Felicitas Kraemer draws on the experiences of patients undergoing deep-brain stimulation (DBS) to propose two distinct and potentially conflicting principles of respect: for an individual's autonomy (interpreted as mental competence), and for their authenticity. I argue instead that, according to commonly-invoked justifications of respect for autonomy, authenticity is itself in part constitutive of an analysis of autonomy worthy of respect; Kraemer's argument thus highlights the shortcomings of practical applications of respect for autonomy that emphasise competence while neglecting other important dimensions of autonomy such as authenticity, since it shows that competence alone cannot be interpreted as a reliable indicator of an individual's capacity for exercising autonomy. I draw from relational accounts to suggest how respect for a more expansive conception of autonomy might be interpreted for individuals undergoing DBS and in general.
Statistics from Altmetric.com
In a recent article in this journal, Felicitas Kraemer discusses a case in which deep-brain stimulation (DBS) treatment compromises a person's decision-making competence but reduces feelings of alienation.1 She proposes that, since the patient may find these two conditions mutually exclusive options, this generates a potential conflict between a patient's ‘autonomy’—understood as competence—and their ‘authenticity’. Against this I shall argue that ‘competence’ and ‘authenticity’ are conditions necessary to a satisfactory analysis of autonomy; Kraemer's observation that the two may conflict thus renders problematic interpretations of respect for autonomy that rely solely on the former. Integrating the importance of competence and authenticity conditions into respect requires that we move beyond the simple doctrine of non-interference with a patient's decisions to the more sensitive approaches advocated by some theorists of relational autonomy.2 ,3
Deep-brain stimulation, competence and authenticity
Kraemer's ‘case of the Dutch patient’ concerns a man undergoing DBS treatment—insertion of a ‘brain pacemaker’ device—for Parkinson's disease. With the pacemaker on, the patient experienced significant improvement in Parkinsonian and depressive symptoms. However, the switched-on state also induced a permanent manic state unresponsive to medication. With the device on, the patient was physically able and less dependent on others, but faced psychiatric hospitalisation due to manic behaviour. When off, the patient was bedridden and depressed but ‘possessed a rational and accountable state of mind’.
The healthcare professionals involved in the patient's treatment ultimately decided to ask the patient, with the device switched off, what he would prefer: to leave it off; or to have it switched on under the condition that he sign an advance directive agreeing to remain under psychiatric care while the device was activated. The patient opted for the latter.
Drawing from experiential narratives of other patients undergoing DBS or psychopharmacological treatment,4 Kraemer proposes an analysis of this decision in terms of a greater felt authenticity with the pacemaker switched on. (Kraemer notes that this is a speculative interpretation, since the authors of the case study from which her example is drawn do not report the patient's feelings.) She writes:
In one[…]case, a DBS patient says: ‘During all these years of illness, I was asleep. Now I am stimulated, stimulated to lead a different life’. Imagine that[…]the Dutch patient had said[…]: ‘The person that drives his car too fast, that leads a promiscuous life and that runs into debts is really me![…]before stimulation, I did not dare do such unreasonable things. I lived a well-adapted life[…]which I see now was never really mine. But now, I have the chance to be who I really am’. 1
On this interpretation, the patient's decision comes to a choice between alienated mental competence and authentic mania. In Kraemer's terminology, it presents a dilemma between autonomy and authenticity. On her account, to be ‘autonomous’ is to be deemed mentally competent, an interpretation consistent with its common employment in medicine (the demands of respect for autonomy are widely interpreted as the requirement to secure informed consent,5 or the obligation to follow patients’ care-related decisions unless there is clear evidence of either external coercion or an inability to understand, employ or evaluate relevant information6). Authenticity, meanwhile, is a characteristic of an individual's interpretation of their emotions, desires, motivations etc.: they are authentic ‘if the individuals experiencing [them] recognize their own feelings really as their own and identify with them’.7
Kraemer presents the case as a dilemma between conflicting obligations of respect for a patient's authenticity and their autonomy. However, the very idea of a conflict between authenticity and autonomy presents a prima facie puzzle, since prominent philosophical accounts of autonomy and its moral significance take authenticity as in part constitutive of autonomy. These accounts attempt to make precise an intuitive idea of autonomy as the capacity for self-governance, living one's life according to one's own reasons and not manipulative external influences. John Christman distinguishes two kinds of condition generally involved in such analyses: competence and authenticity.8 Competence is similar to Kraemer's ‘autonomy’, demanding that the agent exhibit capacities to comprehend relevant information and employ it to reach a ‘rational’ decision. Authenticity conditions concern an agent's attitude toward their own values and motivations, intending to specify what makes them the agent's own. These have been variously described as hierarchical identification with one's desires (not just desiring something, but wanting to desire, or otherwise accepting, it),9 reflective endorsement of one's desires,10 or non-alienation from the process of their formation.11
The concept of autonomy discussed by Christman and others is a more expansive notion than that adverted to by Kraemer, encompassing her ‘autonomy’ and ‘authenticity’. (Note that the latter is not simply the result of an unfortunate homonymy of technical terms; Kraemer's definition of authenticity quoted above is a form of hierarchical identification.) Before discussing the implications of this for the principle of respect for autonomy, however, it is important to examine which conception that principle takes as its subject. I shall argue that, at least according to two important and frequently-cited approaches, respect for autonomy concerns not just Kraemer's minimal version, but the more expansive conception.
The principle of respect for autonomy is frequently taken as self-evident or a truism,12 but where justification is considered important, ‘Kantian’ or ‘Millian’ arguments dominate (the Principles of Biomedical Ethics, for example, employs both13). Kant and Mill alike, however, demand significantly more from autonomy than the minimal conception allows. We can reconstruct their arguments by noting that Mill's ‘character’ and Kant's ‘humanity’ map closely to an intuitive understanding of autonomy. (In both cases, these are terms that: (a) apply to individual persons, (b) make explicit reference to free choosing, (c) concern themselves fundamentally with the ‘self-rule’ central to contemporary discussions of autonomy and (d) are furnished with arguments in favour of their being respected.)
The Millian justification stems from two claims: that exercising free choice helps persons to develop their capacities for reason and become ‘more well-developed human beings’14; and that the choices of persons of well-developed character tend to promote the utility of all, because they are capable of greater acts of individual good and of engaging in ‘experiments of living’ that may highlight the shortcomings of prevailing norms, or at least encourage reflective endorsement of those norms rather than dogmatic acceptance.14 However, this only applies when the choices are made as a result of the person's ‘character’, where:
A person whose desires and impulses are his own—are the expression of his own nature, as[…]developed and modified by his own culture—is said to have a character. One whose desires and impulses are not his own, has no character, no more than a steam-engine has a character.14
Mill suggests that respecting individuals’ self-governance invites a range of social benefits; in order for these benefits to be realised, however, this must be true self-governance, proceeding from a stable, independent nature. The Millian argument thus invokes competence and authenticity conditions in its picture of autonomy.
Kantian ‘humanity’ meanwhile (which the categorical imperative in its Formula of Humanity requires agents to treat as an end in itself) is not just the ability to make unconstrained choice, but the capacity to decide, through practical reasoning, what is valuable and can be set as an end.15 ,16 Christine Korsgaard expands on this interpretation, arguing that all normativity flows from our ‘practical conceptions of our own identity’: as reflective beings, humans need reasons to act, and these must come from ‘the conceptions of ourselves that are most important to us’. The move from here to respect for others’ autonomy proceeds by claiming that practical reasons are intrinsically shared between agents; they exist in the shared ‘space of linguistic consciousness’ and thus reasons arising from an individual's self-conception are reasons for all.10
With Mill's respect for character and Kant's for humanity we find that the object of respect is more than the minimal conception of autonomy. They invoke the importance of choices that are not just free from coercion or incapacity, but that are true to the chooser and expressive of their identity—in short, authentic decisions. This has important ramifications for clinical applications of respect for autonomy. The Dutch patient case shows that competence does not entail authenticity (on Kraemer's analysis, in one state the patient is competent, in the other their values authentic) thus autonomy cannot be established on the basis of competence alone. Since a decision made from a competent but inauthentic standpoint is as dubiously autonomous as the authentic but irrational one, respecting patient autonomy must therefore require evaluating the authenticity of a patient's decisions, as well as their competence. Thus, interpretations of respect for autonomy based only in competence cannot do justice to the principle.
A caveat to the above should be noted; this condition is less problematic in the case of the Dutch patient than might initially appear. Though Kraemer supposes the patient to have inauthentic desires and emotions in the switched-off (competent) state, he is nonetheless able to identify that those he has in the switched-on (manic) state feel more authentically his; the decision to have the pacemaker switched on, then, arises from values with which the patient feels some reflective identification, even if they are not experienced at the time of decision-making. However, one can easily imagine the patient in the competent state rejecting the manic-state values, while still not identifying with those experienced in the competent state, such that he views neither state as ‘authentic’. In such cases, respect for autonomy requires at least an attempt to engage with both perspectives, exploring the reasons underlying differing assessments of authenticity and the potential for common ground, as well as promoting the agent's autonomous capacities from both perspectives.
If we accept a Millian or Kantian justification of respect for autonomy, then, we must not merely attend to patients’ competence in evaluating their autonomy, but to their authenticity as well. The Dutch patient case suggests that such authenticity may come apart from the competence conditions more extensively acknowledged in a clinical setting. The independence of competence and authenticity bears on respect for autonomy in a far more general fashion, however, as the next section will attempt to show.
The conclusion that the minimal conception of autonomy-as-competence provides only a very limited principle of respect is not a new one, the imperfect alignment between informed consent and respect for autonomy in particular having received notable attention.5 ,17 ,18 Furthermore, many argue that challenges to authenticity extend beyond the rare cases discussed by Kraemer; they occur wherever an individual's social, political and cultural context can undermine development of the capacities necessary for its exercise. This may occur via material processes (as when individuals or groups are not afforded the educational resources to develop reasoning skills or the economic resources to assert material independence from the wills of others2) or psychological ones (oppressive situations can lead to the internalisation of such oppression,17 ,19 or believing that socially-imposed limitations are right, natural or inevitable and adapting one's desires accordingly20). While it is beyond the scope of this essay to argue at length that such factors pose threats to an agent's autonomy, note that they all involve having ends set by an external influence, rather than the agent's setting such ends for themselves through practical reason or ‘expressing their own nature’; they thus appear to compromise Kantian humanity and Millian character.
If these constitute threats to autonomy then, since the grounds on which an agent may assert hierarchical identification with or reflective endorsement of certain values are subject to the same external influences as those values, they pose a similar threat to authenticity stated in such terms. Several authors have therefore suggested that the authenticity condition might be strengthened by incorporating a further self-respect criterion,3 ,21 demanding that the agent view themselves as a legitimate source of reasons for acting, considering their own evaluations to be worthy grounds for endorsement or identification with desires etc., rather than feeling obliged to turn to some external authority for such grounds. While I do not wish to argue here for a particular substantive analysis of autonomy, I think such a criterion must be at least part of (or entailed by) any satisfactory analysis. An agent lacking self-respect does not even believe that values arising from their interpretation of their own identity carry any normative weight for them; since they do not let such values govern their actions, they cannot exhibit self-governance.
These relational insights provide grounds for critique and refinement of the principle of respect for autonomy. The principle that arises from the minimal conception of autonomy assumes that (a) if a person has capacity, then respect demands complete non-interference with their decisions and (b) if they lack capacity, then paternalistic intervention ‘in the patient's best interests’ is warranted. This treatment implausibly models autonomy as an all-or-nothing affair,3 and the paternalistic response is especially problematic given the relationship between autonomy and oppression: making decisions on behalf of those whose autonomy is compromised through oppressive circumstances only enhances their powerlessness and may compound a self-conception in which they are unfit to make such decisions for themselves.17 Furthermore, the relationship between oppression and autonomy is not straightforward, and the distinction between inauthenticity and autonomous deviance not always obvious. Those who are socially marginalised may have less invested in dominant norms and thus be in a better position to question them and engage in Millian ‘experiments of living’.17 They may also be better situated to identify gaps in collective epistemic resources (insofar as those resources are geared towards making sense of the experiences of those with less-marginal situations).22
Catriona Mackenzie attempts to develop a more nuanced account of respect for autonomy in the clinical setting that takes these concerns seriously. She builds her account around the narrative of ‘Mrs H’:
Mrs H.[…]has just had a leg amputated below the knee as a last resort treatment for aggressive bone cancer. [She] has lost her hair from chemotherapy and is having to come to terms with the prospect of permanent disability[…]although her doctors are fairly confident that her short to medium term prospects for survival are quite good. Her husband has recently left her because her disability would be burdensome and he finds her and her condition an embarrassment. Mrs H.'s practical identity involves a conception of herself that is governed by the norms of traditional femininity that are taken as authoritative within her cultural community, and her husband's abandonment has left her feeling worthless as a person and without a reason to live. Mrs H. informs her treatment team that she wants to die and that she wants no further treatment if the cancer spreads.3
While there is no question raised over Mrs H's competence, the authenticity of her decision is in doubt. The values driving her choice seem to be the product of oppressive social circumstances that have led her to perceive a life in subordination to her husband as the only viable kind to lead. Yet her decision still exhibits a degree of independent evaluation, deliberation and reflection. Neither accepting Mrs H's decision without question nor overruling it in her ‘best interests’ can do justice to this mixed analysis of the extent of her autonomous capacity.
Mackenzie's interpretation of respect for autonomy involves three interconnected obligations: to recognise the patient's humanity; to try to understand her subjective perspective; and to promote her capacities for fuller autonomy. The first of these is a minimal condition for engaging in any kind of relationship with another—the recognition of that other as a distinct individual, with a distinct conception of identity and value—that is shared with respect for autonomy on the minimal conception. The latter two, however, are more complex.
Understanding the patient's subjective perspective is important for two reasons. Firstly, it requires an active engagement with the patient's motivations and identity that permits exploration of how they see themselves in relation to their reasons for action and whether they display the self-respect suggested above to be necessary for authenticity. Mrs H's decision would exhibit a greater degree of autonomy if, on questioning of her decision, she were to hold herself worthy to act as apologist for the values she accepts and explain why apparent considerations that might militate against her position held no normative weight for her, than if she merely repeated a doctrinal response, deferred again to some other authority or failed even to understand the possibility of such critique. Andrea Westlund argues that there is an important general difference between the autonomy of individuals who defer to other authorities in determining what they consider valuable but are willing to hold themselves answerable to critical challenges to those authorities, and that of ‘deeply deferential’ individuals who, in responding to such challenges, can only invoke those authorities to whom they defer.23 The attempt to understand the patient's perspective would involve dialogical engagement that permitted assessment of these differences, aiding in the distinction between significantly autonomous deviance and deep deference to oppression.
Secondly, the attempt to understand the patient perspective is a precondition for a non-coercive approach to Mackenzie's third obligation. She argues that, where a patient's self-referring attitudes may undermine their own autonomy and flourishing, then there is an obligation to aid the patient in changing their attitudes in ways that would better support these capacities; promoting their self-respect, helping them to find and assert reasons for action independently of others and to develop meaningful patterns of existence. Such attempts to promote patient autonomy would, however, be coercive if attempted through unilateral imposition of a ‘better’ way of life rather than by a shared dialogue around conceptions of identity and value. The latter treats the patient as they are hoped to be—a substantially autonomous agent—while the former would only replace deference to one authority with deference to another.
Mackenzie's interpretation of respect highlights how paying proper attention to what matters about autonomy differs from mere non-interference. Implementation of its recommendations presents its difficulties: the material, time and emotional demands on healthcare workers made by such a scheme, particularly in busy departments with staff shortages, must be taken into account; health workers may be ill equipped to and cannot be expected to undo a lifetime's oppression; the obligation to promote autonomy would have to be engaged with sensitively to protect against the risk of coercion; and the legal implications of a more nuanced account of respect for autonomy such as Mackenzie's would be harder to codify. Nonetheless, it demonstrates some important considerations pertaining to a more complete respect for autonomy.
The above holds notable implications for Dutch patient-type cases. I have already argued that acknowledging the role of authenticity in autonomy requires engagement with the patient's perspective (perspectives in the Dutch patient case) similar to that Mackenzie endorses; given the arguments of this section, it is additionally important that such engagement incorporates an exploration of the degree to which these perspectives involve the self-respect necessary for authenticity.
Perhaps a more important lesson, however, is that attending solely to the differences in perspective held by the treated and untreated patient in examining respect for autonomy may disregard the more salient threats to autonomy posed by the patient's social context. This is particularly problematic given the stigma attached to the kinds of disability-causing or mental health conditions that are liable to induce people to seek DBS or psychopharmacological treatment. This argument, developed forcefully by Françoise Baylis,24 may be summarised as follows: if the epistemic resources available for interpreting a given illness within a community construct it as entailing a certain identity—a physically disabled person as passive and invalid, for example, or a depressed person as unstable and dangerous—and a person internalises such a narrative, but does not identify with it, then the very experience of symptoms of that illness may lead to their adopting aspects of that identity, though they feel alienated from it. The alleviation of symptoms of that condition may thus help the individual to feel less compelled to apply to themselves the identity constructed for those with that condition and so result in a greater felt authenticity. The threat to authenticity, however, lies less in the illness than in the discriminatory narrative surrounding it.
Kraemer suggests that cases like that of the Dutch patient, where a patient's mental competence and their felt authenticity apparently decohere, demonstrate two distinct dimensions of a person's humanity that each warrant respect; however, if autonomy demanding of moral respect includes authenticity, what Kraemer's case instead highlights is the insufficiency of competence-based respect to acknowledge all that is important in autonomy. A more complete respect for autonomy, incorporating relational insights, demands a greater sensitivity to patients’ subjective perspectives and social contexts, and a willingness to engage them in the promotion, development and maintenance of capacities for autonomous decision-making—where such capacities involve at least the combination of competence to make decisions on the basis of practical reasons and the authenticity of those reasons to the agent's character. Some particular implications of this view in Dutch-patient type cases are explored above but, as Mrs H's example demonstrates, its lessons are more widely important.
I am very grateful to the two anonymous reviewers for their comments on an earlier version of this article.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.