Autonomy and the Relational Self

Autonomy is a fundamental principle of bioethics. However, there has been a longstanding dissatisfaction with it. This largely has to do with a particular understanding of autonomy, as the right to make one’s own decisions, protected from outside interference. Accompanying this are related ideals such as individual self-determination, individual independence, and self-sufficiency. Because the right to make one’s own choices has played an indispensable role in the rise of bioethics, it is not surprising that it is treated as the primary expression of autonomy. But autonomy understood only in this way has problems.

First, even those who see a central role for autonomy in ethics have had difficulty with the way it is used in bioethics (O’Neill 2002). When autonomy is regarded only as the right to make one’s own choices, other—more fundamental—meanings of the concept are obscured. Although Kant is often blamed for the excessive emphasis on autonomy, his was a different, deeper conception. This is not the place to mount a defense or even an exposition of Kant, but it is important to recall that Kant saw autonomy as the source of all morality, not simply as a specific principle of an individual’s right to make his or her own decisions. Further, Kant’s theory of morality, I would argue, is a theory of human relationships: treating humanity as an end in itself is not merely about setting limits on action but rather about a framework for seeing ‘the other’ as a creative being with whom one shares ends and enters into a relationship.

Second, when the right of self-determination is unmoored from a deeper basis, it becomes difficult to draw principled boundaries around it. Thus, it is not difficult to find autonomy as individual self-determination taken to an excessive level—what Carl Schneider has called the doctrine of “mandatory autonomy,” which unrealistically idealizes self-sufficiency and sense of control (Schneider 1998). This flies in the face of what, in the medical context, patients (very reasonably) want: They want to be informed, but generally want to defer to those who are in a better position to make decisions (Schneider 1998).

Unfortunately, empirical research in bioethics follows the lead of theoretical ethics by privileging the value of being the master of one’s own decisions. Surveys ask patients what their treatment preferences might be in case of incapacity, as though their current preference is all that matters. Such surveys seldom ask how much people value such a preference (which they express to the researcher only because they are being asked). However, most people say that they trust their loved ones to make decisions for them, even to the point of going against their currently expressed preferences (Kim 2011; Sehgal et al. 1992).

Third, and more generally, autonomy unmoored from a deeper understanding of morality remains an abstract principle. The application of such an abstract notion tends to neglect the context in which it is applied. Perhaps the classic instance of this is in informed consent. The patient or the research subject is supposed to ‘make her own fully informed decision’ based on all the relevant facts. However, the practice of informed consent is not typically designed to do this. If it were, it would incorporate the best evidence about how people do in fact use information to make decisions. Far from incorporating what is known about how people make decisions, informed consent processes are notorious for perversely doing the opposite, for example, in using forms that are too long, unreadable, and difficult to use. Institutional Review Boards—those who are charged with ensuring informed consent—themselves commonly require researchers to use texts for informed consent forms that fall short of their own readability standards (Paasche-Orlow, Taylor, and Brancati 2003).

The problem is that it is too easy to use ‘autonomy’ shallowly and without sufficient sensitivity to the context in which it is applied. The problem is not autonomy, but the abuses of autonomy.

In this issue of PPP, Walter and Ross (2013) advance the work of rehabilitating the bioethics...