The modern debate on whether—and why—physicians and hospitals can refuse patient or family demands for treatment on grounds of “futility” will be reaching its 20th anniversary this year (Blackhall, 1987). The early debate focused on the use of CPR, for good historical and clinical reasons, and CPR probably remains the primary target of hospital policy. But the reach of the arguments over futility extends well beyond this context, most vividly illustrated by the case of Helga Wanglie and the many commentaries it spawned (Brody, 1998; Schneiderman, 1998; Trotter, 1999). There seems no obvious reason in principle that questions about futility cannot be raised about virtually any form of treatment that patients or families might demand, including such things as chemotherapy for cancer (Tomlinson, 2001). Still, we might fairly ask what new dimensions and complexities the futility question acquires when it moves onto decisions about treatments other than CPR. This is what I would like to do in this essay. I will use dialysis as the case in point, although I hope to extract lessons from the dialysis example that will be pertinent to other forms of life-sustaining treatment, like ventilators.