Autonomy, Respect, and the Rights of Persons with Disabilities in Crisis

Article 12(2) of the UN Convention on the Rights of Persons with Disabilities guarantees persons with disabilities ‘the right to legal capacity on an equal basis with others in all aspects of life.’ In its General Comment on Article 12, the Committee on the Rights of Persons with Disabilities claims that this guarantee necessitates the abolition of the world’s dominant approach to mental capacity law. According to this approach, when a person lacks the mental capacity to make a particular legal decision at the material time, the state authorises a third-party to make it on her behalf. The Committee declares such substituted decision-making a violation of the Convention’s guarantee of legal capacity on an equal basis, and therefore demands it be replaced by an allegedly non-discriminatory alternative called supported decision-making. This article argues that we should reject the Committee’s demand in its current form, because the most influential version of the new approach to supported decision-making suffers from serious conceptual flaws that make it inferior to the mental capacity approach. However, I then argue that the Committee’s demand stems from a legitimate ethical concern with respect and equality that ought to inform the CRPD’s implementation process.


Introduction
This is a critical moment in international disability law. 1 With the implementation process of the Convention on the Rights of Persons with Disabilities in full swing, signatory states are scrambling to meet their obligations under the treaty. Among the many ethically charged controversies surrounding this process, none is more hotly contested than the debate over the right of persons with cognitive and psychosocial disabilities to make their own decisions in times of crisis. At the centre of this debate is Article 12 of the Convention, which enjoins signatories to 'recognize that persons with disabilities enjoy the right to legal capacity on an equal basis with others in all aspects of life.' 2 According to a growing chorus of critics, 3 the world's dominant approach to legal capacity stands in violation of this demand.
Legal capacity is the capacity to make decisions with legal effect. Today's dominant approach to the concept forges a strong link between mental capacity 4 and legal capacity: when P lacks the mental capacity to make a particular legal decision at the material time, the state restricts (or removes) P's legal capacity to make that decision and authorises a third-party decision-maker to make a substitute decision on P's behalf. Call this the mental capacity approach to legal capacity (henceforth referred to as MC).
The copyright line for this article was changed on 10 March 2017 after original online publication.
Critics object that MC falls foul of Article 12 on grounds of discrimination. Persons with disabilities are far more likely to have their legal capacity restricted (or removed) on the basis of a mental capacity assessment, and, critics claim, such interventions result in trauma, stigma, and objectification. This is not just a dry legal criticism but also an ethical demand made by and for the world's largest minority. 5 On these grounds, the Committee on the Rights of Persons with Disabilities has called on states to abolish MC and to replace all regimes of substitute decision-making with an alternative model called supported decision-making. 6 In what follows, I argue that we should reject the Committee's demand in its current form. For the most influential version of the new approach to supported decisionmaking suffers from conceptual flaws that make it less transparent and respectful than MC. And it seems that MC already has the resources it needs to overcome the apparent legal and ethical problems associated with substitute decision-making. My goals, however, are not purely negative. I disagree with those who dismiss the Committee's demand as muddled radicalism from the fringes of the disability community. 7 Instead, I argue that it stems from a legitimate ethical concern with respect and equality that should inform the current implementation process. Attending sensitively to this concern, I argue, will improve our odds of reducing the harms caused by legal capacity denials, thereby bringing us closer to meeting our legal and moral obligations.

The Mental Capacity Approach
During the second half of the 20 th century, MC emerged from the interaction of the legal literature on informed consent and medical ethics research on patient autonomy. 8 The approach was in part motivated by criticisms of earlier models. For instance, critics accused guardianship regimes of excessive restrictionas they often divest persons with disabilities of the right to make any legal decisionsand statusbased approaches of direct discriminationas they permit the removal of legal capacity on the basis of a diagnosis of disability. MC avoids excessive restriction by limiting interventions to a particular decision at the time it must be made. And it combats direct discrimination by relying on mental capacity assessments that can apply to any member of the general population. That is, it uses a 'functional test' or an 'abilities test' that any person can failwhether she lacks mental capacity due to disability or some other cause (e.g. intoxication)and that persons with disabilities can pass, if they demonstrate certain decision-making abilities. 9 There are different versions of this approach around the world, but they tend to share three core features: 1. When a person P 10 lacks the mental capacity to make a particular legal decision at the time it must be made, the state restricts (or removes) P's legal capacity to make that decision. 2. Judgments that P lacks the requisite mental capacity to make the relevant legal decision are based on a functional test designed to assess P's decision-making abilities. 11 3. Finally, if a functional test demonstrates that P lacks the mental capacity to make the relevant decision, and a decision must be made, a substitute decisionmaker decides on P's behalf.
Some proponents of MC will object to my characterisation of the first core feature here. In response to the criticism that MC restricts legal capacity on an unequal basis, they argue that MC doesn't restrict (or remove) P's legal capacity at all: it merely recognises the factual absence of decision-making abilities, and so acknowledges the factual absence of the ability to make decisions with legal effect. There's logic in this: if we index legal capacity to decision-making abilities, then the absence of the latter does imply the absence of the former. However, the law operates with a presumption of capacity; that is, prior to any intervention, it presumes that P possesses mental capacity. Thus, prior to any intervention, P often enjoys a kind of de facto legal capacity. She makes what are at least presumed to be decisions, and these putative decisions will have legal effect until the state intervenes and restricts (or removes) her right to make them. It is at best highly artificial to suggest that such interventions in no way restrict (or remove)but merely recognise the absence of -P's legal capacity. In many cases, P suffers the loss of legal freedoms she has previously enjoyed and that she would go on enjoying were it not for such interventions. Obscuring this fact forfeits one of MC's greatest ethical strengthsits transparency about the actual tradeoffs substitute decision-making entails. As I will show below, this is a strength that MC's new rival lacks. It would be a shame to abandon this transparency simply to offer a facile rejoinder to critics.
The ideal of transparency also explains why MC forges such a strong link between legal capacity and decision-making ability: the law needs a clear standard to determine when it can (or cannot) fairly hold an agent accountable for her decision. For MC, decision-making ability underwrites this standard, because such ability is a necessary condition for making autonomous decisions. To be clear, the idea is not to establish a global condition of autonomy, but only to strive to ensure that particular decisions are autonomous at the time they are made. This approach to legal accountability is analogous to the rational abilities view of moral responsibility. According to the rational abilities view, an agent can be held responsible for what she does only if she has the abilities i) to recognise good reasons for acting and ii) to translate those reasons into decisions and actions. 12 Analogously, the mental capacity approach maintains that an agent can decide autonomously and so be held legally accountable for her decision only if she has the decision-making abilitiese.g. understanding, reasoning, the ability to weigh risksthat allow her to recognise good reasons and translate them into a decision. 13 As O'Shea and Freyenhagen (2013) point out, this is not a normatively neutral approach but rather a substantive one: to make autonomous decisions an agent must be able to see the world for what it is and be suitably sensitive to the right sorts of reasons. 14 However, the approach is not strongly but rather weakly substantivethe agent need not recognise good reasons as such, whatever those might be, but rather what she considers good reasons. This is the general standard that mental capacity law sets for autonomous decision-making: at the time a decision is made the agent must be able to decide with reason-sensitive decision-making abilities in light of her own sense of what matters. The functional test mentioned above is designed to determine whether agents meet this standard. If an agent fails that test at a given time, then, on this view, she cannot decide autonomously and the law cannot fairly hold her accountable for any decision she makes. Thus, if a decision must be made at that time, it falls to the state to authorise a substitute decision-maker to decide on her behalf and to take accountability for the decision. This approach to legal capacity can be found in several human rights instruments, and it is the most commonly used approach throughout the world. 15 It has been championed in particular by Western democracies because it promotes two fundamental aims of the liberal stateto respect individual autonomy and to protect non-autonomous citizens facing significant risks.

On an Equal Basis
According to the Committee's General Comment on Article 12, even if MC rules out excessive restriction and direct discrimination, it still fails to meet the Convention's anti-discrimination provisions. Article 2 of the CRPD defines discrimination on the basis of disability as 'any distinction, exclusion or restriction on the basis of disability which has the purpose or effect of impairing or nullifying the recognition, enjoyment or exercise, on an equal basis with others, of all human rights . . .' 16 The language of 'purpose or effect' here mirrors the distinction between direct discriminationwhich has discriminatory intentand indirect discriminationwhich has discriminatory effects without the surface signs of intent. Since direct discrimination isn't the problem, the charge of indirect discrimination becomes an obvious candidate.
Indirect discrimination occurs when treatment that is not on its face discriminatory nevertheless disproportionately and unfavourably impacts a particular group. 17 Notice that disproportionate impact alone is not enough to establish indirect discrimination; one must also show that the treatment is unfavourable. This poses a problem in the current case. It's undeniable that the mental capacity approach disproportionately impacts persons with disabilities: the vast majority of legal capacity denials happen to persons with cognitive and psychosocial disabilities. 18 But whether the treatment is unfavourable hinges significantly on the views of those affected by it; and those affected are divided on the subject.
What makes legal capacity denials unfavourable? Answers to this question typically centre on two types of harm: trauma and stigma. With regard to trauma, critics focus on coercive psychiatric interventions, with some arguing that these practices 'violate the universal prohibition of torture'. 19 As evidence for this, they cite cases where persons with disabilities have 'died [in treatment] or been permanently scarred' by it, 20 highlight similarities between legal definitions of torture and the experience of forced drugging and electroshock, 21 and share the stories of service users who recount their time in psychiatric facilities in terms of suffering and humiliation. 22 Although the torture comparison is open to challengesas torture aims to inflict harm while treatment has therapeutic goalsthere's no doubt that such experiences can be traumatic. When it comes to stigma, critics argue that legal capacity denials reinforce stereotypes that persons with disabilities are incompetent, and that they cause self-stigma by teaching persons with disabilities to view themselves through the lens of these stereotypes. 23 This not only can have a devastating impact on their self-regard and psychological wellbeing, but it can also undermine their social life and career goals. 24 But what proportion of those affected find the treatment unfavourable? The empirical evidence, though limited, sheds some light on this question. A large prospective study of involuntary psychiatric patients in 11 countries found that after one month 55% of patients thought their admission was the right decision and after three months that number rose to 63%. 25 The same study also found that in countries with stronger patient autonomy safeguardslike the functional tests described aboveapproval rates were at 71% after one month and 86% after three. Finally, the findings of smaller-scale, qualitative studies report similar results. One found that 47% viewed their forced treatment positively and 20% were ambivalent about it, 26 while another found that up to 77.8% viewed it as beneficial. 27 While this evidence is hardly conclusive, it suggests that at least half of those subjected to coercive interventionsand significantly more when legal mechanisms are in place to protect patient autonomyview the treatment as favourable. It's also worth noting that this data comes from the most ethically fraught subset of legal capacity denials, i.e. forced treatment. The bulk of legal capacity denials won't involve such an extreme conflict between P and substitute decision-makers, which suggests that overall the favourability statistics could be even higher. It seems, then, that we have evidence for disproportionate impact and unfavourable impact, but it's unclear that there's a case for disproportionate unfavourable impact.
However, critics have another line of argument to establish indirect discrimination. This is the claim that legal capacity denials are intrinsically objectifying, because they treat persons with disabilities as 'objects to be pitied and cared for rather than as subjects before the law'. 28 The idea here is that deciding for oneself is essential to being the subject of a life, and so to override someone's decision is to quash her subjectivity and thereby treat her like an object. Legal capacity denials are thus non-remediably unfavourable. Moreover, if legal capacity denials disproportionately impact persons with disabilities and they are intrinsically unfavourable, then they necessarily result in disproportionate unfavourable impact.
But legal capacity denials aren't intrinsically objectifying. Imagine P is admitted to the emergency room suffering from delusions due not to disability but rather to having ingested some psychoactive substance. The medical team determines P needs a lifesaving operation, but P refuses treatment because he believes the doctors intend to kill him. To restrain Pby the least coercive means availableand then operate would not be to treat him like an object. In fact, such an intervention would acknowledge P as the subject of a life worth living, while letting him die due to a temporary loss of autonomy would be inhumanly callous and myopic.
Even if we defuse the universal claim, however, isn't it still possible that in the majority of cases mental capacity denials objectify persons with disabilities? The empirical evidence cited above suggests otherwise. If such denials objectified the majority of those they affect, it's unlikely that so many people affected by them would view them in favourable terms. Moreover, any attempt to discount such positive appraisals with appeals to false consciousness and the like would be viciously self-contradictory in this case, as the core demand of the CRPD is that we respect the views of persons with disabilities. At this point, then, it's unclear whether substitute decision-making in fact results in indirect discrimination. 29 From an ethical point of view, however, this technical legal issue isn't all-important. Perhaps as more evidence comes in, a broader consensus will emerge to the effect that substitute decision-making results in indirect discrimination, creating new legal duties for the Convention's signatories. But even if that never happens, the issues addressed in this section clearly give rise to moral duties vis-a-vis persons with disabilities who feel harmed by legal capacity denials. No amount of harmbe it trauma, stigma, or objectificationis morally permissible if we can prevent it by making what we all agree are reasonable and materially feasible adjustments to our practices. 30 So if this is a case of indirect discrimination, then signatories of the CRPD might have a legal duty to reform certain aspects of MC. But even if it isn't a case of indirect discrimination, we still seem to have a moral duty to make reasonable adjustments to our substitute decision-making practices that will reduce (if not eliminate) unnecessary harms suffered by persons with disabilities.
Critics of MC have proposed what such reasonable adjustments would look likewhat is often referred to in the relevant literature as the 'new paradigm' of supported decision-making. 31 According to some proponents of the new paradigm, a particular model of supported decision-making could overcome the apparent problem of discrimination by drastically reducing (if not eliminating) the harms associated with substitute decision-making.

The Will and Preferences Approach
In general terms, supported decision-making is the practice of helping someone make a decision for herself. Certain proponents of the new paradigm, however, take this benign practice to a controversial extreme. The first step in this direction is to take a relational view of decision-making as a fundamentally collaborative enterprisewe make up our minds with our significant others, and these relationships can enhance (or impair) our decision-making ability. 32 The second step is to embrace the social model of disability inscribed in the Preamble of the Convention. 33 On this view, disability is not the deficit of a particular individual but rather the result of her personal characteristics interacting disadvantageously with her social context. 34 If a person with disabilities is incapable of exercising her legal capacity, it is often due to a mismatch between her personal makeup and the contingent arrangement of her social environment.
So far, none of this necessarily conflicts with MC. In fact, supported decision-making plays an integral role in most mental capacity legal frameworks, 35 and many jurisdictions require that substitute decision-makers take every reasonable measure to help a person make her own decision before deciding on her behalf. 36 Moreover, courts have found that individuals who lack the capacity to make a solo decision can sometimes retain the ability to make their own decisions with support.
The contrast and controversy emerge, however, when defenders of the new paradigm take a third step away from traditional approaches by indexing legal capacity not to cognitive decision-making abilities but rather to the non-cognitive ability to have a will or preference. Call this the will and preferences approach to legal capacity (henceforth referred to as WP). 37 Before outlining how the approach is supposed to work, it's important to note that WP is not the official representative of the new paradigm, which is a big tent under which many people with diverse views gather. Moreover, although its general comments do reflect the influence of WP, the Committee hasn't explicitly endorsed the approach. However, WP merits our attention because it has exercised considerable influence on this debate and it has some worrisome implications.
According to WP, for the state to attribute legal capacity to P, she must meet the following minimum threshold for decision-making capacity: she must have a discernible will or preference. And P meets this minimum threshold when it's possible for P or P's support person to make a decision that is consistent with P's diachronic identity, i.e. P's identity across time. In other words, to have a will or preference is to be able to make a decision that is consistent with your diachronic identity, or for someone else to be able to do this for you. According to WP, where X is a legal decision, P has legal capacity to make decision X if and only if P (or someone else) can make a decision about X that is consistent with P's diachronic identity.
WP thus replaces the rational abilities view of autonomous decision-making with a non-cognitivist mesh theory. According to mesh theories, actions are free when they instantiate the right sort of mesh or fit between elements of a person's psychology. 38 The best-known example of such a theory is Frankfurt's hierarchical view, which understands freedom of the will as the mesh between a person's first-order desirewhat he wantsand his second-order volitionwhat he wants to want. 39 On WP, the mesh required for a decision to be autonomousthat is, for a decision to be P's ownis between the decision and P's identity. An autonomous decision doesn't require any cognitive abilities, then; it only requires that P's decisionwhoever makes itmesh with her diachronic identity. P can make her own decision with supporta supported decision in the standard sense. But it's also possible, on this view, for support to realise P's decision for her without her even being aware of it. This is supposed to be a new kind of supported decision called a facilitated decision.
What if it's impossible to know what a person wants? Say, for example, P is unconscious, a decision must be made, and no one knows anything about P. 40 According to WP, even a facilitated decision made without knowledge of P's preferences should be guided by an effort to discern what P would do. This seems to be what the Committee has in mind when it writes, 'Where . . . it is not practicable to determine the will and preference of an individual, "best interpretation of will and preference" must replace "best interests" determinations.' 41 In all circumstances, then, the facilitator strives to realise P's will and preferences. With this claim, WP can argue that legal capacity is a universal feature of human nature and a fundamental human right. For on this view, P can exercise her legal capacity in every case regardless of her condition. This is why proponents of WP view it as the least coercive means to promote the autonomy of persons with disabilities. As they see it, WP never infringes on P's autonomy.
But respect for autonomy is not the only state aim at stake herethere's also the matter of safety. Does WP's insistence on the right of persons with disabilities to, as the Committee puts it, 'take risks and make mistakes' commit it to a doctrine of total non-interference? 42 And if so, has it purchased the right to legal capacity at the cost of serious harm to vulnerable people? 43 This is not WP's stance on intervention. Even if it left the legal capacity of persons with disabilities intact, total non-interference would have harmful effects on persons with disabilities, violating their right to 'physical and mental integrity', 44 their 'right to life', 45 and/or their right to 'the highest attainable standard of health'. 46 Thus, WP recognises the need for a protective intervention mechanism. 47 But how can we intervene in P's affairs without reneging on the promise to never infringe on her autonomy? WP offers a kind of workaround here: if P's decisions are autonomous only when they mesh with her diachronic identity, then to intervene on her behalf when this mesh fails to obtainin order to re-establish itis not to infringe on her autonomy but rather to restore it. 48 Bach and Kerzner illustrate this with the following example: imagine a young man with a psychosocial disorder becomes hostile, aggressive, and paranoid to the point that his supporters can no longer discern a will or preference consistent with his identity. 49 Since his behaviour fails to satisfy the diachronic criterion, it no longer expresses his agency, and it's permissible for supporters to intervene and initiate procedures to help him recover. 50 WP thus leaves room for 'constraining choice and personal dignity' 51 and admitting people to a facility for assessment 'without their consent'. 52 Like MC, then, WP aims to respect individual autonomy and safeguard non-autonomous agents facing significant risks. If a decision meshes with P's diachronic identityregardless of her mental capacity -WP does not permit interference, even if it poses serious risks. However, if no such mesh obtains, WP permits us to intervene if P faces significant risks. By this logic, WP represents a less harmful alternative to MC because i) it never restricts P's legal capacity, and ii) it safeguards P's welfare by intervening when she is non-autonomous and faces significant risks.
This conclusion hinges crucially on the premise that supported decision-making allows P to retain full legal capacity even when the intervention mechanism kicks in. 53 This premise, I submit, is false. Even if we grant the legitimacy of indexing legal capacity to the mere presence of a will or preferencea point I don't grant and will return to in the next paragraphthere is little reason to think that supported decision-making could realise P's legal capacity in every instance. As just seen, WP allows the state to 'constrain P's choice', to admit her to hospital without her consent, and to treat what she says as non-expressive of her will and preferences. If P explicitly says she wants to X and support workers override her decision because X-ing doesn't mesh with their interpretation of her diachronic identity, they thereby deny her the right to make a decision on the grounds that they understand who she is better than she does. The claim that this is not a coercive denial of P's legal capacity is a kind of doublespeak where coercion doesn't always mean coercion. What WP calls facilitated decisionsdecisions made for P without P's input (or even against P's input)are not a new kind of supported decision. They are substitute decisions masquerading as support.
This criticism only scratches the surface of the objections one could level against WP's diachronic mesh theory. Does it make sense to call a decision your own when you don't understand itor when you don't know that it's been madesimply because it meshes with your identity? The reasons to doubt this are too numerous to canvass here, so I'll just identify five salient worries. First, many decisions seem to lose their character as autonomous without a certain level of understanding. For example, consider how different the Socratic narrative would read if Socrates drank the hemlock without knowing it would kill him, or if he was unconscious and his suicide was a facilitated decision. Would he still be, as Wayne Martin describes him, one of philosophy's greatest 'autonomy heroes'? 54 Secondly, it's unclear why we should privilege diachronic identity as the ultimate criterion for autonomous decisions. People with episodic identities, for instance, might reject such a criterion. Thirdly, in many cases, a diachronic criterion would be difficult, if not impossible, to use in practice. If P's life lacks sufficient narrative coherence, or P's care workers lack sufficient knowledge about P's life, a diachronic criterion could do little to fix P's will and preferences regarding a particular decision. Fourth, this approach cannot furnish a standard to determine when the law can (or cannot) fairly hold an agent accountable for her decisions. If P decides autonomously whenever her decision meshes with her diachronic identity, is she therefore accountable for all such decisions, even when she doesn't understand them or isn't involved in making them? Finally, WP raises serious worries about manipulation. One of the major criticisms of mesh theories is that the source of the relevant psychological mesh can undermine the freedom that that mesh allegedly instantiates. For example, Fischer and Ravizza worry about Frankfurt's so-called 'willing addict'who wants to use his drug of choice and also wants to want to use itasking, What if the drugs cause the relevant second-order desire? 55 In other words, is the willing addict still free if he only wants to want heroin because he's addicted to it? To raise a similar objection in this context, does the autonomy-conferring mesh obtain in a facilitated decision whenever the support person says it does? What happens when manipulation dons the mantle of support?
It's also worth noting here that once it becomes clear that WP authorises substitute decisions (aka facilitated decisions), it loses much of its apparent novelty. It has not invented a new kind of supported decision-making; and it hasn't developed a new decision-making standard either. The diachronic criterion it champions is very similar to a standard commonly used within the mental capacity paradigm, i.e. the substituted judgment standard. This standard is typically interpreted as a rule that substitute decision-makers ought to endeavour to make the decision that P would make if she had mental capacity. And the diachronic standard looks like little more than a particular interpretation of that rulespecifically, it understands 'the decision P would make' as a decision that is consistent with P's will and preferences interpreted in light of her diachronic identity. Moreover, this is remarkably similar to other existing interpretations of the substituted judgment standard, e.g. Brudney's view that the point of this standard is to promote the authenticity of P's life by making decisions consistent with the life she's lived thus far. 56 The most recent in-depth discussion of WP by Szmukler and Bach even deploys this language of authenticity, arguing that the aim of an intervention should be 'to give effect as much as is possible to the "authentic" will and preferences [of P].' 57 The goal behind WPto put P's preferences at the centre of decisions that shape her lifeis ethically laudable, but some of its implications are unacceptable. In particular, WP seems less transparent than MC, because it obscures the trade-off being made in substitute decisions: the coercive denial of P's legal capacity to protect her from harm. As the reader will recall, WP's fault here is similar to the ethical blunder defenders of MC commit when they claim that MC does not restrict but merely recognises the factual absence of P's legal capacity. This lack of transparency is unacceptable in both cases.

Conclusion -Transforming Practice
To challenge WP's dialectical contribution is not to question the ethical impulse that animates it, nor is it to disparage the larger movement it aims to represent. My goal is rather to suggest that, thus far at least, the movement's promise lies not in novel theoretical contributions but in its potential to transform the way we actually treat persons with disabilities in practice. The idea that P's preferences should determine the decisions made on her behalf already exists in the literature on substitute decision-making. But in practice this idea is too often crowded out by legal worries about liability and the well-intentioned but overly paternalistic impulse to protect vulnerable people. Our inability to put this idea into practice has some steep ethical costs, some of which I mentioned above, namely, trauma, stigma, and objectification. But I have yet to address what I take to be its greatest ethical cost, a particular form of harm that advocates of the new paradigm, including proponents of WP, are rightly striving to reduce.
The harm in question has to do with the fact that some persons with disabilities experience the denial of their legal capacity as a fundamental denial of recognition respect. Here I understand recognition respect in Darwall's sense as the fitting response to a person's dignitythe disposition to give appropriate weight in your practical deliberations to the other person's moral worth. 58 To deny a person recognition respect in this sense is to deny her dignity as a person. And the experience of such denial goes hand in hand with the sense that one is not an equal member of the human community. That is, we think people deserve equal respect because in some deep sense they are equals, and so to be denied recognition respect is to be denied status as an equal. 59 Such denial constitutes an insidious and destructive kind of harmthe painful sense that you exist on a moral plane beneath those around you. 60 Why do some persons experience the denial of their legal capacity as a denial of recognition respect while others don't? In most cases, those who view the treatment as respectful do so because they agreein real-time or in retrospectthat it was the right thing to do. To return to my earlier example, the patient who refuses a life-saving operation due to a temporary drug-induced paranoia will not view his forcible restraint as disrespectful once he recovers his wherewithal. He will admit that he wasn't himself and will be grateful for the help. This is surely why WP allows substitute decisions when they mesh with P's diachronic identity. Such interventions don't deny P recognition respect as a person; they only refuse to respect a temporary state of P's mind that he too will come to reject. However, there is a special category of persons with disabilities who will tend to experience the denial of their legal capacity as a denial of recognition respectpeople subjected to coercive treatment who cannotin the moment or afterwardssee it as a legitimate form of assistance. This category includes a) people who are in fact subjected to illegitimate coercive treatment and b) people who objectively need help but whose cognitive predicament prevents them from recognising that need. This is backed by a simple intuition: whether you're factually right (case a) or wrong (case b), if you see yourself as capable of making your own decisions, then having those decisions forcibly overridden by someone who claims to know better cannot but be experienced as a denial of recognition respect and a fundamental insult to the sense of yourself as a person of equal social standing. It's not hard to see how the prolonged experience of this sense of disrespect and demotion could destroy a person's sense of self worth. As Judge Eldergill wrote recently in her decision in the Court of Protection of England and Wales, . . . individual liberty is of the same fundamental importance to incapacitated people who still have clear wishes and preferences about where and how they live as it is for those who remain able to make capacitous decisions . . . most individuals wish to determine and develop their own interests and course in life, and their happiness often depends on this. 61 The harms associated with the experience of being denied recognition respect and the loss of one's sense of self-determination can at times be worse than the significant risks P would face were the state not to intervene. The failure to appreciate this fact in practiceoften driven by liability worries and an overweening paternalistic impulseis one of the major failings of contemporary care practice that the current disability rights movement stands poised to correct. And it's best chance of doing so, in my view, is not to abolish MC and replace it with WP, but rather to highlight ideas within the mental capacity paradigm that need strengthening in order to achieve the laudable ethical goal of putting P's preferences at the centre of the decisions that shape her life. This will call not for novel theoretical constructs but rather for concrete policies and practices that motivate care professionals to treat persons with disabilities as persons entitled to recognition respect who prize their sense of liberty as much as any other person. Instead of scrapping the world's dominant approach to legal capacity, then, we need to devote more time to thinking about how to make the essential practice of substitute decision-making as respectful as possible, especially when it comes to the trade-off between recognition respect and individual welfare. capacity, treatment consent capacity, competence, decision-making capacity, decision-making skills, and so on. These terms designate the particular mental functions or abilities required to make decisions, and whatever differences there may be between them, for my purposes here I will treat them as, in effect, synonymous.