terça-feira, 3 de dezembro de 2013

Texto a ser lido, criticado e pensado.

The Tortured Patient

A Medical Dilemma

Chiara Lepora, Joseph Millum
The Hastings Center Report 

Abstract and Introduction


Doctors sometimes find themselves presented with a grim choice: abandon a patient or be complicit in torture. Since complicity is a matter of degree and other moral factors may have great weight, sometimes being complicit is the right thing to do.


Torture is unethical and usually counter-productive. It is prohibited by international and national laws. Yet it persists: according to Amnesty International, torture is widespread in more than a third of countries.[1] Physicians and other medical professionals are frequently asked to assist with torture. For example, a recently declassified report from the Central Intelligence Agency on interrogation at Guantanamo Bay states: "OMS [Office of Medical Services] provided comprehensive medical attention to detainees … where Enhanced Interrogation Techniques were employed with high value detainees."[2]

Physicians and Torture

The United Nations Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment sets out a basic definition of torture:
the term "torture" means any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind, when such pain or suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity. It does not include pain or suffering arising only from, inherent in or incidental to lawful sanctions.[7]
Torture therefore encompasses cases ranging from exposing a prisoner to electroshock to extract information, to beating or slapping to "induce surprise, shock, or humiliation,"[8] and cutting off a prisoner's healthy ear or limb as punishment.[9]
Prohibitions on physicians participating in torture are a relatively recent development. From the Middle Ages through to the modern era, physician involvement in torture was a professional requirement. This ended only when torture itself ceased to be legally and socially acceptable.[10] In the last century, international agreements prohibited all forms of torture.[11] The prohibition on torture, including complicity in torture, was explicitly extended to medical professionals. For example, Article 3 of UN Resolution 37 states:
It is a gross contravention of medical ethics, as well as an offence under applicable international instruments, for health personnel, particularly physicians, to engage, actively or passively, in acts which constitute participation in, complicity in, incitement to or attempts to commit torture or other cruel, inhuman or degrading treatment or punishment.[12]
Medical participation in torture is similarly condemned by all professional codes of ethics, including the World Medical Association's Tokyo and Malta declarations, the American Medical Association's Resolution 10, the American College of Physicians' conclusions and recommendations, a joint position statement from U.S. psychiatry and psychology associations, and the World Psychiatry Association's Madrid Declaration.[13]
Legal, ethical, and medical condemnation have not been as effective as their proponents hoped: torture is widespread in more than a third of countries,[14] and medical implication is described in at least 40 percent of reported torture cases.[15] Doctors are frequently required to be on hand for acts ranging from falsifying death certificates to the amputation of detainees' limbs.

Some of these doctors may simply be engaged in torture, or at least sympathetic to the aims and methods of the torturing regime. But others who oppose torture find themselves in a difficult situation. While the torturers may ask them to provide some form of medical attention for purposes unrelated to the prisoner's health, the prisoner may actually need that medical attention in order to be properly treated. In some circumstances, a prisoner may be better off cared for by a doctor, despite the complicity entailed. If the state is going to amputate a limb as punishment, regardless of the international prohibitions, it is surely better for the victim that the amputation be performed in a surgical theater, under anesthesia administered by a qualified surgeon, than without anesthetic in the public square by an untrained official. Thus, doctors may be conflicted about the right course of action to take.

This conflict also arises from the international instruments and codes of medical ethics. While they extend a blanket prohibition on all forms of participation in torture, they also exhort physicians to treat the interests of their patients as a guiding concern. The same U.N. resolution that condemns medical complicity in torture also states that "Medical and other health personnel have a duty to provide competent medical service in full professional and moral independence, with compassion and respect for human dignity, and to always bear in mind human life and to act in the patient's best interest."[16]
The tension between these two directives has been neglected by the substantial literature addressing the ethics of torture and medical complicity. A literature search of philosophical, medical, and legal journals over the last ten years yielded more than four hundred papers that mentioned "physicians" and "ethics" along with "torture," "interrogation," or "forced treatment."[17] But despite deep and divergent views, only a couple of publications present the issue of medical participation in torture as any sort of dilemma;18 the majority propose or repropose exceptionless prohibitions on physician complicity in torture,[19] discuss whether specific mentioned acts are indeed tantamount to torture,[20] argue about whether torture is justified in some exceptional cases when national security is threatened,[21] or consider whether medical participation is necessary and even morally required for some cases of torture.[22]

Complicity and Wrongdoing

Before we can address the specific problem of medical complicity in torture, we need a clear analysis of what it means to be complicit in wrongdoing. The most basic case of complicity in wrongdoing involves a principal actor who carries out a wrongful act and an accessory who does not actually perform the wrongful act but is in some way involved in it.[23] Complicity comes in degrees: someone can be more or less complicit in an act. The degree to which someone is complicit is a function of two factors: assistance and shared intention (corresponding to the Catholic concepts of material and formal complicity24). Assistance is a function of the complicit agent's expected causal contribution to the act. Shared intention is a function of the extent to which she has the same wrongful ends as the principal.

The idea of assistance should be relatively straightforward, even though exactly how to measure the extent of someone's causal contribution is complex. The intuitive notion is that the more the complicit agent's acts are expected to help in achieving the wrongful ends, the more complicit she is. (Of course, as with other cases of moral responsibility, it must be the case that she acts voluntarily and that she knows, or should know, that she is assisting the wrongful act.) Consider an arms dealer who sells weapons to terrorists: the more weapons he sells them, the greater his complicity in the acts they perform with the weapons. Or, to take a medical example, contrast two psychologists who examine a prisoner and record their assessment in his medical records, knowing that the records will be read by the torturers. One psychologist reports the patient's extreme fear of spiders; the other reports only that the patient suffers from anxiety disorder. Although both reports are technically correct, the first, by giving the torturers specific information, thereby helps them more with their interrogation. With the information she gives them, the torturers are able to exploit the prisoner's fears: confining him in a cramped box and inserting insects. Such an experience was designed by interrogators at Guantanamo Bay.[25]
Complicity is not just a matter of voluntarily and knowingly providing assistance to the principal's wrongdoing; the intentions with which the accessory acts are important, too. To amend a famous example of Bernard Williams, there is something morally better about the actions of George, who takes a job at a chemical weapons factory as a last resort to pay his bills, than Henry, who takes the same job because he wants to advance the effectiveness of chemical warfare.[26] Focusing on whether intentions are shared allows us to distinguish a case of two people who are engaged in the same activity (even if their actions take place at different times) from a case in which the accessory's acts simply enable or make it easier for the principal to engage in the activity. This explains the different intuitions about the chemical weapons employees. It can also explain why simply being associated with an activity, without causally assisting it, may entail complicity. Suppose Victor joins a neo-Nazi party (again, voluntarily and knowingly). He may then be judged complicit in the racially motivated violence it incites even if he does nothing to facilitate it himself. A natural explanation of why we regard him as complicit is that his membership signifies that he shares the party's goals. Similarly, a doctor who agrees to attend a waterboarding torture session is complicit in torture regardless of whether she actually intervenes at any point in the process, since her presence can be plausibly interpreted as implicit endorsement of the procedure.

What does it mean to share intentions? To share someone's intentions is to act for the same reasons as that person. Thus, if we are dance partners and you step left in order to waltz and I step right in order to waltz, then we share the joint intention to waltz. Likewise, if one person plants the bomb in the basement and his partner lights the fuse, they share an intention to blow up the building.[27]
Complex acts like torture involve a number of distinct intentions. The torturer must intend each of the component acts that constitute an instance of torture—for example, to secure the prisoner's restraints, attach the wires, check the circuit, turn the switch, and so on. Moreover, the same act may be performed with multiple intentions, under different intentional descriptions; for example, the torturer may turn the switch in order to make the current flow, but also in order to cause the prisoner pain and in order to make him give up information. This entails that, depending on the number of component intentions that are shared, it is possible to share the intentions of another to a greater or lesser degree. Thus, as with providing assistance, complicity through shared intention comes in degrees, depending on how many of the intentions to commit wrongful acts are shared. This will prove important when we consider the different motivations that might lead a physician to be complicit in torture.

To summarize, there are two dimensions to complicity, assistance and shared intention, both of which are a matter of degree. Most cases of complicity involve someone being complicit to some degree on both dimensions, though it is possible to be complicit only by assisting or, through acts with symbolic meaning, only by sharing intentions. Roughly speaking, the further along each dimension one lies, the greater one's total complicity. How bad it is to be complicit in a wrongful act is then a function of both the extent of one's complicity in that act and of how bad the act is (since the wrong of complicity is derived from the wrong of the act with which one is complicit).

s Complicity in Wrongdoing Always Wrong?

When someone is complicit in wrongdoing, she does not herself commit the wrong. Thus, the wrongfulness of the primary act does not entail that the complicit act is itself wrong, all things considered. The act may have other features that speak in favor of it; for example, it might be expected to produce a greater balance of benefits over harms than other acts. Alternatively, it may be the best option among the choices available to someone, all of which are problematic. Moreover, as we just saw, complicity comes in degrees. Someone's actions could be only slightly complicit in wrongdoing (and so, depending on the principal's act, only slightly prima facie wrong). It is therefore possible that other morally relevant features of a complicit act could outweigh the wrong of complicity and make that act permissible or obligatory, all things considered.
If the torturing authorities demand that a prisoner be treated and the prisoner also asks for treatment, then the doctor, in treating, will inevitably be complicit in the torture. But if she treats because of the prisoner's request and not the torturer's, the degree to which she is complicit will be low.
This theoretical point can be illustrated with a well-known example. Oskar Schindler was a member of German Military Intelligence and a businessman who took advantage of the German invasion in 1939 to acquire a bankrupt Polish factory. Schindler created strong and long-lasting friendships with members of the Wehrmacht and the SS, and became their trusted source of cognac and cigars. Until his encounter with Itzhak Stern, a Jewish accountant, Schindler exhibited interest only in business. As a respected and well-connected member of Nazi high society, Schindler was able to hire and keep Jewish workers in his factories, eventually saving more than 1,200 from deportation and death.[28] His workers were glad of his position and requested that he maintain it.

There is no doubt about Schindler's early complicity in the Nazi regime and the ongoing war, which he fueled with the products of his factories. But, on the commonly held assumption that the good he did by saving Jews outweighed the negative consequences of his compliance with the Nazi regime, there is also little doubt that Schindler did the right thing. Given the circumstances, he would have been mistaken to refuse complicity and thus be unable to help his employees.

Cases like Schindler's show that complicity in even the most heinous of acts may not be wrong, all things considered. The prima facie wrongness of complicity in wrongdoing can be outweighed by other moral reasons in favor of the act. But this can apply to medical complicity in torture, just as it did to Schindler's complicity in the Nazi war machine. In certain circumstances, patient-centered considerations will be important enough to outweigh complicity in torture.

The following sections elucidate the two moral considerations that we regard as most important in the context of medical complicity: consequences and patient preferences. We then consider how these considerations relate to a doctor's potential complicity in torture.


The consequences of our actions clearly affect their moral evaluation. In Schindler's case, the good of helping 1,200 people survive was sufficient to outweigh the wrong of being complicit with the Nazis. Likewise, there will be a point at which the beneficial consequences of an act that is complicit in torture will outweigh the prima facie wrong of the complicity. However, exactly how and how much consequences matter in moral decision-making is controversial. It is notoriously hard to weigh the importance of different states of affairs against each other, let alone against very different values, such as avoiding complicity. Here we have space only to indicate the types of consequences that ought to be taken into account.

Three broad classes of relevant consequences may be distinguished: personal consequences, consequences for the prisoner, and social consequences. Personal consequences are those that affect the doctor herself (or other people who are significant in her life). Some should clearly not be given moral weight. For example, if a doctor stands to profit or to be promoted as a result of her complicity with a torturing institution, this is no justification for complicity at all. On the other hand, credible vital threats to the doctor or her family might excuse her complicity. The Iraqi doctor who was executed for refusing to participate in torture might have done a noble thing,[29] but many people would judge his action beyond the call of duty—where someone is threatened with death, his complicity in acts he cannot prevent is excusable.

Such reasoning should not be taken too far, though. The fact that we excuse people who assist in wrongdoing when they are under great pressure should not be taken to excuse all actions taken under any pressure at all. Doctors should accept moderate risks in the service of right action. Quite apart from the general duty that people have to accept moderate risks to preserve the rights of others, physicians are usually thought to have special duties to take risks for the sake of their patients—for example, by risking exposure to nosocomial infections.[30]
Whether a doctor should be taking personal risks by refusing to cooperate also depends on the consequences of her cooperation or refusal for other parties. Consider those occasions when the complicit acts that doctors are asked to perform are also in the medical interests of the prisoner being tortured. For example, the surgeon who is asked to perform an amputation as part of a court-ordered punishment may rightly judge that the prisoner will be better off if she complies than if she refuses and leaves the punishment in the hands of someone with no medical training. Benefits to the prisoner should count in favor of doing as the authorities request. However, what counts as being in the prisoner's interests is a complicated question: medical benefit does not exhaust what constitutes well-being, and frequently, what someone subjectively values makes a difference to what is good for him.

A doctor's complicity in torture may also affect the interests of people outside the doctor-prisoner dyad, and doctors should also take into account these broader social consequences. This point is not about the possible social benefits of torture—we assume that torture is wrong and also that it is not socially beneficial.[31] Instead, the issue is about the possible political consequences if doctors refuse to be complicit. For example, one might argue that an effective physician boycott of all forms of association with torture might limit a government's ability to torture.[32] If a doctor's refusal to comply can have a foreseeable impact on whether torture occurs, then she ought to take this consequence into account. In many cases, however, the social benefits of noncooperation are likely to be speculative at best: a doctor will often lack any real evidence concerning the beneficial or harmful long-term effects of her actions. In such cases, she should not neglect someone's immediate medical needs.

Prisoner Preferences

In considering the consequences of complicity, the interests of the victims are of great importance. However, as in standard cases of medical care, a physician's judgment of what is in a patient's interests may not be sufficient for her to decide whether and how to treat him. Instead, where a patient is competent to make decisions about medical care, his own treatment preferences should normally be respected.[33] This is for three reasons: first, because people are usually knowledgeable about what is in their own interests; second, because what people value partly determines what is in their interests; and third, because respect for autonomy extends to respecting a patient's decisions about what is or is not done to his body.

Consider the following case. A doctor is called to provide treatment to a prisoner who has been severely beaten during interrogation. The prisoner's current prognosis is quite poor but could be significantly improved with immediate, expert treatment. However, if the prisoner's health improves sufficiently, then the doctor expects that he will be tortured again. Should she treat him or leave him? It seems to us that this question cannot be answered without finding out what the prisoner wants. Only he can decide whether it is preferable to survive and be tortured, or to avoid further torture but increase his chances of dying. Further, by soliciting and following his decision, the doctor allows the prisoner some degree of control over what happens to him, and thereby respects his autonomy.

Doctors might wonder how standards of care and informed consent can possibly be respected in a setting such as a prison, where obvious violations of rights are being perpetrated and where open complaints about torture may be punished. Several eyewitness accounts of doctors involved in torture report the presence of security guards at medical examinations. Nonetheless, in most cases, doctors remain able to talk to their patient-prisoners, and they are able to ask whether they wish to receive medical care.[34] For instance, in the case quoted at the beginning of the article, a doctor is reported to have examined the prisoner more than twenty-five times and conversed with him on more than half of those occasions.[35] Admittedly, eliciting treatment preferences from prisoners in places where they are tortured is unlikely to reach the same standards for informed consent that we aim for in more typical clinical care. But it is still far better for doctors to seek their patients' views to the best of their ability than to ignore them entirely.

What should a doctor do if the prisoner is unconscious? In such a case, she should follow the same principles laid out in guidelines for emergency rooms and for the treatment of hunger strikers: in the absence of an expressed preference from the patient, the doctor should promote what is in the presumed best medical interests of the patient.[36] However, if and when the patient is conscious and competent, his preferences trump the principle of medical beneficence. Once he has been revived, these preferences should be elicited.

Someone might object that patients who are also prisoners do not have medical rights as extensive as those of other patients, and so their preferences should not always be respected even when they can be elicited. For example, prisoners may not refuse treatment for a medical condition such as active tuberculosis—a condition that poses a risk to other inmates or to the security of the institution. But such limits on the right to refuse treatment are no different than limits that also apply to nonprisoners living in confined settings.[37] Both the Geneva Convention on the rights of war prisoners and the preponderance of U.S. case law reaffirm that competent prisoners should be afforded the same rights to refuse treatment as patients outside a prison.[38] Furthermore, doctors and other medical personnel have a duty to provide care to prisoners at the same standards as for nonprisoner patients.[39]
Finally, one might object that talk of autonomy is misplaced in the context of torture. If the patient is not only a prisoner, but a prisoner who has been or will be tortured, then one might argue that she faces too much coercion to be capable of autonomous action. However, this objection conflates autonomy with liberty. Someone is autonomous—in the sense that his choices should be respected—when he is capable of reasoning about what to do in the light of his values and making decisions on that basis. This is a capacity that does not rely on having the ability to carry out his decisions—that is, on having sufficient liberty. So long as the prisoner is capable of making an autonomous choice about his care, that choice should be respected; the fact that his liberty is very constrained is no reason to deny him this piece of control over his life.

Patient-centered Reasons and Complicity in Torture

In working out the ethics of a particular complicit act, it is important to note the relationship between respecting the prisoner's welfare or preferences and a doctor's degree of complicity in torture. To return to the previous example, if the prisoner asks for treatment, the same action is simultaneously the one requested by the torturing authorities and by the prisoner. If the doctor wishes to carry out the prisoner's will (which is what is involved in respecting someone's autonomy), then she must do what the torturers request. Inevitably, then, she will be complicit in the torture. However, if that the doctor treats the prisoner just because it is the prisoner's request, then the degree to which she is complicit will actually be quite low. This is because her intention is not to have the prisoner tortured, but to follow his health care wishes. (This assumes that if the prisoner asked for treatment that differed from what the authorities had requested, then the doctor would follow that course instead, and if the authorities requested treatment contrary to the patient's wishes, then the doctor would refuse.) Thus, in these cases, the doctor may provide some assistance to the torturers, but, not sharing their wrongful intentions, she is minimally complicit.

This case can be helpfully contrasted with an alternative motivation. Consider a second doctor, who does as the torturers request and treats the prisoner because that is what she is paid to do. Imagine this doctor defending her actions by pointing to her benign intentions: "I was just doing my job—I didn't want the prisoner to be tortured!" Such a defense would seem fake, and our earlier analysis of complicity can explain why. This doctor may indeed have the ultimate goal of being paid. But a necessary proximate intention for reaching this goal is that she carries out the orders of her superiors, and this requires that she intentionally facilitate torture. Thus, she intends a wrongful act: helping people carry out torture. Our first doctor, on the other hand, need not intend anything of the sort. She does what the torturers request, but not because they request it, and so need not share any of their wrongful intentions. Her contribution is only instrumental.

These are fine distinctions, but important: with them we can separate hypocritical doctors who are really part of the torturing institution from doctors who are struggling to serve their patients under difficult circumstances.

Potential Objections

Someone might accept the analysis given so far, agree that ordinary people faced with difficult dilemmas like the ones we describe sometimes ought to be complicit in wrongdoing, but deny that the analysis applies to physician complicity in torture. Physicians have general ethical duties like everyone else, but they have additional special duties in virtue of being physicians. (Similarly, nurses, psychologists, and so forth each have their own role-based duties.) Some commentators believe that these duties imply that they should never be complicit in torture.[40] For example, some argue that the physician's role as healer entails that she has a special duty to refrain from actions that cause harm, and this includes any form of support for torture.[41]
We believe that such objections miss the force of the problem with which we began. The dilemmas we describe arise because different principles, all of which are internal to the role of the physician, come into conflict. Codes of professional ethics give physicians duties to act in the interests of their patients (even at some risk to themselves), to respect patient autonomy, and to refrain from any form of association with torture. But sometimes it is not possible to fulfill all of these duties at once. Reference to the role morality of physicians therefore does not resolve these dilemmas; rather, it shows why they are so difficult.

A related possible objection is that complicity in torture could require doctors to sacrifice their personal integrity. Here the objection is not that complicity in torture is inconsistent with the values that make up the role morality of a physician, but that it may be inconsistent with the deeply held values of individual physicians. Arguments like this have been developed to defend limited forms of conscientious objection for physicians,[42] and to argue against moral theories that require individuals to sacrifice their personal projects whenever doing so could attain a greater good.[43] In both cases, the form of argument is the same: to ask someone to act in a way that is inconsistent with her deeply held values threatens her identity as a moral agent. Hence, people have a prerogative not to act in such ways. Might a physician legitimately refuse to be complicit in torture on the grounds of personal integrity in cases like the ones we describe? Maybe, but such a refusal is neither easily defended nor morally decisive if defended.

Note first that an appeal to personal integrity must cite more than the doctor's moral opposition to torture. The arguments of this paper start from the premise that the torture we are considering is immoral, and we assume that the physicians we address agree with this judgment. We have argued that even if this is true, there are cases in which a physician ought to act in a way that is complicit in acts of torture. Someone who rejects this conclusion on the grounds of integrity must therefore argue that there is something particular about her values that makes acts complicit in torture worse for her than for other similarly situated people. Further, she must argue that complicity in torture would violate her integrity more than would abandoning a patient in need. After all, another doctor may be equally appalled by torture yet believe that she ought to act in a way that minimizes the damage torture causes, whether that makes her complicit or not.[44]
Second, even in a case in which we can make sense of someone appealing to her integrity in spite of our arguments, it does not follow that she is ethically permitted to refuse to be complicit. Even those philosophers who defend the importance of integrity acknowledge that there can come a point when other factors outweigh the importance of maintaining integrity and that an agent therefore ought to act contrary to her personal values.[45] Hence, integrity becomes just another of the considerations that must be factored into the complex moral calculus and weighed with the disvalue of complicity, the consequences of different courses of action, and the patient's preferences.

Moral integrity is an important concern, and one that should not be dismissed out of hand. But the appeal to integrity in the face of another's wrongdoing is neither always applicable nor decisive where it is applicable.

Dealing with Medical Complicity in Torture

Other things being equal, it is better for a physician not to be complicit in torture. But other things are rarely equal, and as we have argued, a physician ought sometimes to accept complicity in torture for other moral reasons. Even in such cases, however, she should do what she can to minimize her complicity in wrongdoing. This can be achieved by assessing and minimizing the two component parts of complicity: shared intentions and assistance.

The first important way to minimize complicity is to ensure that wrongful intentions are not shared with the wrongdoers. In the case of medical complicity in torture, this may be achieved primarily through the doctor taking as her intentions just those reasons that justify her complicit actions. If, for example, the reason that she should provide immediate supportive care is that this is in the medical interests of an unconscious patient, then she should be resuscitating him only because it is in his interests. Or, if the reason that she should treat a condition that would otherwise preclude the patient from interrogation on medical grounds is that this is exactly what the patient requested, then she should be treating him because it is what he requested. The physician and torturers may then share some of the same subsidiary goals, such as keeping the patient/prisoner alive, but will have quite different ultimate goals, whose moral evaluations are diametrically opposed.

The second feature of complicity concerns the assistance provided by the physician to the torturer. Consider the example of a doctor who is asked to provide a certificate of fitness for a prisoner. The doctor is aware that her certificate will be used to tailor the torture to the prisoner's health condition, so that it will be as "effective" and "safe" as possible. She also knows that refusing to write the certificate would put the patient at undue risk because (let's say) of a preexisting heart problem. In the course of the doctor's routine examination in the prisoner's cell, with a guard waiting outside, she asks the prisoner whether he wants to receive medical care. When the prisoner expresses a strong preference to be kept alive despite the torture, the physician accepts her complicity and writes the certificate mentioning the heart condition. In this case, however, in order to minimize complicity, the physician should not write a standard certificate, which would cover all aspects of the patient's health and might therefore unnecessarily expose weaknesses to the torturers. Instead, she should focus her report on the risks of death the patient would be exposed to, and avoid any additional information that might abet the torture, such as the patient's fear of death.

A physician can further reduce her complicity if, while complicit, she carries out acts that mitigate, prevent, or help redress acts of torture. For example, one way to compensate for complicity is to secretly collect data that can be used for reporting the occurrence of torture and to provide them to investigative bodies as soon as possible. Where physicians have been coerced into assisting with torture, they have often been among the first sources of essential information for international tribunals pursuing justice.[46]
Medical associations also have a role to play in dealing with medical complicity. Medical participation in torture is blankly condemned by all associations, all professional codes of ethics, and a majority of legal codes worldwide. Should these codes be changed, given the arguments in this paper, to reflect the complexities faced by physicians working in extreme conditions? Alternatively, should these codes be strongly enforced in every case, despite the ethical reasons some doctors may have to be complicit in torture? We believe that both of these options would be mistaken.

First, we do not think that these arguments provide sufficient reason to alter the clear, simple rules currently promulgated in the codes. The value of these rules is threefold. First, they constitute a powerful condemnation of torture. Second, they have an aspirational character: they look forward to a world in which there is never a reason for a medical professional to be associated with torture. And third, they provide a defense for doctors who should not be involved in torture, and should be able to cite binding rules that forbid them from being involved.[47]
However, the enforcement of the codes is a different matter. History suggests that proper enforcement of the prohibition on medical participation in torture is very unlikely. From the Nuremberg trial to the present, only thirty-five physicians are known to have been held accountable for involvement in torture—a trivial number compared to the number of physicians reported as being involved, and even more trivial compared to the number of physicians who have been involved in torture but have not been reported at all.[48] But even if enforcement were possible, and so physicians who were involved with torture could expect to be excluded from the medical community, this would not fully solve the problem. Excluding from the medical community any physician who assisted with torture, no matter what the justification, would penalize physicians who have to work in countries where torture is widespread and would be unfair to doctors willing to compromise themselves for the sake of their patients. These considerations suggest that a more nuanced, case-by-case approach to enforcement would be much preferable and have a greater prospect of being effective.

One possible option would be to create an international self-reporting system—a sort of "ethical ombudsman" whom physicians could confidentially approach to report cases of coercion or special circumstances that prompted medical complicity in torture. Such a system could provide the necessary support for physicians who face complex choices and strengthen their witnessing capacities for international tribunals. It would also constitute a body that could help differentiate cases that require and deserve support from the plainly criminal cases of willing or careless participation in torture.

This is just one suggestion; the key point is that whatever system is used, it should be designed to take into account the ethical complexities of the situations in which doctors can find themselves when they work in contexts where torture takes place. While it may be unflagging in its denunciation of torture, it should provide support to doctors who want to do the right thing in difficult circumstances.
Physicians who assist in torture without regard for its victims may rightly be condemned. However, doctors sometimes find themselves presented with the grim choice of either abandoning a patient or being complicit in torture. Such doctors face a genuine ethical dilemma. Here, we have outlined the factors that should be considered when deciding how to respond to these dilemmas: the expected consequences of the doctor's actions, the wishes of the patient, and the extent of the doctor's complicity with wrongdoing. Since complicity is a matter of degree and other moral factors may have great weight, sometimes the right action involves medical complicity in torture. Consequently, the problem of medical involvement in torture will not be resolved by blanket denunciations of complicity. Instead, associations of medical professionals should take into account the circumstances we have described and provide more supportive and efficacious systems of reporting for medical professionals who face such dilemmas.