Woodchopper Revisited 25

guidance concerning contemporary medical dilemmas. By making his assumption clear, Jakobovits responds to one of the two principal criticisms that Newman raises against some Jewish bioethicists who have addressed our topic.

With respect to Newman's second critique, Jakobovits is also clear. He acknowledges the limits of the analogy, that the ruling of Isserles applies only to a patient who is already a goses. A different theoretical approach is required to justify the discontinuation of medical treatment for a choleh noash, a patient who, though not yet in the very last stages of life, has been diagnosed as terminal with no hope of recovery. Jakobovits offers that rationale in the form of a chidush, an idea of his own derivation that he uses to resolve a conflict between Nachmanides and Maimonides concerning the toraitic source of the mitzvah to practice medicine. Whereas the intricacies of that chidush do not concern us here,* Jakobovits uses it to distinguish between permitted and obligatory medical treatment. While a terminal patient is permitted to undertake medical measures“that keep him alive in a state of suffering,” he is not obligated to do so. One practical conclusion of this distinction, writes Jakobovits, applies to the instance of a diabetic who develops terminal cancer. The insulin injections that she takes for her diabetes and that heretofore have been regarded as part of a successful regimen of treatment can be said now to function so as to extend her agony. Though she is permitted to continue with those injections,“one should not object” should she decide to cease them. In his essay, then, Jakobovits makes extensive though not unlimited use of the analogies that figure prominently in Jewish bioethical discussions of the treatment of the terminally ill. He addresses“the problem of importance” by arguing forcefully that the similarities between the baseline cases and the target case outweigh the differences between them. He also recognizes that the analogies have their limits and should not be extended farther than they can plausibly take us.