Indexing Burdens and Benefits of Treatment to Age: Revisiting Paul Ramsey’s “Medical Indications” Policy

This essay reconsiders Paul Ramsey’s “medical indications” policy and argues that his reconstruction of the case of Joseph Saikewicz demonstrates that there is more room for caretakers to decline treatments for “voiceless dependents” than his interlocutors have sometimes thought. It further-more draws on Ramsey’s earlier work to propose ways that Ramsey might have improved his policy, and argues that the shortcomings of Ramsey’s view arise from his bracketing of age in making determinations about what form of medical care is owed. The reading of Ramsey set forth here suggests Cathleen Kaveny’s depiction of the ‘medical indications’ policy in Ethics at the Edges of Law is too rigid and inflexible, even while it affirms other aspects of her critique.


I. INTRODUCTION
Paul Ramsey's reflections on bioethics underwent a significant shift in emphasis between the publication of The Patient as Person (PP) (2002b(PP) ( ) in 1970(PP) ( and 1978s Ethics at the Edges of Life (EE) (1978a). 1 Ramsey's earlier work attempts to delineate the shape of our responsibility to only care for the dying (though by no means merely care for them). His concern is to prevent the responsibility to save life from ossifying into "an absolute and inflexible norm," which would mean certain life-perpetuating treatments might become "a hardship inhumanly applied" (Ramsey, 2002b, 141). Yet, Ramsey concluded that this earlier work provided insufficient safeguards to prevent it from being conscripted for purposes that he thought morally noxious. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
In the intervening essay "The Indignity of 'Death with Dignity,'" Ramsey writes that he found "altogether too many people were agreeing with me," which "caused qualms" (1974,47). 2 His emphasis on only caring for the dying was being deployed in ways that would sanction the under-treatment of nondying patients, thereby permitting what would effectively be "manslaughter in some degree" (Ramsey, 1978a, 188). To foreclose this possibility, Ramsey develops a "medical indications" policy in EE that emphasizes the benefits of treatments (rather than lives), and permits only mature, competent adults to prescind from treatments that are "medically indicated. " Ramsey links his new emphasis to what he saw as shifting cultural tides regarding the treatment of those who are terminally ill. In a article delivered in 1986 reflecting on the debate, Ramsey acknowledges that he may have developed a different policy in "another time and place," but that perhaps "in this place and time, voiceless patients require the special protection that a medical indications policy provides" (2002a, 152). In her recent book Ethics at the Edges of Law, Cathleen Kaveny suggests his transformation reveals the contours of the "cultural conservative playbook over the next several decades" (2018,140). While Kaveny's argument is primarily focused on Ramsey's understanding of the law, she also contends that Ramsey's "medical indications" policy is too narrow, in that it fails to account for the full range of values required to properly navigate end-of-life contexts (2018,.
While I question aspects of Kaveny's reading of Ramsey, her contention that his shift has much in common with contemporary "culture war" conflicts is indisputable. Ramsey wrote in the midst of a ferment of ideas and questions that continues to cast a long shadow over our own time. Wrestling with these early debates helps contemporary bioethicists understand the contours of the discipline as it has developed, making us alive to how our assumptions were once contested. For those discontent with the shape of today's debates, such historical work helps remind us of paths of reasoning and action that were once open to us-and which may have been more beneficial than the one on which we find ourselves.
In this essay, I want to revisit Ramsey's effort to navigate the Scylla and Charybdis of over-and under-treatment for voiceless, dependent individuals. On Kaveny's reading, Ramsey fails to satisfactorily address the reasons for which treatment might be declined for voiceless individuals because he is inattentive to how a person's history and social location affects our understanding of the benefits and burdens of treatment. I argue that Ramsey's policy offers more flexibility to decline treatment for voiceless dependents than Kaveny allows-as is evident from his response to the case of Joseph Saikewicz. Ramsey develops a policy of treating voiceless individuals that admits of exceptions, provided that those exceptions are advanced for "medical" reasons (as Ramsey understands them). While aspects of the policy are doubtlessly animated by Ramsey's worries about the slippery slope he saw medical ethics hurtling down, his deflationary account of what medicine should do for patients-namely, heal bodies without attending to their quality of expected life-is consistent across his work. Even so, Ramsey's medical indications policy is still underspecified. As such, I attempt to sharpen its contours, and propose that it would have benefited from attending to age as a default or prior against which the benefits and burdens of treatment are weighed.
First, then, I recapitulate Ramsey's medical indications policy as Ramsey describes it in EE. Ramsey's policy turns on his distinct understanding of both the aims of medicine and the peculiar importance of the value of "life." As Gilbert Meilaender (1991) has observed, such a policy seems to leave ethicists in a quandary either about furthering euthanasia or of making burdensome treatments obligatory-hardships "inhumanely applied." I subsequently explore this putative dilemma through the lens of Ramsey's discussion of Joseph Saikewicz and argue that in practice Ramsey's policy was less stringent than it might have seemed. At the same time, I argue that Ramsey ought to have incorporated age into his policy, because doing so would have enabled it to secure maximal protections against under-treatment for voiceless dependents in the earliest stages of their lives while lowering the threshold to justify declining life-perpetuating treatments if a patient is older. I then consider Kaveny's critique and argue that Ramsey's view is both more flexible than her account suggests and supplies strong reason to reject her proposal to make social ethics relevant for medical decisions for voiceless individuals.

II. A MEDICAL INDICATIONS POLICY
Ramsey's "medical indications" policy is motivated by his interest in safeguarding voiceless, incompetent individuals from being denied treatments that they are owed. 3 In PP, Ramsey employs the distinction between ordinary and extraordinary treatments in the context of end-of-life decisions, defending the freedom of individuals to decline treatments that were "useless" (2002b, 127). In EE, Ramsey acknowledges that locating his discussion in that context misconstrued the force of the distinction, which bears as much on nondying individuals as it does on those whose "dying process" has begun. Yet, rather than rehabilitate the distinction in EE, Ramsey argues it is inadequate for making treatment decisions for dependent individuals. Ramsey worries that the distinction between ordinary and extraordinary treatments was being filtered through "quality of life" considerations, and so was being employed to effectively treat nondying voiceless individuals as though they were dying. We have begun to "do unto others," Ramsey writes, "as conscious competent patients in former times could do for themselves . . ." The possible refusal of extraordinary or heroic measures of care by competent adults was being deployed as quality of expected life judgments in cases of "the desperately ill or 'incurables' who are voiceless," effectively transforming those who are not dying into those who are (Ramsey, 1978a, 155).
The "medical indications" policy Ramsey develops in EE as an alternative to the ordinary/extraordinary distinction has a number of contours. First, Ramsey's policy presupposes a sharp distinction between medical and what I call prudential reasons. 4 The medical indications policy directs our attention, Ramsey suggests, toward "the objective condition of the patient, not to abstract classification of treatments or to the wishes of any of the parties concerned . . ." (1978a, 159). Ramsey's concern to limit medicine's aims is continuous with the approach he develops in PP, where he objects to expanding physicians' responsibility from the "duty to heal and to save life" to include the "full range of human moral considerations." There, Ramsey distinguishes between the medical and the prudential imperatives in dealing with the dying: medicine should be ordered toward a narrow construal of the body's health, rather than toward a person's well-being-a more expansive category that would seem to give doctors more power than they ought to have (2002b, 118, 123).
While Ramsey might have a narrow understanding of medicine's aims, it is augmented by an emphatic prioritization of the value of bodily life. Ramsey is acutely concerned throughout his writings about bioethics about the instrumentalization of the body for the sake of social or other ends. He vociferously denounced the Cartesian dualism that he thought enabled such instrumentalization and thought it uniquely the task of Protestant ethics to reinvigorate the significance of the natural bonds between sex and procreation. 5 In addressing the question of organ donation in PP, Ramsey worried that the "war against the alleged 'physicalism' of past moral theology" would lead to a notion of the human person that "breaks the bonds of physical existence in quelling the emptiness that is the hallmark of his finitude by faith and love toward God" (2002b, 181). The "only human life we know to respect, protect, and serve in medical care," Ramsey wrote, is "irremediably physical, and presented to us with its moral claims solely within the ambience of a bodily existence" (2002b, 191). Even then, though, Ramsey saw such a stance as a "losing battle of the barricades". Those moralists who "approve of acts transcending physical life itself" and free humanity from the "brutal materiality of his island existence in the limited physical space of the body" will be "crowned victors" (Ramsey, 2002b, 192). 6 Still, Ramsey's concern for the life of the body pervades his medical indications policy in EE. "Life" is something like an architectonic value for Ramsey, which is incommensurate with other goods. "When we speak of good," he writes in EE, we "speak of life's well-being" (Ramsey, 1978a, 207). To choose death would be wrong because of "our religious faith that life is a gift," which is also a trust (Ramsey, 1978a, 146-7). 7 Such a stance imposes stark limits on how we treat one another: we "are not given dominion or co-regency over humankind" alongside God (Ramsey, 1978a, 205). In his discussion of when it is licit to perform surgery on infants with spina bifida, Ramsey adopts the dictum that medical care does not "add years to their life but life to their years." The added time an intervention might supply is incidental: Ramsey focuses instead on surgeries' effort to "reduce disability and to improve a person's capabilities whatever years his life has to run." 8 Medicine is ordered toward the intrinsically valuable end of preserving and restoring an individual's bodily capabilities, without paying any reference or attention to what an individual might do with them. "Whether life will be beneficial to its possessors is not a question" the doctor should ask, Ramsey contends. Rather, medical treatment is ordered toward the "enhancement of human bodily life, or prevention of its deterioration," which is not "merely an instrumental value." 9 Ramsey's understanding of life seems to be especially detached from time. If a patient has a terminal condition but is not yet dying, then a treatment is medically indicated if it "improves their condition or adds meager life to their days" (Ramsey, 1978a, 187)-rather than more time. While his emphasis on preserving bodily life might seem like it commits Ramsey to indefinite life-extension, Ramsey rejects the necessity of putting a patient on a ventilator as a life-prolonging measure in an essay evaluating the Saikewicz case. Cardiac surgery for a nondying patient, though, is a "lifesaving procedure" (Ramsey, 1978b, 41). Medically indicated treatments, then, improve a person's capacity for life, rather than (merely) extend it.
Third, Ramsey's "medical indications" policy does allow for competent adults to decline treatments, but only in a narrow range of situations or cases. In EE, Ramsey challenges a broad basis for declining treatment: there is no "moral right to refuse" treatment, he writes, but only a "relative right" (1978a, 156). Instead of emphasizing the freedom of adults to decline treatment, Ramsey stresses that they should be afforded "free and informed participation in medical decisions affecting him when there are alternative treatments" (1978a, 157). Empirically, Ramsey seems to suggest that new technologies and shifting social conditions have narrowed the range within which competent adults might licitly forgo medically indicated treatments. Progress in medicine has rendered a wider range of treatments 'ordinary,' and subsequently obligatory. Moreover, Ramsey argues the "psychological burden" of death on neighbors, family, and friends can be great, which raises the threshold for reasonably declining treatments (1978a, 156). 10 Fourth, Ramsey's "medical indications" policy precludes appealing to prudential reasons to decline treatment for voiceless individuals. Ramsey raises two objections against making 'substituted judgments' to decline treatments: the norms we develop through that process will be too subjective to adopt as a policy, or they will permit us to devalue the patient's life. On the first, Ramsey rejects Robert Veatch's appeal to "reasonableness" as a standard for weighing up treatment from the patient's standpoint. On Veatch's approach, we assess the burdens and benefits of treatment from the patient's point of view, weighing them against "familial, social, or economic concern, or religious belief." Ramsey worries that the "subjectivism of the first move controls the meaning of the second." If declining treatment on the basis of such nonmedical reasons is licit for competent adults, the approach is "without chart or compass" as a "medical ethical policy" (Ramsey, 1978a, 165). Ramsey argues that as a "rule of practice" there is an obligation to supply medically indicated treatments: for "voiceless nondying patients . . . there remains an undiminished obligation first of all to sustain life" or to use palliative treatments if there are no medically indicated treatments available (1978a, 165). Such a rule establishes a default course of action for doctors, who must favor "life when the patient has no capacity to refuse treatment or to share in physicians' or family decisions to omit or neglect treatment or otherwise initiate the incompetent, incurable patient's dying." That rule of practice "may be rebuttable in rare exceptions." But those exceptions cannot weaken either the moral rule nor, when determined judicially, "the law's protection of life" (Ramsey, 1978a, 165). 11 Ramsey seems to adopt such sharp constraints around declining care for voiceless individuals for normative reasons, and not only because he is worried about abuses happening. We lack a standard by which we might decide whether to decline treatment for another. That competent adults decline treatment for themselves is "no standard at all for assessing such an adult's moral or legal right to refuse treatment for another" (Ramsey 1978a, 166). Because children have not developed their objectives or projects against which they might measure the "usefulness" of a treatment, only the "parent's sense of these things would count-and count," Ramsey adds, "as cryptoprotection" (1978a, 166-7). 12 As Gilbert Meilaender glosses Ramsey's view, decisions to decline life-sustaining treatment are "more like vocational choices, which no one imagines ought to be generalized"-much less made on behalf of someone else (Meilaender, 1991, 150). Any decision to decline treatment must exclusively be that individual's, and as such-as Ramsey would later clarify-a "proper expression of the limited but genuine freedom which the Creator has bestowed upon the human creature" (2002a, 151). 13 What, though, of Ramsey's worry that "substituted judgments" might be used to devalue the lives of voiceless individuals? Ramsey vociferously critiques Richard McCormick's view that we may decline treatment for those who have lost the capacity "for attaining a share, even if reduced, in the 'higher, more important good'" of relationships with others and with God (McCormick, 1974, 10). "Quality of life" considerations, Ramsey contends, are too expansive for use in medical decisions: they fail to account for whether such considerations are medically or socially determined. As Ramsey argues, if it is licit to deny a defective newborn treatment because the expected quality-of-life will be low, it is similarly licit to do so for the "offspring of the 13-year-old unmarried girl from the ghetto." In this way, defenses of the "benign neglect of defective newborns are also arguments for the benign neglect of at least some of the environmentally deprived." In end-of-life contexts, Ramsey argues that quality-of-life judgments are either irrelevant, or they erode the distinction between "voluntary and involuntary euthanasia, and between letting die and actively accelerating death" (1978a, 176-7). 14 McCormick fails to see that the "next logical step in moral reasoning" of his subordination of physical health or life to the possibility of active participation in relationships with others or God "must be to hasten their deaths" (Ramsey, 1978a, 172). Ramsey, then, wants to quarantine decisions to decline treatment from being made on the basis of broader evaluative judgments about the worth of a patient's life or other socially determined reasons.
As with Ramsey's first worry about substituted judgments, his objection to McCormick is also based on his normative understanding of why we owe others treatment. McCormick goes awry, Ramsey argues, by attempting to distinguish between the types of valuable life. By contrast, a medical indications policy requires "no comparison of patient-persons or of different stages or conditions of the same patient-person in order to determine his quality-oflife struggles or prospects" (Ramsey, 1978a, 178). In PP, Ramsey had stressed that the end of treatment is the person, rather than the disease. While he does not reject this in EE, he emphasizes instead that each person is equally valuable, regardless of age and competence, and that the only comparative judgment permitted is whether treatments are beneficial "in any other way other than prolonging dying" (Ramsey, 1978a, 178). The nondying receive treatments that those who are terminally ill do not only on the basis of the distinction "between beneficial and non beneficial treatments, treatment indicated and no further curative or salvic treatment indicated" (Ramsey, 1978a, 180). Treatment decisions should make no reference to the value of the lives being treated.
Ramsey worries, though, that for voiceless patients there is a tendency to collapse burdensome treatments into a burdensome life. Ramsey is especially concerned about the plight of vulnerable dependents: God "has special care for the weak and the vulnerable among us earth people," he writes (1978a, 205). As such, he raises the threshold for protection for those who are in our charge. In principle, a medical indications policy counts the burdens of treatment. Treatment, he says, may be stopped when "excessively burdensome qualities are (1) directly associated with the means used (e.g., pain), or (2) caused by the life-sustaining maiming" (Ramsey, 1978a, 181). 15 Yet, the secondperson nature of decisions for voiceless incompetent dependents raises the risks of abuse to such a point that Ramsey seems to seriously discount their significance. When we "slide from burden of treatment to burden of life," Ramsey would later write, "we conceal from ourselves the real nature of the choice we are making: a choice not of life, but a choice of death" (2002a, 152). Unlike the other dimensions of his thought, Ramsey's approach on this score is entirely contextual: "If in the present moral climate it were possible to consider burdens of treatment alone for such voiceless patients," he writes in this later essay, "a different rule of practice might be compatible with the steadfast care we owe such patients. But we have ample evidence that burden of treatment is regularly elided into burden of life" (Ramsey, 2002a, 152).
Ramsey's medical indications policy, then, aims to secure maximal protections for voiceless individuals at every stage of life. It does this, first, through a narrow construal of medicine's aims, and, second, through a corresponding emphasis on the intrinsic value of bodily life. Third, it affirms the limited freedom of adult individuals to decline treatment, provided they choose life, but grounds this freedom in the person's freedom before God to choose the course of their life. Yet in both morality and policy, Ramsey denies any possibility of deriving decisions to decline treatment for voiceless dependents from our own determinations about what they might or might not do if they were competent or from assessments of their quality of expected life. Finally, it limits our focus to whether treatments are beneficial or nonbeneficial for voiceless individuals, and precludes any consideration of whether their lives are valuable or whether such treatments might be burdensome to them. While Ramsey's approach is normative, he allows that this final constraint is animated by specifically contextualized concerns about the abuses he saw arising around him. Whether this medical indications policy secures protections without engendering obligations to impose "hardships inhumanly applied" on voiceless dependents is the theme of the rest of this article.

III. THE DILEMMA, AMBIGUITIES, AND CONCERNS
Ramsey's two-tiered approach to declining medical treatment seems to put us in a dilemma. As Meilaender distills the problem, failing to adopt Ramsey's medical indications policy "may commit ourselves to a path that will, in fact, end in euthanasia," as quality-of-life considerations or other non-medical concerns allow us to treat the nondying as though they were dying and so deprive them of care. But adopting a medical indications policy "may seem to lose some of the humane wisdom of PP, may seem to make medical intervention 'a hardship inhumanly applied' to the voiceless when we would reject its application ourselves" (Meilaender, 1991, 151). As Meilaender argues, Ramsey's deliberate inattentiveness to the burdens of treatment for voiceless dependents because of the possibility of abuse risks losing the 'middle course' between over-and under-treatment that Ramsey had attempted to chart. As Meilaender reads him, Ramsey's interest in developing a policy leads him to accept the essentially tragic byproduct of requiring overtreatment in some (perhaps small) number of cases. Ramsey trades moral flexibility for maximal protections for voiceless individuals, in order to prevent his own view from being conscripted on behalf of euthanasia.
Yet as I argue in the remainder of this article, Ramsey's medical indications policy plausibly permits declining treatment for individuals within a broader range of cases than it might first appear to, as is evident from his analysis of the case of Joseph Saikewicz. 16 Saikewicz was a 66-year-old man who was described as having a mental capacity of a 3-year-old. While Saikewicz died before the Massachusetts Supreme Court issued its opinion, they eventually concluded that life-prolonging treatments for his leukemia could be reasonably withheld from him. In his analysis of the case, Ramsey opposes any judgment that would admit quality-of-life assessments as part of its reasoning. On Ramsey's critique, if the court appealed to the gap between Saikewicz's age and mental competence as the basis for not extending life, then it shifts the basis of declining treatment for Saikewicz from the fact that his death is impending to the "irrecoverability" of his cognitive life. Such an approach would allow a "quality-of-life criterion" to reach back into the prognosis of cure and recovery, effectively sanctioning involuntary euthanasia. Ramsey worries that the "patient's incompetence (his reason for needing protection) became the reason for not protecting his life" (1978a, 315). That is the "worst reading of the outcome" (Ramsey, 1978a, 302). Now Ramsey also suggests the Saikewicz ruling "could have been a balancing judgment falling within a medical indications policy" (1978a, 302). That is, Ramsey's main concern with the judgment is the court's reasoning-rather than their conclusion. 17 On Ramsey's reconstructed opinion, Saikewicz's "severe retardation" would not have been admitted as a consideration. Instead, Ramsey suggests that the judgment to not proffer medical care might have been made as a "quality-of-life-during-treatment" judgment, rather than a qualityof-life judgment per se. Despite narrowing the aim of medicine to health, rather than well-being, Ramsey adopts a broad account of what constitutes "treatment," which depends in part on Ramsey's understanding of medicine as a joint enterprise in which the patient must be a cooperating agent. "Lack of cooperation from anyone," Ramsey writes, "is, after all, part of medical treatment to be taken into account" (1978a, 316). Ramsey acknowledges the differences between Saikewicz and an ordinary 3-year-old, admitting that he was "probably not as docile" as a child, and muses on this basis that there are "probably some quite specific counterindications of treatment" (1978a, 316). Yet such "counterindications" seem to include for Ramsey Saikewicz's quality-of-life while receiving treatment: it "could have been this particular patient's condition of life during treatment . . . that was placed in the balance" of the court's opinion (1978a, 316). While Ramsey does not employ the language of "burdensome treatment," the concept seems to be at work: he worries about the "life to be taken from his months or a year undergoing treatment" (1978a, 316). In other words, Ramsey's concern seems to extend beyond the sheer effectiveness of treatment, to the manner in which that treatment is received. Behind this is Ramsey's attentiveness to Saikewicz's context: "Saikewicz had lived in state institutions since 1923," he notes. "His inability to cooperate might cause some problems." (Ramsey, 1978a, 302). In other words, Ramsey acknowledges the possibility that "treatment itself imposed unacceptable hardship," and thus thinks it reasonable on a "medical indications" policy to withhold it from him (1978a, 316). If the courts are incapable of deciding on Saikewicz's nontreatment for the reasons Ramsey specifies, then the "ethical rule of practice should be the treatment of incompetents always in the way normal patients would be treated" (1978a, 317).
In Ramsey's own hands, then, a medical indications policy offers more flexibility to decline treatment for voiceless individuals than appears at first glance. Ramsey's opposition to the state's judgment to withhold treatment has everything to do with the reasons they supplied, and the moral environment those reasons establish for future cases. His emphatic discounting of the "burdens of treatment" in his moral reasoning arises from his worry that such a category effortlessly slides into "burdensome lives." Even so, Ramsey leaves the reasons on which the court might licitly withhold treatment from Saikewicz underspecified. While the Saikewicz case is an exception to the norm, Ramsey does not specify how it affirms the norm of treating voiceless dependents-as he frequently does in finding exceptions in other arenas of moral reasoning. In other words, there is a question about how Ramsey's deployment of the medical indications policy in the case of Joseph Saikewicz might generalize without its undermining the controls against treating nondying individuals as though they were dying that Ramsey developed his policy to institute. While Ramsey's application of the medical indications policy to the Saikewicz case is opaque, it could be sharpened in two ways. First, one might stress the medical nature of the burdens, rather than their prudential nature. Such a distinction is, as we saw above, one Ramsey develops in PP-but which he barely employs or develops in his account of when it might be licit to decline treatment for voiceless, dependent individuals. Had he made more use of it and amplified the medical burdens treatment caused Saikewicz, he might have carved a narrow path between resisting views that permit euthanasia on the one side and sanctioning the overtreatment of incompetent individuals on the other. To put the point differently, Ramsey might have emphasized the medical nature of the burdens in judgments about voiceless dependents-rather than discounting burdens out of a concern that they would slip into rejecting judgments of burdensome lives. Ramsey's terminology of "quality-of-life-during-treatment" in his analysis of Saikewicz comes the nearest in his mature work to specifying burdens of treatment in an explicitly medical way. By resisting the terminology of "burdensome treatments" because it drifts so easily into judgments about "burdensome lives," Ramsey correlates "burdens" with psychological and other prudential dimensions of "well-being" that often serve as rationales for declining treatment, rather than with the collateral damage that medical intervention causes an individual's organic, physical life. In short, Ramsey yields to his critic terrain he should have fought harder to preserve.
Second, Ramsey left the connection between the weight we give to particular categories of treatment and the weight we give to a person's individuating characteristics in deciding whether or not to treat a voiceless incompetent opaque. Ramsey's judgment that Saikewicz could reasonably be denied treatment is surprising, because it seems to violate the terms of the medical indications policy as Ramsey lays them out. On the one side, Ramsey allows a comparison of treatments in order to decide what is indicated. On the other side, we "ought not to compare and contrast the persons . . . with one another in other respects" (Ramsey 1978a, 192). Ramsey's policy precludes, for instance, determining to treat or not treat an infant on the basis of whether they are cognitively disabled (1978a, 190ff.). Yet in his discussion of the Saikewicz case, Ramsey does not seem to be able to escape some reference to Saikewicz's disability as contributing to the "burdensome treatments." So, Ramsey suggests that Saikewicz's "severe retardation" need not have been elevated "into the decisive consideration" (1978a, 316, emphasis mine). It might have instead "played only remotely a part through more proximate indications that were relevant" (Ramsey, 1978a, 316). Ramsey's difficulty is similar to that which he brought on himself in PP by arguing that some people might be beyond human care, and so serve as an exception to the norm to provide care for the dying (2002b, 163). In EE, Ramsey explained that because his is an "ethics of agent agape and, in medical ethics, a strong sense of agent care, [he] never suggested that one should base moral judgments in any degree upon an evaluation of the patient-subject as such" (1978a, 219). Ramsey acknowledges that the question of the agent's capacity to convey care has some connection to whether the patient-subject has any capacity for relation. But "instead of asking when agent morality reaches its limits or exceeds its capacity to convey care and a human presence to the dying, we now must ask primarily whether the patient-subject has 'any potential for human experience or relating'" (Ramsey, 1978, 226). Ramsey frames his own concern for the "voiceless" in his later commentary in similar terms: the "wisest rules of practice," he writes, "are those which, at any given time and place, best sustain a love not unlike that which God has shown all of us who-coram deo-are helpless and unable to speak in our own behalf" (Ramsey, 2002a, 152). Agape motivates us to raise the standards of protection for the voiceless-but when it comes to identifying when quality-of-life-during-treatment would be too burdensome for an individual, the nature of the treatment must make some indirect reference to the patient-subject and the particular qualities that make treatment burdensome. While Ramsey's medical indications policy offers more flexibility than it might initially seem to have, Ramsey might have secured similar protections without the conundrums that arise in his handling of the Saikewicz case by admitting in to the medical indications policy some reference to age as the backdrop against which treatments are assessed. Ramsey's emphasis on the equal value of every individual life entails a principle of equal treatment for individuals at every stage of life, such that the 3-year-old and the 70-year-old have the same claim on our resources. "All our days and years are of equal worth whatever the consequence; death is no more a tragedy at one time than at another," he writes (Ramsey 1978a, 191). Ramsey sets his account against those who would discount the claims of the very young relative to the old because their lives have so far "led to nothing along a time line of earthly achievement" (1978a, 191). 18 This principle of parity is, I think, partially responsible for creating the impression that Ramsey's medical indications policy would require overtreatment, especially of voiceless elderly individuals. Ramsey had argued in PP that appropriate treatment is not differentiated by youth or age. As his concern was over-treatment, though, he suggests there is no reason that might require anencephalic infants to be respirated, while not requiring it for elderly individuals. 19 In other words, we could decline treatment for both. As Ramsey's concern in EE flips to under-treatment, the "medical indications" policy seems to entail that the young and old are owed identical treatments, regardless of the resources available or the conditions of the quality-of-life-under-treatment that such treatments might respectively impose. The slow erosion of the body's capacities to self-repair that constitutes aging, though, means the burdens of treatment will in many cases (though not all) be worse for 60-year-olds than 6-year-olds. 20 If nothing else, individuals in their 60s take longer to heal from a particular surgery than individuals in their teens, thereby making it more plausible that an otherwise equivalent treatment would fail to "add life to their days." While Ramsey is acutely focused on the centrality of the body to bioethics, he never explores how ethics should address the aged reality of the body, its quality as a span of time. 21 Doing so would have changed the risk-threshold for decisions to treat or not treat: the benefits of a treatment in terms of the life it might give newborns would more easily outweigh the burdens of treatment than would be the case for an 80-year-old. While reasons to decline treatment that diminished the value of each life would be proscribed in both cases, a medical indications policy might allow declining treatment in a broader range of cases of 66-year-olds than it would in 3-yearolds. At the same time, such a policy only establishes the default for treatment: it is defeasible by other particulars in the individual's situation. After all, a 65-year-old smoker and alcoholic may be in worse physical health than an 80-year-old. While each individual might be equally valuable, the prior against which we weigh benefits and burdens is conditioned by the age of the individual being treated. 22 Though Ramsey does not consider whether age has a place in a medical indications policy, he does seem to allow for the selective distribution of resources on the basis of certain classes of patients-though not on the basis of comparing individuals. When considering the allocation of resources when not everyone can be saved, Ramsey contends that it is licit to stipulate that "certain sorts or categories of life-saving or life-prolonging technology will not be offered to anyone" (1978,263). Under such conditions, individuals who are denied treatment will only be singled out on the basis of the "category of illness that sadly our possible policy decides not to treat" (Ramsey, 1978, 263). Yet Ramsey takes one step further, and suggests that when confronted by finitude, a policy might make reference to classes of patients as a reference for determining who to treat: "A medical indications policy," he writes, "could go so far as to stipulate arbitrary lines to be drawn-for example, that no neonate below a designated weight and gestational age should be saved" (1978,264). However that line is drawn, physicians "could be free within limits to the one side or the other to try to save or not save the infant life" (Ramsey, 1978, 264). Such a policy counts as equal treatment because it is a categorical determination of a class of patients, rather than an individual one; as long as one patient does not have treatment determined on the basis of a comparison against another treatment, then no fundamental norms of justice or equality are violated (Ramsey, 1978, 264). 23 Insofar as Ramsey is serious about his medical indications policy being a policy, he leaves room for setting thresholds for treatment that also give doctors flexibility or individual judgments. He would have done well to make age a background consideration against which the particulars of a person's life and the benefits of treatment might have been weighed.

IV. KAVENY AND RAMSEY
The reconstructed account of Ramsey's medical indications policy I have advanced argues that there is more flexibility within the policy to decline treatments for voiceless individuals than critics have sometimes thought. If the policy requires treatment, it also admits of exceptions. Ramsey is aware that treatment might diminish an individual's quality-of-life-under-treatment, such that it would be reasonable to decline-provided that one does so for this reason, rather than on account of their mental incompetence, social status, or any other reason. Such a reading of Ramsey runs contrary to that which Cathleen Kaveny sets forth in her recent volume Ethics at the Edges of Law. Kaveny's central argument is that Ramsey's account of the law shifts, as cultural pressures move him to focus on protecting the sanctity of life to the exclusion of all other values. As she puts it, the fence Ramsey builds around mistreating incompetent individuals leaves "other values insufficiently protected" (Kaveny, 2018, 160). As a result, Ramsey overlooks highly relevant details about particular situations in the case of Joseph Saikewicz, details that might permit declining treatment in a much broader range of cases than Ramsey would admit. Kaveny points to the Saikewicz case as indicative of this problematic narrowness: Ramsey's failure to attend to the horrendous conditions of life at Belchertown State School for the Feeble-Minded distorts his moral judgment. The question of what treatment Saikewicz was owed cannot be "addressed without straightforwardly addressing the abusive situation in which he did his living" (Kaveny 2018, 160). Such a context demands, for Kaveny, taking into account the non-medical dimensions of an individual's life. As she writes, Ramsey "ought to question the sharp delineation of [medical ethics] from broader issues of moral concern." Kaveny's own reconstruction of the Saikewicz case focuses on those horrendous conditions, which prompts her to suggest that Saikewicz "would likely experience treatment as torture" (2018, 164). Ramsey's position, by contrast, imposes a "mask" on Joseph Saikewicz. Kaveny reads Ramsey's stance on the case through his suggestion that incompetent and competent individuals might be required to receive the same treatments, regardless of any individuating particulars of the former (2018, 164). Despite Ramsey's protestations that persons must not be treated as general categories, Kaveny contends that Ramsey fails to respect Saikewicz's individual, nonfungible personhood.
My reading of Ramsey, though, brings his "medical indications" policy nearer to what Kaveny thinks should be the case. Kaveny never acknowledges that Ramsey constructs an argument that the court could employ for withholding treatment in his analysis of the Saikewicz case. While Kaveny argues that Ramsey's monolithic interest in protecting life obscures the complexity of Saikewicz's position, Ramsey argues that it was the "great glory of American law to be constantly engaged in attempting such individual determinations" rather than reducing the case to what a statistical majority of others "similarly situated" might choose. His objection to the court is not that they decided Saikewicz did not need to be treated, but that they did so for the wrong reasons. In assessing Saikewicz's situation, they "made up their subjective minds, without foundation in any adequate discovery or finding or test . . ." (Ramsey, 1978b, 38, 40). 24 By treating Saikewicz "as if he were a competent patient who . . . would 'reasonably' take into account the fact that he was incompetent," the court brought "background quality of life . . . into the foreground" (Ramsey, 1978, 336). Additionally, as I argued above, Ramsey thinks the co-operation is central to treatment. Ramsey's account thus seems to be (slightly) less reactionary than Kaveny depicts it: it can generate the conclusion about Saikewicz Kaveny thinks is warranted, as his medical indications policy can only be applied when taking into account the conditions that will inhibit cooperation that Kaveny thinks are central to medical ethics.
At the same time, Kaveny's concern about the role age plays comports with my own worries about Ramsey's underdeveloped appreciation for the way age indexes thresholds for making determinations about treatment. Kaveny notes that Saikewicz was neither a "normal" 67-year-old nor a "normal" 3-year-old-he was an "elderly man with the mental age of a three year old." It is the peculiar combination of these two that combine to make the judgment to decline treatment reasonable. As Kaveny argues, valuing "each life equally does not entail refusing to recognize that there are different seasons of life" (2018,165). A normal 60 year old might be treated or not, for his own purposes-and the parents of a 3-year-old accept more unpleasant treatment "in large part because an entire life stretches out in front of the child" (Kaveny, 2018, 165). The equal value of human beings still allows us to differentiate the claim they make on our care. Yet, where Kaveny thinks such a judgment belongs in the particular case, I think Ramsey ought to have included it in his medical indications policy, as doing so would have simultaneously offered stronger protections for voiceless newborns and given more flexibility for declining treatment for voiceless adults like Joseph Saikewicz.
Yet methodologically, Ramsey is averse to expanding the scope of medical judgments for incompetent individuals to the extent that Kaveny thinks necessary. As I noted above, Ramsey has a narrow understanding of "health" and the task of medicine which is derived from his understanding of the value of bodily life: the life of an individual is incommensurable with other values, which quarantines the aims of medicine from being employed for social ends. Medical ethics on Kaveny's view, though, are "ultimately questions of social ethics. Medical ethics should not be treated as a rigidly defined or hermetically sealed field" (Kaveny, 2018, 164). Understanding the cruelty and inhumane treatment that Saikewicz likely experienced in his time in Belchertown State School "sheds crucial light on the end-of-life decision." Such a history would likely make Saikewicz "experience treatment as torture" (Kaveny, 2018, 164). While Ramsey admits that Saikewicz's putative lack of docility supplies a reason to not treat, his reluctance to allow social ethics purchase on medical decisions has real wisdom. Though Richard McCormick's position differs in important ways from Kaveny's, Ramsey's objection to his account of "quality of life" considerations could be applied without remainder against Kaveny's attempt to expand medical ethics into social ethics. Folding social concerns into medical judgments opens up the possibility that we might decline to offer medical care to escape the task of remedying the social problems that might make medical treatment burdensome. As Ramsey worries, declining treatment on such grounds is "one way to empty our mental institutions by attrition" (1978a, 301)-a temptation that becomes more attractive if we have already treated occupants unjustly, making them averse to cooperating with our care. On Ramsey's approach, such a stance would only add a second wrong to the first, rather than offer real relief. A history of mistreatment might make medical treatment seem like torture: but in such a case, the appropriate remedy is not declining to treat, but rather offering compensatory benefits and care while treating so as to ensure the experience of treatment is distanced from that of the mental institution. Keeping the medical and the social distinct, as Ramsey does, ensures that past institutional failures in one dimension do not become the grounds for declining to offer medical care in the other.
Additionally, accentuating age within a medical indications policy would help safeguard individuals from being undertreated for "social" reasons that have no bearing on whether the treatment might add "life to their years." Saikewicz's history of mistreatment might have made medical care seem like torture-but the same may be true for a 44-year-old with similar cognitive capacities and a similar history of institutionalization. If the decision to not treat the 66-year-old Saikewicz seems intuitively plausible, the decision to withhold treatment from the 44-year-old seems less so. If the benefits and quality-of-life-within-treatment are indexed by age, then the burdens of treatment are weightier in Saikewicz's case than a 44-year-old's case. If we distinguish our social and medical responsibilities in such cases, there would be considerably more time for a 44-year-old to receive compensatory care and comfort to offset the harms treatment might impose. We might also expect that the self-repairing quality of a younger body would shift the risk threshold for treatment: even if the odds of "success" are similar, the burdens of treatment are discounted for the young. In that way, the account of a "medical indications" policy I have offered secures Ramsey's conclusion that Saikewicz might have reasonably been denied treatment, while preserving strong protections for other voiceless, incompetent individuals who have reasons to expect to experience the treatment differently.

V. CONCLUSION: RAMSEY, BENEFITS AND BURDENS, AND INDIVIDUALS
Ramsey is averse to allowing evaluative descriptions of individuals' lives any place in determining the licit course of treatment, for understandable reasons. Yet, weighing the licitness of treatment can only be done with some reference to the individual being treated. The firewall Ramsey builds against abuse establishes two thresholds for declining treatment: competent adults may decline medically indicated treatments for vocational or prudential reasons if they otherwise choose life, while voiceless individuals must be provided any treatments that would "add life to their years." While this framework seems to require imposing burdensome treatments on voiceless individuals, Ramsey's response to Joseph Saikewicz indicates the policy is more flexible than it appears. It admits of exceptions: Ramsey acknowledges that the quality-oflife-under-treatment matters, and may be sufficient to withhold treatment from voiceless individuals. Ramsey's strategy of distinguishing medical from prudential reasons simultaneously secures the conclusion that withholding treatment from Saikewicz was just, even while it safeguards a policy from "emptying our mental institutions by attrition." Yet it does so only in an obscure way: Ramsey does not offer much to help us know why Saikewicz is an exception, or how we can generalize from his case to other exceptions. I have argued that age ought to have been a factor in Ramsey's medical indications policy, because it would have more easily secured protections for newborns while offering a more reasonable explanation for why it was licit to decline treatment to Joseph Saikewicz. While Saikewicz was owed better care than he received and a different environment in which he could live, his age made it unlikely that such restitution could be fully paid. That would not necessarily be the case, though, with someone 20 years his junior. If the burdens of treatment are weighted according to age, the threshold to decline treatment is much higher at 44 than at 66. Such an approach secures the highest level of protections for voiceless infants, who do not yet have a history on which their lives might be judged. Additionally, it is commensurate with our responsibility to provide any and all care that would add life to an individual's years, regardless of how meager the life might be relative to the number of years remaining. Now as we age, the burdens within treatment itself become more costly, and the possibility of compensating benefits diminishes. Structuring Ramsey's medical indications policy this way secures, I think, the conclusions he was concerned to preserve without requiring treatments that would be inhuman for their burdens.

NOTES
The 'sanctity of life' is among those "moral and religious premises" that he must make explicit (Ramsey, 2002b, xlv-xlvi). Three years after Patient as Person, Ramsey moves the theological foundation into the center of the expression: "Human beings are a sacredness," he writes, "under God, in the biological order" (Ramsey, 1973, 181). 7. In 1985, Ramsey reiterated and expanded this position in critiquing McCormick's understanding of commensurable values. When "we ask the meaning of good we are asking about the good for human beings, the good for their life," he wrote there; "when we ask about value, we ask about the value for human beings, the goods they serve and may attain." As such, life is not a value per se, but the condition of value. As the gift of God, life is constituted by "everyone's unrepeatable opportunity to praise and glorify God," and as the image of God, is "priceless, incomparable, unrepeatable." That entails that embodied life is incommensurate with other values: "Because a human life is so basic, it is therefore set apart" (Ramsey, 1985, 92-3).
8. While time and capabilities are in some respects related, they are distinct. A missing finger might not reduce a person's lifespan, but it would reduce the ability to perform certain functions. Ramsey does not develop an account of human capabilities, but something like it seems necessary in order to make a "medical indications policy" serviceable. He also includes, though, preventing a "more severe impairment of biologic function" as a legitimate grounds for intervention. Such a standard would presumably include interventions like ventilation or hydration for a person in a coma, which extend a person's life but also forestall worse impairments.
9. In his writings about bioethics, Ramsey was especially worried about the instrumentalization of the body for social or psychological ends. In his queries about organ donation, for instance, Ramsey worried that there might not be anything left "of the notion that a human being has (or better, is) a bodily integrity, and that out of the respect due also to this there are some actions that must be judged to be wrong . . ." (Ramsey, 2002b, 190).
10. Ramsey's concern here is almost the inverse of PP, where he had bemoaned the hiddenness of death on children as a reason why secularism is advancing. Ramsey writes that secularism is expanding in part because childhood is "unhappily sheltered from the dying in all our advanced societies." See Ramsey (2002b, 136, and also 156).
11. As Ramsey notes in a later essay, he "made a career out of discussing exceptions and writing them into rules!" While he thinks his medical indications policy for voiceless dependents is the "wisest and most prudent rule of practice for medicine in a day when considering burdens of treatments for voiceless patients had-and has-become a method for judging their lives too burdensome to be worth sustaining," he also thinks his rule points us to what the received norm of ordinary/extraordinary treatments "meant us to affirm" (Ramsey, 2002a, 152).
12. The medical indications policy constrains medicine until "there is a patient capable of himself refusing treatment lest his established life projects be overwhelmed by continuing the struggle for survival" (Ramsey, 1978, 186).
13. Ramsey stresses the fact that those patients who are "participating in treatment decisions are rendering comparative judgments-their own judgments-about a range of life choices." Provided that they choose life, they do no wrong. It is not a choice for a third party, or between their life and no life at all (Ramsey, 2002a, 151). As such, Meilaender's use of "vocation" to describe Ramsey's focus is apt.
14. These are not predictive claims on Ramsey's part, and as such cannot be dismissed as an illegitimate "slippery slope" argument. Ramsey is obviously concerned about the putative degradation of human life that he sees around him. And in that sense, his reasoning here is motivated-as I acknowledge in the opening of the article. But, his focus is on the logical outcomes of the principles and concepts embedded in certain forms of reasoning. Ramsey makes this explicit in his critique of the Quinlan opinion, pointing out that he has "not used an external or causal slippery slope argument," but that "ample concepts or principles have been adduced from the opinion that are quite capable of greasing the slope or driving the wedge deeper." This problem is especially acute in legal contexts, where precedents "replicate themselves" (Ramsey, 1978a, 296).
15. The standards are originally Warren Reich's, but Ramsey suggests they are equivalent to his medical indications policy. 16. Surprisingly, Meilaender (1991) does not consider the case. 17. Ramsey fills out his objection to the court's ruling in an essay that was published just after EE. See Ramsey (1978b).
18. See also Ramsey (1978a, xii): "I see no conclusive reason for saying that six months of babyhood or 2 years of infancy are of less ultimate worth than 60 years of manhood or womanhood." 19. The position is continuous between PP and EE. In his prior work, Ramsey had suggested that it is "not obvious that an anencephaletic baby should be respirated while a grown man in prolonged coma should no longer be helped to breathe" (1978a, 132). 20. As an astute anonymous commentator observes, many 60-year-olds will have resources to maintain their "quality of life during treatment" that a 6-year-old might not. My suggestion is that if the burdens of treatment count-as Ramsey seems to permit in his analysis of Saikewicz-they may be variably weighted according to (in part) the age of the patient. My gratitude to the commentators for their astute critiques. 21. As Meilaender (1991, 141), observes as well, Ramsey was heavily influenced by Barth. He cites his anthropology favorably at various points in his corpus, and learned his understanding of the "exception" from him. See, for instance, Ramsey (1968, 75ff). Yet Ramsey never seems to interact with, or have learned from, Barth's discussion of humanity's nature as a temporal creature in Church Dogmatics III/2, §47-never mind that such a section structures Barth's end-of-life ethics in important ways. 22. A Ramsey-an account of how age matters in a medical indications policy would have to be consistent with his rejection of treating bodily life as commensurable with other goods, and so as capable of being violated for the sake of those other goods. Ramsey's focus on the body as the lived reality of the person and as the criterion for medical action would plausibly entail, I think, that his account of "age" would be antithetical to the "biographical" account that Daniel Callahan has set forth. For Callahan, a medical notion of age is unreliable as a guide for treatment. Age instead is a "biographical trait" that does not reduce to the possession of 'age-associated' traits. Callahan's emphasis on the biographical aspects of age rests on his rejection of Leon Kass's understanding of medicine as ordered toward the health and wholeness of the body-a view that seems to come very close to Ramsey's understanding of the body and of medicine's aims. While Ramsey might allow for a "biographical" standard for determining treatment for competent adults, as they are capable of determining the significance of age to themselves, for incompetent adults he would have to adopt a medical standard. That is, I take it, the force of the "medical indications" policy. See Callahan (1995, 61-5 and 164-71). My thanks to an anonymous commentator for bringing Callahan to my attention.
23. In a footnote, Ramsey argues that declining to bring aid based on the categorization of cases is commensurate with an ethic in which killing is only protective, and would not imply that one could kill a particular person who is "arguably a social liability or who may even be preventing us from bringing lifesaving treatment to others." Ramsey's policy follows Gene Outka, who distinguishes between the grounds for receiving care and the grounds for distributing it. Ramsey's stance here is more flexible than would be permitted in contemporary culture wars, as it seems to permit the kind of rationing of resources that animated the infamous "death panels" discussion in disputes about the Affordable Care Act.
24. It is the combination of the singularity of the court's focus, their focus on a "subjective" reason, and their acknowledgment of Saikewicz's incompetence that leaves the court without the resources to make the right judgment with the right reasons. As Ramsey writes, "it seems, then, that the substituted judgments of courts of equity must be utterly subjective because utterly singular to the individual case, unless the court can discover some specific trait or interest of the incompetent on which to base its decision. This once was the doctrine of substituted judgment, I suggest; courts purported to find an incompetent's interest. They did not project the unknowable into the unknown. They did not impute, with no specific facts to guide them, to incompetent subjects the courts' judgment concerning what these subjects would choose if they were rational and took into account their irrationality" (Ramsey, 1978b, 39).