Abstract

Medical decisions concerning noncompetent patients that are most morally problematical are those that involve life and death choices. In making these choices for others, I urge that decision-makers carefully attend to the degree and history of a person's noncompetence, and distinguish four relevant categories of competence: partial, potential, lost and never possessed. Attending to these will help enable us to sort out when and how autonomous choice is possible and desirable and when and how to rely upon a judgment of a patient's best interests instead. This conflict is explored with particular reference to the Quinlan, Fox and Saikewicz cases. I argue that respect for autonomy, wherever possible, has presumptive priority, and that it is more often possible than frequently supposed. On the other hand, the notion of autonomous choice itself is perverted when forced upon inappropriate cases, such as those presented by the never competent. Here it is not merely legitimate, but inescapable to use a best interest criterion that ignores autonomy in making treatment decisions. Specific cases in which these normally distinct guidelines for decision-making concur, conflict and need resolution are discussed. In the end it is often more important to determine who will choose than by what substantive criteria he or she will be guided. In considering this issue, I try to show that the substantive criteria have implications for who the proper decision-makers are in many problematical cases.