Introduction
It is widely believed in the West that in crafting health policy care should be taken to ensure that persons are informed about the expected risks as well as the anticipated advantages of the procedures or treatments that they are considering undergoing or receiving. It is, for example, widely accepted that it should be a matter of health policy that persons give their informed consent to their medical treatment if they are competent to do so.Footnote 1 It is similarly widely held that the prospective subjects of medical research should be informed of any health risks that it might involve.Footnote 2 It is also generally accepted that persons who are considering donating body parts to others (for example, plasma or a kidney) should be allowed to do so only after they have demonstrated that they understand the medical risks involved.Footnote 3
Just as it is widely believed that health policy should be crafted so that persons are able to give their informed consent to their medical treatment, so too is it widely believed that this requirement is justified by a concern for the moral value of autonomy. Understood broadly, a person is autonomous with respect to her decisions or actions if she directs herself to make or to perform them. This account of autonomy is broad enough to be acceptable to all autonomy theorists.Footnote 4 The reasoning that supports this belief is simple. It begins with the claim that “[t]o the extent that a person’s understanding is based on false beliefs about that which would otherwise be relevant to an understanding of the action, performance of that action is less than fully autonomous.”Footnote 5 This claim is often then supplemented with a second: “[W]hen I choose to remain ignorant of relevant information, I am choosing to leave whatever happens to chance. I am following a path without autonomy.”Footnote 6 A person who acts on the basis of false beliefs or out of ignorance of information that is relevant to her decision has her autonomy compromised with respect to the decisions that she thereby makes and the actions that she consequently performs.Footnote 7 Autonomy is of great moral value. To protect it, persons must have the opportunity to acquire all of the available information necessary for them to be autonomous (or, at least, sufficiently autonomous) with respect to their health-care decisions.Footnote 8 Hence, health policy should be crafted to ensure that this occurs.Footnote 9
This argument is simple, elegant, persuasive, and widespread. It is also unsound. As I will argue in this essay it is not true that a person’s autonomy with respect to her decisions (or her consequent actions) will be compromised if she makes them on the basis of false beliefs or out of ignorance. It does not follow, though, from the failure of the above argument that there is no reason to provide persons with at least the opportunity to have access to the available information relevant to their decisions concerning their medical treatment. This is because the epistemic requirements for a person to give her consent to her treatment are more stringent than those that must be met for her to be autonomous with respect to her consent. For a person to consent to her prospective treatment she will (typically) need to be provided with information about it. However, this move away from autonomy and toward consent to justify the requirement that health policy ensures that persons are able to give their informed consent to their treatment is not merely to offer a different theoretical justification for a standard practice. Recognizing that the moral concern for autonomy cannot justify the ethical requirement of informed consent raises a fundamental question for health policy: If concern for autonomy cannot be the guiding principle that undergirds the requirement of informed consent, then what is?
The standard view is plausible
I term the view that a person’s autonomy with respect to her decisions and consequent actions will be compromised if she acts on false beliefs or out of ignorance of information she would deem relevant to her decision the “Standard View.” The Standard View is both intuitive and theoretically plausible. A patient makes it clear to his physician that he will only consent to take medications that will have no adverse sexual side effects. Believing that it is critical that his hypertension is treated, his physician prescribes him medication that she knows has a common side effect of causing erectile dysfunction. Knowing that he would be unwilling to take this medication if he were to realize this, she falsely informs him that it will have no adverse sexual side effects. As a result of this deception, he takes the medication. Because he was deceived into doing so, he was not autonomous with respect to this action.Footnote 10
How does successfully subjecting a person to deception compromise her autonomy? One explanation will begin by noting that in this case it was the physician, not the patient, who was the font of the decision that the patient should take the prescribed medication to address his hypertension. She took steps to ensure that his beliefs were such that he would make the decisions and perform the actions that she desired him to make and to perform. To the extent that this was so, he was under her control. He was thus not self-directed, autonomous, with respect to these decisions and actions.Footnote 11
This is not the only possible explanation of why this patient’s autonomy with respect both to his decision to take the medication and to his consequent act of taking it was compromised. One might hold that it was compromised owing to his false beliefs about the medication’s side effects. Support for this can be drawn from an example developed by Tom Dougherty.Footnote 12 Dougherty outlines two cases in which one person, Candace, asks another, Courtney, to store her antique skis for her. The skis were once owned by Josef Stalin. Courtney is unwilling to store items that were previously owned by dictators. In the first case, Candace knows both that the skis were owned by Stalin and that Courtney is unwilling to store the former property of dictators. To secure Courtney’s consent to store the skis she withholds information about their history from her. As in the case of the patient with hypertension, Courtney’s autonomy with respect to her consent to store the skis was compromised by Candace’s successful deception.
In the second case, neither Candace nor Courtney is aware of the skis’ history. Dougherty notes that, in this case, we would hold Candace blameless for thwarting Courtney’s interest in not storing the property of former dictators, as she did not deceive her into storing the skis. Candace was “justifiably ignorant” that the skis were owned by Stalin.Footnote 13 Candace’s blamelessness here is thus the result of features of Candace, not Courtney. However, because this is so, we cannot infer from Candace’s blamelessness that Courtney was autonomous with respect to her (putative) consent to store the skis. After all, Courtney’s mental states are no different in this case than in the first, where (owing to Candace’s deception) she was not autonomous with respect to her consent to store them. It is thus plausible to hold that the question of whether Courtney is autonomous with respect to her consent “depends on facts about Courtney—it depends on the nature of her mental attitudes or utterances.”Footnote 14 In both cases, Courtney’s mental attitudes and utterances were the same. Hence, because Courtney was not autonomous with respect to her consent to store the skis when Candace deceived her into doing so, she was not autonomous with respect to her consent to store them when Candace was blameless. It is thus Courtney’s ignorance of the history of the skis that renders her nonautonomous with respect to her consent to store them. Hence, when a person is successfully deceived into making a decision and consequently performing an action, her autonomy with respect to them is compromised owing to their being made and performed on the basis of her ignorance of “deal-breakers”—that is, facts about a situation that would make a decisive difference to someone’s decision to perform an action.
It is plausible that neither the deceived patient nor Courtney (in either case) was autonomous with respect to their relevant decisions or actions. The claim that false beliefs about (or ignorance of) deal-breakers compromise a person’s autonomy with respect to it is thus intuitively plausible.Footnote 15 But the Standard View is not only intuitively plausible; it also has theoretical support.Footnote 16 In writing of how ignorance could be “a threat to self-government” (that is, autonomy), Sarah Buss states:
If doing Y is constitutive of doing Z, then if I authorize myself to be moved by the desire to do Y because I mistakenly believe that doing Y is a way of not doing Z, then there is an obvious sense in which I have not authorized myself to do what I am now doing when I am moved by the desire to do Y.Footnote 17
Gerald Dworkin holds that Buss’s account of how a person’s ignorance could compromise her autonomy with respect to the actions that she subsequently performs “seems to me the strongest reason for saying that … being deceived … [is] incompatible with a decision being one’s own.”Footnote 18 He also notes that Buss’s account provides reason to hold that a person who acts on a false belief that has not been brought about through the deception of another has diminished autonomy with respect to her actions.Footnote 19 If a person authorizes herself to perform a certain action because she mistakenly believes that that action will achieve a particular end, then “there is an obvious sense in which” she has not authorized herself to do what she is doing. On this analysis, the patient who took the medication prescribed to him because he believed that this was a way of treating his hypertension without affecting his sexual performance did not authorize himself to take it when he was moved to do so by the desire to treat his hypertension. He was thus not fully self-directed with respect to this action, and so his autonomy with respect to it was compromised.
This analysis of why a person’s ignorance of relevant deal-breakers would compromise her autonomy with respect to her consequent decisions and actions also accounts for why persons who are not subject to deception could still have their autonomy with respect to their decisions and actions compromised owing to their ignorance. Courtney’s storing of the skis constituted storing the former property of a dictator. She authorized herself to be moved by the desire to store the skis as she mistakenly believed that this would be an act that did not involve storing the former property of a dictator. Since this is so, “there is an obvious sense in which” she did not authorize herself to store the skis when she was moved by the desire to agree to do so.Footnote 20 Courtney’s mistaken belief that the skis were the sort of thing that she was willing to store thus compromised her autonomy with respect to her agreement to store them. She attempted to direct herself to consent to store skis that had no connection to a dictator. She did not direct herself to consent to store skis that were formerly owned by a dictator. Because she did not direct herself to do what she actually did, her autonomy, her self-direction, was compromised with respect to her relevant decisions and consequent actions. It was thus her false belief about a deal-breaker that compromised her autonomy with respect to her consent to store the skis and her consequent storage of them. How she came to have this belief—for example, as a result of being deceived into having it or a result of making an unwarranted assumption about the history of the skis—is irrelevant.
The revisionary view is also plausible
The Standard View is both intuitively plausible and has theoretical support, but the same claims can be made about what I will call the Revisionary View. According to the latter, that a person has false beliefs about deal-breakers or makes decisions in ignorance of their presence does not in itself compromise her autonomy with respect to the decisions that she makes (or the acts that she consequently performs) on the basis of these false beliefs or in ignorance of these deal-breakers.
The Revisionary View is intuitively plausible. Consider the Canadian Health Coalition’s (CHC) opposition to the legalization of offering financial compensation to prospective blood or plasma donors in Canada.Footnote 21 Some (perhaps all) members of the CHC believe that allowing donor compensation will compromise Canada’s ability to be “self-sufficient in blood and blood products,” that it will compromise Canada’s ability to provide “free and universal” access to blood and blood products, and that it will compromise the safety of the Canadian blood supply system.Footnote 22 These beliefs motivated some of the members of the CHC to advocate the legal prohibition of donor compensation. They also led them to lobby Health Canada to revoke the licenses of for-profit plasma collection agencies operating in Canada as well as to lobby Health Canada and Canadian Blood Services to work together to increase uncompensated plasma donation within Canada.Footnote 23 To achieve these ends the members of CHC engaged in a series of activities, including lobbying the Canadian Government, campaigning to educate the public about their views, and soliciting donations to fund their political efforts.Footnote 24
However, the beliefs held by these members of the CHC about the effects that donor compensation will have on the Canadian supply of plasma are all false. If Canada were to offer sufficient financial compensation to current and prospective plasma donors, the amount of plasma that it would secure would be greater than that which it can secure absent such compensation.Footnote 25 Allowing donors to be compensated would facilitate rather than compromise Canada’s ability to be self-sufficient in plasma. Similarly, compensating plasma donors would not compromise Canada’s ability to provide “free and universal” access to plasma-derived medical products (PDMPs) to those patients within its borders who need them. Offering donor compensation would no more compromise Canada’s ability to provide “free and universal” access to PDMPs than would the offer of compensation to health-care professionals to entice them to work in Canada’s health-care system compromise its ability to offer “free and universal” health care. Finally, compensating plasma donors would not compromise the safety of the Canadian plasma supply.Footnote 26
Yet despite the falsity of their beliefs concerning the effects that allowing donor compensation would have on the Canadian plasma supply, the members of the CHC who act on them still appear to be fully autonomous with respect to their decisions to oppose donor compensation and their consequent actions in support of this.Footnote 27 They identify the goals that they wish to achieve, determine the steps that they need to take to realize them, plan accordingly, and then put their plans into action. That the beliefs on which they act are false does not appear to compromise their autonomy, their self-direction, at all.
The intuitive plausibility of the Revisionary View gains further support from the observation that persons with religious convictions often make medical decisions based on their faith, but many of these religious convictions conflict. Jehovah’s Witnesses, for example, believe that the Bible prohibits accepting blood transfusions.Footnote 28 This belief motivates them to decline them. Catholics, however, do not believe that the Bible prohibits blood transfusions, and so they accept them. At least one of these beliefs is false.Footnote 29 However, both Jehovah’s Witnesses and Catholics can, intuitively, be fully autonomous with respect to their respective decisions to decline or accept transfusions. The Revisionary View is thus intuitively plausible.
The Revisionary view also has theoretical support. It is possible that a member of the CHC who opposes compensating plasma donors volitionally endorses, for example, her first-order desire to write to the Canadian Health Minister to oppose compensating plasma donors; she both wanted to possess the desire and wanted it to be that desire that moved her to act. She could thus meet the conditions that Harry Frankfurt outlines in his seminal article “Freedom of the Will and the Concept of a Person” for her to be autonomous with respect to the desire that moves her to act.Footnote 30 Similarly, it is possible that the decisions she makes concerning her efforts to oppose donor compensation were not the result of “manipulation, deception, the withholding of relevant information, and so on” and that she had not previously renounced her “independence of thought or action” prior to developing them. She could thus be fully autonomous with respect to them, on Gerald Dworkin’s account of autonomy.Footnote 31 She could also qualify as autonomous with respect to the desires that she acted on that led her to work against donor compensation, on John Christman’s analysis of autonomy. She did not resist (and would not have resisted) the development of the desire to oppose compensating plasma donors had she attended to the process by which she formed it. This lack of resistance did not occur under the influence of factors that would inhibit self-reflection. Her self-reflection involved no self-deception and was minimally rational. She had no conflicting beliefs or desires that would significantly alter her behavior.Footnote 32 She could also qualify as autonomous with respect to her relevant preferences, on Laura Waddell Ekstrom’s analysis of autonomy. A preference, Ekstrom explains, is a desire “(i) for a certain first-order desire to be effective in action … and (ii) that is formed in the search for what is good.”Footnote 33 For Ekstrom, a person will be autonomous with respect to those of her preferences that cohere with her “other preferences and acceptances.”Footnote 34 When a person (such as a member of the CHC) acts on such preferences, she will be acting in a way that is characteristic of her—for example, she is acting in a way that manifests her compassion toward her fellow Canadians—and she will be able to provide reasons for her action, such as donor compensation will adversely affect the Canadian supply of plasma.Footnote 35 On Ekstrom’s account, she will thus be autonomous with respect both to her preferences and her consequent actions.
Objections to the standard view
We are now at an impasse. Both proponents of the Standard View and proponents of the Revisionary View can provide equally intuitively plausible—but contrary—accounts of the relationship between false beliefs, ignorance of deal-breakers, and autonomy. Furthermore, the proponents of both accounts can provide theoretical support for their respective views.
The above discussion has served a useful purpose, though, for it has identified the core disagreement between proponents of the Standard View and proponents of the Revisionary View. Proponents of the Standard View assert that the question of whether or not a person is autonomous with respect to her decisions and consequent actions can be settled solely by appeal to facts about her. This is the lesson to be drawn from the discussion of Candace and Courtney. Proponents of the Revisionary View, however, deny this. This is the lesson to be drawn from the example of the deceived patient.
Seamus and the monstrous raccoons
Identifying this core disagreement enables us to see why the Standard View is mistaken. On the Standard View, a person’s autonomy with respect to a decision (and a consequent action) will be diminished if she makes that decision (and hence performs the action) on the basis of a false belief about a “deal-breaker.” It does not matter how she came to have that false belief, for example, whether she was deceived into having it or not. The question of whether or not she is autonomous with respect to her decisions (and consequent actions) depends solely on facts about her. In this discussion the pertinent fact about her is whether or not she has a false belief about a deal-breaker. But whether a person has a false belief about something external to her would not depend merely on facts about her, but on facts about the external world. The falsity of a CHC member’s belief about the safety of PDMPs produced from plasma sourced from compensated donors depends on facts about PDMPs, not on facts about the CHC member who holds this belief. The (possible) falsity of a Jehovah’s Witness’s beliefs about the will of God depends on facts about the will of God, not on facts about the Jehovah’s Witness. It is thus possible on the Standard View for two persons with identical mental states to differ in the degree to which they are autonomous with respect to a decision that they make. Consider a revised version of the example of the deceived patient. This time, the medication is just as his physician presents it to him: an effective treatment for hypertension with no adverse sexual side effects. Because this patient is now not ignorant of facts about the medication that would be deal-breakers for him, he could (on the Standard View) be fully autonomous with respect to his decision to take the medication. This is not because of facts about him, though, but because of facts about the medication.
The simple observation that the truth or falsity of a person’s beliefs will depend not only on facts about her but also on facts about the external world should lead us to reject the Standard View. To see this, consider the (true-ish) example of Seamus and the Monstrous Raccoons.Footnote 36
Growing up in Scotland, Seamus’s only exposure to North American fauna was through Disney cartoons. As a result, he came to believe that raccoons and chipmunks were roughly the same size—about 3–5 inches long, weighing about 3–5 ounces. The first time he encountered (regular-sized) raccoons in America, he thought they were far, far larger than they should be—over 100 times as large as they should be, in fact. Deeply concerned (and not a little frightened) he called the police to report the presence of monstrous raccoons.
Seamus’s decision to call the police and his consequent actions were motivated by his false beliefs about raccoons. On the Standard View, then, his autonomy with respect to them was compromised. On its face, this claim is plausible. That the raccoons he saw were normal-sized was a deal-breaker for Seamus. Had he been aware of this, he would not have called the police. Seamus’s ignorance of the presence of this deal-breaker ensured that he failed to perform the act that he appeared to have authorized himself to perform: to alert the police to the presence of monstrous raccoons. Because there “is an obvious sense” in which he failed to authorize himself to perform the act that he actually performed—that is, alerting the police to the presence of regular-sized raccoons, when he did not desire to do this—Seamus did not do what he directed himself to do; he did not alert the police to the presence of monstrous raccoons. Because he did not direct himself to alert the police to the presence of regular-sized raccoons, his autonomy with respect to his actions was compromised.
So far, the Standard View’s account of why Seamus’s autonomy was compromised with respect to his decision and consequent action is plausible. But the veneer of plausibility that the Standard View enjoys with respect to this example should not lead us to lose sight of the fact that the question of whether a person is autonomous with respect to a decision that she makes or an action that she performs is a question about whether she directs herself to make or perform it. It is a question about whether she is self-directed with respect to it. To answer this question, then, we must make essential reference to facts about her. We need to know whether she was self-directed with respect to this decision or this action—that is, whether she directed herself to make it or perform it—or not. This question cannot be settled solely by reference to facts about the world that are external to her that make no essential reference to her.
Consider Seamus and the raccoons again. On the Standard View, Seamus’s autonomy with respect to his decisions and actions was compromised because he had false beliefs about the size of raccoons. This is a fact about Seamus; it is a fact about his mental states. So far, so good. This approach to answering the question of whether Seamus was autonomous with respect to his decisions and actions appears to be addressing the question by essentially referring to facts about Seamus. However, on the Standard View, it is the falsity of Seamus’s beliefs that compromises his autonomy with respect to the actions that he performed as a result of his having them. This is where things go awry. The falsity of Seamus’s beliefs was determined not by facts about Seamus, but by facts about raccoons, just as, in the example, above, the falsity of the patient’s belief about his medication was determined by facts about its side effects. Proponents of the Standard View are thus committed to claiming that the answer to the question of whether Seamus is autonomous, self-directed, with respect to his act of informing the police about the presence of what he believed to be monstrous raccoons is determined by facts about raccoons. But the question of whether Seamus was self-directed, autonomous, with respect to his decisions and actions cannot be settled by reference to facts about raccoons. It can only be settled by reference to facts about Seamus. Similarly, the question of whether a patient is autonomous, self-directed, with respect to her taking a medication that she has been prescribed cannot be settled by reference to facts about that medication. The Standard View must thus be wrong. It locates in the wrong place the answer to the question of whether a person is autonomous, self-directed, with respect to her decisions and actions.
Rejecting internalist accounts of autonomy
In rejecting the Standard View in this way, I am not claiming that the question of whether a person is autonomous with respect to a decision or an action can be settled solely by reference to her motivational state. I am not endorsing an internalist approach to autonomy on which the question of whether or not a person is autonomous with respect to, for example, a desire, decision, or action can be settled by reference only to the structural features of her motivational set.Footnote 37
Consider again the example of the physician who deceived her patient into taking medication for his hypertension by deliberately falsely informing him that he would have no adverse sexual side effects. Let us call this example A. The patient’s autonomy with respect to his decision to take the medication was compromised, as it was his physician, not he, who was the font of the decision that he should take it. It was she, not he, who directed his decision-making. To the extent that this was so his self-direction, his autonomy, with respect to his decisions and consequent actions was compromised.
Now consider an alternative version of this example in which the physician believes that the medication will have no adverse sexual side effects. Let us call this example B. The external behavior of this second physician and her initial counterpart are identical, as is that of their patients. However, in this case the physician does not deceive her patient into taking the medication; she simply provides him with the facts about it as she understands them. In taking the medication he, and not she, is the font of his decision to do so. His autonomy is not usurped; he remains self-directed with respect to his decisions. The difference in the degree to which these patients are autonomous with respect to their decisions and consequent actions is determined in part by the mental states of their respective physicians. The degree to which a person is autonomous with respect to their decisions and consequent actions thus appears to be determined in part by the mental states of others.Footnote 38
While examples A and B provide us with reason to believe that (to put it crudely) my autonomy can depend on your mental states, they do not tell the whole story of how this could be so. The physicians in examples A and B differed in their beliefs about the effects of the medication. A realized that it had adverse sexual side effects while B did not. That A knew that the medication had these adverse side effects and failed to disclose these to her patient does not, however, establish that she thereby usurped his autonomy with respect to his decision to take the medication.
Consider a third example, C. In this example the physician provides her patient with a list of the side effects of the medication that has been compiled by a third party. This list inadvertently fails to mention its adverse sexual side effects. Like the physician in example A, the physician in example C knows that these side effects are a possibility. She also resembles the physician in example A in that she fails to disclose these side effects to her patient. But unlike the physician in example A, her failure to disclose is not motivated by a desire that the patient take the medication. She is completely indifferent as to whether or not he takes it. Her failure to disclose is simply the result of laziness.Footnote 39
Unlike the physician in example A, the physician in example C has not decided to make it the case that her patient will take the medication that she has prescribed; she has merely recommended that he do so. By refraining from deciding what he will do, she does not usurp his autonomy with respect to the decision as to whether or not he should take the medication. To see this, consider the different responses of the physicians in examples A and C if their patients were to question them directly about the possibility of adverse sexual side effects. In A, the physician would lie, reassuring her patient that they would not occur. In C, however, the physician would tell the truth. The physician in example A would deceive her patient so that he will make the decisions that she has decided that he will make, while the physician in example C would not. The physician in example A is thus attempting to usurp her patient’s autonomy, his self-direction—and if, as a result of this deception, his patient acts as he desires, he will succeed in this usurpation—while the physician in example C is not. A person’s (true) belief that another person has a false belief that the first person believes that she will rely on to make a decision is thus not sufficient for the first person to have usurped the second person’s autonomy with respect to that decision. For this usurpation to occur the first person must also intend to control the decision that the second person makes, for example, through deceiving him into making the decision that she desires that he makes. For this intended usurpation of his autonomy to be successful the decisions that he made must have been made as a result of the actions that she performed to fulfill her intent to control him.
It is important to note that in the above examples the facts about the external world—that is, the mental states of the physicians—that are relevant to assessing whether the patient is autonomous with respect to his decisions and consequent actions still involve his (that is, the patient’s) mental states in that they make essential reference to them. The facts about the external world that are relevant to assessing whether or not this patient’s autonomy was compromised with respect to his decision to take the medication were some of the mental states of his physician: those which had as their intentional objects the relevant mental states of the patient, such as his beliefs concerning the effects of the medication. Facts about the external world can thus be relevant to determining whether or not a person is autonomous with respect to his actions or decisions, but those facts must make essential reference to his mental states. To return to the crude characterization I offered above, my autonomy can depend on your mental states, provided that your mental states are about mine. An account of autonomy on which the question of whether or not a person is autonomous with respect to her decisions or actions can be settled without reference to her mental states is just as mistaken as internalist accounts that refer only to the structural relationships that hold between a person’s mental states.
Facts about nonintrinsic properties can be deal-breakers
Proponents of the Standard View mislocating the answer to the question of whether Seamus was autonomous with respect to his decisions and actions is not, however, the only problem that their view of the relationship between autonomy, false beliefs, and ignorance faces. The Standard View faces a further problem that stems from its acknowledgment that deal-breakers are not limited to the intrinsic properties of the things that feature in a person’s decision-making—for example, that this medication has an ingredient that causes erectile dysfunction—but can also include their nonintrinsic properties, for example, that these skis were once owned by Stalin.
Because they are not limited to the intrinsic properties of persons or objects, a deal-breaker could be the fact that an option that I consider preferable to that which I am considering is available to me. For example, if I would prefer to date Jack rather than Jill in a situation where I would be willing only to date one person, Jack’s willingness to date me would be a deal-breaker for my willingness to date Jill. On the Standard View, then, if I were ignorant of the fact that Jack is willing to date me or if I were falsely to believe that he would not date me, then my autonomy with respect to my agreeing to date Jill would be compromised.
Proponents of the Standard View are thus committed to holding that unless a person consents to an event that is ideal for her given her available options, she will not be autonomous with respect to her consent.Footnote 40 This implies that many of the choices that a person makes when she is presented with a range of options—for example, dating partners, diet, or medical care—would be ones with respect to which her autonomy is compromised. This is implausible; it treats autonomy as a success concept. But autonomy is not a success concept.Footnote 41 Persons can be fully autonomous with respect to their decisions—for example, to pay for name-brand medication—even if they would not have done so had they been aware of other options—for example, for their insurance to pay for generic medication that was chemically identical to the name brand. This implausible implication of the Standard View provides us with a second reason to reject it.Footnote 42
Autonomy and consent
The Standard View is thus mistaken. A person acting out of ignorance or on the basis of false beliefs is no bar to her being fully autonomous with respect to her actions, even if she is ignorant of or has false beliefs about deal-breakers.
This conclusion might appear to have rather alarming implications for health policy. As noted at the start of this essay, persons who work on health policy often claim that health-care professionals should secure their patients’ informed consent to their treatment to protect their ability autonomously to choose for themselves.Footnote 43 On the Revisionary View of the epistemic conditions required for autonomy, securing a person’s informed consent is no longer necessary for this. Others who work on health policy express ethical qualms about certain forms of medical research involving human subjects where their risks are unknown, and so the subjects must agree to participate in ignorance of what they are letting themselves in for.Footnote 44 On the Revisionary View, though, a valuer of autonomy should hold that such qualms are misplaced. Such ignorance would not compromise the ability of potential research subjects to be fully autonomous with respect to their consent to participate. Many persons object to the buying and selling of human body parts (such as plasma or kidneys) on the grounds that the prospective sellers will typically be ignorant of the risks they will incur through such sales, and so concern for their autonomy should lead us to preclude them from entering into such contracts. In response, defenders of such markets note that prospective participants in them should be educated about the risks that they might run.Footnote 45 On the Revisionary View, though, such ignorance is no bar to one’s selling a kidney.
The Revisionary View, then, seems to have startling and implausible implications for health policy. However, accepting this view’s account of the epistemic conditions that a person must meet to be autonomous with respect to her decisions and consequent actions does not have the startling implications for health policy that it might appear to have. This is because the epistemic conditions that must be met for a person to be autonomous with respect to her decision to perform an action and her consequent attempted performance of that action might differ from those that are required for her successfully to perform that action. (The implications that this has for health policy will become clear below.) Consider marriage. In many traditions, in order to marry, a person has to utter a certain form of words. Someone who fails to do so does not become married. A person who utters the wrong form of words due to not knowing the correct form would fail to marry; she would fail to perform the act that she authorized herself to perform. The possession of false beliefs can thus preclude a person from performing an action that she autonomously attempted to perform. Strictly speaking, then, persons will be autonomous with respect to their decisions to perform actions and autonomous with respect to their consequent attempts to perform those actions. Her false beliefs would not compromise her autonomy with respect to her attempt to marry, but they would preclude her from succeeding in doing what she attempted to do, that is, marry.
The same point can be made with respect to consent. Although one might assume that the epistemic conditions that must be met for a person to be autonomous with respect to her (attempt to) consent are the same as those that must be met for her to consent, this assumption is unwarranted.Footnote 46 Consider a person who has a religious conviction that human skin should never be broken intentionally. She does not realize that a biopsy on a growth on her lungs involves an incision, believing instead that it involves the placing of a tube down her throat.Footnote 47 Her attending physician explains the advantages and disadvantages of performing a biopsy on the growth. He is unaware of her misconception concerning the nature of a biopsy and so does not correct it. She considers the information that she has been given and appears to consent to the procedure. Because she directed herself to make the decisions and perform the actions related to this event, she was autonomous with respect to them. However, when she wakes from the anesthesia and realizes what occurred, she would rightly deny that she had consented to being cut open. While her ignorance of this deal-breaker did not preclude her from being autonomous, self-directed, with respect to her (attempt to) give her consent to the procedure, it did preclude her from consenting to the procedure. The epistemic conditions that must be met for a person to be autonomous with respect to a decision and her attempt to perform the actions consequent to it are thus weaker than those that must be met for her to consent.
What are the epistemic conditions that must be met for a person to consent to an event (for example, a biopsy or the donation of her plasma)? Loosely put, a person will have an idea in mind of to what she intends to consent. To the degree that the events that transpire as a result of her (attempted) consent differ from what she had in mind, the less likely it is that she consented to what occurred.Footnote 48 At first blush, then, a person whose attempted consent to the occurrence of an event was based on a false belief about a deal-breaker related to that event would not have successfully consented to its occurrence. This is so in the biopsy case. Owing to her ignorance of what a biopsy was, the patient did not consent to the procedure that she underwent. But it is not only facts about an event’s intrinsic properties that could be deal-breakers. Facts about relational properties could be, too. This is shown in Dougherty’s example of Candace, Courtney, and Stalin’s skis. The history of the skis was a deal-breaker for Courtney, and so she did not consent to their storage.Footnote 49 However, not every deal-breaker will defeat a person’s attempt to consent to an event. Consider again the observation that on the Standard View autonomy is treated as a success concept. On the Standard View, owing to my ignorance of Jack’s romantic availability to me, I was not autonomous with respect to my decision to date Jill. This implication of the Standard View is implausible. It is similarly implausible to hold that I did not consent to date Jill, owing to my ignorance of Jack’s romantic availability. A person’s ignorance of alternatives to an event to which she is considering consenting when the availability of those alternatives would be a deal-breaker for her is thus not relevant to determining whether she consented to the occurrence of the event in question.Footnote 50
A person will be precluded from consenting to the occurrence of an event if she is ignorant of facts about the intrinsic properties of that event that are deal-breakers for her. She will also be precluded from consenting to the occurrence of an event if she is ignorant of facts about the relational properties of that event that are deal-breakers for her, for example, that these skis were owned by Stalin. The epistemic conditions that must be met for a person to consent to an event are thus more stringent than those that must be met for her to be autonomous with respect to her attempt to consent to that event. Certain facts about the external world can preclude a person from consenting to an event when she attempts to do so, even though they do not similarly preclude a person from autonomously attempting to perform an action. Careful attention to the scope of the concepts of consent and autonomy reveals that this is not surprising. When I consent to the occurrence of an event, my consent is conditional on the world being a certain way. If the world transpires not to be that way, then my attempt to consent is unsuccessful. By contrast, when I autonomously make a decision or perform an action, I am directing myself to attempt to make the world a certain way. If my attempt fails and the world does not become the way I attempted to make it, this will not retroactively render my attempt any less an exercise in self-direction.
It might be tempting to conclude that the concern for autonomy in discussions of health policy should be replaced by a concern with securing consent, but this temptation should be resisted. A person who is deceived or manipulated into giving her consent could still meet the conditions for her consent to be valid if the deception or manipulation did not engender within her any false views concerning either the intrinsic or relational properties of the event in question. My consent to date Jill could still be genuine, even if it was the result of Jenny’s deceiving me into thinking that Jack was not romantically available to me. If I were later to discover Jenny’s deception, I would likely say that I would not have consented to date Jill had I known that Jack was available; I would not say that I now realize that I did not consent to date Jill at all. But owing to its deceptive origins, we would not say that such consent was authoritative. We also would not say that it was authoritative, because the person concerned was not autonomous, self-directed, with respect to her consent. Remember, successfully subjecting a person to manipulation or deception will compromise her autonomy with respect to the decisions or the actions that she was manipulated or deceived into making and performing. Rather than jettisoning our moral concern with autonomy and replacing it with a moral concern for consent, then, we should instead be concerned with both autonomy and consent. We should be concerned with ensuring that persons do not merely consent to their treatment, but that they are autonomous with respect to their consent, that is, that they give their autonomous consent.
Conclusion
This has been a long essay and some readers might now feel rather cheated. Such a thicket of words to arrive at the apparently anodyne conclusion that health policy should be crafted to ensure that persons can give their informed consent to their treatments! But this conclusion is not as anodyne as it might appear. First, it is often claimed that persons should give their “autonomous consent” to their treatment (medical or otherwise).Footnote 51 However, until now there has been little or no recognition of the fact that this is a portmanteau requirement, with the epistemic conditions that must be met for a person to consent to her treatment differing from those that must be met for her to give her autonomous consent. Second, the above discussion has established that the common justification for requiring that persons give their informed consent to their treatment—that is, that this is required to protect their autonomy—cannot be correct. The epistemic conditions that must be met for a person to be autonomous with respect to her decisions and actions are weaker than those that must be met for her to give her autonomous consent. A person could thus be autonomous with respect to her (attempt to) consent, even if she has sufficient information to be able to give her autonomous consent. The more robust epistemic requirements that are required for a person to give her autonomous consent to her treatment are thus not required (as they are too stringent) for the protection of autonomy. The justification for these robust epistemic requirements thus cannot be grounded on a concern for autonomy.
This last observation leads to the third reason why the conclusion of this essay is not as anodyne as it might appear. It raises a fundamental question for health policy. If a concern for the moral value of autonomy cannot justify requiring persons to be able to give their autonomous consent to their treatment, what can? Two possibilities suggest themselves. Such consent could be justified out of a concern to protect and promote well-being. This view is based on the defeasible empirical claim that securing a person’s autonomous consent to their treatment would be (in rule-utilitarian fashion) the best way to protect and promote their well-being.Footnote 52 Alternatively, one might eschew such empirical claims to argue that the need to secure a person’s autonomous consent to their treatment is required by respect for personhood.Footnote 53 Which of these two approaches is correct (if, indeed, either is) is the subject for another day.
Acknowledgments
This essay’s great-grandfather was presented at a Bioethics Workshop organized by Samuel Director in January 2023; its grandmother was then presented at the Romanell Bioethics Workshop at the University of Buffalo in March 2023. Its father was then discussed by other contributors to this volume in May 2023. (Unlike ships, academic essays in English are gender-fluid.) I thank the discussants on all three occasions for their very helpful comments and criticisms. I also thank an anonymous referee for their very constructive suggestions on how to improve the father of this essay.
Competing interests
The author declares none.
