brief note, I argue that John, Millum and Wasserman (JMW) fail to establish this conclusion. I propose alternative considerations that explain why disability discrimination in the lifesaving cases JMW discuss is morally problematic.

According to telic prioritarianism, the moral value of a benefit diminishes as the absolute level of well-being of a recipient increases. Accordingly, benefits to the worse off may have greater moral value, and do more to make the outcome better, than larger benefits to the better off (Parfit Reference Parfit, Clayton and Williams2002; Adler and Holtug Reference Adler and Holtug2019). This axiology is typically combined with a version of act consequentialism such that the rightness of an act for telic prioritarians is determined by whether the action maximizes moral value relative to alternative acts an agent could have performed.

It is important to note that when making their arguments, John et al. (Reference John, Millum and Wasserman2017: 176) endorse lifetime prioritarianism: prioritarian concern is only directed at some aggregate of a person’s well-being throughout her whole life (see Arneson Reference Arneson2000 and Adler Reference Adler2012 for other endorsements of lifetime prioritarianism). The concern for lifetimes is rooted in respect for the unity of life and the possibility of intrapersonal compensation. Yet some authors have argued that a person’s level of well-being at particular times is morally relevant independently of her lifetime level (McKerlie Reference McKerlie1997; Parfit Reference Parfit2012; Soto Reference Soto2024). These competing views on the temporal scope of prioritarian concern can have different implications for the distribution of aid, but it would take us too far afield to discuss these differences here. I will set aside the question of whether a version of prioritarianism that gives independent weight to levels at particular times can address all of the concerns I raise below about JMW’s account (I am doubtful that it can). Operating within a lifetime prioritarian framework, we shall see that JMW’s position still faces serious challenges.

JMW have argued that telic prioritarianism often would give priority to persons with disabilities in the distribution of health-care resources. According to John et al. (Reference John, Millum and Wasserman2017: 169–170), able-bodied life generally contains greater well-being than life with a disability, other things equal, so a benefit of a given size received by an abled-bodied person generally will count for less than if it were received by a person with a disability. Though some authors have contested the notion that disability itself reduces well-being overall, I will assume for the sake of argument that JMW are correct in their assessment that it often does reduce overall well-being.Footnote ¹

How, then, might telic prioritarianism discriminate against persons with disabilities? According to telic prioritarianism, the moral value of a benefit depends not only on the absolute level of well-being of a potential beneficiary, but also on the size of a benefit (McKerlie Reference McKerlie1994: 39; Arneson Reference Arneson2000: 343; Parfit Reference Parfit, Clayton and Williams2002: 101). A disability, in addition to lowering a person’s absolute level, could lower a person’s capacity to benefit from aid. In some cases, providing scarce aid to a better-off non-disabled person would be considered morally better – maximize moral value – than providing that aid to a worse-off disabled person, since the non-disabled person would benefit more from the aid.Footnote ²

Imagine a scale of well-being where 0 signifies death and 1 signifies perfect health. The benefit of lifesaving treatment could be described as moving someone from 0 – death from non-treatment – to some point higher on the scale – the outcome of treatment. Accordingly, for those employing such quality-of-life scales, saving the non-disabled person moves this person from 0 to some point higher on the scale compared with saving the disabled person, thereby achieving greater benefit (Kamm Reference Kamm, Anand, Peter and Sen2004). This additional good in the outcome has moral value for telic prioritarians. Thus, when neither candidate for lifesaving aid has been worse off, but one person will have a disability in the future if saved, telic prioritarianism, other things equal, directs agents to save the person with the higher capacity to benefit, as the moral value of saving this person will be greater. These cases can arise, for example, if someone has a genetic condition that inevitably results in a debilitating state – which does not affect longevity – but has yet to manifest. The same holds true for someone who only recently became disabled and has not had adequate time to experience the disability, e.g. a trauma victim.

JMW have acknowledged these implications of telic prioritarianism, but they deny that these results are wrongfully discriminatory. As part of their defence, JMW note that standard cost-effectiveness analysis (CEA), in which benefits to the worse off are not given greater weight, also will discriminate against persons with disabilities. But the discrimination resulting from CEA, JMW claim as do others (Brock Reference Brock2009; Kamm Reference Kamm2009), is invidious. The difference between CEA and prioritarianism, according to John et al. (Reference John, Millum and Wasserman2017: 178), is that CEA frequently – or as JMW put it, “systematically” – will discriminate against persons with disabilities while prioritarianism will do so occasionally.Footnote ³

However, JMW do not indicate how much discrimination is too much. Such vagueness places JMW in a precarious position in light of the fact that there are no constraints internal to telic prioritarianism to limit the frequency of discrimination against persons with disabilities. As Parfit (Reference Parfit, Clayton and Williams2002: 101) noted, prioritarianism is an intuitionist theory, i.e. how much weight the worse off should receive is a matter of intuitive judgement. Hence, there could be different versions of telic prioritarianism that vary according to how they weight benefits. Depending on the number of distributive decisions that need to be made, the condition of potential beneficiaries, the size of potential benefits, how long an individual has been disabled, and the weighting of benefits, telic prioritarianism might result in a large amount of discrimination against persons with disabilities. In light of these variables, “occasional discrimination” is not guaranteed within a prioritarian framework. The disability discrimination mandated by telic prioritarianism may be less than that of CEA, but how much less is unclear.

Even if one postulates a strong version of prioritarianism in which benefits to the worse off are heavily weighted, this is irrelevant in lifesaving contexts for those who have only very recently become disabled or will soon become disabled. Smaller benefits that might accrue to these individuals if saved would produce less moral value, which works to their disadvantage on account of their current or imminent disability. In such cases, the verdicts of strong prioritarianism would be equivalent to maximizing unweighted benefits.

A system that sometimes works to the advantage of some members of a social group does not, by virtue of this fact, entail that it cannot wrongly discriminate against other members of this group. JMW must concede this point, otherwise proponents of CEA could also justifiably deny that CEA wrongly discriminates against persons with disabilities since it sometimes benefits them.

JMW might counter that they can both limit the frequency with which telic prioritarianism discriminates against persons with disabilities and provide further justification for its verdicts that is acceptable to these individuals. They note that the quality-of-life scores employed in health-care allocation are biased against persons with disabilities when these scores are generated without the input of persons living with disabilities (John et al. Reference John, Millum and Wasserman2017: 169). This input, it might be argued, would help curb disability discrimination, if it were the case that life with a disability is systematically undervalued by the non-disabled constructing these scores, which indeed seems to be the case. Given this input, John et al. (Reference John, Millum and Wasserman2017: 180–181) suggest that it would be justified, in part, to sometimes count reductions in well-being against persons with disabilities since the claim that they have lower levels of well-being stems from their own valuations.

Whether or not these valuations are uniform and unanimous – which is itself a dubious claim – this reply is inadequate. An employer relying on data might correctly judge that female applicants for a job are more likely than males to request time off for child-rearing purposes, but the veracity of this judgement does not mitigate discrimination against them.Footnote ⁴ Using accurate and unbiased information does not ensure that discrimination is permissible.

Additionally, proponents of CEA could also claim that they can limit the frequency of disability discrimination by requiring that quality-of-life scores include the input of persons living with disabilities. And like JMW, they can count this input as part of the justifying ground for disadvantaging persons with disabilities. They can state, along with JMW, that “[O]ur approach is consistent with treating disabilities no differently than other health and non-health conditions that reduce overall quality of life” (John et al. Reference John, Millum and Wasserman2017: 181). Would these claims exonerate CEA from the charge of invidious disability discrimination? If so, then how can JMW regard telic prioritarianism as superior to CEA with respect to such discrimination? If not, then why would these claims exonerate telic prioritarianism from the charge of invidious disability discrimination?

If, as JMW argue, telic prioritarianism is morally superior to CEA with respect to how frequently each view discriminates against persons with disabilities, then why might this be so? JMW might think that it is because telic prioritarianism ultimately determines the acceptable amount of discrimination. For if there were an independent standard that sets the acceptable amount, then the fact that telic prioritarianism discriminates less than CEA is not itself evidence that the former does not result in invidious discrimination; the independent standard is what informs and adjudicates. But if the acceptable amount of discrimination is merely posited to be whatever results from the application of telic prioritarianism, then this begs the question against other distributive theories and other accounts of wrongful discrimination.

Let us examine more closely an important supposition in JMW’s overall argument. That is, JMW presume that the frequency of discrimination determines whether or not an instance of discrimination is invidious. But when evaluating at least some other acts, we do not take their moral status to be determined by how frequently they occur, e.g. theft or deception. JMW do not indicate why discrimination is different. In a world in which discrimination has been virtually eliminated, suppose that the last living member of the KKK commits a final act of discrimination before his death. It would be implausible to deny that such discrimination is wrongful on the grounds that it is a one-off event.

Nevertheless, suppose that discrimination is unique in that the wrongness of an individual act of discrimination against a member of a particular social group depends on the performance of (many?) other acts of discrimination against members of this group.Footnote ⁵ Now, disability discrimination occurs in other distributive spheres, e.g. employment.Footnote ⁶ If there is poor compliance with telic prioritarianism in other distributive spheres (assuming telic prioritarianism is applicable in other spheres), or if there is moderate compliance that nonetheless results in a significant amount of discrimination, then disability discrimination in health care that maximizes moral value might be morally wrong according to this model because of the wider occurrence of discrimination against persons with disabilities.

Suppose instead that high compliance with telic prioritarianism is ultimately achieved in all distributive spheres. This still does not ensure that disability discrimination that maximizes moral value is morally right, according to the above model of wrongful discrimination. For if what others do at other locations and at other times determines the rightness of individual discriminatory acts, then why shouldn’t the occurrence of widespread discrimination in the past figure in an assessment of the acceptable amount of discrimination and determine the moral status of individual discriminatory acts? If JMW cannot on principled grounds exclude the relevance of past acts, then their concern about the amount of discrimination may be self-defeating.

The arguments JMW use to defend a telic prioritarian justification of disability discrimination seem to either (a) support a competing theory of resource allocation in health care, i.e. CEA, or (b) undermine the very justification of disability discrimination JMW propose.

A major flaw in a telic prioritarian justification of discrimination is a failure to fully appreciate how historical patterns can determine the significance of current discriminatory acts. The meaning or expressive value of acts matters for discrimination theory.Footnote ⁷ That one distributive theory might discriminate less than a competing theory does not indicate that the former theory has no troubling expressive import. For instance, in some of the life-or-death cases discussed above, the raw, unweighted lives of persons with disabilities are portrayed as having too little value and are less worth preserving, a portrayal that resonates with invidious views responsible for the historical mistreatment of persons with disabilities. It would be a mistake to respond that this expressive disvalue is offset by the fact that telic prioritarianism often benefits persons with disabilities. For this is to apply a type of cost-benefit analysis to a framework that operates independently of, and sometimes in opposition to, such analyses.Footnote ⁸

Alternatively, prioritarians might deny that there is any expressive disvalue because of the benefits telic prioritarianism accords to persons with disabilities. However, suppose that Robinson treats members of a particular social group favourably Monday through Friday, but discriminates against them on Saturday and Sunday. Even if Robinson treats the group favourably overall, it does not follow that Robinson’s weekend actions have no expressive disvalue or that these acts are morally permitted because of what he does – more frequently – during weekdays. But this is precisely the same form of flawed reasoning offered in support of telic prioritarianism.

When saving the life of a person with a disability is deemed to be non-optimific, telic prioritarianism – as conceived by JMW – requires that this person be deprived of a chance at receiving a lifesaving resource.Footnote ⁹ But a denial of a chance in these circumstances, many believe, would be unfair.Footnote ¹⁰ A fundamental complaint of many victims of discrimination has nothing to do with the sum of weighted benefits, but rather with the fact that they have been treated unfairly, often in the form of a denial of opportunities.Footnote ¹¹

If the denial of chances to those with disabilities is a form of unfairness, then this consideration seems to extend to the denial of chances to the non-disabled in the distribution of lifesaving resources, as fairness is a symmetrical value. In connection with this unfairness, some might even argue that telic prioritarianism will often discriminate against the abled.Footnote ¹² Suppose for the sake of argument that denying the abled a chance at receiving lifesaving aid is both unfair and discriminatory. Nevertheless, this discrimination does not have the same expressive significance as discrimination against persons with disabilities. When the abled are discriminated against on telic prioritarian grounds, it is because their lives are deemed better and more valuable in an important respect than the lives of persons with disabilities. That is the opposite valuation than occurs when persons with disabilities are discriminated against. The valuation of the lives of persons with disabilities who are discriminated against, in light of historical precedent and the meaning such precedent imbues to these valuations, appears to be more invidious, particularly in lifesaving contexts.

Unfairness coupled with expressive disvalue can explain both (1) how persons with disabilities can be subjected to discriminatory wrongs by being denied a chance at lifesaving aid, and (2) why the non-disabled would not suffer the same kind of wrong if they are denied this chance though they still could be treated unfairly.Footnote ¹³ JMW’s telic prioritarian account of wrongful discrimination does not adequately explain either.