Disability, Options and Well-Being

Abstract

Many endorse the Bad-Difference View (BDV) of disability which says that disability makes one likely to be worse off even in the absence of discrimination against the disabled. Others defend the Mere-Difference View (MDV) of disability which says that, discounting discrimination, disability does not make one likely to be worse (or better) off. A common motivation for the BDV is the Options Argument which identifies reduction in valuable options as a harm of disability. Some reject this argument, arguing that disabled people's prospects aren't hindered by having fewer options. In this article, I defend the Options Argument by arguing that, in disability cases, possessing a greater number of valuable options seems to overall improve well-being prospects. As such, the Options Argument appears to be sound and – although it doesn't establish the BDV – it lends plausibility to the BDV by identifying a potentially significant cost of disability.

Many (e.g. Shakespeare 2013; McMahan 2005; Singer 2005) endorse the Bad-Difference View (BDV) of disability which says, roughly, that disability makes one likely to be worse off even in the absence of discrimination against the disabled or ‘disablism’. Others (e.g. Amundson 2005; Barnes 2016) endorse the Mere-Difference View (MDV) of disability which says, roughly, that, discounting disablism, disability does not make one likely to be worse (or better) off. A common motivation for the BDV is the Options Argument which identifies reduction in valuable options as a harm of disability. Some (e.g. Wasserman & Asch 2013) argue instead that disabled people's prospects aren't hindered by possessing fewer options because the non-instrumental value of the options to which disability precludes access is multiply realisable. Call this the Multiple Realisability Reply.

In this article, I defend the Options Argument and reject the Multiple Realisability Reply by arguing that, in disability cases, possessing a greater number of valuable options seems to overall improve well-being prospects, even though the value of the options that disabled people cannot access is multiply realisable. As such, the Options Argument appears to be sound and to support the BDV by identifying a potentially significant cost of disability. I don't argue that the BDV is true. The purpose of my arguments is to shift the burden of proof onto those who deny that option reduction hinders disabled people's well-being prospects and, more generally, MDV defenders. Until now, the onus seemed to be on BDV proponents to engage with MDV replies to the Options Argument. I do so here, shifting the onus back to MDV advocates either to reject my arguments or to identify a benefit of disability that counterbalances the cost of option reduction. If they cannot, then the Options Argument offers good grounds for belief in the BDV.

In §1, I offer preliminaries. In §2, I present the Options Argument and discuss two MDV replies, arguing that the Multiple Realisability Reply is most promising and that this reply is inconsistent with another disability-positive argument, leaving MDV proponents with a dilemma. In §§3–4, I defend the Options Argument and reject the Multiple Realisability Reply by (i) identifying benefits of possessing a greater number of options and arguing that these are relevant to disability (§3), and (ii) arguing that the costs involved in possessing additional options appear unlikely to offset the benefits in disability cases (§4). Finally, in §5, I respond to objections and note residual issues.

1. Preliminaries

Whether possessing more options improves prospects is a general issue. One debate in which other thinkers have discussed it (e.g. Singer 2001: 56; Moller 2011: 198–99), and to which it appears intuitively relevant, is that between the MDV and BDV:

BDV: a person is (in virtue of their disability, other things equal and minus the effects of disablism) likely to be overall worse off with a disability than they would have been without.

MDV: a person is (other things equal and minus the effects of disablism) likely to be overall neither worse (nor better) off with a disability than they would have been without.

Barnes (2009a; 2016) argues against probabilistic interpretations of the views (although cf. Campbell & Stramondo 2017). However, it's clearly false that being disabled always makes one worse off. Someone could benefit from disability (by, for example, avoiding conscription) and could do worse because of disability (by, for example, experiencing chronic pain). We thus need weaker, probabilistic claims to capture the views.

Both views consider well-being in non-disablist worlds. Roughly, these worlds include no disability stigma and no policies that unjustly disadvantage disabled people, and all public accommodations, media, etc. are made accessible as far as is possible within budget and technological constraints. More specificity requires consideration of what counts as unjust discrimination, which is beyond my scope.¹ However, I needn't be more specific, as the option reductions I discuss are (fairly) uncontroversially not due to disablism.

I follow the tradition of discounting only disablism and not other harms such as (non-disablist) failure to accommodate disability because of lack of resources (cf. Barnes 2016; Kahane & Savulescu 2009). We discount disablism, first, because everyone agrees that disability tends to hinder prospects in disablist worlds and, second, because this debate originated in the claims of disabled people that, without prejudice (but not other harms), disability would not hinder prospects.²

These understandings might face issues involving identity. If disabilities are identity-determining, then we cannot compare a particular person's prospects with a disability and without (Campbell and Stramondo 2017: 161). I won't discuss whether disabilities are identity-determining. What I'll say is that ‘a person’ and ‘they’ in the definitions above needn't rigidly designate the same numerically identical person. If disability is identity-determining, we may be comparing the prospects of two metaphysically different but relevantly similar people. And we needn't think numerical identity must be preserved for a comparison between individuals to bear relevance to moral considerations (Kahane & Savulescu 2009: 37). As such, I'll assume that the relevant comparisons are possible and at least potentially morally relevant. This move seems dialectically warranted, as all who understand the views as comparing the well-being prospects of individuals with disabilities and without (e.g. Barnes 2016: 86–88) face similar issues and none respond to the Options Argument or defend the MDV by invoking problems of non-identity.

I understand ‘disability’ as a condition in which a person has atypical bodily or psychological features which, in interaction with intrinsic features of the person (e.g. their personality) and features of their environment (built, social or attitudinal), cause them limitations (cf. Shakespeare 2013: 74–75; World Health Organisation 2001).³ However, it shouldn't matter whether you endorse this account as I concentrate on paradigm cases.

My arguments apply to physical and psychological disability. Some (e.g. Barnes 2016; Wasserman & Asch 2013) think that psychological disabilities raise quite different issues; for example, perhaps it's unclear how to evaluate the well-being of severely cognitively disabled people. If you think that psychological disability should be considered separately, then you may take my arguments to apply to only physical disability.

I understand ‘well-being’ as the non-instrumental prudential value a life has for the person whose life it is. I won't endorse any theory of well-being; my discussion aims to be theory-neutral. To do this, I focus on the substantive goods in life (e.g. achievement, personal relationships, etc.) – which every plausible theory of well-being agrees typically make life better (Kahane & Savulescu 2009: 43) – and assume that (other things equal) the more of these substantive goods we get, the better our prospects (cf. Stoner 2016; Campbell & Stramondo 2017).⁴

I understand S’s prospects for well-being, roughly, as how likely S is to have a given level of well-being. I take idealised subjective probability to be the relevant notion (cf. Jackson 1991: 463–65). That is, S’s prospects are determined by the prediction a rational agent would make if they considered all relevant evidence.

I focus on the effects on prospects of quantity of valuable options. I won't discuss the potential effects of quantity of bad options, option quality or diversity, although all are potentially relevant (Schroeder 2018: 16). I narrow my focus, first, because the discussion about quantity of valuable options (henceforth, options) is that which motivates the Options Argument and, second, because these other issues deserve more attention than I can provide here.

Finally, well-being and well-being prospects are an established philosophical topic. However, questions about the prospects of different groups are in part empirical, and there might be other methods of answering them, such as through empirical investigation. I won't argue that philosophical investigation is the best way of identifying prospects. Nor do I take what I say to be the definitive word on the nature of the connection between options and well-being or on the quality of disabled people's lives. My aim is merely to contribute to an ongoing debate and encourage further discussion of this topic.

2. Options and disability

In this section, I explain the relevance of the connection between options and well-being to the debate. First, I present the Options Argument. Then, I discuss two MDV replies.

Intuitively, having options is good. People generally value autonomy and one way we exercise autonomy is by making choices. Choices are possible only if we have options from which to choose (Ullmann-Margalit and Morgenbesser 1977). Accordingly, we think that it would be terrible to have no options as this would curtail our freedom. If some choice is good, perhaps more is better: possessing more options might allow me access to one that I would benefit more from exercising than I would from any in my original set. If not, then perhaps I can simply ignore the additional options. This reasoning might lead one to think that possessing more options always improves prospects. Nozick (1969: 462) and Rawls (1971: 143) seem to have thought so (Dworkin 1988: 64). Rawls reasons that more choice is always preferable because people ‘are not compelled to accept more if they do not want to, nor does a person suffer from greater liberty’.

If having more options is better, this prima facie supports the BDV. On any reasonable understanding, disabilities involve limitations: a blind person cannot become a surgeon; someone with total body paralysis often cannot eat solid foods; severely cognitively disabled people sometimes cannot communicate etc. Some (e.g. Singer 2001: 56; Kahane & Savulescu 2016: 776) think that option reduction is a harm of disability. They might think, as Rawls did, that having more options is always beneficial, or that disability in particular reduces options in a prospect-hindering manner. I'll focus on this second (more charitable) interpretation.

Here's one formulation of the Options Argument:

1. 1. If something reduces valuable options then, to the extent that it does, it can have both negative and positive effects on well-being prospects.

2. 2. Disability reduces valuable options.

3. 3. In disability cases, the negative effects of option reduction are greater than the positive effects.

4. 4. Therefore, insofar as it reduces valuable options, disability hinders well-being prospects.

This argument doesn't establish the BDV, as it doesn't show that disability overall hinders prospects. What it purports to do is support the BDV by identifying a harm of disability.

2.1. The Multiple Realisability Reply

(1) is a modest claim that seems difficult to deny. So, those who want to reject (4) can deny either (2) or (3). Many (e.g. Asch & Wasserman 2010: 206–09; Campbell & Stramondo 2017: 157–58; Moller 2011: 198–99; Schroeder 2016: 224–25; Wasserman & Asch 2013: 149–52) deny (3) on the following grounds. Of course, it would be bad to be unable to access any of life's non-instrumental goods, but disabilities typically only foreclose certain avenues by which one can access them. On any plausible interpretation, life's non-instrumental values are multiply realisable and, although disabled people cannot realise them in certain ways, they can still experience them all. A blind person can get her fill of aesthetic experience by listening to Mozart and a paraplegic person can get his fill of play through wheelchair basketball. Therefore, although disabled people may have fewer options, they still have a perfectly good option set that allows them access to all life's non-instrumental goods – which, the reply claims, means that their prospects are not affected by option reduction and (3) is false.

In defence of the MDV, then, thinkers have rejected (3) by advancing the Multiple Realisability Reply.⁵ This rejection appeals to the following premise:

Restricting goods (RG): For a subject S and some multiply realisable non-instrumental good G, restricting S’s options to access G won't affect S’s well-being prospects as long as S still possesses some/enough option(s) which enable(s) her to access G.

RG is ambiguous between two interpretations. RG-some claims that restricting S’s options will not affect her prospects as long as S still has one option that enables access to each good. RG-enough claims that restricting S’s options will not affect her prospects as long as S still has sufficiently many options to access each good. I'll focus on the weaker (more charitable) interpretation: RG-enough. On RG-enough, the Multiple Realisability Reply amounts to saying that both disabled and non-disabled people have sufficiently many options, so have equal prospects insofar as these are determined by the sizes of their option sets.

2.2. The DSO-argument and inconsistency

Some defend the MDV by endorsing RG. However, arguments that purportedly support the MDV deny principles similar to RG. Campbell and Stramondo (2017) and Barnes (2009a; 2009b; 2016) argue that disability is good for prospects to the extent that it grants access to options exclusive to disabled people:

The experience of disability … isn't just one of absence … It's, rather, one of absence in particular areas that creates (in virtue of that very absence) opportunities in other areas – opportunities not open to the non-disabled. And some disabled people report that the resulting experiences disability creates mean that, on the whole, disability is of great benefit to them. (Barnes 2009b: 15 original emphasis)

Examples of disability-specific options include things like access to the disabled community (Barnes 2016: 116), being able to work peacefully in loud environments (Schroeder 2018: 15) and being permitted to avoid long queues (Campbell and Stramondo 2017: 158).⁶

This disability-positive argument – call it the disability-specific options argument or DSO-argument – purportedly supports the MDV by identifying a benefit of disability. However, the following is a natural extension of RG:

Expanding goods (EG): For a subject S and some multiply realisable non-instrumental good G, adding to S’s options to access G won't affect S’s well-being prospects as long as S already possessed some/enough option(s) which enabled her to access G.

EG follows from RG if there is symmetry between the effects of adding and removing options. It seems clear that there is. If I benefit S by enlarging her option set from O to O*, then I disadvantage her by reducing her options from O* to O. Likewise, if going from O to O* makes no difference, then going from O* to O won't either.

So, RG and EG have identical truth values. This is a problem for MDV defenders. RG provides them with a response to the Options Argument. However, EG undermines the disability-positive DSO-argument. According to EG, the possession of disability-specific options won't improve disabled people's prospects because (i) disabled people already have access to all life's goods prior to the addition of disability-specific options, and (ii) the value of disability-specific options is multiply realisable. This means that non-disabled people can realise the relevant non-instrumental goods through alternative means. For instance, the value of accessing the disabled community can presumably be realised through accessing other communities. There doesn't appear to be any non-instrumental value uniquely experienced by disabled people. Thus, EG implies that possessing disability-specific options doesn't improve disabled people's prospects. MDV proponents, therefore, must deny EG for their DSO-argument to run.

This leaves MDV proponents with a dilemma. Endorsing RG while denying EG is inconsistent. They can either reject both, which would deprive them of their reply to the Options Argument, or they can endorse both, which would deprive them of their DSO-argument about the benefits of disability. Neither option is appealing for MDV advocates.

2.3. The sufficiency interpretation

Perhaps one interpretation exists on which the DSO-argument can be reconciled with the Multiple Realisability Reply: the sufficiency interpretation. To explain it, I must introduce some terminology. Call the option sets of disabled people without the addition of disability-specific options their basic option sets; call their option sets with the addition of disability-specific options their enhanced option sets; and call the options unavailable to disabled people non-disability-specific options.

The DSO-argument is straightforwardly undermined by the RG-some version of the Multiple Realisability Reply, as EG-some (which is true if RG-some is true) holds that disability-specific options won't improve disabled people's prospects because disabled people have in their basic option sets at least one option that enables access to each good. However, EG-enough says that expanding options will improve prospects as long as the size of a person's option set is not already at or above the sufficiency threshold T. So, on the RG-enough version of the Multiple Realisability Reply, perhaps MDV proponents can consistently hold that disability-specific options do improve disabled people's prospects by enlarging their (basic) option sets closer to T, while believing that disabled people would not benefit from gaining non-disability-specific options because the size of their enhanced option sets is at or above T, meaning that additional options make no difference. That is, perhaps, on this sufficiency interpretation, MDV advocates can consistently endorse the DSO-argument and (the RG-enough interpretation of) the Multiple Realisability Reply.

I will now argue that the DSO-argument and the Multiple Realisability Reply are inconsistent even on the sufficiency interpretation. If MDV proponents want to reject the Options Argument, as they do, it's crucial that they maintain that disabled people wouldn't benefit from possessing non-disability-specific options. For this to be so, the size of disabled people's enhanced option sets must be at T. If not, then possessing more options would improve disabled people's prospects. But the view that the size of disabled people's enhanced option sets is at T seems implausible given the significant variation in the sizes of different disabled people's basic option sets and in the amount of disability-specific options different disabled people possess. Different people with different disabilities in different locations etc. possess different numbers of options. This implies that, wherever T is placed, there will probably be a significant number of disabled people who don't reach it, even with the addition of disability-specific options. For these disabled people, non-disability-specific options would improve their prospects (on the view we are considering), meaning that the Multiple Realisability Reply – which states that possessing non-disability-specific options wouldn't improve disabled people's prospects – would be unsound.

Seemingly, the only way to keep the Multiple Realisability Reply and make it such that the addition of non-disability-specific options wouldn't improve disabled people's prospects would be to stipulate that T is located such that even disabled people with the smallest enhanced option sets reach it. There are problems with this move. First, the MDV advocate must motivate this placement of T to avoid seeming ad hoc. Second, placing T so low would nullify the purported benefits of many disability-specific options. Placing T low enough to keep the Multiple Realisability Reply would imply that the majority of disability-specific options have no effect on prospects, as it implies that most disabled people's basic option sets would already be at T, meaning that, for them, disability-specific options would have no benefit.

Now, disability-specific options are presented as being general benefits of disability. That is not to say that thinkers appeal to them as evidence that disability is overall good for prospects. I mean that valuable options exclusive to disabled people are presented as benefitting the disabled people who possess them. No thinker suggests or implies that disability-specific options benefit only a subset of disabled people – namely, those with the fewest options. But the sufficiency interpretation implies exactly this, meaning that, on this interpretation, disability-specific options are not generally beneficial. In fact, this interpretation implies that they are typically of no benefit. As such, attempting to reconcile the Multiple Realisability Reply and the DSO-argument using the sufficiency interpretation ends up negating virtually all of the purported benefits of possessing disability-specific options, nullifying the DSO-argument: on this interpretation, it ceases to be an argument about the benefits of disability. Therefore, the Multiple Realisability Reply and the DSO-argument about the benefits of disability are inconsistent, even on the sufficiency interpretation.

2.4. Denying (2)

In this section, I briefly consider a different reply to the Options Argument. Some might have thought that the option set available to the disabled person is simply a subset of that available to the non-disabled person, which can be roughly represented as Figure 1 (where 0 represents the options available to the disabled person and 1 represents those available to the non-disabled person).

Figure 1. The unequal number subset model of options.

However, disabled people possess disability-specific options, and perhaps there are a similar number of disability-specific options as non-disability-specific ones (see Figure 2).

Figure 2. The equal number disparity model of options.

If this is an accurate representation, then disabled people don't possess fewer options and (2) is false.

One significant problem with this reply is that there are good reasons for thinking that, although there are some disability-specific options, the option sets of non-disabled people are (typically, other things equal) larger than those of disabled people. There is an important asymmetry between disability-specific options and non-disability-specific options: many supposedly disability-specific options are, in fact, possessed by non-disabled people. A hearing person can choose to work peacefully in noisy environments by wearing noise-cancelling headphones; a non-paraplegic person can play wheelchair-basketball, and so on (Kahane & Savulescu 2016: 777). In general, non-disabled people can experience many allegedly disability-specific options by ‘making themselves disabled’ in the relevant sense for a period, but disabled people cannot do the reverse. The claim is not that by wearing headphones the non-disabled person knows what it is like to be deaf. It is merely that many purportedly disability-specific options are also open to non-disabled people.

Once this asymmetry is highlighted, it seems that there are fewer truly disability-specific options than one might have thought. I don't dispute that there might be some options truly unavailable to non-disabled people. But this asymmetry suggests that there are more options to which disability precludes access than truly disability-specific options. This asymmetry may sound trivial to some, but I don't think it is. It highlights the crucial point that disability typically involves a net-decrease in options and that (2) is true.

One might object further that the substitute options available to non-disabled people are quite different from those available to disabled people, so shouldn't be classed as the same options. If so, then there might be room to reject (2). My response is that being able to work peacefully in noisy environments because one is deaf is of course different in some respects from being able to because one wears noise-cancelling headphones. But whether we class these as separate options depends upon our chosen level of granularity. We don't want our specification to be too fine-grained, otherwise there would be no overlap between people's option sets. And defenders of RG certainly don't want too much granularity, since their reply to the Options Argument presupposes a coarse-grained account of life's goods (Asch and Wasserman 2010: 208). However, we also don't want it to be too coarse-grained, such that importantly different options are understood as being the same. So, the question is: are these options importantly different? It's hard to see the important difference between working peacefully in loud environments because one is deaf and because one wears noise-cancelling headphones. What seems relevant is the ability to work peacefully in loud environments. As such, the onus seems to be on those who want to reject (2) to argue both that disability-specific options and the substitute options often available to non-disabled people shouldn't be classed as the same options and, further, that disabled people have as many options as non-disabled people.

So, I submit that (2) is true: the option set available to the non-disabled person is (typically, other things equal) larger than that available to the disabled person, even though there exist some truly disability-specific options. We can thus think of the options available to members of the two groups in (something like) the way depicted in Figure 3.

Figure 3. The unequal number disparity model of options.

In sum, rejecting (2) doesn't seem to be a plausible response to the Options Argument. As such, the Multiple Realisability Reply is the MDV's best response. But this reply is inconsistent with the disability-positive DSO-argument, leaving MDV advocates with a dilemma. They can advance the Multiple Realisability Reply and give up their DSO-argument, or they can keep their argument about the benefits of disability and lose their reply to the Options Argument. Either way, the MDV is weakened.

3. The benefits of more options applied to disability cases

So far, I have presented the Options Argument and discussed two MDV replies, arguing that the Multiple Realisability Reply is most promising. Here and in §4, I defend the Options Argument and reject the Multiple Realisability Reply. In this section, I present three general benefits of possessing more options – even when, without these options, one can access all of life's goods – and argue that these are relevant to disability. This suggests that the Multiple Realisability Reply is unsound.

The first benefit of possessing a greater number of options is that, with more options, a person is more likely to possess options which optimally suit their tastes. Of all ways of realising a particular good, some will suit an individual's tastes better than others. Both listening to music and playing sport provides me with pleasure, but I prefer music, so listening to music allows me to realise pleasure more effectively. As such, if I don't have the option of listening to music, this hinders my well-being prospects even though I can still realise pleasure through playing sport.

In general, while reducing options might not prevent one from realising any non-instrumental good, it might stop one from achieving as full a realisation of it as easily as one might have with more options available. This will occur when the removed options are among those that would have suited a subject's tastes better than those open to her. Now, this wouldn't happen in every case of option reduction, but it's a distinct possibility, which suggests that, in at least some cases, restricting options will hinder prospects by preventing one from being able to achieve as much value as easily as one could have with more options available, even if one can still access all of life's non-instrumental goods.

Is this benefit relevant to disability? It seems so. Although a blind person might find her occupation extremely enjoyable, she might have found more pleasure in being a pilot or a surgeon; while the person with a communicative disability might be able to make and maintain satisfying friendships without verbal communication, he might have forged more satisfying friendships were he able to communicate verbally. In general, we cannot assume that the options a person's disability removes would not suit her tastes better than those she has (Andric & Wundisch 2015: 15). So, because non-disabled people have more options, they appear (other things equal) more likely to possess options that optimally suit their tastes, improving their access to life's goods and, thus, their prospects.

The second benefit is that possessing more options makes one more likely to possess options well-suited to a variety of contexts. Certain activities are better suited to particular situations or moods. Suppose that pleasure is non-instrumentally good. Depending on my mood, different activities are more effective at achieving pleasure. Sometimes a discussion with friends works well. Other times, I feel like being alone, so a conversation with friends is not pleasurable, but I can still access pleasure through reading. Both are ways of accessing the same non-instrumental good, but possessing various options benefits me because activity a is more pleasurable at certain times while activity b is at others.

Many cases appear analogous, where two ways of realising the same good are each more effective in different contexts. For instance, one kind of aesthetic experience might be more valuable at one time than another. Mozart might be more valuable in a given context (or mood) than appreciating a Picasso, and vice versa. The point is that there is contextual variation in what is good for me and the extent to which things are good for me. If true, this suggests that reducing ways of realising non-instrumental goods can hinder well-being prospects, even if a person still has access to a given good. If I cannot talk with my friends, sometimes this won't affect me because I can read my novel instead, but there are times when this will decrease my well-being (when a conversation would have been most effective at achieving pleasure).

This benefit appears relevant to disability. Men with Spinal Cord Injury often have difficulty getting and maintaining erections, experiencing orgasms and ejaculating (Ramos & Samso 2004). This makes penetrative sex (with a natural penis) difficult for some and impossible for others, even with treatment. Penetrative sex is one method of realising pleasure and expressing emotional intimacy – both plausibly non-instrumental values. But, while some men with Spinal Cord Injury cannot have penetrative sex, they can realise pleasure through other means, and they can express emotional intimacy through words, cuddles and other kinds of sexual activity. However, if they were not disabled, they could express emotional intimacy and achieve pleasure in all these ways and through penetrative sex. And it's reasonable to think that there would be contexts or moods where these substitute activities wouldn't provide as much pleasure, or express emotional intimacy as effectively, as penetrative sex would have. The key point is that one method of realising a particular non-instrumental value cannot replace another in all contexts. Thus, it appears better for prospects to have more options to access a given non-instrumental good, which non-disabled people do.

When considering the impact of options on prospects, we must be sensitive to the way options combine. The third benefit is that having more options allows for more combinations. Different ways of combining options can be prudentially significant. One consideration regarding combinations is that certain options enhance others when exercised together. Getting married is nice, but the experience is enhanced if one is surrounded by loved ones. For many, exercising these options separately doesn't add up to the amount of pleasure they get from exercising them together. Thus, while removing any one option would not prevent one from experiencing pleasure, it might prevent one from achieving as much pleasure as one could have with more combinations of options available.

This suggests a kind of value holism. Even if one has access to options A, B and C, it seems that experiencing them separately sometimes doesn't equate to as much value as combining them. It's reasonable to think that many cases are such that options combine to realise goods more effectively than if they were exercised separately, such as when a drink enhances the enjoyment of an evening. If this is true, then increasing options to access a given good will, in some cases, allow a person to achieve value more effectively by combining options in holistically valuable ways.

This point also appears relevant to disability. Perhaps listening to certain sounds enhances certain visual experiences, or vice versa. Evidence shows that changes in lighting and music affect how food tastes.⁷ So, blindness or deafness might preclude one from experiencing certain flavours. More generally, disability might prevent one from experiencing holistically valuable combinations.

Moreover, certain disabilities – while not removing options – make options difficult or impossible to exercise together. For example, a deaf person who relies on lip-reading and sign language to communicate must choose between communicating and anything that requires explicit visual attention. Thus, they cannot communicate while (e.g.) working on a computer, watching a movie or navigating a busy city, whereas non-disabled people can do these things, potentially allowing them access to a holistically valuable combination or, less strongly, to achieve two valuable things simultaneously. The general point is that disability might prevent one from experiencing certain enjoyable combinations, either by removing options or by making options impossible to exercise together. This plausibly prevents disabled people from experiencing as much value as they would have with more options, hindering their prospects.

Another benefit of possessing more combinations is that this can allow one to schedule options in advantageous ways. Suppose that there are only two non-instrumental values: pleasure and knowledge. I can get knowledge by studying and pleasure by running, and both options are open to me. Merely exercising both options isn't all that matters; there might be benefits to scheduling them in different ways. Perhaps if I run first I feel less motivated to study afterwards, resulting in less knowledge. Whereas, if I study first, I have plenty of motivation for running and gain additional pleasure at anticipating my future pleasure. So, if I have only the option of running first and studying later, then my prospects are worse than if I also have the option of studying first and running later. The point is that possessing a greater number of combinations might help resolve some of the difficulties of scheduling options in the best ways.

This third consideration about combinations also appears relevant. The effects of many disabilities are not constant. These dynamic disabilities can make scheduling activities in the best ways problematic. Taking the example of running and studying, suppose that I have Chronic Fatigue Syndrome which causes me to experience chronic joint pain and extreme tiredness in the evenings. As such, I must perform any physically demanding activity in the first hours of the day. In this case, the combination of studying first and running later isn't open to me. If I want to run (which I do), I must do so soon after I wake, and I rarely have the energy or inclination to study afterwards. I can study, but I am likely to achieve less knowledge than if I studied in the morning. To use another example, suppose that I want to care for my family, which involves cooking my children dinner after school, helping them with their homework and putting them to bed. This option is theoretically open to me, but school finishes at 3 p.m., by which time my Chronic Fatigue Syndrome causes me to be unable to get out of bed. In this case, my disability, in conjunction with other people's schedules, causes me to be unable to exercise the option of caring for my family in the way I want to. These examples illustrate the general point that the option restriction involved in disability might prevent one from scheduling (and, more generally, combining) options in the best ways, hindering well-being prospects.

In this section, I have presented various advantages of possessing more options, even when one can already access all life's goods. I have applied these benefits to disability cases, arguing that non-disabled people seem to benefit from having more options. Taking these benefits together, it seems that possessing more options allows non-disabled people to access a larger degree of life's goods, improving their prospects as compared to disabled people, even though disabled people can still access all of life's goods. If true, this shows that the Multiple Realisability Reply is unsound. The reply says that non-disabled people don't benefit from having more options than disabled people as both groups have ‘enough options’. That is, a number large enough such that additional options would not improve prospects. But, if my arguments are correct, then disabled people don't have ‘enough options’ in this sense.

Rejecting the Multiple Realisability Reply doesn't imply that disabled people must have overall bad lives. Clearly, many disabled people live excellent lives. Also note that my argument doesn't imply that no threshold exists whereby additional options above it make no difference to prospects. It merely implies that if such a threshold exists, either disabled people are typically below it whereas non-disabled people aren't, or both disabled and non-disabled people are typically below it. Either way, non-disabled people benefit from having more options than disabled people.

Relatedly, it seems likely that there will be diminishing marginal utility of possessing additional options. That is, (other things equal) one is going to benefit more from possessing additional options when one had fewer options originally. But this does not undermine my argument; I presented three general benefits of possessing more options and offered realistic and generalisable examples of how disabled people would (and non-disabled people do) benefit from having more options. This indicates that, although diminishing marginal utility of options is likely, the benefit gained by non-disabled people is not negligible; the utility curve does not seem to sufficiently flatten out such that disabled people possess enough options to experience all the benefits of additional options. In fact, the benefits for non-disabled people seem reasonably significant. However, the fact that non-disabled people benefit from possessing more options doesn't yet show that (3) is true, or that the Options Argument is sound, as there may be offsetting costs to possessing more options.

4. The costs of more options applied to disability cases

I have argued that non-disabled people benefit from possessing more options than disabled people. However, there are sometimes costs involved in possessing additional options (Dworkin 1988; Schwartz 2016) which might outweigh the benefits. If so, then having more options would not be overall good for prospects in disability cases and (3) would be false. In this section, I complete my defence of the Options Argument by arguing that the costs don't seem likely to outweigh the benefits in disability cases. I do this by arguing that non-disability cases are disanalogous to various scenarios in which having more options hinders prospects. I'll discuss three; I don't claim this is an exhaustive taxonomy.

The first situation in which possessing more options is costly is when this causes deliberation costs to rise. When presented with options, one must understand their natures to make an informed choice. Gaining this information and deliberating about it takes time and effort, and can be stressful (Schwartz 2016). In some situations, these costs are not outweighed by the marginal utility gained from the additional options. Perhaps one example is buying a car. There are so many options and combinations: which brand? Which colour? Do I want tinted windows? If so, which level of tint? Plausibly, any marginal utility gained from choosing (e.g.) tint level seven as opposed to six is so slight that it doesn't compensate for the time, effort and stress required to make the decision.

Another deliberation cost is difficulty; additional options could make deliberation so difficult that one spends too much time deliberating or is so torn with indecision that one never makes a choice so never experiences the value of any option. Anita Silvers (2003: 482) makes this point: ‘[w]e know, for instance, that the absence of limitations can also lead to suffering. People often are better off focusing on a few fulfilling options than being torn with indecision by many glittering ones.’ The point is that additional options can be detrimental to prospects if they cause deliberation costs to increase too much.

The second situation in which additional options are costly is when possessing more options causes one to experience more regret. With no choice, there is no possibility for regret at options not exercised. However, if offered a choice, I may eventually regret my decision (Schwartz 2016). Regret is typically unpleasant. Therefore, I might be better off with fewer options. This will be so when the disvalue of regret at options not exercised outweighs the marginal utility gained from possessing and exercising additional options.

Do the additional options possessed by non-disabled people cause them to systematically experience higher deliberation costs, or more regret, than disabled people? It's hard to see why that would be true. Neither proclivity for regret nor the deliberation costs one experiences appear to be closely correlated with one's quantity of options. Option quantity might be one relevant factor, but various others influence these things.

Much depends on the nature of the choice. Plausibly, high-stakes choices typically impose higher deliberation costs. It's also reasonable to think that one is more likely to regret certain kinds of choices – perhaps high-stakes choices, or other kinds of choice. But it's unclear that non-disabled people systematically face more high-stakes (or any other relevant kinds of) choices merely because of their increased number of options. This would need to be argued for.

Another relevant factor is the kind of person one is. Schwartz (2016: 79–80) distinguishes between maximisers (those that aim for the best possible choice) and satisficers (those that aim for ‘good enough’ choices). He argues that maximisers have a higher proclivity for regret (Schwartz 2016: 88–90) because they will always wonder whether they made the optimal decision, whereas satisficers are typically happy that their decisions are ‘good enough’. A similar point seems to apply to deliberation costs. If one is a maximiser (or neurotic, has OCD, etc.), then one seems more likely to experience costly deliberation because one wants to select the best possible option (cf. Ullmann-Margalit & Morgenbesser 1977: 780). However, if one is a satisficer (or carefree, nonchalant, etc.), then this seems less likely (Schwartz 2016: 150). If this is true, then unless there is reason for thinking that disabled people are more likely to be satisficers and non-disabled people are more likely to be maximisers, we shouldn't assume that non-disabled people have a higher likelihood of experiencing regret or significant deliberation costs merely because they possess more options. We might expect this if disabled people had no choice whatsoever. But, crucially, the range of relevant cases does not include people with no choice. The relevant range is between non-disabled individuals with many options and disabled people with fewer, but still many. Disabled people typically have a large number of options – more than enough to experience regret and deliberation costs – and whether they experience as much of these things as non-disabled people will greatly depend on what kinds of choices they face and the kinds of people they are.

If we assume – as seems reasonable given lack of evidence – that these other relevant factors are distributed evenly among disabled and non-disabled people, we can still ask whether non-disabled satisficers (or maximisers) are more likely to experience regret or hefty deliberation costs than disabled satisficers (or maximisers). It's not clear why they would be. Satisficers typically don't experience much regret unless it's obvious they made a bad decision (Schwartz 2016: 88) and we're not justified in thinking that non-disabled people typically make worse decisions than disabled people. So, we turn to the question of whether non-disabled maximisers are more likely to experience regret than disabled maximisers. It seems unlikely that proclivity for regret (even for maximisers) is closely correlated with one's number of options. People (especially if they're maximisers) often feel regret even when they chose between only a few options. In fact, when we experience regret at options not exercised, we typically focus on a small number of options. We regret not choosing that job or that car. We typically don't regret not choosing dozens of options (unless we obviously made a bad decision). We therefore need only a small number of attractive options to experience regret, which disabled people (maximisers) typically have. So, it seems unlikely that they will experience significantly less regret than non-disabled maximisers.

Similarly, we need only a small number of options to experience high deliberation costs. People rarely deliberate about every available option. Out of dozens of options, familiar considerations of taste, ease of access, cost, etc. often rule out most of them without much deliberation. After this initial ‘screening out’ seems to be when the costliest deliberation typically takes place. We generally agonise over a small number of options, if we agonise at all. If this is true, it's unlikely that non-disabled people (maximisers) generally experience significantly more deliberation costs than disabled people (maximisers), because both groups typically have more than enough options to experience hefty deliberation costs.

To put the point another way, it's plausible that there is diminishing marginal disutility of additional options when it comes to deliberation costs and proclivity for regret. That is, supposing that possessing more options does make one more susceptible to experiencing regret or high deliberation costs, it seems that (other things equal) the difference in deliberation costs, or proclivity for regret, between having two and three options is going to be greater than that between having fifty and fifty-one. These costs don't seem likely to increase uniformly with number of options, but instead at a diminishing rate. And because disabled people have a substantial number of options, and need only a small number to experience these costs, the difference between the deliberation costs and regret that they experience and those experienced by non-disabled people doesn't seem like it is going to be significant enough to outweigh the benefits of additional options outlined in §3.

In sum, many factors other than number of options influence the deliberation costs one experiences and how likely one is to experience regret, including the nature of the decision and the kind of person one is. These factors will likely interact with each other – and with one's number of options – in a complex interplay, and there is no evidence that they are distributed such that either group is likely to experience significantly more costs. Moreover, if we assume that these other relevant factors are evenly distributed among the groups, my suggestion is that the disutility curve of these costs seems to flatten out sufficiently such that non-disabled people aren't likely to experience many more negatives than disabled people. Disabled people typically have enough options to experience these costs, if indeed they are the kinds of people disposed to experience them, and additional options don't seem likely to make a significant negative difference in the range of relevant cases. This contrasts with the benefits of options I discussed in §3. I argued that disabled people would benefit from possessing the additional options they would possess if they were non-disabled, even though diminishing marginal utility of options is likely: the suggestion is that the utility curve of the benefits seems likely to be steeper for longer than the disutility curve of the costs, such that the benefits non-disabled people gain from possessing a larger number of options are more significant.

I am thus arguing that, as it stands, there is no good case to be made that non-disabled people typically experience (i) more deliberation costs or regret than disabled people, or (ii) enough extra deliberation costs or regret to counterbalance the benefits of having more options.

I have discussed two costs involving the mere possession of options. I will now discuss a cost involving the value gained from exercising options. I said above that a benefit of possessing more options is access to additional valuable combinations. However, valuable options might also combine to create disvalue. If non-disabled people are likely to experience more disvalue holism in virtue of having more options, then perhaps possessing more options doesn't overall improve their prospects.

Suppose that additional options do typically grant one access to additional disvaluable combinations. One might think that, if access to additional valuable combinations improves prospects (as I argued), then access to additional disvaluable combinations hinders prospects symmetrically. However, I think this is false because of an important asymmetry: people seek out value and seek to avoid disvalue. This has two implications.

First, through various mechanisms – such as education, reflection and reasoning – we often seek to learn, and come to know prior to experience, which combinations are disvaluable. Thus, while additional disvaluable combinations might be accessible to me because of additional options, I am unlikely to experience many of them as I will seek to avoid them.⁸ By contrast, I am more likely to experience the valuable combinations open to me as I will seek them out.

There might be some combinations that I must experience to learn of their disvalue. This is where the second implication becomes relevant. Once I experience a combination, I typically know whether it's valuable or disvaluable, and the rational person will continue combining valuable combinations and avoid combining disvaluable combinations again. Disvaluable combinations are therefore likely to be experienced far fewer times than valuable combinations. Thus, the net effect of having more (valuable and disvaluable) combinations is likely to be positive. A rational person will seek out and repeatedly experience valuable combinations but seek to avoid and rarely return to disvaluable ones. As such, non-disabled people seem to benefit overall from having more combinations available.

In this section, I argued that non-disability cases are disanalogous to three situations in which having more options hinders prospects. Together with my arguments in §3, this leads me to conclude that (3) seems true: the net effect of option reduction in disability cases seems to be negative. I have motivated this by arguing that non-disabled people gain greater access to life's goods in virtue of having more options than disabled people (§3) but don't seem likely to experience offsetting costs (§4). I haven't established anything about the size of the difference in prospects between the groups or offered completely conclusive arguments that the negatives of option reduction must be greater than the positives in disability cases. The subject matter of this debate – that is, the prospects of counterfactual individuals – makes that almost impossible. What I've argued is that the Options Argument appears to be sound and to identify one potentially weighty cost of disability. My primary aim is to shift the burden of proof onto those (e.g. Moller 2011; Schroeder 2016; Wasserman & Asch 2013) who deny that the option reduction involved in disability hinders prospects and, more broadly, MDV proponents. Until now, the onus seemed to be on BDV proponents to engage with MDV replies to the Options Argument, particularly the Multiple Realisability Reply. Given my arguments, the onus is back on MDV proponents either to reject them or to identify something good about disability that counterbalances the negative effects of option reduction.

Now, this may be possible given the scope of my argument. I only considered the effects of valuable option quantity, but many other things might influence the prospects of each group, such as option quality and diversity, as well as other non-option-related things. I have tried to isolate the effects of one factor, insofar as that is possible. And consideration of other factors might imply that disabled people's prospects are not overall worse, especially given that disabilities are ‘high-impact traits’ that tend to have a substantial impact on how one's life unfolds (Campbell and Stramondo 2017: 166).

Defending the Options Argument supports the BDV but doesn't establish its truth. To do that, one must identify all costs and benefits of disability. However, if MDV proponents cannot reject my arguments or identify a counterbalancing benefit of disability, then the Options Argument gives good grounds for belief in the BDV. My argument shows that, as it stands, the Options Argument provides a defeasible argument for the BDV.⁹

5. Objections, Replies and Residual Issues

In this section, I respond to two objections and note residual issues.

5.1. Adaptive preference

Recall, I said that because non-disabled people have more options, they are more likely than disabled people to possess options which optimally suit their tastes. One might object by pointing out that people's preferences tend to adapt depending on their circumstances and capabilities. So-called ‘adaptive preference’ might mean that (e.g.) blind people don't prefer things that require sight. Perhaps, then, the options that optimally suit a disabled person's tastes are going to be those that they possess. If so, then disabled people are just as likely as non-disabled people to have access to options that optimally suit their tastes and the first benefit discussed in §3 is undermined.

It's true that disabled people are among those affected by adaptive preference – almost everyone somewhat adapts their preferences to their capabilities, which is probably best explained by the human ability to respond to environmental change (Mitchell 2018).¹⁰ But absolute preference adaptation isn't ubiquitous. That is, disabled people at least sometimes prefer things that their disability makes difficult or impossible. For example, people with Ocular Albinism (a visual disability causing one to ‘see the world [as though] through a cloudy plastic bag’) sometimes desire to create visual art, which is made more difficult by their condition.¹¹ We know, further, that disabled people sometimes desire to do things their disability makes impossible (especially if they acquire disability). We know that some quadriplegic people desire to navigate the world unaided, that some people with Chronic Fatigue Syndrome desire to be more active than their condition allows, that some blind people desire to see the faces of their children, and so on. Thus, although disabled people's preferences do adapt, adaptive preference doesn't undermine the first benefit mentioned in §3 as disabled people don't only prefer things that they can do.

5.2. Other relevant factors

In §4, I considered whether other factors (such as the nature of the choices one faces and the kind of person one is) affect the disvalue experienced because of possessing more options, arguing that there was no good reason to think these factors are weighted in such a way as to have a systematic and disproportionate effect on the disvalue experienced by members of each group. However, I did not consider other factors in §3, and it's possible that other factors influence the value gained from possessing additional options such that non-disabled people do not benefit from this.

First, note that when I considered other factors in §4, I was considering the disvalue of merely possessing options. But my discussion in §3 concerned the value of exercising options, and it's less clear to me how, and which, other factors might influence this. However, my response is similar to my argument in §4: it's likely that other factors affect the value gained from exercising options. These factors might include a person's tastes, which options their disability removes and gives access to, which options they have access to for other reasons, how many options they exercise, their decision-making capabilities and more. But there appears to be no compelling reason to think that these factors are distributed among disabled and non-disabled people in such a way as to mitigate the benefits of possessing more options outlined in §3. This would need to be argued for and I leave it open to MDV proponents to do so.

I cannot make a convincing case about the effects of all possible relevant factors. My argument is not bulletproof, but that was not my aim. My aim was to engage with the Options Argument and the Multiple Realisability Reply and to encourage more discussion on this topic. If MDV advocates point out factors that I have not considered which mitigate the benefits I mention in §3, then I have succeeded.

5.3. Residual difficulties

Finally, I'll briefly note two residual issues.

First, I discussed various situations in which having more options is worse for prospects and argued that non-disability cases are disanalogous; however, I didn't present an exhaustive list. This leaves the possibility open that there are situations I didn't consider which are analogous to non-disability and imply that the option reduction involved in disability isn't correlated with worse prospects.

Finally, it's possible that how a person thinks of themselves impacts the value they gain from possessing options or, more generally, their well-being. And disability is something that often enters into someone's self-conception; perhaps, then, conceiving of oneself as a ‘disabled person’ or a ‘deaf person’ in some way influences how the possession of options affects prospects. This potential complication is not accounted for in my discussion.

Conclusion

In this article, I defended the Options Argument and rejected the Multiple Realisability Reply by arguing that, in disability cases, possessing a larger quantity of valuable options overall improves well-being prospects. As such, the Options Argument appears to be sound and provides a defeasible argument for the BDV by identifying a potentially significant cost of disability. This shifts the burden of proof onto those who reject the Options Argument and, more generally, MDV proponents. They must reject my arguments here or identify something good about disability that counterbalances the negative effects of option reduction. If they cannot, then the Options Argument offers good grounds for belief in the BDV.¹²